This content has been downloaded from IOPscience. Please scroll down to see the full text.

Download details:

IP Address: [Link]
This content was downloaded on 06/03/2022 at 01:11

Please note that terms and conditions apply.

You may also like:

Physics in Food Manufacturing

Modeling and Computation in Vibration Problems, Volume 1

The National Mechatronic Platform. The basis of the educational programs in the knowledge society
V Maties

Design of mechatronic simulators to improve the quality of students learning outcomes in

mechatronics subject
J Kustija, H Hasbullah and Y Somantri

Conceptual Design of Mechatronics System Based on Intelligent Control

Jinming Li and Lining Liu

Research on Virtual Platform of Electrical Control of Mechatronics Equipment Based on VRML

Chongqian Huang

Integrated DC Electrical Machine for All-Electric and Hybrid-Electric Vehicles

J Tutaj and B Fijalkowski
 Mechatronics
Dynamical systems approach and theory of holors
 Mechatronics
Dynamical systems approach and theory of holors

B T Fijalkowski
Cracow University of Technology and
State Higher Vocational School in Nova Sandec, Poland

IOP Publishing, Bristol, UK

ª IOP Publishing Ltd 2016

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system
or transmitted in any form or by any means, electronic, mechanical, photocopying, recording
or otherwise, without the prior permission of the publisher, or as expressly permitted by law or
under terms agreed with the appropriate rights organization. Multiple copying is permitted in
accordance with the terms of licences issued by the Copyright Licensing Agency, the Copyright
Clearance Centre and other reproduction rights organisations.

Permission to make use of IOP Publishing content other than as set out above may be sought
at permissions@[Link].

B T Fijalkowski has asserted his right to be identified as the author of this work in accordance with
sections 77 and 78 of the Copyright, Designs and Patents Act 1988.

ISBN 978-0-7503-1350-6 (ebook)

ISBN 978-0-7503-1351-3 (print)
ISBN 978-0-7503-1352-0 (mobi)

DOI 10.1088/978-0-7503-1350-6

Version: 20161201

IOP Expanding Physics

ISSN 2053-2563 (online)
ISSN 2054-7315 (print)

British Library Cataloguing-in-Publication Data: A catalogue record for this book is available
from the British Library.

Published by IOP Publishing, wholly owned by The Institute of Physics, London

IOP Publishing, Temple Circus, Temple Way, Bristol, BS1 6HG, UK

US Office: IOP Publishing, Inc., 190 North Independence Mall West, Suite 601, Philadelphia,
PA 19106, USA
Once again, to Madeleine with love
 Contents

Preface xvii
Acknowledgements xxvi
Author biography xxvii
Acronyms xxviii

Part I Towards a unified theory exploiting a dynamical

systems approach for physical heterogeneous
continuous dynamical hypersystems

1 General considerations 1-1

1.1 Dynamical hypersystem definition 1-1
1.2 Conclusions 1-4
1.3 Summary 1-6
Reference 1-7

2 Model definition 2-1

2.1 Introduction 2-1
2.2 Physical model 2-4
2.3 Deterministic mathematical model 2-5
2.3.1 Synthetic deterministic mathematical model 2-5
2.3.2 Chaotic deterministic mathematical model 2-6
2.4 Statistical mathematical model 2-6
2.5 Stochastic mathematical model 2-6
References 2-6

3 Dynamical systems approach 3-1

3.1 Introduction 3-1
3.2 The dynamical systems approach to physical heterogeneous 3-8
continuous dynamical hypersystems
3.3 Summary 3-10
References 3-10

4 Algorithm for a formulation of the mathematical model 4-1

5 Developmental systems approach 5-1

Reference 5-5

vii
 Mechatronics

6 Synthetic mathematical model of the abstract functional 6-1

heterogeneous continuous dynamical hypersystem
6.1 Abstract functional heterogeneous continuous dynamical hypersystem 6-2
6.2 Independent generalised coordinates 6-3
6.3 Unconstrained and constrained nonlinear functional heterogeneous 6-4
continuous dynamical hypersystems—classification of the constraints
6.4 Principle of stationary action 6-6
References 6-16

Part II Towards a unified theory exploiting the theory of

holors for physical heterogeneous continuous
dynamical hypersystems

7 Physical heterogeneous continuous dynamical hypersystems 7-1

as generalised physical commutation matrixers
7.1 Introduction 7-1
7.2 The generalised physical commutation matrixer’s input- and 7-6
output-signal holor functions
7.3 Multi-port, multi-input/multi-output, generalised physical 7-13
commutation matrixers
7.4 The principal physical commutation matrixer dynamical components 7-26
7.5 Input- and output-signal holors of physical commutation matrixers 7-27
7.6 Physical and mathematical modelling methodologies 7-31
7.7 Signal holors of physical commutation matrixers 7-32
References 7-33

8 Physical commutation matrixer dynamical components 8-1

8.1 Introduction 8-1
8.2 Electrical commutation matrixer dynamical components 8-3
8.2.1 Electrical-signal holors 8-4
8.2.2 Electrical-resistance holors 8-6
8.2.3 Electrical-inductance holor 8-8
8.2.4 Electrical-capacitance holor 8-11
8.3 Magnetic commutation matrixer dynamical components 8-14
8.3.1 Magnetomotive-force and magnetic-flux holors 8-16
8.3.2 Magnetic-reluctance holor 8-16
8.3.3 Magnetic-energy-source holors 8-17
8.4 Mechanical commutation matrixer dynamical components 8-19

viii
 Mechatronics

8.4.1 Mechanical-signal holors 8-19

8.4.2 Mechanical-resistance or damping holor 8-23
8.4.3 Mechanical inductance holor of the spring 8-26
8.4.4 Mechanical-capacitance holor of the mass 8-34
8.5 Fluidic commutation matrixer components 8-37
8.5.1 Fluidic-signal holors 8-38
8.5.2 Fluidic-resistance holor 8-41
8.5.3 Fluidic-inertance holor 8-45
8.5.4 Fluidic-capacitance holor 8-48
8.5.5 Fluidic energy sources 8-57
8.6 Thermal commutation matrixer dynamical components 8-57
8.6.1 Thermal-signal holors 8-58
8.6.2 Thermal-resistance holor 8-60
8.6.3 Thermal-inertance holor 8-62
8.6.4 Thermal-capacitance holor 8-63
8.6.5 General thermal dynamical components 8-65
8.7 Energy sources and loads 8-66
8.7.1 Energy-potential-difference sources 8-67
8.7.2 Energy-transfer sources 8-69
8.7.3 Loads 8-71
8.7.4 Coulomb friction 8-72
8.7.5 Weight 8-74
8.8 Physical dynamical components as mathematical operators 8-75
8.9 Summary 8-75
Reference 8-78

9 Principal physical commutation matrixers as physical models 9-1

9.1 Introduction 9-1
9.2 Creation of physical commutation matrixers 9-2
9.3 Principal physical commutation matrixer laws 9-2

10 Generalised impedance and admittance holors of physical 10-1

commutation matrixers
10.1 Sinusoidal responses to energy-potential-difference holor and 10-1
energy-transfer holor functions
10.2 Impedance and admittance of principal physical commutation 10-1
matrixer dynamical components
10.3 Generalised impedance and admittance holors in combination 10-4
10.3.1 Driving-point impedance and/or admittance holors 10-7
ix
 Mechatronics

10.4 Physical heterogeneous continuous dynamical hypersystems’ 10-9

transmittance or transfer-function holors
10.5 General formulation of physical-heterogeneous-continuous- 10-11
dynamical-hypersystem holor equations
10.6 Some properties of linear physical heterogeneous continuous 10-13
dynamical hypersystems
10.6.1 Two-port single-input/single output physical commutation 10-13
matrixers
Summary 10-24
10.6.2 Physical commutation matrixer having three independent nodes 10-25
10.6.3 Superposition theorem 10-27
10.6.4 Reciprocity theorem 10-29
10.6.5 Thévenin’s theorem 10-31
10.6.6 Norton’s theorem 10-37
10.6.7 Millman’s theorem 10-40
Reference 10-42

11 Electrical homogeneous continuous dynamical systems as 11-1

DC and AC electrical commutation matrixers
11.1 Introduction 11-1
11.2 Electrical homogeneous continuous dynamical systems’ holor 11-6
algebra, steady-state DC analysis
11.2.1 DC holor algebra, steady-state analysis 11-6
11.2.2 Node-voltage holor method 11-9
11.2.3 Loop-current holor method 11-10
11.2.4 Superposition method 11-14
11.2.5 The DC energy source 11-15
11.2.6 The generation of DC energy 11-16
11.3 Electrical homogeneous continuous dynamical systems holor 11-23
algebra, steady-state AC analysis
11.3.1 Sinusoidal alternating-current holor algebra, 11-23
steady-state analysis
11.3.2 The sinusoidal AC energy source 11-23
11.3.3 The generation of AC 11-23
11.3.4 Sinusoidal alternating current 11-28
11.3.5 Sinusoidal voltage and current waveforms—holor 11-29
representation
11.3.6 The passive AC electrical-homogenous-continuous- 11-31
dynamical-system components in the holor domain

x
 Mechatronics

11.3.7 An impedance-holor representation 11-36

11.3.8 An admittance-holor representation 11-38
11.3.9 AC power holor representation 11-39
11.3.10 A geometrical interpretation of the voltage, current 11-42
and power holors
11.3.11 Conclusion 11-44
11.4 The single-phase AC electrical homogeneous continuous 11-45
dynamical system as the single-phase AC electrical
commutation matrixer
11.4.1 Introduction 11-45
11.4.2 Series and/or parallel transformations 11-47
11.4.3 Delta-to-wye or pi-to-tee equivalent 11-50
11.5 The polyphase AC electrical homogeneous continuous 11-54
dynamical system as the polyphase AC electrical
commutation matrixer
11.5.1 Introduction 11-54
11.6 The two-phase AC electrical homogeneous continuous dynamical 11-56
system as the two-phase AC electrical commutation matrixer
11.6.1 Introduction 11-56
11.7 The three-phase AC electrical homogeneous continuous dynamical 11-61
system as the three-phase AC electrical commutation matrixer
11.7.1 Introduction 11-61
11.7.2 Three-phase energy sources 11-63
11.7.3 Balanced three-phase AC electrical homogeneous 11-65
continuous dynamical systems
11.7.4 Wye connection of the AC electrical homogenous 11-70
continuous dynamical system
11.7.5 Delta connection of the AC electrical homogeneous 11-73
continuous dynamical system
11.7.6 Unbalanced wye-connected load of the AC electrical 11-76
homogeneous continuous dynamical system
11.7.7 Unbalanced delta-connected load of the AC electrical 11-78
homogeneous continuous dynamical system
11.7.8 Power holor computations in balanced three-phase AC 11-79
electrical homogeneous continuous dynamical systems
11.7.9 Delta three-phase AC electrical homogeneous 11-84
continuous dynamical systems
11.7.10 Wye three-phase AC electrical homogeneous continuous 11-87
dynamical systems
11.7.11 Holor algebra analysis of the wye–wye three-phase AC 11-88
electrical homogenous continuous dynamical system

xi
 Mechatronics

11.7.12 Holor algebra analysis of the wye-delta three-phase AC 11-94

electrical homogeneous continuous dynamical system
11.7.13 Holor algebra analysis of the delta–wye three-phase AC 11-96
electrical homogenerous continuous dynamical system
11.7.14 Holor algebra analysis of the delta–delta three-phase AC 11-97
electrical homogeneous continuous dynamical system
11.7.15 Delta-to-wye transformations 11-98
11.7.16 Repetitive example I 11-99
11.7.17 Repetitive example II 11-106
11.8 Parallel-series resistance-inductance-capacitance AC electrical 11-109
commutation matrixer
11.9 Application of Thévenin’s theorem 11-111
11.10 Application of superposition theorem 11-115
References 11-115

12 Mechanical homogenous continuous dynamical systems as 12-1

mechanical commutation matrixers
12.1 Simple plane-motion mechanical homogenous continuous 12-1
dynamical system
12.2 Simple pendulum mechanical homogeneous continuous 12-2
dynamical system (approximate solution)
12.3 Bicycle mechanical homogeneous continuous dynamical system 12-4
12.4 Damper–spring–mass mechanical homogeneous continuous 12-4
dynamical system (I)
12.5 Damper–spring–mass mechanical homogeneous continuous 12-6
dynamical system (II)
12.6 Automotive vehicle’s suspension mechanical commutation matrixer 12-9
12.7 Determination of the analogous impedance holor of mechanical 12-10
commutation matrixers
12.8 Damper–spring–mass mechanical homogeneous continuous 12-11
dynamical system (III)
12.9 Damper–mass mechanical commutation matrixer 12-14
References 12-18

13 Fluidic homogeneous continuous dynamical systems as 13-1

fluidic commutation matrixers
13.1 Fluidic-transmission line 13-1
13.2 Node-to-datum holor equations for the fluidic commutation matrixer 13-2
13.3 Loop and mesh holor equations for the fluidic commutation matrixer 13-4

xii
 Mechatronics

Part III Physical Matrixers as Physical Commutators

14 Mechanical commutation matrixer commutators for 14-1

conventional DC and AC magneto-mechano-dynamical
electrical machines
14.1 Introduction 14-1
14.2 MCM AC–DC/DC–AC commutator dynamotors 14-7
14.3 Schrage MCM AC–AC commutator motor 14-9
14.4 Exemplary applications of mechanical commutation matrixer 14-11
(MCM) commutators
14.4.1 AC commutatorless motor—an MCM AC–DC commutator 14-11
generator for converting AC–DC
14.4.2 MCM AC–DC commutator amplidyne 14-12
14.4.3 MCM AC–DC/DC–AC ring-commutator single-armature 14-13
frequency converter
14.4.4 MCM DC–AC/AC–DC ring-commutator single-armature 14-16
frequency converter and adjustable-ratio EE transformer
connected to an AC commutatorless wound-rotor motor
for rotational-speed-control purposes
14.4.5 MCM DC–AC/AC–DC ring-commutator single-armature 14-18
frequency converter and MCM DC–AC commutator motor
coupled to the shaft of the AC commutatorless wound-rotor
motor for speed-control purposes
References 14-21

15 Electrical commutation matrixer commutators for modern 15-1

DC and AC magneto-mechano-dynamical electrical machines
15.1 Introduction 15-1
15.2 Status and trends 15-5
15.3 Physical and mathematical models of a generalised ECM 15-19
AC–AC and/or AC–DC/DC–AC commutator
15.3.1 Introduction 15-19
15.3.2 Hybrid electrical commutation matrixer commutators 15-35
15.3.3 Monolithic electrical commutation matrixer commutators 15-36
Summary 15-49
15.4 MCM and ECM AC–AC and AC–DC/DC–AC commutator 15-51
dynamotors—a basic application

xiii
 Mechatronics

15.5 New-concept ECM AC–AC and AC–DC/DC–AC 15-53

commutator dynamotors
15.6 Exemplary applications of electrical commutation matrixer 15-61
commutators
15.6.1 A 2 × 2 ECM AC–DC/DC–AC commutator 15-61
15.6.2 A 3 × 3, 3 × 5 or 5 × 5 ECM AC–AC commutator 15-61
15.6.3 A 3 × 2 ECM AC–DC commutator 15-64
15.6.4 A 3 × 3 ECM AC–AC commutator 15-65
15.6.5 A single-phase ECM AC–AC and/or AC–DC/DC–AC 15-66
commutator
15.6.6 A 2 × 3/3 × 2 ECM DC–AC/AC–DC commutator and 15-67
2 × 2 ECM DC–DC commutator
15.6.7 A 2 × 5/5 × 2 ECM DC–AC/AC–DC commutator 15-70
15.6.8 A 2 × 3 ECM DC–AC/AC–DC commutator and 15-70
2 × 2 DC–DC commutator
15.6.9 A 2 s(2 × 5)/2 s(5 × 2) ECM DC–AC/AC–DC commutator 15-71
15.7 Electrical commutation matrixer commutators—a look into the future 15-72
15.8 Conclusion 15-74
References 15-78

16 Electrical commutation matrixer keyboards for computers 16-1

16.1 Introduction 16-1
16.2 How electrical commutation matrixer keyboards for computers work 16-4
16.3 Electrical commutation matrixer keyboard fundamentals 16-4
16.4 Electrical commutation matrixer keyboard’s electrical valves 16-6
16.5 Unconventional electrical commutation matrixer keyboards 16-15
16.6 Virtual electrical commutation matrixer keyboards 16-16
16.7 Canesta keyboard 16-16
16.8 Samsung’s Scurry keyboard 16-17
16.9 Exemplary electrical commutation matrixer keyboard 16-17
16.10 Conclusions 16-21
References 16-22

17 Programmable-logic and/or generic-logic electrical 17-1

commutation matrixers for digital devices
17.1 Introduction 17-1

xiv
 Mechatronics

17.2 Programmable-logic electrical-commutation-matrixers 17-2

and classifications
17.2.1 The OR electrical commutation matrixer 17-2
17.2.2 The AND electrical commutation matrixer 17-2
17.2.3 Classifications of programmable-logic electrical 17-3
commutation matrixers
17.3 Programmable ROMs (PROMs and EPROMs) 17-3
17.4 Programmable matrixer logic (PML) 17-5
17.5 Generic matrixer logic (GML) 17-6
17.6 Electrical commutation matrixer crossbar 17-8
17.7 Electrical commutation matrixer tactile sensor 17-10
17.8 Electrical commutation matrixer seven-segment display 17-11
17.9 Fluidic commutation matrixer combinational-chemistry 17-13
microreaction
17.10 Exemplary applications of programmable-logic electrical 17-14
commutation matrixers for digital devices
17.10.1 PCM implementing a 2048 × 8 EPROM 17-14
17.10.2 PCM implementing a sum-of-products (SOP) expression 17-14
17.10.3 GML implementing a sum-of-products (SOP) expression 17-17
17.10.4 Electrical commutation matrixer seven-segment display 17-18
to put on show ‘5’
17.11 Conclusion 17-18
References 17-20

18 Nano-magneto-rheological fluido-mechanical commutation 18-1

matrixer for internal combustion engines
18.1 Introduction 18-1
18.2 The Fijalkowski engine concept 18-5
18.2.1 Magnetic-field-exciter’s electric-current sequencing for 18-12
rotary motion
18.2.2 Engine output-shaft angular velocity control 18-14
18.2.3 Engine output-shaft deceleration and reverse 18-15
18.3 Fijalkowski engine cooling 18-15
18.4 The Fijalkowski engine advantages versus conventional 18-16
internal combustion engines
18.5 Conclusions 18-18
References 18-19

xv
 Mechatronics

19 Conclusion and future trends 19-1

19.1 Concluding remarks 19-1
19.2 Future work 19-4
References 19-7

Glossary 20-1

xvi
 Preface

To make progress, science should not accept the limitations placed on discovery by
traditional methods, conventional approaches, or existing infrastructure. Moselio
Schaechter, Roberto Kolter and Merry Buckley 2004 Microbiology in the 21st
Century: Where Are We and Where Are We Going? (Washington DC: AMS)
The real voyage of discovery consists not in seeking new landscapes but in having new
eyes.
Marcel Proust
A particular new scientific result does not usually gain a victory in a way that the
opponents suffer a defeat and declare that they are converted but much rather
the opponents gradually die out and the new generations grow ab ovo familiar with
the truth.
Max Planck
The resistance to a new idea increases as the square of its importance.
Bertrand Russell
The title of this textbook requires some explanation. ‘Mechatronics’ is the integration
of physics—mechanics, fluidics, electrics and electronics, etc—and computer tech-
nologies into the research and development of physical heterogeneous continuous
complex and/or simple dynamical hypersystems (Fijalkowski 2010, 2011). Thus,
mechatronics consists of the synergistic combination of different physical disciplines
like mechanics, fluidics, electrics and electronics, as well as informatics, etc. Synergy
is a very creative and therefore dynamic process, far more so than the usual
co-operation of the different physical disciplines, and even more so than a close
integration of machine hardware and software.
Mechatronics offers new solutions and unprecedented flexibility in transportation
systems, industrial production processes, and aerospace, automotive and traction
components, etc. Its economic success is based on functional integration, i.e. the
multiple use of mechano-mechanical (MM), fluido-mechanical (FM) and/or electro-
mechanical (EM) actuators (machines); the decentralisation of intelligence into physical
heterogeneous continuous dynamical hypersystems, e.g. aerospace, automotive and/or
traction single-shaft propulsion mechatronical control systems; and the inherent
options for sensorless self-monitoring and system protection.
The ‘dynamical systems approach’ is a holistic description of observable physical
reality. The whole of the dynamical systems approach is not composed of the sum of its
elementary units—the multiplicity of elementary units is a result of the diversification
of the whole.
The idea of a holor (from the Greek δλοσ, as in the English holistic, pertaining to
an entity or a whole) seems to have been originated by Wallis (1673) to describe a
mathematical entity that is made up of one or more merates (from the Greek μεροσ ́ ,
part of an entity) or coordinates (independent quantities) of the holor. It includes
complex numbers, scalars, vectors, matrices, tensors, quaternions and other hyper-
numbers. The works of Grassmann (1862, 1844, 1894) strongly influenced the

xvii
 Mechatronics

subsequent development of holors, including the works of Gibbs (1881, 1884) and
Heaviside (1893, 1925, 1950).
Holors, thus defined, have been known for centuries, but each type has been
developed more or less independently, accompanied by a separate nomenclature and
theory (Moon and Spencer 1986).
The ‘theory of holors’ demonstrates how these complicated subjects can be made
simple by using a single notation that applies to all holors, both tensors and non-
tensors. Moon and Spencer (1965, 1986) consider all possible types of holors and
develop holor algebra and holor calculus in the most general sense.
Thus, the reader will learn to develop a new holor that fits the application, rather
than forcing an application onto a holor representation that is known but does not
perfectly describe it.
The discussion includes non-tensors that have no transformation and holors that
transform in more complicated ways than are allowed with ordinary tensors. This
opens up the possibility of devising a holor for a new physical application without
being limited to a few conventional types of holor. The theory of holors should
establish a method through which students and teachers can learn vector and tensor
analysis via a uniform treatment (Moon and Spencer 1963, 1965, 1986; Fijalkowski
(1967)).
This textbook deals with the identification of natural Lagrangian dynamical
systems (Arnold, 1978), that is, the formulation of the mathematical description, i.e.
the mathematical model of natural Lagrangian dynamical hypersystems. They have
inputs and outputs that are related by means of dynamical relationships in a form of
the Euler–Lagrange second-order differential equation of dynamics in the original
dynamical systems approach and holor notation. Physical heterogeneous continuous
complex and/or simple dynamical hypersystems with holonomic and unholonomic
constraints are generalised physical heterogeneous continuous dynamical objects
and devices. This book takes into consideration a single ‘identification method’,
which uses the dynamical systems approach and the matrix notation of the Euler–
Lagrange second-order differential equations of dynamics. Identification is defined
as the formulation of a mathematical description, i.e. it is the mathematical model of
the physical heterogeneous continuous dynamical hypersystem. As the mathematical
model is a generalised presentation of a significant feature of real or designed
physical heterogeneous continuous dynamical hypersystems, it is in a convenient
form for the researcher. A mathematical model is as a rule a simplification of reality.
In physics and engineering, the author discriminates three fundamental aims for a
formulation of mathematical models of physical heterogeneous continuous dynam-
ical hypersystems: analytical study, design and control.
In an identification of some real (existing) physical heterogeneous continuous
complex and/or simple dynamical hypersystems there is as a rule a great degree of
indeterminacy, connected with the existence of stochastic disturbances and often an
inaccurate knowledge of the physical heterogeneous continuous dynamical hyper-
system’s structure. This indeterminacy often limits statistical identification methods
(Manczak and Nahorski 1983).
In this book, the author has confined himself exclusively to ‘deterministic
identification methods’, because in the considered mathematical models of physical

xviii
 Mechatronics

heterogeneous continuous dynamical hypersystems, the stochastic disturbances may

be neglected.
Deterministic identification methods for physical heterogeneous continuous
dynamical hypersystems are interesting for practitioners and designers.
Dynamical process identification is a complex work that comprises (Wegrzyn
1974): the formulation of mathematical models, experimental studies (collection of
measurement data), analytical studies (physical-parameter computer-simulation of
mathematical models) and the verification (and inspection) of mathematical models.
The most important and most difficult aspect in physically continuous dynamical
hypersystems is the formulation of mathematical models, which is why this is a
principal aim of this textbook.
In principle, the problems that exist upon a formulation of a simple mathematical
model of the dynamical system may be considered to have been sufficiently studied.
In the case of a formulation of a synthetic mathematical model or a functional
mathematical model of the physical heterogeneous continuous dynamical hyper-
system, a suitable methodology of its formulation should be applied.
In the holor theory of abstract functional heterogeneous continuous dynamical
hypersystems (including physically continuous dynamical hypersystems, among
others) two modes of mathematical modelling methodology may be selected: an
analytical mode with decomposition methods and a synthetic mode comprising
aggregation methods.
The full advantages of both modes are taken into consideration in this book. That
is why when considering physical heterogeneous continuous dynamical hypersystems
two kinds of mathematical models may be created. Namely, synthetic mathematical
models for a structure of the physical heterogeneous continuous dynamical hyper-
system, as well as simple mathematical models for individual, functional and
structural dynamical components of physical heterogeneous continuous dynamical
hypersystems, that is, elemental homogeneous continuous dynamical systems, or
dynamical hyposystems.
The formalisation of physical heterogeneous continuous dynamical hypersystems
provides everything about the mathematical models’ methodology, comprising
different operations, phases and connections in a synthetic dynamical process.
However, this is not the case for identification, which most often resolves itself in the
identification of the homological, functional and structural dynamical components
of physically continuous dynamical hypersystems, that is, physically homogenous
continuous dynamical systems, or dynamical hyposystems, by means of well-known
methods.
Stating a generality of knowledge on a considered problem precisely in the domain
of mathematical formalism is termed ‘mathematical model formulation’, and
concrete mathematical relationships are termed ‘mathematical models’. Because
the mathematical model only simulates these features, which specify a purpose for
which it has been created, a given physical heterogeneous continuous dynamical
hypersystem may have no single mathematical model, but instead several, simulating
heterogeneous viewpoints. It may be affirmed, going out from most generalised
assumptions, that each of the simple ‘component’ mathematical models may be led

xix
 Mechatronics

out from a certain synthetic mathematical model of the physical heterogeneous

continuous dynamical hypersystem, in which they are connected to each other.
A synthetic mathematical model that is solely in the domain of abstract
mathematical models and uses the Euler–Lagrange second-order differential equa-
tions of dynamics for its formation may be formatted in the dynamical systems
approach and holor matrix notation. In compliance with this, it is good to determine
the physical heterogeneous continuous dynamical hypersystem by means of a
structure, representing the so-called ‘synthetic mathematical model’.
The physical heterogeneous continuous dynamical hypersystem as a whole may
be concerned or it may be divided into physical homogeneous continuous dynamical
systems on the grounds of which form of energy is concerned (e.g. kinetic energy:
radiant, thermal, motion, sound and electrical homogeneous continuous dynamical
systems; potential energy: chemical, nuclear, stored mechanical, fluidic, and elec-
trical and gravitational homogeneous continuous systems). Next, they may be
divided as physical homogeneous continuous dynamical hyposystems and, as
follows, they may be divisible as functional and structural physical homogeneous
continuous dynamical components.
The author’s intention in writing this textbook is to submit ‘generalised physical
commutation matrixer holor analyses’ in the easiest practicable dynamical systems
approach (systems thinking). The author supposes that this can be best realised
through the application of physical commutation matrixers and the functional and
structural physical homogeneous continuous dynamical components.
Nearly all mechanical and electrical engineering students have been educated
concisely regarding physical commutation matrixers during their university physics
courses. In view of this, the physical model represents an exceptional structure with
which to form a junction between static and dynamical analysis.
It is probably particularly important that the generalised physical commutation
matrixer concept lends itself systematically to analogy. Owing to the application of the
physical commutation matrixer method, one may confirm that it is practicable to
consider the area under discussion in mechatronics (kinetic energy—radiant, thermal,
motion, sound and electrical—as well as potential energy—chemical, nuclear, stored
mechanical, fluidic, electrical and gravitational). For instance, the analogy between
charging an electrical capacitor and saturating a fluidic container of water is made
evident by equating the physical and mathematical models of these physical processes.
Naturally, one might have identified this analogy instinctively. In spite of this, one
becomes aware that these physical processes are also comparable to the variation in
velocity of a mass when a force is relevant. This is not practically so evident. As a
result, the physical commutation matrixer concept confirmation is exceptionally
convenient in the absence of physical situations in various physical heterogeneous
continuous dynamical hyposystems for a generalised physical heterogeneous contin-
uous dynamical hyposystem from which a universal type of solution may be
completed.
A new definition of the ‘physical commutation matrixer’, based on matrix-
interconnection physical heterogeneous continuous dynamical hypersystem theory
concepts, is proposed. The definition may be used to evaluate physical commutation

xx
 Mechatronics

matrixers, which perform multivalent logical functions, and continuously operating

physical commutation matrixers. The serial and parallel connections of two physical
commutation matrixers and simplified formulae are defined, which are valid for high
component physical commutation matrixers.
A comparison is made between results arrived at using the new and conventional
definitions. It is found that the results differ little in the case of equal probability
output letters. For instance, a physical commutation matrixer is a general term
referring to a physical heterogeneous continuous dynamical hypersystem or part of a
physical heterogeneous continuous dynamical hypersystem of matrixery conductive
parts and their matrixery interconnections through which an energy-transfer holor is
intended to flow.
A physical commutation matrixer, i.e. a configuration of physically (electrically,
magnetically, optically, or radiationally) connected dynamical components or
analogue and/or digital devices, is made up of active and passive physical, functional
and structural dynamical components or an assemblage of physical, functional and
structural dynamical components, and their matrix-interconnected row and column
conductive collectors. Thus, a physical commutation matrixer is a physical device
powered by physical energy.
The active physical, functional and structural dynamical components are the
sources of physical energy for the physical commutation matrixer; for instance, they
may be chemo-electrical/electro-chemical (ChE/ECh) storage batteries, direct cur-
rent (DC) or alternating current (AC) mechano-electrical/electro-mechanical (ME/
EM) dynamotors (generators/motors), photovoltaic cells, or fuel cells.
The passive physical, functional and structural dynamical components are
impeders or admitters, i.e. resistors, inductors and capacitors.
The physical commutation matrixer described by a matrixer physical model
(matrixery signal-flow diagram or map) shows the active and passive physical,
functional and structural dynamical components and their matrixery interconnected
row and column conductive collectors (conductors).
For the purposes of analysis, apparatus (equipment) and devices with an
individual physical identity are often represented by equivalent physical commuta-
tion matrixers. These equivalent physical commutation matrixers are made up of the
basic passive and active physical, functional and structural dynamical components
listed above. For instance, electrical commutation matrixers are used to transmit
electrical energy in high-voltage power lines and electro-electrical (EE) transformers
or in low-voltage distribution in factories and homes. They are used to convert
energy from or to its electrical form, for example, as in EM motors, ME generators,
microphones, loudspeakers and lamps; to communicate information, as in tele-
phones, radios, televisions and internet systems; to process and store data and make
logical decisions, as in computers; and to form systems for the automatic control of
equipment.
Physical commutation matrixer theory includes the study of all aspects of
matrixers, including analysis, design and application. In it, the fundamental
quantities are the ‘energy-potential-difference holors’ (i.e. generalised force holors)
between various points, the ‘energy-transfer holors’ (i.e. generalised velocity holors)

xxi
 Mechatronics

flowing in a number of row and column conductive collectors, and the parameters,
which describe the passive physical, functional and structural dynamical components.
Other important physical commutation matrixer quantities, such as power, energy
and time constants, may be computed from the fundamental physical variables. For a
discussion of these parameters, physical commutation matrixer theory is often divided
into special topics. This can be based on how the energy-potential-difference holors
(i.e. voltage holors and current holors) in the physical commutation matrixer vary
with time (i.e. DC or AC, sinusoidal, non-sinusoidal, digital and transient physical
commutation matrixer theory). Moreover, they can be divided based on the arrange-
ment or configuration of the energy-transfer holors’ (i.e. electrical current holors) row
and column conductive collectors (series, parallel, series parallel, parallel series,
coupled, open circuited and short-circuited physical commutation matrixers).
The physical commutation matrixer can be divided into special topics according
to which physical devices form the matrixer, or the application and use of physical
commutation matrix theory can also be divided (power, communication, macro-
and microelectronic, solid state, integrated computers and programmable commu-
tation matrixers).
This textbook provides a comprehensive and careful development of the physical
commutation matrixer base and the use of holor theory formulations of analytical
methods.
No attempt is made to provide the required mathematical tools: matrixer theory,
the functions of a complex variable, Laplace transforms, numerical methods and
programming. However, some references in these areas have been included in the
bibliography.
Holor analysis has been shown beyond doubt to be of prime importance in
physics and engineering. Every current scientist and engineer must be systematically
well acquainted with the symbolism and methods of manipulation.
Because both the nomenclature and routine manipulation are quite uncompli-
cated and may be learned in a short time, the question emerges as to whether an
independent one-term subject on holors is adequate for a programme of scientific
study.
If the subject is on the experiential level, it may be integrated into one of the
necessary physics subjects, such as electronics and magnetics, mechanics, fluidics,
photonics, electronics, etc.
However, if one expects to use matrices as a systematically dependable tool, one
must really understand them. This not only necessitates that an independent subject
is taught in the programme of study, but also that there is a textbook that is
dissimilar from the old school one.
An examination of the handbooks on matrix analysis since 2000 confirms that
virtually all of them are founded on the same physical models. One considers these
physical models to be inadequate on two counts (Moon and Spencer 1965):
• they do not refer to ‘invariance’, which is in reality the imperative factor that
makes a matrix a matrix, a vector a vector and a scalar a scalar;
• they give too much weight to rectangular coordinates.

xxii
 Mechatronics

Matrix analysis as column matrix or vector analysis is the infant of quaternions

and Ausdehnungslehre. The former are principally algebraic, and in them questions
of invariance do not arise. The latter deals with geometrical figures that are tacitly
assumed to be unconcerned by coordinate transformations. However, precise
consideration of invariance under coordinate transformation (which has assumed
such importance in the 20th century, especially after the advent of relativity) is not
present in either parents or offspring. This is why some mathematicians have
considered column matrix analysis or vector analysis to be a marginal subject: an ad
hoc combinatory logic foundation. The inexperienced confidence that a column
matrix or a vector maintains its form and size is transformed when the coordinates
are not adequate. One must also inquire, ‘under what group of transformations?’.
Because matrix analysis has disregarded this question, it is concerned by impreci-
sion. The unacceptable dissimilarity among free, bound and sliding column matrices
or vectors is a paradigm. In addition, the old school treatment of matrices gives too
much weight to orthogonal Cartesian coordinates. Even the definitions of ‘gradient’,
‘divergence’ and ‘curl’ are normally only provided for particular cases of rectangular
coordinates, as if no other coordinates ever emerge in mathematics or physics.
The treatment presented in this interdisciplinary textbook differs from the old
school one by introducing invariance into the theory and providing general
definitions that are sensible in all coordinate systems. In this approach, matrix
analysis is endowed with a concrete logical foundation. Irrespective of the forward-
thinking aspects of the book, we do not sense that the treatment is too difficult for
ordinary readers.
This interdisciplinary subject has been taught to undergraduates and postgrad-
uates at the Cracow University of Technology in Krakow and the State Higher
Vocational School in Nowy Sacz, Poland. This textbook has been written primarily
for undergraduate students of physics and electrical, mechanical, fluidic and thermal
engineering. Its logical structure, however, should also make it valuable for
mathematicians, and it may serve as a helpful review source for graduate students.
The first few chapters of the book introduce a variety of concepts that may be
unknown to some readers, but mastery of these concepts may provide a much deeper
knowledge of matrix analysis than could otherwise be acquired. Index notation is
employed where necessary, but applications to electronics and mechanics, etc, are
performed in the well-known physical commutation matrixer nomenclature. Various
problems are known to provide readers with satisfactory preparation in using holors.
I am the sole author of this textbook and all the text contained herein is of my
own conception unless otherwise indicated. Any text, figures, theories, results, or
designs that are not of my own devising are appropriately referenced in order to give
acknowledgment to the original authors. All sources of assistance have been
assigned due acknowledgement. In this book, all the information has been obtained
and presented in accordance with academic rules and conduct. I have fully cited and
referenced all the material and results that are not original to this book.
I am also indebted to my international and national colleagues who contributed
indirectly to this book. In addition, I am grateful to all of you who have adopted this

xxiii
 Mechatronics

text for your interdisciplinary classes or for your own use. Without you I should not
be in business.
I hope that you find this textbook to be a valuable learning tool and reference for
students. The author is also grateful to his friends for helpful comments. The above
declaration should not be interpreted as a suggestion that these benevolent friends
concur with all the iconoclastic propositions presented in my text; quite the contrary!
For all the radicalisms and imperfections, I assume total responsibility. I would
like to conclude this preface by sharing my life motto—with focus, motivation and
concerted efforts, one can make the seemingly impossible, possible.
Bogdan Thaddeus Fijalkowski
Krakow, 21 August 2015

References
Arnold W I 1978 Mathematical methods of classical mechanics (Graduate Texts in Mathematics
vol 60) (Berlin: Springer)
Cajori F 1912 Historical notes on the graphic representation of imaginaries before the time of
Wessel Am. Math. Mon. 19 167
Fijalkowski B 1967 Zastosowanie rachunku holorowego w teorii pradow przemiennych
(Application of a holor calculus in the theory of alternating currents) Przeglad elektrotech-
niczny (Electrotechnics Review) Rok XLIII, Zeszyt 2, Warszawa, 1967 (In Polish)
Fijalkowski B T 2010 Automotive Mechatronics: Operational and Practical Issues.— vol I.
International Series on Intelligent Control, and Automation Science and Engineering—
Vol 47 (Berlin: Springer)
Fijalkowski B T 2011 Automotive Mechatronics: Operational and Practical Issues.— vol II.
International Series on Intelligent Control, and Automotive Science and Engineering—
Vol 52 (Berlin: Springer)
Gibbs J W 1881 Elements of Vector Analysis privately printed, New Haven pp 17–50; 1881
pp 50–90 1884; 1906 The scientific papers of J Willard Gibbs Vol. II (London: Longmans
Green and Co.) 17
Gibbs J W 1884 Elements of Vector Analysis privately printed, New Haven, 1881 pp 17–50; 1884
pp 50–90; 1906 The scientific papers of J Willard Gibbs Vol. II (London: Longmans Green
and Co.)
Grassmann H 1862 Die Ausdehnungslehre Enslin, Berlin, 1862 (In German)
Grassmann H 1844 Die lineale Ausdehnungslehre, ein neuer Zweig der Mathematik Ottp Wigand,
Leipzig
Grassmann H 1894 Grassmann’s gesammelte math. u. phys. Werke 3 vols (Leipzig: B G Teubner)
Heaviside O 1893 Electromagnetic Theory (London: Electrician Printing and Pub. Co.), 3 vols.
1950 (New York: Dover) Heaviside’s treatment is given in vol 1, 3 132–305: The elements of
vectorial algebra and analysis
Heaviside O 1925 Electromagnetic theory 3 vols (London: The Electrician Printing and Publishing
Company Ltd.)
Heaviside O 1950 Electromagnetic theory 3 vols contains a critical and historical introby ed Ernst
Weber (New York: Dover), see vol 11 pp 192–305 on vector analysis

xxiv
 Mechatronics

Manczak K and Nahorski Z 1983 Komputerowa identyfikacja obiektow dynamicznych

(Computerised Identification of Dynamical Objects) Panstwowe Wydawnictwo Naukowe,
1983 (In Polish)
Moon P H and Spencer D E 1963 A new mathematical representation of alternating currents
Tensor 110
Moon P H and Spencer D E 1965 Vectors (Princeton, NJ: Van Nostrand)
Moon P H and Spencer D E 1986 Theory of holors A generalization of Tensors (Cambridge:
Cambridge University Press)
Wallis J 1673 Letter to Collins, 6 May 1673 published in Treatise of algebra (London 1685) p 264
Wallis J 1685 Letter to Collins, 6 May 1673 Treatise of algebra (London, 1685) p 264
Wegrzyn S 1974 Systemy automatyki kompleksowej. Teoria i doswiadczenie Prace VI Krajowej
Konferencji Automatyki 1974 (In Polish)

xxv
 Acknowledgements

The author gives his sincere thanks to the many friends and students who not only
inspired his initial interest in the subject, but also gave many helpful suggestions
during the writing of the book. Particularly, the author wishes to express his
gratitude to Dr Jozef Tutaj, who has been indispensible in the writing of this book,
and whose enthusiasm, advice, and encouragement helped make all of it possible.
Special thanks go to my grandson Marcel for his very helpful discussions and for the
preparation of many of the figures in this book. I am grateful to the many authors
referenced in this book from whom, during the course of writing, I learned so much
regarding the subjects covered. I am also indebted to my international and national
colleagues who contributed indirectly to this book. I would like to take this
opportunity to also thank the IOP Publishing crew for their persistence in making
this book a reality. In addition, I am grateful to all of you who have adopted this text
for your interdisciplinary classes or for your own use. Without you I would not be in
business.

xxvi
 Author biography

B T Fijalkowski
Professor Dr Bogdan T Fijalkowski was born in 1932 in Poland. He
obtained an MSc in Electrical Engineering, PhD in Power
Electronics and DSc in Physical Heterogeneous Continuous and
Discrete Dynamical Systems from Szczecin University of
Technology in 1959, Academy of Mining and Metallurgy in 1965
and Poznan University of Technology in 1988, respectively. He
began his professional and electrical engineering carrier in 1955. He
has been working 5 years in industry and 50 years in academia; he was the Director
of the Electrotechnics & Industrial Electronics Institute, Faculty of Electrical and
Computer Engineering and he was the Head of Automotive Mechatronics
Institution, Faculty of Mechanical Engineering at the Cracow University of
Technology, Poland. He was a visiting professor at several well-known universities.
He serves as Consultant to several organizations in Poland and the United States.
He is the signatory of the MoU for the establishment of World Electric Vehicles
Association (WEVA). He was Guest Editor of Journal of Circuits, Systems and
Computers—Special Issue on Automotive Electronics, reviewer and referee, IEEE
Transactions on Circuits and Systems—Part I, IEEE Transactions on Fuzzy Systems,
International Journal of Vehicle Design, USA, International Journal on Advanced
Robotic Systems and International Journal of Technology Management (IJTM),
USA. He is a reviewer of books and book chapters on automotive electrics and
electronics, electric drives systems dynamics and wireless information transmission,
as well as information security, and is listed in Who‘s Who in the World, Who‘s Who
in Science and Engineering and so on. He has published 29 books and book chapters
as well as over 200 technical papers and 25 patents on mining and automotive
electrics and electronics as well as mechatronics. Recent publications by him include
books (monographs), book chapters, journal articles, and conference proceedings on
topics such as mathematical models of selected aerospace and automotive discrete
dynamical hypersystems as well as the civil and military, wheeled and tracked
all-electric and hybrid-electric vehicles and also nanomagneto-rheological fluid
(NMRF) mechatronic commutator, ‘crankless’ internal combustion engines with
energy storage, termed the Fijalkowski engines and automotive gas turbines that are
based on the Fijalkowski turbine boosting (FTB) system. Recently, Springer
published two volumes of his book entitled Automotive Mechatronics: Operational
and Practical Issues and IOP Publishing published his interdisciplinary book entitled
Mechatronics: Dynamical Systems Approach and Theory of Holors. He made
valuable contributions to several of ASME, ISATA, ISTVS, SAE, IEEE SMC, as
well as EVS and WEVA, conferences as a track and/or session chair and speaker.
His hobby is yachting. He has the rank of Ocean-Going Yacht Master.

xxvii
 Acronyms

ABEL Advanced Boolean expression language

ABW Absorb-by-wire
AC Alternating current
AEV All-electrical vehicle
AI Artificial intelligence
AWA All-wheel-absorbed
AWB All-wheel-braked
AWD All-wheel-driven
AWS All-wheel steered
BBW Brake-by-wire
BDC Bottom dead centre
b–k Damper–spring
b–m Damper–mass
b–k–m Damper–spring–mass
CASE Computer-aided system engineering
C2 M Crankshaft and connecting rod (conrod) mechanism
CECM Coordinate electrical commutation matrixer
CHINT Charge injection transistor
CHP Combined heat and power
ChE Chemo-electrical
Ch-Th Chemo-thermal
Ch-Th-FM Chemo-thermo-fluido-mechanical
Ch-Th-F-ME Chemo-thermo-fluido-mechano-electrical
CMOS Complementary MOS
COTS Commercially-off-the-shelf
CPU Central processing unit
CR Common rail
CS Computer system
CUPL Computer for universal programmable logic
D Differential
DBW Drive-by-wire
DC Direct current
DDPWM Direct duty-ratio PWM
DG Distributed generation
DoF Degrees-of-freedom
DTC Direct torque control
ECE External combustion engine
ECL Emitter-coupled logic
ECM Electrical commutation matrixer
ECU Electronic control unit
E2CMOS Electrically erasable complementary metal-oxide semiconductor
E2D Electrical energy distribution
EFD Electro-fluido-dynamical
EIC Electrical integrated circuit
EIM Electrical integrated matrixer
EMD Electro-mechano-dynamical
EOF Electro-osmotic flow

xxviii
 Mechatronics

EPD Electro-plasmo-dynamical
EPROM Erasable PROM
EMU Electrical multiple unit
EE Electro-electrical
ECh Electro-chemical
EM Electro-mechanical
E-Th Electro-thermal
FBW Fly-by-wire
FCM Fluidic commutation matrixer
FEM Finite elements method
FET Field effect transistor
FFET Flow field effect transistor
FL Fuzzy-logic
FM Fluido-mechanical
FM Frequency modulation
FOC Field-oriented control
FPLA Field programmable logic array
FM Fluido-mechanical
GCM Generalised commutation matrixer
GECM Generic-logic electrical commutation matrixer
GERF Giant electro-rheological fluid
GML Generic matrix logic
GTO Gate-turn-off
HBT Hetero-junction bipolar transistor
HDI High-pressure direct injection
HDL Hardware description language
HE Hybrid electrical
HEMT High electron injection transistor
HEV Hybrid electrical vehicle
HPS Hybrid power source
HVDC High-voltage direct current
H&RD Human and/or robotic-driver
HE Hydro-electrical
I Integral
ICE Internal combustion engine
IEEE Institution of Electrical and Electronics Engineers
IGBT Insulated-gate bipolar transistor
IP Internet protocol
IPM Interior permanent magnet
IT Information technology
I/O Input/output
JFET JUGFET Junction-gate FET
KFL Kirchhoff’s first law
KSL Kirchhoff’s second law
LAN Local area network
LED Light-emitting diode
LM–RM Linear reciprocating motion to rotary motion
LTQ Light triggered and quenched
MIMO Multi-input/multi-output
MCM Mechano-electrical commutation matrixer

xxix
 Mechatronics

MCT MOS controlled thyristor

MCU Microcontroller unit
MEMS Micro-electro-mechanical system
MESFET Metal-semiconductor FET
MFD Magneto-fluido-dynamical
MISFET Metal-insulator-semiconductor FET
MMC Man–machine communication
MMD Magneto-mechano-dynamical
MNOS Metal-nitride-oxide silicon
MOS Metal-oxide semiconductor
MOSFET Metal-oxide semiconductor FET
MPD Magneto-plasmo-dynamical
MRV Martian Roving Vehicle
MSI Medium scale integration
MSI Mild soft iron
ME Mechano-electrical
MF Mechano-fluidic
MM Mechano-mechanical
MP Mechano-pneumatic
MV Mechano-vacuum
N North
NAEMS Nano-electro-mechanical system
NF Neural network and fuzzy logic
NMOS Non-complementary metal-oxide semiconductor
NMRF Nano-magneto-rheological fluid
NN Neural-network
OFET Organic field effect transistor
OTS Off-the-shelf
P Proportional
PBC Passively based control
PC Personal computer
PC Predictive control
PC Process control
PECM Programmable-logic electrical commutation matrixer
PDA Personal digital assistant
PF Power factor
PHEV Plug-in hybrid-electrical vehicle
PI Polarisable interference
PLM Programmable logic matrixer
PML Programmable matrix logic
PROM Programmable read-only memory
PWM Pulse-width modulation
PM Pneumo-mechanical
QUIT Bipolar quantum interference transistor
RB Reverse blocking
RBSOA Reverse-biased safe-operating area
RBW Ride-by-wire
RCGTO Reverse conducting GTO
RM–LM Rotary motion to linear reciprocating motion
R&D Research and development

xxx
 Mechatronics

R-C Resistor-capacitor
R-L Resistor-inductor
R-L-C Resistor-inductor-capacitor
S South
SBW Steer-by-wire
SCR Semiconductor controlled rectifier
SISO Single-input/single-output
SET Single electron transistor
SIT Static induction transistor
SLI Starting, lighting and ignition
SMES Superconducting magnetic energy storage
SM&GW Steered, motorised and/or generatorised wheel
SPWM Sinusoidal PWM
SR Semiconductor rectifier
SSI Standard scale integration
SOP Sum-of-product
SUPER-HET Bipolar superconductor hot-electron transistor
SVM Space vector modulation
SVPWM Space-vector PWM
TAB Tape automated bonding
TDC Top dead centre
TTL Transistor–transistor logic
UPS Uninterruptible power supply
UV Ultra-violet
VAP Very advanced propulsion
VAT Very advanced technology
VDU Video display unit
VLSI Very large scale integration
VoIP Voice over Internet protocol
VSC Voltage-source chopper
VSI Voltage-source inverter
XBW X-by-wire
Δ2Y Delta-to-wye
Π2T Pi-to-tee
2WD Two-wheel-drive
2WS Two-wheel-steered
2D Two dimensional
3D Three dimensional
4WA Four-wheel-absorbed
4WB Four-wheel-braked
4WD Four-wheel-driven
4WS Four-wheel-steered
2×1 Two-by-one
2×2 Two-by-two
3×2 Three-by-two
3×3 Three-by-three
3×5 Three-by-five
5×5 Five-by-five
5×6 Five-by-six

xxxi
 Part I
Towards a unified theory exploiting a dynamical
systems approach for physical heterogeneous
continuous dynamical hypersystems
 IOP Publishing

Mechatronics
Dynamical systems approach and theory of holors
B T Fijalkowski

Chapter 1
General considerations

CONSILIENCE

Most of the issues that vex humanity daily cannot be solved without integrating
knowledge from the natural science with that of the social sciences and human-
ities. Fluency across the boundaries will provide a clear view of the world as it
really is, not as seen through the lens of ideologies and religious dogmas or
commanded by myopic response to immediate needs.
E O Wilson

1.1 Dynamical hypersystem definition

Dynamical hypersystems unified theory is a mathematical area that describes the
behaviour of heterogeneous and/or homogeneous, complex and/or simple, contin-
uous and/or discrete, dynamical hypersystems, usually by applying differential
equations and/or the difference equations of dynamics. When the differential
equations of dynamics are applied, the theory is termed ‘continuous dynamical
hypersystems’. When the difference equations of dynamics are applied, the theory is
termed ‘discrete dynamical hypersystems’. When the time variable runs over a time
set that is discrete over some intervals and continuous over other intervals, or is any
arbitrary time set such as a Danish mathematician Georg Cantor (1845–1918) set,
one obtains dynamical equations on time scales. Some situations may also be
primarily physically and secondarily mathematically modelled by miscellaneous
‘operators’, such as the differential–difference equations of dynamics. This theory
deals with the long-term qualitative behaviour of dynamical hypersystems, and
studies the solutions of the dynamical equations of motion of dynamical hyper-
systems that are primarily mechanical in nature—although this includes both
planetary orbits and the behaviour of systems that are electrical, magnetic, fluidic

doi:10.1088/978-0-7503-1350-6ch1 1-1 ª IOP Publishing Ltd 2016

 Mechatronics

or thermal in nature—and the solutions to partial differential equations that arise in

biology. Much of modern research is focused on the study of chaotic dynamical
hypersystems. This subject of study is also termed just ‘dynamical hypersystems’,
‘mathematical dynamical hypersystems theory’ or ‘the mathematical theory of
dynamical hypersystems’. The term ‘physical heterogeneous continuous dynamical
hypersystems’, in its most generic form, means the dynamical hypersystems of
dynamical systems, hyposystems and components that change over time. The more
technical usage, ‘physical heterogeneous continuous dynamical hypersystems’, refers
to a class of mathematical differential equations of dynamics that describe time-
based dynamical hypersystems with particular properties.
A ‘physical heterogeneous continuous dynamical hypersystem’ is complex
or simple when it is compiled from many dynamical parts (dynamical systems,
hyposystems and components, respectively) that interconnect in intricate manner.
This definition has to do with the number and nature of the interconnections. The
metric for intricateness is the amount of information contained in the physical
heterogeneous continuous dynamical hypersystem.
A physical heterogeneous continuous complex and/or simple dynamical hyper-
system (see figure 1.1) presents dynamical complexity when the cause and effect are
subtle over time.
Examples: a dramatic variety of effects in the short and long run; dramatically
different effects locally and in other dynamical parts (dynamical systems, hypo-
systems and components, respectively) of the physical heterogeneous continuous
dynamical hypersystem; evident interventions cause non-evident consequences.
A physical heterogeneous continuous dynamical hypersystem (figure 1.1) is
complex and/or simple when it is composed of a collection of related constituent
dynamical parts (dynamical systems, hyposystems and components, respectively),
for which the degree and nature of the relationships are imperfectly known.

Figure 1.1. Characteristics of physical heterogeneous continuous complex and/or simple dynamical
hypersystems.

1-2
 Mechatronics

A physical heterogeneous continuous complex and/or simple dynamical

hypersystem has a collection of various dynamical systems, hyposystems and
components so connected or related as to perform a particular function not
performable by the dynamical systems, hyposystems and components alone.
This necessitates different problem-solving modi operandi at various levels
of abstraction. Thus, a physical heterogeneous continuous complex and/or
simple dynamical hypersystem means a grouping of various dynamical systems,
hyposystems and components that operate together for a common purpose.
Scientific dynamical complexity correlates to the behaviour of the macroscopic
compilation of parts with the potential to change in time. This is altered from
mathematical dynamical complexity, i.e. the number of mathematical operations
indispensable to resolve a problem exercised in computer science.
Complexity theory and chaos theory both attempt to reconcile the unpredict-
ability of nonlinear physical heterogeneous continuous complex and/or simple
dynamical hypersystems with a sense of underlying order and structure. The
implications are: a paradigm where the short-term is predictable but long-term
planning is impossible; dramatic unexpected alterations; orderliness can be altered to
be more innovative and adaptive.
The Newtonian paradigm is founded on Cartesian reductionism (the machine
metaphor) and Cartesian dualism. The body is a physical machine and the mind is
separate from the body. The intuitive concept of a machine created from continuous
parts that can be reduced to those parts without losing its machine-like character is
Cartesian reductionism.
The Newtonian paradigm and the three laws of motion—the general laws of
motion—are applied as the basis of the modern scientific method. Dynamics is the
centre of the physical idea, which leads to trajectory.
Dynamical complexity is a post-Newtonian paradigm for a unified dynamical
systems approach to physical heterogeneous continuous dynamical hypersystems
composed of various dynamical constituent parts (dynamical systems, hyposystems
and components, respectively).
Dynamical complexity is rooted in the fundamental laws of nature. This opens
the way to a systematic study beyond mere analogies and metaphors, a multi-level
hierarchical dynamical systems approach with an interdisciplinary dimension.
Dynamical complexity has benefits beyond nonlinear dynamics and statistical
physics. By contrast to traditional fundamental science-rooted approaches,
dynamical complexity addresses a class phenomenon in which a dynamical
hypersystem and an observer may evolve at comparable scales of time and space.
Dynamical complexity comes from the failure of the Newtonian paradigm to be
general. Complex and/or simple dynamical hypersystems are disjointed compila-
tions that deal with all of nature, but the dynamical systems approach is invented
from complex things and the simple mechanisms are imagined and created by
science; experiments involve reducing the dynamical hypersystem to its constituent
dynamical parts (dynamical systems, hyposystems and components, respectively)
and then studying these parts in a background that is formulated on the subject of
dynamics.

1-3
 Mechatronics

How is science pursued? The senses provide information and mental activity
makes sense of that sensory information. Arrange a natural dynamical hypersystem
into a formal dynamical hypersystem and then manipulate the formal dynamical
hypersystem to imitate the causal change in the natural dynamical hypersystem.
From the formal dynamical hypersystem, derive an implication that relates to the
causal event in the formal dynamical hypersystem and then interpret the formal
dynamical hypersystem and verify its ability to represent the causal event in the
natural dynamical hypersystem and its constituent dynamical parts (dynamical
systems, hyposystems and components, respectively) before studying those parts
against a background formulated on the subject of dynamics.
For years, we complied with the Newtonian paradigm as the formal dynamical
hypersystem, not even remembering about there being encoding and decoding, and
gradually the ontology began to alter so that the Newtonian paradigm in reality
replaced or became the physical universe.
As we began to look more deeply into the dynamical systems approach, we came
up with aspects that the Newtonian paradigm failed to capture. Then we considered
an explanation to be necessary. Dynamical complexity was created. This can be
formalised easily. It has an acutely philosophical meaning.
Dynamical complexity is a phrase that expresses that it is laborious to formulate
the overall behaviour, even when almost comprehensive information is known about
its microscopic dynamical components and their interrelations. A commonly heard
spontaneous definition of dynamical complexity is that it is the ‘edge of chaos’
between order and randomness. A simple set of laws can lead to physical
heterogeneous continuous complex and/or simple dynamical hypersystems, corre-
lated to entropy (disorder) and the second law of thermodynamics. That law states
that entropy always increases in any closed physical heterogeneous continuous
dynamical hypersystem not in equilibrium, and remains constant for a hypersystem
which is in equilibrium.

1.2 Conclusions
There are two main modes of physical heterogeneous complex and/or simple
dynamical hypersystems: differential equations and difference equations of
dynamics (also known as iterated maps). The differential equations of dynamics
describe the evolution of physical heterogeneous continuous dynamical hyper-
systems in continuous time, whereas difference equations (iterated maps) arise in
physical heterogeneous discrete dynamical hypersystems where time is discrete. The
differential equations of dynamics are applied much more widely in science and
engineering, and the author shall therefore concentrate on them. Confining his
attention to differential equations of dynamics, the main distinction is between
ordinary and partial differential equations of dynamics. His concern here is purely
with temporal behaviour, and so he will deal exclusively with ordinary differential
equations of dynamics in a holor notation.
In its general conception, a physical heterogeneous continuous complex dynam-
ical hypersystem may be described as a complex of interacting dynamical parts

1-4
 Mechatronics

(dynamical systems, hyposystems and components, respectively), together with the

relationships among them that permit the identification of a boundary-maintaining
entity or process. From this perspective any given physical heterogeneous contin-
uous complex dynamical hypersystem can be described as a collection of dynamical
parts (dynamical systems, hyposystems and components, respectively), together with
a set of relationships between these dynamical parts. This is the origin of set
notation:
hypersystem = (systems, hyposystems, components)
with respect to this notation, ‘hypersystem (hypermatrix), systems (matrices),
hyposystems (hypomatrices), components (scalars)’, denoting, respectively, a
dynamical hypersystem, a set of dynamical parts distinguished within a dynamical
hypersystem and a relation (or, possibly, a set of relations) defined on dynamical
parts.
Clearly, the parthood and hypersystemhood properties of dynamical hypersystems
reside in dynamical systems, hyposystems and components, respectively. Having
defined the concept of dynamical hypersystems we then identify the domain of inquiry
of the dynamical systems approach (systems thinking) to be those aspects of
dynamical hypersystems that derive from their hypersystemhood properties as
opposed to their parthood properties.
A physical heterogeneous continuous complex dynamical hypersystem is
non-fragmentable. If it were fragmentable, it would be a machine. Its reduction to
dynamical parts (dynamical systems, hyposystems and components, respectively)
annihilates important dynamical hypersystem characteristics irreversibly.
A physical heterogeneous continuous complex dynamical hypersystem comprises
real dynamical components that are separate from its dynamical parts. There are
functional dynamical components defined by the dynamical hypersystem, whose
definition relies on the context of the dynamical hypersystem. Outside the dynamical
hypersystem, they have no meaning. If removed from the dynamical hypersystem,
they lose their original identity.
Physical heterogeneous continuous complex dynamical hypersystems have
mathematical models, analytical or synthetic. However, the implements differ. If a
synthetic mathematical model can replace an analytical mathematical model, the
dynamical hypersystem is fragmentable. There is no ‘major mathematical model’. If
there were a major mathematical model, all other mathematical models could be
created from it and fragmentability would result. Causalities in the dynamical
hypersystem are combined when they are dispersed over the parts. The nature of
causality necessitates feedbacks and feedforwards (closed loops) that are excluded in
the Newtonian paradigm. The important attributes of the physical heterogeneous
continuous dynamical hypersystem are beyond algorithmic definition or realisation:
a path to negate the well-known thesis that all the mathematical models
of computation that have been developed, and all those that may be developed
in the future, are equivalent in power. One will never find a more powerful
mathematical model.

1-5
 Mechatronics

Dynamical complexity is the feature of a physical heterogeneous continuous

dynamical hypersystem that is evident in the inability of any formalism to depict all
its properties adequately. It is necessary to find a ‘particularly different’ mode of
interacting with dynamical hypersystems.
Particularly different in the sense that when one creates successful physical models,
the formal physical heterogeneous continuous dynamical hypersystems considered
necessary to describe each continuous aspect are not derivable from each other.
The Newtonian approach was primarily guided by immediate observations,
which is quite natural in such a pioneer’s situation. Force, relative motion and
momentum formed the cornerstones of Newton’s mechanics. Energy, as a more
abstract concept, was added to classical mechanics through several steps as the
integrated force (Antippa 2000).
Newton’s equations of motion were perhaps too excellent for their time, making
motion more or less independent of its origin. When combined with Galilean
relativity, Newtonian motion found its reference at the observer, an obvious
misconception that was inherited by the theory of relativity. Relative velocity is
well justified in a kinematical sense as the rate of the change in the distance between
an object and the observer. Relative velocity does not, however, relate the quantity
of effective mass and kinetic energy stored in a moving object; the build-up of
effective mass and kinetic energy is related to the work done in obtaining the motion
in the relevant energy frame. Classical ether was implicitly fixed to the observer,
which made Newton’s laws of motion applicable to any observer as such.
The observer-oriented approach claiming the ‘same observations’ for any
observer was reflected in the relativity principle, which was presented to justify
the modification of the local ether environment so that observations looked the same
for any observer. The result was Lorentz modified inertial frames and space–time
modified gravitational frames.
The dynamical systems approach’s ‘multi-level ether’ is built up from the
dynamical hypersystem of cascaded energy frames. Motion gained from potential
energy in a local frame is paid back in the same frame.

1.3 Summary
The basic quantities of physical heterogeneous complex or simple continuous
dynamical hypersystems are defined below.
• Dynamical quantity: a physical quantity (position, momentum, phase, electric
field…) that depends on time.
• Dynamical hypersystem: a set of dynamical quantities that evolves together
under a coupled set of equations or maps.
• Continuous-time versus discrete-time dynamical hypersystems: this depends on
whether time is treated as a continuous or discrete variable:
* continuous-time dynamical hypersystems are governed by (sets of
coupled) ordinary differential equations (ODEs), partial differential
equations (PDEs), integro-differential equations, delay-differential equa-
tions of dynamics, etc; e.g. Newton’s second law, Maxwell’s equations,
etc;

1-6
 Mechatronics

* discrete-time dynamical hypersystems are governed by maps (iterate);

e.g. xn+1 = F (xn ).
• Linear dynamical hypersystem: a dynamical hypersystem, whose dynamical
quantities respond under the influence of a driving field or driving force
linearly with respect to that driving field or force. Such continuous dynamical
hypersystems are characterised by linear integro-differential equations of
dynamics.
• Nonlinear dynamical hypersystem: a dynamical hypersystem, whose dynam-
ical quantities respond under the influence of a driving field or driving force
nonlinearly with respect to that driving field or force. Such continuous
dynamical hypersystems are characterised by nonlinear integro-differential
equations of dynamics.

Reference
Antippa A F 2000 Can. J. Phys. 78 883–99

1-7
 IOP Publishing

Mechatronics
Dynamical systems approach and theory of holors
B T Fijalkowski

Chapter 2
Model definition

THEORY VERSUS MODEL

Insofar as the propositions of mathematics give an account of reality they are not
certain; and insofar as they are certain they do not describe reality.
Albert Einstein

A theory has only the alternative of being right or wrong. A model has a third
possibility; it may be right but irrelevant.
Manfred Eigen

The model is rational and can be understood by anyone with basic knowledge in
physics and mathematics.
Ari Lehto

2.1 Introduction
A basic aim of this textbook is to present a methodology for the formation of physical
and mathematical models of physical heterogeneous continuous complex and/or simple
dynamical hypersystems. Such models are necessary for the synthesis of these dynamical
hypersystems, and help to simplify the computer simulation analytical studies.
In science, models have been known for a very long time. For instance, models (not
then so-called) were conceived by the Greek philosopher, Democritus (circa 400 BCE–
circa 370 BCE), for the structure of the atom; a model structure of the geocentric
universe, the ‘Ptolemaic system’, was developed by the Greek mathematician,
astronomer and geographer, Ptolemy (Claudius Ptolemaeus, circa 90 BCE–circa
168 BCE) in Almagest (‘The Greatest’); and the model of the heliocentric universe, the
‘Copernican system’ was developed by the Polish astronomer, Nicolaus Copernicus

doi:10.1088/978-0-7503-1350-6ch2 2-1 ª IOP Publishing Ltd 2016

 Mechatronics

(1473–1543). Copernicus’s famous book, De Orbium Caelestium Revolutionibus (1543)

was banned by the Roman Catholic Church until 1758.
Ptolemaic astronomy described observations relative to the observer at rest on the
Earth. For describing the planetary orbits, a system of epicycles was necessary.
Copernican astronomy described planets as objects orbiting the Sun, which meant
that the observer was moving with the moving Earth. Such a holistic view created the
basis for a practical understanding of the system and the celestial mechanics behind
the motion. A German mathematician, astronomer, and astrologer Johannes Kepler
(1571–1630) found the mathematics in Copernicus’s planetary systems and Newton
the physics behind Kepler’s equations. Further, an English physicist, and mathema-
tician Issac Newton (1642–1727) formulated the concept of gravitation, and the laws
of motion, which were fixed to inertial observers’ frames, following the Italian
scientist Galileo Galilei’s (1564–1648) relativity. A German-born physicist Albert
Einstein extended the use of Galilean–Newtonian relativity and observer-oriented
inertial frames by redefining the concepts of the coordinate quantities, time and
distance. The roots of the energy system-based frames of reference and the zero-
energy principle were laid down by a German polymath and philosopher Gottfried
Leibniz (1646–1716) when he proposed (in 1694–96) that the build-up of vis viva
(living force or kinetic energy) is obtained against the release of vis mortua (dead force
or potential energy) by reflecting the balance of motion and gravitation in spherical
closed space; a localised energy object in the dynamical universe framework is like
Leibnitz’s monad—a perpetual living mirror of the Universe (Suntola 2006).
In his Baltimore Lectures, William Thomson (Lord Kelvin, 1824–1907 AD)
affirmed that to understand a phenomenon you need to construct its mechanical
model. The models are means to and a form of scientific perception and they have
found applications in strictly theoretical, analytical studies as well as experimental
studies. A notion of the model is in equivalence, which is used as a default here as an
unambiguous definition and a generally received classification.
Applying different methods of model formation, a multitude of models of
physical heterogeneous continuous dynamical hypersystems may be obtained. The
obtained multitude ought to be isomorphic against the functional physical hetero-
geneous continuous dynamical hypersystem, that is, it ought to exist as an abstract
homology between the multitude of models and a modelled physical heterogeneous
continuous dynamical hypersystem. Hence, for the same outputs of physical
quantities, the dynamical hypersystem and its model ought to be homologically
synonymous (in the sense of the selected criterion).
We will take ‘identification’ in this textbook to mean the formation process for the
models of the physical heterogeneous continuous dynamical hypersystems, which
aim to estimate their dynamical and static characteristics while using real informa-
tion on the physical quantities acting on inputs of the physical heterogeneous
continuous dynamical hypersystems and their interaction concerning these physical
quantities. All the models by convention are divisible into two classes: ‘physical
models’ and ‘mathematical models’, as shown in figure 2.1.
By a ‘physical model’ of the physical heterogeneous continuous dynamical
hypersystem we mean a model in which the physical processes occurring are

2-2
 Mechatronics

Figure 2.1. Classification of models (Fijalkowski 1987).

dealt with in a way that is equivalent to describing them through mathematical

relationships.
Physical models may be divided into two categories (figure 2.1): homonatural
mathematical models, which have the same physical nature as the physical
heterogeneous continuous dynamical hypersystem, and heteronatural mathe-
matical models, whose physical nature is different from the physical nature of the
physical heterogeneous continuous dynamical hypersystem. Such a model turns out,
for example, to be a physical model of the physical heterogeneous continuous
complex and/or simple dynamical hypersystem, instead of a physical model of the
physical heterogeneous continuous dynamical hypersystem’s physical commutation
matrixer, which is described in chapter 7.
In the 20th century something new emerged that was more adaptable to present-
day scientific methodologies—the notion of the ‘mathematical model’.
Mathematical models may be divided into two groups (figure 2.1). The first
comprises ‘deterministic’ mathematical models, and the second comprises ‘statis-
tical’ (random) mathematical models.
A deterministic mathematical model of the physical heterogeneous continuous
dynamical hypersystem is a model whose description is possible in the form of the
function relationships between inputs and outputs of the physical heterogeneous
continuous dynamical hypersystem.
As a rule, a mathematical model is adequate for a physical process, functional
and structural dynamical components and superimposed constraints.
A statistical mathematical model of the physical heterogeneous continuous
dynamical hypersystem is a model that is formulated while taking into consideration
the statistical parameters and the probability expansion function.

2-3
 Mechatronics

2.2 Physical model

Physical models are designated to reflect the physical processes of the physical
heterogeneous continuous complex and/or simple dynamical hypersystem. Models
of the first kind differ from the physical heterogeneous continuous dynamical
hypersystem only by their geometrical dimensions and a range of parameter
variations. This is why processes that have a place in the physical heterogeneous
continuous dynamical hypersystem and their models only differ quantitatively. This
makes it possible to the study the physical nature of the physical heterogeneous
continuous dynamical hypersystem, and its transient processes and boundary
conditions, using the best computation methods and fundamental theorems.
To estimate the applicability of a formatted physical model that has the same
physical nature as the physical heterogeneous continuous dynamical hypersystem,
one ought to make a preliminary estimate of the degree of the analogy between the
physical processes in the model and those in the physical heterogeneous continuous
dynamical hypersystem.
By the formation of physical models, it is possible to use a mathematical
description of these physical processes that characterises the physical heterogeneous
continuous dynamical hypersystem. The identity of the physical processes occurring
in the real physical heterogeneous continuous dynamical hypersystem and its model
may be provided by using a probability theory without an aforementioned
mathematical description. Physical processes are called ‘homothetic’ (reciprocal),
if the relations of physical quantities that characterise them are constant. The
formation process for physical models resolves itself into the selection of model
parameters that concern a probability theory and the selection of homological initial
and boundary conditions.
The theory of probability is based on the three theorems presented, for example, in
the works of Vienikov (1956, 1961). These theorems constitute the mutual relations
between the physical quantities which characterise the nature of physical processes.
In compliance with the first theorem of probability, in similar physical processes the
probability criteria are the same numerically.
To develop physical models that are compatible with the second probability
theorem, the number of degrees of freedom (DoF) cannot be greater than the number
of fundamental units for the measurement system adapted to describe a given
physical phenomenon. This means that if a physical process in the physical
heterogeneous continuous dynamical hypersystem depends upon q physical quanti-
ties, of which b are fundamentals and d = q – b are derivatives, it may be possible to
create d probability criteria. In this way, at the formation of the physical models, the
physical quantities d should be given going out from the probability criteria.
According to the third probability theorem, for physical heterogeneous continuous
dynamical hypersystems and physical models in which the conditions of unambi-
guity are similar, the probability criteria created from the conditions of unambiguity
are numerically the same. According to this theorem there should be additional
conditions to select a single physical model from all-possible physical models and in
this way unambiguously formulate a physical model of the physical heterogeneous

2-4
 Mechatronics

continuous dynamical hypersystem. A physical model is ‘synchronous’ with

a physical nature in this case if a physical process runs across in a natural time
scale. If a physical model functions in an artificial time scale, then it is ‘homogenous’
(Pukhov and Khatiashvili 1971).
The ratios between the probability constants are called the ‘indicators’; however, the
probability criteria are sometimes called the ‘invariants’. For the same results, it is
allowed to apply an analysis of dimensionality. The formation of physical models on
functional and structural dynamical components that differ from the physical nature of
these dynamical components is based on the probability of mathematical relationships,
which describe physical processes occurring in dynamical systems and ‘heteronatural
physical models’, i.e. with a varying physical nature. So as in ‘homonatural physical
models’, i.e. with the same physical nature, heteronatural physical models are created
in this way, and for each real functional and structural dynamical element of the
physical heterogeneous continuous dynamical hypersystem there will be a correspond-
ing determined functional and structural dynamical component of the physical model.

2.3 Deterministic mathematical model

Mathematical modelling is a more generalised form of modelling than physical
modelling. The approach in mathematical modelling differs somewhat from that in
physical modelling. In physical modelling a physical model may have the same
physical nature as the heterogeneous continuous dynamical system. In the case
where a physical model has been created with different functional and structural
dynamical components from those of the real heterogeneous continuous dynamical
system, there invariably remains a requirement for maximal identity between
the structure and the mathematical relationships’ parameters. In mathematical
modelling, this requirement is rejected. The real heterogeneous continuous complex
and/or simple dynamical hypersystem and its mathematical model ought to be
equivalent in the sense of the principle of operation.
For instance, in a mathematical model it is permitted to perform an analytical
study by means of a computer, although the processes occurring in the computer
may be completely different. Deterministic mathematical models in their essence
differ from statistical mathematical models. These differences between the two
aforementioned kinds of mathematical models are conditioned by the specifics of
different classes of problems, which for their solution require one model’s features
but not another’s. A deterministic mathematical model is one whose behaviour is
entirely predictable. The physical heterogeneous dynamical hypersystem is perfectly
understood, and then it is possible to predict precisely what will happen.

2.3.1 Synthetic deterministic mathematical model

A synthetic deterministic mathematical model of the physical heterogeneous
continuous dynamical hypersystem, relating to dynamical and static states,
establishes a set of nonlinear differential equations with partial derivatives, when
it imitates in an exact manner the real physical heterogeneous continuous complex
and/or simple dynamical hypersystem.

2-5
 Mechatronics

2.3.2 Chaotic deterministic mathematical model

A chaotic deterministic mathematical model is one where the behaviour of the
physical heterogeneous continuous complex and/or simple dynamical hypersystem
cannot be entirely predicted. The detailed study of chaos is beyond the topic of this
textbook; here we only introduce a definition of the chaotic deterministic mathe-
matical model. The most important components of chaotic behaviour are the
following.
• Sensitive dependence on initial conditions in a bounded set;
• aperiodicity.

2.4 Statistical mathematical model

A statistical mathematical model has a considerably more formal character than a
deterministic mathematical model, because it reflects the physical heterogeneous
continuous complex and/or simple dynamical hypersystem, disregarding its concrete
physical features.

2.5 Stochastic mathematical model

A stochastic mathematical model is one where the behaviour of the physical
heterogeneous continuous complex and/or simple dynamical hypersystem cannot
be entirely predicted.

References
Fijalkowski B 1987 Modele matematyczne wybranych lotniczych i motoryzacyjnych mechano-
elektro-termicznych dyskretnych nadsystemów dynamicznych (Mathematical Models of
Selected Aviation and Automotive Continuous Dynamical Hyperystems), Monografia 53
Politechnika Krakowska imienia Tadeusza Kościuszki, Krakow, 276 p. (In Polish)
Pukhov G E and Khatiashvili C S 1971 Lokalnyie matematicheskoye modeli sistem upravleniya
(Local Mathematical Models of Control Systems) (Kiiev: Naukova dumka) (In Russian)
Suntola T 2006 The Dynamical Universe: Relativity defines the locally available share of total
energy Physical Interpretations of Relativity Theory, PIRT X (London: Imperial College)
Vienikov V A 1956 Fizicheskiye modeli elektricheskikh sistem (Physical models of electrical
systems) (Moscow: GEI) (In Russian)
Vienikov V A 1961 K voprosu o klasifikaciy modeley i metodov modelirovaniya (To question on
classification of models and methods of modelling). Izv. Vuzov, Energetika, No. 10, (In
Russian)

2-6
 IOP Publishing

Mechatronics
Dynamical systems approach and theory of holors
B T Fijalkowski

Chapter 3
Dynamical systems approach

Always strive to construct definitions, or at the least a rudimentary image of that

which has arisen in your mind; strip it bare and focus on its true nature,
considering its entirety through its elements and its elements through its entirety.
Marcus Aurelius

Less emphasis on analysis

More emphasis on synthesis.
Anonymous

3.1 Introduction
The first advance of modern Western science occurred during the Renaissance, a
cultural movement that spanned the period roughly from the 14th to the 17th
century, beginning in Italy in the late Middle Ages and spreading to the rest of
Europe. The prime attributes are ‘analytical vision’ and ‘Cartesian reductionism’.
The philosophy of Descartes was not only of prime importance for the advance of
science, but it has also had an impressive impact on the Western ‘universal’
approach to thinking up to the present day. The Eastern ‘universal systems
thinking’, derived from biology and medicine, is more holistic, based on the view
that all things and events deduced by the senses are interrelated and connected, and
cannot be divided during study; hence it focuses more on the relations between
dynamical parts than on the essence of each dynamical part.
The modern universal systems approach or universal systems thinking was
founded in the 20th century, but primitive versions emerged several thousand years
before. In the ancient period, both Chinese and Greek philosophers created holistic
visions of the Universe. At that time, science, philosophy and religion were not
separated. These types of holistic visions are intuitive and from a contemporary

doi:10.1088/978-0-7503-1350-6ch3 3-1 ª IOP Publishing Ltd 2016

 Mechatronics

viewpoint they are founded on primitive systems intuition. As for the science,
as Einstein pointed out, ‘The only valuable thing is intuition’.
Intuition can be considered as a mode of thinking and also as a kind of knowledge
exploitation and creation. Intuition is an ability possessed by everybody. It is strictly
related to tacit knowledge. Intuition is a preliminary step when scientists take a
universal systems approach to a project. It has been suggested that this kind of
intuition be called ‘systems intuition’ and it has more significance in systematics
(systems engineering). Systems intuition or systems intuitive thinking is characterised
by both holistic and intuitive vision. The scientific foundation of systems intuition is
based on the noetic (cognitive) science initiated in the 20th century by the famous
Chinese scientist, Xuesen Qian (Hsue-Shen Tsien), the father of space technology in
China. He argued that the limitation of traditional cognitive science is that it focuses
largely on analytical (logical) thinking and ignores imaginative thinking, which
has the principal role in decisions and innovation. He is also the originator of
‘systemology’ and a science of thinking.
The neotenic features of systems intuition lie in the concept of the synthesis
of imaginative thinking and creative thinking based on experience (Wang 2011).
An idea of the ‘dynamical systems approach’ in the methodology of the
formulation of mathematical models of physical heterogeneous continuous
dynamical hypersystems was originated by the author of this book under the
influence of a book written by Weinberg (1975, 1979) entitled An Introduction
to General Systems Thinking, which described different situations and methodo-
logical operations appearing in research and development (R&D) works
step-by-step, showing them against a background of problems for a solution
(Fijalkowski 1987).
The dynamical systems approach in the methodology for the formulation
of mathematical models of physical heterogeneous continuous dynamical hyper-
systems facilitates the formation of mathematical models and offers a concept for
their analytical study by computer, just using systems thinking, enabling skilful
methodological progress.
General systems thinking is an approach to integration that is based on the
belief that the dynamical parts of a dynamical hypersystem will act differently
when isolated from the dynamical hypersystem’s environment or other parts of
the dynamical hypersystem. Standing in contrast to positivist and reductionist
thinking, systems thinking sets out to view dynamical hypersystems in a holistic
manner.
Consistent with systems philosophy, systems thinking concerns understanding a
dynamical hypersystem by examining the linkages and interactions between the
dynamical parts (dynamical systems, hyposystems and components, respectively)
that comprise the whole of the dynamical hyposystem.
When scientists and engineers encounter situations that are complex and messy,
systems thinking can help one to understand the situation systemically. This helps
scientists and engineers to see the complex dynamical hypersystem—from which
they may identify multiple leverage points that can be addressed to support
constructive change. It also helps them see the connectivity between the dynamical

3-2
 Mechatronics

parts (dynamical systems, hyposystems and components, respectively) in the

situation, so as to support joined-up actions.
The never-enough-to-be-regretted Professor Bohdan Walentynowicz (past editor
of the Polish Electrotechnical Review) in a preface to the first Polish edition of the
book written by Weinberg (1979) wrote:

Maybe what most accurately indicates the nature of the book written by
Weinberg are the words printed on the book jacket of the original. One reads
in them about how young workers in science can learn systems thinking in two
ways. The first way—the specialist way—depends on mastering the detailed
knowledge of a certain field, and then on crossing to generalisations, however at
the comprehensible limitation of the field of sight because of the bending domain.
The second way—the generalist way—involves learning about thinking, and then
thinking in the categories of generalised systems, before the thinking domain,
and sometimes it is concurrent with or also integrates specialist methods. This
second way of thinking it is not limited to young workers in science, but also
applies to mature scholars, engineers, organisers, students and different people.

The author of this book gives up on defining the notion of the system, since
this just deals with many authors of prominent books (Wilson 1969, Klir 1979,
Weinberg 1979).
However, so the reader of this book can assimilate the notion of the dynamical
hypersystem more easily, we will return to the words of Professor Bohdan
Walentynowicz (Weinberg 1979), that this is a notion of ‘some whole—entering
into the composition of the larger wholes, created from parts (the smaller wholes),
related in the way a certain structure is assigned, and distinguished with regards to
some functions assigned to these wholes’.
The dynamical systems approach in the methodology of formatting mathematical
models of physical heterogeneous continuous dynamical hypersystems concerns
their analysis and also the projection of their relations, which because of the aim of
dynamical systems are interesting.
In getting to know a physical heterogeneous continuous dynamical hypersystem, it
is not desirable to examine the whole at once, one should decide on a limitation and
investigate it step by step through its dynamical components, its functional and
structural dynamical components. It is a matter of choosing what belongs to a given
dynamical hypersystem and what does not, and in doing this, a matter of defining what
that dynamical hypersystem is.
The physical heterogeneous continuous dynamical system is the thought construc-
tion created by the researcher, and so it is—Weinberg (1979) underlines this several
times in his book—a certain ‘way of seeing the world’, for example, the ‘physical
universe’ of the physical heterogeneous continuous complex and/or simple hypersystem.
When considering any functional physical heterogeneous continuous complex
and/or simple dynamical hypersystem, encountered as a functional structure, there
are a great number of diverse phenomena, which relate to various fields of
theoretical and applied sciences.

3-3
 Mechatronics

Figure 3.1. A hierarchical structure of a physical model of an abstract functional continuous complex
dynamical hypersystem (Fijalkowski 1987).

From a definite point of view, one can perceive similar functional and structural
dynamical components in these heterogeneous phenomena (e.g. physical phenom-
ena), which sometimes seem various. To investigate these phenomena from the same
point of view or with mutual correlations we introduce in this book an abstract
notion that relates to all phenomena in the same degree. This is the notion of the
‘abstract functional continuous complex and/or simple dynamical hypersystem’
(figure 3.1). However, the principles established for these phenomena were
termed the ‘dynamical systems approach’ in the formulation methodology of
the mathematical models for abstract functional continuous complex and/or simple
dynamical hypersystems. Thanks to this, in a methodology for the formulation of
mathematical models of functional objects (e.g. machine aggregates and executive
mechanisms), simplicity was found in the complexity.
Introducing the dynamical systems approach into the methodology for the
formation of mathematical models for abstract functional continuous complex and/
or simple dynamical hypersystems complies with the idealisation of this textbook—
the ‘continuous’—letting us describe a dynamical state of a considered complex and/or
simple dynamical hypersystem through a finite number of certain physical quantities.
The complex and/or simple dynamical hypersystems in which such an idealisation is
admissible are termed ‘continuous’.
As an example, the abstract functional continuous complex and/or simple
dynamical hypersystem (figure 3.1) can be folded, respectively, with the finite number
of continuous complex and/or simple dynamical systems, respectively, consisting of
the finite number of continuous complex and/or simple dynamical hypersystems,
folded with the finite number of continuous complex and/or dynamical hyposystems,
or functional and structural dynamical elements, respectively.
The basic requirement of the dynamical systems approach in the methodology for
the formation of mathematical models of physical heterogeneous continuous
complex and/or simple dynamical hypersystems is the existence of: a researcher
studying the functional object (e.g. a machine aggregate or an executive mechanism)
and the solved problem, marking the point of the researcher’s view of the studied

3-4
 Mechatronics

functional object. So now, created by the researcher, the abstract functional

continuous complex and/or simple dynamical hypersystem (figure 3.1) is the physical
model of the original ‘considered’ by him from a definite point of view. What
applicability does the described hierarchical structure of the physical model of the
abstract functional continuous complex dynamical hypersystem have?
Firstly—it contains on every hierarchical level the full set, possible for this level,
of the abstract functional continuous complex dynamical systems, complex
dynamical hyposystems, simple dynamical hypersystems, simple dynamical systems,
simple dynamical hyposystems and simple dynamical components. This means that
it can be neither more nor less on the given level by even one abstract functional
continuous complex dynamical system, complex dynamical hyposystem, simple
dynamical hypersystem, simple dynamical system, simple dynamical hyposystem, or
simple dynamical component, and that there can be such kinds that only this
physical model simulates. This assures in practice the complexity of the analysis
and synthesis, for example, of any functional continuous complex dynamical
hypersystem.
Secondly—in the aim of a simulation of a concrete functional object, for example,
a physical heterogeneous continuous complex dynamical hypersystem, one selects
on every graph branch of the physical model (or one skips certain graph branches),
the most distant from the basis of the graph tree (from a continuous complex
dynamical hypersystem), such a continuous simple dynamical hypersystem which
for researchers still has practical meaning in considerations, for example, of the
physical heterogeneous continuous simple dynamical hypersystem. All the chosen
(sorted) graph branches of the physical model end on a significant continuous simple
dynamical hypersystem and still create the set of elementary functional continuous
simple dynamical systems and relationships among them, serving to build, for
example, the physical heterogeneous continuous simple dynamical hypersystem.
One can build the physical model on this basis, simulating the concrete (e.g.
physical) heterogeneous continuous simple dynamical hypersystem, considered
from the definite point of vision.
Thirdly—it lets us compare among themselves various concrete (e.g. physical)
functional heterogeneous continuous dynamical hypersystems (folded with
elementary functional homogeneous continuous simple dynamical systems, simple
dynamical hyposystems and simple dynamical components), of the same level by
transformation to their functional objects. The aforementioned theses are not only
valid for a received division and classification of the physical heterogeneous
continuous simple dynamical hypersystems, but also for every different division
under the condition that will be kept in the method of formation (introduced in this
textbook) for the successive hierarchical levels. The hierarchical structure of the
physical model of the abstract heterogeneous continuous simple dynamical hyper-
sphere favours certain proprieties. Namely, when its level (beyond the level making
up the complex dynamical hypersystem, that is the object, e.g. the aeroplane,
helicopter, automobile, tractor, etc) contains the functional continuous complex
dynamical systems as subjects acting on the things of their objects, they are on the
level directly below and simultaneously the objects of the subjects, and acting

3-5
 Mechatronics

Figure 3.2. A hierarchical structure of a physical model of the aviation or automotive functional
heterogeneous continuous complex dynamical hypersystem (Fijalkowski 1987).

of course directly on the higher level. A hierarchical structure on every level,

moreover, contains the full set of possible abstract functional homogeneous
continuous complex dynamical systems, complex dynamical hyposystems, simple
dynamical hypersystems, simple dynamical systems, simple dynamical hyposystems
and simple dynamical components created according to the previously described
principles. Moreover, this hierarchical structure can at last be extended vertically in
an unrestricted way. Therefore it has a theoretical character, because for practical
aims the analysis incorporates the set of selected (by the author of this book)
holonomic and unholonomic physical heterogeneous continuous complex dynam-
ical hypersystems that are present in this structure (Fijalkowski 1987). It answers the
requirements of these definitions for aircraft, helicopters, automobiles and tractor,
and one can classify them to a class of phenomena called the physical universe, for
instance, of aviation and automotive functional heterogeneous continuous complex
dynamical hypersystems (figure 3.2).
In this physical universe, one distinguishes several functional classes of the
physical heterogeneous continuous complex dynamical hypersystems, which may
be presented in order of growing complexity, although in this book we are only
interested in the three initial levels, namely:
• the level of static functional structures incorporating the ‘anatomy’ of
aviation or automotive complex dynamical hypersystems (automobile, trac-
tor, aeroplane and helicopter), i.e. the name of the functional continuous
complex dynamical systems and settlement of their mutual positions, that is—
among others—the aviation or automotive continuous complex dynamical
system, i.e. aviation and automotive electrical equipment;

3-6
 Mechatronics

• the level of functional continuous complex dynamical hyposystems, that is—

among others—the aviation or automotive continuous complex dynamical
hyposystem, i.e. the aviation or automotive system electro- or mechano-
energetic system incorporating the foreseen and repeated influence on itself of
concrete simple dynamical hypersystems;
• the level of functional continuous simple dynamical hypersystems, that is—
among others—aviation or automotive simple dynamical hypersystems,
incorporating elementary continuous simple dynamical systems with
definite basic dynamics, that is deterministic, continuous simple dynamical
hypersystems, making up aviation or automotive devices and/or machines.

This classification of the complex and/or simple dynamical hypersystem of the level
of the higher order contains, respectively, the complex and/or simple dynamical
systems of the level of the lower order. Moreover, just this hierarchy of complex and/
or simple dynamical hypersystems makes it possible to use the methods applied in
the synthesis and analysis of simpler dynamical hypersystems for the formulation of
the mathematical models, opinion, building and investigations of more complex
and/or simple dynamical hypersystems.
To obtain the picture of the hierarchical function and structure of the concrete
physically continuous complex dynamical hypersystem and its functional continu-
ous simple dynamical systems and their functional continuous dynamical simple
hypersystems its decomposition was executed. The physical model of the concrete
physical heterogeneous (mechano-electro-thermal) continuous complex dynamical
hypersystem whose hierarchical structure was built is shown in figure 3.3.
It should be noted that in the process of the successive decomposition of the
concrete physical heterogeneous continuous complex dynamical hypersystem we
distinguished its functional simple dynamical systems built up from its functional

Figure 3.3. A hierarchical structure of the physical model of the physical heterogeneous (mechano-
electro-thermal) continuous complex dynamical hypersystem (Fijalkowski 1987).

3-7
 Mechatronics

continuous simple dynamical hyposystems, built-up from its functional continuous

simple dynamical components.
Functional heterogeneous continuous complex dynamical hypersystems,
simple dynamical systems, simple dynamical hyposystems and simple dynamical
component are thus exchangeable notions. Depending on the hierarchical level,
prepare the road for the use of the dynamical systems approach in the formulation
methodology of the mathematical models for computer analytical studies of both the
dynamical and the static processes of the physical heterogeneous continuous
complex dynamical hypersystem. A review should be carried out of the classic
interpretation and behaviour during a continuous sequence of transformations of
the abstract functional continuous simple dynamical hypersystem (figure 3.1).
Attention should be paid nearby to the principal result of the statement about
unambiguous solutions of the differential equations of dynamics, because this
defines a feature of the half-group called the ‘right of causality’. It is shown below
that the variation process is a generalisation of this concept.
To illustrate the use of the dynamical systems approach in the formation
methodology of mathematical models of functional objects (e.g. machine aggregates
and executive mechanisms) in its simplest generalised form, the ways that it leads to
mathematical models for the case of selected physical homogeneous continuous
simple dynamical hypersystems are shown in later chapters. Finally, to acquaint the
reader with the dynamical systems approach in the formulation methodology of
mathematical models we present the results of analytical studies of these selected
physical heterogeneous continuous dynamical hypersystems.

3.2 The dynamical systems approach to physical heterogeneous

continuous dynamical hypersystems
The dynamical systems approach is useful in the development of physical
heterogeneous continuous complex and simple dynamical hypersystems.
The premise of the dynamical systems approach is that the development of a
physical heterogeneous continuous complex or simple dynamical hypersystem
should begin with a thorough understanding of what is intended, including the
heterogeneous continuous dynamical hypersystem’s dynamical behaviour. If such an
understanding can be gained in the early stages of a project, it is possible to eliminate
much of the uncertainty now inherent in the process, and to develop physical
heterogeneous continuous dynamical hypersystems that are more successful yet less
expensive. The understanding of a physical heterogeneous continuous dynamical
hypersystem’s dynamical behaviour in its environment over time is most often
addressed in the design implementation and testing phases of a project, through
iterative trial and error.
The dynamical systems approach provides an alternative to this costly process.
It allows one to build a usable and comprehensive physical heterogeneous contin-
uous dynamical hypersystem—a physical model that is accurate enough to be relied
on and clear enough to be useful. The physical model addresses the usual functional

3-8
 Mechatronics

and data flow issues, but also covers the dynamical and behavioural aspects of a
physical heterogeneous continuous dynamical hypersystem. This physical model can
then be tested with the mathematical model analysis and retrieval tools, which
provide extensive mechanisms for inspecting and debugging the specification and
retrieving information from it. By testing the physical model systems scientists and
engineers can see how the physical heterogeneous continuous dynamical hyper-
system as specified would behave if implemented. One can answer detailed ‘what if?’
questions, following specific scenarios, checking that certain desirable situations
will occur in the lifetime of the physical heterogeneous continuous dynamical
hypersystem and that other undesirable ones will not.
In this sense, systems scientists and engineers can be said to be playing the role of
the eventual user of the physical heterogeneous continuous dynamical hypersystem
as well as that of the environment. Moreover, one can also instruct the analytical
tools themselves to play the role of the user and the environment. This kind of self-
inspection capability has the advantage of testing scenarios that the systems
scientists and engineers might not have thought of or might not want to check
out individually. Often these scenarios represent unpredictable behaviour.
The physical model must be well structured, intuitive and comprehensive if it is
to be useful to the variety of systems scientists and engineers involved, and it must
be rigorous and formal if it is to serve as the basis for useful analysis and retrieval.
This calls for representing as much of the physical model as possible in visual
formalisms—diagrammatical languages with precise meaning.
The systems scientists and engineers model the physical heterogeneous continuous
dynamical hypersystem using natural pictorial descriptions that can be understood
by physicists, but which also have unambiguous underlying semantics that are
meaningful to the computer and can thus be processed by the analytical tools.
The dynamical systems approach meets all of these requirements. It is consistent
with existing structured methods, but is more powerful and more flexible.
This textbook explains the philosophy behind the dynamical systems approach.
When one says ‘approach’, one has in mind an entire spectrum of concepts,
methods, languages and tools.
The systems scientists and engineers work with a set of computerised tools that
support a number of graphical physical commutation matrixers and non-graphical
languages, including languages for both physical and mathematical modelling and
the theory of holor analysis.
Many of the constituents in the dynamical systems approach are similar to those
appearing in other approaches and methodologies for physical heterogeneous
continuous dynamical hypersystem design, a fact that should be taken as evidence
of their wide appeal. However, there are many aspects of the dynamical systems
approach that go far beyond what is currently available elsewhere.
The author has discussed the dynamical systems approach as it applies to all
phases of physical heterogeneous continuous dynamical hypersystem development,
from physical and mathematical modelling through simulation and analysis to
design and implementation.

3-9
 Mechatronics

The dynamical systems approach is based on the preparation and extension of a

rigorous physical model, e.g. a physical commutation matrix that captures not only
functionality but also dynamical behaviour. These models are constructed using
visual formalisms that are clear and intuitive, but precise and rigorous, and thus
amenable to computerised analysis.
The dynamical systems approach allows systems scientists and engineers to
address the most crucial problem in the successful development of physical
heterogeneous continuous dynamical hypersystems: the ability to gain an early
understanding of a system’s dynamical behaviour and to carry that understanding
through the entire development cycle.
The dynamical systems approach will continue to develop as systems scientists
and engineers gain the experience needed to advance the state of the art.
The computer aided systems engineering (CASE) environment will also continue
to evolve: its ability to address future needs is the result of its being based on the
firm foundation of comprehensive physical and mathematical modelling through
simulation and analysis.
The principal heuristic innovation of the dynamical systems approach is what
may be called ‘reduction to dynamics’ in contrast to ‘reduction to dynamical
components’ as practised in the methodologies of classical science.

3.3 Summary
The dynamical systems approach:
• offers a holistic—from whole to local—approach for describing physical
reality;
• produces a comprehensive picture of physical reality;
• produces accurate predictions with straightforward mathematics and clear
logic.

The future of the dynamical systems approach is going to be very stimulating over
the coming years. Science is developing, changing and expanding—and that is what
maintains our enthusiasm.

References
Fijalkowski B 1987 Modele matematyczne wybranych lotniczych i motoryzacyjnych mechano-
elektro-termicznych dyskretnych nadsystemów dynamicznych (Mathematical Models of
Selected Aviation and Automotive Continuous Dynamical Hyperystems), Monografia 53
Politechnika Krakowska imienia Tadeusza Kościuszki, Krakow, 276 p. (In Polish)
Klir K 1972 Trends in General System Theory (New York: Wiley).
Wang Z 2011 Study of System Intuition from Noetic (cognitive) Science Founded by Xuesen
Qian. DUT Presentation, 2011, In Memorandum of 100 Aniversary of Academician Xuesen
Qian. Research Center of Knowledge Wane Sciences and Technology, Dalian University of
Technology, 116085 Dalian, China, ss. 1-14
Weinberg G M 1975 An Introduction to General Systems Thinking (New York: Wiley) 1975

3-10
 Mechatronics

Weinberg G M 1979 Myslenie systemowe (Systems Thinking). Wydawnictwo Naukowo-

Techniczne
Wilson W E 1969 Projektowanie techniczne w ujeciu systemowym (Technical Design in the Systems
Approach). Wydawnictwo Naukowo-Techniczne (In Polish)

3-11
 IOP Publishing

Mechatronics
Dynamical systems approach and theory of holors
B T Fijalkowski

Chapter 4
Algorithm for a formulation of the
mathematical model

THEORY VERSUS EXPERIMENT

It is the mark of an educated mind to rest satisfied with the degree of precision,
which the nature of the subject admits, and not to seek exactness where only
approximation is possible.
Aristotle (384 BCE–322 BCE)

Experience is the best teacher.

Marcus Tullius Cicero

Who was it that invented this in-depth analogy, unseeable with the eyes but which
the mind can grasp? It is mathematical thought, which holds matter in contempt,
attracted to pure form.
Henri Poincaré

While experiment reveals the particular features of every process, the theory
permits the establishment of the general laws and thus contributes to the
fundamental knowledge of the process.
Rosenthal

Great theoreticians know that hypothesis must be confirmed with experiment.

Albert Einstein

doi:10.1088/978-0-7503-1350-6ch4 4-1 ª IOP Publishing Ltd 2016

 Mechatronics

The proposal of physical heterogeneous continuous complex or simple dynamical

hypersystems is introduced as energy and/or information handling ‘physical
commutation matrixers’, which are identified as physical models. Sets of generalised
energy-potential-difference holors and generalised energy-transfer holors and phys-
ical homogeneous continuous dynamical systems, hyposystems and components,
respectively, are developed. The proposals are then made more precise with
reference to mechanical, electrical, fluidic (hydraulic and/or pneumatic), magnetic
and thermal homogeneous continuous dynamical systems.
The energetics and informatics restrictions introduced are applied to intercon-
necting physical homogeneous continuous dynamical systems, hyposystems and
components, respectively, and used to identify a selection of interconnectivity
constraints and hence obtain mathematical models.
An algorithm for the formulation of the mathematical model of any physical
heterogeneous continuous complex or simple dynamical hypersystem comprises not
only the steps required for the arrangement of the differential equations of dynamics,
but also all those necessary for the identification and verification of the model,
namely computer-aided analytical studies, e.g. a computer simulation of the physical
heterogeneous continuous complex or simple dynamical hypersystem and the
comparison of the results with those of experimental studies. In the general case,
the algorithm contains four steps (figure 4.1).

Figure 4.1. An algorithm for the formulation of a mathematical model of the existing and/or designed physical
heterogeneous continuous dynamical hypersystem.

4-2
 Mechatronics

• The recognition of the structure, principle of operation and features of the

studied physical heterogeneous continuous dynamical hypersystem and the
assumptions for simplifying the created physical model. This step of physical
modelling has to bring the dynamical and static characteristics of the
physical model and the existing (real) physical heterogeneous continuous
dynamical hypersystem into sufficient agreement.
• The formulation of the mathematical model comprising a set of differential
equations describing the dynamics and statics of the physical model. This step
of mathematical modelling does not introduce important new (content-related)
information to a description of the physical heterogeneous continuous
dynamical hypersystem, and only realises the assumptions of the physical
model in mathematical language.
• The analytical studies of the dynamics of the mathematical model. This step
consists of the solution of previously arranged differential equations and the
design of the dynamical characteristics of the considered mathematical
model.
• The comparison of the foreseen dynamics of the physical heterogeneous
continuous dynamical hypersystem, which are the results of the analytical
studies of the mathematical model with the assumed characteristics of the
designed physical heterogeneous continuous dynamical hypersystem, with the
results from experimental studies of the already existing (real) physical
heterogeneous continuous dynamical hypersystem. This step of verifying
the mathematical model in a case of unsatisfactory agreement of the
analytical studies’ results for the mathematical model with the assumed
characteristics has the aim of executing a suitable alteration of the parameters
in the physical model, or in the case of the designed physical heterogeneous
continuous dynamical hypersystem, the alteration (change) of the physical
model, and even the constructional solution (see chapter 3). One should
repeat the next step of the algorithm after executing the last step and the
settlement of the preliminary project of a construction of the dynamical
hypersystem until one has obtained sufficient agreement between the results
of the analytical studies of the dynamics and the assumed characteristics of
the designed dynamical hypersystem.

In figure 4.1 the four steps of the formulation algorithm for a mathematical model
of any existing and/or designed physical heterogeneous continuous dynamical
hypersystem are presented by means of a signal-flow graph.
In relation to the obviousness of the subordination the lines of the graph’s
influence on the respective steps of the algorithm here do not give a precise
description of the signal-flow graph, and only direct the attention to stepping out
in the feedbacks and feedforwards, realising the verification of the mathematical
model with reference to the existing (real) or designed physical heterogeneous
continuous dynamical hypersystem.

4-3
 IOP Publishing

Mechatronics
Dynamical systems approach and theory of holors
B T Fijalkowski

Chapter 5
Developmental systems approach

There is nothing permanent except change.

Herbalists of Ephesus (circa 525–475 BCE)
Change alone is unchanging.
Heraclitus
Observe constantly that all things take place by change, and accustom thyself to
consider that the nature of the Universe loves nothing so much as to change the
things which are, and to make new things like them.
Marcus Aurelius
Nothing happens until something moves.
Albert Einstein
To improve is to change; to be perfect is to change often.
Winston Churchill
The world hurries forward—everything changes.
Anonymous
This book attempts to introduce the developmental systems approach into the
methodology of physical models of the physical heterogeneous continuous dynamical
hypersystems, in addition to the dynamical systems approach. So far this has met
with indifference not only from some theorists on the research side of mathematical
models, but also some practitioners making experimental studies. In the struggle to
improve both the physical model, as well as the (actual or projected) physical
heterogenous continuous dynamical hypersystem, the dynamical systems approach
still predominates. This leads to some analytical studies researchers falling for the
illusion that they are in control of actual physical models. The developmental systems
approach turned out not to have been a panaceum for the routine of physical model
creation. To break this routine, a considerably stronger impulse is necessary, an

doi:10.1088/978-0-7503-1350-6ch5 5-1 ª IOP Publishing Ltd 2016

 Mechatronics

impulse that would push researchers to create the macro-conditions needed for
computerised analytical studies of mathematical models. If the theorists—systems
scientists and engineers—and researchers of mathematical models are interested
in their computerised analytical studies as well as their verification (checking)
while taking into account the feedback of an algorithm for the formulation of
a mathematical model, then sooner or later, they must look towards having a
repertoire of methods to improve their physical models. This is an opportunity to
recognise the developmental systems approach in the methodology for improving
physical models—the idea of improving the physical model, and the comprehensive
trouble-shooting processes for dynamical and static problems, which are of interest
to researchers. What, therefore, is this idea, and what might be the terms of its
usefulness? The basic concept in a glossary of the developmental systems approach
in the methodology to create excellent physical models is ‘change’. Mathematical
models continuously improve due to the effect of the alteration of physical models.
Lack of alteration means they are inadequate and unfit. Changes to physical models
are therefore in the world of mathematical models, and are important functions of
the control of their adequacy. This requires that the mechanisms for responding to
the physical model are understood. To be able to control the inadequacy of the
mathematical model by means of alterations of the physical model, the theorists—
researchers of mathematical model—need diagnostic skills for intrinsic operational
states (dynamical and static) or interpretation of the originated observations, based
on the resources of knowledge and the capacity to implement the relevant alterations
to the physical model. Because recognition of the developmental systems approach in
the methodology to improve physical models is a unique strategy for changing them,
it requires widespread propagation among mathematical model researchers, who
during computerised analytical studies, are working on the principles of skilled
incompetence, using a routine to create a physical model and a dullness in systems
thinking as a way to reduce a mathematical model’s inadequacy. It creates a chance
for the researchers of mathematical models to expand their knowledge of the complex
solving of dynamical and static problems. It makes it clear that the process of
creating a mathematical model is a system of ‘learning oneself’. Employing the
development systems approach in the methodology for physical model excellence is
also a departure from the scheme of the dynamical systems approach. The alteration
requires great patience because it is a repetitious process with a multi-step algorithm
(see chapter 4). This has disappointed some of the mathematical model researchers,
who want to see the immediate effects of great alteration programs on the physical
model. At the basis of the diagnosis and intervention that includes the development
systems approach in the methodology to improve the physical models lies the
mathematical model of ‘computer analytical studies in action’, based on data derived
from feedback (see figure 4.1) between the ideal (calculated in computer analytical
studies) and actual (measured in standard experimental studies) operational states
(dynamical and static), so there is therefore emptiness, there is a constant flow of
alteration data improving the physical model.
The so-called ‘external verifier’ improvement of a physical model cooperates in
a continuous manner with the so-called ‘internal verifiers’ of the mathematical

5-2
 Mechatronics

model—the originators of changes (corrections) or reporters of the problem—and

tries to help them to learn about the functional object (the interior of the object and
its surroundings) and see the possibilities for the alteration of the physical model, at
which they are less sensitive, and address the uncertainty resulting from poor
knowledge of physical phenomena, which may include the dynamical systems
approach in the methodology to create mathematical models of physical functional
objects. The idea of the recognition of the developmental systems approach in the
methodology to improve physical models is limited to changes (corrections) to the
structure and values of the parameters of a physical model. In this sense, one thinks
of alteration of the physical model as a strategy for research intervention. Verifiers
also do not forget the difficulties of increasing the physical model, related to the
synthesis of the modelled physically continuous dynamical hypersystem, because
they cause changes (corrections) in other parts of the physical model of the physical
heterogeneous continuous dynamical hypersystem and reciprocal links: the dynam-
ical hypersystem–environment. A comparison of the developmental systems
approach in the methodology to create physical models of functional objects and
the dynamical systems approach in the methodology to formulate mathematical
models of these functional objects shows the sharp differences between them, and
these are compiled in table 5.1.
This comparison shows why the dynamical systems approach in the methodology
to create the mathematical models of physical objects may not be an alternative to
the developmental systems approach, which looks for opportunities to develop the
theory of improving not only the physical model, but also a specific existing or
designed physical heterogeneous continuous dynamical hypersystem.
The above-mentioned ideas are essentially different intellectual sources and they
operate in two different planes: the mathematical model and the original. It is proper
to consider whether they can act as a mirror, showing what is important, what is
missing in this context and what the next step should be.

Table 5.1. Differences between the developmental systems and dynamical systems
approaches (Fijalkowski 1987).

Developmental systems approach Dynamical systems approach

Alteration Stability
Concretism Abstractionism
Empiricism Naturalism
Emotionalism Rationalism
Normativism Descriptivism
Inductionism Deductionism
Informalism Formalism
Behaviourism Cognitivism
Intuitionism Principlism
Disaggregation Aggregation

5-3
 Mechatronics

The existing ideas that include the development systems approach in the
methodology to improve physical models mostly come from behaviourism. When
computer-based analytical studies are performed, it may only be the behaviours of
the mathematical models of the function objects that are studied, together with the
accompanying impact on various dynamical forces, and these are considered in
direct standard experimental studies that just concern the behaviour of a researched
functional object.
The researcher is the creator of the changes to the physical model and determines
their effective implementation, and therefore they ‘get to know each other’, which
offers the oldest technique for improving physical models (e.g. simulation and
analogical modelling), in which a mathematical model is an intellectual source of
energy, which is necessary to learn how to maintain computer analytical studies on
different qualities of a mathematical model.
The dynamical systems approach in the methodology to create mathematical
models of function objects acquires significance for researchers through the
processes of perception and valuation. In the dynamical systems approach,
researchers will check the structure of the hierarchical physical model and examine
what the relation is in this selection of compulsion, using their own knowledge and
imagination. They will not think about concepts that are original (the actual
function object), but rather about a mathematical model (relationships) to better
understand how they define the dynamical systems approach, in which they operate
and select a hierarchical structure of the physical model.
The improvement of the physical model has two stages: diagnostics and
intervention. The techniques for collecting data at the diagnostic stage include
computer analysis and the study of mathematical models and the standard
experimental study behaviours of real (existing) or designed physical functional
objects in different operational states (dynamic and static). The collected data are
classified by the problems, which are solved together with practitioners,
i.e. researchers carrying out experimental studies. These diagnoses have an ongoing
character (based on the mathematical model ‘computerised analytical studies in
action’), and therefore are constantly actualised based on data derived from
feedback (see figure 4.1). The intervention stage is conceived as a development
concept and the implementation of the alteration of the physical model, which can
affect the evolution of the alternative behaviours hitherto existing, should be
effective and focused on preferring a routine for the physical modelling and
invariance. Initially, one tried to alter a physical model using simulation and
analogue modelling, but since that yielded small effects, one attempts to make
more complex alterations, especially regarding the structures of the hierarchical
physical models that condition the mathematical models. Then these hierarchical
structures begin to be examined in the broader terms of a functional environment,
and are also more generally understood in the meteorological environment.
Why the structure of the hierarchical physical models adapts to the environment,
but does not change diametrically, is also reflected on. In view of the above, the
attention of researchers shifts from the inside to the outside of the mathematical
model. The change to the physical model is the problem. Many of the improvement

5-4
 Mechatronics

techniques for physical models therefore stress the methodology for the complex
solving of problems. The overall emergent behaviour is difficult to predict, even
when the behaviour of the dynamical system, hyposystem and components can be
predicted without difficulty. Minor changes to inputs or parameters can cause large
changes in behaviour.
Attempts to prevent researchers in mathematical models from ‘swinging round
each other with their backs’ led them to stop disregarding the physical model, not as
an exercise in useless conformity, but to resolve issues based on knowledge of
logical, dynamic or static processes, and knowledge of the principles of cooperation
of a variety of physical heterogeneous continuous dynamical hypersystems. The idea
of using the developmental systems approach in the methodology to improve the
physical models of physical functional objects was built under the influence of an
article by Maslyk (1985): ‘Change: the recipe for progress’.

References
Fijalkowski B 1987 Modele matematyczne wybranych lotniczych i motoryzacyjnych mechano-
elektro-termicznych dyskretnych nadsystemów dynamicznych (Mathematical Models of
Selected Aviation and Automotive Continuous Dynamical Hyperystems), Monografia 53
Politechnika Krakowska imienia Tadeusza Kościuszki, Krakow, 276 p. (In Polish)
Maslyk E 1985 Zmiany: recepta na postep. Przeglad techniczny, 1985, No. 6, pp 20–21 (In Polish)

5-5
 IOP Publishing

Mechatronics
Dynamical systems approach and theory of holors
B T Fijalkowski

Chapter 6
Synthetic mathematical model of the abstract
functional heterogeneous continuous dynamical
hypersystem

MATHEMATICAL SUPERIORITY

Mechanics is the paradise of mathematical sciences [&] Mechanical science is

most noble and useful beyond all others, since by means of it, all animated bodies
that have motion perform all their operations.
Leonardo DaVinci

An intelligence knowing all the forces acting in nature at a given instant, as well
as the momentary positions of all things in the Universe, would be able to
comprehend in one single formula of the motions of the largest bodies as well as
of the lightest atoms in the world, provided that its intellect were sufficiently
powerful to subject all data to analysis; to it nothing would be uncertain, the
future as well as the past would be present to its eyes.
Pierre Simon Laplace
Mathematics is the most beautiful and most powerful creation of the human
spirit.
Stefan Banach

Mathematics is the art of constructing theories that aid in creating an image of

reality.
Stanislaw Lojasiewicz

doi:10.1088/978-0-7503-1350-6ch6 6-1 ª IOP Publishing Ltd 2016

 Mechatronics

6.1 Abstract functional heterogeneous continuous dynamical

hypersystem
A synthetic mathematical model of an abstract functional heterogeneous continuous
dynamical hypersystem based on the Hamilton–Ostrogradsky variational principle
of stationary action initiates an algorithm to formulate a real mathematical model of
a physical hypersystem in the dynamical systems approach. This principle may be
found in the work of the Irish mathematician, William Rowan Hamilton (1805–65),
as published 1830–35. Hamilton assumed that the output dynamical system is
‘scleronomic’, i.e. it is only subject to stationary constraints, in which the equation of
time t does not appear explicitly. In the general case of non-stationary constraints,
the principle was formulated and substantiated by the Russian mathematician,
Mikhail Vasilevich Ostrogradsky (1801–62), in 1845. In relation to the Hamilton
principle, it is sometimes called the ‘Hamilton–Ostrogradsky principle’. In the 1760s
a Swiss mathematician, Leonhard Euler (1707–83), developed a way of laying out
the differential equations of dynamics based on the methods of variational calculus
(Euler 1766), while in the 1780s the French mathematician Joseph Louis de
Lagrange (1736–1813) developed a stacking of differential equations based on the
energetic methods for mechanical homogenous continuous dynamical systems
(Lagrange 1788), and the differential equations of both Euler and Lagrange are
analogous as they may be applied in the case of a general dynamical hypersystem of
many degrees of freedom (DoF). In the 1870s the British physicist, James Clark
Maxwell (1831–79), developed a way of stacking the differential equations of
dynamics that was based on Lagrange’s energetic methods for electromechanical
continuous dynamical hyposystems that depended on supplementing the Lagrange
potential with an analogue of kinetic and potential energy, as well as on finding an
analogue of the generalised force (Carvallo 1902, Liennard 1902, Maxwell 1873).
By including in the textbook such fundamental concepts of dynamics for the
methodology to create a mathematical model of an abstract functional heteroge-
neous continuous dynamical hypersystem—like the abstract functional continuous
dynamical component (figure 3.1)—instead of using the concept of mechanical
motion applied by Lagrange, as a material point, for mechanical homogeneous
continuous dynamical systems, below we will leave the Euler–Lagrange second-
order differential equation of dynamics in the dynamical systems approach and
hypermatrix notation. The dynamical systems approach and holor hypermatrix
notation take into account the work done by all the generalised forces acting on the
abstract functional heterogeneous continuous dynamical hypersystem, regardless of
whether or not they are conservative generalised forces.
In addition, the dynamical systems approach and hypermatrix notation have
difficulty in distributing the physical variables by applying partial differentiation.
In many cases, the structure of the nonlinear abstract functional heterogeneous
continuous dynamical hypersystem denies its dynamical components the ability to
change their dynamical states freely. A number of conditions and restrictions can
hamper their dynamical states. It is said then that constraints are imposed on the
dynamics of the dynamical hypersystem. The specific form of these constraints can

6-2
 Mechatronics

be very varied, for example, through the combination of two different continuous
dynamical hyposystems of direct current (DC) or alternating current (AC) in the
commutator electrical machine by means of an arbitrary physical commutation
matrixer (e.g. a mechanical or electrical commutation matrixer)—see chapter 11. If
the constraints impose restrictions on the dynamical status of respective parts of the
nonlinear abstract functional continuous dynamical hypersystem, they are called
‘geometrical constraints’. Moreover, if they impose restrictions on the dynamics of
the dynamical hypersystem, i.e. on the values of the generalised velocities of the
respective parts of the dynamical hypersystem, then they are called ‘kinematical
constraints’. However, an inverse affirmation may not be true, i.e. the relationships
between the generalised velocities of different respective parts of the dynamical
hypersystem, and in particular the limit values of the generalised velocities, may
not place restrictions on the dynamical status of the dynamical hypersystem.
A convincing example of this is the arbitrary physical commutation matrixer.
Relatively recently, it was noted that the existence of constraints does not impose
any restrictions on the dynamical state of the nonlinear abstract functional
continuous dynamical hypersystem. In his Mécanique Analytique (1788) Lagrange
did not believe that such constraints could exist, noting that for any mechanical
homogeneous continuous dynamical system using ordinary differential equations
resulting from the structure of the dynamical system, one may select independent
coordinates with independent variations. This oversight became known after many
years (Appell 1899a, 1899b, 1893, 1899d, 1903, Bobylev 1892, Beltrami 1895,
Tchaplygin 1897a, 1897b, Zhukovsky 1893) when considering rolling solid bodies
without slipping after plane or complex surfaces. In 1894, the German physicist,
Heinrich Hertz (1857–94), introduced the division of mechanical homogeneous
continuous dynamical systems into holonomic and unholonomic ones (Hertz 1894).
A nonlinear abstract functional heterogeneous continuous dynamical hypersystem
with unholonomic constraints and no integral that does not reduce to the geo-
metrical constraints is called an ‘unholonomic physical heterogeneous continuous
complex or simple dynamical hypersystem’.

6.2 Independent generalised coordinates

One of the basic concepts of the dynamics of nonlinear abstract functional
heterogeneous continuous dynamical hypersystems is the concept of the elementary
‘abstract functional homogeneous continuous dynamical system’. This means the
dynamical component elementary abstract homogeneous continuous dynamical
system whose physical, economic, social, etc structure one can omit when creating
a synthetic mathematical model of its dynamics. It is understood that the possibility
of such an omission is dependent on the existing conditions of that or another
problem. For instance, a planet can be seen as elementary astronomical homoge-
neous continuous dynamical systems of the sunny homogeneous continuous
dynamical hypersystem in the investigation of their movement around the Sun,
but one cannot be seen in the investigation of their day and night movement.
The dynamical status of elementary abstract functional homogeneous continuous

6-3
 Mechatronics

dynamical systems of a nonlinear abstract functional heterogeneous continuous

dynamical hypersystem—that is, their position in the generalised multi-dimensional
spaces—is determined by a hypermatrix of the independent generalised coordinates

q = ⎡⎣ q1 q2 q3 … qϑ ⎤⎦ .
T T T T T

A derivative of the hypermatrix q over time

dq
q̇ = =v (6.1)
dt
is a hypermatrix of the generalised velocities, and the second derivative
d2q dq ̇
q̈ = = =a (6.2)
dt dt
is a hypermatrix of the generalised accelerations of the nonlinear functional
heterogeneous continuous dynamical hypersystem.
To determine the dynamical status of the nonlinear functional heterogeneous
continuous dynamical hypersystem consisting of N elementary functional homoge-
neous continuous dynamical systems in multi-dimensional spaces, we should
introduce N matrices of the independent generalised coordinates. The total number
of matrices of the independent generalised coordinates, which is necessary to
introduce the unique identification of the nonlinear functional continuous dynamical
hypersystem, is simply the number of DoF. The hypomatrices of these matrices
are independent generalised coordinates of elementary functional homogeneous
continuous dynamical systems of the nonlinear functional heterogeneous continuous
dynamical hypersystem. The mathematical relationship relating the hypermatrix of
generalised accelerations with the hypermatrix of the independent generalised
coordinates and the hypermatrix of the generalised velocities is called the ‘synthetic
mathematical model’ or the ‘differential equation of dynamics’ of the nonlinear
functional heterogeneous continuous dynamical hypersystem. In relation to the
hypermatrix, q , it is the second-order differential equation in the hypermatrix
notation, and their successive integration allows us, in principle, to determine these
hypermatrices, i.e. the trajectories of the dynamics of the nonlinear functional
heterogeneous continuous dynamical hypersystem.

6.3 Unconstrained and constrained nonlinear functional

heterogeneous continuous dynamical hypersystems—
classification of the constraints
The subject covered in this textbook on dynamical analytical studies is the dynamics
of the nonlinear abstract functional heterogeneous continuous dynamical hyper-
system (consisting of elementary functional homogeneous continuous dynamical
systems of this dynamical hypersystem Sυ (υ = 1, 2, 3, …, N) relative to a set of
matrices of the independent generalised coordinates. On the matrices of the
independent generalised coordinates (the generalised positions) and the matrices

6-4
 Mechatronics

of the generalised velocities of these elementary homogeneous continuous func-

tional dynamical systems are imposed some restrictions called ‘constraints’. A
nonlinear functional heterogeneous continuous dynamical hypersystem with such
constraints is called an ‘unconstrained’ one as distinct from ‘constrained’ dynamical
hypersystems, which are not subject to any constraints. The constraints can be
expressed in hypermatrix notation by means of an equation1
F(q , q ,̇ t ) = 0, (6.3)
where after the left side there is a time t and a hypermatrix of the independent
generalised coordinates q of all the elementary nonlinear functional homogeneous
continuous dynamical systems, Sυ, of the nonlinear functional heterogeneous
continuous dynamical hypersystem (υ = 1, 2, 3, …, N).
In a particular case, when the hypermatrix of the generalised velocities q ̇ is not
present in the equation of constraints (6.3), the constraints are called ‘finite’ ones.
Analytically, they may be represented with the hypermatrix notation as follows:
F(q , t ) = 0. (6.4)
In the general case, the constraints (6.3) are called ‘differential’ ones. In the case
where there are finite constraints of the form (6.4), the nonlinear functional
heterogeneous continuous dynamical hypersystem may not find itself in an arbitrary
dynamical position, or may take an arbitrary position in multi-dimensional space at
a given moment t. However, in a case where there are only differential constraints,
the dynamical hypersystem may at a given moment t find itself in an arbitrary
dynamical state (position) in multi-dimensional space. However, in these dynamical
states, the matrices of the generalised velocities of the elementary functional
homogeneous continuous dynamical systems may not be arbitrary, because the
differential constraints impose the restrictions of the generalised velocities on these
matrices. The finite constraints (6.4) are called ‘stationary’ in the case where in their
equation a time t does not appear explicitly. A nonlinear functional heterogeneous
continuous dynamical hypersystem that consists of elementary functional homogeneous
continuous dynamical systems of this dynamical hypersystem is called a ‘holonomic
dynamical hypersystem’ if the elementary functional heterogeneous continuous
dynamical systems are not subject to any integral differential constraints.
For example, the holonomic dynamical hypersystem is every one of the uncon-
strained nonlinear functional heterogeneous continuous dynamical hypersystems
consisting of elementary functional homogeneous continuous dynamical systems, as
well as the constrained nonlinear functional heterogeneous continuous dynamical
hypersystem with ‘finite’ and ‘differential’ constraints, but no integral. In the case
of the holonomic nonlinear functional heterogeneous continuous dynamical
hypersystem, all constraints can be represented in the form of the finite one.
In the case where a nonlinear functional continuous dynamical hypersystem is not

1
( )
The next F(q, q,̇ t ) = 0 is an abbreviated function designation F [ q1 T … qϑ T ]T , [ q1̇ T … qϑ̇ T ]T . An
abbreviation of this kind is often used in this book. It is assumed that the functions found in this book are
continuous with their derivatives, which are found in the relevant chapters.

6-5
 Mechatronics

subject to any integral differential constraints, it is called an ‘unholonomic’

dynamical hypersystem. While those with no integral differential constraints are
the so-called unholonomic ones, those with integral differential constraints
are called ‘semi-holonomic’. A nonlinear functional heterogeneous continuous
dynamical hypersystem is called a ‘scleronomic’ one if it is only subject to stationary
constraints. If not, the dynamical hypersystem is called a ‘rheonomic’ one.
Nonlinear functional heterogeneous continuous dynamical hypersystems, in which
processing takes place in energy generation, conversion, dissipation, or energy
storage, divide into:
• the conservative, in which the work of the external generalised forces changes
without loss into kinetic or potential energy;
• the dissipative, in which the work of the generalised external forces dissipates
into thermal energy.

6.4 Principle of stationary action

The most general law of the dynamics of the conservative nonlinear functional
heterogeneous continuous dynamical hypersystem in this textbook is based on the
variational principle of stationary action, which originates from the Hamilton–
Ostrogradsky principle of least action as defined for conservative dynamical hyper-
systems after its generalisation to dissipative dynamical hypersystems. In accordance
with this principle, each of the nonlinear abstract functional heterogeneous
continuous dynamical hypersystems is characterised by some kinetic potential,
that is, a function called ‘Lagrange’s energy function’ (or the ‘Lagrangian’), which
is equal to the surplus of the conservative kinetic co-energy T⁎ over the potential
energy U of each of the nonlinear abstract functional heterogeneous continuous
dynamical hypersystems, and has been derived, for example, by White and
Woodson (1959), in the conventional (Newtonian paradigm) systems approach
and scalar notation, but in the dynamical systems approach and hypermatrix
notation it is defined as:
⁎
L(q , q ,̇ t ) ≜ T (q , q ,̇ t ) − U(q , q ,̇ t ), (6.5a )

or simply
⁎
L ≜ T − U. (6.5b)
Moreover, a function called ‘virtual’ or ‘prepared’ work, which is equal to the
⁎
surplus of the dissipative kinetic co-energy (dissipation) TR over the dissipative
potential energy U0 of each of the nonlinear abstract functional heterogeneous
continuous dynamical hypersystems, is defined in the dynamical systems approach
and hypermatrix notation as:
⁎
A(q , q ,̇ t ) ≜ TR(q , q ,̇ t ) − UF (q , q ,̇ t ), (6.6a )

or simply
⁎
A ≜ TR − UF . (6.6b)

6-6
 Mechatronics

The variables q, q,̇ t through which the Lagrange energy function and virtual
work are expressed are called the ‘Lagrange variables’. A set of the values of these
variables is characterised by a moment of time and the correspondence to them of
the dynamical status of each of the nonlinear abstract functional heterogeneous
continuous dynamical hypersystems, i.e. the generalised positions and generalised
velocities of the elementary abstract functional heterogeneous continuous complex
or simple dynamical systems.
The conservative kinetic co-energy of any of the nonlinear abstract functional
heterogeneous continuous dynamical hypersystems is defined in the dynamical
systems approach and the hypermatrix notation as:
q̇ q̇
⁎
T≜ ∫0 ̇ q̇≜
pT (q , q )d ∫0 pT (q , v)dv , (6.7a )

where p(q, q )̇ = p(q, v ) is the hypermatrix of conservative generalised momenta and

dp dp
K (q, q )̇ = K (q, v ) = dq ̇ = dv is the hypermatrix of the conservative kinetic differ-
ential coefficients (generalised inertia).
The conservative kinetic co-energy of the arbitrary linear abstract functional
heterogeneous continuous dynamical hypersystem in the dynamical systems
approach and hypermatrix notation is in a homogeneous square form, namely:
⁎ 1 1
T≜ K (q , q )̇ ≜ K (q , v). (6.7b)
2 2
The conservative potential energy of the nonlinear abstract functional heteroge-
neous continuous dynamical hypersystem in the dynamical systems approach and
hypermatrix notation is defined as:
q
U≜ ∫0 f T dq ,̇ (6.8a )

where f is the hypermatrix of conservative generalised forces and P (q, q )̇ = P (q, v ) =

df
is the hypermatrix of the conservative potential differential coefficients (general-
dq
ised stiffnesses).
The dissipative potential energy of any linear functional heterogeneous contin-
uous dynamical hypersystem in the dynamical systems approach and hypermatrix
notation is of a homogeneous square form, namely:
1 T 1
U≜ q P(q , q )̇ q ≜ qT P(q , v)q . (6.8b)
2 2
The conservative kinetic co-energy of any of the nonlinear abstract functional
heterogeneous continuous dynamical hypersystems in the dynamical systems
approach and hypermatrix notation is defined as:
t
⁎
TR ≜ ∫0 Rdt , (6.9)

6-7
 Mechatronics

where the integral of the Rayleigh function, i.e. the function of the thermal losses
R ≜ RV + RC , (6.10)

is a sum of the generalised losses on the viscous friction

q̇ v̇
RV ≜ ∫0 pR (q )d
̇ q̇ ≜ ∫0 pR (v)dv (6.11)

and the generalised losses on the dry friction

q̇ v
RC ≜ ∫0 qṪ (Δf sgn q )d
̇ q̇ ≜ ∫0 vT (Δf sgn v)dv, (6.12)

where pR (q )̇ = pR (v ) is the hypermatrix of the dissipative generalised momenta

dpR dpR
(dissipation), D(q )̇ = D(v ) = dq ̇
= dv
is the hypermatrix of the dissipative kinetic
differential coefficients of the thermal losses caused by the viscous friction and Δf is
the hypermatrix of the dissipative generalised force drops of the thermal losses
caused by the dry friction.
The Rayleigh dissipation function of any of the linear abstract functional
heterogeneous continuous dynamical hypersystems is the sum of the generalised
losses on the viscous friction, establishing a homogeneous square form
1 T 1
RV ≜ q ̇ D(q )̇ q ̇ ≜ vT D(v)v , (6.13)
2 2
and the generalised losses on the dry friction
1 1
RC ≜ (Δf sgn q )̇ q ̇ ≜ (Δf sgn v)v , (6.14)
2 2
T T T
where sgn q ̇ = [sgn q1̇ … sgn qϑ̇ and
]

sgn q ̇ ≜ sgn v ≜ { + 1 v >0;

− 1v <0.
The dissipative potential energy of any of the nonlinear abstract functional
heterogeneous continuous dynamical hypersystems in the dynamical systems
approach and hypermatrix notation is assumed as:
q t t
UF ≜ ∫0 f (t )T dq = ∫0 f (t )T q d
̇ t= ∫0 f (t )T v dt , (6.15)

where f (t ) is the hypermatrix of the dissipative generalised forces (dissipation) as

a function of time.
If a nonlinear abstract functional heterogeneous continuous dynamical hyper-
system transmits from the dynamical state described by a hypermatrix of the
generalised coordinates
q = q (1)

6-8
 Mechatronics

occurring in the period t = t1 to the dynamical state described by a hypermatrix of

the generalised coordinates
q = q (2)
occurring in the period t = t2, then this transition is done in such a way that the
integral
t2
J≜ ∫t 1
[L(q , q ,̇ t ) + A(q , q ,̇ t )]dt = minimum. (6.16)

Let q = q(t) be this hypermatrix of generalised coordinates for which the functional
J has the smallest value. This means that J rose when changing the hypermatrix
q(t) on any hypermatrix of the form
q(t ) + δq(t ), (6.17)
where δ q(t) is the hypermatrix, in which the matrices and hypomatrices are the
elements of the hypermatrix q(t), and are small functions in the entire time period
from t1 to t2 (called the ‘variation’ of the hypermatrix q(t)). Because at t = t1 and
t = t2 the matrices, hypomatrices and elements of the hypermatrix (6.17) should
assume one and the same values of the matrices, hypomatrices and components of
the hypermatrix q (1) and q (2), then it should be the case that
δq(t1) = 0 (6.18a )
and
δq(t2 ) = 0. (6.18b)
A variation I at a permutation q into q(t) + δ q(t) may be obtained by the differences
t2 t2
∫t 1
L(q + δq , q ̇ + δq ,̇ t )dt − ∫t 1
L(q , q ,̇ t )dt

and
t2 t2
∫t1
A(q + δq , q ̇ + δq ,̇ t )dt − ∫t 1
A(q , q ,̇ t )dt .

A distribution of these differences according to δq and δq ̇ (in the integral

expressions) begins from the first-order terms.
A necessary condition for the integral minimisation of J is the transformation of
all its terms to zero; this is called a ‘first variation’ (or simply a variation of the
integral). In this way, in unison with a symbol of the variational calculus and matrix
computation, the Hamilton–Ostrogradsky principle of the stationary action for any
nonlinear abstract functional heterogeneous continuous dynamical hypersystem can
be written in the form:
t2
J≜δ ∫t1
[L(q , q ,̇ t ) + A(q , q ,̇ t )]dt (6.19)

6-9
 Mechatronics

or as a variation
t2 ⎛ ∂L ∂L ∂A ∂A ⎞
∫t
1
⎜
⎝ ∂q
δq +
∂q ̇
δq ̇ +
∂q
δq + δq⎟̇ dt = 0.
∂q ̇ ⎠
d
Taking into consideration the fact that δq ̇ = dt
δq and integrating the second and
fourth terms by parts, one obtains

∂L t2 t1 ⎛ ∂Ld ∂L ⎞
δJ =
∂q
δq +
t1 ∫t
⎜
1 ⎝ ∂q
δq + ⎟δq dt
dt ∂q ̇ ⎠
(6.20)
∂A t2 t1 ⎛ ∂A d ∂A ⎞
+
∂q
δq +
t1 ∫ ⎜
t1 ⎝ ∂q
δq + ⎟δq dt = 0.
dt ∂q ̇ ⎠

After one considers conditions (6.18), the first and third terms in the above
expression disappear. Only the integrals remain, which should be equal to zero
for any of the values of the matrices, hypomatrices and elements of the hypermatrix
δq (t). This is only possible in this case when the under-integral expression will
identically be zero. In this way, we obtain the synthetic mathematical model of any
nonlinear abstract functional heterogeneous continuous dynamical hypersystem,
i.e. a differential equation of dynamics in the dynamical systems approach and
hypermatrix notation,
⎡ d ⎛ ∂L ⎞ ∂L d ⎛ ∂A ⎞ ∂A ⎤
⎢ ⎜ ⎟− + ⎜ ⎟− ⎥δq = 0. (6.21a )
⎣ dt ⎝ ∂q ̇ ⎠ ∂q dt ⎝ ∂q ̇ ⎠ ∂q ⎦
A differential equation of dynamics (6.21a) with hypermatrix notation in the
dynamical systems approach in the methodology for the formation of mathematical
models of abstract functional heterogeneous continuous dynamical hypersystems is
unknown in the literature.
The author decided to apply this hypermatrix notation because it is comfortable
for the dynamical systems approach in the methodology for the formation of
mathematical models. However, this differential equation in the non-matrix, i.e.
scalar notation, is well known and in variational calculus it is called ‘Euler’s
differential equation’ (1766), while in analytical mechanics it is called ‘Lagrange’s
differential equation’ (1788), and it has a somewhat different form (White and
Woodson 1959, Szklarski et al 1963):
d ⎛ ∂L ⎞ ∂L
⎜ ⎟− + fD = fE , (6.21b)
dt ⎝ ∂qi̇ ⎠ ∂qi
where qi , qi̇ are the ith generalised coordinate and velocity, i is the number of DoF,
fD = ∂R is the generalised dissipation forces, R is the Rayleigh function and fE = fE(t)
∂qi̇
is the generalised external forces.

6-10
 Mechatronics

If the Lagrange function of the energy and virtual work of an arbitrary nonlinear
abstract functional heterogeneous continuous dynamical hypersystem are known,
then the Euler–Lagrange second-order differential equation of dynamics in the
hypermatrix notation establishes a relationship between the relevant matrices of the
hypermatrices of generalised accelerations, velocities and coordinates, that is, it
represents a synthetic differential equation of the dynamics, and therefore a synthetic
mathematical model of any nonlinear abstract functional heterogeneous continuous
dynamical hypersystem.
Substituting expressions (6.5) and (6.6) into the equation of dynamics (6.21a), one
obtains a synthetic mathematical model of any nonlinear abstract functional
heterogeneous continuous dynamical hypersystem that is more appropriate for
computerised analytical studies of the dynamics, namely:
⎡ d ⎛ ∂T⁎ ⎞ ∂T⁎ ∂U d ⎛ ∂TR ⎞
⁎ ⎤
⎢ ⎜ ⎟− + + ⎜ ⎟ − f T ⎥ δq = 0, (6.22a )
⎣ dt ⎝ ∂q ̇ ⎠ ∂q ∂q dt ⎝ ∂q ̇ ⎠ ⎦
or
⎡ d ⎛ ∂T⁎ ⎞ ∂T⁎ ∂U
⁎
d ⎛ ∂TR ⎞ ⎤
⎢ ⎜ ⎟ − + + ⎜ ⎟ − f T ⎥ δq = 0, (6.22b)
⎣ dt ⎝ ∂v ⎠ ∂q ∂q dt ⎝ ∂v ⎠ ⎦
where
∂A ∂UF
f T = (grad A)T = = (grad UF )T =
∂q ∂q
is the hypermatrix of the generalised dissipative input (excitation) forces.
In the above Euler–Lagrange second-order differential equation of the dynam-
ics, one can use hypermatrix notation for the generalised coordinates and velocities
⁎
and the scalar functions of the conservative kinetic co-energy T and conservative
potential energy U as homogeneous square forms, where the partial derivatives in
relation to the hypermatrices of the generalised coordinates and velocities form
matrices. The reality of the subsistence of respective matrices of these hypoma-
trices must be verified in each case. Verification of this will ensure the completeness
of the matrices and hypomatrices acting significantly in the abstract functional
heterogeneous continuous dynamical hypersystem. This is the advantage of an
accepted hypermatrix notation for differential equations of the dynamics of any
nonlinear abstract functional heterogeneous continuous dynamical hypersystem
and consistent computation in this notation.
A synthetic mathematical model of any nonlinear abstract functional heteroge-
neous continuous dynamical hypersystem, establishing the Euler–Lagrange second-
order differential equation of dynamics in the dynamical systems approach, is
written in the hypermatrix form, since the abstract functional heterogeneous
continuous dynamical hypersystem is often a set of many functional homogeneous
continuous dynamical systems, whose own mathematical models, establishing a set
of the Euler–Lagrange differential equations, are written in the matrix form, since
each one is often a set of many functional heterogeneous continuous dynamical
hyposystems whose mathematical models, forming a subset of the Euler–Lagrange

6-11
 Mechatronics

Figure 6.1. A hierarchical structure of a physical model of an abstract functional continuous dynamical
hypersystem.

second-order differential equations in the dynamical systems approach, are written

in the hypomatrix form, since they are usually a few or more elements of a
dynamical hyposystem, which elementary mathematical models, constituting only
a single Euler–Lagrange second-order differential equation, do not write in the
matrix form of a matrix, but rather in a scalar one (figure 6.1).
Taking into consideration equations (6.7)–(6.15), the Euler–Lagrange second-order
differential equations of dynamics (6.22) in the dynamical systems approach can be
written in the hypermatrix form as follows:
⎡d 1 ∂K (q , v)
⎢ [vT K (q , v)] − vT v
⎣ dt 2 ∂q
(6.23a )
⎤
+ q P(q , v) + v D(q , v) + sgn v Δf − f ⎥ δq = 0,
T T T T
⎦

or simply
⎡d 1 ∂K ⎤
⎢ (vT K ) − vT v + qT P + vT D + sgn vT Δf − f T ⎥ δq = 0. (6.23b)
⎣ dt 2 ∂q ⎦

Assuming that K , P and D are not also the functions of q as q ̇ = v , the Euler–
Lagrange second-order differential equations of dynamics (6.23) in the dynamical
systems approach can be written in the hypermatrix form as:
⎡d T T T T T
⎤
⎢⎣ v K + q P + v D + sgn v Δf − f ⎥⎦ δq = 0, (6.24a )
dt

6-12
 Mechatronics

or
⎡d T t ⎤
⎢⎣ v K +
dt
∫0 vT dtP + vT D + sgn vT Δf − f T ⎥ δq = 0,
⎦
(6.24b)

⎡ T T P T T T
⎤
⎢⎣v (s )sK + v (s ) + v (s )D + sgn v (s )Δf (s ) − f (s )⎥⎦ δq(s ) = 0, (6.25a )
s
or
⎡ ⎛ P ⎞ ⎤
⎢vT (s )⎜sK + + D⎟ + sgn vT (s )Δf (s ) − f T (s )⎥ δq(s ) = 0, (6.25b)
⎣ ⎝ s ⎠ ⎦

where s = σ + jω is the Laplace operator.

After taking into account the notion of the impedance Z (s ) and/or the admittance
Y (s ), this yields:
P
Z (s ) = D + sK + (6.26a )
s
and
1 1 1 s B
Y (s ) = = + + =G+ + sS . (6.26b)
Z (s ) D sK P s
The Euler–Lagrange second-order differential equations of dynamics (6.25) in the
dynamical systems approach can be written in the hypermatrix form as follows:
[vT (s )Z (s ) + sgn vT (s )Δf (s ) − f T (s )] δq(s ) = 0, (6.27a )

or
⎡ 1 ⎤
⎢vT (s ) + sgn vT (s )Δf (s ) − f T (s )⎥ δq(s ) = 0. (6.27b)
⎣ Y (s ) ⎦

The mathematical representation of any measurable physical heterogeneous

continuous dynamical hypersystem is greatly facilitated by representing it by a
holor2 that fits the physical concept perfectly.
The term holor is a generalisation of the more familiar mathematical concepts of
tensors, vectors, matrices, complex numbers and scalars. Rather than forcing all
measurable physical heterogeneous continuous dynamical hypersystems to be
imperfectly represented by a few families of mathematical quantities, the study of
holors permits the development of a mathematical representation that fits the

2
Definition from Wikipedia: ‘A holor (/ˈhoʊlər’/; Greek ὅλος “whole”) is a mathematical entity that is made up of one or
more independent quantities (“merates” as they are called in the theory of holors). Complex numbers, scalars, vectors,
matrices, tensors, quaternions, and other hypercomplex numbers are kinds of holors. If proper index conventions are
maintained then certain relations of holor algebra are consistent with that of real algebra; i.e. addition and
uncontracted multiplication are both commutative and associative.’ [Link]

6-13
 Mechatronics

physical concept perfectly. The set of numbers that specify a holor are termed
‘merates’.
The physical quantity holor ought to satisfy two fundamental criteria. The holor
must be defined in terms of physically realisable measurements that are significant at
physical frequencies. The holor formulation is very simple and modern computer
techniques make the design of physical heterogeneous continuous dynamical hyper-
systems in terms of the generalised external forces holor (termed by the author the
‘energy-potential-difference holor’, Fi ) and the generalised velocities holor (termed
by the author the ‘energy-flow holor’, V j ) in two-dimensional (2D) spaces, and the
generalised power holor, P κ , in a three-dimensional (3D) spaces, a challenge of only
moderate difficulty.
The mathematical model, represented by the Euler–Lagrange second-order
differential equations of dynamics (6.27) for any physical heterogeneous continuous
dynamical hypersystem with energy storage, may originate from holor equations.
For that reason, the resolution of physical heterogeneous continuous dynamical
hypersystems for a step-function or sine-wave input-signal holor Fi necessitates the
solution of a set of holor equations:
⎡⎣V kZjk + sgn V k ΔFjk − Fk ⎤⎦ δQ j = 0, (6.28a )

or
⎡ δ ki ⎤
⎢V k ij + sgn V k ΔFjk − Fk ⎥ δQ j = 0. (6.28b)
⎣ Y ⎦

This appears for the physical and mathematical approaches presented in chapters 7
and 8. One may consider if it is possible to model a physical heterogeneous
continuous dynamical hypersystem by algebraic equations from which the holor
equations can be created. To deal with this problem one must take the result of the
generalised input-signal holor V j function into account. This nature of the driving
holor function in its normal mathematical appearance is not conventionally realised
by a physical heterogeneous continuous dynamical hypersystem.
In spite of this, this generalised input signal-holor function has a particular form
(sine waveform) that symbolises the conventional generalised driving holor function
applied in studies.
After taking into account the notion of the impedance holor, Zij , and/or the
admittance holor, Y ji , it yields
⎡ Z − Zcs ⎤ ⎡ D − X ⎤
Zij = ⎢ cc ⎥= (6.29a )
⎣ Zsc Zss ⎦ ⎢⎣ X D ⎥⎦

and
δ ki ⎡ ss sc ⎤ ⎡ ⎤ 1 ⎡ D X⎤
Y ij = = ⎢ Y cs Y cc ⎥ = ⎢ G B ⎥ = , (6.29b)
Zjk ⎣−Y Y ⎦ ⎣− B G ⎦ (D ^2) +(X ^2) ⎣⎢−X D ⎦⎥

6-14
 Mechatronics

where the reactance, X , and the susceptance, B, are as follows:

1
X = ωK − P, (6.30a )
ω
1 1 1
B=ω − = ωS − H , (6.30b)
P ωK ω
and
Y ijZjk = δ ki . (6.30c )

The Euler–Lagrange second-order differential equations of dynamics (6.28) in the

dynamical systems approach can be written in the hypermatrix holor form as
follows:
⎡⎣V kZjk + sgn V k ΔFjk − Fk ⎤⎦δQ j
⎧ ⎡ ⎤ ⎡ ΔF 0 ⎤ ⎫ ⎡Q c ⎤ (6.31a )
= ⎨[V c V s ] ⎢ D −X ⎥ + sgn[V c V s ] ⎢ c ⎥ − [ Fc Fs ]⎬ δ⎢ s ⎥ ,
⎩ ⎣X D ⎦ ⎣ 0 ΔFs ⎦ ⎭ ⎣Q ⎦

or
⎡ δ ki ⎤
⎢V k ij + sgn V k ΔFjk − Fk ⎥δQ j
⎣ Y ⎦
⎧ 1 ⎡G −B ⎤
= ⎨[V c V s ] (6.31b)
⎩ (G ^2) + (B ^2) ⎢⎣ B G ⎥⎦
⎡ ΔF 0 ⎤ ⎫ ⎡Q c ⎤
+ sgn[V c V s ] ⎢ c ⎥ − [ Fc Fs ]⎬ δ ⎢ s ⎥ ,
⎣ 0 ΔFs ⎦ ⎭ ⎣Q ⎦

where
⎡ F ⎤ ⎡ F cos ψF ⎤
Fi = (Fc , Fs ) = [Fc Fs ]T = ⎢ c ⎥ = ⎢ ⎥ = F ∠ψF , (6.32a )
⎣ Fs ⎦ ⎣ F sin ψF ⎦

⎡ c ⎤ ⎡V cos ψV ⎤
V j = (V c , V s ) = [V c V s ]T = ⎢V s ⎥ = ⎢ ⎥ = V ∠ψV , (6.32b)
⎣V ⎦ ⎣V sin ψV ⎦

and
F c = F cos ψF , F s = F sin ψF (6.32c )

V c = V cos ψV , V s = V sin ψV (6.32d )

⎡ ΔF 0 ⎤
ΔFij = ⎢ c ⎥. (6.32e )
⎣ 0 ΔFs ⎦

6-15
 Mechatronics

References
Appell P 1893 Traité de Mécanique Rationnelle 4 vols (Paris: Gauthier-Villars) (In French)
Appell P 1899a Les mouvements de roulement en dynamique. Scientia, No. 4, Paris (In French)
Appell P 1899b Sur les mouvements de roulement, equations du movement [Link] es &
celles de Lagrange. C.R., T. 129, Paris (In French)
Appell P 1899d Sur une forme gènèrale des èquations de la Dynamique J. Reine Angew. Math. Bd.
121, Berlin (In French)
Appell P 1903 Remarques sur les systems non holonomes J. Math. Pures Appl. T. 9, pp 27–28 (In
French)
Beltrami E 1895 Sulle equazioni diminiche di Lagrange. Rendiconti dell’Istituto Lombardo, Ser. II,
V. XXVIII, Fasc. XIV, pp. 745–52 (In Italian)
Bobylev D K 1892 O share s giroskopom vnutriy, katyashchemsya po gorizontalnoi ploskosti bez
proskalzyvaniya (On a sphere with a gyroscope inside it rolling after horizontal plane without
slipping). Mat, Sb., T. 16, Vyp. 3, (In Russian)
Carvallo E 1902 L’Electricité deduite d’experience et remenée an pinxipe des travaux virtuels.
Coll. Scientia, No. 19, Paris (In French)
Euler L 1766 Elementa Calculi Variatiorum (The Elements of the Calculus Variations) Originally
published in Novi Commentari Academiiae scientium Petropolitanae 10 51–93
Hertz H 1894 Die Prinzipen der Mechanik. Gesammeite Werke, T. 3, (In German)
Lagrange J L 1788 Mécanique Analytique
Lagrange J L 1788 Mécanique Analytique (Paris: Chez la Veuve DESAINT, Libraire)
Liennard A 1902 Sur l’application des equations de Lagrange aux phenomènes èlectrodynamik et
electromagnetiques. C.R., T. 134, No. 163 (In French)
Maxwell J C 1873 A Treatise of Electricity and Magnetism (Oxford: Clarendon)
Szklarski L, Pelczewski W, Kolendowski J, Puchalka T and Komarzewska M 1963 Dynamika
ukladow nieholonomicznych (Dynamics of Electromechanical Systems) Komitet Elektrotechniki
Polskiej Akademii Nauk, Panstwowe Wydawnictwo Naukowe 259 p (In Polish)
Tchaplygin 1897a O dvizheniy tiazhalogo tela vrashcheniya na gorizontalnoy ploskosti (On a
motion of the solid body on a horizontal plane). Trudy otdeleniya fizicheskih nauk obshchva
liubiteley iestestvoznaniya, T. 9, (In Russian)
Tchaplygin 1897b O nekotorom vozmozhnom obobshcheniy teoremy ploshchadey s primene-
niem k zadache o kataniy sharov (On some possible restriction of a space theory with the
application to the exercise on rolling spheres). Matem. Sb., T. 20, (In Russian)
White D C and Woodson H H 1959 Electromechanical Energy Conversion (New York: Wiley)
Zhukovsky N E 1893 O giroskopicheskom share D K Bobyleva (On a gyroscopic sphere of D K
Bobylev). Trudy ob-va lyubitieley iestiestvoznaniya, antropologii i etnografii. T.6, Vyp. 1, (In
Russian)

6-16
 Part II
Towards a unified theory exploiting the theory
of holors for physical heterogeneous
continuous dynamical hypersystems
 IOP Publishing

Mechatronics
Dynamical systems approach and theory of holors
B T Fijalkowski

Chapter 7
Physical heterogeneous continuous dynamical
hypersystems as generalised physical
commutation matrixers

In most sciences one generation tears down what another has built, and what one
has established another undoes. In mathematics alone each generation builds a
new story for the old structure.
Hermann Hankel
The fascinating impressiveness of rigorous mathematical analysis, with its
atmosphere of precision and elegance, should not blind usto the defects of the
premises that condition the whole process
T C Chamberlin
The most excellent practical solutions stem from the most scrupulous theory
Rene Antoine de Reaumur

7.1 Introduction
The majority of high-level engineering activities involve the physical assembly of
physical equipment or equipment for which the term ‘physical heterogeneous
continuous complex or simple dynamical hypersystem’ is usually a convenient
definition, as it is more difficult to put together a set of physical equipment (systems
and hyposystems) or physical parts (components) that create a whole and provide a
complete function than it is to give a physical model of their applications as a
‘generalised physical commutation matrixer’. The pieces of physical equipment
(dynamical components) or the set of physical equipment (dynamical systems and
hyposystems) that come together to create a physical heterogeneous continuous
dynamical hypersystem are originated on observable facts that students habitually
learn in their earliest physics lessons.

doi:10.1088/978-0-7503-1350-6ch7 7-1 ª IOP Publishing Ltd 2016

 Mechatronics

Physical heterogeneous continuous complex or simple dynamical hypersystems have

to be modelled so that a multitude of physical domains are taken into account. This is
amplified when micro- or nanouniverse phenomena are studied. Therefore, new
approaches and paradigms are needed to simulate these strategically important areas.
Scientists and engineers will face multi-domain modelling and simulation chal-
lenges in the future as an everyday occurrence. Currently, many physical heteroge-
neous continuous dynamical hypersystems already require modelling and simulation
from a multi-domain point of view. For instance, the mechano-electrical/electro-
mechanical (ME/EM) dynamotor (generator/motor) is a heterogeneous continuous
simple dynamical hypersystem requiring both mechanical and electrical modelling.
Since it heats up, thermal modelling is also required. When EM drives are added,
the overall system also requires the modelling of electronics, etc. In such physical
heterogeneous continuous dynamical hypersystems multi-domain issues become
even more natural and evident, as noted by Senturia (1998).
Consider the liquid cooling of electronics, which is becoming a viable option in
portable computers and even cellular phones. In addition to modelling the
electronics, it is necessary to model the fluidics and the electrostatic and magnetic
effects. Radiation must also be taken into account. The task of multi-domain
modelling and simulation is formidable.
One clear problem is that although individual micro- and nano-components can be
modelled and simulated using finite element method (FEM) approaches, there will be
more and more systems on a chip. For example, ‘lab-on-a-chip’ requires not only fluid
channel modelling, but also the modelling of micropumps, microvalves and mixers.
FEM mathematical models are not suitable for system studies when the number of
dynamical components is in the hundreds or thousands, which will be the case in a
nanouniverse. Finding new ways to combine problems in multi-domain mathematical
models and simulations for individual nanosystems and nanocomponents is imperative.
The author’s intention here is to initiate the application of a dynamical systems
approach and holor theories to create firstly a physical model and secondly a
mathematical model, and then perform holor algebra analysis for generalised
physical heterogeneous continuous complex and/or simple dynamical hypersystems.
The hypothesis of a generalised physical heterogeneous continuous dynamical
hypersystem is exceptionally wide. One tries to be more precise by confining the
discussion to what this author calls generalised physical commutation matrixers. On
the other hand, the term physical commutation matrixer was created to designate
the singular nature of a physical heterogeneous continuous dynamical hypersystem
in which the generalised physical commutation matrixer’s dynamical components
correspond to only one physically homogenous continuous dynamical system.
Therefore, we will concern ourselves with physical heterogeneous forms of energy,
which fall into two main well-known states, namely kinetic energy and potential
energy, as shown in table 7.1. For instance, one has to do with electrical, mechanical,
fluidic, magnetic or thermal commutation matrixers. The plus point of the dynamical
systems approach (systems thinking) is that the modus operandi applied for the holor
analysis of electrical and magnetic commutation matrixers is very well developed
(Fijalkowski 1973, 2011).

7-2
 Table 7.1. Forms of energy.

Forms of energy
All forms of energy fall under two main states

Kinetic Potential

Kinetic energy is energy in motion. It is the motion of waves, electrons, Potential energy is stored energy and the energy
atoms, molecules and objects or substances. of position (gravitational).

Motion The movement of objects or substances from one place to Stored Stored mechanical energy is energy stored in objects
mechanical another is motion, e.g. wind and hydropower; water in mechanical by the application of a force. It must push or pull on
energy waterfalls; moving automotive vehicles; and blood energy an object. Compressed springs and stretched rubber
travelling through a body. bands are examples of gravitational potential
energy. When the water is released to spin the
turbines, it becomes kinetic energy.
Electrical Electrical energy is the movement of electrons, e.g. Gravitational Gravitational energy is the energy of place or position;
energy lighting and electricity. Power lines carry electricity. energy of anything ‘up high’. Water in a reservoir behind a
Electric motors are driven by electromagnetic energy. hydropower dam is an example of gravitational
potential energy. When the water is released to spin

7-3
the turbines, it becomes kinetic energy
Mechatronics

Thermal Thermal energy (or heat) is the internal energy in Chemical Chemical energy is the energy stored in the bonds of
energy substances; it is the vibration and movement of atoms energy atoms and molecules; it holds molecules together.
and molecules within substances. The faster the When bonds are broken, chemical energy is
molecules move, the greater the thermal energy released. Biomass, petroleum, natural gas, propane
created. Examples include: geothermal energy, viscous and coal are examples of stored chemical energy.
and dry friction, and changes in the states of matter.
Radiant Radiant energy is electromagnetic energy that travels in Nuclear Nuclear energy is the energy stored in the nucleus of
energy transverse waves. Radiant energy includes visible and energy an atom; it is the energy that holds the nucleus
laser lights, x-rays, gamma rays and radio waves. together. The nucleus of a uranium atom is an
Solar energy is an example of radiant energy. example of nuclear energy When the nucleus of an
atom splits, nuclear energy is released. Nuclear
energy is the most concentrated form of energy. It
comprises both fission and fusion.
Sound energy Sound is the movement of energy through substances in
longitudinal (compression/rare faction) waves.
 Mechatronics

The same modus operandi may be used for problems in mechanical, fluidic and
thermal commutation matrixers. Thus, the analogies between problems in dissimilar
physical heterogeneous continuous dynamical hypersystems are evident.
Consequently, solutions in one physical heterogeneous continuous dynamical system
are convertible by comparison to another one.
The physical model of a physical commutation matrixer ought to be qualitative,
and the mathematical model of a physical commutation matrixer ought to be
quantitative. If it is the latter, there is normally predictability for the experimentation,
either to distribute indispensable data or to adapt a physical commutation matrixer
to do what it is expected to do.
The generalised dynamical systems approach to presenting a physical situation is
normally dealt with firstly as physical modelling and secondly as mathematical
modelling. To create a right and proper physical model necessitates a predisposition
to examine the physical situation and express it first in qualitative terms.
It is also indispensable, secondly, to think in terms of a mathematical model that
can transfer quantitative information regarding the physical situation. To formulate
such a mathematical model, on the other hand, it is necessary to have an accurate
qualitative depiction of the physically observable facts.
It is just when the qualitative depiction (physical model) is terminated that the
quantitative representation (mathematical model) may be completed (Fijalkowski
1987, 2010, 2011).
There is, naturally, a stimulus to accept that when the quantitative representation
(mathematical model) has been completed (i.e. the holor algebra formula) the
conundrum has truly been resolved. The hazard in this dynamical systems approach
is that the physical situations may change. The simple and effective protection
against this is to observe the physical situation regularly.
Surely early man observed electrical energy (electricity), whether in the benign
form of simple static electricity or in the destructive form of radiant energy
(lightning), and wished that one could somehow harness that energy and control it.
At present, we not only know how to generate electrical energy, but how to
convert it into useful forms. However, generation and conversion are not enough.
We are learning that we do not have an unlimited supply of electrical energy, and
nor do we have an unlimited supply of other resources, such as energy supply
engineers’ time. We need more than just energy, we need intelligent energy.
In the generalised sense, artificial intelligence (AI) can be considered to be the ability
to adapt to one’s physical universe, i.e. environment. The interesting (non-trivial)
universe tends to be dynamical, i.e. non-static.
As the physical universe changes, intelligent physical heterogeneous continuous
complex or simple dynamical hypersystems adapt to the changes. In this adaptation,
there is predictably some notion of a ‘cost function’ that is observed and optimised
as a means of controlling the adaptation.
Adaptation can cease when the physical universe impedes change and the
cost function is optimised to an acceptable point. Thus, intelligence has three
basic elements: the physical universe (environment), the cost function and
adaptation.

7-4
 Mechatronics

In terms of energy conversion, the physical universe may include, for example,
things like component values: input energy-potential-difference value and quality,
output energy-transfer value and temperature. The cost function may include, for
example, things like loss or efficiency, ripple, stability, noise, component temper-
ature and pole or zero location.
Adaptation may include, for example, things like changing the relative timing
of commutation, adjusting the energy-potential-difference and/or energy-transfer
levels, changing the compensation and changing the commutation frequency.
For instance, when the electrical energy supply load increases, an intelligent
energy commutator and controller will adjust (adapt to) the dead time between
commutation actions to minimise losses (cost function). A non-intelligent electrical
energy supply will simply sit and get hot.
One can determine, at least on a qualitative level, the relative AI of energy
commutators and controllers within their own species by how they adapt to the
physical universe’s (environment) changes.
Unregulated commutators and controllers are the least intelligent of the commu-
tators and controllers. Any adaptations to physical universe changes are incidental.
As an example, if the load impedance drops, the current only changes according to
Ohm’s law and is defined by the impedance of the conversion commutator.
Feedback has been introduced to electrical energy supplies to increase AI, that is,
to improve their ability to respond to changes in the physical universe (environ-
ment). If the load impedance drops, the feedback network senses a change in the
output voltage and adjusts the operation of the conversion commutator to adapt to
this physical universe change. This level of AI is still limited. For example, the
feedback is designed for a point or narrow range of operating conditions while
assuming given values of the dynamical components. That is, the feedback
parameters do not change. The parameters must be selected to give acceptable
results over the entire operating range. Because of this constraint, an electrical
energy supply designer may spend a considerable amount of time adjusting
parameters to satisfy the operational specification over the entire range of operation.
As the years have progressed, so has the AI of energy conversion. Small
improvements to the intelligence evolved with the addition of feedforward control,
energy-transfer value feedback and temperature correction. For many of these, a
cost function can readily be identified and they can be categorised as true
adaptations.
In this new era of energy, supplies with embedded macroelectronic commutators
(macrocommutators) and microelectronic controllers (microcontrollers) are provid-
ing an entirely new level of AI. No longer are static commutators and micro-
controllers limited to fixed or narrowly constrained parameters. The parameters can
be changed easily from design to design, during operation, and can even be changed
intelligently by the static commutator and microcontroller. Numerous cost functions
can be tracked (like loss, temperature, noise, etc) and the operation can be adapted
based on these cost functions as the physical universe’s conditions change. As an
example of the AI in the best of today’s macrocommutators and microcontrollers,
the relative loss during the energy conversion process is monitored as a cost function.

7-5
 Mechatronics

As the physical universe’s conditions change and losses are no longer minimal, the
energy macrocommutators and microcontrollers adjust or minimise the loss. In
the case of less intelligent energy macrocommutators and microcontrollers, as the
physical universe’s conditions change and impact on efficiency, they simply sit there
and run inefficiently, relatively speaking.
Energy conversion is enjoying a new era in terms of intelligent power supply. The
new era has come along at the perfect time. The physical universe is changing for
energy supply designers; there are fewer designers, more challenging designs that
require higher densities and faster time-to-market, greater cost pressures and higher
performance targets. These novel intelligent energy macrocommutators and micro-
controllers optimise performance so that the designers needs less time to optimise
design over all the physical universe’s conditions. With their high levels of
integration, they provide higher densities and higher performance with an equivalent
or even lower cost of ownership.

7.2 The generalised physical commutation matrixer’s input- and

output-signal holor functions
The aim of this author is to obtain in physical and mathematical form a description
of the dynamical behaviour of a physical heterogeneous continuous dynamical
hypersystem in terms of some physically significant variables called holors. As the
nature of the dynamical hypersystem changes, its holors change. For example, the
holors used in electrical homogeneous continuous dynamical systems are voltage
holors and current holors; in mechanical homogeneous continuous dynamical
systems they are force holors and velocity holors; and in fluidic homogeneous
continuous dynamical systems they are pressure holors and volumetric-flow-rate
holors. Despite the differences in the physical holors used to characterise physical
homogeneous continuous dynamical systems in various physical disciplines, certain
fundamental similarities exist, and it is in the analyst’s interest to seek out and
exploit these similarities in such a manner that the task of physical and mathematical
modelling is eased and one’s overall insight into the dynamical performance of
physical heterogeneous continuous dynamical hypersystems is increased. A suitable
unifying conception that may be used for this purpose is energy. A physical
heterogeneous continuous complex or simple dynamical hypersystem can be thought
of as operating upon a pair of holors whose univalent tensor product is a power
holor.
The physical homogeneous continuous dynamical systems, hyposystems and
components that make up the physical heterogeneous continuous dynamical hyper-
system may thus be thought of as energy commutators and regulators which,
depending upon the way they are interconnected, process the energy injected into the
physical heterogeneous continuous dynamical hypersystem in a characteristic
fashion that is observed as the dynamical hypersystem response.
In the same spirit, the input/output behaviour of a physical heterogeneous
continuous dynamical hypersystem follows if energy can be considered as being

7-6
 Mechatronics

injected into a dynamical hypersystem via energy ports, with similar energy ports
applied to read out the dynamical hypersystem response.
In this book a class of physical heterogeneous continuous complex or simple
dynamical hypersystems is identified that a pair of holors has associated with energy
conversion.
The generalised forms of these dynamical hypersystem holors are the abstract
quantities termed by the author the energy-potential-difference holor and the
energy-transfer holor. In this convention, energy-potential-difference holors are
referred to as ‘transholors’ and energy-transfer holors are denoted as ‘perholors’.
Elemental dynamical hypersystems’ dynamical systems, dynamical hyposystems
and dynamical components are then postulated for the fundamental phenomena of
energy generation, energy storage and energy dissipation.
The concept of physical heterogeneous continuous dynamical systems as energy
static commutators and microcontrollers that interact with inputs and outputs via
energy ports is a conceptual physical model that may be represented in terms of what
the author calls a physical commutation matrixer, which encompasses a wide range
of physical heterogeneous continuous dynamical systems.
In order to develop this notion, however, it is necessary to examine the
mechanism of energetic interactions in terms of so-called dynamical hypersystem
holors, which determine how and in what sense energy is transmitted.
The generalised physical commutation matrixer’s forms for the input- and output-
signal holor functions may be obtained from equations (6.25)–(6.27), namely:
Fi = ZijV j (7.1a )

or
V j = Y jiFi . (7.1b)
If the time domain is constrained to t > 0, one may obtain the energy-potential-
difference-holor (i.e. generalised force) and energy-transfer-holor (i.e. generalised
velocity) functions, Fi and V j , respectively, as

T Fc F cos ψF
Fi = (Fc , Fs ) = Fc Fs = = = F ∠ ψF , (7.2a )
Fs F sin ψF

and
c V cos ψV
V j = (V c , V s ) = V c V s T
= Vs = = V ∠ψV , (7.2b)
V V sin ψV
where, as a general rule, Fi and V j are univalent holors with plethos 2, with merates
or coordinates (independent components) Fc and Fs as well as V c and V s , which may
be expressed, respectively, as

Fc = F cos ψF ⎫ ⎪
⎬, (7.3a )
Fs = F sin ψF ⎭ ⎪

7-7
 Mechatronics

and

V c = V cos ψV ⎫ ⎪
⎬. (7.3b)
V s = V sin ψV ⎭ ⎪

The form of the input-signal holors given in equation (7.2) depends on the values Fc
and Fs as well as Vc and Vs .
Figures 7.1 and 7.2 show the energy-potential-difference-holor and energy-
transfer-holor functions for four possible special combinations of Fc and Fs as
well as Vc and Vs.
For ψF = 0

Fc = F cos ψF = F cos 0 = F ⎫
⎬. (7.4a )
Fs = F sin ψF = F sin 0 = 0 ⎭

For ψF = π /2
π ⎫
Fc = F cos ψF = F cos = 0⎪
2 ⎬. (7.4b)
π
Fs = F sin ψF = F sin = F ⎪
2 ⎭

Thus, if one can obtain the sinusoidal response of a given physical system to the
generalised energy-potential-difference-holor and energy-transfer-holor functions Fi
and V j, one may, by the specification of phase angles ψF and ψV, find the response for
all the special combinations. Even if one does not specify the form of ψF and ψV in
the following sinusoidal responses, the examples given in the textbook will examine

Figure 7.1. A geometrical representation of the energy-potential-difference holor.

7-8
 Mechatronics

Figure 7.2. A geometrical representation of the energy-transfer holor.

the effect of an unquestionably real quantity. We will continue with the specification
of ψF and ψV as holor quantities until chapter 10, when we will discuss frequency
response. An instantaneous value of the step- or sine-wave-input power can be
obtained from a product

p(t ) = f (t )v(t ), (7.5)

where

f (t ) = F sin(ωt + ψF ) = 2 Frms sin(ωt + ψF ) ⎫

⎪
⎬. (7.6)
v(t ) = Vmax sin(ωt + ψV ) = 2 Vrms sin(ωt + ψV )⎪
⎭

Substituting the energy-potential-difference and energy-transfer instantaneous val-

ues from equations (7.6) to equation (7.5), one obtains

p(t ) = FV cos φ − cos⎡⎣2ωt + (ψF + ψV )⎤⎦ ,

{ } (7.7)

where
φ = ψF − ψV .

To obtain a holor representation of the power one must write equation (7.7) in the
form

p(t ) = FV ⎡⎣cos φ − cos + 2ωt cos(ψF + ψV ) + sin 2ωt sin(ψF + ψV )⎤⎦ . (7.8)

7-9
 Mechatronics

If one designs an independent term by Pp and dependent terms from a doublet

frequency by Pq and Pr, then one obtains

P p = FV cos ⌊ψF − ψV ⌋ = FV cos φ = P ⎫

⎪
P q = FV cos(ψF + ψV ) ⎬. (7.9)
r
⎪
P = FV sin(ψF + ψV ) ⎭

In this way equation (7.8) can be rewritten in the form

p(t ) = P p − P q cos 2ωt + P r sin 2ωt . (7.10)
The quantities Pp, Pq and Pr are independent elements (merates) or coordinates of
the three-merates monovalent holor of a step- or sine-wave-input power
P κ = (P p, P q, P r ), (7.11)
which allows us to eliminate time from equation (7.10).
From the first expression of equations (7.9) it is obtained that an independent
element (merate) of the power holor is an active power, which can be measured, for
example, in an electrical heterogeneous continuous dynamical system by means of
an electrodynamic wattmeter. Residual independent elements (merates) can be also
measured or computed with help from physical metrology methods.
In particular circumstances:
a) for an energy-potential-difference-holor phase angle ψF = 0
P q = FV cos φ = P p = P;
P r = −FV sin φ = −Q; and

b) for an energy-transfer-holor phase angle ψV = 0

P q = FV cos φ = P p = P;
P r = FV sin φ = Q.

Thus, for the energy-potential-difference-holor phase angle ψF = 0, or the energy-

transfer-holor phase angle ψV = 0, an independent element (merate) of the holor
power P q is the active power P, however, an independent element (merate) of the
power holor P r is the reactive power Q. To obtain an equation that takes into
account a power holor from the energy-potential-difference holor and the energy-
transfer holor, one must use an adequate law for the multiplication of these holors.
However, in holor algebra, the conventional tensor product in the form FV j
i
(from Moon and Spencer 1986) is not applied, and in the case of the power holor one
uses the particularly useful univalent tensor product
P κ = Fi ⊗ V j = γ..κjiFV
i
j
, (7.12)

7-10
 Mechatronics

where i, j = c, s; κ = p, q, r. The univalent tensor product is designed using the

symbol ⊗, and gives a univalent holor as a result.
From equations (7.12) a power holor
P κ = Fi ⊗ V j = γ..κjiFV
i
j
.

However, from equation (7.1a) an energy-potential-difference holor

Fi = ZijV j .
Hence, the power holor
P κ = Fi ⊗ V j = γ..κjiFV
i
j
= γ..κjiZijV kV j . (7.13)

and its independent elements (merates)

P p = γ..pij ZijV kV j , P q = γ..qij ZijV kV j , P r = γ..rij ZijV kV j . (7.14)

Taking into consideration equations (7.13) and (7.14), one obtains

P p = R⌈(V c ∧ 2) + (V s ∧ 2)⌉ ⎫
⎪
P q = R⌈(V c ∧ 2) − (V s ∧ 2)⌉ − 2XV cV s ⎬ . (7.15)
⎪
P r = X⌈(V c ∧ 2) − (V s ∧ 2)⌉ + 2RV cV s ⎭

Substituting equations (7.9) into the first expression from equations (7.15), one
obtains
P p = R(V ∧ 2) = P. (7.16)
In this way, one obtains a well-known expression for a sine-wave-input physically
continuous dynamical system analysis from which once more one can see that an
independent element (merate) of the holor power P p is a real or active power.
Adding the geometrically independent elements (merates) P q and P r of the power
holor, one derives a well-known expression for a physical effective value of the
apparent power S in the form

sqrt⎡⎣(P q ∧ 2) + (P r ∧ 2)⎤⎦ = sqrt⎡⎣(R∧ 2) + (X ∧ 2)⎤⎦⎡⎣(V c ∧ 2) + (V s ∧ 2)⎤⎦

(7.17a )
= Z (V ∧ 2) = FV = S .

Thus, due to the application of holor algebra for a sine-wave-input physically

continuous dynamical system analysis, it is possible to have an operation with holor
quantities for physically continuous dynamical systems that are analogous to each
other as with real quantities for step-input physically continuous dynamical systems.
A geometrical representation of the power holor needs 3D space. The coordinates
are independent elements (merates) of the power holor Pp, Pq and Pr. The
coordinates Pq and Pr are orthogonal to each other, as in equation (7.17a). The
third axis Pp establishes an angle φ = ψV − ψi with a vector of the apparent power,

7-11
 Mechatronics

which a module S = F V. The power holor P κ is hence a vector sum of its

independent elements (merates) Pp, Pq and Pr (figure 7.3).
A partial representation of the holor power needs 2D space. If a phase angle of
the double-frequency term is small, then a module of the double frequency power
can be obtained from a part of the holor diagram shown in figure 7.3, which depicts
a well-known holor diagram for the power holor established by independent
elements (merates) of the real or active power Pp = P and the reactive power Q.
A partial holor diagram of the holor power, however, does not represent the holor
power exactly, but a strict representation needs three independent elements (merates)
and 3D space. Thus a physical effective value of the apparent power P κ between two
points would be:

P κ = sqrt⎡⎣(P p∧ 2) + (P q ∧ 2) + (P r ∧ 2)⎤⎦ = S . (7.17b)

But what would happen if one moved a physically continuous dynamical system or
body from one point to another point in universal space? There would be a change in
power between the physically continuous dynamical system or body and the
reference point and there would be a rate of change in the power holor P κ per
unit interval, which can be expressed mathematically as ωP κ . Here, a new dimension
is introduced; ω, which is an element of time. One cannot deny that there must be
time, because one can perceive that what has been is different from what is present
and what will be. So one has a past, a present and a future, all of which refer to time.
In the Universe where one exists, time appears to flow equably in one direction,
from the past to the future. This is our own timeline. It is, however, also possible that
time flows in at least three directions. If this is the case, then time can have its own
set of Cartesian coordinates that are similar to but not necessarily the same as those
of space. One may assign Pp, Pq and Pr as the coordinates of time. Here one has a
universe of at least nine dimensions.
We have always used a notion of the sense of direction. We say that a thing goes
forward or backward, or is positive or negative. These senses of direction, therefore,
are dimensions in their own right and may be denoted by ‘+’ for forward and ‘−’ for
backward.

Figure 7.3. A geometrical representation of the power holor.

7-12
 Mechatronics

So far one has only looked at the possible dimensions governing the relation of a local
physically continuous dynamical system or body with a reference point in universal space.
Without a concept of relativity, one presumes that rigid bodies are possible, where
a rigid body remains unchanged when it is moved from one point to any of the
infinite possible locations in universal space. In other words, one assumes that space
is independent of its contents. In such a view, our physics is basically Newtonian.
The existence of a particular physically continuous dynamical system or body
however is also governed by other parameters. Careful observation will give one the
idea that all things are in motion. Motion or momentum, therefore, is a property of
matter. One could then deduce that nothing exists that does not move. If such is the
case, then space could not exist separately from matter, hence all matter has space
and all spaces have matter. There is, therefore, no empty space.

7.3 Multi-port, multi-input/multi-output, generalised physical

commutation matrixers
In 1820, the marquis de Laplace wrote that ‘We ought to regard the present state of
the universe as the effect of its antecedent state and as the cause of the state that is to
follow. An intelligence knowing all the forces acting in nature at a given instant, as
well as the momentary positions of all things in the universe, would be able to
comprehend in one single formula the motions of the largest bodies as well as the
lightest atoms in the world, provided that its intellect were sufficiently powerful to
subject all data to analysis; to it nothing would be uncertain, the future as well as the
past would be present to its eyes. The perfection that the human mind has been able
to give to astronomy affords but a feeble outline of such an intelligence’ (Laplace
1820). This image had a great impact on the dynamical systems approach (systems
thinking) in the 1980s. Nowadays, scientists believe that in nature chaos develops in
most phenomena and that predicting the future is impracticable in the long term.
Controlling chaos is very important from the scientific point of view.
There has been a very rapid revolution in the fields of physics and engineering—
especially mechatronics (mechanics, fluidics, pneumatics, photonics and electronics)—
because stereotyped ideas in the field of knowledge have been overcome. All progress is
stimulated by thinking that is free from dogma and not indiscriminately accepting of
existing concepts. It is also well known that perfect practical solutions are born out of
ingenious theories. ‘There is nothing more practical than a good theory,’ said
L Boltzmann (1844–1906).
In physics, the concept of matrix-interconnected physical networks, termed physical
commutation matrixers (figure 7.4) by the author of this book, was developed from the
mathematical matrix notation of the dynamical systems approach (systems thinking)
for the formation methodology of physical and mathematical models of mechatronical
(e.g. mechanical and/or electrical) heterogeneous continuous dynamical hypersystems
with energy-transfer holor conducting horizontal-row and vertical-column conductive
collectors (conductors), which are crossed with each other but galvanically
separated (insulated) in-between by means of uni- and/or bipolar dissimilar passive
or active physical dynamical components and structural devices (e.g. physical
valves) (Fijalkowski 1967, 1973, 1982, 1985, 1987, 1995).

7-13
 Mechatronics

Figure 7.4. A multi-port MIMO generalised physical commutation matrixer (generalized physical model)
(Fijalkowski 1967, 1973, 1982, 1985, 1987, 1995).

The author’s concept relates to l physical dynamical components that are

interconnected in a (m × n) matrix design with m horizontal-row conductive
collectors and n vertical-column conductive collectors, with l, m and n being natural
numbers unequal to zero. A first conductive collector end of the m row conductive
collectors and n column conductive collectors is connected to an evaluation system
and every l physical dynamical component is interconnected between two of the row

7-14
 Mechatronics

conductive collectors and the column conductive collectors. The concept of the
matrixer arrangement further involves at least one return conductive collector, whose
first conductive collector end is connected to the evaluation system and whose second
conductive collector end is in contact with the second conductive collector end of one
of the row or column conductive collectors (bars or wires). Theoretically and
practically, the multi-port, multi-input/multi output (MIMO) generalised physical
commutation matrixers are the changers of an energy-potential-difference holor (e.g.
a generalised force holor); an energy-transfer holor (e.g. a generalised velocity holor);
the frequency and number of phases (virtually, static matrix converters, which
alter one kind of physical energy with an energy-potential-difference holor, FiP
(e.g. a voltage holor – UiP )); an energy-transfer holor, VPj (e.g. a current holor, IPj );
the frequency, fP, and number of phases, mP, into another kind of physical energy
with an energy-potential-difference holor, Fis , (e.g. a voltage holor, Uis ); an
energy-transfer holor, Vsj (e.g. a current holor, Isj ); and the frequency, fs, and
number of phases, ms, performing the definite matrix function. The multi-port
MIMO generalised physical commutation matrixer concept is an extension of the
one- or two-port MIMO generalised physical commutation matrixer concept.
In figure 7.4 there is a generalised physical model (a matrix diagram or matrix map) of
the multi-port MIMO generalised physical commutation matrixer realised on unipolar
and/or bipolar different passive and/or active physical components and devices (e.g.
physical devices). In this physical commutation matrixer, the horizontal collectors (matrix
rows) A, B, C,…, N are termed primary or input conductive collectors, but the vertical
conductive collectors (matrix columns) a, b, c,…, n are termed secondary or output
conductive collectors. The differentiation of the primary (input) conductive collectors and
the secondary (output) conductive collectors in the multi-port MIMO generalised
physical commutation matrixer is relative, since the physical energy may be charged as
well as discharged (emitted) through the same conductive collectors at any time.
In the general case, the primary conductive collectors are connected with the
secondary conductive collectors by means of any uni- and/or bipolar continuous
physical dynamical components and structural devices (figure 7.4). Physical energy
conversion in any multi-port MIMO generalised physical commutation matrixers
takes place in advance, specified or adapted for the given multi-port MIMO
generalised physical commutation matrixers’ rule algorithms on the basis of the
phenomena’s modus operandi, termed commutation, i.e. an energy-transfer-holor
passage (i.e. electric-current-holor passage) switching changeover among preselected
conductive collectors of the multi-port MIMO physical commutation matrixer,
which connect the terminals of two dissimilar physical heterogeneous continuous
dynamical hypersystems—the primary and secondary.
In order to concretise a concept of the physical commutation matrixer it is
convenient to use the concept so generalised, which after a reduction in consid-
erations is carried out might formulate specific physical commutation matrixer
mathematical models of any physical homogeneous continuous dynamical systems
that are applied in practice. This approach permits generalised physical commuta-
tion matrixers in a synthetic systems approach on the examination of any specific
multi-port MIMO.

7-15
 Mechatronics

Assuming that the excitations of a multi-port MIMO generalised physical

commutation matrixer are given in the form of the column hypermatrices or
hypervectors of energy sources among the primary (input) and secondary (output)
conductive collectors (figure 7.5):

Figure 7.5. A multi-port MIMO generalised physical commutation matrixer with energy-potential-difference
holor sources (physical model I) (Fijalkowski 1987).

7-16
 Mechatronics

T T T T T
T
pT sT
T FiAB FiBC FiC (N − 1) ⋯ Fi(N − 1)N FiNA
Fi = Fi Fi = T T T T T
(7.18)
Fiab Fibc Fic(n −1) ⋯ Fi(n −1)n Fina

as the responses of the multi-port MIMO generalised physical commutation

matrixer, the column hypermatrices or hypervectors of the outer-mesh energy-
transfer holors,

T T T T T
T
j j j
T T T VAB VBC VCj(N − 1) ⋯ V(Nj −1)N VNA
Vj = VPj Vsj = T T T T T
, (7.19)
Vabj Vbcj Vcj(n − 1) ⋯V(nj −1)n Vnaj

a set of the differential equations of dynamics, i.e. the mathematical model of the
multi-port MIMO generalised physical commutation matrixer as the physical
heterogeneous continuous dynamical hypersystem, may be formulated in the
dynamical systems approach and matrix notation as follows:

Fi = ZijV j , (7.20)

where the i × j hypermatrix Zij has a dimension of the impedance holor.

To denote the state variables associated with the primary (input) conductive
collectors by an index P, and those associated with the secondary (output) collectors
by an index s, a set of the differential equations of dynamics, i.e. the mathematical
model of the multi-port MIMO generalised physical commutation matrixer, may be
expressed as follows:

FiP ZijPP ZijPs VPj

= , (7.21a )
Fis ZijsP Zijss Vsj

or in detail

ZijAA ZijAB ZijAC ZijAN ZijAa ZijAb ZijAc ZijAn j

VAB
FiAB
FiBC ZijBA ZijBB ZijBC ⋯ ZijBN ZijBa ZijBb ZijBc ⋯ ZijBn j
VBC

FiC (N − 1) ZijCA ZijCA ZijCC ZijCN ZijCa ZijCb ZijCc ZijCn VCj(N −1)
⋮ ⋮ ⋱ ⋮ ⋮ ⋱ ⋮ ⋮
FiNA ZijNA ZijNB ZijNC ⋯ ZijNN ZijNa ZijNb ZijNc ⋯ ZijNn j
VNA
= . (7.21b)
Fiab ZijaA ZijaB ZijaC ZijaN Zijaa Zijab Zijac Zijan Vabj
Fibc ZijbA ZijbB ZijbC ⋯ ZijbN Zijba Zijbb Zijbc ⋯ Zijbn Vbcj
Fic(n −1) ZijcA ZijcB ZijcC ZijcN Zijca Zijcb Zijcc Zijcn Vcj(n −1)
⋮ ⋮
⋮ ⋱ ⋮ ⋮ ⋱ ⋮
Fina Vnaj
ZijnA ZijnB ZijnC ⋯ ZijcN Zijna Zijnb Zijnc ⋯ Zijnn

7-17
 Mechatronics

The coefficients
∂Fk(k −1)
Zkl = (7.21c )
∂V l (l −1) V j (j − 1)

have a dimension of the impedance holor.

To adapt a mathematical model of the multi-port MIMO generalised physical
commutation matrixer to the theory of physical heterogeneous continuous dynam-
ical hypersystems demands that one refers to a simplified and very general
mathematical model of the multi-port MIMO generalised physical commutation
matrixer, derived from the physical model that is shown in figure 7.6.
In order to fulfil the demands of converting one kind of physical energy with the
energy-potential-difference holor, FiP (e.g. a voltage holor, UiP ), the energy-transfer
holor, VPj (e.g. a current holor, IPj ), the frequency, fP, and the number of phases, mP,
of a primary physical heterogeneous continuous dynamical hypersystem of the
multi-port MIMO generalised physical commutation matrixer into another kind of
physical energy with the energy-potential-difference holor, Fis (e.g. a voltage holor,
Uis ), the energy-transfer holor, Vsj (e.g. a current holor, Isj ), the frequency, fs, and the
number of phases, ms, of a secondary physically continuous dynamical system of
the multi-port generalised physical commutation matrixer—as well as the free
passage of physical energy between these dynamical hypersystems—they ought to
be connected by means of the multi-port MIMO generalised physical commutation
matrixer’s uni- and/or bipolar passive and active physical dynamical components
and devices.
A physical model of the multi-port MIMO generalised physical commutation
matrixer (see figure 7.6) fulfils the aforementioned demands. Uni- and bipolar
passive and active physical dynamical components and devices connect both
physical heterogeneous continuous dynamical hypersystems of the multi-port
MIMO generalised physical commutation matrixer and make free passage of
physical energy between them possible.
In the general case, a physical commutation matrixer such as this can connect
many physical heterogeneous continuous dynamical hypersystems.
In this book we consider a case where there are only two physical heterogeneous
continuous dynamical hypersystems—the primary and the secondary. It is sufficient
to understand the multi-port MIMO physical commutation matrixer concept.
Both physical heterogeneous continuous dynamical hypersystems of the multi-
port generalised physical matrixer may be considered to be autonomous, i.e. when
disconnected from the multi-port MIMO generalised physical commutation
matrixer, they have determined potentials on their terminals.
They have energy-potential-difference-holor sources (e.g. voltage-holor sources),
energy-transfer-holor sources (e.g. current-holor sources) and physical dynamical
components (e.g. inductances, capacitances and resistances), as well as continuous
uni- and/or bipolar physical valves.
Assuming that the excitations of a multi-port MIMO generalised physical
commutation matrixer are given in the form of the column hypermatrices or

7-18
 Mechatronics

Figure 7.6. A multi-port MIMO generalised physical commutation matrixer with energy-transfer holor
sources (physical model II) (Fijalkowski 1987).

hypervectors of source-energy-transfer holors (e.g. current holors) flowing in the

primary (input) and secondary (output) conductive collectors (figure 7.6):
T T T T
T
T T T VAj VBj VCj ⋯ VNj
Vj = VPj Vsj = T T T T
, (7.22)
Vaj Vbj Vcj ⋯ Vnj

7-19
 Mechatronics

as the responses of the multi-port physical commutation matrixer, the column

hypermatrices or the hypervectors of node energy-potential-difference holors
(e.g. voltage holors)
T T T
T
T T T FiA FiB FiT ⋯ FiN
Fi = FiP Fis = T T T T
, (7.23)
Fia Fib Fic ⋯ Fin

a set of the differential equations of dynamics, i.e. the mathematical model of the
multi-port MIMO generalised physical commutation matrixer as the physical
heterogeneous continuous dynamical hypersystems, may be formulated in the
dynamical systems approach and matrix notation as follows:

V j = Y jiFi , (7.24)

where the n × m hypermatrix Y ji has a dimension of the admittance holor.

Node energy-potential-difference holors (e.g. voltage holors) consisting of phys-
ical elements of the hypermatrix Fi are measured as regards the reference node,
which is located on the outside of the multi-port MIMO generalised physical
commutation matrixer. For example, in a star connection of the multi-phase
generalised physical heterogeneous continuous dynamical hypersystem, the refer-
ence node may be a null point of the multi-phase star.
To denote the state variables associated with the primary (input) collectors by an
index P, and those associated with the secondary (output) collectors by an index s,
the set of differential equations of dynamics, i.e. the mathematical model of the
multi-port MIMO generalised physical commutation matrixer, may be expressed as
follows:

ji
VPj YPP YPsji FiP
= , (7.25a )
Vsj YsPji Yssji Fis

or in detail as

VAj ji
YAA ji
YAB ji
YAC ji
YAN ji
YAa ji
YAb ji
YAc ji
YAn FiA
VBj ji
YBA ji
YBB ji
YBC ⋯ ji
YBN ji
YBa ji
YBb YBcji ⋯ ji
YBn FiB
ji ji ji ji ji ji
VCj YCA YCA YCC YCN YCa YCb YCcji ji
YCn FiC
⋮ ⋮ ⋱ ⋮ ⋮ ⋱ ⋮ ⋮
ji ji
VNj ji
YNA ji
YNB ji
YNC ⋯ YNN ji
YNa ji
YNb ji
YNc ⋯ YNn FiN
= . (7.25b)
Vaj ji
YaA ji
YaB ji
YaC ji
YaN Yaaji Yabji Yacji Yanji Fia
Vbj ji
YbA ji
YbB ji
YbC ji
⋯ YbN Ybaji Ybbji Ybcji ⋯ Ybnji Fib
Vcj ji
YcA YcBji ji
YcC ji
YcN Ycaji Ycbji Yccji Ycnji Fic
⋮ ⋮ ⋱ ⋮ ⋮ ⋱ ⋮ ⋮
Vnj ji
YnA ji
YaB ji
YnC
ji
⋯ YcN Ynaji Ynbji Yncji ⋯ Ynnji Fin

7-20
 Mechatronics

The coefficients
∂V k
Y kl = (i ≠ l ), (7.25c )
∂Fl
Fi = 0

have a dimension of the admittance holor.

As with the mathematical models of physical heterogeneous continuous dynam-
ical hypersystems of generalised physical commutation matrixers shown in figures
7.5 and 7.6, consisting of a set of differential equations of dynamics (7.20) and (7.24),
there exists a formulation possibility for the mathematical model of physically
continuous dynamical systems of the generalised physical commutation matrixer
(figure 7.7) acting as an energy-potential-difference commutator (matrix converter)
with an arbitrary number N of primary collectors (inputs), equal to the number m of
physical matrix phases and an arbitrary number n of secondary collectors (outputs).
Summing up, a hybrid mathematical model of the multi-port MIMO generalised
physical commutation matrixer may be written as follows:

∂Fk(k −1) ∂V k
Zkl = l (l −1)
(j ≠ l ) and Ykl = (i ≠ l ) (7.27)
∂V j (j −1) ∂Fl
V Fi = 0

having dimensions of the impedance or admittance holor, respectively.

Assuming that the energy-potential-difference-holor sources applied across the
primary collectors (inputs) A, B, C, …, N of a generalised physical commutation
matrixer commutator (matrix converter) are given in the form of the energy-
potential-difference holors’ column matrix or vector (input holors)
T T T T T T
FiP = FiA FiB FiC ⋯ FiN − 1 FiN , (7.28)

as the primary physical state variables (inputs), and energy-potential-difference-

holors’ sources, applied across the secondary collectors (outputs) a, b, c, …, n of a
generalised physical commutation matrixer commutator (matrix converter), in the
form of the energy-potential-difference holors’ matrix or vector (output holors)

7-21
 Mechatronics

Figure 7.7. Physical model of a generalised physical commutation matrixer commutator (Fijalkowski 1987).

T T T T
Fbj
T
Fsj = Faj Fcj
T
⋯ Fnj−1 Fnj , (7.29)

as the secondary physical state variables (output holors).

Well, a mathematical model of the physical heterogeneous continuous dynamical
hypersystem of a generalised physical commutation matrixer commutator (matrix

7-22
 Mechatronics

converter) acting as an energy-potential-difference commutator, making the outputs

dependent on the inputs, in the dynamical systems approach and a matrix notation
may be written in the form:

Fsj = C sPjiFiP, (7.30)

or in detail as

Faj ji
C aA ji
C aB ji
C aC C aji(N −1) C aN
ji
FiA
j
Fb nm
C bA nm
C bB nm
C bC ⋯ C bnm
(N −1)
nm
C bN FiB
nm nm nm
Fcj C cA C cB C cC C cnm nm
(N −1) C cN FiC
= . (7.31)
⋮ ⋮ ⋱ ⋮ ⋮
Fnj−1 C (nm nm nm C (nm nm
n −1)(N −1) C (n −1)N
FiN −1
n −1)A C (n −1)B C (n −1)C
Fnj ⋯ FiN
nm
C nA nm
C nB nm
C nC C nnm nm
(N −1) C nN

However, a mathematical model of the physically continuous dynamical system of a

generalised physical commutation matrixer commutator (matrix converter) acting as
an energy-potential-difference commutator, making the inputs dependent on the
outputs, may be written in the form:

FiP = CijPs Fsj , (7.32)

or in detail as

A(n −1)
C ijAa C ijAb C ijAc C ij C ijAn
FiA Faj
FiB C ijBa C ijBb C ijBc ⋯ B (n −1)
C IJ Bn
C IJ Fbj
Ca Cb Cc
FiC C IJ C IJ C IJ C (n −1)
C IJ Cn
C IJ Fcj
= , (7.33)
⋮ ⋮ ⋱ ⋮ ⋮
FiN −1 (N −1)a (N −1)b (N −1)c (N −1)(n −1) (N −1)n Fnj−1
C IJ C IJ C IJ C IJ C IJ
FiN Na Nb Nc
⋯
N (n −1) Nn Fnj
C IJ C IJ C IJ C IJ V IJ

JI
where C kL and CijKl are the energy-potential-difference coefficients of the commu-
tation nodes kL and Kl, respectively, which may be expressed as follows:
⎧ Δ j L
⎪ + 1 → F k = + Fi
∂Fkj ⎪
FiI = 0 = ⎨ 0 → F k = 0
ji Δ j
C kL =
∂FiL ⎪
⎪ Δ j L
⎩ − 1 → F k = − Fi

7-23
 Mechatronics

and
⎧ Δ k L
⎪ − 1 → Fi = − Fi
∂FiK ⎪
Fi j = 0 = ⎨ 0 → Fi = 0
Δ
CijKl = k .
∂Fl j ⎪
⎪ Δ k j
⎩ + 1 → Fi = + Fl
In the case of the generalized physical commutation matrixer with bipolar (sym-
metrical) physical valves, the energy-potential-difference coefficients of the commu-
ji
tation nodes C kL = ±1 and CijKl = ∓1 represent an ideal physical-valve impedance
equal to zero, which is incorporated between the primary collector L and the
secondary collector k, or between the secondary collector l and the primary collector
K, and correspond to a turn-on state of the physical valves with a sense of the
energy-transfer flow direction (polarisation) that is dependent on a sign of these
coefficients; however, the energy-potential-difference coefficients of the commuta-
ji
tion nodes C kL = 0 and CijKl = 0, in which the ideal physical-valve impedance equals
infinity, correspond to a turn-off state of the physical valves. Analogously, a
mathematical model of the physical heterogeneous continuous dynamical system
of a generalised physical commutation matrixer commutator (matrix converter)
(figure 7.7), as an energy-transfer commutator with an arbitrary number N of
primary collectors (inputs), equals a number m of physical phases and an arbitrary
number n of secondary collectors (outputs). Assuming that an energy-transfer
holors’ sources, flowing in the primary collectors (inputs) A, B, C, …, N of a
physical commutation matrixer of the physical commutation matrixer commutator
(matrix converter), are given in the form of the energy-transfer holors’ column
matrix or vector (input holors)
T T T T
ViP = ViA ViB ViC
T T
⋯ ViN −1 ViN , (7.34)

as the primary physical state variables (inputs), and the secondary energy-transfer
holors’ sources of a generalised physical commutation matrixer of the physical
commutator, flowing in the secondary collectors’ (outputs) a, b, c, …, n of a physical
commutation matrixer, in the form of the secondary energy-transfer holors (output
holors)
T T T T
Vsj = Vaj Vbj
T
Vcj
T
⋯ Vnj−1 Vnj , (7.35)

as secondary physical state variables (inputs).

Well, a mathematical model of the physical heterogeneous continuous dynamical
hypersystem of a generalised physical commutation matrixer commutator (matrix
converter) acting as an energy-transfer commutator, making the outputs dependent
on the inputs, in the dynamical systems approach and a matrix notation, may be
written in the form:
Vsj = C sPjiViP , (7.36)

7-24
 Mechatronics

or in detail as

ji
C aA ji
C aB ji
C aC C aji(N −1) C aN
ji
Vaj ViA
Vbj
ji
C bA ji
C bB ji
C bC ⋯ C bji(N −1) ji
C bN
ViB
ji ji ji
Vcj C cA C cB C cC C cji(N −1) ji
C cN ViC
= . (7.37)
⋮ ⋮ ⋱ ⋮ ⋮
Vnj−1 C (nji−1)(N −1) C (nji−1)N ViN −1
C (nji−1)A C (nji−1)B C (nji−1)C
Vnj ⋯ ViN
ji
C nA ji
C nB ji
C nC C nji(N −1) C nN
ji

However, a mathematical model of a physically continuous dynamical system of the

generalised physical commutation matrixer commutator (the matrix converter
shown in figure 7.7), acting as an energy-transfer commutator, making the inputs
dependent on the outputs, in the dynamical systems approach and matrix notation
may be expressed in the form:

ViP = CijPsVsj, (7.38)

or in detail as

A(n −1)
C ijAa C ijAb C ijAc C ij C ijAn
ViA Vaj
B (n −1)
ViB C ijBa C Bb C ijBc ⋯ C ij C ijBn Vbj
ViC C ijCa C ijCb C ijCc C (n −1)
C ij C ijCn Vcj
= , (7.39)
⋮ ⋮ ⋱ ⋮ ⋮
ViN −1 (N −1)a (N −1)b (N −1)c (N −1)(n −1) (N −1)n Vnj−1
C ij C ij C ij C ij C ij
ViN ⋯
N (n −1)
Vnj
C ijNa C ijNb C ijNc C ij C ijNn

ji
where C kL and CijKl are the energy-transfer coefficients of the commutation nodes kL
and Kl, respectively.
To recapitulate, it may be affirmed that an action of a generalised physical
commutation matrixer commutator as an energy-transfer commutator (shown in
figure 7.7) may be presented by means of its mathematical model, consisting of the
equations of unholonomic constraints of an energy-transfer holor in matrix
notations (7.38) and (7.39).
For equations of unholonomic constraints of energy-transfer holors in matrix
notation formulated in this way, isochronic variations of the generalised coordinates
are not already independent of each other and that is why we apply the dynamical
systems approach in the methodology for the formulation of mathematical models
of physical commutation matrixer commutators (matrix converters) in a matrix
notation of the Euler–Lagrange second-order differential equation of dynamics.

7-25
 Mechatronics

7.4 The principal physical commutation matrixer dynamical

components
One of our primary objectives is the determination of suitable physical models for
physical situations. Specifically, one wants mathematical models based on estab-
lished natural laws. However, one limits one’s consideration to physical situations
where the mathematical modelling can be accompanied by linear constant coef-
ficient holor equations.
Mathematical models will only be developed for the most important physical
commutation matrixer dynamical components. More complicated physical situa-
tions can then be described with combinations of basic physical commutation
matrixer dynamical components. It is feasible to depict nearly all of the principal
physical situations using single-port (two-terminal) physical commutation matrixer
dynamical components, but there are physical situations that necessitate a two-port
physical model (with two pairs of terminals).
The common multi-port (two-terminal pairs) physical commutation matrixer
dynamical components are described in chapter 8. In spite of this, at this stage one
will only consider the single-port (two-terminal) physical commutation matrixer
dynamical components.
A general single-port physical commutation matrixer dynamical component is
shown schematically in figure 7.8.
The energy-potential-difference holor Fi symbolises a difference in the physical
situation on either interface of the physical commutation matrixer dynamical
component and the energy-transfer holor V j symbolises the quantity transmitted
through the physical commutation matrixer component.
The value of the single-port depiction lies in the fact that it allows the physical
commutation matrixer dynamical components to come together into physical
commutation matrixers that can be analysed by a very advanced physical commu-
tation matrixer theory. In this manner, one may apply a universal holor method of
analysis for all the physical commutation matrixers.
Normally, there are three basic physical commutation matrixer dynamical
components in each of the physical heterogeneous continuous dynamical
hypersystems:
• a single energy-dissipation physical commutation matrixer dynamical com-
ponent in which energy is lost;

Figure 7.8. Single-port (two-terminal) physical commutation matrixer dynamical component.

7-26
 Mechatronics

Figure 7.9. Principal physical commutation matrixer dynamical components: (a) energy dissipation; (b) kinetic
energy storage; (c) potential energy storage.

• double energy-storage physical commutation matrixer dynamical components

in which energy is retained.

Figure 7.9 demonstrates the physical commutation matrixer symbol for each
physical commutation matrixer dynamical component performance. The symbol for
the energy-dissipation physical commutation matrixer dynamical component (figure
7.9(a)), for instance, symbolises electrical resistance. In spite of this, one applies this
symbol to symbolise a physical commutation matrixer dynamical component in any
of the physical systems in which the energy-potential-difference holor Fi and the
energy-transfer holor V j are directly proportional to each other.
This physical situation fits into one’s observation of the fluidic transmission line.
Thus, the fluidic transmission line may be considered as a fluidic resistance.
Correspondingly, the storage physical commutation matrixer dynamical component
symbols (figures 7.9(b) and (c)), as a rule, delegate the electrical inductor and capacitor.
For one physical solution, one may apply these identical symbols for physical
commutation matrixer dynamical components in other physical homogeneous
continuous dynamical systems that create the identical relationship between the
energy-transfer holor, V j , and the energy-potential-difference holor, Fi .
We find, for example, that mechano-mechanical (MM) springs produce the same
signal holor relation as an inductance and that a mass produces a capacitive-type
relation. These physical commutation matrixer dynamical components are considered
in detail in chapter 8.
For the state-of-the-art, on the other hand, it will be adequate to examine whether
the relations between the energy transfer holor, V j , and the energy-potential-difference
holor, Fi , are proportional (P), differential (D), or integral (I). The D and I physical
commutation matrixer dynamical components realise time-domain-dependent
operations on the signal holors. These mathematical operations are considered in
detail in part II.

7.5 Input- and output-signal holors of physical commutation matrixers

Principally, the physical commutation matrixers that one contends with are com-
posed of specific matrix permutations of single-port physical-commutation-matrixer
dynamical components. These matrix permutations when combined together create a
more sophisticated mathematical function than is created by a single-port physical

7-27
 Mechatronics

commutation matrixer dynamical component. In spite of this, in chapter 10 we

discuss mutually coupled physical commutation matrixer dynamical components
that have more than one port (two terminals). A conventional physical commutation
matrixer with single-port physical commutation matrixer dynamical components is
shown in figure 7.10.
The input symbolises a signal holor (either energy-potential-difference or energy-
transfer form) that is relevant to the physical commutation matrixer. The input is,
for that reason, principally an input-signal holor. The physical commutation
matrixer operates on the input-signal holor in relation to the physical commutation
matrixer mathematical function and generates an output-signal holor.
The output-signal holor is occasionally termed the sinusoidal response of the
physical commutation matrixer. It can be of either the energy-potential-difference or
the energy-transfer form.
In conventional applications, a physical commutation matrixer or physical
heterogeneous dynamical hypersystem generates particular effects or results, namely
those required by the human being operating the assembly of physical equipment or
individual equipment.
The normal term used for such effects or results is ‘output’. For instance, the
output of an automotive vehicle could be its travel velocity. However, in order to
obtain an output, there must be some process of giving energy or information (or
both) to the physical heterogeneous continuous complex dynamical hypersystem.
This is the ‘input’. Inputs to an automotive vehicle may include the energy stored in
the liquid or gas fuel, the position of the steering wheel, the on/off-road surface and
the accelerator- and/or brake-pedal positions. These hypotheses are, as a rule,
represented graphically using a physical commutation matrixer of the heterogeneous
continuous complex or simple dynamical hypersystem shown in figure 7.11
(Fijalkowski 1987, 2010, 2011).
For an automotive vehicle, if efficiency is of no concern, there is no need to
consider the output energy as one of the outputs and nor is there any need to
consider the chemical energy of the liquid fuel as one of the inputs. In this case, the

Figure 7.10. Conventional single-port physical commutation matrixer.

7-28
 Mechatronics

Figure 7.11. Physical commutation matrixer representation of the physical heterogeneous continuous complex
or simple dynamical hypersystem.

output torque of the internal combustion engine (ICE) or external combustion engine
(ECE) would be an input to the automotive powertrain system.
The sense of the term ‘physical heterogeneous continuous complex or simple
dynamical hypersystem’ as a result comprises the hypothesis of a boundary within
all the dynamical components that create the physical heterogeneous continuous
complex or simple dynamical hypersystem.
In this example of the automotive vehicle, it must be noted that it is only the
energy and information conversion process that would not be included in the
physical system; the mass of the ICE or ECE (both the case and moving parts) would
still be included in the physical system.
To some extent, the careful selection of the boundary for a specific problem is an
area of expertise, the knowledge of which is one of the qualities of top physical
system scientists and engineers.
The terms ‘input-signal holor’, ‘output-signal holor’ and ‘physical heterogeneous
continuous dynamical hypersystem’ may as a result give the impression of being
somewhat fuzzy.
However, it is elementary to be cognisant of what activates the function (cause) in
a physical heterogeneous continuous dynamical hypersystem (input) and what
constitutes the effects or results (output).
Normally, there is a minor problem in indicating the inputs and outputs of a
physical heterogeneous continuous dynamical hypersystem if it is practical to
provide an acceptable qualitative depiction (physical model) of it.
The estimation of the performance of a physical heterogeneous continuous
dynamical hypersystem requires the creation of the response (output) to particular
expected causes (inputs).
Physical heterogeneous continuous dynamical hypersystems normally traverse a
dynamical or transient state in which the holor algebraic values for their outputs
vary.
Occasionally, this dynamical state is the decisive point in creating the adequacy of
the operation and then it is the dynamical response that ought to be exposed.

7-29
 Mechatronics

In view of the fact that the holor algebraic values of the output vary with time, the
easiest method to express the dynamical response is to use a mathematical function
of time. Normally, the inputs also vary with time and are expressed by mathematical
time functions.
In spite of this, there is a natural thought that the input function can be identified.
It is the cause. The output function, by contrast, is the effect, and it relies on both the
input and the physical commutation matrixer.
Figure 7.12 displays two conventional input functions, the ‘step function input’
(figure 7.12(a)) and the ‘sine wave input’ (figure 7.12(b)).
In a real physical heterogeneous continuous dynamical hypersystem, the input
may be recognised a priori. For instance, some human and/or robotic drivers
(H&RD) step down hard on the accelerator pedal, while others drive sensibly. In
spite of this, the input functions exposed are in fact trial inputs. They allow one to
evaluate the performance of physical heterogeneous continuous dynamical hyper-
systems and physical commutation matrixers. The output that can be found by
computation using the step function input is termed the step response.
The output from a sine-wave input is the sinusoidal response. It is made up of a
transient member and a sinusoidal ‘steady-state’ member.
The sine-wave member of the output has an unchanged frequency, but normally a
changed amplitude and phase angle compared to the sine-wave input.
The performance in which the amplitude or effective (root-mean-square, rms)
value and phase angle of the outputs for the input sine-wave alter over a selection of
various frequencies is termed the frequency response.
In analyses of the physical environment of various conventional responses there
should be evidence of good design. As examples, consider the mechanical commu-
tation matrixers of a simple pendulum with a mass at the end of a string, or a bicycle
being pedalled along a horizontal road on a calm day, both of which are described in
chapter 12 (sections 12.1 and 12.2). Other examples in which the forced and natural
responses may readily be identified by a similar argument are the heating and
cooling of an oven and the switching on and off of a large fan.
The mathematical model of any physical heterogeneous continuous complex or
simple dynamical hypersystem ought to be able to submit solutions that represent
the physical situations taken into account above sufficiently well.
The mathematical model of a physical heterogeneous continuous complex or
simple dynamical hypersystem is normally created into the set of one holor or more

Figure 7.12. Conventional input functions: the step function (a) and the sine wave (b).

7-30
 Mechatronics

scalar equations, the particular expressions of which formulate each particular or

principal effect within the physical heterogeneous continuous complex or simple
dynamical hypersystem. The solution of a differential equation has two members:
the particular integral solution and the complementary solution. The particular
integral solution is familiar with the natural response.
Of course, in engineering education it is normal to deal with these solutions using
their physical metaphors, i.e. the forced and natural responses. For linear physical
heterogeneous continuous dynamical hypersystems (i.e. physical dynamical hyper-
systems in which each principal physical commutation matrixer dynamical compo-
nent has the property that its response is directly proportional to the input or its time
derivative or time integral), the two component solutions are quite independent of
each other. The forced response has, physically and mathematically, a perfectly
equivalent structure to the input or driving function. The natural response has a
structure that is autonomous of the driving function, even if the magnitude does rely
on the input and additionally on the initial stored energy. Alas, nonlinear physical
dynamical systems do not have these simplifying properties and the solutions are
noticeably more elaborate and out of the scope of this textbook. The solution of the
linear holor equations that are normally met in physical commutation matrixers is
taken into account in chapters 7 and 8.

7.6 Physical and mathematical modelling methodologies

The principal effects within a physical heterogeneous continuous complex or simple
dynamical hypersystem are physical events that may be focused at one point in space
or single DoF, but are normally distributed over an area and, as might be
anticipated, it is more laborious to express a distributed physical event in detail
than it is to depict the physical situation at a point. On the other hand, in various
physical events it is evident that taking a limited number of DoFs (points) into
account and creating a physical model that reflects the action occurring at these
continuous DoFs (points) still offers a sufficiently accurate depiction of the relation-
ships with the remainder of the physical heterogeneous continuous dynamical
hypersystem.
This technique of lumping each physical event into an unsophisticated principal
physical situation creates a ‘lumped’ physical model in preference to a ‘distributed’
physical model. In view of the fact that distributed mathematical models, as a rule,
necessitate the solution of sets of partial differential equations and lumped physical
models involve ordinary differential equations, there is an unquestionable predi-
lection for using lumped physical models when sensible. Without doubt, there are
physical situations in which it is not evident that the circumstances are such that a
lumped mathematical model is correct. In such circumstances, it may be indis-
pensable to arrange both lumped and distributed mathematical models and to
contrast the solutions. The lumped mathematical model results may be equated
with measured sensible records from experiments. Either technique is required for
the procedure that is necessary for the formulation of sensible mathematical
models.

7-31
 Mechatronics

7.7 Signal holors of physical commutation matrixers

We have mentioned the signal holors of physical commutation matrixers in a few
words in section 7.4 and they will be described in chapter 8. They are so essential to
physical commutation matrixer analysis that we consider them once more now.
Both lumped and distributed physical and mathematical models necessitate the
application of mathematical holors to symbolise physical quantities. Despite the fact
that the field of study of this textbook is constrained to lumped physical models, the
description of physical holors is also relevant to distributed physical models.
Principally, there are two sets of physical quantities: those that flow along or through
a constrained path, and those that necessitate measurements that must be carried out
across a constrained single port or between two terminals. In connecting a measuring
instrument across a single port or between two terminals it must be located across the
whole thing that is positioned between them, and for that reason such physical
quantities are expressed as ‘energy-potential-difference holors’.
In connecting a measuring instrument to measure the stream along or through a
constrained path it is indispensable to cut the constrained path and include the
measuring instrument so that the stream is restrained to flow out through it, and for
that reason such physical quantities are expressed as ‘energy-transfer holors’. The
hypothesis of all physical quantities being physically modelled as either ‘energy
transfer’ or ‘energy potential difference’ offers a dominant medium for the
promotion of analogous physical situations. Normally, these analogies are con-
verted into something understandable only after the mathematical model (in the
appearance of a set of equilibrium holor equations) has been completed. In
particular, it is possible for a specific mechanical dynamical system, fluidic
dynamical system, thermal dynamical system and electrical dynamical system to
have equilibrium holor equations, which have the same mathematical model.
Having completed a solution for each one of them, the solutions for the others
have been successfully determined. The advantage of these analogies can be realised
at once if they can be disclosed a priori in the physical and mathematical modelling
modus operandi.
The hypothesis of the energy-transfer and energy-potential-difference holors is the
basis for the creation of the analogous physical commutation matrixer diagram,
which for practical purposes creates a correlation between the physical model and
the perfect mathematical model. This diagram has the configuration of an electrical
commutation matrixer.
The major benefit of the analogous physical commutation matrixer diagram is that
a diagram with the same conventional configuration physically models each aspect of
the physical heterogeneous continuous dynamical hypersystem. For that reason, the
procedure for creating a set of equilibrium differential equations for any physical
commutation matrixer is an unsophisticated one that is especially easy to submit an
application to, and which has the result of presenting a generalised dynamical systems
approach (systems thinking) to the analysis of mechanical, electrical, fluidic, magnetic
and thermal homogeneous continuous dynamical systems. In the following chapters,
the particulars of this generalised dynamical systems approach are worked through.

7-32
 Mechatronics

The energy-transfer and energy-potential-difference holors are recognised for

physical heterogeneous continuous dynamical hypersystems concerning single- and/
or multi-port MIMO electrical commutation matrixers, translation mechanical
commutation matrixers, rotation mechanical commutation matrixers, fluid flow
commutation matrixers and thermal problems’ commutation matrixers (chapter 8).
As soon as the procedure for realising the analogous physical commutation
matrixers has been considered (chapter 9), some particularly important physical
situations are taken into account.
Thus, the methodical procedures (chapters 10 to 13) are elaborated with
examples considered from any or all these variants of physical heterogeneous
continuous dynamical hypersystems so that by the conclusion of the textbook the
reader should have the same expertise and self-sufficiency when dealing with
problems of physical and mathematical modelling and the theory of holor analysis
for any of the physical heterogeneous continuous dynamical hypersystems included
in this textbook.

References
Fijalkowski B 1967 Zastosowanie rachunku holorowego w teorii pradow przemiennych
(Application of a holor calculus in the theory of alternating currents) Przeglad elektrotech-
niczny (Electrotechnics Review), Rok XLIII, Zeszyt 2, Warszawa (In Polish)
Fijalkowski B 1973 Kierunki rozwoju napedu elektrycznego lokomotyw (Development trends of
the electric drive of locomotives). Materiały konferencyjne—I Krajowa Konferencja—Pojazdy
Szynowe. Instytut Pojazdow Szynowych Politechniki Krakowskiej, Krakow, pp. 109–128
(In Polish)
Fijalkowski B 1987 Modele matematyczne wybranych lotniczych i motoryzacyjnych mechanoelek-
tro-termicznych dyskretnych nadsystemów dynamicznych (Mathematical Models of Selected
Aviation and Automotive Continuous Dynamical Hyperystems), Monografia 53 Politechnika
Krakowska imienia Tadeusza Kościuszki, Krakow, 276 p (In Polish)
Fijalkowski B T 1982 Trigistor frequency changer-controlled toothed gearless propulsion systems
for electric and hybrid vehicles. Proc. Europe’s International Conference on Electric Road
Vehicle Systems DRIVE ELECTRIC (Amsterdam, 25–28 October 1982) pp 554–567
Fijalkowski B T 1985 On the new concept MMD electrical machines with integral macro-
electronic commutators—Development for the future Proc. First European Conf. on POWER
ELECTRONICS AND APPLICATIONS (Brussels, 16–18 October 1985) vol 2 pp 3.377–
3.384
Fijalkowski B T 1995 Emerging and future intelligent aviation and automotive applications of
MIMO ASIM static-commutators and ASIC microcontrollers. Proc. of the NATO Advanced
Workshop on Future Trends in Microelectronics: Reflections on the Road to Nanotechnology
(Ile de Bendor, France 17–21 July 1995). Also a chapter in the book edited by S Luryu, J Xu
and A Zaslavsky 1996 Future Trends in Microelectronics—Reflections on the Road to
Nanotechnology (NATO ASI Series, Series E: Applied Sciences) vol 323 (Dordrecht: Kluwer)
Fijalkowski B T 2010 Automotive Mechatronics: Operational and Practical Issues.— Vol. 1
(International Series on Intelligent Control, and Automation Science and Engineering—
Vol 47) (Berlin: Springer)

7-33
 Mechatronics

Fijalkowski B T 2011 Automotive Mechatronics: Operational and Practical Issues.— Vol 1.

(International Series on Intelligent Control, and Automation Science and Engineering—
Vol 52) (Berlin: Springer)
Laplace P-S 1820 Theorie Analytique des Probabilités (Paris: Mme Ve Courcier)
Moon P H and Spencer D E 1986 Theory of holors. A generalization of Tensors (Cambridge:
Cambridge University Press)
Senturia S D 1998 CAD challenges for microsensors, micro-actuators, and microsystems Proc.
IEEE 86 1611–26

7-34
 IOP Publishing

Mechatronics
Dynamical systems approach and theory of holors
B T Fijalkowski

Chapter 8
Physical commutation matrixer dynamical
components

A mathematician is a person who can find analogies between theorems; a better

mathematician is one who can see analogies between proofs and the best
mathematician can notice analogies between theories.
Stefan Banach
One can imagine that the ultimate mathematician is one who can see analogies
between analogies.
Stefan Banach

8.1 Introduction
In this textbook, we consider electrical, mechanical, fluidic and thermal physical
commutation matrixer dynamical components. A mathematical model of any
physical situation requires the use of mathematical variables, for instance, holors,
to represent certain physical quantities associated with its behaviour. There are
only two principal types of such quantities (chapter 7): those which flow along
or through a constrained path (such as water flowing through a pipe), and those
which require, whether explicitly or implicitly, measurements that must be made
between two points (such as the distance between the points), e.g. generalised
coordinates.
When one connects an instrument between two points, it must be placed across
everything that lies between the points, and such physical quantities may reasonably
be described by energy-potential-difference holors. In measuring the flow along a
path, it is necessary to cut the path and insert the instrument so that the flow is
constrained to pass through it. Such physical quantities are therefore described
by energy-transfer holors. All the physical heterogeneous continuous dynamical
hypersystems (consisting of electrical, mechanical, fluidic, magnetic and thermal

doi:10.1088/978-0-7503-1350-6ch8 8-1 ª IOP Publishing Ltd 2016

 Mechatronics

homogeneous continuous dynamical systems) will have an energy-potential-difference

holor and an energy-transfer holor.
The holors are different for each physical heterogeneous continuous dynamical
hypersystem. The classification of all physical quantities as either energy-potential-
difference holors or energy-transfer holors is the basis for the recognition of analogous
physical situations in different physical heterogeneous continuous dynamical hyper-
systems. This grouping is especially useful if it is done in a manner that ensures that
the method of computing the energy for one type of physical heterogeneous
continuous dynamical hypersystem, for instance, electrical continuous dynamical
systems, is the same as that used for another type of physical heterogeneous
continuous dynamical hypersystem, for instance, fluidic dynamical systems. This,
however, is not a necessary condition. The essential feature when choosing holors is
that the resulting holor equations must have exactly the same mathematical model for
all physical heterogeneous continuous dynamical hypersystems.
This book considers electrical, magnetic, mechanical, fluidic, magnetic and
thermal homogeneous continuous dynamical systems. In the first four of these,
the choice of the signal holors is made so that the univalent tensor product of the
energy-transfer and energy-potential-difference holors represents a univalent power.
Thermal continuous dynamical systems are different, however, at the
energy-transfer holor itself, the heat flow rate is power.
In choosing the energy-transfer and energy-potential-difference holors it must
be kept in mind that the objective in the end is to produce a physical
commutation matrixer representation for a physical heterogeneous continuous
dynamical hypersystem. This simplifies the problem of describing equilibrium
conditions. However, before considering equilibrium in general, it is necessary to
identify certain elementary physical situations and derive the elemental holor
equations that describe their performance. These correspond to idealised physical
situations in which only one physical phenomenon is considered to be present.
One physically models these phenomena with two terminal physical commutation
matrixer dynamical components. Figure 8.1 is a schematic representation of the
principal single-port or two-terminal physical commutation matrixer dynamical
component.

Figure 8.1. Principal single-port or two-terminal physical commutation matrixer dynamical component:
(a) commutation node; (b) component.

8-2
 Mechatronics

We designate the point at which the energy transfer enters as terminal A and the
point at which the energy transfer leaves as terminal a. There is an energy-potential
difference associated with each terminal, which is a measure of the energy in the
physical commutation matrixer at that location. The energy-potential-difference
holor is the potential difference between terminals A and a. The energy-transfer
holor, representing a transmitted physical quantity, passes through the physical
dynamical component without change. There are only three principal elemental
relations and they are expressed mathematically in holor algebra as:
FiP = DijPVPj , (8.1a )

FiD = ωKijDVDj, (8.1b)

1 I j
FiI = Pij VI , (8.1c )
ω
where DijP , KijD and PijI are proportionality constants that depend on the physical
matrixer’s geometrical configuration and material properties.
It should be emphasised that equation (8.1) is merely a general form for physical
commutation matrixer dynamical components. The principal physical laws of
electrics, mechanics, fluid mechanics and heat transfer can be placed in this format.
It is these laws that determine the appropriate signal holors and the specific values of
the bivalent holors DijP , KijD and PijI . In the following sections, we introduce the
principal components in mechanical, electrical, fluidic, magnetic and thermal
homogeneous continuous dynamical systems.

8.2 Electrical commutation matrixer dynamical components

Electronics (Greek Εληκτρική) is the science and technology relating to or associated
with electricity, as well as current or voltage for transmitting electrical energy and
signals. It has been a subject of physics since at least the early 17th century. The first
electrical engineer was probably the British physician, physicist and natural
philosopher, William Gilbert (1544–1605), who designed the versorium (Latin
‘turn around’), which was the first simple electroscope, the first instrument that
could detect the presence of static electrically charged objects. He was also the first
to describe the evident dissimilarity between magnetism and static electricity and is
recognised as having originated the term ‘electricity’. More to the point, objects that
develop electrification (the act of electrifying, or the state of being charged with
electricity) through friction or rubbing, as Gilbert, the originator of the term, applied
it, indicate ‘dielectrics’. He did not know that, if insulated, any object is one of his
‘electrics’. A piece of copper held by a glass handle becomes electrified by friction.
In 1775, a professor at the Royal School in Como, Alessandro Volta (1745–1827),
devised the electrophorus, a device that produces a static electrical charge, and by
1800 he had developed the voltaic pile, a forerunner of the chemo-electrical/electro-
chemical (ChE/ECh) storage battery. However, it was not until the 19th century that
research into the subject started to intensify. Notable developments from that period

8-3
 Mechatronics

include the work of the Bavarian (German) physicist and mathematician, Georg
Ohm (1789–1854), who in 1827 quantified the relationship between electric current
and voltage potential difference in a conductor; the work of the British scientist,
Michael Faraday (1791–1867), who discovered electromagnetic induction in 1831;
and the work of the British mathematical physicist, James Clerk Maxwell (1831–79),
who in 1873 published a unified theory of electricity and magnetism in his treatise
Electricity and Magnetism (Maxwell 1873). During these years, the study of
electricity was largely considered to be a subfield of physics. It was not until the
late 19th century that universities started to offer degrees in electrical engineering.
Darmstadt University of Technology founded the world’s first chair and faculty of
electrical engineering in 1882. In 1883 Darmstadt University of Technology and
Cornell University introduced the world’s first courses of study in electrical
engineering, and in 1885 University College London founded the UK’s first chair
of electrical engineering. The University of Missouri subsequently established the
USA’s first department of electrical engineering in 1886.
The term electronics is used to cover electrical engineering, including the
generation, transmission, control and use of all forms of electrical energy. The
term electronic engineering is used to include the expanding fields of electronic
communications (including computer networks), computers (both hardware and
software) and electronic components. These components include microcomputer
chips and, increasingly, the optical devices now being used for many applications. In
addition, the field of mechatronical control engineering is on the electrical/electronic
border and, with its use of computer systems, has a general base. Manufacturing
engineering—involving as it does, computer techniques, mechatronical control and
electrical engineering—is another important and elementary area within the field of
electrics and electronics. Much of the very advanced equipment found in the
workplace or home, such as robots and machine tools, domestic appliances,
entertainment systems, personal computers and data processing equipment, is
made possible by electrical and electronic engineering. This extensive, expanding
and exciting subject has interfaces with physics (in the component field), computer
science (in software engineering) and mechanics (in manufacture and control).

8.2.1 Electrical-signal holors

The generalised concept of the energy-potential-difference holor, Fi , and the energy-
transfer holor, V j , as the energy transporting mechanism finds a ready parallel in the
voltage holor, V, j
i and the current holor, I , of electrical homogeneous continuous
dynamical systems. The conceptual similarity of the holor sets leads naturally to the
assignment:
the voltage holor, Vi , is analogous to the energy-potential-difference holor, Fi ;
the current holor, I j , is analogous to the energy-transfer holor, V j .
This assignment is compatible with the notions of the energy-potential-difference
holor, Fi , and the energy-transfer holor, V j , and is used universally. Although the
electrical charge holor, Q j , is perhaps the most important electrical quantity from a

8-4
 Mechatronics

theoretical viewpoint, the most useful holors for electrical homogeneous continuous
dynamical systems are the current holor, I j , and the voltage holor, Vi . The current
holor, I j , is the rate of the flow of the electrical charge holor, Q j , and is clearly
identified as an energy-transfer holor, V j .
The unit of the electrical charge holor, Q j , is the coulomb (C), and the electrical
current holor, I j is the rate of the flow of this charge in coulombs per second (C s−1)
or amperes (A).
Thus, in mathematical terms

I j = ωQ j .

The voltage-potential-difference holor, EiA − EiN , also termed the electrical-potential-

difference holor, is the energy-potential-difference holor. The electrical-potential-
difference holor is the work done per unit charge in moving the charge between two
points at the potentials EiA and EiN . The fact that two points are required in the
determination of a voltage-potential-difference holor, or a voltage-potential drop
holor, clearly identifies it as the energy-potential-difference holor.
The unit for electrical-potential-difference holor is the volt (V). This is the
electrical-potential difference holor when the work involved in moving one coulomb
is one joule (J).
Thus the energy holor, or work, ; , done in moving a charge holor, Q j , between two
points from electrical-potential-difference holors EiA to EiN is ; ϰ = (EiA − EiN ) ⊗ Q j .
The rate at which this work holor is expended is the power holor, P ϰ . The power
holor is:

( )
P ϰ = ω ; ϰ = Ei A − Ei N ⊗ I j . (8.2)

Note that the power holor is the univalent tensor product of the energy-potential-
difference holor and the energy-transfer holor. Since the rate of the change of work
holor is defined as the power holor, this is the well-known general expression for the
power holor, P ϰ, in an electrical commutation matrixer.
The unit for the power holor is the watt (W), which is one joule per second (J s−1).
It also follows that the determination of the energy holor or work holor, ; ϰ, is
simply a matter of integrating equation (8.2):

P = ⎡⎣ EiA − EiN ⊗ I j ⎤⎦
1 ϰ 1
<ϰ =
ω ω
( ) (8.3)

Three elementary physical situations are found in electrical commutation matrixers

(equation (8.1)) that do not involve an energy source being put into the electrical
homogeneous continuous dynamical system. These are commonly known as the
ideal resistor, the ideal inductor and the ideal capacitor.
Although real electrical dynamical components only approach the ideal, they can
often be physically modelled as a combination of two or more of these ideal
electrical dynamical components. All the electrical dynamical components must
originate from physically realisable phenomena and obey physical laws.

8-5
 Mechatronics

8.2.2 Electrical-resistance holors

An electrical device in which the terminal voltage holor, Vi , is statically related to the
current holor, I j , flowing through the device is a dissipater of electrical energy. If the
device is free from storage and source effects, it is said to have a pure electrical
resistance.
The general constitutive relation for a resistance is
Vi = -(I j ). (8.4)

The linear form of equation (8.4) was first discovered by Georg Ohm and bears his
name. Ohm’s law states that
Vi = Rij I j , (8.5)

where Rij is termed the resistance holor of the electrical dissipater.

For an electrical homogeneous continuous dynamical system of length l and
uniform cross-sectional area S, the resistance is given by
R = ρl / S , (8.6)
where ρ is the resistivity of the device material.
The total power dissipated by an electrical resistor is the univalent tensor product
of the terminal voltage holor, Vi , and current holor, I j , which is the sum of the
content, G, and the co-content, J, of the electrical resistor
P ϰ = Vi ⊗ I j . (8.7)
For a linear device, the content, G, and the co-content, J, of the electrical resistor are
equal and are given by
1
G=J= R(i ∧ 2). (8.8)
2
The total power dissipated is
1
Pp = R(i ∧ 2) = (v ∧ 2)/ R . (8.9)
2
From experience with many metallic electrical commutation matrixer parts, one
observes that the voltage-potential-difference holor, EmA − EmN , established in such
an electrical commutation matrixer is directly proportional to the current holor, I j .
This observation may be expressed in mathematical form as:

EcA EcN Ic .
EiA − EiN = Rij I j = − = R 0 (8.10)
EsA EsN 0 R Is

This relation is known as the generalised Ohm equation, or Ohm’s law.

Note that equation (8.10) is identical in form to the general elemental relation
given in equation (8.1a), with Rij = DijP .

8-6
 Mechatronics

Figure 8.2. Characteristics of ideal resistance.

The physical quantity symbolised by R is termed ‘electrical resistance’ and is a

property of the material and geometry of the electrical commutation matrixer part.
The ideal electrical resistor has the voltage-potential-difference holor/current-
holor characteristic shown in figure 8.2.
In this case, the voltage-potential-difference holor is directly proportional to the
current holor, as indicated by the line of the constant slope. The slope of this line is
the resistance R. The unit of resistance is the ohm (Ω). There are occasions, however,
when the reciprocal, G, is more convenient. This is the conductance for which the
unit is the siemens (S) or mhos (℧ ). Note that the ideal resistance dynamical
component is bilateral or has two senses of direction. If potential A is greater than
potential N, then the energy-potential-difference holor, EiA − EiN , is positive and
this defines the positive sense of direction of the current holor. When the potential N
is greater than the potential A, the energy-potential-difference holor is negative and
the current holor reverses.
In a bilateral resistive dynamical component, a given positive or negative voltage-
potential-difference holor causes a current holor of the same magnitude but in the
opposite direction. All objects have electrical resistance. However, in electrical
commutation matrixers that are used to produce some desired effect on the signal
holors, one is dealing with an electrical component that is specially made for the
purpose. In general, electrical resistance is obtained from the resistivity and
geometry of the wire material.
The resistivity of a wire is expressed as:
l
R = ρ , [Ω], (8.11)
S
where ρ is the resisitivity (Ω m) of the wire material, l is the wire length (m) and S is
the wire cross-sectional area (m2).
The energy dissipated in the resistance can be determined by rearranging equation
(8.2) in the form
1 ϰ
; ϰ = ;⁎ϰ = P .
ω

8-7
 Mechatronics

When the voltage-potential-difference holor and the current holor are not changing
with time, the energy is
⎡(E A∧ 2) − (E N ∧ 2)⎤
⎣ c c ⎦ (8.12)
; = ;⁎ = R(I c ∧ 2)t = t.
R
Note that the energy increases with time. The resistance is, therefore, an energy-
dissipation dynamical component.

8.2.3 Electrical-inductance holor

In the 19th century Joseph Henry and Michael Faraday discovered that the
electricity flowing in an electrical homogeneous continuous dynamical system
displayed properties that are analogous to the momentum in a mechanical
continuous dynamical system. The electricity flowing in a long electrical homoge-
neous continuous dynamical system displays what Maxwell termed an ‘electro-
kinetic momentum’ holor, which is proportional to the current holor, I j , flowing,
and constant depending on the physical arrangement of the electrical homogeneous
continuous dynamical system.
The quantity of the electrokinetic momentum holor is in time known as the
magnetic-flux-linkage holor, λi , and is the total magnetic flux holor, Φi , linked by
the electrical homogeneous continuous dynamical system.
Similarly, the physical constant, which together with the current holor, Lij ,
determines the flux-linkage holor, λi , is the inductance holor, Lij ,
λi = Lij I j . (8.13)

The magnetic flux holor, which is linked by the electrical homogeneous continuous
dynamical system, determines the voltage holor, Vi , developed across the electrical
inductor via Faraday’s law:
Vi = ωλi . (8.14)
Equation (8.14) is the linear form of the inductive constitutive relation; in general,
the relation is nonlinear and may be multi-valued:
λi = -(I j ) . (8.15)

For an electrical inductor composed of a coil of N turns wound on a homogeneous

core of permeability μ and of uniform cross-sectional area S, the inductance is given
by
μS ∧
L= (N 2), (8.16)
l
where S is the cross-sectional area of the magnetic homogeneous continuous
dynamical system and l is the length of the magnetic homogeneous continuous
dynamical system.

8-8
 Mechatronics

For non-magnetic core materials the magnetic permeability, μ, is constant,

However, for ferrous-cored inductors the permeability, μ, is a function of the
current holor, I j , such that the inductive constitutive relation is nonlinear.
The energy stored in an inductive field is given by the integral
λ
;= ∫0 idλ . (8.17)

The inductive co-energy is given by the complementary integral

i
;⁎ = ∫0 λ di . (8.18)

If the electrical inductor is linear then the energy, ; , and the co-energy, ;⁎, are equal
and are given by
1 ∧ 1
; = ;⁎ = (λ 2) = L(i ∧ 2). (8.19)
2L 2
The most common form of inductor is a coil of wire wrapped around a core. The
core contains a high level of the magnetic-flux holor, Φi , due to the current holor, I j ,
in the conducting wire coil. When the wire coil contains more than one loop, each
loop contributes to the magnetic-flux-linkage holor, λiAN , between the two terminals
of the coil inductor. When the physical arrangement is such that the magnetic-flux-
linkage holor, λ mAN , is a single-valued function of the current holor, I j , the dynamical
component is termed a pure inductor. This is expressed as:
λ iAN = Lij I j , (8.20)

where λiAN is the magnetical-flux-linkage holor in webers (We) and Lij is the
inductance holor in henries (H).
Thus, the pure or ideal inductor has the magnetic-flux-linkage holor, λi /current
holor, I j , characteristics shown in figure 8.3.
Faraday’s law states that the voltage-potential-difference holor induced in coils
by a changing magnetic field is

Figure 8.3. Characteristics of ideal inductance.

8-9
 Mechatronics

EiA − EiN = ωλ iAa . (8.21)

The combination of equations (8.20) and (8.21) yields a relation between the signal
holors that is:
Ei A − Ei N =

EcA EcN 0 −ωL Ic .

ωλ iAa = ωLij I j = − = (8.22)
E sA E sN ωL 0 Is

Equation (8.22) has the same form as is given in the general elemental relation in
equation (8.1b). In this case, KijD = Lij . From the ideal characteristics shown in figure
8.3, one may observe that the ideal inductor is also a bilateral electrical dynamical
component.
The magnitude of inductance for a wire wound over a cylindrical core is:

4π (10)−7μS (N )2
L= H, (8.23)
l

where μ is the relative permeability of the core material, S is the area of the core
(m2), l is the length of the core (m) and N is the number of coils. For an air core the
relative permeability μ = 1.
When an iron rod is used for the core, the relative permeability may be as high as
(10)4, with the actual value depending on the magnetic-flux density.
In actual electrical dynamical components, it is not possible to achieve the ideal
characteristics given for the resistances and inductances. In both cases, the
dynamical component consists of a length of wire. Thus, the actual inductance
always has some resistance and the actual resistance has some inductance. However,
we assume throughout this book that the dynamical components are ideal and
represent only one phenomenon.
The energy into an inductor is from equations (8.2) and (8.22),
i 1
;= L ∫i 0
i′di′ =
2
L[(i ∧ 2) − (i 0∧ 2)], (8.24)

where i 0 is the initial current.

For the inductor the energy is not directly a function of time. This is because the
inductor stores energy rather than dissipating it. The energy stored by the inductor is
due to the flow of current and the expression given in equation (8.24) is valid whether
the current is constant or not.
The energy stored is not directly dependent on the voltage-potential-difference
holor. Indeed, equation (8.24) applies even when the voltage-potential-difference
holor is zero. The inductor is in a class of energy-storage dynamical components
known as D-type (for energy-potential-difference holor) storage.

8-10
 Mechatronics

8.2.4 Electrical-capacitance holor

The electrical inductor stores energy in a magnetic field. The complementary
electrical dynamical component, the electrical capacitor, stores energy in an
electrical field. Whenever two electrical conductors are at different potentials there
is storage of charge upon them. The amount of the electrical-charge holor, Q j , that is
accumulated in this way is determined by the voltage-potential-difference holor, Vi ,
across the conductors and the physical properties of the conductors and the medium
(dielectric) that separates them.
The constitutive relationship of an electric capacitor therefore has the general
form:
Q j = - (Vi ). (8.25)

In the linear case this becomes

Q j = C jiVi , (8.26)

where C ji is termed the capacitance holor of the electrical capacitor.

For a pair of parallel plates separated by a distance, d, and with a homogeneous
filling of permittivity, ε , the capacitance is given by
εS
C= , (8.27)
d
where S is the effective cross-sectional area of the plates.
For almost all the dielectric materials used in electrical capacitors, the permit-
tivity, ε , is a constant such that the capacitive constitutive relation is a constant. The
electrical-charge holor, Q j , is related to the current holor, I j , flowing in the
capacitor’s electrical homogeneous continuous dynamical system by the dynamical
relationship:
I j = ωQ j . (8.28)

The electrical energy stored in an electrical capacitor is given by an integral

q
<= ∫0 v dq . (8.29)

The electrical co-energy is defined in a complementary integral

v
<⁎ = ∫0 q dv . (8.30)

If the electrical capacitor is linear, the capacitive energy, U, and the co-energy, U ⁎,
are equal and are given by
1 1 ∧
< = <⁎ = C (v ∧ 2) = (q 2). (8.31)
2 2C

8-11
 Mechatronics

Figure 8.4. Characteristics of ideal capacitance.

Two conductors separated from each other by a dielectric (an insulating material)
form a capacitor. In the capacitor, an electrical field is established between the
conductors. When the physical arrangement is such that the electrical-charge holor,
Q j , on the conductors is a single-valued function of the voltage-potential-difference
holor, EiA − EiN , the dynamical component is termed a pure capacitor. This is
expressed as:

(
Q j = C ji EiA − EiN , ) (8.32)

where Q j is the electrical-charge holor in coulombs and C ji is the capacitance-

admittance holor in farads (F).
Thus, the pure or ideal capacitor has the charge-holor/voltage-potential-differ-
ence holor characteristics shown in figure 8.4.
In terms of the signal holors, one may replace the electrical charge-holor, Q j , in
equation (8.32) by the integral of the current-holor, I j , so that:
1
0
δ ki EcA EcN ωC Ic .
Ei A − Ei N = Ij = − = (8.33)
ωCjk E sA E sN 1 Is
− 0
ωC
For this type of electrical dynamical component, the governing equation (8.33) has
the same mathematical form as the general elemental relation of equation (8.1c),
δi
with P jkI = kij . The ideal capacitor is a bilateral dynamical component in the same
C
way as the ideal resistor and ideal inductor.
For a parallel plate capacitor, the magnitude of the capacitance is:
εS
C= F, (8.34)
d
where ε is the permeability of the dielectric in farads per meter (F m−1), S is the
cross-sectional area (m2) of the plates and d is the thickness (m) of the dielectric.

8-12
 Mechatronics

The energy associated with a capacitance may be determined from equations (8.2)
and (8.33) as:
Δe 1 ⎡
< ∫Δe e′de′ = C ⎣(Δe ∧ 2) − (Δe0∧ 2)⎤⎦ , (8.35)
0 2

where Δe is the voltage-potential difference which equals e1 − e2 and Δe0 is the initial
voltage difference.
Since the capacitor stores energy, it is also an energy-storage dynamical
component. In this case, energy is stored due to the presence of a voltage difference.
The expression for energy given in equation (8.35) is valid whether the voltage-
potential difference is constant or not.
The energy stored is not directly dependent on the current. Indeed, equation
(8.35) applies even when the current is zero.
The electrical capacitor is in the class of energy storage devices known as I-type
(for energy-potential-difference holor) storage.
The energy stored by an inductance (equation (8.24)) and by a capacitance
(equation (8.33)) may be formulated in a similar way for the analogous storage
dynamical components in mechanical and fluidic homogeneous continuous dynam-
ical systems.
The normal symbols for the three electrical commutation matrixer dynamical
components are shown in figures 8.2–8.4. The arrow associated with the current
holor shows a positive sense of direction for conventional current-holor flow
through the dynamical component. The terminal at which the positive current
holor enters is positive for each of the dynamical components and polarity
indicators must be set up as in the electrical commutation matrixers if the positive
signs implied in the elemental holor equations are to apply. At this stage, it can be
noted that each of the three electrical dynamical components is significantly
different.
One of them, the resistor, dissipates energy, which cannot be recovered by the
electrical commutation matrixer. The others store energy, which can be recovered
when the electrical commutation matrixer conditions are available. However, in one
case, the capacitor (I-type storage), the energy is stored by virtue of there being a
voltage-potential-difference holor between the two terminals irrespective of whether
there is a current holor flow. In the other case, the inductor (D-type storage), the
opposite statements holds: the energy is stored by virtue of there being a current
holor flow through the dynamical component irrespective of whether there is a
voltage-potential-difference holor across it.
In this chapter, the electrical commutation matrixer physical models of the
storage dynamical components do not include initial storage. That is, i0, for
the inductor, and Δe, for the capacitor, have been assumed to be zero. The
analogous mechanical, fluidic and thermal storage dynamical components will
also be introduced without initial storage. The physical commutation matrixer
models for storage dynamical components with initial storage have been
considered in chapter 7.

8-13
 Mechatronics

8.3 Magnetic commutation matrixer dynamical components

Magnetics is the science and technology relating to or associated with a magnetic
medium, as well as a magnetic flow (magnetic movement of electrons) for the
transmission of magnetic energy and signals. It concerns the production and trans-
ferral of magnetic energy by means of various physical phenomena.
One of the great discoveries in the history of physics was made by the British
physicist, James Clerk Maxwell (1831–79), in the 19th century. Maxwell found that
the two major forms of energy known as electricity and magnetism are not really
different from each other, but are instead closely associated. That is, every electrical
current has a magnetic field associated with it and every changing magnetic field
creates its own electrical current. Maxwell’s equations state that magnetic energy
and electrical energy are related to one another by nature. Magnets produce
potential energy through magnet motion and additionally through the magnetic
field emitted. The magnetic moment of a magnet is a quantity that determines the
force that the magnet can exert on electrical currents and the torque that a magnetic
field will exert on it. A loop of electrical current, a bar magnet, an electron, a
molecule and a planet all have magnetic moments. Both the magnetic moment and
magnetic field may be considered to be vectors with a magnitude and sense of
direction. The sense of direction of the magnetic moment points from the south (S) to
the north (N) pole of a magnet. The magnetic field produced by a magnet is also
proportional to its magnetic moment well. More precisely, the term ‘magnetic
moment’ normally refers to a magnetic homogeneous continuous dynamical
system’s magnetic dipole moment, which produces the first term in the multipole
expansion of a general magnetic field.
The dipole dynamical component of an object’s magnetic field is symmetric about
the direction of its magnetic dipole moment, and decreases as the inverse cube of the
distance from the object. The magnetic dipole moment μ can be considered to be a
vector quantity with a direction perpendicular to the electrical-current loop in the
right-hand-rule direction (see figure 8.5).
Considering magnetic torque as a vector quantity, this can be written as the
vector product by
τ = μ × B. (8.36)

Figure 8.5. Moment μ of a planar current with a magnitude I enclosing an area S.

8-14
 Mechatronics

As seen in figure 8.5, which shows the geometry of an electrical-current loop, this
magnetic torque τ tends to line up the magnetic dipole moment μ with the external
magnetical field B, so this represents its lowest energy configuration.
Since this magnetic torque τ acts perpendicularly to the magnetic moment μ, it
can cause the magnetic moment μ to precess around the magnetical field B at a
characteristic frequency called the ‘Larmor frequency’.
A magnetic dipole moment μ in a magnetic field B will have potential energy
<(θ ) that depends upon its orientation with respect to the external magnetic field B.
Since magnetic-energy sources are inherently dipole sources that can be visualised as
an electrical current I and area S, the energy <(θ ) is usually expressed in terms of the
magnetic dipole moment μ:
<(θ ) = −μB, (8.37)
where μ = IS (or μ = NIS for N electrical-current loops).
The energy <(θ ) is expressed as a scalar product and implies that the energy is
lowest when the magnetic dipole moment μ is aligned with the magnetic field B. The
difference in energy Δ< between aligned and anti-aligned is
Δ< = 2μB. (8.38)

These relationships for a finite electrical-current loop extend to the magnetic dipoles
of electron orbits and the intrinsic magnetic moment μ with electron spin. Nuclear
magnetic moments are also important.
The potential energy < of a permanent magnet of magnetic moment μ in a
magnetic field B is defined as the action of magnetic force (actually of magnetic
torque) on the realignment of the vector of the magnetic dipole moment μ, and is
equal to:
< = − μB, (8.39)

while the magnetic energy stored in an inductor (of inductance L) when electrical
current I is passing via it is
1
<= L(I ∧ 2). (8.40)
2
This second expression forms the basis for superconducting magnetic energy storage
(SMES).
Magnetic energy < is also stored in an external magnetic field B. The magnetic
energy < per unit volume in a sphere of space of permeability μ containing magnetic
field B is:
1 (B∧ 2)
<= . (8.41)
2 μ

In a case where the external magnetic field is non-uniform, there will be a force
(proportional to the magnetic field gradient) acting on the magnetic moment itself.

8-15
 Mechatronics

There has been some discussion regarding how to calculate the force acting on a
magnetic dipole. There are two expressions for the force acting on a magnetic dipole,
depending on whether the physical model used for the dipole is a current loop or two
monopoles (analogous to the electrical dipole).

8.3.1 Magnetomotive-force and magnetic-flux holors

In almost all cases, the properties of magnetic homogenous continuous dynamical
systems can be physically and mathematically modelled indirectly via their reflected
electrical continuous dynamical system properties. Situations arise, however, where
the magnetic continuous dynamical system ought to be considered in its own right.
The extensive variable in a magnetic homogeneous continuous dynamical system
is the magnetomotive-force holor M i. The intensive variable is associated with the
total magnetic flux holor Φ j.
Hence, according to the energy-potential difference holor Fi and the energy-
transfer holor V j analogy: the magnetomotive-force holor M i is analogous to the
energy-potential difference holor Fi, and the magnetomotive-flux holor Φ j is
analogous to the energy-transfer holor V j.
The storage of energy in a magnetic homogeneous continuous dynamical system
is usually physically and mathematically modelled by the reflected electrical
characteristics of the inductance holor Lij.
In magnetic terms, the property of the inductance holor Lij is a flow storage
phenomenon in that it is associated with an accumulation of the magnetomotive-flux
holor Φ j. The complementary form of magnetic energy storage is not normally
discernible.

8.3.2 Magnetic-reluctance holor

The magnetising force at any point in a magnetic homogeneous continuous
dynamical system is the magnetomotive force M per unit length required at that
point to maintain the magnetomotive flux Φ. Lines of magnetomotive flux Φ form
closed loops and if H represents the magnetising force then the total magnetomotive
force M is the integral of H along the centre flux line, i.e.

M= ∮ H dl . (8.42)

Now consider a closed-iron magnetic homogeneous continuous dynamical system

with a cross-sectional area of S (m2) and a mean length l (m).
The total magnetomotive flux is Φ and is the product of the magnetomotive-flux
density B and the cross-sectional area S.
Hence
Φ BS
= . (8.43)
M Hl
However, the ratio B/H defines the magnetic permeability μ of the medium
supporting the magnetomotive flux Φ, giving

8-16
 Mechatronics

Mc R 0 Φc ,
Mi = Rij Φ j = = (8.44)
Ms 0 R Φs

where R = 1/μS is the reluctance of the magnetic homogeneous continuous

dynamical system and is the magnetic form of energy dissipation.
Magnetic reluctance, or ‘magnetic resistance’, is a concept used in the analysis of
magnetic homogeneous continuous dynamical systems. It is analogous to resistance
in an electrical homogeneous continuous dynamical system, but rather than
dissipating electrical energy, it stores magnetic energy.
As an electrical field causes an electrical current to follow the path of least
resistance, so a magnetic field causes magnetic flux to follow the path of least
magnetic reluctance. It is a scalar, extensive quantity, akin to electrical resistance.
The units for magnetic reluctance are inverse henries (H−1). The term was coined in
1888 by Oliver Heaviside.
The notion of magnetic resistance was first mentioned by James Joule and the
term ‘magnetomotive force’ was first used by Bosanquet.
The idea for a magnetic flux law, similar to Ohm’s law for closed electrical
homogeneous continuous dynamical systems, is attributed to H Rowland.
Reluctance is usually represented by a cursive capital R.
Equation (8.44) is the constitutive relation of a magnetic dissipater. Recalling that
the permeability μ is a function of the magnetomotive-flux level, magnetic dissipaters
are usually nonlinear in nature with the constitutive relation
M = - (Φ). (8.45)
Figure 8.6 shows a typical constitutive relationship for a magnetic dissipater,
together with the magnetic commutation matrixer adopted.

8.3.3 Magnetic-energy-source holors

Fundamentally, contributions to any magnetic homogeneous continuous dynamical
system’s magnetic moment may come from magnetic-energy sources of two forms:
(a) the motion of electrical charges, such as electrical currents, and (b) the intrinsic
magnetism of elementary particles, such as the electron.

Figure 8.6. Magnetic commutation matrixer and constitutive relation for a magnetic dissipater.

8-17
 Mechatronics

Contributions due to the magnetic-energy sources of the first form can be

calculated by knowing the distribution of all the electrical currents (or, alternatively,
all the electrical charges and their velocities) inside the magnetic homogeneous
continuous dynamical system. On the other hand, the magnitude of each elementary
particle’s intrinsic magnetic moment is a fixed number, often measured experimen-
tally to a great precision. The sense of direction of the magnetic moment of any
elementary particle is entirely determined by the sense of direction of its spin (the
minus sign in front of the value above indicates that any electron’s magnetic moment
is antiparallel to its spin).
The net magnetic moment of any magnetic homogeneous continuous dynamical
system is a vector sum of contributions from one or both forms of the sources. For
example, the magnetic moment of an atom of hydrogen 1H (the lightest hydrogen
isotope, consisting of a proton and an electron) is a vector sum of the following
contributions:
• the intrinsic moment of the electron;
• the orbital motion of the electron around the proton;
• the intrinsic moment of the proton.

Similarly, the magnetic moment of a bar magnet is the sum of the intrinsic and
orbital magnetic moments of the unpaired electrons of the magnet’s material. The
most basic form of magnetic-energy-source holor is the permanent magnet, which is
a source of the magnetomotive-force holor Mi . Every magnet contains two distinct
regions, one known as the N pole and the other as the S pole. As with electrical
charges, unlike poles attract each other and like poles repel each other.
A study of magnets leads to the introduction of a fascinating hypothesis in energy,
the concept of an energy field. An energy field is a sphere in space in which a
magnetic, electrical or some other form of force can be considered. The concept of
an energy field was some time ago a very intricate one for physicists to understand
and accept. How could one object exert a force on another object if the two were not
in contact with each other? In time, it became understandable that forces can act at a
distance from each other. Electrical charges and magnetic poles seem to exert their
forces throughout an energy field along passageways identified as lines of force.
Most magnetic homogeneous continuous dynamical systems display some
element of permanent magnetism as either an explicit source or a residual magnet-
ism in the ferrous medium that supports the magnetomotive-flux holor Φ j .
Another form of magnetic energy source is an electrical inductor (coil) of N turns
carrying an electrical current holor I j coupled to the magnetical homogeneous
continuous dynamical system such that the magnetomotive-force holor Mi of the
equivalent magnetic energy-source holor is
Mc Ic .
Mi = NijI j = = N 0 (8.46)
Ms 0 N Is
When electrons move through an inductor wire, a magnetic field B is created around
the wire. If the wire carrying the electrons is wound around a metal bar of iron, the
iron will become magnetic, and it is then called an ‘electromagnet’.

8-18
 Mechatronics

As a result of Maxwell’s work, it is often more correct to speak of electromagnetic

energy, a form of energy that has both electrical and magnetic dynamical compo-
nents. Physicists now know that a number of seemingly distinct types of energy are all
actually forms of electromagnetic energy. These include x-rays, gamma (Γ) rays,
ultraviolet (UV) light, visible light, infrared (IR) radiation, radio waves and micro-
waves. These forms of electromagnetic energy differ from each other in terms of the
wavelength and frequency of the energy wave on which they travel. The waves
associated with x-rays, for example, have very short wavelengths and very high
frequencies, while the waves associated with microwaves have much longer wave-
lengths and much lower frequencies.

8.4 Mechanical commutation matrixer dynamical components

Mechanics (Greek Μηχανική) is the science and technology relating to or associated
with motion, as well as the force or torque for transmission of mechanical energy
and signals. In physics it is concerned with the behaviour of physical bodies when
exposed to forces or displacements, and the successive effects of the bodies on their
environment. The subject has its origins in ancient civilisations. From the Middle
Ages on, scientists such as the Polish Renaissance astronomer, Nicolaus Copernicus
(1473–1543), the Italian physicist, mathematician, astronomer and philosopher,
Galileo Galilei (1564–1642), the German mathematician, astronomer and astrol-
oger, Johannes Kepler (1571–1630), and in particular, the British physicist,
mathematician, astronomer, natural philosopher, alchemist and theologian, Isaac
Newton (1642–1727), prepared the basis for what is currently identified as classical
mechanics.
Mechanics is a subject of classical physics that relates to particles that are either
moving with low velocity or are at rest. In addition, it may also be identified as a
subject of physics that relates to the motion and force of a particular object. The
procedure for determining mechanical commutation matrixer physical models for
mechanical dynamical components is similar to that used for electrical commutation
matrixer dynamical components. In this case, one applies Newton’s laws to
mechanical homogeneous continuous dynamical systems and then tries to fit the
resulting holor equations into the standard single-port or two-terminal forms. Here
again, the holor equations do not always fit the standard forms exactly, and one
must use fictitious terminals.

8.4.1 Mechanical-signal holors

The dynamical behaviour of mechanical homogeneous continuous dynamical
systems is specified by a set of holor displacements, velocities, forces and torques.
An appropriate set of these mechanical-signal holors is sufficient to specify the
general motion of a mechanical assembly moving in 3D space. In general, the
movement combines the translation and rotation of the mechanical homogeneous
continuous dynamical system with respect to some reference framework. In practice,
however, a wide variety of situations exist when the motion is either translational or
rotational.

8-19
 Mechatronics

Alternatively, the translation and rotation of a mechanical homogeneous con-

tinuous dynamical system may be considered as separate but interacting energetic
mechanisms.
In any event, it is convenient to distinguish between two kinds of one-port
mechanical-commutation matrixer components:
• one-port mechanical-commutation matrixers for translation along a fixed
direction coordinate;
• one-port mechanical-commutation matrixers for rotation about a fixed
coordinate axis.

The translational motion of a translational mechanical homogeneous continuous

dynamical system is graphically represented by a mechanical commutation matrixer
of energetically interacting components, where the nature of the interaction is
determined by a set of applied and reaction force holors and a corresponding set of
component velocity holors. For instance, one thinks of a force holor applied to an
ideal mass resulting in an acceleration of the mass such that an equal and opposite
reaction force holor is generated. Moreover, the tensor product of the force holor
and the resultant velocity holor is the power holor being delivered to the mass.
Hence, in the dynamical systems approach (systems thinking), the holors—the force
holor and the velocity holor—form a valid energy-potential-difference holor/energy-
transfer holor pair. According to the energy-potential-difference holor and the
energy-transfer holor notion, the velocity holor is an energy-potential-difference
holor, since two points are required to specify the velocity holor of the object, an
inertial reference and a point on the object. Similarly, the force holor can be readily
visualised as an energy-transfer holor, since it may be measured by reference to one
point alone. With this reasoning in mind, the following assignment of holors results.
For the mobility analogy:
• the velocity holor is analogous to the energy-potential-difference holor;
• the force holor is analogous to the energy-transfer holor.

Despite the rationale provided by the energy-potential-difference holor and

energy-transfer holor concept, this assignment is completely arbitrary.
The dual assignment is equally valid as a way of conceiving and physically and
mathematically modelling mechanical homogeneous continuous dynamical systems. In
fact, historically the analogy of the force holor with the energy-potential-difference
holor preceded the mobility analogy. For this reason, the two analogies are used
interchangeably in this book, although the development in this chapter is predominantly
carried out in terms of the mobility analogy. For the classical analogy:
• the velocity holor is analogous to the energy-transfer holor;
• the force holor is analogous to the energy-potential-difference holor.

Thus, the key to the mechanical commutation matrixer representation of mechan-

ical homogeneous continuous dynamical systems is the recognition of the force holor
as the energy-potential holor for the mechanics of translation and the torque holor as
the one for the mechanics of rotation. Here, one notes that it is quite common to

8-20
 Mechatronics

speak of force holors being transmitted through certain parts or of torque holors
being transmitted along mechano-mechanical (MM) or electro-mechanical (EM) or
fluido-mechanical (FM) shafts. This transmission does not require any motion. In the
case of rotation, the torque holor is transmitted in the axial sense of direction and is
not directly associated with the rotary motion. In addition, there is no physical
quantity of fluid or charge being transmitted, and in this sense, the mechanical
homogeneous continuous dynamical system is rather different from the electrical
homogeneous continuous dynamical systems considered so far. Nevertheless, the
concept of a ‘flow’ of force holor or torque holor is a great help in providing a
systematic mechanical commutation matrixer view of mechanical homogeneous
continuous dynamical systems.
Consider the modules ( , ) and * (figure 8.7(a)) in the equilibrium of a smooth
table. A horizontal force holor F j is applied to module ( , as shown.
To demonstrate the transmissibility of the force holor, one draws free-body
diagrams of modules A, B, and C in figure 8.7(b) (vertical force holors have been
neglected for clarity). Since the modules are in equilibrium, there must be no
unbalanced force holor acting on them. The applied force holor F j acting to the right
on module ( must therefore be balanced by a force holor of the same magnitude
acting to the left on module ( . This latter force holor is applied to module ( by
module ) .

Figure 8.7. Transmission of force holor.

8-21
 Mechatronics

From Newton’s third law, every action has an equal and opposite reaction. Since
module ) pushes against module ( , there is a reaction of module ( against module
) . Thus, on the free-body diagram of module B, the force holor from module A acts
to the left. To maintain block ) in equilibrium requires a force holor from
module * . The process of action and reaction shows that the force holor is
transmitted through the modules to the wall.
Rotational motion about a fixed axis of a rotational mechanical homogeneous
continuous dynamical system is graphically represented by a mechanical commu-
tation matrixer of energetically interacting dynamical components, where the nature
of the interaction is governed by a set of torques and angular velocities. For instance,
for the mobility analogy, the torque holor is a moment of the force holor about a
point and is consequently an energy-transfer holor in the same way as the force holor
is. In a similar fashion, the angular-velocity holor about a fixed axis is readily seen to
be an energy-potential-difference holor, thus for rotational motion it is comple-
mentary to the mobility analogy. The angular-velocity holor is analogous to the
energy-potential-difference holor. The axial torque is analogous to the energy-
transfer holor. Thus, in a similar way, one may demonstrate the energy-transfer
holor nature of the torque holor. Figure 8.8(a) shows three discs, ( , ) and * ,
connected together by dowel pins.

Figure 8.8. Transmission of torque holor.

8-22
 Mechatronics

A torque holor, T j , is applied to disc ( . The free-body diagrams of each disc

(figure 8.8(b)) show that the torque holor T j = D jiFi is transmitted through the discs
to the wall. The balancing torque holor, T j , on each disc comes from consecutive
applications of the action and reaction principle.
The energy-potential-difference holor in mechanical homogeneous continuous
dynamical systems can be a distance holor, a displacement holor, a velocity holor, or
an acceleration holor, with the choice being made to ensure that the three elemental
physical situations are mathematically modelled by elemental holor equations of the
same mathematical form as those in the electrical commutation matrixers, and that
the univalent tensor product of the energy-potential-difference holor and the energy-
transfer holor is a univalent power holor.
The appropriate energy-potential-difference holor is, therefore, the velocity
potential-difference holor ΔVi and the dimensional analysis of the univalent tensor
product is shown for confirmation
⎡m⎤ ⎡J ⎤
P κ = ΔVi ⊗ F j = ⎢ ⎥ ⊗ [N ] = ⎢ ⎥ = [W ].
⎣s⎦ ⎣s⎦

Thus, the univalent tensor product of the velocity-potential-difference holor ΔVi in

m s−1 and the force holor F j in N gives us the power holor directly in W. For
rotation, the tensor product of the angular-velocity-potential-difference holor ΔΩi is
in rad s−1 and the torque holor T j is in N · m and W.

8.4.2 Mechanical-resistance or damping holor

A mechanical homogeneous continuous dynamical system that requires a steady
force holor F j to maintain a certain velocity holor ΔVi displays dissipative effects.
Usually, the dissipation of power occurs because energy is being transformed from
kinetic energy to thermal energy by viscous friction. Viscous-force holors F j have to
be overcome whenever neighbouring bodies have a relative-velocity holor ΔVi . Thus,
any arrangement that involves the relative motion of adjacent objects incurs power
dissipation. A pure dissipater is one in which the kinetic and potential energy-storage
phenomena are absent. Thus, a light, rigid object moving through a viscous fluid or
sliding along a rough surface has a constitutive relation that statically relates to the
applied force holor F j and the relative velocity ΔVi of the object
F j = - (ΔVi ). (8.47)

In the linear case the constitutive relation becomes

F j = B ji ΔVi . (8.48)
The pure dissipater is symbolically represented by the dashpot. This schematically
evokes the dashpot devices used to damp the motions of an automotive-vehicle
suspension’s mechanical homogeneous continuous dynamical systems and electrical
meter movements.
The univalent power holor P κ absorbed by a dissipator is the univalent tensor
product of the velocity holor ΔVi and force holor F j , and is obtained from the

8-23
 Mechatronics

constitutive relation as the sum of the dissipater content G and the co-content J,
namely:
⎡m⎤ ⎡J ⎤
P κ = ΔVi ⊗ F j = ⎢ ⎥ ⊗ [N ] = ⎢ ⎥ = [W ].
⎣s⎦ ⎣s⎦

For a linear dissipater with the constitutive relation of equation (8.48), the content
and co-content are given by the Rayleigh function
1
G=J= b(v ∧ 2). (8.49)
2
The net power absorbed is

P p = b(v ∧ 2) = (F ∧ 2)/. (8.50)

Thus, the elemental mechanical situation in which energy is dissipated is that

involving friction. Unfortunately, friction effects are varied, although generally three
principal forms are recognised:
• static friction;
• Coulomb friction;
• viscous friction.

Static friction is the effect whereby the force holor required to move a stationary
object is greater than that required to move the same object when it is moving very
slowly. Clearly, this does not conform to the linear relationship, which is analogous
to the electrical resistance. Similarly, Coulomb friction, that is the force holor
independent of the velocity holor, does not conform to the linear situation in which
the force holor is proportional to the velocity holor, and it will be considered in
section 8.7. However, there are frictional effects and devices in which the force holor
is proportional to the velocity-potential-difference holor, with this being termed
viscous friction. These frictional dynamical components are usually termed damp-
ers, although the most common example is probably the shock absorber on an
automotive vehicle.
A schematic drawing of a translational damper is shown in figure 8.9(a). The
damper has a piston-cylinder arrangement. However, the damping effect is due to a
viscous liquid (oil) in the cylinder. The piston and cylinder may move at different
velocity holors. Thus, the damper is a single-port or two-terminal device. One
terminal is at the piston velocity holor ViA and the other terminal is at the cylinder
velocity holor ViN . The shear force holor FTj developed in the damper is proportional
to the velocity-potential-difference holor ViA − ViN of the piston and cylinder.
From the free-body diagram of figure 8.9(b), one finds that

FTc ⎛ A N ⎞
b 0 ⎜ Vc − Vc ⎟,
( )
FTj = BTji ViA − ViN =
FTs
=
0 b ⎜⎝ VsA VsN ⎟ (8.51)
⎠

8-24
 Mechatronics

Figure 8.9. Schematic drawing and mechanical commutation matrixers of translational damper.

where B Tji is the translational damping-admittance holor and b is the translational

damping in N·s m−1.
Consideration of the analogous electrical commutation matrixers holors shows
that the damping b is therefore analogous to the electrical conductance G = 1/R .
From equations (8.1a) and (8.51) one may describe damping as:
1 πdz
b=G= = μ, (8.52)
R h
where d is the piston diameter, z is the piston length, h is the shear stress and μ is the
viscosity of the oil.
A rotational dissipater of energy display material behaviour determines the device
angular-velocity holor ΔΩi as a function of the device torque holor Τ j . The
dissipater actually converts mechanical energy into thermal energy by the phenom-
ena of viscosity or friction.
Symbolically, a rotational dissipater is represented by a rotational dashpot with a
constitutive relation
Τ j = - (ΔΩi ), (8.53)
where the angular-velocity holor ΔΩi is measured across the dissipator terminals.
In its linear form the constitutive relation of the rotational dissipater is
T j = B ji ΔΩi . (8.54)
The power dissipated by a rotational dissipater is the sum of the dissipater content G
and the co-content J:
P = G + J.

8-25
 Mechatronics

In the general case, G and J are equal and are given by

1
G=J= B(ω ∧ 2). (8.55)
2
In rotational mechanical homogeneous continuous dynamical systems, the rota-
tional damping-admittance holor, BRji , is defined by

TRc ⎛ A N ⎞
B 0 ⎜ Ωc − Ωc ⎟,
( )
TRj = BRji ΩiA − ΩiN =
TRs
=
0 B ⎜⎝ Ω As ΩN ⎟ (8.56)
s ⎠

where ΩiA − ΩiN is the angular-velocity-potential-difference holor.

The rotational damper has essentially the same geometry as the translational
damper. The rotational damping is approximately
1 π (d )2 z
B=G= = μ. (8.57)
R 2h
The unit of rotational damping is N · m · s.
Figure 8.10 shows the symbols used for damping in translational rotational
mechanical homogeneous continuous dynamical systems. The symbols on the left
are descriptive indications of the process physical model. As a mechanical commu-
tation-matrixer dynamical component, however, both dampers are represented by
the symbol for electrical resistance. Remember, however, that B is analogous to the
conductance G = 1/R .
In general, frictional effects do not conform exactly to any one of the three
physical models of friction. However, other than the further consideration of dry
(Coulomb) friction (section 8.7), at this stage, it is quite reasonable to approximate
the actual friction characteristics with the nearest equivalent viscous damping. When
this is done, the analytical techniques used to predict mechanical homogeneous
continuous dynamical system behaviour could then be the same as those used in
electrical commutation matrixers.
The power dissipated by a damper is equal to the univalent tensor product of the
energy-potential-difference and energy-transfer holors associated with the mechan-
ical commutation matrixer component.

8.4.3 Mechanical inductance holor of the spring

A mechanical homogeneous continuous dynamical system which when subject to a
force holor, F j , either compresses or elongates without significant acceleration of its
elastic dynamical component parts or loss of mechanical energy due to the friction
or unrecoverable deformation is a pure translational spring. The energy storage
mechanism of a pure spring is the net displacement holor, Xi , of the spring from its
quiescent state. The variable displacement holor is defined as
Xi = XiA + ΔXiA − XiN − ΔXiN . (8.58)

8-26
 Mechatronics

Figure 8.10. Symbols and mechanical commutation matrixers for translational and rotational damping.

The material properties of a spring are specified by a constitutive relation that relates
the applied force holor, F j , and the resultant spring deformation holor, Xi ,
Xi = - (F j ). (8.59)

A linear ideal spring obeys Hooke’s law and has the constitutive relation
δ ki j
Xi = F, (8.60)
Kjk

where
0 −k
Kij =
k 0
is known as the translational spring stiffness holor, while the inverse multiplied by
the Kronecker delta,
δ ki 1 0 k 0 1/ k ,
= ∧ =
Kjk (k 2) −k 0 −1/ k 0

is commonly known as the ‘translational compliance’.

8-27
 Mechatronics

From the above-mentioned argument it is clear that an ideal spring is (according

to the mobility analogy) an energy-potential-difference holor, F j , store, since the
displacement holor, Xi , represents an accumulation of the energy-potential-
difference holor, F j . The stored energy in a spring is kinetic energy, ; , and both
the potential energy, ; , and the kinetic co-energy, ; *, can be evaluated from the
constitutive relation.
In the linear case, the stored energy and co-energy are given by
1
; = ;⁎ = k (f ∧ 2). (8.61)
2
Examples of mechanical homogeneous continuous dynamical systems that can be
physically and mathematically modelled as a spring are:
• the leaf spring used in automotive-vehicle suspensions’ mechanical homoge-
neous continuous dynamical systems;
• the pipe U joint common in thermal homogeneous continuous dynamical
systems;
• the familiar coil spring used in clock and meter movements.

In fact, in any mechanical homogeneous continuous dynamical system, where

compliance effects are very dominant, the power absorbing characteristics can be
physically and mathematically modelled as a spring. Another mechanical situation is
that of a spring (figure 8.11). The ideal spring (negligible mass) is a direct application
of Hooke’s law; the extension or compression is directly proportional to the force
holor, FHj , producing it. Normally, this is expressed in the holor form

Figure 8.11. Helical and spiral spring configurations with their mechanical commutation matrixers.

8-28
 Mechatronics

(
FHj = PHji XiA + ΔXiA − XiN − ΔXiN )
FHc ⎛ A A N N ⎞ (8.62a )
= = 0 k ⎜ Xc + ΔX c − Xc + ΔX c ⎟,
FHs −k 0 ⎜⎝ X sA + ΔX sA X sN + ΔX sN ⎟
⎠

where k is the spring stiffness of the translational spring, and XiA and XiN are the
position holors of the ends of the helical spring relative to some fixed reference when
the force holor, FHj = 0. ΔXiA and ΔXiN are the displacement holors of the ends when
the force holor, FHn , is applied.
If one differentiates equation (8.62a) with respect to time (frequency), since Xi1
and Xi2 are constants, one obtains

( ) (
ωFHj = PHjiω ΔXiA − ΔXiN = PHji ViA − ViN )
FHc ⎛ A N ⎞
= ω = 0 k ω⎜ ΔX c − ΔX c ⎟
FHs −k 0 ⎜⎝ ΔX sA ΔX sN ⎟
⎠ (8.62b)
⎛ VA VcN ⎞
= 0 k ⎜⎜ A − ⎟.
c
−k 0 ⎝ Vs VsN ⎟⎠

This is the same holor form as equation (8.1b). Thus, a spring is the analogous
situation to an inductor, although it must be noted that the stiffness, k , is analogous
to the reciprocal of inductance, 1/L .
The principal unit of stiffness is N m−1, which is rather small for most useful
springs, and therefore k is more likely to be expressed in kN m−1 or MN m−1. For a
helical spring, the spring constant may be computed from the formula

1 (d ) 4
k = =G , (8.63)
L 8N (D )3

where G is the shear modulus of elasticity, d is the wire diameter, D is the coil
diameter and N is the number of coils.
Any mechanical homogeneous continuous dynamical system that is subject to a
torsional torque holor, T j , has a resultant angular-displacement holor, Δθi ,
measured across the object, and along an axis displays the essential features of a
rotational spring. If the object is nominally massless and moves without dissipation,
then it is a pure rotational spring. The constitutive relation of a pure torsional
spring is
Δθi = - (T j ). (8.64)

Equation (8.64) states that the angular-displacement holor, Δθi , measured across the
spring is statically determined by the torque holor, T j , applied at the two ends. The
angular-displacement holor, Δθi , is defined in terms of the angular-velocity holor,
ΔΩi , by the dynamical relationship

8-29
 Mechatronics

ΔΩi = ωΔθi . (8.65)

In equation (8.65), the relative angular velocity, ΔΩi , is defined as positive if the
terminal angular velocities, ΔΩiA and ΔΩiN (themselves measured with respect to an
inertial framework), are such that ΔΩiA is larger in an anticlockwise sense than ΔΩiN .
If the constitutive relation of the spring is linear, then
T j = K ji Δθi , (8.66)
where

K ji = 0 −K
K 0
is known as the torsional spring stiffness holor, while the inverse multiplied by the
Kronecker delta,
δ ki 1 0 K 0 1/ K ,
ij
= =
K (K ∧ 2) −K 0 −1/ K 0
is commonly known as the ‘torsional compliance’.
According to the mobility analogy, the rotational spring is an energy-potential-
difference holor, Δθi , store with stored potential energy < and co-energy < *.
If the store is linear, the potential energy and co-energy are given by
1
< = <⁎ = K (Δθ ∧ 2). (8.67)
2
For a torsional spiral spring (figure 8.11(b)), the torque holor, TSj , is related to the
change in the angular-displacement holor, ΔθiiS, as:
ji
TSj = PSji Δθ iS = PS (ΔθiA − ΔθiN )

TSc ⎛ A N ⎞
= 0 K ⎜ Δθc − Δθ c ⎟. (8.68)
s =
TS −K 0 ⎜⎝ ΔθsA ΔθsN ⎟
⎠

Thus, the spring constant for the torsional spiral spring has the same units as the
torque holor (N · m).
Now, deriving equation (8.68) with respect to time (frequency) yields
ji
( )
ωTSj = PSjiωΔθ iS = PS ω ΔθiA − ΔθiN = PSji ΩiA − ΩiN( )
TSc ⎛ A N ⎞
= ω = ω 0 K ⎜ Δθc − Δθc ⎟
TSs −K 0 ⎜⎝ ΔθsA ΔθsN ⎟
⎠ (8.69)
⎛ A N ⎞
= 0 K ⎜ Ωc − Ωc ⎟ ,
−K 0 ⎜⎝ Ω As ΩN ⎟
s ⎠

8-30
 Mechatronics

where Ω1A and Ω 2N are the angular velocity holors of the inner and outer terminals of
the spiral spring. Note, also, that the helical spring may also be used in torsion.
The spring constant for a torsional spring with round wire is:
1 π (d ∧ 4)
K = =E , (8.70)
L 64l
where E is the modulus of elasticity, d is the wire diameter and l is the total length of
the spring.
Figure 8.12 shows the mechanical and mechanical commutation matrixer
symbols for springs, i.e. elastic dynamical components. The mechanical commuta-
tion matrixer representation for both types is the same as for the electrical
inductance and the fluidic inertance. Note, however, that K is analogous to 1/L .
The energy stored by a spring depends on the energy-transfer holor and the spring
constant. Its form is the same as that given for the energy stored in the electrical
inductance (equation (8.69)).

Figure 8.12. Symbols for springs with their mechanical commutation matrixers.

8-31
 Mechatronics

The spring produces a force holor that is proportional to the displacement holor.
For some special geometry, a force holor that is proportional to a displacement
holor can be obtained without a spring.
Consider the harmonic motion shown in figure 8.13. When a mass is connected to
a spring (figure 8.13(a)), the mechanical homogeneous continuous dynamical system
will oscillate if the mass is initially displaced and then released.
The spring creates a restoring force holor that is always proportional to the
displacement holor. A similar restoring force holor may occur due to the position
holor without a spring present. Figure 8.13(b) shows a mass suspended from a string
in the form of a pendulum.
Now, if the mass is displaced and released, the mass will oscillate. The position
holor of the mass provides the restoring force holor in this case. This force holor can
be computed by considering a free-body diagram of the mass (figure 8.14(a)).

Figure 8.13. Harmonic motion and its mechanical commutation matrixer.

8-32
 Mechatronics

Figure 8.14. Pendulum restoring-force holor and its mechanical commutation matrixer.

From a summation of the force holors (figure 8.14(b)), the relation between the
restoring force holor and displacement holor is:

FRc XcR
FRj = PRjiX iR = s = 0 k , (8.71)
FR −k 0 XsR

and in addition, the equivalent spring constant for the pendulum is:
1 mg
k = = ,
L l
where l is the length of the string in metres (m) and m is the mass in kilograms (kg).
If equation (8.71) is differentiated with respect to time (frequency), the result is:

ωFRj = ωPRjiX iR = PRjiViR

FRc XcR VcR (8.72)
=ω 0 k
0 k
= s =
FR −k 0 XsR −k 0 VsR

The restoring force holor and the velocity holor have the relationship indicated by
equation (8.1b), except that again, one has only one terminal.
One must add a fictitious terminal to make this fit into the single-port or two-
terminal physical model. In this case, one chooses the ground at the zero velocity
holor as the other terminal.
Figure 8.14(c) shows the mechanical commutation matrixer representation for the
restoring force holor due to the pendulum position holor. The mechanical dynamical
component is shown as ‘dashed’ because of the assumption of a fictitious terminal.
The equivalent gravitational force for the pendulum is:
F G = sqrt[(F R ^2) + (F E ^2)]. (8.73)

8-33
 Mechatronics

8.4.4 Mechanical-capacitance holor of the mass

A pure translational mass is a mechanical homogeneous continuous dynamical
system’s rigid object that is moving through a non-dissipative environment. Then,
according to Newton’s second law, the momentum holor, P j , of the mass is linearly
related to the rigid object’s velocity holor, Vi

P j = M jiVi , (8.74)

where M ji is the Newtonian mass holor of the rigid object.

For the quantity, P j , the momentum holor is defined by

1 j
Pj = F or F j = ωP j , (8.75)
ω

where F j is used to signify the force holor.

The mobility analogy indicates that the quantity momentum holor, P j , is formally
analogous to the energy-transfer holor, V j , accumulation, and thus a pure transla-
tional mass can be classified as an energy-transfer store. Observe that the velocity
holor, Vi , is considered to be positive if there is a net movement of terminal N in the
sense of direction indicated.
The Newtonian mass, i.e. an inertial dynamical component, has an intrinsically
linear constitutive relation and consequently the stored kinetic energy, ; , and the
kinetic co-energy, ; *, are equal:
1
; = ;⁎ = m(v ∧ 2). (8.76)
2

The ideal translational mass moving without resistance is completely described by

the dynamical relation given in equation (8.75) or the constitutive or material
relation given in equation (8.74). The remarkable thing about the material behaviour
of mass as specified by Newton’s second law is its linearity.
From Einstein’s theory of relativity, it is now established that this is only an
approximation that is valid for velocities much lower than the velocity of light.
When this condition does not prevail, the constitutive relation for the momentum,
p, of the mass yields
mv
p= ,
sqrt[1−(v ∧ 2)/(c ∧ 2)]

where c is the velocity of light.

The elemental mechanical situation of an isolated mass (figure 8.15(a)) can be
considered through a simple application of Newton’s law:

c Vc
F j = ωM jiVi = F s = 0 −ωm , (8.77)
F ωm 0 Vs

8-34
 Mechatronics

Figure 8.15. Symbols for mass and their mechanical commutation matrixers.

or by integrating with respect to time (frequency):

δ ki
Vi = F j + V0. (8.78)
ωM ij
The initial velocity holor, V0 = (0,0), equation (8.78), almost fits the mathematical
model described in equation (8.1c). The difference is that the mass in not a single-
port or two-terminal device. There is only one velocity holor associated with the
mass impedance holor. To place this in a mechanical dynamical-component form,
one must add the ground terminal at the zero velocity holor to equation (8.78).

8-35
 Mechatronics

Then equation (8.78) becomes:

δ ki Vc VcG 0 1/ ωm Fc ,
Vi − ViG = F j
= − = (8.79)
ωM ij Vs VsG − 1/ ωm 0 Fs

where

δ ki 0 1/ ωm .
=
ωM ij −1/ ωm 0

For translational mechanical homogeneous continuous dynamical systems, the mass

is directly analogous to electrical capacitance. From a mechanical commutation
matrixer viewpoint, the mass always represents capacitance to a fictitious ground
terminal. A pure rotational mass store is a rigid mechanical homogeneous
continuous dynamical system rotating without resistance about a fixed axis. Then,
according to Newton’s second law, the angular-momentum holor,

H j = J ji Ωi , (8.80)

where J ji is defined as the moment of inertia of the object about its fixed axis.
The axial torque holor, T j , applied to the object is defined by the dynamical
relation

T j = ωH j . (8.81)

Note that the energy-transfer accumulation, the angular-momentum holor, H j , is

linearly related through the constitutive relation to the energy-potential difference
holor, i.e. the angular-velocity holor, Ωi . As a result, the stored kinetic energy, ; ,
and the kinetic co-energy, ; *, are equal and are given as

1
; = ;⁎ = J (Ω∧2). (8.82)
2

As with the translational mass, the linear constitutive relation given by equation
(8.80) is a direct consequence of neglecting relativistic effects. This is shown in figure
8.15(a). A rotating mass admittance holor (figure 8.15(b)) is also related to its signal
holors by a specialisation of Newton’s second law.
Thus:

T j = ωJ ji Ωi , (8.83)

where J ji is the polar-moment-of-inertia admittance holor in kg · m2.

For the rotational mechanical continuous dynamical system, the capacitance
equals the polar-moment-of-inertia holor. Figure 8.15 shows the mechanical
commutation matrixer representation.

8-36
 Mechatronics

With a fictitious non-rotating ground terminal, equation (8.83) becomes:

δ ki j
Ωi = ΩiA − ΩGi = T, (8.84)
ωJ ij
and moreover, the rotating mass also represents the electrical capacitance to a
fictitious ground terminal.
The energy stored by a mass in a translational mechanical homogeneous
continuous dynamical system depends on the velocity holor and the mass-admittance
holor. For a rotational mechanical homogeneous continuous dynamical system, the
energy stored is a function of the angular-velocity holor and the moment-of-inertia
admittance holor. The formulation for the energy is the same as that given for the
electrical capacitor in equation (8.74).

8.5 Fluidic commutation matrixer components

Fluidics is the science and technology relating to or associated with a compressible
medium for transient fluidic energy and signals. It involves using the flows and
pressures of fluids in sensing, control and information-processing systems with no
moving parts. The physical premises of fluidics are hydraulics and pneumatics, based
on the theoretical foundation of fluid dynamics. The term ‘fluid’ refers to a substance
with particles that move and change their relative position easily without a
separation of mass and that yields easily to pressure and is capable of flowing, or
a substance tending to flow or conform to the outline of its container and that
includes a compressible medium (liquids, gases, aerosols and suspensions).
One of the newest technologies, fluidics has in recent years come to compete with
mechanical and electrical homogeneous complex and simple dynamical systems.
In the 1930s, a Romanian inventor, aerodynamics pioneer and the builder of an
experimental aircraft, Henri Coandă (1886–1972), described what is now known as
the Coandă effect, a major contribution to fluidic technology. He observed that as a
free jet emerges from a jet nozzle, the stream tends to follow a nearby curved or
inclined surface. It also ‘attaches’ itself to and flows along this surface if the
curvature or angle of inclination is not too sharp. Coandă explained this tendency as
being caused by the jet stream’s entraining (picking up) nearby fluid molecules.
When the supply of these molecules is limited by an adjacent surface, a partial
vacuum develops between the jet and the surface. If the pressure on the other side of
the jet remains constant, the partial vacuum, which is a lower pressure region, will
force the jet to bend and attach itself to the wall. Because fluidics is not as rapid as
electronics, it is unlikely to compete in fields with ultrahigh-speed requirements. On
the other hand, in many applications fluidics is especially advantageous. It is now
possible to detect, interlock and power complex operations by using air throughout a
system. Controls can be installed by a competent fitter who might not be capable of
dealing with electrical homogeneous continuous complex and simple dynamical
systems. The elimination of contacted electrical valves (switches) prevents a possible
fire hazard. A fluidic valve (i.e. a fluidic diode) is composed solely of a fixed
structure, which structure defines an input fluid passage and an output fluid passage,

8-37
 Mechatronics

and through these passages fluid normally passes through the input fluid passage
into the output fluid passage in a desired or forward sense of direction. The fixed
structure comprising fixed fluid-flow control means that it is effective to inhibit the
reverse or backward flow of fluid from the output fluid passage through the input
fluid passage in a fashion that is analogous to that seen in an electrical valve (i.e. an
electrical diode). Although fluidic principles are fairly old, it was not until about
1960 that researchers attempted to use fluidics commercially. The demand for
reliable controls in space research stimulated progress.
The 1960s saw the application of fluidics to complex control systems, with the
introduction of the fluidic amplifier. A jet of fluid can be deflected by a weaker jet
striking it at the side. This provides nonlinear amplification, similar to the controlled
electrical valve (i.e. the transistor) used in electrical digital logic. It is mostly used in
environments where electrical digital logic would be unreliable, as in fluidic
homogeneous continuous complex and simple dynamical systems exposed to high
levels of electromagnetic interference or ionising radiation. Nanotechnology con-
siders fluidics to be one of its instruments. In this domain, effects such as fluid–solid
and fluid–fluid interface forces are often highly significant. Fluidics has also been
used for military applications.
To develop mathematical models for fluidic dynamical components with fluid
flow one must apply the holor equations of fluid mechanics to them and try to
adapt them to the holor forms given for the principal dynamical components in
equation (8.1). In some cases, the fluidic dynamical components analysed by the
fluid mechanics holor equations do not match the principal holor forms exactly.
When that happens, restrictions are placed on the resulting fluidic dynamical
components. One should therefore be aware that the analogy between fluidic
dynamical components and electrical dynamical components is not perfect.

8.5.1 Fluidic-signal holors

In a similar way to mechanical homogeneous continuous dynamical systems, fluid
motion can be physically and mathematically modelled in terms of elementary
source, storage and dissipative mechanisms. However, fluids are generally less dense
and more compliant than solids, so the distinction between the massive and
compliant parts of fluidic homogeneous continuous dynamical systems cannot
generally be made.
Nevertheless, a wide variety of physical homogeneous continuous dynamical
systems which use fluids as a working medium are susceptible to the lumped
parameter treatment given here.
The kind of fluidic homogeneous continuous dynamical systems dealt with here
are the most simple to analyse, consisting of one-dimensional (1D) flow in pipe-work,
with the energy handling phenomena isolated as specific points.
In contemplating the energy transport mechanism in fluidic homogeneous
continuous dynamical systems, one is led to conclude that the fluid-pressure holor,
Pi , and fluid-flow-rate holor, Q j , form a valid pair of holors with which to associate
the generalised energy holor pair:

8-38
 Mechatronics

• the fluid-pressure holor, Pi , is analogous to the energy-potential-difference

holor, Fi ;
• the fluid-flow-rate holor, Q j , is analogous to the energy-transfer holor, V j .

Consider a closed pipe in which the fluid pressure is P at a certain cross-section.

Pressure is defined as the fluid force per unit area; thus if the cross-sectional area of
the pipe is S, the total fluid pressure F is

F = ∫S P dS. (8.85)

If the pressure is constant over the area, the total force is the product of the fluid
pressure and the cross-sectional area. The energy associated with fluid flow can now
be determined using the relation (cf the potential energy < of a translational or
rotational spring):
x1 x1
<= ∫0 F dx = ∫0 PS dx , (8.86)

where x is the displacement of fluid along the axis of flow past the reference cross-
section, S.
By recalling that the incremental fluid-flow-rate holor, δQ j , past the reference
cross-section, S, is
δQ j = ω(SδX j ), (8.87)

the work-done holor, < κ , can be written

1 κ 1
<κ =
ω
P =
ω
(
Pi ⊗ Q j . ) (8.88)

This expression means that, in addition to being intuitively valid, the fluid-pressure
holor, Pi , and the fluid-flow-rate holor, Q j , are true energy holors with a univalent
tensor product that equals the univalent fluid-power holor, P κ , and the time integral,
which is the fluidic energy holor, < κ . There are several fluid dynamical components
that store flow energy, with the two most important being the fluid reservoir and the
pressurised tank. The quantity that is conserved in the flow of fluid at a junction is
the mass-flow holor, Mi (in kg s−1). Strictly speaking, this ought to be the energy-
transfer holor. However, it is more convenient for measurement purposes to use the
volume-flow holor, Q j (in m3 s−1), as the energy-transfer holor. The relation between
the mass-flow holor and the volume-flow holor is
Mi = ρij Q j ,

where ρij is the fluid-density impedance holor in kg m−3.

Thus, the volume-flow holor will be proportional to the mass-flow holor when the
fluid-density impedance holor remains constant through the fluidic commutation
matrixer. Under this condition, the volume-flow holor becomes an exact energy-
transfer holor. The constant fluid-density condition is approximated very closely

8-39
 Mechatronics

when one has a gas medium with low velocity or a liquid medium. One therefore
restricts the fluidic dynamical components to liquid flow or incompressible gas flow
when one selects the volume-flow holor as the energy-transfer holor. The choice of a
suitable energy-potential-difference holor is suggested by the condition that the
tensor product of energy-transfer and energy-potential-difference holors should
represent the power holor. From this one determines that the units of an appropriate
energy-potential holor are N m−3 or pascals (Pa). Thus, the pressure-difference
holor, PiA − PiN , is clearly indicated as the energy-potential-difference holor in
fluidic homogenous continuous dynamical systems.
One interesting result is that it is possible to define the pressure holor in a manner
that is directly analogous to the definition of the voltage holor. That is, the pressure-
difference holor, ΔPi , is the work done, Δ< , in moving the unit volume, ΔV,
between two points. This is illustrated by considering the movement of an
incremental volume of fluid between two points along a pipe of uniform cross-
section, S, as shown in figure 8.16.
The work-done, Δ< , according to this alternative definition is given by
Δ<
Δp = . (8.89)
ΔV
However, the volume may be expressed as ΔV = S ΔX , and the work done is the net
force on the incremental volume multiplied by the distance moved. That is, Δ< =
ΔF ΔX. The substitution of the work done and volume into equation (8.89) yields:
Δ< ΔF ΔX . ΔF
Δp = = .= (8.90)
ΔV S ΔX S
In the limit, this is the force per unit area, which is the normal definition of pressure.
Just as there are three elemental physical situations with electrical commutation
matrixers, there are three directly analogous situations in problems of the flow of
fluids. However, depending on the fluid involved, the linear relationship between
energy-transfer and energy-potential-difference holors does not exist over as wide a
range of values, and the resulting physical and mathematical models are therefore not
quite as accurate. Nevertheless, a significant amount of information can be derived
from such models, offering a considerable insight into how fluidic homogeneous
continuous dynamical systems behave.
Since the amount of effort involved in obtaining solutions for linear fluidic
homogeneous continuous dynamical systems is significantly less than that required

Figure 8.16. Elemental volume within a pipe.

8-40
 Mechatronics

for obtaining them for nonlinear fluidic homogeneous continuous dynamical

systems, the linear physical and mathematical model is an attractive tool for
determining the general nature of the performance of fluidic homogeneous con-
tinuous dynamical systems. It is, of course, necessary to have some appreciation of
the conditions under which the linear physical and mathematical model ceases to be
accurate, and part of the following discussion is directed towards making this
assessment.

8.5.2 Fluidic-resistance holor

Since fluid flow always involves the relative motion of close fluidic homogeneous
continuous dynamical systems, there are many fluid devices in which viscous effects
can cause the conversion of fluidic kinetic energy to thermal energy and hence
dissipate power.
There are, however, two basic mechanisms that fluid dissipaters employ:
• viscous-force holors between the fluid and the retaining pipework;
• viscous-force holors between fluid particles.

The fluid dissipaters that occur because of the former effect are as follows.
(i) Flow through a porous medium: D’Arcy’s law for incompressible flow
through a porous medium inserted in a fluid line is linear. It relates the
relative-pressure holor, PiAN , across the plug to the effective fluid-flow-rate,
Q j , through it as

PiAN = Rijf Q j , (8.91)

where Rijf is the fluidic-resistance impedance holor.

(ii) Laminar flow through a capillary pipe. For small values of Reynolds
numbers, the Hagen–Poisenille law predicts that power dissipation follows a
linear constitutive relation with the fluidic resistance given by

128μl
Rf = , (8.92)
π (d ∧ 4)
where μ is the fluid viscosity, l is the capillary pipe length and d is the
capillary pipe diameter.
The fluid dissipaters that rely upon viscous-force holors between fluid particles are
as follows.
(i) Turbulent flow through a long pipe. For incompressible flow at Reynolds
numbers in excess of ∼2,000, the flow is normally turbulent. In this situation
fluid power is dissipated and the dissipater constitutive relation is approximately
P AN = aQ Q 3/4 , (8.93)
where a is a constant dependent upon the fluid properties and pipe
geometry.

8-41
 Mechatronics

(ii) Incompressible flow through an orifice. A sudden restriction or change in

cross-sectional area over a short length of a pipe constitutes an orifice. Fluid
is rapidly accelerated through the orifice with resultant turbulent flow
downstream of the restriction. An approximate expression for the con-
stitutive relation of the orifice dissipater is
ρ
P AN = QQ, (8.94)
2(Cd ∧ 2)(S0∧ 2)
where ρ is the fluid density, Cd is the orifice coefficient of the discharge, S0 is
the cross-sectional area of the orifice and P AN is the pressure difference
measured upstream and downstream of the dissipater.

Formally, the constitutive relation is given by

PiAN = - (Q j ) . (8.95)

The power holor, P κ , dissipated by a fluidic resistance holor, Rijf , is

P κ = PiAN ⊗ Q j , (8.96a )

or only the dissipated power

P p = P AN Q = G + J , (8.96b)

where the dissipater content, G, and the co-content, J, are defined in the normal way:
Q P AN
G = ∫0 P AN dQ , J = ∫0 Q dP AN . (8.97)

For linear dissipaters, the power dissipated is given by

P p = P AN Q = R f (Q ∧ 2). (8.98)

The fluidic resistance physically models the frictional effects that impede the flow of
fluid along a pipe. These exist at the interface between the fluid and the pipe, and
also between layers of the fluid, which are not moving at the same velocity.
In certain circumstances, the flow of all the particles of fluid is in the sense of
direction of the tube, such as the one indicated in figure 8.17, with this form of flow
being termed ‘laminar’.
The profiles of the fluid velocity holor are shown at three locations in the pipe. At
the pipe entrance, the velocity holor profile is uniform. The viscous properties of the
fluid cause shear force holors, which act to retard the velocity holor at the
boundaries and increase it along the centre-line.
At some distance down the pipe, a stabilised parabolic shape profile is reached.
This distance is known as the entrance length, le , and it usually varies from about
80 to 120 diameters. Thereafter, the velocity holor profile remains unchanged.

8-42
 Mechatronics

Figure 8.17. Pipe flow with viscous friction and its fluidic commutation matrixer.

In the region where the profile does not change, the steady state momentum
equation of the fluid mechanics yields:

PcA PcN Qc j
PiA − PiN = Rij Q j = − = R 0 (8.99)
PsA PsN 0 R Q sj

128μ(l − le )
R= , (8.100)
π (d ∧ 4)
where μ is the dynamical viscosity of the fluid (Pa s) and d is the diameter of the
pipe (m).
Thus, the laminar flow alone, the pressure-drop holor along a pipe, is propor-
tional to the flow-rate holor and the fluidic resistance is defined in exactly the same
manner as the electrical resistance.
The linear portion of the pipe resistance only occurs after the entrance length.
Since there is no way to eliminate the nonlinear entrance effects, the usual procedure
is to select pipe lengths, l, that are at least four times the entrance length. In this way,
the entrance length effects are minimised.

8-43
 Mechatronics

In fact, the fluidic resistance is usually approximated by replacing the distance,

l −le , in equation (8.100) with the length, l, alone. Thus, as a good approximation,
when l is much larger than le ,

PcA PcN Qc
PiA − PiN = Rij Q j = − = R 0 (8.101)
PsA PsN 0 R Qs

128μl
R= , (8.102)
π (d ∧ 4)

where PiA is now the pressure holor at the entrance to the pipe.
One should emphasise that equation (8.101) only holds for laminar flow. When
the flow-rate holor increases beyond some critical value the flow breaks up and
contains circulating currents termed ‘vortices’. This type of flow is termed
‘turbulent’ and results in an increased rate of fluidic energy dissipation within the
fluid. The critical point between laminar and turbulent flow is not precisely known,
but may be estimated by means of the Reynolds number (R ), which is a non-
dimensional expression involving several properties of the fluid and the geometry of
the pipe,

4ρq
R= , (8.103)
πμd

where q is the velocity of the fluid in m s−1, ρ is the density of the fluid in kg m−3, μ is
the dynamical viscosity of the fluid in Pa s and d is the diameter of the pipe in m.
As a rule, if R ⩽ 2 × 103, the flow is laminar, but it becomes fully turbulent when
R = 5 × 103. No special name has been given to the unit of the fluidic-resistance
holor and probably the most convenient unit is N s m5.
The laminar-flow fluidic resistance given in equation (8.102) is applicable for both
liquid and gas flows. A fluidic resistance holor is bilateral if a flow in either sense of
direction develops the same magnitude of pressure-difference holor between the end
terminals.
In fluidic homogeneous continuous dynamical systems, one may assume at once
that the resistance holors are bilateral if only one fluid medium is present. However,
in cases where liquid (e.g. water) pipes empty into gas (e.g. air), the resistance holor
is not bilateral. For instance, when the resistance is connected between the mechano-
fluidic (MF) pump and the fluido-fluidic (FF) tank, it is bilateral. There is only one
medium, the liquid (e.g. water). On the other hand, when the resistance is connected
between the liquid MF pump and the gas (e.g. air), the resistance is not bilateral. A
reversal of the MF pump will draw in gas (e.g. air).
One may still apply this fluidic commutation matrixer as a physical model,
however, if one knows in advance that the flow always remains in the proper
unilateral sense of direction from the MF pump to the atmosphere.

8-44
 Mechatronics

The power holor dissipated by a fluidic-resistance holor is equal to the tensor

product of the pressure-drop holor and the flow-rate holor associated with the fluidic
component, namely:
j
( ) ( )
P κ = PiA − PiN ⊗ Q j = γ..κji PiA − PiN Q = γ..κjiRij Q kQ j = (P p, P q, P r ).

However, a pressure-drop holor

PiA − PiN = Rij Q k.

This may also be simpler expressed as

P p = R(q ∧2 ) ,

or

P p = ⎡⎣(P A − P N )∧ 2⎤⎦ R.

8.5.3 Fluidic-inertance holor

The fluid mechanism for storing the energy-potential-difference holor, Fi , is the
kinetic energy associated with a moving body of fluid. Consider an incompressible
fluid flowing without dissipation in a tube and suppose that a plug of fluid of length l
can be considered as a rigid mass with a uniform velocity holor, V j .

( )
Fi = S PiA − PiN = Slωρij V j , (8.104)

where PiA − PiN is the pressure-difference holor across the fluid plug, ρij is the fluid-
density holor and S is the cross-sectional area of the tube.
The constitutive relation of the fluidic energy-potential-difference holor store can
be obtained by manipulating equation (8.104):

l
ΓiAN = ρ Q j, (8.105)
S ij

where ΓiAN is defined as the fluid momentum holor and is related to the pressure-
difference holor, PiAN , across the plug by

PiAN = PiA − PiN = ωΓiAN . (8.106)

The material properties of a fluidic energy-potential-difference holor store are

defined by its fluidic inertance, Lf , defined from equation (8.105) as
Lf = ρl / S . (8.107)

8-45
 Mechatronics

The kinetic energy, ; , stored in a fluidic energy-potential-difference holor store is

given by
ΓAN ΓAN
;= ∫0 Q dΓAN = ∫0 - −1
(ΓAN )dΓAN . (8.108)

The stored kinetic co-energy, ;⁎, is defined by the complementary integral

Q Q
;⁎ = ∫0 ΓAN dQ = ∫0 - (Q )d . (8.109)

In the linear case these energy functions are equal and given by
1
; = ;⁎ = Lf (Q ∧ 2). (8.110)
2

Another physical situation in fluidic homogeneous continuous dynamical systems

relates to the inertia effect in the fluid. This is of particular importance with liquids,
although it is not necessarily negligible in pneumatic continuous dynamical systems.
Consider a tube such as that indicated in figure 8.18, in which the fluid volume
flow-rate holor is Q j (m3 s 1).
If one restricts one’s attention to a section of length l and uniform cross-sectional
area S, the mass holor of the fluid is Slρij .

Figure 8.18. Physical model of the fluidic inertance.

8-46
 Mechatronics

The velocity holor of the fluid is Q j

PcA PcN 0 −ωL Qc

PiA − PiN = ωLij Q j = − = . (8.111)
PSA PSN ωL 0 Qs

Equation (8.111) has the same form as the general relation given in equation (8.1b).
The pressure-difference holor, PiA − PiN , like the energy-potential-difference holor,
is directly analogous to the voltage-source holor.
One therefore notes that this equation has the same form as the elemental
equation for inductance (equation (8.111)). This effect can therefore be termed
‘fluidic inductance’, although usually the term ‘inertance’ is used, since it is the
inertia effect that is being physically and mathematically modelled. The fluidic
inertance of a pipe is computed from
ρl
L= . (8.112)
S
The fluidic inertance unit is N s2 m−5.
The fluidic energy stored by a fluid inertance is dependent upon the flow-rate
holor (energy-transfer holor) and inertance is in the same form as given for the
electrical inductance holor in equation (8.24). Fluidic inertance has been derived
under the assumption of the flow of a frictionless fluid in a pipe. Previously, fluidic
resistance was also derived in a pipe for a fluid with friction but no inertia. Both
physical dynamical components share the same principal geometry: the pipe. For
holor analysis purposes, it is accepted that the friction and inertia effects in the pipe
are separate. In any practical case, however, both effects exist simultaneously. Thus,
in a real pipe there is both fluidic resistance and fluidic inertance. There is only one
flow holor, Q j , passing through the pipe. The complete physical model for a pipe is
therefore:

PcA PcN R −ωL Qc

PiA − PiN = ZijQ j = − = (8.113)
PSA PSN ωL R Qs

The pressure-difference holor, PiA − PiN , is the sum of the pressure-difference holor
caused by fluidic resistance and fluidic inertance. This physical model as a fluidic
commutation matrixer is shown in figure 8.19.
One uses the same symbols for the fluidic dynamical components that one used
previously for the electrical dynamical components. The physical model of a fluid
line or fluid pipe consists of a resistance and inertance in series.
The order of the fluidic dynamical components does not matter for the fluid line
because there is no definable pressure terminal between the resistance and inertance.
This occurs because the effects are not separable in a physical sense, although they
were separated analytically. Sometimes one of the effects is much smaller than the
other and can be neglected. We indicate in each problem the appropriate physical
model for each line fluidic component. However, in general, both effects must be
considered.

8-47
 Mechatronics

Figure 8.19. Fluidic commutation matrixer as a fluid-line physical model.

8.5.4 Fluidic-capacitance holor

The momentum equation of fluid mechanics was used to define fluidic resistance and
inertance. Fluidic capacitance is due to an effect one may observe in the continuity
equation of fluid mechanics.
A simple way to demonstrate the formulation of the signal variables in the form
of equation (8.1c) is to consider the closed chamber (figure 8.20(a)).
The mass holor, Mi , of the fluid in the chamber is equal to the product of the
density holor, ρij , and the volume holor, V j (Mi = ρij V j ).
If one takes the time derivative of the mass holor, Mi , one obtains an expression
for the mass flow, m. In terms of the volume flow, the result is:
dV V dρ
q= + . (8.114)
dt ρ dt
Equation (8.114) is a form of the continuity equation of fluid mechanics. When the
volume of the chamber is fixed, the first term on the right-hand side is equal to zero.
The resulting equation relates the volume-flow holor to the fluid-density holor.
However, the principal signal holors are the volume-flow holor and the pressure-
difference holor. To find a relation between the signal variables one must specify the
connection between pressure and density. This connection depends on the fluid.
For a liquid one may apply Hooke’s law to obtain:
1 1
dρ = dp , (8.115)
ρ β
where β is the bulk modulus of the liquid in Pa.
When the fluid is a gas, the relation between pressure and density depends on the
thermodynamic process. The appropriate results are:
1 1
dρ = dp (isothermal process) (8.116a )
ρ p + pa

8-48
 Mechatronics

Figure 8.20. Closed-tank fluidic capacitance and its fluidic commutation matrixer.

1 1
dρ = dp (adiabatic process), (8.116b)
ρ γ (p + pa )
where pa is the atmospheric pressure and γ is the ratio of the specific heats.
One emphasises the atmospheric pressure here because the absolute pressure is
required in equation (8.114) and p is usually the gauge pressure. If equations (8.113)
and (8.114) are substituted into equation (8.112) and integration is performed
β
p=
V
∫ q dt + p(0) (liquid) (8.117a )

p + pa
p=
V
∫ q dt + p(0) (gas, isothermal) (8.117b)

γ (p + pa )
p=
V
∫ q dt + p(0) (gas, adiabatic), (8.117c )

where p(0) is the value of the pressure in the chamber at t = 0.

8-49
 Mechatronics

If one assumes that p(0) = 0, equation (8.112) almost has the form desired in
equation (8.1c). The difference between equations (8.114) and (8.117c) is that there is
only one pressure in equation (8.116) and a pressure-difference holor is required to
fit the standard form. This is a consequence of the fact that the closed tank has only
one terminal, which is the internal pressure holor, Pi . In effect, one is trying to make
a two-terminal fluidic-component physical model out of a physical dynamical
component that has only one terminal.
One accomplishes this by assuming a fictitious terminal, PiG , at zero potential and
expressing equation (8.116) as:

δ ki PcA PcN 0 1/ωC Qc

PiA − PiG = Qj = − = , (8.118)
ωCjk PsA PsN − 1/ωC 0 Qs

V V V
where C = β
for a liquid, C = p + pa
for an isothermic gas and C = γ (p + pa )
for an
adiabatic gas.
The unit of fluidic capacitance is m5 N−1.
Thus, the fluidic capacitance for a gas is a function of pressure, p. When the fluidic
homogeneous continuous dynamical system operates with small pressure changes,
the quantity p + pa does not change appreciably and the fluidic capacitance is
essentially constant. When there are large pressure transients, the fluidic capacitance
value is variable. In general, a fluidic component that varies as a function of one of
the signal holors is termed a ‘nonlinear fluidic component’. Such fluidic dynamical
components are beyond the scope of this textbook. For our purposes, all problems
relating to gas operation will be restricted to small pressure transients. Then one can
treat the fluidic capacitance as a constant.
The fluidic energy stored by a fluidic capacitance is dependent on the pressure-
difference holor (energy-potential-difference holor) and has the same form as that
given for the electrical capacitance in equation (8.35). The fictitious terminal
adopted to complete the two-terminal form is at zero potential. Thus, a closed
tank always represents a fluidic capacitance ground. However, since the ground
terminal is fictitious, it cannot be located physically. One denotes a fictitious
ground terminal by making the fluidic-commutation-matrixer line going to the
ground ‘dashed’.
A gas-filled tank is bilateral up to a point. It ceases to be bilateral when all the gas
has been removed. Figure 8.21(a) shows a fixed volume tank with inlet and outlet
flow. The previous reasoning is still applicable. There is only one pressure terminal
in the tank and its mathematical model (equation (8.1)) requires two pressure
terminals for each fluidic component.
Figure 8.21(b) shows the fluidic commutation matrixer representation for a closed
tank capacitor. Note that the liquid filled closed tank is not bilateral. If liquid is
removed from the tank, cavitation occurs and one is no longer dealing with a
homogeneous medium.

8-50
 Mechatronics

Figure 8.21. Closed-tank fluidic capacitance with inlet and outlet.

Holor analysis of the tank with the inlet- and outlet-flow holors, Q Aj and Q Nj ,
leads to:

δ ki PcA PcN
PiA − PiN =
ωCjk
( Q Aj − Q Nj ) =
PsA
−
PsN
(8.119)
⎛ c c ⎞
= 0 1/ωC ⎜ Q A − Q N ⎟
−1/ωC 0 ⎜⎝ Q As Q Ns ⎟
⎠

The fluidic commutation matrixer equivalent for this tank is shown in figure 8.21(b).
The tank again represents a fluidic capacitance to ground. An open tank or reservoir
that is fed from the bottom with working fluid is a store of fluid flow energy. As fluid
flows into the reservoir, the potential energy of the fluid mass is increased by the
rising fluid level in the tank. The increase in energy manifests itself in an increase in
the relative fluid-pressure holor, Pi , measured at an arbitrary fixed point in the tank.
The principle of conservation of material can be applied to obtain

(
ρij Q j = ω Sρij H j , ) (8.120)

8-51
 Mechatronics

where ρij is the mass-density impedance holor of the fluid, S is the cross-sectional
area of the reservoir and H j is the fluid-height holor in the reservoir.
This expression can be rearranged in the form:
V j = C jiP,
i (8.121)
which is the constitutive relation for the reservoir and relates the fluid-pressure
holor, Pi , to the fluid-volume holor, V j , in the reservoir.
The fluid-volume holor, V j , is related to the fluid-flow-rate holor, Q j , by the
dynamical relation
Q j = ωV j = ωC jiP.
i (8.122)

Note that the constitutive relation given by equation (8.121) is linear; this is because
incompressible flow and vertical tank sides have been assumed.
The coefficient of a flow store is termed the fluidic capacitance, C; for the open
reservoir this is given by
S
C = . (8.123)
ρg
Consider now the open tank in which a liquid is stored, as indicated in figure 8.22.
The previous reasoning is still applicable.

Figure 8.22. Liquid open-tank fluidic capacitance with its fluidic commutation matrixer.

8-52
 Mechatronics

There is only one pressure terminal in the tank and its mathematical model
(equation (8.1)) requires two pressure terminals for each fluidic dynamical component.
Holor analysis of the tank with the inlet- and outlet-flow holors, Q Aj and Q Nj ,
leads to:

δ ki PcA PcN
PiA − PiG =
ωCjk
( Q Aj − Q Nj ) =
PsA
−
PsN
(8.124)
⎛ c c ⎞
= 0 1/ ωC ⎜ Q A − Q N ⎟.
−1/ ωC 0 ⎜⎝ Q As Q Ns ⎟
⎠

In this case, the pressure holor, PiA, is indicated at the bottom of the tank. The
pressure is the gravitational force per unit area of the column of liquid above it. If
the horizontal area of the tank is S, then the mass holor, Mi , of the column of height
holor, H j , is
S A
Mi =
g
( )
Pi − PiG = Sρij H j ,

from which
PiA − PiG = gρij H j , (8.125)

where g = 9.81 m s−2 is the gravitational constant and PiG is the pressure holor above
the liquid column.
Before proceeding to identify the fluidic capacitance effect, it may be noted that
equation (8.125) provides the justification for using the height of the column of
liquid as an alternative measure of pressure, commonly termed ‘head’. However,
from the discussion at the beginning of this subsection, it should be evident that the
product of the head and flow rate cannot possibly result in power as expressed in any
acceptable unit.
The resulting confusion with expressions involving fluidic resistance, capacitance
and inductance is such that head is not used as an energy-potential-difference holor
in the fluidic commutation matrixers analysed in this textbook.
Now, from equation (8.114), when the density does not change but the volume
does, q = dV/dt. In addition, since V = Sh, one may express equation (8.125) in terms
of the pressure-difference holor and the flow-rate holor as:

δ ki PcA PcN 0 1/ωC Qc

PiA − PiG = Qj = − = (8.126)
ωCjk PsA PsN −1/ωC 0 Qs

Comparison with the standard form in equation (8.1c) shows that the fluidic
capacitance of the open tank is
S
C = . (8.127)
ρg

8-53
 Mechatronics

This fluidic dynamical component is essentially a two-terminal dynamical compo-

nent without the assumption of a fictitious terminal. One terminal exists at the bottom
of the tank and the other is the pressure over the liquid column. Usually, this pressure
will be zero gauge pressure and the open tank like the closed tank will represent a
fluidic capacitance to ground (figure 8.22(b)). However, this is not necessarily the case.
A second fluid flow store is the pressurized tank; these devices store fluid flow energy
by compressing the fluid in a sealed chamber. Suppose the tank is filled with fluid of
uniform density holor, ρij , with pressure holor, P,i and absolute temperature, T. The
principle of conservation of material may be applied to obtain
ρij Q j = ωρij VTj , (8.128)

where Q j is the fluid-flow-rate holor into the tank and VTj , is the volume holor of the
tank, which is assumed constant, i.e. the rigid tank.
The constitutive relation of the pressurised tank now depends upon the nature of
the working fluid. More specifically, it depends upon how the mass density holor, ρij ,
of the fluid is related to the absolute fluid-pressure holor, Pi .
For liquids, a good approximation to this dependence is given by
1
dρij = ρ dPi , (8.129)
β ij
where β is the fluid bulk modulus of elasticity and is a measure of the ‘stiffness’ of the
fluid.
For gases used in pneumatic homogeneous continuous dynamical systems, the
situation is more complex at a given temperature and the pressure holor, Pi , is
dependent upon the precise specification of the heat transfer that accompanies the
detrain upon the gas. Only two specifications are normally considered—isothermal
heat transfer and adiabatic heat transfer. These determine two distinct thermody-
namic constraints, respectively, the constant temperature processes and the zero
heat-exchange processes.
Under these conditions, two distinct bulk modules arise—the isothermal elastic-
ity, k I , and the adiabatic elasticity, k A.
For an ideal gas there are:
bulk modulus at constant temperature ⇛ isothermal elasticity, k I = p ;
(8.130)
bulk modulus at constant heat ⇛ adiabatic elasticity, k A = γp ;

where γ = Cp /Cv is the ratio of the principal specific heats and p is the gaseous
pressure.
After manipulating equations (8.128) and (8.129), the constitutive relation of a
pressurised tank flow-store holor is
1
Vκ =
β
(
Pi ⊗VTj , ) (8.131)

8-54
 Mechatronics

where β is a constant for most fluids, β = k I = p for isothermal gaseous processes

and β = k A = γp for adiabatical gaseous processes.
The total fluid-volume holor, V j in the pressurised tank is determined from the
fluid-flow-rate holor

Q j = ωV j . (8.132)

The pressurised tank fluid flow store is characterised by the fluidic capacitance, Cf ,
given by
VT
Cf = . (8.133)
β

The stored flow energy, < , is given by

V V
<= ∫0 P dV = ∫0 - −1
(V ) dV . (8.134)

Similarly, the flow co-energy < * is given by

P
<⁎ = ∫0 - (P ) dP (8.135)

In the linear case, the two energy functions are equal and given by
1
< = <⁎ = Cf (P∧ 2). (8.136)
2

Consider figure 8.23(a), in which a closed tank is a partially filled tank; the addition
of flow, Q j , compresses the gas trapped above the liquid column. As a result, the
pressure above the liquid column is no longer constant. The equivalent fluidic
commutation matrixer for this circumstance is shown in figure 8.23(b). From the
liquid fluidic commutation matrixer viewpoint, there are two fluidic capacitances in
series. The resulting fluidic capacitance is reduced. Thus, sealing an open tank will
reduce the fluidic capacitance. The liquid open tank fluidic capacitance is only
bilateral if there is liquid in the tank. When there are flow reversals, the liquid level in
the tank must be checked to ensure that the tank has not emptied. There is still
another type of fluidic capacitance. This type features a moving part between two
chambers.
Figure 8.24(a) shows a spring-loaded piston separating two chambers at pressure
holors, PiA and PiN .
The change in volume depends on the pressure-difference holor and the spring
constant, k. If the area of the chamber is S, the volume holor change may be
expressed as:

(
ΔV j = C ji PiA − PiN . ) (8.137)

8-55
 Mechatronics

Figure 8.23. Fluidic capacitance for a liquid tank closed at the top.

Moreover, from equation (8.114), for density change, one may relate the pressure-
difference holor and the volume-flow holor for this fluidic component as:

A δ ki PcA PcN 0 1/ωC Qc

Pi − PiG = j
Q = − = (8.138)
ωCjk PsA PsN −1/ωC 0 Qs

Now, once again, comparison with the standard form (equation (8.1c)) shows that
this is a two-terminal fluidic capacitance with a capacitance value of

S2
C = . (8.139)
k

Figure 8.24(b) shows the equivalent fluidic commutation matrixer. The fluidic
capacitance computed above is designated C2. There is a compressibility capacitance
associated with each chamber: C1 and C3. Most often, the fluidic capacitances C1
and C3 have small values. In that case, the spring-loaded piston represents a
two-terminal fluidic capacitance.

8-56
 Mechatronics

Figure 8.24. Spring-loaded piston-type fluidic capacitance.

8.5.5 Fluidic energy sources

The most frequently encountered source of fluidic energy is the MF pump or MF
compressor. In general, these MF machines are neither pure flow nor pressure
sources and they are physically and mathematically modelled by a pure source with
auxiliary dynamical components to represent imperfections. For a variety of
applications, positive-displacement MF pump or MF compressor types can be
physically and mathematically modelled as pure flow sources.
If a constant pressure fluidic source is required, a fluid flow store is frequently
employed to smooth the pressure variations from a constant-displacement MF
pump or MF compressor.

8.6 Thermal commutation matrixer dynamical components

Thermics is the science and technology relating to or associated with a heat (or cold)
medium, as well as the use of a heat flow (thermal movements of molecules)
or temperature to transmit thermal energy and signals. It concerns producing heat

8-57
 Mechatronics

(or cold) and its transfer by means of various physical phenomena, in particular,
conduction, convection and radiation.
Thermal phenomena are numerous, complex and often interconnected.
Conduction generates thermo-mechanical phenomena, convection thermo-fluidic
phenomena and radiation, which brings one back to the atom. There are various
sources of thermal energy, from electricity dissipation and radiofrequency through
to combustion with thermochemistry and thermofluidics.
The main thermal control dynamical systems are based on the electrical effects of
thermoelasticity, which were discovered by a French physicist, Jean Charles
Athanasy Peltier (1785–1845), i.e. changes of phase (capillary pumped two-phased
fluidic loops) and chemical phenomena (endo- and exothermic effects). Materials do
not behave in the same way at very low temperatures (cryogenics) and very high
temperatures. Peltier also discovered the calorific effect of electrical current passing
through the junction of two different metals, as well as the calorific effect of sunlight.
This is now termed the Peltier effect, or the Peltier–Seebeck effect.
The Peltier effect, where electrical current is forced through a junction of two
different metals, forms the root of the 12, 24 or 42 V heater/coolers for automotive
vehicle use.
Thermics physicists possess exceptional know-how for understanding and master-
ing all the thermal phenomena that affect basic dynamical components, such as
thermal homogeneous continuous dynamical hyposystems, or even complete ther-
mal homogeneous continuous dynamical systems (satellites, aircraft, automobiles,
trains, buildings, etc). This knowledge is vital for their design and construction, and
for keeping them in action, especially in the case of thermal homogeneous
continuous dynamical systems confronted with severe environments (aeronautics,
space and defence applications, etc).
Since in the dissipation of energy in all of the physical homogeneous continuous
complex and simple dynamical systems considered so far there appears a thermal
energy called ‘heat’, it would be logical to extend this physical and mathematical
modelling process to thermal commutation matrixers.
The transfer of heat is a complex process, which may occur through a combination
of conduction, convection and radiation. Conduction is heat transfer by direct
molecular communication. In convection, heat is transported by the movement of
fluid masses. Radiation heat transfer, such as the transfer of heat from the Sun to the
Earth, operates by wave motion.
We restrict our consideration here to conduction and convection, since these
processes fall more naturally into the linear physical commutation matrixer’s
physical models (equation (8.1)). In this case, one applies Fourier’s law of heat
conduction and Newton’s law of cooling to obtain the appropriate thermal
commutation matrixer’s physical models.

8.6.1 Thermal-signal holors

The concept of heat arises from an intuitive judgement of how objects ‘feel’ when
touched. The experience of somebody’s tactile sensations is that objects can be

8-58
 Mechatronics

ordered in a sequence of ‘hotter’ or ‘colder’. Furthermore, this conceptual property

of ‘heat’ exhibits flow properties. Two bodies, with one hotter than the other, tend to
retain this difference if isolated from one another, but if they are brought into close
contact they tend to lose any temperature difference. These notions are systematised
by the introduction of ‘temperature’ as a measure of the degree of hotness and the
quantity ‘heat’, which tends to flow from hotter to cooler bodies. A quantitative
measure of temperature is obtained by fixing a scale of measurement against some
controlled events that are characterised by specific degree of hotness. The temper-
ature scales in common use, Celsius and Fahrenheit, are fixed with respect to the
characteristic temperature of water. The amount of heat can be similarly quantified
by an interaction experiment between two bodies of holors, TiA and TiN , that are
initially at different temperatures. Then, postulating that the amount of heat
associated with a body is proportional to the product of mass and temperature,
the conservation principle of energy can be invoked to determine the constant of
proportionality that characterises the substance.
In this way, the phenomenon of heat can be rationalised as follows. If a mass
(admittance holor, m ji ) of a substance is heated from the temperature holors, TiA to
TiN , the amount of the heat holor, H j , required is given by

A
(
H j = Cpm ji T i − TiN , ) (8.140)

where Cp is the specific heat of the substance.

Thus the quantities of the heat-flow-rate holor, Φ j , and the temperature holor, Ti ,
apparently qualify as an energy-potential-difference holor/energy-transfer holor
pair. The heat holor, H j , is readily thought of as an energy-transfer holor, and
the temperature holor, Ti , is an energy-potential-difference holor, since a temper-
ature datum is required for thermometer calibration. This notion is valid, but should
be used with care, since the heat holor, H j , is an energy holor, and so generalisations
concerning energy will not carry over.
One may recognise immediately that the heat-flow-rate holor, Φ j (e.g. through
the walls of buildings), is an energy-transfer holor. However, it is significantly
different from the other energy-transfer holors that one has encountered in the
electrical, mechanical and fluidic commutation matrixers.
The difference is that the heat-flow holor is the rate at which thermal energy is
transferred and has units of power by itself. As a result, the product of the energy-
transfer and energy-potential-difference holors is of no particular significance and
the energy-potential-difference holor is chosen without any regard to it.
The most convenient energy-potential-difference holor is the temperature-
difference holor, ΔΤi , expressed in degrees Celsius (°C). Despite the aforementioned
differences, the analogy between the thermal and electrical commutation matrixers
is real and useful. The quantity of heat is directly analogous to the quantity of
electrical charge and thus, the rate at which electrical charge flows, namely the
electrical current. An additional similarity is that temperature is proportional to the
energy per unit of heat, whereas voltage is the energy per unit of charge.

8-59
 Mechatronics

8.6.2 Thermal-resistance holor

The flow of heat through a substance is accomplished by thermal resistance and a
consequent temperature gradient.
The material relation that governs the energy-transfer holor, Φ j , of heat by
conduction is Fourier’s law:
S ji
Φj =
l
(
σC ΤiA − ΤiN , ) (8.141)

where σCji is the thermal-conductivity admittance holor of the material, S is the cross-
sectional area of the object, l is the length of the object and ΤiA − ΤiN is the
temperature-difference holor across the ends of the object.
If the transport of heat is by convection, the thermal resistance to heat transfer is
difficult to analyse. In this instance, the notion of an overall heat transfer coefficient
for a specific physical homogeneous continuous dynamical system is employed. The
resultant constitutive relation is

(
Φ j = SC Hji ΤiA − ΤiN , ) (8.142)

where the heat-coefficient admittance holor, C Hji , is defined in terms of the

constitutive relation of a particular object.
The third mechanism for heat transfer is radiation; this is described analytically
by the Stefan–Boltzmann law, which describes the power radiated from a black body
in terms of its temperature:

Φ = CR[(T A∧ 4) − (T N ∧ 4)], (8.143)

where CR is a constant determined by the geometry of the surfaces radiating and

receiving heat.
Like other kinds of physical dissipater components, thermal dissipaters do not
lose energy. The heat flow is associated with increase in the entropy of the
transporting medium and a net decrease in ‘useful’ heat energy.
Thermal resistance mathematically models the relationship between the temper-
ature-difference holor and the heat-flow-rate holor for a material of a specific
geometry when there is no storage of heat taking place. The absence of storage is
merely an approximation, which depends on the properties of the material. We
discuss this point more thoroughly at the end of this subsection.
From Fourier’s law of heat conduction in the steady state (no storage), one finds
that
δ ki
(
Φ j = G ji ΤiA − ΤiN =) Rjk
(ΤiA − ΤiN ) (8.144a )

δ ki = RjkG ij (8.144b)

8-60
 Mechatronics

S
G = σC , (8.144c )
l
where Φ j is the heat-flow holor in W, Τi is the temperature holor in °C, G ji is the
thermal conductance holor in W °C−1, δ ki is the Kronecker delta, Rjk is the
termal resistance holor in °C W−1, σC is the thermal conductivity of the material
in W m−1 °C −1, S is the area through which heat flows in m2 and l is the length of
material in m. Let us clarify thermal resistance by referring to the specific geometry
shown in figure 8.25.
Here, a round piece of insulating material has a diameter, d, and a length, l. A
collar of perfect insulation that is shown cross-hatched surrounds the piece. The
purpose of the collar is to keep the heat flow in the axial direction. Although perfect
insulation is not physically realisable, it can be closely approximated. In this case, it
is a valuable concept because it helps to define the area through which heat flows
clearly. The left face of the piece is at temperature ΤiA, the right face is at temperature
ΤiN , and the ambient temperature is Τia . For this configuration, equation (8.144a) can
be placed in the form of equation (8.1a) as:

ΤiA − ΤiN = Rij Φ j . (8.145)

Thus the thermal resistance of the cylindrical piece with axial conduction heat
transfer is:

4l
R= . (8.146)
π (d ∧ 2)k

There is heat transferred from the right face of the piece to the ambient air
temperature, Τia . This heat transfer takes place by convection.
From Newton’s law of cooling, one may express this heat transfer by:

(
Φ j = G ji ΤiA − ΤiN ) (8.147a )

G = hS , (8.147b)

where h is the surface coefficient (film conductance) in W m−2 °C−1.

For the convection heat transfer the thermal resistance has the form:
1 4
R= = . (8.148)
hS π (d ∧ 2)h

The surface coefficient, h, depends on the motion of the fluid medium and is given in
heat transfer texts for various geometric configurations and surface conditions.
Figure 8.25 shows the thermal resistance physical model as the thermal commu-
tation matrixer for the solid piece and the film. Together they may be physically
modelled as two thermal resistances in series.

8-61
 Mechatronics

Figure 8.25. Thermal resistance with its thermal commutation matrixers.

8.6.3 Thermal-inertance holor

From the signal holors for thermal commutation matrixers and equation (8.1b), one
requires a relation of the type
ΤiA − ΤiN = ωLij Φ j . (8.149)

8-62
 Mechatronics

Equation (8.149) would define thermal inertance. However, no physical situation is

known to produce such a relation between the temperature-difference holor,
ΤiA − ΤiN , and the heat flow-rate holor Φ j . As a result, there is no thermal inertance
component analogous to an electrical inductor. There is no thermal component that
displays an energy storage mechanism that is complementary to the energy-transfer
store.

8.6.4 Thermal-capacitance holor

The ability of a specific material to store heat is a measure of its thermal capacity.
The constitutive relation given by equation (8.140) for a thermal-flow store relates in
a linear fashion the total heat transferred and the temperature change.
The total heat holor, H j , is related to the heat-flow-rate holor, Φ j , by the
dynamical relation

Φ j = ωH j . (8.150)

The total heat energy stored in the energy-transfer store is the time integral of the
heat-flow-rate holor:
1 j
Hj = Φ . (8.151)
ω

The storage of heat within a mass has long been recognised as a physical process in
which thermal energy is regarded almost as a fluid, or as charge, as noted previously.
As a result, the terms ‘thermal mass’ and ‘thermal capacity’ appear frequently.
The basic equation of heat storage in a mass is:

Φ j = ωC jiΤi . (8.152)

By comparison with equation (8.1c), one finds that the thermal capacitance, C, is
equal to:
C = ρcV , (8.153)

where C is the thermal capacitance in W s °C−1, ρ is the mass density in kg m−2, c is

the specific heat of the mass in W s kg−1 °C-1 and V is the volume of the mass in m3.
The temperature holor, Τi , in equation (8.152) refers to the temperature of the
mass. For pure heat storage components, the temperature is uniform throughout the
mass. Thus, there is only one temperature associated with a mass storage compo-
nent. To put the mass into the form required for two terminal thermal capacitive
components one must assume a frictionless ground terminal at constant temperature
holor, ΤiG . Then equation (8.152) is written as:
δ ki
ΤiA − ΤiN = Φ j. (8.154)
ωCjk

8-63
 Mechatronics

Figure 8.26. Completely insulated thermal capacitance and its thermal commutation matrixer.

Consider the cylindrical mass of the material enclosed by perfect insulation in

figure 8.26(a). For pure heat storage, the temperature of the mass is uniform
throughout. This means that the materials used for heat storage will be very good
heat conductors.
The thermal conductivity of such materials will be high and they will have very
low thermal resistance. If heat is applied to the cylindrical mass, the temperature of
the mass increases in accordance with equation (8.154). Figure 8.26(b) shows the
analogous thermal commutation matrixer’s thermal dynamical component.
Once again, the assumption of a fictitious potential terminal leads to a thermal
capacitance to ground. All thermal capacitance has this fictitious terminal and,
therefore, it always represents thermal capacitance to ground in the thermal-
capacitance-matrix sense. Note that because the terminal is fictitious, there is no
heat transferred to ground.
If the insulation does not completely surround the heat storage dynamical
component, as shown in figure 8.27(a), the thermal capacitance effect is still to
ground.
The heat flow that acts to raise the temperature holor, Τi , is now Φ Aj − Φ Nj , and
the thermal dynamical components may be represented as shown in figure 8.27(b).

8-64
 Mechatronics

Figure 8.27. Partially insulated thermal capacitance and its thermal commutation matrixer.

8.6.5 General thermal dynamical components

In reality, all thermal dynamical components have both resistance and capacitance.
The magnitude of the resistance is inversely proportional to the thermal conductivity
of the material.
The magnitude of the capacitance depends on the density-specific heat product.
The physical size of the dynamical component is a factor in both the resistance and
the capacitance.
Some combinations of material and configuration are predominantly resistive or
predominantly capacitive. Such are the physical models already described in section 8.6.
Figure 8.28, however, shows the generalised thermal commutation matrixer
physical model for all thermal dynamical components.
The temperature holors ΤiA and ΤiN are the temperatures of the left and right
boundaries, and Τi is the mid-plane temperature holor.
The generalised thermal commutation matrixer physical model has half the
resistance on each side of the mid-plane and a capacitance to ground from the mid-
plane temperature. When the thermal material has a large thermal conductivity, the
dynamical component reduces to the thermal commutation matrixer shown in figure
8.27(c). In this case, the resistance is negligibly small and the dynamical component
is a capacitance to ground.

8-65
 Mechatronics

Figure 8.28. General thermal commutation matrixer physical model.

On the other hand, a material with a lower thermal conductivity has resistance. If
the specific heat-density product is low, the capacitance will be small and the
dynamical component will be purely resistive, as shown in figure 8.25. In the general
case, however, both the thermal conductivity and the specific heat product must be
considered to physically model a thermal dynamical component.

8.7 Energy sources and loads

The fundamental situations and physical machines and devices discussed so far have
implied a source of energy that is quite independent of the physical dynamical
component. These dynamical components have been either dissipative or storage
physical machines and devices that can only act on the energy while it is within a
physical heterogeneous continuous complex and/or simple dynamical hypersystem,
and they are therefore called ‘passive physical dynamical components’.
The passive physical dynamical components (resistance, inductance and capaci-
tance) are often also called ‘physical loads’. By contrast, ‘active physical dynamical
components’ are physical machines and devices that imply situations where energy
is supplied to or taken from the physical heterogeneous continuous complex and/or
simple dynamical hypersystem. Active physical dynamical components often

8-66
 Mechatronics

involve some energy conversion process such as takes place in a chemo-electrical/

electro-chemical (ChE/ECh) storage battery, mechano-electrical (ME) generator,
mechano-fluidic (MF) pump, mechano-pneumatic (MP) compressor, or chemo-
thermal (Ch-Th) furnace. That is, the source of energy for any physical heteroge-
neous continuous complex and/or simple dynamical hypersystem is physically
modelled by means of active physical dynamical components. Thus, in general, a
source supplies energy and a load receives energy. In this section, we discuss various
modes of energy sources and loads.

8.7.1 Energy-potential-difference sources

The details of the origin of the energy are often not required; only the form in which
it is fed into a physical heterogeneous continuous complex and/or simple dynamical
hypersystem is necessary. For example, most ChE/ECh storage batteries are capable
of supplying energy to an electrical homogeneous discrete dynamical system with the
voltage-difference holor, Ei between its input terminals only changing slightly if the
current holor, I j , changes.
The ideal situation would presumably be one in which the voltage-difference
holor, Ei , is constant, irrespective of the current holor, I j , flowing from the ChE/ECh
storage battery. In electrical commutation matrixer analysis, this leads to the idea of
‘ideal’ voltage source, usually referred to simply as a ‘voltage-difference source’.
Pursuing energy-potential-difference holor and energy-transfer holor analogues,
one would expect to find ideal pressure-difference sources, velocity sources and
temperature sources, which for convenience can be called ‘energy-potential-differ-
ence sources’.
Figure 8.29(a) shows the static characteristics (the energy-potential-difference
holor–energy-flow holor relation) of an ideal energy-potential-difference source.
Such a source can supply energy flow without changing potential. The electrical
commutation matrixer representation of an ideal energy-potential-difference source
(figure 8.29(b)) is a circle with the polarity indicated by a plus arrow sign inside it.
When the polarity is reversed, the device is sometimes termed a ‘sink’. However, one
may also look on this situation as a negative energy-potential-difference source.
Of course, real energy-potential-difference sources deviate from ideal ones in that
a normal pressure MF pump, for example, does not quite maintain constant

Figure 8.29. Ideal energy-potential-difference source.

8-67
 Mechatronics

Figure 8.30. Real energy-potential-difference source.

pressure, irrespective of the flow rate. However, just as a ‘real’ ChE/ECh storage
battery is regarded as a constant voltage with the internal resistance connected in
series (physically modelled by a voltage-difference source and series connected
resistance), the analogous energy-potential-difference sources can be physically
modelled by an energy-potential-difference source and series dissipative dynamical
component.
Figure 8.30(a) shows the static characteristics of the real energy-potential-
difference source.
The equivalent electrical commutation matrixer representation of the real energy-
potential-difference source is shown in figure 8.30(b). The electrical commutation
matrixer representation has an ideal energy-potential-difference source and an
internal resistance, Rs , in series. It should be noted that for dynamical studies the
physical model of an energy-potential-difference source might require at least one
storage dynamical component in addition to the energy-potential-difference source
and dissipative dynamical component. This should become evident when Thévenin
and Norton’s theorems are considered in chapter 10.
Situations involving an ideal velocity source are not often encountered in
unsophisticated mechanics. The main point is to look to see whether the velocity
at a point in the mechanical continuous dynamical system is constrained to some
specific value or function of time.
The situation where a body is struck and carried by a much larger mass that is
already moving at a constant velocity can be physically modelled by a velocity
source, since the velocity of the larger mass is virtually unchanged after contact is
made with the smaller mass.
The resulting forces may be sufficiently large to cause damage, and therefore great
care is required when subjecting any part of a mechanical homogeneous continuous
dynamical system to a suddenly applied constraint of constant velocity. The ideal
temperature source is a reasonable physical model of a thermo-statically controlled
electro-thermal (E-Th) oven, provided that the item to be heated can be inserted
without any change in E-Th oven temperature. This is inevitably very difficult to
achieve exactly.
Nevertheless, the times involved are such that in many such situations the
physical model of the temperature-controlled E-Th oven as an ideal temperature
source gives reasonably accurate results.

8-68
 Mechatronics

In general, an ideal energy-potential-difference source is the correct physical model

for any situation in which the energy-potential-difference holor is constrained to
some specified value or function of time. This is true whether the constraint is
produced naturally or whether it is the result of some mechatronical control system.
If the end result is a value or functional expression for the energy-potential-difference
holor that is independent of the energy-transfer holor, it is appropriate to use an ideal
energy-potential-difference source as a physical model. The characteristic holor
equation of an energy-potential-difference source for electrical homogeneous con-
tinuous dynamical systems may be written as:
Fi = Ei − RijsI j , (8.155)

where Ei is the voltage-difference holor when there is no current holor, I j .

Note that for the ideal energy-potential-difference source, the internal resistance
holor, Rijs = [0], and the voltage-difference holor, Fi = Ei , are independent of the
current holor, I j .

8.7.2 Energy-transfer sources

As might be expected, situations arise in which the value of the energy-transfer holor
is constant or is some specific function, independent of the energy-potential-
difference holor.
Most unsophisticated mechanics are based on the premise that it is possible (at
least theoretically) to apply a force holor that is independent of the velocity holor of
the point at which it is applied. Such situations can be physically modelled as an
ideal force-source holor, usually termed simply a ‘force source’.
In electrical homogeneous continuous dynamical systems, there are very few
situations where the current holor from a source is naturally constant, but it is quite
common for the current holor from a DC power supply to be mechatronically
controlled, such that these supplies may be represented by current sources. The ideal
current source is one where the current holor is entirely independent of the voltage
holor across it.
Figure 8.31(a) shows the static characteristics of an ideal energy-transfer holor or
energy-transfer source and its physical commutation matrixer.

Figure 8.31. Ideal energy-transfer source and its physical commutation matrixer.

8-69
 Mechatronics

The energy-transfer holor remains constant for all values of the energy-potential-
difference holor. The physical commutation matrixer symbol to represent the ideal
energy-transfer source is a cornerwise square with an arrow indicating the sense of
flow direction along the diagonal line.
At this stage, it may be wise to emphasize the fact that ‘source of current’ and
‘current source’ have significantly different meanings. A source of current has
absolutely no implications concerning whether the voltage holor or current holor are
constant (or neither is). However, a current source unequivocally implies an ideal
energy-transfer source, namely one in which the current holor is constant or some
function that is independent of the voltage holor.
Similarly, a ‘source of force’ must not be confused with a ‘force source’. For
convenience, sources in which the energy-transfer holor is considered are termed
‘energy-transfer sources’. In fluidic homogenerous continuous dynamical systems,
when this latter term is particularly applicable to MF pumps in which clearly
identifiable quantities of liquid are passed through the pump, it is termed ‘flow
source’. Provided that such MF pumps are driven at constant speed, the rate at
which the liquid is delivered is constant.
In the case of thermal continuous dynamical systems, an ordinary E-Th heater
without any thermo- control produces heat at a rate that is reasonably independent
of the temperature. An ideal flow source static is therefore used to physically model
this situation.
In general, an energy-transfer source is used to physically model any situation in
which the energy-transfer holor is constrained to some specified value or function of
time. This is true regardless of whether the constraint is produced naturally or is the
result of some mechatronical control system. If the end result is a value or functional
expression for the energy-transfer holor that is independent of the energy-potential-
difference holor, it is appropriate to use an ideal energy-transfer source as a physical
model for this situation.
In reality, there are no ideal energy-transfer sources. Any real energy-transfer
source is affected by the energy-potential-difference holor. Thus, for example, a
constant displacement MF pump would slow down if the pressure-difference holor
became too large.
Figure 8.32(a). shows the static characteristics of a real energy-transfer source and
its physical commutation matrixer.
As a fluidic homogenous continuous dynamical system’s component, the real flow
source may be considered as an ideal flow source in parallel with an internal
resistance, Rs . This is shown in figure 8.32(b).
The equation of the real flow source for fluidic continuous dynamical systems is:

δ ki
Vj = Qj − Pi = Q j − GsjiP,
i (8.156)
R jks

where Q j is the value of the flow-source holor when the pressure-difference holor,
Pi = [0].

8-70
 Mechatronics

Figure 8.32. Real energy-transfer source and its physical commutation matrixer.

In this case, an ideal flow source has an infinite value of the internal resistance
holor, R jks . Note that under this condition the flow holor does not depend on the
pressure-difference holor, Pi . The characteristics of real energy-transfer sources and
real energy-potential-difference sources are very similar (compare figures 8.30(a) and
8.32(a)).
For a unity value of the internal resistance, equations (8.152) and (8.153) are in
the same form and can be identical when Ei = Q j . In this case, one would not
designate the dynamical component as an energy-transfer source or an energy-
potential-difference source, one would merely designate it as a source.
In general, the active dynamical component can be used to physically model
either a source or a sink of energy. If the physical machine or device is acting as a
source, positive flow will emerge from the positive terminal for both the energy-
potential-difference and the energy-transfer sources. If the physical situation is one
that absorbs energy, positive flow will enter at the positive terminal. Although it is
not entirely consistent, it is nevertheless convenient to term the active dynamical
component a source, whether it is actually acting as a source of energy or a sink.

8.7.3 Loads
Resistance, inductance and capacitance are all load dynamical components.
However, resistance is the only load that has a static characteristic. Inductance
and capacitance generally have time-dependent signal holors. Figure 8.33(a) shows
the characteristics of a resistive load
For a linear resistance, the characteristic is a straight line passing through the
origin. The slope of the line is the value of the resistance. Figure 8.33(b) indicates the
analogous physical commutation matrixer for the load resistance, RL .
The relation between the holors for a resistive load can be obtained from equation
(8.1a) for this case as:
Fi = RijLV j . (8.157)

The interaction of a source and a load involves some principles of holor theory.
Although holor theory is not considered until chapter 10, one may gain some insight
now by superimposing the source and load characteristics. For example, suppose
that the characteristics of an ideal energy-potential-difference source and a resistive

8-71
 Mechatronics

Figure 8.33. Resistive load and its physical commutation matrixer.

Figure 8.34. (a) Energy-potential-difference source/resistive load interaction. (b) Physical commutation
matrixer connecting an ideal energy-potential-difference source to a resistive load.

load are superimposed (figure 8.34(a)). This is equivalent to the physical commu-
tation matrixer connecting an ideal energy-potential-difference source to a resistive
load (figure 8.34(b)).
Let us recall that one may interpret a source as representing the locus of all
possible operating states of the energy dynamical component. Similarly, the load
represents the locus of all possible values of the signal holors for a resistance.
The physical commutation matrixer configuration is such that both loci must be
applicable simultaneously. The only point that permits both loci to coexist is the
intersecting point of the characteristics. This point is termed the operating point
because it shows the value of the signal holors in the physical commutation matrixer.
This is a graphical interpretation of the simultaneous solution of two equations,
equations (8.152) and (8.153).
The energy-potential-difference source to load method is especially valuable in
linear physical commutation matrixers, where the results are more difficult to obtain
analytically.

8.7.4 Coulomb friction

When we considered friction, it was noted that the characteristic of Coulomb friction
is that the frictional-force holor is independent of the velocity holor. Accordingly,
the appropriate physical model of this situation is a force source (a torque source if

8-72
 Mechatronics

rotating) connected such that the flow of the force holor enters at its positive
terminal. This physical model is exact, provided that the motion or rotation is
restricted to a single sense of motion or rotation direction. If the motion reverses, the
polarity of the force source must be reversed, because the frictional-force holor
reverses.
Let us examine, for example, the case of a mass holor, M ji , moving along a rough
horizontal surface with the velocity holor, Vi (figure 8.35).
The normal friction between the mass holor, M ji , and the surface holor, N j , is
equal to the weight holor, G j = gM ji . The frictional-force holor retarding motion,
F j , is μN j (or μG j ), where μ is the coefficient of the sliding or kinetic friction. Figure
8.36 shows a plot of the velocity-difference holor, Vi − ViG , to the force holor, F j ,
characteristics of the friction.
When the velocity-difference holor, Vi − ViG = [0] (no motion), the frictional
force, F j , is somewhat larger than when the mass is moving. This occurs because the
coefficient of the static friction is larger than the coefficient of the kinetic friction.
In reality, the frictional-force holor, F j , decreases slightly with increased velocity-
difference holor, Vi − ViG . Note, however, that the characteristic shown in figure 8.36
is quite similar to the ideal energy-transfer source characteristic shown in figure
8.31(a). As an approximation, therefore, one uses a force source physical model to
represent Coulomb friction. If the motion reverses in its sense of direction, the force
source is reversed. This is a nonlinear effect that is outside the scope of this text; the

Figure 8.35. Sliding friction.

Figure 8.36. Physical model of the Coulomb friction by the force (energy-transfer) source.

8-73
 Mechatronics

analysis of such a situation is usually introduced in texts on mechatronical control

systems.

8.7.5 Weight
Another situation involving a constant force holor, F j , which is not necessarily a
source of energy, is the weight holor, G j , of a mass holor, M ji , subject to vertical
motion.
In figure 8.37(a), an applied force holor, F j , is required to balance the weight
holor, G j = gM ji , if the mass holor, M ji , is to remain stationary, this being the
means by which one is aware of the gravitational-force holor, G j , on an object.
However, the weight holor of the object is the gravitational force holor, G j , which
should be expressed in newtons (N).
Provided that the distances from the centre of mass of the Earth do not vary
significantly, the weight holor, G j , is constant and can therefore be physically
modelled by means of a force-source holor.
The polarity of the force-source holor will depend on the sense of direction chosen
for the positive velocity holor, Vi . If the velocity holor, Vi , is positive when the motion
is upwards, the polarity is that shown in figure 8.37(b).

Figure 8.37. Physical modelling of weight.

8-74
 Mechatronics

8.8 Physical dynamical components as mathematical operators

From the fundamental holor equations of all of the passive dynamical components, it
can now be noted that the energy-transfer holor is either a multiple, or a time derivative,
or a time integral of the energy-potential-difference holor. That is, the voltage-drop
holor, Vi , across a resistor, for example, is simply the current holor, I j , multiplied by a
constant, i.e. the resistance holor, Rij . This is true for ideal resistance no matter how the
current holor, I j , varies, and therefore the appearance of a graph of the voltage-drop
holor, Vi , against time is identical to that of a graph of the current holor, I j , against
time, except for the scale factor that the value of resistance introduces.
The voltage-difference holor, Vi , across an inductor is the time derivative of the
current holor, I j , flowing through it with the scale factor of the value of inductance
holor, Lij , included.
If the current holor, I j , is constant, the voltage-difference holor, Vi , across an ideal
inductance holor, Lij = [0]. If the current holor, I j , is some function of time, the
voltage-difference holor, Vi , is then the derivative of that function multiplied by the
value of the inductance holor, Lij , as a scale factor.
The voltage-difference holor, Vi , across a capacitor, on the other hand, is the time
integral of the current holor, I j , flowing through it with the scale factor of the
capacitance, Cik , included.
If the current holor, I j , is constant, the voltage-drop holor, Vi , across the capacitor
will increase at a uniform rate.
If the current holor, I j , is some function of time, e.g. sinusoidal, the voltage-drop
holor, Vi , is then the integral of that function multiplied by the Kronecker delta, δ ki ,
and divided by the capacitance reactance holor, ωCjk .
The process of integration generally requires evaluation of the constant of
integration. However, in this context, the value of the integral is usually known as
the instant in time at which the process is considered to start, i.e. t = 0 in most cases.
By using the format of a particular integral, the value of the constant of
integration is usually obtained automatically by inspection. For example, if one
considers a capacitor that is initially charged when a current holor, I j , is passed
through it, the resulting voltage-drop holor, Vi , across the capacitor may be
developed from the fundamental holor equation (8.1c):
δ ki j
Vi − Vi0 = I . (8.158)
ωCjk

The second term of the left-hand side represents the voltage-drop holor, Vi0 , resulting
from the initial charge on the capacitor.
One should be aware that time integration always adds the integral in an interval
to the value of the voltage-drop holor at the beginning of the interval.

8.9 Summary
The process of determining analogous physical situations can now be seen to have
two main parts. The first of these involves identifying the analogous energy-
potential-difference and energy-transfer holors.

8-75
 Mechatronics

Table 8.1. The analogous energy-potential-difference and energy-transfer holors.

Energy-potential- Energy-transfer Tensor product

difference holor holor ⊗
Electrical homogeneous Voltage holor (V) Current holor (A) Power holor (W)
continuous dynamical
system
Magnetic homogeneous Electromotive-force Magnetic-flux Power holor (W)
continuous dynamical holor (A turns) holor (Wb)
system
Mechanical homogeneous Velocity holor (m s−1) Force holor (N) Power holor (W)
continuous dynamical
system—translation
Mechanical homogeneous Angular-velocity holor Torque holor Power holor (W)
continuous dynamical (rad s−1) (N m)
system—rotation
Fluidic homogeneous Pressure holor (Pa) Volume-flow-rate Power holor (W)
continuous dynamical holor (m3 s−1)
system
Thermal homogeneous Temperature- Heat-flow-rate Power holor (W)
continuous dynamical difference holor (°C) holor (W)
system

Although at first these are selected on the basis that the univalent tensor product
will be a univalent power holor, there is no reason not to use the univalent power
holor as the energy-transfer holor, if the fundamental holor equations turn out to
have the same form, as in the thermal continuous dynamical system. The final choice
is shown in table 8.1.
The fundamental physical situations are physically modelled by means of the
physical commutation matrixer components. With the physical heterogeneous
continuous complex and/or simple dynamical hypersystems having a power holor
as the tensor product of the energy-potential-difference and energy-transfer holors
(electrical, mechanical and fluidic homogeneous continuous dynamical systems),
there are only three physical commutation matrixer dynamical components.
• One dissipates energy, usually as heat with P-type energy dissipation, and the
other two store energy.
• One of the storage physical dynamical components stores energy when the
energy-potential-difference holor exists, whether or not the energy-transfer
holor has some value. For convenience, these physical dynamical components
can be categorised as D-type or kinetic-energy storage physical dynamical
components.
• The other physical dynamical component stores energy by virtue of the energy-
transfer holor, whether or not the energy-potential-difference holor has some
value. For convenience, these physical dynamical components can be categor-
ised as I-type or potential-energy storage physical dynamical components.

8-76
 Mechatronics

For thermal homogeneous continuous dynamical systems, there are only two
fundamental physical dynamical components: dissipative and kinetic energy
storage.
The major forms of fundamental holor equations with impedance holors are:
Fi = ZijPV j = DijPV j P -type energy dissipation, (8.159a )

j
Fi = ZijDV j = ωK ijDV D-type energy storage, (8.159b)

1 I j
Fi = ZijI V j = Pij V I-type energy storage. (8.159c )
ω
The appropriate interpretations of DijP , KijD and PijI are summarised in table 8.2.
The major forms of fundamental holor equations with admittance holors are:
δki
V j = YPjiFi = Fi P -type energy dissipation, (8.160a )
D jkP

1 δki
V j = YDjiFi = Fi I -type energy storage, (8.160b)
ω K jkD

δki
V j = YIjiFi = ω Fi D-type energy storage. (8.160c )
PjkI
δ ki δ ki δ ki
The appropriate interpretation of , and is summarised in table 8.3.
DP D
jk K jk P Ijk

Table 8.2. The analogous impedance physical components.

P-type DijP D-type K ijD I-type PijI

Electrical homogeneous continuous dynamical system Rij Lij δ ki

Cjk
Magnetic homogeneous continuous dynamical system Rij — —
Mechanical homogeneous continuous dynamical δ ki δ ki δ ki
bjk kjk mjk
system—translation
Mechanical homogeneous continuous dynamical δ ki δ ki δ ki
system—rotation Bjk Kjk J jk
Fluidic homogeneous continuous dynamical system Rij Lij δ ki
C jk
Thermal homogeneous continuous dynamical system Rij — δ ki
Cjk

8-77
 Mechatronics

Table 8.3. The analogous admittance physical components.

δ ki δ ki δ ki
P-type P I-type D-type
D jk D
K jk I
P jk

Electrical homogeneous δ ki δ ki C ji
G ij = Rjk Ljk
continuous dynamical system
Magnetic homogeneous δ ki — —
G ij = Rjk
continuous dynamical system
Mechanical homogeneous b ji k ji m ji
continuous dynamical system—
translation
Mechanical homogeneous B ji K ji J ji
continuous dynamical system—
rotation
Fluidic homogeneous continuous δ ki δ ki C ji
G ij = Rjk Ljk
dynamical system
δ ki
Thermal homogeneous continuous G ij = — C ji
Rjk
dynamical system

Reference
Maxwell J C 1873 A Treatise of Electricity and Magnetism (Oxford: Clarendon)

8-78
 IOP Publishing

Mechatronics
Dynamical systems approach and theory of holors
B T Fijalkowski

Chapter 9
Principal physical commutation matrixers as
physical models

The fascinating impressiveness of rigorous mathematical analysis, with its

atmosphere of precision and elegance, should not blind us to the defects of the
premises that condition the whole process.
T C Chamberlin

9.1 Introduction
There are two principal procedures to show a physical heterogeneous continuous
dynamical hypersystem in a diagram. In one, the particulars of the physical hetero-
geneous continuous dynamical hypersystem connections are exposed. Diagrams of
these features are not easy to put into practice. In the other, the particulars of the
connections are neglected. As an alternative, the dynamical components are arrayed in
a physical commutation matrixer for maximum intelligibility, and to display how
energy-transfer physical holors and energy-potential-difference holors integrate or
separate. The physical commutation matrixer diagrams for this depiction ought to be
those generally used for analysing the action of a physical heterogeneous continuous
dynamical hypersystem. There are conventional procedures for creating equilibrium
holor equations by well-defined reference to these physical commutation matrixer
diagrams. This eases the formulation of equilibrium holor equations to represent the
signal holors.
Physical commutation matrixer diagrams with the same form as electrical-
commutation-matrixer diagrams may symbolise the interactions of the essential
physical dynamical components in mechanical, fluidic and thermal homogeneous
continuous dynamical systems. This creates the same analytical procedures that are
accessible for the above-mentioned systems. The effect is that equilibrium holor
equations for physical heterogeneous continuous dynamical hypersystems of

doi:10.1088/978-0-7503-1350-6ch9 9-1 ª IOP Publishing Ltd 2016

 Mechatronics

mechanical, fluidic and thermal dynamical components can be created with the same
knowledge and dependability.

9.2 Creation of physical commutation matrixers

The aim of physical commutation matrixer creation is to arrange mechanical,
electrical, fluidic and thermal dynamical component configurations into physical
commutation matrixer diagrams.
The core of the problem is to detect those points in the physical heterogeneous
continuous dynamical hypersystem that may have various values of the energy-
potential-difference holor. More to the point, each physical commutation matrixer
must have a ‘datum’ or ‘reference’ value for the energy-potential-difference holor.
The datum is principally termed the ‘ground’ of the physical commutation
matrixer. The other physical commutation matrixer points are termed ‘nodes’.
For physical heterogeneous continuous dynamical hypersystems, the number of
nodes (excluding a datum) is the same as the number of degrees of freedom (DoF).
The procedure for physical-matrixer creation necessitates the detection of the
principal effects. In particular, one must detect all the energy-potential-difference
holor nodes and then authenticate the adequate physical dynamical component or
dynamical components that connect them together. The nodes are then converted
into the connections of the physical commutation matrixer branches. Each branch
displays one physical dynamical component and relates to one principal physical
situation. The procedure is displayed in the examples given in chapters 10–13.

9.3 Principal physical commutation matrixer laws

The objective of physical-commutation-matrixer holor analysis is to derive
mathematical relations between output-signal holors and input-signal holors.
The physical commutation matrixer representation of a physical heterogeneous
continuous dynamical hypersystem permits one to accomplish this analysis through
the application of two principal physical commutation matrixer laws. One of these
laws, Kirchhoff’s first law (KFL), refers to physical conditions at any junctions, and
is expressed as:
ΣV j = 0. (9.1)
For instance, the summation of current holors at a junction of an electrical commu-
tation matrixer is analogous to the summation of force holors in a mechanical
commutation matrixer or to the summation of fluid-flow-rate holors in a fluidic
commutation matrixer. The fact that the holor-algebraic sum of each of these
energy-transfer holors at a junction equals zero is simply an application of the principal
ideas of continuity and equilibrium. Since the energy-potential-difference holor in
mechanical systems is the force holor, Kirchhoff’s second law (KSL) can be recognised
as being analogous to d’Alembert’s principle.

9-2
 Mechatronics

The other physical commutation matrixer law, KSL, refers to physical conditions
around a closed path in the physical commutation matrix, and is:
ΣFi = 0. (9.2)
For instance, the summation of voltage-drop holors in an electrical commutation
matrixer is analogous to the summation of velocity-difference holors in mechanical
commutation matrixer or the summation of pressure-drop holors in a fluidic
commutation matrixer.
The summation of these energy-potential physical holors is sometimes referred to
as the compatibility relation.
Although both of Kirchhoff’s laws are usually stated (equations (9.1) and (9.2)) as
the holor-algebraic sum being zero, it is often simpler to equate one of the physical
holors to the sum of the others. For instance, the energy-transfer holor flow into a
junction may be set equal to the energy-transfer holor flow out of the junction.
When the principal physical commutation matrixer laws are used in this way it is
usually possible to set up the equilibrium holor equations in a form that is suitable
for direct mnemonic solution. The generalisation of this procedure for complex
physical commutation matrixers is considered in chapter 10. However, for simple
series, parallel, series-parallel and parallel-series physical commutation matrixers,
one may apply the principal physical commutation matrixer laws without formal
procedures.

9-3
 IOP Publishing

Mechatronics
Dynamical systems approach and theory of holors
B T Fijalkowski

Chapter 10
Generalised impedance and admittance holors
of physical commutation matrixers

Always surpass the one you were yesterday.

Yagyu Sekishusai

10.1 Sinusoidal responses to energy-potential-difference holor

and energy-transfer holor functions
The holor equations for the generalised physical commutation matrixers were given
in equations (8.159) and (8.160), and are
Fi = ZijV j (10.1)

and
V j = Y jiFi (10.2)
where Zij is the generalised impedance holor for general physical commutation
matrixers and Y ji is the generalised admittance holor for general physical commu-
tation matrixers.

10.2 Impedance and admittance of principal physical commutation

matrixer dynamical components
It should be possible to obtain a procedure for writing equilibrium holor equations
that is homogeneously right and proper for finding the natural response, the forced
response, or the complete response.
One may realise this by describing the principal physical commutation matrixer
dynamical components in holor-algebraic terms rather than as time derivatives or

doi:10.1088/978-0-7503-1350-6ch10 10-1 ª IOP Publishing Ltd 2016

 Mechatronics

integrals. The holor-algebraic substitution applied to express the principal physical

commutation matrixer dynamical components is termed the ‘generalised impedance
holor’.
The generalised impedance holor of a physical commutation matrixer dynamical
component is defined as the ratio of the amplitude or effective (rms) value of the
energy-potential-difference holor to the amplitude or effective (rms) value of the
corresponding energy-transfer holor.
The generalised impedance holor, Zij , is a function of the exponent holor,
s = (σ , ω ), and is expressed mathematically as:
effective value of an energy-potential-difference holor
Zij = . (10.3)
effective value of the corresponding energy-transfer holor
To find the impedance holors of the three principal physical commutation matrixer
dynamical components we begin with the elementary relationships given in equation
(8.1). The relationships are
P
Fi = ZijPV j = Dij V j P -type energy dissipation, (10.4a )

j
Fi = ZijDV j = ωKijDV D-type energy storage, (10.4b)

1 I j
Fi = ZijI V j = Pij V I -type energy storage. (10.4c )
ω
Now we assume that both the energy-potential-difference holor and the energy-
transfer holor can be symbolised by holor functions. That is,

T Fc F cos ψF
Fi = (Fc , Fs ) == F ∠ψF = Fc Fs = = = ZijV j, (10.5a )
Fs F sin ψF

and
c V cos ψV
V j = (V c , V s ) = V ∠ψV = Vc Vs T
= Vs = = Y jiFi . (10.5b)
V V sin ψV

When equations (10.5) are substituted into equations (10.4), we are left with a
relationship between the effective (rms) values F and V. If we use the definition of
generalised impedance holors as the ratio of F/V, we obtain that:
ZijP = DijP P -type energy dissipation, (10.6a )

ZijD= ωK ijD D-type energy storage, (10.6b)

1 I
ZijI = Pij I -type energy storage, (10.6c )
ω

10-2
 Mechatronics

where
⎡D 0 ⎤
DijP = ⎢ , (10.7a )
⎣ 0 D ⎥⎦

⎡ ⎤
K ijD = ⎢ 0 −K ⎥ , (10.7b)
⎣K 0 ⎦

and
⎡ ⎤
PijI = ⎢ 0 P ⎥ . (10.7c )
⎣−P 0 ⎦
Equations (10.6) are therefore the holor-algebraic description of the principal
physical commutation matrixer dynamical components that we previously described
in equations (10.4) with holor frequency differentials and integrals.
Table 10.1 lists the generalised impedance holors in the various physical
homogeneous continuous dynamical systems.
The reciprocal of the generalised impedance holor multiplied by a Kronecker
delta is normally a more appropriate procedure to express the principal physical
commutation matrixer dynamical components. This is termed the ‘generalised
admittance holor’, Y ji , and is defined as
effective value of an energy-transfer holor
Y ji = . (10.8)
effective value of the corresponding energy-potential-difference holor

Before proceeding further, it is indispensable to know that conventionally mechanical

engineers have used the term ‘mechanical impedance holor’ with a certain perception
of force—the larger the force necessitated to cause movement, the greater the holor
mechanical impedance. With this assumption, the mechanical impedance holor is

Table 10.1. The analogous impedance physical dynamical components.

1 I
P-type DijP D-type ωK ijD I-type P
ω ij
i
Electrical homogeneous continuous dynamical Rij ωLij 1 δk
ω Cjk
system
Magnetic homogeneous continuous dynamical Rij — —
system
Mechanical homogeneous continuous dynamical δ ki δi 1 δk
i
-
bjk
ω kk ω mjk
system—translation jk
Mechanical homogeneous continuous dynamical δ ki δ ki 1 δk
i
ωK
system—rotation Bjk jk ω J jk
Fluidic homogeneous continuous dynamical Rij ωLij 1 δk
i

system ω Cjk
Thermal homogeneous continuous dynamical Rij — 1 δk
i

system ω Cjk

10-3
 Mechatronics

determined by V/F or F/V. Having preferred the theory of the energy-transfer/energy-

potential-difference holors as the principal assumption, this textbook uses the
electrical commutation matrixer concept of interpreting the generalised impedance
holor as the ratio F/V or V/F. In some of the mechatronics literature, the ratio F/V or
V/F is termed the ‘mobility’.
In spite of this, this text will be understood much more easily if the term
‘generalised impedance holor’ is used solely to symbolise the ratio of the energy-
potential-difference holor to the energy-transfer holor, while the term ‘generalised
admittance holor’ is used solely for the ratio of the energy-transfer holor to the
energy-potential-difference holor. This procedure is respected throughout this book.

10.3 Generalised impedance and admittance holors in combination

Generalised impedance and admittance holors can be combined in much the same
way as equivalent resistance holors are obtained for series and parallel connections
of resistors in electrical dynamical systems. The essential difference is that the
energy-storage physical commutation matrixer dynamical components, being terms
in ω or 1/ω, must have their impedance or admittance holors added algebraically so
that in general, an equivalent impedance holor or admittance holor is a polynomial
in ω.
The determination of a series or parallel connection must only be made by inspection
of the analogous physical matrixer. That is, it is only when the analogous physical
commutation matrixer shows unequivocally that the flow of the energy-potential-
difference holor is identical for two or more physical commutation matrixer dynamical
components that they may be considered as a series physical commutation matrixer.
Similarly, it is only when the analogous physical commutation matrixer shows
unequivocally that the energy-potential holor is identical to two or more physical
commutation matrixer dynamical components that they may be considered as a
parallel physical matrixer.
Figure 10.1(a) shows three impedance holors (Zij1, Zij2 and Zij3) interconnected so
that there is a common energy-potential holor. This is, therefore, a series physical
commutation matrixer.
Suppose one wishes to represent the physical commutation matrixer using the one
with the single equivalent impedance holor (ZijE ) shown in figure 10.1(b). This
requires that the applied energy-potential-difference holor (FiT ) produce the same
energy-transfer holor in both physical commutation matrixers.
The applied energy-potential-difference holor in figure 10.1(a) may be expressed
as:
FiT = Fi1 + Fi2 + Fi3. (10.9)

Alternatively, in terms of impedance holors (equation 10.3), equation (10.9) is:

FiT = Zij1V j + Zij2V j + Zij3V j = (Zij1 + Zij2 + Zij3)V j . (10.10)

10-4
 Mechatronics

Figure 10.1. Impedance holors in series.

Figure 10.2. Impedance holors in parallel.

The relation between the applied energy-potential difference holor and the energy-
transfer holor is in the equivalent physical commutation matrixer (figure 10.1(b)) is:

⎡ c ⎤ ⎡ Zcc −Zcs ⎤⎡Vc ⎤ ⎡ D E −X E ⎤⎡Vc ⎤

E E
FiT = ZijE V j = ⎢ F s ⎥ = ⎢ E ⎥⎢ ⎥ = ⎢ ⎥⎢ ⎥ . (10.11)
⎣ F ⎦ ⎢⎣ Z Z ssE ⎥⎦⎣Vs ⎦ ⎣ X E D E ⎦⎣Vs ⎦
sc

A comparison of equations (10.10) and (10.11) shows that the equivalent impedance
holor for a series physical commutation matrixer is:
ZijE = Zij1 + Zij2 + Zij3. (10.12)

Let us now consider the equivalent impedance holor for a parallel physical
commutation matrixer. Figure 10.2(a) shows three impedance holors (Zij1, Zij2 and
Zij3 ) that have a common energy-potential-difference holor and are therefore in
parallel to replace this physical commutation matrixer with a single equivalent
impedance holor (ZijE ).
Figure 10.2(b), an energy-transfer physical holor VTj , must produce the same
energy-potential-difference holor in each physical commutation matrixer.

10-5
 Mechatronics

For the parallel physical matrixer in figure 10.2(a):

⎛ δ ki(1) δ ki(2) δ ki(3) ⎞ T
VTj = ⎜ 1 + 2 + 3 ⎟⎟Fi .
⎜ (10.13)
⎝ Z jk Z jk Z jk ⎠

For the equivalent physical commutation matrixer shown in figure 10.2(b):

⎛ δ ki (E ) ⎞ ⎡VTs ⎤ ⎡Y Ess Y Esc ⎤⎡ FcT ⎤
VTj = ⎜⎜ E ⎟⎟FiT = YEjiFiT = ⎢ c ⎥ = ⎢ cs cc ⎥⎢ T ⎥
⎝ Z jk ⎠ ⎣VT ⎦ ⎣Y E Y E ⎦⎢⎣ F s ⎥⎦
(10.14)
⎡ GE BE ⎤ ⎡ FcT ⎤
=⎢ ⎥ ⎢ ⎥.
⎣−BE GE ⎦ ⎢⎣ F sT ⎥⎦

From equations (10.13) and (10.14) one may observe that the equivalent admittance
holor for a parallel physical commutation matrixer is:
δ ki (E ) δ ki(1) δ ki(2) δ ki(3)
= + + . (10.15)
Z jkE Z 1jk Z jk2 Z j3−k
Figure 10.3(a) shows the admittance holors (Y1ji , Y2ji and Y3ji ) in series and
figure 10.3(b) shows an equivalent admittance holor YEji representation.
One may determine the relation between the equivalent admittance holor and the
individual admittance holors by proceeding directly from the series physical
commutation matrixer summation of energy-potential-difference holors given in
equation (10.9).
However, since we know that the admittance holor is the multiplied-by-
Kronecker-delta reciprocal of the impedance holor, we may use the series physical
commutation matrixer result given in equation (10.12) to obtain:
δ ki (E ) δ ki(1) δ ki(2) δ ki(3)
= + + . (10.16)
YEij Y1ij Y2ij Y3ij

Figure 10.3. Admittance holors in series.

10-6
 Mechatronics

In a similar way, one may find the equivalent admittance holor in parallel. Thus, if
one inverts the equivalent impedance holor and the individual impedance holors for
the parallel physical commutation matrixer given in equation (10.15), the result is:
ji
YEji = Y1 + Y2ji + Y3ji . (10.17)

Admittance holors in parallel are therefore additive, just like impedance holors in
series.
Physical commutation matrixer dynamical components may be represented by
either impedance holors or admittance holors, whichever is more convenient for a
particular physical commutation matrixer. Care must be exercised, however, to
indicate clearly whether an impedance holor or an admittance holor is being used.

10.3.1 Driving-point impedance and/or admittance holors

The ratio of an energy-potential-difference holor’s effective value to an energy-
transfer holor’s effective value as observed across the input port of the physical
commutation matrixer is termed the ‘driving-point’ impedance holor (ZijDP ).
One may obtain the driving point impedance holor by combining impedance
holors in series and parallel until the entire physical commutation matrixer is
represented by a single equivalent impedance holor. The driving-point admittance
holor is merely the multiplied-by-Kronecker-delta (reciprocal) of the driving-point
impedance holor. The following examples will clarify the procedure.
Figure 10.4(a) shows a physical matrixer with impedance holors, Zij1, Zij2, Zij3 and Zij4 .
One ought to find the driving-point-impedance holor across the port A–N. The
procedure is to apply the rules of series and parallel physical commutation matrixers
(equations (10.12) and (10.15)) in sequence until only one equivalent impedance
holor remains. Accordingly, one begins by combining the series connection of Zij3
and Zij4 into an equivalent impedance holor, Zij3 + Zij4 . The physical commutation
matrixer then appears as shown in figure 10.4(b).

Figure 10.4. Determination of the driving-point-impedance holor.

10-7
 Mechatronics

Now the parallel combination of Zij2 and Zij3 + Zij4 produces an equivalent
impedance holor, according to equation (10.15), of:
δ ki (E ) δ ki(2) δ ki(3&4)
= + ,
Z jkE Z jk2 Z jk3 + Z jk4

or
δ ki
ZijE = Zij2 (Zij3 + Zij4 ) .
Z jk2 + Z jk3 +Z jk4

With this combination, the physical commutation matrixer reduces to the simple
series physical commutation matrixer shown in figure 10.4(b).
Finally, the addition of the two impedance holors in series permits us to represent
the physical commutation matrixer by a single impedance holor. This is the driving-
point impedance holor
δ ki
ZijDP = Zij1 + 2
4 Zij
Z jk + Z 3jk +Z jk
2 (Zij3 + Zij4)
δ ki
=
Z jk + Z 3jk +Z jk
2 4 (Zij1Zij2 + Zij1Zij3 + Zij1Zij4 + Zij2Zij3 + Zij2Zij4).
Figure 10.5(a) shows a physical commutation matrixer in which the dynamical
j
components have been represented by their admittance holors Y1j , Y2 , Y3j and Y4j .
One ought to find the driving-point-admittance holor across the port A–N.
Thus, one must apply the rules for combining admittance holors (equations
j
(10.16) and (10.17)). The series combinations of Y1j , Y2 , and Y3j , Y4j are shown in
figure 10.5(b). This follows directly from equation (10.16). The two resulting

Figure 10.5. Determination of the driving-point-admittance holor.

10-8
 Mechatronics

admittance holors are in parallel and one may be treated by direct addition
(equation (10.17)).
The result is:
δ ki δ ki
YEji = ij
ji ji
ij Y1 Y2 + ij
ji
ij Y3 Y4
ji
Y1 + Y2 Y3 + Y4

δ ki ji ji ji
= ij ij ij j ij ij (
ij ij Y1 Y2 Y3 + Y1jiY2jiY4ji + Y1jiY3jiY4ji + Y2jiY3jiY4ji ) .
Y1 Y3 + Y2 Y3 + Y1 Y4 + Y2 Y4

10.4 Physical heterogeneous continuous dynamical hypersystems’

transmittance or transfer-function holors
In the previous examples, the ratio of the sinusoidal response to the excitation holor
has been either a generalised admittance holor or a generalised impedance holor,
since a suitable energy-transfer holor was chosen as the sinusoidal response to an
energy-potential-holor excitation, or a suitable energy-potential-difference holor
was chosen as the response to an energy-transfer-holor excitation.
In previous chapters, there have been examples where this has not been the case
(e.g. the excitation and sinusoidal response are either both energy-potential-
difference holors or both energy-transfer holors). It is usual to define a ‘physical
heterogeneous continuous dynamical hypersystem’s transmittance or transfer-
function holor’, Gij , for the ratio of two polynomials in s = (σ, ω).

bi ωi + bi −1ωi −1+⋯+ b0
Gij = (10.18a )
aj ω j + aj −1ω j −1+⋯+ a 0

or

(ω − ωa )(ω − ωb)(ω − ωc ) ⋯ (ω − ωi )
Gij = , (10.18b)
(ω − ω1)(ω − ω2 )(ω − ω3) ⋯ (ω − ωj )

where aj , bi are the coefficients of the polynomials and ωa , ω b, …, ω1, ω2 refer to the
roots of the polynomials.
Although, strictly speaking, a physical heterogeneous continuous dynamical
hypersystem’s transmittance or transfer function is obtained for the forced response
only, it is a simple matter to ‘recover’ the original differential equation by recalling
the association between ω and d/dt, and 1/ω and ∫ dt .
It should also be noted that a physical heterogeneous continuous dynamical
hypersystem’s transmittance or transfer function implies a single exponential energy
source. As a result, the physical heterogeneous continuous dynamical system’s
transmittance or transfer-function concept is valid only when the energy-storage
physical commutation matrixer dynamical components have no initial energy. This
is because the initial energy storage is physically modelled with an additional energy

10-9
 Mechatronics

source (chapter 8). There are two situations—the energy-potential-difference-holor

divider and the energy-transfer-holor divider—that often appear in physical com-
mutation matrixers and are readily treated with physical heterogeneous continuous
hypersystem energy-transfer-holor functions.
Figure 10.6 shows the energy-potential-difference-holor divider and the energy-
transfer-holor divider physical matrixers.
The energy-potential difference-holor divider physical commutation matrixer
(figure 10.6(a)) is basically a series physical commutation matrixer and has only a
single energy-transfer holor associated with it.
The physical heterogeneous continuous dynamical hypersystem’s transmittance
or transfer-function holor that relates the energy-potential-difference-holor output,
(Fiout ), and the energy-potential-difference-holor input, (Fiin ), is:
Fiout δ ki
Gij = = Zij2 (10.19a )
Fiin Z 1jk + Z jk2

or

Figure 10.6. Energy-potential-holor divider and energy-transfer-holor divider physical commutation matrixers.

10-10
 Mechatronics

Fiout δi
Gij = in
= ij k ij Y1ji , (10.19b)
Fi Y1 + Y2

where equation (10.19b) represents the physical heterogeneous continuous dynam-

ical hypersystem’s transmittance or transfer-function holor for the energy-potential-
difference-holor divider in terms of admittance holors.
The energy-transfer-holor divider’s physical commutation matrixer (figure
10.6(b)) is a parallel physical commutation matrixer. There is only one energy-
potential difference-holor across both impedance holors. The physical heteroge-
neous continuous dynamical hypersystem’s energy-transfer holor for the ratio
j
of the energy-transfer-holor output, (Vout ), to the energy-transfer-holor input,
j
(Vin ), is:
j
Vout δ ki
Gij = = Y2ji , (10.20a )
Vinj Y1ij + Y2ij

or
j
Vout δ ki
Gij = j = 1 2
Zij1, (10.20b)
Vin Z jk + Z jk

where equation (10.20b) is the physical heterogeneous continuous dynamical

hypersystem’s transmittance or transfer-function holor for the energy-transfer-holor
divider’s physical commutation matrixer in terms of impedance holors.

10.5 General formulation of physical-heterogeneous-continuous-

dynamical-hypersystem holor equations
One of the main reasons for seeking a physical commutation matrixer representation
of a physical heterogeneous continuous dynamical hypersystem is that it facilitates
the procedure of determining the physical heterogeneous continuous dynamical
hypersystem equilibrium holor equations.
Simple algorithms exist from which these equations can be mnemonically written,
often directly from an inspection of the physical commutation matrixer diagram,
which represents a physical model.
In this way, the principal techniques of physical commutation matrixer holor
analysis can be applied to physical heterogeneous continuous dynamical hyper-
systems, other than an electrical homogeneous continuous dynamical system, and
the analogous properties of such physical homogeneous continuous dynamical
systems become quite evident. Physical commutation matrixer holor analysis is
based on the two Kirchhoff laws. These were considered in chapter 9 (equations (9.1)
and (9.2)), but are restated here in the form that is most convenient for the physical
commutation matrixer holor analysis of more complicated physical commutation
matrixers than have been considered so far.

10-11
 Mechatronics

For Kirchhoff’s first law:

∑VPj = ∑VSj , (10.21a )

which means that the sum of all the energy-transfer holors leaving a junction
through the passive physical commutation matrixer components connected directly
to the junction, VPj , is equal to the sum of all the energy-transfer holors entering the
junction from the energy-transfer-holor sources, VSj .
For Kirchhoff’s second law:

∑FiP = ∑FiS, (10.21b)

which states that the sum of the energy-potential difference-holor drops across the
passive physical commutation matrixer dynamical components, FiP, around a closed
physical commutation matrixer path is equal to the sum of the energy-potential
difference holor increases due to the energy-potential difference-holor sources, FiS.
Up to this point, the most complicated physical heterogeneous continuous
dynamical hypersystem that has been described in detail (by means of one or two
equilibrium holor equations) only has three physical commutation matrixer
dynamical components. One is now in a position to develop systematic procedures
to obtain sets of equilibrium holor equations from any physical homogeneous
continuous dynamical system, no matter how complicated, using its analogous
physical commutation matrixer diagram as the common intermediate step.
The physical commutation matrixer diagram is at first simply a graphical
representation of the interactions that take place among all the physical commuta-
tion matrixer dynamical components of a physical system in qualitative terms, i.e.
a physical model.
However, each interaction can be modelled mathematically by means of a simple
elemental mathematical relationship and thus the physical commutation matrixer
effectively becomes a graphical representation of the equilibrium holor equations.
The problem is, therefore, a matter of finding a method whereby the information
contained in the physical commutation matrixer diagram may be qualified quickly
and reliably in the form of equilibrium holor equations, which, in general, will be a
set of simultaneous holor equations. If one uses the ideas developed in this chapter,
one will be able to derive the holor equations more confidently and with less effort.
One therefore implies that all the driving holor functions are sinusoidal, and uses
the generalised impedance-holor notation. It should be clearly understood that this
does not in any way limit the validity or generality of the resulting holor equations.
With some practice, it is a simple matter to write mnemonically the equilibrium
holor equations of many physical heterogeneous continuous dynamical hyper-
systems directly in this form (i.e. ‘by inspection’ of the physical commutation
matrixer), thus reducing the likelihood of error.
We clarify the procedure in the following examples, which are given in
chapters 11–13. One may now also formulate the equilibrium holor equations
for energy-transfer holors using the mesh method and through inspection of the

10-12
 Mechatronics

impedance-holor physical commutation matrixer. We demonstrate the procedures in

the examples given in chapter 14, section 14.4.

10.6 Some properties of linear physical heterogeneous continuous

dynamical hypersystems
Much of the simplification of the procedure for writing equilibrium holor equations
in section 10.5 is due to the basic property of components of a linear physical
heterogeneous continuous dynamical hypersystem: holor responses are directly
proportional to the driving function (or its derivative or integral). Indeed, this is
why the equivalent impedance holor of a series connection is simply the sum of the
individual impedance holors.
It should therefore be expected that these basic attributes of the components,
which make up a physical heterogeneous continuous dynamical hypersystem, should
result in characteristic properties of a linear physical heterogeneous continuous
dynamical hypersystem.
Some of these properties are expressed as ‘physical commutation matrixer
theorems’, which greatly simplify the task of eliminating unwanted holors. Others
are useful in that they give guidance as to the kinds of measurements that must be
taken to obtain a valid mathematical model of the physical heterogeneous
continuous dynamical hypersystem.

10.6.1 Two-port single-input/single output physical commutation matrixers

Up to this point, we have frequently focused our attention on the behaviour of a
physical commutation matrixer at a specified pair of terminals, i.e. conductive
collectors.
Recall that we introduce the Thévenin and Norton equivalent physical commu-
tation matrixer solely to simplify the holor analysis of some physical heterogeneous
continuous dynamical systems, to focus our attention on two pairs of terminals, i.e.
conductive collectors. One finds this viewpoint convenient in situations where a
signal is fed into one pair of terminals, i.e. matrixer-input conductive collectors, and
is then extracted at a second pair of terminals, i.e. matrixer-output conductive
collectors, after being processed by the physical commutation matrixer. Because the
terminal (conductive-collector) pairs represent the points in the physical commuta-
tion matrixer where signals are either fed in or extracted, they are referred to as the
‘ports’ of the physical heterogeneous continuous dynamical hypersystem.
In this section, the author will limit his discussion to physical commutation
matrixers in which one has one matrixer-input port and one matrixer-output port.
The basic two-port single-input/single-output (SISO) physical commutation matrixer
is illustrated in figure 10.7.
There are several restrictions on using this physical commutation matrixer.
Firstly, no energy can be stored within the physical commutation matrixer.
Secondly, there can be no independent energy sources within the physical commu-
tation matrixer. Thirdly, the energy-transfer holor into the port must be equal to the
energy-transfer holor out of the port, that is, VAj = VNj and Vaj = Vnj . The fourth and

10-13
 Mechatronics

Figure 10.7. The two-port SISO physical commutation matrixer.

final restriction is that all external connections must be made to either the matrixer-
input port or the matrixer-output port. No external connections between ports are
allowed. In other words, no external connections can be made between the
terminals, i.e. the conductive collectors A–a, A–n, N–a, or N–n. These restrictions
simply limit the range of physical commutation matrixer problems to which the
two-port SISO formulation is amenable.
The fundamental principle underlying two-port SISO physical commutation
matrixer modelling is that only the terminal energy-potential-difference holors and
energy-transfer holors, that is VAj, FiA, Vaj and Fia , are of interest. There is no interest
in computing the energy-transfer holors and energy-potential-difference holors that
exist inside the physical commutation matrixer. In this section, we formalise the
approach by introducing two-port parameters.

Two-port SISO physical commutation matrixer holor equations

In viewing a two-port physical commutation matrixer, we are interested in relating
the energy-transfer holor and the energy-potential-difference holor at one port to the
energy-transfer holor and energy-potential-difference holor at the second port. The
reference polarity of the terminal energy-potential-difference holors and the senses of
direction for the references of the energy-transfer holors are shown in figure 10.7. It is
worth noting that the references at each port are symmetrical with respect to each
other, that is, at each port, the energy-transfer holor is sensed into the upper terminal
and each port energy-potential-difference holor is a step up from the lower to the
upper terminal. This symmetry makes it easier to generalise the holor analysis of a
physical commutation matrixer in terms of two-port SISO black boxes and this is the
reason for its nearly universal use in the mechatronics literature. The most general
description of the two-port SISO physical commutation matrixer may be carried out
in holor calculus. For instance, in the case of purely passive resistive physical
commutation matrixers the holor analysis reduces to solving passive resistive physical
commutation matrixers. Sinusoidal steady-state problems can be solved by direct
analysis in the holor domain, finding the appropriate holor expressions.

10-14
 Mechatronics

In this book, all the equations are written in the holor domain. Thus passive
resistive physical commutation matrixers and sinusoidal steady-state solutions
simply become special cases. The basic black box in terms of the holor domain
variables, i.e. the holors, VAj, FiA, Vaj and Fia is shown in figure 10.8.
Of the four terminal holors, VAj, FiA, Vaj and Fia , only two are independent. Thus,
for a given physical commutation matrixer, once one specifies two of the holors, one
can find the two remaining unknowns. For instance, given FiA and Fia plus the
physical commutation matrixer as the black box, one can determine VAj and Vaj . It
follows that one can describe a two-port SISO physical commutation matrixer with
just two simultaneous holor equations. Although one needs just two simultaneous
holor equations, one has a choice of six different ways in which to combine the four
holors.
The six sets of holor equations are:

FiA = ZijAaVAj + ZijAnV a ⎫

j
⎪
⎬, (10.22a )
Fia = ZijNaVAj + ZijNnV a ⎪
j
⎭

ji a ⎫
VAj = YaAji A
Fi + YnA Fi ⎪
⎬, (10.22b)
Fi ⎪
ji ji a
Vaj = YaN FiA + YnN ⎭

Fi − AijAn V a ⎫
ji a j
FiA = AaA ⎪
⎬, (10.22c )
Fi − AijNn V a ⎪
a j
VAj = AaN
ji
⎭

Fi − BijAnV A ⎫
ji A j
Fia = BaA ⎪
⎬, (10.22d )
Fi − BijNnV A ⎪
ji A j
Vaj = BaN ⎭

Figure 10.8. The holor-domain two-port SISO physical commutation matrixer.

10-15
 Mechatronics

j ji a ⎫
FiA = HijAaV A + HnA Fi ⎪
⎬, (10.22e )
Fi ⎪
j ji a
Vaj = HijNaV A + HnN ⎭

VAj = G aA Fi + G ijAnVaj ⎫
ji a
⎪
⎬. (10.22f )
ji
Fia = G aN FiA + G ijNnVaj ⎪
⎭

These six sets of holor equations can also be thought of as three pairs of mutually
inverse relations.
The first set (equations 10.22a) gives the input and output energy-potential-
difference holors as functions of input and output energy-transfer holors, whereas
the second set (equations 10.22b) gives the inverse relationship, that is, the input
and output energy-transfer holors as functions of input and output energy-
potential-difference holors. Further study of the remaining four sets will show
that equations (10.22c) and (10.22d) are inverse relations, as are equations (10.22e
and (10.22f). The coefficients of the energy transfer and/or energy potential
difference holors on the right-hand side of equations (10.22a)–(10.22f) are called
the ‘parameters’ of the two-port SISO physical matrixer. Thus, when using equations
(10.22a), one refers to the Zij – parameters of the physical matrixer. Similarly, one
refers to the Y ij – parameters, the Aij and A ji – parameters, the Bij and B ji – parame-
ters, the Hij and H ji – parameters, and the Gij and G ji – parameters of the two-port
SISO physical commutation matrixer.

The two-port SISO physical commutation matrixer parameters

The parameters for any given physical commutation matrixer can be found by either
computations or measurements. The computation, or measurement, to be made is
determined directly from the parameter holor equations. For instance, suppose that
one’s problem is to find the Zij – parameters for a given physical commutation
matrixer. It follows directly from equations (10.22a.1) that

FiA
ZijAa = , (10.22g )
VAj Vaj =0

FiA
ZijAn = , (10.22h)
Vaj VAj=0

Fia
ZijNa = , (10.22i )
VAj Vaj =0

and

10-16
 Mechatronics

Fia
ZijNn = . (10.22j )
Vaj VAj=0

A study of equations (10.22g)–(10.22j) reveals that the four Zij – parameters can be
described as follows:
• ZijAa is the impedance holor seen looking into port A when port a is open;
• ZijAn is a transfer impedance holor. It is the ratio of the port A energy-potential-
difference holor to the port a energy-transfer holor when port A is open;
• ZijNa is a transfer impedance holor. It is the ratio of the port a energy-
potential-difference holor to the port A energy-transfer holor when port a is
open;
• ZijNn is the impedance holor seen looking into port a when port A is open.

One sees, then, that the impedance parameters can be either computed or
measured by first opening port a and determining the ratios FiA /VAj and Fia /VAj,
and then opening port A and determining the ratios FiA /Vaj and Fia /Vaj . Each
parameter is the ratio of an energy-potential-difference holor to an energy-transfer
holor and is therefore an impedance holor with a dimension of ohms (Ω).
Vj ⎫
ji
YaA = AA [℧] ⎪
Fi F a=0 ⎪
i
⎪
ji V j ⎪
YnA = Aa [℧] ⎪
Fi F A=0 ⎪
i
⎬ (10.22k )
ji V j ⎪
YaN = aA [℧] ⎪
Fi F a=0 ⎪
i

j ⎪
[℧] ⎪
ji Va
YnN = a
Fi F A=0 ⎪
i ⎭

FiA ⎫
ji
AaA = ⎪
Fia Vaj
⎪
⎪
An FiA ⎪
Aij =− j [Ω]⎪
Va F a ⎪
i
⎬ (10.22l )
j
ji V ⎪
AaN = Aa [℧] ⎪
Fi V j
A ⎪
j ⎪
Nn VA ⎪
Aij =− j
Va F A ⎪
i ⎭

10-17
 Mechatronics

Fia ⎫
ji
BaA = ⎪
FiA j
VA=0
⎪
⎪
F a ⎪
BijAn = − i j [Ω]⎪
VA F A=0 ⎪
i
⎬ (10.22m)
ji Vaj
⎪
BaN = A [℧] ⎪
Fi V j =0 ⎪
A
⎪
Vj ⎪
BijNn = − aj
VA F A=0 ⎪
i ⎭

⎫
FiA
HijAa = [Ω] ⎪
VAj Fia =0
⎪
⎪
ji F A ⎪
HnA = ia ⎪
Fi V j =0 ⎪
A
⎬ (10.22n )
Na Vaj ⎪
Hij = j ⎪
VA F a=0 ⎪
i
⎪
Vaj
ji
HnN = a [℧] ⎪
Fi V j =0 ⎪
A ⎭

VAj ⎫
ji
GaA = [℧ ] ⎪
FiA V j =0 ⎪
a
⎪
An VAj ⎪
Gij = j ⎪
Va F A=0 ⎪
i
⎬. (10.22o)
ji F a ⎪
GaN = iA ⎪
Fi V j=0 ⎪
a
⎪
Fia
Nn
Gij = j [Ω] ⎪
Va F A=0 ⎪
i ⎭

The remaining port parameters are either computed, or measured, using the same
thought process that one used to find the Zij – parameters. A given port parameter
is obtained by either opening or shortening a port. Furthermore, a given port
parameter will be an impedance, an admittance, or a dimensionless ratio. The
dimensionless ratio will be either the ratio of two energy-potential-difference holors

10-18
 Mechatronics

or the ratio of two energy-transfer holors. These observations are summarised by

equations (10.22k)–(10.22o). The two-port SISO physical commutation matrixer
parameters are also described in relation to the reciprocal sets of holor equations.
The impedance and admittance parameters are grouped into ‘immittance’
parameters. The term immittance holor is used to denote a quantity that is either
an impedance holor or an admittance holor. The Aij and A ji or Bij and B ji
parameters are called the ‘transmission’ parameters because they describe the
energy-potential-difference holor and energy-transfer holor at one end of the two-
port in terms of the energy-potential-difference holor and energy-transfer holor at
the other end.
Thus the Aij and A ji or Bij and B ji parameters give a measure of how the energy-
potential-difference holor and energy-transfer holor are transmitted through the
physical commutation matrixer. The immittance and transmission parameters are
characterised as being the natural choices for relating the port holors. That is,
they relate either energy-potential-difference holors to energy-transfer holors or
matrixer-input holors to matrixer-output holors.
The Hij and H ji and Gij and G ji – parameters, on the other hand, relate cross-
holors, that is, a matrixer-input energy-potential-difference holor and matrixer-
output energy-transfer holor to a matrixer-output energy-potential-difference
holor and matrixer-input energy-transfer holor. Therefore, the Hij and H ji and
the Gij and G ji – parameters are called ‘hybrid’ parameters.

Analysis of the terminated two-port SISO physical commutation matrixer

In the typical application of a two-port physical commutation matrixer, it is driven
at port A and loaded at port a. The holor domain physical commutation matrixer
diagram for the typically terminated two-port physical commutation matrixer is
shown in figure 10.9, where ZijG represents the internal impedance holor of the
internal energy-potential-difference source holor, FiG , and ZijL is the load impedance
holor.
The analysis of the physical commutation matrixer in figure 10.9 involves
expressing the terminal energy-transfer holors and energy-potential-difference holors
as functions of the two-port SISO physical commutation matrix parameters, FiG , ZijG
and ZijL .

Figure 10.9. A terminated two-port SISO physical model of the physical commutation matrixer.

10-19
 Mechatronics

Six characteristics of the terminated two-port SISO physical commutation

matrixer define its terminal behaviour:
• the matrixer-input impedance holor, ZijA = FiA /VAj , or admittance holor, YAji =
VAj /FiA;
• the matrixer-input energy-transfer holor Vaj ;
• the Thévenin energy-potential-difference and impedance holors, VAj(TH), ZijA(TH),
with respect to port a;
• the energy-transfer gain holor, Vaj /VAj ;
• the energy-potential-difference gain holor, Fia /FiA;
• the energy-potential-difference gain holor, Fia /FiG .

To illustrate how these six characteristics are derived, we develop the expressions
using the Zij – parameters to mathematically model the two-port SISO physical
commutation matrixer portion. The derivation of the desired expressions involves
the algebraic holor manipulation of any one of the two-port SISO physical
commutation matrixer holor equations, along with the two constraint holor
equations imposed by the terminations.
If one uses the Zij – parameter holor equations, the four-holor equations that
describe the physical commutation matrixer in figure 10.9 are:
FiA = ZijAaVAj + ZijAnVaj , (10.23)

Fia = ZijNaVAj + ZijNnVaj , (10.24)

FiA = FiG − ZijGVAj, (10.25)

Fia = − ZijLVaj . (10.26)

Expressions (10.23) and (10.24) describe the constraints imposed by the terminations.

Interconnected two-port SISO physical matrixers

In the design of large physical heterogeneous continuous complex dynamical
hypersystems, it is usually easier to synthesise the complex dynamical hypersystem
by first designing simple dynamical systems and/or dynamical hyposystems of them.
The physical heterogeneous continuous complex dynamical hypersystem is then
fabricated by interconnecting these simpler, easier-to-design, smaller physical
commutation matrixer units. If the physical heterogeneous continuous simple
dynamical systems and/or dynamical hyposystems are physically modelled by
two-port physical commutation matrixers, then the synthesis of the physical
heterogeneous continuous complex dynamical hypersystem involves the analysis
of interconnected two-port SISO physical matrixers.
There are five fundamental ways of interconnecting two-port SISO physical
commutation matrixers: (a) cascade, (b) series, (c) parallel, (d) series-parallel and

10-20
 Mechatronics

Figure 10.10. The five basic interconnections of two-port SISO physical commutation matrixers: (a) cascade,
(b) series, (c) parallel, (d) series-parallel and (e) parallel-series.

(e) parallel-series interconnections. These five basic interconnections are illustrated

in figure 10.10.
We only analyse and illustrate the cascade connection in this section. However, in
passing, one can obtain the parameters that describe the interconnected physical
commutation matrixers by the simple addition of the parameters of the individual
physical commutation matrixers.
In particular, the Zij – parameters describe the series connection, the Y ji – parameters
describe the parallel connection, the Hij and H ji – parameters describe the series-parallel
connection, and the Gij and G ji – parameters describe the parallel-series connection.
The interested reader can find a detailed discussion of these four interconnections,
e.g. for electrical commutation matrixers, in the literature (Ruston and Bordogna 1966).
The cascade connection is important because of its frequent occurrence in the
physical modelling of physical heterogeneous continuous complex dynamical
hypersystems.
Unlike the other four basic interconnections, there are no restrictions on using the
parameters of the individual two-port SISO physical commutation matrixers to
obtain the parameters of the interconnected physical commutation matrixers.
The Aij and A ji – parameters are best suited for describing the cascade connections
using the physical commutation matrixer shown in figure 10.11, where one has used
a single prime to denote the Aij and A ji – parameters of the first physical commu-
tation matrixer and a double prime to denote the Aij and A ji – parameters of the
second physical commutation matrixer.
The matrixer-output energy-potential-difference holor and the energy-transfer
holor of the first physical commutation matrixer are labelled FaI and VIa , and the
matrixer-input energy-potential-difference holor and the energy-transfer holor of the

10-21
 Mechatronics

Figure 10.11. Cascade connection of two-port SISO physical commutation matrixers.

second physical commutation matrixer are labelled FiA(I ) and VAj (I ). The problem is
to derive the Aij and A ji – parameter holor equations that relate Fia(I ) and Vaj(I ) to
FiA(I ) and VAj (I ). That is, one seeks the pair of holor equations
ji a
FiA = AaA Fi − AijAn Vaj , (10.27)

a
VAj = AaN
ji
Fi − AijNn Vaj , (10.28)

where the Aij and A ji – parameters are given explicitly in terms of the Aij and A ji –
parameters of the individual physical commutation matrixers.
One begins the derivation by noting from figure 10.11 that
a (I )
ji
FiA = AaA (I ) Fi − AiAn
j
(I )
Vaj(I ), (10.29)

and
a (I ) j
VAj = AaA
ji
(I ) Fi − AijNn(I ) Va(I ). (10.30)

Next, one observes that the interconnection means that Fia(I ) = FiA(I ) and Vaj(I ) = VAj (I ).

10-22
 Mechatronics

When these constraints are substituted into equations (10.29) and (10.30), one
obtains
A(I )
ji
FiA = AaA (I ) Fi − AijAn(I ) VAj (I ), (10.31)

and
A(I )
VAj = AaA
ji
(I ) Fi − AijNn(I ) VAj (I ). (10.32)

The energy-potential-difference holor, FiA(I ), and the energy-transfer holor, VAj (I ),

through the Aij and A ji – parameters of the second physical commutation matrixer,
thus
FiA(I ) = AaA
ji a An(II )
(II ) Fi − Aij Vaj (10.33)

and
Nn(II )
VAj (I ) =AaA
ji a
(II ) Fi − Aij Vaj . (10.34)

One can substitute equations (10.33) and (10.34) into equations (10.31) and (10.32)
to generate the sought-after relationships between FiA, VAj and Fia , Vaj .
One obtains

(
ji
FiA = AaA ji An(I ) ji
(I ) AaA(II ) + Aij )
a
ji
AaA(II ) F i − AaA (
(I ) A
An(II )
ij )
+ AijAn(I ) AijAn(II ) Vaj (10.35)

and

(
VAj = AaA
ji ji Nn(I )
(I ) AaA(II ) + Aij
ji
AaA ) a ji
(
(II ) Fi − AaA(I ) A
Nn(II )
ij )
+ AijNn(I ) AijNn(II ) Vaj . (10.36)

By comparing equations (10.35) and (10.36) with equations (10.27) and (10.28), one
obtains the desired expressions for the Aij and A ji – parameters of the interconnected
physical commutation matrixers, namely,
ji ji ji An(I ) ji
AaA = AaA(I ) AaA (II ) + Aij AaA (II ) , (10.37)

ji An(II )
AijAn = AaA(I ) Aij + AijAn(I ) AijNn(II ) , (10.38)

ji ji ji Nn(I ) ji
AaN = AaA(I ) AaA (II ) + Aij AaA (II ) , (10.39)

ji Nn(II )
AijNn = AaA(I ) Aij + AijNn(I ) AijNn(II ) . (10.40)

One can use equations (10.27)–(10.40) to analyse the cascade connection of the two
physical commutation matrixers. If more than two physical commutation matrixer

10-23
 Mechatronics

units are connected in cascade, the Aij and A ji —parameters of the equivalent two-
port physical commutation matrixer can be found by successively reducing the
original set of two-port SISO physical commutation matrixers one pair a time.

Summary
Physical models of two-port SISO physical commutation matrixers are useful when
it is convenient to describe the performance of the physical commutation matrixer in
terms of its behaviour at two pairs of terminals. Each pair of terminals is referred to
as a port.
The port where the signal is fed into the physically continuous dynamical system
is called the ‘matrixer input’ and the port where the signal is extracted is called
‘matrixer output’.
The physical model of the two-port SISO physical commutation matrixer is
restricted to physical homogeneous continuous dynamical systems where: (i) there
are no independent energy sources in the internal physical commutation matrixer
between the ports; (ii) there is no energy stored in the physical commutation
matrixer; and (iii) the passive components (e.g. b–k–m for a mechanical commuta-
tion matrixer or R–L–C for an electrical commutation matrixer) are linear and time-
invariant. It is also understood that no external connections are made between the
matrixer-input and matrixer-output ports.
The physical model of the two-port SISO physical commutation matrixer is
described in terms of four holors: the matrixer-input—energy-potential-difference
holor (FiA) and energy-transfer holor (VAj), and the matrixer-output—energy-
potential-difference holor (Fia ) and energy-transfer holor (Vaj ).
The mathematical model of the two-port SISO physical commutation matrixer is
formulated in the holor domain so that the physical commutation matrixers and holor
domain are treated as special cases, that is, ω j = 0 for the translational-motion
physical commutation matrixers (e.g. for DC electrical commutation matrixers) and
ω > 0 for sinusoidal steady-state rotary-motion physical commutation matrixers (e.g.
for AC electrical commutation matrixers). Only two of the four holors are independ-
ent, which means that there are six possible ways to relate the four holors.
The six sets of holor equations that group the holors are:
(i) the energy-potential-difference holors as functions of the energy-transfer
holors (Zij – parameters);
(ii) the energy-transfer holors as functions of the energy-potential-difference
holors (Y ji – parameters);
(iii) the matrixer-input port holors as functions of the matrixer-output port
holors (Aij and A ji – parameters);
(iv) the matrixer-output port holors as functions of the matrixer-input port
holors (Bij and B ji – parameters);
(v) the matrixer-input energy-potential-difference holor and the matrixer-out-
put energy-transfer holor as functions of the matrixer output energy-
potential-difference holor and the matrixer-input energy-transfer holor
(Hij and H ji – parameters);

10-24
 Mechatronics

(vi) the matrixer-output energy-potential-difference holor and the matrixer-

input energy-transfer holor as functions of the matrixer-input energy-
potential-difference holor and the matrixer-output energy-transfer holor
(Gij and G ji – parameters).

These six sets of holor equations can be grouped into three sets of reciprocal pairs
and identified by the parameters as:
(i) the immittance parameters (Zij , Y ji );
(ii) the transmission parameters (Aij and A ji , Bij and B ji );
(iii) the hybrid parameters (Hij and H ji , Gij and G ji ).

Because all six sets of holor equations involve the same four holors, all the
parameters are interrelated. Thus if one knows one set of parameters, one can
compute the remaining five sets.
If the two-port SISO physical commutation matrixer satisfies the principle of
reciprocity, it is referred to as a reciprocal two-port SISO physical commutation
matrixer. For reciprocal physical commutation matrixers, special relationships exist
ji ji
among the port parameters. In particular, ZijAn = ZijNa, YaA = YaN , ΔA = Δa = 1,
Na ji An ji
Hij = −HnA and G ij = −G aN .
If a reciprocal two-port SISO physical commutation matrixer is symmetrical, the
matrixer-input and matrixer-output ports can be interchanged without distributing
the values of the port holors. For symmetrical two-port SISO physical commutation
matrixers, ‘ZijAa = ZijNn, YnA
ji ji
= YaN ji
, AijNa = AaA ji
, BijNn = B aA and ΔA = Δa =1.

10.6.2 Physical commutation matrixer having three independent nodes

Consider a set of node holor equations, the basic form of which we have seen is:

V j = Y jiFi . (10.41)

In the case of a physical commutation matrixer with three independent nodes, the
details are:

⎡V j ⎤ ⎡Y ji Y ji Y ji ⎤⎡ F A ⎤
⎢ a ⎥ ⎢ aA aB aC ⎥⎢ i ⎥
⎢V j ⎥ = ⎢Y ji Y ji Y ji ⎥⎢ FiB ⎥ . (10.42)
⎢ b ⎥ ⎢ bA bB bC ⎥⎢ ⎥
⎢⎣Vcj ⎥⎦ ⎢⎣YcAji YcBji YcCji ⎥⎦⎢⎣ FiC ⎥⎦

The solution has the basic form:

δ kj j
Fi = V = zikV i , (10.43)
Y ji

10-25
 Mechatronics

where
⎡ z Aa z Ab z Ac ⎤
⎢ ij ij ij
⎥ ⎡ Δ − ΔBa ΔCa ⎤
⎢ Ba Bb Bc ⎥ 1 ⎢ Aa ⎥
zIJ = zij zij zij = ⎢− ΔAb ΔBb − ΔCb ⎥ ,
⎢ ⎥ Δ
⎢⎣ zijCa zijCb zijCc ⎥⎦ ⎣ ΔCa − ΔCb ΔCc ⎦

where Δ is the determinant of the admittance holor, Y ji , and Δjk is the co-factor of
the admittance holor, y jk .
The terms in the inverse physical-holor matrix have some physical significance, in
that they are driving point and transfer impedance holors. That is, the term zJj is the
impedance holor between the node j and ground when the energy sources connected
to all the other nodes are suppressed (i.e. set to zero), which can be recognised as the
driving point impedance holor discussed in section 10.3.
The term zik is the ratio of the energy-potential difference holor between the node
i and ground to the energy-transfer holor entering the node k when all the other
energy sources are suppressed, and is usually termed the transfer impedance holor. It
is therefore theoretically possible to take appropriate measurements on a physical
heterogeneous continuous dynamical hypersystem to yield values for the entire array
of the driving point and transfer impedance holors so that the impedance-holor
matrix, zij , may be measured.
By inverting it, the impedance-holor matrix, Y ji , of the corresponding node holor
equations can be obtained. Reversal of the procedure to obtain the node holor
equations ‘by inspection’ of the physical commutation matrixer will then produce a
mathematical model of the physical heterogeneous continuous dynamical
hypersystem.
In practice, it is rarely necessary to carry out this inversion process because the
form of the mathematical model can be determined if one knows the number of
holors required. However, this does indicate the form of the measurements that must
be taken on a physical heterogeneous continuous dynamical hypersystem, including
the constraints that must be applied.
It should be evident that if one considers a set of mesh holor equations with the
form
Fi = ZijV j , (10.44)

the solution is
⎡δ i ⎤
⎣ k⎦
V =j
Fi = y ji Fi , (10.45)
Zjk

where y ji is the array of driving point and transfer admittance holors.

Note that there is no simple relation between the y ij (i.e. the driving point and
transfer admittance holors) and Y ji (i.e. the admittance holor array of the node holor

10-26
 Mechatronics

equations). Nor is there any simple relation between the driving point and transfer
impedance holor matrix, zik , and the mesh equation impedance holor matrix, Zij .

10.6.3 Superposition theorem

Returning to equation (10.43), one can write out the solution for FiA as:
FiA = zijAaVaj + zijAbVbj + zijAcVcj. (10.46)

One notes that for this physical heterogeneous continuous dynamical hypersystem
with three energy sources the solution has three components, each of which is the
value of FiA, while the other energy sources are set zero.
The superposition theorem uses the property of a linear physical heterogeneous
continuous dynamical hypersystem to obtain the total sinusoidal-response holor as
the sum of the response holors for each driving function holor or energy source holor
taken one at a time while all the other energy sources are suppressed. As an example,
a physical commutation matrixer consisting of two impedance holors is driven by an
energy-potential-difference holor and an energy-transfer-source holor, as shown in
figure 10.12(a).
Find the energy-potential-difference across, Zij2 . The procedure is simply to obtain
the holors of Fia when only one energy source is acting at a time, and to add those
solutions. In this case, there are two energy sources, so the problem is to find Fia in
each of figures 10.12(b) and 10.12(c) and add them.
From figure 10.12(b), the first component of Fia can be found directly by noting
that this is an energy-potential-difference holor divider,
δ ki
Fia = Zij2FiA.
Z 1jk + Z jk2

From figure 10.12(c), the second component of Fia is found by noting that this is the
energy-potential-difference holor across the parallel combination of Z 1jk and Z jk
2
.

δ ki
Fia = Zij2Zij1Vaj .
Z 1jk + Z jk2

The complete response is the sum of these two expressions:

δ ki δi δi
Fia = Zij2FiA + 1 k 2 Zij2Zij1Vaj = 1 k 2 Zij2(FiA + Zij1Vaj ) .
Z 1jk 2
+ Z jk Z jk + Z jk Z jk + Z jk

Some physical commutation matrixers require more than one energy-potential-

difference source holor. For instance, certain types of physical amplifiers require a
positive and a negative energy-potential-difference source holor for proper oper-
ation, as well as an alternating (e.g. sinusoidal) signal energy source.
In this section, a general method is presented for evaluating series-parallel
physical commutation matrixers having multiple energy-potential difference source

10-27
 Mechatronics

Figure 10.12. Superposition with its physical commutation matrixers.

holors. The method is illustrated with three-energy-source physical commutation

matrixers, but it can be extended to physical commutation matrixers with any
number of energy sources, where the basic principle will be the same.
The superposition theorem is a way to determine energy-transfer holors and
energy-potential-difference holors in a linear physical commutation matrixer that

10-28
 Mechatronics

has multiple energy sources by taking one energy source at a time. The other energy
sources are replaced by their internal impedance holors.
Recall that the ideal energy-potential difference source holor has a zero internal
impedance holor. All energy-potential difference source holors will be treated as
ideal in order to simplify the coverage. A general statement of the superposition
theorem is:

The energy-transfer holor in any given branch of a multiple source linear

physical commutation matrixer can be found by determining the energy-transfer
holors in that particular branch, produced by each energy source acting alone,
with all the other energy sources replaced by their internal impedances.

The total energy-transfer holor in the branch is the holor algebraic sum of the
individual energy-transfer source holors in that branch. The steps for applying the
superposition theorem are as follows.
(i) Take one energy-potential difference source holor at a time and replace
each of the other energy-potential-difference source holors with a short,
which represents the zero internal impedance holor.
(ii) Determine the particular energy-transfer-holor or energy-potential-difference
holor that one wants to find, just as if there were only one energy-potential-
difference source holor in the physical commutation matrixer. This is a
component of the total energy-transfer holor or energy-potential-difference
holor that one is looking for.
(iii) Take the next energy-potential difference source holor in the physical
commutation matrixer and repeat steps (i) and (ii) for each energy source.
(iv) To find the actual energy-transfer-holor in a given branch, add or subtract
the component energy-transfer holors to or from each individual energy
source.
(v) If the energy-transfer holors are in the same sense of direction, add them,
and if they are in the opposite sense of direction, subtract them, with the
sense of direction of the resulting energy-transfer holor being the same as
that of the largest of the individual energy-transfer holors. Once the energy-
transfer holor has been found, the energy-potential difference holor can be
determined by a generalised holor equation.

An example of the dynamical systems approach to superposition is illustrated in

figure 10.13 for a series-parallel physical commutation matrixer with two energy-
potential-difference source holors. Study the steps in this figure.

10.6.4 Reciprocity theorem

We previously noted that when all of the dynamical components in a physical system
or physical commutation matrixer are bilateral, the terms Y jk and Y kj are equal in
the node holor equations, and Zjk and Zkj are equal in the mesh holor equations, and
therefore the admittance-holor and impedance-holor matrices [Y ji ] and [Zij ] are

10-29
 Mechatronics

Figure 10.13. Illustration of the superposition theorem.

symmetrical. As a result, their inverses, [zij ] and [y ji ], are also symmetrical. Although
this is of little value as an aid to analyse a physical commutation matrixer, a
corollary of the reciprocity theorem is of some value in physical modelling.
If a set of measurements on a physical system shows that any pair of transfer
impedance holors or admittance holors is not reciprocal (i.e. zik ≠ zki or y jk ≠ y kj )

10-30
 Mechatronics

the physical heterogeneous continuous dynamical hypersystem cannot be physically

modelled using only bilateral dynamical components and independent energy
sources such as those being considered in this textbook.

10.6.5 Thévenin’s theorem

Thévenin’s theorem is a very convenient method of obtaining a simple physical
commutation matrixer that is equivalent to a section of a complete physical
commutation matrixer. To some extent, it is an extension of the physical modelling
of real energy sources as an ideal energy source associated with an impedance holor.
The section of the physical commutation matrixer to be physically modelled or
simplified will always have at least one energy source and a port (a pair of terminals)
at which it is connected to the remainder of the complete physical commutation
matrixer. This is shown symbolically in figure 10.14(left).
The Thévenin theorem states simply that the physical commutation matrixer
shown in figure 10.14(right) will produce exactly the same energy-transfer holor/
energy-potential-difference holor relationship at port terminals a and n provided that:
• FiTH is the energy-potential-difference holor between a and n when the energy-
transfer holor V j = (0, 0), i.e. the open-output physical commutation matrixer
energy-potential-difference holor Fi ;
• ZijTH is the bivalent impedance holor between port terminals a and n when all
the energy sources within the section of the physical commutation matrixer
are suppressed, i.e. the driving-point impedance holor.

Thévenin’s theorem provides a method for simplifying a physical commutation

matrixer to a standard equivalent form. In many cases, this theorem can be used to
simplify the holor analysis of series-parallel physical commutation matrixers.
The Thévenin equivalent form of any two-terminal impedance-holor physical
commutation matrixer of an equivalent energy-potential-difference-holor source
(FiTH) and an equivalent impedance holor (ZijTH) is arranged as shown in figure 10.14.
The values of the equivalent energy-potential-difference holor and the impedance
holor depend on the values in the original physical matrixer. Any two-terminal

Figure 10.14. The general form of a Thévenin equivalent physical commutation matrixer is an energy-
potential difference source holor in series with an impedance holor.

10-31
 Mechatronics

impedance holor can be simplified to a Thévenin equivalent, regardless of its

complexity.

Thévenin’s equivalent energy-potential-difference holor and equivalent

impedance holor
The equivalent energy-potential-difference holor, FiTH, is one part of the complete
Thévenin equivalent physical commutation matrixer. The other part is an equivalent
impedance holor, ZijTH.
The Thévenin equivalent energy-potential-difference holor (FiTH) is the open-
output (no-load) holor between two specified output terminals in a physical
commutation matrixer. Any dynamical component connected between these two
terminals effectively ‘sees’ FiTH in series with ZijTH, a s defined by Thévenin’s theorem.
The Thévenin equivalent impedance holor (ZijTH) is the total impedance holor
appearing between two specified output terminals in a physical commutation
matrixer with all of the energy sources replaced by their internal impedance holors
(which for ideal energy-potential-difference source holors are zero).

What equivalency means in Thévenin’s theorem

Although a Thévenin equivalent physical commutation matrixer is not of the same
form as the original physical commutation matrixer, it acts the same in terms of the
output energy-potential-difference holor and the energy-transfer holor. This condition
is sometimes known as ‘terminal equivalency’, because both physical commutation
matrixers look the same from the ‘viewpoint’ of the two output terminals.

The Thévenin equivalent of a physical commutation matrixer

To find the Thévenin equivalent of any physical commutation matrixer, one must
determine the equivalent energy-potential-difference holor, FiTH, and the equivalent
impedance holor, ZijTH. For instance, the Thévenin equivalent for the physical
commutation matrixer between output terminals a and b is developed in figure 10.15.
In figure 10.15(a), the energy-potential-difference holor across the designated
terminals a and n is the Thévenin equivalent energy-potential-difference holor. In
this particular physical commutation matrixer, the energy-potential-difference holor
from a to n is the same as the energy-potential-difference holor across Zij2 , because
there is no energy-transfer holor through Zij3 and, therefore, no energy-potential
holor drop across it. FiTH is expressed as follows for this particular example:

δ ki S
FiTH = Zij2F i .
Z 1jk + Z jk2

In figure 10.15(b), the impedance holor between terminals a and n with the energy
source replaced by a short to represent its zero internal impedance holor is the
Thévenin equivalent impedance holor. In this particular physical commutation

10-32
 Mechatronics

Figure 10.15. The simplification of a physical commutation matrixer by Thèvenin’s theorem.

matrixer, the impedance holor from a to b is Zij3 in series with the parallel
combination of Zij1 and ij .
Therefore, ZijTH is expressed as follows:
δ ki
ZijTH = Zij3 + Zij1Zij2.
Z 1jk + Z jk2

The Thévenin equivalent physical commutation matrixer is shown in figure 10.15(c).

Thévenin equivalency depends on the viewpoint

The Thévenin equivalent for any physical commutation matrixer depends on the
location of the two output terminals from between which the physical commutation
matrixer is ‘viewed’. Any given physical commutation matrixer can have more than
one Thévenin equivalent, depending on how the output terminals are designated.
In figure 10.16(a), when FiTH is viewed from between terminals a and c, it is the
energy-potential-difference holor across Zij2 + Zij3 and can be expressed using the
energy-potential-difference-holor divider formula as

10-33
 Mechatronics

Figure 10.16. A physical commutation matrixer Thévenised from two different sets of terminals: (a), (b) and
(c) illustrate one set of terminals and (d) (e), and (f) illustrate another set of terminals. (The FiTH and ZijTH
holors are different for each case.)

δ ki
FiTH
(ac ) =
Z 1jk + Z jk2 + Z jk3
(Zij2 + Zij3)FiS.
Also, as shown in figure 10.16(b), the impedance holor between the terminals a and c
is Zij2 + Zij3 in parallel with Zij1 (the energy source is replaced by a short) and can be
expressed as
δ ki
(ac ) = Zij (Zij + Zij ) =
ZijTH Zij1(Zij2 + Zij3) .
1 2 3
Z 1jk 2 3
+ Z jk + Z jk

The resulting Thévenin equivalent physical commutation matrixer is shown in

figure 10.16(c). When it is viewed from between terminals b and c, as indicated in
figure 10.16(d), FiTH 3
(bc ) is the energy-potential-difference holor across Zij and can be
expressed as
δ ki S
FiTH
(bc ) = 1 2 3
Zij3F i .
Z jk + Z jk + Z jk

As shown in figure 10.16(c), the impedance holor between terminals b and c in

parallel with the series combination of Zij1 + Zij2 :
δ ki
(bc ) = Zij (Zij + Zij ) =
ZijTH Zij3(Zij1 + Zij2 ) .
3 1 2
Z 1jk + Z jk2 + Z jk3

Thévenising a bridge physical commutation matrixer

The usefulness of Thévenin’s theorem can be illustrated when it is applied, for
example, to a bridge physical commutation matrixer. For instance, consider the case

10-34
 Mechatronics

Figure 10.17. A bridge physical commutation matrixer with a load impeder connected between the output
terminals is not a straightforward series-parallel physical commutation matrixer.

where a load impeder is connected to the output terminals of a bridge, as shown in

figure 10.17.
The bridge physical commutation matrixer is very difficult to analyse because it is
not a straightforward series-parallel arrangement when an impeder, such as the very
small impeder of a galvanometer, is connected between the output terminals a and n.
There are no impeders that are in series or in parallel with another impeder. Using
Thévenin’s theorem, one can simplify the bridge physical commutation matrixer
to an equivalent physical commutation matrixer viewed from the load impeder, as
shown step-by-step in figure 10.18. Study the steps in the figure carefully. Once the
equivalent physical commutation matrixer for the bridge is found, the energy-
potential-difference holor and the energy-transfer holor for any value of load
impeder can be determined easily using a generalised Ohm holor equation.

Summary of Thévenin’s theorem

Remember, the Thévenin equivalent physical commutation matrixer for any
impedance-holor physical commutation matrixer is always an equivalent energy-
potential-difference source holor in series with an equivalent impedance holor,
regardless of the original physical commutation matrixer that it replaces.
The significance of a Thévenin theorem is that the equivalent physical commu-
tation matrixer can replace the original physical commutation matrixer as far as any
external load is concerned. Any load impeder connected between the terminals of a
Thévenin equivalent physical matrixer will have the same energy-transfer holor
through it and the same energy-potential-difference holor across it as if it were
connected to the terminals of the original physical commutation matrixer.
The steps for applying Thévenin’s theorem are as follows.
(i) Open the two terminals (remove only load) between which one wants to
find the Thévenin equivalent physical commutation matrixer.

10-35
 Mechatronics

Figure 10.18. Simplifying a bridge physical commutation matrixer with Thévenin’s theorem.

(ii) Determine the energy-potential-difference holor (FiTH) across the two open
terminals.
(iii) Determine the impedance holor (ZijTH) between the two terminals with all
the energy sources replaced by their internal impedance holors. An ideal
energy-potential-difference-source holor is replaced by a short.
(iv) Connect FiTH and ZijTH in series to produce the complete Thévenin
equivalent for the original physical commutation matrixer.

10-36
 Mechatronics

(v) Replace the load removed in step (i) across the terminals of the Thévenin
equivalent physical matrixer. The load energy-transfer holor and the load
energy-potential-difference holor can now be computed using only Ohm’s
law, and they have the same value as the load energy-transfer holor and the
load energy-potential holor in the original physical commutation matrixer.

10.6.6 Norton’s theorem

Like Thévenin’s theorem, Norton’s theorem provides a method for reducing a more
complex physical commutation matrixer to a simpler form. The basic difference is
that Norton’s theorem gives an equivalent energy-transfer-source holor in parallel
with an equivalent impedance holor.
The form of Norton’s equivalent physical commutation matrixer is shown in
figure 10.19. Regardless of how complex the original physical commutation matrixer
is, it can always be reduced to this equivalent form.
The equivalent energy-transfer-source holor is designated VNj , and the equivalent
impedance holor is designated ZijN .
To apply Norton’s theorem, one must know how to find the two quantities VNj
and ZijN . Once one knows them for a given physical commutation matrixer, one
simply connects them in parallel to obtain the complete Norton physical commu-
tation matrixer.

Norton’s equivalent energy-transfer holor

As stated, VNj is one part of the complete Norton equivalent physical commutation
matrixer; ZijN is the other part. VNj is defined as the short-circuit energy-transfer holor
between two points in a physical commutation matrixer. Any dynamical component
connected between these two points effectively ‘sees’ an energy-transfer-holor source
of value VNj in parallel with ZijN . To illustrate, suppose that a holor-impedance
physical commutation matrixer of some kind has an impeder connected between two
points in the physical commutation matrixer, as shown in figure 10.20(a).

Figure 10.19. The form of Norton’s equivalent physical commutation matrixer.

10-37
 Mechatronics

Figure 10.20. Determining the Norton equivalent energy-transfer holor, VNj .

One wishes to find the Norton physical commutation matrixer that is equivalent
to the one shown to be ‘seen’ by the load impedance holor, ZijL . To find VNj , compute
the energy-transfer holor between points a and b with these two points shorted, as
shown in figure 10.20(b).
To determine VNj for the physical commutation matrixer in figure 10.20, short
terminals a and b, as shown in figure 10.20(b). VNj is the energy-transfer holor
through the short and is computed as follows.
First, the total impedance holor seen by the energy-potential-difference source
holor is

δ ki
ZijT = Zij1 + Zij2Zij3.
Z jk2 + Z jk3

The total energy-transfer holor from the energy-potential-difference-source holor is

δ ki S
VTj = ji S
T Fi = YT Fi .
Z jk

10-38
 Mechatronics

Now apply the energy-transfer-holor divider formula to find VNj (the energy-transfer
holor through the short),

δ ki j
VNj = Zij2V T .
Z jk2 3
+ Z jk

This is the value for the equivalent Norton energy-transfer-holor source.

Norton’s equivalent impedance holor

One defines a Norton equivalent impedance holor ZijN in the same way as ZijTH: it is
the total impedance holor appearing between two terminals in a given physical
commutation matrixer with all the energy sources replaced by their internal
impedance holors.
We find ZijN for the physical commutation matrixer in figure 10.20(b). First, we
reduce FiS to zero by shorting it as shown in figure 10.20(b).
Looking in at terminals a and b, one can see that the parallel combination of
Zij and Zij2 is in series with Zij3.
1

Thus,
1 1
ZijN = Zij3 + Zij .
2
Norton’s theorem is a dual of Thévenin’s theorem in that the equivalence for a
section is expressed in terms of an energy-transfer-holor source and a shunting
admittance holor. This is indicated in figure 10.19, where the equivalent is provided.
• VNj is the energy-transfer holor flow when the energy-potential-difference
holor between a and n is zero, i.e. the short-circuit energy-transfer holor.
• YNji is the admittance holor between terminals a and b when all the energy
sources within the section of the physical commutation matrixer are sup-
pressed. i.e. the driving-point admittance holor.

The Norton equivalent can be obtained by series-parallel combinations, by

the solution of mesh holor equations, or by a transformation of the Thévenin
equivalent.
In all the equivalent physical commutation matrixers, one has considered it is
important to note that some transformation is lost when one is working with the
equivalent. The equivalence is valid only at the terminals we have designated a and b.
In effect, these equivalences are convenient methods to eliminate unwanted
physical holors, thereby simplifying some of the mathematical manipulation,
which is required. We have shown above how to find the two equivalent
components of a Norton equivalent physical commutation matrixer, VNj and ZijN
or YNji . Keep in mind that these physical holors can be found for any linear
physical heterogeneous continuous dynamical hypersystem. Once these are
known, they must be connected in parallel to form the Norton equivalent physical
commutation matrixer, as illustrated in figure 10.20

10-39
 Mechatronics

Summary of Norton’s theorem

Any load impeder connected between the terminals of a Norton equivalent physical
commutation matrixer will have the same energy-transfer holor through it and the
same energy-potential-difference holor across it that it would if it were connected to
the terminals of the original physical commutation matrixer.
The steps for applying Norton’s theorem are as follows.
(i) Short the two terminals between which one wants to find the Norton
equivalent physical commutation matrixer.
(ii) Determine the energy-transfer holor (VNj ) through the shorted terminals.
(iii) Determine the impedance holor (ZijN ) or the admittance holor (YNji ) between
the two terminals (opened) with all energy-potential-difference-holor sour-
ces shorted and all the energy-transfer-holor sources opened (ZijN = ZijTH).
(iv) Connect VNj and ZijN or YNji in parallel to produce the complete Norton
equivalent for the original physical matrixer. This is Norton’s equivalent
physical commutation matrixer via the energy source conversion method.

10.6.7 Millman’s theorem

Millman’s theorem permits any number of parallel energy-potential-difference
source holors to be reduced to a single equivalent energy-potential-difference source
holor. It simplifies finding the energy-potential-difference holor across, or the
energy-transfer holor through, a load impeder.
Millman’s theorem gives the same results as Thévenin’s theorem for the special
case of parallel energy-potential-difference source holors. A conversion via
Millman’s theorem is illustrated in figure 10.21.
Millman equivalent energy-potential holor and equivalent impedance holor
Millman’s theorem gives us a formula for computing the equivalent energy-
potential-difference holor, FiEQ. To find FiEQ, convert each of the parallel energy-
potential-difference-holor sources into an energy-transfer-holor source, as shown in
figure 10.22.
In figure 10.22(b), the total energy-transfer holor from the parallel energy-transfer
source holors is
VTj = V1j + V2j + V3j + ⋯ +VXj .

Figure 10.21. Reduction of parallel energy-potential source holors to a single equivalent energy-potential
source holor.

10-40
 Mechatronics

Figure 10.22. Parallel energy-potential source holors converted to energy-transfer source holors.

The total admittance holor between terminals a and b is

YTji = Y1ji + Y2ji + Y3ji + ⋯ + YXji ,

where YTji = δ ki /Z jk
T
, Y1ji = δ ki /Z 1jk , and so on.
Remember, the energy-transfer sources are effectively open. Therefore, by
Millman’s theorem, the equivalent impedance holor, ZijEQ, is the total impedance
holor, ZijT.
δ ki δ ki
ZijEQ = ij = .
YT (δ ki Z 1jk ) + (δ ki Z jk2 ) + (δ ki Z jk3 )+ ⋯ + (δ ki Z jkX )
By Millman’s theorem, the equivalent energy-potential-difference holor is
FiEQ = ZijEQVTj ,

where VTj is expressed as follows:

VTj = V1j + V2j + V3j + ⋯ +VXj

δi δi δi δi
= k
F1 + k
F2 + k
V 3+ ⋯+ k
ViX
Z1 i Z2 i Z3 i ZX
jk jk jk jk

= Y1jiFi1 + Y2jiFi2 + Y3jiFi3+ ⋯ + YXjiFiX .

The following is the formula for the equivalent energy-potential-difference holor:

δ ki
FiEQ =
Y1ij + Y2ij + Y3ij + ⋯ + YXij
(
Y1jiFi1 + Y2jiFi2 + Y3jiFi3+ ⋯ + YXjiFiX . )

10-41
 Mechatronics

The equivalent energy-potential-difference source holor has a polarity such that the
total energy-transfer holor through a load impedance holor will be in the same sense
of direction as in the original physical commutation matrixer.

Reference
Ruston H and Bordogna J 1966 Electric Networks: Functions, Filters, Analysis (New York:
McGraw-Hill)

10-42
 IOP Publishing

Mechatronics
Dynamical systems approach and theory of holors
B T Fijalkowski

Chapter 11
Electrical homogeneous continuous
dynamical systems as DC and AC electrical
commutation matrixers

How would people on the Earth do,

Thrown among the endless stars
Without light and warmth,
Electric beams and magnetic forces?
How would it be? It’s easy to guess:
A gloomy time of chaos and cold would come.
So, sunbeams be praised!
Let’s raise a glass to our Sun!
Adam Mickiewicz

11.1 Introduction
Direct current (DC) is a continuous flow of electricity in one sense of direction
through a wire or conductor. DC is created by generators, such as fuel or
photovoltaic cells; by static electricity and lightning; by ChE/ECh storage batteries;
and by DC commutatorless homopolar generators and MCM or ECM AC–DC
commutator generators.
It flows from a high to a low potential; for example, in a ChE/ECh storage
battery, it flows from a positive to a negative pole. Any device that relies on ChE/
ECh storage batteries—e.g. a flashlight, portable CD player, or laptop computer—
operates on DC. When represented graphically, DC voltage appears as a straight
line, usually flat (see figure 11.1—‘A direct current’).
Alternating current (AC) is electricity that changes its sense of direction at regular
intervals. It builds up to a maximum voltage in one sense of direction and decreases

doi:10.1088/978-0-7503-1350-6ch11 11-1 ª IOP Publishing Ltd 2016

 Mechatronics

Figure 11.1. DC versus AC (EPRI 2006).

to zero; then builds up to a maximum in the opposite sense of direction; and then
returns to zero once more (see figure 11.1—‘An alternating current’). This complete
sequence, or cycle, repeats, and the rate at which it repeats is called the ‘frequency’ of
the current. In Europe, the AC energy provided to a home outlet has a frequency of
50 cps, and in the USA the frequency is 60 cps. This is expressed as 50 or 60 Hz, with
hertz (Hz) being a unit that is equal to one cycle per second (cps).
The steady-state behaviour laws of DC electrical homogeneous continuous
dynamical systems that are energised by DC energy sources are an important area
of study for several reasons.
First, the DC energy generation, high voltage direct current (HVDC) trans-
mission, distribution and consumption of electrical energy occurs under essentially
DC steady-state conditions.
Second, it is possible to predict the behaviour of DC electrical homogeneous
continuous dynamical systems to AC sinusoidal energy sources if the DC behaviour
is understood.
Third, it is often easier to carry out the design of DC electrical homogeneous
continuous dynamical systems based on steady-state DC behaviour. Thus, design
specifications are spelled out in terms of a desired steady-state DC response.
The DC electrical homogeneous continuous dynamical system is designed to meet
these DC characteristics. If these specifications are satisfied, the designer knows that the

11-2
 Mechatronics

DC electrical homogeneous continuous dynamical system will respond satisfactorily to

DC inputs.
In view of the fact that one has been able to analyse relatively simple resistive DC
electrical homogeneous continuous dynamical systems through the intelligent
application of Kirchhoff’s laws in combination with Ohm’s law, this approach
can also be used for all AC electrical homogeneous continuous dynamical systems,
but as the AC electrical homogeneous continuous dynamical systems become
structurally more complicated and involve more and more structural elements,
one will soon find this direct method becoming quite cumbersome.
One purpose is to introduce two powerful techniques of DC and AC electrical
homogeneous continuous dynamical system holor algebra steady-state analysis that
facilitate the holor algebra analysis of sophisticated AC homogeneous continuous
dynamical system’s structures:
• the node-voltage holor method;
• the mesh-current holor method.

In addition to these general holor methods of holor algebra and steady-state

analysis, we will also discuss some additional techniques for simplifying DC and AC
electrical homogeneous continuous dynamical systems. We have already seen how
series-parallel reductions and pi-to-tee (Π2T) or delta-to-wye (Δ2Y) transformations
can be used to simplify a given structure. We now add energy-source transformation
and Thévenin–Norton equivalent DC and AC electrical homogeneous continuous
dynamical systems to our list of simplification holor techniques.
Before we begin our discussion of the node-voltage and mesh-current holor
methods of DC and AC electrical homogeneous continuous dynamical system
steady-state analysis, let us take a breather and think about the foundations that
have been laid to this point.
Firstly, we introduce a voltage holor and a current holor as the parameters of the
impedance holor. These three types of basic structural components of DC and AC
electrical homogeneous continuous dynamical systems were chosen to start our
discussion of DC and AC electrical homogeneous continuous dynamical system
steady-state analysis because all the basic analytical techniques can be explored
using interconnections of these structural components.
Thus, we introduce DC and AC electrical homogeneous continuous dynamical
system steady-state analysis through the direct application of Ohm’s law and
Kirchhoff’s laws. Ohm’s law is critical because it describes the relationship between
the current and the conductors of an impeder, and Kirchhoff’s laws are important
because they describe the constraints imposed on currents and voltages due to the
interconnections of the basic structural components.
One is now ready to introduce the node-voltage and mesh-current holor methods
of DC and AC electrical homogeneous continuous dynamical system steady-state
analysis. Keep in mind that these two holor methods are of interest because they
give us two systematic holor methods for describing DC and AC electrical
homogeneous continuous dynamical systems with the minimum number of simulta-
neous equations.

11-3
 Mechatronics

In order to discuss the more elegant holor methods of DC and AC electrical

homogeneous continuous dynamical system analysis, one must define a few basic
terms that will enable one to give a clear, concise description of the important
features of a given DC and AC electrical homogeneous continuous dynamical
system.
Thus far, all of the DC and AC electrical homogeneous continuous dynamical
systems have been ‘planar’ DC and AC electrical homopolar continuous dynamical
systems, that is, dynamical systems that can be drawn on a plane such that no
branches cross over each other. A DC and an AC electrical homogeneous
continuous dynamical system that can be drawn with branches crossing over each
other is still considered ‘planar’ if it can be redrawn with no crossover branches.
The node-voltage holor method can be applied to both planar and nonplanar
DC and AC electrical homogeneous continuous dynamical systems, whereas in the
nonplanar case the mesh-current holor method may be replaced by a technique
known as the loop-current holor method.
Although we will not discuss the loop-current holor method, we wish to mention
the fact that once one understands the mesh-current holor method, the transition to
the loop-current holor method is not difficult. One has already defined what is meant
by an ideal DC and AC electrical homogeneous continuous dynamical system’s
basic structural components. And when the basic DC and AC electrical homoge-
neous continuous dynamical system components are interconnected to form a DC or
an AC electrical homogeneous continuous dynamical system, the resulting inter-
connections are described in terms of nodes, paths, branches, loops and meshes. One
defined both a node and a closed path, or loop.
One purpose here is to review those definitions and then expand one’s vocabulary
to include the terms path, branch and mesh.
• A ‘node’ is a point in a DC and an AC electrical homogeneous continuous
dynamical system where two or more of its structural elements join.
• A ‘path’ is formed whenever a set of a DC and AC electrical homogeneous
continuous dynamical system’s adjoining basic structural components is
traced, in order, without passing through a connecting node more than once.
• A ‘closed path’, or ‘loop’, is created by starting at a selected node and then
tracing through a set of a DC and AC electrical homogeneous continuous
dynamical system’s connected structural elements in such a manner that one
returns to the original starting node without passing through any intermediate
node more than once.
• A ‘branch’ is a path that connects two nodes.
• A ‘mesh’ is a special type of loop, that is, it does not contain any other loops
within it.

The greatest advantage of both the node-voltage and mesh-current holor methods
is that they reduce the number of simultaneous equations that must be manipulated.
They also require the analyst to be quite systematic in organising and writing the
required simultaneous equations. A first possible approach is to compare the
number of simultaneous equations required for each holor method and then select

11-4
 Mechatronics

the one that requires the fewest. A second is to analyse the location of energy sources
(voltage and/or current sources) within the DC and AC electrical homogeneous
continuous dynamical system structure.
Voltage sources may require extra effort in formulating holor node-voltage
equations, whereas current sources may require extra effort in formulating
mesh-current equations.
Another point to consider when choosing between the two holor methods is what
information about the DC or AC electrical homogeneous continuous dynamical
system being analysed is of primary interest. In other words, a complete solution of a
DC and an AC electrical homogeneous continuous dynamical system may not be
needed, and therefore the particular piece of information that is of interest may
influence what method is used.
Even though the node-voltage and mesh-current holor methods are powerful
techniques for solving DC and AC electrical homogeneous continuous dynamical
systems, one is still interested in holor methods that can be used to simplify DC and
AC electrical homogeneous continuous dynamical systems.
One begins to expand our list of simplifying holor techniques with energy source
transformations. An ‘energy source transformation’ allows one to replace a voltage
source in series with a resistor or impeder by a current source in parallel with the
same resistor or impeder, or vice versa.
If one’s interest in a DC and AC electrical homogeneous continuous dynamical
system’s behaviour is focused on a pair of conductors, the Thévenin and Norton
theorem equivalent DC and AC electrical homogeneous continuous dynamical
systems, which one is about introduce, are also extremely valuable aids in holor
algebra analysis. However, it ought to be noted that Thévenin and Norton theorem
equivalent DC and AC electrical homogeneous continuous dynamical systems can
be used to represent any electrical homogeneous continuous dynamical system
made up of linear structural components. This equivalence will hold for all
possible values of load resistor or impedance. The Thévenin theorem resistor or
impedance is the ratio of the open-circuit voltage holor to the short-circuit
current holor.
The Norton theorem equivalent DC and AC electrical continuous dynamical
system consists of an independent current source in parallel with the Norton
theorem equivalent resistor or impedance. It can be derived from the Thévenin
theorem equivalent DC and AC electrical homogeneous continuous dynamical
system simply by making an energy source transformation. Thus the Norton current
holor equals the short-circuit current holor at the conductive conductors of interest
and the Norton resistor or impedance holor is identical to the Thévenin resistor or
impedance holor.
The most distinguishing characteristic of a linear DC and AC electrical homoge-
neous continuous dynamical system is the principle of ‘superposition’, which states
that whenever a linear DC and AC electrical homogeneous continuous dynamical
system is excited, or driven by more than one independent energy source, one can
find the total response by finding the response of each independent energy source
separately and then summing the individual responses.

11-5
 Mechatronics

Since one is dealing with DC and AC electrical homogeneous continuous

dynamical systems made up of interconnected linear structural components, one
can apply the principle of superposition directly to the holor algebra analysis of such
DC and AC electrical homogeneous continuous dynamical systems when they are
driven by more than one independent energy source.
Summing up, the purpose of this textbook has been to introduce extremely useful
holor algebra techniques of DC and AC electrical homogeneous continuous
dynamical system analysis. The node-voltage and mesh-current holor techniques
are important because they enable us to minimise the number of simultaneous
equations needed to describe a DC and an AC electrical homogeneous continuous
dynamical system. These two techniques are also valuable because they force us to
take a systematic holor approach to writing DC and AC electrical homogeneous
continuous dynamical system equations.
Energy source transformation and Thévenin–Norton equivalent DC and AC
electrical homogeneous continuous dynamical systems—two additional methods of
simplifying DC and AC electrical homogeneous continuous dynamical systems—are
useful when the performance of a DC and an AC electrical homogeneous
continuous dynamical system focuses on its behaviour at a specific pair of
conductors.
The principle of superposition is an important concept when two or more energy
sources are present in the DC or AC electrical homogeneous continuous dynamical
system, because it enables one to isolate the effect of each Thévenin or Norton
source on the total response of the DC or AC electrical homogeneous continuous
dynamical system.
It is also important to keep in mind that although we have used the impedance
holor in AC electrical homogeneous continuous dynamical systems to illustrate the
various analytical techniques, the techniques themselves can also be applied to DC
electrical homogeneous continuous dynamical systems. Thus, in this chapter, we
begin discussing the holor algebra analysis of linear, lumped-parameter, single-
and/or polyphase AC electrical homogeneous continuous dynamical systems that
contain an impedance holor with inductances and capacitances as well as
resistances.

11.2 Electrical homogeneous continuous dynamical systems’ holor

algebra, steady-state DC analysis
11.2.1 DC holor algebra, steady-state analysis
The simplest DC electrical homogeneous continuous dynamical system (figure 11.2)
consists of an electrical energy source (e.g. a ChE/ECh storage battery, a DC
homopolar generator, or a DC heteropolar generator, i.e. an electrical commutation
matrixer (ECM) AC–DC commutator generator), a receiver (or simply load) of an
electrical energy (e.g. an electrical lamp, electrical heating device, DC homopolar
motor, or DC heteropolar motor, i.e. an ECM DC–AC commutator motor), and
four conductive collectors connecting the receiver and energy-source single port (two
terminals). The energy source transforms chemical, mechanical, thermal or other

11-6
 Mechatronics

Figure 11.2. Simple DC electrical commutation matrixer.

energy into electromagnetic energy, which, for brevity, is often simply called
‘electrical energy’; the electrical energy receiver (load), by contrast, transforms the
electrical energy to other forms of energy: mechanical, thermal, radiant and so on.
The energy source, together with the conductors and receiver (load) connected to it,
forms a closed loop, along which an uninterrupted stream of electrical charges (an
electrical current) flows. This is called a DC ‘electrical commutation matrixer’. With
respect to the energy source, the conductive collectors and receiver (load) form the
so-called ‘external circuit’.
The direct cause of the flow of electrical current in an electrical commutation
matrixer is the self- or mutual-induced voltage source of the power source.
Conventionally, the sense of the current direction in the DC electrical commuta-
tion matrixer is taken to coincide with the sense of the induced voltage (in figure
11.2 the senses of these directions are shown by arrows). The terminal of the
voltage source through which the current ‘leaves’ for the external circuit is called
the ‘positive’ (+) one, and the terminal through which the current ‘enters’ the
voltage source from the external circuit is called the ‘negative’ (−) one. The sense of
the current direction is thus ‘from to’ in the external circuit and ‘from to’ inside the
voltage source. Both the external circuit and the voltage source possess resistance,
the value of which depends upon the material, shape and dimensions of the
conductive collectors comprising these parts of the electrical matrixer. The
magnitude or intensity of the current (or, simply, the current) in the electrical
circuit depends on the magnitude of the induced voltage and the resistance of the
whole DC electrical commutation matrixer. The relationship between these three
quantities is expressed by Ohm’s law, which plays a very important role in
electrical engineering, being the basis of many practical calculations. According
to this law, the current holor I j in an electrical commutation matrixer is directly
proportional to the voltage-source holor Ei and ‘inversely’ proportional to the

11-7
 Mechatronics

resistance holor Rij (or proportional to the conductance holor G ji ) of the whole DC
electrical commutation matrixer:
δ ki
Ij = Ei = G jiEi , (11.1)
Rjk
where the conductance holor G ji matrix is the inverse of the resistance holor Rjk matrix:
G ijRjk = δ ki ,
and moreover, the current holor I j , the voltage-source holor Ei , the conductance
holor G ji and the resistance holor Rij matrices can be written directly from
relationships:
⎡ c⎤
I j = (I c,0) = ⎢ I ⎥;
⎣ 0⎦
⎡ c⎤
Ei = (E c,0) = ⎢ E ⎥;
⎣ 0⎦
⎡R 0 ⎤
Rij = ⎢ ;
⎣ 0 R ⎥⎦
⎡ ⎤
G ji = ⎢G 0 ⎥ .
⎣ 0 G⎦

Ohm’s law can be applied to the DC electrical commutation matrixer as a whole

and, separately, to any portion of the DC electrical commutation matrixer outside
the energy source. In this case, the induced voltage-source holor Ei in equation (11.1)
must be replaced by the voltage holor across the portion Vi (i.e. the difference in
potential between the ends of the portion), while Rij is the resistance holor of the
given portion and not of the whole DC electrical commutation matrixer. Thus,

δ ki
Ij = Vi = G jiVi , (11.2)
Rjk

i.e. the current holor in the given portion of the DC electrical commutation matrixer
is directly proportional to the voltage across it, and inversely proportional to the
resistance holor Rij , or proportional to the conductance holor G ji of the portion.
If the internal resistance holor of an energy source is significant, then the voltage
holor (or potential-difference holor) Vi across the port terminals of the induced
voltage-source holor Ei is less than its induced-voltage holor Ei by the ‘internal
voltage-drop’ holor, which, according to Ohm’s law, is equal to the product of the
current holor I j multiplied by the ‘internal resistance’ holor RijE of the induced
voltage-source holor Ei , i.e.
Vi = Ei − RijE I j . (11.3)

Often the internal resistance holor RijE of the induced voltage-source holor Ei is
relatively low, and the last term on the right-hand side of equation (11.3) may be

11-8
 Mechatronics

disregarded. In this case, the voltage holor Vi of the energy source is approximately
equal to the induced voltage-source holor Ei .
When the external circuit is open, i.e. under so-called ‘no-load conditions’ of the
induced voltage source, I j = (0,0), and, irrespective of the magnitude RijE , equation
(11.3) gives
Vi = Ei . (11.4)
If the resistance holor Rij of the external circuit (‘external resistance holor’) becomes
low compared with the internal resistance holor RijE of the induced voltage source,
then one may say that Vi = (0,0) and that the energy source is short-circuited, in
which case, according to equation (11.3), the current holor
δ ki
Ij = Ei = G EjiEi (11.5)
R jkE

can reach a very high value, since RijE is usually many times less than the resistance
of the loads. In certain computations, it is more convenient to use the reciprocal
multiplied by the Kronecker delta of the resistance holor R jkE , i.e. the conductance
holor G Eji .

11.2.2 Node-voltage holor method

The direct use of Kirchhoff’s laws for determining the currents in complicated DC
electrical commutation matrixers demands the simultaneous solution of a consid-
erable number of equations, making the computations extremely time-consuming.
However, there are a number of methods, based on the same Kirchhoff laws that
obviate the solving of a set of equations or reduce the number of equations and,
hence, significantly cut the computations.
One of these is the ‘node-voltage method’, which may be applied when the DC
electrical commutation matrixer contains only two nodes (figure 11.3), where the
internal conductance (or resistance) holors of the energy sources are assumed to be
included in the conductances (or inductances) of the branches, or may be reduced to

Figure 11.3. DC electrical commutation matrixer with two nodes.

11-9
 Mechatronics

such a form by converting deltas into wyes or vice versa. The number of branches N
connecting the two nodes may be arbitrary.
The voltage-source holor ViAN between the nodes (the ‘node-voltage holor’) may
be considered as the difference between the induced voltage-source holor EiE of any
( jth) branch (which also includes the internal voltage-drop holor RijE I j in the
resistance holor RijE of that branch (which also includes the internal resistance holor
of the energy source)), i.e.
ViAN = EiE − RijE I j
if positive senses of directions from node N to node A are chosen for all current
holors. Whence the current holor in the jth branch is
δ ki
Ij =
R jkE
( ) (
EiE − ViAN = G Eji EiE − ViAN . ) (11.6)

But, according to Kirchhoff’s first law (KFL), the sum of the currents meeting at
node A must be equal to zero, i.e.
δ ki
R jkE
( ) (
EiE − ViAN = G Eji EiE − ViAN = 0, )
or
δ ki E δ ki AN
E i = Vi ; G EjiEiE = G EjiViAN ,
R jkE R jkE
whence
⎡V AN ⎤ G jiE E G E + G 2E2 + ⋯ + GN EN
ViAN = ⎢ c ⎥ = E jii = VcAN = 1 1 . (11.7)
⎣ 0⎦ GE G1 + G 2 + ⋯ + GN
Equation (11.7) gives us the node voltage; it is then easy to find the required current
from equation (11.6). If the induced voltage of one of the branches has a reverse
sense of direction to that shown in figure 11.3 or is missing altogether, then it must be
included with a minus sign or omitted entirely in equations (11.6) and (11.7),
respectively. If in one of the branches EiE < ViAN , then, according to equation (11.6),
the current holor I j will be negative, i.e. sensed in direction from node A to node N.
In this case, it is more convenient to write equation (11.6) in the form
δi
( )
I j = kE ViAN − EiE = G Eji ViAN − EiE
R jk
( )
in order to obtain the absolute magnitude of the required current.

11.2.3 Loop-current holor method

A quite different application occurs in electrical commutation matrixer theory (Staff
1940, Kron 1959, Bewley 1961). If a DC electrical commutation matrixer with

11-10
 Mechatronics

several energy sources has more than two nodes, the currents in it may be found by
the ‘method of loop-currents’. This is one of the most universal methods for solving
DC electrical commutation matrixers.
In a number of cases, a DC electrical commutation matrixer can be considered as
consisting of a set of adjoining loops (two in figure 11.4), each of which forms a
polygon made up of several branches of the DC electrical commutation matrixer
(without any diagonals).
Some branches of the DC electrical commutation matrixer are common to two
adjacent loops, while others form an external circuit where each branch only occurs
in one loop.
The required branch current holor I J of the DC electrical commutation matrixer
must satisfy a set of holor equations based on KFL. If a certain current holor I J ,
known as the ‘loop’ holor, is assigned to each loop in an arbitrary sense of direction,
and the actual current in the common branch of two adjacent loops is regarded as
the algebraic sum of the corresponding loop currents, then KFL will be automati-
cally satisfied, since all the loop currents in one branch flow towards a node, and in
the other branch they flow away from the node.
When determining the loop currents it is only necessary that the actual currents
should satisfy Kirchhoff’s second law (KSL). The DC electrical commutation matrixer
shown in figure 11.4 consists of, e.g. two loops and consequently has two loop-current
holors: IAj and `Iaj . The current in any external branch will be equal in magnitude to the
corresponding loop-current holor (as the only current in that branch) and the sign will
be dependent on whether its sense of direction is the same as that of the loop-current
holor or the opposite. If for all j loops one writes holor equations based on KSL, the
result will be a set of j equations with j unknown loop-current holors. The solutions of
these equations will first give all the loop-current holors and then all the actual-current
holors. For instance, figure 11.4 shows a two-loop DC electrical commutation
matrixer.
The applied voltage holors are ViA, and Via and the resulting current holors are IAj
and `Iaj . The holors are related by
Vi = Rij I J , (11.8)

Figure 11.4. Two-loop DC electrical commutation matrixer.

11-11
 Mechatronics

where the voltage holor Vi , the current holor I J and the resistance holor Rij in matrix
notation are

⎡ ⎤
⎢ VcA ⎥
⎡ A⎤ ⎢ ⎥
0 ⎥ ⎡V A⎤
Vi = ⎢Vi ⎥ = ⎢
⎢ V a ⎥ ⎢ a ⎥ ⎢⎣V a ⎥⎦
= ;
⎣ i ⎦
⎢ Vc ⎥
⎢⎣ 0 ⎥⎦
⎡ c ⎤
⎢ IA ⎥
⎡ I j⎤ ⎢ 0 ⎥ ⎡ I ⎤
IJ = ⎢ j ⎥ = ⎢ ⎥ = ⎢ A⎥;
A
⎢⎣ Ia .⎥⎦ ⎢ c ⎥ ⎣ Ia .⎦
⎢ Ia ⎥
⎢⎣ 0 ⎥⎦
⎡ R Aa R An ⎤ ⎡ Aa
Rij = ⎢ Na
ij ij
⎥ = ⎢R R An ⎤ = ⎡ R1 + R3 −R3 ⎤ .
⎥ ⎢ ⎥
⎢⎣ R ij R ij ⎥⎦ ⎣ R
Nn Na
R Nn ⎦ ⎣ −R3 R2 + R3⎦

Thus, the loop linear matrix equation is

⎡V A⎤ ⎡ R Aa R An ⎤⎡ IA⎤ ⎡ R1 + R3 −R3 ⎤⎡ IA⎤
⎢⎣V a ⎥⎦ = ⎢⎣ Na ⎥⎢ ⎥ = ⎢ ⎥⎢ ⎥ ,
R R Nn ⎦⎣ Ia ⎦ ⎣ −R3 R2 + R3⎦⎣ Ia ⎦

or the loop linear algebraic equations are

V A = R AaIA + R AnIa = (R1 + R3)IA − R3Ia;

V a = R NaIA + R NnIa = −R3IA + (R2 + R3)Ia .

A DC electrical commutation matrixer of any number of loops may also be expressed

by equation (11.8), although, of course, the matrix Rij will be different for each DC
electrical commutation matrixer. In the usual problem of the holor analysis of an
N-loop DC electrical commutation matrixer (Moon and Spencer 1986), Rij are given
and the current holors I k are to be determined. The result can be written directly from
equation (11.8):

I k = G EkiVi , (11.9)

where the conductance holor G ji is the inverse of the resistance holor Rik matrix:

G kiRij = δ jk .

Evidently, the customary problem of DC electrical commutation matrixer holor

analysis does not involve any question of coordinate transformation (Moon and
Spencer 1986), so Vi , I J and, Rij are merely holors

11-12
 Mechatronics

Vi = (V A, V B, … ,V N );
I J = (Ia , Ib, … , In);
⎡ R Aa R Ab … R An ⎤
⎢ ⎥
⎢ RBa RBb … RBn ⎥
Rij = ⎢ .
⋮ ⋱ ⋮⎥
⎢ ⋮ ⎥
⎣ R Na R Nb … R Nn ⎦

For a DC electrical commutation matrixer consisting of N loops, one uses equation

(11.8), namely:
Vi = Rij I J ,
or

⎡ V A⎤ ⎡ R Aa R Ab … R An ⎤⎥⎡ Ia ⎤
⎢ B ⎥ ⎢ Ba ⎢ ⎥
⎢V ⎥ = ⎢ R RBb … RBn ⎥⎢ Ib ⎥ .
⎢ ⋮⎥ ⎢ ⋮ ⋮ ⋱ ⋮⎥⎢ ⋮⎥
⎢⎣V N ⎥⎦ ⎢ Na ⎥⎢ ⎥
⎣R R Nb
… R Nn ⎦⎣ In ⎦

Thus, for an N-loop DC electrical commutation matrixer, one obtains the following
set of linear algebraic loop equations:
V A = R AaIa + R AbIb+⋯+ R AnIn;
V B = RBaIa + RBbIb+⋯+ RBnIn
… … … ……………………;
V N = R NaIa + R NbIb+⋯+ R NnIn.
The solution of which, for an arbitrary current In , has the form
ΔN
In = , (11.10)
Δ
where the determinant is

R Aa R Ab ⋯ R An
RBa RBb ⋯ RBn
Δ= . (11.11)
⋮ ⋮ ⋱⋮
R Na R Nb ⋯ R Nn

Moreover, the determinant ΔN is obtained from Δ by substituting the applied

voltages V A, V B, … ,V N for the resistances of the Nth column.
The determinants Δ and ΔN may be computed by rules that may be found in reference
books on mathematics. Equation (11.10) serves to determine any loop current. The
extension to AC electrical commutation matrixers is obvious. In AC electrical commu-
tation matrixer holor synthesis, however, the equation of transformation may arise.

11-13
 Mechatronics

11.2.4 Superposition method

The determinant ΔN in equation (7.26) can be given in the form:
ΔN = ΔNa E A + ΔNbE B + ⋯ +ΔNmE M + ⋯ +ΔNnE N
by expanding it in terms of components of the Nth column.
Here, the algebraic complements ΔNm are obtained from the determinant if one
strikes out the Nth column and the mth row and multiplies the determinants thus
obtained by (1)N+m.
Substituting this expression for ΔN into equation (11.10), one obtains
ΔNa A ΔNb B ΔNc C ΔNn N
IN = E + E + E +⋯+ E . (11.12)
Δ Δ Δ Δ
Every term on the right-hand side of this expression represents the current in the
loop, provided applied voltage is present in one loop and absent in the other loops.
Thus, equation (11.12) shows the current in any loop and, hence, in any branch it
can be considered as the superimposition of the partial currents arising in that
branch due to each applied voltage in the loop, and, consequently, to each applied
voltage in the branch separately. This is known as the ‘superposition principle’, and
represents a special case of the principle of the independent action of forces, which is
widely used in physics.
The superposition principle can serve as the basis for a method of solving
electrical commutation matrixers—the ‘superposition method’.
In a DC electrical commutation matrixer of many energy sources, determinations
are successively made of the partial currents produced by the action of each applied
voltage (the other applied voltages each time being put equal to zero).
Next, the partial currents produced by separate applied voltages in the same
branches are algebraically added (superimposed). In this way, one finds the resultant
currents produced by the combined action of all the applied voltages. For instance,
in the DC electrical commutation matrixer in figure 11.5(a), which contains two
energy sources, the applied induced voltage Eia is initially eliminated (but the
internal resistance R Ea of the energy source is retained).
This makes it possible to consider the DC electrical commutation matrixer as one
series parallel connected matrixer component and to then simply determine all the
partial currents I1(I ), I2(I ) and I3(I ) caused by the action of the applied induced voltage
EiA alone (figure 11.5(b)).
Then in a similar manner one determines the partial currents I1(II ), I2(II ) and I3(II )
produced by the single applied induced voltage Eia (figure 11.5(c)).
Finally, determinants produced by the independent actions of the applied induced
voltages are superimposed, and the resultant currents in the DC electrical commu-
tation matrixer (which contains two energy sources) are determined:
I1 = I1(I ) − I1(II ); I2 = I2(I ) − I2(II ); I3 = I3(I ) − I3(II ).

However, the superposition method has a serious shortcoming: the partial currents,
which flow in different senses of directions in a given branch, are often several times

11-14
 Mechatronics

Figure 11.5. Solution by superposition method.

greater than the resultant current. Because of this, a small error (percentage-wise) in
computing the partial currents leads to an impermissibly great error in the
magnitude of the resultant current. For this reason, the superposition-holor method
is used comparatively rarely.

11.2.5 The DC energy source

A DC ‘energy source’ produces electrical energy. It may act as a ‘voltage source’
(independent or dependent) that produces a voltage constant with time, or as a
‘current source’ (independent or dependent) that produces a current that is also

11-15
 Mechatronics

constant with time. We use the constant function throughout in the discussion
below.

11.2.6 The generation of DC energy

A current or voltage is termed ‘direct’ if it does not periodically change its sense of
direction and magnitude with time. DC energy is obtained from AC–DC static or
rotary converters (AC–DC rectifiers) or an MCM AC–DC commutator generator,
or an AC commutatorless alternator with an ECM AC–DC commutator acting as
the AC–DC rectifier.
An MCM AC–DC commutator generator is an electrical machine that converts
mechanical energy into electrical energy. For its operation, an MCM AC–DC
commutator generator depends upon electromagnetic induction. In an MCM
AC–DC commutator generator, the armature 1 (figure 11.6), which carries a
winding, is driven by a prime mover in the magnetic field of the north (N) and
south (S) poles of electromagnet 2. The AC induced voltage generated in the
armature winding is collected by means of an MCM AC–DC commutator 3 and
brushes, and delivered into an external circuit. The current delivered into the
external circuit is DC because of the action of the MCM AC–DC commutator.
The prime mover may be a steam engine, a steam turbine, a gas turbine, an
external or internal combustion engine, a windmill, an AC commutatorless
synchronous motor or an MCM or ECM DC–AC commutator synchronous or
induction motor.
In 1889, the noted US inventor Thomas Alva Edison (1843–1931) wrote in
Scientific American that ‘My personal desire would be to prohibit entirely the use of
alternating currents. They are as unnecessary as they are dangerous, I can therefore
see no justification for the introduction of a system which has no element
of permanency and every element of danger to life and property.’ Edison’s
19th-century electrical energy distribution system relied on DC energy generation,
delivery and use. This pioneering system, however, turned out to be impractical and
uneconomical, largely because in the 19th century DC energy generation was limited

Figure 11.6. Diagram of an MCM AC–DC commutator generator (Kuznetsov 1960).

11-16
 Mechatronics

to a relatively low voltage potential and DC energy could not be transmitted above a
small amount for more than a kilometre or mile.
Edison’s electrical energy plants had to be local affairs, sited near the load, or the
load had to be brought close to the DC commutatorless homopolar generator or
MCM AC–DC commutator heteropolar generator.
AC energy distribution was far superior for the needs of a robust electrical
infrastructure. Unlike DC electrical energy, the voltage of AC could be stepped up
with relatively simple EE transformer devices for distance transmission and
subsequently stepped down for delivery to appliances and equipment in the home
or factory. As a consequence, the Hungarian-born US electrical engineer and
inventor Nikola Tesla’s (1856–1943) invention of a relatively simple polyphase
AC commutatorless hyposynchronous (induction) motor meant end users needed AC,
which could be generated at large central electrical energy plants for high-voltage bulk
delivery over long distances. (See section 11.3.3 for more on the attributes of AC
energy and DC energy, and why AC energy originally prevailed.)
Despite a vigorous campaign against the adoption of alternating current, Edison
could not overcome the shortcomings of his DC energy system. The AC energy
system won out, and today utilities generate, transmit and deliver electricity in the
form of AC energy.
Despite the fact that high-voltage direct current (HVDC) is now a viable means of
long-distance electrical energy transmission and is used in above a hundred
applications worldwide (Woodford 1998, Barthold 2006), no one is advocating a
wholesale change of the infrastructure from AC–DC, as this would be wildly
impractical.
Although in Edison’s time, DC was impractical for transmission beyond a
distance of a kilometre or mile, today HVDC can transmit bulk energy over very
long distances and also enables the interconnection of incompatible power grids.
Electrical valves that can convert high voltage from AC–DC and back again were
needed for HVDC to work, so such conversion was enabled with the development of
high-voltage ECM AC–DC/DC–AC commutators (static converters) with unilateral,
mercury-arc electrical valves capable of handling high voltages.
This technology was first deployed in 1954 by the noted Swedish electrical
engineers Uno Lamm and Gunnar Engström (ASEA), for a transmission system
between the island of Gotland and the Swedish mainland. The system was rated
at 20 MW and 100 kV and transmitted power over 91 km (57 miles).
Development in the 1960s of high-voltage unipolar semiconductor-controlled
electrical valves, called ‘thyristors’, helped boost transmission capacity, increas-
ing the cost-effectiveness of HVDC. Advances in the technology continue apace,
with refinements in microprocessor control and other developments enhancing
performance.
Over 100 HVDC systems, including many in Europe, Africa, Asia, Asia–Pacific,
and North and South America, are now in service around the world. The largest, in
Brazil, is rated at 600 kV. According to ABB, a key supplier of HVDC systems,
newer designs have expanded the power range, extending the economic power range
of HVDC systems from 90 MW to 1 GW (EPRI 2006).

11-17
 Mechatronics

HVDC can be the most economical option in some cases. Depending on the
distances involved, an overhead DC transmission line and towers may be less
expensive per unit of length than an overhead AC line. Although AC–DC/DC–AC
converter stations are more costly than AC line terminating stations, in some cases
transmission lengths as low as 640 km (400 miles) are sufficient to make the DC
energy system more economical. The ‘break even’ distance needed for economical
deployment of HVDC may be less when submarine or underground DC cable
transmission is involved. In these cases, AC cable transmission can be no longer than
about 48 km (30 miles), but distances of 288 km (180 miles) have been achieved with
HVDC cables, and systems covering over 1.122 km (700 miles) are in the planning
stages (EPRI 2006).
HVDC cables may make it easier and more economical to site high-voltage lines.
Although electrical energy demand in certain areas may increase, installing
high-voltage lines, or electrical energy plants near to load centres, typically meets
with public opposition. Using compact HVDC underground cables as city infeeds
from remote areas may be a means to overcome the political problems associated
with building such new systems.
Moreover, in some circumstances, by upgrading or replacing existing AC trans-
mission lines with HVDC ones, the electrical energy transfer capability of existing
rights of way may be substantially increased.
Greater electrical energy flow control may be afforded to marketers and system
operators. Various restructuring and deregulation schemes have forced the electrical
energy system to operate in ways that it was not designed to. The electrical energy
system was, in effect, designed to operate as a private toll road, and is now expected
to operate as an open-access highway, handling millions of transactions daily
(Galvin Electricity Initiative 2005).
AC networks do not easily accommodate the desired electrical energy flow
control for today’s system, so power marketers and system operators may require
the electrical energy flow control capability provided by HVDC transmission.
HVDC may be the only feasible means of interconnecting electrical energy net-
works. HVDC can interconnect two separate electrical energy networks so that
electrical energy can be exchanged between them. Two AC energy systems can be
connected by installing an AC–DC/DC–AC converter station in each system, with
an interconnecting superconductive DC link between them, making it possible to
transfer electrical energy even though the AC energy systems so connected remains
asynchronous. For instance, an AC energy system may not be synchronized to
neighbouring networks, even though the physical distance between them is quite
small. This is the case in Japan, which has asynchronous networks: half the country
is served by a 60 Hz network, and the other half by a 50 Hz system. It is physically
impossible to connect the two together by direct AC methods—only DC works.
HVDC can be a barrier to cascading blackouts. Perhaps one of the most vivid
examples of a benefit of HVDC is related to its effect on grid reliability. HVDC lines
cannot be overloaded and energy flow can be controlled for grid stability. This was
illustrated during the ‘Northeast blackout’ of 2003, when the Quebec/Canada grid
was unaffected, since it is interconnected to the neighbouring system with HVDC.

11-18
 Mechatronics

As a result, more HVDC interconnections between asynchronous networks such as

those in the east and west of the USA and in Texas have been advocated. Additional
HVDC links are being considered, most notably the Trans-America Generation Grid
(TAGG), a connection of the eastern and western networks (EPRI 2006).
However, a new debate is arising over AC versus DC: should DC energy delivery
systems displace or augment the AC energy distribution system in buildings or other
small, distributed applications?
Edison’s original vision for a system that has DC generation, energy delivery
and end-use loads may come to fruition—at least for some types of installations.
Facilities such as data centres, campus-like groups of buildings, or building
hyposystems may find a compelling value proposition in using DC energy. Several
converging factors have spurred the recent interest in DC energy delivery. One of
the most important is that an increasing number of microprocessor-based
electronic devices use DC energy internally, converted inside the device from
standard AC supply. Another factor is that new distributed resources such as
solar photovoltaic cells (PVC) and fuel cells (FC) produce DC energy, while
ChE/ECh storage batteries and other technologies store it. So why not have a DC
energy distribution system as well? Why not eliminate the equipment that converts
DC–AC for electrical energy distribution and then back again to DC at the
appliance?
Advocates point to the greater efficiency and reliability of a DC energy delivery
system. Eliminating the need for multiple conversions could potentially prevent
energy losses of up to 35%. Less waste heat and a less complicated conversion system
could also potentially translate into lower maintenance requirements, longer-lived
system components and lower operating costs. In a larger context, the deployment of
DC energy delivery systems as part of AC–DC ‘hybrid’ buildings—or as a DC
energy micro-grid ‘island’ that can operate independently of the bulk electrical
energy grid—could enhance the reliability and security of the electrical energy
system.

Benefits and drivers of DC energy delivery systems

Several potential benefits are driving new-found interest in DC energy delivery
systems.
• Increasingly, equipment operates on DC, requiring conversion from AC
electrical energy sources. All microprocessors require DC and many devices
operate internally on DC energy since it can be regulated precisely for
sensitive components. Building electrical systems are fed with AC that is
converted to DC at every fluorescent ballast, and then there are computer
systems and other electronic devices. As one specialist electronics manufac-
turer put it, ‘DC is the blood of electronics’ (Savage 2003). AC–DC
conversions within these devices waste power. The electrical energy supplies
that convert high-voltage AC power into the low-voltage DC energy needed
by the electronic equipment used in commercial buildings and data centres
typically operate at roughly 65–75% efficiency, meaning that 25–35% of all
the electrical energy consumed is wasted.

11-19
 Mechatronics

About half the losses are from AC–DC conversions, the rest from stepping
down DC voltage in DC–DC conversions. Simply getting rid of the losses
from AC–DC conversion could reduce energy losses by about 10–20%.
Likewise, an increasing number of portable gadgets such as cell phones and
personal digital assistants (PDAs) require an AC–DC adapter, which also
results in electrical energy losses during conversion. Considered in aggregate,
the millions of AC–DC conversions necessitated for the operation of
electronics extract a huge electrical energy loss penalty.
• Distributed generation systems produce DC energy. Many distributed gener-
ation electrical energy sources, such as PVCs and FCs, and advanced energy
storage systems (ChE/ECh storage batteries, flywheels and ultra capacitors),
produce electrical energy in the form of DC energy. Other devices can also be
suited to DC output, such as Ch-Th-FM microturbines and PM wind turbines.
Even hybrid electric vehicles (HEV) could serve as MCM AC–DC commutator
generators in emergencies with the right equipment to connect them to the
electrical energy system. The electrical energy losses entailed in converting DC–
AC energy for distribution could be eliminated with DC energy delivery,
enhancing efficiency and reliability and system cost-effectiveness. For instance,
EPRI Solutions estimated that the total lifecycle cost of PVC electrical energy
for certain DC applications could be reduced by more than 25% compared to
using a conventional DC–AC approach—assuming that the specific end-use
applications are carefully selected (EPRI PEAC Corp 2002, Barker et al 2003,
Crudele et al 2004). The costs of new distributed generation sources such as
PVCs are still high, so the optimisation of designs with DC energy delivery may
help spur adoption and efficient operation.
• Storage devices such as ChE/ECh storage batteries, flywheels and capacitors
store and deliver DC energy. This again helps to avoid unnecessary con-
versions between AC and DC.
• DC electrical energy could help power hybrid electric vehicles, transit buses, and
commercial fleets. Plug-in hybrid electric vehicles (PHEV) can travel greater
distances on electricity than today’s hybrids since they have larger ChE/ECh
storage batteries. These batteries store DC energy, so charging them with
electricity from solar PVCs and other distributed electrical energy sources could
reduce reliance on gasoline, enhancing security and emergency preparedness.
• DC energy delivery could potentially enhance electrical energy efficiency in
data centres, a pressing need. One of the most promising potential applica-
tions of DC energy delivery is in data centres, which have densely packed
racks of servers that use DC electrical energy. In such centres, AC is
converted to DC at the uninterruptible power supply (UPS) to facilitate
storage, and then converted back again to push out to the servers, before
being converted one more time to DC at each individual server. These
conversions waste power and generate considerable heat, which must be
removed by air-conditioning systems, resulting in high electricity costs. For
instance, in a 10–15 MW data centre, as much as 2–3 MW may be lost
because of electrical energy conversions (EPRI 2006). As these centres install

11-20
 Mechatronics

ever more dense configurations of server racks, DC energy delivery systems

may be a means to reduce their rocketing electrical energy needs.
• Improved ECM DC–AC commutators, acting as DC–AC inverters and macro-
electronics, allow DC energy to be converted easily and efficiently to AC energy
and to different voltage levels. Component improvements enable greater
efficiency than in the past, and improve the economics of hybrid AC–DC
systems. Although improved macro-electronics also enhance AC-only elec-
trical energy systems, such enabling technology makes the DC energy
delivery option feasible as well.
• The evolution of central electrical energy architecture in computers and other
equipment simplifies DC energy delivery systems. At present, delivering DC
energy to a computer requires input at multiple voltages to satisfy the
electrical energy requirements of various internal components (RAM, pro-
cessors, etc). The development of central electrical energy architecture, now
underway, will enable the input of one standardised DC voltage at the port,
streamlining delivery system design.
• DC energy delivery may enhance micro-grid system integration, operation and
performance. A number of attributes make DC energy delivery appealing for
use in micro-grids. With DC energy distribution, solid-state commutation
(switching) can quickly interrupt faults, making for better reliability and
power quality. If tied into the AC transmission system, a DC energy micro-
grid makes it easy to avoid back-feeding surplus generation and fault
contributions into the bulk utility system (by the use of an ECM AC–DC
commutator acting as an AC–DC rectifier that only allows unilateral (one-
way) electrical energy flow). In addition, in a low-voltage DC energy system,
such as would be suitable for a home or group of homes, a line of a given
voltage rating can transmit much more DC energy than AC energy.
Of course, while DC electrical commutation matrixers are widely used in
energy-consuming devices and appliances, DC energy delivery systems are not
commonplace, and therefore face the obstacles any new system design or
technology must overcome. For any of the benefits outlined above to be realised,
testing, development and demonstration are needed to determine the true
potential and market readiness of DC energy delivery, as outlined in (EPRI 2006).

High-density computer environments, utilising AC energy supply blade server

technology, are presenting major challenges to operators and owners. Typical
facility infrastructures cannot accommodate the added heat loads and reliability
requirements of such installations (Baldwin and Geary 2006).
Servers equipped with DC energy supplies, instead of AC energy supplies, operate
with 30–40% less heat, reduce electrical energy consumption up to 30%, increase server
reliability, offer flexibility to installations and decrease maintenance requirements.
Utilising current off-the-shelf (OTS) components, it is possible to install and operate a
high-reliability DC energy distribution system for a DC powered server environment—
without ChE/ECh storage batteries or complicated AC energy system drawbacks.

11-21
 Mechatronics

High-reliability DC energy distribution systems could be the next-generation

electrical energy systems for server environments. The benefits of DC energy systems
are as follows (Baldwin and Geary 2006).
• Lower component count equals higher system efficiencies, greater reliability,
lower maintenance costs and lower total costs.
• They are modular and flexible: systems can grow with load requirements.
• They are environmentally flexible: system front-end components can be
located in non-conditioned spaces or on the raised floor near the load.
• Flywheel energy storage eliminates the need for ChE/ECh storage batteries.
• Universal starline busway provides a double end feed feature to permit
redundant DC energy sources at critical loads.
• No downstream static or transfer switches are required, as voltage-matched
DC systems can inherently be coupled together.
• DC energy distribution eliminates harmonics.
• There is simplified grounding—positive grounding or negative grounding.
• A very advanced technology (VAP) maintenance management system con-
tributes to higher system reliability and cost savings by managing equipment
documentation effectively.
• System configuration is a unique application and partnership of existing and
proven commercially off-the-shelf (COTS) equipment.
• This DC energy system has no requirement for an uninterruptible power
system (UPS) in order to provide high system reliability with utility power
outage ride through.
• DC energy distribution eliminates power factor concerns.
• The heat load at server racks is reduced by 20–40%.
• Electrical energy consumption is reduced by up to 30%.
• Electrical energy reliability is increased by as much as 27%.

Another information technology (IT) application that may lend itself to DC power
delivery is supply for power over the Ethernet (PoE). According to the Institution of
Electrical and Electronics Engineers (IEEE), PoE technology allows internet
protocol (IP) telephones, wireless local area network (LAN) access points and other
appliances to receive power as well as data over LAN cabling without needing or
modifying the existing ethernet infrastructure.
The LAN cables carry both data and power, just as traditional telephone lines
carry both voice and power on the same lines. PoE can supply power to computer
phones (also known as voice over internet protocol (VoIP)), as well as other devices,
such as web cameras, electronic badge readers and even electric guitars and other
musical instruments. The voltage for PoE is 48 V, with about 12 to 15 W of available
power at the device. A DC–DC converter transforms the 48 V to the lower voltages
needed for electronics. The power infrastructure must be able to support the
increased power requirements of PoE.
In Cisco Systems (2006) it is stated that the overall power budget for both the
switch and powered devices is 5–6 kW. The ultimate source of the power may be
AC, but as with the increased power requirements of data centres, facility managers

11-22
 Mechatronics

must consider how best to deliver this energy, taking into consideration component
costs, efficiency and cooling.
Cisco Systems (2006) noted that its technology can support DC power delivery
for the network that is suitable ‘for installations where highly available power
delivery is critical and where an investment in DC power infrastructure is considered
to be a business benefit.’

11.3 Electrical homogeneous continuous dynamical systems holor

algebra, steady-state AC analysis
11.3.1 Sinusoidal AC holor algebra, steady-state analysis
The steady-state behaviour laws of AC electrical homogeneous continuous dynamical
systems energised by AC sinusoidal energy sources are an important area of study for
several reasons. First, the generation, transmission, distribution and consumption
of electrical energy occur under essentially AC sinusoidal steady-state conditions.
Second, it is possible to predict the behaviour of AC electrical homogeneous
continuous dynamical systems to non-sinusoidal energy sources if the sinusoidal
behaviour is understood. Third, it is often easier to carry out the design of AC
electrical homogeneous continuous dynamical systems based on steady-state sinus-
oidal behaviour. Thus, the design specifications are spelled out in terms of a desired
steady-state sinusoidal response. The AC electrical homogeneous continuous dynam-
ical system is designed to meet these sinusoidal characteristics. If these specifications
are satisfied, the designer knows that the AC electrical homogeneous continuous
dynamical system will respond satisfactorily to non-sinusoidal inputs.

11.3.2 The sinusoidal AC energy source

A sinusoidal AC electrical energy source produces electrical energy. It may act as a
sinusoidal voltage source (independent or dependent) that produces a voltage that
varies sinusoidally with time, or as a sinusoidal current source (independent or
dependent) that produces a current that also varies sinusoidally with time. One can
express a sinusoidally varying function using either the sine function or the cosine
function. There is no clear-cut choice for either function. Although either works
equally well in sinusoidal steady-state analysis, it is important to recognise that both
functional forms cannot be used simultaneously. We will use the sine function
throughout in the discussion below.

11.3.3 The generation of AC

A current or voltage is termed ‘alternating’ if it periodically changes its sense of
direction and magnitude. The complete period of a current variation is usually
termed simply the period T of the AC, and the number of such periods per second
f = T −1 (11.13)
is termed the frequency. The unit of frequency is s−1, cycles per second (cps), and it is
sometimes, especially in Europe, termed hertz (Hz).

11-23
 Mechatronics

In a more restricted sense, AC is a periodically varying current, the average value of

which, over a period, is zero. The early electrical energy systems developed by Thomas
Edison generated and delivered DC. However, DC energy systems had many
limitations; most notably, the power typically could not be practically transmitted
beyond a distance of a little more than a kilometre or approximately a mile.
Moreover, because changing the voltage of DC was extremely inefficient, the delivery
of DC energy in Edison’s time meant that separate electric lines had to be installed to
supply power to appliances and equipment of different voltages, an economically and
physically impractical approach. Another limitation was that DC current incurred
considerable power losses. Edison’s concept for electrification of the USA—which
included royalties from his patents on DC energy systems—was to deploy relatively
small-scale individual DC energy plants to serve small areas, such as Pearl Street
Station, which powered a part of New York City’s financial district (EPRI 2006).
By contrast, the US inventor George Westinghouse’s (1846–1914) polyphase AC
energy system—invented by Nikola Tesla and used with EE transformers developed
by the US electrician William Stanley, Jr (1858–1916), who also worked for
Westinghouse—proved to be far superior technically and economically. The voltage
of AC could be stepped up or decreased to enable long-distance power transmission
and distribution to end-use equipment.
Thomas Alva Edison fought vociferously against the use of AC-based systems,
which he claimed would be dangerous because of the high voltage at which power
would need to be transmitted over long distances. Edison even went so far as to
demonstrate the danger of AC by using it to electrocute a Coney Island elephant
named Topsy, who had killed three men. He also electrocuted numerous cats and
dogs procured from neighbourhood boys. However, despite proving that AC could
be an effective means of electrocuting these hapless creatures, its superiority to DC
for transmission and distribution was compelling. AC energy can be produced by
large ME generators, and the AC voltage can be stepped up or down for
transmission and delivery. The distance limitation of DC energy and the difficulties
of changing DC voltages proved critical factors in abandoning DC energy systems in
favour of those based on AC.
With DC energy systems, electrical energy had to be generated close to where it
was used. This resulted in problematical reliability and economics. If the local plant
failed, the entire system was down. And because initial DC energy systems were
devoted to lighting loads and the systems only generated electrical energy at times of
high usage, the cost of energy was high—often more than $1 (US dollar) per kWh
when adjusted for inflation to present dollars (2005)—compared to an average cost
for residential electricity today of 8.58¢ (US cents, $0.0858) per kWh (Energy
Market Profiles database 2006).
Engineers wanted to interconnect systems to improve reliability and overcome the
economic limitations of DC energy systems. If one area’s power were out because of
a problem at the ME generator, then the adjacent town would be available to pick
up the load. In addition, by interconnecting isolated systems, a greater diversity of
load was obtained, which would improve load factor and enable more economical
operation of the generation plants.

11-24
 Mechatronics

Another major driver was the desire to make use of hydro-electrical (HE) energy
sources located far from urban load centres, which made long-distance transmission—
and therefore AC—essential. By using EE transformers, the AC voltage can be
stepped up to high levels so that electricity can be distributed over long distances at
low currents, and hence with low losses. EE transformers that could efficiently adjust
voltage levels in different parts of the system and help minimise the inherent energy
losses associated with long-distance distribution were a critical enabling technology
that led to today’s AC-dominated energy distribution system. EE transformers do not
work with DC energy. William Stanley of the Westinghouse Company (IEEE
HISTORY CENTER 2006) first demonstrated effective EE transformers in 1886.
Stanley first demonstrated the potential of EE transformers to enable AC energy
transmission at Main Street in Great Barrington, Massachusetts, USA.
Similar AC energy systems that used EE transformers eventually replaced
Thomas Edison’s DC energy transmission systems. Stanley’s installation in Great
Barrington was the first such system to include all of the basic features of large
electrical energy systems, as they still exist more than one hundred years later.
Other factors led to the preference for AC energy transmission instead of DC
energy delivery—most notably a desire for large-area grids relying on centralised
electrical energy plant, such as HE dams. Having a transmission and distribution
system that could provide HE to cities or to remotely located industries was also an
economic imperative. Such development relied not only on EE transformers, but
also on the development of polyphase AC commutatorless synchronous generators.
Up to the end of the 1870s, AC was not used. Soon after, AC became widely used
in Europe, especially in France. The first AC energy generation plant was built at
Ames, Colorado, USA. Electricity produced here in the spring of 1891 was
transmitted 4 km (2.6 miles) over rugged and at times inaccessible terrain to provide
electrical energy to operate the EM motor-driven mill at the Gold King mine. This
pioneering demonstration of the practical value of transmitting electrical energy was
a significant precedent in the USA for much larger plants at Niagara Falls (in 1895)
and elsewhere. Electricity at Ames was generated at 3 kV, 133 Hz, single-phase AC,
by a 74 kW (100 hp) Westinghouse AC commutatorless synchronous alternator
(EPRI 2006).
In 1891, the eminent Polish electrician, Mikhail Dolivo-Dobrovolsky (1862–1919),
was the genius who designed the first long-distance three-phase AC energy transmission
in the world. The electrical energy transmission was from Lauffen am Neckar waterfall
(near Heidelberg, Germany), where there was a hydro-mechanical (HM) turbine with a
three-phase AC commutatorless synchronous generator and a step-up EE transformer,
to Frankfurt am Main, the site of the 1891 International Electro-Technical Exhibition.
Here, there was a step-down three-phase EE transformer that powered a three-phase
AC commutatorless hyposynchronous (induction) EM motor, which drove a
mechano-hydraulic (MH) pump (see figure 11.7).
In this three-phase AC energy transmission system, the three-phase EE trans-
formers, the three-phase AC commutatorless hyposynchronous (squirrel-cage) EM
motor and the system itself were all invented by Dolivo-Dobrovolsky. Here, 149 kW
(200 hp) electrical energy was transmitted over a distance of 175 km (109 miles) at a

11-25
 Mechatronics

Figure 11.7. The three-phase AC commutatorless synchronous generator installed at Lauffen am Neckar,
Germany. Credit: Historisches Museum, Frankfurt, Germany.

line voltage of 15 kV and a frequency of 30–40 Hz. The electrical energy trans-
mission efficiency was 75% (Hoseman et al 1999). This was the first long-distance
transmission of electrical energy for utility. This event also confirmed the first
efficient long-distance transmission system that used a three-phase AC dynamical
system.
The confirmation of the effectiveness of a three-phase AC energy transmission
system was a sign of the end of the ‘war of currents’. This episode confirmed the
dominance of Dobrovolsky’s three-phase AC energy transmission system over other
systems. Three-phase AC energy transmission systems would prove over the next 20
years to be better practice for transmitting electrical energy than Edison’s DC and
Tesla’s two-phase AC energy transmission systems.
In the generation, transmission and distribution of electrical energy, AC has two
basic advantages over DC.
• The possibility (through using EE transformers) of increasing and reducing
voltage simply and economically; this is especially important for long-
distance electrical energy transmission.
• Simple design and, hence, cheaper electrical machines.

At present, AC of various frequencies is applied in diverse fields of electrical

engineering. The highest frequencies are used in radio and television: from 100 kHz
(long waves) up to 20 MHz (short waves), and then even up to several GHz
(centimetric waves). A wide frequency band (from 500 Hz to 50 MHz) is used in
various electrical furnaces.
One finds frequencies of the order of hundreds and thousands of cycles per second
in telephone networks. Frequencies of 150–2000 Hz are used in powering high-speed
AC EM motors. The lowest frequencies 50 or 60 Hz are used for lighting and for
powering EM motors with rotation speeds not exceeding 3000 or 3600 revolutions
per minute (rpm), respectively. Thus, electric railways use 15 or 162/3–50, or 25–60 Hz
AC. However, such low frequencies cannot be used for lighting purposes, since at 15,
162/3, or 25 Hz, the light intensity varies with the current variation, producing a
noticeable effect on the eye at frequencies below 40 cps.

11-26
 Mechatronics

On the other hand, as will be shown later, the voltage losses in AC energy
transmission increase with increasing frequency. For this reason, a standard
frequency of 50 Hz has been established for all power stations in Europe. Electric
power stations in the USA operate at frequencies of 25 and 60 Hz.
At present, electrical-valve oscillators usually obtain high-frequency alternating
currents. On the other hand, rotary or linear AC commutatorless ME generators
produce industrial 50 or 60 Hz AC. Alternating currents of higher frequency (from
500 Hz to 10 kHz) are obtained from specially designed AC commutatorless
generators and/or electrical-valve DC–AC, AC–DC–AC, or AC–AC commutators
(converters).
The basis of AC generation is the law of electromagnetic induction. For instance,
rotary AC commutatorless generators are ordinarily designed to produce the AC in
the armature phase windings of the stationary part of the electric machine (the
‘stator’). These phase windings are fitted into ‘slots’ into the inner surface of the
stator (figure 11.8), and the separate leads are interconnected in such a way that the
energy-source voltages produced in them combine.
The moving part of the AC generator (the ‘rotor’) is a flywheel with north (N) and
south (S) magnetic pole pieces arranged alternately around the rim. The magnetic-
field excitation windings for these magnetic poles are connected to two conductive
commutators’ ‘slip rings’, mounted on the shaft but insulated from it. Carbon
brushes against the conductive commutators’ slip rings supply DC to the magnetic-
field windings. The DC is obtained from AC–DC converters (rectifiers) or a separate
MCM AC–DC commutator generator, termed a ‘DC exciter’, often mounted on the
same shaft as the AC commutatorless alternator. Use is also made of AC generators
with interior permanent magnets (IPM), which, of course, do not have any rotor
magnetic-field windings.
One turn of a two-pole AC commutatorless generator’s rotor corresponds to one
cycle of an AC energy-source voltage in the stator armature phase windings, and if
the rotor makes n revolutions per second (rps), the frequency of the induced energy-
source voltage is
f = n. (11.14)

Figure 11.8. Schematic cross-section of a six-pole AC generator (Kasatkin and Perekalin 1970).

11-27
 Mechatronics

Since the angular velocity of the AC ME generator is, in this case,

ω = 2πn ,
a simple relationship exists between it and the frequency of the induced energy-
source voltage:

ω = 2πf = 2π / T . (11.15)

For an AC ME generator with p pairs of magnetic poles,

f = pn. (11.16)

Independently of the AC generator construction, the value ω determined from

equation (11.15) is termed the ‘angular frequency’ of the AC. Thus, the angular
frequency is the angular velocity of a two-pole AC generator, the energy-source
voltage of which has a frequency f. The unit of angular frequency is s−1.
The standard 50 or 60 Hz frequency corresponds to an angular frequency

ω = 2πf = 2π 50 = 314 s−1,

or

ω = 2πf = 2π 60 = 377 s−1.

11.3.4 Sinusoidal alternating current

The alternating energy-source voltage of a commutatorless AC generator and the
AC can have highly diversified variation waveforms, which in large measure depend
on constructional factors, for instance, the shape of the rotor magnetic-pole pieces.
The most important periodic waveforms are the sine or cosine waveforms, which
are the only periodic functions with derivatives like themselves. The technical
significance of this fact is that the shape of the waveforms of voltages and currents
(figure 11.9) is the same in all the components of an AC electrical continuous
dynamical system, i.e. sinusoidal or harmonic.
The analytical expressions for the sinusoidal functions of instantaneous values of
the voltages and currents are

v(t ) = Vmax sin(ωt + ψv ) = 2 Vrms sin(ωt + ψv )⎫⎪

⎬, (11.17)
i (t ) = Imax sin(ωt + ψi ) = 2 Irms sin(ωt + ψi )⎪
⎭

where the small letters v(t) and i(t) denote the instantaneous values of voltage and
current, i.e. their values at an arbitrary time t, the capital letters Vmax and Imax denote
the corresponding amplitudes, i.e. the maximum values, Vrms = V and Irms = I
denote the corresponding effective values, and ψv and ψi denote the corresponding
phase angles, depending on the values of the voltage and current at the instant
chosen as the time origin (t = 0).

11-28
 Mechatronics

Figure 11.9. Sinusoidal waveform of voltage or current.

If one takes the initial time to be when the current passes through its maximum,
sin(ωt + ψi ) at t = 0 must be equal to unity:
sin(ωt + ψi ) = 1 and ψi = π /2.
The various stages of the processes described by the sinusoidal functions, namely,
growth, decay, passage through zero or a maximum, and so on, are termed phases.
The phase of a process at any time t is defined by the angle ωt + ψ of the sine and is
termed the ‘phase angle’.
The angle ψ that determines the phase at a given initial instant (t = 0) is termed
the ‘initial phase’, and the difference between the initial phases ψv and ψi of the two
functions is termed the ‘phase difference’ or ‘phase shift’.

11.3.5 Sinusoidal voltage and current waveforms—holor representation

In an AC electrical homogeneous continuous dynamical system one usually deals
with sinusoidal voltage and current waveforms that are all of the same frequency,
but that may be shifted from each other in time. We start by discussing sinusoidal
voltage and current waveforms. Very simple sinusoidal voltage and current
instantaneous values from using the trigonometric identity are (Fijalkowski 1967):

v (t ) = 2 U sin(ωt + ψv ) = 2 V [sin(ωt cos ψv + cos ωt sin ψv )]⎫⎪

⎬, (11.18)
i (t ) = 2 I sin(ωt + ψi ) = 2 I [sin(ωt cos ψi + cos ωt sin ψi )] ⎪
⎭

in which ψv is the voltage phase angle and ψi is the current phase angle.
Now, one must be able to convert equations (11.18) of the forms

v (t ) = 2 [Vs sin ωt + Vc cos ωt ]⎫ ⎪

⎬
⎪
. (11.19)
i (t ) = 2 [I s sin ωt + I c cos ωt ] ⎭

11-29
 Mechatronics

Then, comparing equations (11.19) with equations (11.18), one obtains

Vc = V cos ψv , Vs = V sin ψv . (11.20)

The quantities Vc and Vs are independent elements (merates) of an effective value of

the voltage holor
Vi = (Vc , Vs ). (11.21)
The above-written holor fully represents a sinusoidal voltage.
Analogically, a sinusoidal current can be represented by an effective value of the
current holor
I j = (I c , I s ), (11.22)
where
I c = I cos ψi , I s = I sin ψi . (11.23)

Thus, the sinusoidal voltage holor Vi and/or the current holor Ij are defined as
mathematical entities or ‘wholes’ built from two independent elements (merates) Vc
and Vs, as well as Ic and Is, which carry the effective values and phase angle
information of sinusoidal functions to the doublet (dipole) functions of two
independent quantities that cannot be added directly.
Due to the fact that independent elements of the voltage holor as well as the
current holor depend solely on the effective values and phase angles of these
quantities, then the voltage holor and the current holor can be represented in the
same way as in complex numbers, namely:
Vi = V ∠ψv (11.24)

and
I j = I ∠ψi . (11.25)

As in equations (11.21)–(11.25), a holor quantity will be represented throughout the

text by a boldface letter.
In studying equations (11.24) and (11.25), we observe that the shorthand
abbreviation ∠ in the angle notation of the phase angles belongs to holors that
carry the effective values and phase angles of the given sinusoidal functions. We use
this notation extensively in the material that follows. These holors are by definition
the ‘holor representation’, or ‘holor transform’, of the given sinusoidal functions.
Thus, the holor transform transfers the sinusoidal function from the time domain to
the holor domain. Thus far, we have emphasized moving from the sinusoidal
function to its holor transform. However, it should be apparent that one can also
reverse the process. That is, given the holors, one can write the expressions for the
sinusoidal functions.
Thus, if one is given that Vi = V ∠ψv and I j = I ∠ψi , the expressions are
v(t ) = 2 U sin(ωt + ψv ) and i (t ) = 2 I sin(ωt + ψi ), since one has decided to use
the sine function for all sinusoids.

11-30
 Mechatronics

Observe that one cannot deduce the value of ω from the holor. The holors only
carry the magnitude of the effective values and the phase angle information. The
step of going from the holor transform to the time-domain expression is referred to
as finding the inverse holor transform and is formalised by the equation

−1{V ∠ψu} = 2 {V ∠ωt + ψv}⎫⎪

⎬, (11.26)
−1{I ∠ψi} = 2 {I ∠ωt + ψi} ⎪
⎭

where the notations −1{V ∠ψv} and −1{I ∠ψi} are read as the inverse transforms of
V ∠ψu and I ∠ψi .
Equation (11.26) tells us that to find the inverse holor transform one multiplies the
holors by 2 {V ∠ωt} and 2 {I ∠ωt} and then extracts the second merates Vs and Is
of the products, respectively. The holor transform is so useful in electric system
analysis because it reduces the problem of finding the maximum amplitude and
phase angle of the steady-state sinusoidal response to the algebra of holors. In a
linear electrical continuous dynamical system driven by sinusoidal energy sources,
the steady-state response is also sinusoidal.
The holor transform is also useful in electric system analysis because it applies directly
to the sum of sinusoidal functions. Since electrical continuous dynamical system analysis
involves summing voltages and currents, the importance of this observation is obvious.
The holor representation of the sum is the sum of the holors of the individual
independent elements (merates). The superiority of the holor approach to adding
sinusoidal functions should be apparent. One should note that the holor approach
requires the ability to move back and forth between the polar and rectangular forms.

11.3.6 The passive AC electrical-homogenous-continuous-dynamical-system

components in the holor domain
The systematic application of the holor transform in AC electrical homogeneous
continuous dynamical system analysis requires two steps.
First, one must establish the relationship between the current holor and the
voltage holor of the passive AC electrical homogeneous continuous dynamical
system’s components.
Second, one must develop the holor-domain version of Kirchhoff’s laws, which
we discuss in the next section. In this section, we establish the relationship between
the current holor and the voltage holor at the conductive collectors of the resistor,
inductor and capacitor. We begin with the resistor. We use the passive sign
convention in all the derivations. It follows directly from Ohm’s law that if the
current in a resistor varies sinusoidally with time, that is, if i (t ) = 2 I sin(ωt + ψi ),
then the voltage at the conductive collectors of the resistor (figure 11.10) will be
i (t ) = R 2 I sin(ωt + ψi ), (11.27)

where Imax = 2 I is the maximum amplitude of the current in amperes (A) and ψi is
the phase angle of the current.

11-31
 Mechatronics

Figure 11.10. A resistive AC electrical-homogeneous-continuous-dynamical-system’s dynamical component

carrying a sinusoidal AC.

Figure 11.11. The holor domain equivalent AC electrical homogeneous continuous dynamical system for a
resistor.

The holor transform of this voltage holor is

Vi = Rij I ∠ψi . (11.28)

However, I ∠ψi is the holor representation of the sinusoidal current holor; thus, one
can write equation (11.28) as
⎡ ⎤ c
Vi = Rij I j = ⎢ R 0 ⎥ I s , (11.29)
⎣ 0 R⎦ I

which tells one that the voltage holor at the conductive collectors of a resistor is
simply the resistance holor Rij times the current holor.
An AC electrical homogeneous continuous dynamical system diagram for a
resistor in the holor domain is shown in figure 11.11.
This phase relationship is shown in figure 11.12.
The signals are said to be in time phase because they both reach corresponding
values on their respective waveforms at the same time (for example, they are at their
positive maxima at the same instant).

11-32
 Mechatronics

Figure 11.12. A resistive AC electrical-homogeneous-continuous-dynamical-system’s dynamical component

carrying a sinusoidal AC.

The relationship between the current and voltage at the conductive connectors of
the inductor is derived by assuming a sinusoidal current and using Ldi/dt to establish
the corresponding voltage.
Thus, given i (t ) = 2 I sin(ωt + ψi ), one sees that the expression for the voltage is
di ( t )
v (t ) = L = ωL 2 I cos(ωt + ψi ). (11.30)
dt
Now one rewrites equation (11.30) using the sine function:
di ( t )
v (t ) = L = ωL 2 I sin(ωt + ψi + π /2) . (11.31)
dt
The holor representation of the voltage holor given by equation (11.31) is
⎡ ⎤ ⎡ ⎤
Vi = XijLI j = ⎢ 0 −ωL ⎥I ∠ψi + π /2 = ⎢ 0 −ωL ⎥I j . (11.32)
⎣ ωL 0 ⎦ ⎣ ωL 0 ⎦
From equation (11.32), one sees that the voltage holor at the conductive conductors
of an inductor equals the inductive-reactance holor XijL times the current holor.
The holor domain equivalent AC electrical homogeneous continuous dynamical
system is shown in figure 11.13. The phase relationship between the current holor
and the voltage holor at the conductive conductors of an inductor is contained in
equation (11.32), from which we can see that the voltage and current will be out of
phase by exactly π /2.
In particular, the voltage will lead the current by π /2 or, what is equivalent, the
current will lag the voltage by π /2. This concept of ‘voltage leading current’ or
‘current lagging voltage’ is made clear by the plot shown in figure 11.14, where one
notes that as time increases from zero, the voltage comes to a particular instanta-
neous value on the sine function ahead of the current.
For instance, it reaches its negative peak exactly π /2 before the current reaches its
negative peak.

11-33
 Mechatronics

Figure 11.13. The holor domain equivalent AC electrical homogeneous continuous dynamical system for an
inductor.

Figure 11.14. An inductive AC electrical-homogeneous-continuous-dynamical-system’s dynamical component

carrying a sinusoidal AC.

The same observation can be made with respect to the ‘zero-going positive’
crossing or the positive peak. Thus, one says that the voltage holor ‘leads’ the
current holor by π /2 or, what is equivalent, the current holor ‘lags’ behind the holor
voltage by π /2. The phase shift can also be expressed as a period T in seconds (s).
A phase shift of π /2 corresponds to a quarter of a period, hence the voltage holor
leads the holor current by T/4, or 1/4f s.
The relationship between the current and voltage at the conductive conductors of
a capacitor follows directly from the derivation of equation (11.32).
That is, if one notes that for a capacitor,
dv ( t )
i (t ) = C , (11.33)
dt
and then assume
v (t ) = 2 V sin(ωt + ψu ),

11-34
 Mechatronics

it follows that
δ ki c ⎡ −ωC ⎤ Vc
I j = YCjiVi = V = Is = ⎢ 0 .
0 ⎦⎥ Vs
C i ⎣ (11.34)
Z jk I ωC

Now, if one writes equation (11.34) for the voltage holor as a function of the current
holor, one obtains
⎡ 1 ⎤
⎢ 0 ⎥
Vc ωC ⎥ I c
Vi = XijC I j = =⎢ . (11.35)
Vs ⎢ 1 ⎥ Is
−
⎢⎣ ωC 0 ⎥⎦

From equation (11.35), one sees that the equivalent AC electrical continuous
dynamical system for the capacitor in the holor domain is as shown in figure 11.15.
The voltage holor across the conductive conductors of a capacitor will lag behind
the capacitor current holor by exactly π/2.
One can easily see this by rewriting equation (11.35) as
⎡ 1 ⎤
⎢ 0 ⎥
Vc ωC ⎥ I c
Vi = XijC I j = =⎢ , (11.36)
Vs ⎢ 1 ⎥ Is
−
⎢⎣ ωC 0 ⎥⎦

where I j = I c I s T = I ∠ψi − π /2.

The alternative way to express the phase relationship contained in equation
(11.36) is to say that the current holor leads the voltage holor by π /2.
The phase relationship between the current holor and voltage holor at the
conductive conductors of a capacitor is shown in figure 11.16.
We conclude our discussion of the passive AC electrical-homogeneous-continu-
ous-dynamical-system’s dynamical components in the holor domain with a very
important observation. When one compares equations (11.29), (11.32) and (11.35),
one notes that they are all of the form

Figure 11.15. The holor domain equivalent AC electrical continuous dynamical system for a capacitor.

11-35
 Mechatronics

Figure 11.16. A capacitive AC electrical-homogeneous-continuous-dynamical-system’s dynamical component

carrying a sinusoidal AC.

Vi = ZijI j , (11.37)

where Zij represents the ‘impedance holor matrix’ of the AC electrical-continuous-

dynamical system components.
Thus, the impedance of a resistor is R, the impedance of an inductor is ωL and the
impedance of a capacitor is 1/ωC . In all cases, the impedance holor is measured in
ohms (Ω). The concept of impedance is very powerful in sinusoidal steady-state
analysis and we will have much more to say about the usefulness in subsequent
sections.
The term ‘inductive reactance’ is used to describe the product ωL . The reactance
of an inductor is ωL (Ω), whereas the impedance of an inductor is ωL (Ω).
The term ‘capacitive reactance’ is used to describe the quantity −1/ωC. The
reactance of the capacitor is 1/ωC (Ω), whereas the impedance of a capacitor is
−1/ωC (Ω). Here, we will make one final comment about the impedance holor. Note
that although the impedance holor is a bivalent holor, it is a matrix Zij .
Finally, we remind the reader that if the reference sense of direction for the
current holor in a passive AC electrical-homogeneous-continuous-dynamical-sys-
tem’s dynamical component is in the sense of direction of the voltage holor increase
across the dynamical component, a minus sign must be inserted into the equation
that relates the holor voltage to the current holor.

11.3.7 An impedance-holor representation

Applying KSL to the R–L–C AC electrical homogeneous continuous dynamical
system shown in figure 11.17 gives one
t
di ( t ) 1
v(t ) = Ri (t ) + L
dt
+
C
∫0 i (t )dt . (11.38)

Substituting to equation (11.38) an instantaneous value of the current from

equations (11.18), one has

11-36
 Mechatronics

Figure 11.17. Holor equivalent ECM of the R–L–C AC electrical homogeneous continuous dynamical system.

v (t ) = 2 I ⎡⎣(R cos ψi − X sinψi )sinωt + (R sinψi + X cos ψi )cosωt ⎤⎦ , (11.39)

where the passive reactance is given as

X = ωL − 1/ωC . (11.40)
Substituting to equation (11.39) an instantaneous value of the voltage from
equations (11.18), one has
(V cos ψv − RI cos ψi + XI sin ψi )sin ωt
(11.41)
+ (V sin ψv − RI sin ψi − XI cos ψi )cosωt = 0.

But if sinωt and cosωt are the independent linear functions, one has
V cos ψv = RI cos ψi − XI sin ψi (11.42)

V sin ψv = RI sin ψi + XI cos ψi . (11.43)

Comparing equations (11.21) and (11.24) with equations (11.42) and (11.43), one
obtains the relationship between the independent elements (merates) of the voltage
holor Vi and independent elements (merates) of the current holor I j , namely:
Vc = RI c − XI s , Vs = XI c + RI s . (11.44)
Assuming that
Zcc = R , Zcs = −X , Zsc = X , Zss = R ,
equation (11.44) can be rewritten as
Vc = ZccI c − ZcsI s , Vs = ZscI c + ZssI s . (11.45)
Thus, the relationship between the current holor I j and the voltage holor can be
described by an expression
Vi = ZijI j , (11.46)

which defines a new bivalent holor Zij , termed the bivalent impedance holor.

11-37
 Mechatronics

The impedance holor written in a general form establishes a matrix:

⎡ Z −Zcs ⎤
Zij = ⎢ cc ⎥. (11.47)
⎣ Zsc Zss ⎦

In particular circumstances, the following applies.

a) For a series R–L–C AC electrical homogenous continuous dynamical
system, the impedance holor Zij is well known in the (Fijalkowski 1967)
matrix form of the impedance holor matrix [Z ], namely:
⎡ ⎤ ⎡ R 1/ωC + ωL ⎤ ,
Zij = ⎢ R −X ⎥ = ⎢ ⎥⎦ (11.48)
⎣ X R ⎦ ⎣ ωL − 1/ωC R
where X = ωL − 1/ωC .
b) For an ideal active resistance R, the passive reactance X = 0, hence
⎡ ⎤ ⎡ ⎤
Zij = ⎢ R 0 ⎥ = R⎢1 0 ⎥ = R1.
⎣ 0 R⎦ ⎣o 1 ⎦

c) For a series R–L AC electrical homogeneous continuous dynamical system,

the passive reactance X = ωL, hence
⎡ ⎤
Zij = ⎢ R −ωL ⎥ .
⎣ ωL R ⎦

d) For a series R–C AC electrical homogeneous continuous dynamical system,

the passive reactance X = 1/ωC, hence

⎡ 1/ωC ⎤ .
Zij = ⎢ R
⎣−1/ωC R ⎥⎦
Applying the holor algebra laws that govern the bivalent holors, one also obtains
the impedance holors for other very complicated AC electrical homogeneous
continuous dynamical systems.

11.3.8 An admittance-holor representation

Equation (11.46) for the current holor can be also rewritten in the form
I j = Y jiVi , (11.49)
where the admittance holor Y ji is the inverse of impedance holor:
Y ijZjk = δ ki , (11.50)

where δ kj is a Kronecker delta.

For instance, for a series R–L–C AC electrical homogeneous continuous
dynamical system,

11-38
 Mechatronics

⎡ ss sc ⎤ ⎡ ⎤ 1 ⎡ R X⎤
Y ji = ⎢ Y cs Y cc ⎥ = ⎢ G B ⎥ = ∧ . (11.51)
⎣−Y Y ⎦ ⎣−B G ⎦ (R 2) + (X 2) ⎢⎣−X R ⎥⎦
∧

11.3.9 AC power holor representation

An instantaneous value of the AC power can be obtained from a product
p(t ) = u(t )i (t ). (11.52)
Substituting to equation (11.52) instantaneous values of the voltage and current
from equations (11.18), one obtains

{
p(t ) = VI cosφ − cos[2ωt + (ψv + ψi )] , } (11.53)

where
φ = ψv − ψi .

To obtain a holor representation of the AC power one must write equation (11.53) in
the form

p(t ) = UI ⎡⎣ cosφ − cos + 2ωt cos(ψv + ψi ) + sin 2ωt sin(ψu + ψi )⎤⎦ . (11.54)

If one designs an independent term by Pp and dependent terms from a doublet

frequency by Pq and Pr, then one obtains
P p = UI cos ⌊ψv − ψi ⌋ = UI cos φ = P ⎫
⎪
P q = UI cos(ψv + ψi ) ⎬. (11.55)
r ⎪
P = UI sin(ψv + ψi ) ⎭

In this way, equation (11.54) can be rewritten in the form

p(t ) = P p − P q cos 2ωt + P r sin 2ωt . (11.56)
The quantities P p, P q and P r are independent elements (merates) or coordinates of
the three-merate monovalent holor of an AC power
P κ = (P p, P q, P r ) , (11.57)

which allows us to eliminate time from equation (11.56).

From the first expression of equations (11.55) it results that an independent
element (merate) of the power holor is an AC active power, which can be measured
by means of an electrodynamic wattmeter.
Residual independent elements (merates) can also be measured or computed with
the aid of electrical metrology methods.
In particular circumstances, the following applies.
a) For a voltage-holor phase angle ψv = 0
P q = VI cos φ = P p = P;

11-39
 Mechatronics

P r = −VI sin φ = −Q .

b) For a current-holor phase angle ψi = 0

P q = VI cos φ = P p = P;

P r = VI sin φ = Q .
Thus, for the voltage-holor phase angle ψv =0 or the current-holor phase angle
ψi = 0, an independent element(merate) of the holor power P q is the active AC
power P, however, an independent element (merate) of the power holor P r is the
reactive AC power Q.
To obtain an equation that takes a power holor from the voltage holor and the
current holor into account one must use an adequate law for the multi-application
of these holors. However, in holor algebra, the conventional tensor product in the
form VI j (due to Moon and Spencer (1986)) is not for application in the case of
i
the power holor, where one uses the particularly useful univalent tensor product
P κ = Vi ⨂I j = γ..κjiVI j
i , (11.58)

where i, j = c, s and κ = p, q, r. The univalent tensor product is designed using a

symbol ⨂, giving a univalent holor as a result.
Hence, applying a holor representation of the power in the AC electrical
homogeneous continuous dynamical system analysis, one obtains a full analogy
with the DC electrical current analysis.
Independent elements (merates) of the power holor can be obtained from
equation (11.55), namely:
P p = γ..pij VI j
i , P q = γ..qij VI j
i , P r = γ..rij VI j
i , (11.59)

where
⎡ γ pc γ pc ⎤ ⎡ ⎤ ⎡ γ qc γ qc ⎤ ⎡ ⎤
γjpi = ⎢ ps ps ⎥ = ⎢1 0 ⎥ , γjqi = ⎢ qs qc ⎥ = ⎢1 0 ⎥ ,
..c ..s ..c ..s
⎢⎣ γ..c γ..s ⎥⎦ ⎣ 0 1 ⎦ ⎢⎣ γ..c γ..s ⎥⎦ ⎣ 0 −1⎦
⎡ γ rc γ rc ⎤ ⎡ ⎤
γjri = ⎢ rs rc ⎥ = ⎢1 0 ⎥ .
..c ..s
⎢⎣ γ..c γ..s ⎥⎦ ⎣ 0 1 ⎦

Applying a convention of holor summation, one obtains

s⎫
P p = γ..pc c pc s ps
c VcI + γ..s VcI + γ..c VsI + γ..s VsI
c ps
⎪
P q = γ..c VcI c + γ..s VcI s + γ..c VsI c + γ..s VsI s ⎬ .
qc qc qs qs
(11.60)
⎪
P r = γ..rcc VcI c + γ..rcs VcI s + γ..rsc VsI c + γ..rss VsI s ⎭

11-40
 Mechatronics

Substituting independent elements (merates) of the holor voltage from equations

(11.20) and independent elements (merates) of the holor current from equations
(11.23) to equations (11.60), one obtains
P p = γ..pc pc ⎫
c cos ψv cos ψi + γ..s sin ψv sin ψi
ps ps
⎪
+ γ..c sin ψv cos ψi + γ..s sin ψv sin ψi ⎪
P q = γ..qcc cos ψv cos ψi + γ..qcs sin ψv sin ψi ⎪
⎪
qs qs ⎬. (11.61)
+ γ..c sin ψv cos ψi + γ..s sin ψv sin ψi ⎪
P r = γ..rcc cos ψv cos ψi + γ..rcs sin ψv sin ψi ⎪
⎪
+ γ rs sin ψ cos ψ + γ rs sin ψ sin ψ ⎪
..c v i ..s ⎭
v i

Comparing equations (11.61) with equations (11.55), one obtains

γ..pc ps
c = γ..s = 1, γ..pc ps
s = γ..c = 0,

γ..qcc = 1, γ..qss = −1, γ..qcs = γcqs = 0,

γ..rcc = γ..rss = 0, γ..rcs = γ..rsc = 1.

The aforementioned independent elements (merates) of a trivalent holor γ can also

be rewritten in the form of the following matrices:
⎡ ⎤ ⎡ ⎤ ⎡ ⎤
γ..pij = ⎢1 0 ⎥ , γ..qij = ⎢1 0 ⎥ , γ..rij = ⎢ 0 1 ⎥ . (11.62)
⎣0 1 ⎦ ⎣ 0 −1⎦ ⎣1 0 ⎦

Taking into account equations (11.62), one can express equations (11.60) as follows:
P p = VcI c + VsI s , P q = VcI c − VsI s , P r = VcI s + VsI c . (11.63)
Routinely, to compute an AC power, an analogous expression to that for computing
a DC power is needed in the form
P = R(I ∧ 2).
From equations (11.55), a power holor
P κ = Vi ⨂ I j = γ..κjiVI j
i .

However, from equation (11.37) a voltage holor

Vi = ZijI j .
Hence, the power holor
P κ = Vi ⨂ I j = γ..κjiVI
i
j
= γ..κjiZijI kI j , (11.64)

and its independent elements (merates)

P p = γ..pij ZijI kI j , P q = γ..qij ZijI kI j , P r = γ..rij ZijI kI j . (11.65)

11-41
 Mechatronics

Applying a convention of holor summation, one obtains

P p = γ..pm c∧ pm c s pm s c pm s∧ ⎫
n Zcc(I 2) + γ..n ZccI I + γ..n ZcsI I + γ..n Zcs(I 2)
⎪
+ γ..pm Z ( I c ∧ 2) + γ pmZ I cI s + γ pmZ I sI c + γ pmZ (I s ∧ 2) ⎪
n sc ..n sc ..n ss ..n ss
qm qm qm qm
⎪
P = γ..n Zcc(I 2) + γ..n ZccI I + γ..n ZcsI I + γ..n Zcs(I s ∧ 2) ⎪
q c ∧ c s s c
⎬. (11.66)
+ γ..qm qm qm qm
n Zsc(I 2) + γ..n ZscI I + γ..n ZssI I + γ..n Zss(I 2) ⎪
c∧ c s s c s∧
⎪
P r = γ..rm c∧ rm c s rm s c rm
n Zcc(I 2) + γ..n ZccI I + γ..n ZcsI I + γ..n Zcs(I 2)
s∧
⎪
rm ∧ rm c s rm s c rm ∧ ⎪
+ γ..n Zsc(I 2) + γ..n ZscI I + γ..n ZssI I + γ..n Zss(I 2) ⎭
c s

Taking into consideration equations (11.48) and (11.61), one obtains

P p = R⌈(I c ∧ 2) + (I s ∧ 2)⌉ ⎫
⎪
P q = R⌈(I c ∧ 2) − (I s ∧ 2)⌉−2XI cI s ⎬ . (11.67)
P r = X⌈(I c ∧ 2) − (I s ∧ 2)⌉+2RI cI s ⎪
⎭

Substituting equations (11.66) into the first expression from equations (11.67), one
obtains
P p = R(I ∧ 2) = P. (11.68)
In this way, one obtains a well-known expression for an AC electrical continuous
dynamical system analysis, from which we can again see that an independent
element (merate) of the holor power P p is an AC real or active power.
Adding geometrically independent elements (merates) P q and P r of the power
holor, one derives a well-known expression for an AC effective value of the apparent
power S in the form

sqrt⎡⎣ (P q ∧ 2) + (P r ∧ 2)⎤⎦ = sqrt⎡⎣ (R∧ 2) + (X ∧ 2)⎤⎦⎡⎣ (I c ∧ 2) + (I s ∧ 2)⎤⎦

(11.69a )
= Z (I ∧2 ) = VI = S .

Thus, due to an application of the holor algebra for a sinusoidal AC electrical

continuous dynamical system analysis, an operation is possible with holor quantities
for the AC electric continuous dynamical systems that are analogous with real
quantities for the DC electrical continuous dynamical systems.

11.3.10 A geometrical interpretation of the voltage, current and power holors

The voltage, current and power holors introduced can be represented geometri-
cally in 2D or 3D spaces. If one assumes that independent elements (merates) of
the current holors Ic and Is are rectangular coordinates in 2D space (figure
11.18), then the first equation of equations (11.44) can be regarded as an
equation of the straight line Vc = const, whose slope is described by Zij. The
second straight line Vs = const results from the second equation of equations
(11.44).

11-42
 Mechatronics

An impedance holor Zij can be represented using a pair of orthogonal straight

lines (see figure 11.18). The current holor stretches out from the beginning of the
coordinate system to cut each of these straight lines. In particular circumstances, for
instance, for an ideal active resistance, these orthogonal straight lines run across
vertically and horizontally.
The voltage holor may also be represented geometrically in a similar way to the
current holor, as shown in figure 11.19.
A geometrical representation of the power holor needs 3D space. The coordinates
are independent elements (merates) of the power holor Pp, Pq and Pr. The
coordinates Pq and Pr are orthogonal to each other with regard to equation
(11.69). The third axis Pp establishes an angle φ = ψV − ψi with a vector of the
apparent power, with a module S = V I.
The power holor P κ is hence a vector sum of its independent elements (merates)
P , Pq and Pr (figure 11.20).
p

A partial representation of the holor power needs 2D space. If a phase angle of

the double-frequency term is small, then a module of the double-frequency power
can be obtained from a part of the holor diagram shown in figure 11.20, which
depicts a well-known holor diagram for a power holor established by independent
elements (merates) of the real or active power Pp = P and the reactive power Q.
A partial holor diagram of the holor power, however, does not represent the holor
power exactly, and a strict representation needs three independent elements
(merates) and 3D space. Thus a physically effective value of the apparent power
P κ between two terminals would be:
P κ = sqrt[(P p∧ 2) + (P q ∧ 2) + (P r ∧ 2)] = S . (11.69b)

Figure 11.18. A geometrical representation of the current holor.

11-43
 Mechatronics

Figure 11.19. A geometrical representation of the voltage holor.

11.3.11 Conclusion
Holor algebra, like complex number algebra, allows one to eliminate time from
equations describing a physical action of the sinusoidal physical variables, as well as
a geometrical interpretation of the fundamental quantities of the alternating current
as real quantities in 2D and 3D spaces.
Holor algebra allows one to isolate fundamental electrical quantities into
individual terms, such as the voltage holor, current holor, impedance holor,
admittance holor and power holor. It seems that a basic value of holor algebra
lies not only in computing processes, but also in the theoretical domain, as it allows
the sinusoidal AC fundamental quantities to be represented in a way that is logical
and approved by traditional mathematics.

Figure 11.20. A geometrical representation of the power holor.

11-44
 Mechatronics

11.4 The single-phase AC electrical homogeneous continuous

dynamical system as the single-phase AC electrical
commutation matrixer
11.4.1 Introduction
Almost all AC electrical homogeneous continuous dynamical systems operate with
sinusoidal alternating currents and voltages. Fortunately, AC electrical homogeneous
continuous dynamical system analysis can be essentially the same as the analysis
presented for DC electrical homogeneous continuous dynamical systems. It is only
necessary to develop some understanding of current, voltage, impedance and power
holors (and of the holor algebra by which they are computed) to be able to apply the
same DC electrical homogeneous continuous dynamical system knowledge to them.
Before starting, it might be helpful to look at a standard and simple single-phase
AC electrical homogeneous continuous dynamical system to see how much like DC
it really is, and to see what new ideas have to be developed to apply DC techniques
to AC. Consider the single-phase AC electrical homogeneous continuous dynamical
system in figure 11.21(a) and the similar DC electrical homogeneous continuous
dynamical system in figure 11.21(b).
Figure 11.21(a) shows a series with an AC energy-source (voltage source) and
three passive components in series. In addition to the resistor R there are two other
components. C is a capacitor and L is an inductor.
Figure 11.21(b) shows a similar DC electrical homogeneous continuous dynam-
ical system with the same configuration. All three passive components here are
resistors. From DC electrical homogeneous continuous dynamical system analysis,
it is possible to come to the solution for it (figure 11.21(b)) as follows:
1 1
I= V= V,
R R1 + R3 + R3
where
R = R1 + R3 + R3.
The solution for a single-phase AC electrical homogeneous continuous dynamical
system (figure 11.21(a)) is almost the same:
δ ki δ ki
Ij = Vi = 1 Vi ,
Zjk Z jk + Z jk2 + Z jk3

where
I j = Ic Is T
= I ∠ψi;
T
Vi = Vc Vs = V ∠ψu;
Zij = Zij1 + Zij2 + Zij3.

Here the boldface letters Vi and I j stand for AC voltage holors and current holors.
As might be guessed, the AC voltage holor produces an AC current holor. The

11-45
 Mechatronics

Figure 11.21. Single-phase AC electrical homogeneous continuous dynamical system and a similar DC
electrical homogenous continuous dynamical system.

angles ψi and ψu describe the phase angle associated with AC current holors and
voltage holors, respectively, as explained in section 11.3.
The AC impedance holor (Z 1jk ) of a component is a more general form of DC
resistance (R), namely:
⎡ ⎤ ⎡ ⎤
Zij1 = ⎢ R 0 ⎥ = R⎢1 0 ⎥ .
⎣ 0 R⎦ ⎣0 1 ⎦

Alternatively, to put it another way, in addition to R, single-phase AC electrical

homogeneous continuous dynamical systems have two other forms of impedance
holor (Zij2 and Zij3), which distinguish between the resistive kind of impedance holor
and that produced with inductors L or capacitors C, namely:
⎡ ⎤
Zij2 = ⎢ 0 −ωL ⎥
⎣ ωL 0 ⎦
and
⎡ 1/ωC ⎤ .
Zij3 = ⁎⎢ 0
⎣−1/ωC 0 ⎥⎦

11-46
 Mechatronics

11.4.2 Series and/or parallel transformations

The rules for combining impedance holors in series and/or parallel are the same as
those for resistors. The only difference is that combining impedance holors involves
the algebraic manipulation of hypernumbers, i.e. holors.
Impedance holors in series can be combined into a single impedance holor Zij by
simply adding the individual impedance holors.
Thus, the series single-phase AC electrical homogeneous continuous dynamical
system’s equivalent impedance holor Zij is
⎡ R 1/ωC − ωL ⎤ = ⎡ R −X ⎤ ,
Zij = ⎢ ⎥⎦ ⎢⎣ X R ⎥⎦
⎣ ωL − 1/ωC R
where X = ωL − 1/ωC .
Put another way, the problem is defined in general terms by the series AC single-
phase electrical homogeneous continuous dynamical system in figure 11.22, where
the impedance holors Zij1, Zij2, …, Zijs are connected in series between the conductors
A, N.
The impedance holors are in series because they carry the same current holor I j .
From equation (VI = ZIJ I j ) one knows that the voltage-drop holor across each
impedance holor will be
Zij1I j , Zij2I j , … , ZijsI j

and from Kirchhoff’s second law (KSL), one has

ViAN = Zij1I j + Zij2I j + ⋯ + ZijsI j = (Zij1 + Zij2 + ⋯ + Zijs )I j . (11.70)

The equivalent impedance holor between the conductive conductors A, N is

ViAB
ZijAN = = Zij1 + Zij2 + ⋯ + Zijs . (11.71)
Ij

Figure 11.22. Holor impedances in series.

11-47
 Mechatronics

The impedance holor connection in parallel for the AC electrical homogeneous

continuous dynamical system can be reduced to a single equivalent impedance holor
by multiplying by a Kronecker delta relationship
δ ki (AN ) δ ki(1) δ ki(2) δ ki (p)
= + +⋯+ , (11.72a )
Z jkAN Z 1jk Z jk2 Z jkp

or the impedance holor connection in parallel for the AC electrical homogeneous

continuous dynamical system can be reduced to a single equivalent holor
admittance
ji
YAN = Y1ji + Y3ji + ⋯ +Y pji . (11.72b)

The parallel connection of holor impedances is illustrated in figure 11.23. Note that
the impedance holors are in parallel because they have the same holor voltage across
their conductors A, N.
The derivation of equation (11.72) follows directly from figure 11.23. One simply
combines Kirchhoff’s first law (KFL) with the holor-domain version of Ohm’s law,
that is

I j = I1j + I2j + ⋯ + I pj ,

or

δ ki (AN ) AN
δ ki(1) AN
δ ki(2) AN
δ ki (p)
Ui = Ui + Ui +⋯+ UiAN , (11.73a )
Z jkAN Z 1jk Z jk2 Z jkp

or
ji AN
YAN UiAN = Y1jiUiAN + Y2ji Ui + ⋯ + Y pji UiAN . (11.73b)

Figure 11.23. Holor impedances in parallel.

11-48
 Mechatronics

Once the common voltage holor term is cancelled out of equation (11.37b) one sees
that one has equation (11.72).
Equation (11.72) can also be expressed in terms of an admittance holor, which is
defined as the reciprocal of the holor impedances multiplied by a Kronecker delta δki
ji
and is denoted by the boldface letter YAN . Thus

ji
⎡Y AN
ss sc ⎤
Y AN
i
⎡ G B ⎤ δ k(AN ) 1 ⎡ R X⎤
YAN = ⎢ cs ⎥ = ⎢ ⎥ = AN = , (11.74)
⎣ ⎦ (R 2) + (X 2) ⎣⎢−X R ⎦⎥
cc
⎣Y AN Y AN ⎦ −B G Z jk ∧ ∧

where
X = ωL − 1/ ωC .
ji
Like conductance G, the admittance holor YAN is measured in siemens (S) and
commonly expressed in mhos (℧).
When one uses equation (11.74) in equation (11.72), one has

ji
YAN = Y1ji + Y2ji + ⋯ + Y pji . (11.75)

One also notes from equation (11.72) that for the exceptional case of just two
impedance holors in parallel,

δ ki
ZijAN = Zij1Zij2. (11.76)
Z 1jk + Z jk2

It is also worth noting the holor admittance of each of the ideal passive single-phase
AC electrical homogeneous continuous dynamical system’s components. In partic-
ular, the admittance holor of an individual resistor is
⎡1 ⎤
⎡YRss YRsc ⎤ ⎡G 0 ⎤ δ ki (R ) 1 ⎡R 0 ⎤ ⎢ R 0 ⎥
YRji = ⎢ cs cc ⎥ = ⎢ ⎥= = ∧ ⎢ ⎥= [℧].
⎣YR YR ⎦ ⎣ 0 G ⎦ Z jkR (R 2) ⎣ 0 R ⎦ ⎢ 0 1 ⎥
⎣ R⎦

We have already noted in our discussion of resistive single-phase AC electrical

homogeneous continuous dynamical systems that the reciprocal of the resistance
holor multiplied by a Kronecker delta δ ki is termed the ‘conductance’ holor.
The admittance holor of the individual inductor is
⎡ 1 ⎤
⎡Y Lss Y Lsc ⎤ ⎡ 0 B ⎤ δ ki (L ) 1 ⎡ 0 ωL ⎤ ⎢ 0 ωL ⎥
YLji = ⎢ cs cc ⎥ = ⎢ ⎥= = ⎢ ⎥= ⎥[℧].
⎣Y L Y L ⎦ ⎣−B 0 ⎦ Z jkL (ω ∧ 2)(L∧ 2) ⎣−ωL 0 ⎦ ⎢− 1 0⎦
⎣ ωL

The quantity above is referred to as the ‘inductive-susceptance’ holor of the

inductor.

11-49
 Mechatronics

The admittance holor of the individual capacitor is

⎡ 1 ⎤
⎡YCss YCsc ⎤ ⎡ i
⎤ δ k(C ) 1 ⎢ 0 − ωC ⎥
YCji = ⎢ cs cc ⎥ = ⎢ 0 B ⎥ = = ⎢ 1
⎣YC YC ⎦ ⎣−B 0 ⎦ Z Cjk 0 ⎥⎦
( 2)( )
1
ω
∧
1
C
∧
2 ⎣ ωC
⎡ ⎤
= ⎢ 0 −ωC ⎥[℧].
⎣ ωC 0 ⎦
The quantity above is referred to as the ‘capacitive-susceptance’ holor of the
capacitor.

11.4.3 Delta-to-wye or pi-to-tee equivalent

The AC electrical homogeneous continuous dynamical system’s bridge configura-
tion in figure 11.24 introduces an intersection of holor impedances that warrants
further discussion.
An interesting thing about this AC electrical homogeneous dynamical system is
that the interconnected impeders cannot be reduced to a single equivalent holor
impedance across the conductive conductors of the energy source if one is restricted
to the simple series or parallel equivalent AC electrical homogeneous continuous
dynamical systems that were introduced earlier in this chapter.
The interconnected impeders can be reduced to a single equivalent impeder by
means of a delta-to-wye (Δ2Y) or pi-to-tee (Π2T) equivalent AC electrical
continuous dynamical system. The Δ and Y structures occur in a variety of single-
and/or polyphase AC electrical homogeneous continuous dynamical systems and
therefore one will still find Δ2Y transformation a helpful tool in AC electrical
continuous dynamical system analysis.
The ZijAa, ZijAn, and Zijan or ZijNa, ZijNn and Zijan in the AC electrical homogeneous
continuous dynamical system shown in figure 11.24 are referred to as a delta (Δ)
interconnection because the interconnection looks like the Greek letter Δ. It is also
referred to as a pi (Π) interconnection because the Δ can be shaped into a Π without

Figure 11.24. The AC electrical homogeneous continuous dynamical system’s bridge configuration.

11-50
 Mechatronics

disturbing the electrical equivalence of the two configurations. The electrical

equivalence between the Δ and the Π interconnections is apparent from figure 11.25.
The Δ2Y equivalent AC electrical homogeneous continuous dynamical system
means that the Δ configuration can be replaced by a Y configuration, such that the
conductive conductors’ behaviour for the two configurations will be identical.
The Δ2Y transformation is illustrated in figure 11.26, where the two AC electrical
homogeneous continuous dynamical systems shown are equivalent at the conductive
conductors a, b and c, provided that
δ ki
Zijan = Z caZijab;
ab ij (11.77)
Z jkbc ca
+ Z jk + Z jk

δ ki
Zijbn = Z abZijbc;
ab ij (11.78)
Z jkbc ca
+ Z jk + Z jk

Figure 11.25. A Δ configuration viewed as a Π configuration.

11-51
 Mechatronics

Figure 11.26. The Δ2Y transformation.

δ ki
Zijcn = Z bcZijca .
ab ij (11.79)
Z jkbc ca
+ Z jk + Z jk

The Y configuration is also referred to as the tee (T) configuration, because the Y
structure can be replaced by the T structure without disturbing the electrical
equivalence of the two structures.
Equations (11.77)–(11.79) give the Y-connected impeders as functions of the
Δ-connected impeders. It is also possible to reverse the Δ2Y or Π2T transformation.
That is, one can start with the Y structure and replace it with an equivalent Δ structure.
The electrical equivalence of the Y and T configurations is apparent from figure
11.27.
The expressions for the three Δ-impeder holors, Zijbc, Zijca and Zijab, as functions of
the three Y-impeder holors Zijan, Zijbn and Zijcn are
δ ki
Zijab = (
Z jkcn
)
ZijanZijbn + ZijbnZijcn + ZijcnZijan ; (11.80)

δ ki
Zijbc = (
Z jkan
)
ZijanZijbn + ZijbnZijcn + ZijcnZijan ; (11.81)

11-52
 Mechatronics

Figure 11.27. A Y configuration viewed as a T configuration.

11-53
 Mechatronics

δ ki
Zijca = (
Z jkbn
)
ZijanZijbn + ZijbnZijcn + ZijcnZijan . (11.82)

The derivation of equations (11.77)–(11.82) can be accomplished by noting that the

two AC electrical systems are by definition equivalent with respect to their conductive
collectors’ behaviour. Thus if each AC electrical homogeneous continuous dynamical
system is placed in a black box one cannot tell by external measurements whether the
black box contains a set of Δ-connected impeders or a set of Y-connected impeders.
This can only be true if the impedance holor between the corresponding conductive
conductor pairs is the same for each black box. For instance, one must see the same
impedance holor between conductive conductors a and b whether one uses the
Δ-connected set or the Y-connected set. It follows directly that
δ ki
Zijab =
Z jkbc + Z jkca + Z jkab
( )
Zijab Zijbc + Zijca = Zijan + Zijbn; (11.83)

δ ki
Zijbc =
Z jkbc + Z jkca + Z jkab
( )
Zijbc Zijca + Zijab = Zijbn + Zijcn; (11.84)

δ ki
Zijca =
Z jkbc + Z jkca + Z jkab
( )
Zijca Zijab + Zijbc = Zijan + Zijcn. (11.85)

One can obtain equations (11.77)–(11.82) by straightforward holor algebraic

manipulation of equations (11.83)–(11.85).
In addition to single-phase AC electrical homogeneous continuous dynamical
systems, other polyphase ones are also used for special purposes. For instance, a
‘conventional two-phase’ AC electrical homogeneous continuous dynamical system
consists of two energy-source voltage holors. The minimum phase difference
between these voltage holors is π.
An ‘unconventional two-phase’ AC electrical homogeneous continuous dynam-
ical system does not follow the preceding ideas. It consists of two energy-source
voltage holors whose phase difference is π/2.
The ‘five-phase’ electrical homogeneous continuous dynamical system consists of
five energy-source voltage holors. The minimum phase difference between them is
2π/5, etc.

11.5 The polyphase AC electrical homogeneous continuous

dynamical system as the polyphase AC electrical commutation
matrixer
11.5.1 Introduction
In addition to single-phase AC electrical homogeneous continuous dynamical
systems, other polyphase AC electrical homogeneous dynamical systems are also
used for special purposes. For instance, a three-phase AC electrical homogeneous

11-54
 Mechatronics

continuous dynamical system consists of three voltage-source holors. The minimum

phase difference between these voltage-source holors is 2π/3. Similarly, a five-phase
electrical homogenous continuous dynamical system consists of five voltage-source
holors. The minimum phase difference between them is 2π/5. The unconventional
two-phase AC electrical homogeneous continuous dynamical system does not follow
the preceding ideas. It consists of two voltage-source holors whose phase difference is
π/2. Note that this is not the conventional two-phase one discussed below in section
11.6. ‘Polyphase AC electrical homogeneous continuous dynamical system’ is the name
given to an assemblage of AC electrical homogenous continuous dynamical systems in
which the voltage-source holors have the same frequency but are shifted in phase.
The entirety of voltage-source holors, load voltage holors and current holors in a
polyphase AC electrical homogeneous continuous dynamical system are also termed
a ‘polyphase AC electrical homogenous continuous dynamical system’s holors’. The
separate parts comprising a polyphase AC electrical homogeneous continuous
dynamical system are termed the ‘phases’. Note: do not confuse them with the
phase angles of the sinusoidal waveforms of the voltage and/or current phases, which
are likewise termed ‘phase’ for short.
A polyphase AC electrical homogeneous continuous dynamical system is termed a
‘two-, three-, five-phase…’ one, depending on the number of phases. If the voltage-
source holors, load-voltage holors, or current holors forming a polyphase AC electrical
homogeneous continuous dynamical system have identical amplitudes in all phases of
the AC electrical homogeneous continuous dynamical system and are shifted in phase
relative to one another by equal angles, then the AC electrical continuous dynamical
system is termed ‘balanced’ or ‘symmetrical’ (figures 11.28 and 11.29).
The letters A, B, C denote the separate phases of a polyphase AC electrical
homogeneous continuous dynamical system, etc. It is usual to have the sequence or
order of the phases such that the voltage-source holor in each successive phase lags
the voltage-source holor of the preceding phase, i.e. the voltage-source holor of
phase B lags behind the voltage-source holor of phase A, the voltage-source holor of
phase C lags behind the voltage-source holor of phase B, and so on. Such a phase
sequence is termed a ‘positive sequence’. However, there are occasions when several
voltage-source holors of the AC electrical homogeneous continuous dynamical

Figure 11.28. Balanced voltage-source holors of five- and three-phase AC electrical homogeneous continuous
dynamical systems.

11-55
 Mechatronics

Figure 11.29. Balanced voltage-source holors of a three-phase AC electrical homogeneous continuous

dynamical system.

systems act in the same polyphase AC electrical homogeneous continuous dynam-

ical system and not all of them have the same sequence.
An AC electrical homogeneous continuous dynamical system in which the
voltage-source holor in each successive phase leads the preceding voltage-source
holor is termed a ‘negative sequence’ AC electrical homogeneous continuous
dynamical system. For instance, the phase sequence determines a sense of the
rotation direction of polyphase EM motors.

11.6 The two-phase AC electrical homogeneous continuous

dynamical system as the two-phase AC electrical commutation
matrixer
11.6.1 Introduction
By far the most widespread system is the conventional two-phase AC electrical
homogeneous continuous dynamical one, since, of all the AC electrical homoge-
neous continuous dynamical systems, it is the most economical for electrical energy
transmission. Electrical energy supply to households, commercial establishments
and other small consumers is usually distributed as single phase, but with two
available voltages—220 V or 110 V, or 440 V or 220 V. The energy source in a
conventional two-phase electrical homogeneous continuous dynamical system is a
single-phase ME generator, which supplies the primary winding of a small (often
only 15–25 kVA) EE transformer with centre-tapped secondary winding. Figure
11.30(a) shows how the two-phase voltages are derived from the centre-tapped
winding of an EE transformer for two-phase AC electrical homogeneous continuous
dynamical systems. The advantages of this two-phase AC electrical homogeneous
continuous dynamical supply system is that small loads (such as electric lamps and
hand tools) can be accommodated with more safety 220 V or 110 V, while larger
loads (particularly heavy electric heating devices and machines) can have the
advantage of higher voltage and lower current. Furthermore, if the 220 V or
110 V loads are reasonably balanced between the two phases, some of the
advantages of lower current are retained even for them. For safety reasons, the
centre (neutral) conductor is grounded, as shown in figure 11.30(a).

11-56
 Mechatronics

Figure 11.30. Two-phase AC electrical homogeneous continuous dynamical systems and three-phase AC
electrical homogeneous continuous dynamical systems are easily distinguished by taking the voltmeter between
three pairs of conductors.

A further advantage is that no part of the two-phase AC electrical homogeneous

continuous dynamical system is more than 220 V or 110 V above ground, even for
440 V or 220 V appliances. It is easy to determine whether a conventional two-phase
AC electrical homogeneous continuous dynamical system is connected according to
this scheme (figure 11.30(b)). Of the three voltage readings between the conductors,
two will be equal and the third will be twice as great.
The two-voltage, three-conductor conventional two-phase AC electrical homoge-
neous continuous dynamical system is also used infrequently with the two-phase DC
electrical homogeneous continuous dynamical system. The phase voltage-source
waveforms of a conventional balanced two-phase AC electrical homogeneous
continuous dynamical system are shifted relative to each other by equal phase
angles of π (figure 11.31).
Accordingly, the instantaneous values of the phase voltage-source waveforms are
expressed by the following formulae:

11-57
 Mechatronics

Figure 11.31. The phase voltage-source waveforms of a conventional balanced two-phase AC electrical
homogeneous continuous dynamical system.

eAN (t ) = E max sin ωt = 2 E rms sin ωt ⎫

⎪
⎬
⎪
, (11.86)
eBN (t ) = E max sin(ωt + π ) = 2 E rms sin(ωt + π )⎭
if t = 0 corresponds to the instant when the voltage of phase A passes through zero and
becomes positive. The last formula in equations (11.86) can also be written in the form
eBN (t ) = E max sin(ωt − π ) = 2 E rms sin(ωt − π ). (11.87)

In the holor notation of the phase-voltage holor, the real axis is usually made to
coincide with the energy-source-voltage holor of phase A.
The effective values (Erms = Eϕ ) of the conventional balanced two-phase AC
electrical homogeneous continuous dynamical system are represented by the holors

EijAN = Eϕ ∠0 ⎫
⎪

BN
⎬. (11.88)
Eij = Eϕ ∠π ⎪ ⎭
Figure 11.32 shows the waveforms of the instantaneous energy-source voltages
for both phases; it is easy to see that at any instant the algebraic sum of the two
energy-source voltages is zero.
This result can also be established analytically with the aid of the relations in
equations (11.86), but it is easier to use the holor diagram (figure 11.33) for this
purpose. Indeed, combining the voltage holor EijBN , one finds that, due to the
symmetry, the voltage holor EijBN is directed vertically downwards, i.e. it is opposite
to the holor voltage EijAN .
Forming π angle with the voltage holor EijBN , this is equal in magnitude to the
voltage holor EijAN . Therefore,

(
EiAN + EiBN = EiAN + −EiAN = 0, ) (11.89)

11-58
 Mechatronics

Figure 11.32. Voltage-source holors of the two-phase EE transformer’s secondary winding.

Figure 11.33. Holor diagram of the voltage-source holors of the conventional balanced two-phase AC
electrical homogeneous continuous dynamical system.

and since the sum of the instantaneous values eAN (t ) + eBN (t ) represents the sum of
the EiAN and EiBN holor projections multiplied by 2 , or the projections of their sum
on the vertical, one has, at any instant,
eAN (t ) + eBN (t ) = 0. (11.90)
It is obvious that the positive senses of the directions of the energy-source voltages
and currents must be chosen the same for all phases, for instance, away from the
reader into the plane of the image shown in figure 11.33.
Let us decide then to consider the ‘beginning’ of each phase as the point where the
positively directed current comes out of the EE transformer’s secondary winding
into the external AC electrical homogenous continuous dynamical system, and the
‘end’ of the phase as the point where the positively directed current leaves
the external circuit and enters the EE transformer’s secondary winding. Thus, the
energy-source voltages in the EE transformer secondary phases are positive when
they act from the ends of phases (N) to their beginning (A and B). In AC electrical
homogenous continuous dynamical system diagrams, a two-phase EE transformer’s
secondary winding is ordinarily depicted in the form of two secondary windings NA
and NB (figure 11.32). If the load phases connected to the beginnings and ends of
each phase have identical (in magnitude and waveform) holor impedances ZijAN and

11-59
 Mechatronics

Figure 11.34. The phase energy-source voltage waveforms of an unconventional balanced two-phase AC
electrical homogeneous continuous dynamical system.

j j
ZijBN , the current holors IAN and IBN will likewise be identical in magnitude, and
shifted in phase relative to the corresponding holor energy-source voltage holors by
equal angles φ. They will also be shifted in phase relative to each other by π, i.e. the
mutual distribution of their waveforms will not differ from that of the waveforms of
the voltage-source holors in figure 11.32, and, hence, the holors of the effective
values of the holor currents will be connected by the relation
j j
IAN + IBN = 0, (11.91)

and the instantaneous values of the currents by

iAN (t ) + iBN (t ) = 0. (11.92)
Such a load for a two-phase AC electrical homogeneous continuous dynamical
system is termed a ‘balanced’ or ‘symmetrical’ load. A two-phase electrical
homogeneous continuous dynamical system, where two conductors to its own
load phase independently connect each EE transformer’s secondary winding phase,
is termed an ‘independent’ two-phase AC electrical homogeneous continuous
dynamical system. Thus, an independent two-phase AC electrical homogeneous
continuous dynamical system demands the use of six conductors to transmit
electrical energy from the EE transformer to the loads.
The phase energy-source voltages of an unconventional balanced two-phase AC
electrical homogeneous continuous dynamical system are shifted relative to each
other by equal phase angles of π/2 (figure 11.34).
Accordingly, the instantaneous values of the phase energy-source voltages are
expressed by the following formulae:
eAN (t ) = E max sin ωt = 2 E rms sin ωt ⎫
⎪
⎬
⎪
, (11.93)
eBN (t ) = E max sin(ωt + π /2) = 2 E rms sin(ωt + π /2) ⎭

11-60
 Mechatronics

Figure 11.35. Holor diagram of the voltage-source holors of the unconventional balanced two-phase AC
electrical homogeneous continuous dynamical system.

EiAN = Eϕ ∠0 ⎫
⎪
⎬. (11.95)
EiBN = Eϕ ∠π /2 ⎪
⎭

if t = 0 corresponds to the instant when the voltage of phase A passes through zero
and becomes positive.
The last formula in equations (11.94) can also be written in the form
⎛ π⎞ ⎛ π⎞
eBN (t ) = E max sin⎜ωt − ⎟ = 2 E rms sin⎜ωt − ⎟ . (11.94)
⎝ 2⎠ ⎝ 2⎠
In the holor notation of the phase-voltage holor, the real axis is usually made to
coincide with the energy-source voltage holor of phase A. The effective values
(Erms = Eϕ ) of the unconventional balanced two-phase AC electrical homogeneous
continuous dynamical system are represented by the holors (figure 11.35).
It is most interesting that one advantage of the Δ–Δ or Δ–Y three-phase AC
electrical homogeneous continuous dynamical system in industry is the possibility of
using the ‘open delta’ (i.e. unconventional unbalanced two-phase AC electrical
homogeneous continuous dynamical system) in the case where one energy-source
phase has been lost. In figure 11.36(a), the three-phase energy sources are shown as
three EE transformer secondaries (secondary phase windings)—the common indus-
trial situation in the USA.
If one transformer is burned out or faulty, as shown in figure 11.36(b), it can
simply be left out and the AC electrical homogeneous continuous dynamical system
will still be capable of supplying a reduced amount of three-phase electrical energy.
In this situation, the Δ2Δ or Δ2Y three-phase AC electrical homogeneous contin-
uous dynamical system operates as an unconventional unbalanced two-phase AC
electrical homogeneous continuous dynamical system.

11.7 The three-phase AC electrical homogeneous continuous

dynamical system as the three-phase AC electrical commutation
matrixer
11.7.1 Introduction
The generation, transmission, distribution and utilisation of large quantities of electrical
energy can be achieved using polyphase AC electrical homogeneous continuous
dynamical systems. The most common is the three-phase AC electrical homogeneous

11-61
 Mechatronics

Figure 11.36. The Δ2Δ or Δ2Y three-phase AC electrical homogeneous continuous dynamical system.

continuous dynamical system, since, of all the AC electrical homogeneous continuous

dynamical systems, it is most economical for electrical energy transmission.
The basic structure of a three-phase AC electrical homogenous continuous
dynamical system consists of primary energy sources connected to loads via EE
transformers and transmission lines. One can reduce the problem to the analysis of
an AC electrical homogeneous continuous dynamical system consisting of a primary

11-62
 Mechatronics

Figure 11.37. Cross-section of the two-pole three-phase AC ME generator (Kasatkin and Perekalin 1970).

voltage source connected to a load via a transmission line. The omission of the EE
transformer as a structural element in the AC electrical homogeneous continuous
dynamical system simplifies the discussion without jeopardising a basic understanding
of the computations involved.
The primary energy source in a three-phase AC electrical homogenous continuous
dynamical system is a three-phase AC generator, which differs from a single-phase AC
generator in that on the inner surface of the stator (figure 11.37) it has three separate
armature-phase windings (‘phases’), the axes of which (A, B, and C) form angles of 2π/3
between each other (if the rotor of the AC generator has one pair of magnetic poles).
Each phase consists of two groups of conductors, the energy-source voltages of
which are opposite in direction at any given instant. Hence, the groups A, B and C,
shifted relative to each other by 2π/3, will be termed the ‘positive sides’ or
‘beginnings’ of the phases, and the groups diametrically opposite to them, X, Y
and Z, will be termed the ‘negative sides’ or ‘ends’ of the phases A, B and C.
All the conductors of the positive and negative sides of each phase are connected
in series to each other (on the end surfaces of the stator), so that their energy-source
voltages are added together, forming the phase energy-source voltages. Note: by the
phase axis is understood the direction of the radius that passes through the centre of
the positive portion of the phase winding.

11.7.2 Three-phase energy sources

As mentioned above, three-phase energy sources consist of AC generators that have
three separate armature-phase windings distributed around the periphery of the
stator. Each armature-phase winding comprises one phase of the AC generator.
The rotor of the ME generator is an electromagnet with magnetic-field winding,
driven at a synchronous speed by a prime mover such as an external combustion engine
(i.e. a steam or gas turbine) or an internal combustion engine (i.e. a diesel engine).
As the electromagnet is rotated past the three armature-phase windings, a
sinusoidal AC armature-phase voltage is induced in each armature-phase winding.
The armature-phase windings are designed so that the sinusoidal AC armature-
phase voltages induced in them are equal in amplitude and out of phase with each
other by exactly 2π/3.

11-63
 Mechatronics

Since the armature-phase windings are stationary with respect to the rotating
electromagnet, the frequency of the armature-phase voltages induced in each
armature-phase winding is the same.
Normally, the impedances of each armature-phase winding on a three-phase AC
generator are very small compared to the other impedances in the AC electrical
homogenous continuous dynamical system.
Therefore, to an approximation, each armature-phase winding can be physically
modelled in an AC electrical homogeneous continuous dynamical system by an ideal
sinusoidal energy source.
There are two ways of interconnecting the separate armature-phase windings to
form a three-phase energy source. The armature-phase windings can be connected
together in either a triangle or delta (Δ) or a three-ray star or wye (Y) configuration.
The delta and wye (Y) connections are shown in figure 11.38, where ideal energy
sources are used to physically model the armature-phase windings of the three-phase
AC generator.
The common terminal in the Y-connected energy source, labelled n in figure
11.38(a), is known as the ‘neutral terminal’ of the three-phase energy source. The
neutral terminal may or may not be available for external connections.
If the impedances of each armature-phase winding are not negligible, the
three-phase energy source is physically modelled by placing the armature-phase
winding impedance in series with an ideal sinusoidal energy source.
In view of the fact that all armature windings on the AC generator are of the same
structure, the armature-phase winding impedances are assumed to be identical. The
armature-phase winding impedance of three-phase AC generators is inductive.
A physical model of a three-phase energy source, including the armature-phase
winding impedance, is shown in figure 11.39, in which Rw is the winding resistance
and Xw is the inductive reactance of the winding.
The fact that a three-phase energy source can be either Δ-connected or Y-connected
means that the basic AC electrical homogeneous continuous dynamical system in
figure 11.40 can take four different configurations, since the three-phase loads can also
be either Δ- or Y-connected.

Figure 11.38. The two basic connections of an ideal three-phase energy source. (a) Δ-connected energy source.
(b) Y-connected energy source.

11-64
 Mechatronics

Figure 11.39. A physical model of a three-phase energy source with armature-phase winding impedance. (a)
Δ-connected energy source. (b) Y-connected energy source.

Figure 11.40. Basic AC electrical homogeneous continuous dynamical system.

The four possible arrangements are:

• a Δ-connected energy source and a Δ-connected load;
• a Δ-connected energy source and a Y-connected load;
• a Y-connected energy source and a Y-connected load;
• a Y-connected energy source and a Δ-connected load.

We begins our holor algebra analysis with the third arrangement. After holor
analysing the Y–Y three-phase AC electrical homogeneous continuous dynamical
system, we show how for balanced three-phase AC electrical homogeneous
continuous dynamical systems the remaining three arrangements can be reduced
to a Y–Y equivalent three-phase AC electrical homogeneous continuous dynamical
system.
In other words, holor algebra analysis of the Y–Y three-phase AC electrical
homogeneous continuous dynamical system is the key to solving all balanced three-
phase arrangements.

11.7.3 Balanced three-phase AC electrical homogeneous continuous dynamical

systems
The phase or line-to-neutral voltage-source holors of a balanced three-phase AC
electrical homogeneous continuous dynamical system are shifted relative to each
other by equal phase angles of 2π/3 (figure 11.41).

11-65
 Mechatronics

Figure 11.41. Phase or line-to-neutral voltage-source holors of a balanced three-phase AC electrical homogeneous
continuous dynamical system.

In discussing three-phase AC electrical homogeneous continuous dynamical

systems, it is standard practice to refer to the three phases as A, B and C.
Furthermore, the A phase is almost always used as the reference phase. The three
voltages that comprise the three-phase set are referred to as the A-phase energy-
source voltage, the B-phase energy-source voltage and the C-phase energy-source
voltage. Since the phase or line-to-neutral energy-source voltages are out of phase by
exactly 2π /3, two possible phase relationships can exist between the A-phase energy-
source voltage and the B-phase and C-phase energy-source voltages. One possibility
is for the B-phase energy-source voltage to lead the A-phase energy-source voltage
by 2π /3, in which case the C-phase energy-source voltage must lead the A-phase
energy-source voltage by 2π /3. This phase relationship is known as the ABC or
‘positive phase sequence’. The only other possibility is for the B-phase energy-source
voltage to lead the B-phase energy-source voltage by 2π /3, in which case the C-phase
energy-source voltage must lag the A-phase energy-source voltage by 2π /3. This
phase relationship is known as the ACB or ‘negative phase sequence’. Accordingly,
the instantaneous values of the phase or line-to-neural energy-source voltages for
negative phase sequences are expressed by the following formulae:

eAN (t ) = E max sin ωt = 2 E rms sin ωt ⎫

⎪
eBN (t ) = E max sin(ωt + 2π /3) = 2 E rms sin(ωt + 2π /3) ⎬ , (11.96)
⎪
eCN (t ) = E max sin(ωt + 4π /3) = 2 E rms sin(ωt + 4π /3) ⎭

corresponding to the instant when the voltage of phase A passes through zero and
becomes positive.
The last formula in equations (11.96) can also be written in the form

eCN (t ) = E max sin(ωt − 2π /3) = 2 E rms sin(ωt − 2π /3). (11.97)

In the holor notation of the phase or line-to-neutral energy-source voltage, the real
axis is usually made to coincide with the energy-source voltage of phase A. In holor
notation, the two possible sets of effective values (Erms = Eϕ ) of the phase or line-to-
neutral energy-source voltages of the balanced three-phase AC electrical homoge-
neous continuous dynamical system are represented by the following holors.

11-66
 Mechatronics

a) For negative phase sequences:

EiAN = Eϕ ∠0 ⎫
⎪
BN 2π ⎪
Ei = Eϕ ∠ ⎬. (11.98a )
3
⎪
4π 2π
EiCN = Eϕ ∠ = Eϕ ∠− ⎪
3 3⎭

b) For positive phase sequences:

EiAN = Eϕ ∠0 ⎫
⎪
EiBN = Eϕ ∠ − 2π /3 ⎬. (11.98b)
⎪
EiCN = Eϕ ∠ − 4π /3 = Eϕ ∠ + 2π /3⎭

The phase sequence of the energy-source voltage holors given in equations

(11.98a) is the ACB, or negative, sequence. The phase sequence of the energy-source
voltage holors given in equations (11.98b) is the ABC, or positive, sequence. The
holor diagram representations of the voltage-holor sets given by equations (11.98a)
and (11.98b) are shown in figure 11.42, from which one can determine the phase
sequence by noting the order of the subscripts as one moves clockwise around the
diagram.
The fact that a three-phase AC electrical homogeneous continuous dynamical
system can have one of two possible phase sequences is an important characteristic
that must be taken into consideration whenever two separate polyphase AC
electrical homogeneous continuous dynamical systems are operated in parallel.
The two polyphase AC electrical homogeneous continuous dynamical systems can
only operate in parallel if they have the same phase sequence. Another important

Figure 11.42. Holor diagrams of a balanced set of three-phase voltage holors. (a) ABC (positive) sequence. (b)
ACB (negative) sequence.

11-67
 Mechatronics

characteristic of a set of balanced three-phase line-to-neutral energy-source voltage

holors is that the sum of the line-to-neutral voltage-source holors adds to zero.
Thus, using either equations (11.98a) or equations (11.98b), one has
EiAN + EiBN + EiCN = 0. (11.99)

The fact that the sum of the line-to-neutral energy-source voltage holors adds to
zero also means that the sum of the instantaneous values of the line-to-neutral
energy-source is zero, that is,
eAN (t ) + eBN (t ) + eCN (t ) = 0. (11.100)
Another noteworthy observation is that if one knows the phase sequence and one
voltage holor in the set, one knows the entire set. Thus if one has a balanced
three-phase AC electrical homogeneous continuous system, one can focus on
determining the voltage holor (or current holor) in one phase, because once one
knows one phase quantity, one automatically knows the corresponding quantity in
the other two phases.
Figure 11.43 shows the waveforms of the instantaneous line-to-neutral energy-
source voltages for all three phases; it is easy to see that at any instant the algebraic
sum of the three line-to-neutral energy-source voltages is zero.
This result can also be established analytically with the aid of the relations in
equations (11.100), but it is easier to use the holor diagram (figure 11.44) for this
purpose. Indeed, combining the holors EiBN and EiCN , one finds that, due to the
symmetry, the holor EiBN + EiCN is sensed vertically downwards, i.e. it is opposite to
the holor EiAN .
Forming π/3 angles with the EiBN and EiCN holors (or NB and BD), this result
becomes one of the sides of the equilateral triangle NBD and, consequently, is equal
in magnitude to the holor EA.

Figure 11.43. Energy-source voltages of the three-phase ME generator.

11-68
 Mechatronics

Figure 11.44. The sum of the voltage-source holors.

Therefore,

(
EiAN + EiBN + EiCN = EiAN + −EiAN = 0, ) (11.101)

and since the sum of the instantaneous values eAN (t ) + eBN (t ) + eCN (t ) represents the
sum of the EiAN , EiBN and EiCN holor projections multiplied by 2 , or the projections
of their sum on the vertical, one has, at any instant,
eAN (t ) + eBN (t ) + eCN (t ) = 0. (11.102)
It is obvious that the positive senses of the directions of the energy-source line-to-
neutral voltages and currents must be chosen the same for all phases, for instance,
away from the reader into the plane of the image shown in figure 11.44.
Let us decide, then, to consider the beginning of each phase to be the point where
the positively directed current comes out of the ME generator’s stator winding into
the external AC electrical homogenous continuous dynamical system, and let us
consider the end of the phase to be the point where the positively directed current
leaves the external AC electrical homogeneous continuous dynamical system and
enters the ME generator’s stator winding. Thus, the energy-source voltages in the
ME generator phases are positive when they act from the ends of phases (X, Y and
Z) to their beginnings (A, B and C). In AC electrical homogeneous continuous
dynamical system diagrams, a three-phase ME generator is ordinarily depicted in
the form of three stator windings XA, YB and ZC (figure 11.45).
If the loads connected to the beginnings and ends of each phase have identical (in
magnitude and waveform) impedance holors ZijAN , ZijBN , and ZijCN , the current holors
j j j
IAa , ,IBb and ICc will likewise be identical in magnitude, and shifted in phase relative
to the corresponding voltage-source holors by equal angles φ (figure 11.46).
They will also be shifted in phase relative to each other by 2π/3, i.e. the mutual
distribution of their waveforms will not differ from that of the waveforms of the line-
to-neutral voltage-source holors in figure 11.43, and hence the values of the effective
values of the line-current holors will be connected by the relation
j j
IAa + IBb + ICcj = 0 (11.103)

11-69
 Mechatronics

Figure 11.45. Schematic diagram of a three-phase generator feeding three loads.

Figure 11.46. Holor diagram of balanced three-phase AC electrical homogeneous continuous dynamical system.

and the instantaneous values of the line currents will be connected by

iAa(t ) + iBb(t ) + iCc(t ) = 0. (11.104)
Such a load for a three-phase electrical homogeneous continuous dynamical
system is termed a ‘balanced’ or ‘symmetrical load’.
A three-phase electrical homogeneous continuous dynamical system where each
ME generator phase is independently connected by two conductors to its own load is
termed an ‘independent’ three-phase AC electrical homogeneous continuous
dynamical system. Thus, an independent three-phase AC electrical homogeneous
continuous dynamical system demands the use of six conductors to transmit
electrical energy from the ME generator to the loads.

11.7.4 Wye connection of the AC electrical homogenous continuous dynamical system

The number of conductors can be reduced if, instead of three separate return
conductors, one common conductor is used (figure 11.47), and one joins to it all the
phase ends of the ME generator or EE transformer secondaries and load phases (let
us agree that the positive sense of the current direction in the load phases is from the
beginning of the phase to its end). The operating conditions of the load phases are in
no way changed, since the voltage between the beginning and end of each phase, as
in an independent three-phase AC electrical homogeneous continuous dynamical
system, is equal to the voltage of the corresponding phase of the ME generator or
EE transformer secondaries (neglecting the impedances of the connecting

11-70
 Mechatronics

Figure 11.47. A three-arm star or wye (Y) connection of the three-phase AC electrical homogeneous
continuous dynamical system.

conductors). In this way, a four-conductor Y-connected three-phase AC electrical

homogeneous continuous dynamical system is obtained.
In circuit diagrams, it is customary to represent the phases of the ME generator or
EE transformer secondaries and load phases in the form of a three-arm star, or more
often, a wye (Y), placing the load phases parallel to the corresponding ME generator
phases or EE transformer secondaries (figure 11.47).
The common point where all the phase terminals of the ME generator (or EE
transformer secondaries) and load phases are joined is termed the ‘neutral point’ or
‘neutral’, and the common return conductor joining the neutral points of the ME
generator (or EE transformer secondaries) and load is termed the ‘neutral conductor’.
The current in this conductor, by KFL, is equal to the sum of the current holors
j j
IAa , IBb and ICcj:
j j j j
INn = IAa + IBb ′ + ICc .
(11.105)
j j j
For balanced load phases when IAa + IBb ′ + ICc = 0
, there will be no current in
the neutral conductor (hence the term ‘neutral’), and the neutral conductor becomes
unnecessary. By omitting the neutral conductor, one obtains a three-wire
Y-connected three-phase AC electrical homogeneous continuous dynamical system.
Such an AC electrical homogeneous continuous dynamical system can only function
normally when the load is distributed equally among the phases. Otherwise, the
phase or line-to-neutral effective (rms) values of the load-voltage holors will no
longer be identical and equal to the ME generator (or EE transformer secondaries)
phase or line-to-neutral effective (rms) values of the energy-source voltage holors.
The effective (rms) values of the voltages between the beginnings of the ME
generator (or EE transformer secondaries) or load phases and the neutral point are
designated VΦan, VΦbn and VΦcn and are termed the ‘phase’ or ‘line-to-neutral load
voltages’.
In three-phase AC electrical homogeneous continuous dynamical system analysis,
the phase or line-to-neutral load voltages are often given instead of the phase or line-
to-neutral energy-source voltages EΦAN , EΦBN and EΦCN , and they are equal to the

11-71
 Mechatronics

phase or line-to-neutral energy-source voltages when the ME generator or EE

transformer winding impedances can be neglected as compared to the load phase
impedances.
The effective (rms) values of the voltages between the phase beginnings a and b, b
and c, and c and a are designated VΦab, VΦbc and VΦca and are termed the ‘line’ or ‘line-
to-line load voltages’. If the neutral point of the ME AC generator or EE AC
transformer is chosen as a point of zero potential, the potentials of points a, b and c
will be equal to the corresponding phase or line-to-neutral load voltages
VΦan, VΦbn and VΦcn .
The line or line-to-line load voltage holors, regarded as differences of potential,
will be related to the phase or line-to-neutral load-voltage holors by the expressions

Viab = Vian − Vibn, Vibc = Vibn − Vicn, Vica = Vicn − Vian. (11.106)

This subtraction is performed on the holor diagram (figure 11.48) by holor addition
of the holors Vian, Vibn and Vicn to the holors –Vibn, − Vicn and −Vian, respectively,
which are equal and opposite to the holors Vibn, Vicn and Vian.
It can be seen from the holor diagram that the magnitude and direction of the
line-voltage holors can be obtained more simply (by joining by a straight line) from
the ends of the holors of the corresponding phase-load-voltage holors; the line-
voltage holor is directed from the end of the subtrahend holor to the end of the
minuend holor. The three line-voltage holors are equal in magnitude and shifted in
phase (relative to each other) by 2π/3, in the same way as the phase-voltage holors.
Here, the line-voltage holor three-ray star (Y) is rotated forward relative to the
phase-voltage three-ray star or wye (Y), through an angle of π/6. The relationship
between the absolute value Uλ of any line voltage and the absolute value VΦ of any
phase voltage can be found by considering the isosceles triangle nab (or nbc or nca)
with a vertex angle of 2π/3.
Dropping a perpendicular from the vertex of the obtuse angle to the opposite side,
for example ab, one has

Figure 11.48. Phase or line-to-neutral and line or line-to-line load-voltage holors.

11-72
 Mechatronics

V λab π 3
= VΦan cos = VΦan , (11.107)
2 6 2
whence
Vλ = 3 VΦ. (11.108)

This important relation refers equally to a three-phase AC electrical homogeneous

continuous dynamical system with a neutral conductor and one without one.

11.7.5 Delta connection of the AC electrical homogeneous continuous dynamical

system
If three load impedance holors Zijab, Zijbc and Zijca are connected directly between the
conductors of a three-conductor line from the terminals of a three-phase ME
generator, without using a neutral conductor, the result is a triangle, or more often a
delta (Δ), connection (figure 11.49).
Here, one need not distinguish between the phase and line or line-to-line load
voltage holors, since the voltage holor between the beginning and end of each load
phase is at the same time the line or line-to-line load-voltage holor. But there is a
difference between the phase load-current holors Iabj, Ibcj and Icaj , whose positive
senses of directions are chosen from a to b, b to c, and c to a, and the line-load
current holors, which, as before, are considered positive in the senses of directions
from the ME generator or EE transformer to the load phases.
From KFL,
j
IAa = Iabj − Icaj, j
IBb = Ibcj − Iabj, ICcj = Icaj − Ibcj , (11.109)

that is, any one of the line-current holors is equal to the holor difference in the
current holors of two phases directly connected to the given conductor of the line;

Figure 11.49. A triangle or delta (Δ)-connected three-phase AC electrical homogeneous continuous dynamical
system.

11-73
 Mechatronics

the minuend is the phase-current holor directed from the conductor, and the
subtrahend is the phase-current holor directed to the line.
From the relationships (11.109) it follows that, for any values of the phase
currents, the holor sum of the line-current holors is
j j
IAa + IBb + ICcj = 0. (11.110)

For balanced phases (that is, for load phase impedance holors Zijab, Zijbc and Zijca
identical in magnitude and character), when the phase or line-to-neutral current
holors Iabj, Ibcj and Icaj are equal in magnitude and differ in phase by the same angles
relative to the corresponding line or line-to-line load voltage holors Viab, Vibc and Vica
(figure 11.50), forming a symmetrical three-arm wye (Y), the line-current holors
j j
IAa , IBb and ICcj also become equal and symmetrically distributed. Here, the wye (Y)
line-current holor is rotated backwards relative to the wye (Y) phase-current holor
through a π/6 angle.
Each of the line-current holors thus represents the base of an isosceles triangle
with a 2π/3 vertex angle, and for this reason the same relationship exists between the
line- and phase-current holors for a symmetrical load as between the line voltage and
phase voltage in a wye (Y) connection, namely,
Iλ = 3 IΦ. (11.111)

This important relation makes it rather easy to compute the line currents in a
symmetrical load. If the impedances of conductors joining delta (Δ)-connected load
phases to the energy source or to any apparatus connected in these conductors must
be taken into consideration, one should transform the impedance delta into an
equivalent wye (Y), which, in the case of a symmetrical delta, is also symmetrical.
The impedance of each line then becomes connected in series with the impedance
of each phase of the wye (Y), and it may be replaced by overall impedance. In this
case, each energy-source phase must be directly connected between two conductors
of the three-conductor line. If, in this case, phase AB of energy source (EiAB ) is
connected between conductors A and B (figure 11.51), phase (EiBC ) between
conductors B and C, and phase (EiCA) between conductors C and A, and in such a
way that the energy-source voltages have the positive senses of directions shown in

Figure 11.50. Line-current holors under balanced loading of phases.

11-74
 Mechatronics

Figure 11.51. Triangle or delta (Δ)-connected three-phase ME generator.

figure 11.51, then the following relationships will hold (neglecting the winding
impedances):
Viab = EiAB, Vibc = EiBC , Vica = EiCA,
and consequently, the usual line-voltage sequence, in which Vibc lags Viab by 2π/3 and
Vica lags Vibc by 2π/3, is retained.
Thus, the three energy-source phases must be connected in series with each other,
that is, the end of phase AB must be connected with the beginning of phase BC, and
the end of phase CA with the beginning of phase AB.
At first sight it may seem that the series connection of the three energy-source
phases in a closed delta (Δ) is equivalent to a joint short circuit, as the case would be,
for instance, in such a connection of DC energy sources.
However, since in a three-phase AC electrical continuous dynamical system
e(t ) AB + e(t )BC + e(t )CA = 0 at any instant, not only is there no short circuit in the
given case, but, in the absence of a load, the current in the closed delta (Δ) of the
energy-source is zero. However, this only holds true when the sum of the phase
energy-source voltages is zero. But often the phase energy-source voltages are not
strictly sinusoidal waveforms and do not fully satisfy the relationship. Therefore,
delta (Δ)-connected three-phase ME generators are rarely encountered. On the other
hand, delta (Δ)-connected EE transformers are just as common as wye
(Y)-connected ones.
In the case of a symmetrical (balanced) load when the line-current holors
j j
IAa , IBb , and ICcj are equal in magnitude and shifted in phase relative to each other
by 2/3, any one of the energy-source phase currents is
Iλ
Iϕ = ,
3
as was also the case in a balanced three-phase AC electrical homogeneous
continuous dynamical system for the phase currents of the loads, and it follows
that if the phases of both the energy source and the load are delta (Δ) connected,
then, in a balanced three-phase AC electrical homogeneous continuous dynamical
system, the currents in the corresponding energy-source and load phases will be
equal in magnitude and phase.

11-75
 Mechatronics

11.7.6 Unbalanced wye-connected load of the AC electrical homogeneous continuous

dynamical system
Turning to a holor algebra analysis of unbalanced three-phase loads, let us first
consider the case of a wye (Y)-connected energy source and load where the line
conductors, including the neutral, possess a finite (and not negligibly small)
impedance.
Incidentally, the impedances of the phase conductors connected in series with the
load phases may be grouped with the impedances of these phases, and then one only
needs to take the neutral-conductor impedance into account.
Let us choose the neutral of the energy source as zero potential and then aim to
compute the mode of operation of the three-phase AC electrical homogeneous
continuous dynamical system (figure 11.52), if one knows the energy-source phase-
voltage holors EiAN , EiBN and EiCN , that is, the potentials at points A, B and C, or a,
b and c, the load-phase impedance holors Zijan, Zijbn and Zijcn, and the impedance
holor ZijNn of the neutral conductor.
Since the given AC electrical homogeneous continuous dynamical system has two
nodes (the energy source and load neutrals), the problem is easily solved using the
node-voltage holor method.
One first determines the voltage holor between the neutral points, which is termed
displacement of the neutral, that is, the potential of the neutral point n of the load,
from the formula

δ ki ⎛ δi δj δi ⎞
ViNn = ⎜⎜ kan EiAN + kbn EiBN kcn EiCN ⎟⎟ .
δ ki δ ki δ ki δ ki
⎝ Z jk Z jk Z jk ⎠ (11.112)
Z an
+ + Z cn
+ Nn
jk Z bn
jk jk Z jk

Figure 11.52. Three-phase AC electrical homogeneous continuous dynamical system with a finite impedance
of the neutral conductor.

11-76
 Mechatronics

δ ki
Replacing all the s by the corresponding admittance holors Y ji , one writes the
Zjk
formula in a more concise form:
δ ki
ViNn = ij
Yan ij
+ Ybn ij
+ Ycnij + YNn
ij
Yan (
EiAN + YbnjiEiBN + YcnijEiCN . ) (11.113)

ji
If there is no neutral conductor, that is, if ZikNn = ∞, then YNn = 0, and the last term
in the denominator is removed. If the energy source is given a symmetrical energy-
source voltage-holor three-phase AC electrical homogeneous continuous dynamical
system
⎫
(
EiAN = EcAN , E sAN = E ∠0, ) ϕ ⎪
⎪
EiBN = (E BN
c , E sBN )=E ∠ − 2π /3, ⎬ (11.114)
ϕ
⎪
EiCN = (E CN
c , E sCN )=E ∠ − 4π /3 = Eϕ ∠2π /3,⎪
ϕ ⎭

and the current holors

j j
IAa = YanjiEiAN , IBb = YbnjiEiBN , ICcj = YcnjiEiCN , (11.115)

then by substituting expressions (11.115) into formula (11.104), one obtains a

formula more suitable for computation
δ kj
ViNn =
Yanji + Ybnji ji
j
( j j
ji I Aa + IBb + ICc .
+ Ycn + YNn
) (11.116)

For a balanced load, when Yanji = Ybnji = Ycnji = Y ji , UiNn = 0, since EiAN + EiBN +
EiCN = 0 (as in figure 11.52).
When the neutral displacement VmNn has been found, it becomes easy to determine
the phase-voltage holors of the load:

Vian = EiAN − ViNn, Vibn = EiBN − ViNn, Vicn =EiCN − ViNn. (11.117)

The current holor in the neutral conductor may be found as the holor sum of the
line- or phase-current holors,
j j j
INn = IAa + IBb + ICcj ,

or from the formula

j ji
INn = YNn ViNn.
If the impedance holor of the neutral conductor can be ignored, ZikNn = 0, then
ji
YNn = ∞ and, from equation (11.103), ViNn = 0, that is, the neutral point potentials
of the energy source and load are the same. Consequently, the current holors may be
computed directly from formula (11.115).

11-77
 Mechatronics

11.7.7 Unbalanced delta-connected load of the AC electrical homogeneous continuous

dynamical system
Let us consider the problem of an unbalanced delta (Δ)-connected load, the vertices
of which are connected to the energy source through impedance holors
ZikAa, ZikBb and ZikCc, which may be identical or different (figure 11.53).
This problem is simply solved by the delta (Δ) to wye (Y) transformation
according to the familiar formulae:
j
δk
Zikan = bc + Z ca + Z ab ZijcaZijab;
Zik ik ik
j
δk
Zikbn = bc + Z ca + Z ab ZijabZijbc;
Zik ik ik
j
δk
Zikcn = bc + Z ca + Z ab ZijbcZijca .
Zik ik ik

Knowing the admittance holors of each phase,

ji δ kj ji δ kj ji δ kj
YAa = , YBb = , YCc = ,
ZikAa + Zikan ZikBb + Zikbn ZikCc + Zikcn
it becomes easy (from equation (11.112)) to find the neutral displacement ViNn by
ji
putting YNN = 0 and then compute the line-current holors:
j
IAa ji
= YAa (
EiAN − ViNn , )
j
IBb ji
= YBb (E i
BN
− Vi Nn
),
ICcj = YCcji (Ei
CN
− ViNn ),

Figure 11.53. Unbalanced delta (Δ)-connected load of a three-phase AC electrical homogeneous continuous
dynamical system.

11-78
 Mechatronics

which are, at the same time, the phase-current holors of the equivalent wye (Y)-
connected load. Then, the phase-voltage holors of this equivalent wye (Y)-connected
load will be
j j
Vian = ZijanIAa , Vibn = ZijbnIBb , Vicn = ZijcnICcj,

and its line-voltage holors, which are also the delta (Δ) line-voltage holors, will be

Viab = Vian − Vibn, Vibc = Vibn − Vicn, Vica =Vicn − Vian.

Finally, the phase-current holors of the load are

δ kj δ kj δ kj ca
Iabj = V ab, Ibcj =
ab i
V bc, Icaj =
bc i
Vi .
Zik Zik Zikca

11.7.8 Power holor computations in balanced three-phase AC electrical homogeneous

continuous dynamical systems
An instantaneous value of the AC phase power can be obtained from a product
pϕ (t ) = vϕ(t )iϕ(t ). (11.118)

Substituting to equation (11.118) instantaneous values of the phase voltage and

phase current from equations (11.18), one obtains

pϕ (t ) = VϕI ϕ cosφϕ − cos⎡⎣2ωt + (ψvϕ − ψiϕ )⎤⎦ ,

{ } (11.119)

where
φϕ = ψvϕ − ψiϕ.

To obtain a holor representation of the phase power one must write equation
(11.117) in the form

pϕ (t ) = VϕI ϕ⎡⎣cos φϕ − cos 2ωt cos(ψvϕ + ψiϕ ) + sin 2ωt sin(ψuϕ + ψiϕ )⎤⎦. (11.120)

In this way, equation (11.118) can be rewritten in the form

pϕ (t ) = Pϕp − Pϕq cos 2ωt + Pϕr sin 2ωt . (11.121)

The quantities P p, P q and Pϕr are independent elements (merates) or coordinates of

the three-merate monovalent holor of an AC phase-power holor,
Pϕκ = (Pϕp, Pϕq, Pϕr ) , (11.122)

and this allows time to be eliminated from equation (11.123).

11-79
 Mechatronics

If one designed an independent term by Pϕp and dependent terms from a doublet
frequency by Pϕq and Pϕr , then one would obtain for the balanced load the following
expressions for the phase-power holor independent elements (merates):

Pϕp = VϕI ϕ cos(ψuϕ − ψiϕ ) = VϕI ϕ cos φϕ = Pϕ ⎫

⎪
⎪
Pϕ = VϕI cos(ψu − ψi )
q ϕ ϕ ϕ
⎬. (11.123)
⎪
Pϕr = VϕI ϕ sin(ψuϕ − ψiϕ ) ⎪
⎭

From the first expression of equations (11.121) it results that an independent

element (merate) of the power holor is an AC phase average or active power, which
can be measured by means of an electrodynamic wattmeter.
Residual independent elements (merates) can be also measured or computed with the
aid of electrical metrology methods. In particular circumstances, the following apply.
a) For a voltage phase angle ψv = 0
Pϕq = UϕI ϕ cos φϕ = Pϕ0 = Pϕ;

Pϕr = −UϕI ϕ sin φϕ = −Q ϕ.

b) For a current phase angle ψi = 0

Pϕq = UϕI ϕ cos φϕ = P 0 = Pϕ;

Pϕr = UϕI ϕ sin φϕ = Q ϕ.

Thus, for the voltage phase angle ψVϕ = 0 or the current phase angle ψiϕ = 0, an
independent element (merate) of the power holor Pϕq is the AC average or active
power Pϕ ; however, an independent element (merate) of the power holor Pϕr is the AC
reactive power Qϕ .
A phase-power holor can be also postulated by special tensor multiplying the
phase-voltage holor and phase-current holor together so that

Pϕκ = Viϕ⨂Iϕj = γ..κjiViϕIϕj = γ..κjiZijϕIϕkIϕj = (Pϕp, Pϕq, Pϕr ) . (11.124)

However, a voltage holor

Viϕ = ZijϕIϕk.

Thus the power holor associated with any one phase of Y-connected load can also be
computed using the holor algebra techniques introduced in chapter 11 or from
equation (11.122); thus

Pϕκ = (Pϕp, Pϕq, Pϕr ) . (11.125)

11-80
 Mechatronics

Thus far, holor algebra analysis of balanced three-phase AC electrical continuous

dynamical systems has been limited to the determination of voltage holors and
current holors in a given AC electrical homogeneous continuous dynamical system.
One begins by discussing the average or active power delivered to a balanced
Y-connected load. A balanced Y-connected load, along with the pertinent voltage
holors and current holors, is shown in figure 11.54.
The average or active power associated with any one phase can be computed
using the holor algebra techniques introduced in chapter 11.
Using equation (11.123) as a starting point, one finds that one can express the
average or active power associated with the A phase of the load as

PAp = Vijan IAa

j
cos(ψvan − ψi Aa ) , (11.126)
j
where ψvan and ψiAa denote the phase angles of Uijan and IAa , respectively.
Using the notation introduced in equation (11.124), one finds that the powers
associated with the B and C phases are
j
PBp = Vibn IBb (
cos ψvbn − ψiBb ) (11.127)

and

PCp = Vicn ICcj cos(ψvcn − ψiCc ) . (11.128)

In equations (11.124)–(11.126), all currents and voltages are written in terms of the
effective (rms) values of the sinusoidal function that they represent.
In a balanced three-phase AC electrical homogeneous continuous dynamical
system, the effective value of each line-to-neutral voltage is the same, as is the
effective value of the phase current. The argument of the cosine functions is also
the same for all three phases. To emphasise these observations, one introduces the
following notation to facilitate further discussion of power computations in
balanced three-phase AC electrical homogeneous continuous dynamical systems.
Vϕ = Vian = Vibn = Vicn , (11.129)

Figure 11.54. A balanced Y-connected load used to introduce average or active power computations in a
three-phase AC electrical homogeneous continuous dynamical system.

11-81
 Mechatronics

j j
I ϕ = IAa = IBb = ICcj , (11.130)

and
φϕ = ψvan − ψi Aa = ψvbn − ψiBb = ψvcn − ψiCc . (11.131)

One also notes from the above observations that for a balanced three-phase AC
electrical homogeneous continuous dynamical system the power delivered to each
phase of the load is the same. Thus
PAp = PBp = PCp = Pϕp = VϕI ϕ cos φϕ , (11.132)

where Pϕp stands for the average or active power per phase.
Power is a scalar quantity, and so the three-phase total, average or active power is
simply the average power in three phases. For this reason, the total average or active
power delivered to the balanced Y-connected load is simply three times the power
per phase. Thus
PTp = 3Pϕp = 3VϕI ϕ cos φϕ . (11.133)

But for a delta,

ϕ
Vλ = Vϕ and I λ = 3I
and for a wye (Y)
I λ = I ϕ and Vλ = 3 Vϕ.
Hence, for either it is also desirable to express the total power in terms of the
effective value of the line voltage and the effective value of the line current.
If one lets V represent the effective value of the line voltage and I λ represent the
effective value of the line current, then one can modify equation (11.134) as follows:
⎛ Vϕ ⎞
PTp = 3⎜ ⎟I λ = 3 VλI λ cos φϕ . (11.134)
⎝ 3⎠

In deriving equation (11.132), one uses the fact that for a balanced Y-connected
load, the effective value of the phase voltage is the effective value of the line
voltage divided by 3 , and the effective value of the line current is equal to the
effective value of the phase current. In using equation (11.132) to compute the
total power delivered to the load, it is important to remember that φϕ is the phase
angle between the phase voltage and the phase current. Thus equation (11.132) is
the basic average or active power formula for a balanced three-phase AC
electrical continuous dynamical system in either wye (Y) or delta (Δ). Note that
the term power factor of a three-phase AC electrical continuous dynamical
system, cosφϕ, is the ‘power factor of the phase’. Thus, the term ‘power factor’
has no meaning for a badly unbalanced three-phase AC electrical continuous
dynamical system.

11-82
 Mechatronics

If the load is Δ-connected, the computation of holor power is basically the same
as for a Y-connected load. A Δ-connected load, along with the pertinent holor
voltages and holor currents, is shown in figure 11.55.
From figure 11.55 it follows that the phase average or active power associated
with each phase is

(
PAp = Viab Iabj cos ψvab − ψiab , ) (11.135)

where ψvab and ψiab denote the phase angles of Vijab and Iabj , respectively.
Using the notation introduced in equation (11.133), one finds that powers
associated with the B and C phases are

(
PBp = Vibc Ibcj cos ψvbc − ψibc ) (11.136)

and

PCp = Vica Icaj cos(ψvca − ψica ).

For balanced Δ-connected load,

Vϕ = Viab = V = Vica , (11.138)

I ϕ = Iabj = Ibcj = Icaj , (11.139)

and
φϕ = ψvab − ψiab = ψvbc − ψibc = ψvca − ψica . (11.140)

One also notes from the above observations that for a balanced three-phase AC
electrical continuous dynamical system the power delivered to each phase of the load
is the same. Thus
PAp = PBp = PCp = Pϕp = UϕI ϕ cos φϕ . (11.141)

Figure 11.55. A Δ-connected load used to discuss power computations.

11-83
 Mechatronics

It is worth noting that equation (11.129) is the same as equation (11.130). This is
equivalent to saying ‘in a balanced load the average or active power per phase is
equal to the product of the effective (rms) value of the phase voltage, the effective
(rms) value of the phase current and the cosine of the angle between the phase
voltage and phase current’.
The total power delivered to a balanced Δ-connected load is
⎛ Iλ ⎞
PTp = 3Pϕp = 3VϕI ϕ cos φϕ = 3Vλ⎜ ⎟cosφϕ = 3 VλI λ cos φϕ . (11.142)
⎝ 3⎠

Note that equation (11.140) is the same as equation (11.132).

For the phase real or average power Pϕp a third factor, cosφϕ , that is, the power
factor (PF) enters the equation, so that
Pϕp = VϕI ϕ cos φϕ ,

where φ is the power factor angle between Vϕ and I ϕ. Note that these equations yield
results in watts (W).

11.7.9 Delta three-phase AC electrical homogeneous continuous dynamical systems

Electrical energy is universally produced, transmitted and distributed via three-
phase AC electrical homogeneous continuous dynamical systems with three or four
conductors. Such an arrangement is simply three single-phase (two collectors) AC
electrical homogeneous continuous dynamical systems combined. The three single-
phase AC electrical homogeneous continuous dynamical systems are constrained in
a certain relationship with one another.
Figure 11.56 illustrates the development of a delta (Δ), three-phase AC electrical
homogeneous continuous dynamical system.
In the three single-phase AC electrical homogeneous continuous dynamical
systems (figure 11.56(a)), the energy-source phase voltage-source holors EiAN , EiBN

Figure 11.56. The three single-phase AC homogeneous continuous dynamical systems can be combined into a
Δ three-phase AC homogeneous continuous dynamical system in (a). The voltage-source holors must be
related in phase, as sketched. EiBY and EiCZ may be interchanged.

11-84
 Mechatronics

and EiCN are equal in amplitude and 2π/3 apart in the n phase, as shown in the holor
diagram (figure 11.56(b)).
If each energy-source phase voltage holor is equal in amplitude and if each of
j j j
their load impedance holors are equal, the phase-current holors IAN , IBN and ICN will
be also equal. The phase-current holors will be along the corresponding voltage
holors and will consequently also be 2π/3 apart.
Suppose now that in the interest of saving conductors, the six conductors are
combined into three, as shown in figure 11.57(a).
Figure 11.57(b) shows that to keep the load voltage holors uniform under
unbalanced load, the wye (Y) three-phase AC electrical homogeneous continuous
dynamical system requires a fourth neutral conductor. None of the three load
current holors (or their corresponding current holors in the three energy sources,

Figure 11.57. Delta (a) and wye (b) three-phase AC electrical homogeneous continuous dynamical systems.
Note the carefully chosen symmetrical notation (signs and arrows) for the voltage and current holors.

11-85
 Mechatronics

which in power plants are usually the secondary windings of EE transformers) will
be affected. The energy sources and loads should act as before.
It might be objected that the three energy sources EiAB , EiBC and EiCA have been
short-circuited on each other. However, adding three equal voltages that are 2π/3
apart around the energy source loop on the left side of the circuit makes a zero sum.
In practice, when these AC electrical homogeneous continuous dynamical systems
are set up, great care must be exercised to ensure that one holor voltage is not
connected backwards, destroying the zero sum effect. In the three-phase AC
electrical homogeneous continuous dynamical system it will be noted that the
‘line’ current holors, labelled IAa
j j
, IBb and ICcj , are each the difference of two phase
j j j
current holors, for instance IAa = IAN − ICN .
j
However, a carefully drawn holor diagram will show that these holors (IAa , and
−ICN ), when combined, are π/3 apart, so that with three equal loads
j

I λj = 3 Iϕj . (11.143)

Moreover, in any Δ three-phase AC electrical homogeneous continuous dynamical

system,

Viϕ = Vi λ. (11.144)

Note that 3 = 2 cos π/3 = 1.732.

Three-phase AC electrical homogeneous continuous dynamical systems have two
principal advantages. Each conductor carries 1.732 times as much current as any of
the six conductors of figure 11.56(a). If the same size conductors were used, the
R(I ∧ 2) losses would be three times as great as for any of the six conductors.
However, there are only three conductors. Increasing their cross-sectional area by
only 50% would reduce the losses to the original case. Thus 25% of the copper is
saved. This saving of copper in AC three-phase transmission lines and electrical
machines is significant.
The other advantage is even more important: electrical energy flow is
smoothed. A single-phase AC electrical homogeneous continuous dynamical
system with a frequency of 50 or 60 Hz power pulses 100 or 120 times a second,
respectively.
Thus, all single-phase electrical machines must be constructed to withstand not
only peak torques and stresses, but also a continual 100 or 120 Hz vibration.
It can be easily shown that in three-phase electrical machines the power flow is
completely uniform. Electrical machines can be smaller for the same ratings;
vibration problems are reduced. Three-phase AC electrical homogeneous continu-
ous dynamical systems with equal phase loads, as shown in figure 11.57, are termed
‘balanced’. Phase loads in balanced three-phase AC electrical homogeneous con-
tinuous dynamical systems must have the same in phase angle as well as magnitude.
Equation (11.143) is only true for balanced delta (Δ) three-phase AC electrical
homogeneous continuous dynamical systems.

11-86
 Mechatronics

11.7.10 Wye three-phase AC electrical homogeneous continuous dynamical systems

There is a second way to connect three-phase AC electrical homogeneous contin-
uous dynamical systems; it is termed wye (Y) and illustrated in figure 11.57(b).
The central node to which all energy sources are connected (and referred) is
termed the ‘neutral’ or N point. It is often referred to as ‘ground’, and is frequently
actually grounded for safety reasons. Similarly, there is a neutral point in the load
j j j
connection. Here the phase-current holors (IAN , IBN and ICN ) are equal to the line-
j j j
current holors (IAB , IBC and ICA). However, the line voltage holors (sometimes
termed ‘line-to-line’ voltage holors) EiAB , EiBC and EiCA are 3 times the phase
voltage holors EiAN , EiBN and EiCN . If this Y three-phase AC electrical homogeneous
continuous dynamical system were to have only three conductors, there would be
some interesting problems with unbalanced loads.
The holor voltage drops on the load phases could adjust themselves differently for
each phase (to enable KFL to be satisfied). To overcome these unbalance problems a
fourth conductive conductor is usually run in Y three-phase AC electrical homoge-
neous continuous dynamical systems (see figure 11.57(b)).
This neutral conductive conductor carries the current holor unbalances and keeps
the voltage-drop holor magnitudes equal for all load phases. Since the Y three-phase
AC electrical homogeneous continuous dynamical system is designed to be quite
well balanced, the neutral conductive conductor’s current holor will be small, and
there is no need to use a large conductor for this purpose. Moreover, since the
neutral conductive conductor is run at ground potential, it has no need for expensive
and bulky insulation.
Single-phase loads on Y three-phase AC electrical homogeneous continuous
dynamical systems are generally taken from one hot conductor and the neutral.
The term ‘hot conductor’ refers to one of the three-phase conductors as opposed
to the ground or neutral one.
Thus for four-conductor Y three-phase AC electrical homogeneous continuous
dynamical systems, whether balanced or not, and for balanced three-conductor Y
three-phase AC electrical homogeneous continuous dynamical systems,

l
Viϕ = 3 Vi (11.145a )

and
I λj = Iϕj . (11.145b)

While Y three-phase AC electrical homogeneous continuous dynamical systems

have the disadvantage of requiring this fourth (usually smaller) conductive con-
ductor, they have several advantages over Δ three-phase AC electrical homogeneous
continuous dynamical systems. There is a natural ground in the Y three-phase AC
electrical homogeneous continuous dynamical system, which is often quite helpful
for single-phase loads. For a given phase voltage holor, transmission is at higher
voltage holors, resulting in smaller line-current holors and further copper savings.

11-87
 Mechatronics

Older industrial Δ three-phase AC electrical homogeneous continuous dynamical

systems can have their capacity increased by almost 75% by adding a fourth
conductor, raising the distribution voltage holor and reconnecting load in wye (Y).
The Y and Δ three-phase AC electrical homogeneous continuous dynamical
systems of figure 11.57 can be generalised as follows. All three-phase AC electrical
homogeneous continuous dynamical systems comprise three holor voltages that are
equal in magnitude but 2π/3 apart. Their current holors depend, however, on the
nature of load.
If the three load impedance holors are equal in magnitude and phase angle, the
three phase-current holors will be equal in magnitude and the three holor line
currents will be equal in magnitude. This is termed a balanced three-phase AC
electrical homogeneous continuous dynamical system.
Most such three-phase AC electrical homogeneous continuous dynamical systems
can usually be analysed one phase at a time, or, at worst, by mesh-current holor
methods. There are many single-phase loads—lights are a prime example. There are
two possible orders of ‘phase sequence’.
Standing at the point marked x in figure 11.58, one sees the three holors (2π/3
apart) rotating counter-clockwise. First Vian passes one, then Vibn , then Vicn .
The phase sequence seen is a, b, c. If Vibn and Vicn were unchanged in position, the
sequence would be a, c, b. These two sequences are often termed ‘positive’ and
‘negative’, respectively.

11.7.11 Holor algebra analysis of the wye–wye three-phase AC electrical homogenous

continuous dynamical system
One begins holor algebra analysis of the Y2Y three-phase AC electrical homoge-
neous continuous dynamical system by assuming that the system is not a balanced
(symmetrical) three-phase AC electrical homogeneous continuous dynamical
system.
One does this so that one can show for a three-phase AC electrical homogeneous
continuous dynamical system what one means by a balanced system, and what the
consequences of being balanced are in terms of the holor algebra analysis of the Y2Y
three-phase AC electrical homogeneous continuous dynamical system.
The general Y2Y three-phase AC electrical homogeneous continuous dynamical
system is illustrated in figure 11.59, where one has included a fourth conductor that
connects the energy source neutral N to the load neutral n. The fourth conductor is
only possible in the Y2Y arrangement.
We also mention that for convenience in drawing the holor diagram, one has
transformed the Y-connections into ‘tipped over tees’.
In figure 11.59, ZikAN , ZikBN and ZikCN represent the internal impedance holors
associated with each phase winding of the energy source; ZikAa, ZikBb and ZikCc
represent the impedance holor associated with each phase conductor of the trans-
mission line connecting the energy source to the load; ZikNn is the impedance holor of
the neutral conductor that connects the energy-source neutral to the load neutral;
and Zikan, Zikbn and Zikcn represent the impedance holors of each phase of the load.

11-88
 Mechatronics

Figure 11.58. The m – 1 wattmeter method—two wattmeters for three lines. The sum of the wattmeter
indications is the total active power in the three loads.

The general Y2Y three-phase AC electrical homogeneous continuous dynamical

system in figure 11.59 can be described by a single node-voltage holor equation.
Using the energy-source neutral as the reference node and letting EiNn denote the
node voltage holor between the nodes N and n, one finds that the node-voltage holor
equation is
δ ki δ ki
Nn Ei
Z jk
Nn
= an Aa AN Ei
Z jk + Z jk + Z jk
Nn
(− E jAN )
δ ki
+ (
Z jkbn + Z jkBb + Z jkBN
EiNn − EiBN ) (11.146)

δ ki
+
Z jkcn + Z Cc
jk + Z CN
jk
(
EiNn − E −CN = 0.)
Before pursuing equation (11.142) any further, let us pause to observe that the AC
electrical homogeneous continuous dynamical system holor analysis techniques

11-89
 Mechatronics

Figure 11.59. A Y2Y three-phase AC electrical homogeneous continuous dynamical system.

discussed in the earlier chapters are directly applicable to three-phase AC electrical

homogeneous continuous dynamical systems. Thus, it is not necessary to introduce
new holor analytical techniques in order to analyse three-phase AC electrical
homogeneous continuous dynamical systems.
However, as we see in the remainder of this chapter, if a three-phase AC electrical
homogenous continuous dynamical system is balanced, one can take some signifi-
cant holor analytical shortcuts to predict the behaviour of the system.
The general Y2Y three-phase AC electrical homogeneous continuous dynamical
system in figure 11.59 is a balanced one if it satisfies all of the following four criteria:
• EiAN , EiBN and EiCN form a set of balanced three-phase energy-source winding
voltage holors;
• Z jkAN = Z jk
BN
= Z CNjk form a set of balanced three-phase internal energy source
winding impedance holors;
• Z jkAa = Z jk
Bb
= Z Cc
jk form a set of balanced three-phase transmission line
impedance holors;
• Z jk
an bn
= Z jk cn
= Z jk form a set of balanced three-phase load-phase impedance
holors.

There is no restriction on the neutral-conductor impedance holor (ZikNn ), its value

has no effect on whether the Y2Y three-phase AC electrical homogeneous
continuous dynamical system is balanced.
If the Y2Y three-phase AC electrical homogeneous continuous dynamical system
is balanced, equation (11.146) tells one that UiNn must be zero.
To see that, let
Z jkϕ = Z jkan + Z jkAa + Z jkAN (11.147)

and then rewrite equation (11.142) as

⎛ δi δi ⎞ δi
⎜⎜ Nn k
+ 3 kϕ ⎟⎟UiNn = kϕ EiAN + EiAN + EiAN .
( ) (11.148)
⎝ Z jk Z jk ⎠ Z jk

11-90
 Mechatronics

The right-hand side of equation (11.144) is zero because by hypothesis the

numerator is a set of balanced three-phase voltage holors and Zikϕ is not zero. The
only value of UiNn that satisfies equation (11.144) is zero. Therefore, for a Y2Y
balanced three-phase AC electrical homogeneous continuous dynamical system,
ViNn = 0. (11.149)

Equation (11.145) is an extremely important result. If ViNn is zero, there is no

difference in potential between the energy-source neutral N and the load neutral n,
and consequently the current holor in the neutral conductor is zero. These
observations tell us that one can either remove the neutral conductor from a
j
balanced Y2Y configuration (INn = 0) or replace it by a perfect short circuit between
Nn
the nodes N and n (Vi = 0).
One finds both equivalents convenient to use when physically modelling balanced
three-phase AC electrical homogeneous continuous dynamical systems.
Now let us turn our attention to what effect balanced conditions have on the three
transmission-line-current holors. It follows directly from figure 11.59 that when the
three-phase AC electrical homogeneous continuous dynamical system is balanced,
the three transmission-line-current holors will be
δ ki δ ki AN
j
IAa = (
Z jkan + Z jkAa + Z jkAN
E i
AN
− E i )
Nn
=
Z jkϕ
Ei = YϕjiEiAN ; (11.150)

δ ki δ ki BN
j
IBb = (
Z jkan + Z jkAa + Z jkAN
E i
BN
− E i )
Nn
=
Z jkϕ
Ei = YϕjiEiBN ; (11.151)

and
δ ki δ ki BN
ICcj = (
Z jkan + Z jkAa + Z jkAN
E i
CN
− E i )
Nn
=
Z jkϕ
Ei = YϕjiEiBN , (11.152)

from which one sees that in a balanced, Y2Y three-phase AC electrical homoge-
neous continuous dynamical system the three transmission-line-current holors form
a balanced set of three phase-current holors. Thus, a current holor in each
transmission-line will be equal in amplitude and frequency and will be exactly
2π/3 out of phase with the other two transmission-line-current holors. This tells us
j
that if one computes the current holor IAa , one can write down the transmission-
j j
line-current holors IBb and ICc without further computation. One implies by this
statement that the phase sequence is known.
One can use equation (11.146) to construct an equivalent single-phase AC
electrical continuous dynamical system of the balanced Y2Y three-phase AC
electrical homogeneous continuous dynamical system.
It follows from equation (11.146) that the current holor in the A-phase conductor
line is simply the voltage-source holor generated in the A-phase armature winding
of the AC EM generator divided by the total impedance holor in the A phase of the
Y2Y three-phase AC electrical homogeneous continuous dynamical system.

11-91
 Mechatronics

Thus, equation (11.146) describes the simple single-phase AC electrical homoge-

neous continuous dynamical system shown in figure 11.60, where a perfect short
circuit has replaced the neutral conductor. Here, we offer a word of caution.
The current holor in the neutral conductor of figure 11.60 is not the current holor
in the neutral conductor of a balanced three-phase AC electrical homogeneous
continuous dynamical system.
The current holor in the neutral conductor is
j j j
INn = IAa + IBb + ICcj , (11.153)

whereas the current holor of the neutral conductor in figure 11.60 is IAa j
.
Thus, the AC electrical homogeneous continuous dynamical system in figure
11.59 gives the correct value of the transmission-line-current holor, but only the
A-phase component of the neutral current holor.
Whenever the equivalent single-phase AC electrical homogeneous continuous
dynamical system in figure 11.60 is applicable, the transmission-line-current holors
form a balanced three-phase set and the right-hand side of equation (11.149) sums to
zero.
Once one knows the transmission-line-current holors in the AC electrical
homogeneous continuous dynamical system in figure 11.59, it is a relatively simple
task to compute any voltage holors that are of interest. In particular, one is
interested in the relationship between the line-to-line voltage holors and the line-
to-neutral voltage holors. One will establish this relationship at the load terminals.
The observations one makes will also apply at the energy-source terminals. The
line-to-line voltage holors at the terminals of the load in terms of the line-to-neutral
voltage holors at the three-phase load are
Viab = Vian − Vibn; (11.154)

Vibc = Vibn − Vicn; (11.155)

and
Vica = Vicn − Vian. (11.156)
The double-subscript notation in the voltage holor equations indicates that the
voltage holor is a drop from the first subscript to the second subscript.
The relationships given by equations (11.150)–(11.152) are shown in figure 11.61.
Because one is interested in the balanced state, one has to omit the neutral conductor

Figure 11.60. An equivalent single-phase AC electrical homogeneous continuous dynamical system.

11-92
 Mechatronics

Figure 11.61. Line-to-line and line-to-neutral voltage holors.

from figure 11.61. To show the relationship between the line-to-line voltage holors
and the line-to-neutral voltage holors, one assumes a negative, or acb, sequence.
One arbitrarily chooses the line-to-neutral voltage holor as the reference. Thus,
Vian = Vϕ ∠0; (11.157)

Vibn = Vϕ ∠2π /3; (11.158)

and
Vicn = Vϕ ∠4π /3 = Vϕ ∠ − 2π /3, (11.159)
where UΦ represents an effective value of the line-to-neutral voltage holor.
When one substitutes equations (11.153)–(11.155) into equations (11.150)–
(11.152), respectively, one obtains
2π 11π
Viab = Vian − Vibn = Vϕ ∠0−Vϕ ∠ = 3 Vϕ ∠ .; (11.160)
3 6

2π 4π π
Vibc = Vibn − Vicn = Vϕ ∠ – Vϕ ∠ = 3 Vϕ ∠ ; (11.161)
3 3 2
and
Vica = Vicn − Vian = Vϕ ∠4π /3−Vϕ ∠0 = 3 Vϕ ∠7π /6. (11.162)

From equations (11.156)–(11.158), one sees that:

• the effective value of the line-to-line voltage holor is 3 times the effective
value of the line-to-neutral voltage holor;
• the line-to-line voltage holors form a balanced three-phase set of voltage holors;
• the set of line-to-line voltage holors lags the set of line-to-neutral voltage
holors by π/6.

These observations are summarised in the holor diagram of figure 11.62.

11-93
 Mechatronics

Figure 11.62. Holor diagrams showing the relationship between line-to-line voltage holors and line-to-neutral
voltage holors in a balanced Y–Y three-phase AC electrical homogeneous continuous dynamical system: (a)
acb sequence and (b) abc sequence.

One now observes that if the line-to-neutral voltage holor is known at some point
in the balanced Y2Y three-phase AC electrical homogeneous continuous dynamical
system, the line-to-line voltage holors at the same point are also known, and vice
versa. In addition, one must make some additional comments on terminology. In the
Y2Y three-phase AC electrical homogeneous continuous dynamical system, the
line-to-neutral voltage holor is also termed the ‘phase voltage holor’. For concision,
the line-to-line voltage holor is also termed the ‘line voltage holor’.
The ‘phase-current holor’ is defined as the current holor in each phase of the load
or, at the energy-source end of the AC electrical homogeneous continuous
dynamical system, the current holor in each phase of the AC EM generator.
The ‘line-current holor’ is defined as the current holor of each phase of the
transmission line. For the Y–Y arrangement, the phase-current holor and line-
current holor are identical. Because three-phase AC electrical homogeneous con-
tinuous dynamical systems are designed to handle large quantities of electrical
energy, all voltage-holor and current-holor specifications and computations are
given in terms of effective values or root-mean-square (rms) values. Finally, the
Greek letter phi (Φ) is usually used in the literature to denote a per-phase quantity.

11.7.12 Holor algebra analysis of the wye-delta three-phase AC electrical

homogeneous continuous dynamical system
If the load in a three-phase AC electrical homogeneous continuous dynamical
system is connected in a delta (Δ), it can be transformed into a wye (Y) by using
the Δ2Y transformation discussed in section 11.7. When the load is balanced, the
impedance holor of each leg of the Y-load is one-third of the impedance holor of the
Δ load. Thus
1 Δ
ZijY = Zij , (11.163)
3
which follows directly from equations (11.166)–(11.168).

11-94
 Mechatronics

Once the Δ load has been replaced by its Y equivalent, the Y-energy-source,
Δ-load, three-phase AC electrical homogeneous continuous dynamical system can
be physically modelled by the equivalent single-phase AC electrical homogeneous
continuous dynamical system in figure 11.60.
After one has the equivalent single-phase AC electrical homogeneous continuous
dynamical system to compute the line or line-to-line-current holors, one can find the
current holor in each leg of the original Δ load by simply dividing the line or line-to-
line-current holors by 3 and shifting them through π/6.
This relationship between line or line-to-line-current holors and phase or line-to-
neutral current holors in the Δ load can be derived using the AC electrical
homogeneous continuous dynamical system in figure 11.63.
When a load or energy source is connected in a delta (Δ), the current in each leg of
the delta is the phase current and the voltage across each leg is the phase voltage.
One can see from figure 11.63 that in the Δ configuration, the phase voltage is
identical to the line voltage. To see the relationship between the phase-current holors
and the line-current holors, one will assume a negative phase sequence and let Iϕ
represent an effective value of the phase current. It follows, then, that
j
IAB = Iϕ ∠0; (11.164)

j
IBC = Iϕ ∠2π /3; (11.165)

and
j
ICA = Iϕ ∠4π /3 = Iϕ ∠ − 2π /3. (11.166)
n
In writing these equations, one has arbitrarily selected IAB as the reference current
holor. One can write the line-current holors in terms of the phase-current holors by
direct application of KFL. Thus

Figure 11.63. AC electrical homogeneous continuous dynamical system used to establish the relationship
between line-current holors and phase-current holors in a balanced delta (Δ) load.

11-95
 Mechatronics

j j j
IAa = IAB − ICA = Iϕ ∠0 − Iϕ ∠4π /3 = 3 Iϕ ∠π /6; (11.167)

j j j
IBb = IBC − IAB = Iϕ ∠2π /3 − Iϕ ∠0 = 3 Iϕ ∠7π /6; (11.168)

and
ICcj = ICA
j j
− IBC = Iϕ ∠4π /3 − Iϕ ∠2π /3 = 3 Iϕ ∠11π /6. (11.169)

When one compares equations (11.163)–(11.165) with equations (11.160)–(11.162),

one sees that an effective value of the line-current holors is 3 times an effective
value of the phase-current holors, and the set of line-current holors ‘leads’ the set of
phase-current holors by π/6. One will leave it to the reader to verify that for a
positive phase sequence, the line-current holors are 3 times larger than the phase-
current holors and ‘lag’ the phase-current holors by π/6. The relationship between
the line-current holors and the phase-current holors of a Δ-connected load is
summarised in figure 11.64.

11.7.13 Holor algebra analysis of the delta–wye three-phase AC electrical

homogenerous continuous dynamical system
In the Δ2Y three-phase AC electrical homogeneous continuous dynamical system,
the energy source is Δ connected and the load is Y connected. One can obtain the
equivalent single-phase AC electrical homogeneous continuous dynamical system by
replacing the balanced Δ-connected energy source by a Y equivalent. One can
obtain the Y equivalent of the energy source by dividing the internal phase voltage
holors of the Δ energy source by 3 and shifting this set of three-phase voltage
holors by π/6 if the phase sequence is negative and by – π/6 if the phase sequence is
positive. The internal impedance holor of the Y equivalent is one-third of the
internal impedance holor of the Δ energy source.
The Y-equivalent AC electrical homogeneous continuous dynamical system of a
positive-sequence Δ-connected energy source is illustrated in figure 11.65.

Figure 11.64. Holor diagrams showing the relationship between line-current holors and phase-current holors
in a delta-connected load. (a) Negative sequence. (b) Positive sequence.

11-96
 Mechatronics

Figure 11.65. The Y equivalent of a balanced three-phase Δ-connected energy source of a positive phase
sequence. (a) Δ energy source and (b) Y equivalent.

For a positive phase sequence, the set of Δ energy-source current holors (IBA j j
, ICB
and IAC in figure 11.65) leads the set of line or line-to-line-current holors
j

j
IAa j
, IBb and ICcj by π/6. An effective value of the phase or line-to-neutral currents
is 1/ 3 times an effective value of the line or line-to-line currents. For a negative
phase sequence, the phase or line-to-neutral currents in the energy source lag the line
or line-to-line currents. To show that the Y energy source of figure 11.65(b) is
equivalent to the Δ energy source of figure 11.65(a), it is only necessary to show that
the two AC electrical homogeneous continuous dynamical systems produce the same
terminal conditions for any balanced ‘external’ connections applied to the terminals
A, B and C. The two test conditions that are easiest to prove are open circuit and
short circuit. For open-circuit conditions, the three line-current holors are zero and
the two AC electrical homogeneous continuous dynamical systems are equivalent if
they deliver the same voltage holors between the terminals A, B and C. For an
external short circuit connecting the terminals A, B and C, the line or line-to-line
voltage holors are zero and the two AC electrical homogeneous continuous
dynamical systems are equivalent if they deliver the same line or line-to-line-current
holors.

11.7.14 Holor algebra analysis of the delta–delta three-phase AC electrical

homogeneous continuous dynamical system
In the Δ–Δ three-phase AC electrical homogeneous continuous dynamical system,
both the energy source and the load are Δ connected. The equivalent single-phase AC
electrical homogeneous continuous dynamical system of balanced Δ2Δ three-phase
AC electrical continuous dynamical system is obtained by replacing both the energy
source and the load with their Y equivalents. As before, the Y-equivalent AC
electrical homogeneous continuous dynamical system is used to solve for line or line-
to-line-current holors and phase line-to-neural voltage holors. Once one knows the
line or line-to-line-current holors, one can find the phase or line-to-neutral current
holors in both the load and the energy source using the holor techniques described in
sections 11.6 and 11.7. The phase or line-to-neutral voltage holors can be converted
to line or line-to-line voltage holors, as described in section 11.5.

11-97
 Mechatronics

11.7.15 Delta-to-wye transformations

The rules of combining impedance holors for making delta-to-wye (Δ2Y) trans-
formations are the same as those for resistor holors. The only difference is that
combining holor impedances involves the algebraic manipulation of hypernumbers,
i.e. holors. The Δ2Y discussed in section 11.7 concerning resistive AC electrical
homogeneous continuous dynamical systems also applies to impedance holors. The
Δ-connected impedance holors along with the Y-equivalent AC electrical homoge-
neous continuous dynamical system are defined in figure 11.66.
The Y-impedance holors as functions of the Δ-impedance holors are given by the
following:
δ ki
Zijan = Z caZijab;
ab ij (11.170)
Z jkbc ca
+ Z jk + Z jk

δ ki
Zijbn = Z abZijbc ;
ab ij (11.171)
Z jkbc ca
+ Z jk + Z jk
δ ki
Zijcn = Z bcZijca .
ab ij (11.172)
Z jkbc ca
+ Z jk + Z jk

It is also possible to reverse the Δ2Y transformation. That is, one can start with the
Y structure and replace it with an equivalent Δ structure.
The Δ-impedance holors as functions of the Y-impedance holors are given by the
following:
δi
( )
Zijab = kcn ZijanZijbn + ZijbnZijcn + ZijcnZijan ;
Z jk
(11.173)

δ ki
Zijbc = (
Z jkan
)
ZijanZijbn + ZijbnZijcn + ZijcnZijan ; (11.174)

Figure 11.66. The delta-wye (Δ2Y) transformation.

11-98
 Mechatronics

δ ki
Zijca = bn
Z jk
( )
ZijanZijbn + ZijbnZijcn + ZijcnZijan . (11.175)

The derivation of equations (11.166)–(11.168) or equations (11.169)–(11.171)

follows the same thought process as was used to derive the corresponding equations
for pure resistive AC electrical homogeneous continuous dynamical systems. In fact,
if one compares equations (11.77)–(11.79) with equations (11.166)–(11.168), and
equations (11.80)–(11.82) with equations (11.169)–(11.171), one can see that the
resistance holor Rij has been replaced by the impedance holor Zij. The usefulness of
the Δ-Y transformation in holor-domain AC electrical homogeneous continuous
dynamical system analysis will be illustrated in some examples below.

11.7.16 Repetitive example I

In section 11.4, one discussed the simple electrical energy distribution system illustrated
in figure 11.67. This is termed a ‘single-phase energy distribution system’. It utilises a
single energy source, a single-phase ME generator, or an EE transformer secondary.
We shall now discuss an electrical energy distribution system that utilises three
energy sources, or a three-phase ME AC generator, or EE AC transformer
secondaries that differ in phase from each other by 2π/3. This is termed a ‘three-
phase AC electrical homogeneous continuous dynamical system’.
A physical model for a three-phase ME generator or EE transformer secondaries
is shown in figure 11.68.
For the most part, three-phase EM generators and EE transformers are
constructed as one single ME generator or EE transformer. However, figure 11.68
is an accurate physical model for an ideal three-phase ME generator or EE
transformer secondaries.
In figure 11.68 one shows the transmission line leaving the ME generator or EE
transformer secondaries. The conductors A–a, B–b and C–c are termed the lines,
while the conductor N–n is termed ‘neutral’. One can talk of line or line-to-line
energy-source voltages and phase or line-to-neutral energy-source voltages. Let us
compute these holor values.

Figure 11.67. A simplified electrical energy distribution system.

11-99
 Mechatronics

Figure 11.68. A physical model of a three-phase ME generator or EE transformer secondaries.

Figure 11.69. A holor representation of the holor phase or line-to-neutral energy-source voltages.

The phase or line-to-neutral energy-source voltage holors for a positive phase

sequence are

EiAN = Eϕ ∠0 ⎫
⎪
EiBN = Eϕ ∠ − 2π /3 ⎬. (11.176)
⎪
EiCN = Eϕ ∠ − 4π /3 = Eϕ ∠ + 2π /3⎭
A holor diagram representing these voltage holors is shown in figure 11.69. Note
that three voltage holors differ in phase by 2π/3.
Now let us compute the line or line-to-line energy-source voltage holors.
EiAB = EiAN − EiBN = E∅ − E∅∠ − 2π /3. (11.177)

Proceeding similarly, one has

EiAB = 3 E∅∠π /6 ⎫
⎪
EiBC = 3 E∅∠π /6 − 2π /3⎬ . (11.178)
⎪
EiCA = 3 E∅∠π /6 − 4π /3⎭

The line or line-to-line energy-source voltage holors are also three-phase voltage
holors. That is, they are equal and their phase angles differ by 2π /3.
The line or line-to-line energy-source voltage holors are displaced by π /6 from
the phase or line-to-neutral energy-source voltage holors. The line or line-to-line

11-100
 Mechatronics

energy-source voltage holors are 3 times as large as the phase or line-to-neutral

energy-source voltage holors.
Let us now analyse the three-phase AC electrical homogeneous continuous
dynamical system of figure 11.70. This represents a three-phase AC generator or
EE AC transformer secondaries connected to a three-phase load. The internal
energy-source impedance holors ZijAN , ZijBN and ZijCN are included to physically
model the three-phase AC generator losses or the losses of the EE AC transformer
secondaries.
The three-phase load consists of impedance holors Zijan, Zijbn and Zijcn, and this is
termed a Y-connected three-phase load. Note that its configuration resembles the
letter Y (which is upside down in this case).
The impedance holors ZijAa, ZijBb, Zmn
Cc
and ZijNn are included to physically model
the transmission-line losses.
If one applies mesh-current holor analysis to this three-phase AC electrical
homogeneous continuous dynamical system, one obtains

EiAN = Eϕ ∠0 = (ZijAN + ZijAa + Zijan + ZijNn)I Aa

j j
+ ZijNnIBb ⎫
+ ZijNnICcj
⎪
2π ⎪
Nn j ⎪
EiBN = Eϕ ∠−
3
j
= ZijNnI Aa ( BN Bb bn Nn j
)
+ Zij + Zij + Zij + Zij IBb + Zij ICc ⎬ . (11.179a )
⎪
4π j ⎪
EiCN = Eϕ ∠−
3
j
= ZijNnI Aa Nn j
(CN Cc cn Nn
+ Zij IBb + Zij + Zij + Zij + Zij ICc ⎪
⎭ )
Equations (11.179a) can also be written in matrix notation as follows:

⎡ AN ⎤ ⎡ Z AN + Z Aa + Z an + Z Nn ZijNn ZijNn
⎤⎡ j ⎤
⎢Ei ⎥ ⎢ ij ij ij ij ⎥ ⎢ I Aa ⎥
⎢ BN ⎥ ⎢ Nn BN + Z Bb + Z bn + Z Nn Nn
⎥⎢ j ⎥
⎢ iE ⎥ = ⎢ Z ij Z ij ij ij ij Z ij ⎥ ⎢ I ⎥, (11.179b )
⎢ CN ⎥ ⎢ ⎥ ⎢ Bb ⎥
⎢⎣ E i ⎥⎦ ⎢ ZijNn Nn
Zij CN
Zij + Zij + Zij + Zij ⎥ ⎢ I j ⎥
Cc cn Nn
⎣ ⎦ ⎣ Cc ⎦

or

⎡ E AN ⎤ ⎡ ZijϕAA ZijNn ZijNn ⎤⎡ I j ⎤

⎢ iBN ⎥ ⎢ Nn ⎥⎢ Aa ⎥
⎢
⎢ Ei ⎥ = Zij
ϕBB
Zij Nn ⎥
Zij ⎢ IBb j
⎥, (11.179c )
⎢ CN ⎥ ⎢ Nn ⎥
ϕCC ⎢ j ⎥
⎣ Ei ⎦ ⎢⎣ Zij Zij Nn
Zij ⎥⎦⎣ ICc ⎦

where
ZijAN + ZijAa + Zijan + ZijNn = ZijϕAA,

ZijBN + ZijBb + Zijbn + ZijNn = ZijϕBB ,

and
ZijCN + ZijCc + Zijcn + ZijNn = ZijϕCC .

11-101
 Mechatronics

Figure 11.70. A three-phase AC electrical homogeneous continuous dynamical system with a wye (Y)-
connected three-phase load.

11-102
 Mechatronics

These can be solved in the same way as any of the sets of simultaneous holor
equations that we have considered. An especially interesting case occurs when the
three-phase AC electrical homogeneous continuous dynamical system is balanced.
In this case, one assumes that

Zijan = Zijbn = Zijcn = ZijφY . (11.180)

In addition, one assumes that the three internal energy-source impedance holors and
line impedance holors, as well as the neutral impedance holor, respectively, are
equal.
ZijAN = ZijBN = ZijCN = Zijγ . (11.181a )

ZijAa = ZijBb = ZijCc = Zijλ, and ZijNn = Zijν. (11.181b)

Equations (11.179a) for the Y–Y balanced three-phase AC electrical homogeneous

continuous dynamical system now become

EiAN = Eϕ ∠0 = (Zijγ + Zijλ + ZijφY + Zijν )IAa

j j
+ ZijνIBb + ZijνICcj ⎫
⎪
⎪
2π ⎪
EiBN = Eϕ ∠− j
= ZijνIAa + (Zijγ + Zijλ + ZijφY + Zijν )IBb
j
+ ZijνICcj ⎬ . (11.182a )
3 ⎪
4π ⎪
EiCN = Eϕ ∠− j
= ZijνIAa j
+ ZijνIBb + (Zijγ + Zijλ + ZijφY + Zijν )ICcj ⎪
3 ⎭

Equations (11.182a) can also be written in matrix notation as follows:

⎡ ⎤⎡ j ⎤
⎡E AN⎤ ⎢ Zijγ + Zijλ + ZijφY + Zijν Zijν Zijν ⎥⎢ I Aa ⎥
⎢ i ⎥ ⎢ ⎥⎢ j ⎥
⎢Ei ⎥ = ⎢
BN ν γ λ φY ν ν
Zij Zij + Zij + Zij + Zij Zij ⎥⎢ I Bb ⎥, (11.182b )
⎢ CN ⎥ ⎢ ⎥ j
⎢⎣Ei ⎥⎦ Zij + Zij + Zij + Zij ⎥⎦⎢⎣ ICc ⎥⎦
ν ν γ λ φY ν
⎢⎣ Zij Zij

or

⎡ E AN ⎤ ⎡ ZijϕY Zijν Zijν ⎤⎡ I j ⎤

⎢ iBN ⎥ ⎢ ν ⎥⎢ Aa ⎥
⎢ Ei ⎥ = ⎢ Zij Zij Zijν ⎥⎢ IBb
ϕY j
⎥, (11.182c )
⎢ CN ⎥ ⎢ ⎥
ϕY ⎢ j ⎥
⎣ Ei ⎦ ⎢⎣ Zij ν
Zij Zij ⎥⎦⎣ ICc ⎦
ν

where

Zijγ + Zijλ + Zijφ + Zijν = ZijϕY .

If one assumes that Zijγ = 0, because the energy-source winding impedance holors
are small and may be neglected for the Y–Y balanced three-phase AC electrical

11-103
 Mechatronics

homogeneous continuous dynamical system, equation (11.182b) can be written in

matrix notation as follows:
⎡ E AN ⎤ ⎡ Zijλ + ZijφY + Zijν Zijν Zijν ⎤⎡ j ⎤
⎢ iBN ⎥ ⎢ ⎥⎢ IAa ⎥
⎢ Ei ⎥ = ⎢ ⎥⎢ IBb
ν λ φY ν ν j
Zij Zij + Zij + Zij Zij ⎥ . (11.182d )
⎢ CN ⎥ ⎢ ⎥⎢ j ⎥
⎣ Ei ⎦ ⎢⎣ Zijν Zijν Zijλ + ZijφY + Zijν ⎥⎦⎣ ICc ⎦

If one assumes that Zijλ = 0, because the transmission-line impedance holors are
neglected for the short transmission lines for the Y–Y balanced three-phase AC
electrical homogeneous continuous dynamical system, equation (11.182d) can be
written in matrix notation as follows:
⎡ E AN ⎤ ⎡ ZijφY + Zijν Zijν Zijν ⎤⎡ j ⎤
⎢ iBN ⎥ ⎢ ⎥⎢ IAa ⎥
⎢ Ei ⎥ = ⎢ ⎥⎢ IBb
ν φY ν ν j
Zij Zij + Zij Zij ⎥. (11.182e )
⎢ CN ⎥ ⎢ ⎥⎢ j ⎥
⎣ Ei ⎦ ⎢⎣ Zijν Zijν ZijφY + Zijν ⎥⎦⎣ ICc ⎦

If one assumes that Zijν = 0, because the neutral-conductor impedance holor does
not exist for the Y–Y balanced three-phase AC electrical homogeneous continuous
dynamical system, equation (11.182e) can be written in matrix notation as follows:
⎡ E AN ⎤ ⎡ ZijφY 0 0 ⎤⎡ IAa j ⎤
⎢ iBN ⎥ ⎢ ⎥⎢ ⎥
⎢ Ei ⎥ = ⎢ 0 Zij 0 ⎥⎢ IBb
φY j
⎥. (11.182f )
⎢ CN ⎥ ⎢ ⎥
φY ⎢ j ⎥
⎣ Ei ⎦ ⎢⎣ 0 0 Zij ⎥⎦⎣ ICc ⎦

Now add the three equations (11.183)

EiAN + EiBN + EiCN = Eϕ(∠0 + ∠ − 2π /3 + ∠ − 4π /3)
(11.183)
(
= (Zijγ + Zijλ + ZijφY + 3Zijν ) IAa
j j
+ IBb )
+ ICcj .

Noting that
1(∠0 + ∠ − 2π /3 + ∠ − 4π /3) = 0, (11.184)
one has
j j
IAa + IBb + ICcj = 0. (11.185)

But,
j
INn (j
= − IAa j
+ IBb + ICcj .)
Hence
j
INn = 0. (11.186)

11-104
 Mechatronics

Thus, in a Y–Y balanced three-phase AC homogeneous electrical continuous

dynamical system, the current holor in the neutral conductor is zero. It is very
advantageous if the current in a transmission line is zero. Then there is no energy
loss in that conductor, and nor is there any voltage drop across it.
Let one contrast three-phase energy transmission with single-phase energy trans-
mission. Suppose that one has three ME generators or three EE transformer
secondaries and three loads.
If one connects them as three isolated single-phase AC electrical homogeneous
continuous dynamical systems, and three loads, then one needs six sets of
conductors. That is, each transmission line consists of a pair of conductors (see
figure 11.67).
If each conductor of the transmission line has a resistance Rλ, and the current in
each line has an effective value Irms, the energy dissipated will be 6Rλ(Irms ∧ 2).
Now let us contrast this with the balanced three-phase case. Only four conductors
are used, and in one of them the current is zero. Thus the energy dissipated by all
transmission lines is 3Rλ.(Irms ∧ 2). Thus, the energy loss is halved. Not only is there a
saving in energy, but there is also a saving in conductors, since three-phase
transmission requires four conductors rather than six.
Since there is no current in the neutral conductor, it seems as though it could be
omitted. If the load were always balanced, one would indeed not need the neutral
conductor. However, if the load is not exactly balanced, the neutral current will not
be zero.
Consider that figure 11.70 represents an energy distribution system;
Zijan, Zijbn and Zijcn could represent three different neighbourhoods. Statistically,
they would be equal loads, but actually they would differ slightly. In this case, the
neutral current would be much smaller than the line current, but would not be zero.
For this reason, the neutral conductor is always present in a three-phase energy
distribution system.
We have not shown any EE transformers in the three-phase energy-distribution
system model of figure 11.70. However, the ideas presented in figure 11.67 are
incorporated into the three-phase energy distribution system.
Let us continue with holor analysis of figure 11.70. Assume that the three-phase
j
electrical homogeneous continuous dynamical system is balanced. Since INn = 0, the
Nn ν
voltage drop across Zij = Zij = 0.
Then, from equation (11.183), one has

δ ki δ ki AN ⎫
j
IAa = E i
AN
= E i = Y ji AN
ϕ iE ⎪
Z jkγ + Z jkλ + Z jkφ Z jkϕ ⎪
⎪
ji BN ⎪
i i
j δk BN δ k BN
IBb = γ E i = E i = Yϕ i E ⎬, (11.187)
Z jk + Z jkλ + Z jkφ Z jkϕ ⎪
i i ⎪
j δk CN δ k CN ji CN ⎪
ICc = γ Em = ϕ Ei = Yϕ Ei
Z jk + Z jkλ + Z jkφ Z jk ⎪
⎭

11-105
 Mechatronics

where
Z jkγ + Z jkλ + Z jkφ = Z jkϕ ,

and
δ ki δ ki
= = Yϕji .
Z jkγ + Z jkλ + Z jkφ Z jkϕ

Applying loop-current holor analysis, one has

⎛ 2π ⎞ ⎫
⎝ 3 ⎠
( j
) j j
EiAB = Eϕ⎜∠0 − ∠− ⎟ = ZijAN + ZijAa + ZijBb + Zijab I Aa − ZijBbI Bb − ZijAaICc ⎪
⎪
⎛ 2π 4π ⎞ ⎪
BC
⎝ 3 3 ⎠
Bb j BN
(
Bb bc Cc j
) Cc j
Ei = Eϕ⎜∠− −∠− ⎟ = − Zij IAa + Zij + Zij + Zij + Zij I Bb − Zij ICc ⎬. (11.188a )
⎪
⎛ 4π ⎞ j j j
⎪
CA
⎝ 3 ⎠
Aa Cc CN
(Aa Cc
Ei = Eϕ⎜∠− −∠0⎟ = − Zij I Aa − Zij I Bb + Zij + Zij + Zij + Zij ICc ⎪ ca
⎭
)
Equations (11.188a) can also be written in matrix notation as follows:
⎡ E AB ⎤ ⎡ Z AN + Z Aa + Z Bb + Z ab −ZijBb −ZijAa ⎤⎡ j ⎤
⎢ i ⎥ ⎢ ij ij ij ij ⎥⎢ I Aa ⎥
⎢ E BC ⎥ = ⎢ Bb BN + Z Bb + Z bc + Z Cc Cc ⎥⎢ j ⎥
⎢ i ⎥ ⎢ ⎥⎢ I Bb ⎥, (11.188b)
− Zij Zij ij ij ij − Zij
⎢ E CA ⎥ ⎢ ⎥⎢ ⎥
⎣ i ⎦ ⎢⎣ −ZijAa Cc
−Zij Zij + Zij + Zij + Zij ⎥⎢⎣ I j ⎥⎦
CN Aa Cc ca
⎦ Cc

or
⎡ E AB ⎤ ⎡ ZijϕΔ −ZijBb −ZijAa ⎤⎡ I j ⎤
⎢ i ⎥ ⎢ ⎥⎢ Aa ⎥
⎢ EiBC ⎥ = ⎢ −ZijBb ZijϕΔ −ZijCc ⎥⎢ IBb
j
⎥, (11.188c )
⎢ CA ⎥ ⎢ ⎥
ϕΔ ⎢ j ⎥
⎣ Ei ⎦ ⎢⎣−Zij −Zij
Aa Cc
Zij ⎥⎦⎣ ICc ⎦

where
ZijAN + ZijAa + ZijBb + Zijab = ZijϕΔ,

ZijBN + ZijBb + Zijbc + ZijCc = ZijϕBC ,

and
ZijCN + ZijAa + ZijCc + Zijca = ZijϕCA.

The procedures discussed for solving simultaneous equations can be used here.
An especially interesting case occurs when the three-phase AC electrical homoge-
neous continuous dynamical system is balanced.

11.7.17 Repetitive example II

Figure 11.71 shows a three-phase AC homogenous electrical continuous dynamical
system with a triangle or delta (Δ)-connected three-phase load. The procedures

11-106
 Mechatronics

Figure 11.71. A three-phase AC electrical homogeneous continuous dynamical system with a delta (Δ)-
connected three-phase load.

discussed for solving simultaneous equations can be used here. An especially

interesting case occurs when the three-phase AC electrical homogeneous continuous
dynamical system is balanced.
In this case, one assumes that

Zijab = Zijbc = Zijca = ZijφΔ. (11.189)

In addition, one assumes that the three internal energy-source impedance holors
and line impedance holors are equal.
ZijAN = ZijBN = ZijCN = Zijγ , (11.189a )

11-107
 Mechatronics

ZijAa = ZijBb = ZijCc = Zijλ. (11.189b)

Equations (11.188a) for the Y2Δ balanced three-phase AC electrical homoge-

neous continuous dynamical system now become

⎛ 2π ⎞ ⎫
⎝ 3 ⎠
( j
E iAB = Eϕ⎜∠0 − ∠− ⎟ = Zijγ + Zijλ + Zijλ + ZijφΔ I Aa )j
− ZijλIBb j
− ZijλICc ⎪
⎪
⎛ 2 π 4 π ⎞ ⎪
E iBC = Eϕ⎜∠− −∠− ⎟ = − ZijλIAa
⎝ 3 3 ⎠
j
(
+ Zijγ + Zijλ + ZijφΔ + Zijλ IBbj j
− ZijλICc)⎬ (11.190a )
⎪
⎛ 4π ⎞ ⎪
CA
⎝ 3 ⎠
λ j λ j γ λ
(
E i = Eϕ⎜∠− −∠0⎟ = − Zij I Aa − Zij IBb + Zij + Zij + Zij + Zij ICc
λ φΔ j
⎪
⎭
)

⎡ φΔ ⎤⎡ j ⎤
⎡ E AB ⎤ ⎢ Zijγ + Zijλ + Zijλ + Zij −Zijλ −Zijλ ⎥⎢ I Aa ⎥
⎢ i ⎥ ⎢ ⎥⎢ j ⎥
φΔ
⎢ EiBC ⎥ = ⎢ −Zijλ Zijγ + Zijλ + Zij + Zijλ −Zijλ ⎥⎢ I Bb ⎥. (11.190b)
⎢ CA ⎥ ⎢ ⎥ j
⎢⎣ Ei ⎥⎦ Zijγ + Zijλ + Zijλ + Zij ⎥⎦⎢⎣ ICc ⎥⎦
φΔ
⎢⎣ −Zijλ −Zijλ

The neutral is not connected. The load is called a delta (Δ) configuration because of
its resemblance to the Greek letter delta.
If one assumes that Zijγ = 0, and Zijλ = 0, because the energy-source winding
impedance holors and the transmission-line impedance holors, respectively, are
small and may be neglected for the Y2Δ balanced three-phase AC electrical
homogeneous continuous dynamical system, equation (11.190b) can be written in
matrix notation as follows:

⎡ E AB ⎤ ⎡ ZijφΔ 0 0 ⎤
⎢ i ⎥ ⎢ ⎥
⎢ EiBC ⎥ = ⎢ 0 ZijφΔ 0 ⎥ . (11.190c )
⎢ CA ⎥ ⎢ ⎥
⎣ Ei ⎦ ⎢⎣ 0 0 ZijφΔ ⎥⎦

Let us now consider the case of the Y2Δ balanced three-phase AC electrical
homogeneous continuous dynamical system. Note that equations (11.185) and
(11.186) apply.
Substituting these equations and adding equation (11.184), one has
j j
0 = ZijφΔ(IAa + IBb + ICcj ) . (11.191)

Hence,
j j
IAa + IBb + ICcj = 0. (11.110)

In fact, it can be shown that if the three-phase load is balanced, the load current
holors will be equal in magnitude and will differ in phase by 2π/3.

11-108
 Mechatronics

11.8 Parallel-series resistance-inductance-capacitance AC electrical

commutation matrixer
One will find the node-to-datum holor equations in matrixer form for the parallel-
series resistance-inductance-capacitance (R–L–C) AC electrical commutation
matrixer shown in figure 11.72(a). One will also express the effective (rms) value of
Ei2 in terms of the electrical commutation matrixer components and input sources.
Figure 11.72(b) shows the admittance-holor diagram representation of the AC
electrical commutation matrixer. There are a total of four nodes or terminals.
However, two of them (the ground terminal and the voltage-source terminal holor,
Ei ) are already defined.

Figure 11.72. AC electrical commutation matrixer.

11-109
 Mechatronics

Thus, there are only two independent nodes and the admittance-holor matrix will
be of the form:
⎡ Ij ⎤
⎡Y ji ji ⎤⎡ 1 ⎤ ⎢ ⎥
Node1 ⟹ YAn E ⎢ 0 1
⎥
⎢ Aa ⎥⎢ i ⎥ = ωL Ei .
Node2 ⟹ ⎢⎣YNa
ji ji
⎥
YNn ⎦⎣ Ei2 ⎥⎦
⎢ ⎢ 1
⎥
⎢ − ωL 0 ⎥
⎣ ⎦

First, let us consider the flow into the nodes from the inputs. The flow into node 1
from the input voltage-source holors is merely the effective value of the current-
holor source, I j . However, in this case the flow into node 2 depends on the voltage-
δ ki
source holor, Ei and the inductive admittance holor, . This is a flow holor,
ωLjk
δ ki
ωLjk i
E. As a result, the known column-vector flow-holor is as indicated above.
The matrix admittance-holor component Y11ji is the parallel combination of 1/R1
and ωC. This is the condition that exists when all terminals except Ei1 are grounded.
Thus

ji 1/R1 ωC
YAa = .
−ωC 1/R1

To find the matrix admittance-holor element YNn jii

one reverses the procedure. Now
2
all the terminals except Ei are grounded.
The admittance holor from node 2 in this case is the parallel combination of ωC,
1/R2 and 1/ωL or

ji 1/R2 1/ ωL + ωC
YNn = .
−1/ ωL − ωC 1/R2

In this case, there is only one common term between Y11ji and Y12ji , and it is ωC, so
that:

ji ji 0 −ωC .
YAn = YNa =
ωC 0

Moreover, the complete hypermatrix representation is:

⎡ 1 ⎤
⎢ R1 ωC 0 −ωC ⎥⎡ ⎤
⎢ ⎥⎢ Ec1 ⎥ ⎡ Ic ⎤
⎢ −ωC R
1 ωC 0 ⎥⎢ 1 ⎥ ⎢ ⎥
Es Is
⎢ ⎥⎢ ⎥=⎢ ⎥.
1

⎢ 1 1
− ωC ⎥⎢ Ec2 ⎥ ⎢⎢ 0 1/ ωL Ec ⎥
⎢ 0 −ωC R2 ωL ⎥⎢ ⎥ Es ⎥⎦
E s2 ⎥⎦ ⎣
−1/ ωL 0
⎢ ωC 0 1 1 ⎥⎢⎣
⎢⎣ − ωL + ωC ⎥⎦
R2

11-110
 Mechatronics

The effective value of Ei2 in terms of determinant is:

1/R1 ωC Ic
−ωC 1/R1 Is
0 −ωC 0 1/ ωL Ec
Ec2 ωC 0 −1/ ωL 0 Es
Ei2 = =
E s2 1/R1 ωC 0 −ωC
−ωC 1/R1 ωC 0
0 −ωC 1/R2 1/ ωL + ωC
ωC 0 −1/ ωL − ωC 1/R2

and, evaluating the determinants,

Ec2
Ei2 =
E s2
1/R1 ωC 0 1/ ωL Ec Ic
− 0 − C
ω
−ωC 1/R1 −1/ ωL 0 Es ωC 0 Is
= .
1/R1 ωC 1/R2 1/ ωL + ωC
− 0 − C 0 −ωC
ω
−ωC 1/R1 −1/ ωL − ωC 1/R2 ωC 0 ωC 0

11.9 Application of Thévenin’s theorem

We may now obtain the electrical commutation-matrixer transfer function EiL /Ek
for the electrical commutation matrixer shown in figure 11.73(a). The procedure is to
obtain the Thévenin equivalent of the section of the original electrical commutation
matrixer reproduced in figure 11.73(b). Since there is no current-holor flow through
Zij3, the open output electrical-matrixer’s voltage holor is obtained directly from the
voltage-holor divider relation:

δki
EiTH = Zij2Ei .
Z 1jk + Z jk3

The equivalent impedance holor is the driving-point impedance holor between

port terminals a and b when the voltage-source holor is suppressed and this is
readily seen to be the equivalent of Zij3 in series with the parallel combination of Zij1
and Zij3.

δ ki δ ki
ZijTH = Zij1 +
Z 1jk + Z jk2
Z 1 2
Z
ij ij =
Z 1jk + Z jk2
(Zij1Zij2 + Zij2Zij3 + Zij3Zij1).

11-111
 Mechatronics

Figure 11.73. Thévenin equivalent electrical commutation matrixer.

Using these values in figure 11.73(c), one obtains:

EiL δ ki
= ZijL,
E kTH Z jkTH + Z jkL

and substituting for EiTH and ZijTH gives

EiL δi δi
= 1 k 2 Zij2. TH k L ZijL
Ek Z jk + Z jk Z jk + Z jk
δ ki
= Zij2ZijL.
Z 1jkZ jk2 + Z jk2 Z jk3 + Z jk3 Z 1jk + Z jkL (Z 1jk + Z jk2 )

11-112
 Mechatronics

This electrical commutation matrixer is sufficiently simple to check this answer by

obtaining the solution to the node holor equations of the entire electrical commu-
tation matrixer.
⎡ δ ki δ ki δ ki δi ⎤
⎢ Z1 + Z 2 + Z 3 − k3 ⎥⎡ 1 ⎤ ⎡ δ ki ⎤
Z jk
⎢ jk jk jk ⎥⎢ Ei ⎥ = ⎢ Z 1 E ⎥ .
⎢ δ k ⎥⎢ ⎢ jk i ⎥
⎣ Ei ⎥⎦ ⎣
j j j L
δk δk
⎢ − 3
Z 3jk
+
ZL
⎥ 0 ⎦
⎣ Z jk jk ⎦

The solution is
δ ki δ ki δ ki
EiL = ⎛ δi ⎞ ⎛ ⎞ ⎛ ⎞ ⎛ ⎞ 3 Z 1 Ei
δi δ i δi δ i δ i δ i Z
⎜ k + k + k ⎟⎜ k + k ⎟ − ⎜ k ⎟⎜ k ⎟ jk jk
⎜Z 1 Z2 Z 3 ⎟ ⎜ Z3 ZL ⎟ ⎜ Z 3 ⎟⎜ Z 3 ⎟
⎝ jk jk jk ⎠⎝ jk jk ⎠ ⎝ jk ⎠⎝ jk ⎠
δ ki
= δi
Zij2Ei
ZL
jk
(
k Z1 Z 2 + Z 2 Z 3 + Z 3 Z1 + Z1 + Z 2
jk jk jk jk jk jk jk jk )
δ ki
= Zij2ZijLEi .
Z 1jkZ 2jk + Z 2jkZ 3jk + Z 3jkZ 1jk + Z L 1
( 2
jk Z jk + Z jk )

Figure 11.74. Application of Thévenin’s theorem. (a) Original electrical commutation matrixer. (b) Thévenin
equivalent electrical commutation matrixer.

11-113
 Mechatronics

In some cases the process of obtaining the Thévenin equivalent by the simple
procedure used in this example of the electrical commutation matrixer becomes
complicated and it is simpler to obtain the open-output voltage holor and driving-
point impedance holor directly from a set of node holor equations.
A simple application of Thévenin’s theorem that is often very useful is that of
transforming an energy-transfer-source holor and the impedance holor connected
across its terminals (figure 11.74(a)) to an equivalent series connected energy-

Figure 11.75. Superposition theorem.

11-114
 Mechatronics

potential-source holor and impedance holor (figure 11.74(b)). The electrical com-
mutation matrixers are equivalent provided that EiTH = ZijI j and ZijTH = Zij .

11.10 Application of superposition theorem

An electrical commutation matrixer consisting of two resistors is driven by a
voltage-source holor and a current-source holor, as shown in figure 11.75(a).
One will find the voltage holor across the resistor R2 . The procedure is simply to
obtain the values of Ei2 when only one energy source is acting at a time and then add
these solutions.
In this case, there are two energy sources, so the problem is to find Ei2 in each of
the electrical commutation matrixers shown in figures 11.75(b) and (c) and add
them. From figure 11.75(b), the first component is a voltage divider holor
⎡E 2⎤ 1 ⎡ R2 0 ⎤⎡ Ec ⎤ R2
Ei2 = ⎢ 2 ⎥ =
c
⎡ ⎤ ⎡ ⎤ ⎢ ⎥⎢ ⎥ = Ei .
⎢⎣ E s ⎥⎦ R1 0 R2 0 ⎣ 0 R2 ⎦⎣ Es ⎦ R1 + R2
⎢ ⎥ ⎢
+ ⎥
⎣ 0 R1⎦ ⎣ 0 R2 ⎦

From figure 11.75(c), the second component of Ei2 is found by noting that this is
the voltage holor across the parallel combination of R1 and R2
⎡E 2⎤ 1 ⎡ R1 0 ⎤⎡ R2 0 ⎤⎡ I c ⎤ R1R2 j
Ei2 = ⎢ 2 ⎥ =
c
⎡ ⎤ ⎡ ⎤ ⎢ ⎥⎢ ⎥⎢⎣ s ⎥⎦ = I .
⎢⎣ E s ⎥⎦ R1 0 R2 0 ⎣ 0 R1⎦⎣ 0 R2 ⎦ I R1 + R2
⎢ ⎥ ⎢
+ ⎥
⎣ 0 R1⎦ ⎣ 0 R2 ⎦

The complete sinusoidal response is the sum of these two expressions:

R2 R1R2 j R2
Ei2 =
R1 + R2
Ei +
R1 + R2
I =
R1 + R2
(
Ei + R1I j . )

References
Baldwin M and Geary D 2006 High reliability DC power distribution system (HRDCPDS) for
critical environments (DCUPS). A patent pending system presentation, Baldwin
Technologies Inc., Environmental Support Solutions, [Link] (46 slides)
Barthold L 2006 Current-Modulated HVDC Transmission: An Efficient Way to Convert AC
Lines to DC PowerPoint presentation by Ram Adapa, PhD, EPRI
Bewley L V 1961 Tensor Analysis of Electric Circuits and Machines (New York: Ronald Press)
Cisco Systems 2006 Power over Ethernet: Cisco Inline Power and IEEE 802.3af White Paper,
available for download at [Link]
Crudele F D, Key T and Mansoor A 2004 Commercial and Industrial Applications Getting
Ready for Direct-Current Power Distribution IEEE Power Systems Conf. and Exposition
(New York)
Energy Market Profiles database 2006 Energy Use Strategic Service Online Tool EPRI Solutions
(Primen), [Link].
EPRI 2006 DC Power Production, Delivery and Utilization. An EPRI White Paper, June 2007.
[Link]/documents/EPRI_DCpower_WhitePaper_June_2006.pdf

11-115
 Mechatronics

EPRI PEAC Corp 2002 Development and Demonstration of DC Photovoltaic Applications: Task
1 Project Results Draft report EPRI Solutions
Fijalkowski B 1967 Zastosowanie rachunku holorowego w teorii pradow przemiennych
(Application of a holor calculus in the theory of alternating currents) Przeglad elektrotech-
niczny (Electrotechnics Review) (Warszawa: Rok XLIII, Zeszyt 2). (In Polish)
Galvin Electricity Initiative 2005 Transforming Electricity Service Reliability and Value for the
21st Century, available for download at [Link].
Hoseman G, Krolikowski L, Raatz E, Shapilov E D, Turek-Kwiatkowska L and Vieselovsky O N
1999 Zhizn i delyatelnost Mikhaila Dolivo-Dobrovolskogo—vydaiushchegosia inzhenera elek-
trika evropeiskogo masshtaba Proc. of the 4th Int. Conf. on Unconventional Electromechanical
and Electrical Systems (St Petersburg) vol. 1 pp 59–70 (In Russian)
IEEE History Center 2006 History Center. Rutgers-The State University, Contact history@ieee.
org. Material available at the IEEE website, [Link]/organizations/ history_center
Kasatkin A and Perekalin M 1970 Basic Electrical Engineering (Moscow: Mir Publishers)
Kron G 1959 Tensors for Circuits (New York: Dover)
Kuznetsov M 1960 Fundamentals of Electrical Engineering (Moscow: Mir Publishers)
Moon P H and Spencer D E 1986 Theory of holors. A generalization of Tensors (Cambridge:
Cambridge University Press) (Digitally printed, 2005)
Savage P 2003 What would Thomas Edison do? Energy Pulse, September 2003
Staff of the Department of Electrical Engineering at MIT 1940 Electric circuits (New York:
Wiley) ch. 8
Woodford D 1998 HDVC Transmission (Manitoba: HVDC Research Centre) [Link]

11-116
 IOP Publishing

Mechatronics
Dynamical systems approach and theory of holors
B T Fijalkowski

Chapter 12
Mechanical homogenous continuous dynamical
systems as mechanical commutation matrixers

Mechanics is the paradise of mathematical sciences … Mechanical science is

most noble and useful beyond all others, since by means of it, all animated bodies
that have motion perform all their operations.
Leonardo DaVinci

12.1 Simple plane-motion mechanical homogenous continuous

dynamical system
Consider a simple plane-motion mechanism in txerms of a parameter holor. In this
case, it is possible to express the coordinates x and y of all the significant points of
the mechanism by means of simple analytical expressions containing a single
parameter holor. It may be advantageous in such a case to determine the absolute
velocity and the absolute acceleration of the various points of the mechanism
directly, since the components of the velocity and acceleration of a given point may
be obtained by differentiating the coordinates x and y of that point.
Let one consider the rod-length holor Li , whose extremities slide, respectively, in a
horizontal and a vertical track (figure 12.1).
The coordinates xA and yB of the extremities of the rod AB may be expressed in
terms of the angle θ the rod AB forms with the vertical:

T lc
Li = (lc , ls ) = lc ls =
ls
(12.1)
y l cos θ
= (yB , xA) = yB xA T
= B = = l ∠θ .
xA l sin θ

doi:10.1088/978-0-7503-1350-6ch12 12-1 ª IOP Publishing Ltd 2016

 Mechatronics

q l
yB

xA A

Figure 12.1. Plane motion (adapted from Beer and Johnston 1984).

Differentiating equations (12.1) twice with respect to t, one writes

vB = yḂ = −θl̇ sin θ

aB = yB̈ = −(θ ∧̇ 2)l cos θ − θ l̈ sin θ

vA = xȦ = θl̇ cos θ

aA = xÄ = −(θ ∧̇ 2)l sin θ + θ l̈ cos θ .

Recalling that θ ̇ = ω and θ ̈ = α , one obtains
vB = −ωl sin θ vA = xȦ = ωl cos θ (12.2)

aB = −(ω ∧ 2)l cos θ − αl sin θ aA = −(ω ∧ 2)l sin θ + αl cos θ . (12.3)

One notes that a positive sign for vA or aA indicates that the velocity vA or the
acceleration aA has a sense of direction to the right; a positive sign for vB or aB has a
sense of direction upward. Equation (12.2) may be used, for example, to determine
vB and ω when vA and θ are known. Substituting for ω in equation (12.3), one may
then determine aB and α if aA is known.

12.2 Simple pendulum mechanical homogeneous continuous

dynamical system (approximate solution)
Most of the vibrations encountered in mechanical homogeneous continuous
dynamical systems may be represented by a simple harmonic motion. Many others,
although of a different type, may be approximated by a simple harmonic motion,
provided that their amplitude remains small.
Consider a simple pendulum mechanical homogeneous continuous system,
consisting of a bob of mass m attached to the end of a string or cord of length l,
which may oscillate on a vertical plane (figure 12.2(a)).
If a constant force is applied periodically, the pendulum can be made to swing
with constant amplitude. If the force is no longer applied, the pendulum will

12-2
 Mechatronics

Figure 12.2. Response of pendulum.

continue to swing, although its amplitude will decrease due to air resistance
(figure 12.2(b)).
At a given time, the string forms an angle θ with the vertical. The force holors
acting on the bob are its weight holor Gi and the force holor Pi exerted by the string.
Resolving the holor Ri = MijA j into tangential and normal components with
Ri = MijATj , the sense of direction is to the right, i.e. to the sense of direction
T

corresponding to increasing values of an angle θ , and observing that ATj = lA j = lθ ,̈

one writes
−G sin θ = mlθ .̈
Noting that G = mg , and dividing through by ml, one obtains
g
θ ̈ + sin θ = 0. (12.4)
l
For oscillations of small amplitude, one may replace sin θ by θ , expressed in radians,
and write
g
θ ̈ + θ = 0. (12.5)
l
Equation (12.5) shows that the equation obtained is that of a simple harmonic
motion, and that the circular frequency ω of the oscillations is equal to sqrt (g /l).
Taking into consideration the fact that the period T = 2π /ω, one obtains the
period of the small oscillations of a simple pendulum of length l:
2π ⎛l ⎞
T= = 2π sqrt⎜ ⎟ . (12.6)
ω ⎝g⎠

The main point to note is that there is motion (output) even when there is no applied
force (input). The form this motion takes is a property of the pendulum and is
termed the ‘natural response’ or ‘force-free response’.
The force is applied periodically to take advantage of this fundamental property of
the mechanical homogeneous continuous dynamical system, the difference between

12-3
 Mechatronics

Figure 12.3. Response of bicycle (Neimark and Fufaiev 1972).

the natural response and the actual response being the component of the response
directly attributable to the applied force. It is usually termed the ‘forced response’.

12.3 Bicycle mechanical homogeneous continuous dynamical system

An example in which the forced response is more easily seen is that of a bicycle being
pedalled along a horizontal road on a calm day. If the rider applies a constant force
to the pedals, the velocity will increase until a steady value is reached (figure 12.3).
This velocity is directly related to the applied force and is the forced response of the
mechanical homogeneous continuous dynamical system. After the rider stops pedal-
ling, the bicycle still moves, although at a decreasing speed. Again, there is a response
without any stimulus. This is the natural response of the mechanical homogeneous
continuous dynamical hypersystem, which can also be identified as being the differ-
ence between the forced response and the actual response during acceleration.

12.4 Damper–spring–mass mechanical homogeneous continuous

dynamical system I
Take into account the damper–spring–mass (b–k–m) mechanical homogeneous
continuous dynamical system (I) depicted in figure 12.4(a), in which a human being
submits a force holor FAj to one end of the damper. The initial action is to detect all
the components of the mechanical homogeneous continuous dynamical system,
which are constrained to move at the same values of velocity holors, and to signify
the sense of direction in which these velocity holors will be considered as positive.
This action is shown in figure 12.4(b). All the velocity holors are taken to be positive
in the same sense of direction.
This eases the creation of equilibrium holor equations and formulates results that
are considerably easier to deduce.
For the mechanical homogeneous continuous dynamical system under consid-
eration, three velocity holors are not effectively the same. These are the velocity
holor at the force holor end of the damper (Vi1), the velocity holor at the point
between damper and spring (Vi2 ) and the velocity holor of the mass (Vi3).
For this reason, the equivalent mechanical commutation matrixer diagram will
have three nodes or junctions other than the one indicating the datum velocity holor

12-4
 Mechatronics

Figure 12.4. Damper–spring–mass mechanical homogeneous continuous dynamical system (I).

(Vig ). These junctions, depicted in figure 12.5(a), are the configuration on which the
mechanical commutation matrixer components are incorporated. The junctions
symbolise the physical holors Vi1, Vi2 , Vi3 and Vig . For simplicity and intelligibility, the
ground terminal, Vg may be prolonged along a line at the same potential. The passive
components are positioned in the mechanical commutation matrixer diagram by
detecting the velocity difference (energy potential) that is necessitated for each
principal holor equation. For instance, the damping, b, physically models an effect
that is determined by the difference between Vi1 and Vi2 .
Thus, the damping component is connected between junctions 1 and 2. Damping is
mechanical resistance and is a dissipative component. One represents it in a physical
commutation matrixer diagram by an electrical resistance symbol. The spring
stiffness, k, also physically models an effect that depends on velocity difference. In
this case, the velocities at the ends of the spring are Vi2 and Vi3. The spring component
is therefore connected between these two nodes. The spring is analogous to the
electrical inductance and has the same mechanical commutation matrixer symbol.
Now the physical holor equation that describes the motion of the mass (equation
(3.48)) shows that it is only dependent on one velocity. In this case, the velocity holor
is Vi3. Since all mechanical commutation matrixer components require two terminals,
one adds a terminal that does not change the physical relations. This terminal is
the ground junction at zero velocity holor. When a fictitious terminal is used, the
mechanical commutation matrixer diagram contains a dashed line to show this. The
mechanical mass contains a dashed line to show this. The mechanical mass is
analogous to an electrical capacitance with one terminal always connected to ground.
Figure 12.5(b) shows the mechanical-matrixer diagram for the damper–mass–
spring (b–m–k) mechanical continuous dynamical system. The driving force holor
FAj exerted by the human being is physically modelled by an energy-transfer-source
holor connected between junctions Vi1 and Vig . Since the force is applied in the sense
of direction of the positive velocity holor, the flow must be shown as entering
junction Vi1. Note that the force holor representation is complete because it also
indicates the necessary reaction force holor on the datum in the opposite sense of
direction (away from junction Vig ).
The electrical analogue of this particular mechanical commutation-matrix is a
series R–L–C electrical commutation matrixer with a current-holor source input.
However, it must be emphasised that other damper–mass–spring (b–m–k) mechan-
ical homogeneous continuous dynamical systems in which the components appear in

12-5
 Mechatronics

Figure 12.5. Damper–mass–spring (b–m–k) mechanical homogenous continuous dynamical system (I) and its
mechanical commutation matrixer.

tandem do not necessarily result in a series mechanical commutation matrixer. This

point is illustrated in the following example.

12.5 Damper–spring–mass mechanical homogeneous continuous

dynamical system (II)
Figure 12.6(a) shows a damper–spring–mass (b–k–m) mechanical homogeneous
continuous dynamical system (II) in which the force is applied to the mass. Here
again, the mechanical homogeneous continuous dynamical system has the compo-
nents in tandem. However, in this case, the order is different. With this mechanical
commutation matrixer, there are only two velocity holors that can be identified as
independent.
These are the velocity holor of the mass (designated as Vi1) and the velocity holor
of the point between the spring k and damper b (designated as Vi2 ).
Once again, the reference terminal is the velocity holor of the ground (ViG ). The
mass m represents a capacitance between the Vi1 and ViG . The spring is equivalent to
an inductance between Vi1 and Vi2 , and the damper b is a resistance that goes from Vi2
to ViG . The resulting mechanical commutation matrixer is shown in figure 12.6(b). In
this case, the passive components do not form a series mechanical commutation

12-6
 Mechatronics

Figure 12.6. Damper–spring–mass (b–k–m) mechanical homogeneous continuous dynamical system (II) and
its mechanical commutation matrixer.

matrixer. Thus, it is important not to jump to conclusions about series or parallel

equivalents based solely on damper–mass–spring sketches.
The analogous mechanical commutation matrixer representation can only be
obtained by determining the terminals and connecting them with the appropriate
component. Before proceeding to more complicated examples with more than one
energy source, it will be useful to discuss the polarity of energy sources further.
Polarity concepts are especially pertinent when the energy sources are not all acting
in the same sense of direction.
Consider the isolated mass shown in figure 12.7, in which several possible physical
situations are indicated. In all cases, the sense of direction selected for the velocity holor
(Vi ) is conceptually positive. Then the application of the fundamental holor equation
⎡ ωm ⎤
Fj = ⎢ 0 V
⎣−ωm 0 ⎥⎦ i

requires that the force holor F j is positive from the terminal Vi through the
component to the ground terminal. Thus, the physical situations presented in
figures 12.7(a) and (b) are identical in the sense that the force holor F j applied to
the mass m acts in the same sense of direction as the assumed positive sense of
direction for the velocity holor Vi . In other words, the force holor F j tends to increase
the velocity holor Vi .
On the other hand, the force holor F j in figure 12.7(c) is directed to the left,
while the assumed positive velocity holor Vi ’s sense of direction is to the right. As a
result, the corresponding mechanical commutation matrixer representation shows a
force-source holor that is directed away from the terminal. When two force holors
FAj and Faj act on a mass m, the mechanical commutation matrixer has two energy
sources and therefore contains three branches, which meet at a junction. In the case

12-7
 Mechatronics

Figure 12.7. Polarity of energy sources.

of two force holors FAj and Faj in the same sense of direction (figure 12.7(d)), the
previous discussion indicates that each energy source should have its positive flow
towards the junction.
Of course, this is because both force holors FAj and Faj act to increase the velocity
holor Vi . The continuity of flow requires that the total force holor passing through
the mass is the sum of the flows entering the junction. That is,
F j = FAj + Faj .
In conventional mechanics terms, the sum of the unbalanced force holors FAj and Faj
determines the acceleration of the mass m. The mechanical commutation matrixer
approach, therefore, is merely another way of recognising and developing principal
physical relations.
Suppose now that the force holors FAj and Faj are acting in opposite senses of
direction, as in figure 12.7(e).
The force-source holor (Faj ) in the sense of direction of the positive velocity holor (Vi )
is physically modelled by an energy-transfer holor (Fa ) directed towards the terminal
(Faj ). This is the force-source holor (F aj ) that tends to increase the velocity holor Vi .
The opposite force holor FAj is physically modelled by an energy-transfer-source
holor directed away from the terminal and has an energy-transfer holor at the
terminal that yields
F j = Faj − FAj.
Both physical situations (figures 12.7(d) and (e)) are in exact agreement with physical
reality. The junction of the mechanical commutation matrixer representation
corresponds to the free-body diagram of mechanics. However, for complicated
mechanical commutation matrixers, several free-body diagrams may be required.
An advantage of the mechanical commutation matrixer representation is that it is
completely described by a single diagram.
For physical situations that are physically modelled by velocity-source holors, the
procedure is simply to observe whether the energy source is in a positive or negative sense
of direction. If it is positive, the positive terminal of the velocity-source holor is connected
to the junction. When a velocity-source holor is directed opposite to the assumed positive

12-8
 Mechatronics

Figure 12.8. Automotive vehicle’s suspension damper–spring–mass (b–k–m) mechanical homogeneous

continuous dynamical system.

velocity-holor sense of direction, the negative terminal of the energy source is connected
to the junction. These points are illustrated in the following examples.

12.6 Automotive vehicle’s suspension mechanical commutation

matrixer
Figure 12.8(a) shows an automotive vehicle approaching a bump in the on/off road
surface. The suspension mechatronical control system of the automotive vehicle
supports the vehicle body by means of springs (k1) and shock absorbers (b1).
The wheel-tyres exhibit a spring effect (k2 ) that is close to the ideal spring, but the
hysteresis normally associated with rubber pneumatic springs is physically modelled
by an equivalent damper (b2 ).
The constraint at the point of contact between the wheel-tyre and the on/off road
surface is that the position of this point is defined.
If the automotive vehicle is driven at constant forward velocity along the on/off
road surface, the vertical velocity holor constraint is, therefore, appropriately
physically modelled by a velocity-source holor (ViS).
In addition, there are force-source holors to physically model the mass of the
automotive vehicle and the mass of the wheels. These considerations lead to the
damper–spring–mass (b–k–m) configuration shown in figure 12.8(b).
To obtain the mechanical commutation matrixer diagram one must first determine
the number of independent velocity holors associated with the mechanical commuta-
tion matrixer. Each mass may move independently and this leads to terminals
representing Vi1 and Vi2 . The point of contact with the ground is Vi3. The velocity holor
of the ground is ViG .
Thus, there are three terminals plus a ground on the mechanical commutation
matrixer diagram (figure 12.9). The connection of components between the
terminals is obtained by referring to the image in figure 12.8(b).

12-9
 Mechatronics

Figure 12.9. Mechanical homogeneous continuous dynamical system.

The velocity-source holor ViS acts in the sense of direction of the positive velocity
holor. As a result, the positive terminal of this energy source is connected to Vi3. This
means that ViS = Vi3 − ViG .
Had the reference-velocity holor sense of direction been selected as down, the
negative terminal of VS would have been attached to Vi3.
In that case, ViS = ViG − Vi3. The force-source holors due to the weight of the
automotive vehicle (m1g) and the weight of the wheels (m2g) are directed away from
their respective terminals (Vi1 and Vi2 ). Obviously, if the reference velocity holor was
down, these sources would be pointed toward the terminals.

12.7 Determination of the analogous impedance holor of mechanical

commutation matrixers
Since the physical commutation matrixer diagram is the basis on which the generalised
dynamical systems approach rests, it is appropriate to review the process by which the
physical commutation matrixer was obtained in previous chapters.
Let us consider the damper–spring–mass (b–k–m) mechanical homogeneous con-
tinuous dynamical system in figure 12.10(a).
The first step is to identify all the parts, which are constrained to move at the same
velocity and show the sense of direction chosen to be positive. Number these points,
since they will form the junctions or nodes of the mechanical-commutation matrixer
diagram. In this case, there are three (3) independent velocity holors, Vi1, Vi2, Vi3,
and a reference velocity holor, Vig .
The holor equations are usually easier to write and the results easier to interpret if
the positive holor velocity is taken in the same sense of direction at each part of the
mechanical homogeneous continuous dynamical system. Now draw the ‘matrix
skeleton’ of the mechanical commutation matrixer, i.e. the three (3) nodes corre-
sponding to the three (3) parts of the mechanical continuous dynamical system at
which the velocity holors are different, and the datum or reference node. In figure
12.10(b), these are indicated as Vi1, Vi2, Vi3 and Vig .
The impedance-holor blocks corresponding to the passive mechanical commu-
tation matrixer components are added to the velocity-difference holor, which is
required in the force-velocity relation of the mechanical-commutation-matrixer

12-10
 Mechatronics

Figure 12.10. Determination of the analogous mechanical circuit, showing impedances.

component. The damper, b1, physically models an effect that is dependent on the
relative velocity holor between nodes 1 and 2. Therefore, the impedance-holor block
(1/b1) is inserted between these nodes. The damper, b2, and the spring, k, both
depend on the relative velocity holor between nodes 2 and 3. Thus, the damper
block (1/b2) is connected in parallel with the spring impedance-holor block (ω/k)
between nodes 2 and 3. The mass, m, on the other hand, physically models an effect
that is dependent on the absolute velocity holor of node 2. To make this a two-
terminal mechanical commutation matrixer component, one uses the ground as a
fictitious terminal and connects the mass impedance holor block (1/ωm) between Vi2
and Vig . Now the branches with the energy sources can be added. The velocity-
source holor at node 1 constrains node 1 to move to the right at a velocity holor Vi
relative to the datum. Therefore, on the mechanical commutation matrixer diagram
this is represented by a velocity-source holor connected between node 1 and the
datum. It is positive at node 1 because the velocity-source holor constrains node 1 to
move in the positive sense of direction.
The force-source holor, F j , acting at node 3 would have moved node 3 in
the negative sense of direction if it had been acting separately. Thus, the ‘transfer
of force’ into node 3 must also be negative and the arrow is fixed away from node 3.

12.8 Damper–spring–mass mechanical homogeneous continuous

dynamical system (III)
The damper–spring–mass (b–k–m) mechanical homogeneous continuous dynamical
system (III) shown in figure 12.11(a) consists of three dampers, one spring and two

12-11
 Mechatronics

masses. There are two known force-holor inputs (Faj and Fcj ). One must present the
node-to-datum holor equations in hypermatrix form for this mechanical homoge-
neous continuous dynamical system. Figure 12.11(b) shows the mechanical-
commutation-matrixer representation of the b–m–k mechanical homogeneous
continuous dynamical system. Since there are three unknown velocity holors, the
required hypermatrix must be of the form:

Figure 12.11. Damper–spring–mass (b–k–m) mechanical homogeneous continuous dynamical system (III) and
its mechanical commutation matrixer.

12-12
 Mechatronics

⎡Y ji Y ji Y ji ⎤⎡V A⎤ ⎡ j ⎤
node1 ⟹ ⎢ aA aB aC ⎥⎢ i ⎥ ⎢ Fa ⎥
node2 ⟹ ⎢YbA
ji ji
YbB ji
YbC ⎥⎢ViB ⎥ = ⎢ 0 ⎥ .
⎢ ji ji ⎥⎢ C ⎥
⎣YcA YcB YcC ⎦⎣Vi ⎦ ⎢⎣−Fc ⎥⎦
j
node3 ⟹ ji

The column vector of unknowns has sinusoidal effective values at the velocity-holor
nodes; these are designated as ViA and ViC . The column vector of knowns consists of
sinusoidal effective values of the force-holor inputs. The force-holor effective value-
entering node A due to the force-holor inputs is merely Faj . There is no force-holor
input entering node B exactly and thus there is a zero in the known column vector at
node B. For node C one may observe that the force-holor effective value, Fa, acts
away from the node. As a result, the force-holor effective value is −Fc.
ji
One may now determine the admittance-holor matrix. Matrix component YaA
represents the admittance holor from terminal A (ViA) with all the other terminals
grounded (ViB = ViC = 0). This is the parallel combination of b1, ωm1 and b3, so that:
ji b1 + d3 −m1ω
YaA = .
ωm1 b1 + b3
ji
In a similar way, YbB represents the equivalent admittance holor from terminal
B
B (Vi ) with all the other terminals grounded (ViA = V C = 0):

ji b3 k / ω
YbB = ,
− k / ω b3

and YcCji applies to terminal C (ViC ) with the other terminals grounded

(V i
A
= ViB = 0 :)
b2 −ωm2 + k / ω
YcCji = .
ωm2 − k / ω b2

The other admittance holors are merely the negative of all the common terms. For
instance, to find YaB
ji ji
and YbA , there is only one common term, b3, in

ji ji − b3 0
YaB = YbA = .
0 − b3

To find YbC
ji ji
and YcBji , the only common term in YbB and YcCji is k/ω, and thus
ji
YbC = YcBji = 0 k /ω .
−k / ω 0
Since there are no common terms between Y aA and Y cC , the admittance holors
ji
YaC = YcAji = 0 0 .
0 0

12-13
 Mechatronics

The complete hypermatrix is therefore:

⎡ ⎤
⎡ b + b − ωm ⎤ ⎢ VcA ⎥ ⎡ Fc ⎤
− b3 0 0 0 ⎢ a ⎥
⎢ 1 3 1
⎥ ⎢ VsA ⎥ s
⎢ ωm 1 b1 + b 3 0 − b3 0 0 ⎥ ⎢ ⎥ ⎢ Fa ⎥
⎢ ⎥ ⎢ ⎢ ⎥
b 0 b3 k / ω VcB ⎥
⎢ − 3 0 k /ω ⎥ ⎢ ⎥= ⎢ 0 ⎥.
⎢ 0 − b3 − k / ω b3 −k /ω 0 ⎥ ⎢ VsB ⎥ ⎢ 0 ⎥
⎢ ⎥ ⎢ ⎥ ⎢ c ⎥
⎢ 0 0 0 k /ω b2 − ωm2 + k / ω ⎥ ⎢ VcC ⎥ ⎢ − Fc ⎥
⎢⎣ 0 0 −k /ω 0 ωm2 − k / ω b2 ⎥⎦ ⎢ ⎥ ⎢ −F s ⎥
VsC ⎥⎦ ⎣ c ⎦
⎢⎣

To express the effective values at any terminal in terms of the basic mechanical
commutation matrixer components and the known input effective values, one must
use determinants, for example:
Fac −b 3 0 0 0
Fas 0 −b 3 0 0

0 0 b3 k/ω 0 k/ω
0 0 −k / ω b 3 −k / ω 0
−Fcc 0 k/ω b2 −ωm 2 + k / ω
VcA −Fcs −k / ω 0 ωm 2 − k / ω b2
ViA = = ,
V sA b1 + b3 −ωm1 −b 3 0 0 0
ωm1 b1 + b3 0 −b 3 0 0
−b 3 0 b3 k/ω 0 k/ω
0 −b 3 −k / ω b 3 −k / ω 0
0 0 0 k/ω b2 −ωm 2 + k / ω
0 0 −k / ω 0 ωm 2 − k / ω b2

and alternatively, if one evaluates the determinants

⎡ b ⎤ a ⎡ −b ⎤ c
⎢ 3 k /ω 0 k /ω 0 k /ω ⎥ Fc 3 0 0 k /ω ⎥ − Fc
D− a +⎢
⎣ − k /ω b 3 − k /ω 0 − k /ω 0 ⎦ Fs ⎣ 0 −b 3 − k /ω 0 ⎦ − Fsc
ViA = ,
⎡ b ⎤ ⎡ ⎤
k /ω 0 k /ω 0 k /ω ⎥ b1 + b 3 − ωm1 ( − b 3) 2 0
−⎢ D⎥
3
⎢ D−
⎣ − k /ω b 3 − k /ω 0 − k /ω 0 ⎦ ωm1 b1 + b 3 ⎢⎣ 0 (− b 3)2 ⎥⎦

where
k
b2 −ωm2 +
D= ω .
k
ωm2 − b2
ω
One may find expressions for ViB and ViC in a similar way.

12.9 Damper–mass mechanical commutation matrixer

Figure 12.12(a) shows a damper–mass (b–m) mechanical homogeneous continuous
dynamical system with two dampers and a mass. Either of the two impedance-holor
mechanical commutation matrixers shown in figures 12.12(b) and (c) may represent

12-14
 Mechatronics

this mechanical homogeneous continuous dynamical system, showing that the force-
holor that acts to accelerate the mass is the same in both mechanical commutation
matrixers.
In the mechanical commutation matrixer shown in figure 12.12(b), the mesh holor
equation hypermatrix has the form:
⎡ Z Aa Z An ⎤⎡ j ⎤ ⎡ A⎤
⎢ ij ij
⎥⎢ FA ⎥ = ⎢Vi ⎥ ,
⎢⎣ Zij Zij ⎥⎦⎢⎣ Faj ⎥⎦ ⎣ 0 ⎦
Na Nn

where the energy source has no contribution to mesh (Faj ) and contributes a velocity-
holor rise (ViA) to mesh (FAj).

Figure 12.12. Mechanical commutation matrixer.

12-15
 Mechatronics

The sum of the impedance holor around the meshes yields:

1/b1 1/ωm 1/b2 1/ωm
ZijAa = , ZijNn = ,
−1/ωm 1/b1 −1/ωm 1/b2
and the common term is 1/ωm, so that

1
0 −
ωm
ZijAn = ZijNa = .
1
0
ωm
Therefore the matrix becomes:

⎡ ⎤ ⎡ F Ac ⎤ ⎡ VcA ⎤
⎢ 1/b1 1/ωm 0 1/ωm ⎥ ⎢ ⎥ ⎢ ⎥
⎢ −1/ωm 1/b1 −1/ωm 0 ⎥ ⎢ F As ⎥ ⎢ V A ⎥
⎢ ⎥ ⎢ ⎥=⎢ s ⎥.
⎢ 0 1/ωm 1/b2 1/ωm ⎥ ⎢ Fac ⎥ ⎢ 0 ⎥
⎢ −1/ωm 0 −1/ωm 1/b2 ⎥⎦ ⎢⎣ Fas ⎥⎦ ⎢⎣ 0 ⎥⎦
⎣

In this mechanical commutation matrixer (figure 12.12(b)) the force holor applied to
the mass is FMj = FAj − Faj .
One must therefore solve the matrix for FAj and Faj and then subtract the
results.

1
VcA
0
ωm
V sA −
1
0
ωm
1 1
0 b2 ωm
0 1 1
−
j F Ac ωm b2
FA = =
F As 1 1 1
0
b1 ωm ωm
1 1 1
− − 0
ωm b1 ωm
1 1 1
0
ωm b2 ωm
1 1 1
− 0 −
ωm ωm b2

1 1
b2 ωm VcA

−
1 1 V sA
ωm b 2
= ,
1 1 1 1 1 1
0 0
b1 ωm b2 ωm ωm ωm
−
1 1 1 1 1 1
− − − 0 − 0
ωm b1 ωm b 2 ωm ωm

12-16
 Mechatronics

1 1
b1 ωm VcA
− ω1m VsA
1
b1

1
0
ωm 0
0
Fac − ω1m 0
Faj = =
Fas 1
b1
1
ωm 0 1
ωm

− ω1m 1
− ω1m 0
b1

1 1 1
0 b2 ωm
ωm

− ω1m 0 − ω1m 1
b2

1
0
ωm VcA
− ω1m 0 VsA
= 1 1 1 1 1 1
,
b1 ωm b2 ωm 0 0
ωm ωm
−
− ω1m 1
− ω1m 1
− ω1m 0 − ω1m 0
b1 b2

and thus
b2 0 VcA
0 b2 VsA
FMj = FAj − Faj = 1 1
+
1 1 1 1
0 0
b1
ωm b2 ωm ωm ωm
−
1 1 1 1 1 1
− − − 0 − 0
ωm b1 ωm b 2 ωm ωm

b2 0 VcA
0 b2 V sA
− .
1 1 1 ⎡ 1 1 ⎤
0 0 0 ⎢ 0 0 ⎥
b1 b2 ωm ⎢ b1 b2 ⎥
+ +
1 1 1 ⎢ 1 1 ⎥
0 0 − 0 ⎢ 0 0 ⎥
b1 b2 ωm ⎣ b1 b2 ⎦
Let us solve the same problem using the mechanical commutation matrixer shown in
figure 12.12(c).
The mesh holor equations by inspection are:

⎡ 1 1 1 ⎤
⎢ b1 0 0 −b 0 ⎥ ⎡ F Ac ⎤ ⎡ VcA ⎤
b2 2
⎢ + ⎥ ⎢ ⎥ ⎢
1
⎢ 0 b1 0 1
0 −b
1
⎥ ⎥
b2 ⎢ F As ⎥ ⎢ V A ⎥
⎢ 2
⎥ ⎢ ⎥ = ⎢ s ⎥.
⎢ −b
1
0
1 1 ⎥ ⎢ Fnc ⎥ ⎢ 0 ⎥
⎢ b2 ωm ⎥ ⎢⎣ Fns ⎥⎦ ⎢⎣ 0 ⎥⎦
2
⎢ 1 1 1 ⎥
⎢⎣ 0 −b − ωm ⎥⎦
2 b2

12-17
 Mechatronics

Note that this matrix is different from the matrix determined for the mechanical
commutation matrixer shown in figure 12.12(b). In this mechanical commutation
Fnc
matrixer, the force holor applied to the mass is Fnj = alone. Thus if one solves
Fns
for Fnj one obtains:
1 1
0 0 VcA
b1 b2
+
VsA
1 1
0 0
b1 b2

− b1 0
2 0
− b1 0
j Fnc 0
2
FMj = Fn = =
Fns 1
b1
0 1
b2
0 − b1 0
2
1
+ 1
0
b1
0
b2
0 − b1
2
1 1
− b1 0 b2 ωm
2

0 − b1 − ω1m 1
2 b2

1
0 VcA
b2

VsA
1
0
b2
= ⎡ 1 0
,
⎢ b1
1
0 ⎤⎥ 1 1 1
0 1
0
b2 b2 ωm b2 b2
⎢ + 1 ⎥
−
1
⎢⎣ 0 b1 0
b2 ⎥
− ω1m 1
0 1
0 1
⎦ b2 b2 b2

or
1
0 VcA
b2

0 1
V sA
b2
FM = .
1 1 ⎡ 1 1 ⎤ 1
0 0 ⎢ b 0 0 ⎥ 0
b1 b2 b2 ωm
1 1
+⎢ 1 1 + ⎥
0 0 ⎢⎣ 0 b1 0 1 ⎥ − 1 0
b1 b2 b2 ⎦ ωm

Moreover, the results are the same in both mechanical commutation matrixers.

References
Beer F P and Johnston E R Jr 1984 Vector Mechanics for Engineers—Statics and Dynamics
(New York: McGraw-Hill)
Neimark J I and Fufaiev N A 1972 Dynamics of non-holonomic systems. (Providence, RI:
American Mathematical Society) (Translated from Russian ed, 1967)

12-18
 IOP Publishing

Mechatronics
Dynamical systems approach and theory of holors
B T Fijalkowski

Chapter 13
Fluidic homogeneous continuous dynamical
systems as fluidic commutation matrixers

Windmills, which are used in the great plains of Holland and North Germany to
supply the want of falling water, afford another instance of the action of velocity.
The sails are driven by air in motion—by wind.
Hermann von Helmholtz

13.1 Fluidic-transmission line

Let us consider one of the basic fluidic commutation matrixer components—the
fluidic-transmission line. Figure 13.1(a) shows a mechano-fluidic (MF) pump.
How are we to describe this physical commutation matrixer component math-
ematically? To answer this question we must first decide what we wish to know
about the passage of the fluid to a transmission line. One thing we want to know is
the rate at which liquid is passing through the fluidic-transmission line. We call this
the flow, and for a fluidic-commutation matrixer, it is designated as q. Another thing
we might need to know is the size of the MF pump required to produce the flow.
Instead of physical size, the pressure often indicates the capability of an MF pump,
pA, it can develop.
If we were able to relate the pressure, pA,, to the flow, q, we would have a type of
mathematical model of the fluidic transmission line. But can we do this? We cannot,
because the flow through the fluidic-transmission line also depends on the pressure,
pa, at the end of the fluidic-transmission line.
If pa were equal to pA, there would be no flow through the fluidic-transmission
line. Thus the mathematical model must relate the flow, q, to the pressure difference,
pA – pa.

doi:10.1088/978-0-7503-1350-6ch13 13-1 ª IOP Publishing Ltd 2016

 Mechatronics

Figure 13.1. Fluidic-transmission line fluidic-commutation matrixer component.

These quantities, flow and pressure difference, are the signal holors for fluidic
commutation matrixers. Flow is the energy-transfer holor (V j) and pressure differ-
ence is the energy-potential-difference holor (Fi).
To determine the mathematical model of the fluidic-transmission line, one could
perform some experiments on the fluidic-transmission line in which the pressure
difference was changed and the resulting flow measured.
Figure 13.1(b) shows some typical results for data points. As the pressure
difference increases, the flow increases. The mathematical model, of course, must
hold for an infinite number of pressure differences—flow combinations—not just the
three data points.

13.2 Node-to-datum holor equations for the fluidic commutation

matrixer
The node-to-datum holor equations are usually referred to as node holor equations.
The method is a very convenient way of formulating holor equations when the
desired output is of the energy-potential-difference holor type. When we apply the
node-to-datum method, the coefficients turn out simplest if admittance holors are
used. Accordingly, consider the fluidic homogeneous continuous dynamical system
shown in figure 13.2.
Suppose it is the analogous fluidic commutation matrixer for a particular fluidic
homogeneous continuous dynamical system that has flow-source holors, Q A and Q a .
One seeks to find the equilibrium holor equations that describe the pressure holors,
PA and Pa .
One begins by applying equation (10.1) (KCL) to nodes A and a.
node A → Q1j + Q 2j = Q Aj, (13.1a )

13-2
 Mechatronics

Figure 13.2. Analogous fluidic homogeneous continuous dynamical system involving admittance holors.

node a → Q 3j − Q1j = Q aj , (13.1b)

where the flow holors away from the mode are positive on the left-hand side of the
holor equation and negative on the right-hand side.
Once all the node holor equations are written (in this case the two equations
(13.1a) and (13.1b), one formulates the fluidic-component relations. The three
fluidic-component relations are:

(Pi
A
)
− Pia Y1ji = Q1j (13.2a )

(Pi
A
)
− PiG Y2ji = Q 2j (13.2b)

(Pi
a
)
− PiG Y3ji = Q 3j . (13.2c )

Now, if equations (13.2) are substituted into equations (13.1) and one notes that the
reference pressure holor, PiG , is zero, the result is as follows.
For node A:

(Y 1
ji
)
+ Y2ji PiA − Y1jiPia = Q Aj. (13.3a )

For node a:

(
Y1jiPiA + Y1ji + Y3ji Pia = Q aj . ) (13.3b)

Let us consider the individual terms in these two holor equations and try to interpret
them.
Node A: (Y1ji + Y2ji )PiA—total flow holor away from node A due to PiA only,
with all other pressure holors set to zero;
Y1jiPia —total flow holor away from node a due to Pia only, with all
other pressure holors set to zero;
Q Aj —total flow holor into node A from all flow-holor sources con-
nected to node A.

13-3
 Mechatronics

Node a: (Y1ji + Y3ji )Pia —total flow holor away from node a due to Pia , with all
other pressure holors set to zero;
Y1jiPiA—total flow holor away from node a due to PiA, with all other
pressure holors set to zero.

At this stage, it is valuable to note the coefficients in the holor equations, and the
matrix form of the holor equations is useful for this purpose.

⎡ ji −Y1ji ⎤⎡ PiA⎤ ⎡Q A ⎤
ji j
node A ⟹⎢Y1 + Y2 ⎥⎢ a ⎥ = ⎢ j ⎥ . (13.4)
node a ⟹ ⎢⎣ −Y1ji Y1ji + Y3ji ⎥⎦⎣ Pi ⎦ ⎢⎣Q a ⎥⎦

In this array of holor admittances, the terms on the main diagonal are positive and
all the others are negative. This is true in general for node-to-datum equilibrium
holor equations.
The general form is:
⎡Y ji Y ji ⎤⎡ P A⎤ ⎡Q j ⎤
⎢ Aa An
⎥⎢ i a ⎥ = ⎢ Aj ⎥ . (13.5)
⎢⎣YNa YNn ⎥⎦⎣ Pi ⎦ ⎣⎢Q a ⎥⎦
ji ji

ji
Note that YAa is the sum of the admittance holor of the positive branches connected
ji ji
to node A and YNn is the sum of admittance holors at node a. YAn is the negative of
the admittance holor of the branch connecting nodes A and a. If every mechanical
commutation matrixer component connecting two nodes is ‘bilateral’ (i.e. the
pressure/flow relationship is the same for flow in both senses of direction),
ji ji
YAn = YNa or Y jk = Y kj , and the matrixer is symmetrical. With some practice, it is
a simple matter to write mnemonically the equilibrium holor equations of many
physical heterogeneous continuous dynamical hypersystems directly in this form (i.e.
‘by inspection’ of the physical commutation matrixer), thus reducing the likelihood
of error. One will clarify the procedure in the examples given in section 13.3.

13.3 Loop and mesh holor equations for the fluidic commutation
matrixer
The loop and mesh holor equations method is a convenient means of formulating
holor equations when the unknowns are of the energy-transfer-holor type. This
method favours the use of impedance-holor parameters.
Let us consider, therefore, the fluidic commutation matrixer shown in figure 13.3.
The fluidic commutation matrixer represents a particular fluidic homogeneous
continuous dynamical system that has input pressure-source holors of sinusoidal
type with effective values of pressure-source holors PiA and Pia .
One wants to find holor equations that describe the sinusoidal flow holors
through all the impedance-holor components. In this case, there are three distinct
impedance holors and three distinct flow holors. However, only two of these are

13-4
 Mechatronics

Figure 13.3. Analogous physical commutation matrixer (fluidic commutation matrixer) involving impedance
holors.

independent. For instance, if Q Aj is the flow holor through Zij2 , and Q aj is the flow
holor through Zij3, then the flow holor through Zij1 is Q Aj − Q aj .
The flow holors Q Aj and Q aj are the mesh flow holors and they are the most
convenient choice for the unknowns, since one may apply equation (10.1b) around
the closed mesh.
If one applies equation (10.1b) around meshes A and a:

( ) ( ) (
mesh A ⟹ PiA − Pi y + Pi y − Pig = PiA − Pig , ) (13.6a )

( ) ( ) (
mesh a ⟹ Pig − Pi y + Pi y − Pia = Pig − Pia . ) (13.6b)

The meshes are the open spaces in the impedance-holor fluidic commutation matrixer.
One always takes the flow holor in each mesh in a clockwise sense of direction.
One may now substitute the fluidic commutation matrixer component relations into
the mesh equations (13.6a) and (13.6b) to obtain:

mesh A ⟹ Zij2Q Aj + Zij1(Q Aj − Q aj ) = PiA, (13.7a )

mesh a ⟹ Zij1(Q aj − Q Aj ) + Zij3Q aj = Pia . (13.7b)

Alternatively, if one groups the flow holors together in equation (13.7) the result is:

mesh A ⟹ (Zij1 + Zij2 )Q Aj − Zij1Q aj = PiA, (13.8a )

mesh a ⟹ Zij1Q Aj + (Zij1 + Zij3)Q aj = Pja . (13.8b)

13-5
 Mechatronics

One may interpret the terms in equation (13.8) as follows.

In mesh A: (Zij1 + Zij2 )Q Aj —the sum of the pressure-holor drops across each
impedance holor in a sense of the flow-holor direction and due to
Q Aj only; all the other flow holors are set to zero;
Zij1Q aj —the sum of the pressure-holor drops across impedance
holors, in a sense of direction Q Aj , but due to flow holor Q aj ; all
the other flow holors are set to zero;
In mesh a: −Zij1Q Aj —the sum of the pressure-holor drops in mesh a in a sense of
flow direction due to Q Aj ; all the other holor flows are set to zero;
(Zij1 + Zij3)Q 2j —the sum of the pressure-holor drops due to holor flow
Q aj ; all the other flow holors are set to zero.

In matrix form, equation (13.8) may be written as:

⎡ 1 2
−Zij1 ⎤⎡Q Aj ⎤ ⎡ PiA ⎤
mesh A ⟹⎢ Zij + Zij ⎥⎢ ⎥ = ⎢ ⎥ . (13.9)
mesh a ⟹ ⎢⎣ −Zij1 Zij1 + Zij3⎥⎦⎢⎣Q aj ⎥⎦ ⎣ Pi ⎦
a

In the impedance-holor array, the terms on the main diagonal are positive and all
others are negative. This is true for mesh holor equations when the assumed senses of
flow direction are all clockwise or all counter clockwise.
The general form for the mesh holor equations is:
⎡ Z Aa Z An ⎤⎡Q j ⎤ ⎡ A⎤
⎢ ij ij
⎥⎢ A ⎥ = ⎢ Pi ⎥ . (13.10)
⎢⎣ Zij Zij ⎥⎦⎢⎣Q aj ⎥⎦ ⎣ Pia ⎦
Na Nn

The fluidic-commutation-matrixer component ZijAa is the sum of all the impedance

holors in mesh A and ZijNn is the sum of all the impedance holors in mesh a. The
impedance holors ZijAn and ZijNa are equal and are the negative of all the common
terms in ZijAa and ZijNn , provided that all the circulating flow holors are taken in the
same sense of direction (e.g. all clockwise).

13-6
 Part III
Physical Matrixers as Physical Commutators
 IOP Publishing

Mechatronics
Dynamical systems approach and theory of holors
B T Fijalkowski

Chapter 14
Mechanical commutation matrixer commutators
for conventional DC and AC magneto-
mechano-dynamical electrical machines

Everything should be made as simple as possible, but not simpler.’

Albert Einstein

14.1 Introduction
The ‘Gramme machine’ or ‘Gramme dynamo’ is an AC–DC commutator synchro-
nous generator designed by Belgian inventor Zénobe Gramme in 1871. It was
the first magneto-mechano-dynamical (MMD) electrical machine to be used for
industrially generated DC electrical energy. His innovation was to use many armature
windings, wound on a doughnut shaped armature and commutated (switched) with a
many-segmented mechanical ring commutator, to rectify (smooth) the output AC
waveforms, generating nearly constant DC electrical energy. Figure 14.1 shows a
hand-cranked physical model from the 1870s, designed for use in laboratories. It used
the laminated permanent magnets invented by Jamin (Fontaine 1878).
The phenomenon of unipolar electrical-current conductivity had already been
applied in 1883, when the British physicist, W Ritchie, constructed an MMD electrical
machine, i.e. an AC–DC/DC–AC commutator dynamotor (figure 14.2), with a rotary
mechanical commutation matrixer (MCM) commutator, i.e. an MCM AC–DC/DC–
AC commutator, which can be taken for the prototype of a contemporary mechanical
ring- or disc-type polysectorial MCM AC–DC/DC–AC commutator.
MCM AC–AC and AC–DC/DC–AC commutators are characteristic components
that are common to a large group of DC electrical machines. MCM AC–AC and
AC–DC/DC–AC commutators (figure 14.3) are conventionally made by assembling
hard-drawn copper sectors (bars or segments) interleaved with 0.7–2.0 mm sheet mica
or micanite, with these separators being ‘undercut’ by about 1 mm.

doi:10.1088/978-0-7503-1350-6ch14 14-1 ª IOP Publishing Ltd 2016

 Mechatronics

Figure 14.1. The Gramme dynamo (Fontaine 1878).

Figure 14.2. Ritchie’s MCM AC–DC/DC–AC commutator dynamotor (Veselovsky and Kopylov 1984).

14-2
 Mechatronics

The stationary or movable brushes of a suitable carbon/graphite content are

mounted in boxes with spring loading to hold them against the MCM commutator
surface, with a medium to strong pressure depending on the application. The
circumferential brush width is typically 2–3 sector widths (10–20 mm) and about
30 mm axially. One brush-arm per pole is employed, except for in certain four-pole
wave-wound MCM AC–DC/DC–AC commutator dynamotors (generator/motors),
which have two brush-arms in adjacent positions to facilitate maintenance.
The MCM AC–DC/DC–AC commutator is a necessary part of a DC commutator
electrical machine, while the use of an MCM AC–AC commutator made possible AC
commutator motors that have certain valuable features. Finally, MCM AC–DC/DC–
AC commutators are used in single-armature converters, the function of which is to
convert AC–DC or DC–AC. Thus, the first MCM with ‘mechanical-contact’
electrical valves became an MCM connection of sliding contacts, setting up, for
example, collector phosphorus bronze springs or carbon/graphite brushes, sliding,
respectively, on the surface of the collector split-rings or hard-drawn copper sectors
(bars or segments). Rotary sliding-contacts, mechano-electrical (ME) AC–DC com-
mutators, termed ‘MIMO MCM DC–AC commutators’ by the author (figures 14.4–
14.8)—which are widely used in MCM AC–DC commutator generators, according to
the dynamical systems approach on physical heterogeneous continuous simple
dynamical hypersystems, that is, MMD electrical machines—are a series of copper
sectors (bars or segments), insulated from one another and forming matrixer-input
primary conductive collectors of an MCM AC–DC commutator.
The copper sectors, i.e. matrixer-input primary conductive collectors, are connected
to the commutating sections, i.e. phase-winding terminals, of an armature wind-
ing mounted on the rotor of an electrical machine. Stationary carbon brushes are
placed on the MCM DC–AC commutator cylindrical surface with copper sectors.
Carbon brushes forming the matrixer-output secondary conductive collectors of
an MCM commutator dynamical system are connected by means of sliding-
contacts, i.e. electrical valves (copper-sector/carbon-brush contacts), to the copper
sectors, i.e. the matrixer-input primary conductive collectors of an MCM DC–AC
commutator.

Figure 14.3. Section through an MCM AC–AC and/or AC–DC/DC–AC commutator (Kuznetsov 1960,
Kasatkin and Perekalin 1970, Fijalkowski 1973).

14-3
 Mechatronics

Figure 14.4. The single-phase, half-wave MCM AC–DC commutator, with a single homogeneous cathode
commutation group and a neutral, acting as a one-way MCM AC–DC rectifier.

Figure 14.5. The two-phase, half-wave MCM AC–DC commutator, with a single homogeneous cathode
commutation group and a neutral, acting as a one-way MCM AC–DC rectifier.

By means of these sliding contacts acting as electrical valves, it is possible to

connect the phase windings of the rotary armature winding in the electrical machine
to an external electrical commutation matrixer dynamical system.
The operating principle of the MIMO ECM AC–DC/DC–AC commutator will
be considered in more detail later on. The MIMO MCM AC–DC/DC–AC

14-4
 Mechatronics

Figure 14.6. The two-phase full-wave MCM AC–DC commutator, with double heterogeneous cathode and
anode commutation groups and a neutral, acting as a double-way MCM AC–DC rectifier.

commutator was in the past, and to date still is, a necessary component of
AC–DC/DC–AC commutator dynamotors.
In addition, the use of an MIMO MCM AC–AC and/or AC–DC/DC–AC
commutator made possible yesterday and today’s polyphase MCM AC–AC and/
or AC–DC/DC–AC commutator electrical machines.
Finally, MIMO MCM AC–DC commutators have been and are used in
single-armature MCM AC–DC commutator rotary converters, the function of which
is to convert three-phase alternating voltages to direct voltage, or vice versa.
On the other hand, MCM AC–AC and AC–DC/DC–AC commutator dynamo-
tors are quite expensive to manufacture compared with AC and DC commutatorless
machines, while they demand special care and are not so rugged.
MCM AC–AC and AC–DC/DC–AC commutator dynamotors are not as
maintenance-free as AC and DC commutatorless dynamotors, due to brush and
MCM commutator wear.
The roundness of the MIMO MCM AC–AC and AC–DC/DC–AC commutator
is the most important design parameter to reduce the brush noise of the MCM
AC–DC/DC–AC commutator dynamotor. The roundness of the MIMO MCM
AC–AC or AC–DC/DC–AC commutator needs to be maintained in order to
maintain the brush noise under a certain level throughout the life span of the
MCM AC–AC and AC–DC/DC–AC commutator dynamotor.

14-5
 Mechatronics

Figure 14.7. The three-phase, half-wave MCM AC–DC commutator, with a single homogeneous cathode
commutation group and a neutral, acting as a one-way MCM AC–DC rectifier.

AC and DC commutatorless electrical machines do not permit smooth and

economical control of voltage and current (for MMD generation) or torque and
speed control (for MMD motoring) of the kind that is necessary for some forms of
industrial and traction propulsion complex dynamical hypersystems.
The use of the integral MCM commutator in MCM AC–AC and AC–DC/DC–AC
commutator dynamotors makes it possible to produce variable-voltage and
variable-current or variable-torque and variable-speed dynamotors with flat or
drooping voltage–current and torque–speed characteristics and with the power
factor in some of them brought to unity.
All of these useful characteristics of MCM AC–AC and AC–DC/DC–AC
commutator dynamotors are achieved at the expense of complexity and the extra

14-6
 Mechatronics

Figure 14.8. The three-phase full-wave MCM AC–DC commutator, with double heterogeneous cathode and
anode commutation groups and a neutral, acting as a double-way MCM AC–DC rectifier.

cost of the MMD electrical machine, with reduced reliability and worse operating
conditions.
The main difficulty in building MCM AC–AC and AC–DC/DC–AC commutator
dynamotors is to obtain satisfactory ME commutation of alternating and direct
currents, especially under starting conditions.
In order to make an MCM AC–AC commutator dynamotor suitable for AC
supply it is necessary to adapt its magnetic circuit to an alternating magnetic field,
reduce the reactance of the windings to a minimum and, finally, improve the
conditions of the contacted commutation in the MCM AC–AC commutator
dynamotors. The first two requirements may be considered as being suitably fulfilled
at the expense of considerable complexity and the increased cost of the MMD
electrical machine. As for securing good commutation conditions, nothing that has
been done in this direction has satisfied scientists, who do not think that the problem
has been properly resolved and are looking for a better solution.

14.2 MCM AC–DC/DC–AC commutator dynamotors

Industry has used MCM AC–DC/DC–AC commutator dynamotors (generators/
motors) for years because of their outstanding voltage and/or current as well as

14-7
 Mechatronics

torque and/or speed control capabilities, down to essentially zero (0) revolutions
per second (rps), and because of their good starting and low-speed torque. MCM
AC–DC/DC–AC commutator dynamotors are usually much more expensive
than standard AC commutatorless dynamotors, and the high-current part (the
‘armature’) of the MCM AC–DC/DC–AC commutator dynamotor is on the rotor.
In getting from the DC terminals into the armature, these high currents must pass
through brushes gearing on an MCM AC–DC/DC–AC commutator, a rotating
AC–DC rectifier/DC–AC inverter.
Wear on the MCM AC–DC/DC–AC commutator and the brushes adds
substantially to maintenance costs and reduces reliability. MCM AC–DC/DC–AC
commutator motors have the further disadvantage that large DC power supplies are
generally not available unless especially provided for.
In the last few years, cheaper and higher power electronic-control equipment has
been developed, which converts AC–DC right at the dynamotor.
Thus, small and moderate-sized MCM AC–DC/DC–AC commutator motors can
be supplied in effect right from the AC mains, eliminating the need for special
power supplies. Such an electronic controller can also readily adjust input DC
voltage to the MCM AC–DC/DC–AC commutator motor. Typical applications are
propulsion motors in electric traction railroads or automotive vehicles, aircraft and
ships, and also electrical drills and hand tools.
The use of MCM AC–DC/DC–AC commutators gives the electrical machines
several valuable properties, of which the first is the possibility of AC–DC rectifying
the polyphase AC induced voltages in the rotating armature winding into a single-
phase DC induced voltage at the carbon brushes, and/or DC–AC inverting the DC
induced voltage at the carbon brushes into polyphase AC induced voltages in the
rotating armature winding.
The generation of an AC induced voltage in the armature winding of an MCM
AC–DC commutator generator is exactly the same as in an AC commutatorless
generator. As the armature winding rotates, an alternating current is produced in it.
Some rotating AC–DC rectifier must be inserted between the armature winding and
the terminals so that the terminal AC induced voltage can always act upon the
external load in the same sense of direction.
This rotating AC–DC rectifier must reverse the connection of the winding
conductors to the external circuit at the instant when the AC induced voltage of
the conductors is zero and changing its sense of direction. This rotating AC–DC
rectifier is termed by the author the ‘MCM AC–DC commutator’, and the rotating
DC–AC inverter is termed the ‘MCM DC–AC commutator’. The Achilles’ heel for
an MCM AC–DC/DC–AC commutator dynamotor is its MCM commutator.
Consistently transferring DC from a rotating shaft to the stationary point, or,
vice versa, from a stationary point to the rotating shaft, requires materials that can
carry DC yet withstand friction and arcing. MCM commutators require periodic
maintenance, decrease dynamotor life and limit the maximum speed at which the
dynamotor can turn.

14-8
 Mechatronics

14.3 Schrage MCM AC–AC commutator motor

The AC commutatorless, hyposynchronous wound-rotor motor, frequency con-
verter and adjustable-ratio transformer were first combined into a single structure by
K H Schrage of Sweden to provide a complete, self-contained, adjustable-speed
EM motor (Laughton and Say 1985). The electrical machine consists of a special
phase-wound motor, whose ‘rotor’ acts as the ‘primary’ (it is connected to the
three-phase electrical energy source through a set of slip rings) and the ‘stator’,
which performs as the ‘secondary’. This is just the reverse of the functions of the
stator and rotor windings in the standard AC commutatorless, hyposynchronous
wound-rotor motor. In addition, there is a third winding that functions as an
electromagnetic link between the primary and the secondary and as a source of slip-
frequency voltage for the secondary. This so-called MCM commutator ‘tertiary’
winding is placed directly over the primary and is connected to an MCM AC–AC
commutator mounted on the shaft end opposite the slip rings. Instead of being
connected to a resistance controller, the secondary winding is wired to three pairs of
carbon brushes that ride on the MCM AC–AC commutator. Speed control is
effected by moving a lever or turning a handwheel that shifts the three pairs of
carbon brushes. Thus, the recovery of slip energy can be achieved in a single
electrical machine by incorporating an MCM AC–AC commutator in the rotor.
‘Schrage’ (rotor-fed) and ‘doubly fed’ (stator-fed) MCM AC–AC commutator
motors have some commercial importance. Both have a ‘shunt’ speed–torque
characteristic and can operate both above and below synchronous speed. A three-
phase MCM AC–AC commutator motor with a ‘series’ characteristic is also
available.
Summing up, the Schrage MCM AC–AC commutator motor has its primary
winding on the rotor, connected to the supply through slip rings and carbon brushes.
The rotor also carries a low-voltage MCM commutator tertiary winding with
conductor located in the same slots as, and above, those of the primary. The
secondary winding is on the stator. The primary and secondary windings are similar
to those of the conventional AC commutatorless induction motor. The brush gear
comprises two movable rockers, each fitted with three carbon-brush spindles per pair
of poles. The two rockers are geared together, each being fitted with a toothed
segment. The two segments mesh with pinions fitted to a short shaft to which either a
handwheel or a small pilot actuator is connected. The gearing is so arranged that the
movement of the handwheel or pilot actuator causes the carbon-brush rockers to
move in opposite senses of direction.
The carbon brushes attached to each rocker move over separate portions of
the MCM commutator surface to enable these brushes to be placed ‘in line’ or to be
moved in either sense of direction, so that more or fewer commutator sectors are
included between a brush on one rocker and the corresponding carbon brush on
another rocker. The bus-rings of each rocker are connected to the secondary (stator)
winding as in figure 14.9, which represents a two-pole electrical machine.

14-9
 Mechatronics

Figure 14.9. Schrage MCM AC–AC commutator motor (Laughton and Say 1985).

As the primary winding is on the rotor, AC induced voltages at the brushes of slip
frequency are induced in the secondary (stator) winding. The AC induced voltage at
the carbon brushes is also of this frequency.
The AC induced voltage induced in each coil of the MCM tertiary commutator
winding is constant at all speeds, and therefore the AC induced voltages injected
via the carbon brushes into the secondary winding are proportional to the number
of MCM commutator sectors included between the corresponding carbon brushes
on the two rockers, i.e. between the brushes connected to a particular phase of the
secondary. Thus when these carbon brushes are in line, the secondary winding has
no AC induced voltage injected into it and the MCM AC–AC commutator motor
will run with its natural slip. When the carbon brushes are moved so that the
injected AC induced voltage opposes current, the angular velocity is reduced
(slip positive); for the opposite movement, the angular velocity is increased (slip
negative).
In the Schrage MCM AC–AC commutator motor with a synchronous angular
velocity ωs and a carbon brush-separation electrical angle θ , the no-load angular
velocity is
⎛ 1 ⎞
ω = ωs ⎜1 − k sin θ ⎟ ,
⎝ 2 ⎠
where k is a constant, depending on the number of turns in the secondary and
commutator windings.
Because of its simplicity of operation and the ease with which a large stepless
angular-velocity range can be obtained (above and below synchronism), Schrage
MCM AC–AC commutator motors have been and are still used for many
moderately small industrial applications.

14-10
 Mechatronics

14.4 Exemplary applications of mechanical commutation

matrixer (MCM) commutators
14.4.1 AC commutatorless motor—an MCM AC–DC commutator generator
for converting AC–DC
In order to obtain DC electrical energy, one can first convert AC electrical energy
into mechanical energy by means of an AC commutatorless EM motor, and then
convert the mechanical energy back into electrical energy (but this time in the form
of DC) by means of an MCM AC–DC commutator generator. The necessary motor-
generator (MG) system (Ward Leonard system) must, therefore, consist of two
electrical machines, connected together mechanically (figure 14.10).
A valuable advantage of this system is the complete electrical independence of the
DC mains from the AC mains; in many practical applications, it is very important to
be able to control, over a wide range, the DC voltage delivered by the MG system.
This system is, for example, an essential part of the generator-motor (GM) system,
called the ‘rotary dynamo electrical Ward Leonard system’, which was, in the past,
used in cases where it was necessary to regulate the speed of an actuating mechanism
over a very wide range. The disadvantage of the rotary Ward Leonard system is,
above all, its low efficiency, since the losses in the unit are made up of the losses in
both electrical machines.
MG systems, in which a change in a relatively small control input-signal power
produces a significant change in a large controlled output-signal power of the
system, are called ‘dynamoelectrical amplifiers’. Of course, the dynamoelectrical
amplifier must receive electrical energy from some energy source whose power is
essentially controlled by the input-signal power. In this sense, a separately excited
MCM AC–DC commutator generator can serve as a dynamoelectrical amplifier;
a small change in electrical energy applied to the exciting winding produces a greater
change in the electrical energy delivered to the external circuit. However, the
amplification in such an MG system is relatively small, of the order of 30 to 40, and

Figure 14.10. The AC commutatorless motor—an MCM AC–DC commutator generator for converting
AC–DC (Ward Leonard system).

14-11
 Mechatronics

the variation process proceeds slowly because of large inductance in the exciting
winding.

14.4.2 MCM AC–DC commutator amplidyne

The development of industrial electrification required the production of more up-to-
date dynamoelectrical amplifiers on the basis of separately excited MCM AC–DC
commutator generators. The previously widely used dynamoelectrical amplifier also
known as the ‘MCM AC–DC commutator amplidyne’ is basically a separately excited
MCM AC–DC commutator generator, but the basic magnetic flux which determines
the load voltage is produced by the armature magnetic field. This method of exciting
the main magnetic field was, in the past, used in several special-purpose MCM AC–
DC commutator generators (for railway trains, welding, lighting, etc). These electrical
machines are called ‘cross-field excitation’ electrical machines. Figure 14.11 shows the
basic schematic diagram of an MCM AC–DC commutator amplidyne.
In contrast to normal MCM AC–DC commutator generators, its magnetic-field
system is usually a non-salient magnetic pole with slots of differing sizes for several
exciting windings, control windings and a compensating winding.
The MCM AC–DC commutator amplidyne operates as a generator with a
constant rotational speed, and its prime mover is usually an AC commutatorless
motor. One of the exciting windings is connected across some small DC energy

Figure 14.11. MCM AC–DC commutator amplidyne with basic control and compensating fields (Siskind 1963).

14-12
 Mechatronics

source and the input-signal voltage at its terminals produces a relatively small
exciting current. The latter excites a relatively small longitudinal magnetic flux,
which links the armature winding. The first pair of carbon brushes placed on the ring
commutator across the cross axis of the amplidyne are short-circuited, and so the
small DC induced voltage produces a considerable current in the armature winding.
A second pair of carbon brushes is situated on the ring commutator on the vertical
axis of the electrical machine, but the DC induced voltage does not cause any
potential difference between them.
The armature current passes through the armature winding and produces a
relatively large cross magnetic flux; the magnetising force of this current is large,
and because of the non-salient magnetic-pole construction of the MCM AC–DC
commutator amplidyne frame, the reluctance is a salient magnetic-pole DC
electrical machine. The cross-magnetic flux induces a second relatively large
DC induced voltage in the armature winding. This DC induced voltage produces
a DC output voltage between the brushes placed on the longitudinal axis, resulting
in a certain output current being produced in the controlled load circuit of the MCM
AC–DC commutator amplidyne.
The controlled electrical energy at the amplidyne output will be many times
greater than the input electrical energy. Of course, a certain resultant armature
current is produced in the armature winding, but it is convenient on the basis of the
superposition principle to consider the armature current and output current
separately. It is necessary that the magnetising force of the output current should
not change the longitudinal magnetic flux. In order to ensure this, in the frame slots
of the amplidyne there is (in addition to the interpole commutating winding) a
compensation winding situated on the frame of the amplidyne and connected in
series with the current circuit, the function of which is to compensate for the
magnetising force of the output current.
At present, the field of application of the MCM AC–DC commutator amplidyne
is very narrow. To date, they have been employed in old installations for controlling
reversing rolling mills, cold rolling mills, blast-furnaces and mine winders, and for
metal-cutting and paper machines, etc.

14.4.3 MCM AC–DC/DC–AC ring-commutator single-armature frequency

converter
In the past, a natural development of the idea of converting AC electrical energy
directly into DC electrical energy was to do away with the mechanical link and
perform the conversion in one electrical machine. This was first achieved in a rotary
dynamoelectrical MCM AC–DC/DC–AC ring-commutator single-armature
frequency converter, which is also known as an MCM AC–DC/DC–AC commu-
tator dynamotor. This is an MCM AC–DC/DC–AC commutator generator with
slip rings on the extended shaft at the end opposite to the MCM commutator. These
slip rings are connected to equidistant points of the armature winding in the sense of
rotation direction. Carbon brushes are placed on the rings and are connected across
an AC voltage (figure 14.12).

14-13
 Mechatronics

Figure 14.12. MCM AC–DC ring-commutator single-armature frequency converter (Kasatkin and Perekalin
1970).

Figure 14.13. Circuit of a three-phase MCM AC–DC ring-commutator single-armature frequency converter
(Kasatkin and Perekalin 1970).

The number of slip rings depends on the number of phases of AC supply that
must be converted. For a single-phase supply, two slip rings are necessary, for a
three-phase supply, three are required, and for five-phase supply, five are needed.
The diagram in figure 14.13 shows a three-phase MCM AC–DC/DC–AC
ring-commutator single-armature frequency converter; for simplicity, it is two
magnetic poles and has a ring armature winding.
The points at which the slip rings are connected to the armature winding in a
three-phase MCM AC–DC ring-commutator single-armature converter must be
spaced 2π /3 electrical radians apart.
If the armature winding consists of 2a parallel branches, each pair of parallel
branches, in succession round the armature winding, must be divided into three parts
by the points of construction of the slip rings. It follows that a point of the armature
winding is connected to each slip ring. Thus, relative to the AC side, the armature

14-14
 Mechatronics

Figure 14.14. Circuit connections of the windings of a slip-ring MCM AC–DC ring-commutator single-armature
frequency converter (Kasatkin and Perekalin 1970).

winding is connected in delta (Δ), each side of which consists of a parallel branch
(figure 14.14).
In converting AC to DC, the MCM AC–DC/DC–AC ring-commutator single-
armature frequency converter receives AC on the slip-ring side and delivers DC from
the MCM AC–DC commutator side; it operates as an AC commutatorless
synchronous motor with respect to the AC mains, and as an MCM AC–DC
commutator generator with respect to the DC mains. If the DC side of the MCM
AC–DC/DC–AC ring-commutator single-armature converter is disconnected and
the excitation left connected, one obtains an AC synchronous electrical machine
with stationary magnets and rotating armature.
It can operate as an AC commutatorless synchronous motor at the expense of
electrical energy taken from the AC mains, and as an AC synchronous generator if it
is driven by some prime mover.
MCM AC–DC/DC–AC ring-commutator single-armature converters are also
used for converting DC into AC, acting as a rotating DC–AC inverter.
In this case, it operates as an MCM DC–AC commutator motor with respect to
the DC mains and an AC commutatorless synchronous generator with respect to the
AC mains. If the AC side is disconnected, the electrical machine will be able to
operate as an MCM DC–AC commutator motor at the expense of the electrical
energy taken from the DC mains, and as an MCM AC–DC commutator generator
if it is driven by some prime mover.
The armature current in the armature winding of the MCM AC–DC/DC–AC
ring-commutator single-armature frequency converter may be regarded as the result
of a superimposition of the alternating and direct currents, consequently, the actual
current in the winding will be the result of the alternating and direct currents.
The armature current reacts with the magnetic field of the poles and produces
a corresponding torque.
A great advantage of the MCM AC–DC/DC–AC ring-commutator single-
armature frequency converter is the fact that the armature winding is heated less
by the resultant current than would be the case if there were only alternating or

14-15
 Mechatronics

direct current flowing through it. The more phases there are in the MCM AC–DC/
DC–AC ring-commutator single-armature frequency converter, the smaller the
resultant current. When it is operating as a polyphase MCM AC–DC/DC–AC
ring-commutator single-armature frequency converter, the electrical machine has
considerably smaller losses than it does when operating as an MCM AC–DC
commutator generator. This fact makes it possible to obtain more electrical energy
from an MCM AC–DC/DC–AC ring-commutator single-armature frequency
converter on the DC side than from an MCM AC–DC commutator generator of
the same size.
In the past, MCM AC–DC/DC–AC commutator single-armature frequency
converters were widely used as high-, medium- and low-power rotating DC–AC
inverters for converting DC to AC. This was required in DC supply systems for
powering AC devices and radio apparatus (e.g. in aircraft and automotive vehicles),
as well in high-power, high-frequency MCM DC–AC commutator generators for
long-wave radio stations (e.g. in the old radio station in Nauen, Germany).
A few years ago, the high-power MCM AC–DC ring-commutator single-
armature frequency converters were also used in DC supply systems for urban
electrical traction substations for tramways and trolleybuses (e.g. in the old electrical
traction substation in Szczecin, West Pomerania, Poland).

14.4.4 MCM DC–AC/AC–DC ring-commutator single-armature frequency

converter and adjustable-ratio EE transformer connected to an AC
commutatorless wound-rotor motor for rotational-speed-control purposes
Something that is important for speed-control by frequency variation is the need for
an injected voltage whose frequency must always be the same as the slip frequency.
Since the rotor frequency increases progressively as the speed drops below or rises
above synchronism, the equipment supplying the injected voltage must automatically
adjust itself to generate slip-frequency AC induced voltage as the main AC
commutatorless wound-rotor motor changes speed.
In one fundamental design, the electrical machine that supplies the adjustable
voltage for the phase-wound rotor is somewhat similar to a rotary converter, except
that the stator is a laminated core with no winding, acting merely as part of a good
magnetic circuit. The rotor contains the usual DC armature winding that is connected
to an MCM commutator at one end and slip rings at the other (Siskind 1963).
As illustrated in figure 14.15, the brushes riding on the slip rings are connected to
the three-phase power lines through a set of adjustable-ratio EE transformers, while
another set of brushes, spaced 2π /3 electrical radians apart, rests on the MCM
commutator and is wired to the wound-rotor slip rings.
The special electrical machine is coupled to the main AC commutatorless
wound-rotor motor and therefore always operates at the same speed as the latter.
To understand why the slip-ring frequency of the MCM DC–AC/AC–DC single-
armature converter always matches the slip frequency of the AC commutatorless
wound-rotor motor, it is necessary to recognise the following conditions of
operation.

14-16
 Mechatronics

Figure 14.15. MCM DC–AC ring-commutator single-armature frequency converter and adjustable-ratio
EE transformer connected to an AC commutatorless wound-rotor motor for rotational-speed-control purposes
(Siskind 1963).

• Line frequency currents enter the AC end of the MCM DC–AC/AC–DC

single-armature frequency converter to set up a revolving magnetic field that
rotates at a synchronous rotational speed with respect to the armature-winding
conductors.
• If the rotor is at rest, the magnetic field thus produced revolves in space at
a synchronous rotational speed.
• If the rotor is turned at a synchronous rotational speed in a sense of direction
that is opposite to the motion of the magnetic field with respect to the moving
conductors, the magnetic field is stationary in space. Under this condition, the
magnetic field is, in effect, rotated backward as rapidly as it is moved forward.
• At any rotational speed intermediate between rest and synchronism, the
magnetic field revolves at a rotational speed that is the difference between
synchronism and the rotational speed of the rotating electrical machine.

It follows, then, that the frequency at the MCM commutator brushes is always
identical to that of the interconnected wound-rotor slip rings. In particular, the
MCM commutator-brush frequency and the slip frequency are both f (cps or Hz)
when the rotor is at rest, and the two frequencies are zero (0, cps or Hz) when the
rotor turns at synchronous rotational speed in a sense of direction opposite to the
rotation of the magnetic field with respect to its own conductors.
As in all rotary converters, the voltage at the MCM commutator end always
bears a very definite ratio to the impressed AC induced voltage at the slip-ring end.

14-17
 Mechatronics

This means that, with an adjustable-voltage EE transformer to supply electrical

energy to the slip-ring end, the correct voltage, for the desired rotational speed, will
be available at the MCM commutator.
The Scherbius method permits the operator to control the rotational speed of
a constant torque application over an extremely wide range above and below
synchronism and to do so steplessly.
Figure 14.13 illustrates the arrangement of the electrical machines and auxiliary
equipment and shows how the control circuit interconnects the various units. The
propulsion mechatronical control system, the so-called ‘rotary dynamoelectrical
Scherbius system’ is put into operation by first bringing the main AC commutator-
less wound-rotor motor up to rotational speed. As previously explained, this is done
by closing the main contactor’s line contacts M, and then closing the starting
contacts S and the accelerating contacts A. The slip-ring brushes are next transferred
to the MCM commutator brushes by closing the C contacts and opening the S and A
contacts. Finally, the T contacts are closed to connect the adjustable-ratio
EE transformer to the slip rings of the MCM DC–AC/AC–DC single armature
frequency converter, and manual settings are made at the EE transformer to provide
the required injected voltage for the desired rotational-speed adjustment.
Although this scheme was rarely used as described in the past, it is basic to all
other propulsion mechatronical control systems in which an AC commutatorless
wound-rotor motor is the main drive and voltage injection in the secondary is behind
some of the advanced propulsion mechatronical control systems presently in use,
as well as the special Schrage MCM AC–AC commutator motor discussed in this
chapter.

14.4.5 MCM DC–AC/AC–DC ring-commutator single-armature frequency

converter and MCM DC–AC commutator motor coupled to the shaft
of the AC commutatorless wound-rotor motor for speed-control purposes
For variable-speed propulsion mechatronic control systems requiring very large AC
commutatorless wound-rotor motors, from about several megawatts (MW) of
power, the wound-rotor is preferred. When used in such installations, it is, however,
associated with auxiliary equipment, which unlike resistance controllers, returns
much of the slip-frequency electrical energy of the rotor to the system. One method
of doing this was described and illustrated in figure 14.15.
Another variable-speed propulsion mechatronic control system that was
frequently employed in the past in, for instance, steel-mill drives, dredges and other
heavy-machinery loads whose rotational speeds must be adjustable over extremely
wide ranges, is the so-called ‘rotary dynamo-electrical Kraemer system’, two
variations of which are used to convert the slip frequency electrical energy from
the main AC commutatorless wound-rotor motor into useful electrical energy.
In one, the constant-power propulsion mechatronical control system, DC
electrical energy from the MCM DC–AC/AC–DC ring-commutator single-
armature frequency converter is supplied to an MCM DC–AC commutator motor
coupled to the shaft of the main AC commutatorless wound-rotor motor. In the

14-18
 Mechatronics

other, the constant-torque propulsion mechatronical control system, the DC output

of the MCM DC–AC/AC–DC ring-commutator single-armature frequency con-
verter ultimately drives an AC commutatorless synchronous or hyposynchronous
(induction) generator that feeds back into the AC supply lines.
Figure 14.16 shows the wiring connections for the constant-power propulsion
mechatronical control system, called the rotary dynamoelectrical Kraemer system.
With the MCM DC–AC/AC–DC ring-commutator single-armature frequency-
converter’s contactor C open, the unloaded main AC commutatorless wound-rotor
motor is brought up to rotational speed in the usual way with the rotor-resistance
starter. Contacts S and A are then opened, normal excitation is applied to the
frequency converter, and contact C is quickly closed to impress slip-frequency
voltage across its slip rings to permit the electrical machine to rotate slowly.
The magnetic field of the MCM DC–AC commutator motor is excited
next, whereupon that unit, which is coupled to the main AC commutatorless
wound-rotor motor and running at nearly synchronous rotational speed, generates
a DC voltage that is impressed across the MCM commutator end of the MCM
DC–AC/AC–DC ring-commutator single-armature frequency converter. Because
a definite voltage ratio always exists between the AC and DC ends of the MCM
DC–AC/AC–DC ring-commutator single-armature frequency converter, a certain

Figure 14.16. Rotary dynamoelectrical Kraemer system for rotational-speed control for constant-power
operation (Siskind 1963).

14-19
 Mechatronics

slip-frequency potential appears at the slip ring and is injected into the rotor of the
main AC commutatorless wound-rotor motor. The latter therefore slows down,
automatically adjusting its own input current for the particular load that was
applied after starting.
With the rotational speed of the main AC commutatorless wound-rotor motor
reduced and its slip frequency and voltage increased, the MCM ring-commutator
single-armature frequency converter speeds up to deliver electrical energy to the
armature of the MCM DC–AC commutator motor, which in turn converts it to
mechanical energy that aids the main AC commutatorless wound-rotor motor
in driving the load. The set now stabilises at the adjusted rotational speed for
the particular load, delivering a mechanical-energy output equivalent to the
mechanical-energy input, minus the losses.
To decrease the rotational speed of the main drive, the excitation of the MCM
DC–AC commutator motor must be increased. This raises the DC induced voltage
of that electrical machine, causes the voltage at the slip rings to increase, and results
in a higher injected slip-frequency potential, which brings about the desired drop in
the rotational speed. The reverse is true when the excitation is reduced.
In fact, if the excitation to the MCM DC–AC commutator motor is reduced to
zero (0) when the main AC commutatorless wound-rotor motor operates with small
slip, and is then reversed with slightly increased excitation, the drive will run at
synchronous rotational speed. Under this condition, the MCM DC–AC/AC–DC
ring-commutator single-armature frequency converter stops, the slip frequency is
zero (0), and the main AC commutatorless wound-rotor motor becomes an AC
commutatorless synchronous motor with DC excitation. Further increasing the
reverse excitation causes the main AC commutatorless wound-rotor motor to
operate above synchronous rotational speed, and the sense of the rotation direction
of the MCM DC–AC/AC–DC ring-commutator single–armature frequency con-
verter reverses. For rotational speeds above synchronism, the DC electrical machine
becomes an MCM AC–DC commutator generator, imposes an additional load on
the main AC commutatorless wound-rotor motor, and feeds its electrical energy
through the MCM DC–AC/AC–DC ring-commutator single–armature frequency
converter into the rotor of the main AC commutatorless wound-rotor motor.
A useful feature of the propulsion mechatronical control system is the fact that
it is possible to control the power-factor by adjusting the excitation of the
rotary dynamoelectrical MCM DC–AC/AC–DC ring-commutator single–armature
frequency converter; correction up to about 0.95 is generally practical through
over-excitation.
Compared with other propulsion mechatronic control systems for heavy power
drives—notably the rotary dynamoelectrical Scherbius system, which uses such
special types of equipment as ohmic drop exciters, AC commutatorless hyposyn-
chronous (induction) electrical machines, EE transformers and MCM DC–AC
commutator motors, the rotary dynamoelectrical Kramer system is economical.
This is because the MCM DC–AC commutator motors and rotary dynamoelectrical
DC–AC/AC–DC ring-commutator single–armature frequency converters were in
the past, for the most part, similar to those of the standard design.

14-20
 Mechatronics

References
Fijalkowski B 1973 Kierunki rozwoju napedu elektrycznego lokomotyw (Development trends of
the electric drive of locomotives) Materiały konferencyjne—I Krajowa Konferencja—Pojazdy
Szynowe (Krakow: Instytut Pojazdow Szynowych Politechniki Krakowskiej) pp 109–28
(In Polish)
Fontaine H 1878 Electric Lighting: A Practical Treatise (Translated from French by Higgs P.)
(London: E and FN Spon) p 86
Kasatkin A and Perekalin M 1970 Basic electrical engineering (Moscow: Mir Publishers) p 480
Kuznetsov M 1960 Fundamentals of Electrical Engineering (Moscow: Peace Publishers)
Laughton M A and Say M G 1985 Electrical Engineer's Reference Book (London: Butterworth)
Siskind Ch S 1963 Electrical control systems in industry (New York: McGraw-Hill)
Vesolovsky O N and Kopylov I P 1984 U istokov elektromekhaniki [K 150-letiyu elektro-
dvigatelya Yakobi i principa obratimosti] Elektrotekhnika (In Russian)

14-21
 IOP Publishing

Mechatronics
Dynamical systems approach and theory of holors
B T Fijalkowski

Chapter 15
Electrical commutation matrixer commutators
for modern DC and AC magneto-mechano-
dynamical electrical machines

A scientific truth does not triumph by convincing its opponents and making them
see the light, but rather because its opponents eventually die and a new
generation grows up that is familiar with it.
Max Planck

15.1 Introduction
The world must solve important challenges to convert electrical energy and trans-
form information for mechatronic control systems in an efficient way. Examples of
this are transportation, renewable energies and industrial processing applications.
These problems can be solved using energy converters based on modern semi- and/or
superconductive electrical commutation matrixer (ECM) AC–AC, AC–DC–AC,
AC–DC/DC–AC, DC–DC and DC–AC–DC commutator applications in magneto-
mechano-dynamical (MMD) electrical machines.
The ideal ECM commutator in many of these applications may have the
following characteristics:
• sinusoidal input and output currents;
• operation with unity power factor;
• regeneration capability;
• compact design with a good power-to-mass ratio.

Semi- and/or superconductive ECM AC–AC, AC–DC–AC, AC–DC/DC–AC,

DC–DC and DC–AC–DC commutators (matrix converter) can fulfil all these
characteristics and this is the reason for the tremendous interest in this topology.

doi:10.1088/978-0-7503-1350-6ch15 15-1 ª IOP Publishing Ltd 2016

 Mechatronics

In the last decade, many advances in the development of this topology have been
presented, including industrial applications up to megawatt level. The use of the
semi- and/or super-conductive ECM AC–AC, AC–DC–AC, AC–DC/DC–AC,
DC–DC and DC–AC–DC commutator (matrix converter) in real applications
and the challenges that these applications present are very topical and important.
In view of the fact that sources of conventional energy are gradually being
depleted, the general public is increasingly interested in discussions about novel
types of future-oriented energy-saving industrial and traction propulsion complex
dynamical hypersystems. These energy-saving complex dynamical hypersystems
have already become an alternative to be taken seriously. The development of
today’s industrial and traction propulsion complex dynamical hypersystems is
closely tied up with developments in electronic technology.
The MMD electrical machines (dynamotors) currently in use have remained
unchanged since their conception before the turn of the 20th century; new electrical
dynamotor concepts have only gained marginal acceptance.
Today, there are good possibilities available to select optimal industrial and
traction propulsion complex dynamical hypersystems with due consideration
being paid to performance, lifetime costs, environment, influence on the storage
battery, etc. This very advanced propulsion (VAP) technology is based on two
important disciplines—electrical-integrated-matrixer (EIM) commutator and
electrical-integrated-circuit (EIC) processor techniques. Some of the results achieved
higher performance and other operational benefits. The microcontroller unit (MCU)
gives the necessary conditions for further improvements, such as diagnostics.
Continued research and development (R&D) work on ECM commutators will
lead to greater competitiveness for AC–AC and AC–DC/DC–AC commutator
dynamotors (generators/motors) for industrial and traction propulsion complex
dynamical hypersystems.
Recent years have not only witnessed rapid progress in ‘microelectronics’ (i.e. the
area of electronics dealing with the design, realisation and application of various
low-power EICs made in a solid (continuous medium) or on its surface), but also in
‘macroelectronics’ (i.e. the area of electronics dealing with the design, realisation and
application of various high-power ECMs made in a solid (continuous medium) or on
its surface).
Above all, it refers to the developments that have taken place in the field of
MCUs (i.e. low-power EICs with a central processing unit (CPU), memory and
input/output (I/O) capability, made in the shape of individual chips containing very
large scale of integration (VLSC) EICs) and ECM commutators (i.e. electronic
commutators containing standard scale integration (SSI) and medium-scale integra-
tion (MSI) ECMs). The latter, implemented by the author in 1973, have been widely
popularised, particularly in the field of industrial and traction propulsion complex
dynamical hypersystems (Fijalkowski 1973, 1985, 1987, 1988, 2010, 2011).
Over 40 years ago, the author started R&D works on mono- and poly-crystalline
as well as amorphous semiconductor electronic commutators with uni- and/or
bipolar electrical valves, which he called for the first time ‘ECM commutators’,
and these aroused the interest of the greatest world consortia. These consortia are

15-2
 Mechatronics

aware of the fact that the author is one of the few scientific men who initialised the
avalanche of concepts, the consequences of which are difficult to foresee. It is
obvious that these concepts will not only change contemporary integrated power
electronics (now termed by the author ‘macroelectronics’) to a certain degree, but
also modern industrial and traction propulsion complex dynamical hypersystems
and, consequently, newly designed energy-saving electrical machines. The scale of
these changes remains unknown for the time being. Behind these concepts hides the
harbinger of the new universe—the ‘universe of macroelectronics’.
Is the new universe—the ‘universe of power electronics’—going to be better than
the existing one? This is unknown. Certainly, it offers new possibilities that are
difficult not to take advantage of.
Power electronics and industrial and traction propulsion complex dynamical
hypersystems are again queuing for the results achieved in solid-state physics
laboratories.
The scientific revolution in the field of power electronics has taken place very rapidly,
because stereotypical ideas in this area have been overcome.
All progress is stimulated by thinking that is free from dogmas that are opposed
to challenges to existing concepts. It is also known that the best practical solutions
are born out of ingenious theory. ‘There is nothing more practical than a good
theory’ said Ludwig Boltzmann (1844–1906).
The concept of a generalised intelligent multi-input/multi-output (MIMO) ECM
AC–AC and/or AC–DC/DC–AC commutator with discrete and/or continuous uni-
and/or bipolar electrical valves was born due to the mathematical hypermatrix
notation of the dynamical systems approach for the formation methods of
mathematical models of ECM AC–AC and/or AC–DC/DC–AC commutator
dynamotors (Fijalkowski 1973, 1985, 1987).
MIMO ECM AC–AC and/or AC–DC/DC–AC commutators are the key to
overcoming the contactless (brushless), i.e. sparkless commutation (changing-over
the way of electrical current flow), and so the static conversion of one kind of
electrical energy into another one, with the application of the phenomena inducing
uncontrolled or controlled electrical current conductivity (carrying of electrical
charges by positive or negative carriers into a definite medium under the action of
electrical field).
MIMO ECM AC–AC and/or AC–DC/DC–AC commutators have introduced
the era of newly designed energy-saving AC–AC and AC–DC/DC–AC commutator
dynamotors, enabling continuous (stepless) and contactless control of their current
and voltage (during generating), or their torque and speed (during motoring).
These ECM AC–AC and/or AC–DC/DC–AC commutators with uni- and/or
bipolar electrical valves constitute a successive breakthrough in the development of
ECM AC–AC and AC–DC\DC–AC commutator dynamotors, and multiply their
possibilities enormously. There are fantastic perspectives that will probably become
reality as soon as in the current century. However, the scientific sense of concepts
conceived for the first time by this author forty years ago does not consist of this. It is
hope that attracted the attention of the greatest world consortia. The most essential
value of this new concept was the achievement of a cardinal breakthrough in how

15-3
 Mechatronics

one thinks about the static conversion of one kind of electrical energy into another
one, with the aid of continuous uni- and/or bipolar electrical valves—a change of
paradigm regarding the kind of thinking that is standard in the field of power
electronics. In order to understand it, it is necessary to look into the past.
The physical science of electricity and electrical machinery is one of the youngest
branches of physical knowledge. Electrical machine building is actually above
100 years old. However, in this relatively short period a tremendous amount of
R&D work has been carried out, which has revolutionised the technical and
economic features of modern industry, traction and our social lives.
The construction of different types of magneto-mechano-dynamical (MMD),
magneto-fluido-dynamical (MFD), magneto-plasmo-dynamical (MPD), electro-
mechano-dynamical (EMD), electro-fluido-dynamical (MFD) and electro-plasmo-
dynamical (EPD) electrical machines is a subject for a special course of study and is
beyond the scope of this textbook. Here the author considers only this fundamental
principle underlying the MMD conversion of mechanical energy into electrical
energy and vice versa.
In the dynamical systems approach to MMD electrical machines, one classifies
them as physical heterogeneous continuous simple dynamical hypersystems, i.e.
mechano-electrical and/or electro-mechanical (ME/EM) heterogeneous continuous
simple dynamical hypersystems (Fijalkowski 1985).
An MMD conversion involves the interchange of energy between a mechanical
continuous dynamical system and an electrical continuous dynamical system, or vice
versa, through the medium of coupling magnetic-field energy, and utilises the effect
of electromagnetic induction and the mechanical action of the coupling magnetic-
field energy for its accomplishment.
The process is essentially reversible, except that a small amount of the thermal
energy of a thermal continuous dynamical system is lost as heat.
Thus, MMD electrical machines are the principal rotary and linear converters
of mechanical energy into electrical energy, and/or electrical energy into mechan-
ical energy, and they are also the principal rotary converters of one form of
electrical energy into another form, differing in voltage, current and sometimes
frequency.
When the MMD conversion takes place from mechanical into electrical energy,
the MMD electrical machine is called an ‘ME generator’ (or an ‘ME dynamo’).
When electrical energy is converted into mechanical energy, the MMD electrical
machine is called an ‘EM motor’.
When bilateral energy MMD conversion is necessary, the MMD electrical
machine is called an ‘ME/EM dynamotor’.
The constant improvements in the design of MMD electrical machines have made
possible many new practical applications, and have provided the impulse for further
progress, while fostering the most diverse uses of electrical energy.
This explains why MMD electrical machines were given great attention by
scientists and engineers, and why MMD electrical machines quickly attained
technical perfection of design.

15-4
 Mechatronics

15.2 Status and trends

The properties of unipolar electrical-current conductivity, i.e. the nonlinearity of the
current–voltage characteristics of polycrystalline semiconductors, were discovered
by Adams and Day in the 1880s. In 1883, Fritsch described a construction of
selenium electrical valves, and these were not applied on a large scale until the end of
the 1920s. In 1926, Grondahl discovered cuprous-oxide electrical valves, which
became popular in the 1930s. At the same time, titanium dioxide electrical valves
with an admissible working temperature of up to 473 K were also applied. In 1882,
Jamin and Manevrieu published a work on the properties of unipolar electrical-arc
conductivity in a mercury vapour, i.e. mercury-arc or ionic electrical valves, which
were broadly applied at the beginning of the 1930s. Fleming invented the first
electron diode in 1904, which he named the ‘kenotron’. In parallel to polyvalvular
(multi-anode) mercury-arc electrical valves, called ‘mutators’, the manufacturing
technique for ‘gasotrons’ and, a little later, ‘thyratrons’ expanded. In the 1950s
‘ignitrons’ and ‘exitrons’ were also applied on a large scale. The development of
the manufacturing technology for monocrystalline semiconductors, and then
germanium and silicon unipolar electrical valves, took place in the 1940s (uncon-
trolled discrete electrical valves—germanium and silicon diodes) and 1950s
(controlled discrete electrical valves—germanium and silicon triodes: transistors,
insulated-gate bipolar transistors (IGBT), metal-oxide-semiconductor field-effect
transistors (MOSFET), thyristors, trigistors, magnetic-field controlled thyristors
(MCT), etc—as well as germanium and silicon tetrodes: spacistors, tetristors, etc).
In addition to silicon unipolar electrical valves and silicon bipolar electrical valves
with single or two control gates—including, for example, bipolar field transistors
and photo-transistors, as well as symistors (symmetrical thyristors), photo-symistors
and bipolar trigistors (gate-turn-off (GTO) thyristors)—reverse-blocking insulated-
gate bipolar transistors (RB IGBT) have also been applied. Progress in the
recognition of the electrical properties of glassy amorphous materials is leading to
a series of promising new applications for amorphous semiconductors in macro- and
microelectronics, and thus in the field of discrete or continuous uni- and bipolar
(uni- and bilateral) electrical valves.
In 1958 S R Ovshinsky discovered the phenomenon of switching over in
amorphous semiconductors, i.e. a rapid changeover of electrical-current conductiv-
ity when the applied electrical field attains its critical value. In the case of
chalcogenide glasses, nearly the same phenomenon appeared in electrical field
with an intensity of 105 V cm−1, with the changeover time yielding a few
nanoseconds (Holmberg and Shaw 1974), while the electrical-current conductivity
can be changed approximately 106 times. The results obtained were not published
before 1970 (Fijalkowski 1988c), while references for them can be found in the
monograph (Fijalkowski 1987). However, some technological difficulties, as well as
military applications, contributed to the limitation of their initial development.
More than 50 years have passed since Ovshinsky demonstrated the ‘ovonic’
semiconductor device (the name is a contraction of ‘OVshinsky electrONIC’), which
was the first electrical valve made from amorphous semiconductors.

15-5
 Mechatronics

For 19 years, although the ovonic existed and was in operation, nobody precisely
knew the principle of its operation. There have been several well-known cases in the
history of electronics where semiconductor devices have been in common use before
their operating principles were exactly known and their theoretical basis created.
This was the case for the protoplast of the galena detector, as well as the best
polycrystalline semiconductor electrical valve in the past—the selenium rectifier.
It was not until the 1970s that three physicists—P W Anderson and J van Vleck of
the USA and N F Mott of the UK—worked out new fundamentals for the theory
and technology of amorphous semiconductors, which showed that unipolar
electrical-current conductivity could not only exist in mono- and polycrystalline
semiconductors, but also in amorphous semiconductors. The theory gained common
approval, and its creators were honoured with the Nobel Prize for physics in 1977. If
everybody had assented to a theory that had been formulated earlier (Seitz 1940;
Kittel 1956; Mott and Davis 1971), which anticipated that unipolar electrical-
current conductivity could only exist in unipolar mono- and polycrystalline
semiconductors, nobody would have an inkling about some of the technologies
that humanity could aspire to. Fortunately, a group of scientists doubted the
correctness of this theory (Mott and Davis 1971). Due to their obstinacy, it is
known today that the earlier theory was only valid for mono- and polycrystalline
semiconductors.
All the aforementioned discrete uni- and bipolar electrical valves were or are used
in MIMO ECM AC–AC, AC–DC–AC and AC–DC/DC–AC commutators with
non-integrated ECMs (Fijalkowski 1973, 1982, 1985, 1987). In the field of
macroelectronics, they are named by the author ‘polyvalvular MIMO ECM
AC–AC and/or AC–DC/DC–AC commutators’. It is obvious that in the field of
power electronics these polyvalvular MIMO ECM commutators are still called
‘static converters’. The same thing also took place in the field of MCM AC–AC and
AC–DC/DC–AC commutator dynamotors, which in the field of macroelectronics
are named by the author ‘polysectorial MIMO MCM AC–AC and/or AC–DC/
DC–AC commutators’ (Fijalkowski 1973, 1985, 1987).
Less expensive, lighter and smaller than electrical machines’ polysectorial MIMO
MCM AC–AC and/or AC–DC/DC–AC commutators, polyvalvular MIMO ECM
AC–AC and/or AC–DC/DC–AC commutators lie at the very heart of the
conversion and control of electrical energy, which in turn lie at the heart of making
that energy useful. From domestic appliances to space-faring and automotive
vehicles, the applications of macroelectronics are virtually limitless. Until now,
however, the same could not be said for access to up-to-date reference books devoted
to power electronics (see the references and bibliography).
MIMO ECM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and DC–AC–
AC commutators are ECMs of controlled semiconductor electrical valves that
directly connect each matrixer-input phase to each matrixer-output phase, especially
without any intermediate DC link. The MIMO ECM AC–AC, AC–DC–AC,
AC–DC/DC–AC, DC–DC or DC–AC–AC commutator is usually fed at the
matrixer-input side by a voltage source and it is connected to inductive and/or
capacitive loads at the matrixer-output side. The main advantage of the ECM

15-6
 Mechatronics

commutator is the absence of bulky reactive components, which are subject to

ageing and reduce the MIMO ECM AC–AC, AC–DC–AC, AC–DC/DC–AC,
DC–DC and DC–AC–AC commutator reliability. One significant advantage of
this arrangement is that the ECM commutator does not carry current to the rotor,
which eliminates the brushes and their wear-related drawbacks.
Furthermore, MIMO ECM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC
and DC–AC–AC commutators provide uni- and/or bilateral (bipolar) power flow,
nearly sinusoidal matrixer-input and matrixer-output waveforms, a controllable
matrixer-input power factor, a regenerative capability and the origin function to
generate magnitude-frequency controllable matrixer-output voltages. Therefore,
ECM AC–AC, AC–DC/DC–AC and DC–DC commutators have received
considerable attention as an advantageous alternative to an ECM AC–DC–AC
commutator acting in voltage-source inverter (VSI) topology, or an ECM DC–AC–
DC commutator acting in voltage-source chopper (VSC) ECM topology.
The author showed in 1973 in Fijalkowski (1973) that each static converter or
MIMO MCM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and DC–AC–DC
commutator of electrical machines cannot only be considered theoretically, but must
also be made practically as polyvalvular MIMO ECM AC–AC and AC–DC/DC–
AC and DC–DC commutators with built-in discrete and/or continuous unipolar
electrical valves called ‘thyristors’ (see figure 15.1), but he did not expect then that
the results of this would have any practical consequences in the future.
In addition, the author showed once more in 1982 (Fijalkowski 1982) that each
MIMO MCM AC–AC and/or AC–DC/DC–AC commutator of electrical machines
cannot only be considered theoretically, but must also be made practically as MIMO
ECM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and DC–AC–DC

Figure 15.1. MIMO ECM AC–AC, AC–DC/DC–AC and DC–DC commutators with built-in continuous
unipolar electrical valves (‘thyristors’) (Fijalkowski 1973).

15-7
 Mechatronics

commutators with built-in discrete and/or continuous bipolar, triggered and

quenched electrical valves called ‘trigistors’ (see figure 15.2).
The development of three-phase by three-phase (3 × 3) ECM AC–AC commu-
tators (called ‘direct frequency converters with nine electrical valves’) was started
by Zagorski (1969), and the development of each and every one of the m × n ECM
AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and DC–AC–DC commuta-
tors (with m × n bipolar, triggered and quenched electrical valves) was started
afterwards by Fijalkowski (1973, 1982, 1985, 1987), while the development of 3 × 3
ECM AC–AC commutators (called ‘matrix converters with nine bipolar electrical
valves’) continued soon after, as performed by Gyugyi and Pelly (1976), Cheron
et al (1985), and Alesina and Venturini (1989), who proposed the basic principles
of their operation.
The 3 × 3 ECM AC–AC commutators are superior to the up-to-now applied
ECM AC–AC commutators (called ‘thyristor cycloconverters with eighteen or
thirty-six unipolar electrical valves’) in being able to produce unity power factor (PF)

Figure 15.2. MIMO ECM AC–AC, AC–DC/DC–AC and DC–DC commutators with built-in discrete and/or
continuous bipolar, triggered and quenched electrical valves (‘trigistors’) (Fijalkowski 1982).

15-8
 Mechatronics

line current. However, compared to PWM voltage-fed AC–DC–AC ECM commu-

tators, the parts count is significantly higher. Consequently, the R&D works in this
field show two trends (see the references and bibliography).
On the one hand, there was the need for reliable discrete and continuous uni-
and/or bipolar electrical valves (electronic switches). On the other hand, the initial
modulation strategy was unsuitably in favour of ultra-modern ECM solutions,
allowing higher energy (voltage) transfer ratios and better power (current) quality.
The performance of MIMO ECM AC–AC, AC–DC–AC, AC–DC/DC–AC,
DC–DC and DC–AC–DC commutators is strictly dependent upon the pulse-width
modulation (PWM) strategy employed to control uni- and/or bipolar, triggered and
quenched electrical valves. Since the introduction of MIMO ECM commutators,
numerous modulation methods have been developed.
In Alesina and Venturini (1989) and Tenti et al (1992), off-line global duty-ratio
functions are mathematically derived, which can be modulated by a carrier-
frequency signal. This modulation method necessitates a formidable number of
complicated computations at the implementation stage.
The space vector PWM (SVPWM) for an ECM commutator explores a more
systematic approach to understand the operation of the ECM (Huber and Borojevic
1995; Casadei et al 2002). However, the SVPWM is far from spontaneous and
requires look-up tables with the a priori prepared and stored commutation patterns.
The carrier-frequency based PWM is the latest modulation strategy for ECM AC–
AC commutators (Yoon and Sul 2006; Li et al 2008, 2009a, 2009b).
The carrier-frequency based PWM employs the carrier-frequency and reference
signals and can be implemented without complicated computations and look-up
tables.
The modulation method described in Yoon and Sul (2006), however, necessitates
proper offset voltages and discontinuous carrier-frequency signals, which implies
relatively indirect understanding, and cannot be applied to ECM topologies with a
neutral connection between the matrixer-input and matrixer-output neutrals.
In Li et al (2008), a carrier-frequency-based PWM strategy, termed the direct
duty-ratio PWM (DDPWM), is presented for the three-phase to three-phase (3 × 3)
ECM AC–AC commutator, which does not require the reference offset and employs
a continuous triangular carrier-frequency waveform. The desired matrixer-output
phase voltages can be synthesised by utilising the matrixer-input phase voltages
based on the per-output-phase average concept over one commutation period, i.e.
per carrier cycle. A maximum gain of 0.866 can be obtained simply by injecting third
harmonic components into matrixer-output voltage references in the three-phase
electrical energy system.
In Li et al (2009a), the DDPWM method is extended in order to use it for the
various ECM commutator topologies, e.g. three-phase to one-phase 3 × 1 or 3 × 2
ECM AC–AC and/or AC–DC/DC–AC commutators, three-phase to two-phase
3 × 2 or 3 × 3 ECM AC–AC commutators, three-phase to three-phase 3 × 3 or 3 × 4
ECM AC–AC commutators, three-phase to five-phase 3 × 5 or 3 × 6 ECM AC–AC
commutators, and five-phase to five-phase 5 × 5 or 5 × 6 ECM AC–AC
commutators.

15-9
 Mechatronics

In Li et al (2009b), a systematic method to generate the reference signals in

various ECM commutators is presented. Before the DDPWM is used to synthesise
the matrixer-output voltages, the per-phase matrixer-output voltage references
should be given at each commutation time. The global minimum of the matrixer-
input voltage range is used to determine the matrixer-output voltage magnitude. The
feasibility and validity of the proposed DDPWM method applied in various EMC
commutators was verified with simulation and experiment results. When, over 40
years ago, the author had the idea of MCM and ECM commutators realised on
discrete and continuous uni- and/or bipolar electrical valves, only a few people
believed in the feasibility of the R&D programme started by him. It is clear that
amorphous semiconductor electrical valves will be the basis of the next great
advance in macro-, meso-, micro- and nanoelectronics.
The author intends to show why the field of power electronics is presently in a
state of crisis, and therefore historically ripe for a new basic approach. A character-
istic component of a large group of MMD electrical machines is an integral MCM
or ECM AC–AC and/or AC–DC/DC–AC commutator that is in the dynamical
systems approach for MMD electrical machines, i.e. a physical heterogeneous
continuous dynamical hypersystem is classified as a generalised commutation
matrixer (GCM) (Fijalkowski 1973, 1985).
The concept of designing fully static power electronic electrical-valve ECM AC–
DC/DC–AC commutators—named ‘ECM commutators’ by the author—for the
conversion of polyphase alternating voltages that are constant in both amplitude
and frequency, or direct voltages that are constant in magnitude, to polyphase
alternating voltages, modulated in both amplitude and frequency in order to be able
to control the torque and speed of ECM AC–AC and AC–DC/DC–AC commutator
dynamotors, is by no means new (see figures 15.3–15.10). If one studies the older
literature on static converters, one can observe that such static power electronic
electrical-valve ECM AC–DC/DC–AC commutators were described as far back as
the early 1890s.
The chronological background of MIMO ECM AC–DC commutators dates
back to the end of the 19th century, when in 1895 (date of application) the notable
Polish electrician, Charles (Karol Franciszek) Pollak (honorary doctor of Warsaw
University of Technology), was issued with UK (24,398 AD 1895) and German

Figure 15.3. The Pollak bridge rectifier with four aluminium electrical valves (Pollak 1895, 1896).

15-10
 Mechatronics

Figure 15.4. A single-phase electrical commutation matrixer (ECM) AC–AC commutator motor: 1—an EE
transformer; 2—a stator winding of the traction EM motor; 3—a rotor winding; 4—a mercury-arc controlled
electrical valve; 5—a rotary-commutator; and 6—a control system (Kern 1938).

Figure 15.5. A single-phase electrical commutation matrixer (ECM) AC–AC commutator motor: 1—a stator
winding; 2—thyratrons; and 3—rotary-commutator (Willis 1933).

Figure 15.6. A three-phase electrical commutation matrixer (ECM).

15-11
 Mechatronics

Figure 15.7. A single-phase electrical commutation matrixer (ECM) AC–AC commutator motor: 1—an EE
transformer; 2—a stator winding; and 3—thyratrons (Zavalishin and Vegner 1936).

Figure 15.8. A 6 × 2/2 × 6 mechanical commutation matrixer (ECM) AC–DC/DC–AC commutator

generator/motor (left) and a 6 × 2/2 × 6 electrical commutation matrixer (ECM) AC–DC/DC–AC
commutator generator/motor (right) (Schilling 1940).

(DRP 96564) patents for an aluminium electrical valve bridge rectifier (Pollak 1895,
1896).
Descriptions of the patents included a scheme of the Pollack bridge rectifier
(figure 15.3) as a single-phase full-wave MIMO ECM AC–DC commutator in a
single-phase double-way connection, now known as the Graetz bridge rectifier.
These descriptions were published in the Elektronische Zeitung (1897) with notation
from the editor that at that time Leo Graetz was working on rectifiers with a similar
principle of operation. However, the solution of Leo Graetz was published in 1897, a
year and a half after Karol Pollak’s first patent had been issued (Strzelecki and
Benysek 2008).

15-12
 Mechatronics

Figure 15.9. A 6 × 2/2 × 6 electrical commutation matrixer (ECM) AC–DC/DC–AC commutator generator/
motor (right) compared with a 6 × 2/2 × 6 electrical commutation matrixer (ECM) AC–DC/DC–AC rectifier/
inverter (left) (Schilling 1940).

Figure 15.10. A 2(2 × 3) electrical commutation matrixer (ECM) AC–AC commutator generator/motor
(Schilling 1940).

In 1901, the Belorussian academician Vladimir Fedorovich Mitkevich (1872–

1951) published on two-phase and three-phase half-wave ECM AC–DC commuta-
tors in a two- and/or three-phase single-way connection (now known as Mitkevich
rectifiers) (Neiman 1972). Soon after, in 1924, the Russian academician Andrei

15-13
 Mechatronics

Nikolaevich Larionov (1889–1963) published on a three-phase full-wave ECM

AC–DC commutator in a three-phase double-way connection (now known as the
Larionov bridge rectifier) (Anon 1950). Other subsequent inventions that are of
importance to ECM AC–AC and AC-DC/DC–AC commutators (see figures 15.4–
15.10) were developed in the 1930s (Kloninger 1932; Willis 1933; Alexanderson and
Mittag 1934; Kern 1938; Schilling 1940).
Improved ECM AC–AC and AC–DC/DC–AC commutators were frequently
proposed during the 1940s and 1950s (Schilling 1940; Rémy 1954). However, such
ECM AC–AC and AC–DC/DC–AC commutators did not gain wide acceptance
and most of the concepts probably did not leave the laboratory. The fact that this
was so must naturally be attributed to the characteristics of the electrical valves
available to the scientists and engineers in those days—electronic tubes and mercury-
arc electrical valves of different kinds. For technical and economic reasons, it was
not until the 1960s that scientists could solve this problem by introducing the
semiconductor-controlled rectifier (SCR) (later called the ‘thyristor’) into the power
electronics world. In France, the first ECM commutator for an MMD electrical
machine (probably the first in the world) was conceived by Kloninger (1932), while
in the UK the first complete ECM commutator based on thyratron electrical valves
was developed by Alexanderson and Mittag (1934).
AC–AC commutator motor: 1—a stator winding; 2—a rotor winding; and
3—thyratrons (Alexanderson 1931).
The author started R&D work on ECM AC–DC/DC–AC commutator
dynamotors (figure 15.11) for high-power EM propulsion systems in the early
1960s (Fijalkowski 1962, 1964).
The 1960s witnessed rapid progress in ECM high-power electronics (called
‘macroelectronics’ by the author), above all when it came to the developments
that took place in the field of bipolar, triggered and quenched electrical valves such
as trigistor, gate-turn-off (GTO) thyristors and/or bipolar transistor ECMs.
New applications for macroelectronics have thus evolved and established
applications have advanced further. This textbook briefly presents some of the
progress and new developments that occurred within the macroelectronics field over
the previous century.

Figure 15.11. A 2 × 12/12 × 2 electrical commutation matrixer (ECM) AC–DC/DC–AC commutator

dynamotor (Fijalkowski 1964).

15-14
 Mechatronics

By definition, macroelectronics is the specific application of electronics using

ECM commutators (called multi-input/multi-output (MIMO) ECM commutators by
the author) to the electrical power field, mainly through the new concept of ECM
AC–AC and AC–DC/DC–AC commutator dynamotors for industrial and traction
EM propulsion systems. It followed the developments that have taken place within
macro- and microelectronics as a whole.
The introduction of the ECM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC
and/or DC–AC–DC commutator in combination with macro- and microelectronic
based technology has been the most significant factor in the tremendous
improvement of the power handling capacity, voltage and current data, etc, of the
new-concept power-handling components, i.e. ECM commutators.
At the same time, it has been possible to fully exploit the rapid integration of
components by miniaturising the static high-power MIMO ECM AC–AC, AC–DC–
AC, AC–DC/DC–AC, DC–DC and/or DC–AC–DC commutators in MMD
electrical machines as well as low-power (small signal) electronic computer process-
ors, i.e. microprocessors in the higher- and lower-level controllers included in macro-
and microelectronics equipment (or ECM AC–AC, AC–DC–AC, AC–DC/DC–AC,
DC–DC and/or DC–AC–DC commutators, as they are usually called), and in this
way the ECM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and/or DC–AC–
DC commutator has evolved from a rather unqualified integral static high-power
ECM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and/or DC–AC–DC com-
mutator into EM propulsion process equipment that incorporates qualified electrical
integrated circuits (EIC) for control, monitoring, operation and protection.
Up to now, the use of analogue EIC techniques has predominated, but the trend
towards digital EIC techniques is both justified for cost reasons and essential in
this case—for example, with very advanced ECM AC–AC, AC–DC–AC, AC–DC/
DC–AC, DC–DC and/or DC–AC–DC commutators of the sinusoidal pulse-width-
modulated (SPWM) type.
In the 1970s, as ECM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and/or
DC–AC–DC commutators were developed and put into practical use, demand grew
for fast-switching thyristor ECMs with high performance, such as higher blocking
voltage, larger current capability and shorter turn-off times (Fijalkowski 1973).
The current trends in aerospace avionics and automotive vetronics are towards
intelligent electronically commutated (reciprocal and rotational) EM actuators,
namely electrical machines and other digital devices with much higher performance
and greater compactness, and not only at the interconnection stage.
The increasing complexity of intelligent electronically commutated actuators and
digital devices is leading to large chip sizes, and the greater degree of integration
means that the power dissipated by the working chip is also rising. Increasingly, the
tendency is to package several chips together in the form of a multichip module,
using techniques such as wire bonding, tape automated bonding (TAB) and flip chip
bonding to reduce the inter-connection length and hence the overall size.
There ought to be continuing demand for high-performance intelligent MIMO
ECM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and/or DC–AC–DC
commutators of far greater complexity than before.

15-15
 Mechatronics

Artificial neural networks and fuzzy logic (NF) are increasingly being incorpo-
rated into intelligent MIMO ECMs to provide robust and effective control in a wide
range of aerospace and automotive applications. The method is based on
many-valued-logic, which enables general principles and expert knowledge to be
used to provide control rules and procedures. The intrinsic nonlinearity and
variability of operating conditions make NF control an ideal method for this area.
The author is actively pursuing the integration of matrixery and circuitry on the
same transparent substrate as the intelligent MIMO ECM AC–AC, AC–DC–AC,
AC–DC/DC–AC, DC–DC and/or DC–AC–DC commutators and NF controllers,
respectively.
The most obvious advantage of integrated matrixery and circuitry on the same
transparent substrate is the reduction of the number of mono- and polycrystalline
‘work-horse’ components, as well as amorphous Si or GaAs super- and/or
semiconductors, fast-switching discrete and/or continuous uni- and/or bipolar
electrical valves, e.g. diodes, triodes, tetrodes, etc, namely charge-injection transistors
(CHINT), complementary metal-oxide-semiconductor (CMOS) transistors, hetero-
junction bipolar transistors (HBT), high-electron injection transistors (HEMT), metal-
semiconductor field-effect transistors (MESFET), metal-insulator-semiconductor
FETs (MISFET), metal-oxide-semiconductor FETs (MOSFET), organic FETs
(OFET), junction-gate field-effect transistor (JFET or JUGFET), bipolar quantum
interference transistors (QUIT), static induction transistors (SIT), bipolar super-
conductor-base hot electron transistors (SUPER-HET), insulated-gate bipolar
transistors (IGBT), gate-turn-off (GTO) thyristors, light-triggered-and-quenched
(LTQ) thyristors, MOS controlled thyristors (MCT), reverse-blocking insulated-
gate bipolar transistors (RB IGBT), ovonics, etc, and also external connections.
Discrete electrical valves suffer from strongly temperature-dependent performance.
The author therefore carried out a detailed programme of research into intelligent
MIMO ECM commutators using continuous electrical valves in which the electrical-
current conduction is much less temperature-dependent than in discrete electrical
valves. The external connections are also one of the dominant causes of MIMO
ECM commutator failure. They will be lighter and more compact because of their
proportion of size and mass. It is difficult to estimate the cost and reliability of these
ECM commutators, as they are not yet in manufacture.
Although the R&D and initial manufacture costs are high, the leading electronic
manufacturers will be investing significantly in this key technology and in intelligent
MIMO ECM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and/or DC–AC–
DC commutators that are capable of truly inaudible (>20 kHz) operation.
NF controllers will soon extend their application from aerospace avionics and
automotive vetronic intelligent electronically commutated EM actuators and other
digital device markets into other areas and they have every chance of completely
revolutionising the ECM universe.
Recently, attention has been focused on the bipolar reverse-conducting (RC) GTO
thyristor, MCT, MOSFET and IGBT ECM, called the RC trigistor ECM, as
fast-switching semiconductor controlled electrical-valve ECMs, which may feasibly
overcome these difficulties. The electrical-valve ECMs proved to be successful for

15-16
 Mechatronics

practical ECM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and/or DC–AC–

DC commutator applications, and important application data were obtained.
These bipolar trigistor and bipolar transistor ECMs have been used in ECM AC–
AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and/or DC–AC–DC commutators
for industrial and traction propulsion systems and have the advantage of making
them more compact, less expensive and of lighter mass than the unipolar reverse-
blocking (RB) thyristor with freewheeling diode ECMs.
The Nobel Prize in Physics 2010 was awarded jointly to Andre Geim and
Konstantin Novoselov ‘for groundbreaking experiments regarding the two-
dimensional material graphene’1 (the most promising and versatile material at
present). Graphene has attracted attention as a promising two-dimensional honey-
comb lattice composed of sp2-bonded carbon-atoms material for very advanced
ECM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and/or DC–AC–DC
commutators, including high-speed single-electron transistors (SET), which play a
central role in LSI technology.
The SET is based on the idea of controlling the number of electrons in a quantum
dot using its electrical potential. The energy consumption of SET is thought to be
very small, because the unit of information is a single electron.
Graphene is a one-atom-thick layer of carbon atoms arranged in a hexagonal
lattice. It is the building block of graphite (which is used, among others things, for
pencil tips), but graphene is a remarkable substance on its own, with a multitude of
astonishing properties that have repeatedly earned it the title ‘wonder material’.
Graphene is the thinnest material known to man, at one atom thick, and is also
incredibly strong—about 200 times stronger than steel. On top of that, graphene is
an excellent conductor of heat and electricity and has interesting light absorption
abilities. It is truly a material that could change the world, with unlimited potential
for integration in almost every industry.
Graphene has a potential advantage in that quantum dots can be formed easily
just by cutting out small regions via etching, because graphene begins with a
thickness of only a single atomic layer. However, controllable manufacture with a
precision of several nanometers still remains a challenge. For example, the SET
commutation (switching) performance at room temperature is not yet well reprodu-
cible due to the lack of precise manufacture technology.
Therefore, SET is expected to be an electrical valve (electronic switch) that will be
used widely in the near future, and much research effort has been put into it.
Today, the author is designing ECM commutators, which, compared with the
first thyristor ECM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and/or DC–
AC–DC commutators, will be designed for a 40 × higher ECM AC–AC, AC–DC–
AC, AC–DC/DC–AC, DC–DC and/or DC–AC–DC commutator power rating.
The integral ECM commutators now being conceived and developed for current
projects of ECM AC–AC and AC–DC/DC–AC commutator dynamotors belong to
the third generation. The first two generations were air-cooled and air-circulated

1
[Link]/nobel_prizes/physics/laureates/2010/

15-17
 Mechatronics

thyristor ECM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and/or DC–AC–

DC commutators, while oil has been selected as the coolant for the third generation.
The change to oil cooling was a natural consequence of the fact that ECM AC–
AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and/or DC–AC–DC commutators
show considerably superior performance. Further, oil-cooling systems have sub-
stantially lower inherent losses, which offers significant gains with the rapid
escalation of the evaluation costs.
The performance and efficiency of MIMO ECM commutators and MMD
electrical machines (ME/EM dynamotors and EE transformers) is limited by
thermal constraints in the cooling ECM channels and MMD electrical-machine
windings. These constraints can be reduced somewhat by circulating liquid through
the liquid-cooling ECM channels, or air through the air gaps in the MMD electrical
machines. However, such approaches necessitate the addition of pumping
mechanisms and are limited by the thermal conductivity of air and liquid.
Recently, a novel approach has been taken to cool MIMO ECM commutators
and MMD electrical machines, using nano-magneto-rheological (NMR) or
giant-electro-rheological (GER) fluids as the cooling media.
The aim is to take advantage of the thermal and magnetic or electrical characteristics
of NMR or GER fluids, respectively, such that not only will the NMR or GER fluid
circulate through the MIMO ECM commutators and MMD electrical machines, but
the existing thermal and magnetic or electrical fields in the MIMO ECM commutators
and MMD electrical machines will provide the energy required to pump the NMR or
GER fluid. The cooling mechanism will therefore not require additional moving
mechanical parts, MF pumps, MP compressors, or sensors. The behaviour of the
NMR or GER fluid will provide the method to pump and cool the MIMO ECM
commutators and MMD electrical machines (thus increasing the MIMO ECM
commutators and MMD electrical machines’ potential energy density and efficiency).
It is generally the need for a high short-circuit capability and low losses rather
than the performance during normal operation that determines, e.g. the choice of
trigistor cathode area for the ECM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–
DC and/or DC–AC–DC commutator’s electrical valves.
A high short-circuit current capability allows one to optimise the armature-
winding reactance concerning reactive power consumption, MMD electrical
machine costs, and losses.
Today, there are no plans for MIMO ECM AC–AC, AC–DC–AC, AC–DC/DC–
AC, DC–DC and/or DC–AC–DC commutators where the data suggest that parallel
connection of electrical valves will be necessary.
The bipolar trigistors with a cathode area exceeding 100 cm2 now available give a
single chain of trigistors connected in series. This results in high reliability and
minimal complexity, costs and losses. The triggering pulses for MIMO ECM
AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and/or DC–AC–DC
commutator electrical valves, where a large number of electrical valves are
connected in series to obtain a sufficient voltage rating, must be transmitted by
non-galvanic means. For instance, for this purpose, one uses light guides, which
transmit light pulses for the electrical-valves’ electronic control unit (ECU) at earth

15-18
 Mechatronics

potential to the individual electrical valves. Because the electrical valves require an
electrical triggering pulse, the conversion from light to electrical current pulses is
performed in an ECU by each electrical valve (electronic switch).

15.3 Physical and mathematical models of a generalised

ECM AC–AC and/or AC–DC/DC–AC commutator
15.3.1 Introduction
In the existing theory of ECM AC–AC and AC–DC/DC–AC commutator dyna-
motors, both the rotary and linear roles, which in these MMD electrical machines are
played by the sliding-contact mechano-electrical commutation matrixer (MCM) AC–
AC and/or AC–DC/DC–AC commutator, are completely devalued. Up to now, the
theory for a generalised electrical commutation matrixer (ECM) commutator of
MMD electrical machines had also not been worked out and its mathematical model
in the author’s dynamical systems approach for MMD electrical machines and
matrix notation had not been formulated. A complex variety of phenomena taking
place with the changeover of the sense of current direction polarity in individual
commutating sections, i.e. the phase coils of a polyphase armature winding short
circuited through the carbon brush, i.e. the matrixer-input primary collectors or a
rotary sliding-contact MCM AC–AC and/or AC–DC/DC–AC commutator, have
merely been considered (Kostenko and Piotrovsky 1968).
As mentioned above, up to now, the theory of the MCMs of MCM AC–AC and AC–
DC/DC–AC commutator dynamotors, both rotary and linear, has not been worked out.
The mathematical model formulation to fill this theoretical gap was first carried out by the
author (Fijalkowski 1973, 1982, 1985, 1987, 1988) and will be presented below.
The generalised ECMs of MMD electrical machines (figure 15.12, both rotary
and linear), sliding-contact MCM AC–AC and/or AC–DC/DC–AC commutators,
and solid-state semi- or superconductor electrical-valve ECM AC–AC and/or AC–
DC/DC–AC commutators always act as static frequency and phase-number
changers, converting one form of electrical energy of (e.g. with the voltage holor
UiP , the frequency f P and the phase number mP) into another form of electrical
energy (e.g. with the voltage holor Uis , the frequency f s and the phase number ms)
without losses. At the same time, they take into account the properties of the uni- or
bipolar current-conduction of electrical valves.
For instance, sliding contacts (copper-bar–carbon-brush contacts) form the
components of a sliding-contact MCM AC–AC and/or AC–DC/DC–AC commu-
tator, or solid-state semi- and/or superconductor electrical valves, both monocrystal
and amorphous, form the components of an MIMO ECM.
In the simplest way, the generalised ECM AC–AC and/or AC–DC/DC–AC
commutator of MMD electrical machines, both rotary sliding-contact MCM
AC–AC and/or AC–DC/DC–AC commutators and static ECM AC–AC and/or
AC–DC/DC–AC commutators, can be presented as generalised ECM AC–AC
and/or AC–DC/DC–AC commutators (figure 15.12) consisting of crossed-over
but non-galvanically connected (by means of electrical valves) current-conduct-
ing matrixer-input and matrixer-output conductive collectors.

15-19
 Mechatronics

In this generalised ECM, the horizontal collectors (matrixer rows) A, B, C, …, N

are called the ‘matrixer-input’ or ‘primary collectors’, and the vertical collectors
(matrixer columns) a, b, c, …, n are called the ‘matrixer-output’ or ‘secondary
collectors’. Discrimination of the matrixer-input or primary collectors is relative,
since at an evident moment of time electrical energy can be both absorbed and
recovered through the same collectors. This generalised ECM AC–AC and/or
AC–DC/DC–AC commutator may only be connected with two other electro-
energetic systems, namely the armature winding of an MMD electrical machine
and the AC or DC supply line, by means of self-collectors.
Considering an ECM AC–AC and/or AC–DC/DC–AC commutator of MMD
electrical machines as the ME/EM dynamical hypersystem, it is necessary to take
into account its cooperation with the armature winding of an MMD electrical
machine by means of either voltage sources or current-holor sources led to the
matrixer-input and matrixer-output conductive collectors.
The generalised ECM AC–AC and/or AC–DC/DC–AC commutator of MMD
electrical machines is the MCM or ECM with the matrixer-input or primary
collectors A, B, C, …, N connected to the machine’s polyphase armature winding
terminals, which are connected as poly-star or polygon (e.g. wye or delta), and the
matrixer-output or secondary conductive collectors a, b, c, …, n connected to the
AC or DC supply line. The individual collectors of this ECM are connected to each
other by means of uni- and/or bipolar electrical valves.
The impedance of such a real electrical valve is nonlinear, but the impedance of
such an ideal electrical valve in the turn-on state is equal to zero, while in the turn-off
state it is equal to infinity.
The most obvious difference between a generalised ECM and other types of
MCMs and ECMs is the electrical-valve action. In contrast to digital ECMs, the
electrical valves do not indicate a logic level. Control is effected by determining the
times at which the electrical valves should turn-on. Whether there is just one
electrical valve or a large group of them, there is a complexity limit: if a converter
has P primary collectors (inputs) and s secondary conductive collectors (outputs),
even the densest possible collection of electrical valves would have a single electrical
valve between each input conductive collector (column line) and each output
conductive collector (row line). The P × s electrical valves in the ECM AC–AC
and/or AC–DC/DC–AC commutator can be arranged according to their
connections.
The pattern suggests an ECM, as shown in figure 15.12. MIMO ECM AC–AC,
AC–DC/DC–AC and/or DC–DC commutators fall into two classes.
• Direct ECM AC–AC, AC–DC/DC–AC and/or DC–DC commutators. In these
ECMs, energy-storage components are only connected to the ECM at the
primary (input) and secondary (output) of its collectors (ports). The electrical
energy-storage components effectively become part of the energy source or
load. For instance, an ECM AC–DC commutator (rectifier) with an external
low-pass filter is an example of a direct ECM AC–DC commutator. In the
literature, only direct 3 × 3 ECM AC–AC commutators are called ‘matrix
converters’.

15-20
 Mechatronics

Figure 15.12. Generalised ECM AC–AC, AC–DC/DC–AC and/or DC–DC commutator (Fijalkowski 1987).

• Indirect ECM AC–DC–AC and/or DC–AC–DC commutators, also termed

embedded electrical commutators. These ECM AC–DC–AC and/or DC–AC–
DC commutators, for instance, the polarity-reverser example, have energy-
storage components connected within the ECM structure. There are usually a
very small number of energy-storage components. Indirect ECM AC–DC–
AC and/or DC–AC–DC commutators are most commonly analysed as a
cascade connection of direct ECM AC–AC and/or AC–DC/DC–AC com-
mutators with the energy-storage in between.

15-21
 Mechatronics

From such a dynamical systems approach for MMD electrical machines with a
generalised MIMO ECM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and/or
DC–AC–DC commutator (Fijalkowski 1973, 1982, 1985, 1987, 1988), in referring
to AC commutatorless MMD electrical machines with which this generalised
MIMO ECM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and/or DC–
AC–DC commutator cooperates and establishes an ME/EM simple dynamical
hypersystem, two different mathematical models may be formulated, and therefore,
two different types of matrix notation can be obtained for the generalised MIMO
ECM commutator’s holor equations.
Taking into account ECM excitations (forcings) in the form of the voltage-source
holor matrix, and taking as the ECM responses the external mesh-current holor
matrix, the mathematical model of a generalised MIMO ECM AC–AC, AC–DC–
AC, AC–DC/DC–AC, DC–DC and/or DC–AC–DC commutator of MMD elec-
trical machines has been formulated in the dynamical systems approach. The matrix
notation is as follows (figure 15.13):
T
Ei = EiPT EisT
(15.1)
T T T
= EiANT EiBAT EiCBT ⋯ EiN (N −1) EianT EibaT EicbT ⋯ Ein(n−1) .

are the responses of the multi-port MIMO generalised MIMO ECM, the column
hypermatrices or the hypervectors of the outer-mesh current holors,
T
I j = I PjT I sjT
(15.2)
j T j T j T T T T T T
= I AN I BA ICB ⋯ I Nj (N −1) I anjT I baj Icbj ⋯ I nj(n−1) .

A set of the differential equations of dynamics, i.e. the mathematical model of the
multi-port MIMO generalised ECM as the electrical heterogeneous continuous
dynamical hypersystems may be formulated in the dynamical systems approach and
holor notation as follows:
Ei = ZijI j , (15.3)

where i × j hypermatrix Zij has a dimension of the impedance holor.

To denote the state variables associated with the primary (input) conductive
collectors by an index P, and those associated with the secondary (output)
conductive collectors by an index s, a set of the differential equations of dynamics,
i.e. the mathematical model of the multi-port generalised MIMO ECM may be
expressed as follows:

EiP ZijPP ZijPs IPj

= , (15.4a )
Ei s ZijsP Zijss Isj

or in detail as

15-22
 Mechatronics

EiAB
EiBC
C (N −1)
Ei
⋮
EiNA
=
Eiab
Eibc
c(n −1)
Ei
⋮
Eina

ZijAA ZijAB ZijAC ZijAN ZijAa ZijAb ZijAc ZijAn

ZijBA ZijBB ZijBC ⋯ ZijBN ZijBa ZijBb ZijBc ⋯ ZijBn
ZijCA ZijCA ZijCC ZijCN ZijCa ZijCb ZijCc ZijCn
⋮ ⋱ ⋮ ⋮ ⋱ ⋮
ZijNA ZijNB ZijNC ⋯ ZijNN ZijNa ZijNb ZijNc ⋯ ZijNn

ZijaA ZijaB ZijaC ZijaN Zijaa Zijab Zijac Zijan (15.4b )

ZijbA ZijbB ZijbC ⋯ ZijbN Zijba Zijbb Zijbc ⋯ Zijbn

ZijcA ZijcB ZijcC ZijcN Zijca Zijcb Zijcc Zijcn
⋮ ⋱ ⋮ ⋮ ⋱ ⋮
ZijnA ZijnB ZijnC ⋯ ZijcN Zijna Zijnb Zijnc ⋯ Zijnn

j
I AB
j
I BA
j
ICB
⋮
j
I N (N −1)
.
j
I an
j
I ba
j
Icb
⋮
j
I n(n −1)

The coefficients
∂Ek(k−1)
Zkl = I j (j −1) (j ≠ l ) (15.5)
∂I l (l −1)
have a dimension of the impedance holor.

15-23
 Mechatronics

Figure 15.13. A multi-port generalised MIMO ECM with voltage-source holors (physical model I)
(Fijalkowski 1987).

To adapt a mathematical model of the multi-port generalised MIMO ECM to the

theory of the electrical heterogeneous continuous dynamical hypersystems’
demands, one refers to a simplified and very general mathematical model of the
multi-port generalised MIMO ECM, derived from the physical model shown in
figure 15.12.
In order to fulfil the demands of physical energy conversion of one kind of
electrical energy with the voltage-source holor EiP , the current holor IPj , the
frequency fP and the number of phases mP of a primary electrical heterogeneous

15-24
 Mechatronics

continuous dynamical hypersystem of the multi-port generalised MIMO ECM into

another kind of electrical energy with the voltage-source holor Eis , the current holor
Isj , the frequency fs and the number of phases ms of a secondary electrical continuous
dynamical system of the multi-port generalised physical commutation matrixer—as
well as the free passage of electrical energy between these dynamical hypersystems—
they ought to be connected by means of the multi-port generalised MIMO ECM’s
uni- and/or bipolar passive and active electrical components and devices.
A physical model of the multi-port generalised MIMO ECM (figure 15.14) fulfils
the aforementioned demands. Uni- and bipolar passive and active electrical
components and devices connect both electrical heterogeneous continuous dynam-
ical hypersystems of the multi-port generalised MIMO ECM and they make free
passage of electrical energy between them possible.
In the general case, this MIMO ECM can connect many electrical heterogeneous
continuous dynamical hypersystems. In this textbook we consider a case where there
are only two electrical heterogeneous continuous dynamical hypersystems—the
primary and secondary. This is sufficient to understand the multi-port MIMO ECM
concept. Both electrical heterogeneous continuous dynamical hypersystems of the
multi -port generalised MIMO ECM may be considered to be autonomous, i.e.
when disconnected from the multi-port generalised MIMO ECM, they have
determined potentials on their terminals. They have voltage-source holors, cur-
rent-source holors and electrical components (e.g. inductances, capacitances and
resistances), as well as discrete or continuous uni- and/or bipolar physical valves.
Assuming that the excitations of a multi-port generalised MIMO ECM are given in
the form of the column hypermatrices or hypervectors of current-source holors
flowing in the primary (input) and secondary (output) conductive collectors (figure
15.14):
T T
I j = I PjT I sjT = I AjT I BjT ICjT ⋯ I Nj T I ajT I bjT IcjT ⋯ I njT . (15.6)

are the responses of the multi-port MIMO ECM, the column hypermatrices or the
hypervectors of the node voltage holors
T T
Vi = ViPT VisT = ViAT ViBT ViT ⋯ ViNTViaT VibT VicT ⋯ VinT . (15.7)

A set of the differential equations of dynamics, i.e. the mathematical model of the
multi-port generalised MIMO ECM as the electrical heterogeneous continuous
dynamical hypersystems, may be formulated in the dynamical systems approach and
matrix notation as follows:
I j = Y jiVi , (15.8)
where n × m hypermatrix Y ji has a dimension of the admittance holor.
Node voltage holors, consisting of electrical components of the hypermatrix Vi ,
are measured with regard to the reference node, which is located on the outside of
the multi-port generalised MIMO ECM. For example, in a star connection of the

15-25
 Mechatronics

multi-phase generalised electrical heterogeneous continuous dynamical hypersys-

tem, the reference nodes may be a null point of the multi-phase star connection.
To denote the state variables associated with the primary (input) collectors by an
index P, and those associated with the secondary (output) collectors by an index s,
the set of differential equations of dynamics, i.e. the mathematical model of the
multi-port generalised MIMO ECM, may be expressed as follows:
ji
IPj YPP YPsji ViP
= , (15.9a )
Isj YsPji Yssji Vis

or in detail as
ji ji ji ji ji ji ji ji
IAj YAA YAB YAC YAN YAa YAb YAc YAn
ji ji
IBj
ji
YBA ji
YBB ji
YBC ji
⋯ YBN YBa YBb ji
YBcji ⋯ YBn
ICj
ji
YCA ji
YCA ji
YCC ji
YCN YCaji YCbji YCcji YCnji
⋮ ⋮ ⋱ ⋮ ⋮ ⋱ ⋮
ji ji ji ji ji ji ji ji
INj YNA YNB YNC ⋯ YNN YNa YNb YNc ⋯ YNn
=
Iaj ji
YaA ji
YaB ji
YaC ji
YaN Yaaji Yabji Yacji Yanji
Ibj ji
YbA ji
YbB ji
YbC ji
⋯ YbN Ybaji Ybbji Ybcji ⋯ Ybnji
Icj YcAji YcBji YcCji ji
YcN Ycaji Ycbji Yccji Ycnji
⋮ ⋮ ⋱ ⋮ ⋮ ⋱ ⋮
Inj ji ji ji
YnA YaB YnC ⋯ YcN
ji
Yna Ynb Ync ⋯ Ynnji
ji ji ji

(15.9b)
ViA
ViB
ViC
⋮
ViN
.
Via
Vib
Vic
⋮
Vin

The coefficients
∂j k
Y kl = Ui = 0 (i ≠ l ) (15.9c )
∂Ul
have a dimension of the admittance holor.

15-26
 Mechatronics

Figure 15.14. A multi-port generalised MIMO ECM with current-source holors (physical model I)
(Fijalkowski 1987).

Summing up, a hybrid mathematical model of the multi-port generalised MIMO

ECM may be written as follows:

15-27
Mechatronics

15-28
 Mechatronics

kl
∂Ek(k −1) ∂J k
Zkl = (j ≠ l ) and Y = (i ≠ l ) (15.11)
∂I l (l −1) I j (j −1)
∂Vl
Vi = 0

have dimensions of the impedance or admittance holors, respectively.

Similar to the mathematical models of physical heterogeneous continuous
dynamical hypersystems of generalized MIMO ECMs shown in figures 15.13
and 15.14, consisting of a set of differential equations of dynamics (15.4) and
(15.9), is the possible formulation of the mathematical model of physically
continuous dynamical systems of generalized MIMO ECMs (figure 15.14) acting
as voltage-source ECM commutators (matrix converters) with an arbitrary
number N of primary collectors (inputs), equal to the number m of physical
commutation matrixer phases, and an arbitrary number n of secondary collec-
tors (outputs).
Assuming that the voltage-source holors applied across the primary collectors
(inputs) A, B, C,…, N of a generalised MIMO ECM commutator (matrix converter)
are given in the form of the voltage-source holor column matrix or vector (input
holors), then
T
EiP = EiAT EiBT EiCT ⋯ EiN −1T EiNT (15.12)

is the primary electrical state variables (inputs), and a voltage-source holor, applied
across the secondary collectors (outputs) a, b, c,…, n of a generalised MIMO ECM
commutator (matrix converter), in the form of the voltage-source holor matrix or
vector (output holors)
T
Esj = EajT EbjT EcjT ⋯ Enj−1T EnjT (15.13)

is the secondary electrical state variables (output holors).

A mathematical model of the electrical heterogeneous continuous dynamical
hypersystem of a generalised MIMO ECM commutator (matrix converter) acting
as a voltage-source ECM commutator, making the outputs dependent on the
inputs, in the dynamical systems approach and matrix notation may be written in
the form:

Esj = C sPjiEiP , (15.14)

15-29
 Mechatronics

or in detail as

ji
C aA ji
C aB ji
C aC C aji(N −1) C aN
ji
Eaj
Ebn
ji
C bA ji
C bB ji
C bC ⋯ C bji(N −1) C bN
ji

Eci CcAji CcBji CcCji Ccji(N −1) CcN

ji
=
⋮ ⋮ ⋱ ⋮
Enj−1 C(nji−1)A C(nji−1)B C(nji−1)C C(nji−1)(N −1) C(nji−1)N
Enj ⋯
ji
C nA ji
C nB ji
C nC C nji(N −1) C nN
ji
. (15.15)
Ei A
EiB
EiC
⋮
Ei N −1
Ei N

However, a mathematical model of the electrical continuous dynamical system of a

generalised MIMO ECM commutator (matrix converter) acting as a voltage-source
ECM commutator, making the inputs dependent on the outputs, may be written in
the form:

EiP = CijPs E sj , (15.16)

or in detail as

Ei A CijAa CijAb CijAc CijA(n−1) CijAn

EiB CijBa CijBb CijBc ⋯ C IJB(n−1) C IJBn
FiC C IJCa C IJCb C IJCc C IJC (n−1) C IJCn
=
⋮ ⋮ ⋱ ⋮
Ei N −1 C IJ(N −1)a C IJ(N −1)b C IJ(N −1)c C IJ(N −1)(n−1) C IJ(N −1)n
Ei N ⋯
C IJNa C IJNb C IJNc C IJN (n−1) VIJNn (15.17)
Eaj
Ebj
Ecj
,
⋮
Enj−1
Enj

15-30
 Mechatronics

JI
where C kL and CijKl are the voltage-source coefficients of the commutation nodes kL
and Kl, respectively, which may be expressed as follows:

⎧ Δ j L
⎪ + 1 → E k = + Ei
∂Ekj ⎪
JI
C kL = =⎨ 0 → Δ
E kj = 0 .
∂EiL ⎪
E iI = 0
⎪ Δ j L
⎩ − 1 → E k = − Ei

and

⎧ Δ K j
⎪ − 1 → Ei = − El
∂EiK ⎪
CijKl = =⎨ 0 → Δ
EiK = 0
∂El j Ei j = 0
⎪
⎪ Δ K j
⎩ + 1 → Ei = + El .

In the case of the generalized MIMO ECM with bipolar (symmetrical) electrical
ji
valves, the voltage-source coefficients of the commutation nodes C kL = ±1 i CijKl = ∓1
represent an ideal electrical-valve whose impedance equals zero, and which is
incorporated between the primary collector L and the secondary collector k, or
between the secondary collector l and the primary collector K, which corresponds to
a turn-on state of the electrical valves with a sense of the current flow direction
(polarisation) that is dependent on a sign of these coefficients. However, for voltage-
ji
source coefficients of the commutation nodes C kL = 0 and CijKl = 0, an ideal
electrical-valve impedance equals infinity, which corresponds to a turn-off state of
the electrical valves.
Analogically, a mathematical model of the physical heterogeneous continuous
dynamical system of a generalised MIMO ECM AC–AC and/or AC–DC/DC–AC
commutator (matrix converter) shown in figure 15.14, as a current-source ECM
commutator with an arbitrary number N of primary collectors (inputs), equals a
number m of physical phases and an arbitrary number n of secondary collectors
(outputs).
Assuming that an energy-transfer holor’s sources, flowing in the primary
collectors (inputs) A, B, C, …, N of an MIMO ECM commutator (matrix
converter), are given in the form of the current-source holor column matrix or
vector (input holors), then
T
IiP = IiAT IiBT IiCT ⋯ IiN −1T IiNT (15.18)

are the primary electrical state variables (inputs), and the secondary current-source
holors of a generalised MIMO ECM commutator, flowing in the secondary

15-31
 Mechatronics

collectors (outputs) a, b, c, …, n of an MIMO ECM, in the form of the secondary

current-source holors (output holors)
T
Isj = I ajT I bjT IcjT ⋯ I nj−1T I njT (15.19)

are the secondary physical state variables (inputs).

A mathematical model of the physical heterogeneous continuous dynamical
hypersystem of a generalized MIMO ECM commutator (matrix converter) acting as
a current-source ECM commutator, making the outputs dependent on the inputs, in
the dynamical systems approach and a matrix notation may be written in the form:

Isj = C sPjiIiP , (15.20)

or in detail as

Iaj
Ibj
Icj
⋮
Inj−1
Inj

ji ji ji C aji(N −1) C aN
ji (15.21)
C aA C aB C aC IiA
ji
C bA ji
C bB ji
C bC ⋯ C bji(N −1) C bN
ji
IiB
ji
CcA CcBji ji
CcC Ccji(N −1) CcN
ji
IiC
.
⋮ ⋱ ⋮ ⋮
C(nji−1)A C(nji−1)B C(nji−1)C C(nji−1)(N −1) C(nji−1)N IiN −1
⋯ IiN
ji
C nA ji
C nB ji
C nC C nji(N −1) C nN
ji

However, the mathematical model of an electrical continuous dynamical system of

the generalised MIMO ECM commutator (matrix converter) shown in figure 15.15,
acting as a current-source ECM commutator, making the inputs dependent on the
outputs, in the dynamical systems approach and matrix notation may be expressed
in the form:
IiP = CijPsIsj , (15.22)

or in detail as

15-32
 Mechatronics

A(n −1)
C ijAa C ijAb C ijAc C ij C ijAn
IiA
B (n −1)
C ijBa C Bb C ijBc ⋯ C ij C ijBn
IiB
IiC C ijCa C ijCb C ijCc C (n −1)
C ij C ijCn
=
⋮ ⋮ ⋱ ⋮
IiN −1 (N −1)a (N −1)b (N −1)c (N −1)(n −1) (N −1)n
C ij C ij C ij C ij C ij
IiN ⋯
N (n −1)
C ijNa C ijNb C ijNc C ij C ijNn (15.23)
j
Ia
j
Ib
j
Ic
,
⋮
j
I n −1
j
In

ji
where C kL i CijKl are the current-source coefficients of the commutation nodes kL and
Kl, respectively.
To recapitulate, it may be affirmed that an action of a generalised MIMO ECM
commutator as a current-source ECM commutator (figure 15.15) may be presented
by means of its mathematical model, consisting of equations of the unholonomic
constraints of current-source holors in matrix notation ((15.24) and (15.25)). For
thus-formulated equations of the unholonomic constraints of current-source holors
in matrix notation, the isochronic variations of the generalised coordinates are not
already independent of each other, and that is why it is effective to apply the
dynamical systems approach in the methodology for formulating the mathematical
models of MIMO ECM commutators (matrix converters) in the matrix notation of
the Euler–Lagrange second-order differential equation of dynamics. In the case of
an ideal bipolar electrical valve, the voltage-holor coefficient of a commutating
junction, moreover, concerns the electrical-valve resistance connected between the
primary collector P and the secondary collector s. In this case, the electrical-valve
resistance is equal to zero, which corresponds to the turn-on state of an electrical
valve with the sense of the current direction dependent upon the sign of this voltage-
source coefficient. On the other hand, the voltage-holor coefficient of a commutating
junction regards the electrical-valve resistance as equal to infinity, which corre-
sponds to the turn-off state (blocking) of an electrical valve.
As future spacecraft grow in size, multiple electrical energy sources may be
required to supply the electrical distribution network. To supply the various
electrical loads, these energy sources must be combined to derive an ECM DC–
AC–DC commutator with several different load voltages. Such an ECM is
important from a reliability standpoint, since the electrical loads can still be supplied
even if some of the electrical energy sources fail.

15-33
 Mechatronics

Figure 15.15. Physical model of the generalised MIMO ECM AC–AC and/or AC–DC/DC–AC commutator
of an AC and/or DC MMD electrical machine (Fijalkowski 1987).

One method of implementing such an ECM DC–AC–DC commutator, where

several DC–AC inverters are supplied from different electrical energy sources to
drive a common AC bus, may also be indicated by figure 15.15. Because of their high
output impedance, these DC–AC inverters share the electrical load in a predictable
manner, and they provide an ECM DC–AC–DC commutator that can be designed
to operate in a short circuit and thus provide a fault-tolerant system. These DC–AC
inverters are operated synchronously through a common microcontroller unit
(MCU) and regulated by monitoring one of the AC–DC rectifier outputs. The
output DC voltage is regulated in the usual manner by variation of the operating
frequency. The use of a common AC bus appears to be very advantageous for
those applications where multiple ECM outputs must be driven from multiple DC
inputs. This type of ECM DC–AC–DC commutator provides the high efficiency of
a resonant circuit, and at higher power levels it can still be operated in the 25 kHz
range to minimise the mass of the magnet. For example, in a personal computer

15-34
 Mechatronics

(PC) energy supply, there are commonly five (5) separate DC loads, and the ECM
is 2 × 10.
A small number of applied ECM commutators have more than 36 electrical
valves, and most designs use fewer than 12. An ECM is a means to organise ECM
AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and/or DC–AC–AC commuta-
tors for a given application.
It also helps to focus efforts into three major spheres. Each of these spheres must
be addressed effectively in order to create a useful macro- and microelectronic
continuous dynamical system.
• The hardware sphere—create an ECM; this involves the selection of appro-
priate super- and/or semiconductor electrical valves and the auxiliary
elements that drive and protect them.
• The software sphere—activate the ECM to achieve the desired AC–AC, AC–
DC–AC, AC–DC/DC–AC, DC–DC and/or DC–AC–DC conversion; all
operational decisions are implemented by adjusting the electrical-valve
timing.
• The interface sphere—insert energy-storage components to provide the low-
pass filters or intermediate energy-storage necessary to meet the application
requirements; unlike with most passive filter applications, lossless active filters
with simple structures are required.

15.3.2 Hybrid electrical commutation matrixer commutators

The hybrid ECM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and/or DC–
AC–DC commutator, realised in continuous uni- and/or bipolar electrical valves,
establishes a mediate component amidst the polyvalvular ECM AC–AC, AC–DC–
AC, AC–DC/DC–AC, DC–DC and/or DC–AC–DC commutators, also realised on
continuous uni- and/or bipolar electrical valves. The reactive components of ECM,
and so primary (matrixer input) and secondary (matrixer output) collectors, i.e. the
connections between the continuous uni- and/or bipolar electrical valves, are made
on ceramic or glassy bases using vacuum evaporation, printing or injection casting
techniques.
The active elements of ECMs, and so continuous uni- and/or bipolar (bilateral)
electrical valves—diodes, triodes (transistors, thyristors, trigistors, etc) or tetrodes
(spacistors, tetristors, etc)—without packages are soldered or pressed into the ECM
manufactured in this manner.
The hybrid ECM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and/or
DC–AC–DC commutator has a number of essential advantages, especially when
it is made using the thick-film technique:
• relatively short design and production preparation time;
• moderate cost of production, even in small series;
• possibility of integrating continuous uni- and/or bipolar electrical valves and
protection circuits;

15-35
 Mechatronics

• high overcurrent capability;

• high over-voltage capability;
• high output power.

By way of example, rotary ECM AC–DC/DC–AC commutators, which are

designed to be built into the rotor of the ECM AC–DC/DC–AC commutator
dynamotor and work in the most severe conditions, are now most frequently made
using the hybrid technique.

15.3.3 Monolithic electrical commutation matrixer commutators

The monolithic ECM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and/or
DC–AC–DC commutator, realised on continuous uni- and/or bipolar electrical
valves, is an intelligent MIMO ECM commutator in which all the active elements
(continuous uni- and/or bipolar electrical valves) are made in a single wafer of the
crystalline or amorphous material in the shape of single chips. Thus, the technique
has assured high packing density for the active components.
According to the number of individual single active components integrated on the
single wafer of crystalline or amorphous material, ECMs with an SSI (comprising
several to 36 continuous uni- and/or bipolar electrical valves) and ECMs with an
MSI (from 24 to hundreds of continuous uni- and/or bipolar electrical valves) can be
distinguished.
The monolithic intelligent MIMO ECM AC–AC, AC–DC–AC, AC–DC/DC–
AC, DC–DC and/or DC–AC–DC commutator, realised on mono-crystalline semi-
conductor continuous uni- and/or bipolar electrical valves, used in the static
converter technique and ECM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC
and DC–AC–DC commutator dynamotors, show far more limitations, namely:
• difficulties with the integration of continuous uni- and/or bipolar electrical
valves;
• limited resistance against over-voltages;
• low cost only for series production on a large scale (100 000 ECMs per
annum).

Of late, in hybrid and monolithic ECM AC–AC, AC–DC–AC, AC–DC/DC–

AC, DC–DC and DC–AC–DC commutators, two basic types of ECMs have been
used: unipolar and/or bipolar.
The author announced the expansion of the pioneering family of compact
intelligent MIMO ECM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and
DC–AC–DC commutators with integral press-pack ECMs.
These novel macroelectronic devices have recently shown a continuous electrical
valve voltage rating of 8 kV and a current rating of 2.5 kA. The compact matrixer-
built package design gives the new macroelectronic device unrivalled megawatt
power density, with the advantages of high reliability and efficiency.
The unique matrixer-built hermetic ceramic package and internal construction
provide double-sided cooling for the electrical valves, resulting in superior thermal

15-36
 Mechatronics

power cycling and power ratings when compared to alternative module

technologies.
The compact matrixer-built, fully-hermetic package design is mechanically
compatible with cooling systems, including total immersion systems such as oil,
currently employed for 3 or 4 kA GTO thyristors, this feature allowing for either
retrofitting or design continuation of existing systems.
The high current rating allows the conversion of old bipolar static converter
drives to state-of-the-art ECM technology, without the need to replace the basic
mechanical structure. Adopting a refurbishment approach to component obsoles-
cence is both economical and environmentally friendly, as it reduces the consump-
tion of materials and energy compared to the replacement of equipment.
The high power density of compact intelligent MIMO ECM AC–AC, AC–DC–
AC, AC–DC/DC–AC, DC–DC and DC–AC–DC commutators with integral
presspack ECMs, and their unique matrixer-built mechanical construction, make
them ideal for new designs—not only in MCM AC–AC and AC–DC/DC–AC
commutator dynamotors, but also in green power applications, such as the matrix
converters normally used for wind and/or wave power generation in the high
megawatt range.
The novel ECM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and DC–
AC–DC commutator dynamotors with intelligent self-protecting compact MIMO
ECMs offer size and mass reduction, together with improved reliability, among
others benefits.
ECM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and DC–AC–DC
commutator dynamotors are highly integrated and provide key functions, including
overcurrent, over-temperature and short-circuit protection, and supply ultra-violet
(UV) lock functions.
The novel ECM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and DC–
AC–DC commutator dynamotors are high-performance MMD electrical machines
with built-in current-sense electrical valves in the ECM stage, which is housed in a
thin and compact matrixer-built package, thanks to the insulated base-plate with
high thermal conductivity. They are the first ECMs capable of truly inaudible
(>20 kHz) operation.
The high reliability and exceptional thermal ruggedness are also well suited to the
demands of high-power ECM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC
and DC–AC–DC commutator dynamotors for industrial and traction propulsion
mechatronical control systems, e.g. for both high rail and main line applications
such as high-speed electrical multiple units (EMU). For instance, full-diffusion
technology with loose mono- and/or poly-silicon wafers may be used to manufacture
MOS-controlled thyristor (MCT) type intelligent MIMO ECM commutators. Up to
now, neutron-doped silicon of the so-called ‘float-zone’ type has mainly been used.
Neutron doping means that the silicon rod is irradiated with neutrons in a nuclear
reactor prior to being sliced into wafers. A number of silicon atoms are then
converted into phosphorus atoms and the material becomes weakly n-doped.
The major advantage, however, is that after heat treatment the silicon rod will
have a very homogeneous resistivity. This is valid for achieving high-voltage

15-37
 Mechatronics

capability for the ECMs. To obtain an acceptable yield in the ECM manufacture, i.e.
the number of MCT-type ECMs fulfilling the specifications in relation to the number
of MCT-type ECMs manufactured, the environment must be extremely clean.
Diffusion technology means that the silicon wafer is coated with gallium,
aluminium or phosphorus. These substances are then allowed to diffuse into the
silicon wafer in diffusion furnaces under high temperature. In certain ECM
manufacturing stages, only parts of the silicon wafer surface are coated for
subsequent diffusion. For this purpose, photolithography is utilised. After comple-
tion of the diffusion and the coating of the aluminium contacts, the silicon wafers are
sliced into ECM chips with the aid of high-power lasers. Depending on the silicon
wafer area of the chips, the intelligent MIMO ECMs, e.g. for AC–AC commutator
dynamotors (generators/motors), incorporate 3 × 3 or 3 × 4 ECMs, and AC–DC/
DC–AC commutator dynamotors incorporate 3 × 2/2 × 3 ECMs. By controlling the
lifetime of the charge carriers, it is possible to determine the position of the MCT-
type ECM on the recovery charge–minimum on-state voltage drop curve. For this
purpose, gold irradiation is applied, which provides considerably closer tolerances.
With electron irradiation, the ECM components are bombarded with high-energy
electrons (>10 MeV) from electron guns.
On the basis of the large volume of ECM commutators (static converter) that
were manufactured in the 20th century (primarily diode- or thyristor-type MIMO
ECMs for AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and/or DC–AC–DC
commutators), it was possible to introduce more reliable dimensioning criteria. In
combination with closer MIMO ECM manufacturing tolerances, above all, this
substantially raised the maximum reverse voltage/off-state voltage. This lowered the
on-state voltage drop, and as a consequence reduced the on-state charge and thus the
high power losses in protective circuits around the intelligent MIMO ECM AC–AC,
AC–DC–AC, AC–DC/DC–AC, DC–DC and/or DC–AC–DC commutators.
The author also recently worked out a concept for an ECM commutator with a
bipolar ECM, realised on super- and semiconductor hybrid continuous bipolar
electrical valves, based on the use of heterojunctions, i.e. a contact of amorphous
and mono- and/or polycrystalline materials. The properties of these heterojunctions
seem to be for various reasons competitive in relation to structures made from
mono- and/or poly-crystalline semiconductors.
A concept for the bipolar ECM, realised on continuous bipolar power transistors
of the n–p–n type—in which the emitters will be made from chalcogenide glass and
the bases will be made from monocrystalline silicon—or continuous bipolar
heterojunctions and dielectrically insulated gates in field-effect transistors, is being
worked out.
In addition, there is a concept for a bipolar ECM, realised on continuous bipolar
power superconductor-base hot electron transistors (SUPER-HET) on the base of
GaAs/Nab/Ins, continuous bipolar power quantum interference transistors (QUIT),
or continuous bipolar power and very high-speed static induction transistors (SIT)
and/or graphene single electron transistors (SET) with low power-delay product and
very high transconductance. Bipolar ECMs of that type can be used, owing to
specific conditions of the electrical-current super- and/or semi-conductivity, as basic

15-38
 Mechatronics

trivalent-logic functions of the ECM AC–AC, AC–DC–AC, AC–DC/DC–AC,

DC–DC and/or DC–AC–DC commutator, which seem to be the simplest and
most effective static converters for conversion of one kind of electrical energy into
another. This trivalent-logic is based on the Lukasiewicz–Tarski algebra.
Semi- and/or superconductive ECM AC–AC, AC–DC–AC, AC–DC/DC–AC,
DC–DC and/or DC–AC–DC commutators are now particularly indispensable in
modern EM actuator engineering, and for energy conversion purposes, mainly in
fly-by-wire (FBW) aircraft and ride-by-wire (RBW) or x-by-wire (XBW) automotive
vehicles. Compared with previous polysectorial (polysegmental) MCM commuta-
tors and polyvalvular ECM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and/
or DC–AC–DC commutators (static converters), they are more practice-oriented,
especially from the point of view of operating convenience and the space require-
ments of ultra-modern super- and/or semiconductor ECM techniques.
Numerous semi- and/or superconductive ECM AC–AC, AC–DC–AC, AC–DC/
DC–AC, DC–DC and/or DC–AC–DC commutators have been developed for
modern MMD electrical machines and power applications. When selecting these
commutators, various criteria have to be borne in mind, i.e. the cost of the semi-
and/or superconductive ECM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC,
or DC–AC–DC commutator, the driver requirements, the switching and network,
mounting and wiring needs, together with the ECM commutator’s electrical
characteristics and performance.
The cost of semi- and/or superconductive ECM AC–AC, AC–DC–AC, AC–DC/
DC–AC, DC–DC and/or DC–AC–DC commutators, when compared to that of the
non-MCM wiring requirements, has been reducing and will continue to reduce
significantly.
The development aims for semi- and/or superconductive ECM AC–AC, AC–
DC–AC, AC–DC/DC–AC, DC–DC and/or DC–AC–DC commutators are: to
manufacture ECM commutators that have uncomplicated drivers, and to reduce
circuitry and produce a simplified method for the ECM wiring and mounting in a
realistic process.
Newly developed insulated ECMs meet the requirements for modern MMD
electrical machines and power applications.
Insulated ECMs reduce mounting costs:
• ECM wiring through printed circuits;
• ECM wiring using copper bus-bars;
• a common heat-sink for all ECM components.

Insulated ECMs limit RFI phenomena:

• extremely low parasitic capacitance;
• the heat-sink is grounded to the ECM’s shield and chassis.

Insulated ECMs increase safety:

• internal insulation of 2.5 kV (50 HZ, 1 min);
• safer construction, no mica insulation, as is needed for MCMs;
• very low thermal resistance between the junction and case of electrical valves.

15-39
 Mechatronics

The semi- and/or superconductive ECM AC–AC, AC–DC–AC, AC–DC/DC–

AC, DC–DC and/or DC–AC–DC commutators for modern MMD electrical
machines and power applications are specifically designed to reduce the number
of components, especially electrical valves.
In the case of very high-power semi- and/or superconductive ECM AC–AC, AC–
DC–AC, AC–DC/DC–AC, DC–DC and/or DC–AC–DC commutators for modern
MMD electrical machines and power applications, the use of continuous electrical
valves permits a reduction in parasitical inductance and an improvement in cooling
characteristics.
The semi- and/or superconductive ECM AC–AC, AC–DC–AC, AC–DC/DC–
AC, DC–DC and/or DC–AC–DC commutators for modern MMD electrical
machines and power applications operate at high-current values and, due to their
high-current gain, can be controlled by low-power gate or base drivers. Their
switching safe operating areas, namely reverse-biased safe-operating areas (RBSOA),
allow repeatable operation without use of snubbers.
For high-performance ECM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC
and/or DC–AC–DC commutator dynamotors, ECM-mode power supplies, storage-
battery chargers and welding equipment, the main requirements are:
• noiseless operation;
• large band-pass;
• a reduction in the size and mass of the passive components.

The result is the choice of switching frequencies f ⩾ 15 kHz, or even more for
amorphous electrical valves, e.g. ovonics.
The ECM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and/or DC–AC–
DC commutator has a number of advantages over conventional power converter
technologies when it is used on an AC bus system. As higher temperature power
semi- or superconductive electrical valves, such as silicon carbide ones, become
available, the technology will become a good candidate for power conversion in
applications where volume and mass are at a premium. For example, the monolithic
3 × 3 ECM commutator with a bipolar ECM, realised on only nine (9) continuous
reverse-blocking insulated-gate bipolar transistors (RB IGBT), or metal-oxide semi-
conductor field-effect transistors (MOSFET), or bipolar symmetrical trigistors, that
is, gate turn-off (GTO) symmetrical thyristors (symistors) in a 3 × 3 ECM
connection, is shown in figure 15.16.
This monolithic 3 × 3 ECM AC–AC commutator makes the bilateral flow of
secondary electrical energy possible during generating (absorbing) as well as during
motoring (driving) of the AC–AC commutator dynamotor with variable angular
velocity. Six operating modes are possible in this MMD electrical machine.
Potentially attractive features of the monolithic 3 × 3 ECM, such as the absence
of DC link energy storage with reactive components, high power density, high
quality of input and output waveforms, are leading to continued research efforts
aimed at translating their promise into reality (Fijalkowski 1973, 1982, 1985;
Wheeler et al 2004).

15-40
 Mechatronics

Figure 15.16. The monolithic 3 × 3 ECM AC–AC commutator with nine (9) RB IGBTs.

Although monolithic 3 × 3 ECMs were first introduced as a class of frequency and

phase-number changer realised using controlled electrical valves (electronic
switches) by Fijalkowski (1973, 1982, 1988) and then later by Gyugyi and Pelly
(1976), significant progress began with the introduction of high-frequency synthesis,
as proposed by Venturini (1980).
Although the initial approach was limited to a voltage transfer ratio of one half,
improved modulation strategies with a voltage transfer ratio of 3/2 were published on
by Alesina and Venturini (1989). Further along, indirect modulation methods based on
a fictitious DC link concept were proposed (Rodriguez 1984; Alesina and Venturini
1989; Oyama et al 1989; Neft and Schauder 1988a; 1988b; Ziogas et al 1986). With the
exception of the programmed commutation (switching) pattern in (Ziogas et al 1986),
none of these modulation strategies for monolithic 3 × 3 ECMs based on high-
frequency synthesis have been able to break the upper bound given by 3/2 p.u.
Since AC–AC commutator motor output power is proportional to the square of the
applied voltage, this limitation leads to reduced output power, limited to 3/4 of the AC–
AC commutator motor rating, when fed from a monolithic 3 × 3 ECM AC–AC
commutator operating from the line nominally rated at AC–AC commutator motor
voltage. Otherwise, a custom-designed AC–AC commutator motor designed for the
appropriate output voltage level would be required. This remains one of the many
roadblocks limiting the viability of the monolithic 3 × 3 ECM AC–AC commutator.
A low commutation (switching) frequency six-step modulation strategy for
frequency conversion using 3 × 3 ECM AC–AC commutators capable of providing
more than unity voltage transfer factor was presented by Wang and Venkataramanan
(2006).
The modulation strategy may be more easily developed on the basis of the
operation of the dual monolithic 3 × 2 plus 2 × 3 ECM topology with a fictitious DC
link by using commutation-matrixer holor multiplication of these two cascaded
connected ECMs (see sub-section 10.6.1), as previously introduced by the author of
this book (Fijalkowski 1967, 1982, 1985, 1987).
However, the approach may also be applicable for the monolithic 3 × 3 ECM
(nine-electrical valve) topology, as may be demonstrated through appropriate
topological mapping.

15-41
 Mechatronics

Figure 15.17(a) illustrates the 3 × 3 ECM using three monolithic single-pole-triple-

valvular (SPTV) commutation groups Va, Vb, Vc with nine (9) RB IGBTs. Figure
15.17(b) illustrates the reduced electrical valve (switch) two monolithic SPTV
commutation groups Vp and Vn with six (6) RB IGBTs and three monolithic
single-pole-double-valvular (SPDV) commutation groups Va, Vb, Vc with six (6) RB
IGBTs. The two realisations are labeled as conventional ECM and indirect ECM,
respectively. If the commutation function of the RB IGBTs, as illustrated in
figure 15.17, is defined to be C sPji , the input-output transposition properties of the
ECM can be expressed as
Esj = C sPjiEiP , (15.24)

or in detail as
ji ji ji
Eaj C aA C aB C aC Ei A
Ebj = ji
C bA ji
C bB ji
C bC EiB . (15.25)
Ecj CcAji CcBji CcCji EiC

However, a mathematical model of an electrical continuous dynamical system of the

3 × 3 ECM AC–AC commutator shown in figure 15.17, acting as a current-source
ECM AC–AC commutator, making the input holors dependent on the output
holors, in the dynamical systems approach and matrix notation may be expressed in
the form:
IPj = C PsjiIis , (15.26)

or in detail as
ji ji ji
IAj C Aa C Ab C Ac Iia
IBj = ji
C Ba ji
C Bb C Bcji Iib , (15.27)
ICj CCaji CCbji CCcji Iic

where
T
Esj = Eaj Ebj Ecj ;

T T T
EiP = EiA EiB EiC ; IPj = IAJ IBJ ICJ ; Iis = IIa IIb Iic

C sPji is the voltage-source holor commutation matrixer with the elements being the
commutation functions of various throws shown in figure 15.17(a):
ji ji ji
C aA C aB C aC
C sPji = ji
C bA ji
C bB ji
C bC .
CcAji CcBji CcCji

15-42
 Mechatronics

Figure 15.17. Representation of the monolithic 3 × 3 ECM AC–AC commutator (a) and the monolithic 3 × 2
plus 2 × 3 ECM AC–DC–AC commutator (b).

15-43
 Mechatronics

C Psji is the current-source holor commutation matrixer with the elements being the
commutation functions of various throws shown in figure 15.17(a):
ji ji ji ji ji ji T
C Aa C Ab C Ac C aA C aB C aC
C Psji = ji
C Ba ji
C Bb C Bcji = ji
C bA ji
C bB ji
C bC .
CCaji CCbji CCcji CcAji CcBji CcCji
JI
C kL are the voltage-source coefficients of the commutation nodes kL, which may be
expressed as follows:
⎧ Δ j L
⎪ + 1 → E k = + Ei
∂Ekj ⎪
JI
C kL = =⎨ 0 → Δ
E kj = 0 .
∂EiL ⎪
E iI = 0
⎪ Δ j L
⎩ − 1 → E k = − Ei

CijKl are the current-source coefficients of the commutation nodes Kl, which may be
expressed as follows:
⎧ Δ K j
⎪ − 1 → Ei = − El
∂Ei K ⎪
CijKl = =⎨ 0 → Δ
EiK = 0 .
∂El j Ei j = 0
⎪
⎪ Δ K
⎩ + 1 → Ei = + El
j

The input–output transposition properties of the 3 × 2 plus 2 × 3 EMC AC–DC–AC

commutator can be expressed as
P
EsJ = C sPji (E )CijsP(I )E i , (15.28)

or in detail as
ji ji
Eaj C pA C nA Ei A
ji ji
Cijap Cijbp Cijcp
Ebj = C pB C nB EiB , (15.29)
Cijan Cijbn Cijcn
Ecj ji
C pC ji
C nC EiC

where CijsP(I ) is the commutation matrixer of the current-source 2 × 3 EMC DC–AC

commutator connected with the stiff voltage holors. C sPji (E ) is the commutation
matrixer of the voltage-source 3 × 2 EMC AC–DC commutator connected with the
stiff current holors. They are given by
ji ji
C pA C nA
ji ji
C sPji (E ) = C pB C nB
ji ji
C pC C nC

15-44
 Mechatronics

and

Cijap Cijbp Cijcp

CijsP(I ) = .
Cijan Cijbn Cijcn

The input–output transposition properties of the 3 × 2 EMC AC–DC commutator

can be expressed as
E sj(E ) = C sPji (E ) EiP(E ), (15.30)

or in detail as
ji ji
C pA C nA EiA(E )
E pj(E ) ji ji
= C pB C nB EiB(E ) , (15.31)
Enj(E ) ji ji
C pC C nC EiC (E )

and the input–output transposition properties of the 2 × 3 EMC DC–AC commu-

tator can be expressed as
Eis(I ) = CijsP(I )EPj(I ), (15.32)

or in detail as

Eia(I )
Cijap Cijbp Cijcp E pj(I )
Eib(I ) = . (15.33)
Cijan Cijbn Cijcn Enj(I )
Eic(I )

However, a mathematical model of an electrical continuous dynamical system of the

3 × 2 plus 2 × 3 ECM AC–DC–AC commutator shown in figure 15.17, acting as a
current-source ECM AC–DC–AC commutator, making the input holors dependent
on the output holors, in the dynamical systems approach and matrix notation may
be expressed in the form:
Isj = C Psji (I )CijPs(E )IiP , (15.34)

or in detail as

IAj C paji C naji Iia

CijAp CijBp CijCp
IBj = C pbji C nbji Iib . (15.35)
CijAn CijBn CijCn
ICj C pcji C ncji Iic

The input–output transposition properties of the 3 × 2 EMC AC–DC commutator

can be expressed as
Isj(I ) = C Psji (I )IiP(I ),

15-45
 Mechatronics

or in detail as

C paji C naji Iia(I )

I pj(I )
= C pbji C nbji Iib(I ) ,
Inj(I )
C pcji C ncji Iic(I )

and the input–output transposition properties of the 2 × 3 EMC DC–AC

commutator can be expressed as

IiP(E ) = CijPs(E )Isj(E ),

or in detail as

IiA(E )
CijAp CijBp CijCp I pj(E )
IiB(E ) = ,
CijAn CijBn CijCn Inj(E )
IiC (E )

where

T
Esj(E ) = Epj(E ) Enj(E )
T
EiP(E ) = EiA(E ) EiB(E ) EiC (E )
T
Esj(I ) = Eaj(I ) Ebj(I ) Ecj(I )
T
EiP(I ) = Ei p(I ) Ein(I )
T
Esj(E ) = Epj(I ) Enj(I )
T
Isj(I ) = IAj (I ) IBj(I ) ICj(I )
T
IiP(I ) = Ii p(I ) Iin(I )
T
IiP(E ) = IiA(E ) IiB(E ) IiC (E )
T
Isj(E ) = I pj(E ) Inj(E )

C Psji (E ) is the commutation-matrixer holor of the voltage-source 3 × 2 EMC AC–DC

commutator connected with the stiff voltage holors and C Psji (I ) is the commutation-

15-46
 Mechatronics

matrixer holor of the current-source 2 × 3 EMC DC–AC commutator connected

with the stiff current holors. They are given by

C paji C naji
C Psji (I ) = C pbji C nbji
C pcji C ncji

and

CijAp CijBp CijCp

CijPs(E ) = .
CijAn CijBn CijCn

By matrix multiplication of the commutation-matrixer holor of the voltage-source

3 × 2 EMC AC–DC commutator and the commutation-matrixer holor of the
current-source 2 × 3 EMC DC–AC commutator, we may receive mathematical
confirmation that a product of the commutation-matrixer holors of the 3 × 2 plus 2 ×
3 EMC commutator is equal to the commutation-matrixer holor of the 3 × 3 EMC
AC–AC commutator.

ji ji
C pA C nA
sP (I ) ji ji
Cijap Cijbp Cijcp
C sPji (E )C = C pB C nB
ij
ji ji
Cijan Cijbn Cijcn
C pC C nC
ji ji ji ji ji ji
C pA Cijap + C nA Cijan C pA Cijbp + C nA Cijbn C pA Cijcp + C nA Cijcn
ji ji ji ji ji ji
= C pB Cijap + C nB Cijan C pB Cijbp + C nB Cijbn C pB Cijcp + C nB Cijcn
ji ji ji ji ji ji
C pC Cijap + C nC Cijan C pC Cijbp + C nC Cijbn C pC Cijcp + C nC Cijcn
ji ji ji
C aA C aB C aC
ji ji ji
= C bA C bB C bC .
CcAji CcBji CcCji

Similarly, by matrix multiplication of the commutation-matrix holor of the current-

source 2 × 3 EMC AC–DC commutator and the commutation-matrix holor of the
voltage-source 3 × 2 EMC DC–AC commutator, we may receive mathematical
confirmation that a product of the commutation-matrix holors of the 3 × 2 plus 2 × 3
EMC commutator is also equal to the commutation-matrix holor of the 3 × 3 EMC
AC–AC commutator.

15-47
 Mechatronics

C paji C naji
Ps(E ) CijAp CijBp CijCp
C Psji (I )C = C pbji C nbji
ij CijAn CijBn CijCn
C pcji C ncji

C pajiCijAp + C najiCijAn C pajiCijBp + C najiCijBn C pajiCijCp + C najiCijCn

= C pbjiCijAp + C nbjiCijAn C pbjiCijBp + C nbjiCijBn C pbjiCijCp + C nbjiCijCn
C pcjiCijAp + C ncjiCijAn C pcjiCijBp + C ncjiCijBn C pcjiCijCp + C ncjiCijCn
ji ji ji
C Aa C Ab C Ac
ji ji
= C Ba C Bb C Bcji ,
CCaji CCbji CCcji

where
ji ji ji ji ji ji T
C aA C aB C aC C Aa C Ab C Ac
ji ji ji ji ji
C bA C bB C bC = C Ba C Bb C Bcji
CcAji CcBji CcCji CCaji CCb
ji
CCcji

Commutation functions and voltage/current holors

Let the three-phase matrixer-input-voltage holors and three-phase matrixer-output-
current holors be assumed to be balanced stiff AC holor quantities and expressed as
EA = E ∠α I a = I∠β
EB = E ∠α + 2π /3 I b = I∠β + 2π /3 (15.36)
EC = E ∠α − 2π /3 I c = I∠β − 2π /3,

where E is the voltage-source holor amplitude; I is the current-source holor

amplitude; αP is the phase angle of the voltage-source holor, given by
αP = ωPt + αP0 ; and βs is the phase angle of the current-source holor, given by
βs = ωst + βs0 .
Let the modulation functions for the current-source 3 × 2 ECM and voltage-
source 2 × 3 ECM be defined as
MA = 1∠α M a = 1∠β
MB = 1∠α + 2π /3 M b = 1∠β + 2π /3 (15.37)
MC = 1∠α − 2π /3 M c = 1∠β − 2π /3.

The displacement angle φP on the current-source 3 × 2 ECM primary side and the
displacement angle φs on the voltage-source 2 × 3 ECM secondary side are defined as
φP = αP 0 − βP 0, φs = αs 0 − βs 0 . (15.38)

15-48
 Mechatronics

The commutation (switching) functions for the various throws of the current-source
3 × 2 ECM and voltage-source 2 × 3 ECM are determined by equations (15.37) and
(15.38), respectively.
⎧1 if (MA > MB ) ∧ (MA > MC )
ji
C pA =⎨
⎩ 0 otherwise
⎧1 if (MB > MC ) ∧ (MB > MA)
ji
C pB =⎨
⎩ 0 otherwise
⎧1 if (MC > MA) ∧ (MC > MB )
ji
C pC =⎨
⎩ 0 otherwise
(15.39)
⎧1 if (MA < MB ) ∧ (MA < MC )
ji
C nA =⎨
⎩ 0 otherwise
⎧1 if (MB < MC ) ∧ (Mb < Ma )
ji
C nB =⎨
⎩ 0 otherwise
⎧1 if (MC < MA) ∧ (MC < MB )
ji
C nC =⎨
⎩ 0 otherwise

⎧ 1 if MA > 0
⎪
Cijap=⎨ ; Cijan = 1 − Cijap
⎪
⎩ 0 otherwise
⎧ 1 if MB > 0
⎪
Cij = ⎨
bp
; Cijbn = 1 − Cijbp (15.40)
⎪
⎩ 0 otherwise
⎧ 1 if MC > 0
⎪
Cij = ⎨
cp
; Cijcn = 1 − Cijcp .
⎪
⎩ 0 otherwise

The commutation functions for the current-source 3 × 2 ECM and the voltage-
source 2 × 3 ECM are depicted in figure 15.18.
With this choice of commutation functions, the resulting output-voltage holors
and input-current holors achieve unity displacement factor operation (cos φ = 1) at
the input and output simultaneously.

Summary
Conventional modulation strategies for ECM commutators based on trigonometric
α − β or d − q transformations and space-vector (SV) techniques have their voltage
transfer capability limited at about 0.866 p.u. This leads to a derating of AC–AC
commutator motor capability by 25% in the realisation of variable frequency
electromechanical drives using off-the-shelf (OTS) line-fed commutatorless AC
motors, or requires custom-designed AC–AC commutator motors for operation at

15-49
 Mechatronics

Figure 15.18. Waveforms of various commutation functions for the current-source 3 × 2 ECM AC–DC
commutator (a) and the voltage-source 2 × 3 ECM DC–AC commutator (b) for unity power factor operation
at both sets of matrixer terminals or both matrixer ports.

reduced voltage. Thus, a six-step variable-frequency modulation strategy for ECM

commutators that are capable of providing more than a 1 p.u. fundamental
component of output voltage is presented.
In this sub-section a new modulation/control strategy for the 2 × 3 plus 3 × 2
ECM commutator featuring up to unity voltage transfer ratio is presented. If such a
low-frequency modulation approach is combined with proper filter design and/or
multi-pulse techniques, practical designs may be realized for applications using
electrical valves as RB IGBTs.
The combination of the presented modulation with conventional pulse-width
modulation (PWM) provides independent specification of power factor at both the
matrixer-input and matrixer-output ports of the MCM commutator. Furthermore,
the presented approach makes it possible to develop the limits of conventional scalar
and vector PWM approaches in the transition region between linear operation and
six-step operation.
A six-step operation that is suitable for the conventional nine-electrical-valve
topology and the reduced electrical valve (switch) dual monolithic ECM top-
ologies is presented, and it is capable of unity displacement-factor operation
(cos φ = 1) at the matrixer input and matrixer output simultaneously. It is
demonstrated that by varying the commutation (switching) angles, control of
the fundamental component of output-voltage holor and input-current holor is
possible, with concomitant generation of reactive power at the matrixer-input and
matrixer-output terminals of the monolithic ECM AC–DC–AC commutator,
with bipolar (bilateral) power flow.

15-50
 Mechatronics

15.4 MCM and ECM AC–AC and AC–DC/DC–AC commutator

dynamotors—a basic application
Macro- and microelectronics technology led to the change-over from today’s power
electronics static converter torque and speed control systems for MCM DC–AC
commutator motors and AC commutatorless motors to tomorrow’s ECM AC–AC
and DC–AC commutator motors.
In the past, power electronics technology led to the change-over from the rotary
dynamoelectric Ward Leonard torque and speed control systems for MCM DC–AC
commutator motors to static-converter Ward Leonard torque and speed control
systems. This commenced in the 1930s with the introduction of mercury-arc
converters and in the 1950s with the introduction of transductor-controlled semi-
conductor rectifiers (SR), i.e. power diodes.
Today, semiconductor controlled rectifiers (SCR), i.e. thyristor converters for
MCM DC–AC commutator motor advanced propulsion (traction dynamical
hypersystems) have matured, and this technology is well established in many
applications. Because of their simple and robust design, AC commutatorless motors,
and then all the AC hyposynchronous (induction) squirrel-cage-rotor motors
invented by the Polish electrician, M O Dolivo-Dobrovolsky (1862–1919), in
1888, are preferable to the MCM DC–AC commutator motor for many torque
and speed control applications, particularly in difficult and hazardous environments
(Hoseman et al 1999). However, torque and speed control for AC commutatorless
motors have only been a dream for many years. It is true that this technique has
existed for a long time and indeed manufacturers have also been manufacturing
static converters (static frequency changers) for AC MCM commutatorless motors
for today’s industrial and traction propulsion complex dynamical hypersystems.
In recent years, the cost trend for ECM commutators has now made it
increasingly interesting from the commercial viewpoint to use ECM commutators
more widely, in ECM AC–AC and AC–DC/DC–AC commutator dynamotors for
variable-speed industrial and traction propulsion complex dynamical hypersystems.
This trend is still in its infancy and this technique should experience its definite
breakthrough during the coming years.
The design of ECM commutators is closely tied to the complexity of the MMD
electrical machine. Beyond doubt, electrical energy conversion is technically the
most advanced today. It involves the static conversion of electrical energy.
It has been, and still is, a challenge for the application of newly designed energy-
saving ECM AC–AC and AC–DC/DC–AC commutator dynamotors (generators/
motors), particularly if industrial and traction propulsion complex dynamical
hypersystems are considered. It is consequently not an exaggeration to state that
ECM AC–AC and AC–DC/DC–AC commutator dynamotors are technically in
advance of those for the other MMD electrical machines.
The author has been very active during recent years in the field of ECM AC–AC
and AC–DC/DC–AC commutator dynamotors. Viewed from a historical perspec-
tive, the author has kept up with the technical developments in this area and on

15-51
 Mechatronics

many occasions has quickly and successfully exploited new technical advances.
Figures 15.19 and 15.20 show some important landmarks schematically.
As far back as 1881, the electrical machinery industry supplied the first MCM
AC–DC/DC–AC commutator dynamotors for industrial and traction propulsion
complex dynamical hypersystems.
The technology was based on ME energy conversion. Development then
continued with the introduction of mercury-arc and semiconductor converters.
The first fully static ECM commutators and ECM AC–AC and AC–DC/DC–AC
commutator dynamotors were launched in the early 1930s (figures 15.16 and 15.17).
Recently, there has been considerable interest in the use of ECM AC–AC,
AC–DC–AC, AC–DC/DC–AC, DC–DC and DC–AC–DC commutator technol-
ogy for industrial and traction propulsion mechatronical control systems, as well
as for aerospace and automotive applications.
The aerospace and automotive industries have also shown considerable interest in
ECM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and DC–AC–DC com-
mutators due to the inherent potential advantages they have in aircraft and
automotive applications. This textbook will give an in-depth introduction to the
full range of ECM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and DC–
AC–DC commutator technologies and investigate the suitability and design of ECM
AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and DC–AC–DC commutators.
The potential advantages of ECM AC–AC, AC–DC–AC, AC–DC/DC–AC,
DC–DC and DC–AC–DC commutator technology will be examined, and the
factors that have so far limited commercial exploitation of the ECM will be
discussed (see the references and bibliography).
The book will use appropriate practical results from various prototypical MIMO
ECM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and DC–AC–DC com-
mutators to illustrate the topics under discussion. These examples will focus on

Figure 15.19. Some development milestones in the field of MCM AC–DC/DC–AC commutator dynamotors
(Fijalkowski 1988).

15-52
 Mechatronics

Figure 15.20. Some development milestones in the field of MCM AC–AC commutator motors (Fijalkowski
1988).

application areas where the advantages of the ECM AC–AC, AC–DC–AC,

AC–DC/DC–AC, DC–DC and DC–AC–DC commutator carry a large premium,
for example industrial and traction propulsion mechatronical control systems,
especially for aerospace and automotive applications.
The textbook is aimed at anybody interested in the area of MIMO ECM AC–AC,
AC–DC–AC, AC–DC/DC–AC, DC–DC and DC–AC–DC commutators. It will
start with a basic introduction to the topology and lead into in-depth discussion on
the critical research developments and topical issues.
Today, macro- and microelectronics technology has become widely accepted for
new-concept integral ECM AC–AC and AC–DC/DC–AC commutator dynamo-
tors. Such technology has resulted in considerably improved industrial and traction
propulsion complex dynamical hypersystems, providing fast, smooth and precise
torque and speed control. For instance, for railway and automotive vehicles this
means better slip controls, which have made it possible to take out high tractive
efforts close to the limit of adhesion.

15.5 New-concept ECM AC–AC and AC–DC/DC–AC commutator dynamotors

The development of modern ECM AC–AC and AC–DC/DC–AC commutator
dynamotors is closely connected with the further development of macro- and
microelectronics technology. Moreover, it is necessary to pay attention to the
interesting fact that contemporary MCM AC–AC and AC–DC/DC–AC commu-
tator dynamotors have not changed their basic construction and appearance since
the late 1980s. R&D work on ECM AC–AC and AC–DC-/DC–AC commutator
dynamotors initially concentrated on designing their details (component parts) and a
deeper understanding of the physical phenomena appearing during their working.
Modern ECM AC–AC and AC–DC/DC–AC commutator dynamotors with
ECM commutators have up to now been used very rarely, mainly in aviation and
automotive industrial and traction propulsion complex dynamical hypersystems
(Fijalkowski 1985b).

15-53
 Mechatronics

Another factor that contributed to such a state of affairs in power electronics was
the fact that in the 1940s the rapid development of mercury-arc converters took
place, followed by the development of thyristor converters in the 1950s.
Thus, scientists were only interested on a large scale in the development of
electronic commutators, i.e. polyvalvular ECM AC–AC and/or AC–DC/DC–AC
commutators with unipolar ECMs, realised on continuous power diodes and
thyristors—up to now called ‘static converters’ in the field of power electronics.
The new-concept integral ECM AC–AC and/or AC–DC/DC–AC commutator
dynamotors conceived by the author can be used, e.g. in advanced cradled
dynamometer simple continuous dynamical systems for testing combustion engines,
and electrical, fluidic and pneumatic machines, as well as other rotary machines, etc,
while industrial and traction propulsion complex dynamical hypersystems operate
in sinusoidal-pulse width modulation (SPWM) mode with integral ECM commuta-
tors, just like conventional MCM AC–AC and AC–DC/DC–AC commutator
dynamotors.
A dynamometer that is designed to be driven is called an ‘absorption dyna-
mometer’. A dynamometer that can either drive or absorb is called a ‘universal
dynamometer’. A dynamometer can also be used to determine the torque and power
required to operate a driven machine, for example, a mechano-fluidic (MF) pump or
a mechano-pneumatic (MP) compressor. In that case, a motoring or driving
dynamometer is used.
A motoring dynamometer acts as an MCM AC–AC or a DC–AC commutator
motor that drives the equipment being tested. It must be able to drive the
combustion engine, EM, FM or PM motor, or other rotating prime mover, at
any angular speed and develop any level of torque that the test requires. Only torque
and angular speed can be measured directly.
In most dynamometers, power is not measured directly, it must be computed
from the torque and angular-speed holors, or force and linear-velocity holors,
according to the following holor product:
P κ = Ti ⊗ Ω j = γ..κjiTi Ω j ,

or
P κ = Fi ⨂ V j = γ..κjiFV
i
j
,

where P κ is the power holor (W), Ti is the torque holor (Nm), Ω j is the angular-velocity
holor (rad s−1), Fi is the force holor (N) and V j is the linear-velocity holor (m s−1).
A growing feature of industrial and traction dynamical systems is the sense of the
power-flow direction in the ECM commutator of the MMD electrical machine being
arbitrary, that is, it can be used for both EM motoring and ME generation
(regenerative braking). The change to the sense of the power-flow direction can
take place without any reconnections. Other characteristics that can be achieved
with industrial and traction propulsion dynamical systems are a power factor that
can be controlled in an optimum way and low machine armature-winding phase

15-54
 Mechatronics

current harmonics. These characteristics are perhaps more important than the
industrial and traction propulsion complex dynamical hypersystem as such.
Combined with reasonable forecasts for the ever more widespread application of
new-concept integral ECM AC–AC and AC–DC/DC–AC commutator dynamotors
in industrial and traction propulsion complex dynamical hypersystems, constant
attention has been devoted to studying and updating these initial new-concept
integral ECM AC–AC and AC–DC/DC–AC commutator dynamotors and integral
lower- and higher-level microprocessor controllers, as broken down into the
following integral MMD electrical machine types (figures 15.21 and 15.22).
• Three- or two-phase ECM AC–AC commutator hypersynchronous orthog-
onal-two-phase wound-rotor dynamotor (triply fed brushed MMD electrical
machine)—figures 15.21(a).
• Three- or two-phase ECM AC–AC commutator hypersynchronous orthog-
onal-two-phase wound-rotor dynamotor (triply fed brushless MMD electrical
machine)—figures 15.21(b).
• Three- or two-phase ECM AC–AC commutator hypersynchronous three-
phase wound-rotor dynamotor (doubly fed brushed MMD electrical
machine)—figures 15.21(c).
• Three- or two-phase ECM AC–AC commutator hypersynchronous three-
phase wound-rotor dynamotor (doubly fed brushless MMD electrical
machine)—figures 15.21(d).
• Three- or two-phase ECM AC–AC commutator synchronous orthogonal-
two-phase wound-rotor electromagnetically excited dynamotor (triply fed
brushed MMD electrical machine)—figure 15.21(e).

Figure 15.21. Three- or two-phase ECM AC–AC commutator MMD electrical machines (Fijalkowski 1987).

15-55
 Mechatronics

Figure 15.22. Single-phase ECM AC–AC and/or AC–DC/DC–AC commutator MMD electrical machines
(Fijalkowski 1987).

• Three- or two-phase ECM AC–AC commutator synchronous orthogonal-

two-phase wound-rotor electromagnetically excited dynamotor (triply fed
brushless MMD electrical machine)—figure 15.21(f).
• Three- or two-phase ECM AC–AC commutator synchronous single-phase
wound-rotor electromagnetically excited dynamotor (doubly fed brushed
MMD electrical machine)—figure 15.21(g).
• Three- or two-phase ECM AC–AC commutator synchronous single-phase
wound-rotor electromagnetically excited dynamotor (doubly fed brushless
MMD electrical machine)—figure 15.21(h).
• Three- or two-phase ECM AC–AC commutator synchronous magneto-
electrically excited dynamotor (doubly fed brushless MMD electrical
machine)—figure 15.21(i).
• Three- or two-phase ECM AC–AC commutator synchronous reluctance
dynamotor (singly fed brushless MMD electrical machine)—figure 15.21(j).
• Three- or two-phase ECM AC–AC commutator hyposynchronous (induc-
tion) squirrel-cage-rotor dynamotor (singly fed brushless MMD electrical
machine)—figure 15.21(k).
• Three- or two-phase ECM AC–AC commutator hyposynchronous (induc-
tion) orthogonal-two-phase stator and hollow-rotor dynamotor (doubly fed
brushless MMD electrical machine)—figure 15.21(l).
• Single-phase ECM AC–AC and/or AC–DC/DC–AC commutator hyper-
synchronous orthogonal-two-phase wound-rotor dynamotor (triply fed
brushed MMD electrical machine)—figure 15.22(a).

15-56
 Mechatronics

• Single-phase ECM AC–AC and/or AC–DC/DC–AC commutator hyper-

synchronous orthogonal-two-phase-wound-rotor dynamotor (triply fed
brushless MMD electrical machine)—figure 15.22(b).
• Single-phase ECM AC–AC and/or AC–DC/DC–AC commutator hyper-
synchronous three-phase-wound-rotor dynamotor (doubly fed brushed
MMD electrical machine)—figure 15.22(c).
• Single-phase ECM AC–AC and/or AC–DC/DC–AC commutator hyper-
synchronous three-phase-wound-rotor dynamotor (doubly fed brushless
MMD electrical machine)—figure 15.22(d).
• Single-phase ECM AC–AC and/or AC–DC/DC–AC commutator hyper-
synchronous orthogonal-two-phase-wound-rotor electromagnetically excited
dynamotor (triply fed brushed MMD electrical machine)—figure 15.22(e).
• Single-phase ECM AC–AC and/or AC–DC/DC–AC commutator syn-
chronous orthogonal-two-phase wound-rotor electromagnetically excited
dynamotor (triply fed brushless MMD electrical machine)—figure 15.22(f).
• Single-phase ECM AC–AC and/or AC–DC/DC–AC commutator synchro-
nous single-phase wound-rotor electromagnetically excited dynamotor (dou-
bly fed brushed MMD electrical machine)—figure 15.22(g).
• Single-phase ECM AC–AC and/or AC–DC/DC–AC commutator syn-
chronous single-phase wound-rotor electromagnetically excited dynamotor
(doubly fed brushless MMD electrical machine)—figure 15.20(h).
• Single-phase ECM AC–AC and/or AC–DC/DC–AC commutator syn-
chronous single-phase wound-rotor magnetoelectrically excited dynamotor
(doubly fed brushless MMD electrical machine)—figure 15.22(i).
• Single-phase ECM AC–AC and/or AC–DC/DC–AC commutator synchro-
nous reluctance dynamotor (singly fed brushless MMD electrical machine)—
figure 15.22(j).
• Single-phase ECM AC–AC and/or AC–DC/DC–AC commutator hyposyn-
chronous (induction) squirrel-cage-rotor dynamotor (singly fed brushless
MMD electrical machine)—figure 15.22(k).
• Single-phase ECM AC–AC and/or AC–DC/DC–AC commutator hyposyn-
chronous (induction) orthogonal-two-phase wound-stator and hollow-rotor
dynamotor (doubly fed brushless MMD electrical machine)—figure 15.22(l).

These new-concept integral MMD electrical machines with ECM commutators

will have a fully laminated stator and all the electrical insulation will be of class H.
The material utilisation, which is the ratio between the mass and the output
power, will be better for the new-concept ECM AC–AC and AC–DC/DC–AC
commutator dynamotors compared with the preceding MMD electrical machines.
This improvement will be achieved without having to increase either the electrical
stresses or the relative level of the material used. This means that, for example, the
margin for a harmful temperature rise in connection with overloads is at least as
great as for the older designs.
ECM AC–AC and AC–DC/DC–AC commutator dynamotors with their acces-
sories and above all intelligent MMD electrical machines represent one of the main

15-57
 Mechatronics

themes of this textbook. Concepts such as artificial intelligence and microcomputer

capacity are factors that are increasingly putting their imprint on our world. The use
of industrial and traction propulsion complex dynamical hypersystems with integral
ECM AC–AC and AC–DC/DC–AC commutator dynamotors is growing rapidly.
This is largely due to the increasingly severe demands being made on these
dynamical hypersystems.
Today’s and tomorrow’s traction propulsion complex dynamical hypersystems
must be capable of being adjusted to cope with the varying operation conditions and
handle a wide range of current and voltage/torque and speed controls.
In addition, they must be flexible to meet operation modifications, where new
operating conditions might be needed in existing and traction propulsion complex
dynamical hypersystems. To meet these demands, integral ECM AC–AC and AC–
DC/DC–AC commutator dynamotors are becoming more advanced and their range
of accessories is growing. These dynamotors will be available in the near future as
complete ‘function packages’ for different applications. Their accessories will
include flexible microprocessor-based ECM commutator controllers and a new
artificial intelligence dynamical system.

Advantages and disadvantages of integral ECM AC–AC and AC–DC/DC–AC

commutator dynamotors
Here follows a short list of the advantages and disadvantages of integral ECM
AC–AC and AC–DC/DC–AC commutator dynamotors.
Some of the advantages are as follows.
• Variable/adjustable voltage and/or current, as well as torque and/or speed
controls, which may increase the efficiency of the load. Increased efficiency
leads to reduced electrical energy costs.
• Easy installation, leading to reduced installation costs.
• Easy commissioning, leading to reduced commissioning costs.
• No space is required for an ECM AC–AC or AC–DC/DC–AC commutator
cabinet.
• Reduced EMC problems, both radiated and line-carried, due to the contain-
ment of the ECM and the cable from the ECM AC–AC and/or AC–DC/DC–
AC commutator to the dynamotor.
• Reduced stock inventory, since one integral ECM AC–AC and AC–DC/DC–
AC commutator dynamotor can replace, for example, several AC induction
motors with different pole numbers for a certain torque.

Some of the disadvantages are as follows.

• Novel integral ECM AC–AC and AC–DC/DC–AC commutator dynamotor
concept scares potential buyers.
• Slightly more expensive to buy than an AC or a DC dynamotor with a
separate static converter.
• The AC–AC and/or AC–DC/DC–AC commutator’s ECM is exposed to the
same environment as the dynamotor, regarding vibrations and heat.

15-58
 Mechatronics

Modern industrial and traction propulsion complex dynamical hypersystems with

an integral ECM AC–AC or AC–DC–AC commutator dynamotor as their chief
and only component are generally built up as hierarchical physical heterogeneous
continuous complex dynamical hypersystems (figure 15.23).
The dynamotor’s microprocessor-based ECM commutator controllers (lower-
level controllers) communicate with higher-level propulsion controllers, which
handle the coordination within the industrial and traction propulsion complex
dynamical hypersystem and, in their turn, the propulsion controllers communicate
with the ECM commutator controllers. Further down in the hierarchy is a relatively
small volume of data for the different ECM AC–AC and AC–DC/DC–AC
commutator dynamotors’ ECM commutator controllers. On the other hand, greater
demands are frequently made on the signal-processing rate than in the higher-level
propulsion controllers. The need for the signal-processing rate in the control
functions decreases at the higher level.
In the case of higher-level control and monitoring with operator presentation, the
microcomputer must be able to handle large volumes of data, while the demand
made on the processing rate decreases.
There are three main function categories in each type of process computer system:
• process control (PC) functions;
• man–machine communication (MMC) functions;
• computer system (CS) functions.

Figure 15.21 shows the main functions and signal routes of the industrial or
traction propulsion complex dynamical hypersystem. The control equipment is

Figure 15.23. Main functions of an ECM AC–AC and or AC–DC–AC commutator dynamotor (Fijalkowski
1985).

15-59
 Mechatronics

divided into a ‘higher-level propulsion controller’ and a ‘lower-level static commu-

tator’. Both sets of control equipment are located in the ECM commutator
compartment.
The ECM commutator controller incorporates circuits for controlling and
monitoring the armature and field ECMs. In principle, the input signal to the
ECM commutator controller is a torque reference (set point), which is compared
with a torque signal obtained by means of suitable control of the armature and field
ECMs.
The propulsion controller incorporates circuits for controlling and monitoring the
object driven by the integral ECM AC–AC or AC–DC/DC–AC commutator
dynamotor. The control mode is generally speed control, but also position or
some other variables; for example, tension, pressure, flow and loop may also be
controlled. The reference (set point) for the controlled variable may be formed in the
propulsion controller or may be obtained from the higher-level industrial or traction
process commander (control system). Today’s microprocessor-based ECM commu-
tator controllers consist of microcomputers of varying degree of complexity working
in real time. They are frequently linked in hierarchical structures to satisfy different
functional requirements in the best possible way. Modern industrial and traction
propulsion complex dynamical hypersystems impose demands on advanced data
communication methods.
The operator controls and supervises his/her control process with the aid of an
MMC system, which essentially consists of multi-colour video display units (VDU)
and keyboards. These VDUs reproduce in a quick, correct and easily under-
standable way what is happening in the integral ECM AC–AC and AC–DC/DC–
AC commutator dynamotors, i.e. in the industrial and traction propulsion complex
dynamical hypersystems. This highlights the current requirements and demands for
new-concept integral MMD electrical machines with ECM AC–AC and/or AC–DC/
DC–AC commutators.

Placement of the ECM AC–AC and/or AC–DC/DC–AC commutators

Some of the possibilities may place the ECM AC–AC and/or AC–DC/DC–AC
commutator on top of the integral AC–AC and/or AC–DC/DC–AC commutator
dynamotor, while some of them also allow it to be placed on the left or right side of
the dynamotor.
Placement at the top conserves the footprint of the integral ECM AC–AC and/or
AC–DC/DC–AC commutator dynamotor, but increases the height. The ECM AC–
AC and/or AC–DC/DC–AC commutator can also be placed at the non-drive end.
This placement extends the axial length of the ECM AC–AC and/or AC–DC/DC–
AC commutator dynamotor, but maintains the existing height and reduces the risk
of harmful vibrations. In this case, the AC–AC and/or AC–DC/DC–AC commu-
tator’s ECM is better protected from heat, especially the rising heat of the integral
ECM AC–AC and/or AC–DC/DC–AC commutator dynamotor after it is turned
off.
Modern integral ECM AC–AC and/or AC–DC/DC–AC commutator dynamo-
tors should have ECMs, together with single-chip microcontrollers (designed for

15-60
 Mechatronics

control and monitoring), built into the newly designed terminal boxes of integral
MMD electrical machines.

15.6 Exemplary applications of electrical commutation matrixer

commutators
15.6.1 A 2 × 2 ECM AC–DC/DC–AC commutator
The roles of ECM AC–DC/DC–AC commutators in realistic applications are
major. A 2 × 2 ECM AC–DC/DC–AC commutator, e.g. covers all possible cases
with a single-port input energy source and a two-terminal load.
The ECM AC–DC/DC–AC commutator is commonly drawn with a single
homogeneous anode or cathode commutation group, or two heterogeneous anode
and cathode commutation groups, i.e. the well-known H-bridge converter (rectifier/
inverter) shown in figure 15.24.

15.6.2 A 3 × 3, 3 × 5 or 5 × 5 ECM AC–AC commutator

This book, among other subjects, especially concerns 3 × 3, 3 × 5 and/or 5 × 5 ECM
AC–AC commutator dynamotors as alternative voltage-source inverter (VSI) AC
motor drives (see figure 15.25).

Figure 15.24. A 2 × 2 ECM AC–DC/DC–AC commutator with a single homogeneous anode or cathode
commutation group, or two heterogeneous anode and cathode commutation groups (H-bridge configuration).

15-61
 Mechatronics

Figure 15.25. The 3 × 3, 3 × 5 and 5 × 5 ECM AC–AC commutator dynamotors for steered, motorised and/or
generatorised wheels (SM&GW) (Fijalkowski 1987, 1982, 1985, 1988, 1997).

The 3 × 3, 3 × 5 and 5 × 5 ECM commutators are direct ECM AC–AC

commutators with no DC link. The lack of reactive components in the DC link is
one of the salient advantages of the 3 × 3, 3 × 5 and 5 × 5 ECM commutators.
Furthermore, the 3 × 3, 3 × 5 and/or 5 × 5 ECM features full four-quadrant
operation and sinusoidal ECM input current. The ECM output voltage is limited to
87% of the ECM input voltage.
The 3 × 3, 3 × 5 or 5 × 5 ECM AC–AC commutator needs nine (9), 15 or 25
bipolar electrical valves, respectively, to connect two three- or five-phase voltage
systems in all possible combinations.
Most R&D work about 3 × 3, 3 × 5 and 5 × 5 ECMs has so far concerned the
modulation and control of the ECM AC–AC commutator dynamotor. The
implemented 3 × 3 ECM AC–AC commutator shows a reduction of the reactive
elements by more than a factor of 12 compared to a standard VSI inverter AC motor
drive at the same power rating.
The power flow through the 3 × 3, 3 × 5 or 5 × 5 ECM AC–AC commutator is
bipolar and it draws sinusoidal input current in the generative as well as the
regenerative mode. The input power-power factor is above 0.98. It may be
concluded that VSI converter AC motor drives will become standard in the near
future. For example, in civil and/or military all-electric vehicles (AEV) and hybrid-
electric vehicles (HEV), mass and size reduction is critical. High-temperature
operation is desirable to ease the problem of the thermal management of 3 × 3,
3 × 5 and/or 5 × 5 ECM AC–AC commutator dynamotors with single cooling heat-
sinks (Fijalkowski 1988b, 1988e, 1988g, 1997).
It is believed that the 3 × 3, 3 × 5 or 5 × 5 ECM AC–AC commutator dynamotor
may have significant advantages over the conventional ECM AC–DC–AC commu-
tator dynamotor in all of these areas, since it is possible to eliminate the DC link
capacitor, especially if an AC primary power system is used.
An interesting example of the design and construction of a snubberless 150 kVA
experimental 3 × 3 ECM AC–AC commutator induction dynamotor, built using
600 A, 1400 V continuous IGBTs (Powerex CM600HA-28H) is shown in
figure 15.26 (Wheeler et al 2004).
The 150 kVA experimental 3 × 3 ECM AC–AC commutator is controlled with a
microprocessor (Infineon SAB80C 167). Despite the relatively low-performance

15-62
 Mechatronics

Figure 15.26. A snubberless 150 kV experimental 3 × 3 ECM AC–AC commutator induction motor (Wheeler
et al 2004).

control platform, it has been possible to achieve full closed-loop vector control of the
ECM AC–AC commutator induction motor load. From the output-voltage and
output-current waveforms of the 3 × 3 ECM AC–AC commutator induction motor
shown in figure 15.27, it can be noted that the output power of the 3 × 3 ECM AC–
AC commutator is 150 kVA (Wheeler et al 2004).
It is very important to minimise the parasitic inductance within the ECM AC–AC
commutator. In this ECM AC–AC commutator, the parasitic inductance has been
minimised using laminate input power planes with distributed input filter
capacitance.
Another interesting example of the design and construction of a static Scherbius
system with a 3 × 3 ECM AC–AC commutator wound-rotor induction dynamotor,
which is used to control the rotor-side currents, is shown in figure 15.28 (Altun
2005).
In such a static Scherbius system, the slip power is taken from the rotor circuit
and given to the grid through the 3 × 3 ECM AC–AC commutator. The static
Scherbius system combines a 3 × 3 ECM AC–AC commutator and a commutator-
less AC wound-rotor induction dynamotor in a slip energy recovery EM propulsion
mechatronic control system, as shown in figure 15.28. This static Scherbius system is
inherently capable of bilateral power flow and also offers virtually sinusoidal input
currents, without the harmonics usually associated with present commercial ECM
AC–AC and AC–DC–AC commutators, the so-called ‘cycloconverters’ and/or
‘inverters’.
In the static Scherbius system, a 3 × 3 ECM AC–AC commutator has been used
to extract the slip energy from the rotor into the mains in the hyposynchronous

15-63
 Mechatronics

Figure 15.27. The 3 × 3 ECM AC–AC commutator induction motor’s output-voltage and output-current
waveforms (Wheeler et al 2004).

region, and to supply the rotor in the hypersynchronous region. It is known that the
use of the 3 × 3 ECM AC–AC commutator in the static Scherbius system has the
advantages of single-stage power conversion, operation in both hyposynchronous
and hypersynchronous speed regions, sinusoidal voltage and current waveforms,
and simpler microelectronic gate-control drivers than other ECM AC–AC and/or
AC–DC–AC commutators (cycloconverters and/or inverters).
At present, the static Scherbius system has an important place in wind-power
generation applications, where a variable-speed, constant-frequency static Scherbius
system produces electrical energy for a wide range of wind speeds.

15.6.3 A 3 × 2 ECM AC–DC commutator

A slightly more complicated example of a 3 × 2 ECM AC–DC commutator
generator (alternator), with a two heterogeneous anode and cathode commutation
groups, i.e. the well-known three-phase bridge rectifier, is shown in figure 15.29
(Fijalkowski 1987).
There are three possible primary collectors (inputs) and two secondary collectors
(outputs) of the DC circuit, which gives a 3 × 2 ECM.

15-64
 Mechatronics

3-Phase A–C Supply

A B C
imA imB imC

isA isB isC

VA VB VC
of AC–AC Matrix Converter
Switching time functions

Va_target SAa SBa SCa

Va Va

Vb_target SAb SBb SCb ira A B C

Vb Vb

Vc_target SAc SBc SCc irb

Vc Vc

Matrix Converter irc IM

Calculation of the target
output voltages of the
matrix converter ωref

Figure 15.28. Slip energy recovery static Scherbius system with a 3 × 3 ECM AC–AC commutator that is a
matrix converter (adapted from Altun 2005).

Figure 15.29. A 3 × 2 electrical commutation matrixer (ECM) AC–DC commutator generator (right)
compared with a 3 × 2 mechanical commutation matrixer (MCM) AC–DC commutator generator (left)
(Fijalkowski 1987, Tutaj 1996).

Because ECMs may be used in ECM AC–AC and AC–DC/DC–AC commutator

dynamotors, two important recent achievements may now be applied with great
success: contactless (sparkless) commutation and contactless (brushless) electro-
magnetic excitation.

15.6.4 A 3 × 3 ECM AC–AC commutator

Three-phase ECM AC–AC commutator hypersynchronous dynamotors (figure 15.30)
with three-phase wound rotors consist of the ECM commutator with a bipolar ECM,
realised on only nine (9) continuous reverse-blocking insulated-gate bipolar transistors

15-65
 Mechatronics

Figure 15.30. Three-phase ECM AC–AC commutator hypersynchronous dynamotor with a 3 × 3 ECM
commutator and three-phase AC wound rotors (Fijalkowski 1988).

(RB IGBT), metal-oxide semiconductor field-effect transistors (MOSFET) or bipolar

symmetrical trigistors, that is, gate turn-off (GTO) symmetrical thyristors (symistors)
in a 3 × 3 ECM connection (Fijalkowski 1988).
This 3 × 3 ECM AC–AC commutator (figure 15.31) also makes the bilateral flow
of secondary electrical energy possible during generation (absorbing) and motoring
(driving) with the hypersynchronous angular velocity.
Six operating modes are possible in this MMD electrical machine.

15.6.5 A single-phase ECM AC–AC and/or AC–DC/DC–AC commutator

The single-phase ECM AC–AC and/or AC–DC/DC–AC commutator synchronous
(electromagnetically excited) dynamotor (figure 15.32) with two three-phase arma-
ture windings, one of which is wye-connected, and another of which is delta-
connected, consists of a non-rotary main ECM commutator with two bipolar
ECMs, realised on 12 continuous bipolar trigistors in a 2 s(2 × 3)/2 s(3 × 2) ECM
connection in series, and a rotary auxiliary ECM commutator with a unipolar ECM,
realised on six (6) continuous diodes in a 2 × 3 ECM connection. It has a contactless
(brushless) electromagnetic excitation.
The single-phase ECM AC–AC and/or AC–DC/DC–AC commutator synchro-
nous (magnetoelectrically excited) dynamotor (figure 15.33) consists of an ECM
commutator with a single or two bipolar ECMs, realised on six (6) or 12 continuous
bipolar trigistors in a 2 × 3 ECM connection or a 2 s(2 × 3) ECM connection in
series, by means of permanent-magnet poles (for example, alloys of neodymium,
iron and boron).
The single-phase ECM AC–AC and/or AC–DC/DC–AC commutator hyposyn-
chronous (induction) dynamotor (figure 15.34) with a squirrel-cage rotor, consists of

15-66
 Mechatronics

Figure 15.31. Three-phase 3 × 3 ECM AC–AC commutator with RB IGBT.

Figure 15.32. A single-phase ECM AC–AC and/or DC–AC/AC–DC commutator synchronous (electro-
magnetically excited) dynamotor with a 2 s(2×3)/2 s(3×2) ECM commutator (Fijalkowski 1988, Tutaj 2012).

an ECM commutator with a single or two bipolar ECMs, realised on six (6) or 12
continuous bipolar trigistors in a 2 × 3 ECM connection or a 2 s(2 × 3) ECM
connection in series.

15.6.6 A 2 × 3/3 × 2 ECM DC–AC/AC–DC commutator and 2 × 2 ECM DC–DC

commutator
An advantage of the hybrid-electric drive-by-wire (DBW two-wheel-drive (2WD)
and/or four-wheel-drive (4WD) propulsion mechatronical control system is that a
novel form of external combustion engine (ECE) or internal combustion engine (ICE)
can be used, designed specifically for high-efficiency constant engine-shaft angular
velocity and power operation and minimum emission (Fijalkowski 1997).

15-67
 Mechatronics

Figure 15.33. A single-phase ECM AC–AC and/or AC–DC/DC–AC commutator synchronous (magneto-
electrically excited) dynamotor with a 2 × 3/3 × 2 or 2 s(2 × 3)/2 s(3 × 2) ECM commutator (Fijalkowski 1988).

Figure 15.34. A single-phase ECM AC–AC and/or AC–DC/DC–AC commutator hyposynchronous (induc-
tion) dynamotor with a 2×3/3×2 or a 2(2×3)/2(3×2) ECM commutator and squirrel-cage rotor (Fijalkowski
1988).

The DBW 4WD propulsion mechatronical control system using an EM/ME

steered, motorised and/or generatorised wheel (SM&GW) with ECM DC–AC/AC–
DC commutator wheel-hub dynamotors will eliminate the need for axles and drive
shafts, which will allow a separate suspension for each independently sprung
SM&GW (figure 15.35). The rotating housing is made in the form of a wheel
hub, designed to fit standard rim sizes on the market. The ECM DC–AC/AC–DC

15-68
 Mechatronics

Figure 15.35. Layouts of three different independently sprung, planetary-gearless steered, motorised and/or
generatorised wheels (SM&GW) with 2 × 3/3 × 2 ECM DC–AC/AC–DC commutator in-wheel-hub
dynamotors and tubular, linear 2 × 2 ECM DC–DC commutator wheel-drum or wheel-disc brake-actuator
motors (Fijalkowski 1997).

commutator wheel-hub motor’s transmission offers complete freedom in the design

of the hybrid-electric vehicle (HEV).
A DBW 2WD and/or 4WD propulsion mechatronical control system may
provide the HEV with good climbing ability (up to π /4 rad on firm surfaces). The
two SM&GWs on each side of the HEV’s hull-wheeled unit may have an
independent absorb-by-wire (ABW) four-wheel-absorbed (4WA) suspension.

15-69
 Mechatronics

Figure 15.36. Layout of the steered, motorised and/or generatorised wheel (SM&GW) (A) and the physical
model of the ECM DC–AC/AC–DC commutator reluctance and IPM magnetoelectrically excited in-wheel-
hub motor/generator (B) (Fijalkowski 1984, 1990).

15.6.7 A 2 × 5/5 × 2 ECM DC–AC/AC–DC commutator

The two front and the two rear SM&GWs (figure 15.36) are driven and/or absorbed
individually by four 2 × 5/5 × 2 ECM DC–AC/AC–DC commutator reluctance and
IPM magnetoelectrically excited in-wheel-hub dynamotors, respectively, and the
angular velocity of each SM&GW may be arbitrarily controlled by a driver-vehicle
and a terrain-vehicle real-time expert system, incorporating a mathematical model
following fuzzy-logic (FL) programmable and neural-network (NN) learning motion
mechatronical control (Fijalkowski 1990, 2010).

15.6.8 A 2 × 3 ECM DC–AC/AC–DC commutator and 2 × 2 DC–DC commutator

The author considered a novel mobility and steerability concept for HEVs with EM
differentials. The use of these novel triad hybrid mobility and triad hybrid steer-
ability enhancing concept DBW 2WD and/or 4WD propulsion mechatronical
control systems for HEV (figure 15.37) opens up wide possibilities for improving
fossil and non-fossil fuel economy, cutting initial and whole-life DBW 4WD
propulsion costs, protecting the environment and improving the distribution of
terrain thrust (gross tractive effort), as well as keeping the net motion resistance of
the DBW 4WD propulsion and the SM&GW sinkage (rut depth) low—not only by
increasing the velocity of travel, but also by decreasing the rolling resistance of all
the SM&GWs.
HEVs may be able to retain a sufficient level of mobility even if they lose one or
more SM&GWs. Using EM differential torque and/or angular-velocity controls of
the outer SM&GWs, as well as current and/or voltage controls of the inner
SM&GWs, can increase the lateral motion control effect, especially when recovering
from braking with the inner SM&GWs acting as the ECM AC–DC commutator in-
wheel-hub generators, because the front gravitational forces on the HEVs become
greater than the respective rear ones. At the same creep, this leads to greater
horizontal (longitudinal and lateral) forces.
The DBW 4WD propulsion mechatronical control systems used on HEVs satisfy
nearly all the same requirements that the running gear systems used on conventional
vehicles have (Fijalkowski 1986, 1990, 1994, 2010):
• to apply an EM ‘single-shaft’ DBW 4WD propulsion mechatronical control
system to a complete number of SM&GWs;

15-70
 Mechatronics

Figure 15.37. Elementary wiring connection for the axleless, full-time DBW 4WD transmission arrangement
with the torque proportioning front-wheel drive (FWD) and rear-wheel drive (RWD), as well as centre-wheel
drive (CWD) electro-mechanical (EM) differentials (Fijalkowski 1997, 2010).

• to allow a positive propelling (driving) torque to the outer side of the curve
and a negative dispelling (braking) torque to the inner side, achieving their
maximum value for pivot-skid steering;
• to occupy the minimum volume within the space envelope of the HEV;
• to distribute the mass of the HEV over a relatively spacious ground surface or
soil area.

The requirement of the first feature may contribute to a very good soil perform-
ance for HEVs. The third feature may tend to conflict with requirement of the first
feature.

15.6.9 A 2 s(2 × 5)/2 s(5 × 2) ECM DC–AC/AC–DC commutator

The layout of a particularly interesting planetary-gearless SM&GW with a wire-
mesh tyre for the Martian Roving Vehicle (MRV) with a 2 s(2 × 5)/2 s(5 × 2) ECM
DC–AC/AC–DC commutator reluctance and a magneto-electrically excited
(unwound MSI single ring-shape outer-rotor and wound and IPM twin ring-shaped
inner stator) in-wheel-hub motor/generator is shown in figure 15.38 (Fijalkowski
2001a, 2001b).

15-71
 Mechatronics

Figure 15.38. Planetary-gearless steered, motorised and/or generatorised wheel (SM&GW) with a 2 s(2 × 5)/2 s
(5 × 2) ECM DC–AC/AC–DC commutator reluctance and a magnetoelectrically excited in-wheel-hub motor/
generator (Fijalkowski 2001b).

It is evident that there are no windings on the mild soft iron (MSI) single ring-
shaped outer-rotor, while there are no slip rings and there is no MCM DC–AC/AC–
DC commutator with sliding-copper-segments and carbon brushes. What is not
evident is the nature of the twin ring-shaped inner-stator five-phase windings. If the
twin ring-shaped inner-stator is a six-pole, five-phase structure with a large number
of ‘teeth’, then the single ring-shaped outer-rotor will also have teeth and it will be
magnetised so that the south pole occupies one-half of the periphery, while the north
pole occupies the other half. Both members afford a large number of opportunities
for positional lock-up of the single ring-shaped outer rotor. Although the twin ring-
shaped inner-stator only has six poles, and the single ring-shaped outer-rotor only
has three poles, the presence of teeth on both members affords a large number of
opportunities for positional lock-up of the single ring-shaped outer-rotor.
The 2 s(2 × 5)/2 s(5 × 2) ECM DC–AC/AC–DC commutator reluctance and the
magnetoelectrically excited in-wheel-hub motor/generator with different pole num-
bers on the unwound MSI single ring-shaped outer-rotor, and the wound and IPM
ring-shaped inner-stator, can provide traction if the stator five-phase winding phase-
coils are sequentially energised. Here the unwound MSI single ring-shaped outer-
rotor is made to ‘chase’ the sequentially switched wound and IPM twin ring shaped
inner-stator’s poles. As soon as magnetic alignment occurs, the attracting pole is de-
energised. A variation of this basic scheme employs Hall-effect elements to directly
sense the position of either the single ring-shaped outer-rotor itself, or a suitably
magnetised ring on the shaft.

15.7 Electrical commutation matrixer commutators—a look into the

future
The ECM commutator (conceived by the author in 1988) represents the ‘super-
conductive ECM commutator’ in which all the active (continuous bipolar electrical
valves) and reactive elements (primary and secondary collectors, i.e. connections) of

15-72
 Mechatronics

the ECM are manufactured in a super- and/or semiconducting single wafer of

amorphous material in the shape of single chips (Fijalkowski 1988).
In 1987, the Nobel Prize in the field of physics was awarded to the Swiss physicist,
K A Müller, and the German physicist, J G Bednorz, who both worked for IBM in
Zürich. In early 1987, they discovered the properties of electrical-current conduc-
tivity (a rapid loss of resistance) in barium, lanthanum, copper and oxygen
compounds at a temperature of 35 K. A little later, the American physicist,
P Chu, who followed the path indicated by Müller and Bednorz, went beyond the
‘magic’ limit of the nitrogen boiling point temperature of 77 K.
Today, the electrical-current superconductivity phenomenon has already been
registered even at the ambient temperature of the natural (human) environment. The
methods for approaching electrical-current superconductivity have changed com-
pletely. Amorphous materials may be used similarly to mono- and polycrystalline
conductors (e.g. metals), non-conductors and superconductors (Kazmerski 1980).
The electrical-current conductivity of amorphous germanium and silicon can be
controlled easily by means of adequate doping, and the remaining properties of these
materials are in principle analogous to those existing in their mono- and polycrystal-
line equivalents. Thus, it is possible to perform junctions of the p–n type from
amorphous germanium and silicon, which constitute the basic elements of the
majority of semiconductor electrical valves.
Cheap and easy to attain, amorphous material can be used for the construction of
both active (super- and/or semiconductor amorphous continuous bipolar electrical
valves) and reactive elements (superconductor amorphous primary and secondary
collectors, i.e. connections).
If the transistor revolution took place due to the discovery of the properties of
mono- and polycrystalline semiconductors, the next revolution in electronics will
make use of semiconductor amorphous materials.
From the point of view of industrial and traction technology, amorphous
materials are characterised above all by their enormous heterogeneity of physical
properties and by the fact that they require no complex technological processes.
Amorphous materials are usually manufactured in the shape of glasses by means
of rapid quenching of the material from the liquid phase in the shape of thin films,
which are vacuum evaporated on adequate quenched substrates. These technologies
are distinctly competitive compared to the manufacturing technologies for mono-
and polycrystalline materials.
ECM AC–AC and AC–DC/DC–AC commutators with bipolar ECMs, of which
primary and secondary collectors are realised on ceramic superconductors, are
contributing to one more revolution in the history of both macroelectronics and
ECM AC–AC and AC–DC/DC–AC commutator dynamotors. Although the ECM
AC–AC and AC–DC/DC–AC commutator with a bipolar ECM, realised on
continuous bipolar ovonics, is made of glassy material that does not have a
crystalline structure, it is a super- and/or semiconductor static converter, which
provides a solution to the problem of its cooling.
Continuous bipolar ovonics are characterised by the short time-lag of the
transition between the on state (conduction-state) of the electrical valve and its off

15-73
 Mechatronics

state (non-conductive-state) of 0.1 μs. The ratio of the off-state to the on-state
resistance yields 107 and more.
The triggering signal applied to the triggering gate may have an arbitrary
polarisation. Continuous bipolar ovonics can operate at a frequency below 500 kHz
and at an ambient temperature range up to 573 K (Ovshinsky 1958; Young 1961;
Cohen et al 1969; Ovshinsky 1970; Holmberg and Shaw 1974; Ovshinsky and Adler
1978; Kazmerski 1980).
Continuous bipolar ovonics, under the influence of some control voltage, the
value of which depends upon the chemical composition and dimensions of the glassy
wafer, undergo a metamorphosis from a resistor with a resistance of 107 Ω into a
conductor with a resistance of 10−2 Ω.
After the metamorphosis into a conductor, a relatively high electrical current
starts to flow in the ovonic. As long as the electrical current is flowing, the glassy
wafer remains a conductor. It maintains this ability without any change, even when
the electrical current changes its instantaneous value over a small range. However, a
single power pulse opposite to the primary is sufficient for the amorphous conductor
(without crystalline structure) to undergo a metamorphosis back into a resistor with
a high resistance.
Since the electrical current practically does not flow in this resistor, the heat is not
released in it. However, if the ovonic metamorphosises itself into a conductor and
conducts the electrical current, its resistance is so insignificant that the quantity of
heat that is released can be neglected.
In this way, the dream of electronics scientists and engineers to make a great
number of continuous electrical valves on the single carrying wafer and to realise in
practice uni- and/or bipolar ECMs for the ECM commutators becomes realisable.

15.8 Conclusion
Macroelectronics is the study of ECM commutators for the conversion and control
of electrical energy. The technology is a critical part of energy infrastructure and
supports almost all important electrical applications. For macroelectronics design,
one considers only those matrixers and devices that, in principle, introduce no loss
and can achieve near-perfect reliability. The two key characteristics of high
efficiency and high reliability are implemented with ECMs, supplemented with
energy storage. ECMs in turn can be organised as electrical-valve matrixers. This
facilitates their analysis and design.
In a macroelectronic continuous dynamical system, the three primary challenges
are the hardware problem of implementing an ECM, the software problem of
deciding how to operate that ECM, and the interface problem of removing
unwanted distortion and providing the user with the desired clean electrical-energy
source. The hardware is implemented using special types of power super- and/or
semiconductors. These include several types of SiC and AsGa diodes, n–p–n and
p–n–p transistors, especially MOSFETs, IGBTs, RB IGBT, OT IGBTs and HV
IGBTs, and several types of thyristor, especially SCRs, PPTs, PPCSTs and GTOs.
The software problem can be represented in terms of commutation (switching)

15-74
 Mechatronics

functions. The frequency, duty ratio and phase of the commutation functions are
available for operational purposes. The interface problem is addressed by means
of lossless filters. Most often, these are lossless LC passive filters to smooth out
ripples or reduce harmonics. More recently, active filters have been applied to
make dynamical corrections in power conversion waveforms.
For the time being, it is difficult to pronounce optimistic opinions on the
prospects for practical applications of ECM commutators. The author believes
that scientists should not promise users too much. As an example to put users into an
optimistic mood, one ought to mention the invention of the transistor by J Bardeen,
W G Brattain and W B Shockley in 1948. At that time, people assumed that the
application of transistors resolved itself into the replacement of electron lamps,
especially triodes. Nobody could foresee that soon after this invention the unpre-
cedented miniaturisation of electronics and the creation of EICs with larger scales of
integration would follow. In the author’s opinion, such a thing can occur with ECM
commutators, but for the time being it is impossible to foresee some of the future
practical applications.
The results of the invention of ECM commutators and the R&D work carried out
in this field were that solid-state physics began life anew. Not long ago, people
believed that in the field of power electronics everything had already been invented,
whereas this is only half-true. Recently, there has been enormous demand for R&D
work in the field of solid-state physics.
Scientific research concerning ECM and commutators will be transformed into a
great race by scientific teams from great world consortia and its results, which will
revolutionise the universe of macroelectronics, may be expected at any moment.
Very similar races by scientific teams have already taken place today in the field of
solid-state physics, where a series of discoveries concerning high-temperature
electric-current conductivity in materials of a new kind took place.
These particular experiments terminated in the invention of ECM commutators
with bipolar ECMs, realised on continuous bipolar ovonics; hybrid transistors with
heterojunctions; SUPER-HETs with a superconductive base; QUITs; and very high-
speed SITs.
These ECM commutators will be used in newly designed energy-saving MMD
electrical machines with the application of high-temperature superconductors, for
example, made from Y–Cu–Ba–O ceramics. Superconductivity is an exciting and
rapidly developing technology, especially since the recent discovery of high TC
materials.
The basic technology and performance of IECs based on Nb Josephson junctions
were implemented, and then followed by an assessment of the prospects for IECs
incorporating high-TC materials. These include uni- and bipolar semiconductor
triodes (three-terminal electrical valves) and hybrid semiconductor/superconductor
electrical valves, with the challenge being to compete successfully with current
semiconductor technology, which is itself developing rapidly. ECM commutators
are now indispensable in modern MMD, MEM, MFD and MPD electrical machine
engineering, and for energy converting processes in industrial and traction propul-
sion complex dynamical hypersystems. Compared with the previous polyvalvular

15-75
 Mechatronics

technology, ECM commutators (static converters) are more practice-oriented,

especially from the point of view of operating convenience and the space require-
ments of ultra-modern semiconductor/superconductor ECM techniques.
A major step forward has been taken within electrical machinery technology
with the development of macroelectronic commutators, called ‘ECM commutators’
by the author. The ECM commutator is a new product family based on the ECM
and the EIC computer controller. The latest advances made within the field of
macroelectronic components, such as powerful ECM commutators, and in the field
of 64-bit, 128-bit and 256-bit microprocessors, have been exploited during the R&D
work.
Components in the form of hardware and software have been developed with the
aid of these new macro-, meso-, micro- and nanoelectronic components, and
adapted to new modern control system structures.
Today, the ECM on its own may be used for all the new-concept integral ECM
AC–AC. AC–DC–AC, AC–DC/DC–AC, DC–DC and DC–AC–DC commutator
dynamotor’s industrial and traction propulsion complex dynamical hypersystems,
previously covered by several earlier modular dynamical hypersystems. The method
used to build industrial and traction propulsion complex dynamical hypersystems
has varied over the years, depending on the types of components that were available.
A changeover from extremely distributed industrial and traction propulsion
complex dynamical hypersystems, where each object had its own ECU, to modern
integrated industrial and traction propulsion complex dynamical hypersystems took
place. The torque and speed control functions were grouped together in ECM
commutator dynamotors to provide central access to all information.
The world must solve major challenges to transform and control electrical energy
in an efficient way. Examples of this can be seen in aerospace, automotive and
railway transportation applications, and in renewable energy and industrial proc-
essing applications. These problems can be solved using ECM AC–AC, AC–DC–
AC, AC–DC/DC–AC, DC–DC and DC–AC–DC commutators based on modern
power super- and/or semiconductor amorphous electrical valves.
The ideal ECM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and DC–
AC–DC commutator in many of these applications might have the following
characteristics:
• compact design with a good power-to-mass ratio;
• regeneration capability;
• operation with unity power factor;
• sinusoidal input and output currents.

ECM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and DC–AC–DC

commutators can fulfil all of these characteristics, and this is the reason for the
tremendous interest in this matrix topology.
Over the last decade, many advances in the development of this matrix topology
were presented, including industrial applications up to the megawatt (MW) level.

15-76
 Mechatronics

The use of ECM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and DC–

AC–DC commutators in real applications and the challenges that these applications
present is very topical and important.
This book will present the most recent advances with topics to the macro-, meso-,
micro- and nanoelectronics community. These include, but not limited to, the
following.
• ECM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and DC–AC–DC
commutators for aerospace, automotive and railway transportation applica-
tions, and renewable energy and industrial applications.
• ECM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and DC–AC–DC
commutator derived matrix topologies.
• New super- and/or semiconductor amorphous electrical valves for use in
ECM AC–AC, AC–DC–AC, AC–DC/DC–AC, DC–DC and DC–AC–DC
commutators.
• Power quality issues and ECM AC–AC, AC–DC–AC, AC–DC/DC–AC and
DC–DC and DC–AC–DC commutator reliability and stability.
• Implementation of artificial intelligence (AI) current commutation strategies
in application examples.
• New control/modulation methods for ECM AC–AC, AC–DC–AC, AC–DC/
DC–AC, DC–DC, DC–AC–DC, DC–DC or DC–AC–DC commutator
applications, including field-oriented control (FOC), space vector modulation
(SVM), direct torque control (DTC), passively based control (PBC) and
predictive control (PC).
• Comprehensive comparative evaluation of the ECM AC–AC and AC–DC/
DC–AC commutator concepts versus the ECM AC–DC–AC or DC–AC–
DC commutator concepts, with DC or AC link energy storage components,
respectively.

However, the practical applications of the 3 × 3 ECM AC–AC, commutator, as

of now, are very limited.
The main reasons are:
• non-availability of bilateral fully controlled monolithic electrical valves
capable of high-frequency operation;
• complex control law implementation;
• an intrinsic limitation of the output-input voltage ratio;
• commutation (Case 2000) and protection of the switches.

To date, the electrical valves are assembled from existing continuous devices,
resulting in increased costs and complexity, and only experimental ECMs with a
capacity well below 1 MVA have been built. However, with advances in device
technology, it is hoped that these problems will eventually be solved, and 3 × 3 ECM
AC–AC super or static commutators will not only replace naturally commutated
cyclo-converters in all the applications, but will also take over from the PWM
rectifiers/inverters.

15-77
 Mechatronics

In Mahlein et al (1999), it was shown that with space-vector (SV) PWM control
using over-modulation, the voltage transfer ratio may be increased to 1.05 at the
expense of more harmonics and large filter capacitors.
Improvements in devices and advances in control concepts have led to steady
improvements in ECMs and macroelectronic continuous dynamical systems. This is
driving a tremendous expansion of their application. For instance, PCs would be
unwieldy and inefficient without macroelectronic DC energy supplies. Portable
communication devices and computers would be impractical. High-performance
HVDC transmission links, ME/EM generator/motor (voltage/torque) controls,
lighting systems and a wide range of industrial controls depend on macroelectronics.
In the near future, one can expect strong growth in automotive applications, in
DC energy supplies for communication systems, in portable applications and in
high-end ECMs for advanced microprocessors.
During the next generation, we will reach a time when almost all electrical energy
is processed through macroelectronics somewhere in the path from generation to end
use.

References
Alesani A and Venturini M G B 1989 Analysis and design of optimum-amplitude nine-switch-
direct AC–AC converters IEEE Trans. Power Electron. 4 101–12
Alexanderson E F W and Mittag A H 1934 The ‘Thyratron’ Motor Electr. Eng. 53 1817
Altun H 2005 Application of a matrix converter in a slip energy recovery drive system Eng.
Modelling 18 3–4, 69–80
Anon 1950 Professor AN Larionov k 60-letiiu so dnia rozhdeniia i 30-letiiu nauchno-pedago-
gicheskoi deiatelnosti Elektrichestvo 9 (In Russian)
Casadei D, Serra G, Tani A and Zarri L 2002 Matrix converter modulation strategies: a new
general approach on space-vector representation of the switch state IEEE Trans. on Ind.
Electron. 49
Case M J 2000 The commutation process in the matrix converter Conf. Proc. EPE-PEMC 2000
(Kosice, Slovakia) 2 109–12
Cheron Y, Foch H and Perin A 1985 An improved direct frequency changer using power
transistors Proc. First European Conf. on Power Electronics and Applications 1 1.123–8
Cohen M H, Fritzsche H and Ovshinsky S R 1969 Phys. Rev. Lett. 22 1065
Fijalkowski B 1962 Sterowanie maszyny wyciagowej pradu stalego z przeksztaltnikami rtęciowymi
(Control of the DC mine winder using mercury-arc converters) Zeszyty naukowe Politechniki
Szczecinskiej, Nr 36, Prace Monograficzne, Nr 9 (Politechnika Szczecinska: Szczecin) p 59
(In Polish)
Fijalkowski B 1964 Sterowany przeksztaltnikowy naped elektryczny i perspektywy jego rozwoju
(Converter drive and perspectives of its development) Zeszyty Naukowe Politechniki
Szczecinskiej Elektryka VI, Nr 53, 1964, pp 6–34 (In Polish)
Fijalkowski B 1967 Zastosowanie rachunku holorowego w teorii pradow przemiennych
(Application of a holor calculus in the theory of alternating currents) Przeglad elektrotech-
niczny (Electrotechnics Review) Rok XLIII, Zeszyt 2, Warszawa (In Polish)
Fijalkowski B 1973 Kierunki rozwoju napedu elektrycznego lokomotyw (Development trends of
the electric drive of locomotives) Materiały konferencyjne—I Krajowa Konferencja—Pojazdy

15-78
 Mechatronics

Szynowe (Krakow: Instytut Pojazdow Szynowych Politechniki Krakowskiej) pp 109–128

(In Polish)
Fijalkowski B 1984 Electronic commutator AC/DC motor-driven tracked all-terrain vehicles with
extremely high mobility Proc. 8th Int. ISTVS Conf.: The performance of off-road vehicles and
machines (Churchill College, Cambridge University, 5–11 August 1984) J. Terramech 1045–63
Fijalkowski B 1985a New concept MACRO- and MICRO-electronics cradled dynamometer
systems for testing of combustion engines Proc. ISATA 85: Int. Symp. on Auto-motive
Technology and Automation (Graz, Austria) 2 587–616
Fijalkowski B 1986 Future hybrid electromechanical very advanced propulsion systems
for civilian wheeled and tracked all-terrain vehicles with extremely high mobility Proc.
1986 EVS-8: The Eight International Electric Vehicle Symp. (Washington DC) pp 428–43
Fijalkowski B 1987 Modele matematyczne wybranych lotniczych i motoryzacyjnych mechano-
elektro-termicznych dyskretnych nadsystemów dynamicznych (Mathematical Models of
Selected Aviation and Automotive Continuous Dynamical Hyperystems) (Krakow:
Monografia 53 Politechnika Krakowska imienia Tadeusza Kościuszki) p 276 (In Polish)
Fijalkowski B 1988a Choice of hybrid propulsion systems—wheeled city and urban vehicle—
tracked all-terrain vehicle. Part II Electr. Vehicle Dev. 7 31–4
Fijalkowski B 1988b Odnochipnyi makrokommutator—Tehnologiia v stadii razvitiia—Uzhe nie
teoriia, no eshche ne promyshlennost’ Zbornik Prednasok z VIII. celostatnej konferencje so
zahranicnou ucastou Elektricke Pohony A Vykonova Elektronika (Kosice, Kosicka Bela,
Czecho-Slovakia: I. DIEL) pp 182–94 (In Russian)
Fijalkowski B 1988c Single-chip static-commutator—an indispensable module of new concept AC
and/or DC dynamotors for automotive very advanced propulsion systems Proc. EVS9: The
9th Int. Electric Vehicle Symposium (Toronto, 13–16 November 1988) EVS88-031 1-6
Fijalkowski B 1988d Amorphous and polycrystalline semiconductor single-chip super-
commutator—an indispensable module of new concept AC and/or DC electrical machines
Proc. Int. Conf. and Intensive Tutorial Course on Semiconductor Materials (New Delhi) 8–16
Fijalkowski B 1990 Very advanced propulsion spheres for high-speed tracked vehicles Proc. of the 10th
Int. Conf. of the International Society for Terrain-Vehicle Systems (ISTVS) (Kobe) 3 783–97
Fijalkowski B 1994 Front- or rear-wheel-drive and inter-wheel-drive electromechanical differ-
entials for full time 4WD × 4WB × 4WA × 4WS supercars with steerable & motorized wheels
Proc. ISATA 94: Int. Dedicated Conf. on ELECTRIC, HYBRID & ALTERNATIVE FUEL
VEHICLES and SUPERCARS (ADVANCED ULTRALIGHT HYBRIDS) in conjunction
with the 27th Int. Symp. on Automotive Technology and Automation, Aachen (Germany, 31
October–4 November 1994) Paper No. 94UL024 681–8
Fijalkowski B T 1982 Trigistor frequency changer-controlled toothed gearless propulsion systems
for electric and hybrid vehicles Proc. Europe’s Int. Conf. on Electric Road Vehicle Systems
DRIVE ELECTRIC (Amsterdam) 554–67
Fijalkowski B T 1985b On the new concept hybrid and bimodal vehicles for the 1980s and 1990s
Proc. DRIVE ELECTRIC Italy ’85 (Sorrento) pp 4.04.2–8
Fijalkowski B T 1985c On the new concept MMD electrical machines with integral macro-
electronic commutators—Development for the future Proc. First European Conf. on POWER
ELECTRONICS AND APPLICATIONS (Brussels, 16–18 October 1985) 2 pp 3.377–3.384
Fijalkowski B T 1997 Intelligent Automotive systems: development in full-time chassis motion
spheres for intelligent vehicles Advanced Vehicles and Infra-structure Systems—Computer
Applications, Control and Automation ed C O Nwagboso (New York: Wiley) pp 125–142

15-79
 Mechatronics

Fijalkowski B T 2001a Articulated triad Martian roving vehicle The Fourth Int. Mars Society
Convention (Stanford, CA) (Lakewood, CO: The Mars Society Inc.)
Fijalkowski B T 2001b A concept of the All-Electric Pressurized Articulated Triad Martian Roving
Vehicle On To Mars 2 ed R M Zubrin and F Crossman (Lakewood, CO: The Mars Society)
Fijalkowski B T 2010 Automotive Mechatronics: Operational and Practical Issues.— Vol 1
(International Series on Intelligent Control, and Automation Science and Engineering vol. 47)
(Berlin: Springer)
Fijalkowski B T 2011 Automotive Mechatronics: Operational and Practical Issues.— Vol. 2
(International Series on Intelligent Control, and Automotive Science and Engineering vol 52)
(Berlin: Springer)
Gyugyi L and Pelly B R 1976 Static Power Frequency Changers: Theory, Performance and
Application (New York: Wiley)
Holmberg S N and Shaw M P 1974 Proc. of the 5th Int. Conf. Amorphous Liquid Semiconductors
(London: Taylor and Francis)
Hoseman G, Krolikowski L, Raatz E, Shapilov E D, Turek-Kwiatkowska L and Vieselovsky O N
1999 Zhizn i delyatelnost Mikhaila Dolivo-Dobrovolskogo—vydaiushchegosia inzhenera
elektrika evropeiskogo masshtaba Proc. of the 4th Int. Conf. on Unconventional
Electromechanical and Electrical Systems (St Petersburg) 1 59–70 (In Russian)
Huber L and Borojevic D 1995 Space vector modulated three-phase to three-phase matrix
converter with input power factor correction IEEE Trans. Ind. Appl. 31 1234–46
Kazmerski L L (ed) 1980 Polycrystalline and amorphous thin films and devices (New York: Academic)
Kern E 1938 Der Dreiphasenstromrichtermotor und seine Steuerung bei Betrieb als Umkehr-
motor ETZ 59 467 (In German)
Kittel Ch 1956 Introduction to Solid State Physics (New York: Wiley)
Kloninger M J 1932 Applications des redresseurs á grilles polarisées. Séance S.F.E. 6 Fevrier
Bulletin S.F.E. Fev. (In French)
Kostenko M and Piotrovsky L 1968 Electrical Machines 2 vols (Moscow: Mir Publishers) vol 1
chs 5 and 6 pp 201–67; vol 2 ch 27 649–54
Li Y and Choi N-S 2009a Carrier based pulse width modulation for matrix converter Conf.
Record of IEEE Applied Power Electronics Conf. 1709–15
Li Y, Choi N-S and Han B-M 2009b DDPWM based control of matrix converters. Power
Electron. 9 535–43
Li Y, Choi N S, Han B-M, Kim K-M, Lee B and Park J-H 2008 Direct duty ratio pulse width
modulation method for matrix converter Int. J. Control Autom. Syst. 6 660–9
Mahlein J, Simon O and Braun M 1999 A matrix converter with space vector control enabling
overmodulation Conf. Proc. EPE 99 (Lausanne) pp 1–11
Mott M F and Davis E A 1971 Electronic Processes in Mono-crystalline Materials (London and
New York: Oxford University Press)
Neft C L and Schauder C D 1988a Theory and design of a 30-hp matrix converter Conf. Record of
the 1988 Industry Applications Society Annual Meeting (Pittsburgh, PA) (IEEE Cat. No.
88CH 2565-0)
Neft C L and Schauder C D 1988b Theory and design of a 30-hp matrix converter Proc. of the
IEEE Industrial Application Society Annual Meeting (Pittsburgh, PA) pp 248–53
Neiman L R 1972 Akademik Vladimir Fedorovich Mitkevich ego trudy i progressivnye idei.
K 100-letiiu so dnia rozhdeniya Elektrichestvo 8 (In Russian)
Ovshinsky S R 1958 Phys. Rev. Lett. 21 1450

15-80
 Mechatronics

Ovshinsky S R 1970 J. Non-Cryst. Solids 2 99

Ovshinsky S R and Adler D 1978 Contemp. Phys. 10 100
Oyama J, Higuchi T, Yamada E, Koga T and Lipo T 1989 New control strategy for matrix
converter Record of the 20th Annual IEEE Power Electronics Specialists Conf.—PESC ‘89
(Milwaukee, WI) Cat. No.89 CH2721-9 pp 360–7
Pollak Ch 1895 English Patent (24,398 A.D)
Pollak Ch 1896 Elektrische Flüssigkeitskondensator mit Aluminium-elektroden. Kaiserliches
Patentamt, Patentschrift No 92564, Klasse 21: Elektriche Apparate, Patentirt im
Deutschen Reiche vom 14. Januar 1896 ab. Ausgaben den 19. Mai 1897.
Rémy E 1954 Le collecteur magnétique Bulletin E.S.E.,Nõel 1954, et - Dynamo sans balais. Revue
court-circuit de l’amicale E.S.E (In France)
Rodriguez J 1984 A new control technique for AC–AC converter Proc. of the 3rd IFAC Symp. on
Control in Power Electronics and Electrical Drives (Lausanne) p 203
Schilling W 1940 Die Wechselrichter und Umrichter—Ihre Berechnung und ArbeitweiseVerlag von
R Oldenbourg) (In German)
Seitz P 1940 Modern Theory of Solids (New York: McGraw-Hill)
Strzelecki R M and Benysek G (ed) 2008 Power Electronics in Smart Electrical Energy Networks
(Berlin: Springer)
Tenti P, Malesani L and Rossetto L 1992 Optimum control of N-input K-output matrix
converters IEEE Trans. Power Electron. 7 707–13
Tutaj J 1996 Estymacja jakosci energii samochodowych komutatorowych pradnic mechano-
elektrycznych pradu stalego z komutatorami elektronicznymi (Energy quality estimation of
DC mechano-electrical generators with electronic commutators) PhD Thesis (Promotor:
Bogdan Fijalkowski, Automotive Vehicles and Combustion Engines Institute, Cracow
University of Technology, Krakow) (In Polish)
Tutaj J 2012 Ujecie systemowe dynamiki wielofunkcyjnego pradnico-rozrusznika silnika spalino-
wego pojazdu samochodowego (Dynamical Systems Approach of a Multifunctional Generator/
Starter for the Combustion Engine of an Automotive Vehicle) Scientific Editor: Bogdan
Fijalkowski, Seria Mechanika, Monografia, 409, Wydawnictwo Naukowe Politechniki
Krakowskiej im. Tadeusza Kosciuszki, Krakow (In Polish)
Wang B and Venkataramanan G 2006 Six Step Modulation of Matrix Converter with Increased
Voltage Transfer Ratio Proc. of the 37th IEEE Power Electronics Specialists Conf. (Jeju,
Korea) 49 930–6
Wheeler P W, Clare J C, Katsis D, Empringham L, Bland M and Podlesak T 2004 Design and
construction of a 150 kVA matrix converter induction motor drive Proc. of the 2nd IEE Int.
Conf. on Power Electronics Machines and Drives (Edinburgh) 498
Willis C H 1933 A study of the thyratron commutation motor Gen. Elect. Rev. 76
Yoon Y-D and Sul S K 2006 Carrier-based modulation technique for matrix converter IEEE
Trans. Power Electron. 21 169–703
Young L 1961 Anodic Oxide Films (New York: Academic)
Zagorsky V T 1969 Tekhniko-ekonomicheskiye pokazateli neposredstvennykh preobrazo-vateley
chastoty s prinuditelnoy kommutaciey Elektrichestvo pp 35–41 (In Russian)
Zavalishin D A and Vegner O G 1936 Ionnye skhemy ventilnykh dvigateley Elektrichestvo 3 1936
(In Russian)
Ziogas P D, Khan S I and Rashid M H 1986 Analysis and design of forced commutated
cycloconverter structures with improved transfer characteristics IEEE Trans. Ind. Electron.
33 271

15-81
 IOP Publishing

Mechatronics
Dynamical systems approach and theory of holors
B T Fijalkowski

Chapter 16
Electrical commutation matrixer keyboards
for computers

People who are really serious about software should make their own hardware.
Alan Kay
Information: the negative reciprocal value of probability.
Claude Shannon

16.1 Introduction
In this chapter, electrical commutation matrixer (ECM) keyboards for computers are
introduced. Computers are electronic machines that take input from a human user,
process the given input and generate output in the form of useful information. ECM
keyboards are the most commonly used input devices. One can use ECM keyboards
to type data and text and execute commands (see figure 16.1).
Figure 16.2 illustrates a typical laptop computer ECM keyboard layout, where all
the common keys found on the large desktop computer ECM keyboard have been
miniaturised to laptop size. The design is intended to let human users type without
the risk of friction from their fingers rubbing together.
The layout of the keys was borrowed from the typewriter with a number of new
keys added. Various ECM keyboards have different extra keys. An ECM keyboard
for a Windows computer has different extra keys than one for an Apple computer.
Laptop computers have fewer keys, but have an additional ‘Fn’ key that can be
combined with other keys to replace the missing ones. ECM keyboards can be
customised for other languages, too.
ECM keyboards (as shown in figures 16.1 and 16.2) are the most common input
device. Several kinds of ECM keyboards are available, but they resemble each other,
with minor variations. The ECM keyboard in most common use is the QWERTY

doi:10.1088/978-0-7503-1350-6ch16 16-1 ª IOP Publishing Ltd 2016

 Mechatronics

Figure 16.1. Standard MS Windows keyboard layout—Windows Natural Keyboard.

Figure 16.2. Typical laptop keyboard layout.

board. Generally, a standard keyboard has 104 keys. In these ECM keyboards, the
cursor control keys are duplicated to allow easier use of the numeric pad.
A virtual ECM keyboard is actually a key-in device, roughly the size of a fountain
pen, which uses highly advanced laser technology, to project a full-sized ECM
keyboard onto a flat surface. Since their invention, computers have undergone rapid
miniaturisation. Disks and components grew smaller in size, and the only compo-
nent that remained the same for decades was the ECM keyboard. Since the
miniaturisation of a conventional ECM keyboard is very difficult, one opts for a
virtual ECM keyboard. Here, a camera tracks the finger movements of the typist to
obtain the correct keystroke. A virtual ECM keyboard is one that a human user
operates by typing on or within a wireless or optical detectable surface or area rather
than by depressing physical keys.
ECM keyboards have changed very little in layout since their introduction. In
fact, the most common change has simply been the natural evolution of adding more
keys that provide additional functionality. There are various types of ECM
keyboards, such as the 101-key enhanced keyboard, the 104-key Windows keyboard,

16-2
 Mechatronics

the 82-key Apple standard keyboard and the 108-key Apple extended keyboard.
Portable computers, such as laptops, quite often have custom ECM keyboards that
have slightly different key arrangements from a standard ECM keyboard. In
addition, many system manufacturers add specialist buttons to the standard layout.
A standard ECM keyboard consists of the following commutation groups of
electrical valves acting as keys.
• Alphanumerical keys. The alphanumeric or ‘typewriter’ keys include the
number keys and alphabet keys. These keys are arranged in the same style
as in normal typewriters, popularly known as the ‘QWERTY’ layout.
• Functional keys. Arranged in a row on the top of the ECM keyboard, they
help perform specific tasks, such as searching for a file or refreshing a web
page.
• Central keys. Used for controlling the movement of the cursor and screen
display, they include arrow keys and modifier keys, such as ‘ALT’, ‘CTRL’
‘SHIFT’.
• Numerical keypad. Located on the right side of the keyboard, this looks like a
calculator’s keypad.
• Special purpose keys. ‘Delete’ or ‘Del’; ‘End’; Escape’; ‘Home’; ‘Insert’ or
‘Ins’; ‘Page Down’ or ‘PgDn’; ‘Page Up or ‘PgUp’; ‘Pause’; ‘Print Screen’;
‘Spacebar’; and ‘Tab’.

Many changes to the conventional ECM keyboard design are attempts to make
them safer or easier to use. For example, some human users have associated
increased ECM keyboard use with repetitive stress injuries like carpal tunnel
syndrome, although scientific studies have produced conflicting results. They are
conventional ECM keyboards.
• Das keyboard is a completely black keyboard with weighted keys that require
more pressure from a human user’s strongest fingers and less pressure from
the weaker ones.
• The Virtual Laser Keyboard projects a representation of an ECM keyboard
onto a flat surface. When used successfully, a human user’s fingers pass
through the beam of infrared (IR) light above the projected surface and a
sensor interprets this as a keystroke.
• The True-Touch Roll-Up keyboard is flexible and can be rolled up to fit in a
backpack or bag.
• The Ion Illuminated keyboard uses light-emitting diodes or electroluminescent
film to send light through the keys or the spaces between keys.
• The Optimus keyboard has organic light-emitting diodes (OLED) in the keys.
Users can change which letter, command or action each key represents, and
the OLED can change to display the new information.

With the exception of the Virtual Laser Keyboard, which has its own sensing
system, each of these ECM keyboards uses the same technology as conventional
ECM keyboards do to communicate with the computer. We shall look at that
technology next.

16-3
 Mechatronics

16.2 How electrical commutation matrixer keyboards

for computers work
A computer ECM keyboard is a hardware device that functions in accordance with the
instructions made by the human user. It comprises ECMs, electrical valves (switches)
and processors that help in transferring keystroke messages to the computer.
In the contemporary technology-driven world, everyone is aware of computing and
the use of computer ECM keyboards. Everyone knows that the ECM keyboard is an
input device that functions in accordance with the instructions of the human user.
Computer ECM keyboards are used to perform various tasks, such as typing on a
word processor or text editor, accessing menus and playing games. Although pressing
a key corresponds to typing a letter, sign or symbol, it may also represent computer
commands. For example, pressing the ‘F5’ key is a command for refreshing. For
some characters, it is necessary to press and hold more than one key at the same time.
Speaking about the history of computer ECM keyboards, they were modeled in
the 1940s based on the technology of the typewriter keyboard. Similar to a
typewriter keyboard, a computer keyboard consists of a set of keys engraved with
symbols and letters. However, the productivity of working on a computer ECM
keyboard is much greater than that for working a typewriter keyboard.
In general, there are 80–110 keys on a computer keyboard. The keys may vary
depending upon the brand and the type of operating system (OS). Nevertheless, the
shape, size and spacing of the keys are almost the same for all ECM keyboards. In
addition, the layout/arrangement of the keys that represent letters, signs and
symbols is the same, and this is referred to as QWERTY.
The working of a computer ECM keyboard can be compared to that of a
miniature computer. Inside the keyboard there is a metallic plate, a circuit board
(key ECM) and a processor, and these are responsible for transferring information
from the ECM keyboard to the computer.
When one looks at all the extras and options that are available for modern
computer ECM keyboards, it can be hard to believe that their original design came
from mechanical typewriters that did not even use electricity. Now, one can buy
ergonomic ECM keyboards that bear little resemblance to the flat, rectangular
models with ordinary square keys.
Some flashier ECM keyboards light up, roll up or fold up, and others offer
options for programming one’s own commands and shortcuts. But no matter how
many bells and whistles they offer, most ECM keyboards operate using similar
technology. They use electrical valves acting as keys to translate a human user’s
keystrokes into signals that a computer can understand. In this chapter the author
will explore ECM keyboard technology, together with different key layouts, options
and designs.

16.3 Electrical commutation matrixer keyboard fundamentals

An ECM keyboard’s primary function is to act as an input device. Using an ECM
keyboard, a human user can type a document, use keystroke shortcuts, access

16-4
 Mechatronics

menus, play games and perform a variety of other tasks. ECM keyboards can have
different electrical valves acting as keys, depending on the manufacturer, the OS
they are designed for and whether they are attached to a desktop computer or are
part of a laptop.
But, for the most part, these keys, also called ‘keycaps’, are the same size and
shape from keyboard to keyboard. They are also placed at a similar distance from
one another in a similar pattern, no matter what language or alphabet the keys
represent.
Most ECM keyboards have between 80 and 110 keys, including: typing
alphanumerical keys; a numerical keypad; functional keys; and control keys. The
typing alphanumerical keys include the letters of the alphabet, generally laid out in
the same pattern that is used for typewriters. According to legend, this layout,
known as ‘QWERTY’ for its first six letters, helped keep mechanical typewriters’
metal arms from colliding and jamming as users typed. Some people question this
story, but whether it is true or not, the QWERTY pattern had been standard for a
long time when computer keyboards were first developed. ECM keyboards can also
use a variety of other typing alphanumerical key arrangements. The most widely
known is ‘Dvorak’, named after its creator, August Dvorak.
The Dvorak layout places all of the vowels on the left side of the ECM keyboard,
and the most common consonants on the right. The most commonly used letters are
all found along the home row. The home row is the main row where one places one’s
fingers when one begins typing. Users who prefer the Dvorak layout say it increases
their typing speed and reduces fatigue. Other layouts include: ‘ABCDE’, ‘XPeRT’,
‘QWERTZ’ and ‘AZERTY’. Each is named after the first keys in the pattern.
The QWERTZ and AZERTY arrangements are commonly used in Europe. The
numerical keypad is a more recent addition to the computer ECM keyboard. As the
use of computers in business environments increased, so did the necessity for speedy
data entry. Since a large part of the data was made up of numbers, a set of 17 keys,
arranged in the same configuration found on adding machines and calculators, was
added to the ECM keyboard.
In 1986, IBM further extended the fundamental ECM keyboard with the addition
of functional and control keys. Applications and OSs can assign specific commands
to the functional keys. Control keys provide cursor and screen control. Four arrow
keys arranged in an inverted ‘T’ formation between the typing alphanumerical keys
and numerical keypad move the cursor on the screen in small increments.
The Windows ECM keyboard added some extra control keys: two (2) Windows
or Start keys and an Application key. Apple keyboards, on the other hand, have
Command (also known as ‘Apple’) keys.
An ECM keyboard developed for Linux users features Linux-specific hot keys,
including one marked with ‘Tux’ the penguin—the Linux logo/mascot. An ECM
keyboard is a lot like a miniature computer. It has its own processor and circuitry
that carries information to and from that processor. A large part of this circuitry
makes up the key ECM keyboard.
The key ECM keyboard is a grid of circuits underneath the keys. In all ECM
keyboards (except for capacitive ones, which we shall discuss in the next section),

16-5
 Mechatronics

each circuit is broken at a point below each key. When one presses a key, it presses
an electrical valve, completing the circuit and allowing a tiny amount of current to
flow through.
The mechanical action of the electrical valve causes some vibration, called ‘bounce’,
which the processor filters out. If one presses and holds a key, the processor recognises
this as the equivalent of pressing a key repeatedly. When the processor finds a circuit
that is closed, it compares the location of that circuit on the key ECM keyboard to the
character map in its read-only memory (ROM).
A character map is basically a comparison chart or look-up table. It tells the
processor the position of each key in the ECM keyboard and what each keystroke
or combination of keystrokes represents. For example, the character map lets the
processor know that pressing a key by itself corresponds to a small letter ‘a’, but the
‘Shift’ and ‘a’ keys pressed together correspond to a capital ‘A’.
A computer can also use separate character maps, overriding the one found in the
ECM keyboard. This can be useful if a user is typing in a language that uses letters
that do not have English equivalents on an ECM keyboard with English letters.
Users can also set their computers to interpret their keystrokes as though they were
typing on a Dvorak keyboard, even though their actual keys are arranged in a
QWERTY layout. In addition, OSs and applications have ECM keyboard acces-
sibility settings that let human users change their keyboard’s behaviour to adapt to
disabilities.

16.4 Electrical commutation matrixer keyboard’s electrical valves

ECM keyboards use a variety of electrical valve (switch) technologies. Capacitance
electrical valves acting as keys are considered to be non-mechanical because they do
not physically complete a circuit like most other ECM keyboard technologies.
Instead, current constantly flows through all the parts of the key ECM keyboard.
Each key is spring-loaded and has a tiny plate attached to the bottom of it. When
one presses a key, it moves this plate closer to the plate below it. As the two plates
move closer together, the amount of current flowing through the ECM keyboard
changes. The processor detects the change and interprets it as a key press for that
location.
Capacitive electrical-valve ECM keyboards are expensive, but they have a longer
life than any other keyboard. In addition, they do not have problems with bounce,
since the two surfaces never come into actual contact. All of the other types of
electrical valves used in ECM keyboards are mechanical in nature. Each provides a
different level of audible and tactile response—the sounds and sensations that typing
creates.
Depending upon the working principle, there are twelve main types of keys,
namely, metal contact, capacitance, metal dome, rubber dome, membrane, buckling
spring, roll-up, Hall effect, reed, foam component, laser and photo-optics. Let us
discuss the functioning of these key electrical and optical valves (switches) in brief.
• Metal contact. A metal (hard) contact key is attached to a metallic plate that
helps in connecting the circuit board. When the metal key is pressed, it pushes

16-6
 Mechatronics

a metallic plate, which in turn touches the metallic portion of the circuit plate.
This overall process of completing a circuit results in a circuit flow, allowing
the transfer of the message to the central processing unit (CPU), which is
further transmitted to the software. Metal ECM keyboards are growing less
common. Metal electrical valves (switches) simply have a spring-loaded key
with a strip of metal on the bottom of the plunger. When the key is pressed,
the metal strip connects the two parts of the circuit. This ECM keyboard
technology has good tactile response, making satisfyingly audible ‘clicks’, and
is inexpensive to manufacture. The simplest switch mechanism employs
mechanical contacts that are brought together as the key plunger is depressed.
An example of a mechanical-contact key electrical valve (switch) is shown in
figure 16.3.
Mechanical-contact ECM keyboards use separate complete electrical
valves (switches) underneath every key. Each electrical valve (switch) is
composed of a base, a spring and a stem. Depending on the shape of the
stem, such ECM keyboards have varying actuation and travel distances.
Depending on the resistance of the spring, the key requires different amounts
of pressure to actuate. Mechanical-contact ECM keyboards allow for the
removal and replacement of keycaps. Mechanical-contact ECM keyboards
have proven to be popular amongst the PC gaming community and are
gaining in popularity among typists. Although low in cost, one problem with
mechanical contacts is that they suffer from contact bounce, which can
generate false signals. This problem usually necessitates the use of appro-
priate denounce circuitry. Another disadvantage of mechanical-contact

Keycap

Return spring

Mechanical
contacts

Plunger
Printed circuit
board

Figure 16.3. An example of a mechanical-contact key electrical valve (switch) (adapted from McLoughlin 2011).

16-7
 Mechatronics

electrical valves (switches) is that contacts can be contaminated by particles

or oxidation. Such contamination results in increased contact resistance,
causing unreliable operation or even total switch failure. The other problem is
that the contacts tend to wear out or corrode faster than on ECM keyboards
that use other technologies. There is no barrier that prevents dust or liquids
from coming into direct contact with the circuitry of the key ECM. Key life is
in the range of 5–10 million actuations, but the survival rate often drops
rapidly above 5 million operations (McLoughlin 2011).
• Capacitance electrical valves (switches). One of the most popular ECM
keyboard technologies uses capacitance changes to signal key activation. In
this type of keyboard, pressing the key changes the capacitance of a pattern of
capacitor pads. Unlike in ‘dome switch’ ECM keyboards, the pattern consists
of two D-shaped capacitor pads for each electrical valve (switch), printed on a
printed circuit board (PCB) and covered by a thin, insulating film of solder
mask, which plays the role of a dielectric. The mechanism of the capacitance
electrical valves (switches) is very simple, compared to that of mechanical
ones. Its movable part ends with a flat foam element (with dimensions near to
those of an aspirin tablet) finished with aluminum foil below. The opposite
side of the electrical valve (switch) is a PCB with capacitor pads. When a key
is pressed, the foil clings tightly to the surface of the PCB, forming a daisy
chain of two capacitors between contact pads and itself separated with thin
solder mask, thus ‘shorting’ the contact pads with an easily detectable drop of
capacitive reactance between them. Usually this permits a pulse or pulse train
to be sensed. The keys do not need to be pressed fully to be fired on, which
enables some typists to work faster. In its most common form, the capaci-
tance commutation (switching) mechanism employs a foam pad that is
bonded to the bottom of the key plunger. Figure 16.4 illustrates an example
of a capacitance key electrical valve (switch). A metal plunger is fixed on the
underside of a capacitive key, which helps in activating the circuit flow. When
a capacitive key is pressed, the metal plunger applies gentle pressure to the
circuit board.
The pressure is identified by the computer and the circuit flow is initiated,
resulting in the transfer of information from the circuit to the currently
installed software. A sheet of conductive foil is attached to the bottom of the
foam pad. As a key is pressed, the foil, which acts as a capacitive plate, is
brought into contact with a printed circuit board mounted under the array of
key plungers. The increase in capacitance as the foil contacts the circuit board
passes a low-level signal onto detector circuits, and this signals key activation.
Because capacitance switching does not depend on the closing of a circuit
through low resistance contacts, capacitance keyboards exhibit high reliabil-
ity. Additional circuits and AC signal are needed to sense electrical valve
(switch) closure. Typical key life characteristics range from 10–50 million
keystrokes (McLoughlin 2011).
• Metal dome electrical valve (switch). One popular electrical valve (switch)
technology is the dome electrical valve (switch). Dome electrical valve

16-8
 Mechatronics

Keycap

Plunger Return spring

Foil Foam pad

Printed circuit Etched capacitive

board plates

Figure 16.4. An example of a capacitance key electrical valve (switch) (adapted from McLoughlin 2011).

(switch) ECM keyboards are a hybrid of flat-panel membrane and mechan-

ical ECM keyboards. They bring two circuit board traces together under a
rubber or silicone keypad using either metal ‘dome’ electrical valves
(switches) or polyurethane formed domes. The metal-dome electrical valves
(switches) are formed pieces of stainless steel that, when compressed, give the
human user a crisp, positive tactile feedback. These metal types of dome
electrical valves (switches) are very common, are usually reliable for over
5 million cycles, and can be plated in either nickel, silver or gold. Dome
electrical valve (switch) ECM keyboards are also called direct electrical valve
(switch) ECM keyboards. As shown in the cross-section of figure 16.5, in a
metal dome electrical valve (switch), the metal dome provides a spring action,
and when pressed, it snaps to the closed position (McLoughlin 2011).
This provides some travel and good tactile response. Metal dome electrical
valves (switches) are available as a single push button and in an ECM
keyboard. A combination of membrane and metal dome technologies is
widely used. This combines the low cost and environmental immunity of the
membrane technology with the tactile response of the metal dome electrical
valves (switches)
• Rubber dome electrical valves (switches). These are very common. They use
small, flexible rubber domes, each with a hard carbon center. When the user
presses a key, a plunger on the bottom of the key pushes down against the
dome, and the carbon center presses against a hard, flat surface beneath the
key ECM. As long as the key is held, the carbon center completes the circuit.
When the key is released, the rubber dome springs back to its original shape,
forcing the key back up to its at-rest position. The rubber dome electrical

16-9
 Mechatronics

Keycap
Front panel

Metal
Contact dome

Figure 16.5. An example of a metal dome key electrical valve (switch) (adapted from McLoughlin 2011).

valves (switches), most commonly referred to as polydomes, are formed

polyurethane domes where the inside bubble is coated in graphite. While
polydomes are typically cheaper than metal domes, they lack the crisp snap of
the metal domes and usually have a lower life specification. Polydomes are
considered very quiet, but purists tend to find them ‘mushy’ because the
collapsing dome does not provide as much positive response as metal domes.
For either metal or polydomes, when a key is pressed, it collapses the dome,
which connects the two circuit traces and completes the connection to enter
the character. The pattern on the PC board is often gold-plated. Both are
common switch technologies and are used in mass-market ECM keyboards
today. This type of electrical valve (switch) technology is most commonly
used in handheld controllers, mobile phones, automotive vehicles, consumer
electronics and medical devices. Rubber dome electrical-valve ECM
keyboards are inexpensive, have pretty good tactile response and are fairly
resistant to spills and corrosion because of the rubber layer covering the key
ECM.
• Membrane. Rather than having an electrical valve (switch) for each key,
membrane ECM keyboards use a continuous membrane that stretches from
one end to another. A pattern printed in the membrane completes the circuit
when one presses a key. Some membrane ECM keyboards use a flat surface
printed with representations of each key rather than keycaps. There are two
types of membrane-based ECM keyboards, flat-panel membrane ECM
keyboards and full-travel membrane ECM keyboards. Flat-panel membrane
ECM keyboards are most often found on appliances like microwave ovens or
photocopiers. A common design consists of three layers. The top layer (and

16-10
 Mechatronics

the one the user touches) has the labels printed on its front and conductive
stripes printed on the back. Under this it has a spacer layer, which holds the
front and back layer apart so that they do not normally make electrical
contact. The back layer has conductive stripes printed perpendicularly to
those of the front layer. When placed together, the stripes form a grid. When
a user pushes down at a particular position, his/her finger pushes the front
layer down through the spacer layer to close a circuit at one of the
intersections of the grid. This indicates to the computer or ECM keyboard
control processor that a particular button has been pressed. Generally, flat-
panel membrane ECM keyboards do not have much of a ‘feel’, so many
machines that use them issue a beep or flash a light when the key is pressed.
They are often used in harsh environments where water- or leak-proofing is
desirable. Although they were used in the early days of the personal computer
(on the Sinclair ZX80 and ZX81 and the Atari 400), they have been
supplanted by the more tactile dome and mechanical switch keyboards.
Full-travel membrane-based ECM keyboards are the most common com-
puter ECM keyboards today. They have one-piece plastic keytop/switch
plungers which press down on a membrane to actuate a contact in an ECM.
Although essentially a metal electrical valve (switch) technology, full travel
membrane ECM keyboards present a low-cost and reliable keyboard alter-
native. An ECM of full travel (approximately 4 mm) key button assembly is
mounted above a three-layer sandwich of circuit materials that comprise the
commutation (switching) mechanisms for all the keys on the ECM keyboard
(see figure 16.6).
The top and bottom layers of the sandwich are thin sheets of flexible film
(typically, clear polymer) with a conductive silver composition deposited onto
facing sides of the flexible film. The silver composition is ‘printed’ in a pattern
that forms the electrical contacts for each key station and all the electrical
interconnections. Between the conductive layers is a third layer that provides
a gap between the contacts through holes in the layer. The entire three-layer
switch assembly is bonded to a sheet metal base plate. When a key is pressed,
a switch closure is accomplished by deflecting the top film layer through the
hole in the separation layer so that the conductive material on the top film
layer comes in contact with the conductive material on the bottom layer.
Typical key life now exceeds 10 million keystrokes (McLoughlin 2011).
Unfortunately, this type of ECM keyboard also suffers from switch bounce.
Membrane ECM keyboards do not have good tactile response, and without
additional mechanical components they do not make the clicking sound that
some human users like to hear when they are typing. However, they are
generally inexpensive to make.
• Buckling spring. Many typists prefer buckling spring ECM keyboards. The
buckling spring mechanism1 atop the electrical valve (switch) is responsible

1
Expired US patent 4,118,611.

16-11
 Mechatronics

Membrane

Spacer

Substrate

Figure 16.6. Exploded view of a full travel membrane ECM keyboard showing the unitised frame, space layer,
adhesive layers and base plate (adapted from McLoughlin 2011).

for the tactile and aural response of the keyboard. This mechanism controls a
small hammer that strikes a capacitive or membrane switch.
• Roll-up. Some ECM keyboards are designed using flexible materials that can
roll up in a moderately tight bundle. Normally, the external materials are
either silicone or polyurethane. Although manufacturers may claim that roll-
up ECM keyboards are foldable, they cannot be folded without damaging the
membrane that holds the circuitry. Typically, they are completely sealed in
rubber, making them watertight like membrane ECM keyboards. Like
membrane ECM keyboards, they are reported to be very hard to get used
to, as there is little tactile feedback, and silicone will tend to attract dirt, dust
and hair.
• Hall effect. Hall-effect ECM keyboards use interior permanent magnets (IPM)
and Hall-effect sensors instead of an actual electrical valve (switch). When a
key is depressed, it moves an IPM, which is detected by the solid-state sensor.
In Hall-effect ECM keyboards, a magnetic transducer (typically an IC) is
mounted near an IPM, which is affixed to the key plunger (see figure 16.7).
Depression of a key moves the IPM closer to the magnetic transducer,
causing the transducer to output a voltage change, signaling key activation.
Because the commutation (switching) mechanism employs solid-state elec-
tronics, Hall-effect ECM keyboards exhibit extremely high reliability. Key
life of more than 100 million actuations per key is typical (McLoughlin 2011).
These ECM keyboards are extremely reliable, and are able to accept millions
of keystrokes before failing. They are used for ultra-high reliability

16-12
 Mechatronics

Magnetic
field

Extrinsic I
semiconductor A

Hall voltage
(positive for n-type material
negative for p-type material)

Figure 16.7. Hall effect (adapted from McLoughlin 2011).

applications, in locations like nuclear power plants or aircraft cockpits. They

are also sometimes used in industrial environments. These ECM keyboards
can easily be made totally waterproof. They also resist large amounts of dust
and contaminants. Because an IPM and sensor are required for each key, as
well as custom control electronics, they are very expensive. In addition,
because individual key electrical valves (switches) can be sealed, Hall effect
ECM keyboards are often used in military, marine, industrial, and aerospace
and automotive applications, where reliability is a critical factor. The
disadvantages of Hall-effect electrical valves (switches) are their relatively
high cost and the large size of the individual key electrical valve (switch)
modules.
• Reed electrical valves (switches). These are essentially mechanical contacts;
the problems caused by contamination are eliminated by enclosing the
electrical valve (switch) contacts in a sealed capsule, which is usually made
of glass. Depression of a key moves an IPM close to the encapsulated reed
electrical valves (switches). The attraction of the IPM on the reed electrical
valves (switches) causes them to be brought together until the contacts touch.
Because the sealed capsule prevents contamination of the electrical valve
(switch) contacts, the long-term reliability is generally good. Key life of 50
million operations is possible (McLoughlin 2011). The biggest disadvantage
of reed electrical valves (switches) is their high cost relative to other
technologies.
• Foam element. These ECM keyboards are growing less common. The foam
element electrical valve (switch) is basically the same design as that used for
metal ECM keyboards, but with a small piece of spongy foam between the
bottom of the plunger and the metal strip, providing a better tactile response.
This ECM keyboard technology has good tactile response, make satisfyingly
audible ‘clicks’, and is inexpensive to manufacture. The problem is that the
contacts tend to wear out or corrode faster than on ECM keyboards that use

16-13
 Mechatronics

other technologies. In addition, there is no barrier that prevents dust or

liquids from coming into direct contact with the circuitry of the key ECM.
• Laser. A laser projection device approximately the size of a computer mouse
projects the outline of the ECM keyboard keys onto a flat surface, such as a
table or desk. This type of laser ECM keyboard is portable enough to be used
easily with PDAs and cellphones, and many models have retractable cords
and wireless capability. However, sudden or accidental disruption of the laser
will register unwanted keystrokes. Also, if the laser malfunctions, the whole
unit becomes useless, unlike conventional ECM keyboards, which can be
used even if a variety of parts (such as the keycaps) are removed. This type of
ECM keyboard can be frustrating to use, since it is susceptible to errors, even
in the course of normal typing, and its complete lack of tactile feedback
makes it even less user-friendly than the lowest quality membrane ECM
keyboards.
• Photo-optics. This is for photo-optic ECM keyboards, light-responsive ECM
keyboards, photo-electric ECM keyboards and optical key actuation detec-
tion technology. Optical ECM keyboard technology was introduced in 1962
by Harley E Kelchner for use in a typewriter machine with the purpose of
reducing the noise generated by actuating the typewriter keys. Optical ECM
keyboard technologies utilise light-emitting devices and photo sensors to
optically detect actuated keys. Most commonly, the emitters and sensors are
located at the perimeter, mounted on a small PCB. The light is directed from
one side of the keyboard interior to the other, and it can only be blocked by
the actuated keys. Most optical ECM keyboards require at least two beams
(most commonly a vertical beam and a horizontal beam) to determine the
actuated key. Some optical keyboards use a special key structure that blocks
the light in a certain pattern, allowing only one beam per row of keys (most
commonly a horizontal beam). The mechanism of the optical ECM keyboard
is very simple—a light beam is sent from the emitter to the receiving sensor,
and the actuated key blocks, reflects, refracts or otherwise interacts with the
beam, resulting in an identified key. Some earlier optical ECM keyboards
were limited in their structure and required special casing to block external
light, while no multi-key functionality was supported and the design was
limited to a thick rectangular case. The advantages of optical keyboard
technology are that it offers a truly waterproof keyboard, is resilient to dust
and liquids, and uses about 20% PCB volume, compared with membrane or
dome electrical valve (switch) ECM keyboards, significantly reducing elec-
tronic waste.

In all the key types, the circuit signals the processor to read and/or identify the
character that has been pressed. For example, in a metal (hard) contact key, the
processor reads that pressing the ‘Shift’ and ‘a’ keys at the same time corresponds to
‘A’. Hence, the letter, sign or symbol is displayed on the screen accordingly.
Releasing the pressed key breaks the circuit flow, after which the key retains its
original position. The communication between the computer ECM keyboard and

16-14
 Mechatronics

the main computer is bilateral, meaning that messages or information can be sent to
each other.
Different manufacturers have used these standard ECM keyboard technologies,
and a few others, to create a wide range of unconventional ECM keyboards. We
shall look at some of these unconventional ECM keyboards in the next section.

16.5 Unconventional electrical commutation matrixer keyboards

Many of the modifications to conventional ECM keyboard designs are an attempt to
make them safer or easier to use. For example, some human users have associated
increased keyboard use with repetitive stress injuries (RSI) like carpal tunnel
syndrome, although scientific studies have produced conflicting results.
Ergonomic ECM keyboard designs are intended to keep a human user’s hands in
a more natural position while they are typing in an attempt to prevent injuries. While
these ECM keyboards can certainly keep human users from holding their hands in a
‘praying mantis’ position, studies disagree on whether they actually prevent injury.
The simplest ergonomic ECM keyboards look like conventional ones that have
been divided down the middle, keeping a human user’s hands further apart and
aligning the wrists with the forearms. More complex designs place the two halves of
the keyboard at varying angles to one another and to the surface on which the ECM
keyboard rests. Some go even further, placing the two halves of the ECM keyboard
on the armrests of chairs, or making them completely perpendicular to the desk
surface. Others, like the ‘Datahand’, do not look much like ECM keyboards at all.
Some modifications, while not necessarily ergonomic, are designed to make ECM
keyboards more portable, more versatile, or just ‘cooler’. Some unconventional
ECM keyboards are designed to reduce fatigue for RSI, or for one-handed use. e.g.
the Maltron left-handed keyboard shown in figure 16.8.

Figure 16.8. The Maltron left-handed keyboard.

16-15
 Mechatronics

16.6 Virtual electrical commutation matrixer keyboards

Since their invention, computers have undergone rapid miniaturisation, from being
a ‘space saver’ to ‘as tiny as your palm’. Disks and components grew smaller in size,
but one component remained the same for decades—the ECM keyboard. The
miniaturisation of ECM keyboards proved to be a nightmare for users.
Human users of PDAs and smartphones are annoyed by the tiny size of the keys.
The innovative new virtual ECM keyboard uses advanced technologies to project a
full-sized computing ECM keyboard onto any surface. This device has become the
solution for mobile computer users who prefer to touch-type rather than cramping
over tiny keys. Typing information into mobile devices usually feels about as natural
as a linebacker riding a Big Wheel. The virtual ECM keyboard is a way to eliminate
finger cramping. All that is needed to use the ECM keyboard is a flat surface. Using
laser technology, a bright red image of a keyboard is projected from a device such as
a handheld. Detection technology based on optical recognition allows users to tap
the images of the keys, so the virtual ECM keyboard behaves like a real one. It is
designed to support any typing speed.
Theoretically, with either approach, the ECM keyboard can be in space and the
human user can type by moving fingers through the air. The regular QWERTY
keyboard layout is provided. Several products have been developed that by ‘virtual
ECM keyboard’ mean one that has been put on a display screen as an image map. In
some cases, the ECM keyboard can be customised. Depending on the product, the
human user (who may be someone who is unable to use a regular ECM keyboard)
can use a touch screen or a mouse to select the keys. A virtual ECM keyboard is one
that a user operates by typing (moving fingers) on or within a wireless or optical-
detectable surface or area rather than by depressing physical keys. In one
technology, the ECM keyboard is projected optically onto a flat surface and, as
the human user touches the image of a key, the optical device detects the stroke and
sends it to the computer. In another technology, the ECM keyboard is projected
onto an area and selected keys are transmitted as wireless signals using short-range
Bluetooth technology. With either approach, a virtual keyboard makes it possible
for the human user of a very small smartphone or a wearable computer to have full
ECM keyboard capability. The advantages of the virtual ECM keyboard are as
follows:
• keyboard layouts can be changed by software, allowing for ‘foreign’ or
alternative keyboard layouts;
• ability to minimise the risk of repetitive strain injuries;
• flat or large typing surfaces are unnecessary;
• speed of text entry;
• accuracy;
• portability.

16.7 Canesta keyboard

The Canesta Keyboard IR Light Source (IR-CK100) plays a critical role in any
Canesta Keyboard Perception Chipset implementation by enabling Canesta’s

16-16
 Mechatronics

patented electronic perception technology. The module collects the IR laser diode
and all the associated optics into a single concise package, making it easy to
integrate into a final product.
The IR Light Source operates by emitting a beam of infrared light. This light
beam is designed to overlap the area on which the Canesta Keyboard Pattern
Projector (PP-CK100) displays the ECM keyboard layout, so the human user’s
fingers are illuminated by the infrared light beam. The Canesta Keyboard Sensor
Module (SM-CK100) detects the finger movement and the typing activity is resolved
into the appropriate keystrokes or mouse actions.

16.8 Samsung’s Scurry keyboard

Samsung’s Scurry is also a wearable ECM keyboard. It works by attaching motion
sensors to each finger. It does not detect muscle movement, but rather uses
gyroscopic technology to detect angular movements of fingers through space. This
approach works better. Demonstrations on the show floor were far more impressive
than its competitor’s performance in terms of accuracy. However, the device is too
bulky. Nonfunctional prototypes of the final product are much smaller. The ECM
keyboard demonstrated is wired. The manufacturers say that by the time their
products become commercially available they will support the wireless Bluetooth
protocol. In addition, the product requires specialised software for handheld devices.

16.9 Exemplary electrical commutation matrixer keyboard

Computer ECM keyboards include control circuitry to convert key presses into key
codes (usually scancodes) that the computer’s electronics can understand. The key
electrical valves (switches) are connected via the printed circuit board in an X × Y
ECM, where a voltage is provided sequentially to the Y lines and, when a key is
depressed, detected sequentially by scanning the X lines. The first computer ECM
keyboards were for mainframe computer data terminals and used discrete electronic
parts. The first ECM keyboard microprocessor was introduced in 1972 by General
Instruments, but keyboards have been using the single-chip 8048 microcontroller
variant since it became available in 1978.
The ECM keyboard commutation (switching) matrixer is wired to its inputs, it
converts the keystrokes to key codes, and, for a detached keyboard, sends the codes
down a serial cable (the keyboard cord) to the main processor on the computer
motherboard. This serial ECM keyboard cable communication is only bilateral to
the extent that the computer’s electronics controls the illumination of the caps lock,
number lock and scroll lock lights.
When pressing an ECM keyboard key, the key contacts may ‘bounce’ against
each other for several milliseconds before they settle into metal (hard) contact. When
released, they bounce some more until they revert to the uncontacted state. If the
computer were watching for each pulse, it would see many keystrokes for what the
human user thought was just one. To resolve this problem, the processor in an ECM
keyboard (or computer) ‘debounces’ the keystrokes by aggregating them across time
to produce one ‘confirmed’ keystroke.

16-17
 Mechatronics

The keyboard one is interfacing with here is an ECM keyboard. This ECM
keyboard is designed with a particular rows and columns. These rows and columns
are connected to the microcontroller through its ports of the microcontroller. One
normally uses and X × Y ECM keyboard. So only two ports of the microcontroller
can be easily connected to the rows and columns of the ECM keyboard. Whenever a
key is pressed, a row and a column gets shorted through that pressed key and all the
other keys are left open. When a key is pressed only a bit in the port goes high. What
indicates to the microcontroller that the key has been pressed? It is identified through
this high on the bit key in the corresponding column. Once one is sure that one of the
keys in the keyboard has been pressed, one’s next aim is to identify that key. To do
this, one first checks for a particular row and then checks the corresponding column
in the ECM keyboard.
To check the row of the pressed key in the ECM keyboard, one of the rows is
made high by making one bit in the output port of the microcontroller high. This is
done until the row is revealed. Once one obtains the row, the next task is to ascertain
the column of the pressed key. The column is detected through the contents in the
input ports with the help of a counter. The content of the input port is rotated with
carry until the carry bit is set.
The scan ECM lies beneath the keys, as shown in figure 16.9. Whenever a key is
depressed and released, the control code at the time of the event is sensed, and this is
then translated by the microcontroller into K-scan code. When a key is depressed, a
make code is produced, and when a key is released, a break code is produced.
Figure 16.10 shows the key ECM for the 107 standard keyboard with power
management keys (power, wake and sleep).
Figure 16.11 shows the key ECM for a multimedia keyboard with function keys.
Figure 16.12 shows a block diagram for an exemplary ECM keyboard. The
solution only includes the key button inputs and LED indicator outputs.
Figure 16.13 shows the schematic of the exemplary ECM keyboard, where:
• J1 is used for USB connection;
• J2 is used for PS/2 mouse connection (for future development);
• J3 and J7 are used for in-circuit programming;
• an eight row × 18 column key ECM is implemented;
• eight rows are implemented in PTA[7:0];

Figure 16.9. Scan electrical commutation matrixer.

16-18
 Mechatronics

Figure 16.10. 107-key ECM (Lau 2006).

Figure 16.11. Multimedia key ECM (Lau 2006).

• 18 columns are implemented in PTB[7:0], PTC[7:0], PTE0 and PTE2;

• keyboard LEDs:
○ scroll lock at PTD2;
○ caps lock at PTD3;
○ number lock at PTD4.

The ‘intellect’ of the exemplary ECM keyboard is the commutation matrixery

within the unit that handles the processing of keystrokes and exchanges information

16-19
 Mechatronics

Figure 16.12. Block diagram of the exemplary ECM keyboard (Lau 2006).

Figure 16.13. Exemplary ECM keyboard schematics (Lau 2006).

with the host system. While ECM keyboards are fairly simple devices, one might be
surprised at just how complex the internal commutation matrixery in an ECM
keyboard is (figure 16.14).
In fact, the ECM keyboard has within it small versions of several components
that one finds within the computer as a whole. It has its own microprocessor, which
‘runs the show’, so to speak. This is, of course, a tiny central processor unit (CPU),
not a fully fledged CPU, like the main computer hardware uses. There is also some
read-only memory (ROM) that runs this small processor, similar to the system BIOS
code on the motherboard.

16-20
 Mechatronics

Figure 16.14. Internal commutation matrixer board from a typical ECM keyboard PCB. The chip is the ECM
keyboard controller. Note the LEDs sticking up from the circuit board in this design; they ‘poke through’ holes
in the keyboard case (Lau 2006).

Programmable ECM keyboards also contain some EEPROM memory to hold

programming information; this functions as if it were a tiny hard disk, in a way. This
internal commutation matrixery has two fundamental tasks. First, it is responsible
for sensing the raw signals created by the key electrical valves (switches) as they
make electrical contact, and then translating them into signals that can be sent to the
computer. Due to the electrical characteristics of keystrokes, this is not as simple as
it sounds. Second, it handles the actual interfacing between the ECM keyboard and
the rest of the host system. The interface between the computer and the ECM
keyboard is a fully bilateral serial communications line, with its own special
protocols and commands.

16.10 Conclusions
As one types, the processor in the ECM keyboard analyses the key ECM and
determines what characters to send to the computer. It maintains these characters in
its memory buffer and then sends the data. Many ECM keyboards connect to the
computer through a cable with a PS/2 or universal serial bus (USB) connector.
Laptops use internal connectors. Regardless of which type of connector is used, the
cable must carry power to the ECM keyboard, and it must carry signals from the
ECM keyboard back to the computer. Wireless ECM keyboards, on the other hand,
connect to the computer through infrared (IR), radio frequency (RF) or Bluetooth
connections. IR and RF connections are similar to what one finds in a remote
control. Regardless of which sort of signal they use, wireless ECM keyboards require
a receiver, either built-in or plugged into the USB port, to communicate with the

16-21
 Mechatronics

computer. Since they do not have a physical connection to the computer, wireless
ECM keyboards have an AC power connection or use, e.g. graphene Li-ion or
Mg-ion storage batteries for power. In the field of storage batteries, conventional
storage-battery electrode materials (and prospective ones) are significantly improved
when they are enhanced with graphene. Graphene can make storage batteries that
are light, durable and suitable for high-capacity energy storage, while it also shortens
charging times. Whether it is through a cable or wireless, the signal from the ECM
keyboard is monitored by the computer’s ECM keyboard controller. This is an
integrated circuit (IC) that processes all of the data that come from the ECM
keyboard and forwards them to the OS. When the OS is notified that there are data
from the ECM keyboard, it checks to see if the ECM keyboard data are a system
level command. A good example of this is ‘Ctrl-Alt-Del’ on a Windows computer,
which reboots the system. Then, the OS passes the ECM keyboard data onto the
current application.
The application determines whether the ECM keyboard data are a command, like
‘Alt-f’, which opens the File menu in a Windows application. If the data are not a
command, the application accepts them as content, which can be anything from
typing a document, to entering a URL, to performing a calculation. If the current
application does not accept the ECM keyboard data, it simply ignores the
information. This whole process, from pressing the key to entering content into
an application, happens almost instantaneously. Virtual ECM keyboards are
projected images of the real thing that let typists compose their sentences on any
flat surface. They are inching closer to store shelves. The ECM keyboard will be
convenient for users who travel so they can easily access information through web
pages. This will be efficient for individuals who have physical impairments. This
technology will provide a keyboard that does not require force to activate the keys.
It will serve people with physical impairments better.
Of the various devices available for transmitting information from humans to
computers, the electrical valve (switch) is the simplest form of input device.
However, the standard ECM keyboard is certainly the most generally applied. It
is clearly the device of choice for applications involving significant amounts of
textual input. It is also the computer input technology most affected by historical
precedent. The data entry ECM keyboard achieved some degree of design stand-
ardisation long before its linkage to the computer. While this linkage has had a
substantial effect on the design of current ECM keyboards, the functionality, layout,
mechanics and dimensions of these devices are still in many ways a reflection of
19th-century technology.

References
Lau D 2006 USB and PS/2 Multimedia Keyboard Interface. Designer Reference Manual.
M68HC08 Microcontrollers, DRM014, Rev. 1.0, 4/2006, Freescale Semiconductor Inc.,
Hong Kong 34 p
McLoughlin I 2011 Computer Architecture: an embedded approach (New York: McGraw-Hill)

16-22
 IOP Publishing

Mechatronics
Dynamical systems approach and theory of holors
B T Fijalkowski

Chapter 17
Programmable-logic and/or generic-logic
electrical commutation matrixers for digital
devices

You have to learn the rules of the game. And then you have to play better than
anyone else.
Albert Einstein

17.1 Introduction
In this chapter, programmable-logic electrical commutation matrixer (PECM)
and/or generic-logic electrical commutation matrixers (GECM) are introduced.
The emphasis is on using PECMs to implement combinational logic. The sequential
logic applications of PECMs are discussed. PECMs and GECMs can be used
to replace SSI and MSI digital devices in most types of applications, resulting
in fewer parts and lower costs (Broesch 1991; Floyd 1994; Carter 1997; Sharma
1998).
Programming software, such as the advanced Boolean expression language
(ABEL) and the compiler for universal programmable logic (CUPL), are used to
implement logic design in PECMs.
Once one is familiar with one PECM programming language, the others are easy
to learn because of their similarities. For instance, ABEL can be used to progress
any type of PECM and is therefore a matrix-independent language. ABEL is run on
a computer connected to a language-independent ECM programmer into which the
PECM is inserted (Pellerin and Holley 1994).

doi:10.1088/978-0-7503-1350-6ch17 17-1 ª IOP Publishing Ltd 2016

 Mechatronics

17.2 Programmable-logic electrical-commutation-matrixers

and classifications
All PECM consists of programmable matrixers, which are essentially grids of
conductors that form rows and columns with a fusible link at each cross point.
PECMs can be either fixed or programmable.
The earliest type of programmable logic matrix (PLM), which dates back to the
1960s, has an electrical commutation matrixer (ECM) with an electrical valve as a
diode at each cross point of the ECM.

17.2.1 The OR electrical commutation matrixer

The original diode ECM evolved into the integrated OR ECM, which consists of an
ECM of OR gates (that produce a HIGH output when one or more inputs are
HIGH) connected to a PECM with fusible links at each cross point of a row and
column, as shown in figure 17.1(a).
The OR ECM is programmed by means of electrical valves, i.e. blowing fuses to
eliminate selected variables from the output functions, as illustrated in figure 17.1(b)
for a specific case. For each input to an OR gate, only one fuse is left intact in order
to connect the desired variable to the gate input. Once a fuse is blown, it cannot be
reconnected.

17.2.2 The AND electrical commutation matrixer

This type of ECM consists of AND gates (that produce HIGH output only when all
of the inputs are HIGH) connected to a PECM with fusible links at each cross point,
as shown in figure 17.2(a). Like the OR ECM, the AND ECM is programmed by
means of electrical valves, i.e. blowing fuses to eliminate variables from the output
function, as illustrated in figure 17.2(b).
For each input to an AND gate, only one fuse is left intact in order to connect the
desired variable to the gate input. Also like the OR ECM, the AND ECM with
fusible links is one-time programmable.
Fusible link
A A B B A A B B

X1 X1 = A • B

X2 X2 = A • B

X3 X3 = A • B

Unprogrammed Programmed

Figure 17.1. An example of a basic OR electrical commutation matrixer (adapted from Floyd 1994).

17-2
 Mechatronics

A A B B A A B B

X1 X1 = AB

X2 X2 = AB

X3 X3 = AB

Unprogrammed Programmed

Figure 17.2. An example of a basic AND electrical commutation matrixer (adapted from Floyd 1994).

17.2.3 Classifications of programmable-logic electrical commutation matrixers

PECMs are classified according to their architecture, which is basically the internal
functional arrangement of the elements that give a PECM its particular operating
characteristics.
• Programmable Read-Only Memory (PROM). The PROM consists of a set of
fixed (non-programmable) AND gates connected as a decoder and a
programmable OR ECM. It is used primarily as an addressable memory
and not as a logic matrix because of the limitations imposed by the fixed
AND gates.
• Programmable Logic Matrix (PLM). The PLM is a PECM that consists of a
programmable AND ECM and a programmable OR ECM. The PLM was
developed to overcome some of the limitations of PROM. The PLM is also
called field-programmable logic array (FPLA) because the user in the field, not
the manufacturer, programs it.
• Programmable Matrix Logic (PML). The PML is an ECM that was
developed to overcome certain disadvantages of the PLM, such as longer
delays due to the additional fusible links that result from using two program-
mable matrixers and more circuit complexity. The basic PML consists of a
programmable AND ECM and a fixed OR ECM with output logic. The
PML is the most common on-time programmable logic matrix and is
implemented with bipolar transistor-transistor logic (TTL) or emitter-coupled
logic (ECL) technology.
• Generic Matrix Logic (GML). A more recent development in GE CMs is the
GML. The GML, like the PML, has a programmable AND ECM and a fixed
OR ECM with programmable output logic. The two main differences
between GML and PML matrixers are (a) the GML is reprogrammable
and (b) the GML has programmable output configurations.

17.3 Programmable ROMs (PROMs and EPROMs)

A ROM contains permanently or semi-permanently stored data, which can be read
from the memory but either cannot be changed at all or cannot be changed without

17-3
 Mechatronics

special equipment. A ROM stores data that are used repeatedly in system
applications, such as tables, conversions, or programmed instructions for system
initialisation and operations. ROMs retain stored data when the power is off and are
therefore non-volatile memory.
PROM, or programmable ROM, is the type in which the user with the aid of
specialised instruments electronically stores the data. PROMs are basically the same
as mask ROMs, once they have been programmed. The difference is that PROMs
come from the manufacturer unprogrammed and are custom programmed in the
field to meet the user’s needs.
PROMs use some type of fusing process to store bits, in which a memory link is
burned open or left intact to represent a binary zero (0) or a binary one (1). The
fusing process is irreversible; once a PROM has been programmed, it cannot be
changed. Figure 17.3 illustrates a metal-oxide semiconductor (MOS) PROM matrix
with fusible links.
The fusible links are manufactured into the PROM between the source of each
cell’s transistor and its column line.
In the programming process, a sufficient current is injected through the fusible
link to burn it open to create a stored binary zero (0). The link is left intact for a
stored binary one (1).

+VDD

Rows

Columns

Figure 17.3. MOS PROM matrixer with fusible links (adapted from Floyd 1994). (All drains are commonly
connected to VDD.)

17-4
 Mechatronics

A PROM is normally programmed by inserting it into a special instrument called

a ‘PROM programmer’. Basically, the programming is accomplished by the
simplified setup. An address is selected by the electrical valves on the address lines,
and then a pulse is applied to those output lines corresponding to bit locations where
zeros (0) are to be stored. These pulses blow the fusible links, thus creating the
desired bit pattern. The next address is then selected and the process is repeated. This
sequence is actually done automatically by a software-driven PROM programmer.
Unlike an ordinary PROM, an erasable PROM (EPROM) can be reprogrammed
if an existing program in the memory matrixer is erased first. An EPROM uses an
NMOSFET matrixer with an isolated gate structure.
The isolated transistor gate has no electrical connections and can store an
electrical charge for indefinite periods. The data bits in this type of matrix are
represented by the presence or absence of a stored gate charge.
Two basic types of EPROMs are the ultraviolet (UV) light EPROM, i.e. the UV
EPROM, and the electrically erasable PROM, i.e. the E2PROM.
One can recognise the UV EPROM matrix by the transparent quartz lid on the
package. The isolated gate in the field-effect transistor (FET) or an ultra-violet
EPROM is ‘floating’ within an oxide insulating material.
The programming process causes electrons to be removed from the floating gate.
Erasure is done by exposure of the memory matrix chip to high-intensity UV
radiation through the quartz window on top of the package. The positive charge
stored on the gate is neutralised after several minutes to an hour of exposure time.
Electrically erasable complementary metal-oxide semiconductor (E2C MOS) can
be both erased and programmed with electrical pulses. Since it can be both
electronically written into and electronically erased, the E2PROM can be rapidly
programmed and erased in matrix for reprogramming.
Two types of E2PROMs are the floating-gate MOS and the metal nitride-oxide
silicon (MNOS). The application of a voltage on the control gate in the floating gate
structure permits the storage and removal of charge from the floating gate.

17.4 Programmable matrix logic (PML)

The PML and the GML are the most common EPMs used for logic implementation.
The PML consists of a programmable ECM of AND gates that connects to a fixed
EPM of OR gates. This structure allows any sum-of-products (SOP) logic expression
with a defined number of variables to be implemented. Note that any logic function
can be expressed in SOP forms.
The basic structure of a PML is illustrated in figure 17.4 for two input variables
and one output, although most PMLs have many inputs and many outputs.
As we know, a programmable matrix is essentially a grid of conductors
forming rows and column with a fusible link at each cross point. Each fused cross
point of a row and column is called a ‘cell’ and is the programmable component
of a PML.
Each row is connected to the input of an AND gate and each column is connected
to an input variable or its complement. By using the presence or absence of fused

17-5
 Mechatronics

A A B B

Figure 17.4. Basic structure of programmable matrix logic (PML) (adapted from Floyd 1994).

connections created by programming, any combination of input variables or

complements can be applied to an AND gate to form any desired product term.

17.5 Generic matrix logic (GML)

The GML in its basic form is an EPM with a reprogrammable AND ECM, a fixed
OR ECM and programmable output logic. The GML can be reprogrammed
repeatedly because it uses E2CMOS technology instead of bipolar technology and
fusible links.
Just as in a PML, this structure allows any SOP logic expression with a defined
number of variables to be implemented.
The basic structure of a GML is illustrated in figure 17.5 for two input variables
and one output, although most GMLs have many inputs and many outputs.
The reprogrammable EGM is essentially a grid of conductors forming rows and
columns with an E2CMOS cell at each cross point, rather than a fuse as in
PML. These cells are shown as blocks in figure 17.5. Each row is connected to the
input of an AND gate, and each column is connected to an input variable or its
complement.
By programming each E2CMOS cell to be either ‘on’ or ‘off’, any combination of
input variables or complements can be applied to an AND gate to form any desired
product term.
A cell that is on effectively connects its corresponding row and column, and a cell
that is off disconnects the row and column.
The cells can be electrically erased and reprogrammed. A typical E2C-MOS cell
can retain its programmed state for 25 years or more.
E2CMOS technology is based on a combination of complementary metal-oxide
semiconductor (CMOS) and non-complementary metal-oxide semiconductor (NMOS)
technologies and is used in GML ECMs such as those above discussed and other
PLMs.

17-6
 Mechatronics

A A B B

E2CMOS E2CMOS E2CMOS E2CMOS

E2CMOS E2CMOS E2CMOS E2CMOS

E2CMOS E2CMOS E2CMOS E2CMOS

E2CMOS E2CMOS E2CMOS E2CMOS X

E2CMOS E2CMOS E2CMOS E2CMOS

E2CMOS E2CMOS E2CMOS E2CMOS

Figure 17.5. Basic E2CMOS matrix structure of generic matrixer logic (GML) (adapted from Floyd 2000).

An E2CMOS cell is built around a MOS transistor with a floating gate that is
externally charged or discharged by a small programming current. A physical model
of this type of cell is shown in figure 17.6.
When the flowing gate is charged to a positive potential by removing electrons,
the sense transistor is turned on, storing a binary zero (0). When the floating gate is
charge to a negative potential by placing electrons on it, the sense transistor is turned
off, storing a binary one (1).
The control gate controls the potential of the floating gate. The pass transistor
isolates the sense transistor from the EGM during read and write operations that use
the word and bit lines. The cell is programmed by applying a programming pulse to
either the control gate or the bit line of a cell that has been selected by a voltage on
the word line. During the programming cycle, the cell is erased by applying a voltage
to the control gate to make the floating gate negative. This leaves the sense transistor
in the off state, storing a binary one (1). A write pulse is applied to the bit line of a
cell in which a binary zero (0) is to be stored. This will charge the floating gate to a
point where the sense transistor is on, storing a binary (0). The bit stored in the cell is
read by sensing the presence or absence of a small cell current in the bit line. When a
binary one (1) is stored, there is no cell current because the sense transistor is off.
When is a binary zero (0) is stored, there is a small current because the sense
transistor is on. Once a bit is stored in a cell, it will remain indefinitely unless the cell
is erased or a new bit is written into the cell.

17-7
 Mechatronics

An E2CMOS cell. Bit line

Pass transistor

Word line
Substrate
Floating gate

Control gate
Sense transistor

Cell ground

Figure 17.6. An E2CMOS cell (adapted from Floyd 2000).

17.6 Electrical commutation matrixer crossbar

A type of mid-19th-century telegraph-exchange mechanical ECM crossbar consisted
of a grid of vertical and horizontal brass bars with a hole at each intersection. The
operator inserted a brass pin to connect one telegraph line to another. An ECM
crossbar consisted of metal bars associated with each matrixer input and matrixer
output, together with some means of controlling the movable contacts (electrical
valves) at each cross point.
In the later part of the 20th century, these literal ECM crossbars declined and
the term came to be used symbolically for rectangular array ECMs in general. The
development of reed-relay switch (electrical valve) cross points was crucial to the
advancement of ECM crossbar implemented digital switching.
Modern ‘electrical-commutation-matrixer crossbars’ are usually implemented
with semiconductor technology. An important emerging class of optical crossbars
is being implemented with technology.
An ECM crossbar (also termed an ‘electrical commutation matrixer cross
point’) is an ECM connecting multiple matrixer inputs to multiple matrixer
outputs in a matrix manner. Originally, the term was used literally, for an ECM
controlled by a grid of crossing metal bars, but later it was broadened to describe
an ECM’s electrical valves in general. It is one of the principal commutation
architectures, together with a rotating switch, memory switch and a crossover
switch. An ECM crossbar is an assembly of individual electrical valves between
multiple matrixer inputs and multiple matrixer outputs. The electrical valves are
arranged in an ECM. If the ECM crossbar has m matrixer inputs and n matrixer
outputs, then a crossbar has an ECM with m × n cross points or places where the
‘bars’ cross (see figure 17.7).
At each crosspoint is an electrical valve; when turned on, it connects one of the m
matrixer inputs to one of the n matrixer outputs. A given crossbar is a single-layer
non-blocking switch. Collections of crossbars can be used to implement multiple
layer and/or blocking electrical valves. An ECM crossbar is also termed a coordinate
electrical commutation matrixer (CECM).

17-8
 Mechatronics

Figure 17.7. Electrical commutation matrixer crossbar (Ogorzalek 2007).

ECM crossbars are characterised by the fact that (Ogorzalek 2007):

• the number of cross points grows as the square of number of stations;
• loss of cross points prevents connection;
• there is inefficient use of cross points:
○ all stations are connected, but only a few cross points are in use;
• they are non-blocking.

ECM crossbars are most commonly used in information processing applications

such as telephony and circuit commutation, but they are also used in applications
such as mechanical sorting machines with inputs.
The ECM crossbar is also used in some semiconductor memory digital devices
implemented with nanotechnology. Here the bars are extremely thin metal wires and

17-9
 Mechatronics

the electrical valves are fusible links. The fuses are blown or opened using high
voltage and they are read using low voltage. Such digital devices are called
‘programmable read-only memory’ (Chen et al 2003). Mouttet (2008) presented a
nanoscale ECM crossbar implementation of an adding circuit being used as an
alternative to logic gates for computation.

17.7 Electrical commutation matrixer tactile sensor

A modern ‘electrical commutation matrixer tactile sensor’ of sensing devices consists
of tactile sensors, which are used for several purposes, such as shape recognition,
contact point determination and pressure/force measurement.
A number of ECM tactile sensors have been proposed in the literature, with
several different solutions concerning their realisable features—optical, piezoresis-
tive, piezoelectric and so on; see, for example, (Pugh 1986; Howe and Cutkosky
1989; Nicholls and Lee 1989) for an overview of the technologies and applications.
ECM tactile sensors have in particular been implemented in automation and
robotics. Currently, as with force sensors, ECM tactile sensors are commercially
available devices. Doubtless, they are the most universally implemented sensorial
ECMs for industrial grippers, even though they are often used as advanced on/off
devices to certify whether a grasp or contact condition occurs (Siciliano and Khabit
2008). Usually, they consist of an ECM (array) of sensing devices (figure 17.8(a)).
Each sensing device is usually referred to as a taxel (from tactile device), and the
whole set of information is termed a tactile image (figure 17.8(b)). The key objective
of this set of sensors is to measure the pressure map over the sensing area.
In general, the types of information that may be obtained from an ECM tactile
sensor are (Siciliano and Khabit 2008):
• contact: this is the most simple information given by the sensor, and concerns
the presence or absence of a contact;
• force: each sensing device provides information related to the amount of locally
applied force, which can be used in several ways for successive elaborations;
• simple geometrical information, i.e. the position of the contact area and the
geometrical shape of the contact itself (planar, circular, and so on);
• the main geometrical features of the object: by proper elaboration of the data from
the taxels, it is possible to deduce the type of object in contact with the sensor, for
example, a sphere or a cylinder (data relative to the three-dimensional shape);
• mechanical properties, such as the friction coefficient or roughness, while the
thermal properties of the object may also be measured by a tactile sensor;
• slip condition, i.e. the relative movement between the object and the sensor.

Several technologies have been adopted for the design of tactile sensors, ranging
from the piezoresistive through the magnetic to the optical. Among the most
common, one can mention (Siciliano and Khabit 2008):
• resistive and conductive effects;
• electromagnetic effects;
• capacitive effects;

17-10
 Mechatronics

Figure 17.8. An electrical commutation matrixer (ECM) tactile sensor: (a) scheme and (b) example data
(Siciliano and Khabit 2008).

• piezoelectric effects;
• optical effects;
• mechanical methods.

Each of these technologies has positive and negative aspects. Common drawbacks,
however, include the size of these sensors, which are usually quite large in comparison
with the available space, and the necessity of a large number of electrical connections.

17.8 Electrical commutation matrixer seven-segment display

A common type of display device that uses light-emitting diodes (LED) is the
electrical commutation matrixer seven-segment display. Combinations of the seg-
ments form ten decimal digits, as illustrated in figure 17.9 (Floyd 2000).
Each segment in the electrical commutation matrixer display is an LED. By
forward-biasing selected combinations of segments, any decimal digit and a decimal
point can be formed. Two types of LED electrical commutation matrixers are the
common anode and the common cathode, as shown in figure 17.10.

17-11
 Mechatronics

F B
G

E D C Decimal
point

Figure 17.9. LED seven-segment arrangement and typical device. (adapted from Floyd 2000)

Figure 17.10. Electrical commutation matrixer seven-segment display. (a) Common-anode commutation
group. (b) Common-cathode commutation group.

17-12
 Mechatronics

Standard LEDs are also used for indication lamps and readout displays on a wide
variety of instruments, ranging from consumer appliances through to scientific
apparatus.

17.9 Fluidic commutation matrixer combinational-chemistry

microreaction
In the near future, scientists and engineers face the task of developing designs
for both micro-electro-mechanical systems (MEMS) and nano-electro-mechanical
systems (NAEMS). These can be standalone systems, like chemical microplants.
Combinational chemistry is an important new methodology developed by research-
ers in the pharmaceutical industry to reduce the time and costs associated with
producing effective and competitive new drugs.
Figure 17.11 shows a state-of-the art chemical microplant as a fluidic commu-
tation matrixer combinational-chemistry microreaction (Koivo 2012).
The production of drugs, chemicals and other substances takes place at
much smaller plants, chemical microplants, which produce less waste, use less
energy and cause less environmental damage. Such inventions will grow in
number, but to make strong advances, this microfluidic-matrix domain is in
need of vigorous efforts in physical and mathematical modelling and simu-
lation (Koivo 2012).
The control of electro-osmotic flows (EOF) in microfluidic electrical valves has
been achieved by using a new generation of flow field effect transistors (FFET) with
integrated polarisable interfaces (PI). This new type of microfluidic electrical valve
was compared to more conventional metal-isolator-electrolyte FFETs.
Using a fluidic Wheatstone bridge configuration, precise control of the liquid flow
rate was obtained with very small voltages (<500 mV). By controlling the EOF
independently of the external electrical field, electrophoretic filter effects could be
observed. The ability of scientists and engineers to integrate such fluidic transistors
in serial or parallel architectures opens up new routes for high-selectivity biomole-
cule sorting using programmable chip matrixers.

CombiChem Microreaction Network

C1 C2
C1 Cn
C2

Cn chemical
product

Figure 17.11. Artist’s impression of a microfluidic electrical commutation matrixer based on flow FETs
(adapted from Koivo 2012).

17-13
 Mechatronics

17.10 Exemplary applications of programmable-logic electrical

commutation matrixer for digital devices
17.10.1 PCM implementing a 2048 × 8 EPROM
Its operation is representative of that of other typical EPROMs. As the logic symbol
in figure 17.12 shows, this EPM has 2049 addresses (2011 = 2048), each with eight (8)
bits. Note that the eight outputs are tri-state (V ).
To read from the memory, the output enable input (OE ) must be LOW and the
power-down/program (CE /PGM ) input must be LOW. To erase the stored data, the
EPM is exposed to high-intensity UV light through the transparent lid. As in most
EPROMs after erasure, all the bits are binary ones (1). To program the EPM, a high
DC voltage is applied to VPP, and OE is HIGH.
The eight (8) data bits to be programmed into a given address are applied to the
outputs O0 through O7, and the address is selected on inputs A0 through A10. Next, a
10 ms to 55 ms HIGH-level pulse is applied to the CE /PGM input. The addresses
can be programmed in any order.
Another example of EPM implementing an EPROM is shown in figure 17.13,
where it is programmed by blowing a fuse link to create a binary zero (0).

17.10.2 PCM implementing a sum-of-products (SOP) expression

In its simplest form, each cell in a basic AND ECM consists of a fusible link
connecting a row and a column, as represented in figure 17.14.
When a connection between a row and column is required, the fuse is left intact.
When no connection between a row and column is required, the fuse is blown open

EPROM
2048 × 8
(8)
A0 0
(7) (9)
A1 ∇ O0
(6) (10)
A2 ∇ O1
(5) (11)
A3 ∇ O2
(4) (13)
A4 ∇ O3
(3) (14)
A5 A 0
2047 ∇ O4
(2) (15)
A6 ∇ O5
(1) (16)
A7 ∇ O6
(23) (17)
A8 ∇ O7
(22)
A9
(19)
A10 10
(18)
CE/PGM
& VCC = +5 V Pin 24
(20) EN VPP = +5 V Pin 21
OE VSS = Gnd Pin 12

Figure 17.12. The logic symbol for a 2094 × 8 EPROM (adapted from Floyd 2000).

17-14
 Mechatronics

0 +V
l 2 3 4 5 6 7 8 9

1
11 12 13 14 15 16 17 18

2
19 20 21 22 23 24 25 26 27

3
1 28 29 30 31 32 33 34 35 36

X 2 4

4 37 38 39 40 41 42 43 44 45

5
46 47 48 49 50 51 52 53 54

6
55 56 57 58 59 60 61 62 63

7
64 65 66 67 68 69 70 71 72

28 27 26 25 24 23 22 21 20

V3

Figure 17.13. The EPROM which is programmed by blowing a fuse link (adapted from Floyd 2000).

A A B B

Figure 17.14. Basic structure of a PML (adapted from Floyd 2000).

17-15
 Mechatronics

during the programming process. As one can see, the fusible links are left intact to
connect the desired variables or their complements to the appropriate AND gate
inputs.
As an example, a simple PML matrixer is programmed as shown in figure 17.15
so that the product term AB is produced by the top AND gate, AB is produced by
the middle AND gate and AB is produced by the bottom AND gate.
The fusible links are opened when a variable or its component is not used in a
given product term. The final output from the OR gate is the SOP expression,
X = AB + AB + AB.
The programmed polarity output in a PML matrixer operates by selectively opening
or leasing intact a cell on one input of an XOR gate, causing the gate to invert its
other input or not. See figure 17.16.
The selected matrixer inputs determine the combinational or registered mode,
matrixer input or matrixer output and inverted or non-inverted matrixer output. See
figure 17.17.

A A B B

X = AB + AB + AB

Figure 17.15. PML implementation of a SOP expression X = AB + AB + AB (adapted from Floyd 2000).

A A B B C C A A B B C C

X X

(a) X = ABC + ABC + ABC (b) X = BC + ABC + ABC

Figure 17.16. PML implementation of SOP expressions (adapted from Floyd 2000).

17-16
 Mechatronics

A A B B C C A A B B C C

X X

(a) X = ABC + ABC + AB + BC (b) X = AB + AC + AB + BC

Figure 17.17. PML implementation of SOP expressions (adapted from Floyd 2000).

00X 10
X01 11
11X 10

a b c f1 f2

a a⬘ b b⬘ c c⬘ f1 f2

Figure 17.18. PLM implementation of the expressions f1 = a′b′ + b′c + ab and f2 = b′c (adapted from De
Micheli 2007).

Figure 17.18 shows a PLM implementation of the expressions f1 = a′b′ + b′c + ab

and f2 = b′c (De Micheli 2007).

17.10.3 GML implementing a sum-of-products (SOP) expression

As an example, a simple GML EDM is programmed as shown in figure 17.19 so that
the product term AB is produced by the top AND gate, AB is produced by the
middle AND gate and AB is produced by the bottom AND gate.
As shown, the E2CMOS cells are on to connect the desired variables or their
complements to the appropriate AND gate inputs. The E2CMOS cells are off where
a variable or its complement is not used in a given product term. The final output
from the OR gate is an SOP expression.

17-17
 Mechatronics

A A B B

Off On Off Off

AB
Off Off On Off

On Off Off Off

AB
Off Off On Off X = AB + AB + AB

Off On Off Off

Off Off Off On

AB

Figure 17.19. GML implementation of a SOP expression X = AB + AB + AB (adapted from Floyd 2000).

17.10.4 Electrical commutation matrixer seven-segment display to put on show ‘5’

Figure 17.20 shows how to connect the electrical commutation matrixer seven-
segment display shown in figure 17.18(a) to put on show ‘5’. The maximum value of
the continuous forward current for each LED is 30 mA and a +5 V source is used.
The maximum forward current for each LED is 30 mA and a +5 VDC voltage
source is to be used.

17.11 Conclusion
PMLs are programmed by leasing specific fusible links intact and blowing open all
the others. GMLs are programmed in a similar way, except that the E2CMOS cells
are turned on or off. The logic functions to be implemented determine which cells
are affected. In order to program a PML or GML, the following items are required:
a computer, programming software and a PML or GML programmer.
As mentioned earlier, there are several software packages for implementing logic
design in PMLs and DMLs. ABEL and CUPL are two commonly used hardware
description languages (HDL). Because both languages are similar and produce the
same results in terms of programming a PCM or GCM, use is often a matter of
preference and availability.
When considering tactile sensors, it is useful to begin by considering the
fundamental physical quantities that can only be sensed through contact with the
environment. The most important quantities measured with touch sensors are shape
and force. Each of these may be measured as either an average quantity for some
part of the robot or as a spatially resolved, distributed quantity across a contact
area.

17-18
 Mechatronics

Figure 17.20. The ECM seven-segment display to put on show ‘5’.

PECMs and DECMs are fundamental to modern flat-panel displays. Thin-film-

transistor LCDs have a transistor at each cross point, and so including an ECM
crossbar as part of their structure might be considered.
For video switching in home and professional theatre applications, an ECM
crossbar is used to make the output of multiple video appliances available to every
monitor or every room throughout a building simultaneously.
In a typical installation, all the video sources are located on an equipment rack,
and they are connected as matrixer inputs. Where central control of the

17-19
 Mechatronics

commutation matrixer is practical, a typical rack-mounted ECM crossbar offers

front-panel buttons to allow the manual connection of matrixer inputs to matrixer
outputs. An example of such a use might be in a so-called ‘sports bar’, where
numerous programs are displayed simultaneously.
In order to accomplish this, a sports bar would ordinarily need to purchase a
separate cable or satellite subscription for each display for which independent
control is desired. The ECM crossbar enables the signals to be re-routed on a whim,
thus allowing the establishment to purchase only those subscriptions needed to cover
the total number of unique programs viewed anywhere in the building.
Such ECM crossbars are used in high-end home theatre applications. The video
sources that are typically shared include set-top cable/satellite receivers and DVD
changers, while the same concept also applies to audio devices. The matrixer outputs
are wired to televisions in individual rooms.
The ECM crossbar may be controlled (via an Ethernet or RS-232 serial
connection) by a whole-house automation controller (such as those made by
AMX, Crestron, or Control4), which provides the user interface that enables the
user in each room to select which appliance to watch. The actual user interface varies
according to system brand, and might include a combination of on-screen menus,
touch-screens and handheld remote controls.
The ECM crossbar is necessary to enable users to select the program they wish to
watch from the same room they will watch it from; otherwise it would be necessary
for them to walk to the equipment rack.
The special ECM crossbars used in distributing satellite TV signals are called
‘multimatrixers’. ECM crossbar technology opens up the opportunity to use
computational resources to achieve automatic commutation or switching.

References
Broesch J D 1991 Practical Programmable Circuits: a Guide to PLDs, State Machines, and
Microcontrollers (San Diego, CA: Academic Press)
Carter J W 1997 Digital Programming with Programmable Logic Devices (Upper Saddle River,
NJ: Prentice Hall) p 418
Chen Y, Jung G-Y, Ohlberg D A A, Li X, Stewart D R, Jeppesen J O, Nielsen K A and Stoddart
J F 2003 Nanoscale molecular-switch cross bar circuits Proc. of the 2003 Nano-technology 14
462–8
De Micheli 2007 Two-level Logic Synthesis and Optimization. Presentation Centre SI: <Two-
Level Exact [Link]>, Integrated Systems Centre, EPF Lausanne, Switzerland
Floyd Th L 1994 Digital Fundamentals (New York: Macmillan)
Floyd Th L 2000 Digital Fundamentals (Upper Saddle River, NJ: Prentice Hall)
Howe R D and Cutkosky M R 1989 Sensing skin acceleration for texture and slip perception
Proc. IEEE Int. Conf. Robotics and Automation (Scottsdale, AZ)
Koivo H 2012 Challenging paradigms in modelling and simulation. Whitepaper 104, Theme (T.4),
Helsinki University of Technology, Control Engineering Laboratory, FIN-02015 HUT,
FINLAND [Link]
Mouttet B 2008 Logicless computational architectures with nano-scale crossbar arrays. NSTI
Nanotech 2008 Conf.

17-20
 Mechatronics

Nicholls H R and Lee M H 1989 A survey of robot tactile sensing technology Int. J. Robot. Res.
8 3–30
Ogorzalek M J 2007 Telecommunication Technologies, Week 11, Circuit Switching—Packet
Switching Presentation EIE 325 Telecommunication Technologies (Dept. of EIE, Hong Kong
Polytechnical University)
Pinto S F, Silva J F, Pellerin D and Holley M 1994 Digital Design Using ABEL (Upper Saddle
River, NJ: Prentice Hall)
Pugh A 1986 Robot Sensors: Tactile and Non-Vision vol 2 (Berlin: Springer)
Sharma A K 1998 Programmable Logic Handbook: PLDs, CPLDs, and FPGAs (New York:
McGraw-Hill)
Siciliano B and Khabit O (ed) 2008 Springer Handbook of Robotics (Berlin: Springer)

17-21
 IOP Publishing

Mechatronics
Dynamical systems approach and theory of holors
B T Fijalkowski

Chapter 18
Nano-magneto-rheological fluido-mechanical
commutation matrixers for internal combustion
engines

18.1 Introduction
The automotive world is moving towards hybrid-electric vehicles (HEV) in every
respect other than actual conventional automotive vehicles. Enthusiasm for all-electric
vehicles (AEV), a feature of the early 1990s, has declined a little. It is now certain that
HEVs are within reach of the advanced hybrid power source (HPS), which is an
advanced internal combustion engine (ICE) without crankshaft and connecting rod
(conrod) mechanisms (C2M), conceived and developed by the author (Fijalkowski
1986).
As a result, the generation changes in C2M, which acts as linear reciprocating motion
to rotary motion (LM–RM/RM–LM) and vice versa (nano-magneto-rheological
(NMR) fluido-mechanical commutation matrixer (FCM) LM–RM/RM–LM
commutator), are taking place throughout the world.
The author will focus on the main innovations and trends for tri-mode hybrid-
electric (HE) all-wheel-driven (AWD) drive-by-wire (DBW) propulsion and all-wheel-
braked (AWB) brake-by-wire (BBW) dispulsion mechatronic control systems, as well
as a two- or four- or even five-stroke thermodynamic cycle, twin-opposed-piston,
crankless reciprocating ICE, termed the ‘Fijalkowski engine’, with a nano-magneto-
rheological fluid (NMRF) FCM LM–RM/RM–LM and/or RM–RM commutator.
Despite the recent emphasis on fuel cells, ICEs will continue to be the
chemo-thermo-fluido-mechanical (Ch-Th-FM) energy converters of choice in
automotive applications for the next few decades. However, the decades ahead are
likely to be characterised by a fragmentation of the engine market, with the emergence
of many competing technologies and a blurring of the distinction between engine

doi:10.1088/978-0-7503-1350-6CH018 18-1 ª IOP Publishing Ltd 2016

 Mechatronics

families. The driving forces behind these technological developments are, and will
continue to be, dominated by energy efficiency, emission regulations and drivability,
coupled with great emphasis on reliability, manufacturability, cost, diagnostics,
serviceability and recyclability. Emerging technologies in sensors, actuators and
advanced mechatronical control systems are opening up many parallel paths, and
the future of the ICE is likely to be shaped by the economic and regulatory climate on
a global scale.
This section provides an overview of emerging trends, opportunities and
challenges in ICE technology. Most conventional ICEs are reciprocating ones.
Pistons that move back and forth inside enclosed cylinders power reciprocating IC
engines. The LM of the pistons is converted to RM by a conventional crankshaft or
geared crankshaft, bent axis, swash plate, wobble plate, transition arm or cam
mechanism to convert LM to RM, or vice versa, to perform useful work such as
turning the wheels of an automotive vehicle or the rotors of a helicopter.
The crankshaft, sometimes colloquially abbreviated to ‘crank’, is the part of an
ICE that converts the LM of the pistons into the RM of the crankshaft. It typically
connects to a flywheel to reduce the pulsation characteristic of the two- or four-
stroke thermodynamic cycle, and sometimes to a torsion or vibration damper at the
opposite end to reduce the torsion vibrations that are often caused along the length
of the crankshaft by the cylinders furthest from the output end acting on the torsion
elasticity of the metal.
The crankshaft is subjected to various forces, but it needs to be checked in two
positions. Firstly, failure may occur at the position of maximum bending. In such a
condition, the failure is due to bending and the pressure in the cylinder is maximal.
Secondly, the crank may fail due to twisting, so the crankpin needs to be checked for
shear at the position of maximal twisting. The pressure at this position is not
maximal pressure, but a fraction of the maximal pressure. Thus, the most important
mechanical motions of the ICE are the LM of the pistons and connecting rods
(conrods) and the RM of the crankshaft. The latter converts the LM of the pistons
into the RM of the crankshaft. Then, the RM can be output to the driving wheels of
the automotive vehicle through the clutch, transmission and shafts. During the
combustion and expansion processes, the high cylinder pressure pushes the piston
motion to bottom dead centre (BDC). During compression and intake/exhaust
processes, the inertial of the flywheel and the cylinder pressure of other cylinders
drive the piston motion to top dead centre (TDC).
The novel ICE without a C2M that the author would like to offer is based on
another law of physics. The invention called the ‘nano-magneto-rheological mecha-
tronical commutator internal combustion engine’, more concisely termed the
‘Fijalkowski engine’ elsewhere (Pinto-Silva 1987; De Falco 1989), is based on a
completely novel propulsion engineering solution that has no analogies in the world.
Furthermore, with the elimination of the conventional C2M, its friction and mass
are substantially reduced, which results in even greater mechanical energy output.
The Fijalkowski engine, as conceived in the 1980s and first presented and
published on in October 1986 at the 8th International Electric Vehicle Symposium
in Washington, DC (Fijalkowski 1986), has only one or more moving parts, the

18-2
 Mechatronics

piston-rod assembly. It may be constructed of one or more pairs of directly opposed

pistons with opposed cylinders and heads, and one or more IPM arrays fixed to the
piston rod between them, or driven by an NMR FCM LM–RM/RM–LM and/or
RM–RM commutator, which replaces the conventional C2M. The piston-rod
assembly may shuttle back and forth in a straight line from compression–ignition
to compression–ignition in its opposing cylinders.
Mechanical energy from the combustion and expansion of a fuel mixture is
concentrated in a straight line, and because the pistons do not experience angular
loading in the cylinders (as seen in conventional crankshaft reciprocating ICE
configurations), friction is substantially reduced, which translates into greater
mechanical energy output.
The Fijalkowski engine has major advantages; it can burn a large variety of liquid-
fossil as well as gaseous-nitrogen fuels, and has good potential for depollution.
The new technology, which was discovered by the author (Fijalkowski 1986), is a
highly efficient method for converting LM to RM in advanced reciprocating ICEs,
mechano-fluidic (MF) pumps or compressors, and mechano-vacuum (MV) pumps, as
well as fluido-mechanical (FM) and pneumo-mechanical (PM) motors. This new
method may also provide an effective way to vary the capacity of MF pumps and
MP compressors, as well as the compression ratio of diesel and Otto engines.
The specific result is an advanced opposed piston, opposed cylinder compression
ICE that is much more fuel-efficient than conventional ICEs and may therefore save
users millions of dollars in fuel costs each year. Taking into account the instanta-
neous ICE friction, torque is decisive in recognising which ICE components are
contributing the most towards the overall frictional losses of the ICE. The
attenuation of frictional losses for such components may also rationalise the fuel
consumption and increase the power output of the ICE. Therefore, one of our aims
ought to be the development of a tool for reckoning the instantaneous friction
torque of the C2M.
The invention that is being offered is from the area of propulsion engineering,
specifically the area of the opposed-piston, opposed-cylinder ICE, which is wide-
spread. It is a popularly known fact that ICEs have a C2M, except for rotary ICEs.
The C2M has a crankshaft, which is a key part of that mechanism. Manufacturing
and servicing that part is costly. According to various sources, about 30% of the cost
of the entire ICE comes from the crankshaft alone. The presence of a C2M causes the
reciprocating ICE to vibrate, and this leads to many technical problems. Therefore,
attempts to create an ICE without a C2M have been made for quite a while,
however, to date the majority of ICEs still use them. So-called rotary ICEs are an
exception (e.g. Wankel engines), and they are used extremely rarely because of their
difficult servicing procedures and complicated structure. The necessity of separating
the lubrication system will lead to designs with outside scavenging, such as
mechanical blowers, and maybe a design solution similar to the Fijalkowski engine
can be found.
Figure 18.1 shows the design animation of the Fijalkowski engine’s NMR FCM
LM–RM/RM–LM commutator. The four pairs of opposing pistons of a ‘boxer’
ICE arrangement are solidly connected to each other as well as to an NMR

18-3
 Mechatronics

Figure 18.1. Design of the Fijalkowski engine’s NMR FCM LM–RM/RM–LM commutator.

mechatronical commutator by means of round piston rods. There are separating

walls between the NMR FCM LM–RM/RM–LM commutator case and the
underside of the pistons. Because the piston rods are guided and are supported by
bearings in the separation wall, the pistons hardly touch the cylinder walls, there are
no side forces, and there is no piston tipping at the dead centres. As is the case with
the conventional two-stroke thermodynamic cycle reciprocating ICE, the bottom
side of the pistons may function as charge pumps, one for each cylinder. Because the
ceramic pistons do not need to be lubricated, the combustion air will remain free of
lubricating oily fluid.
As can be seen in this Fijalkowski engine, the two-stroke thermodynamic cycle
may allow us to place great expectations in novel non-metallic composites such as
fibreglasses (e.g. Kevlar and Technora) or ceramic materials. Ceramics are assumed
to construct a highly loaded piston that will be thermally stable. It seems possible
that cylinders constructed from ceramics will develop less thermal energy (heat)
distortion and less wear. The components of the Fijalkowski engine that may be
constructed from ceramic materials are: pistons; piston rings (if any); piston rods’
mover (forcer) blocks inside the NMR FCM LM–RM/RM–LM and RM–RM
commutator; cylinder sleeves; and exhaust channels. There are many advantages in
the design of the Fijalkowski engine.
Because there is no gudgeon pin or short skirt, the piston will only be about half
the mass of its conventional counterpart. This in turn means that the primary
vibration forces may be reduced by about 50%. Since the pistons may be connected
rigidly, because of the absence of a gudgeon pin, the angular motion of the sliding
followers (piston rods), and therefore the secondary vibration forces, may be
eliminated. This all means that such an unconventional reciprocating ICE with
fewer cylinders than a conventional one does not have unacceptable levels of
vibration. Because the combustion-gas forces act perpendicularly to the opposed
pistons, their resultant (as the total outcome of these forces) in the Fijalkowski
engine may always be reacted to by pure compression in the sliding followers (piston
rods). This and the reduction of lateral acceleration forces owing to the light mass of
the opposed pistons may virtually eliminate piston slaps. The value of the inertia

18-4
 Mechatronics

force will be smaller than the value of the combustion-gas force between the rings and
the bores at all times, while their maximum values do not occur simultaneously—the
lateral components of the inertia force are zero at the end of each stroke.
Consequently, ring flutter should not occur and the rate of wear for the rings and
bores should be low. By virtue of the shortening of the piston skirt, the oily-fluid
drag (if any) ought to be about 5% of that of a conventional piston. Therefore, the
cold cranking (starting) of crankless ICEs should be easier, and so a smaller starting,
lighting and ignition (SLI) chemo-electrical/electro-chemical (ChE/ECh) storage
battery and an ECM DC–AC commutator starter motor would be needed.
In addition, the length of the cylinder bore to be machined might be only about
two-thirds of that required conventionally. Many of these advantages would be
of special benefit for HEVs, and the author has also designed two- or four- or even
five-stroke thermodynamic cycle versions with triangular, rectangular and oval
twin-opposed pistons, instead of circular twin-opposed pistons, which may also be
utilised.
The Fijalkowski engine concept presented below may operate on the high-
pressure direct injection (HDI) and constant-volume four-stroke thermodynamic
cycle principle. The twin-opposed pistons move almost symmetrically in their
horizontal ICE cylinders, and each of their LMs may be converted into the RM
of the ICE’s output shaft by means of a pair of sliding followers (piston conrods)
with their bearings, and the main shaft’s NMR rotary ratchets, and the NMR rotary
clutch, with the NMR rotary ratchets’ mechanical harmonic oscillators shifted at π
rad. This engine has a simple and low-cost supercharging system, which may have a
compact volume and the freedom to position its components. The HEV may have
lower values of under-bonnet (under-hood) temperature.
The Fijalkowski engine may have small dimensions, but it will produce much
more power than a conventional reciprocating IC engine. With small cylinder sizes
(three times smaller than a conventional reciprocating IC engine’s), this particular
crankless reciprocating IC engine may develop two–three times more power than a
conventional one. This crankless reciprocating IC engine may be described as
‘square’—a term indicating that the bore and stroke are equal. This engine may have
a high velocity, and it may have no C2M, but it may have only piston rods and
pistons, and it may be successfully utilised in technologies and products that
necessitate high-velocity small-size crankless reciprocating ICEs.

18.2 The Fijalkowski engine concept

An advanced hybrid power source (HPS), termed the Fijalkowski engine, has been
preferred to other solutions with conventional crankshaft reciprocating ICEs. This
HPS, that is, the Fijalkowski engine (see figure 18.2), with its silent ‘four-in-one’,
electronically commutable, magneto-electrically excited, high-temperature super-
conductor, brushless ECM AC–DC/DC–AC commutator, interior permanent
magnet (IPM) flywheel/damper generator/motor, may provide chemo-thermo-
fluido-mechano-electrical (Ch-Th-F-ME) or Ch-Th-FM energy conversion for
HE AWD DBW propulsion.

18-5
 Mechatronics

Figure 18.2. Layout of the Fijalkowski engine with its NMR FCM LM–RM/RM–LM and RM–RM
commutator and two-, four-, or even five-stroke thermodynamic cycle rigidly connected twin-opposed piston
configuration (A) Transverse section and (B) longitudinal section (Fijalkowski 1986, 1988d, 1989, 1999b, 2006).

Figure 18.3. Nano-magneto-rheological fluid (NMRF) action. Credit: Centro Richerche, FIAT (Folonari
2006).

At the very heart of this novel HPS is a nano-magneto-rheological fluid (NMRF).

This is essentially a suspension of micro-sized, magnetically polarisable nano-
magnets embedded on micro-particles in oily fluid or some other liquid. Under
normal operating conditions, NMRF is a free-flowing liquid with a consistency
similar to that of ICE oily fluid (Folonari 2006).
Exposure to a magnetic field, however, may transform the NMRF into a nearly
solid state in milliseconds. Just as quickly, the NMRF may return to its liquid state
with the removal of the magnetic field. The degree of change in an NMRF is
proportional to the magnitude of the applied magnetic field. When it is subjected to
a magnetic field, an NMRF actually develops yield strength and behaves as a
Bingham solid. The flow ratio of NMRFs may vary with the application of a
magnetic field, as shown in figure 18.3 (Folonari 2006).
The advantages of working with nano-magnets as compared to micro-particles
are as follows:
• reduced wear;
• reduced aggregation;

18-6
 Mechatronics

• reduced power consumption;

• faster response time.

The applications may be as follows:

• high-speed NMR mechanical components;
• active NMR shock absorbers (dampers);
• NMR rotary ratchets;
• NMR brakes;
• NMR clutches.

The change may appear as a very large change in effective viscosity. NMRF is
similar to a giant-electro-rheological fluid (GERF), but it is 20–50 times stronger
(Wen et al 2005, Novak 2005). It may also operate directly from a low-voltage
(14 VDC or 28 VDC, or even 42 VDC) automotive onboard electrical energy
distribution (E2D) system, while it is far less sensitive to contaminants and extremes
in temperature. Applied to a novel NMR mechatronical commutator prime mover,
that is, the Fijalkowski engine, NMR technology may provide flexible mechatron-
ical control capabilities in designs that may be far less complicated and more reliable
than conventional crankshaft reciprocating ICEs.
With an efficient low-voltage E2D system and a response time of a few milli-
seconds, two fully controllable NMR rotary ratchets, which are real bipolar
mechanical valves in a novel NMR mechatronical commutator prime mover, offer
one of the most effective means available for interfacing mechanical components
with an HE AWD DBW propulsion mechatronical control system (Fijalkowski
1986, 1997, 2011).
The simple design of the NMR rotary ratchets, which are real NMR fully
controllable bipolar mechanical valves with few moving parts, translates into high
reliability and quiet operation.
NMR rotary ratchets are real NMR rotary-ratchet composite mechanical
harmonic oscillators provided with NMR fully controllable bipolar mechanical
valves, operating with a mechatronical catch or a pawl that prevents the outer
composite mechanical harmonic oscillator from slipping back and allows them to
move in only a single sense of rotary direction (rotation).
They are smooth acting, with proportional brakes that are more compact and
require substantially less power to operate than magnetic-hysteresis or eddy-current
ones. Their simplicity and ease of mechatronical control makes them a cost-effective
choice for a novel NMR prime mover. They may be capable of high torque and low
velocity, and can be used in lock/unlock applications.
The author has focused his attention on the complexity as well as the physical
mass and size of the conventional IC engine’s gudgeon-pin arrangement, the close
tolerances entailed in their manufacture and the time spent on assembly. The author
conceived and developed an advanced HPS, termed the Fijalkowski engine, that is
nearly conventional except that its pistons and their sliding followers (piston rods)
are rigidly connected (see the transverse section in figure 18.2(A)).

18-7
 Mechatronics

The Fijalkowski engine in question may have eight ICE cylinders that are
arranged co-axially with the ICE output shaft and rotate the latter by means of
two NMR rotary ratchets and a single NMR rotary clutch on it, and sliding
followers (piston rods). The two pistons of each pair of ICE cylinders are of the
opposed type and are therefore dynamically balanced. The Fijalkowski engine’s
pistons longitudinal motion, which is LM, may be converted to RM by unconven-
tional cylindrical NMR rotary ratchets, with sliding followers (piston rods),
mounted in the housing, engaging sinusoidal harmonic oscillators around the
shanks of the piston assembly units, while splines couple the pistons’ rotation to
an output shaft passing through the axle of housing, as shown in figures 18.2 and
18.4. If further development proves to be satisfactory, it could have a considerable
impact on the Fijalkowski engine design. For vehicular applications, high thermal
efficiency is important for good fuel economy, but it is also important that the power
plant is simple and compact.
Preliminary design optimization studies resulted in an advanced NMR mecha-
tronical commutator, four-stroke thermodynamic cycle, reciprocating ICE concept
with a relatively high thermal efficiency. This IC engine has not generally been
favoured for HEV applications in the past. However, use of a direct-coupled
automotive silent four-in-one composite flywheel damper onboard mechano-elec-
trical (ME) generator/starter electro-mechanical (EM) motor, with macro- and
microelectronic control of the electrical energy transmission, may make the constant
IC engine’s output shaft angular velocity running feasible, maintaining near-optimal
ICE conditions over a wide power range.
The Fijalkowski engine, that is, the crankless reciprocating IC engine with
reciprocating twin-opposed pistons, may have an output shaft connected to an
NMR rotary clutch with MRF rotary ratchets, and several radially located opposed
pistons engaging NMR rotary ratchets’ mechanical harmonic oscillators, mecha-
tronically controlled so that their power strokes balance the opposing pistons (see
figure 18.4).
The crankless reciprocating ICE may include housing, and an output main shaft
radially connected to the NMR rotary ratchets. Eight ICE cylinders in the housing
may be spaced about the NMR rotary clutch’s axis of rotation, so that each cylinder
may be coupled to a sliding follower (piston rod) that engages the NMR rotary

Figure 18.4. Design of the Fijalkowski engine.

18-8
 Mechatronics

ratchets so that the piston’s movement between the BDC may drive it round, thus
imparting torque through the NMR rotary clutch to the output shaft.
The Fijalkowski engine’s thermodynamic cycle may be mechatronically con-
trolled so that the power stroke of any one piston precedes that of the one that is the
most radially distant from that one and the approaching TDC. This may keep the
crankless ICE well balanced.
The Fijalkowski engine may have low friction vibration, overall mass and
combustion pressure, resulting in high efficiency, low NOx emissions, low compo-
nent wear, and cool and quiet running.
The Fijalkowski engine will be an unconventional ICE, with eight ICE cylinders
of approximately the same length as conventional IC engine cylinders, giving the
same power output as a conventional boxer IC engine, and eight or 12 opposed
pistons with sliding followers (piston rods) and a rotating output shaft, four-in-one,
automotive silent, electronically commutable, magnetoelectrically excited, high-
temperature superconductor, brushless ECM AC–DC/DC–AC commutator, com-
posite-flywheel/damper onboard generator/starter motor that can convert the
mechanical energy supplied to the opposed pistons by the combustion of the usual
hydrogen or hydrocarbons directly into electrical energy.
The Ch-Th-M part of the Fijalkowski engine may have the following advantages:
• use of the constant-volume four-stroke thermodynamic cycle principle;
• use of automotive mechatronics to ensure the starting, synchronization and
delivery of usable mechanical energy, and the ability to ignite either of the
opposed pistons on many short circular ICE cylinders;
• use of ICE output shafts with four-in-one generator/motor rotor inertia to
return the opposed pistons;
• use of sliding-follower/output-shaft NMR FCM LM–RM/RM–LM and
RM–RM commutators to replace the C2M, which may permit the elimination
of the inlet (intake) and outlet (exhaust) valves, pushrods, lifters, camshafts,
timing gear and timing belt or chain required in conventional ICEs; thus it
might require fewer parts and it might be simpler to manufacture, while
lowering the consumption of metal reserves, and it may not require a separate
ICE flywheel.

Then there is the ME/EM part of the Fijalkowski engine, that is, the four-in-one,
automotive silent, electronically commutable, magnetoelectrically excited, high-
temperature, superconductor, brushless-type ECM AC–DC/DC–AC commutator,
composite-flywheel/damper on-board generator/starter motor (see figure 17.20). It
may have two 2 × 5-phase, pentagon-connected, wire-wound (slotted-core), disc-
shaped outer stators and an unwound IPM disc-shaped inner rotor (also see figure
18.5).
Armature alternating voltages, which are real ECM AC–DC commutator-input
alternating voltages, can be rectified to armature direct voltages, that is, ECM AC–
DC commutator-output direct voltages, or alternatively, to armature direct vol-
tages, that is, ECM DC–AC commutator-input direct voltages. The latter can be
inverted to armature alternating voltages, which are real ECM DC–AC commutator

18-9
 Mechatronics

Figure 18.5. Layout of the Fijalkowski engine. (A) The silent four-in-one, electronically commutable,
magnetoelectrically excited, high-temperature superconductor, brushless ECM AC–DC/DC–AC commutator.
(B) IPM flywheel/damper generator/motor.

output alternating voltages, modulated in both amplitude and frequency in sinus-

oidal pulse-width modulation (SPWM) mode to produce armature sinusoidal phase-
current waveforms in the electrical energy, or vice versa, in advantageous conditions
with regard to mass and volume as well as efficiency.
The Fijalkowski engine may be designed to perform three key functions: reduce
fuel consumption by 20% compared to previous reciprocating IC engines, increase
drivability and minimize exhaust emissions.
The first objective may be achieved by implementing the latest fuel HDI
technology—common rail (CR)—which involves mechatronically controlled high-
pressure injection. The Fijalkowski engine architecture may also be designed to
reduce fuel consumption.
The second objective is to persuade consumers to forget about the Fijalkowski
engine’s poor image. The HDI may provide 50% additional torque at low values of
the engine’s output shaft angular velocity and add more power while reducing noise
and vibration. Superior environmental performance may also be achieved.
The Fijalkowski engine may be one of the cleanest reciprocating ICEs on the
automotive market, while offering considerable potential for further development in
the future. The longitudinal section of the NMR reciprocating prime mover, that is,
a crankless, reciprocating, constant-volume, two- or four-stroke thermodynamic
cycle, eight-cylinder, opposed IC engine, called the Fijalkowski engine, is shown in
figure 18.6.
Its transmission, by which power may be transmitted from the ICE to the live axle(s)
in an HEV, may represent a transmission drum, which consists of an outer NMR
rotary ratchet NMR FCM LM–RM/RM–LM commutator (see figure 18.7), acting as
a mechatronical LM–RM rectifier (during ICE operating action), or an NMR FCM
RM–LM inverter (during ICE cranking action), a floating composite mechanical

18-10
 Mechatronics

Figure 18.6. Longitudinal section of the Fijalkowski engine.

harmonic oscillator, and an inner NMR rotary ratchet’s composite mechatronical

common drum.
The reciprocating motion of two parallel acting IC engine pistons’ straight piston
rods of the crankless, constant-volume, four-stroke, thermodynamic cycle, eight-
cylinder opposed reciprocating ICE may be transferred tangentially to a pair of
NMR rotary ratchets, which are a real NMR rotary ratchet’s fully controlled
bipolar mechanical valves.
The NMR rotary ratcheting action may result from locking of the NMR rotary
ratchet’s composite mechanical harmonic oscillator when a magnetic-field exciter’s
current pulses solidify the NMRF. This may cause the inner NMR rotary ratchet
composite mechanical common drum to rotate with uniform RM (see figures 17.18–
18.8).
The uniform RM may in turn be transferred to a central NMR rotary clutch’s
composite mechanical common hub by a continuously variable-velocity NMR rotary
clutch. The latter controls the mechanical energy throughout (see figure 17.18).
The layout of the Fijalkowski engine’s NMR FCM LM–RM/RM–LM and RM–
RM NMR commutators is shown in figure 18.7.
Mechanical energy is mechatronically controlled by variation of the magnetic-field-
exciter current of the outer NMR rotary-clutch’s composite mechanical common
hub’s NMR layer, or by current-pulse controlled ‘quick passage’ of the outer-clutch
composite mechanical common hub from pushing to dragging current pulses.
A very advanced novel HPS that is the Fijalkowski engine (see figures 18.1–18.8)
supplies reciprocating mechanical energy to a pair of its straight piston rods, which
in turn supply reciprocating mechanical energy tangentially to a pair of NMR
rotary-ratchet composite mechanical harmonic oscillators, that is the NMR rotary-
ratchets’ fully controlled bipolar mechanical valves.
Synchronized current pulses of the NMR rotary-ratchet magnetic-field-exciter latch
the presently forward RM outer rotary-ratchet’s composite mechanical harmonic
oscillator to an inner NMR rotary-ratchet’s composite mechanical common drum,
while the alternate outer NMR rotary-ratchet composite mechanical harmonic
oscillator slips backward to reset.

18-11
 Mechatronics

Figure 18.7. A transmission drum that consists of an outer NMR rotary ratchet NMR FCM LM–RM/RM–
LM commutator acting as a floating composite mechanical harmonic oscillator, and an inner NMR rotary
ratchet composite mechatronic common drum.

Upon resetting, the situation is reversed, and the previously latched NMR rotary
ratchet deactivates and slips backward to reset while the previously slipping NMR
rotary ratchet is reactivated to latch onto a forward stroke of the straight piston rod.
In this manner, the sum of the two alternating NMR rotary ratchets action
provides continuous forward RM to the central NMR rotary-clutch’s composite
mechanical common hub (see figures 18.7 and 18.8).

18.2.1 Magnetic-field-exciter’s electric-current sequencing for rotary motion

The reversal of RM can be realised easily by swapping the NMR rotary ratchets’
magnetic-field-exciter’s electric-current-pulse sequencing, as shown in figure 18.9.
Thus, the sense of RM direction (rotation) is changed by skipping one pulse on the
left-hand NMR rotary-ratchet’s fully controllable bipolar mechanical valve to allow
the left-hand NMR rotary-ratchet composite mechanical flywheel cylinder to slip,
while one NMR rotary-ratchet’s magnetic-field-exciter’s electric-current pulse is

18-12
 Mechatronics

Figure 18.8. Layout of the Fijalkowski engine’s NMR FCM LM–RM/RM–LM and RM–RM commutators.
TFR, TFL— NMR rotary-ratchet composite harmonic oscillators’ forward torques for the right- and left-hand
IC engine output-shaft rotations, respectively; TBR, TBL—NMR rotary ratchet composite harmonic oscil-
lators’ backward torques for right- and left-hand IC engine main-shaft rotations, respectively; TMR, TML—
ECM AC–DC/DC–AC commutator flywheel/damper generator/motor electromagnetic torques for right- and
left-hand IC engine main-shaft rotations, respectively; ωFR, ωFL—NMR rotary-ratchet composite harmonic
oscillators’ forward angular velocities for right- and left-hand IC engine main-shaft rotations, respectively;
ωBR, ωBL—NMR rotary-ratchet composite harmonic oscillators’ backward angular velocities for right- and
left-hand IC engine output shaft rotations, respectively; ωCR, ωCL—NMR rotary-ratchets and clutch
composite mechanical common drum angular velocities for right- and left-hand IC engine output shaft
rotations, respectively; and ωSR, ωSL—NMR rotary-ratchets and clutch composite common drum angular
velocities for right- and left-hand IC engine output shaft rotations, respectively.

18-13
 Mechatronics

Figure 18.9. Layout of the Fijalkowski engine’s NMR FCM LM–RM/RM–LM and RM–RM commutation
waveforms for its NMR FCM LM–RM/RM–LM and RM–RM commutator: TFR, TFL—NMR rotary-
ratchet’s composite mechanical harmonic oscillators’ forward torques for right- and left-hand IC engine output
shaft rotations, respectively; TBR, TBL—NMR rotary ratchet’s composite mechanical harmonic oscillators’
backward torques for right- and left-hand IC engine output-shaft rotations, respectively; TMR, TML—ECM
AC–DC/DC–AC commutator IPM flywheel/damper generator/motor electromagnetic torques for right- and
left-hand IC engine output-shaft rotations, respectively; ωFR, ωFL—NMR rotary-ratchet composite mechan-
ical harmonic oscillators’ forward angular velocities for right- and left-hand IC engine output-shaft rotations,
respectively; ωBR ωBL—NMR rotary-ratchet composite mechanical harmonic oscillators’ backward angular
velocities for right- and left-hand IC engine output-shaft rotations, respectively; ωCR, ωCL—NMR rotary-
ratchets and clutch composite mechanical common drum angular velocities for right- and left-hand IC engine
output-shaft rotations, respectively; and ωSR, ωSL—NMR rotary-ratchets and clutch composite mechanical
common drum angular velocities for right- and left-hand IC engine output-shaft rotations, respectively.

added to the right-hand NMR rotary-ratchet’s fully controllable mechanical valve

to latch the right-hand rotary-ratchet’s composite mechanical common drum to the
backward RM. The NMR rotary-ratchet’s magnetic-field-exciter electric-current-
pulse sequence is in the opposite phase and the transmission in the drag mode
supplies reverse torque to the central NMR rotary-clutch’s composite hub.

18.2.2 Engine output-shaft angular velocity control

The inner NMR rotary-ratchet’s composite mechanical common drum contains the
central NMR rotary-ratchet’s composite mechanical common hub and is separated
from its NMR layer by means of the diamagnetic-material layer. Transmission of

18-14
 Mechatronics

torque between the inner NMR rotary-ratchet’s composite mechanical common

drum and a separate clutch magnetic-field-exciter’s electric-current-pulse driver
achieves the central NMR rotary-ratchet’s composite mechanical common hub.
During acceleration, one of two methods for controlling the torque transmission
may be taken into account.
Either the NMR rotary-clutch’s magnetic-field-exciter’s applied voltage may be
progressively raised to solidify the NMR, or the NMR rotary-clutch’s magnetic-
field-exciter’s electric-current-pulse control techniques may be used to vary the
electric-current-pulse width and/or the frequency of the force impulses between the
inner NMR rotary-ratchet’s composite mechanical common drum and the central
NMR rotary-ratchet’s composite mechanical common hub by means of the RM–
RM NMR mechatronical commutator acting as a bipolar four-quadrant RM–RM
NMR mechatronical chopper, with the pulse-width-modulation (PWM), and/or the
frequency modulation (FM).

18.2.3 Engine output-shaft deceleration and reverse

The action for slowing the Fijalkowski engine comprises sequential operations.
Firstly, by switching off the applied control voltage, the NMR layer between the
inner NMR rotary-ratchet’s composite mechanical common drum and the output
central NMR rotary-clutch’s composite common hub (see figures 17.4 and 17.5) is
deactivated.
Secondly, the electric-current-pulse sequence will trip shortly to give reverse RM
to the inner NMR rotary-ratchet’s composite mechanical common drum.
Thirdly, the NMR rotary-ratchet’s composite mechanical common drum will be
progressively reconnected to the output central NMR rotary-clutch’s composite
mechanical common hub by one of the methods mentioned above to transmit
reverse torque to decelerate the Fijalkowski engine. This introduces another unique
feature of the advanced NMR mechatronical commutator, twin-opposed-piston-
type reciprocating ICE. If the deceleration torque continues after the reciprocating
IC engine has come to a halt, the Fijalkowski engine may begin to move backwards.
This can facilitate parking and HEV manoeuvring. It should be pointed out here
that the conventional AWB BBW dispulsion mechatronical control system might
have to be retained, at least for emergency stopping.

18.3 Fijalkowski engine cooling

The performance and efficiency of conventional ICEs and ECEs, as well as MMD
electrical machines (ME/EM dynamotors and EE transformers), is limited by
thermal constraints in the cooling channels and windings. These constraints can
be reduced somewhat by circulating liquid through liquid cooling channels or air
through the air gaps in the ICEs and ECEs and the MMD electrical machines.
However, such approaches necessitate the addition of pumping mechanisms and are
limited by the thermal conductivity of air or liquid.
A novel approach has been taken to cooling the Fijalkowski engine, using
ferrofluids as the cooling media. The aim is to take advantage of the thermal and

18-15
 Mechatronics

magnetic characteristics of ferrofluids, such that not only will the ferrofluid circulate
through the Fijalkowski engine, but the existing thermal and magnetic fields in the
Fijalkowski engine will provide the energy required for pumping the ferrofluid. The
cooling mechanism will therefore not require any additional moving mechanical
parts, pumps, compressors or sensors.
The behaviour of the ferrofluid will provide the pumping and cooling method for
the Fijalkowski engine (thus increasing the Fijalkowski engine’s potential energy
density and efficiency). The necessary challenge has been the synthesis of ferrofluid
magnetic nanoparticles that possess the required ferrofluidic magnetic and thermal
properties. There ought to be particular focus on the chemical and biological
synthesis of ferrofluid magnetic nanoparticles that exhibit the necessary properties.

18.4 The Fijalkowski engine advantages versus conventional internal

combustion engines
As compared to conventional reciprocating ICEs, the Fijalkowski engine has the
following advantages.
• It does not have a C2M (there is no need to utilise the ICE’s oily fluid
lubricant), crankcase, cooling jacket or coolant; the demands made on the
metal or ceramics that the block of cylinders are made of are thereby lowered.
• It does not have a radiator, an MP fan, an MF pump, a cylinder head, an
overhead camshaft or related auxiliary systems.
• The torque of the efficient force in it may be increased many times over
without increasing the piston’s power stroke; for example, if a conventional
reciprocating ICE has 1.5 revs during idle strokes in the working range, there
are no idle strokes in it at all; the torque may be increased sharply in it because
of the jet of the exhausts.
• It will experience almost none of the vibrations that present conventional
reciprocating ICEs with a C2M have—which eventually destroy the drive-
shaft and weaken the joints in HEVs.
• Without the C2M, it may be almost perfectly balanced, reducing the bearing
loads to a minimum.
• Without the side loading common to conventional reciprocating ICEs,
mechanical efficiency may be improved by 5–8%, as demonstrated by
estimates, according to its inventor.
• It may show a threefold increase in power-to-mass over diesel engines, and it
may be run in reverse as an MP compressor or even an MF pump.

What is wrong with the highly developed conventional C2M? Nothing at all is
wrong with the crankshaft. The problem with the mechanism lies in the statement
that the piston connecting rods flip from side to side with each cycle, in effect
shortening their length in the process.
If one imagines that the crank is extremely smooth, the ‘big-end’ bearing may be
described as a sinusoidal motion, in view of the fact that the ‘flipping’ has altered it.
To illustrate this effect, the Fijalkowski engine’s sinusoidal harmonic piston motion

18-16
 Mechatronics

Figure 18.10. The Fijalkowski engine’s sinusoidal harmonic piston motion compared with the conventional
reciprocating ICE’s crankshaft-driven piston motion.

is compared with the conventional reciprocating ICE’s crankshaft-driven piston

motion in figure 18.10.
Are the conventional C2M mechanisms really so defective? Not if drivers do not
necessitate higher values of the ICE’s output crankshaft angular velocity, and they
may take steps to prevent continued dangerous resonance.
Increased values of the IC engine’s crankshaft angular velocity mean increased
ICE power output—without an increase in size and mass. For example, the average
crank reciprocating ICE is approaching its potential at 12 000 rpm, and the average
crankless reciprocating ICE with the NMR FCM LM–RM/RM–LM commutator,
that is, the Fijalkowski engine, may reach 20 000 rpm.
The NMR ratchet mechanism is efficient, it is smooth and it is easy to engineer. It
may reduce the number of significant moving parts, it does not necessitate high-
pressure oily fluid lubrication, it is less expensive, it requires less skill and it takes
precise sinusoidal harmonic motion as the C2M (see figure 18.10), so it does not
generate spurious harmonics, and the piston movement is more efficient, a plus that
may rise with the crankless reciprocating IC engine’s output-shaft angular velocity
to become very significant at 20 000 rpm1.
Controllable NMRF stems are interesting because of their ability to provide
simple, quiet, rapid-response (within few milliseconds) interfaces between mecha-
tronical controls and mechanical systems. The ability of a controllable NMRF to be
utilised directly as an NMR ratchet’s fast-acting, fully controlled bipolar mechanical

1
NB Although 20 000 rpm seems very fast by modern reciprocating IC engine standards, the Fijalkowski
engine can be expected to reach 40 000 rpm with further development.

18-17
 Mechatronics

valves (with no moving parts) in an NMR FCM LM–RM/RM–LM commutator is

one of the principal motivating factors for the development of such NMRFs.
Fijalkowski engines in the 20 000–40 000-rpm range may be used for stationary
applications such as distributed generation (DG), combined heat and power (CHP)
and shaft power, as well as other stationary power plant applications.
Let us look at the negatives. It looks funny, seems to be the most common
reaction, but the author may skip that one. It does necessitate oily-fluid slides for
lateral stabilisation. This is because a piston must have a little ‘play’, that is to say, it
must not scrape the piston walls, and because the rings that may be employed to
effect this gap without loss of compression have to allow for thermal energy (heat)
expansion, a tiny amount of slap may occur at mid-stroke, just when the piston rods
alter their sense of motion direction. This may be dampened out adequately by the
slides (which may be fed constantly with oily fluid), except when the dreaded
resonance occurs. However, because the assembly may have a very simple and
fundamental resonant frequency, and because the whole design of the crankless
reciprocating ICE, MF pump, MP compressor, or FM or PM motor may recognise
the necessity of avoiding this problem, the NMR ratchet mechanism may be
equipped with resonance dampers—the universally accepted best method of dealing
with unavoidable resonance problems.
A quick look at the sketches (see figures 18.1 and 18.4) may reveal all. What the
author might have thought of as a crankless reciprocating ICE’s combustion
chamber with an air inlet and exhaust gas outlets, a fuel injector and a couple of
spark plugs (with an excess of air), may also be a tuneable damper. There may be a
theoretical resonance, hazard, but by tweaking the liquid fuel input and the exhaust
port trimming and duration, resonance amplification may be avoided.
The conventional C2M is a relatively simple mechanical device to build and, with
its soft bearings and high-pressure NMR fluid cooling system, it is made to work
well enough for most applications. On the other hand, the advanced NMR FCM
LM–RM/RM–LM commutator has to be precision-engineered, with significant
attention given to stress relief and cooling.

18.5 Conclusions
A generational change in C2M is taking place throughout the world. The author has
performed R&D work on an advanced NMR FCM LM–RM/RM–LM and RM–
RM commutator, reciprocating IC engine, that is, a two- or four- or even five-stroke
thermodynamic cycle, twin-opposed piston, crankless, reciprocating IC engine (the
‘Fijalkowski engine’), for over 20 years (Fijalkowski 1986). In most conventional
reciprocating IC engines, the reciprocating motion of the pistons drives a rotating
crankshaft. The notable exception to this general configuration, the Fijalkowski
engine, may utilise an NMR FCM LM–RM/RM–LM commutator instead of the
mechano-mechanical commutation matrixer (MCM) LM–RM/RM–LM commuta-
tors of the C2M. Compared with earlier MCM LM–RM/RM–LM commutators,
NMR FCM LM–RM/RM–LM commutators are more practice-oriented, especially
from the viewpoint of operating convenience and the space requirements of ultra-

18-18
 Mechatronics

modern and innovative NMR mechatronical techniques. NMR FCM LM–RM/

RM–LM and/or LM-LM or RM–RM commutators adapt the characteristics of the
mechanical energy to the operational requirements.
The objective of this chapter is to obtain feedback through exciting interest from
HPS designers who are working in the automotive, aerospace, marine and other
industries.
The Fijalkowski engine may be much more practical than conventional recip-
rocating IC engines and are likely to be manufactured by the aerospace, automotive,
marine and other industries in the not-too-distant future. Such is the hope of the
author, who is working towards this goal by utilising several innovative mecha-
tronical techniques that incorporate artificial intelligence.

References
De Falco F 1989 Materiale rotabile e trazione ATTI Relazioni Generali, 1-mo Convegno
Internazionale 'LA FERROVIE NEI TRASPORTI DEGLI ANII 2000 (Bologna 12–14
April) pp 39–50
Fijalkowski B 1986 Future hybrid electromechanical very advanced propulsion systems for civilian
wheeled and tracked all-terrain vehicles with extremely high mobility Proc. 1986 EVS-8: The
Eight International Electric Vehicle Symposium (Washington DC 20–23 October) pp 428–43
Fijalkowski B 1988 Unconventional internal combustion engines for automotive vehicles Motor
Vehicles and Motors ‘88, 3–5 October 1988 (also in Motorna Vozila - Motori Saopstenja
(XIV-83) - Kragujevac, November 1988) pp 265–74
Fijalkowski B 1989 The novel means of very advanced propulsion for 2000s thermo-electric high-
speed trains. Proc. Int. Congress The role of railways in the year 2000 (Bologna 12–14 April
1989) pp 425–31
Fijalkowski B T 1997 Intelligent automotive systems: development in full-time chassis motion
spheres for intelligent vehicles Advanced Vehicles and Infra-structure Systems—Computer
Applications, Control and Automation ed C O Nwagboso (New York: Wiley) pp 125–42
Fijalkowski B T 1999a A novel concept of the air independent propulsion and/or dispulsion for
attack submarines Proc. of the 4th UEES’99: Int. Conf. on Unconventional Electromechanical
and Electrical Systems (St Petersburg 21–24 June 1999) vol 2 pp 357–62
Fijalkowski B T 1999b A novel automotive MRF or ERF reciprocating prime mover Proc.
Barcelona 1999 European Automotive Congress—Vehicle Systems Technology for the Next
Century (Barcelona 30 June–2 July 1999) Paper No. STA99P410
Fijalkowski B T 2006a An advanced reciprocating internal combustion engine without crank-
shaft and connecting rod mechanism Proc. of the ASME Internal Combustion Engine Division
2006 Fall Technical Conference – ICEF2006 (Sacramento, CA 5–8 November) ICE2006-1519
Fijalkowski B T 2011 Automotive Mechatronics: Operational and Practical Issues Volume II
(International Series on Intelligent Control, and Automotive Science and Engineering vol 52)
(Berlin: Springer)
Folonari C V 2006 Nanotechnology for automotive suspensions Presentation to Centro Richerche
FIAT (Turin) 1–19
Novak R 2005 ‘Stiff’ fluid could soon put the brakes on cars New Scientist 11 October 2005 p 23
Pinto-Silva J 1987 Non-conventional land transportation systems J. Terramech. 24 153–57
Wen W, Huang W, Yang S, Lu K and Sheng P 2003 The giant electro-rheological effect in
suspension of nanoparticles Nat. Mater. 2 727–30

18-19
 IOP Publishing

Mechatronics
Dynamical systems approach and theory of holors
B T Fijalkowski

Chapter 19
Conclusion and future trends

19.1 Concluding remarks

The title of this book requires some explanation.
• ‘Mechatronics’ is the integration of physics—mechanics, fluidics, electrics and
electronics, etc—and computer technologies into the research and develop-
ment of physical heterogeneous continuous complex and/or simple dynamical
hypersystems (Fijalkowski 2010, 2011). Thus, mechatronics consists of the
synergistic combination of different physical disciplines like mechanics,
fluidics, electrics and electronics, as well as informatics, etc. Synergy is a
very creative and therefore dynamic process, far more so than the usual
co-operation of the different physical disciplines, and even more so than a
close integration of machine hardware and software. Mechatronics offers new
solutions and unprecedented flexibility in transportation systems, industrial
production processes, and aerospace, automotive and traction components,
etc. Its economic success is based on functional integration, i.e. the multiple
use of mechano-mechanical (MM), fluido-mechanical (FM) and/or electro-
mechanical (EM) actuators (machines); the decentralisation of intelligence
into physical heterogeneous continuous dynamical hypersystems, e.g.
aerospace, automotive and/or traction single-shaft propulsion mechatronical
control systems; and the inherent options for sensorless self-monitoring and
system protection.
• The ‘dynamical systems approach’ is a holistic description of observable
physical reality. The whole of the dynamical systems approach is not composed
of the sum of its elementary units—the multiplicity of elementary units is a
result of the diversification of the whole.
• The idea of a holor (from the Greek δλοσ , as in the English holistic, pertaining
to an entity or a whole) seems to have been originated by Wallis (1673) to

doi:10.1088/978-0-7503-1350-6ch19 19-1 ª IOP Publishing Ltd 2016

 Mechatronics

describe a mathematical entity that is made up of one or more merates (from

́ , part of an entity) or coordinates (independent quantities) of
the Greek μεροσ
the holor. It includes complex numbers, scalars, vectors, matrices, tensors,
quaternions and other hypernumbers. The works of Grassmann (1844, 1894)
strongly influenced the subsequent development of holors, including the
works of Gibbs (1881) and Heaviside (1893, 1925). Holors, thus defined,
have been known for centuries, but each has been developed more or less
independently, accompanied by separate nomenclature and theory (Moon
and Spencer 1986). The ‘theory of holors’ demonstrates how these compli-
cated subjects can be made simple by using a single notation that applies to all
holors, both tensors and non-tensors. Moon and Spencer (1965, 1986)
consider all possible types of holors and develop holor algebra and holor
calculus in the most general sense. The theory of holors should establish a
method by which students and teachers can learn vector and tensor analysis
via a uniform treatment (Moon and Spencer 1963, 1965, 1986; Fijalkowski
1967).

The physical heterogeneous continuous dynamical hypersystem as a whole may

be concerned or it may be divided into physical homogeneous continuous dynamical
systems on the grounds of which form of energy is concerned (e.g. kinetic energy:
radiant, thermal, motion, sound and electrical homogeneous continuous dynamical
systems; potential energy: chemical, nuclear, stored mechanical, fluidic, and
electrical and gravitational homogeneous continuous systems). Next, they may be
divided as physical homogeneous continuous dynamical hyposystems and, as
follows, they may be divisible as functional and structural physical homogeneous
continuous dynamical components.
The author’s intention in writing this textbook is to submit ‘generalised physical
commutation matrixer holor analyses’ in the easiest practicable dynamical systems
approach (systems thinking). The author supposes that this can be best realised
through the application of physical commutation matrixers and the functional and
structural physical homogeneous continuous dynamical components.
Nearly all mechanical and electrical engineering students have been educated
concisely regarding physical commutation matrixers during their university physics
courses. In view of this, the physical model represents an exceptional structure with
which to form a junction between static and dynamical analysis.
It is probably particularly important that the generalised physical commutation
matrixer concept lends itself systematically to analogy. Owing to the application of
the physical commutation matrixer method, one may confirm that it is practicable to
consider the area under discussion in mechatronics (kinetic energy—radiant,
thermal, motion, sound and electrical—as well as potential energy—chemical,
nuclear, stored mechanical, fluidic, electrical and gravitational). For instance, the
analogy between charging an electrical capacitor and saturating a fluidic container
of water is made evident by equating the physical and mathematical models of these
physical processes. Naturally, one might have identified this analogy instinctively. In
spite of this, one becomes aware that these physical processes are also comparable to

19-2
 Mechatronics

the variation in velocity of a mass when a force is relevant. This is not practically so
evident. As a result, the physical commutation matrixer concept confirmation is
exceptionally convenient in the absence of physical situations in various physical
heterogeneous continuous dynamical hyposystems for a generalised physical
heterogeneous continuous dynamical hyposystem from which a universal type of
solution may be completed.
A new definition of the ‘physical commutation matrixer’, based on matrix-
interconnection physical heterogeneous continuous dynamical hypersystem theory
concepts, is proposed. The definition may be used to evaluate physical commutation
matrixers, which perform multivalent logical functions, and continuously operating
physical commutation matrixers. The serial and parallel connections of two physical
commutation matrixers and simplified formulas are defined, which are valid for high
component physical commutation matrixers.
A comparison is made between results arrived at using the new and conventional
definitions. It is found that the results differ little in the case of equal probability
output letters. For instance, a physical commutation matrixer is a general term
referring to a physical heterogeneous continuous dynamical hypersystem or part of a
physical heterogeneous continuous dynamical hypersystem of matrixer conductive
parts and their matrix interconnections through which an energy-transfer holor is
intended to flow.
A physical commutation matrixer; i.e. a configuration of physically (i.e. electri-
cally, magnetically, optically, or radiationally) connected dynamical components or
analogue and/or digital devices was made up of active and passive physical,
functional and structural dynamical components or an assemblage of physical,
functional and structural dynamical components and their matrix interconnected
row and column conductive collectors.
Physical commutation matrixer theory includes the study of all aspects of
matrixers, including analysis, design and application. In it, the fundamental
quantities are the ‘energy-potential-difference holors’ (i.e. generalised force holors)
between various points, the ‘energy-transfer holors’ (i.e. generalised velocity holors)
flowing in a number of row and column conductive collectors, and the parameters,
which describe the passive physical, functional and structural dynamical components.
Other important physical commutation matrixer quantities, such as power, energy
and time constants, may be computed from the fundamental physical variables.
This textbook provides a comprehensive and careful development of the physical
commutation matrixer base and the use of holor theory formulations of analytical
methods. No attempt is made to provide the required mathematical tools: matrix
theory, the functions of a complex variable, Laplace transforms, numerical methods
and programming. However, some references in these areas have been included in
the bibliography.
Holor analysis has been shown beyond doubt to be of prime importance in
physics and engineering. Every current scientist and engineer must be systematically
well acquainted with the symbolism and methods of manipulation. Because both
the nomenclature and routine manipulation are quite uncomplicated and may be
learned in a short time, the question emerges as to whether an independent one-term

19-3
 Mechatronics

subject on holors is adequate for a programme of scientific study. If the subject was
taught on the experiential level, it might be integrated into one of the necessary
physics subjects, such as electrics and electronics, magnetics, mechanics, fluidics,
photonics, etc. The treatment presented in this interdisciplinary textbook differs
from the old school one by including invariance in the theory and providing general
definitions that are sensible in all coordinate systems. In this approach, holor
calculus and matrix analysis is endowed with a concrete logical foundation.
Irrespective of the forward-thinking aspects of the textbook, one does not sense
that the treatment is too difficult for ordinary readers.

19.2 Future work

Further investigation of physical heterogeneous dynamical hypersystems, as pre-
sented in this textbook, may build on the promising cyber-physical heterogeneous
dynamical hypersystems. They are integrations of computation, networking and
physical processes. Computers and networks monitor and control the physical
processes, with feedback loops where physical processes affect computations, and
vice versa. The societal and economic potential of such dynamical hypersystems is
vastly greater than has been realised, and major investments are being made
worldwide to develop this technology. The technology builds on the discipline of
embedded dynamical hypersystems, computers and software embedded in devices
whose principle mission is not computation, such as aircraft, automotive vehicles,
medical devices and scientific instruments. Cyber-physical heterogeneous dynamical
hypersystems integrate the dynamics of the physical processes with those of the
software and networking, providing abstractions and modeling, design and analysis
techniques for the integrated whole (Podelski and Bogomolov 2012; Lee and Seshia
2013; Sanfelice 2014).
The combination of networking and information technology with engineered
physical heterogeneous dynamical hypersystems and associated services has enabled
an innovative generation of ‘smart dynamical hypersystems’. These cyber-physical
heterogeneous dynamical hypersystems combine distributed sensing, monitoring,
actuation and control networks. They have inter-operable dynamical systems,
hyposystems and component integration, very advanced analytics, and user inter-
faces featuring customised degrees of independence, which can facilitate adaptive,
predictive and mutual optimisation of cyber-physical heterogeneous dynamical
hypersystem performance over the entire life cycle of a machine or device (e.g.
design, construction, operation/use, maintenance, and service). The existing and
emerging applications are creating exciting opportunities, especially for the
aerospace and automotive industries. Currently, unmanned aircraft systems (UAS)
are being exploited for search and rescue, surveillance, mapping, crop spraying and
environmental monitoring. Moreover, unmanned ground vehicles (UGV) have very
diverse applications, such as security and mine clearance (Cranfield University
2016).
Cyber-physical heterogeneous continuous dynamical hypersystems integrate
analogue and/or digital devices, interfaces, networks, computer systems and the

19-4
 Mechatronics

like with the natural and man-made physical world. The naturally interrelated and
heterogeneous combination of behaviors in these dynamical systems makes their
analysis and design a challenging task. The safety and reliability specifications
imposed in cyber-physical applications, which are normally translated into stringent
robustness standards, aggravate the matter. Unfortunately, the state-of-the-art tools
for system analysis and design cannot cope with the intrinsic complexity of
cyber-physical heterogeneous dynamical hypersystems. Tools that are suitable for
the analysis and design of cyber-physical heterogeneous dynamical hypersystems
must allow for a combination of the physical or continuous dynamics and the cyber
or computational components, as well as being able to handle a variety of modes
of perturbation, such as exogenous disturbances, time delays and dynamical
hypersystem failures (Sanfelice 2014). For instance, cyber-physical heterogeneous
dynamical hypersystems demonstrate the applicability of cyber-mode control
algorithms to physical heterogeneous dynamical hypersystems without significant
design effort being required, and underline their advantage of high robustness. The
remaining chattering effects, which decrease cyber-mode position control accuracy,
may be reduced by means of faster hardware or sophisticated observers and
disturbance estimators. As a result, a challenging research tendency is, for instance,
the investigation of an adaptive disturbance estimator for the flexible shaft system
that does not require information about the structure of a disturbance or the
frequency of periodical perturbations. The performance of the aforementioned
cyber-mode position control of an AC–AC or AC–DC–AC commutator asynchro-
nous (induction) motor may also be improved by implementing better observers for
magnetic flux in addition to angular velocity. Furthermore, an estimator that
provides information about the unknown load may lead to more accurate control
performance.
Cyber-physical heterogeneous dynamical hypersystems can be described as smart
dynamical hypersystems that encompass computational (i.e. hardware and software)
and physical homogeneous dynamical systems, hyposystems and components,
which are seamlessly integrated and interact closely to sense the changing state of
the real world (Lu et al 2015; Anon 2013). Unlike more conventional embedded
dynamical hypersystems, a cyber-physical heterogeneous dynamical hypersystem
is usually designed to be a network (whole) of interacting dynamical systems,
hyposystems and components, with physical inputs and outputs, rather than a
standalone device (Anon 2008).
Cyber-physical heterogeneous dynamical hypersystems are becoming popular in
many application areas: power networks, the aerospace and automotive industries,
manufacturing, healthcare, critical infrastructure, etc. However, with its popularity
and extensive application in critical and important applications, cyber-physical
heterogeneous dynamical hypersystems are becoming increasingly vulnerable and
more of a target for cyber-physical attacks (Wang et al 2010). Hackers can launch
malicious attacks on power networks and transportation systems (Mills 2009) and be
able to hack into wireless medical devices implanted in the human body (Leavitt
2014). More and more security vulnerabilities are being found in all kinds of
cyber-physical heterogeneous dynamical hypersystems (Fletcher and Liu 2011). The

19-5
 Mechatronics

security of cyber-physical heterogeneous dynamical hypersystems has become a

concern of utmost importance in the research and systems design of cyber-physical
heterogeneous dynamical hypersystems.
Thus, a cyber-physical heterogeneous dynamical hypersystem integrates
computing, communication and storage capabilities with the monitoring and/or
control of entities in the physical world dependably, safely, securely, efficiently and
in real time (Mahapatra 2016).
• Cyber-physical heterogeneous dynamical hypersystems will be an exciting
development area for the coming decades;
• they involve multi-disciplinary research-and-development (R&D) work:
○ they involve high-confidence software;
• cyber-physical heterogeneous dynamical hypersystems have the potential to
change the way people interact with their surroundings;
• future applications for cyber-physical heterogeneous dynamical hypersystems
are only limited by the human imagination;
• affordability and ease of application will drive adoption.

Cyber-physical heterogeneous dynamical systems are expected to play a major

role in the R&D of future physical heterogeneous dynamical systems, with new
capabilities that far exceed today’s levels of autonomy, functionality, usability,
reliability and cyber security. R&D advances can be accelerated by close
collaboration between academic disciplines in computation, communication,
control, and other engineering and computer science disciplines, coupled with grand
challenge applications.
The author’s selected recommendations for R&D in cyber-physical heteroge-
neous dynamical systems are as follows.
• Standardised abstractions and architectures that permit modular design of
cyber-physical heterogeneous dynamical systems are urgently needed.
• Cyber-physical heterogeneous dynamical system applications involve
components that interact through a complex, coupled physical environment.
Reliability and security pose particular challenges in this context—new
frameworks, algorithms and tools are required.
• Future cyber-physical heterogeneous dynamical systems will require hard-
ware and software components that are highly dependable, reconfigurable
and, in many applications, certifiable—and trustworthiness must also extend
to the system level.

In future work, the author will focus on cyber-physical heterogeneous dynamical

hypersystems in many application areas: power networks, the aerospace and
automotive industries, manufacturing, healthcare and critical infrastructure—as
well as electro-mechanical (EM) drives, etc—and measure their performance to
compare against theoretical holor analysis.

19-6
 Mechatronics

References
Anon 2008 University of California at Berkeley, Electrical Engineering and Computer Sciences,
Technical Report No. UCB/EECS-2008, [Link]
[Link]
Anon 2013 Foundations for innovation in cyber physical systems workshop summary report,
[Link]
Anon 2016 Cranfield University Centre for Cyber-physical Systems. 2016, [Link]
[Link]
Fijalkowski B 1967 Zastosowanie rachunku holorowego w teorii pradow przemiennych
(Application of a holor calculus in the theory of alternating currents). Przeglad elektrotech-
niczny (Electrotechnics Review), Rok XLIII, Zeszyt 2, Warszawa 1967 (In Polish)
Fijalkowski B T 2010 Automotive Mechatronics: Operational and Practical Issues—Volume 1
(International Series on Intelligent Control, and Automation Science and Engineering vol 47)
(Berlin: Springer)
Fijalkowski B T 2011 Automotive Mechatronics: Operational and Practical Issues—Volume 2
(International Series on Intelligent Control, and Automotive Science and Engineering vol 52)
(Berlin: Springer)
Fletcher K K and Liu X F 2011 Security requirements analysis, specification, prioritization, and
policy development in cyber-physical systems 5th Int. Conf. on Secure Soft-ware Integration &
Reliability Improvements Companion (SSIFI-C) pp 106–13
Gibbs J W 1881 Elements of Vector Analysis (privately printed, New Haven, pp. 17–50, 1881;
pp. 50–90, 1884) 1906 The scientific papers of J Willard Gibbs (London: Longmans, Green
and Co) vol. II p 17
Grassmann H 1844 Die linealoe Ausdehnungslehre, ein neuer Zweig der Mathematik (Otto
Wigand, Leipzig, 1844), Die Ausdehnungslehre (Enslin, Berlin, 1862)
Grassmann H 1894 Grassmann’s gesammelte math U. phys. Werke, 3 vols (Leipzig: B G
Teubner)
Heaviside O 1893 Electromagnetic Theory (London: Electrician Printing and Pub. Co., London),
3 vols; (New York: Dover). Heaviside’s treatment is given in vol I, Ch III, pp 132–305: ‘The
elements of vectorial algebra and analysis’
Heavside O 1925 Electrical Papers (Boston, MA: The Copley publishers)
Leavitt N 2014 Researchers Fight to keep implanted medical devices safe from hackers Computer
43 11–4
Lee E A and Seshia S A 2013 Introduction to Embedded Systems—A Cyber-Physical Systems
Approach [Link]
Lu T , Zhao J, Zhao L, Li Y and Zhang X 2015 Towards a framework for assuring cyber physical
system security Int. J. Security Appl. 9 25–40
Mahapatra R N 2016 Cyber-physical systems: issues and challenges (presenation adapted from
NSF workshops) Third Edition of Int. Conf. on Wireless Networks & Embedded Systems—
WECON 2011 sls 1–50
Mills E 2009 Hackers broke into FAA air traffic contro system The Wall Street Journal A6
Moon P H and Spencer D E 1963 A new mathematical representation of alternating currents
Tensor p 110
Moon P H and Spencer D E 1965 Vectors (Princeton, NJ: van Nostrand)
Moon P H and Spencer D E 1986 Theory of Holors. A Generalization of Tensors (Cambridge:
Cambridge University Press) (Digitally printed, 2005)

19-7
 Mechatronics

Podelski A and Bogomolov S 2012 Lectures notes. [Link]

532012/cps-hm
Sanfelice R G 2014 Introduction to Cyber-physical Systems (CMPE 142) Fall 2014 (Santa Cruz,
CA: University of California)
Wallis J 1685 Letter to Collins, 6 May 1673 Treatise of algebra (London, 1685) p 264; Cajori F
1912 Historical notes on the graphic representation of imaginaries before the time of Wessel
Am. Math. Mon. 19 167
Wang E K, Ye Y, Xu X, Yiu S M, Hui I C K and Chou K P 2010 Security Issues and challenges
for cyber physical systems 2010 IEEE/ACM Conf. on Green Computers and Communications &
2010 IEEE/ACM Conf. on Cyber Physical and Social Computing pp 733–8

19-8
 IOP Publishing

Mechatronics
Dynamical systems approach and theory of holors
B T Fijalkowski

Glossary

Actuator—The component of an open-loop or closed-loop mechatronical control

system that connects the electronic control unit (ECU) with the process; the
actuator consists of a mechano-electrical (MM) commutation matrixer (MCM)
or an electrical commutation matrixer (ECM) AC–AC, AC–DC–AC or DC–AC
commutator motor and a final-control unit; positioning electrical signals are
converted to mechanical output.
All-electrical vehicle (AEV)—Any ground automotive vehicle whose original source
of energy is electrical power, such as an electrical automobile or electrical
locomotive.
Angular velocity holor is defined as the rate of change of the angular displacement
holor and is a vector quantity (more precisely, a pseudovector) that specifies the
angular speed (rotational speed) of an object and the axis about which the object
is rotating. The SI unit of angular velocity is radians per second (rad-s−1).
Angular velocity holor is usually represented by the symbol omega (Ωj). The sense
of direction of the angular velocity holor is perpendicular to the plane of rotation,
in a sense of direction that is usually specified by the ‘right-hand rule’.
Artificial intelligence (AI) is the synthetic intelligence exhibited by machines or
software, and the domain of computer science that develops machines and soft-
ware with human-like intelligence.
Application specific integrated circuit (ASIC)—An integrated circuit (IC) designed
for a custom requirement, frequently a gate array or single-chip microprocessor
or programmable logic device.
Application specific integrated matrixer (ASIM)—An integrated matrixer (IM)
designed for a custom requirement, frequently a gate array or single-chip static-
commutator or electrical commutation matrixer (ECM) AC–AC, AC–DC–AC,
AC–DC/DC–AC or DC–DC commutator.

doi:10.1088/978-0-7503-1350-6ch20 20-1 ª IOP Publishing Ltd 2016

 Mechatronics

Cartesian dualism—Developed by René Descartes, this states that there are two
basic forms of substance: mental and material. According to Descartes’s philo-
sophy, the mental does not have extension in space, and the material cannot
think. Also called substance dualism, it is credited with giving rise to much
thought concerning the well-known mind-body dilemma.
Cartesian reductionism—The idea of reductionism has existed since the ancient
Greeks, while René Descartes, the 17th-century French philosopher, was the first
to formally state the concept. He stated that the world is like a machine composed
of many smaller parts, and that it can be understood by taking apart and studying
the parts before learning how they all fit into the whole.
Chemo-electrical/electro-chemical storage battery (ChE/ECh)Self-contained ChE/
ECh cell/cells or an accumulator system that converts chemical energy into
electrical energy in a reversible process.
Complexity is the property of a real-world dynamical hypersystem that is perceptible
in the inability of any one formalism to denote all its properties. It necessitates
that one finds different means of interacting with physical heterogeneous con-
tinuous dynamical hypersystems. Different in the sense that when one formulates
adequate physical and mathematical models, the formal physical heterogeneous
continuous dynamical hypersystems required to describe each particular aspect
are not derivable from each other.
Computer-aided systems engineering (CASE)—A tool designed to help a systems
analyst complete development tasks.
Connecting rod—Any straight link that transmits motion or power from one linkage
to another within a mechanism, especially linear to rotary (LM–RM), as in a
reciprocating internal combustion engine (ICE) or mechano-pneumatic (MP)
compressor.
Constraint—A restriction on the natural degrees of freedom (DoF) of a physical
heterogeneous continuous dynamical hypersystem, the number of constraints is
the difference between the number of natural DoF and the number of actual DoF.
Cost function—Can refer to: in mathematical optimisation, the loss function, a
function to be minimised; in artificial neural networks, the function to return a
number representing how well the neural network performed to map training
examples to correct output.
Crankshaft—A shaft about which a crank rotates.
Current holor—A current holor, also known as an electrical energy-transfer holor, is
a flow of electrical charge. In electrical commutation matrixers (ECM) this charge
is often carried by moving electrons in wires, i.e. ECM rows and columns. It can
also be carried by ions in an electrolyte, or by both ions and electrons, as in a
plasma. The SI unit for measuring a current is the ampere (A), which is the flow of
electrical charges through a surface at the rate of one coulomb per second (C s−1).

20-2
 Mechatronics

Cybernetics is the study of human/machine interaction guided by the principle that

numerous different types of systems can be studied according to the principles of
feedback, control and communication. The field has a quantitative component,
inherited from feedback control and information theory, but it is primarily a
qualitative, analytical tool—one might even say a philosophy of technology.
Cybernetics is characterized by a tendency to universalise the notion of feedback,
seeing it as the underlying principle of the technological world. The study of
systems that are open to energy but closed to information and control systems
that are tight is called cybernetics. Cybernetics takes as its domain the design or
discovery and application of the principles of regulation and communication.
Cybernetics does not treat things, but rather ways of behaving.
Cyber-physical heterogeneous continuous complex and/or simple dynamical
hypersystems have collections of various cyber-physical dynamical systems,
hyposystems and components so connected or related as to perform a particular
function not performable by the cyber-physical dynamical systems, hyposystems
and components alone, and they must encompass computational (i.e. hardware and
software) and physical homogeneous dynamical systems, hyposystems and com-
ponents, seamlessly integrated and interacting closely to sense the changing state of
the real world. These cyber-physical heterogeneous dynamical hypersystems com-
bine distributed sensing, monitoring, actuation and control networks.
d’Alambert principle—The principle that the resultant of the external forces and the
kinetic reaction acting on a body equals zero.
Darcy’s Law—The law that the rate at which a fluid flows through a permeable
substance per unit area is equal to the permeability, which is only a property of
the substance through which the fluid is flowing, times the pressure drop per unit
length of flow, divided by the viscosity of the fluid.
Deterministic mathematical model of the physical heterogeneous continuous
dynamical hypersystem is a model whose description is possible in the form of the
function relationships between the inputs and outputs of the physical hetero-
geneous continuous dynamical hypersystem.
Developmental systems approach is a perspective that has been adopted by increasing
numbers of scientists and engineers since it emerged in the twentieth century.
Developmental systems approach into the methodology of physical models of the
physical heterogeneous continuous dynamical hypersystems, in addition to
the dynamical systems approach has met with indifference not only from some
theorists on the research side of mathematical models, but also some practitioners
making experimental studies. It still omnipotently predominates in the excellence
as well the physical model, as the modelled existing or projected physical het-
erogeneous continuous dynamical hypersystem, the dynamical systems approach
in the methodology of a formulation of mathematical models, creating in
researchers who are occupied in analytical studies the illusion of also steering the
physical models. The developmental systems approach in the methodology of the

20-3
 Mechatronics

excellence of physical models did not bring about the panacea on the routine for
creating the physical model. This requires a considerably stronger impulse for the
creation of the macro-conditions for computerised analytical studies of mathe-
matical models. If the theorists—systems scientists and engineers—and researchers
of mathematical models are interested in their computerised analytical studies as
well as their verification (checking) while taking into account the feedback of an
algorithm for the formulation of a mathematical model, then sooner or later, they
must look towards having a repertoire of methods to improve their physical
models. This is an opportunity to recognise the developmental systems approach
in the methodology for improving physical models—the idea of improving the
physical model, and the comprehensive trouble-shooting processes for dynamical
and static problems, which are of interest to researchers. What, therefore, is this
idea, and what might be the terms of its usefulness? The basic concept in a glossary
of the developmental systems approach in the methodology to create excellent
physical models is ‘change’. Mathematical models continuously improve due to
the effect of the alteration of physical models. Lack of alteration means they are
inadequate and unfit. Changes to physical models are therefore in the world of
mathematical models, and are important functions of the control of their
adequacy.
Dynamical systems approach—Offers a holistic from whole to local approach for
describing physical reality; produces a comprehensive image of physical reality;
produces accurate predictions with straightforward mathematics and clear logic.
Dynamical universe (DU) theory is a holistic description of observable physical
reality. It is a unifying theory converting space–time in variable coordinates into
dynamic space in absolute coordinates. The dynamic universe (DU) theory relies
on an overall zero-energy balance in space and the conservation of the total energy
in interactions in space. The whole in the DU is not composed of the sum of the
elementary units—the multiplicity of elementary units is a result of the diversification
of the whole. Many physical processes in the Universe are highly nonlinear and
coupled in a complex way. A deep understanding of the DU often requires detailed
numerical simulations, coupled with sophisticated data analysis and visualisation.
Electrical commutation matrixer (ECM) AC–DC commutator—This commutator is
an ECM AC–DC rectifier for a rotary or transversal AC–DC commutator
generator, the commutator electronically commutes (switches) the armature
windings so that the resultant induced source AC armature phase-voltages always
act with the same sense of voltage polarisation; this requires a reversal of the
armature winding connection every π rad; the induced source AC armature
phase-voltages are mechanically rectified to the induced source DC armature
voltage via bipolar electrical valves, e.g. diodes, MOS-FETs, IGBTs or MCTs.
Electrical commutation matrixer (ECM) DC–AC commutator—This commutator is
an ECM DC–AC inverter for a rotary or transversal DC–AC commutator motor
or actuator, the commutator electronically commutes (switches) the armature
windings so that the resultant torque or force always acts in the same sense of

20-4
 Mechatronics

rotary or transverse direction, respectively; this requires a reversal of the armature

winding connection every π rad; the DC armature supply is via bipolar electrical
valves, e.g. MOSFETs, IGBTs or MCTs.
Electrical capacitance is the ability of a body to store an electrical charge. Any object
that can be electrically charged exhibits capacitance. A common form of energy
storage device is the parallel-plate capacitor. In a parallel plate capacitor, capa-
citance is directly proportional to the surface area of the conductor plates and
inversely proportional to the separation distance between the plates.
Electrical conductance—The inverse quantity of electrical resistance is electrical
conductance, the ease with which an electric current passes. The SI unit of
electrical conductance is measured in siemens (S) or mhos (℧).
Electrical inductance—In electromagnetism and electronics, inductance is the
property of a conductor by which a change in current flowing through it ‘induces’
(generates) an electromotive force (induced voltage) in both the conductor itself
(self-inductance) and in any nearby conductors (mutual inductance).
Electrical resistance is the repulsion of a current within an electrical homogeneous
continuous electrical system. It explains the relationship between voltage
(the amount of electrical pressure) and the current (flow of electricity), e.g. the
electrical resistance of an electrical conductor is the opposition to the passage of
an electric current through that conductor. Electrical resistance shares some
conceptual parallels with the mechanical notion of friction. The SI unit of elec-
trical resistance is the ohm (Ω).
Electrics is the technology that generally deals with the application of electricity,
electronics and electromagnetism. This branch of physics first became distinct in
the latter half of the 19th century, after the introduction of the electrical telegraph,
the telephone and electrical power distribution and use. Subsequently, broad-
casting and recording media made electronics part of everyday life.
Electro-mechanical (EM)—Pertaining to an electro-mechanical (EM) motor or
actuator or device, EM heterogeneous continuous dynamical hypersystem, or
EM process that is electromagnetically or electrostatically actuated or controlled.
Engine—A machine in which power is applied to do work by the conversion of
various forms of energy into mechanical torque, force and motion.
Enthalpy—The sum of the internal energy of a thermal homogeneous continuous
dynamical system plus the product of the system’s volume multiplied by the
pressure exerted on the system by its surroundings. Also known as heat content,
sensible heat or total heat.
Entropy—Function of the state of a thermal homogeneous continuous dynamical
system whose change in any differential reversible process is equal to the heat
absorbed by the system from its surroundings divided by the absolute temperature
of the system. Also known as thermal charge.

20-5
 Mechatronics

Eulerian coordinates—Any system of coordinates in which the properties of a fluid

are assigned to points in space at each given time, without an attempt to identify
individual fluid parcels from one time to the next; a sequence of synoptic charts is
a Eulerian representation of the data.
Eulerian correlation—The correlation between the properties of a flow at various
points in space at a single instant of time. Also known as synoptical correlation.
Eulerian equation—A mathematical representation of the motions of a fluid in which
the behaviour and the properties of the fluid are described at fixed points in a
coordinate system.
Euler method—A method of studying fluid motion and the mechanics of deformable
bodies in which one considers volume elements at fixed locations in space, across
which material flows, the Euler method is in contrast to the Lagrangian method.
Extended finite element method (XFEM) is a numerical technique based on the
generalized finite element method (GFEM) and the partition of unity method
(PUM). It extends the classical finite element method (FEM) approach by
enriching the solution space for solutions to differential equations with dis-
continuous functions.
External combustion engine (ECE)—An engine that operates by the energy of
external combustion of a fuel.
External energy-potential-difference holor—An energy-potential-difference holor
exerted on a physical heterogeneous continuous dynamical hypersystem or on
some physical homogeneous continuous dynamical systems, hyposystems
and components by an agency outside the physical heterogeneous continuous
dynamical hypersystem.
External work—The work done by a physical heterogeneous continuous dynamical
hypersystem in expanding against energy-potential-difference holors exerted from
outside.
Final-control unit—The second or last stage of an actuator to control mechanical
output.
Finite elements method (FEM) is a numerical technique for finding approximate
solutions to boundary value problems for partial differential equations. It uses the
subdivision of a whole problem domain into simpler parts, called finite elements,
and variational methods from the calculus of variations to solve the problem by
minimising an associated error function.
Flow—The forward continuous movement of a fluid, such as gases, vapours, or
liquids, through closed or open channels or conduits.
Flow-rate—Also known as the rate of flow holor. (i) Time necessitated for a given
quantity of flowable material to flow a measured distance. (ii) Weight or volume
of flowable material flowing per unit time.

20-6
 Mechatronics

Flow resistance—Any factor within a conduit or channel that impedes the flow of
fluid, such as surface roughness or sudden bends, contractions, or expansions.
Also known as viscosity.
Friction flow—Fluid flow in which a significant amount of mechanical energy is
dissipated into heat by the action of viscosity.
Friction torque—The torque that is produced by frictional forces and opposes
rotational motion, such as that associated with journal or sleeve bearings in
machines.
Fluid is a substance (a continuum) that continuously deforms the application of a
shear, no matter how small the shear stress may be, and includes both liquids and
gases. A fluid cannot sustain a shear when at rest. Liquids are nearly incom-
pressible, but gases are highly compressible. Liquids are distinguished from gases
by orders of magnitude differences in their density, absolute viscosity and bulk
modulus.
Fluid density—The mass of a fluid per unit volume.
Fluid dynamics—The science of fluids in motion.
Fluid mechanics—The science concerned with fluids, either at rest or in motion, and
dealing with pressures, velocities and accelerations in the fluid, including fluid
deformation and compression or expansion.
Fluid statics—The determination of the pressure intensities and forces exerted by
fluids at rest.
Fluidic capacitance—The volume of fluid in a channel can change just because of a
change in pressure: this is either due to fluid compressibility or channel elasticity.
Fluidic capacitors are one of two types of fluidic energy storing dynamical
components in fluidic homogeneous continuous dynamical systems. Capacitance
in a fluidic homogeneous continuous dynamical system comes from continuous
fluido-fluidic (FF) accumulators, but also from the fluid itself if it is compliant. It
is essential to consider fluid compliance in pneumatic homogeneous continuous
dynamical systems, but it generally does not play a significant role in the physical
models of hydraulic homogeneous continuous dynamical systems unless there is
significant trapped air causing spongy behaviour. The capacitance of the fluid is
captured by its bulk modulus property. The idea of the capacitance of fluid
trapped in a cylinder can be expanded to estimate the capacitance of a plug of
fluid in a hose or pipe, which in turn can be used in a dynamic model. One
application of a physical model involving fluid compression is to understand the
hydro-mechanical (HM) hammer, which is impact loading caused by sudden
changes in flow, such as when a fluidic valve is commuted (switched) from on to
off. One has already seen that a fluid itself, whether a liquid or gas, exhibits fluid
compliance due to its compressibility. Certain fluidic machines or devices may
also introduce flow compliance into a fluidic homogeneous continuous dynamical
system, even if the fluids are absolutely incompressible.

20-7
 Mechatronics

Fluidic equation of motion—One of the set of fluido-dynamic equations representing

the application of Newton’s second law (NSL) of motion to a fluidic homo-
geneous continuous dynamical system, the total acceleration of an individual fluid
particle is equated to the sum of the forces acting on the particle within the fluid.
Fluidic inertance or inductance—The second type of energy-storing dynamical
component is fluidic inertance or inductance. In mechanical homogeneous con-
tinuous dynamical systems, mass and rotary inertia often dominate dynamical-
system behavior and must be physically and mathematically modelled. In fluidic
homogeneous continuous dynamical systems, the inertia of the fluid is generally
insignificant and it is usually ignored in dynamical system physical models. This is
because in hydraulic homogeneous continuous dynamical systems, the pressures
are so high that inertial forces can be neglected, and in pneumatic homogeneous
continuous dynamical systems, the mass of air is so low that inertial forces can
also be neglected. When analysing the high-frequency behaviour of a fluidic
homogeneous continuous dynamical system, for example, with the sudden on–off
commutation (switching) of fluidic valves that causes transients in fluid flow,
fluidic inertance should be included in the physical model.
Fluidic resistance—The force exerted by a gas or liquid opposing the motion of a
body through it. Like mechanical friction and electrical resistance, this fluidic
dynamical component performs an energy-dissipation function. The dissipation
of fluidic energy into thermal energy (heat) occurs in all fluidic machines (FM
motors, MF pumps and compressors) or devices to some extent. Thus, fluidic
resistors are any dynamical component that resists flow. Another way of looking
at fluidic resistors is: any dynamical component that causes a pressure drop when
fluid flows through the component. Fluidic resistors include fluidic valves, filters,
hoses, pipes and fittings.
Fluidic friction—Conversion of mechanical energy in fluid flow into thermal energy.
Fluidics is the technology of using the flow characteristics of liquid or gas to operate
a fluidic homogeneous continuous dynamical system. One of the newest of the
control technologies, fluidics has in recent years come to compete with mechan-
ical and electrical homogeneous continuous dynamical systems.
Fluido-mechanical (FM)—Pertaining to a fluido-mechanical (FM) motor or
actuator or device, FM heterogeneous continuous dynamical hypersystem, or
FM process that is fluido-dynamically or fluido-statically actuated or controlled.
Force holor—That influence on a mechanical homogeneous continuous dynamical
system that causes it to accelerate, quantitatively it is a mechanical energy-
potential-difference holor, equal to the mechanical homogeneous continuous
dynamical system’s rate of change of momentum holor.
Fuzzy-logic (FL)—Software design based upon a reasoning node rather than a fixed
mathematical algorithm, a fuzzy logic design allows the mechatronic engineer to

20-8
 Mechatronics

participate in the software design because the fuzzy language is linguistic and built
upon easy-to-comprehend fundamentals.
Generalised admittance holor is a measure of how easily a physical heterogeneous
continuous dynamical hypersystem will allow an energy-transfer holor to flow. It
is defined as the holor inverse of generalized impedance holor, that is the holor
reciprocal of the generalised impedance holor. Thus, the generalised admittance
holor is the energy-transfer holor-to-energy-potential-difference holor ratio, and
it conventionally carries SI units of siemens (S), formerly called mhos (℧).
Generalised coordinates—A set of variables used to specify the positional orientation
of a physical heterogeneous continuous dynamical hypersystem, in principle
defined in terms of the Cartesian coordinates of the physical heterogeneous
continuous dynamical hyposystem’s particles and of the time to some convenient
manner, the number of such coordinates equals the number of degrees of freedom
(DoF) of the physical heterogeneous continuous dynamical hyposystem. Also
known as Lagrangian coordinates.
Generalised equation of motion—Equation that specifies the coordinates of particles
as functions of time. Also known as a differential equation of dynamics, or one of
several such equations, from which the coordinates of particles as functions of
time can be obtained if the initial positions and velocities of the particles are
known.
Generalised energy-potential-difference holor—The generalised energy-potential-
difference holor corresponding to a generalised coordinate is the ratio of the
virtual work done in an infinitesimal virtual displacement, which alters those
coordinates and no others.
Generalised energy-transfer holor—The derivative with respect to time of one of the
generalised coordinates of a particle. Also known as the Lagrangian generalised
energy-transfer holor.
Generalised finite element method (GFEM) is a direct extension of the standard finite
element method (SFEM, or FEM), which makes it possible to attain the accurate
solution of physical and engineering problems in complex domains that might be
practically impossible to solve using the FEM.
Generalised impedance holor is the measure of the opposition that a physical het-
erogeneous continuous dynamical hypersystem presents to an energy-transfer
holor when an energy-potential-difference holor is applied. In quantitative terms,
it is the holor ratio of the energy-potential-difference holor to the energy-transfer
holor in a physical heterogeneous continuous dynamical hypersystem. The
impedance holor extends the concept of the resistance holor to physical hetero-
geneous continuous dynamical hypersystems, and possesses both magnitude and
phase, unlike the resistance holor, which only has magnitude. When a physical
heterogeneous discrete dynamical hypersystem is driven with a step function,
there is no distinction between the impedance holor and the resistance holor; the

20-9
 Mechatronics

latter can be thought of as an impedance holor with a zero phase angle. The
magnitude of the impedance holor is the holor ratio of the energy-potential-
difference holor amplitude to the energy-transfer holor amplitude. The phase of
the impedance is the phase shift by which the energy-transfer holor lags the
energy-potential difference holor. In general, the generalised impedance will be a
holor, with the same SI units as the physical resistance holor, for which the SI unit
is the ohm (Ω).
Generalised momentum—If qj ( j = 1, 2, 3,…) are the generalised coordinates of a
classical physical heterogeneous continuous dynamical hypersystem, and L is its
Lagrangian, the momentum conjugate to qj is pj = ∂L/∂qj . Also known as
canonical momentum or conjugate momentum.
Graphene is a remarkable substance with a multitude of astonishing properties that
have repeatedly earned it the title of ‘wonder material’. Graphene is the thinnest
material known to man at one atom thick, and it is also incredibly strong—about
200 times stronger than steel. On top of that, graphene is an excellent conductor
of heat and electricity and has interesting light-absorption abilities. It is truly a
material that could change the world, with unlimited potential for integration in
almost any industry.
Hamilton–Ostrogradsky principle—A variational principle that states that the path
of a conservative physical heterogeneous continuous dynamical hypersystem in
configuration space between two configurations is such that the integral of the
Lagrangian function over time is a minimum or maximum relative to near paths
between the same end points and taking the same time.
Hamilton’s equations of motion—A set of first-order, highly symmetrical
equations
.
describing the motion of a classical dynamical system, namely
qj = ∂H /∂pj . pj = −∂H /∂pj , where qj ( j = 1, 2,…) are the generalised coordi-
nates of the system, pj is the momentum conjugate to qj and H is the
Hamiltonian. Also known as canonical equations of motion.
Hamiltonian function—A function of the generalised coordinates and momenta of a
physical heterogeneous continuous dynamical hypersystem, equal in value to the
sum over the coordinates of the product of the generalised momentum corre-
sponding to the coordinate, and the coordinate’s time derivative, minus the
Lagrangian of the physical heterogeneous continuous dynamical hypersystem; it
is numerically equal to the total energy of the Lagrangian and does not depend on
time explicitly, the equations of motion of the physical heterogeneous continuous
dynamical hypersystem are determined by the functional dependence of the
Hamiltonian on the generalised coordinates and momenta.
Heat—Thermal energy is transmitted due to a temperature difference between the
source from which the energy is coming and a sink towards which the energy is
going, other types of energy in transit are called work.

20-10
 Mechatronics

Heat conduction—The flow of thermal energy through a substance from a higher to

a lower temperature region. Also known as heat diffusion.
Heat convection—The transfer of thermal energy by actual physical movement from
one location to another of a substance to which thermal energy is stored. Also
known as thermal convection.
Heat flow—Heat thought of as thermal energy flowing from one substance to
another, quantitatively, the amount of heat transferred in a unit time. Also known
as heat transmission.
Heat radiation—Also known as thermal radiation, this is electromagnetic radiation
generated by the thermal motion of charged particles in matter. All matter with a
temperature greater than absolute zero emits thermal radiation. When the tem-
perature of the body is greater than absolute zero, interatomic collisions cause the
kinetic energy of the atoms or molecules to change. Thermal radiation is one of
the fundamental mechanisms of heat transfer. Heat transfer through radiation
takes place in the form of electromagnetic waves that are mainly in the infrared
region. Radiation emitted by a body is a consequence of the thermal agitation of
its composing molecules.
Heat transfer describes the exchange of thermal energy between physical hetero-
geneous continuous dynamical hypersystems, depending on the temperature and
pressure, by dissipating heat. Thermal homogeneous discrete dynamical systems
which are not isolated may decrease in entropy. Most objects emit infrared
thermal radiation near room temperature. The fundamental modes of heat
transfer are conduction or diffusion, convection, advection and radiation.
Holism—Asserts that everything exists in relationship, in a context of connection
and meaning—and that any change or event causes realignment, however slight,
throughout the entire pattern. The whole is greater than the sum of its parts and
this means that the whole comprises a pattern of relationships that are not con-
tained by the parts, but ultimately define them. Holism stands in stark opposition
to the method of reductionism, which holds that analysis, dissection and strict
definition are the tools for understanding reality. Holism asserts that phenomena
can never be fully understood in isolation; it asserts that reductionism can only
give us a partial view of anything it dissects. Holism is difficult to pin down
precisely, because by its very nature it embraces paradox, mystery and contra-
diction. The holistic world view has its roots in the new physics, systems thinking
and ecology, holism and perennial philosophy.
Holonomic constraints—An integrable set of differential equations of dynamics
which describe the restrictions on the motion of a physical heterogeneous
continuous dynamical hypersystem, a function relating several variables in the
form f(x1,… xn) = 0, in the optimisation of physical problems.
Holonomic physical heterogeneous continuous dynamical hypersystem—A physical
heterogeneous continuous dynamical hypersystem in which the constraints are

20-11
 Mechatronics

such that the original coordinates can be expressed in terms of independent

coordinates and possibly also time.
Holor—A holor (ho℧lər; Greek öλoς whole) is a mathematical entity that is made up
of one or more independent quantities (’merates’, as they are called in the theory
of holors). Complex numbers, scalars, vectors, matrices, tensors, quaternions and
other hypercomplex numbers are kinds of holor.
Hybrid-electrical vehicle (HEV) an automotive vehicle that uses two or more dis-
tinct power sources. However, one of them ought to be a secondary power source
with the possibility of energy storage. The term most commonly refers to hybrid
electrical vehicles (HEVs), which combine an internal combustion engine (ICE)
or external combustion engine (ECE) and one or more electro-mechanical/
mechano-electrical (EM/ME) dynamotors. However, other mechanisms to cap-
ture and utilise energy are included. A hybridisation of a conventional automotive
vehicle and an electrical vehicle combines an ICE or ECE propulsion system with
an electrical propulsion system. The presence of the electrical power train
is intended to achieve either better specific fuel consumption (SFC) than a
conventional automotive vehicle or better performance.
Internal combustion engine (ICE)—A prime mover in which the fuel is burned
within the engine and the products of combustion serve as thermodynamic fluid,
as with gasoline and diesel engines. An engine that operates by the energy of
internal combustion of a fuel.
Kinetic energy—The energy which a physical heterogeneous continuous dynamical
hypersystem possesses because of the motion in classical physics equal to one-half
of the physical heterogeneous continuous dynamical hypersystem’s mass and/or
capacitance times the square of its energy-transfer variable.
Kirchhoff’s first law (KFL)—This law is also known as Kirchhoff’s point rule,
or Kirchhoff’s junction rule (or nodal rule). The principle of conservation of
electrical charge implies that at any commutation node (junction) in an electrical
commutation matrixer (ECM), the holor sum of the current holors flowing into
that commutation node is equal to the holor sum of the current holors flowing out
of that commutation node, or the holor sum of the current holors in an ECM of
conductors (rows and columns) meeting at a point is zero.
Kirchhoff’s second law (KSL)—This law is also known as Kirchhoff’s loop rule,
Kirchhoff’s mesh rule, or Kirchhoff’s second rule. The principle of conservation
of energy implies that the holor sum of the electrical energy-potential-difference
holors (voltage holors) around any closed electrical commutation matrixer
(ECM) is zero, or, more simply, the holor sum of the electromotive force (emf)
holors in any closed loop is equivalent to the holor sum of the potential-drop
holors in that loop, or the holor sum of the products of the resistances of the ECM
conductors (rows and columns) and the current holors in them in a closed loop is
equal to the total emf holor available in that loop.

20-12
 Mechatronics

Lagrange–Hamilton theory—The formalised study of continuous systems in terms of

field variables where a Lagrangian density function and a Hamiltonian density
function are introduced to produce equations of motion.
Lagrange’s equations—Equation of dynamics of a physical heterogeneous
continuous dynamical hypersystem for which a classical (non-quantum
mechanical) physical description is possible, and which relates the kinetic energy
of the physical heterogeneous continuous dynamical hypersystem to the gen-
eralised coordinates, the generalised energy-potential-difference variables and the
time. Also known as the Lagrangian equations of dynamics.
Lagrangian—The difference between the kinetic energy and the potential energy of a
physical heterogeneous continuous dynamical hypersystem, expressed as a
function of generalised coordinates and energy-transfer variables from which
Lagrange’s equations can be derived. Also known as the kinetic potential
Lagrange function.
Lagrangian density—For a dynamical system of fields or continuous media, a
function of the fields, of their time and space derivatives, and the coordination
and time, whose integral over space is the Lagrangian.
Lagrangian function—The function which measures the difference between the
kinetic and the potential energy of a physical heterogeneous continuous dyna-
mical hypersystem.
Lagrangian method—A method of studying of fluid motion and the mechanics of
deformable bodies in which one considers volume elements which are carried
along with the fluid or body, and across whose boundaries material does not flow,
in contrast to the Euler method.
Machine—A combination of rigid or resistant bodies that have definite motions and
are capable of performing useful work.
Magnetic reluctance is a concept used in the analysis of magnetic homogenous
continuous dynamical systems. It is analogous to electrical resistance in an
electrical homogeneous continuous dynamical system, but rather than dissipating
electrical energy it stores magnetic energy. In likeness to the way an electrical field
causes an electrical current to follow the path of least resistance, a magnetic field
causes magnetic flux to follow the path of least magnetic reluctance. It is a scalar,
extensive quantity, akin to electrical resistance. The units for magnetic reluctance
are inverse henries (H−1). Also known as magnetic resistance.
Magnetic resistance—See magnetic reluctance.
Magnetomotive force—In physics, the magnetomotive force (mmf), is a quantity
appearing in the equation for the magnetic flux in a magnetic homogeneous
continuous dynamical system, sometimes known as Hopkinson’s law: M = R Φ,
where Φ is the magnetic flux and R is the magnetic reluctance of the homogeneous
continuous dynamical system. It can be seen that the mmf plays a role in this

20-13
 Mechatronics

equation that is analogous to that of the voltage V in Ohm’s law: V = R I. The

mmf is analogous to electromotive force (emf), i.e. the difference in electrical
potential, or voltage, between the terminals of a source of electricity, e.g. a
ChE/ECh storage battery from which no current is being drawn, since it is the
cause of magnetic flux in a magnetic homogeneous continuous dynamical system,
i.e. (i) M = N I, where N is the number of turns in the coil and I is the electrical
current through the magnetic homogeneous continuous dynamical system, and
(ii) M = R Φ, where Φ is the magnetic flux and R is the magnetic reluctance, and
(iii) M = H l, where H is the magnetising force (the strength of the magnetising
field) and l is the mean length of a solenoid or the circumference of a toroid.
Mass-flow holor—Also known as the mass-transfer holor and bulk-flow holor, it is
the measure of the movement of material matter. In physics, mass flow occurs in
open fluidic homogeneous continuous dynamical systems and is often measured
as occurring when moving across a certain boundary characterised by its cross-
sectional area and a flow-rate holor. In fluidics it may also be a flow of fluids in a
tube or vessel of a certain diameter. A bulk transfer of particles of matter in a
characterised type of flow is also known as bulk flow.
Mass-velocity holor—The mass flow-rate holor of a fluid divided by the cross-
sectional area of the enclosing chamber or conduit, The SI unit is kilogram per
second and square metres (kg s−1 m2).
Mathematical model of the physical heterogeneous continuous dynamical hypersystem
is adequate to a physical process, functional and structural homogeneous continuous
dynamical systems, hyposystems and components and superimposed constraints.
Mechanical commutation matrixer (MCM) AC–DC commutator—This commu-
tator is a mechano-electrical commutation matrixer (MCM) AC–DC rectifier for
a rotary AC–DC commutator generator, the commutator mechano-electrically
(ME) commutes (switches) the armature windings so that the resultant induced
source AC armature phase-voltages always act with the same sense of voltage
polarisation; this requires a reversal of the armature winding connection every π
rad; the induced source AC armature phase-voltages are mechano-electronically
rectified to the induced source DC armature voltage via bipolar mechano-
electrical valves, i.e. commutator segments that contact the carbon brushes.
Mechanical commutation matrixer (MCM) DC–AC commutator—This commutator
is a mechano-electrical commutation matrixer (MCM) DC–AC inverter for a rotary
DC–AC commutator motor or actuator, the commutator mechano-electrically
commutes (switches) the armature windings so that the resultant torque always acts
in the same sense of rotary direction; this requires a reversal of the armature winding
connection every π rad; the DC armature supply is via bipolar mechano-electrical
valves, i.e. carbon brushes that contact the commutator segments.
Mechanics (Greek μηχανική) is concerned with the behavior of physical bodies when
subjected to forces or displacements, and the subsequent effects of the bodies on
their environment. It is a branch of classical physics that deals with particles that

20-14
 Mechatronics

are either at rest or are moving at velocities significantly less than the speed of
light. It can also be defined as a branch of science that deals with the motion of
and forces on objects.
Mechanism—The part of a machine which contains two or more pieces so arranged
that the motion of one compels the motion of the others.
Mechano-electrical (ME)—Pertaining to a mechano-electrical (ME) generator or
device, ME heterogeneous continuous dynamical hypersystem, or ME process
that is mechanically driven or controlled.
Mechano-fluidic (MF)—Pertaining to a mechano-fluidic (MF) pump or compressor
or device, MF heterogeneous continuous dynamical hypersystem, or MF process
that is fluidically driven or controlled.
Microcomputer unit (MCU)—A semiconductor device that has a central processing
unit (CPU), memory and input/output (I/O) capability on the same chip.
Motion—A continuous change of position of a body.
Mechano-fluidic (MF) pump—A machine that draws a fluid into itself through an
entrance port and forces the fluid out through an exhaust port.
Mechano-pneumatic (MP) compressor—A mechano-pneumatic (MP) machine that
increases the pressure of a gas or vapour by increasing the gas density and
delivering the gas or vapour against the connected pneumatic homogeneous
continuous dynamical system resistance.
Newtonian mechanics—A mechanical homogeneous continuous dynamical system
based upon Newton’s laws of motion in which mass and energy are considered as
separate, conservative, mechanical properties, in contrast to their treatment in
relativistic mechanics.
Newton’s first law—The law that a particle not subjected to external forces remains
at rest or moves with constant velocity in a straight line. Also known as the first
law of motion, or Galileo’s law of inertia.
Newtonian paradigm—Newton gave us three ‘laws of motion’, which were intended
to describe the motion of the planets. It turned out that these laws could be
applied in a seemingly perfectly general way. This broader application has been
the foundation of the modern scientific method and will be referred to here as the
Newtonian paradigm. At the centre of this paradigm is ‘dynamics’. Dynamics is
the way the laws of motion get applied.
Newton’s second law—The law that the acceleration of a particle is directly pro-
portional to the resultant external force acting on the particle and is inversely
proportional to the mass of the particle. Also known as the second law of motion.
Newton’s third law—The law that, if two particles interact, the force exerted by the
first particle on the second particle (called the action force) is equal in magnitude
and opposite in sense of direction to the force exerted by the second particle on

20-15
 Mechatronics

the first particle (called the reaction force). Also known as the law of action and
reaction, or the third law of motion.
Partition of unity method (PUM)—Can be regarded as a generalisation of the
classic finite element methods (FEM). Instead of a mesh, the construction of a
global approximation space and the evaluation of the discretisation use a much
more versatile set of patches, the so-called cover. The method enables the
inclusion of arbitrary local function spaces and is particularly well suited for
adaptive refinements with respect to spatial resolution and reproducing the
quality of the function space.
Physical heterogeneous continuous complex and/or simple dynamical hypersystems—
Have collections of various physical dynamical systems, hyposystems and
components so connected or related as to perform a particular function not
performable by the dynamical systems, hyposystems and components alone.
Physical model of the physical heterogeneous continuous dynamical hypersystem is a
model in which passing physical processes are equivalent to each other when
being described by mathematical relationships.
Pneumatics—Fluid statics and behaviour in closed pneumatic homogeneous
continuous dynamical systems when the fluid is a gas.
Potential energy—The capacity to do the work that a physical heterogeneous
continuous dynamical hypersystem has by virtue of its position or configuration.
Pressure-drop holor—The difference in pressure holor between two points in a fluidic
homogeneous continuous dynamical system, usually caused by frictional resis-
tance to a fluid flowing through a conduit, filter media, or other flow-conducting,
fluidic homogeneous continuous dynamical system.
Pressure holor—A type of stress which is exerted uniformly in all senses of
directions; its measure is the force holor exerted on per unit area.
Prime mover—The component of a power plant that transforms energy from the
thermal or the fluidic form to the mechanical form.
Principle of virtual work—The principle that the total work done by all energy-
potential-difference variables acting on a physical heterogeneous continuous
dynamical hypersystem in static equilibrium is zero for any infinitesimal
generalised coordinate from equilibrium that is consistent with the constraints of
the physical heterogeneous continuous dynamical hypersystem. Also known as
the virtual work principle.
Protocol—The rules governing the exchange of information (data) between
networked elements.
Pulse-width modulation (PWM)—The precise and timely creation of negative and
positive waveform edges to achieve a waveform with a specific frequency and
duty cycle.

20-16
 Mechatronics

Ratchet—A wheel operating with a catch or a pawl so as to rotate in only a single

sense of direction.
Ratchet coupling—A coupling between two shafts that uses a ratchet to allow the
driven shaft to be turned in one sense of direction only and also to permit the
driven shaft to overrun the driving shaft.
Semicustom MCU—A microcontroller unit (MCU) that incorporates normal MCU
elements plus application-specific peripheral devices, such as higher-power port
outputs, special times units, etc; mixed semiconductor technologies, such as high-
density CMOS (HCMOS) and bipolar analogues, are available in a semicustom
MCU, generally the HCMOS is limited to 10 VDC, whereas the bipolar-analogue
is suitable to 60 VDC.
Specific fuel consumption (SFC)—The mass flow rate of fuel required to produce a
unit of power or thrust, e.g. kilograms per kilowatt-hour (kg kWh−1).
Statistical mathematical model of the physical heterogeneous continuous dynamical
hypersystem is a model that is formulated by taking into consideration the
statistical parameters and the probability expansion function.
Statistical mechanics—Statistical mechanics is a branch of physics that uses prob-
ability theory to study the average behaviour of a mechanical homogeneous
continuous dynamical system where the state of the system is uncertain. Statistical
mechanics is a collection of mathematical tools that are used to fill this dis-
connection between the laws of mechanics and the practical experience of
incomplete knowledge.
Systems thinking approach involves shifting one’s attention from the dynamical parts
to the whole, from objects to relationships, from structures to processes, from
hierarchies to networks. It also includes shifts of emphasis from the rational to
the intuitive, from analysis to synthesis, from linear to nonlinear thinking. By
contrast to reductionism, systems thinking is a style of thought and reasoning
that seeks to understand a whole dynamical hypersystem by examining the
physical heterogeneous continuous dynamical hypersystem as a whole instead of
disassembling and studying the dynamical parts, i.e. the physical homogeneous
continuous dynamical systems, hyposystems and components. While some
scientists prefer to use one of the two styles of thinking to the exclusion of the
other, it is more common to use whichever style fits a given situation. Quite
simply, some situations call for systems thinking, while others require a closer
look at the dynamical parts of the physical heterogeneous continuous dynamical
hypersystem and are better suited to reductionism. Summing up, systems thinking
is an approach to integration that is based on the belief that the dynamical parts,
i.e. the physical homogeneous continuous dynamical systems, hyposystems and
components of a physical heterogeneous continuous dynamical hypersystem, will
act differently when isolated from the hypersystem’s environment or other parts of
the hypersystem. Standing in contrast to positivist and reductionist thinking,
systems thinking sets out to view physical heterogeneous continuous dynamical

20-17
 Mechatronics

hypersystems in a holistic manner. Consistent with systems philosophy, systems

thinking concerns an understanding of a physical heterogeneous continuous
dynamical hypersystem by examining the linkages and interactions between the
physical homogeneous continuous dynamical systems, hyposystems and compo-
nents that comprise the whole of the physical heterogeneous continuous dynamical
hypersystem. When one encounters situations that are complex and messy, then
systems thinking can help one to understand the situation systemically. This helps
one to see the full-size image from which one may identify multiple leverage points
that can be addressed to support constructive change. It also helps one to see the
connectivity between physical homogeneous continuous dynamical systems,
hyposystems and components in the situation, so as to support joined-up actions.
Theoretical reductionism states that all theories in a field are part of a larger theory
with a broader scope. In theory, this supports the idea of the existence of a ’grand
unified theory’ of physics that combines quantum physics with other observed
phenomena.
Thermal capacitance—The ratio of the entropy added to a body to the resulting rise
in temperature.
Thermal conductance—The amount of heat transmitted by a material divided by the
difference in temperature of the surfaces of the material.
Thermal conductivity—The heat flow through a surface per unit area per unit time,
divided by the negative of the rate of change of temperature with distance to a
direction perpendicular to the surface. Also known as heat conductivity.
Thermal inductance—The product of temperature difference and time divided by
entropy flow.
Thermal energy potential difference is the difference between the thermodynamic
temperatures of two points.
Thermal resistance—A measure of a body’s ability to prevent heat from flowing
through it, equal to the difference between the temperatures of opposite faces of
the body divided by the rate of heat flow. Also known as heat resistance.
Thermal resistivity—The reciprocal of thermal conductivity.
Thermics—Also known as thermal physics, this is the combined study of thermo-
dynamics, statistical mechanics and kinetic theory. This umbrella-subject is typically
designed for physics students and functions to provide a general introduction to
each of the three core heat-related subjects. Others, however, define thermics loosely
as a summation of thermodynamics and statistical mechanics.
Thermodynamics is a branch of physics concerned with heat and temperature and
their relation to energy and work. It defines the macroscopic variables, such as
internal energy, entropy, and pressure, which partly describe a body of matter or
radiation. It states that the behaviour of these variables is subject to general
constraints that are common to all materials, not the peculiar properties of

20-18
 Mechatronics

particular materials. These general constraints are expressed in the four laws of
thermodynamics.
Torque holor—(i) For a single force holor, the cross product of a holor from some
reference point to the point of application of the force holor with the force holor
itself. Also known as the moment of force holor or rotation moment holor. (ii)
For several force holors, the holor sum of the torque holors (first definition)
associated with each of the force holors.
Transfer-function holor—The mathematical relationship between the output holor of
a mechatronical control system and its output holor for a linear system; it is the
output holor divided by the input holor under conditions of zero initial energy
stored. Also known as the transmittance holor.
Transfer-matrix holor—The generalisation of the concept of a transfer-function
holor to a multivariable mechatronical control system; it is the matrix whose
product with the vector representing the input holors yields the vector repre-
senting the output holors.
Unholonomic physical heterogeneous continuous dynamical hypersystem—A physical
heterogeneous continuous dynamical hypersystem that is subject to restraints of such
a nature that the physical heterogeneous continuous dynamical hypersystem cannot
be described by independent coordinates; examples are generalised commutation
matrixer (GCM) commutators, e.g. in mechanical homogeneous continuous dyna-
mical systems are a rolling hoop, or an ice skate which must point along its path.
Vapour is a gas at a temperature below the critical temperature, so that it can be
liquefied by compression, without lowering the temperature.
Velocity holor—The time rate of change of position of a body; it is a vector quantity
with a sense of direction and magnitude. Also known as the linear velocity holor.
Voltage holor—Also known as the electrical energy-potential-difference holor,
electrical tension holor or electrical pressure holor (denoted Vi) and measured in
SI units of electrical potential: volts (V) or joules per coulomb (J C−1). It is the
electrical energy-potential-difference holor between two electrical commutation
matrixers’ (ECM) separated terminals, i.e. between ECM rows and columns, or
the difference in electrical potential energy of a unit charge transported between
two ECM separated terminals.
Volumetric flow-rate holor—In fluidics-in particular, fluid dynamics and hydro-
metry-the volumetric flow-rate holor (also known as the volume flow-rate holor,
the rate of fluid-flow holor or the volume velocity holor) is the volume of fluid that
passes through a given surface per unit time. The SI unit is cubic meters per
second (m3 s−1). It is usually represented by the symbol Qj. The volumetric flow-
rate holor should not be confused with the volumetric flux holor, as defined
by Darcy’s law and represented by the symbol qj, with SI units of m3/m2 s, that is,
(m s−1). The integration of a flux over an area gives the volumetric flow rate.

20-19