Virtual Machine Tool

State of the art, problems and trends in the simulation of machine tools Prof. Dr.-Ing. Christian Brecher Chair of Machine Tools, WZL RWTH Aachen, Germany

Ideko, 27.10.2011, Spain

WZL/Fraunhofer IPT

Virtualization of the machine tool development Product developement process with virtual prototypes

concept Konzept

engineering

physical Prototyp prototype

testing Erprobung

optimization Optimierung

physical Prototyp prototype

startup

traditional development time development using virtual prototypes

VIRTUAL PROTOTYPING VERIFICATION
VNC VPLC

(t) F(t) F (t)

time saving start of production

concept Konzept

engineering

physical Prototyp prototype

Reduction of startup Optimierung

testing Erprobung

development time
Implementation of

further improvements in the design stage

Also useful for a better

commissioning
WZL/Fraunhofer IPT Seite 2

Virtualization in product development Application scenarios for virtual machine tools

Applications for virtual machine tools

Development of machine tools

Operation of machine tools

dx,u dy,u dz,u Tt

Tt 1 nz

fz ae ap

Concept development

Component design

Process force prediction

Machine construction

Coupled simulation

Optimization of dynamic behavior

Virtual set-up (controllers)

NC Consideration program of energy optimization consumption

Consistent virtual manufacturing system Consistent virtual manufacturing system

WZL/Fraunhofer IPT Images with courtesy of: Siemens, Chiron, Index, WZL Seite 3

Overview of the FE based simulation of machine tools

Mechanical behavior of a machine tool
static dynamic thermal

In-process behavior of a machine tool

electro-mechanical cutting process trends

Input: stiffness

Input: mass damping

Input: conduction convection

Input: control loop parameters drive parameters

Input: cutting force coefficients process kinematic

Input: ...

Output

Fstat

Q(t) Fdyn

(t) F(t) F (t)

WZL/Fraunhofer IPT

Seite 4

Static behavior of machine tools

Machine Tool
Parameter
Stiffness of structural components Stiffness of connectors Reduced kinematic The stiffness of the machine components are sometimes not available or their calculation is not trivial (e.g. spindle bearings) Finite- Element- Modell
57742 shell elements 18912 Solid elements 74528 nodes

Model

Y-guide 3,0% drives 3,7% Z-guide 9,1%

bed & column 2,6%

slide 1,8% X- guide 0,7%

Simulation result
Static deformation and stress caused by external forces Dynamical problems cannot be calculated
WZL/Fraunhofer IPT

crossbeam 10,5% Turning bearing 24,4%

Z-ram 44,0%

Seite 5

Dynamic behavior of machine tools I

Machine Tool
Parameter
Stiffness and mass allocation The calculated results are very sensitive to incorrect stiffness values and the mass allocation. This is especially true for complex mode shapes

Calculated Eigen modes and their frequencies

Model

Simulation results
Eigen modes and Eigen frequencies The amplitude of the oscillation cannot be calculated
WZL/Fraunhofer IPT Z Y X

Frequency: 42.4 Hz

Frequency: 73.4 Hz

Frequency: 102.0 Hz
Seite 6

Dynamic behavior of machine tools II Machine Tool

Parameter
Stiffness and mass allocation as well as material damping, gap damping and component damping
10
-1

f5 = 102 Hz f4 = 73 Hz f1 = 42 Hz

No verified damping coefficients are available

Model

complience GXX [m/N]

Y X

10

-2

Simulation results
Frequency response function Calculation in time domain possible
WZL/Fraunhofer IPT

20

40

60

80 100 120 Frequency [Hz]

140

160

180

200

Seite 7

Investigation of damping effects within a cluster Dresden, Munich, Aachen, DFG Project FOR 1087
Damping in mechanical systems is the

irreversible transformation of mechanical energy in other forms of energy. Loss of energy by fluids Loss of energy by friction Loss of energy in solid bodies F
3 2 1/6 1 2 4

Damping in machine tools

Material Damping Gap Damping

Damping is caused by:

Components Damping
F
3

(investigated at the WZL) of linear axis

of the spindle

x
4 5 6

x
5

Bearings
Roller Guide 1 Rolle 35 Ball Guide 1 90 35 80 RollerKugel55 Guide 2 Rolle 70 Ball Guide 2 Kugel 55
60 50 40 30 20 10 0

Ball bushes

Ball Screw

Guides

Damping of linear guides

100

compliance

compliance

k m x F

k m d x F

Gap Damping Ball Contakt Damping

Vertical

frequency
WZL/Fraunhofer IPT

frequency

Vertikal Vertical

Horizontal Horizontal

Horizontal
Seite 8

Thermal machine tool behavior (steady-state)

Machine Tool
Parameter & Boundary conditions
Thermal loads, prescribed temperatures, convection and thermal conduction

Q FT L

Coefficients & Boundary conditions are often hardly to define

LT

LF

Model

Linear elongation of a truss due to rise in temperature T LT = T L

Linear elongation of a truss due to force F LF = 1 LFT AE

problems of the static calculation

Definition of Thermal-equivalent Force

Simulation result
deformation caused by stationary temperature distribution The distribution of the temperature over the time is not calculated
WZL/Fraunhofer IPT

The thermal-equivalent force FT compensates for the change in length L of a body caused by a change in temperature T. 1 LT = LT T L = AE LFT FT = AET
Seite 9

Thermal machine tool behavior (non-steady-state)

Machine Tool
Parameter & Boundary conditions
Thermal loads, prescribed temperatures, convection, thermal conduction & changes in the stored heat quantity

FE-Model of the Spindle Carrier

Heat transfer coefficients a1 = 6 W m2K W a1 a2 = 24 2 mK W a3 = 10 2 mK Constraints Temperature distribution a2 after 15 min run-up period Iso-lines
37W

173W

Coefficients & Boundary conditions are often hardly to define

Not depicted: Additional heat sources in: drives, bearings & losses due to splashing of the oil

Q(t) Model

a3

problems of the static calculation

Simulation result
deformation caused by in stationary temperature distribution
WZL/Fraunhofer IPT

LN PQ P M H J K G= 30 M= 36 H H O J H= 31 N= 37 I G R I = 32 O= 38 I J= 33 P= 39 K= 34 Q= 40 H L= 35 R= 41 I J I J KK
Environment temperature 21C

= 25 m
Deformed structure Seite 10

Cluster project Dresden, Chemnitz, Aachen: DFG-Transregio 96 Thermo-energetic Design of Machine Tools
Motivation: Conflict between accuracy, productivity and energy input
Productivity Accuracy

Universities participating in the project

Energy Input

Correction
Calculating inverse set points for a controlinternal adjustment of the target or rather the actual value for position control of feed axes based on models and experiments

Compensation
Physically influencing the causes by components integrated into assembly groups for muting or rather reducing thermally caused effects at the tool center point
Structure integrated sensors Controlled heat flow

based on characteristic maps

WZL/Fraunhofer IPT

based on property models

based on structure models

Construction material

Motor drive optimization

Seite 11

Interaction of the structure and the control

integrated simulation
Finite Element Analyses (FEM) (FEM) Finite Elemente Analyse
Finite Elemente Analyse (FEM) mred k red
Antrieb reduziert

Finite Element Analyses (FEM)

PT2

Digitalle B lo c k s im u la tio n (DBS) Block Simulation (D B S ) D ig ita

s -

cred

Structure reduced Mechanik reduziert

mechanische Antriebselemente reduziert

kinematic elements
k ges

control loops reduced

FEM

DBS

G e k o p p e lte M e h rk rp e rs im u la tio n (M K S )
MKS

G e k o p p e lte fle x ib le M e h rk rp e rs im u la tio n

MKS

Verfahrsimulation m. flex. MKS-Simulation

MKS

Weg, Geschwindigkeit

Weg, Geschwindigkeit

Weg, Geschwindigkeit

Interaction InterS c h n itts te lle action

Kraft

S c h n itts te lle

x, x

FA

G01 X=212.2; Y=122.6; Z=152.12;

TCP path
Interaction
Kraft

Interaction InterS c h n itts te lle action

x, x

S c h n itts te lle

FA

Interaction InterS c h n itts te lle action

Kraft

[M]; [K];[C]

Coupled Multi Body Simulation (MBS)

WZL/Fraunhofer IPT

x, x

S c h n itts te lle

FA

G01 X=212.2; Y=122.6; Z=152.12;

TCP path

Schnittstelle

Schnittstelle

Interaction
-

DBS

DBS

DBS

Flexible MBS

Moving flexible MBS

co simulation
Seite 12

Differently detailed levels of mechanic control and process modeling

Substitution of machine tool behavior
Measurement + analytics Simulation + analytics Simulation + FEA cutting sim.

msubst ksubst dsubst F (t)

msubst ksubst dsubst F (t)

msubst ksubst dsubst

FEM + analytics

MBS + analytics

FEA + FEA cutting simulation

(t) F (t) F (t)

(t) F (t) F (t)

(t) F (t)

Coupled simulation
WZL/Fraunhofer IPT Seite 13

Problems in modeling Process-Machine-Interactions

machine behavior
dynamic deflections of the TCP
amplitude
z

unstable
y

n x

stable
spindle speed

dynamic process forces

F (t)

Uncertainties
FEM modeling of the exact dynamic Cutting force model Cutting force coefficients Process kinematic Tool geometry Work piece behavior

Uncertainties
Fy

machine tool behavior (damping)

Behavior of the control loops Different dynamic machine tool behavior

Fz

in the working room

Influence of model reduction methods

Fx

WZL/Fraunhofer IPT

Seite 14

process behavior

phase

Cutting depth

Extended simulation of spindle-bearing-systems

integrated simulation
stiffness coefficients

bearing calculation
direct and crosscompliance boundary conditions n, F, preload,
Fv d(Fv) d d(Fa) axial bearing displacement

Extended simulation of spindlebearing systems considering nonlinear bearing characteristics and varying bearing stiffness to provide FRF at operation conditions

axial force Fa

Fa

stiffness

dynamic system behavior

tool

300 200 100 0 0 10000 axial

radial

20000 30000 spindle speed

additional masses bearing stiffness and damping

spindle

Fexc.

Calculation of non-linear

bearing stiffness

Simulation of static and dynamic system behavior Consideration of bearing system characteristics

Consideration of load,

revolution speed and thermal effects

(preload, axial elastic adjustment, etc.)

WZL/Fraunhofer IPT

Seite 15

Co-Simulation of spindle bearing systems and HSC processes

MBS-Simulation k(n)
n

xges yges zges

Tt 1 n z

Tt

dx,u dy,u dz,u dx,u dy,u dz,u

Cutting Edge 1
Complex Force Model

dF,Fx dF,Fy dF,Fz

Tt

Tt

Tt

Cutting Edge 2
Complex Force Model

dF,Fx dF,Fy dF,Fz

G [m/N]

Tt

n]

n[ 1/m i

Tt

f [Hz]

1 n z

Tt

Tt +

Nominal Force
n

fz
Fz Fy Fx
Tt 1 n z

dx,u dy,u dz,u

Cutting Edge n
Complex Force Model

dF,Fx dF,Fy dF,Fz

ae ap

Tt

Tt

WZL/Fraunhofer IPT

Seite 16

Contact and tolerance simulation Example of machine tool interfaces

Machine tool interfaces for combined turning and milling operations:

HSK-T

PSC

TS

Simulation requirements: Technical requirements:

Ability for turning and milling Automatic tool changing High stiffness High degree of changing accuracy Static and dynamic simulation Consideration of motor speed and cutting forces Consideration of the manufacturing tolerances Consideration of the contact situation (friction and

lubrication) under process loads and motor speed

Consideration of the clamping system

WZL/Fraunhofer IPT

Seite 17

Tool Interface Current Research

Static behavior Optimization for rotational speed
FFlieh FZug increasing load decreasing load
kB = MB

Revolver-integration

tilting

n HSK-T

MB MA bending moment MB

PSC

TS

Integrated simulation of the tool interface

WZL/Fraunhofer IPT

Seite 18

Software-in-the-Loop simulation shortens set-up times

PLC software development
Extensions to NC cycle and HMI software development possible
Development of VNCK, VPLC, VHMI and

Machine reaction

Software, testing
Status bits

Controller simulation

VMCP completed
Optimization of drive settings requires

Models, drive data,

VNCK VPLC

Virtual drives
Outputs

Inputs

Mechatronic behavior simulation Long-term aim:

Development of controller software on virtual machines
WZL/Fraunhofer IPT

virtual drives: Inputs: PLC and NC software versions NC setup archives Mechatronic machine model (Simple) drive model Requirements: Adoption of original drive data and drive setup archives Coupling capabilities and interfaces to other models within the virtual machine tool

Seite 19

Who can model and control complex multi-physics problems in an efficient way?
The assembly of a complete FEM-Model becomes

more and more complex and expensive

The series of new designed machine tools is very low
Source: Georg Fetscher

(mostly < 1 each year)

In-house FEM calculation The handling of a 1-for-all CAE-Software becomes

more complicated and expensive Should complete FEM-Models of machine tools be build up in-house or should this FEM-calculation be done by an external supplier?
In-house FEM calculation
Easy optimization of single structure

external FEM calculation

Loosing know-how to the supplier Just the FEM calculation of a single

parts and complete machine tools

Integrated product design

problem
No license costs

A must have for high tech machine tool builders

WZL/Fraunhofer IPT

Has to be done by small and midrange machine tool builders

Source: Georg Fetscher

Seite 20

external FEM calculation

Integrated CAx software with a 1-for-all model Needs and wishes of the end users
Integrated CAx software
One software for the complete production

1-for-all model
The end user dislikes to build up different

process of work pieces or machine systems (CAD, CAE, CAM and CAP)
One software for the design and

models for different problems of one machine system

The end user likes to have a single and

optimization of machine systems (structure parts, electrical and hydraulic systems)

complete model for the design and optimization process

The software has to handle all the

different problems

How to build up a target-oriented and efficient CAx software system which can be used in different branches?
WZL/Fraunhofer IPT Seite 21

Virtualization in product development Closing the simulation chain for production

Optimization NC program
NURBS Set-points x = ... (t) a = ... (t) y = ... (t) b = ... (t) z = ... (t)

No

Yes

Identification of critical manufacturing steps

Data transfer: TCP pos., chipping cross section, current workpiece geometry Statistic and dynamic process behavior
Process model Coupled MBS model
Control

OK?
Evaluation of the process
Stability (chattering) Tolerances m Surface quality

Workpiece model (adaptive)

Hybrid process design

Empiric F(t) Model
f

F(t)

Temperature
max

Thermal-model machine

Workpiece model (adaptive)

Thermal process behavior

FEM model d(t)

th dtotal= dm+ dth

WZL/Fraunhofer IPT
min

Optimal NC program
F(t) xd

CAM

Path optimization

NC simulation (VNCK)

Seite 22

CAx framework within the CoE at RWTH Aachen

Objectives
h A

Integrative connected simulation of control,

machine and process Dimensioning of complex milling and grinding processes Optimising concept for multiaxial NC-path planning

Current work
Integration in CAM-system NX

Further development of milling and grinding simulations (models, algorithms) Structure-dynamic machine modelling Analysis of control performance

WZL/Fraunhofer IPT

Seite 23

Summary and outlook

The FEM based simulation of machine tools is a complex

task, which needs experienced FEM end users and high quality input data.
Especially the simulation of the dynamic and thermal machine

tool behavior is still complex today.

In future the process-machine-interactions will become a key

figure in the design of high quality machine tools.

The market demand for a single CAx system will lead to

solutions for an integrated software framework for design, optimization and process planning of work pieces.

WZL/Fraunhofer IPT

Seite 24

Congratulations to your 25th birthday!!!

DYNXPERTS (2010-2013) Chameleon (2008-2011) Next (2005-2009)

WZL/Fraunhofer IPT Seite 25