Photonic Solutions

RSoft CAD
v2024.09

User Guide

Synopsys, Inc.
 Copyright Notice and Proprietary Information
Copyright © 2025 Synopsys, Inc. All rights reserved. This software and documentation contain confidential and proprietary
information that is the property of Synopsys, Inc. The software and documentation are furnished under a license agreement and
may be used or copied only in accordance with the terms of the license agreement. No part of the software and documentation may
be reproduced, transmitted, or translated, in any form or by any means, electronic, mechanical, manual, optical, or otherwise,
without prior written permission of Synopsys, Inc., or as expressly provided by the license agreement.

Right to Copy Documentation

The license agreement with Synopsys permits licensee to make copies of the documentation for its internal use only. Each copy
shall include all copyrights, trademarks, service marks, and proprietary rights notices, if any. Licensee must assign sequential
numbers to all copies. These copies shall contain the following legend on the cover page:
“This document is duplicated with the permission of Synopsys, Inc., for the exclusive use of
______________________________ and its employees. This is copy number __________.”

Destination Control Statement

All technical data contained in this publication is subject to the export control laws of the United States of America. Disclosure to
nationals of other countries contrary to United States law is prohibited. It is the reader’s responsibility to determine the applicable
regulations and to comply with them.

Disclaimer

SYNOPSYS, INC., AND ITS LICENSORS MAKE NO WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, WITH REGARD TO
THIS MATERIAL, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
FOR A PARTICULAR PURPOSE.

Trademarks

Synopsys’ company and certain product names are trademarks of Synopsys, as set forth at:
[Link] All other product or company names may be trademarks of their
respective owners.

FOSS Notices

Free and Open Source Software (FOSS) notices are located in the \readme subfolder of the RSoft installation folder.
Contents

Preface 1
Foreword .................................................................................................. 1
How to Read This Manual ........................................................................ 3

1 Introduction 7
1.A. Program Installation & System Requirements ................................... 7
1.B. Physical Conventions ........................................................................ 7
1.C. Program Executables ........................................................................ 8
1.D. Example Files ................................................................................... 9
1.E. README Files .................................................................................. 9
1.F. Product Support, Upgrades, and Resources ................................... 10

2 A Simple Tutorial 11
Step 1: Opening the CAD Window......................................................... 11
Step 2: Creating a New Design File ....................................................... 12
Step 3: Defining Variables ..................................................................... 14
Step 4: Drawing a Component ............................................................... 15
Step 5: Moving a Component ................................................................ 16
Step 6: Set Component Properties Exactly ............................................ 17
Step 7: Creating a Tapered Component ................................................ 18
Step 8: Verifying Structure Layout ......................................................... 20
Step 9: Simulation & Advanced CAD Features ...................................... 20

3 Basic CAD Usage 22

3.A. The CAD Window............................................................................ 22
3.A.1. Menubar ........................................................................... 22
3.A.2. Toolbars ........................................................................... 25

RSoft CAD  iii

 3.A.3. Status Bar......................................................................... 30
3.A.4. Tree View Toolbar ............................................................ 30
3.A.5. RSoft Tips......................................................................... 31
3.B. Creating a New Design File ............................................................. 31
3.C. The Symbol Table & Variables ........................................................ 34
3.C.1. Why use the Symbol Table & Variables? .......................... 35
3.C.2. Using the Symbol Table ................................................... 35
3.D. Drawing Components ..................................................................... 37
3.D.1. Drawing Modes ................................................................ 37
3.D.2. Placing Components in a Design File ............................... 38
3.D.3. Drawing Preferences ........................................................ 39
3.D.4. Try It! ................................................................................ 39
3.E. Viewing Components ...................................................................... 39
3.E.1. Single-Pane Mode ............................................................ 40
3.E.2. Multi-Pane Mode (3D Only) .............................................. 40
3.E.3. Zooming ........................................................................... 41
3.E.4. Panning ............................................................................ 41
3.E.5. Display Aspect Ratio:........................................................ 41
3.E.6. Displaying ‘Simulation’ Objects ......................................... 42
3.F. Editing Components ........................................................................ 42
3.F.1. Selecting Components ...................................................... 42
3.F.2. Modifying Components ..................................................... 43
3.G. Calculating Material Profiles ........................................................... 45
3.G.1. Computing a Material Profile ............................................ 45
3.G.2. Saving a Material Profile .................................................. 49
3.G.3. Producing Index Profile Plots from the Command Line ..... 49
3.H. Viewing the Current Value of a Parameter ...................................... 50
3.I. Easily Edit Long Expressions ........................................................... 50
3.J. Editable Drop-Down Lists ................................................................ 50

4 Segment Components 52
4.A. Segment Overview .......................................................................... 52
4.A.1. Basic Segment Properties ................................................ 52
4.A.2. Segment Orientation Overview ......................................... 53
4.A.3. Setting Segment Properties .............................................. 54
4.A.4. Setting Segment Display Color ......................................... 54
4.B. Vertices ........................................................................................... 55
4.B.1. Setting Position................................................................. 55
4.B.2. Index Definition ................................................................. 56
4.B.3. Controlling Geometry ........................................................ 57
4.C. Basic 2D Segments ........................................................................ 58
4.D. 3D Structure Types ......................................................................... 58
4.D.1. Fiber ................................................................................. 59

RSoft CAD  iv
 4.D.2. Channel ............................................................................ 59
4.D.3. Diffused ............................................................................ 60
4.D.4. Rib/Ridge ......................................................................... 62
4.D.5. Multilayer .......................................................................... 64
4.D.6. Polygon Structure Types .................................................. 67
4.D.7. Mixing 3D Structure Types ............................................... 69
4.E. Profiles & Tapers............................................................................. 69
4.F. Segment Types (z-Axis, Seg-Axis, Extended) ................................. 69
4.G. Drawing Common Extended Segments .......................................... 71

5 Other Component Types 74

5.A. General Overview ........................................................................... 74
5.B. Lenses ............................................................................................ 74
5.C. Circles............................................................................................. 77
5.D. Polygons ......................................................................................... 79
5.E. Registration Marks .......................................................................... 81
5.F. Circuit References, Simulation Regions, Vertices, and Monitors ..... 82

6 Advanced CAD Features 83

6.A. Profiles ............................................................................................ 83
6.A.1. Built-in Profile Types ......................................................... 83
6.A.2. Types of User Profiles ...................................................... 85
6.A.3. Creating & Using User Profiles ......................................... 87
6.A.4. Example User Profile ........................................................ 89
6.B. Tapers............................................................................................. 90
6.B.1. Component Options that Support Tapering ....................... 91
6.B.2. Built-In Taper Types ......................................................... 91
6.B.3. Built-In Arc Taper Types ................................................... 92
6.B.4. Creating & Using User Tapers .......................................... 93
6.B.5. An Example User Taper ................................................... 95
6.C. Priority (When Components Overlap).............................................. 96
6.C.1. Combine Mode ................................................................. 96
6.C.2. Priority .............................................................................. 96
6.C.3. General Notes .................................................................. 97
6.D. Global Index Generation Options .................................................... 97
6.D.1. Simulating Curved Segments to Large Angles.................. 98
6.D.2. Sidewall Angle Control ..................................................... 99
6.D.3. Simulating the Effects of Finite Lithographic Resolution ... 99
6.D.4. Simulating the Effects of Lithographic Roughness ............ 99
6.D.5. Interpreting User Profiles Absolutely ............................... 101
6.E. Geometry Import/Export ................................................................ 102
6.E.1. Export ............................................................................. 102

RSoft CAD  v
 6.E.2. Import ............................................................................. 102
6.E.3. Import/Export Settings and Important Notes ................... 103
6.F. Circuit References ......................................................................... 104
6.F.1. Creating a Circuit Reference ........................................... 105
6.F.2. Setting Circuit Reference Properties ............................... 106
6.F.3. Editing a Circuit Reference ............................................. 108
6.F.4. Controlling Variables in Child File From Parent File ........ 108
6.F.5. Hierarchy Exports ........................................................... 109
6.F.6. Creating Python Functions .............................................. 111
6.F.7. Adding Vertices to a Circuit Reference............................ 113
6.F.8. Dynamically Importing GDS files via Circuit References . 113
6.G. Dynamically-Sized Arrays ............................................................. 114
6.G.1. Creating a Dynamic Array .............................................. 115
6.G.2. Setting Lattice Properties ............................................... 115
6.H. Defining Pathways ........................................................................ 116
6.H.1. Creating a Pathway ........................................................ 116
6.H.2. Using Pathways .............................................................. 117
6.I. Importing Symbols or Components from Another Design File ......... 118
6.I.1. Importing Symbols ........................................................... 118
6.I.2. Importing Components ..................................................... 120
6.J. CAD Preferences ........................................................................... 120
6.K. Additional Component Properties .................................................. 122
6.K.1. General Properties ......................................................... 123
6.K.2. Extended Segment Properties ........................................ 124
6.K.3. Crystal Axes ................................................................... 125
6.K.4. Component Tags ............................................................ 126
6.L. Using EIM to Simulate 3D Structures in 2D (2.5D)......................... 128
6.M. Expanded Referencing of Vertices ............................................... 129
6.N. Warning Options ........................................................................... 130
6.O. Component Filters ........................................................................ 130
6.O.1. Defining Filters ............................................................... 130
6.O.2. Using Filters ................................................................... 133
6.P. Simulation Log Files ...................................................................... 133

7 Layout Generation Utilities 135

7.A. Motivation...................................................................................... 135
7.B. Lattice Parameters ........................................................................ 136
7.A.1. 1D Lattice Types ............................................................. 137
7.B.2. 2D Lattice (XZ and XY) Types ........................................ 137
7.B.3. 3D Lattice Types ............................................................. 138
7.C. Unit Cell Parameters ..................................................................... 139
7.C.1. 1D Unit Cells .................................................................. 139
7.C.2. 2D Unit Cells (XZ and XY) .............................................. 140

RSoft CAD  vi
 7.C.3. 3D Unit Cells .................................................................. 141
7.C.4. Additional Options .......................................................... 142
7.D. Creation Options ........................................................................... 143

8 Advanced Materials 144

8.A. Using the Material Editor ............................................................... 144
8.A.1. The Material Tree ........................................................... 145
8.A.2. The Material Tabs ........................................................... 146
8.A.3. Creating and Organizing Materials in the Material Editor 146
8.A.4. Using Materials in a Project (Local Design File) .............. 147
8.A.5. Displaying Material Definition .......................................... 148
8.A.6. Material Properties.......................................................... 148
8.A.7. Tips & Traps ................................................................... 149
8.B. Linear Epsilon Properties .............................................................. 151
8.B.1. Linear Epsilon Overview ................................................. 151
8.B.2. Simple Linear Epsilon ..................................................... 152
8.B.3. Anisotropic Linear Epsilon .............................................. 154
8.B.4. Dispersive Linear Epsilon ............................................... 155
8.C. Non-Linear Epsilon Properties ...................................................... 159
8.D. Linear Mu Properties ..................................................................... 162
8.E. Non-Linear Mu Properties ............................................................. 163
8.F. Electro-Optic Properties ................................................................ 164
8.G. Thermo-Optic Properties............................................................... 165
8.H. Stress-Optic Properties ................................................................. 166
8.I. Semiconductor Properties .............................................................. 167
8.K. Special Materials ........................................................................... 171

9 Non-Uniform Grids 173

9.A. Grid Basics.................................................................................... 173
9.B. Advanced Grid Controls ................................................................ 174
9.C. Overriding Grid Settings for Specific Components or Materials ..... 177
9.D. Viewing the Grid............................................................................ 178

10 Scripting & Scanning 179

10.A. Layout Scripts ............................................................................. 179
10.A.1. Using Scripts within the RSoft CAD .............................. 179
10.A.2. Generating Design Files Completely via Scripts ........... 191
10.B. Parameter Scanning and Optimization ........................................ 195
10.C. Data Manipulation ....................................................................... 195
10.D. Simulation Scripts ....................................................................... 196
10.D.1. Running the Simulation Programs from the Command
Line........................................................................................... 196

RSoft CAD  vii

 10.D.2. Creating a Simulation Script ......................................... 198
10.D.3. Using Schedulers ......................................................... 199

11 Transverse Mode Solving 200

11.A. BPM-Based Mode Solvers .......................................................... 200
11.A.1 Basic Theory of BPM-Based Mode Solving ................... 201
11.A.2. Using the BPM Mode Solvers ....................................... 202
11.B. TMM-Based Mode Solver (TmmSIM) .......................................... 204
11.B.1. Using TmmSIM ............................................................. 204
11.B.2. Output Options ............................................................. 205
11.C. FEM-based Mode Solver (FemSIM) ............................................ 205

Appendix A Tips and Traps 206

A.A. Common RSoft CAD mistakes ...................................................... 206
A.B. Some Good RSoft CAD habits to learn ......................................... 207

Appendix B File Formats 209

B.A. Standard RSoft File Format .......................................................... 209
B.A.1. Syntax Parameters ......................................................... 209
B.A.2. File Syntax ..................................................................... 211
B.A.3. Example Files................................................................. 213
B.B. TDR File Format ........................................................................... 217
B.B.1. Using TDR Files as Inputs to the RSoft Software ........... 217
B.B.2. Enabling TDR File Output from the RSoft Software ........ 218
B.C. User-Profile Functions .................................................................. 218
B.D. User-Taper Functions ................................................................... 219
B.E. Polygon Files ................................................................................ 219

Appendix C RSoft Expressions 220

C.A. Arithmetic Operators & Order of Operations ................................. 220
C.B. Built-in Constants ......................................................................... 221
C.C. Using File Names in Variables/Expressions ................................. 221
C.D. Defining Default Value for Variable when used in Expression....... 221
C.E. Square Bracket Notation for Hierarchy Export Symbols ................ 222
C.F. Built-In Functions .......................................................................... 222
C.F.1. Basic Functions .............................................................. 222
C.F.2. Comparison Functions ................................................... 223
C.F.3. Logical Functions: .......................................................... 224
C.F.4. Step Functions ............................................................... 225
C.F.5. Random Functions ......................................................... 225
C.F.6. Data-File Functions ........................................................ 226

RSoft CAD  viii

 C.F.7. Material Functions .......................................................... 229
C.F.8. Effective Index Functions ............................................... 231
C.F.9. Polygon Functions .......................................................... 232
C.F.10. PDK Functions ............................................................. 232
C.F.11. String Functions ........................................................... 233
C.F.12. Unit Conversion Functions ........................................... 233
C.F.12. Sentaurus Workbench (TCAD) Functions ..................... 234

Appendix D Symbol Table Variables 235

D.A. Startup Window/Global Settings ................................................... 235
D.B. Special Effects .............................................................................. 236
D.C. Advanced Grid Parameters .......................................................... 236
D.D. Advanced Features ...................................................................... 237
D.E. Display Material Profile ................................................................. 238
D.F. CAD Preferences: ......................................................................... 239

Appendix E Utilities 240

bdconv – Matrix Manipulation ............................................................... 242
bdutil – Overlap Integrals, Far Fields, Spot Sizes, and LightTools Ray
Files ..................................................................................................... 246
Syntax....................................................................................... 246
Overlap Integrals: -i................................................................... 247
Coupling Coefficients: -k ........................................................... 249
Field Size Calculation: -s .......................................................... 250
Far Field Calculations: -f ........................................................... 251
LightTools Interface .................................................................. 254
Simple File Arithmetic and Data Transformations: -o, -sc, -tr .... 255
bmp2ind – Convert Bitmap Image to a User Profile .............................. 256
Syntax:...................................................................................... 256
Usage: ...................................................................................... 256
Indexed-Color Bitmap Background ........................................... 258
Examples .................................................................................. 259
bp2codev and codev2bp – Ray Tracing Converter for CODE V ........... 262
Converting from RSoft to CODE V (bp2codev) ......................... 262
Converting from CODE V to RSoft (codev2bp) ......................... 263
BSDF Utilities (bsdfgen and bsdfutil) .................................................... 265
Custom PDK Utility (smatgen) .............................................................. 266
coatuf – Create Coating for User Profile ............................................... 267
disperse – Calculate Dispersion Relations............................................ 269
dxf2ind – Convert DXF Mask to .ind File............................................... 270
ffutil – Far-Field Utility ........................................................................... 271
fhakit – Fast Harmonic Analysis of Complex Exponential Time Series . 272

RSoft CAD  ix
 findpks – Find Peaks Utility .................................................................. 277
fitdisp – Dispersion Fitting Utility ........................................................... 279
fwmodekit – Extract Data From FullWAVE State File............................ 282
ind2gds, ind2dxf – Convert .ind File to Mask ........................................ 284
gds2ind – Convert GDS Mask to .ind File ............................................. 285
killall – Utility to stop processes ............................................................ 286
Mask Conversion Utilities ..................................................................... 287
GDS Utilities ............................................................................. 287
DXF Utilities .............................................................................. 288
mat2bp – Convert Matrix Data to RSoft Format .................................... 289
mathmat – Do Arithmetic with Data Files .............................................. 290
parutil – Generate nk data files from TCAD Sentaurus parameter files . 293
Ray-Tracing Conversion Utility ............................................................. 294
rsanimate – Create Animations from Plot Files ..................................... 295
rscombine – Combine Frequency Results ............................................ 296
rspython – Python Environment ............................................................ 297
shufflemat – Rearrange Data Files ....................................................... 298
smatgen - Custom PDK Utility .............................................................. 301
tailmon – Extract Final Value Utility ...................................................... 302
tdrutil – TDR File Utility ......................................................................... 303
wgmode – Waveguide Mode Utility....................................................... 307

Appendix F Color Scales 308

F.A. What are Scale Files? ................................................................... 308
F.B. Using Color Scales ....................................................................... 309
F.C. Scale File Format.......................................................................... 310

Appendix G Polarization Conventions 313

G.A. Background .................................................................................. 313
G.B. The RSoft Convention .................................................................. 314
G.C. Simple rules to remember ............................................................ 317
G.D. Complications .............................................................................. 317

Appendix H Python Usage 319

H.A. Python Version Support ................................................................ 319
Note: Numeric Array Support .................................................... 320
H.B. Layout API: RSoftCircuit() class.................................................... 320
H.B.1. RSoftCircuit() class ........................................................ 320
H.B.2. RSoftComponent() class ................................................ 328
H.B.3. RSoftVertex() class ........................................................ 332
H.B.4. Using Hierarchy Exports ................................................ 333
H.B.5. Best Practices ................................................................ 333

RSoft CAD  x
 H.B.6. Example Usage.............................................................. 334
H.C. RSoft Data API: RSoftUserFunction() .......................................... 337
H.C.1. Reference ...................................................................... 337
H.C.2. Example Usage ............................................................. 339

Appendix I RSoft CAD Release Notes 341

Index 348

RSoft CAD  xi
Preface

Foreword
The RSoft CAD is the core program in the RSoft Photonic Device Tools, and acts as a control
program for the passive device simulation modules BeamPROP BPM, FullWAVE FDTD,
BandSOLVE, GratingMOD, DiffractMOD RCWA, FemSIM FEM, and ModePROP EME. It is
used to define the most important input required by these simulation modules: the material
properties and structural geometry of a photonic device. A user will typically first design a
structure in the CAD interface and then use one or more simulation engines to model various
aspects of the device performance.
This modular approach to the design and simulation of photonic devices is one of RSoft’s
Photonic Device Tool’s greatest strengths. Each program in the suite is designed to “play nice”
with the other programs, creating an environment in which data can be shared between the
modules. Virtually all the input and output files are in a simple ASCII text format, which allows
even greater user control over program operation as well as third-party programs to be integrated
into the suite.
While the tools are designed to be used via the GUI (Graphical User Interface), command line
operation is also possible. This, coupled with the modularity of the Suite, allows for complex
scripting capability. The tools are not limited to a single scripting language, but rather uses the
native scripting language of your operating system. For example, Windows users can use DOS
batch files, while Unix users can use bash scripts. Additionally, users familiar with languages
such as Perl, Python, C, or C++ can create custom scripts in these languages. The RSoft

RSoft CAD  1
Component Design Suite provides the best of both worlds: it allows for simulations to be
performed via the GUI, and for complicated custom simulations to be performed via a script.
New and advanced users alike are able to realize the full power of the tools.

Associated Simulation Modules

The RSoft CAD is the main control program for a series of simulation modules which are
licensed separately from RSoft. These simulation modules, along with typical applications, are:

BeamPROP BPM
BeamPROP is a simulation engine for the design of integrated and fiber-optic waveguide devices
and circuits. The software incorporates advanced finite-difference beam propagation (BPM)
techniques for simulation. The software has been commercially available since 1994, and is in
use by researchers and development engineers in both university and industrial environments.

FullWAVE FDTD
FullWAVE is ideal for the design of complex photonic devices. The software employs the finite-
difference time-domain (FDTD) method for simulation, which allows analysis of devices, such
as photonic bandgaps and ring resonators, that cannot be modeled with techniques such as the
efficient beam propagation method (BPM).

BandSOLVE
BandSOLVE is a simulation module for generating and analyzing photonic band structures. This
simulation module is based on an advanced optimized implementation of the plane-wave
expansion technique for periodic structures. It is ideal for producing band structures for classic
photonic bandgap structures such as 2D and 3D photonic crystal waveguides and defect sites. In
addition, it can be applied to fiber structures such as photonic crystal fibers and photonic
bandgap fibers, which are particularly challenging for other simulation techniques. It is
especially useful for optimizing the band structure properties of photonic crystal structures,
which can be subsequently simulated in FullWAVE to examine time-dependent properties such
as waveguide loss and coupling.

GratingMOD
This software tool helps model and analyze many devices that incorporate gratings and various
kinds of filters, and is easily integrated with RSoft’s award-winning simulation tools. It provides
both a forward and backward analysis, allowing for the calculation of grating spectra from
known geometries and the synthesis of grating structures from grating spectra.

RSoft CAD  2
DiffractMOD RCWA
DiffractMOD is a general design tool for diffractive optical structures such as diffractive optical
elements, subwavelength periodic structures, and photonic bandgap crystals. It is based on the
Rigorous Coupled Wave Analysis (RCWA) technique, and can simulate both 2D and 3D
structures with an arbitrary lattice structure and unit cell index profile. In addition to dielectric
materials, dispersive and lossy material structures such as metallic systems can also be used.
Typical applications include Diffractive optical elements (DOEs), photonic bandgap structures,
wavelength filters, optical metrology, nano-lithography, polarization sensitive devices, artificial
dielectric coatings, photovoltaic systems, 3D displays, optical interconnections, optical data
storage, spectroscopy, microlens arrays, and beam splitting, combining, and shaping.

FemSIM FEM
FemSIM uses the finite element method to calculate modes, both waveguide and cavity, of any
2D cross-section. A generally non-uniform and irregular mesh is used to solve for all field
components of supported modes, both guided and leaky. FemSIM can easily find modes for
highly hybrid structures, highly leaky structures, and structures with small feature sizes.

ModePROP EME
ModePROP is an eigenmode propagation tool that accounts for both forward and backward
propagation and radiation modes. It provides a rigorous steady-state solution to Maxwell’s
equations that is based on the highly stable Modal Transmission Line Theory. A full array of
analysis and simulation features makes this tool flexible and easy to use.

How to Read This Manual

While this manual can (and should!) be used as a reference manual at times, it is recommended
that you read the first five chapters in their entirety to get a firm foundation from which to
interpret the rest of the documentation. This is not only important for learning to use the RSoft
CAD, but also for using any associated simulation modules as the correct usage of the CAD is
critical to attaining meaningful simulation results.
If you are a new user, please reject the notion that you can simply turn to a section of interest in
this manual or any simulation manuals for BeamPROP, FullWAVE, BandSOLVE,
GratingMOD, DiffractMOD, FemSIM, or ModePROP. It can be hard to resist this urge, but
doing so will reward you with better usage habits, a fuller understanding of how the software
works, and will help you from suffering from common usage problems.

How is This Manual Organized?

This manual can be logically split into several main parts:

RSoft CAD  3
Introductory Information
Chapters 1 and 2 provide an overview of the installation of the program, an introduction to the
program and its components, information about technical support and product upgrades, and an
introductory tutorial. All new users should read these chapters in their entirety.

Basic CAD Usage

Chapters 3-5 explain the basic usage of the RSoft CAD including an overview of the CAD
window, the symbol table, creating new design files, adding/removing components from a
design, setting component properties such as material and geometry information, and the various
component types. All new users should read these chapters in their entirety.

Advanced CAD Usage

Chapters 6 and higher cover advanced topics, automatic layout utilities, advanced material
properties, non-uniform grids, parameter scanning/scripting, the multi-physics utility, and basic
transverse mode calculations. These chapters can be treated as a reference manual and do not
need to be read by all new users.

Appendices
The appendices provide detailed discussion of file formats, expressions, utilities, and other
relevant information. The appendices can be treated as a reference manual and do not need to be
read by all new users.

This manual should be used alongside the simulation tool manuals as described in the next
section.

Where Is The Documentation For…

The documentation for the RSoft Photonic Device Tools is divided into several manuals. The
manuals are structured using a simple rule:

Anything defining geometry and/or material parameters is in the CAD manual. Anything else is
in an appropriate simulation manual.

Using this rule, almost any topic can be found. As with any rule, there are a few exceptions. The
major exceptions are:
• Installation
The installation procedure is covered in the Photonic Solutions Installation Guide.
• Parameter Scanning/Scripting/Batch Operation

RSoft CAD  4
 These topics are very similar, and are shared by all the simulation modules. They are
discussed in Chapter 10 of the CAD manual.
• Computing the Index Profile
Computing the index profile is discussed in Section 3.G of the CAD manual.
• Pathways
Pathways define the location of pathway monitors (used by BeamPROP and FemSIM) and
the location and geometry of launch (initial) fields, in BeamPROP, FullWAVE, and
ModePROP. They are documented in Section 6.G of the CAD manual.
• Command Line Utilities
The RSoft Photonic Device Tools ship with several command line utilities that perform a
variety of tasks. These utilities are documented in Appendix E of the CAD manual.
• RSoft Expressions
Virtually any numeric field can accept an analytical expression involving pre-defined and
user-defined variables. The form of these expressions, including valid arithmetic operators
and functions can be found in Appendix C of the CAD manual.
• Mode Solving
A variety of mode solvers is included, see Chapter 11 in the CAD manual for an overview of
the mode solving methods available.
Anytime this rule is violated, a note will direct the reader to the proper section in the proper
manual.

Where are these manuals located?

All documentation is placed on your computer during the program installation. Online versions
can be accessed through the RSoft CAD via the Help menu item, or the two help buttons on the
right of the top toolbar. The actual files can be found in the subdirectory help in the installation
directory. Additionally, PDF versions can be found in the subdirectory docs. These files require
the Adobe Acrobat Reader, which can be obtained from Adobe ([Link]) at no charge.

Typographical Conventions
A number of typeface and layout conventions are followed in this manual.
• The names of fields and controls in the GUI dialogs are written in boldface
• The values of pull-down menus and radio button controls are written in italics.
• File names and paths, symbol table variables and values, expressions typed in GUI edit
fields, and code snippets are written in monospace.

RSoft CAD  5
• In referring to example CAD files, the installation directory for the CAD tool is specified as
<rsoft_dir>, and should be replaced with the correct value for your installation.

RSoft CAD  6
1

Introduction

This chapter explains the installation procedure for the RSoft CAD, a discussion of physics
conventions used in this manual, information on running the program, and notes about getting
the product updates and technical support.

1.A. Program Installation & System

Requirements
The installation process is outlined in the Photonic Solutions Installation Guide. Please check the
'System Requirements' section of the relevant product on Synopsys' Photonic Solutions website
([Link] for a list of
the specific OS versions we currently support.
Note that once a product has been successfully installed, it may not function correctly after
changes such as OS updates or other software/hardware modifications are made to the computer
system. In such cases, we will attempt to resolve any issues for customers that have a current
annual maintenance contract, but we do not guarantee success.

1.B. Physical Conventions

As with any branch of science, there are a number of concepts in the study of photonic devices
for which there exist several different definitions exist in the literature. These are our
conventions:

RSoft CAD  7
Units
The units used in the CAD are as follows:
• The standard unit of length is measured in microns [µm].
• The angular unit used is degrees.
• The units of time (for FullWAVE) are also in microns [µm], and therefore in units of cT,
where c is the speed of light in a vacuum which is approximately 3e14 µm/s.
• The units of imaginary refractive index are defined as:

nimag =
4
where  is the wavelength and  is the usual exponential loss coefficient defined such that the
power decays as e-z and is given in units of m-1.

Polarization
Since the RSoft CAD is shared by many programs, the polarization conventions used can vary by
tool. Please consult Appendix G as well as the individual simulation tool manuals for
information specific to a particular tool.

1.C. Program Executables

The directory <rsoft_dir>\bin\ contains many executable files. The following is a list of the
executables used by the RSoft CAD, associated simulation engines, the WinPLOT graphing tool,
and MOST, RSoft’s scanning and optimization tool, and other utilities.

Product Windows Name Linux Name

The RSoft CAD tool [Link] xbcad

BeamPROP BPM simulation tool [Link] xbeam

BSDF Utilities See Appendix E See Appendix E

FullWAVE FDTD simulation tool [Link] xfullwave

BandSOLVE simulation tool [Link] xbandsolve

Custom PDK Utility [Link] xsmatgen

GratingMOD simulation tool [Link] xgrmod

DiffractMOD RCWA simulation tool [Link] xdfmod

FemSIM FEM simulation tool [Link] xfemsim

RSoft CAD  8
 LaserMOD CAD tool [Link] Xlasermod

LaserMOD simulation tool [Link] xlmsim

LED Utility [Link] xled

ModePROP EME simulation tool [Link] xmodeprop

Solar Cell Utility [Link] xscutil

Tapered Laser Utility [Link] xlmbpm

WinPLOT graphing tool [Link] xplot

MOST tool [Link] xrsmost

Note that each simulation tool is licensed separately, and you will only be able to run those you
have a valid license for.

These executables have different names under Windows and Linux though in most
documentation the Windows names are used. Also, these programs can be run from the
command line; more information on this can be found in Section 10.D.1.

1.D. Example Files

The subdirectory <rsoft_dir>\examples\ contains example device files. These files are sorted
into subdirectories based on the simulation modules. These files have an extension .ind (named
for the refractive index distribution which describes the device). Each examples directory has a
\Tutorial subdirectory that contains the index files described in the Tutorials chapter(s) of that
product’s manual.
To open an example file, open the RSoft CAD, select File/Open from the CAD menu, and
choose the desired index file. To start a simulation, click the Perform Simulation (green light)
icon and press OK in the simulation parameters dialog.

These example and tutorial files are specific to each simulation engine; you can only perform a
simulation if you are licensed for that particular simulation engine.

1.E. README Files

The latest product information can be found in the README files located in the /readme
subdirectory of the installation directory. These files contain important last-minute information

RSoft CAD  9
about RSoft software that is not contained in other documentation, including new or improved
features and options.

1.F. Product Support, Upgrades, and Resources

The software includes technical support, product updates, and other resources. Software sales
and technical support worldwide is handled by Synopsys, except in those countries where we
have local representatives whom you can contact directly.
Visit [Link] for information about how to
access training resources, knowledge articles, product downloads/updates, documentation, how
to open support cases, and other resources. Many of these resources are in Synopsys’
SolvNetPlus portal ([Link] eligible users can find information about
how to create an account at the main support link above.

RSoft CAD  10
2

A Simple Tutorial

This chapter contains a brief tutorial that shows the basic usage of the RSoft CAD. It breaks the
design process into nine steps in order to illustrate a sample workflow. In practice, some of these
steps may be skipped or repeated several times to achieve a working design.
Even though this tutorial creates a simple fiber structure, the ideas and techniques introduced
will be exactly the same for producing complex structures. While not necessary, it is
recommended that you follow the steps in this tutorial at a computer while reading this chapter.

Step 1: Opening the CAD Window

To open the Rsoft CAD Environment:
• Windows: Use the Windows Start Menu.
• Linux: Enter xbcad at a command prompt.
The CAD program appears as in Fig. 2-1. There is a menubar at the top of the window, two
toolbars with icons just below it, another toolbar along the left edge of the window, a tree-view
toolbar along the right edge of the window, and a status line at the bottom of the window. These
CAD menus and icons allow for standard editing operations as well as other common functions.
Their use is completely documented in Section 3.A.

RSoft CAD  11
 Figure 2-1: The RSoft CAD window, showing the menubars at the top and left, the tree-view
toolbar at the right, and the status line at the bottom.

If this is your first time opening the RSoft CAD, a “Welcome to the RSoft CAD Program!” tip
window will appear, as appears in Fig. 2-1. These tip windows are designed to convey relevant
information about the software to both new and experienced users, and will occasionally pop-up
as you use the software.

Step 2: Creating a New Design File

The RSoft CAD provides the key input for each of the RSoft simulation engines: the refractive
index distribution of the problem to be modeled. This is determined by the geometry,
arrangement, and optical properties of components that the user has placed. The RSoft CAD
allows users to specify this information in a straightforward, user-friendly manner.
To create a new design file, click on the New Circuit icon (the leftmost icon on the top toolbar)
or choose File/New from the menu. The startup dialog will appear as shown in Fig. 2-2 where
basic information about the structure and simulation tool to be used.

After creating a new structure, changes can be made to these settings in the Global Settings
dialog that can be accessed via the Globe icon in the left toolbar.

RSoft CAD  12
 Figure 2-2: The startup window where basic information about the structure is set.

Note the following settings:

Option Description
Model Dimension The dimension of the structure.
Component Delta-N The default index difference between a component and the Background
Index
Background Index The index of the background (white) region.
Simulation Tool The simulation tool to be used. Since this tutorial will only look at the
CAD, the tool chosen does not matter.
Load Settings Loads a pre-defined list of settings, for example typical Silicon waveguide
settings.

For this tutorial, all the default settings are acceptable except the Model Dimension. Change the
Model Dimension to 3D and then click OK to continue to create the layout window within the
main CAD window as shown in Fig. 2-3.
The view shown in Fig. 2-3 shows the multi-pane view. There are three panes, an XZ plane, YZ,
XY pane, and 3D pane. Axes labels are shown in each pane. If this were a 2D structure, only the
XZ pane would be shown. To view a single pane, click the 1P button in the view toolbar (the
second toolbar on the top). Click the X, Y, and Z buttons in view toolbar to switch between the
views. To switch back to multi-pane view, click the 4P button in the view toolbar.

RSoft CAD  13
 Figure 2-3: Illustration of a layout window where components are added to the circuit. The
simulation domain (solid or dashed line) and launch field are also shown.

Finally, note that the mouse appears as a cross-hair, and the coordinate display in the status line
indicates the mouse position in real coordinates [µm].

Step 3: Defining Variables

It is usually beneficial to create variables that will represent various aspects of the structure. As
an example, we will create a variable to represent the length of a component. Click on the Edit
Symbols icon in the left toolbar to open the Symbol Table Editor as shown in Fig. 2-4.

RSoft CAD  14
 Figure 2-4: The Symbol Table Editor; Length has been defined with a value of 42.

The symbol table will have several symbols already defined that correspond to the settings made
in the Startup window. We are going to define a symbol Length, and set it equal to 42: click
New Symbol, enter the symbol Name (Length) and Expression (42), and then click Accept
Symbol. The variable Length should now appear in the symbol table. This variable will be used
in a later step. More information on the usage of the Symbol Table can be found in Section 3.B.

Step 4: Drawing a Component

CAD Components (usually referred to as just ‘components’) are the basic building blocks from
which a design is created. A wide variety of components can be used, including segments
(rectangular, cylindrical, multilayer, diffused, etc.), tapered objects, lenses, user-defined
polygons, spheres, cylinders, and completely user-defined objects. Any number and combination
of these components, which have their own local, geometric and material parameters, can be
embedded in a background material of a specific index to create the refractive index distribution
of the entire structure.
This tutorial will create a simple rectangular segment along the Z axis. The type of component to
be added to a design is selected by the Drawing Mode. For this example, the Segment (In Plane)
mode will be used, and is the default selection on the left toolbar. By default, a 3D segment has a
rectangular cross-section. This can be changed to other cross-sections as discussed in Section
4.D.
Move the cross-hair cursor to the desired start of the fiber component (X=0, Z=0). After
positioning the cursor, press and hold the left mouse button and move the cursor to the desired
end of the fiber component as shown in Fig 2-4. As you move the cursor, the center line of the
component is shown in blue and the status line coordinate display is continuously updated with

RSoft CAD  15
the coordinate of the end point. If a mistake is made when adding any component, use the Undo
icon in the top toolbar.

Figure 2-5: a) A fiber component as it appears in the XZ plane after drawing, and b) Multi-
pane view clearly showing the fiber component. Multi-pane view is not available for 2D
structures. Here, the tree view has been hidden by clicking on the Toggle Component
Tree button.

You may notice that the CAD display does not appear as shown in Figure 2-5. This is most likely
due to a different display aspect ratio. There is a pulldown menu on the view toolbar to control
the aspect ratio, set it to 1:1. You can switch between the single pane view shown in Fig 2-5a and
four pane view in Fig 2.5b by clicking the 1P or 4P button in the view toolbar. Use the pulldown
menu on the view toolbar to choose an appropriate aspect ratio for each view.
In general, the next step would be to continue to add elements to the CAD layout to create the
desired device. This process is described further in Section 3.D. For this example, there are no
more elements to add.

In addition to viewing all components in the CAD at once, it is also possible to view only the
components that lie within a specific cut plane by using the “Cut Mode” option. See Section
3.E.1 for details.

Step 5: Moving a Component

There are two methods to move a component within the CAD:
• Via the Mouse

RSoft CAD  16
 Components can be selected with the mouse and then dragged to a new position. This allows
for the quick movement of a waveguide segment.
• Via the Component Properties window
The exact coordinates of a component can be set in the Component Properties dialog box
allowing for precise control of component position. The use of this box will be discussed in
the next step.

Step 6: Set Component Properties Exactly

It is desirable to set component properties exactly in order to have precise control over the
component properties such as refractive index, position, and size. This is done via the
Component Properties dialog shown in Fig. 2-7 which can be opened by right-clicking on a
component in the CAD window, or in the component’s icon in the tree view.

Figure 2-7: The Component Properties dialog for the fiber component.

The RSoft CAD is an object oriented design environment so each component has a different
Component Properties dialog, and therefore a different set of parameters. A detailed description
of this dialog can be found in Chapter 4.

The current value of any field in the software, including those in the Component Properties
dialog shown in Fig. 2-7, can be found by holding the CTRL key and double-clicking on the field.

RSoft CAD  17
Setting Position
The length of the fiber component was set using the mouse in Step 4. It is possible to control this
length exactly in the Component Properties dialog. This will be done using the variable Length
that has been previously defined and the concept of “references”. In order to set the length of this
component, the ending vertex will be offset by the variable Length from the starting vertex
resulting in a fiber with a length of 42 µm. Set the Reference Type of the Z coordinate of the
Ending Vertex to Offset, the Offset Value to Length, and relative to the Starting Vertex of
component 1. Thus, the Z coordinate of the ending vertex of this component, or component 1,
should be computed as an offset from the starting vertex of this component, or component 1.
This offset will therefore equal the length of the segment.

Setting Refractive Index & Other Material Properties

The refractive index of this component can be set in one of two ways:
• Locally Defined:
Set the Index Difference field of the starting vertex (the ending vertex is equal to the starting
vertex by default). This value is defined relative to the Background Index for the structure
set in the Startup window in Step 2. For this example, the Background Index is 1 and the
Index Difference is 0.1, so the real index of the structure will be 1.1. See Section 4.B.2 for
additional details.
• Via a Material
The Material Editor can be used to define materials that can be simple dielectrics, but can
also have material properties such as anisotropy, non-linearity, and dispersive effects. Note
that these material properties are not enabled/appropriate for all simulation tools. More
information about the specification of materials can be found in Chapter 8.

Step 7: Creating a Tapered Component

Some components such as the fiber in this tutorial have two vertices and allow for properties
such as index, width, height, and position to be tapered, or varied, along the length of the
component.
Right click on the fiber component to open its Component Properties dialog. Change the ending
vertex Component Width, which corresponds to the fiber diameter along X, to width*4. The
built-in variable width was set in the Startup window in Step 2 and can be changed via the
Global Settings window. Click OK and the fiber will appear tapered as in Fig. 2-8: one vertex
will have a circular cross-section, the other elliptical.

RSoft CAD  18
 Figure 2-8: The tapered fiber as seen in the CAD window.

The width taper function, or how the width varies between the two vertices, is linear by default.
It can be changed via the Width Taper setting in the Component Properties dialog. See Section
4.E.

The view shown in Fig. 2-8 shows the entire structure; it is also possible to view ‘slices’ of each
cross-section at a particular position. This is most readily illustrated with the XY view. On the
view toolbar, select Z and click the +/- buttons to begin stepping through the structure: Both the
absolute Z position and the step along Z between slices are shown on the view toolbar, and can
be directly set by the user. As the slices step through the structure, the cross-section will change
from circular to elliptical (and/or back). To view the entire structure again, click the All button on
the view toolbar.

RSoft CAD  19
Step 8: Verifying Structure Layout
It is important to simulate the desired index distribution! Performing a simulation with the wrong
index distribution will result in erroneous results. There are two main methods to verify the
structure layout:
• Via the CAD
Various aspects of the geometry can quickly be seen using the multi-pane view and cut
feature described briefly above. See Section 3.E for more details.
• By Displaying the Index Profile
Index profiles show the index distribution and can help verify that both the geometry and
material properties have been set correctly. To compute an index profile, click the Display
Material Profile icon in the left toolbar to open the dialog shown in Fig. 2-9.

Figure 2-9: The Compute Index Profile dialog.

Pressing OK in this dialog will display the index profile at the Z Domain Min value. More
information on the display of material profiles such as the refractive index can be found in
Section 3.G.

Step 9: Simulation & Advanced CAD Features

Once the structure has been created, the next step is to use the selected simulation tool to perform
a simulation. See the simulation tool documentation for more information.

RSoft CAD  20
This tutorial illustrates basic CAD layout features and a sample workflow for using the RSoft
CAD. The rest of this manual explains the simple concepts introduced in this tutorial in more
detail and also introduces advanced concepts.

RSoft CAD  21
3

Basic CAD Usage

This chapter should be read by all users and covers the CAD window, the symbol table, creating
new structures, creating/editing components, viewing the structure, and calculating material
profiles. This chapter does not cover component types or component properties. See Chapter 4
for segment components and Chapter 5 for other component types.

3.A. The CAD Window

This section describes the menu, toolbars, and status bar in the CAD (Fig. 2-1).

3.A.1. Menubar
Some options can be accessed via the toolbars; keyboard shortcuts are shown in parenthesis if
available.

Under File:

New… (Ctrl-N) Close Import Circuit

Create new design Close current Import mask file: Section 6.E
file: Section 3.B design file

Open… (Ctrl-O) Close All Export Circuit

Open existing Close all

RSoft CAD  22
 design file design files Export mask file: Section 6.E

Save (Ctrl-S) Print (Ctrl-P) Exit

Save current Print current Close RSoft CAD
design file design file
Save As… Printer Setup
Save current design Set printer related
file with a different name parameters

Under Edit:

Undo(Ctrl-Z) Duplicate Flip Horizontal

Reverse last Duplicate Flip selected
change selected components components horizontally

Redo (Ctrl-Y) Select All Flip Vertical

Redo last Selects all Flip selected
change components components vertically
Cut (Ctrl-X) Select/Edit Components… Rotate…
Move selected components
Select or edit a component Rotate selected
to the clipboard
based on its number or name components

Copy (Ctrl-C) Edit Default Tags… Convert to Polygons

Copy selected components Edit default component tags: Convert selected components
to the clipboard Section 6.K.4 to polygons

Paste (Ctrl-V) Edit Hierarchy Exports… Flatten Circuit

Paste clipboard contents Edit hierarchy exports: Section Flatten hierarchy
into design file 6.F.5 into one level
Delete (Del) Change Order Tables
Delete selected Change order in which Access various setting dialogs
components components are drawn.

Under View:
These features are discussed in Section 3.E.

In Last ReGrid
Zoom in Revert viewing window Recalculate
CAD window to the last setting the grid

Out Set View Clear Log

RSoft CAD  23
 Zoom out Directly set the Clear log file, see Section 6.P
CAD window viewing window
Full ReDraw Open Log
Set viewing domain so
Redraw all components Open log file, see Section 6.P
the entire structure is visible

Under Options:

Preferences Index Generation Import Symbols

Open CAD preferences Open Global Index Generation Import symbol table from
dialog: Section 3.D.4 Options dialog: Section 6.D another file: Section 6.I

Global Settings Warnings Insert

Open Global Settings Enable/disable Insert components
dialog: Section 3.B warnings: Section 6.M by type

Under Run:

Go Compute Modes Run Script

Open Simulation Parameters Compute transverse Run an RSoft
dialog to start simulation modes: Chapter 11.A script.

Display Material Profile MOST Optimizer/Scanner Cluster Settings

Open Display Material Open MOST optimization Set FullWAVE
Profile dialog: Section 3.G and scanning utility cluster parameters

Multi-Threading Settings GPU Settings

Set Multi-Threading Options: Set GPU Settings: see
See Simulation Manuals Simulation Manuals

Under Graph:

View Single Plot View S-Matrix File Close All Browsers

Open a single plot Close all
Open an RSoft Matrix File
in WinPLOT DataBROWSER windows

Launch DataBROWSER Close All

Open DataBROWSER Close all WinPLOT and
DataBROWSER windows

View BSDF File Close All Plots

Open an RSoft BSDF Close all

RSoft CAD  24
 file in the BSDF Viewer WinPLOT windows

Under Utility

Array Layout Multi-Physics Q-Finder

Create periodic RSoft’s Multi-Physics RSoft’s Q-Finder Utility:
structures: Chapter 7 Utility FullWAVE manual

WDM Router Layout… Solar-Cell Raytracing-Converters

Create WDM Router or RSoft’s Solar Cell Utility: Convert field data to and from
AWG: AWG Utility Manual Solar Cell Utility manual common ray-tracing formats

WDM Router Layout… LED TMM-Mode-Solver

Simulate WDM Router or RSoft’s LED Utility: TmmSIM mode
AWG: AWG Utility Manual LED Utility manual solver: Section 11.B

GratingMOD Grating Layout Tapered-Laser

Create simple grating RSoft’s Tapered-Laser Utility:
structures for GratingMOD Tapered-Laser Utility manual

Under Window:

Tile Arrange Icons Close All

Tile all open design files Arrange minimized windows Close all open file.
in the CAD window within the CAD window

Cascade Close
Cascade all open design Close the current file
files in the CAD window

Under Help
This menu option provides online help for all RSoft products that share the CAD interface.

3.A.2. Toolbars
The CAD has several toolbars:

The Top Toolbar:

New Circuit Flip Section Vertically

Create new design Flips selected
file: Section 3.B components vertically

RSoft CAD  25
 Open Circuit Rotate Selection
Open existing Rotate selected
design file components

Save Circuit Send Forward/Send Backward

Save current Change component order
design file of selected components

Undo Last Change Send to Front/Send to Back

Reverse last change made Change component order
to the design file of selected components

Redo Last Change View Graphs

Redo last change made Open plot file
to the design file in WinPLOT.

Cut Selection Launch DataBROWSER

Cut selected Open DataBROWSER
components

Copy Selection Open Folder In Working Directory

Copy selected Open a file browser in the working
components directory

Paste Clipboard Contents Open Terminal in Working Directory

Paste clipboard Open a terminal (DOS) window in
contents working directory

Duplicate Selection Help for CAD program

Duplicate Open online help
selected components for RSoft CAD

Flip Section Horizontally Help for Active Sim Program

Flip selected Open online help for
components horizontally current simulation tool

The View Toolbar

Choose Simulation Tool Cut View Controls

Select active Display specific
simulation tool slice of the structure

Single-Pane/Four-Pane Zoom In
Switch between single-pane Zoom in
and multi-pane view CAD window

RSoft CAD  26
 Edit X Axis View Zoom Out
Display YZ plane Zoom out
CAD window

Edit Y Axis View Zoom Full

Display XZ plane Set viewing domain of the
CAD so entire structure is visible

Edit Z Axis View Aspect Ratio

Display XY plane Set aspect ratio for
the selected cut view

Show 3D View Redraw Circuit

Display 3D view Redraw components

Toggle Component Tree

Hide/Reveal tree view toolbar

The Side Toolbar

Select Mode Zoom Mode

Select components: Zoom in/out:
Section 3.F.1 Section 3.E.3

Drawing Tool Options Pan Mode

Set drawing preferences: Pans CAD display:
Section 3.D.3 Section 3.E.4

Segment (In Plane) Segment (Out of Plane)

Draw straight in-plane segment Draw straight out-of-plane segment
component: Section 3.D.1 component: Section 3.D.1

Arc S-Bend
Draw arc segment Draw S-bend segment
component: Section 3.D.1 component: Section 3.D.1

Lens Circle
Draw lens component: Draw circle/cylinder/sphere:
Section 3.D.1 Section 3.D.1

Polygon
Draw polygon objects:
Section 3.D.1

RSoft CAD  27
 Circuit Reference Dynamic Array
Draw circuit references: Section 6.F Draw a dynamic arrays: Section 6.G
Extended Segments Structure Preference
Nested menu to draw extended Nested menu for structure type
segments, see section below for all sub- preference, see section below for all
options sub-options

Simulation Region Monitor

Draw BeamPROP simulation Draw monitor: FullWAVE,
region: BeamPROP DiffractMOD, ModePROP, BeamPROP

Edit Global Settings Edit Symbols

Open Global Settings Open Symbol Table:
dialog: Section 3.B Section 3.C

Edit Materials Edit Layers

Open Material Editor: Open Layer Table Editor:
Chapter 8 Section 4.C.5

Edit User Profiles Edit User Tapers

Open User Profile Editor: Open User Taper Editor:
Section 6.A Section 6.B

Edit Launch Field

Open Launch
Parameters dialog

Edit Pathways Edit Pathway Monitors

Pathways are used to define launch Create and modify pathway monitors:
positions and for analysis: Section 6.H BeamPROP and FemSIM

Edit Global Settings Launch MOST

Open Simulation Parameters Open the MOST optimization
dialog to start simulation and scanning utility

Display Material Profile Compute Modes

Open the Display Material Compute transverse
Profile dialog: Section 3.G modes: Chapter 11.A

Extended Segments:
Extended Segments are discussed in more detail in Section 4.G. Though illustrated with a
circular cross-section, these segments can have any 3D Structure Type.

RSoft CAD  28
 Straight Extended Segment Arc Extended Segment
Draw straight Draw arc
extended segment extended segment

Half-Taper Extended Segment Toroidal Extended Segment

Draw half-taper Draw toroidal
extended segment extended segment

Full-Taper Extended Segment Helical Extended Segment

Draw full-taper Draw helical
extended segment extended segment

Structure Preference:
The structure preference is discussed in Section 3.D.3 and sets the 3D Structure Type of
segments to be drawn. The default choices indicate that the setting in the Global Settings dialog
should be used.

Fiber Structure Preference Triangle Structure Preference

Fiber structure type: Polygon structure type
Section 4.D.1 with 3 sides: Section 4.D.6

Channel Structure Preference Pentagon Structure Preference

Channel structure type: Polygon structure type
Section 4.D.2 with 5 sides: Section 4.D.6

Diffused Structure Preference Hexagon Structure Preference

Diffused structure type: Polygon structure type
Section 4.D.3 with 6 sides: Section 4.D.6

Rib/Ridge Structure Preference Octagon Structure Preference

Rib/ridge structure type: Polygon structure type
Section 4.D.4 with 8 sides: Section 4.D.6

Multilayer Structure Preference N-sided Structure Preference

Multilayer structure type: Polygon structure type
Section 4.D.5 with N sides: Section 4.D.6

Polygon Structure Preference Polygon Datafile Structure

Polygon structure type: Preference
Section 4.D.6 Polygon structure type defined
by a data file: Section 4.D.6

RSoft CAD  29
3.A.3. Status Bar
The status bar is located on the bottom of the CAD window, and provides information about the
current mode of the CAD interface as well as the current coordinates of the mouse.

Figure 3-1: The status bar at the bottom of the CAD window.

3.A.4. Tree View Toolbar

The tree view toolbar is located on the right side of the CAD window, and provides a
hierarchical view of the elements present in the CAD window. Double-clicking on any branch
folder (Components, Monitors, Launch Fields, etc.) in the tree-view will list all elements of that
type that are currently present in the CAD window.

Figure 3-2: Tree view for a simple structure consisting of 2 Segments, a Lens, and a Launch
Field.

Any element in the CAD can be easily modified by right-clicking on it in the tree view toolbar
(for example, Segment #1 or Launch #1 in Figure 3-2). New elements of any type can be created
by right-clicking on the appropriate branch folder in the tree view. A branch folder will be
greyed out if it is not relevant to the simulation tool being used.

RSoft CAD  30
If desired, the tree-view toolbar can be hidden by clicking on the Toggle Component Tree icon
in the view toolbar.
The Component Filters folder can be used to create custom list(s) of components based on user-
defined searches/filters. For example, a filter can be made to list components that have a material
of ‘Si’, or to list all monitors, or all components with a priority of 4. The resulting objects can
also be hidden in the drawing windows, so for example, you can hide all components that have a
material of ’Si’, etc. See Section 6.O for more details.
Multiple elements in the CAD can be selected by right-clicking on them in the tree view toolbar
and holding the Ctrl and/or Shift keys down. Holding down the Ctrl key adds them one by
one whereas holding down the Shift key selects a range. In the case of multiple selections, the
last element is highlighted in the tree view toolbar and the other elements in the selection are
marked by an * in front of their element name. Operations can be performed on the elements in
the multiple selection list through right-clicking the highlighted element. One such operation is
adding these elements to a Group which will then appear as a named Group under the
Component Filters. The resulting Group can now perform similar operations like the Component
Filters, see Section 6.O for more details
Project materials are shown in the tree. If material colors are used (see Section 8.A.6), the
material color is shown in the tree.

3.A.5. RSoft Tips

The RSoft CAD has an automatic learning system which provides context-sensitive tips. These
tip windows are designed to convey relevant information about the software to both new and
experienced users, and will occasionally pop-up as you use the software.
While such systems are by their very nature intrusive, the RSoft Tips system is designed to
minimize this as much as possible. Tips are kept short with rare exceptions, and are located as so
not to obscure the current dialog. Tips also do not have to be dismissed; you can interact with the
main program window or current dialog as if the tip were not even there!
Even if you are an experienced user, we recommend that you use the tip learning system since
the basic information will pass quickly, and you are likely to discover new and useful aspects of
the software as more advanced tips appear.

3.B. Creating a New Design File

Click the New Circuit icon in the top toolbar or select File/New from the CAD menu to create a
new design file. The startup dialog appears where basic information about the structure is
specified. This information can be changed at any time via the Global Settings dialog. Also, all
options may not be relevant for all simulation tools or simulation types.

RSoft CAD  31
The Load Settings button can be used to pre-load a set of simulation settings, for example
typical Silicon wire settings. It is recommended to load the closest settings for your specific case
and then modify as needed. New settings packages can also be saved, see below.

Figure 3-3: The Startup window.

Simulation Tool and Dimensions:

Simulation Tool
All of these products share the RSoft CAD interface and are licensed separately. The simulation
tool can also be chosen from the view toolbar in the CAD.

Model Dimension
The dimension of the structure which can be either 2D or 3D. A 2D structure is in the XZ plane.
Note that in some cases, 3D structures can be automatically simulated in 3D using the Effective
Index Calculation option described below.

Radial Calculation
Performs a 2D radial simulation of structures that have azimuthal symmetry; this option is not
enabled for all simulation tools and is described in relevant simulation tool manuals.

RSoft CAD  32
Effective Index Calculation (2.5D)
The Effective Index Method (EIM) can be used to greatly reduce the computation time and
memory requirements of a simulation by converting a 3D structure into an approximate 2D
structure. This is sometimes called ‘2.5D’. This option only affects the simulation; the definition
of the original 3D structure is unaltered allowing this feature to be safely turned on and off
without changing the defined structure. See Section 6.L for details including supported
simulation tools and other requirements.

Global Default Settings:

These settings control the global default settings for components.

Free Space Wavelength

The wavelength [µm] of light used for simulation. It is measured in free space, i.e. λ=c/f where
c is the velocity of light in vacuum and f is the frequency of the light.

For most simulation tools, the wavelength is an input. Some tools solve for wavelength(s) as a
part of the simulation (BandSOLVE for example). For such tools this setting will, in most cases,
have no effect. Exceptions include using dispersive materials where the material properties are
defined as a function of the wavelength or when variables are defined as a function of the built-
in symbol for this option free_space_wavelength.

Background Index or Background Material

The real refractive index of the background material in which components are embedded. A real
value can be given via the Background Index, or a material can be assigned via a Background
Material. For 3D simulations, the definition of this property depends on the default 3D Structure
Type as described in Section 4.D. See Chapter 8 for more information about materials.

Component Material and Component Delta-N

These options set the default material properties of a component:
• The Component Material is the default material for components. If set, it overrides the
Component Delta-N setting below and any components with the default material properties
will use the defined Component Material.
• Component Delta-N sets the default difference between the real refractive index of a
component and the background index. The default imaginary refractive index (loss) of a
component can be specified through the variable alpha in the symbol table.
Individual components can be assigned different real and/or imaginary values as described in
Section 4.B.

RSoft CAD  33
Component Width and Component Height
The default width and height of a component along the X and Y axes respectively. Individual
components can also have a different width and/or height described in Section 4.B. The height is
only used for 3D simulations and is set equal to the width by default.

Index Profile Type

This option determines the default functional form of the refractive index profile along the
transverse direction of a component. Individual components can have different profiles. See
Section 4.E.

3D Structure Type
The default 3D structure type. See Section 4.D.

Cover Index, Cover Material, Slab Index, Slab Material, and Slab Height
These options set the refractive index when 3D Structure Type is set to Rib/Ridge or Multilayer.
See Section 4.D.4 and Section 4.D.5 for more details.

PDK Info file

Selects the PDK Info file to use to set relevant settings set via the pdkinfo() function. See
Section C.F.10, this option can be automatically set for new files using the Preference described
in Section 6.J. PDK info files can also be selected from one or more RSoft CAD Info folders, see
the Custom PDK manual for more information on using RSoft CAD Info folders.

Symbols, Materials, Loading Settings:

The buttons in the bottom of the Startup Window dialog can be used to set Symbols, select
Materials, or Load pre-defined settings.

Settings Files:
Settings files are .ind file that contain convenient starting points for a structure/simulation.
There are built-in settings files, users can create their own setting files, or can be loaded from
one or more RSoft CAD Info folders. See the Custom PDK Utility manual for more information
on using Settings files and RSoft CAD Info folders.

3.C. The Symbol Table & Variables

The Symbol Table is a powerful concept in the RSoft CAD and is easily the most-used feature.
As such, a proper understanding of the concepts in this section is critical to using the CAD and
associated simulation tools effectively.

RSoft CAD  34
3.C.1. Why use the Symbol Table & Variables?
The Symbol Table has two main uses:
• Modify existing built-in variables
Every option in the RSoft CAD and simulation tools is controlled by a built-in variable in the
symbol table. While most options can be set in the graphical interface, advanced users can
modify these variables in the symbol table to access these features. This concept is the basis
for RSoft’s convenient scripting capabilities: see Chapter 10 for more details.
• Create user-defined variables
User-defined variables can be used to set various aspects of a design. These variables can be
a simple numerical value, or an arbitrary mathematical function of other variables. This is an
extremely powerful and unique feature of the RSoft CAD which allows for the creation of
parametric designs, facilitates parameter scanning and optimization, and simplifies
simulation scripting.

It is recommended that the names user-defined variables start with a capital letter. Variables that
start with a capital letter are displayed on top of the variable list making them easier to find.

3.C.2. Using the Symbol Table

To open the Symbol Table, click the Edit Symbols button in the left toolbar or select
Edit/Tables/Edit-Symbols... from the CAD menu.

Figure 3-4: The Symbol Table Editor dialog which allows access to both built-in and user-
defined symbols.

RSoft CAD  35
Most of the variables shown in Fig. 3-4 correspond to the settings shown in the startup dialog
described in Section 3.B. Altering any of the values will immediately affect the corresponding
parameter in the design file and is equivalent to changing the parameter in the Global Settings
dialog.

Adding a New Variable

To add a new variable to the Symbol Table, simply click New Symbol, type the desired variable
name in the Name field, enter the value in the Expression field, and click Accept Symbol. To
reject a variable without saving it to the variable list, click Reject Symbol instead of Accept
Symbol.

Modifying an Existing Variable

To modify an existing variable, select the desired variable, change the value in the Expression
field, and then click Accept Symbol. To reject a variable without saving the modification, click
Reject Symbol instead of Accept Symbol.

Removing a Variable
To remove a variable from the symbol table, select the variable and click Delete Symbol.

An Example
As an example, open the file <rsoft_dir>\examples\beamprop\[Link]. Open the
symbol table and note that the variable width is set to 4 indicating that the default width of
components in this design is 4 µm. Change the value of this variable to 10 and click OK to return
to the layout window. The structure will have adjusted accordingly.

Additional Notes
• Variables names are case-sensitive.
• Variables can be defined as an analytic expression involving any other variable in the symbol
table. See k0 shown in Fig. 3-4. Appendix C documents the syntax and functions that can be
used.
• Variables can be edited via a text editor using the Text Editor button. Any changes here will
be automatically reloaded back into the Symbol Table when the text file is closed. A default
text editor is used, set the environment variable RSOFT_EDITOR to the path of the text editor
of your choice.
• Numerous built-in variables control all CAD and simulation options. A list of built-in
variables that correspond to CAD options can be found in Appendix D; built-in variables that
correspond to simulation tool settings can be found in similar appendices in each simulation
tool manual.

RSoft CAD  36
• Built-in variables that do not appear in the symbol table are internally set to their default
values. Do not use the name of a built-in variable as a user-defined variable to avoid conflict.
The easiest way to avoid this problem is to start the name of any user-defined symbol with a
capital letter since the names of all built-in symbols are entirely lower case.
• Variables can be used in place of filenames where applicable in the graphical interface. To
use this feature, create the variable and set its value to the name of the data file. Then, enter
$name in the graphical interface where name is the name of the variable.

3.D. Drawing Components

Drawing components in the CAD is very simple. Most geometric shapes (e.g. tapered or curved
structures, periodically varying structures, etc.) can be produced from variants these components
or by overlaying more than one component.

3.D.1. Drawing Modes

The CAD has several drawing modes that control the type of components to be added to the
design. They can be selected through the left CAD toolbar and can be broadly categorized by the
number of vertices (end points):

Two-Vertex Components:
All two-vertex components in the CAD are generalized forms of the same component, the
generalized segment. However, several drawing modes are provided to simplify the creation of
these different types:

See Chapter 4 for a complete description of the segment component type. Also, one type of
segment can be changed into another by modifying its settings.

Segment (In Plane), Segment (Out of Plane), and Extended Segments

These drawing modes produced segments whose axis follows either a straight or curved line in
the drawing plane or perpendicular to the drawing plane. Its properties can be changed so that it
lies in any plane after it is drawn. Extended segments can be arbitrarily curved.

Arc and S-Bend

This drawing mode produces curved segments whose axis follows either a circular arc or an s-
bend shape. The exact shape can be set using the arc properties as described in Section 6.B.3.
Note that custom tapers can be used as well to define custom arc shapes.

RSoft CAD  37
One-Vertex Components:
One-vertex components that can be added to a design file are:

See Chapter 5 for a complete description of these component types.

Lens and Circle

These drawing modes produce lens, cylindrical, or spherical components.

Polygon
This drawing mode produces polygon components defined by a series of points in the XZ plane.

Circuit Reference
This drawing mode produces circuit reference components.

Simulation Region
This drawing mode produces simulation region components and is only available for
BeamPROP.

Monitor
This drawing mode produces monitor components and is only available for FullWAVE,
DiffractMOD, ModePROP, and BeamPROP.

3.D.2. Placing Components in a Design File

Components are drawn by first selecting the desired drawing mode and then using the mouse in
the layout window. One-vertex components are added by simply clicking at the desired location
of the vertex; two-vertex components are added by rubberbanding. Rubberbanding involves left-
click-dragging the mouse to draw the extent of the component. Please note:
• The center line of the component is shown while rubberbanding.
• The drawing process can be canceled by right-clicking or pressing ESC.
• Alternatively you can first click without moving the mouse to define the starting point, move
the mouse, and click again to define the ending point.
• The exact coordinates (as well as other component properties) can be directly set in the
Component Properties dialog discussed in Chapter 4.

Snap Modes
Snap modes provide simple ways to align or connect components:

RSoft CAD  38
• Snap To Drawing Grids
The layout window maintains both a fine and a coarse grid. Drawing coordinates can snap to
either of these grids reducing required hand-eye coordination: the fine grid is used by default,
and the coarse grid can be used by holding Shift. Holding Ctrl disables snapping.
• Snap To Components
When drawing components a vertex will automatically snap to the vertex of another
component if the two vertices are near enough. This is extremely helpful when drawing
components that need to fit together to form a larger structure. The tolerance for this can be
set in the CAD preferences. This relationship is preserved as a reference between the
components by the CAD. When one component snaps to another the coordinates are not
simply set equal but is logically attached to that component. When a component is moved,
any attached components move with it. Even if components aren’t drawn using this mode
they can be logically attached. See Section 4.B.1.

3.D.3. Drawing Preferences

The initial properties of a component are controlled by the drawing preferences. Click the
Drawing Preferences icon in the left toolbar to edit these preferences. By default, the values of
these properties are set by the user-defined values set in the Startup window and the CAD
defaults. The 3D Structure Type drawing preference can also be changed in the left CAD
toolbar.

It is important to understand that role of the Drawing Preferences: changing the Drawing
Preferences will not affect any existing components, only components that will be drawn.
Changing the options in the Global Settings will change all components that have properties
based on those options.

3.D.4. Try It!

Many techniques for adding components using the mouse have been described in this section. It
is best to experiment with these modes in order to better understand their operation.

3.E. Viewing Components

The CAD has several options that control how the components in a design file are displayed.
This includes single-pane and multi-pane mode, full and cut mode, zooming, panning, and the
display aspect ratio.

RSoft CAD  39
3.E.1. Single-Pane Mode
Single-Pane Mode displays a single view of the structure in the layout window and can be
accessed via the 1P button in the view toolbar. The X, Y, and Z buttons in the view toolbar set the
axis which the structure should be viewed along. In each view, the axes are clearly labeled in the
lower left corner. The Y (XZ) view is always used for 2D layouts.

Full & Cut Views:

Each axis view has two modes, Full Mode and Cut Mode. Full Mode displays all the components
at once, regardless of their position, and cut mode displays all the components that lie within a
specific cut plane. This allows, for example, only components that lie in the Y=100 plane to be
shown while in the Y axis mode (viewing the XZ plane).
By default, Full Mode is used and all components will be shown. To enter Cut Mode, click the
+/- buttons in the view toolbar. The current cut position is displayed, as well as the Cut Step
Size. The user can directly set both these fields. Cut mode is only available in 3D.

3.E.2. Multi-Pane Mode (3D Only)

Multi-Pane mode displays the X, Y, and Z axis views as well as a 3D view simultaneously. It can
be accessed via the 4P button in the view toolbar. The X, Y, and Z axis views can be used in
exactly the same way as in single-pane mode. Each view maintains its own viewing window.
The currently selected axis view is illustrated in the CAD window by a blue rectangle around the
active pane, as well as the state of the buttons in the view toolbar. Each axis view can also be in
Full or Cut Mode.

Using the 3D View:

The 3D view shows the structure rendered as 3D polygons and can be manipulated via the
mouse:
• Rotating: The 3D view can be rotated by clicking on it and moving the mouse.
• Zooming: The 3D view can be zoomed in or out by right-clicking on the 3D view and
moving the mouse up and down.
• Panning: The 3D view can be panned by holding CTRL and clicking on the 3D view and
moving the mouse.
• Display Modes: The 3D view supports several different display modes such as solid,
transparent, and wireframe. These modes can be cycled through by double-clicking the
mouse on the 3D view.
Additionally, holding Shift and double-clicking toggles the display of the axes, holding Ctrl
and double-clicking resets the view, and right-double-clicking enters an auto-animate mode.

RSoft CAD  40
3.E.3. Zooming
A layout window can be zoomed using the mouse, toolbar buttons, or menu items.
• Zooming via Toolbar Buttons
There are three zoom buttons on the view toolbar: Zoom In, Zoom Out, and Zoom Full. The
zoom percentage is 10% by default but can be set in the CAD preferences.
• Zooming via the Mouse Wheel
Holding the SHIFT key and using the mouse wheel will zoom in/out. The zoom percentage is
10% by default but can be set in the CAD preferences. This is available for Windows
systems only.
• Zooming via the Mouse
To zoom with the mouse click the Zoom Mode icon in the left toolbar. The mouse will
change to a magnifying glass and clicking the left and right mouse buttons will zoom in and
out respectively as well as center the display at the mouse position. Holding the Shift key
while pressing the left button will zoom out. The zoom percentage is 10% by default but can
be set in the CAD preferences.
• Zooming via Menu Items
It is possible to zoom via the CAD menu with the options under the View menu.

3.E.4. Panning
A layout window can be panned via scroll bars or the mouse.
• Panning via Scroll Bars
Each viewing window has scroll bars which can be used to pan the structure view. The scroll
bars can be removed if necessary via the CAD preferences described in Section 6.J.
• Panning via the Mouse
To pan with the mouse click the Pan Mode icon in the left toolbar or hold the CTRL and
SHIFT keys. The structure can then be moved with the mouse. Alternatively, use the mouse
wheel to pan vertically, hold the CTRL key and use the mouse wheel to pan horizontally.

3.E.5. Display Aspect Ratio:

The display aspect ratio controls how the two axes in the drawing window are scaled. For
structures that are much bigger in one direction than another, or for structures where it is critical
to see the actual dimensions, it can be useful to change the aspect ratio to better view the
structure. The aspect ratio can be set via the view toolbar, or via the View/Set-View-
Parameters… menu item. When Auto is chosen from the pulldown menu, the structure will be
scaled to best fit the viewing window.

RSoft CAD  41
3.E.6. Displaying ‘Simulation’ Objects
The CAD, by default, displays additional objects in the drawing windows:
• Simulation Domain
The simulation domain is displayed as a dark solid line and cannot be selected or moved with
the mouse. For BandSOLVE and GratingMOD, it is displayed as a dashed line which
represents the domain that is used to when displaying material properties. For BandSOLVE,
the actual simulation domain is defined by the lattice vectors and can be viewed by clicking
the View Domain button in the Simulation Properties dialog. For GratingMOD, the actual
simulation domain is defined so as to include the entire structure.
• Launch Fields
Launch fields are displayed as orange rectangles. Right-clicking on these objects in the
layout window will open the Launch Properties. These objects cannot be moved with the
mouse.
• Direction Indicators
Both launch field and monitor objects have associated directions and can be displayed in the
CAD layout window.
The display of these objects can be disabled/enabled via the CAD Preferences window.

3.F. Editing Components

This section discusses how to select, move, scale, flip, and rotate components, as well as
standard editing functions such as cut and paste.

3.F.1. Selecting Components

Use the Select Mode option in the left toolbar to select components. When in this mode the
cursor will appear as an arrow and no drawing is possible.

If reading this section for the first time it might be beneficial to create a simple structure with
several components to experiment with.

To select a single component, simply click on the component. The selected component will be
highlighted with small black handles which can be used for resizing or scaling the component as
discussed below. The component number of the most recently selected component is shown in
the CAD status bar.

RSoft CAD  42
 Figure 3-6: A straight component that has been selected by clicking on it.

To deselect component(s), click an unoccupied area of the layout window.

Selecting Components in Slice and Multi-Pane Mode (3D Only)

Some components will be easier to select in different cross-section views. While two
components might overlap in one cross-section, they might not in another. Furthermore, it is
possible to select segments in both the Full and Cut Modes described in Section 3.E.1.

Selecting Components by Number or Name

It can be convenient to select a component based on its number or name especially when one or
more components occupy the same physical space. To do this, use the CAD menu item
Edit/Select/Edit-Component and enter the number of the desired component in the dialog. See
Section 6.K for more details on setting a component’s name.

Selecting Multiple Components

Multiple components can be edited at once in the CAD by dragging a selection box around the
components with the mouse. Additionally, a component can be added/dropped from the current
selection by holding the Shift button and clicking on the component.

3.F.2. Modifying Components

Once a component or set of components has been selected, they can be operated on in several
ways:

Basic Operations
These are simple mouse-based methods to move, resize, flip, and rotate components. Complete
control over a component can be achieved by directly setting the component properties in the
Component Properties dialog described below.

The snap features described in Section 3.D.3 are not enabled by default when editing existing
components. It can be enabled if needed in the CAD Preferences.

RSoft CAD  43
• Moving
To move the selection, simply drag one of the components in the selection to the desired new
position. If the position of the component(s) has been defined by variables in the Component
Properties dialog, a dialog will be shown where the user can set an offset which will be added
to the current position of the selected components. Both numeric and variable-based
expressions can be used. To force this dialog to be shown hold CTRL while moving the
components.
• Resizing
To resize or scale the selection along a particular direction, simply grab the corresponding
handle of any one of the selected components and move it to the desired position.
• Flipping
The Flip Horizontal and Flip Vertical icons in the top toolbar can be used to flip components
around their center. To flip around their top or buttom, hold the left or right Shift key while
clicking the icon. Alternatively, use the Edit/Flip-Horizontal/ and Edit/Flip-Vertical CAD
menu items.
• Rotating
The CAD also has a limited capability for rotating components via the Edit/Rotate menu
option or the view toolbar . Several limitations on this option exist depending on the
component type.

Standard Editing Operations

The CAD layout system possesses the standard editing functions for cut, copy, paste, and delete.
These functions are accessed via icons in the top toolbar or through the Edit menu item.

Using the Component Properties Dialog to Edit One or More Components

The most complete method for modifying a component’s properties is to directly edit the
component’s properties dialog which can be opened by right-clicking on the component. For a
complete description of this dialog, see Chapter 4.
To edit more than one component at a time, first select the components to be edited. Once a
group selection has been made, right-click on any component in the selection (the primary
component) to open its properties dialog. Generally, any changes made in this dialog will be
applied, if applicable, to all selected components. Some notes:
• If a mixed group of components has been selected (that is a group of components with more
than one component type), the user will be presented with a choice to apply changes to all
components or just those of the same type of the primary component.
• Special consideration must be given if the selected group of components contains both
single-vertex components (lens, circles, monitors, etc) and two-vertex components

RSoft CAD  44
 (segments). In this case, the single-vertex components are considered to have only a starting
vertex. Thus, any changes to the vertex information (refractive index, position, width, and/or
height) of a single-vertex component will be applied to all single-vertex components as well
as the starting vertex of any selected segments. Any changes to the ending vertex of a
segment will only be applied to the ending vertex of any other segments. It is possible to
modify this logic using the previous note about editing mixed groups.
• After changes have been made in the properties dialog, the CAD will, for all changed
options, check if all selected components had the same initial value as the primary selected
component. If they do, the new value will be applied to all selected components. If not, the
user will be presented with a choice to apply the change to all selected components (if
applicable) or just those components that had the same initial value as the primary selected
component.

Changing Component Order

When one component overlaps another component, it can be useful to change the component
numbers so that a certain component can be viewed as on top of another. To do this, select one of
the options under the CAD menu item Edit/Change-Order to move a component forward,
backward, to the front, or to the back. This is done by changing the component number.
By default, the components are drawn in the RSoft CAD by order of the component number
(first component is drawn first, etc.), but this procedure can be overridden using the Draw
Priority setting described in Section 6.K.1.

3.G. Calculating Material Profiles

It is very useful to view and save plots of material properties such as the actual refractive index
profile to be used in a simulation. This option should be used liberally as one of the main causes
of inaccurate simulation results is the use of an undesired index distribution.

Using an inaccurate index distribution results in inaccurate simulation results.

3.G.1. Computing a Material Profile

A material profile can be calculated via the Display Material Profile button in the left toolbar.

RSoft CAD  45
 Figure 3-6: The Compute Index Profile window.

The relevant parameters are discussed in the following sections.

Calculation Parameters
The calculation parameters set in this dialog determine the domain and grid that are used to
calculate the desired profile. While default values for these parameters are given, there are some
guidelines for choosing these parameters:
• Calculation Domain
The domain should be large enough to encompass the area of interest in the structure. The
default values correspond to the whole structure.
• Grid Sizes
The grid sizes should be small enough to resolve the smallest feature in the structure.
• Slice Grid
The index information is saved and displayed on the slice grid. For most calculations, it is
best to set the slice grid and grid size equal to best view the index.
• Material Property
The real or imaginary index profile can be computed, as well as other material parameters for
3D simulations. See Chapter 8 for more information on these material properties. See
discussion at the end of this section for information on displaying anisotropic index profiles.
• Overlay Grid

RSoft CAD  46
 This option, enabled by default, overlays the simulation grid on the material profile. Note
that this only works with the contormap and 3D viewer display modes.
When the OK button is pressed, the desired profile is displayed as a line plot for a 2D profile or a
contour plot for a 3D profile.

Choosing the Display Mode

The Display Mode field sets the display type to be used for the calculation. Several of the
parameters discussed below can be found in the Display Options dialog and are discussed in the
next section. The Display Mode options are:
• Default (Fixed Z) – Computes a cross-sectional profile at the Z value specified in the Z Cut
field. By default, this corresponds to the Z Domain Min.
• Slices – Displays the property along Z as slices every value of Z given by the Z Slice Grid.
• 3D Slices – Displays the property as a function of X and Z as 3D slices along Z at every
value of Z given by the Z Slice Grid. In 3D, the Y position of the slices is specified by the
Slice Position Y.

• WireFrame – Displays the property as a function of X and Z as 3D wireframe graph. In 3D,

the Y position of the slices is specified by the Slice Position Y.
• SolidModel – Displays the property as a function of X and Z as a 3D solid model with
shading to portray an illuminated surface. The color of the surface is specified by the field
Surface Color, and in 3D, the Y position of the slices is specified by the Slice Position Y.

• HeightCoded – Displays the property as a function of X and Z as a 3D contour graph with

color coding to indicate height, or index. The color scale can be set via the options discussed
below.
• ContourMap(XZ) –Displays the property as a function of X and Z as a 2D color-coded
contour map. The color scale can be selected via the options discussed below. In 3D, the Y
position of the slices is specified by the Y Cut.
• ContourMap(YZ) –Only available for 3D calculations. It displays the property as a function
of Y and Z. The color scale can be selected via the options discussed below. The X position
of the slices is specified by the X Cut.
• ContourMap(XY) – This option is only available for 3D calculations, similar to the Default
(Fixed Z) option.
• SurfaceRelief – Displays a surface map for the 3D structure types Rib/Ridge and Multilayer.
• 3D Volume – This option is only available for 3D calculations and displays a 3D rendering of
the property distribution using the MayaVi program. This type of display can be very
memory intensive for structures with many grid points, so it is recommended that you
increase the slice grid to avoid unnecessarily large memory usage.

RSoft CAD  47
• 3D Viewer – This option is only available for 3D FullWAVE designs. It calculates a 3D
refractive index profile, allowing the user to easily browse through different cross-sections.
See the FullWAVE manual for more details.

Setting Display Options

The material profile display options are set in the Index Display Options dialog which can be
opened via the Display… button in the Compute Material Profile window.

Figure 3-7: The Index Display Options dialog.

The index display options are:

Surface Color
Sets the surface color for plots produced when the Display Mode is set to SolidModel.

View Phi and View Theta

Sets the viewing angles Phi and Theta of the plots in the simulation window when the Display
Mode is set to either SolidModel or HeightCoded

Allow Redraw 2D/ Allow Redraw 3D

Indicates that 2D/3D plots displayed in the simulation window should be stored in memory so
that the plots can be redrawn if needed. This option is necessary for computer systems with
minimal RAM; most computers will work fine with this option enabled.

Index Display Min and Index Display Max

Sets the minimum and maximum index used when calculating an index profile. Index values
which are under this value are displayed in black and white respectively.

Outline Color

RSoft CAD  48
This pulldown menu sets the color of the circuit outline to be displayed over the simulated fields.

Color Shades, Color Scale, and Show Color Scale

These options control the colors used to render plots in the simulation window. Color scale files
can be chosen via the Scale Files… button. See Appendix F for more on the use of color scales.

Plot Aspect Ratio

This option sets the aspect ratio of the plot in the simulation window.

Setting Output Options

The index profile output options are set in the Index Output Options dialog which can be opened
via the Output… button in the Display Material Profile dialog.

Output index/epx/mu in complex format

Enables complex output.

Output All index/eps/mu tensor elements

Indicates that all elements on the index tensor should be output. This option was previously
called Output All Index Tensor Elements.

Notes for Anisotropic Designs

For anisotropic designs, the variable index_profile_tensor_component selects the tensor
element to be displayed. This variable can have values from 1 to 9 which correspond to the xx,
xy, xz, yx, yy, yz, zx, zy, and zz elements respectively. This variable can be used with the
options described above if needed.

3.G.2. Saving a Material Profile

In order to save the material property profile simply enter an Output Prefix.. The real part of the
refractive index is saved in the file <prefix>.ipf, and a corresponding WinPLOT command file
for the graph will be saved in the file <prefix>.ppf, where <prefix> is the output prefix
specified. The analogous files for the loss profile are <prefix>.lpf and <prefix>.ppl. Please
refer to Appendix B for further information on these files and their formats.

3.G.3. Producing Index Profile Plots from the Command Line

Index profiles can be produced from the command line by simulating the *.ind file with the
executable for the simulation tool BeamPROP ([Link]/xplot for Windows/Linux) via
the variable index_profile. The value of any relevant calculation parameters will be taken
from the *.ind file if not specified in the command. This variable should be set to 1 to compute

RSoft CAD  49
the index profile and 2 to compute the loss profile. The default of 0 computes the optical field as
usual. Users who are not licensed for BeamPROP can still perform this type of calculation.

3.H. Viewing the Current Value of a Parameter

The current value of any option can be viewed by holding the CTRL key and double-clicking on
the option. For example holding the CTRL key and double-clicking on the input box for the
Component Width will display the current value of the width.

3.I. Easily Edit Long Expressions

When editing long expression, hold the SHIFT key and double-click on the option. A larger pop-
up editor will display. After editing, close the window and the option will be automatically
updated.
It is also possible to edit the Symbol table in a text editor. Open the Symbol Table, click Text
Editor, make any necessary changes and/or add new symbols, then close text editor and Symbol
Table will be updated.

3.J. Editable Drop-Down Lists

Several drop-down lists in the RSoft CAD and simulation tool dialogs are editable, allowing
them to not only be set to one or several pre-defined values, but also to a symbol table
expression. This capability further extends the parametric design capabilities of the RSoft CAD,
allowing these drop-down lists to be set by symbols. This advanced feature enables these options
to be scanned over via MOST and set via the import/export options described in Section 6.I.

Figure 3-8: An editable drop-down list a) during editing, and b) after it has been set.

Note that these options can only be edited with the mouse on Linux, it is only currently possible
to edit these options via the mouse on Windows platforms.

RSoft CAD  50
To edit one of these fields, click on the drop-down list and note that the current value is
highlighted. Simply type the desired expression in the box as you would with any other text
option. Once set, the option will appear as shown in Fig. 3-8a. The current value is shown in
square brackets along with the expression.

It is recommended to use values that correspond to the internal values of the drop-down list. The
list of built-in values for a particular editable drop-down lists is included in the documentation
for that particular option.

As an example, we illustrate a series of symbol table expressions that change the Width Taper as
a function of Model Dimension (i.e. 2D vs. 3D). First, here are the built-in options for the Width
Taper setting shown in Fig 3-9 are:

Width Taper List Option Built-In Value

None TAPER_NONE

Linear TAPER_LINEAR

Quadratic TAPER_QUADRATIC

Exponential TAPER_EXPONENTIAL

User 1 TAPER_USER_1

Here are the expressions that can be used to set the Width Taper as a function of Model
Dimension (i.e. 2D vs. 3D):

Variable Value
Is2D eq(dimension,2)

Width_taper if(Is2D,TAPER_LINEAR,TAPER_QUADRATIC)

The Is2D variable is a ‘Boolean’ symbol which equals 1 when the design is 2D (dimension = 2)
and 0 when the design is 3D (dimension = 3). The Width_taper variable then equals
TAPER_LINEAR in 2D and TAPER_QUADRATIC in 3D. Using this variable to define the Width
Taper as shown in Fig. 3-9 effectively selects Linear for 2D and Quadratic for 3D designs.

RSoft CAD  51
4

Segment Components

CAD components are the basic building blocks from which a design is created. Any number and
combination of different components can be embedded in a background material of a specific
refractive index to create the refractive index of the entire structure.

In this chapter the discussion is intentionally limited to the segment component type. Many of
the concepts in this chapter apply to other component types; see Chapter 5 for a detailed
description of other components and their properties can be found in the next chapter.

A complex design can be achieved by simply adding a combination of components with varying
parameters to a design file. Sometimes there is more than one way to create a specific index
distribution. In theory, it does not matter how the index distribution was created; however a
particular approach might be easier to implement and lend itself to a simpler simulation later on.

4.A. Segment Overview

Segments are the most basic component type used by the CAD. This section provides a
background to understand how segments are defined in the RSoft CAD.

4.A.1. Basic Segment Properties

A segment is composed of two vertices, one at the start and another at the end of the segment.
• 2D Layouts:

RSoft CAD  52
 In 2D (XZ), a segment has a transverse width and the X position, width, and refractive index
can be varied as a function of a normalized coordinate between the two vertices using tapers
and the index structure of the segment cross-section can be set via profiles. See Section 4.E
for details.
• 3D Layouts:
In 3D, a segment has a transverse width and height and the segment X, Y, and Z position,
width, height, index, and psi (twist) can be varied as a function of a normalized coordinate
between the two vertices using tapers (Section 4.E). The index structure of the segment
cross-section is set by the 3D Structure Type (Section 4.D) and profiles (Section 4.E) to
achieve fiber, multilayer, custom, etc. segments.

4.A.2. Segment Orientation Overview

Segment Orientation is covered in detail in Section 4.F, but it is helpful to mention here as well.
It refers to how segment properties are set relative to the segment axis and results in z-axis
segments, seg-axis segments, and extended segments.
As an example of why this is important, consider whether the endfaces of the segment should be
perpendicular to the segment axis or along a specific axis. While this may seem trivial, it has
important implications for different design areas: Waveguide designers want segments that fit
well without gaps and nano-structure designers may want cylinders that are oriented along an
arbitrary axis. Both are valid needs and the RSoft CAD provides ways for both designers to get
what they need. See Section 4.F for more details about Segment Orientation.

RSoft CAD  53
4.A.3. Setting Segment Properties

Figure 4-1: The Segment Properties dialog where individual component properties can be set.

By default, segment properties are set either by the user-specified data in the Startup/Global
Settings window or program defaults. The CAD is, however, an object oriented design
environment and allows the user to individually change these properties as needed.
To access any component’s properties dialog right click on the component in the layout window
(except in Zoom or Pan Mode). The rest of this chapter will contain a detailed description of the
properties shown in Fig. 4-1. Multiple components can be edited at once as described in Section
6.L.

4.A.4. Setting Segment Display Color

Every segment can have a different color. To set the color, use the Display Color option in the
segment properties dialog. This option is an editable drop-down list which, in addition to being
set via the drop-down can also be a Symbol Table expression. See Section 3.J for details about
how to use this feature. The built-in values for this option are:

Display Color List Option Built-In Value

Default Unset. If material
colors are used,
defaults to the material

RSoft CAD  54
 color, see Section
8.A.6.
Black 0

Dark Blue 1

Dark Green 2

Dark Cyan 3

Dark Red 4

Dark Magenta 5

Brown 6

Light Gray 7

Dark Gray 8

Light Blue 9

Light Green 10

Light Cyan 11

Light Red 12

Light Magenta 13

Yellow 14

White 15

4.B. Vertices
A vertex is a component property that defines the position, width, height, and refractive index of
a component. Segments have two vertices: the properties of the starting vertex are set in the
lower left portion of the dialog; the ending vertex in the lower right portion as seen in Fig. 4-1.

Segment properties can tapered, or changed, between the vertices. See Section 4.E for details.

4.B.1. Setting Position

There are three methods to specify the coordinates of a vertex.

RSoft CAD  55
 Figure 4-2: An example of the three reference types possible.

The Reference Type, which sets the method to be used, is set for each coordinate by a dropdown
menu:
• If the Reference Type is None, the coordinate is specified absolutely by entering its value in
the X, Y, or Z coordinate field. This is the simplest method.
• If the Reference Type is Offset, the coordinate value is determined in a relative manner by
adding the offset value, which is specified in the Offset Value field, to the reference
coordinate. The reference coordinate is specified in the Relative To fields. This is the most
commonly used method. Typically these coordinates are defined within the global design
coordinates of the design file, see Section 6.M for details about using the local coordinate
system of the vertex.
• If the Reference Type is Angle, the coordinate value is determined such that a line
connecting the vertex with the reference vertex specified in the Relative To fields makes a
specific angle with respect to the Z axis. The angle is measured clockwise from the Z axis
and is specified in the Angle Value field in units of degrees.

When specifying a reference vertex via Relative To, give a component number and whether the
vertex is the starting vertex or the ending vertex. For components with only one vertex use the
starting vertex. Circuit References can have additional vertices, see Section 6.F.7.

When a component is added to a design file, these settings are filled in appropriately based on
the beginning and ending mouse positions. By editing these fields, the user can explicitly control
the positioning of components with great flexibility. As an example, consider the portion of a
component properties dialog shown in Fig. 4-2 where the X coordinate is set directly, the Y
coordinate is defined via an Offset reference, and the Z coordinate is defined via an Angle
reference.

4.B.2. Index Definition

There are two methods to define the refractive index of a component:
• Via Materials:

RSoft CAD  56
 Materials are defined in the Material Editor and can be based on nk data files, etc. To use a
Material, select a material already in the project, or add a material by selecting New Material
under the Material setting for the component. These materials can also include dispersion,
non-linearity, anisotropy, and electro-optic, thermo-optic, and stress-optic effects. If a
Default Material has been defined in the Global Settings, all components will use that
material if they have the default material settings (Locally Defined, delta, alpha).
See Chapter 8 for more details about materials.
• Via an Index Difference:
Components are embedded in a background material and the real index definition is defined
as:

n(r ) = n0 + nf ( x, y, z )

where n0 is the Background Index and Δn is the local Index Difference. The function f(x,y,z)
is a profile function (Section 4.E). The imaginary index of the vertex can be set via the Index
(imag part) option which defaults to the variable alpha (0 by default).

Several notes:
• The Background Index and the default value for the Index Difference (the variable
delta) is set in the Global Settings window.

• By default, the index of the starting and ending vertices are the same. If different, the
segment’s index will be tapered as described in Section 4.E.
• This definition varies slightly for 3D segments based on the 3D Structure Type. See
Section 4.D for details on a specific 3D structure type.
• The real refractive index can be specified in absolute units. This mode can be enabled in
the CAD Preferences window. When this option is changed in an existing file, the CAD
will attempt to convert all existing components to the new mode. It is important to realize
that internally the CAD uses the index difference method and so it is the preferred
method.

4.B.3. Controlling Geometry

Several vertex properties contribute to the final geometry of a component including the
Component Width and Component Height which set the horizontal (X) and vertical (Y) extent
of the component respectively. These properties use the default values in the Global Settings
dialog (the variables width and height). The width and height of the starting and ending
vertices are equal; if each vertex has a different value, the segment will be tapered (see Section
4.E) For 3D segments, the Profile Type, and the 3D Structure Type can also affect component
geometry (see Section 4.D). Typically the Component Width and Component Height are set
locally, see Section 6.M for details about inheriting these settings from a referenced component.

RSoft CAD  57
4.C. Basic 2D Segments
Basic 2D segments are very simple to define in the CAD: They consist of only a Component
Width, Index Difference, and a starting and ending vertex. By default, the cross section of the
component is assumed to be step-index; however, this can be changed via a profile function as
discussed in Section 4.E. Furthermore, a material can be assigned to a simple 2D component as
described in Chapter 8.

Figure 4-3: The basic 2D component definition.

4.D. 3D Structure Types

The 3D Structure Type controls the basic transverse profile used for a component. It can be set
globally in the Global Settings dialog or individually at the top of a segment properties dialog.
When drawing segments, the Structure Preference icon in the left CAD toolbar indicates the 3D
structure type of the component that will be drawn. See Section 3.D.3 for more information
about Drawing Preferences.
Each structure type has a slightly different definition of how basic parameters are interpreted.
These parameters include Background Index, Index Difference, Material, Component Width,
and Component Height, as well other 3D specific parameters. While all these parameters can be
set in the Global Settings dialog, some of these parameters are set for segments individually in
the segment properties dialog. Additionally, many parameters can be tapered along the segment
and profile functions can be used to modify the refractive index (see Section 4.E).

RSoft CAD  58
4.D.1. Fiber
This structure type corresponds to a circular or elliptical core embedded in a uniform background
material.

Figure 4-4. Cross-section of the built-in 3D fiber structure.

Fiber Structure Parameters:

The following parameters are relevant for this structure type:

Background Index or Background Material

The refractive index, n0, of the background material and is set in the Global Settings dialog.

Index Difference and Index (Imag. part)

The real and imaginary refractive indices of the component. The real part, Δn, is defined as the
difference between the real refractive index of the core and the background material.
Alternatively, a material can be used to define the material properties by setting the Material
property in the Component Properties dialog. See Chapter 8 for more details on the use of
materials.

Component Width and Component Height

The size of the elliptical cross section in the horizontal (X) and vertical (Y) directions
respectively.

4.D.2. Channel
This structure type corresponds to a rectangular core embedded in a uniform background
material.

RSoft CAD  59
 Figure 4-5: Cross-section of the built-in 3D channel structure.

Channel Structure Parameters:

The following parameters are relevant for this structure type:

Background Index or Background Material

The refractive index, n0, of the background material and is set in the Global Settings dialog.

Index Difference and Index (Imag. part)

The real and imaginary refractive indices of the component. The real part, Δn, is defined as the
difference between the real refractive index of the core and the background material.
Alternatively, a material can be used to define the material properties by setting the Material
property in the Component Properties dialog. See Chapter 8 for more details on the use of
materials.

Component Width and Component Height

The size of the rectangular cross section in the horizontal (X) and vertical (Y) directions
respectively.

Additional Settings
A channel can have a sloped sidewall via the Sidewall Angle option. See Section 6.D.2 for
details.

4.D.3. Diffused
This structure type corresponds to a diffused structure in a substrate material. As shown in
Fig. 4-6, the "background" in this case consists of a semi-infinite region referred to as the

RSoft CAD  60
substrate which is defined for y<0, and another semi-infinite region referred to as the cover,
which is defined for y>0.

Figure 4-6: Cross-section of the built-in 3D diffused structure.

The position of the component is controlled by specifying the coordinates of the starting and
ending vertices. By default, the structure is positioned at y=0.

Diffused Structure Parameters

The diffused profile is defined as

n( x, y ) = n0 + ng ( x) f ( y )


where

1  w    w  
g ( x) = erf  2 + x  hx  + erf  2 − x  hx  
2      

Here, n0 is the substrate or background index, Δn is the maximum index change produced by
diffusion from an infinitely extended source, w is the width of the source in the horizontal
direction, and hx and hy are the diffusion lengths in the horizontal and vertical direction
respectively (hx = hy by default).
The vertical profile f(y) is determined by the variable diffusion_shape:
• diffusion_shape = SHAPE_GAUSSIAN [default]

f ( y ) = exp  − y 2 hy2 

• diffusion_shape = SHAPE_EXPONENTIAL

f ( y ) = exp  y hy 

RSoft CAD  61
• diffusion_shape = SHAPE_ERFC

f ( y ) = erfc  y hy 

The parameter γ is set by the variable diffusion_gamma, which equals 1.0 by default. Since it
represents the non-linear interaction between the concentration and index, it is applied after the
index changes for all components have been accumulated.
The following parameters are relevant for this structure type:

Background Index or Background Material

The refractive index, n0, of the background material which is set in the Global Settings dialog.

Cover Index or Cover Material

The refractive index, nc, of the cover material which is set in the Global Settings dialog.

Index Difference and Index (Imag. part)

The real and imaginary refractive indices of the component. The real part, Δn, is defined as the
between the real refractive index of the core and the background material. Alternatively, a
material can be used to define the material properties by setting the Material property in the
Component Properties dialog. See Chapter 8 for more details on the use of materials.

Component Width and Component Height

The width of the diffusion source, w, in the horizontal direction, and the diffusion length in the
vertical direction, or, hy, in the above expressions respectively. The diffusion length in the
horizontal direction is given by hx = hy *diffusion_ratio, where diffusion_ratio is a
symbol table variable which defaults to 1.0.
An example ‘process file’ for a titanium-diffused lithium niobate waveguide is included as
[Link] and is documented in [Link]. This file may be imported into another
.ind file via the Options/Import-Symbols… menu item described in Section 6.I.

4.D.4. Rib/Ridge
This structure type corresponds to what are commonly referred to as rib or ridge structures as
shown in Fig. 4-7. The "background" in this case consists first of a semi-infinite region referred
to as the substrate, which is defined for y<0. On top of the substrate is a thin film of a given
height and index, followed by another semi-infinite region referred to as the cover, which is
defined for y values above the top edge of the film. This background forms a one-dimensional
three-layer slab waveguide in the vertical direction.
The position of the component is controlled by specifying the coordinates of the starting and
ending vertices. In background region, where no structures have been defined, the film layer is

RSoft CAD  62
considered to exist up to a height defined by the Slab Height option in the Global Settings
dialog.

Figure 4-7: Cross-section of the built-in 3D rib/ridge structure.

Rib/Ridge Structure Parameters

The following parameters are relevant for this structure type:

Background Index or Background Material

The refractive index, n0, of the background material, and is set in the Global Settings dialog.

Cover Index or Cover Material

The refractive index, nc, of the cover material and is set in the Global Settings dialog.

Index Difference and Index (Imag. part)

The real and imaginary refractive indices of the component. The real part, Δn, is defined as the
between the real refractive index of the core and the background material. Alternatively, a
material can be used to define the material properties by setting the Material property in the
Component Properties dialog. See Chapter 8 for more details on the use of materials.

Slab Index or Slab Material

The refractive index of the thin film material defining the vertical slab in the background. By
default, this is defined by a formula such that the Slab Index equals the Background Index plus
the Index Difference. This option is set in the Global Settings dialog.

Slab Height

RSoft CAD  63
The height of the thin film material defining the vertical slab waveguide which forms the
background. This option is set in the Global Settings dialog.

Component Width and Component Height

The size of the rib’s cross section in the horizontal (X) and vertical (Y) directions respectively.

4.D.5. Multilayer
This structure type describes a multilayer structure and can be used to define simple rib
structures as well as more complicated layered structures.

Figure 4-8: Cross-section of the built-in 3D multilayer structure. Each layer can have a
unique height and index value.

The position of the component is controlled by specifying the coordinates of the starting and
ending vertices. A multilayer structure is defined using the following concepts:
• Substrate
The ‘background’ in this case consists first of a semi-infinite region referred to as the
substrate, which is defined for y<0.
• Layers
On top of the substrate are one or more layers, or thin films. Each layer is defined by its
height and complex-valued refractive index via a Layer Table, which is discussed in the next
section. Different components can be assigned different Layer Tables, resulting in a complex
arbitrary structure.
• Cover
The layer structure is followed by another semi-infinite region referred to as the cover, which
is defined for y greater than the top edge of the last layer.

RSoft CAD  64
In background region where no structures have been defined the layer structure is considered to
exist up to a height defined by the Slab Height option in the Global Settings dialog.

Multilayer Structure Parameters:

The following parameters are relevant for this structure type:

Background Index or Background Material

The refractive index, n0, of the background material and is set in the Global Settings dialog.

Cover Index or Cover Material

The refractive index, nc, of the cover material and s set in the Global Settings window.

Layer Table
Sets the layer table to be used. See description below.

Slab Height
The layer structure (defined by Layer Table 0) is considered to exist up to this height in the
background where no components are present. This option is useful since the slab structure is
usually similar to the core structure, and so only one layer table need be used to define both the
slab and core structure. This option is set in the Global Settings window.

Component Width and Component Height

The size of the rib structure in the horizontal (X) and vertical (Y) direction respectively. The
layer structure is considered to exist up to the height in the interior of the component. When this
value is less than the total height defined in the assigned layer table, the CAD will truncate the
layer table at this value. It is not recommended to set the height larger than the total height
defined in the assigned layer table as it can cause problems in certain cases.

Defining Layer Tables

The Layer Table Editor can be accessed via the Edit Layers button in the Global Settings dialog,
via the left toolbar, via the Edit/Tables/Edit-Layers... menu item, or via the Edit Layers button in
the segment properties dialog. The Layer Table Editor, shown in 4-9, allows the user to create
and modify layer tables. A layer table contains film height and refractive index, both real and
imaginary, or a material. Once a layer table has been defined, it can then be assigned to a
segment.

RSoft CAD  65
 Figure 4-9: The layer table editor dialog, which is used for defining the layer structure for a
multilayer component.

The upper portion of the Layer Table Editor selects the layer table to be edited. Layer table 0
defines the background layer structure, or regions where there are no components, as well as the
default layer structure within a component unless changed in the component properties dialog.
There are also contains buttons for creating a new layer table, deleting a layer table, and
duplicating an existing layer table. The right portion of this dialog controls the information for
the selected layer table and functions similarly to the symbol table editor described in
Section 3.C. The lower portion of the dialog allows the user to define the properties of the semi-
infinite substrate and cover, and are the same controls as those used in the Global Settings dialog.
Note that when defining a layer table other than the background one, or layer table 0, the user
may specify that the index information for a given layer is to be "transparent" to the background.
This is done by entering the keyword transparent in the corresponding field for the real or
imaginary part of the index. To use this feature, the given layer table must be able to merge with
the background; that is, there must be a one-to-one correspondence between the number of layers
and the layer heights. For example, this allows a component to exist only in a certain layer or
group of layers, as well as allowing for vertically coupled components.

Assigning a Layer Table to a Component

Much like the case of using a material to define the properties of a component, the Index
Difference and Index (imag part) fields in the segment properties dialog serve no function and
are disabled. They are replaced by the Layer Table field, which specify the layer table to be used
for the individual component being edited. This allows different components to be assigned
different layer structures, since the layer structure of the background is replaced by the specified
layer table within the component.

RSoft CAD  66
Creating Complicated Multilayer Structures
Multilayer components can be overlapped to achieve complicated multilayer structures. When
done, it is recommended that the user use as few layer tables as possible. Additionally, the Slab
Height and Component Height options described above should be used so that layer tables can
be ‘reused.’

4.D.6. Polygon Structure Types

This structure type corresponds to a polygon cross-section embedded in a uniform background
material as shown in Fig. 4-10. The polygon is inscribed in an ellipse centered at the vertex
position.

Figure 4-10: Cross-section of two versions of the built-in polygon structure: a) a regular
polygon with 6 sides (hexagon), and b) a user-defined polygon. Any number of sides or a
custom data file can be used.

Polygon Structure Parameters:

The following parameters are relevant for this structure type:

Background Index or Background Material

The refractive index, n0, of the background material which is set in the Global Settings dialog.

Index Difference and Index (Imag. part)

The real and imaginary refractive indices of the component. The real part, Δn, is defined as the
difference between the real refractive index of the core and the background material.
Alternatively, a material can be used to define the material properties by setting the Material
property in the Component Properties dialog. See Chapter 8 for more details on the use of
materials.

RSoft CAD  67
Component Width and Component Height
The size of the bounding circle in the horizontal (X) and vertical (Y) directions respectively.
When these parameters are equal, a regular polygon will be created.

Note that these options set the size of the bounding circle around the polygon. To instead specify
the length of a side of the polygon, specify the Component Width and Component Height using
the expression 2*(sqrt(S^2/(2*(1-cos(360/Num))))) where S is the desired side length and
Num is the number of sides in the polygon.

Additional Settings
A polygon structure can have a user-specified number of sides or use a data file to define the
cross-section shape of the component. These settings can be changed by clicking the Poly Info…
button in the Segment Properties dialog which opens the Polygon Structure Info dialog.

Figure 4-11: The Polygon Structure Info dialog.

Number of Sides
Sets the number of sides for a regular polygon structure. As shown in Fig. 4-10, the polygon is
inscribed inside a circle defined by the Component Width and Component Height. When a file is
specified, it takes precedence over the number of sides.

Polygon File
Sets the data file that contains the data points for the polygon. The data should be in the polygon
format discussed in Section B.E and is interpreted drawn such that a unit circle will be drawn on
the bounding circle shown in Fig. 4-10. If the data in the file should not be scaled, set the
Component Width and Component Height to 2. The polygon data must define a closed surface:
the first and last point in the data file must be the same.
It is possible to use a variable to define a polygon file: set the polygon file option to
$<variable> where <variable> is a variable created in the Symbol Table set equal to the name
of the desired polygon file.

RSoft CAD  68
Date Mode
Sets the load method for the polygon data. A Dynamic Load will load the file from disk every
time; a Static Load will embed the polygon data in the .ind file and not reference the file.

4.D.7. Mixing 3D Structure Types

The different structure types available in the CAD can be combined in the same file to produce
an extremely wide variety of 3D geometries. When combining structure types, the background
definition must be noted. The definition of the background is different for each 3D structure
type; for example the background for fiber and channel structures is uniform, while the
background for diffused and multilayer structures has a substrate and a cover. The background
used is always set by the 3D Structure Type set in the Global Settings window.

4.E. Profiles & Tapers

By default, most components have a step-index profile such that the refractive index inside the
component is the sum of the background index and the index difference and outside is the
background index. However, it is possible to change the profile so that other geometries such as
graded index profiles, etc, can be achieved. Please see Section 6.A for a complete discussion of
profiles.
Tapers allow a segment’s properties to vary between the two vertices. The segment’s width,
height, index, and position cane be tapered independently. Tapers are defined in the top region of
the segment properties dialog. Please see Section 6.B for a complete discussion of tapers.

4.F. Segment Types (z-Axis, Seg-Axis, Extended)

There are three forms of the general segment, each with subtle but important differences. Each
has a different interpretation of the width/height specified by the user, how the segment
terminates, how tapers are calculated, and how segment profiles are interpreted.

RSoft CAD  69
 Figure 4-12: 2D Illustration of segment types: a)z-Axis b) Segment-Axis/Extended.

The type of segment is controlled by the Seg Orientation option in the segment properties
dialog.

Type Width/Height Endfaces Tapers/Profiles Rotation

z-Axis Measured Flush with z axis Interpreted N/A
relative to z axis relative to z axis
Seg-Axis Width relative to Perpendicular Some tapers relative Rotation in
segment axis, to segment axis to segment axis, XZ plane
height relative to profiles relative to z
y axis axis
Extended Width and Perpendicular Tapers and profiles Arbitrary
height relative to to segment axis relative to segment rotation possible
segment axis axis
Some notes:
• Extended segments are used by default for new segments added to a design.
• When using curved seg-axis or extended segments the resolution at which the curves are
resolved can be set via the Import/Export Resolution in the CAD Preferences dialog
described in Section 6.E.
• Extended segments can easily be added using the Extended Segments drawing mode in the
left CAD toolbar. See Section 4.G for more information about these built-in structures.
• Extended segments can be twisted along the segment axis using the X’Y’ Rotation options in
the Additional Segment Properties dialog as described in Section 6.K.

RSoft CAD  70
• Extended segments can be rotated using the Reference Phi, Reference Theta, and Reference
Psi options in the Additional Segment Properties dialog described in Section 6.K.

Which Type of Segment Should I Use?

Seg-axis or extended segments give more control over segment geometry and are generally
preferred. Note that it is possible to produce mismatched width or gaps between components that
are not properly matched. Contact RSoft for recommendations for an appropriate solution to a
specific problem.

4.G. Drawing Common Extended Segments

Common extended segments can easily be drawn using the Extended Segments mode in the left
CAD toolbar (Section 3.A.2). Since all segments are forms of a generalized segment, these
segments can easily be modified to produce other customized segments. These extended
segments are:

Though pictured in the CAD with a circular cross-section, these extended segments can have any
cross-section (3D Structure Type).

• Straight
This mode produces a straight segment.
• Half-taper
This mode produces a segment with a tapered width and height. The ending width/height is
set equal to half the starting width/height; this can be changed by right-clicking on the
segment and changing the relevant properties.
• Full-taper
This mode produces a segment with a full tapered width and height where the ending
width/height is set to 0. This can be used to make cones, triangles, etc.
• Arc
This mode produces 90 degree arc segment. The radius and angular extent of the arc can be
set in the Arc Data dialog which can be opened by clicking Arc Data… in the Segment
Properties dialog.
• Toroid
This mode produces a toroid segment. The radius can be set in the Arc Data dialog which can
be opened by clicking Arc Data… in the Segment Properties dialog. The Arc Reference
setting in that dialog is set to Center so that the position of the starting vertex controls the

RSoft CAD  71
 center of the toroid. Depending on which plane the toroid is drawn, the reference angles
(Section 6.K.2) will be set to achieve the desired orientation. These can be changed to any
value to orient the toroid along any angle.
• Helix
This mode produces a helix segment. The radius can be set in the Arc Data dialog which can
be opened by clicking Arc Data… in the Segment Properties dialog. The Arc Reference
setting in that dialog is set to Center so that the position of the starting vertex controls the
center of the helix. Depending on which plane the toroid is drawn, the reference angles
(Section 6.K.2) will be set to achieve the desired orientation. These can be changed to any
value to orient the toroid along any angle. The default helix has 3 turns, this is controls by the
Start Angle and can be changed by editing the Arc Initial Angle and Arc Final Angle from its
default values of 0 and 3*360.

RSoft CAD  72
 Figure 4-13: Examples of common extended segments: a) helix segment with fiber structure
type, toroid segment with fiber structure type, c) straight segment with polygon cross-section
(hexagon) with width/height taper, d) straight segment polygon cross-section (data-file) with
‘twist’ taper.

RSoft CAD  73
5

Other Component Types

Chapter 4 described the segment component type. This chapter describes other component types.

5.A. General Overview

In additional to the segment component type, which was described in Chapter 4, the CAD
contains several other component types which allow different geometries to be easily created
such as lenses, circles, polygons, circuit references, simulation regions, and time monitors. Keep
the following in mind when using these component types:
• These component types have one vertex which is specified in exactly the same way as a
segment. Rather than repeat the documentation here, see the relevant section in Section
4.B.1.
• These components have only one vertex and so tapers cannot be used; some of these
components can, however, have a 3D Structure Type and Index Profile Type defined.

5.B. Lenses
Lenses are segments that are defined by two radii, a center thickness, and a width. Note that
while lenses can be to create circles, cylinders, spheres, and rings, the circle component is easier
for that purpose.

RSoft CAD  74
 Figure 5-1: A schematic of a lens structure. Two radii, R1 and R2, as well as a center
thickness and lens width are shown. In 3D, this structure can be cylinder-like or sphere-like.

Lenses can be added when the Lens Mode button in the left toolbar of the CAD has been
selected.

Figure 5-2: Properties dialog for Lens component.

Vertex Settings
The left half of the lens properties dialog is identical to the vertex portion of the segment
properties dialog (see Section 4.B). The vertex defines the center of the lens as illustrated in
Fig. 5-1 as well as the lens component's optical properties. The diameter of the lens is set by the
smaller of the Component Width or twice the smallest radius.

RSoft CAD  75
Lens-specific Settings
The right half of the properties dialog defines lens-specific parameters:

Front Radius and Back Radius

The radius of curvature of the front and back radii of the lens, labeled R1 and R2 respectively in
Fig. 5-1. The centers of the circles determining the front and back interfaces are defined
implicitly such that the circles intersect the front and back faces of the uniform central region.
The default value for these radii is inf and –inf respectively, corresponding to a infinite radius.
The sign convention for the radii is the usual one in optics, namely a positive front radius and a
negative back radius yields a convex lens.

Center Thickness
The thickness of the uniform central region, if any. This parameter defaults to 0.

Phi
This controls the phi angle of the lens measured from the Z axis. The default values are 0.
Typically the Phi angle is set locally, see Section 6.M for details about inheriting this setting
from a referenced component.

Lens Shape
This field controls the shape of the lens in 3D. See the next section for more details about its use.

Notes for 3D
In 3D, a lens can have either a cylindrical like-shape, or a sphere-like shape. The vertical extent
of the lens along the Y axis is controlled by the Component Height field.

Figure 5-3: Lenses in 3D, a) Spherical version of lens shown in Fig. 5-1, b) Cylindrical
version of lens shown in Fig. 5-1.

RSoft CAD  76
5.C. Circles
Circles are components that are defined by one radius, and can be used to create circles in 2D
and cylinders and spheres in 3D.

Figure 5-4: The circle component type.

Circles can be added when the Circle Mode button in the left toolbar of the CAD has been
selected.

Figure 5-5: Properties dialog for a circle component.

Vertex Settings
The left half of the circle properties dialog is identical to the vertex portion of the segment
properties dialog (see Section 4.B). The vertex defines the center of the circle as illustrated in
Fig. 5-4 as well as the circle component's optical properties.

RSoft CAD  77
Circle-specific Settings
The right half of the properties dialog defines circle-specific parameters:

Component Diameter
The diameter of the circle.

Component Height
The height of the circle when 3D Shape is set to Cylinder.

3D Shape
This field controls the shape of the lens in 3D. See the next section for more details about its use.

Common Circle Configurations

In 3D, a circle can have either a cylindrical like-shape, or a sphere-like shape. The vertical extent
of the circle along the Y axis is controlled by the Component Height field.

Cylinders & Spheres

Cylinders can be created in 3D by setting 3D Shape to Cylinder, spheres with a 3D Shape of
Sphere.

Figure 5-6: A sphere, cylinder, and ring made with circles.

This cylinder is along the Y axis; a cylinder along the Z axis can be made via the 3D fiber
structure type.

Rings
One way to make a ring in 2D or 3D is to overlapping two circles/cylinders. The inner segment
should have a Index Difference of 0 (or some other desired index value) and a higher Priority
than the outer segment. See Section 6.C for more details about overlapping segments. These
types of objects can also be made with seg-axis or extended segments (see Section 4.F).

RSoft CAD  78
5.D. Polygons
Polygons are segments which are defined via a series of arbitrary vertex points in the XZ plane.

Figure 5-7: Geometry of a polygon component defined by a series of arbitrary vertex points
in the XZ plane. For 3D structures, the structure is simply extruded in the Y direction so that
it has a height given by the Component Height option.

Polygons can be added when the Polygon Mode button in the left toolbar of the CAD has been
selected.

Figure 5-8: Properties dialog for Polygon component.

RSoft CAD  79
Vertex Settings
The left half of the Polygon Properties dialog is identical to the vertex portion of the segment
properties dialog (see Section 4.B). The vertex defines the origin of the polygon relative to the
design file as well as the polygon component's optical properties.

Polygon-specific Settings
The right half of the dialog defines the following polygon specific properties:

X Scale and Z Scale

A scale factor that can be used to scale the polygon data along the X and Z axes respectively. By
default these parameters are 1.

Angle
The angular rotation about the origin (defined by the vertex) to be applied to the polygon data.
The default origin and scale factors are chosen so that the polygon coordinates are interpreted as
absolute coordinates. However, if the polygon data represents a general shape to be used
repeatedly, the scale, angle, and vertex information can be used to size, rotate, and position the
component as required, without the user having to generate a different data file for each polygon
component to be added. Typically the Angle is set locally, see Section 6.M for details about
inheriting this setting from a referenced component.

Specifying the Polygon Data

The data file is selected by pressing the Data File... button and selecting the data file that contains
the point information. The Data Mode controls how the program loads the data: a Static Load
copies the contents of the file into the *.ind file, and a Dynamic Load loads the data every time
the fine is opened. Also, the data file can be stored in a variable as described in Appendix C.

Data Format
The polygon coordinate data must be imported from an ASCII data file containing two columns
defining the X and Z coordinates of each point:
X(0) Z(0)
… …
X(i-1) Z(i-1)
X(i) Z(i)
X(0) Z(0)

The polygon data must define a closed surface: the first and last point in the data file must be the
same. Also, after selecting the file, the coordinate information is stored locally in the *.ind file.

RSoft CAD  80
Notes for 3D
For 3D structures, the polygon is simply extruded or extended with a constant thickness along Y.
This thickness is given by the Component Height field. Polygons can have a 3D Structure Type
as well.

5.E. Registration Marks

Registration marks are useful for mask production and are defined by a width, height, and
thickness. This type of segment has no effect on simulations and is not included in the index
calculation.

Figure 5-9: The Registration Mark component type.

To add a registration mark to the circuit, select Options/Insert/Mark form the CAD menu. This
will open the properties dialog for the new mark.

Vertex Settings
The left half of the mark properties dialog is identical to the vertex portion of the segment
properties dialog (see Section 4.B). The Width/Length and Height fields set the width and height
of the mark. Typically these settings are set locally, see Section 6.M for details about inheriting
this setting from a referenced component.

Mark-specific Settings
The right half of the mark properties dialog specifies mark-specific information:

Mark Type
This field specifies the type of mark to be used.

RSoft CAD  81
Line Thickness
This field specifies the line thickness of the mark as illustrated in Fig, 5-8.

5.F. Circuit References, Simulation Regions,

Vertices, and Monitors
Circuit references, simulation regions, vertices, and time monitors are advanced components.
• Circuit references are a component that contains another design file. It allows one design file
to be nested in another resulting in a hierarchical design. See Section 6.F for more details.
• Simulation regions allow BeamPROP’s simulation parameters to locally be changed. See the
BeamPROP manual for more details.
• Monitors make measurements for FullWAVE, DiffractMOD, and ModePROP (see
appropriate simulation manual) and can be used to define ports for S-Matrix calculations (see
the Custom PDK Utility manual).
• Vertices are control points that can be added to a design, used mostly when defining circuit
references. See Section 6.F.7.

RSoft CAD  82
6

Advanced CAD Features

This chapter covers advanced RSoft CAD topics. It assumes basic familiarity with the RSoft user
interface. If any concept is not clear, please consult the previous chapters.

6.A. Profiles
A component’s profile function provides a crucial part of the refractive index of that component:

n(r ) = n0 + nf ( x, y, z )

where n0 is the Background Index and Δn is the local Index Difference, and f(x,y,z) is a profile
function. The profile function is set by the Index Profile Type option either globally in the
Global Settings window or locally via a component’s properties window. By default the profile
function is generally step index. There are several built-in profile or users can create a user-
defined profile.

6.A.1. Built-in Profile Types

The user profile type is set in the Index Profile Type option in the Component Properties dialog.
can be set All profile types are not available for all component types; see notes at end of this
section. In these equations, w is the component width and x is the distance from the center of the
component.
• Default

RSoft CAD  83
 The default profile is set by the global default in the Global Settings window and is based on
the Model Dimension and 3D Structure Type. See table below for the default profile for each
structure type.
• Inactive
An inactive component does not contribute to the refractive index distribution of the design
file and is useful for defining optical pathways for launch fields and pathway monitors.
• Step Index
A step index profile equals 1 inside the structure and 0 outside. This is the most common
profile.
• Gaussian
A Gaussian profile function is defined as:
x2
−
n ( x ) = n0 + ne a2

where a is defined to be half the Component Width.

• Diffused
A Diffused profile is implemented as a pseudo-effective index profile given by

 w  w 
1   2 + x  2 − x  
n( x) = n0 + n erf   + erf  
2  h   h 
     
where h represents the diffusion depth and is defined by setting the height variable in the
symbol table, or via the Component Height. For additional information and options, refer to
the discussion of 3D Structure Type of Diffused in Section 4.D.
• New Profile…or User #
These choices allow the user to specify the functional form of the index profile arbitrarily.
This feature is further described in Section 6.A.3.

Available Index Profile Types & Defaults

This table summarizes the available profiles for each structure type.

Inactive Step Index Gaussian Diffused User

2D Slab Yes Default Yes Yes Yes
Fiber Yes Default Yes Yes Yes
Channel Yes Default Yes Yes Yes

RSoft CAD  84
 Diffused Yes No No Default No
Rib/Ridge Yes Default No No Yes
Multilayer Yes Default No No Yes
Polygon Yes Default No No No

Notes:
• Using a Diffused profile type for a 2D Slab results in the diffused profile described above.
• Using a Diffused profile type for a Fiber or Channel is the same as a Diffused structure type.
• Using a Gaussian profile type for a Fiber or Channel results in a radial Gaussian profile.
• Using a user-defined profile for a Rib/Ridge or Multilayer results in a Height Profile (Section
6.A.2).

6.A.2. Types of User Profiles

User-defined profiles can be functions of any combination of x, y, and/or z resulting in many
possible equation types. This section describes the possible combinations.

Coordinate Normalization
In the following user profile definitions, the normalized coordinates x', y', and z' are defined as
2x 2y z
x' = , y' = , z'=
w h l
(1)
where x, y, and z are the actual positions along the segment, and w, h, and l are the local width,
height, and length of the segment. Therefore, the normalized coordinates are by default defined
to vary from -1 to 1 for x' and y', and from 0 to 1 for z'.
It is possible to interpret the coordinate and index information in the user-profile absolutely
without applying any coordinate or index normalization using the option described in Section
6.D.5. The data will still be translated to the position of the component.

Arbitrary 3D User Profile

This is the most generic user profile function and can be expressed as:
n ( x ', y ', z ' ) = n0 + nf ( x ', y ', z ' )

(2)

RSoft CAD  85
where n(x', y', z') is the desired index profile, n0 is the background index, and Δn is the local
index change. Both the channel and fiber 3D structure types accept this type of profile.

Arbitrary 2D User Profile

This is a 2D version of (2), and can be defined in either the X-Z plane for 2D structures or in the
X-Y plane for 3D structures:

n ( x ', y ') = n0 + nf ( x ', y ')

n ( x ', z ' ) = n0 + nf ( x ', z ' )

(3)
Both the channel and fiber 3D structure types accept this type of profile. Note that in a 2D
simulation, even though the profile lies in the XZ plane, the function must be entered as a
function of x and y where y represents the z axis and is normalized between 0 and 1.

Arbitrary 1D User Profile

This is a 1D version of (2):
n ( x ') = n0 + nf ( x ')

(4)
This profile type can be used for all 2D structures as well as in the case of a 3D fiber structure
type where it represents a radial profile.

1D/2D Height Profile

This profile type describes the height of a component as a function of x and/or z. This is defined
as:

H ( x ') = hf ( x ')
H ( x ', z ') = hf ( x ', z ')

(5)
where h is the local segment height and H() is the actual component height. This profile type is
used by both the rib/ridge and multilayer 3D structure types. This functionality enables the use of
a surface relief data file to define a surface in the RSoft CAD. The coatuf utility described in
Appendix E can be used to add coatings to these surfaces.

RSoft CAD  86
Other Profile Types
It is possible to achieve other profile types, including anisotropic user-profiles (1D/2D/3D) and
user profiles for epsilon (above profiles are for refractive index). Please contact
photonics_support@[Link] for more information.

6.A.3. Creating & Using User Profiles

User profiles are defined in the User Profile Editor dialog which can be opened by clicking the
Profiles… in a segment properties dialog or by via a button on the left CAD toolbar. Once a
profile has been defined, it can be assigned to a particular component via Index Profile Type in
the component properties window.

Figure 6-1: The User Profile Editor where User Profiles are defined.

The profile functions in Eqs. 2-5 can either be defined by either an expression or a data file.
Once a profile has been defined, it can be viewed in the normalized coordinate system by
clicking the Test button.

Expression
Set the Source Type option shown in Fig. 6-1 to Expression to define a profile via a
mathematical expression. When f(x', y', z') is defined via an expression, several options exist
which allow for further customization of the profile.
• See Appendix C for a complete discussion of the format and capabilities of RSoft
expressions.
• Enter all coordinate variables as x, y, and/or z. The program interprets x as x', etc.
• The expression is interpreted based on the coordinates used. For example, a non-radial 2D
profile must have at least one x and y coordinate in the expression; if only x is found the

RSoft CAD  87
 CAD will interpret it as a radial function. For cases like this, ‘missing’ coordinates can easily
be included as 0*y, etc such as:

f ( x ', y ') = g ( x ') + 0 * y '

(6)
• The number of points used to evaluate the expression can be set via the X', Y', and/or Z'
Points options. This can be useful to resolve rapidly changing functions or periodic
functions.
• By default, the variables x', y', and z' are defined only within the component. From (1), they
are therefore normalized so that they vary between -1 and 1 (x and y) or between 0 and 1 (z).
For the case of a 1D radial profile, they vary between 0 and 1. Outside of this domain, the
function f(x', y', z') is set to 0.
In some cases, it is useful to define the profile outside of this domain. This can be set by
changing the Min and/or Max values for X', Y', and/or Z'. These settings are in the normalized
coordinate system defined in (1).
• For many custom structures such as circles, cylinders, spheres, tori, etc, symmetry along the
z axis is required and it is useful for the z' coordinate to vary from -1 to 1 like the x' and y'
coordinates. This can be done by replacing z' with (2z'-1) in the user profile. This results in a
new definition of the normalized coordinate z' as z' = 2z/l.
• It is possible to interpret a user-defined profile absolutely, that is without refractive index or
coordinate normalization, by enabling the option discussed in Section 6.D.5. In this case, the
profile is interpreted absolutely but it is still translated to the component center.
• Finally, it is recommended that the function f(x', y', z') be normalized to vary between 0 and 1
when x' and y' vary between -1 and 1 and z' between 0 and 1. This additional restriction will
ensure that the maximum value of the index difference in the profile will correspond to the
local Index Difference of the component. While not required, this can simplify the definition
of a profile.

Data File
Set the Source Type option shown in Fig. 6-1 to Data File to define a profile via a data file,

See Appendix B for a discussion of data file formats.

When a data file is used to define the function f(x', y', z'), the variables x', y', and z' are given by
(1) and x, y, and z are defined over the range given in the data file.
It can be useful to remove the normalizations in (1) by setting the component height and width
equal to 2 and set Δn = 1 so that (2) reduces to
n( x ', y ', z ') = n0 + f ( x, y, z )

RSoft CAD  88
 (7)
Because it is usually not possible to set n0, the data contained in the file should be altered so that
the values are less than their actual amount by n0. For more information about generating a file
for use as an index profile, please see the example in the next section.

Note that the Min, Max, and X, Y, and Z Points options in the User Profile window are not used
when using a data file to define a profile. The Min/Max values can be used however. The data
file can be dynamically loaded at run time by entering the filename as $name and setting the
variable name to the actual filename of the profile.

6.A.4. Example User Profile

As an example, consider the parabolic index profile given by

1 − x '2 x' 1

f ( x ') = 
 0 x' 1

Using an Expression
To define this profile via an expression, simply enter the following formula in the appropriate
field of the dialog:
1 - x^2

Note that the variable name x in this expression represents the normalized variable x' which
ranges from -1 to +1 over the width of the segment. The domain of the profile can be changed
via the X Min/Max options shown in Fig 6-1. The range [-1,1] still maps to the width of the
segment, however. Also note that outside the domain of definition the function is set to zero.

Using a Data File

To define the above profile via a data file in the standard RSoft file format, first create a text file
in the standard 2D output file format for real data, as described in Appendix B. The following
data file defines this profile on a coarse grid containing 11 points:
/rn,a,b/nx0
11 -1 1 0 OUTPUT_REAL
0.00
0.36
0.64
0.84
0.96
1.00
0.96
0.84
0.64

RSoft CAD  89
 0.36
0.00

During a simulation, the CAD will interpolate between the points in the data file if necessary.
This example is illustrated in the example file
<rsoft_dir>\examples\BeamPROP\[Link]. More information about the file format for
user profile data files can be found in Appendix B.

6.B. Tapers
A taper function allows a segment’s width, height, index, and position to be varied or tapered
along the segment. The form of a taper function depends on the Segment Type (Section 4.F):
• z-Axis/Seg-Axis Segments: A taper function has the form
a ( z ) = a0 + ( a1 − a0 ) f ( z ' )

(8)

z − z0
z' =
z1 − z 0

(9)
where a(z) represents the segment property to be tapered, a0 and a1 are the parameter values
at the starting and ending vertex, f( z') is the taper function, and z' is a normalized coordinate
along z that varies from 0 to 1 along the Z extent of the segment. Z-position tapers are not
possible for these segment types; use an extended segment.
As an example, a width taper would be defined as:
w ( z ) = w0 + ( w1 − w0 ) f ( z ' )

(10)
where w(z) represents the segment width, w0 and w1 are the width at the starting and ending
vertex, and f( z') is the taper function.
• Extended Segments: A taper function have a more general form of the previous definition:
a ( z ) = a0 + ( a1 − a0 ) f ( t )

(11)
where a(z) represents the segment property to be tapered, a0 and a1 are the parameter values
at the starting and ending vertex, f( t) is the taper function, and t is an arbitrary parametric
coordinate that varies from 0 to 1. By using X, Y, and Z tapers, it is possible to define an

RSoft CAD  90
 arbitrary space curve. Note that while the definition of the taper in Eq. (11) are a function of
t, any user-defined tapers should be entered into the GUI as a function of z.

6.B.1. Component Options that Support Tapering

Several options in the RSoft CAD support tapering. This includes the segment properties (Fig 6-
2: Index Taper, Width Taper, Height Taper, X Pos Taper, Y Pos Taper, and Z Pos Taper), some
Extended Segment Properties (Section 6.K.2) and the Crystal Axes specifications (Section
6.K.3).

Fig 6-2: The portion of the Segment Properties dialog where segment tapers are set. Tapers
are also used elsewhere, including for Extended Segment Properties and Crystal Axes.

The Index Taper, Width Taper, and Height Taper are controlled by editable drop-down lists
which allow then to not only be set via the drop-down, but also by a Symbol Table expression.
See Section 3.J for details about how to use this feature. The built-in values for these taper types
are:

Taper List Option Built-In Value

None TAPER_NONE

Linear TAPER_LINEAR

Quadratic TAPER_QUADRATIC

Exponential TAPER_EXPONENTIAL

User 1 TAPER_USER_1

6.B.2. Built-In Taper Types

The choices for the taper function are:
• None

f ( z' ) = 0

RSoft CAD  91
• Linear

f ( z' ) = z'
• Quadratic

f ( z' ) = z'2
• Exponential

e Bz ' − 1
f ( z ') =
eB − 1
where B is defined as
B = sign( p0 − p1 )

where p0 and p1are the starting and ending values of the tapered parameter respectively.
• New Taper… and User #
These choices allow the user to specify the form of the taper function arbitrarily and are
described in Section 6.B.4.
• Arc (XZ)
This choice indicates that the segment path follows a curved path in the XZ plane and is
discussed in Section 6.B.3.

6.B.3. Built-In Arc Taper Types

The properties for an Arc (XZ) taper are set in the Arc Properties dialog which can be opened by
clicking the Arc Data… button in the segment properties dialog.

Figure 6-3: The Arc Properties dialog.

The possible Arc Type settings are:

• Free

RSoft CAD  92
 The Arc Radius, Arc Initial Angle, and Arc Final Angle of the circular arc can be arbitrarily
set. The program automatically calculates the ending vertex of the segment based on these
parameters and the starting vertex.
• Radius
The Arc Radius of the circular arc can be set and the program fits the arc to lie through the
segment’s starting and ending vertices.
• Initial Angle
The Arc Initial Angle of the circular arc can be set and the program fits the arc to lie through
the segment’s starting and ending vertices.
• Final Angle
The Arc Final Angle of the circular arc can be set and the program fits the arc to lie through
the segment’s starting and ending vertices.
• S-Bend
The program fits together two circular arcs to produce an s-bend through the segment’s
starting and ending vertices.
• Cosine S-Bend
The program uses a cosine taper function for the segment:

f ( z ') =
1
2
(1 − cos ( z ') )
• Raised Sine S-Bend
sin(2 z ')
f ( z ') = z '−
2

To recreate these tapers with a user-defined taper, the arguments for the trigonometric functions
in the above equations need to be rewritten in terms of degrees.

Arc Reference Point

The Arc Reference sets the reference point of the arc to be either at the starting vertex or at the
center of the arc. Putting the reference point at the center is recommended for arcs used to define
rings, toroids, and helixes so that they can be easily positioned, rotated and translated.

6.B.4. Creating & Using User Tapers

A user taper can be defined in the User Taper Editor window which can be opened by pressing
the Tapers… in a segment properties window or by via a button on the left CAD toolbar. Once a

RSoft CAD  93
taper has been defined, it can be assigned to a particular segment via the pull down menus in the
segment properties dialog.

Figure 6-4: The User Taper Editor where User Tapers are defined.

The taper function found in (8) can either be defined by either an expression or a data file.

Expression
To define a taper via a mathematical expression, set the Source Type option shown in Fig. 6-4 to
Expression. When f( z') is defined as an expression, several options exist which allow for further
customization of the taper.
• For a complete discussion of the format and capabilities of RSoft expressions, refer to
Appendix C.
• Enter all coordinate variables as x, y, and/or z. The program interprets x as x'.
• The number of points used to evaluate the expression can be set via the Z' Points options.
This can be useful to resolve rapidly changing functions or periodic functions.
• By default, the variable z' is defined only along the length of the segment. From (9), it is
therefore normalized so that it varies between 0 and 1.
In some cases, it is useful to define the taper outside of this domain. This can be set by
changing the Min and/or Max values for Z'. These settings are in the normalized coordinate
system defined in (1).
• Finally, it is recommended that the function f(z') be normalized to vary between 0 and 1 with
z' between 0 and 1. This additional restriction will ensure that values for the starting and
ending segment parameters retrain their original meaning. While not required, this can
simplify the definition of a taper.

RSoft CAD  94
Data File
To define a taper via a data file, set the Source Type option shown in Fig. 6-4 to Data File.
When a data file is used to define the function f(z'), the variable z' is given by (9), and is defined
over the range given in the data file. For more information on the use of a data file, see the
example in the next section. Also, the data file can be dynamically loaded at run time by entering
the filename as $name and setting the variable name to the actual filename of the profile.

Grating
This taper type is used to quickly define grating structures for the simulation tool GratingMOD.
Its use is discussed in detail in the GratingMOD manual.

6.B.5. An Example User Taper

As an example, consider the parabolic index taper given by

f ( x ') = 1 − z '2 0  z' 1

Using an Expression
To define this taper via an expression, simply enter the following formula in the appropriate field
of the dialog:
1 - z^2

Using a Data File

To define the above taper via a data file using the standard RSoft file format, first create a data
file in the standard 2D output file format for real data, as described in Appendix B.
The following data file defines this taper on a coarse grid containing only 11 points:
/rn,a,b/nx0
11 0 1 0 OUTPUT_REAL
0.00
0.36
0.64
0.84
0.96
1.00
0.96
0.84
0.64
0.36
0.00

Note that during a simulation, the CAD will interpolate between grid points if necessary.

RSoft CAD  95
6.C. Priority (When Components Overlap)
The creation of complicated structures in the CAD sometimes involves overlapping several
components.

It is crucial to understand how the CAD treats overlapping components by default and how to
override this behavior when necessary.

By default, when one or more components overlap, the program merges the components and uses
the index of the component with the higher absolute index difference at each grid point. Several
options exist to override this default behavior. They are:
• Combine Mode - This option indicates the method in which overlapping components are
evaluated.
• Merge Priority - This option sets the relative importance between components when a
Combine Mode of Merge is used.
Both of these settings can be made in a component’s properties window.

6.C.1. Combine Mode

The Combine Mode of a component has three settings:
• Default - This option automatically switches between Merge and Add based on the structure
type.
• Merge - This is the default setting for all structure types except diffused, and corresponds to
the use of a priority scheme in order to determine which index is used. See the next section
for more details on priority.
• Add - This is the default setting for diffused structures. This option allows the component to
act as a local index perturbation: the index is simply added to the background and any other
components. For 3D structures, the structure type of the component being added (i.e. that has
Combine Mode set to Add selected) is restricted to Fiber, Channel, or Diffused structure
types. For example, to create a combination of Multilayer and Diffused profiles, the user can
select Multilayer as the default structure in the Global Settings, and add Diffused components
using Combine Mode Add. However, the reverse is not currently possible.

6.C.2. Priority
Component priority is used when Combine Mode is set to Merge. The priority scheme used to
determine the index at a point where more than one component is defined is:
• If the segments have the same Merge Priority setting, the segment with the higher absolute
index relative to the background is used.

RSoft CAD  96
• If the segments have a different Merge Priority setting, the segment with the higher priority it
used.
By default, every component has a priority of 0, and so the first rule is used. However, this can
be overridden by changing the priority of one the components that overlap. The actual value of
the Merge Priority itself does not matter, only the relative priority of a component to other
components in a design.

This option is only available for 2D Step-Index and User-Defined profiles, and 3D Fiber,
Channel, or Multilayer geometries; if it is needed for other circumstances, please contact RSoft.

6.C.3. General Notes

• Using either Combine Mode Merge or Combine Mode Add to mix structures can, by design,
yield very different results. To ensure that the desired structure has been obtained, the user
should view the resulting index profile as described in Section 3.G.
• For multilayer structures, the desired geometry might not be realized when using overlapping
components. This can usually be fixed by setting the height of the multilayer components to
the exact height of the defined structure in assigned the layer table and/or using priority.

6.D. Global Index Generation Options

The CAD has several options for perturbing the index profile to achieve special effects, such as
simulating a curved space, truncating the index distribution at a boundary, or incorporating
imperfections into the index distribution, such as finite lithographic resolution or roughness.
These are controlled through the Global Index Generation Options dialog which can be accessed
via the Options/Index Generation... menu item.

Figure 6-5: The Global Index Generation Options window which contains several options to
perturb the index profile to achieve special effects.

RSoft CAD  97
6.D.1. Simulating Curved Segments to Large Angles
This feature is primarily used in the simulation tool BeamPROP, and uses a conformal index
mapping technique to transform a curved segment geometry into a straight segment geometry. It
can also be used to find bending modes in FemSIM. The Global Index Generation Options
window sets this feature globally for the entire simulation domain. This feature can also be
enabled locally for a segment to handle multiple bends or a series of bends and straight segments
via the analogous settings in the Advanced Component Properties dialog which can be opened
via the More… button in the Component Properties dialog. The method used is based on:
R. Baets and P. E. Lagasse, “Loss calculation and design of arbitrarily curved integrated-optic
waveguides”, J. Opt. Soc. Am., 73 (1983)

Motivation
Direct simulation of curved segments to large angles is difficult, even with the Padé-based wide-
angle capability available in the simulation tool BeamPROP. This is due to the fact that there are
several factors in addition to the obvious paraxiality limitations that must be considered. These
include the need for finer grid and step sizes as well as a large computational domain, both of
which can increase the computation time significantly. There is, however, an alternate approach
which is appropriate for modeling of a single bend or a series of bends (as in an s-bend for
example). This technique makes use of a coordinate transformation to map a curved segment,
onto a straight segment.

Implementation
The CAD implements the above-mentioned technique, which is accessed via the Simulated Bend
fields in the special effects dialog for the case of a single bend; multiple bends are discussed
below. When the Simulated Bend input field is set to Yes, the entire problem space is
transformed to effectively follow a circular arc, with a radius defined by the value in the numeric
input field labeled Simulated Bend Radius. This is done by transforming the refractive index
profile through multiplication by (1+x/R). This method has been demonstrated to be accurate
when w << R, where w is the width of the segment and R is the radius of curvature.
The Bend Origin controls how the radius is measured. The center of the bend is located at x = -R,
and so if the outer edge of a waveguide is to be located at a radius or R, that edge should be
positioned at x = 0 in the CAD. If the waveguide has a width of w and is centered at x = 0, then
setting Bend Origin to w/2 will effectively position the outer edge of the waveguide at R. In this
mode, the length of the waveguide becomes the arc length.
For diagnostic purposes it can be useful to see the perturbed index profile. This can be done by
setting the variable index_profile_transform to 1.
When using this feature globally via the Global Index Generation Options dialog in a design file
consisting of a single straight segment located at X=0, the Z axis then represents arc length.
When using this feature locally for each segment, as would be done to model an s-bend, one

RSoft CAD  98
would layout two straight segments in series, and assign the first segment a positive radius
indicating a bend curving to the left, while the second segment would be assigned an equal but
negative radius corresponding to curving to the right. If desired, one segment can be offset
laterally from the other to improve alignment of the curved segment modes, which are generally
displaced to the outside of the bend.

6.D.2. Sidewall Angle Control

The sidewalls of a channel or multilayer structure can be tilted using the Sidewall Angle option
in the Additional Component Properties dialog as discussed in Section 6.K. By default, however,
the width of the component is assumed to be constant at the bottom. This can be changed by
setting the Sidewall Angle Ref. option in the Global Index Generation Options window. A value
of 0 corresponds to the bottom and is the default, 1 corresponds to the top. Any value between 0
and 1 can be used.

6.D.3. Simulating the Effects of Finite Lithographic Resolution

The CAD has the ability to simulate the effects of finite lithographic resolution by automatically
rounding all component geometry to a grid, as if the circuit were drawn by filling in squares on a
piece of graph paper. To enable this feature, enter the desired grid resolution in the Litho
Resolution field of the Global Index Generation Options window. This feature is also accessed
via the litho_res variable in the symbol table. The default value of this variable is 0,
corresponding to perfect lithographic resolution.
This option only affects the geometry used in the field propagation. The perturbed geometry is
not used in the CAD display of the circuit, nor is it used in any of the analysis monitors, although
it influences their values through the field.
This feature deals with a systematic perturbation of the index profile; random perturbations, due
to processing-induced roughness for example, are discussed in the next section.

6.D.4. Simulating the Effects of Lithographic Roughness

The CAD has the capability to randomly perturb a component’s sidewalls in order to simulate
the effect of imperfections due to lithographic roughness. There are two parameters, Litho
Roughness and Correlation Length, which control this feature. They are specified in the Global
Index Generation Options dialog. This section discusses the definition of these parameters and
how the program uses them to calculate the random perturbations.
“Theory of Dielectric Optical Waveguides”, by Dietrich Marcuse, Chapter 4.6 in 2nd edition.

Theoretical Background
The two parameters Litho Roughness and Correlation Length are used to describe the random
perturbation produced by the CAD. To better visualize the definitions of these parameters, an

RSoft CAD  99
autocorrelation function will be introduced. These parameters will be defined in the context of
the autocorrelation function, and then related to the perturbation function.
An autocorrelation function is used to describe a stationary random process, and measures the
correlation, or similarity, between a perturbation function h(z) and the same perturbation function
offset a distance u in space. The autocorrelation function R(u) of a function h(z) is defined as:

R (u ) =  h ( z )h ( z − u ) dz
−

(11)
This function R(u) has its maximum value at u=0. As the value of u is increased, R(u) decreases.
For large values of u, R(u) approaches zero since there is no longer any correlation between the
function at the two sample points z and z+u.
Two parameters are defined to characterize the statistical variation:

Litho Roughness
This parameter can be defined as

 = R(0)
Since it is assumed that h(z) averages to 0 over large distances, the square of the lithographic
roughness is the variance of h(z). The lithographic roughness is also called the rms deviation.

Correlation Length
This is the length u=D at which the correlation function R(u) has decreased to a fraction of its
maximum value - usually to 1/e. It is a measure for the length over which the perturbations are
correlated to each other, and is therefore the scale length of the random perturbation. Within this
length, h(z) is not changing randomly.
In the context of the autocorrelation function, these two parameters hold intuitive meaning and
can easily be visualized. The following autocorrelation function is used as a convenient model
for many cases:
u
−
R (u ) =  e 2 D

(12)

The Perturbation Functions

In the RSoft CAD, a random perturbation of a component’s side walls is created in order to
model lithographic roughness. To do this, both the width and position of the component are
perturbed, which results in a non-symmetric perturbation. The width and position taken into
account during a simulation are:

RSoft CAD  100

 x( z ) = x0 ( z ) + x ( z ),
w( z ) = w0 ( z ) + w( z ).

(13)
The width and position are given as a function of w0 and x0, the unperturbed width and position,
and associated perturbations Δx(z) and Δw(z). Both the width and position perturbations are
calculated in the same manner. The program constructs the functions Δx(z) and Δw(z) such that
their resulting autocorrelation functions, as defined in (11), are approximated by (12). In this
way, a component with perturbed sidewalls characterized by the Litho Roughness and the
Correlation Length, is created.
This option only affects the geometry used during a simulation and not in the CAD window
display of the circuit. Also, a current limitation of this feature is that similar component in a
structure at a given z will have the same wall perturbation. This restriction can be relaxed in the
future based on user demand.

Additional Comments
This feature can be accessed via the variables litho_sigma and litho_corlen in the symbol
table. The default value of litho_sigma is 0, corresponding to no roughness and the default
value of litho_corlen is undefined. Also, this option only affects the geometry used in the field
propagation. The perturbed geometry is not used in the display of the circuit, nor is it used in any
of the analysis monitors, although it influences their values through the field.
This feature deals with a random perturbation of the index profile; systematic perturbations, due
to finite lithographic resolution for example, are discussed in the previous section. Additionally,
additional types of lithographic roughness can be created quite easily. The above formulation is
nothing more than a position taper and width taper with a built-in pseudo random function. This
effect can be recreated with any functions of the users choosing as described in Section 6.B, thus
allowing the user additional control over the type of roughness produced.

6.D.5. Interpreting User Profiles Absolutely

By default, user-defined profiles (see Section 6.A) normalize both the coordinate and refractive
index information to achieve the final profile. While this is useful for many cases, it is also
useful to interpret a user-defined profile absolutely. To enable this option, select the Absolute
index in User Profile option in the Global Index Generation Options window.
When this option is used, it is important to note that the profile is translated from the coordinates
defined in the expression/data file to the component center. The easiest way to directly interpret
the defined coordinates, simply position the component with the user-defined profile at the
origin.
This option sets several advanced sub-options which, if needed, can be controlled separately:

RSoft CAD  101

 Symbol Description
user_profile_scale_coords Set to 1 to control the scaling of the X and Y
coordinates.
user_profile_scale_coordsz Set to 1 to control the scaling of the Z
coordinate.
user_profile_offset_coords Set to 1 to control the offset of the X and Y
coordinates.
user_profile_offset_coordsz Set to 1 to control the offset of the Z coordinate
user_profile_scale_index Set to 1 to control the scaling of the index by the
index difference.
user_profile_offset_index Set to 1 to control the offset of the index by the
background index.

6.E. Geometry Import/Export

The RSoft CAD has the ability to import and export files in both RSoft or other CAD formats.
The supported formats are:
• Export: GDS-II, DXF
• Import: GDS-II, DXF, CIF, IND (Synopsys RSoft), TDR (Synopsys TCAD Sentaurus)

6.E.1. Export
The export options are found under the File/Export menu item.
Selecting GDS export will open the Export GDS dialog which sets the output File Name, the Cut
Direction and Cut Position, and Clipping options. The default clip position corresponds to the
simulation domain. Selecting DXF will save a .dxf mask file of the XZ structure with the same
name as the .ind file. This can also be done via the command line utilities described in Section
E.K.
See Section 6.E.3 for details about output to a specific layer and other important notes.

6.E.2. Import
There are three methods to import geometry:
• Convert a GDS-II, DXF, or CIF file into an RSoft .ind design file:

RSoft CAD  102

 The options found under the File/Import menu item provide a one-time conversion from a
GDS-II, DXF, or CIF mask file to an RSoft .ind design file. This can also be done via the
command line utilities described in Section E.K.
• Dynamically reference a GDS-II or IND using a Circuit Reference component:
This option embeds a child design file (.gds or .ind) in a parent .ind file via a Circuit
Reference. Where appropriate, a specific layer can be imported. See Section 6.F for more
details about Circuit References.
• Use Exact Geometry data in a TCAD Sentaurus TDR file in an RSoft .ind design file:
See the Multi-Physics Utility manual for details about importing TDR geometry data from a
Synopsys TCAD Sentaurus simulation.

6.E.3. Import/Export Settings and Important Notes

The primary options controlling related to import/export are contained in the Import/Export tab
of the CAD Preferences dialog.

Details for the rest of the CAD preferences can be found in Section 6.J.

Import/Export Resolution
Sets the resolution at which CAD objects are “fractured”, or converted to polygons in the target
CAD file. This resolution is defined as the maximum distance between any chord of the fractured
object and the true curve. This option is relevant for both import and export. This setting is used
when segment-axis segments are used in a simulation and also controls the “fracturing” of
curved surfaces. It affects both the CAD display and simulation results. See Section 4.F for
information about Segment Orientation.

Export Mask Layers

Sets the mask layer to be exported, and is used along with the Mask Layer option in the
Additional Component Properties window. See Section 6.K for details. When this option is set to
0, components are exported to layers that correspond to the Mask Layer for each component.
When this option is not set to 0, all components with a Mask Layer of 0 will be exported to the
value specified in the Export Mask Layers field.

Export Polygon Mode

This controls how the CAD exports polygon objects within the CAD. The Raster option causes
the CAD to break a polygon into smaller polygons so as to satisfy the 200-point GDS-II limit for
polygons. The Raw option exports polygons as they are, without breaking them up.

RSoft CAD  103

Notes:
It is important to note several aspects of the import/export process:
• When a mask file is imported into the RSoft CAD, all supported objects are converted to
RSoft polygons. Similarly, when the RSoft CAD exports a mask file, all RSoft CAD objects
are converted to polygons in the target file format. This avoids possible ambiguities since
polygons have a fairly consistent interpretation among different file formats and software
packages. As such, only 2D formats can be imported.
• RSoft CAD’s imports the following objects from a DXF file: LINE, SOLID, POLYLINE,
LWPOLYLINE, ARC, and CIRCLE. The following objects can be imported from a GDS-II
file: GDS_BOUNDARY, GDS_BOX.
• Setting the Mask Layer of a specific component to a negative number will exclude that
component from export.

6.F. Circuit References

Circuit references, or hierarchy, embed one design (the child) within another design (the parent).
This nesting can be multiple levels deep. Multiple file formats are supported:
• RSoft .ind files, symbols can be passed from the parent to the child for parametric control.
• GDS .gds mask files, a specific layer can be specified.
• Python .py scripts that generate an RSoft .ind file. See Section 10.A for information about
using Python scripts to create .ind files. Symbols can be passed from the parent to the child
for parametric control.
• TCAD Sentaurus .tdr files created by SDE or S-Process.
The figure below shows an example circuit reference shown with a shaded bounding box.
Elements in the child circuit are not directly editable but double clicking on the child circuit will
open the child file.

RSoft CAD  104

 Figure 6-6: Example Circuit Reference in the RSoft CAD. The ring resonator structure is
defined in a child circuit. The hatched blue box outlines the circuit reference.

The files can be embedded dynamically (remain as two files) or statically (combine two files into
one). Furthermore, this feature allows for child circuits to be periodically repeated in the parent
circuit allowing for dynamically sized arrays of objects.

6.F.1. Creating a Circuit Reference

To add a child circuit to a parent circuit, use the Circuit Reference button in the left toolbar of
the CAD. You will be asked to select the child file (.ind, .gds, or .py) from the current
working directory. Note that a special shortcut is created at the top of the directory listing so that
you can easily select from RSoft’s structure library, select either one of the built-in files or any
other file. Once the file is selected, the Circuit Reference properties dialog is shown.

RSoft CAD  105

 Figure 6-7: The Circuit Reference properties dialog box.

6.F.2. Setting Circuit Reference Properties

There are three main categories of properties: the circuit definition, the position, and the common
information.

Circuit Definition:
The circuit definition options are:

Embedded File or External File

This Embedded File or External File field (the name changes depending on whether the
Embedded option is selected) specifies the file for the child.
The syntax for the filename depends on the file type:
• RSoft .ind file:
Enter the name of the child .ind file. The Common Information described below can be used
to override the defaults in the child file, symbols/variables in the child file can be set from the
parent as described in Section 6.F.5.
• GDS file:
Enter the name of the child .gds file. The common Information described below can be used
to set the material, color, and profile of the components in the child file. See Section 6.F.8 for
more options.
• Python script:

RSoft CAD  106

 Enter the name of the child .py file. This file will be run using the version of Python
included with the RSoft tools to produce an .ind file. See Section 10.A for more details.
Note that symbols defined in the Local Symbols table that have names like ArgN are passed
to the .py script as arguments, see Section 6.F.6.
• TDR Boundary file:

The Import TDR option described in the Multi-Physics and TCAD Sentaurus Interface manual
can be used to automatically import TDR files.

Enter the name of the child .tdr file. The file will be dynamically referenced. For 2D TDR
files, the syntax example_bnd.tdr,xy can be used to orient the TDR file in the XY plane.

Embedded
This option selects if the child file (.ind only) is stored within the parent file. You can browse
the list of files currently stored in the parent file in the pulldown list in the Circuit Reference
dialog, or in the Embedded Circuits folder in the tree view on the right of the RSoft CAD. To
save a copy of an embedded file as a separate file, double-click on the circuit reference
component to open the child file and choose File/Save As from the CAD menu.

XZ Rotation Angle
Sets the angle [degrees] of the child circuit in the XZ plane of the parent circuit, and is measured
from the positive Z axis. Typically this angle is set locally, see Section 6.M for details about
inheriting this setting from a referenced component.

Lattice Info…
This control allows for the creation of dynamically sized arrays and is discussed in Section 6.G.

Position & Common Information:

The right hand side of the dialog defines the Starting Vertex. The starting vertex is defined as the
position of the child circuit’s origin (X=Y=Z=0) in the parent circuit. The specification of the
starting vertex is similar to the standard component dialog, and is not repeated here. Note that
when using the dynamic array option, it is recommended to center the unit cell in the child circuit
to simplify the lattice position in the parent circuit. Note that it also possible to add additional
vertices to the circuit reference, useful for connecting components to ports or other locations of
interest; see Section 6.F.7.
The left hand side of the dialog defines the Common Information. This includes the 3D Structure
Type, Index Profile Type, Combine Mode, Merge Priority, and Display Color. These options
override the global default settings in the child file. Note:

RSoft CAD  107

• Only components that have default values for these options will be overridden. If a
component in the child file has one of the above options specifically set (i.e. 3D Structure
Type set to Channel), then it will not be overridden by these settings.

• Note that Merge Priority will only be used if all components in the child circuit have a 0
priority (i.e. the default).

6.F.3. Editing a Circuit Reference

Elements in the child circuit are not directly editable, double click on a child circuit to open its
child file and edit if appropriate. Child circuits are drawn in the parent circuit with a shaded
bounding box, the display properties of this box can be set by the Circuit Reference Box
preference described in Section 6.J. Child circuits that are defined by scripts will open the script
in a default text editor, set the environment variable RSOFT_EDITOR to the path of the text editor
of your choice.

6.F.4. Controlling Variables in Child File From Parent File

Hierarchy allows for variables to be inherited by the child circuit from the parent circuit. In
order to do this, the Circuit Reference dialog has two symbol tables: a Local Symbol table and a
Global Symbol table.

Global Symbols…
This button opens the symbol table in the parent file, and behaves as expected. This is the same
symbol table that is accessed via the Edit Symbols button on the left menu in the CAD interface.

Local Symbols…
This button pulls up a special type of symbol table, in which the variables on the left will
override values in the child, while the expressions on the right are evaluated in the context of the
parent’s symbol table. By default, expressions such as "width = width" are defined, which
indicate that the child’s "width" variable (on the left) is assigned the value of the parent’s
"width" variable (on the right). This allows the variables in the parent circuit to be used to
control variables in the child circuit. Local symbols can be passed to .py child script files, see
Section 6.F.6.

Symbols from the child circuit can be exported to this special symbol table using the Hierarchy
Export options described in Section 6.F.5. This allows easy control of more common aspects of
the child file design to automatically ‘bubble’ up into the parent file. The square bracketed
expression format [value] is used to denote that the value from the child file should be used in
the parent file. The value between the square brackets will be dynamically updated to show the
current value. See Appendix C for more details on symbol table expressions.

RSoft CAD  108

 Figure 6-8: Local symbol table showing assignment of the values of parent circuit’s variables
(right column) to children circuit’s variables (left column).

Which Settings Take Precedence?

When a child file is used in a parent file, the following rules are used to determine which settings
take precedence:
• Any symbols defined in the Local Symbols table described above will be used in the child
file in the final layout.
• In general, any parameters that are set to default values in the child file are overridden by the
settings in the parent file. One example of this is the 3D Structure Type. The most typical
configuration (which is the default) of the 3D Structure Type is that it is set in the Global
Settings, and individual components are set to this default. If this default is different in the
child file and the parent file, than the parent file’s value is used in the final layout. For
example, if the 3D Structure Type is set to Fiber in the child file, and a component’s local 3D
Structure Type is set to Default in the child file, and the 3D Structure Type is set to Channel
in the parent file, the component will be a channel in the final layout.
• Some parameters in the child file are always overridden by the settings in the parent file out
of necessity. These parameters include the Model Dimension.
• Some properties of the design file must be defined in both the parent and child file, including
materials, tapers, and profiles.

6.F.5. Hierarchy Exports

The Local Symbol table provides an important link between the child and parent files when
using hierarchy. Hierarchy exports provides an important capability to this feature, the ability for
child files to automatically export or broadcast a list of symbols which will be automatically
added to the Local Symbol table. This allows important control variables to be broadcast from
the child file to the parent file to facilitate control and allow child files to more easily be reused
in different circumstances.

RSoft CAD  109

Selecting Symbols to be Exported
The list of symbols that are exported are set at the child circuit level. Open the child circuit in the
RSoft CAD and choose the Edit/Edit Hierarchy Exports… option from the RSoft CAD menu.

Figure 6-9: The Edit Hierarchy Exports dialog.

Once entered in the Edit Hierarchy Exports dialog, symbols are then exported to a parent file
when the child file is used via hierarchy. Multiple symbols can be entered on separate lines as
shown in the above figure.

Using Values in Child File for Exported Symbols

In some cases, it can be advantageous to set the value of an exported symbol to the value in the
child file. The square bracket notation [xx] can be used in the Local Symbols table to indicate
that the value in the child file should be used in the parent file.
Consider the example Local Symbol table shown below. The symbols Color and Material have
been exported from the child to parent file, but, as configured, we wish for the default color and
Material to be used in the child file. The value in inside the square brackets is automatically
updated based on the value set in the child file. In this case, the symbol Color is set to -1 in the
child file which indicates that the default color should be used and the symbol Material is set to
“” which indicates that the Material Properties setting should be set to Locally Defined. Note
that both of these symbols are used in the child file to set the Color and Material Properties
setting respectively.

RSoft CAD  110

 Figure 6-10: Illustration of the square bracket notation.

6.F.6. Creating Python Functions

While it is possible to implement a script that will parse arguments, it is recommended to use the
rsoft_layout() function. The arguments of this function will automatically be added to the
Local Symbol table with the ArgX prefix (i.e. Arg1_name). The function should return a
RSoftCircuit object.

Automatically Parse Script Arguments:

The rsoft_layout() function can be used to automate argument parsing:
• Define the function rsoft_layout() which contains all the code necessary to produce the
design file (.ind for a Python script). If present, this function will be automatically executed
by the RSoft CAD when running the script for each circuit reference block. Note that the
function must compile for this to work.
• The rsoft_layout() function should be defined with all arguments that will be passed to
the script along with default values.
• Symbols in the Local Symbols table for the circuit reference that begin with the prefix ArgX
will be automatically created, and then passed to the corresponding arguments in the
rsoft_layout() function definition along with the design file name. For example, the
following rsoft_layout() Python function declaration has two arguments:
def rsoft_layout(NumRings=int(10),
data_file=’’):

The arguments will be automatically set by Local Symbols that start with Arg1 and Arg2, say
symbols Arg1_N and Arg2_file. Note that if using a string argument as a file name input to
the Python API, the file must be present for the script to compile. In these cases it is safer to

RSoft CAD  111

 use an if statement to check if the file exists (or is not set to ‘’ as shown above) before using
the Python API.
• As a note, it is possible to use any function name, so long as you include a comment after the
function declaration with rsoft_layout, for example:
def any_function(NumRings=int(10), //rsoft_layout
data_file=’’):

Returning a Circuit Object in Python:

The Python script should return a RSoftCircuit object like:
def rsoft_layout(NumRings=int(10),
data_file=’’):

c = RSoftCircuit()
//script to create circuit c
return c

Example Python Scripts:

Further examples can be found here:
<rsoft_dir>\examples\RSoftCAD\scripting\

For Python, see Appendix H for Python support details and Section 10.A for a detailed
description of these examples.

Testing Scripts Outside Circuit Reference Blocks:

Testing scripts outside circuit reference blocks can be done by calling the rsoft_layout()
function to execute the script:
• You can add a line after the rsoft_layout() definition as follows which will save a copy of
the circuit as [Link] when the parent design file is opened in the RSoft CAD. These lines
can be commented out when not needed.
def rsoft_layout(NumRings=int(10),
data_file=’’):

c = RSoftCircuit()
//script to create circuit c
return c
c2 = rsoft_layout(10,’data_file’)
[Link](‘[Link]’)

• You can create a small wrapper script that imports the layout script, saves the
rsoft_layout() as a circuit and writes a .ind file as described above.

In both cases, the script can also be run from a command line using the command:
rspython <script>

RSoft CAD  112

6.F.7. Adding Vertices to a Circuit Reference
Vertices are a core part of the RSoft CAD and allow components to be easily connected. Circuit
references, by default, have one vertex at the local origin of the embedded design file. Some
applications benefit from multiple vertices such as multi-port devices where each port can be
thought of as a vertex. This allows multi-port objects to be easily connected.
To add a vertex, right-click on the Components folder in the component tree and select New
Vertex. The vertex can be positioned as needed in the child file. Once used as a circuit reference
in the parent, additional components can be connected to this vertex as usual. Note that vertex 0
of the circuit reference is the default vertex, it is the local origin of the embedded file. There can
be any number of additional vertices, these extra vertices start at vertex 1 and up. Also note that
vertices must be sequentially numbered starting at 1. This can be automated using the pv option
described below.

6.F.8. Dynamically Importing GDS files via Circuit References

GDS files can be dynamically imported by setting External File to the .gds file. There are
several options that control how the GDS data is imported (layers, cells, etc), and then several
options that control the non-transverse geometry (position, height, material, etc.).

GDS File Options:

At a minimum, the External File should be set to the name of the .gds file. For example:
[Link]

Several optional options can be added which control how the GDS data is imported:
[Link]:<layer>/<data>,pv#,n<cell>

where <layer> is the layer number, <data> is the datatype, the pv# option automatically adds
vertices, and the n option selects a particular named cell. Notes:
• Any option supported by the gds2ind utility can be used, see the Mask Conversion Utilities
section.
• If no layer/data numbers are provided, all layers in the GDS file will be used.
• If no datatype is set, a datatype of 0 is used.
• The pv# option automatically detects ports/vertices and adds vertex objects. See section
below for more details.
• The n option selects a particular named cell from the GDS. If not specified, the entire GDS is
imported.
• (Not shown above) By default, the .gds file is positioned in the XZ plane. The xy option is a
beta option to position it in the XY plane. In this XY mode, the ‘Z’ length can be set by the

RSoft CAD  113

 Length symbol in the Local Symbols table. This is a beta feature, please contact us with
questions or issues.

GDS File Geometry Options:

The GDS file sets the geometry in the transverse plane, the rest of the geometry is set by the
vertical position, the layer height, and the layer material. These can be set via the Y position (Z
when the xy option is used), the Component Height, and the Component Material.

GDS Port/Vertex Detection:

The pv# option enables the automatic detection of vertices. This is very useful when creating
devices that will be connected to other devices, or to place port monitors for S-Matrix calculation
with the Custom PDK Utility.
• Circuit reference objects have 1 vertex (vertex 0, the main X,Y,Z position), and so this option
will add additional vertices. For example, a structure with 3 ‘ports’ will have 4 vertices:
vertex 0 is at the global X,Y,Z position of the component, and vertices 1 through 3 will be at
the detected port positions,
• The number argument will translate the structure so that numbered vertex is at the origin. A
value of 0 does not move the structure.
• By default, the vertex detection will exclude ‘ports’ with a width > 0.999 μm. This limit can
be changed using the pw# option.
• By default, vertices are found on all 4 sides of the structure. This can be changed using the ps
option. The possible sides are l, r, t, b for left, right, top, and bottom. For example, setting
pslrt will look for ports on the left, right, and top sides, excluding the bottom.

• Note that this option only adds vertices, use the Attach Port Monitors option to add port
monitors, if needed. See the Custom PDK Utility manual for more details.

Example:
To dynamically load layer 88/4 from file [Link], enabling automatic vertex detection
for ports on the left and right side, with a maximum width of 1.5 μm, use this setting:
[Link]/4,pv0,pw1.5,pslr

6.G. Dynamically-Sized Arrays

The Circuit Reference feature allows child circuits (the unit cell) to be periodically repeated to
create a lattice.

RSoft CAD  114

6.G.1. Creating a Dynamic Array
There are two ways to create a dynamically-sized array:
• Using the Dynamic Array button
The Dynamic Array button in the left CAD toolbar provides a convenient way to create a
dynamic array. To use this feature, click the Dynamic Array button and then click in the
CAD drawing pane to add the array. You will be prompted to select the .ind design file that
represents the unit cell. You can pick one of the built-in files with simple geometries, or
select another file. After you have selected the file, the RSoft CAD will automatically create
a hierarchy component with the desired unit cell embedded in the parent .ind file. You can
then edit the Lattice Properties as needed.
• Converting a Circuit Reference (Hierarchy block) into a Dynamic Array
You can convert any circuit reference component into a dynamic array be opening the
properties dialog for the circuit reference, clicking the Lattice Info… button, and setting the
options described below. It is recommended to center the unit cell in the child circuit to
simplify the lattice position in the parent circuit.

When using dynamic arrays, be sure to use the Local Symbols… option described above to
logically link options from the parent file to the child file.

6.G.2. Setting Lattice Properties

The lattice properties are set in the Lattice Properties dialog box which can be opened by
clicking the Lattice Info… button in the Circuit Reference Properties dialog box.

Figure 6-11: The Lattice Properties dialog box.

RSoft CAD  115

The lattice properties are:

Lattice Type
Sets the type of lattice including Cubic, Hexagonal (HCP), BCC, FCC, and Diamond lattices.

Bounding Shape
Sets the shape of the bounding box around the lattice. The bounding shape can be Rectangular or
Hexagonal.

Period
Sets the period of the lattice.

Lattice Size
Sets the number of periods along the X, Y, and/or Z directions.

Defect Size
Sets the size of a defect measured in periods) at the center of the lattice along the X, Y, and/or Z
directions.

Scale Factor
Scales the entire lattice in Cartesian coordinates.

6.H. Defining Pathways

Pathways serve two main purposes within BeamPROP, FullWAVE, and FemSIM:
• To serve as paths through the index structure for BeamPROP and FemSIM monitors.
• To specify which component to launch the incident field into.
This section will cover the creation of a pathway in the CAD. Note that the use of BeamPROP
and FemSIM monitors are covered in the BeamPROP and FemSIM manuals respectively, and
the definition of a launch field is discussed in the BeamPROP and the FullWAVE manuals.

6.H.1. Creating a Pathway

To define an optical pathway, click on the Edit Pathways icon in the left toolbar. The program
switches to Pathway Edit Mode as shown in Fig. 6-13. The circuit display remains the same, but
the toolbar has changed and the status line indicates the new mode.
To define a new pathway, click on the New button in the toolbar and the pathway count indicated
at the top of the toolbar increments by one. Then select the desired components using the left

RSoft CAD  116

mouse button; note each selected component is highlighted in green. To deselect a component,
simply click on it again. To select multiple components in one operation, drag a selection box
around the components. To define additional pathways, simply click on New and repeat the
above procedure. When all desired pathways have been defined, click on OK to accept the
definitions and return to the main layout window mode. Note that when in 4P mode, components
can only be added to or removed from a pathway in the pane that was active before entering
Pathway Mode.
At any time the user can add new pathways by entering Pathway Edit Mode and performing the
above steps. To modify or delete existing pathway definitions, click on the << and >> buttons to
select the desired pathway, and select or deselect components to modify the definition, or click
on the Delete button to delete the pathway definition entirely.

Figure 6-12: Edit Pathway Mode, which is used to define optical pathways for analysis. This
example shows a directional coupler with the left pathway selected.

6.H.2. Using Pathways

As stated before, pathways have two uses:
• For BeamPROP and FemSIM monitors: Each pathway is essentially a set of components
grouped together, usually so that they form a continuous path from one part of the device to
another. For example, a directional coupler may consist of two s-bend input arms, two
parallel straight sections, and two s-bend output arms. In this case the left set of components
forms one natural optical pathway, while the right set forms another. Note that it is possible
to define pathways which include inactive components. An inactive component as a Profile

RSoft CAD  117

 Type set to Inactive , as is discussed in Section 6.A. This can be useful if it is necessary to
monitor the power in a region which does not contain or is otherwise different from a true
component region. The use of BeamPROP and FemSIM monitors is discussed in the
BeamPROP and FemSIM manuals.
For BeamPROP, pathways are only clearly defined when the components within that
pathway do not overlap along the Z axis. If a pathway does have multiple components at a
particular Z position, BeamPROP will give a warning during the simulation when it reaches
that Z position.
• For the Launch Field: The use of a pathway for a launch field is discussed in the
BeamPROP, FullWAVE, and ModePROP manuals.

6.I. Importing Symbols or Components from

Another Design File
It may be necessary or desirable to import symbols or components from one design file (*.ind
file) into another. In this section we describe several CAD features for assisting with each of
these tasks.

6.I.1. Importing Symbols

The RSoft CAD provides several features to facilitate the import of symbols between data files.
This allows, for example:
• A simple one-time import of symbols from one file to another.
• A pre-selected list of important symbols can be automatically imported from a ‘primary’
design file into a ‘secondary’ design file every time the ‘secondary’ file is opened. This helps
keep multiple files synchronized when running simulation scripts.
• Selected symbols can be exported from a child design file to allow easy control from the
parent design file. This simplifies the use of circuit references/hierarchy, see Section 6.F.4.

Methods of Import

By default all symbols are imported, see next section for instructions on importing specific
symbols.

• One-Time Import
Symbols can be imported from one design file to another via the Options/Import-Symbols…
menu item. This will copy all of the symbols in the selected *.ind file into the current file.
Symbols with the same name will be overwritten by symbols from the imported file.

RSoft CAD  118

• Via Import Symbol File
The symbol import_symbol_file is used to automatically load from a ‘master’ design file.
Its use is best illustrated by example: Consider two *.ind files [Link] and [Link].
Setting import_symbol_file = [Link] in the [Link] file will automatically load
symbols from [Link] into [Link] every time the [Link] file is loaded. This
feature can be helpful to keep multiple files synchronized when running simulation scripts.
When used, changes to variable values in the primary file are automatically propagated to the
secondary files every time the secondary files are loaded. The automatic option can be
disabled by setting import_symbol_file_auto to 0 even if the import_symbol_file
option is set.

Specifying Symbols to Import

By default, all symbols are imported when using the options described above. The import_list
variable can be used to specify the symbols that should be imported. This variable can be used in
two ways:
• Comma-Separated List
This mode accepts a simple list of comma-separated variables to be imported. Variables not
mentioned in the list will not be imported. For example, defining:
import_list = width,variable1,variable2,variable3

will import only the variables width, variable1, variable2, and variable3.
• Data File
This mode accepts the name of a data file and is useful when the list of variables to import is
long. The ‘@’ character should be used before the data file. For example, defining:
import_list = @symbols_to_import.txt

will import the variables listed in the data file symbols_to_import.txt. If this data file
contains the following:
width
variable1
variable2
variable3

will import only the variables width, variable1, variable2, and variable3.

Advanced Symbol Import Options

By default, an unset symbol will use a default value. The sync mode can be used to replicate this
behavior. The sync mode will:
• If the symbol is defined in both files or only in the primary file, then it will be copied from
the primary to the secondary file.

RSoft CAD  119

• If the symbol is not defined in the primary file, it will remove it from the secondary file.
To use the sync mode, add the keyword ‘:sync” after the variable name. So, for example, when
using the comma-separated list:
import_list = width,variable1:sync,variable2,variable3

or when using a data file:

width
variable1:sync
variable2
variable3

the symbol variable1 will be interpreted in sync mode.

6.I.2. Importing Components

Components can be imported from one design file into the current design file by first opening the
existing *.ind file, selecting the desired components, and copying them to the clipboard. Then
switch back to the current layout via the Window menu, and paste the components into the
current layout. If the selection contains formulas, the variables used in those formulas will need
to be added to the symbol table of the current layout. For this feature to be useful, the layout of
the device in the existing file should be well-designed to make use of references and other
features to allow the device to be smoothly connected to the current layout without minimal
additional editing.

6.J. CAD Preferences

The CAD Preferences window controls several options within the CAD and can be opened via
the Options/Preferences menu item. Changes to some of these options require the CAD to be
restarted.

Display Preferences
Draw Interior, Inactive Color, Pathway Multi-Pane View
Component Color Color, Rubberband Color Enables Multi-Pane mode and
Display/color of Color of inactive components, sets initial state: Section 3.E.2
component interior pathways, and rubberbands.

Draw Edge, Show Simulation Domain, Show 3D Pane

Edge Color Show Launch Fields Enable the 3D pane for Multi-
Display/color of Display simulation domain, Pane mode
component edges launch field (if applicable)

Draw Center Line, Direction Indicators Enable Scroll Bars

RSoft CAD  120

 Center Color Display direction indicators for Enable scroll bars in
Display/color of launch fields and monitors each drawing pane
component centerline
Use Material Colors, Show Start with Tree Expanded Circuit Reference Box
Default Comp as Comp Starts component tree with Sets display properties of
Color, Map File folders expanded circuit reference boxes
Sets the material display
color options, See Section
8.A.6.
Component Tree
Enables component tree and
sets initial state.

Editing Preferences
Attach on Edit Max Level Undo Zoom Factor (>1.0)
Enables the snap-to feature Number of undo levels Percentage used for
when editing components zooming in or out.

Attach Tolerance, Automatically Regrid Zoom/Full Affects All Panes

Select Tolerance after Zoom Zoom/Full is be done for all
The pixel distance for ‘snap-to’ Display grid is viewing panes at once
and component selection recalculated after zoom

Paste Offset Warn when automatically Zoom/Full Uses

resizing grid Simulation Domain
Offset used for pasting objects
from clipboard. CAD will warn when Zoom/Full uses domain rather
display grid has been resized than structure size

Show All Draw Tools Zoom/Full Border Allow Mixed Types,

Display all useful drawing Percentage of space around Allow Mixed Properties
modes in each pane in left zoom border when using Controls group editing
CAD toolbar Zoom/Full; must be between 0 of mixed component types
and 1

Units Preferences
The Index Input Mode sets the method that users define refractive indices in the CAD. By
default, indices are defined via index differences relative to the background index; however in
some cases it can be useful to define them absolutely. When this option is changed in an existing
file, the CAD will attempt to convert all components to the new mode. Since the CAD uses the
index difference method internally it is usually preferred.

RSoft CAD  121

Simulation Preferences
Static Color Shades Dynamic Color Shades Color Scale
Default number of static color Number of dynamic Default color scale to be used;
shades; can be set locally as color shades can be set locally as well
well

Slice Output Use Single Precision MPI Display Whole Domain

Default state for BeamPROP’s Default state for data type: see Default state for FullWAVE
slice output: see BeamPROP FullWAVE, DiffractMOD, Whole Display option: see
manual ModePROP, or FemSIM
FullWAVE
manual

Use Perpendicular Launch Enable Auto Clustering

Width Convention Default state for FullWAVE
Set launch width equal clustering: see FullWAVE
to component width manual

Limit Preferences
These preferences set the limits in the CAD for various items. It is not recommended to increase
these values unless needed.

Import/Export Preferences
The top options are discussed in Section 6.E, the bottom option sets the Group Library Path for
where materials are stored (Section 8.A.1). The choice of PDK used by the pdkinfo() function
described in Section C.F.10 is also chosen here. Note the following:
• Custom PDK parameters can be defined in the <rsoft_dir>\products\rsoftcad\pdkinfo
directory.
• This setting is global for all new .ind files created in the current user’s environment, set the
PDK Info File option in Global Settings pdkinfo_file to the desired PDK file for local
control within a specific .ind file.

6.K. Additional Component Properties

The Additional Component Properties dialog can be opened by clicking the More… button in a
component’s properties dialog.

Not all of these options are appropriate/enabled for all simulation tools or components. Consult
the appropriate simulation tool manual for information about the use of these options.

RSoft CAD  122

 Figure 6-14: The Additional Component Properties dialog which controls infrequently used
component properties.

6.K.1. General Properties

Component Name
This name will appear in the titlebar of the component properties dialog. Components can also be
selected by name via the Edit/Select/Edit Component menu item. For monitors, the name will be
used in some output plots and output file names.

Comment
The comment does not affect the simulation in any way and is solely for the user.

Tag
Tags are used in some cases to pass special instructions to simulation engines. See Section 6.K.4
for more details.

Draw Priority
This option sets the drawing priority for display in the CAD interface. By default, all segments
have a priority of 0 which means that component 1 is drawn first, component 2 next, etc., until
all components are drawn. If a component has a higher Draw Priority, it will be given precedence
and be drawn on top of components with lower priority. It is recommended to use as few draw
priority levels around the default of 0 as necessary.

RSoft CAD  123

Mask Layer
When exporting the design file to a mask file, this information will be used to place different
components in different mask layers. The default mask layer is 0. Setting the layer to -1 for a
specific component will exclude that component from export. For more information about the
export of mask files, see Section 6.E.

Sidewall Angle
The sidewalls of Channel and Multilayer components can be sloped. This sets the angle in
degrees and is measured from the vertical: A positive angle slopes in at the top, while a negative
angle slopes out. The angle is maintained along the length of the segment even if the geometry is
tapered. By default, the width is assumed to be defined at the bottom of the component, but this
can be changed via the Sidewall Angle Ref. option in the Global Index Generation Options
dialog.

Simulated Bend and Simulated Bend Radius

These fields control a CAD option which applies a coordinate transformation useful for
simulating segment bends. This feature is described in Section 6.D.1. These fields apply locally
in Z, that is, only over the length of the segment to which they correspond. Alternatively, this
option can be globally enabled for the entire simulation domain via the Global Index Generation
Options dialog.

Electrode/Heater Type and E/H Control Parameter

These options are used by the Multi-Physics Utility. See the Multi-Physics Utility manual for
more information.

6.K.2. Extended Segment Properties

These properties are valid only for extended segments (Section 4.F).
Reference X, Reference Y, and Reference Z
The reference position offsets the center of the actual object compared to the location of the
starting vertex.
Reference Phi, Reference Theta, and Reference Psi
The reference angles specify an extra rotation about the starting vertex. This rotation is done
after any reference position settings have been made. The rotation is done in the following order:
• Phi (ϕ) rotation is around Y axis and measured from Z axis.
• Theta (θ) rotation is around X' axis and measured from Z' axis.
• Psi (ψ) rotation is around Z' axis and measured from X' axis.
To transform from the unprimed system to the primed system:

RSoft CAD  124

X ' = ( X − X ref ) ( cos cos  + sin  sin  sin ) − (Y − Y ) (sin cos )
ref

+ ( Z − Z ) ( sin sin  cos  − sin  cos )

ref

Y ' = ( X − X ref ) ( cos  sin − sin  sin cos  ) + (Y − Y ) ( cos  cos )

ref

− ( Z − Z ) ( sin sin + sin  cos  cos )

ref

Z ' = ( X − X ref ) ( cos sin  ) + (Y − Y ) (sin  ) + ( Z − Z ) ( cos cos  )

ref ref

To transform from the primed system to the unprimed system:

X = X ' ( cos  cos + sin  sin  sin ) + Y ' ( cos  sin − sin  sin  cos ) + Z ' ( cos  sin  ) + X ref

Y = X' ( − sin *cos  ) + Y ' ( cos  cos ) + Z ' ( sin  ) + Yref

Z = X' ( sin  cos  sin − sin  cos ) − Y ' ( sin  sin + sin  cos  cos ) + Z ' ( cos  cos  ) + Z ref

where Xref, Yref, and Zref are the Reference X, Reference Y, and Reference Z settings.

X’Y’ Rotation Start, X’Y’ Rotation End, and X’Y’ Rotation Taper
This rotation ‘twists’ the component along its local axis. By default, the ending angle is set equal
to the starting angle. If they are not equal, the ‘twist’ can be tapered along the length of the
component. See Section 6.B for more information on tapers.

6.K.3. Crystal Axes

The Crystal Axes Specification dialog, which can be opened via the Crystal Axes button in the
Additional Component Properties dialog controls the local crystal axes of the component
material. This option can only be used when an Anisotropic material has been used, and with
simulators that support full anisotropy. See Section 8.B.3 for more information about defining
anisotropic materials and simulator manuals for full anisotropy support.

Figure: 6-15: The Crystal Axes Specification dialog.

RSoft CAD  125

Simple Crystal Axes Rotation
To simply rotate the crystal axes relative to the orientation definition in the Material Editor, set
via the Start values for Phi, Theta, and Psi:
• A Phi rotation is around Y axis and measured from Z axis.
• A Theta rotation is around X' axis and measured from Z' axis.
• A Psi rotation is around Z' axis and measured from X' axis.
This type of crystal axes rotation is available for all types of components.

Tapered Crystal Axes Rotation for Segments

For segments (i.e. two vertex components), it is possible to taper, or change, the crystal axes
along the segment. The taper has the form:

 ( z ) = 0 + (1 − 0 ) f ( z ')

z − z0
z' =
z1 − z 0
where θ(z) represents the crystal axis to be tapered, θ0 and θ 1 are the Start and End values
specified in the Crystal Axes Specification dialog for that crystal axis, f( z') is the taper function,
and z' is a normalized coordinate along Z that varies from 0 to 1 along the Z extent of the
segment. Individual control over all three angles is possible, and setting End to the keyword
default effectively disables tapering and uses a simple rotation as described above. Several
built-in tapers can be used, as well as user-defined tapers. See Section 6.B for more information
about using tapers.
When using an extended segment, the variable crystal_axis_prime can be set to 1 to align the
crystal axis along the segment Z’ axis.

6.K.4. Component Tags

Tags are used to pass special instructions to simulation engines and are set in the Additional
Component Properties dialog. They should not be set unless needed. Tags are discussed in
appropriate simulation manuals and include:
• skip_est: This tag, when applied to a component, removes it from the default parameter
estimation including domain and grid sizes.
• Several grid-related tags, see individual simulation manuals as well as Section 9.C and the
simulation manuals for more details.
• Several monitor-related tags, see individual simulations manuals for more details.
Multiple tags should be separated by spaces, a tag setting of:

RSoft CAD  126

 skip_est grid_size=Dx

enables the skip_est tag and sets the local grid size to the value of the Dx symbol.

Settings Tags
Tag may also be set in the Edit Default Tags dialog. This method allows tags to be assigned to
segments that are defined by a specific material or when importing geometries from TCAD
Sentaurus. The Edit Default Tags dialog can be opened via the Edit menu item in the RSoft
CAD. The syntax for these tags is:
component <name> = tag1=value tag2=value
region <name> = tag1=value tag2=value
material <name> = tag1=value tag2=value

where the keywords component and region are synonymous and allow tags to be assigned to
particular components in a design and the material tag allows tags to be assigned to
components that are defined by the material specified.

Figure 6-13: The Default Tags dialog where tag assignments can be specified.

Tag Precedence
The order of precedence for tags is, in order of importance:
• Tags set locally for a particular component
• Tags set via the ‘Edit Default Tags’ dialog for a particular component/region
• Tags set via the ‘Edit Default Tags’ dialog for a particular material

RSoft CAD  127

Example Tags
The above figure shows an example that assigns the following tags:
• The tags ‘skip_est=1 and grid_size_x=0.001’ to the component named MyComponent.
Note that while both component and region keywords are used here, only one need to be
used since they are synonymous.
• The tags ‘grid_size_ppw_z=25’ to components that are defined via the SiO2 material.

6.L. Using EIM to Simulate 3D Structures in 2D

(2.5D)
The Effective Index Method (EIM) can be used to greatly reduce the computation time and
memory requirements of a simulation by converting a 3D structure into an approximate 2D
structure. This is sometimes called ‘2.5D’. This option only affects the simulation; the definition
of the original 3D structure is unaltered allowing this feature to be safely turned on and off
without changing the defined structure. This feature is enabled by the Effective Index
Calculation option in the Global Settings dialog.
The use of EIM requires the following:
• Use one of the supported RSoft tools, including FullWAVE, BeamPROP, ModePROP, or
DiffractMOD. See the manuals for these products for additional information and/or
examples.
• Use only components with only one of the supported 3D Structure Types: Rib/Ridge,
Channel, or Diffused. The Y position of these segments must be:
• Rib/Ridge or Diffused components must by positioned at Y=0
• Channel components must be positioned either at the same Y position or such that their
bottom edge is at the same Y position.
Note the following when using EIM:
• The definition of polarization when using EIM is the same as is for 3D simulations allowing
the feature to be turned on and off in a consistent way. When using EIM, the polarization
option in the simulation dialog is displayed as Pol (3D EIM) and should be set using the
standard 3D polarization convention. This polarization is what is used to calculate the
effective 2D indices, the appropriate polarization will be used when actually simulating the
structure in 2D.
• For simulations that contain multiple wavelengths in a single simulation (i.e. FullWAVE
‘pulse’ simulations), the material properties of the effective material are automatically fit to a
dispersive model to account for dispersion.

RSoft CAD  128

6.M. Expanded Referencing of Vertices
One of the core aspects of the RSoft CAD is that a vertex position can be defined relative to
another vertex, allowing for components to be logically connected/positioned within a design.
This feature can be extended to include the width/height of a component as well as the
angle/orientation of the vertex.
Expanded referencing can be enabled by setting:
• The X, Y (for 3D), and
• The Z Reference of the starting vertex must have a Type of Offset and be Relative To the
same vertex
Once enabled, the width, height, or angle can be referenced:
• Width/Height Referencing:
The Component Width and Component Height can be referenced by setting these options to
the keyword default which will inherit the width/height value from the referenced vertex. If
expanded referencing is not enabled, it will be set the values set in the Global Settings dialog.
• Angle Referencing:
The angle (see details below for supported angles) can be referenced by settings this option
to the keyword default which will inherit the angle value from the referenced vertex. If
expanded referencing is not enabled, it will be set to 0.
Expanded references are supported by the following components:
• Segments:
Segments (Chapter 4) support expanded referencing for the Component Width, Component
Height, and the Reference Phi angle (in Additional Segment Properties dialog) options. Note
that Segment Orientation must be set to Extended.
• Polygons:
Polygon components (Section 5.D) support expanded referencing for the Component Width,
Component Height and XZ Angle options.

• Lenses:
Lens components (Section 5.B) support expanded referencing for the Component Height and
XZ Angle options.

• Circles:
Circle components (Section 5.C) support expanded referencing for the Component Diameter,
Component Height and XZ Angle options.

• Monitors and Port Monitors:

RSoft CAD  129

 Monitor components (see Section 5.F and simulation tool manuals) support expanded
referencing for the Component Width, Component Height, and Phi options.
• Marks and Vertices:
Mask (Section 5.E) and vertices (Section 5.F) support expanded referencing for the
Width/Length, Height, and XZ Angle options.

• Circuit References:
Circuit References (Section 6.F) support expanded referencing for the Component Width,
Component Height, and XZ Rotation Angle options.

6.N. Warning Options

The CAD and simulation tools will give warnings in some cases. They can be disabled in a
design file through the Warning Preferences window. These warnings should only be disabled if
necessary.

6.O. Component Filters

Component filters allow you to create custom list(s) of components based on user-defined
searches/filters. This helps to organize and group components, simplifying views, and
Filters help to organize and group components, can be used to produce custom views in the
drawing pans, and generally help to find specific components.
For example, a filter can:
• Show a list of all components that have a Material of ‘Si’
• Show a list all monitors
• Show a list of all components with a Merge Priority of 2
• In the drawing pane(s), hide all components that have a material of ‘Si’ and a Display Color
of Red.

6.O.1. Defining Filters

Filers are defined in the Component Filter folder in the tree view on the left of the RSoft CAD.

RSoft CAD  130

 Fig 6-14: The CAD tree view, showing the Component Filter folder with one filter defined.

Right-clicking on the Component Filter folder gives the option to create a new filter, opening the
Filter Creation dialog.

Figure 6-15: The Component Filter Properties dialog.

To define a component filter, give it a Name and Definition. You can use the Component Fields
and Target Value fields to help define the definition. The syntax is field=target, with boolean
AND or OR operators to combine multiple fields. Consider this example definition which returns
all components that have a ‘Si’ material and a Merge Priority of 1:
Material=Si AND priority=1

Here are the supported search fields, along with special rules for matching:

Component Field Description

type Matches against any of the following (case-insensitive):
• segment – matches segments in the Components tree folder
• polygon – matches polygon components in the Components
tree folder

RSoft CAD  131

 • lens – matches lens components in the Components tree
folder
• mark – matches mark components in the Components tree
folder
• monitor – matches monitors in the Monitors tree folder
• simregion – matches simulation regions in the Simulation
Regions folder
• hierarchy – matches circuit references in the Circuit
Reference tree folder
• launch – matches launch fields in the Launch Fields tree
folder
structure Matches the 3D Structure Type of a component (case-insensitive):
• fiber
• channel
• diffused
• ribridge
• multilayer
• polygon
priority Matches the Merge Priority of a component.
draw_priority Matches the Draw Priority of a component.
mask_layer Matches the Mask Layer of a component.
color Matches the Display Color of the component (case insensitive):
• black
• dkblue
• dkgreen
• dkcyan
• dkred
• dkmagenta
• brown
• ltgray
• dkgray
• blue
• green
• cyan

RSoft CAD  132

 • red
• magenta
• yellow
• white
material Matches the Material of a component (case-sensitive).
name Matches the Name of a component (case-sensitive)
tags Matches the Tag of a component. Note that some tags have values,
some do not:
• To match a tag that does not have a value, use tags=abc
syntax.
• To match a tag that has a value, use tags:var=abc syntax.
groups Matches the Group of a component

6.O.2. Using Filters

Once a filter is defined, it appears in the Component Filters folder in the CAD tree view. You
can use the filters to simply sort components, or use them to show/hide the group, edit the group
properties, etc. Right-click on the filter to select one of these options:
• Edit Filter: Edit the filter settings.

• Delete Filter: Delete the filter

• Select All: Select all components in the filter

• Hide Group: Hide the components in the filter

• Show Group: Show the components in the filter

• Delete Group: Delete all components in the filter

• Group Properties: Edit the group properties of all components in the filter

6.P. Simulation Log Files

Each simulator saves information in a log file. By default, this file is named [Link] and
appended to each time a simulation is run in a particular directory. There are several options to
control this:
• To create a new file each time the simulation runs (i.e.. delete the old file), use the Menu
option in the RSoft CAD View/Clear Log/Always.

RSoft CAD  133

• The variable rsoft_logbyprefix can be used to name the log file using the simulation
prefix. A value of 0 uses the default [Link] name, a value of 1 uses the simulation prefix (i.e.
<prefix>_log.txt) and appends data for subsequent simulations, and a value of 2 uses the
simulation prefix and creates a new file for each simulation

RSoft CAD  134

7

Layout Generation Utilities

Some designs call for complicated structures which can be very tedious to manually create. For
this reason, the CAD includes automatic layout utilities for certain common problems. These can
be accessed via the Utility menu item:
• Array Layout Generator
This utility is used to create periodic arrays of segments and will be described in this chapter.
• WDM Router Layout tool
This utility is used to create layouts for simulation and mask creation for arrayed-waveguide-
grating routers (or AWG’s). This utility is described in detail in the AWG Utility manual.
• GratingMOD Grating Layout tool
This utility is used create grating layouts to be used with RSoft’s grating simulation tool
GratingMOD and is described in detail in the GratingMOD manual
The rest of this chapter will describe the usage of the Array Layout utility.

7.A. Motivation
Dynamically-sized arrays can be created using the hierarchy feature discussed in Section 6.F.
The RSoft CAD system is very powerful and allows expressions to define aspects of the
geometry and component references to define logical relationships between components. Once a
properly designed layout is complete, these features make it very easy to vary the structure and
perform design studies. The Array Layout utility allows the user to quickly create structured

RSoft CAD  135

composed of periodic arrays. It can be opened via the Utility menu item in the CAD window
(Fig. 7-1). The Array Layout Utility can be divided into three sections: Unit cell Parameters,
Lattice Parameters, and Creation Options. These three sections are shown in Fig. 7-1 and are
described in the next sections.

Figure 7-1: The Array Layout Generator interface.

7.B. Lattice Parameters

The lattice parameters define the type of periodic lattice to create from the defined unit cell. This
section will cover the lattice parameters; the next the unit cell parameters.

Dimensions
This sets the dimensions of the lattice to be created. Possible choices are 1D, 2D (XY), 2D (XZ),
and 3D. Each dimension has different possible lattice types which are described below.

Do not confuse this setting with the dimension of the simulation to be performed. This only
defines the periodicity to use. For example, a PBG slab is periodic in two dimensions, but is
finite in the third and thus requires a 3D simulation.

Lattice Size

RSoft CAD  136

The lattice size controls the size of the lattice produced. The definition of L, M, and/or N will
change depending on the dimension and lattice type chosen, and is shown in the GUI. For a
general cubic case, L corresponds to the number of elements along the X axis, M along the Y
axis, and N along the Z axis. However, other lattice types have different definitions.

7.A.1. 1D Lattice Types

The available 1D lattices are:
• 1D X, 1D Y, 1D Z
These options create a 1D lattice along the specified coordinate. One such lattice with a
rectangular unit cell is show here:

The Lattice Size settings correspond to the number of unit cells in the a direction.

7.B.2. 2D Lattice (XZ and XY) Types

The available 2D lattices are shown in Fig. 7-2, and described below.

Figure 7-2: 2D lattice types a) cubic, b) hexagonal, c) cubic rings, and d) hexagonal rings.

• Cubic
This options creates a 2D cubic lattice, and the Lattice Size settings correspond to the
number of unit cells in the a and b directions.
• Hexagonal
This option creates a 2D hexagonal lattice, and the Lattice Size settings correspond to the
number of unit cells in the a and b directions. The lattice may be rotated using the symbol
PhiPattern to achieve the other typical hexagonal lattice orientation.

• Cubic Rings

RSoft CAD  137

 This option creates a 2D lattice formed by cubic rings, and the Lattice Size settings
correspond to the number of outside and inside rings. The example lattice shown above has 3
outside rings and 2 inside rings.
• Hexagonal Rings
This option creates a 2D lattice formed by hexagonal lattice, and the Lattice Size settings
correspond to the number of outside and inside rings. The example lattice shown above has 3
outside rings and 1 inside ring.

7.B.3. 3D Lattice Types

The available 3D lattices are shown in Fig. 7-3, and are described below.

Figure 7-3: Some of the available 3D lattice types (clockwise from top right) a) cubic, b)
FCC, c) BCC, d) woodpile, e) diamond logs, and f) diamond.

• Cubic
This option creates a 3D cubic lattice and the Lattice Size settings correspond to the number
of unit cells in the a, b, and c directions.
• FCC
This option creates a 3D FCC (face centered cubic) lattice and the Lattice Size settings
correspond to the number of unit cells in the a, b, and c directions.
• BCC

RSoft CAD  138

 This option creates a 3D BCC (body centered cubic) lattice and the Lattice Size settings
correspond to the number of unit cells in the a, b, and c directions.
• Diamond
This option creates a 3D diamond lattice, and the Lattice Size settings correspond to the
number of unit cells in the a, b, and c directions.
• Diamond Logs
This option creates a 3D diamond lattice with connecting logs, and the Lattice Size settings
correspond to the number of unit cells in the a, b, and c directions.
• Woodpile
This option creates a 3D woodpile lattice, and the Lattice Size settings correspond to the
number of unit cells in the a, b, and c directions.

7.C. Unit Cell Parameters

This section of the Array Layout Utility window allows the user to define the contents of the unit
cell to be periodically repeated. The contents of the unit cell are set by the pulldown menu
labeled Contents. The options are:

7.C.1. 1D Unit Cells

The available 1D unit cells are
• Sine Wave
This option creates a 1D unit cell composed of a sine wave such as is shown here:

The *.ind file created has one segment with either a user profile or taper depending on
whether the variation is along X, Y, or Z.
• Square Wave
This option creates a 1D unit cell composed of a square wave such as is shown here:

RSoft CAD  139

 The *.ind file created has one segment with either a user profile or taper depending on
whether the variable is along X, Y, or Z.
• Sawtooth Wave
This option creates a 1D unit cell composed of a square wave such as is shown here:

The *.ind file created has one segment with either a user profile or taper depending on
whether the variable is along X, Y, or Z.
• Triangle Wave
This option creates a 1D unit cell composed of a square wave such as is shown here:

The *.ind file created has one segment with either a user profile or taper depending on
whether the variable is along X, Y, or Z.

7.C.2. 2D Unit Cells (XZ and XY)

The available 2D lattices are shown in Fig. 7-4, and are described below:

Note that all options are not available for all lattice dimensions.

Figure 7-4: Some of the available 2D unit cells: a) square, b) hexagon, c) circle, and d)
ellipse.

• Square

RSoft CAD  140

 This option creates a 2D unit cell composed of a singular square object. The segments used
in the *.ind file to create the lattice are standard waveguide segments for both XZ and XY
layouts, and can be easily modified to produce rectangular segments if necessary.
• Hexagon
This option creates a 2D unit cell composed of a singular hexagonal object. The segments
used in the *.ind file to create the lattice are polygons. This option is only available for XZ
layouts.
• Circle
This option creates a 2D unit cell composed of a single circular object. The segments used in
the *.ind file to create the lattice are lenses for XZ layouts and 3D fibers for XY layouts.
• Ellipse
This option creates a 2D unit cell composed of a single elliptical object. The segments used
in the *.ind file to create the lattice are segments with width tapers for XZ layouts and 3D
fiber objects for XY layouts.
• .ind file
This options uses hierarchy (Section 6.F) to periodically repeat a ‘child’ design file to create
a custom lattice in a ‘parent’ file. The contents of this unit cell are defined by whatever is in
the *.ind file used. The *.ind file to be used is specified in the field .ind file.

7.C.3. 3D Unit Cells

The available 3D lattices are shown in Fig. 7-5, and are described below:

Figure 7-5: Selected 3D unit cells: a) cube, b) hexagonal prism, c) sphere, and d) cylinder.

• Cube
This option creates a 3D unit cell composed of a single cubic object. The segments used in
the *.ind file to create the lattice are simply polygons and can be easily modified to produce
rectangular segments if necessary.
• Hexagonal Prism

RSoft CAD  141

 This option creates a 3D unit cell composed of a single hexagonal prism object. The
segments used in the *.ind file to create the lattice are polygon segments.
• Sphere
This option creates a 3D unit cell composed of a single spherical object. The segments used
in the *.ind file to create the lattice are 3D lens objects.
• Cylinder
This option creates a 3D unit cell composed of a single cylindrical object. The segments used
in the *.ind file to create the lattice are 3D lens objects.
• .ind file
This options uses hierarchy (Section 6.F) to periodically repeat a ‘child’ design file to create
a custom lattice in a ‘parent’ file. The contents of this unit cell are defined by whatever is in
the *.ind file used. The *.ind file to be used is specified in the field .ind file.
• Rectangular Rods
This option applies to the Woodpile Lattice Type only and creates a woodpile lattice
composed of rectangular rods. The segments used in the *.ind file are channel segments.
• Elliptical Rods
This option applies to the Woodpile Lattice Type only and creates a woodpile lattice
composed of elliptical rods. The segments used in the *.ind file are fiber segments.

7.C.4. Additional Options

These options further control the definition of the unit cell.

Randomness
It can be useful to incorporate perturbed lattice parameters for each point in the lattice to study
non-ideal crystals. All selected parameters will be defined using the ranseq function. A seed
variable is also created to allow the user to control the set of random numbers used. The
perturbation is defined as
<parameter>+sig_<parameter>*ranseq(seed_<parameter>,$id)

where <parameter> is the parameter to be varied. For example, if a random Refractive Index, or
delta, is used, the Index Difference of each object in the array will have an Index Difference
defined as
delta+sig_delta*ranseq(seed_delta,$id)

where the following table explains the parts of the equation:

Parameter Description

RSoft CAD  142

 delta
The standard Index Difference defined in the Global Settings dialog.
sig_delta
The “amplitude” of the randomness, controls the min/max perturbation.
ranseq
This function returns a random number based on the two seed variables
described below. The value it returns lies in the range (-0.5,0.5). See Appendix
C for more details on this function.
seed_delta
The first seed variable is the same for each segment. Changing this variable in
the symbol table will choose a different “random” set of perturbations for the
design.
$id
The second seed variable is set to a value of $id, which is a special notation for
the segment number. This allows each segment to have a different random
number for a given value of the seed_delta parameter.

Special Effects
These fields allow the user to add additional capability to the design file to use materials.

7.D. Creation Options

The final section of the Array Layout Generator dialog allows the user to set the name of the
output file and to save lattice creation settings.

Output Prefix
The output prefix sets the name of the *.ind file to be created and is required.

Keep these Settings Next Time

This option saves the settings in the Array Layout Generator to for the next time a lattice is
created.

RSoft CAD  143

8

Advanced Materials

RSoft’s Material Editor provides a convenient database of complex material models that can be
used in simulations. These materials can be simple dielectrics, but can also incorporate
dispersion, non-linearity, and anisotropy. A built-in material library is provided; users can also
create specific materials as needed. Materials can be shared between different designs and users.
This chapter will discuss how to define these material properties and assign them to specific
objects.

Since the CAD is shared by several simulation tools, all possible material properties will not
have meaning for all simulation tools. It is critical to understand the underlying simulation
mechanisms and if various properties are relevant.

The first section outlines the usage of the Material Editor; the rest of the chapter is devoted to
documenting specific material parameters.

8.A. Using the Material Editor

The Material Editor dialog, shown in Fig. 8-1, can be opened by clicking the corresponding
button in the left toolbar of the CAD, by clicking the Materials… button in component properties
dialogs, or by selecting New Material in the Material Properties field in a component properties
dialog. This dialog has a material tree control where the user can browse the materials available
for use within the current project, as well as any material libraries that have been defined. When
selected, a material’s detailed information is displayed over several tabs.

RSoft CAD  144

8.A.1. The Material Tree
The Material Tree allows users to quickly navigate through materials which have been defined.

Figure 8-1: The Material Editor dialog. The material tree is located on the left, the toolbar is
on top, and the material property tabs are displayed on the right.

The tree is divided into several main categories:

• Project Materials
These materials are available to use and are stored within the local design file.
• RSoft Library
This built-in, write-protected library is provided by RSoft. It contains typical dielectric,
metallic, semiconductor, and special materials such as Graphene (see Section 8.K) and PEC
(Perfect Electric Conductor). Materials in this library cannot be edited; if changes need to be
made to one of these materials, copy it to a user-defined library make the changes there.
• RSoft CAD Info Folder(s)
Materials from one or more RSoft CAD Info folders can be used in a design file, see the
Custom PDK Utility manual for more information on RSoft CAD Info folders.
• Group Libraries

RSoft CAD  145

 This category holds user-created libraries and materials that are available to multiple users on
a single computer. Users can use this library to ‘share’ materials between .ind files:
materials from the Project Materials group to this group, and copy materials from this group
to the Project Materials. Libraries and materials in this category are stored in a user-defined
directory which should be a shared directory outside the RSoft installation directory. There
are two ways to define this directory:
• Via a Preference
This directory can be easily set by right-clicking on the Group Libraries folder and
selecting New Library option. Click Yes and then select a location to create the new
folder. Alternatively, it can be set via the Group Library Path option in on the
Import/Export tab of the CAD Preferences dialog described in Section 6.J.
• Via an Environment Variable
The environment variable RSOFT_MATLIB_GROUP_PATH can be used to select the location
of this folder. Note that if the ‘Preference’ method described above is used, it will
override this setting. Please contact your system administrator for information on setting
this environment variable.
Users with only read-access to this directory will not be able to create new or modify existing
libraries/materials. Users with write-access must take care to avoid inadvertent changes to
materials as this will permanently change the database for other users. See notes in Section
8.A.7 for using data files in materials defined in this group.

8.A.2. The Material Tabs

The material tabs can be accessed when a material is selected in the material tree. Each tab
corresponds to a different material property such as linear epsilon or electro-optic parameters.
These tabs are described in Sections 8.B and following in this chapter. Not all material properties
can be used by all simulation engines, nor do all properties have to be set.

8.A.3. Creating and Organizing Materials in the Material Editor

The user can create and organize materials in the Material Editor using the buttons on the top
toolbar, or by right-clicking in the tree area. These options are only available when it is
appropriate to use them.
These options are:

New Material Copy Material

Create a new material in selected Copy selected material to the
library. The name of the material can be material clipboard.
changed at the top of the dialog.

RSoft CAD  146

 New Library Paste Material
Create a new library in the User Paste contents of material
Libraries category. The name of the clipboard into selected library.
library can be changed at the top of the
dialog.

Save Library Use Material (Add to Project)

Save library after changes have been Duplicates selected material in the
made. If not saved, the CAD will Project Materials category for use in
prompt to do when exiting the Material local project (design file).
Editor.

Delete Material or Library

Deletes the selected
material or library.

8.A.4. Using Materials in a Project (Local Design File)

It is critical to understand that in order to use a material in a project (the local design file), a copy
of the material must be placed within the Project Materials category of the materials tree. To do
this, either copy/paste the material or use the Add to Project button described in the previous
section. The benefit of this approach is twofold: to keep the design file consistent even when
materials in the Material Editor change, and to simplify the sharing of design files.

Materials must be in the Project Materials category to be used within the design file.

Once in the Project Materials category, the material can be used within the RSoft CAD interface
in places such as the Material Properties setting in a Component Properties dialog, the Layer
Table editor, or the Global Settings dialog.

Using a Symbol to Define Assign a Material

The Material Properties setting can be set via a symbol if desired. To do this, click on the text
part of the Material Properties control and type the symbol name preceded by the $ character,
i.e. $Material_name. The symbol should contain the actual name of the desired material in
quotes, i.e. “Ag”. The empty but still quoted value of “” can be used to select the Locally
Defined option and effectively disable the use of materials.
This can be useful in many circumstances but is especially useful when using hierarchy as a
means to control materials in a child file from the parent file. See Section 6.F.5 for details.

RSoft CAD  147

8.A.5. Displaying Material Definition
It can be beneficial to test specific material definition to ensure that accuracy. This can easily be
done via the Display button which will open a plot of the current material property. Several
options which control the format of the output can be set in the Material Display Options
window which can be opened by clicking the Display Opts… button. These options are:

Display Axes
Sets the axes used to display the material data. Both the X and Y axes can be set.

Plot Range/Resolution
Sets the minimum and maximum values on the X axis, as well as the grid spacing used to display
the material properties. Separate controls are given for wavelength and frequency.

Measurement Data
Allows a data file to be plotted with the computed material data. This is useful when trying to
match a particular data file. The data file must have the same units as the chosen Display Axes
for a direct comparison. This feature is for comparison purposes only.

Output Prefix
Sets the name of the output file used.

8.A.6. Material Properties

The material properties are set in the General Material Properties dialog, which can be opened by
clicking the Properties button in the Material Editor. Here you can set the material color and set
several notes.

Fig 8-2: The General Material Properties dialog.

The options in this dialog are:

RSoft CAD  148

Comment
Sets a comment for the material. This information is not used anywhere in the software, it is
strictly for record keeping purposes.

Origin
Displays the origin of the material. Materials in the RSoft Material library include the
wavelength range and/or references if applicable. This information is not used anywhere in the
software, it is strictly for record keeping purposes.

Display Color

This option is used in specific circumstances only, see notes below.

Sets the display color of components to which the material is assigned. You can select from one
of the standard colors or define RGB values for a custom color. The options are:
• Default: Uses the default color, see options below.
• A standard Color: Select the desired color form the dropdown list, i.e. Red, Dark Red, etc.
• Custom: Select an RGB value for the color.

Note on Display Colors

There are various places to set the color, here is the order of precedence for the options:
• Global Preference Color Map File: The Color Map file set in the Preferences dialog (See
Section 6.J) can be used to set material colors at a group level. This is intended for situations
such as multiple users who work on the same file(s) and want a standard color scheme across
multiple users, projects, and files.
The color map file is a text file, each line sets the default color for a material. Color codes or
hex color values can be used, for example:
Glass = COLOR_RED
SiO2 = COLOR_RED
SiN = #0045234

• Users can change a material color for specific materials in a project using the material
Display Color option described above. This overrides the global color map file.
• The color of specific components can be directly set by the Display Color option in the
component properties dialog. This overrides any other setting.

8.A.7. Tips & Traps

Keep the following in mind when using the Material Editor:

RSoft CAD  149

• All of the toolbar actions described in Section 8.A.3 can also be done with the mouse. Right-
clicking on a library or material gives a menu of the possible actions, and materials can
copied and pasted into other libraries as well as dragging them to the Project Materials
category. Keyboard shortcuts are also available for most actions. These shortcuts are
displayed in the menu shown when right-clicking on a library or material.
• Materials from one or more RSoft CAD Info folders can be used in a design file, see the
Custom PDK Utility manual for more information on RSoft CAD Info folders.
• The Symbol Table button on the bottom of the Material Editor dialog opens the Symbol
Table for the local design file. It is recommended that users not use symbols when defining
materials that will be copied to the Group Library category. This is because these symbols
must not only be defined in any design file that the material will be used in, but also have the
correct values. The one exception to this rule is the built-in symbol
free_space_wavelength which is defined in all design files. This exception allows for the
creation of simple wavelength-dependent material properties such as those described in the
notes in Section 8.B.1.
• If using a nk data file via the userreal() and userimag() functions, the wavelength range
in the data file will be displayed in the Material Editor. A warning will also be displayed if a
simulation is run at a wavelength outside the range. If ignored, the nk data at the end of the
range will be used.
• Synopsys periodically updates material data for materials in the RSoft Material Library.
Archives of older material data, sorted by version, can be found in this folder:
<rsoft_dir>\products\rsoftcad\materials\rsoft\archives

• It is possible to use user-created data files to define material properties as a function of

wavelength via the userdata() function (see the notes in Section 8.B.1 for an example).
These data files must be in specific locations so that the CAD can find them:
• For materials defined in the Project Materials category, no path information is necessary
if the files are in the same directory as the design file. The appropriate expression would
be:
userdata(“material_property.dat”, free_space_wavelength)

• For materials in the Group Library category, any associated files should be located in the
directory defined by the RSOFT_MATLIB_GROUP_PATH environment variable. The special
tag <groupmat> can be used in any expressions to refer to this path. The appropriate
expression would be:
userdata(“<groupmat>\material_property.dat”, free_space_wavelength)

• Each built-in material has notes which can be viewed by clicking the Notes button in the
bottom of the editor. These notes contain important information about the material model.
Users are encouraged to create these notes for user-defined materials as well.

RSoft CAD  150

• It is critical to make sure that the properties of any material(s) used correspond to the actual
physical conditions you wish to simulate. This is especially important when using the built-in
materials since each model is valid only for certain conditions such as wavelength,
temperature, pressure, etc.. Also, they may not be useful for FullWAVE simulations with a
single wavelength, or for multiple wavelengths at a time. Use the display options liberally,
check the material notes, and read through the notes in Section 8.B.1 to ensure that any
material properties are valid for your specific needs. RSoft does not guarantee the accuracy
of any supplied materials and may change them without notice in future releases.

8.B. Linear Epsilon Properties

The linear epsilon, or relative permittivity, properties can be defined in the Eps Linear tab. It can
be either a simple real or complex number (i.e. isotropic/uniaxial and non-dispersive), or either
anisotropic or dispersive.

8.B.1. Linear Epsilon Overview

The Material Editor allows users to create materials with the following definition for linear
epsilon:
D =  0E + P

where

P =  0   ( ) E + non-linear terms 

and χ(w) can be a frequency dependent linear dispersion term. For the rest of this section, we will
assume that there are no non-linear terms; see Section 8.B.2 for more on non-linear epsilon.

Notes on Dispersive Linear Epsilon

Material dispersion, or the change of refractive index as a function of wavelength, is not
automatically included by any of RSoft’s simulation tools. The user needs to define how the
refractive index varies with wavelength. Some materials, such as Silicon, have additional
properties: see Section 8.I.

For simulation tools that perform monochromatic simulations only, a parameter scan over
wavelength is required to study dispersive effects.

There are several ways to define dispersive materials in the CAD.

• Via Simple Linear Epsilon

RSoft CAD  151

 This method sets a component’s complex refractive index to a single value either via the
Component Properties dialog (when Material Properties is set to Locally Defined) or the
Simple Linear Epsilon properties in the Material Editor. This value for real or imaginary
epsilon can be defined as a function of wavelength via an expression or via a data file. One
example of using an expression could be
1+(B*free_space_wavelength^2)/(free_space_wavelength^2-C)

where B and C are coefficients in the symbol table and free_space_wavelength is the
simulation wavelength, and one example of using a data file could be
userdata(“tungsten_real.dat”, free_space_wavelength)

where the file tungsten_real.dat contains refractive index data in the standard RSoft
format and lambda is the simulation wavelength. See the notes in Section 8.A.7 for more
information on using symbols and data files in materials, and see Appendix C for more
details of this and other functions that can be used in the symbol table and Appendix B for
more details on file formats.
This method is valid for single simulations at one wavelength, and if defined as a function of
wavelength, for parameter scans over wavelength.
• Via Dispersive-Class Materials in the Material Editor
This method uses a dispersion-class material in the Material Editor to define a
comprehensive dispersive relation for a material. This method is valid for single simulations
at one wavelength, scans over wavelength and for FullWAVE simulations where multiple
wavelengths are present in a single simulation such as a pulsed excitation.

8.B.2. Simple Linear Epsilon

A simple linear epsilon can be defined when Anisotropic and Dispersive are not selected. This is
equivalent to:
D = E

Dispersive materials can be defined via this method if epsilon (or index) is defined as a function
of the simulation wavelength as noted in Section 8.B.1. All simulation tools support this material
property.

The options for this type of material are:

Input Mode
Sets the way in which the entered data is interpreted. When Epsilon is selected, the user can
input epsilon directly. When Index is selected, the user can input the refractive index, or the
square root of epsilon. This does not automatically convert any entered data between these two
formats.

RSoft CAD  152

Index (real) or Epsilon (real)
Sets the real part of the linear epsilon (or index).

Index (imag) or Epsilon (imag)

Sets the imaginary part of the linear epsilon (or index).

Import NK Data…
This button selects a data file that contains n/k data and automatically sets the Index/Epsilon
options so that they can read in the data from the data file. Specifically, these options are set to:
userreal(“<data_file>”,free_space_wavelength)
userimag(“<data_file>”,free_space_wavelength)

where <data_file> is the name of the selected file. The data file should have three columns, the
first should be wavelength [μm], the second should be the real index (n), and the third should be
the imaginary index (k). The filename should not contain spaces. Alternatively, the data can
contain real/imaginary epsilon. The data will be interpreted according to the Input Mode. The
data can have the standard RSoft 1D real/imag header or no header. See Appendix A for a
description of supported file formats and Appendix C for more information on the userreal()
and userimag() functions.

Figure 8-3: The portion of the Linear Eps tab in the Material Editor where a simple linear
epsilon is defined.

RSoft CAD  153

8.B.3. Anisotropic Linear Epsilon
An anisotropic linear epsilon can be defined when Anisotropic is selected. This is equivalent to:

  xx  xy  xz 
 
D =   yx  yy  yz  E
  zx  zy  zz 


Dispersive materials can be defined via this method if the tensor elements are defined to be a
function of the wavelength as noted in Section 8.B.1. Not all simulation tools support the use of
anisotropic materials or the use of all tensor elements; consult the relevant simulation tool
manual for more information. Instructions on viewing material profiles for specific tensor
elements can be found in the ‘Notes for Anisotropic Designs section in Section 3.G.1.

The options for this type of material are:

Input Mode
Sets the way in which the entered data is interpreted. When Epsilon is selected, the user can
input the epsilon tensor directly. When Index is selected, the user can input the refractive index,
or the square root of epsilon. This square root is done on an element by element basis like nij2 =
εij where i and j are the tensor indices.

xx(re), xx(im), et al.

Sets the real and imaginary parts of epsilon (or index) for the xxth element in the epsilon (or
index) tensor. The other elements are set in similar fashion. The crystal axes of the material can
be explicitly defined via these controls or rotated for a particular CAD component using the
Crystal Axes options described in Section 6.K.3.

RSoft CAD  154

 Figure 8-4: The portion of the Linear Eps tab in the Material Editor where an anisotropic
linear epsilon is defined.

8.B.4. Dispersive Linear Epsilon

A dispersive linear epsilon and/or a linear gain with saturation can be defined when Dispersive is
selected. This is equivalent to:

D =  ( ) + G ( )  E
 k
 ( ) =   + 
k −ak  − bk ( i ) + ck
2

where G(w) is a frequency-dependant gain term (for FullWAVE only) to be discussed later, ε∞ is
the value of epsilon in the limit of infinite frequency, Δεk is the strength of each resonance, ak, bk,
and ck are fitting coefficients, and w is measured in units of radians/μm. The choice of these
parameters sets the type of resonance used.

RSoft CAD  155

 Figure 8-5: The portion of the Linear Eps tab in the Material Editor where a dispersive linear
epsilon is defined.

All simulation tools support the basic use of this material property, and will compute the
dispersive properties at the simulation wavelength. Furthermore, this method provides a true
dispersive environment for FullWAVE simulations with multiple simulation wavelengths (such
as in the case of a pulsed excitation). To use this feature, the Dispersion/Nonlinearity option in
the FullWAVE Simulation Parameters window must be enabled. Also, when using this method
in the context of a FullWAVE CW simulation, note that an optimized form of the dispersion
parameters are used since one wavelength is primarily present. This optimization can be disabled
by setting fdtd_dispersion_opt to 0.

Standard linear dispersion models, such as Lorentz, Drude, and Debye models can be used by
choosing the fitting parameters ak, bk, ck as follows:

ak bk ck
Lorentz 1 free free
Drude 1 free 0

Debye 0 free 1

The free parameters above, plus the coefficient Δεk (delta in the graphical interface), can be used
to match any single dispersion resonance. Multiple resonances can be combined to realize
complicated dispersion relations; to view additional resonances, use the <</>> buttons. The Gain
Term checkbox should not be checked unless the gain feature described later in this section is
used.

RSoft CAD  156

Fitting Data Files to Lorentzian Coefficients

N/k data fits are only used for pulsed FullWAVE simulations. All other simulation types are at a
single wavelength and a data fit is not necessary.

The RSoft Material Editor can automatically fit a data file to Lorentzian coefficients. There are
two types of fits:
• Auto-fit for Each Simulation:
Click the AutoFit button in the Linear Epsilon tab of the Material Editor and select the auto-
fit options. The n/k data will be automatically fit before each simulation. The AutoFit Method
selects the type of fit: A Simple fit matches the value and slope of the n/k data at the Free
Space Wavelength; a Lorentzian fit uses a multiple Lorentzian model to the data. Click the
Try Fit button to view the current fit (note that the actual fit will be calculated before each
simulation).
• One-time Fit:
To fit the data, open the Material Editor, create a new material, and click the Dispersive
button on the Linear Eps tab. Click the Fit NK Data… button to open the Fit NK Data
Options dialog. Select the n/k data file and click Try Fit. The fit will be performed, and then a
comparison between the original data and data fit will be displayed. The fit looks good, but if
it wasn’t you can change the fitting parameters and retry the fit. Click Accept Fit to copy the
calculated coefficients into the material.
In both cases, the fit is done with the fitdisp utility, see Appendix E for more details about this
utility.
Note the following:
• The data file should contain complex index (n/k) data vs. wavelength [μm]. This can be from
any source including experiment or many online sources. The data file should have three
columns like:
wavelength1 n1 k1
wavelength2 n2 k2
wavelength3 n3 k3
etc…

No header is required, though other formats (see Appendix B) can be used.

• If the fit is not adequate, vary the fit options (see the documentation for the fitdisp utility in
Appendix E for more details about these options) and click the Try Fit button again. Repeat
until a good fit is found.

Option Description
Max # Lorentzians Sets the maximum number of Lorenzian resonances

RSoft CAD  157

 used for the fit.
Fit Tol Sets the fitting tolerance.
Lambda Min Sets the range of wavelengths to be fit. If a good fit
Lambda Max cannot be found, try reducing the wavelength range.
When using the auto-fit option, the wavelength range is
automatically determined at run-time based on the
simulation settings.

Fitting Standard Linear Dispersive Models

Assuming the following forms for typical resonances, the fitting parameters for each resonance
type is given below.

These equations represent typical forms of these resonances. Results for similar forms can easily
be found.

• Lorentzian Resonance

Ak k2
 ( ) =   + 
k − 2 − 2i k + k2

where Ak is the resonator strength, δk is the damping factor, and wk is the resonant frequency.
• Drude Resonance

 p2
 ( ) =   +
−2i c −  2

where wp is the plasma frequency and vc is the collision frequency.

• Debye Resonance
Ak
 ( ) =   +
1 − itk

where Ak is the resonator strength and tk is the Debye relaxation time constant.
• Sellmeier Resonance

Ak  2
 ( ) =  +  2
k  − k − i Sk
2

where λk is the resonant wavelength, Ak is the resonator strength, and δSk is the Sellmeier
damping factor.

RSoft CAD  158

The following chart outlines the appropriate choice of fitting parameters to realize a Lorentz
resonance:

Sellmeier
Parameter Lorentz Value Drude Value Debye Value
Value
Δεk Akwk2 wp2 Ak Ak(2π/λk)2
ak 1 1 0 1

bk 2δk 2vc tk (2π/λk) 2δSk

ck wk2 0 1 (2π/λk)2

Dispersive Linear Gain for Epsilon

As previously mentioned above, FullWAVE supports an optical gain term that is defined through
a saturable, frequency dependent conductivity, and is defined as:

1  1 
G ( ) =     k ( )
i 1 + g sat I  k
 b 2 ibk 
 0 k  k − 
 k ( w) =  4 2 
− ak  2 − ibk + ck

where Δσ0k is related to the peak strength of each resonance, ak, bk, and ck are fitting coefficients,
and w is measured in units of radians/μm These fitting coefficients can be chosen for standard
resonance types in the same exact way as described above for non-gain terms.
It is possible to define a material that has standard dispersion and dispersive gain. In order to
define a resonance as a gain term, check the Gain Term box next to the resonance in the Material
Editor. If this box is not checked, the resonance will be applied to the standard dispersion
relation. Also, multiple gain resonances can be defined to realize a complicated dispersion
relation.

8.C. Non-Linear Epsilon Properties

The non-linear epsilon, or relative permittivity, properties can be defined on the Eps Nonlinear
tab and allow control over both second-order and third-order non-linearities.

Overview
The Material Editor allows users to create materials with the following definition for non-linear
epsilon:

RSoft CAD  159

 D =  0E + P
P =  0 linear term+ 2E2 +  3 ( ) E E
2
 
where χ2 is the second-order non-linear susceptibility, and χ3(w) is the saturable dispersive third-
order non-linear susceptibility and is defined as a linear sum of multiple resonance terms
multiplied by a saturation term:

  1 
 3 ( ) =   3 +   k3 ( )   2
 k  1 + csat E 
where χ∞3 is the value of the susceptibility in the limit of infinite frequency and χ3(w) are
dispersion terms. Each dispersion term is defined according to the following:

 k3
 k3 ( ) =
ak ( i ) − bk ( i ) + ck
2

where Δχ3k is the strength of each resonance, ak, bk, and ck are fitting coefficients, and w is
measured in units of radians/μm. The choice of these parameters sets the type of resonance used.

Only BeamPROP and FullWAVE support the use of this material property. Specifically,
BeamPROP supports third-order non-linearity, and computes the susceptibility at the simulation
wavelength. FullWAVE supports both second- and third-order non-linearities and computes the
susceptibility for any wavelengths present in the simulation. To use this feature in FullWAVE,
the Dispersion/Nonlinearity option in the FullWAVE Simulation Parameters window must be
enabled. For both BeamPROP and FullWAVE, the parameter nl_iterations (default = 1) sets
the number of iterations used to compute the non-linear effects.

Setting Non-Linear Epsilon Properties

The options for this material property are:

Chi2
This sets the χ2 susceptibility described above.

Chi3
This sets the χ3 susceptibility described above, and corresponds to the value in the limit of
infinite frequency.

Chi3 Saturation Coef:

This sets the saturation coefficient described above for χ3.

RSoft CAD  160

 Figure 8-6: The portion of the Non-Linear Eps tab in the Material Editor where non-linear
epsilon is defined.

Fitting Standard Non-Linear Models

Standard nonlinear models, such as Raman and Kerr models can be used by choosing the fitting
parameters ak, bk, and ck as follows:

ak bk ck
Lorentz 1 free free
Drude 1 free 0

The free parameters above, plus the coefficient Δχ3k, can be used to match any single resonance.
Multiple resonances can be combined to realize complicated dispersion relations.

A Note on Units for Non-Linear Parameters in FullWAVE:

Since most problems are linear, the field units are essentially irrelevant since monitor results are
normalized to the input power: all results are clearly interpreted as relative. However, when
using nonlinear materials, units become important: the software must be used in a consistent
way to get the correct answer:

This section provides a simple interpretation of units. A detailed approach for FullWAVE can be
found in the appendices of the FullWAVE manual.

• Using field values without units:

RSoft CAD  161

 It is sometimes easier, to keep E unitless and use dimensionless χ2 and χ3 parameters. This is
most easily done by setting the launch Normalization to be None or Unit Peak, leaving the
Launch Power set to 1, and using dimensionless values for these parameters as follows:

 norm
2
=  2 μm V   E0  V μm
 norm =  3 ( μm V )   E02 ( V μm ) 
3 2 2

   
where E0 is the true peak field amplitude in the launch. This approach has the advantage that
the fields are of the order of unity, avoiding difficulties in the graphics display. Similar
normalizations can be done with Unit Power, but are not described here.
• Using field units:
The easiest approach is to assume that all field components have units of V/µm. This choice
requires that: 1) while the unit of E is the traditional MKS unit, all units of length must be
assumed to be µm, and 2) the unit of H is the same as the unit of E.
If the field has units of V/µm, then the nonlinear parameters χ2 and χ3 must have units of
µm/V and (µm/V)2 respectively.
This approach also requires a consistent interpretation of the power contained in the launch
field(s) based on the launch Normalization:

Type Launch Power

None Sets E02 where the electric field E0 [V/µm] multiples
the unnormalized launch field.
Unit Peak Sets E02 where the electric field E0 [V/µm] is the
peak electric field
Unit Power Sets the power of the launch field: By default, the
power is in V2. Setting fdtd_launch_power_unit
= 1 forces the Launch Power to be interpreted in
Watts.

8.D. Linear Mu Properties

The linear mu, or relative permeability, properties can be defined on the Mu Linear tab.

Overview
The Material Editor allows users to create materials with the following definition for linear mu:

RSoft CAD  162

 B = 0 H + M
M = 0   m ( ) H + non-linear terms 

where χm(w) is a frequency dependent linear dispersion term. For the rest of this section, we will
assume that there are no non-linear terms.

Only FullWAVE supports the use of this material property. To enable its use in a simulation,
select the Use Magnetic Materials option in the Advanced Parameters dialog.

Setting Linear Mu Properties

The equations and options for linear mu are exactly analogous to those for dispersive linear
epsilon (except that gain is not supported), and so are not repeated here. See Section 8.C for this
information.

8.E. Non-Linear Mu Properties

The non-linear mu, or relative permeability, properties can be defined on the Mu Nonlinear tab.

Overview
The Material Editor allows users to create materials with the following definition for non-linear
mu:
B = 0 H + M
M = 0 linear term +  m2 H 2 +  m3 ( ) H H 
2
 
where χm is the second-order non-linear susceptibility, and χ3m(w) is the saturable dispersive
third-order non-linear susceptibility

Only FullWAVE supports the use of this material property. To enable its use in a simulation,
select the Use Magnetic Materials option in the Advanced Parameters dialog.

Setting Non-Linear Mu Properties

The equations and options for non-linear mu are exactly analogous to those for non-linear
epsilon, and so are not repeated here. See Section 8.D for this information.

RSoft CAD  163

8.F. Electro-Optic Properties
The electro-optic properties can be defined on the Electro-Optic tab of the Material Editor.

Overview
The Material Editor allows users to create materials whose refractive index is perturbed by the
presence of electrodes in the simulation domain. This perturbation is added to the normal
refractive index of the material. For more information on the use of electrodes and the Multi-
Physics Utility, see the Multi-Physics Utility manual.

Setting Electro-Optic Properties

The options for this material property are:

Eps_xx, Eps_yy, and Eps_zz

Sets the diagonal elements of the DC dielectric constant tensor ε (also known as the RF dielectric
constant) used to compute the spatially dependent DC electric field from any defined electrodes
via the Poisson equation. The zz value of this quantity is currently not used and is provided for
compatibility with a future release.

R_x, R_y, and R_z

Sets the electro-optic coefficients Rx and Ry used to compute the index perturbation from any
defined electrodes. The units of these parameters are µm/V. The z value of this quantity is
currently not used and is provided for compatibility with a future release.

RSoft CAD  164

 Figure 8-7: The Electro-Optic tab in the Material Editor where electro-optic properties are
defined.

8.G. Thermo-Optic Properties

The thermo-optic properties can be defined on the Thermo-Optic tab of the Material Editor.

Overview
The Material Editor allows users to create materials whose refractive index is perturbed by the
presence of heaters in the simulation domain. This perturbation is added to the normal refractive
index of the material. For more information on the use of heaters and the Multi-Physics Utility,
see the Multi-Physics Utility manual.

Setting Thermo-Optic Properties

The options for this material property are:

Figure 8-8: The Thermo-Optic tab in the Material Editor where thermo-optic properties are
defined.

Kappa_xx, Kappa_yy, and Kappa_zz

Sets the diagonal elements of the thermal conductivity tensor κ used to compute the spatially
dependent temperature distribution from any defined heaters via the Poisson equation. The units
of these parameters are W/(µm*K).

RSoft CAD  165

dN/dT
Sets the temperature coefficient dn/dT for index change in units of K-1.

8.H. Stress-Optic Properties

The stress-optic properties can be defined on the Stress-Optic tab of the Material Editor. The use
of this option is licensed separately; contact us with questions.

Overview
The Material Editor allows users to create materials whose refractive index is perturbed by the
stress created during the cooling from the fabrication temperature. This perturbation is added to
the normal refractive index of the material.
When a simulation that incorporates stress is initiated, the displacement is calculated using force-
balanced equations as a function of the temperature differential between operating temperature
(ΔT) and fabrication temperature, which is defined in the Multi-Physics Utility, and the stress
properties defined in the material editor. The spatially-dependent strain distribution ε is then
computed from the displacement. The spatially-dependent stress σ(x) is then computed from this
quantity as:

x   1 − −   x 
  1  
  y  = E  − 1 −   y  +T
   − − 1  
 z    z 
where ε [unitless] is the strain, E is Young’s Modulus, ν [unitless] is Poisson’s Ratio, α [K-1]is
the thermal expansion coefficient, and ΔT [K] is the temperature change. The units of E and the
computed σ(x) values are in inverse units, for example if E is in GPa then σ(x) would be in GPa-
1
. It is important to use the same units for E for all materials in a simulation.
The local index perturbation used in a simulation is computed from the strain ε as:

 nx   P1 P2 P2   x 
    
  ny  =  P2 P1 P2   y 
n  P P1  
 z  2 P2   z 
where P1 and P2 [unitness] are defined in Material Editor and ε is the strain. For more
information on the use of the Multi-Physics Utility, see the Multi-Physics Utility manual.

This material property is supported by all simulation tools. Special care should be given when
using stress with simulation tools that require periodic structures since the resulting perturbation
may render the structure unperiodic. The stress distribution can be viewed using the multi-
physics utilities discussed in the Multi-Physics Utility manual.

RSoft CAD  166

Setting Stress-Optic Properties
The options for this material property are:

Figure 8-9: The Stress-Optic tab in the Material Editor where stress-optic properties are
defined.

Young’s Modulus (E), Poisson’s Ratio (nu), and Thermal Exp. (alpha)
Sets E, ν, and α as defined in the previous section. These parameters are used to compute the
spatially-dependant stress/strain distribution. The output values of the stress will be in the same
units as Young’s modulus.

P1, P2, P3
Sets the strain optic coefficients used to compute the index perturbation as described in the
previous section.

8.I. Semiconductor Properties

The semiconductor properties can be defined on the Semiconductor tab of the Material Editor.
This tab allows users to access the LaserMOD material library and to define certain other
parameters that are required when the effects of electronic transport need to be considered.
Editing the properties in this tab will most likely be necessary only when using RSoft’s Solar
Cell Utility, Tapered Laser Utility, or the Multi-Physics Utility, or when the optical properties of
compound materials are required.

RSoft CAD  167

Setting Optical Semiconductor Properties
The optical properties of a semiconductor are defined on the Eps Linear tab of the Material
Editor. While the options on this tab are described in detail in Section 8.B, it is worth noting
several usage scenarios here.
The refractive index of ternary or quarternary semiconductor material systems, at a specific alloy
composition, can be interpolated between those of their constituent binaries. Interpolation
functions (nkinterp2(), nkinterp4(), etc. described in Section C.F.7) and for extracting the
complex index are available for use in the Eps Linear Tab of the Material Editor. The ‘x’ and ‘y’
arguments for these functions are set by the Alloy X and Alloy Y parameters in the Semiconductor
Tab, respectively. See notes in the next section for the definition of Alloy X and Alloy Y.

Figure 8-10: The Eps Linear tab in the Material Editor for the AlGaAs (ternary) material
system. The complex index is interpolated between those of AlAs and GaAs according to the
“Alloy X” parameter in the Semiconductor Tab, and at the wavelength specified by the
free_space_wavelength symbol.

Setting Semiconductor Properties

For each semiconductor (and certain other materials) available in the RSoft CAD material
library, there is a corresponding LaserMOD material library entry. This entry is selected via the
Material System parameter on the Semiconductor Tab.

RSoft CAD  168

 Figure 8-11: The Semiconductor tab in the Material Editor.

As stated in the previous section, the X Alloy and Y Alloy composition must also be selected for
most semiconductor systems:
• Ternary Material Systems (X only):
The composition is A(x)B(1-x)C. For example, AlGaAs is AlxGa1-xAs.
• Quarternary Material Systems (X and Y):
The composition is A(x)B(1-x)C(y)D(1-y). For example, GaInAsP is GaxIn1-xAsyP1-y.
Furthermore, the doping concentration in a particular region of the device must be specified for
nearly all electronic simulations; donors for n-type regions and acceptors for p-type regions.
There must be a project material in the problem file for each unique combination of material
system, alloy composition, and doping.
The Semiconductor Tab also allows for the specification of how a material region will be treated
by the active simulation; as a (non-optically active) bulk semiconductor, as an optically active
bulk semiconductor, as a quantum well structure, or as an insulator. Lastly, the Edit Table button
allows access to a local symbol table that permits overriding of any of the parameters in the
LaserMOD Material Library.

Including Free Carrier Effects in Silicon

Two options further control the refractive index of Silicon based on the doping levels:

Do not use these options if these effects have already been accounted for in the material model
that you are using (i.e. if the material model has come from a TCAD Sentaurus simulation that

RSoft CAD  169

already includes free-carrier effects). Using these options in such a scenario will result in a
double counting of this effect.

• The complex index perturbation due to free carriers can be included by selecting the Enable
Free-Carrier Index Effects option. This effect requires that a doping concentration be
specified via the Donors (cm-3) or (Acceptors cm-3) option (otherwise the free-carrier
concentration is zero).
• The complex index perturbation is calculated using these equations:
dn = fcn_elcoef*(n^fcn_elexp) + fcn_hlcoef*(p^fcn_hlexp)
da = fca_elcoef*(n^fca_elexp) + fca_hlcoef*(p^fca_hlexp)

where n and p are the doping concentrations in cm-3 Note that dn and da are scaled by
(lambda/ fca_lam0)^ fca_lamexp), and are in dimensionless and cm-1 units respectively.
The ‘fca’ coefficients and exponents can be defined in several ways:
• Using Published Data:
There are two sets of published data available taken from Ref 1 and Ref 2 below. Set
fca_method = 1 to use the data from Ref 1 (default), set fca_method = 2 to use the data
from Ref 2. These symbols should be set in the symbol table in the Semiconductor tab.
• Using Custom Data:
To use custom coefficients/exponents, create a simple text file with the desired
parameters and values. Then, set fca_file to the name of the file in the Symbol table in
the Semiconductor tab. Do not set the fca_method symbol, it takes precedence. An
example file could be:
fcn_elcoef = 0.0
fcn_hlcoef = 0.0
fca_elcoef = 0.0
fca_hlcoef = 0.0
fcn_elexp = 1.0
fcn_hlexp = 1.0
fca_elexp = 1.0
fca_hlexp = 1.0
fca_lam0 = 1.0
fca_lamexp = 0.0

Note that this example shows the default value for each parameter. If a parameter does not
appear in the file, its default value will be used.
• Furthermore, the number of free carrier can be impacted by incomplete ionization. This
model is enabled by the Enable Incomplete Ionization option.

RSoft CAD  170

Using a Custom Doping Profile
It is possible to define the doping profile via data files allowing for custom profiles. The
following symbols control this feature:

Variable Description
dope_file_import Set to 1 to enable the use of custom doping profiles.
dope_file_donor Sets the donor profile.
dope_file_acceptor Sets the acceptor profile.
dope_file_net Sets the net doping profile.

Note that negative values indicate acceptors, while positive values indicate donors. Units are
always cm-3. Also, all profiles should have the same coordinate system as the simulation and
will be applied to the entire domain.

References:
1 Soref, R. et al, IEEE Journal of Quantum Electronics, QE-23 (1987)

2 Nedeljkovic et al, IEEE Photonics Journal 3 (2011)

8.K. Special Materials

Some materials have additional options and/or require additional configuration. These include:
• Graphene
GrapheneCX
GrapheneCY
GrapheneCZ
The ‘Graphene’ material is isotropic and the GrapheneCX, GrapheneCY, and GrapheneCZ
materials are anisotropic. The name of the anisotropic material denotes the out-of-plane or
stacking direction: for example, GrapheneCY is oriented so that the Graphene layer is in the
XZ plane. In general the isotropic Graphene model is preferred, but depending on your
polarization and Graphene orientation it may be necessary to use one of the anisotropic
models. Each design file can contain multiple Graphene models, each with different
properties.
These Graphene models require additional material properties that are set in the special
symbol table in the semiconductor tab described in Section 8.I . Click Edit Table on the
Semiconductor tab to change these values. Note that these material models are based on the
graphene_real() and graphene_imag() functions described in Section C.F.7: see this

RSoft CAD  171

section for a complete description of the mathematical model used and how these parameters
are used to calculate the Graphene properties.
While the built-in material uses reasonable default values for these parameters, it is critical
that you use values that correspond to your specific situation.

Parameter Description & Notes

U_
Sets the chemical potential [eV]
G1_
Sets the intraband Γ1 [eV]. Note:
• Γ [eV] = hbar* Γ [1/sec]
To convert a time constant τ, typically Γ = 1/(2τ)
G2_
The interband Γ2 [eV]
• See notes above for Γ units and conversions
Setting G2_ to -1 makes G2_ = G1_.
T_
Sets the temperature [K]
H_
Sets the Graphene thickness H [μm], typically set to 0.0007
EpsC_
Reference Eps (εref ) for in-plane calculation described in
Section C.F.7
E_
Reference Eps (εref) for out-of-plane direction, only used for
anisotropic Graphene models

RSoft CAD  172

9

Non-Uniform Grids

The RSoft CAD uses an intelligent grid generation algorithm to produce a simulation grid based
on the structure(s) defined in the CAD environment, as well as a variety of grid options
described in detail later in this chapter. The grid generation algorithm automatically calculates a
uniform or non-uniform grid best suited for the defined structure. The grid is composed of
rectangular shaped cells, or elements, of varying sizes, which cover the entire computational
domain, including any PML. Additionally, FemSIM supports a hybrid triangular and rectangular
grid which is described in more detail in the FemSIM manual.

Please note that not all options are available for all simulation programs. BeamPROP and
FemSIM’s capabilities are described in this chapter, FullWAVE’s non-uniform capabilities are
described in the FullWAVE manual.

9.A. Grid Basics

The entire grid is built on a base unit called grid cells. The definition of a grid cell differs for
BeamPROP and FemSIM, specifically how fields are calculated for each cell:
• BeamPROP: A finite-difference algorithm is used to represent field values at specific grid
points. These grid points are located at the center of each grid cell as shown in Fig. 9-1a.
• FemSIM: A finite element method is used which represents the field over the entire cell as a
function of pre-determined basis functions as shown in Fig. 9-1a. Furthermore, FemSIM

RSoft CAD  173

 supports a subdivision of cells into triangular elements. This option is described in further
detail in the FemSIM manual.

Figure 9-1: The grid cell definition for a) BeamPROP and b) FemSIM. The dashed red line
represents a material boundary, the solid black lines represent cell boundaries, and the gray
dots represent the grid points. The FemSIM cells are shaded to indicate that the field is
defined over the entire cell.

9.B. Advanced Grid Controls

The main spatial domain and grid controls are located in the Simulation Parameters dialog. The
Enable Nonuniform option in the Simulation Parameters dialog controls whether a non-uniform
grid is used. This non-uniform grid is based on the structure defined in the RSoft CAD, the
materials used, the simulation wavelength, and several other options. The grid options are set in
the Advanced Grid Controls dialog (shown in Fig. 9-2) which is opened by clicking the Grid
Options… button in the Simulation Parameters window. Each coordinate has separate grid
controls and are described in this section. Note that it is possible to overspecify the grid size, the
final simulation grid used is the union of all grid settings such that the smallest grid specified at
any location is used.

RSoft CAD  174

 Fig 9-2: The BeamPROP Advanced Grid Controls dialog where advanced grid options are
set. Note that FemSIM contains several additional options which are addressed in the
FemSIM manual.

RSoft’s grid generation algorithm is capable of creating a non-uniform rectangular grid based on
the structure defined in the CAD and other grid options that allow user control over the size and
position of grid cells. This section describes these grid options, which include: interface
alignment, grid size at interfaces, grid grading, and required minimum divisions within a region.

Spatial Domain:
The X, Y, and Z Domain Min and Domain Max options set the simulation domain and are
duplicates of the same options in the main Simulation Parameters dialog. See the BeamPROP
and FemSIM manuals for a description of these options.

Grid Sizes:
The simulation grid can be defined in several ways, choose the method that is best for your
specific application. The simplest method to use a non-uniform grid is to set the Grid Size and
the Enable Nonuniform options. It is possible to over specify the grid size, the final simulation
grid used is the union of all grid settings such that the smallest grid specified at any location is
used.
When using a non-uniform grid, the simulation grid is smoothly varied between regions using
the Grid Grading option described below. If needed, the grid within a particular component or
material can be set directly via the tags described in Section 9.C.
• Grid Size (Bulk)

RSoft CAD  175

 This is the primary control for grid generation and are duplicates of the similarly named Grid
Size option in the main Simulation Parameters dialog. See the BeamPROP and FemSIM
manuals for a description of this parameter.
• Grid Edge Size or Grid Edge Factor

These options help accurately model field variations at material interfaces which are typically
larger than in bulk regions. Grid Edge Size sets the grid size at the edge of a material
interface in spatial units [μm], whereas Grid Edge Factor sets the edge grid relative to the
‘bulk’ grid. Only one method may be used per spatial coordinate, use the slider control to set
which option is used. Grid Edge Factor is used by default, use the radio buttons to select
Grid Edge Size if needed. Note that Grid Edge Factor should be between 0 and 1.
• Minimum Divisions
This option sets the minimum number of grid cells used between two material interfaces and
has a default value of 1. This option should be set to an integer (1 or greater); it is
recommended to set it equal to the thickness of the region of interest divided by the desired
“average” grid size.

Grid Grading:
The Grid Grading option automatically grades, or varies, the grid size between regions. The
Grading Ratio sets the rate at which adjacent grid sizes vary and defaults to sqrt(2). A
recommended guideline is to set this parameter to a value between 1.1 and 2 depending on the
structure. A value that is too small increases the computation burden and a value that is too large
can create accuracy issues.

Fig 9-3: A sample non-uniform grid that uses grid grading between the small grid in the
structure core and the background regions.

RSoft CAD  176

Additional Grid Options:
The Interface Alignment option controls how grid cells are aligned to interfaces between
material regions. This option has several possible choices, including Default, None, Centered
Cells, or Straddled Cells. The default is Centered Cells, setting to None can be useful to
approximate a uniform grid along a particular axis when Enable Nonuniform is enabled.

9.C. Overriding Grid Settings for Specific

Components or Materials
While the Grid Size and Grid PPW options give adequate control over the grid for most cases, it
can sometimes be advantageous to set the desired grid size(s) within a particular component or
material. This allows, for example, more/less grid points to be used in a particular region of
interest.
All grid options described above can be enabled/disabled for particular components or materials
using tags. Tags can be assigned to a particular component directly or via the Edit Default Tags
option in the Edit menu item in the RSoft CAD. Multiple tags can be set, use spaces between
tags. The final grid will be the union of the global grid settings and any tags set (the tags take
precedence). Note that a non-uniform grid must be used to enable these options.
The tags that control grid related parameters are:

See sections above for details about what each option does, including recommended usage.

Tag Description
grid_size_x These tags correspond to the X, Y, and Z Grid Size options
grid_size_y
which sets the grid size in μm units.
grid_size_z
grid_ppw_x These tags correspond to the X, Y, and Z Grid PPW options
grid_ppw_y
which sets the grid size in points-per-wavelength units.
grid_ppw_z
grid_edge_x These tags correspond to the X, Y, and Z Grid Edge options
grid_edge_y
which sets the grid edge in μm.
grid_edge_z
grid_edge_factor_x These tags correspond to the X, Y, and Z Grid Edge Factor
grid_edge_factor_y
options which sets the grid edge factor.
grid_edge_factor_z
grid_mindiv_x These tags correspond to the X, Y, and Z Grid Minimum
grid_mindiv_y
Divisions options.
grid_mindiv_z
grid_bulk_nonuniform_x These tags correspond to the X, Y, and Z Grid Grading

RSoft CAD  177

 grid_bulk_nonuniform_y options which enable/disable grid grading. Set to 0 to
grid_bulk_nonuniform_z
disable, 1 to enable.
grid_ratio_x These tags correspond to the X, Y, and Z Grid Ratio options
grid_ratio_y
which sets the grid grading ratio
grid_ratio_z

Grid Tag Example:

To set the X grid size to 35 points-per-wavelength and the Y grid size to 1 nm within a particular
component in the RSoft CAD, set the Tags option in the Additional Component Properties
dialog to:
grid_ppw_x=35 grid_size_y=0.001

9.D. Viewing the Grid

The simulation mesh can be viewed by pressing the View Grid button in either the Simulation
Parameters window or within the Advanced Grid Controls window. For 3D simulation, the
cross-section of the mesh can be selected via the Cut option. Note that PML will be displayed
with the mesh.
Note that the grid can also be viewed on top of the refractive index using the Overlay Grid
option.

RSoft CAD  178

10

Scripting & Scanning

The RSoft device simulation tools share a comprehensive set of scripting features for layout,
simulation, parameter scanning/optimization, and data manipulation. Layout scripting is done via
a Python API, simulations scripting is done via the user’s operating system (ANY language), and
data manipulation is done via a built-in set of utilities and/or a Python API. Additionally, MOST
can be used to automate tedious scanning and optimization problems.

10.A. Layout Scripts

There are two main ways to script device layouts:
• Using scripts within the RSoft CAD
• Generating design files completely via script
The method you use will depend on your exact need. Each method has benefits, and the methods
can also be combined as needed.

10.A.1. Using Scripts within the RSoft CAD

The hierarchical approach of using scripts with the RSoft CAD is beneficial for many situations.
The control file can pass relevant design parameters to the child file(s). Child files can be RSoft
.ind files, GDS files, or Python scripts. This modular approach allows portions of the design to
easily change without having to continuously regenerate the entire design. See Appendix H for a
description of the Python API to generate design files, including supported Python versions.

RSoft CAD  179

 Figure 10-1: A fiber-grating coupler shown in the RSoft CAD. The description of the
structure is below. The colors are chosen to show the etch depth, red is a standard waveguide,
yellow can be either a partial or full-etch.

As an example, consider a standard fiber-grating coupler shown above. This design consists of:
• A series of structure properties, materials, and simulation settings set via symbols such as the
input waveguide width, number of arc segments, grid sizes, etc. (not shown)
• Two static components (straight red waveguide at bottom and tapered yellow waveguide)
that do not need to be scripted, their settings can easily be taken from the symbols in the
control file.
• A launch field and several monitors for measurement purposes that do not need to be
scripted, their properties can again be easily set by the symbols in the control file.
• A series of dynamic components (arc segments shown in red) that need to by dynamically
produced based on the number of rings setting in the control file.
For this example, we define everything except the yellow arc segments in the control file and use
a Circuit Reference block to dynamically call the Python API to produce the arc segments.

See Appendix H for a description of the Python API to generate design files, including supported
Python versions.

RSoft CAD  180

Passing Parameters to a Python Script
There are several ways to pass settings and/or symbols from the control file to the Python script.
The method you choose depends on the parameter or setting type as well as the which method is
more efficient. There are two types of parameters:

While any parameter can be passed as an argument, it is usually better to use the other methods if
possible to keep the Python code simple, clean, and reusable.

• Parameters that, when changed, require a new design file to be generated. These parameters
typically control how many components are in the design file or something similar. This type
of parameters must be passed to the script as script arguments. See Section 6.F.6 for more
details about passing script arguments to the Python script as well as the examples below.
• Parameters that can be changed through Symbols. These parameters typically control the
properties of components. This type of parameters can be passed to the .ind file generated
by the script through the options in the circuit reference dialog and/or the Local Symbols
table for the circuit reference. See Section 6.F.2 for details about setting options in the circuit
reference dialog, see Section 6.F.6 more details about using hierarchy exports with the
Python API, and the examples below.

Example 1-1: Grating Coupler

We begin by going through the example described above in more detail. The files for this
example are located here:
<rsoft_dir>\examples\RSoftCAD\scripting\Python\grating_coupler\

Open the file grating_coupler.ind. This is the master control .ind file that uses the Python
script grating_coupler.py which generates the series of yellow arcs.
We will not cover the creation of the full master control file grating_coupler.ind here, instead
we will focus on the circuit reference component that uses the Python API to draw the yellow arc
segments.

Creating a Circuit Reference using a Python script

Right-click on the yellow arcs to open the Circuit Reference dialog to see how the Python script
is loaded and how the necessary parameters are passed.

RSoft CAD  181

 Figure 10-2: The Circuit Reference dialog used to call the Python script.

Note that the External File is set to grating_coupler.py which will call the Python script to
generate the layout. This script (which we will go through below) generates an .ind file
containing the arc segments.

Passing Parameters to the Script and/or Generated .ind File

Before looking at the script, let’s consider the symbols and settings that are needed to define the
grating arcs:

Symbol/Property Description
N Sets the number of rings
Period Sets the period/spacing of the rings
Angle Sets the half-angle of the rings
Rmin Sets the minimum radius of the rings
Material Properties Sets the material of the rings
Period Sets the periodicity of the rings
Duty Sets the duty cycle of the rings
Depth Sets the height of a ring
Merge Priority Sets the merge priority of the ring, used because the rings have a
lower index than the Si waveguide
Display Color Sets the color of the rings

RSoft CAD  182

There are several ways to pass settings and/or symbols from the control file to the Python script:
• Circuit Reference Component Settings: Many settings can be simply set in the properties
dialog for the Circuit Reference. As shown in Fig. 10-2, we have set the Material Properties
of the rings to SiO2 which matches the upper cladding, we have set the Component Width to
Period*Duty, and Component Height to Depth. These symbols are evaluated in the parent
file and will be automatically passed to the .ind file created by the Python script. Also set
here is the Merge Priority and Display Color.
• Hierarchy Exports: Other settings that can be simply passed to the .ind file created in the
Python script are Period, Duty, Rmin, and Angle. These symbols have been set in the Local
Symbols table (click Local Symbols to open) shown below. They have also been set as
Hierarchy Exports in the Python script so that they automatically appear in the Local
Symbols table.

Figure 10-3: The Local Symbols table for the Circuit Reference.

• Script Arguments: This method is required for properties that cannot be simply passed as
symbols. The only property of the grating coupler that this is true for is N, the number of arcs.
Script arguments are set in the Local Symbol table with names that start with the prefix ArgX.
Here we have one symbol Arg1_NumRings which is set to NumRings. Had there been more
arguments, they should start with names like Arg2_, Arg3_, etc. We’ll note this later, but the
script should have default values for arguments in case they are not defined in the Local
Symbols table.

The Python Script

Open the file grating_coupler.py in the text editor of your choice or by double-clicking on the
red arcs in the RSoft CAD. We will go through this script line by line.

RSoft CAD  183

You can select the editor used when opening from the CAD by setting the environment variable
RSOFT_EDITOR to the text editor of your choice.

Figure 10-4: The Python script for the grating coupler.

See Appendix H for a description of the Python API to generate design files, including supported
Python versions.

• Line 2: Loads the RSoft API

RSoft CAD  184

• Line 4: Define the rsoft_layout() function, the recommended function that will be
automatically run by the hierarchy block to build the .ind file. Note that the arguments to the
rsoft_layout() function will be automatically parsed from the arguments used when
calling the script, automatically set by the Local Symbols that start with the prefix ArgX as
described above. In this case there is just one argument NumRings which has a default value
of 10. It is recommended to always define a default value so that the script will run even if
the argument symbols are not defined in Local Symbols.
• Line 7: Creates the RSoftCircuit() object c.
• Lines 10-18: Sets various symbols and values, and then sets four of these symbols as
hierarchy exports so that they appear in the Local Symbols table.
• Lines 20-28: Defines a for loop to create the arc segments. The starting (x,z) value, radius,
initial and final angle, and width are calculated and then the arc is created using the
add_arc() function.

• Lines 30-39: Defines the starting sector or wedge based on Rmin.

• Line 42: Returns the RSoftCircuit object

Putting it All Together:

Once defined, and once the various parameters and symbols are defined, the grating coupler can
be used just as any other design file. Changing the options/symbols in the master file
grating_coupler.ind will control the entire design. For example, changing N in the symbol
table of the master file will change the number of rings produced by the Python script. Note that
the script is not run every time the display is refreshed for efficiency purposes, you can force the
script to be re-run by selecting View/Refresh Hierarchy from the CAD menu. Also note that the
design is always refreshed before running a simulation.

Example 1-2: Binary Phase Mask

This example uses a data file to define a patterned binary phase mask element. The data file
defines the local thickness of the structure for one period (unit cell). Specifically, a 0 value in the
data file corresponds to fully etching that position, a value of 1 corresponds to not etching that
position. The final grating is shown below.

RSoft CAD  185

 Figure 10-5: Top-view of an example binary phase mask.

This example uses both Python APIs (RSoftUserFunction() and RSoftCircuit()). We begin
by going through the example described above in more detail. The files for this example are
located here:
<rsoft_dir>\examples\RSoftCAD\scripting\Python\binary_phase_mask

Open the file binary_phase_mask.ind. This is the master control .ind file that uses the Python
script binary_phase_mask.py which generates the yellow boxes based on the data in the file
binary_phase_mask.dat.

We will not cover the creation of the full master control file binary_phase_mask.ind here,
instead we will focus on the circuit reference component that uses the Python API to draw the
yellow box segments. The only input is the data file binary_phase_mask.dat, the script
automatically scales the data to fit within the a single unit cell (defined by the symbols PeriodX
and PeriodY variables) in the binary_phase_mask.ind file. The data file is passed as an
argument in the Local Symbol table for the circuit reference.

The Python Script:

Open the file binary_phase_mask.py in the text editor of your choice or by double-clicking on
the yellow boxes in the RSoft CAD. We will go through this script line by line.

RSoft CAD  186

 Figure 10-6: The python script to generate the binary phase mask.

See Appendix H for a description of the Python APIs to read data files and generate design files,
including supported Python versions.

• Lines 2-3: Loads the RSoft APIs.

• Line 5: Define the rsoft_layout() function, There is one argument, data_file. This will
be set by the symbol Arg1_data_file in the Local Symbol table of the circuit reference.
• Line 8: Creates the RSoftCircuit() object c.
• Lines 16-18: Reads data from the file using the RSoftUserFunction() class into three
arrays: the x coordinates, y coordinates, and 2D data. Note that an if function is used to
protect against the case of a missing file

RSoft CAD  187

• Lines 20-24: Define several symbols, including DX and DY which represent the X and Y size
of each of the pixels.
• Lines 27-32: Loop through all elements in the data array and, if set to 1, add a segment at that
position.
• Line 35: Return the RSoftCircuit object

Example 1-3: Metalens Design

This example uses an Python script to draw a simple Metalens design. The design is
characterized by a data file which contains hole radius as a function of lens radius. The final
structure is shown below.

Figure 10-7: The simple metalens design shown in the RSoft CAD.

This example uses both Python APIs (RSoftUserFunction() and RSoftCircuit()). We begin
by going through the example described above in more detail. The files for this example are
located here:
<rsoft_dir>\examples\RSoftCAD\scripting\Python\metalens

Open the file [Link]. This is the master control .ind file that uses the Python script
[Link] which generates the yellow boxes based on the data in the file
diameter_ratio.dat.

We will not cover the creation of the full master control file [Link] here, instead we will
focus on the circuit reference component that uses the Python API to draw metalens.

RSoft CAD  188

 Figure 10-8: The data file used to defined hole radius as a function of lens radius.

The Python Script

Open the file [Link] in the text editor of your choice or by double-clicking on the red arcs
in the RSoft CAD. We will go through this script line by line.

RSoft CAD  189

 Figure 10-9: The Python script used to draw the simple metalens.

See Appendix H for a description of the Python API to generate design files, including supported
Python versions.

• Line 2: Loads the RSoft API

• Line 3: Loads the math library so we can use the sqrt() function
• Line 4: Define the rsoft_layout() function, the recommended function that will be
automatically run by the hierarchy block to build the .ind file. Note that the arguments to the
rsoft_layout() function will be automatically parsed from the arguments used when
calling the script, automatically set by the Local Symbols that start with the prefix ArgX as
described above. Here we set the data file, Period, and Diameter.
• Lines 10-14: Create a new circuit and add symbols and exports.
• Lines 17-18: Read in data file
• Line 21: Calculate the array dimension based on the period and diameter
• Lines 24-30: Draw the array, using the data from the file.
• Lines 33: Return the circuit object.

RSoft CAD  190

10.A.2. Generating Design Files Completely via Scripts
There are some situations where it is benefiical to generate design files completely via scripting
and not invoke the RSoft CAD. Note that all functionalities may not be available when operating
in this mode, please contact technical support with questions. Also note that the designer will
have to make choices such as grid sizes and perform device optimization as usual.

Example 2-1: 1x2 MMI

As an example, consider a 1x2 MMI device. Here we will create this device completely via the
Python API.
<rsoft_dir>\examples\RSoftCAD\scripting\Python\mmi_1x2\

The Python file is make_mmi_1x2.py and can be run by moving to the directory above and
running this command:
rspython make_mmi_1x2.py

The file mmi_1x2.ind will be generated.

Figure 10-13: The 1x3 MMI shown in the RSoft CAD.

The Python Script:

Open the file make_mmi1x2.py in the text editor of your choice, here is a description of each
line:

RSoft CAD  191

See Appendix H for a description of the Python API to generate design files, including supported
Python versions.

• Line 1: Loads the RSoft API

• Line 4: Defines the name of the file to produce.
• Lines 10-17: Defines the properties of the 1x2 MMI, these properties will become symbols.
Note that here we use a Python dictionary to store the options as an illustration, there are
many ways to create this type of script
• Lines 20-44: Sets various simulation options, these properties will become symbols. Again
we use the same Python dictionary.
• Lines 47-50: Creates the RSoftCircuit object, loads a standard Silicon settings file, and sets
the symbols defined in previous lines.
• Lines 53-59: Adds components in to draw the MMI. Note that the position of the components
is not explicitly defined, these components will be attached together to form the structure in
the next section.
• Lines 61-66: The components are attached to form the 1x2 MMI.
• Lines 69-71: Port monitors are created and attached.
• Line 78: Write the .ind file.

Example 2-2: Waveguide Crossing from GDS

Another design flow is to load the structure from a GDS file. Each layer has a corresponding
position, height, and material based on the foundry stack data.
<rsoft_dir>\examples\RSoftCAD\scripting\Python\
waveguide_crossing_from_gds\

The structure is a waveguide crossing contained in the file [Link], and

contains two layers: layer 88/0 which corresponds to a standard Si rib waveguide, and layer 89/0
which corresponds to a partial-etch Silicon waveguide.
The Python script is make_WaveguideCrossing_from_gds.py and can be run by moving to the
directory above and running this command:
rspython make_WaveguideCrossing_from_gds.py

The file [Link] will be generated.

RSoft CAD  192

 Figure 10-14: The waveguide crossing shown in the RSoft CAD.

The Python Script:

Open the file make_WaveguideCrossing_from_gds.py in the text editor of your choice, here is
a description of each line:

Note that in this example we hardcode the layer properties (vertical position, height, and
material), see the next example for how load stack data from a particular PDK. Also, see
Appendix H for a description of the Python API to generate design files, including supported
Python versions.

• Line 1: Loads the RSoft API

• Line 4: Defines the name of the file to produce.
• Line 10: Defines the GDS file to use
• Lines 13-14: Defines the layers to import, and the layer where ports should be added.
• Lines 17-19: Defines the layer properties such as the base (Y position), height, and material.
These arrays are tied to the layers defined above.
• Lines 21-50: Defines the simulation parameters for the waveguide crossing. Note that we use
a Python dictionary to store the options as an illustration, there are many ways to create this
type of script. Also, note that the transverse geometry is set by the GDS file.

RSoft CAD  193

• Lines 53-56: Creates the RSoftCircuit object, loads a standard Silicon settings file, and sets
the symbols defined in previous lines.
• Lines 59-79: Adds layers, setting the position, height, and material. The ports are added ot
the correct layer.
• Line 64: Write the .ind file.

Example 2-3: Waveguide Crossing from GDS using PDK Stack Data
This example is an extension of Example 2-2 above, using stack data directly from the
referenceOpto PDK. This concept can be generalized to any PDK.

Note that you will need access to the referenceOpto PDK to run this example.

The files associated with this tutorial can be found here:

<rsoft_dir>\examples\RSoftCAD\scripting\Python\
waveguide_crossing_from_gds_refopto\

The Python script is make_WaveguideCrossing_from_gds_refopto.py and can be run by

moving to the directory above, ensuring that your environment points to the refOpto PDK, and
running this command:
rspython make_WaveguideCrossing_from_gds_refopto.py

The file roWGCrossing_refopto.ind will be generated.

The Python Script:

Open the file make_WaveguideCrossing_from_gds_refopto.py in the text editor of your
choice. The script is the same as Example 2-2 with the following differences:
• The layers info is not set at the top of the script, this data will come from the pdkinfo file
(refopto stack info)..
• A settings file from the PDK is used.
• The code to set the layer properties is slightly different. Instead of accessing the layer info
arrays, the settings are made via the pdkinfo() function. The data in the refOpto pdkinfo file
are named by layer numbers, for example L_88_0_material = “Si” and L_89_0_height =
0.22,. The script builds strings based on the layer number, automatically loading the stack
data when the .ind file is opened or run.

RSoft CAD  194

10.B. Parameter Scanning and Optimization
The RSoft CAD allows users to automatically scan over design parameters to find an optimal
design. This is done by using RSoft’s MOST package. Details about the usage of MOST can be
found in the MOST manual.

10.C. Data Manipulation

The RSoft tools produce many types of data, and in many cases this data is enough for the
designer’s needs. There are cases, however, where it can be useful to manipulate the data for
post-processing or other purposes. This can be done through WinPLOT, command line utilities,
and/or the RSoftUserFunction() Python API.
• Manipulating data with WinPLOT:
WinPLOT, the plotting software included with the RSoft tools, provides options to scale,
translate, transform, and perform other operations on data before plotting. Note that this is for
display purposes only.
As an example, the /powy2 command will square data before display, the /sx4 command
will scale the X data by 4 before display, and the /try6 command will translate the Y data
by 6 before display. See the WinPLOT manual for more details about these and other
supported commands.
• Using Command Line Utilities:
The RSoft tools include several command line utilities which are documented in Appendix E.
These utilities can be used to calculate overlap integrals (bdutil -i), far-field
transformations (bdutil -f), Cropping/re-gridding/slicing data files (bdconv), bitmap image
to RSoft data format (bmp2ind), conversion to/from ray-tracing formats for Synopsys’ CODE
V, LightTools, and many other functions.
• Using the RSoftUserFunction() Python API:
This API allows users to pragmatically interact with RSoft data files. All standard RSoft file
formats described in Appendix B are supported, including 1D, 2D, 3D, real, complex,
uniform, and non-uniform data formats. Note that files in the TCAD Sentaurus TDR format
are not currently supported. See the example in Section 10.A.2 for an example, see Appendix
H for a complete description of this API, including supported Python versions.

RSoft CAD  195

10.D. Simulation Scripts
Simulation scripts are built around the fact that all of the RSoft tools can be run with complete
functionality from the command line. This section will explore this functionality, and then
provide several examples of such scripts in different programming languages.

The use of simulation scripts for most applications can be automated with RSoft’s MOST
package.

10.D.1. Running the Simulation Programs from the Command

Line
The ability to run the RSoft simulation tools from the command line forms the basis for all
simulation scripts. In this section, we will illustrate this concept and give examples for each
product.

Syntax
The basic syntax for running the simulation programs from the command line is:
SIM_PROGRAM index-file-name [name=value name=value ...] @<symfile>

Where SIM_PROGRAM is the name of the simulation program (the names are listed in Conventions
section in the Preface), the index-file-name is the name of the *.ind file to be simulated, the
name/value entries are symbol table variables and values to be applied to the *.ind file, and
<symfile> is an optional file that contains a list of additional symbols. For example, to run the
FullWAVE example file [Link], which is located in the directory
<rsoft_dir>\examples\fullwave, with an output prefix of ‘run1’, issue the following command
from the directory where the *.ind file is located:
fullwave [Link] prefix=run1

For Linux systems, the command will be:

xfullwave [Link] prefix=run1

Examples for other tools are below. If you get errors that the system can not find the program or
application, read through RSoft Installation Guide to ensure that you have set up the PATH
properly.

Controlling how Simulation Windows Behave

Several options exit can be used to control how simulation windows behave. To indicate that the
simulation window should start minimized, use the –min switch. Similarly, to indicate that the

RSoft CAD  196

simulation window should be run as a background task, use the –hide switch. For example, to
run FullWAVE in the background, use a command like:
fullwave –hide [Link] prefix=run1

When using scripts it can be very useful to close the simulation window automatically after a
simulation has completed so the control is returned to the script and the next item can be run. To
do this, set the variable wait to 0 like:
fullwave –hide [Link] prefix=run1 wait=0

The wait option can be set directly in the symbol table, but since it is only really useful when
scripting it makes most sense to add it to the command line. While this option will work for all
the RSoft simulation tools, MOST uses a slightly different syntax:
rsmost -nowait [Link]

Try It!
The best way to understand this concept is to try it. To open a command prompt on Windows,
either choose the ‘Command Prompt’ option from the start menu, or choose Run from the Start
menu and give the command cmd. To open a command prompt on a Linux system, open a shell
window.
Please choose the appropriate product(s) from the list below. Also, note that the following
commands are for Windows systems. Linux users should consult Section 1.C for the appropriate
program name.

In all these cases, a simulation should open and save the results with a prefix of run1.

• BeamPROP
Move to the directory <rsoft_dir>\examples\beamprop and give the command:
bsimw32 [Link] prefix=run1

• FullWAVE
Move to the directory <rsoft_dir>\examples\fullwave and give the command:
fullwave [Link] prefix=run1

• BandSOLVE
Move to the directory <rsoft_dir>\examples\bandsolve and give the command:
bandsolve [Link] prefix=run1

• GratingMOD
Move to the directory <rsoft_dir>\examples\gratingmod\Tutorial\TutA1 and give the
command:
grmod [Link] prefix=run1

RSoft CAD  197

• DiffractMOD
Move to the directory <rsoft_dir>\examples\diffractmod and issue the command:
dfmod [Link] prefix=run1

• FemSIM
Move to the directory <rsoft_dir>\examples\femsim and issue the command:
femsim [Link] prefix=run1

• ModePROP
Move to the directory <rsoft_dir>\examples\modeprop\plasmon and issue the command:
modeprop [Link] prefix=run1

10.D.2. Creating a Simulation Script

An example of a basic simulation script would be a simple parameter scan. This section will
present the same script in two different languages: as a batch file and in Perl.

The scripts in this section are for BeamPROP but the concepts are the same for other engines.

Using a Batch File

A batch file is simply a text file which contains a list of commands to be run sequentially from
the command line. This example illustrates the process of finding an optimal gap for a directional
coupler using BeamPROP. We will use the file [Link] located in
<rsoft_dir>\examples\beamprop. Several notes about this script:

• This script will perform simulations with values of the variable separation from 1 µm to a
value of 10 µm with a step of 1 µm.
• Data will be stored with a prefix of runXX where XX corresponds to the simulation number.
• This script will not produce a scan result like MOST. To do this, this script would have to be
modified using the scan_output, scan_prefix, and scan_variable variables used in the
Perl script below. For a simple case like this, it would be easier to use MOST; this example is
only meant to illustrate basic scanning concepts.
The batch file is:
bsimw32 [Link] wait=0 prefix=run01 separation=1
bsimw32 [Link] wait=0 prefix=run02 separation=2
bsimw32 [Link] wait=0 prefix=run03 separation=3
bsimw32 [Link] wait=0 prefix=run04 separation=4
bsimw32 [Link] wait=0 prefix=run05 separation=5
bsimw32 [Link] wait=0 prefix=run06 separation=6
bsimw32 [Link] wait=0 prefix=run07 separation=7
bsimw32 [Link] wait=0 prefix=run08 separation=8

RSoft CAD  198

 bsimw32 [Link] wait=0 prefix=run09 separation=9
bsimw32 [Link] wait=0 prefix=run10 separation=10

To run this script, save it in a text file with extension *.bat and run it from the command line.

10.D.3. Using Schedulers

The RSoft tools can be integrated into a scheduler using the command line syntax described
above. See manual for tools that support clustering (FullWAVE FDTD and MOST) for more
specific information about integrating schedulers.

RSoft CAD  199

11

Transverse Mode Solving

RSoft’s Passive Device Design Suite contains several mode solvers that can be used to compute
transverse modes of a structure with an arbitrary index cross section. These modes can then be
used as a launch file or as needed. Two of these methods are based on the Beam Propagation
Method (BPM), another (TmmSIM) is based on the Transfer Matrix Method (TMM), and
another (FemSIM) is based on the Finite Element Method (FEM).

The mode solvers based on BPM and TMM can be used by all users who are licensed for any of
RSoft’s passive device simulation tools. The FEM-based mode solver (FemSIM) is licensed
separately.

11.A. BPM-Based Mode Solvers

All users who use the RSoft CAD can use these mode solvers; a license for BeamPROP is not
needed. This section provides a brief overview of these mode solvers; a full discussion: a
technical background can be found in Chapter 2 (including many references) and usage details in
Chapter 6 of the BeamPROP manual. An online form of this manual can be opened via the Help
menu item in the CAD window and a pdf format can be found in the <rsoft_dir>/docs
directory.

RSoft CAD  200

The RSoft CAD includes two mode solvers that are based on the Beam Propagation Method
(BPM), the same algorithm that is used by RSoft’s BeamPROP simulation engine. The two
BPM-based mode solvers are the iterative method and the correlation method.

11.A.1 Basic Theory of BPM-Based Mode Solving

The BPM algorithm is an approximate numerical approach to solving the exact wave equation
for monochromatic waves, and is based on the scalar steady-state Helmholtz equation. Assuming
that most of the variation in the field is due to the phase variation along the propagation axis, and
applying the slowly varying envelope approximation results in the basic BPM equation:

u i   2u  2u 
=  2 + 2 + k −k u
z 2k  x y
2 2
( )

where k-bar is the reference wavenumber and is a constant number chosen to represent the
average phase variation. This equation is solved numerically by first specifying a launch field at
z=z0, and then integrating in z to obtain the field for z>z0.
During a mode calculation, it should be assumed that the structure is uniform along z, and so the
propagation can be described in terms of the modes of the structure and their propagation
constants. Since we are only going to illustrate the concepts of mode solving, we will consider a
2D propagation; the concepts are exactly the same in 3D.

Iterative Method
Propagation of an electromagnetic field through a structure can be expressed via a simple phase
variation based on the propagation constants, βm, of each mode [ϕm(x)] of the structure. The
iterative method is based on imaginary-distance BPM, and propagates along an imaginary z axis
by replacing the longitudinal coordinate z by z'=iz so that each mode experiences an artificial
exponential increase:

 ( x, z ) =  cmm ( x ) exp  m z 
m

This implies that each mode will exponentially grow with a growth rate equal to its real
propagation constant βm. The mode with the highest propagation constant, which is conveniently
the fundamental mode (m=0), will therefore grow faster than all other modes in the structure
causing the simulation results to converge to the field profile of the fundamental mode. The
propagation constant can then be obtained using a variational-type expression.
Higher order modes can be obtained by using an orthogonalization procedure to subtract
contributions from lower order modes while performing the propagation. Of course, it is
important to launch an arbitrary field which excites all the interesting modes in the structure.
Issues such as optimal choice of launch field, reference wavenumber, grid size(s), etc., are
discussed in the BeamPROP manual. This method is not valid for leaky or lossy modes.

RSoft CAD  201

Correlation Method
The correlation method launches an arbitrary field into the structure and propagates via normal
BPM. During the propagation the following correlation function between the input field and the
propagating field is computed:

P ( z ) =  in* ( x )  ( x, z ) dx

This function can also be expressed, using an expansion of the propagating field in the modes to
be found, in a form such that a Fourier transform of this function produces a mode spectrum with
peaks at the modal propagation constants. To obtain the mode profiles, a second BPM
propagation is performed in which the known propagating constants are beat against the
propagating field.
Additional information on this technique can be found in the BeamPROP manual. While the
correlation method is generally slower than imaginary-distance BPM, it can sometimes be
advantageous because it can be applicable to problems that are difficult or impossible for
imaginary distance BPM, such as leaky or radiating modes.

11.A.2. Using the BPM Mode Solvers

Mode solving via BPM can be thought of as a four step process. These steps are:

Step 1: Create the index structure in the CAD.

In order to solve for the mode of a structure, the structure has to be created. The structure should
be uniform in Z, and have the desired index profile. Use the Display Material Profile option to
check that the profile is indeed correct.

Step 2: Choose the appropriate mode solver and related options.

As discussed above, BeamPROP contains two mode solvers which can be accessed by clicking
the Compute Modes icon in the left toolbar of the CAD. The one you choose depends on the
type of modes you want to find. Generally:
• The iterative method is the default choice and is recommended for most standard waveguide
problems. The calculation will stop when the desired modes are found.
• The correlation method can have advantages for problems such as antiguiding, leaky, lossy,
or radiating modes, or when large numbers of modes need to be calculated. The calculation
consists of two full BPM simulations.
The desired mode solver, along with several related options such as the modes you want to solve
for, can be set in the Mode Options dialog which can be opened by clicking the Mode Options…
button in the Mode Calculation Parameters dialog.

RSoft CAD  202

Step 3: Choose the mode calculation parameters
Once the mode solver and related parameters have been chosen, the mode calculation parameters
need to be chosen. The Mode Calculation Parameters dialog, shown in Fig. 12-1, allows the user
to specify simulation domain and numerical grid values for the mode calculation.
Generally, these parameters should be chosen such that the domain includes the entire structure,
the domain is large enough to let the mode field be sufficiently close enough to 0 at the
boundaries, the transverse grid sizes need to be small enough to resolve the structure all fields,
the Z step size has a special meaning in the case of the iterative method (see the BeamPROP
manual), and the launch field should be set up so that it excites all the modes you wish to find.
Generally, a simple Gaussian launched off-axis will satisfy this need.

Figure 12-1: The Compute Modes dialog box where the mode calculation parameters are set.

Step 4: Perform the mode calculation.

After the mode calculation parameters have been chosen, enter an Output Prefix and click the
OK button. The mode files will be saved to the current working directory. Each mode found will
have two files: the mode data is contained in the file <prefix>.m##, and has a corresponding
WinPLOT command file <prefix>.p##, where ## is the mode number. Also, a *.nef file is
created which contains a list of the effective indices are found. The format of these files are
discussed further in Appendix B.

RSoft CAD  203

11.B. TMM-Based Mode Solver (TmmSIM)
This section outlines the use of the TMM-based 1D mode solver TmmSIM. This mode solver
can calculate the modes of step-index structures with an arbitrary number of layers (either slab or
radial geometry).

TmmSIM can be used by all users who are licensed for any of RSoft’s passive device simulation
tools.

To open TmmSIM, choose the appropriate option from the Utility menu item in the RSoft CAD
interface.

Figure 12-2: The TmmSIM simulation dialog.

11.B.1. Using TmmSIM

To use TmmSIM, first define the structure of interest in the CAD as normal. Then, open the
TmmSIM simulation parameters dialog shown above and set the following:
• Simulation Options
The simulation options control the Polarization (TE or TM) and whether the geometry is a
slab or radial. If the structure has radial symmetry, check the Enable Radial option and enter
the desired Azimuthal Mode #. This mode number has the same definition as RSoft’s other
tools, namely that it sets the m in the HEmn type designation.
• Simulation Domain

RSoft CAD  204

 The simulation domain is defined to be along the X axis (at Y=0), and the Domain Min and
Domain Max can be set by the user. Additionally, the Z Position can also be set.

• Mode Searching Options

Modes can be found via a Local Search, where the user can specify a guess for an effective
index (via Neff (Real) and Neff (imag)), and the program tries to find the closest mode. When
using a Global Search, the user can specify a range in the complex plane to search for modes,
and the program tries to find all the modes within that range.
• The Output Prefix
The Output Prefix sets the name to be used for all output files created.
• Finding Leaky Modes
TmmSIM will usually automatically detect if the structure supports leaky modes and then
find them. However, in some cases if this auto detection does not work, TmmSIM can be
forced to find leaky modes by setting tmm_leaky to 1.

11.B.2. Output Options

TmmSIM outputs modes exactly the same as the BPM-based mode solvers and FemSIM. The
modes are numbered up from 0 (the fundamental mode), and are output as usual. Also, a *.nef
file is created which contains a list of the modes found.
Also, any modes found in a radial calculation can be converted to an XY format automatically
using the option in the Output Options dialog. Note that this is not step is only necessary for
visual purposes, and is not needed if launching these files into a 3D simulation as they will be
automatically converted.

11.C. FEM-based Mode Solver (FemSIM)

FemSIM is based on the Finite Element Method, and allows users to compute both propagating
and cavity modes. Its use is fully described in the FemSIM manual; an online form can be
opened via the Help menu item in the CAD window and a pdf format can be found in the
<rsoft_dir>/docs directory.

FemSIM is licensed separately and is mentioned here for completeness.

RSoft CAD  205

Appendix A

Tips and Traps

This appendix contains some advice on good habits and sources of confusion for novices and
experienced RSoft CAD users alike.

A.A. Common RSoft CAD mistakes

• Misuse of overlapped segments.
Overlapped segments can be used to achieve complicated geometries. However, it is very
important to understand the logic used by the CAD when segments overlap. Please read
through Section 6.C.
• Using the wrong units.
It is important to use the correct units when describing a structure in the CAD. All units of
length are given in µm, angular units are given in degrees, and the units of imaginary index
values are:

nimag =
4
where λ is the wavelength and γ is the usual exponential loss coefficient defined such that the
power decays as e-γz, and is given in units of µm-1.

RSoft CAD  206

A.B. Some Good RSoft CAD habits to learn
Developing the following habits will save you trouble and time in the long run.
• Use the Display Material Profile option frequently
It is easy to make mistakes with geometry and refractive index information. Many apparently
bizarre results are simply the result of simulating the wrong structure.
• Get Full Usage out of the Symbol Table
The Symbol Table is a powerful tool which provides a simple way to parameterize a design.
Virtually all fields in the CAD, including those in the Segment Properties box, can accept
variables and function of variables. This allows the user to create complicated designs in
which each parameter, including waveguide coordinates, simulation specific parameters, and
virtually every program option is logically tied to other parameters in the circuit. When
designing a circuit, it is highly recommended that you utilize the Symbol Table as much as
possible. As tempting as it may be to enter exact coordinates for waveguide vertices or
refractive index values, it is usually better to create variables in the Symbol Table which
equal the desired quantities, and then enter these variable names into the appropriate fields in
the CAD interface. This allows for a design file which is very easy to modify, and easier to
understand.
• Use Referenced Coordinates
The coordinate specification by logical references described in Section 4.B.1 is a unique and
powerful CAD feature that can be used to layout circuits with specific geometric
relationships which can remain constant while various dimensions (lengths, displacements,
separations, etc.) are changed. Coupled with the programmability feature described above
that enables formulas to be used to define coordinates or offset/angle parameter values, this
feature allows the user to define a logical arrangement of waveguide components whose
specific dimensions can be changed by simply altering a few values in the symbol table.
Illustrations of this approach are given in several of the example circuits found in the
<rsoft_dir>examples directory.

• Learn to use the command line utilities included with the RSoft Photonic Suite
You can achieve a lot of tedious work by writing a single command. You may even find uses
for these utilities in your other work. These utilities are documented in Appendix E.
• Learn the scripting capabilities of the RSoft Photonic Suite
While the RSoft CAD is designed to be used via a graphical interface (GUI), complicated
structures can be quite tedious to create in this fashion. You can create a layout script which
automatically creates a desired design file. Additionally, simulation scripts can be created to
automate tedious simulations. See Chapter 10.
• Read the WinPLOT manual

RSoft CAD  207

 WinPLOT’s rather terse command syntax can be daunting at first, but allows complex
plotting features to be described quickly in just a few lines. Being able to quickly launch
plots with old command files from the command line can be a huge time saver.
• User color scales to produce attractive plots.
WinPLOT is capable of producing many types of plots beyond the standard output usually
displayed. Color scale files can be found in the directory, <rsoft_dir>RPLOTDIR and are
used to change the default scale used when displaying contour plots. These files are easy to
create, and provide a quick way to customize the display of data for presentation purposes.
More information on the use and creation of color scales can be found in Appendix F.

RSoft CAD  208

Appendix B

File Formats

This appendix documents file formats used both by the RSoft CAD. Note that some file formats
are also used by simulation engines as well. For example, the Standard RSoft File Format is used
by every simulation engine for a variety of outputs. Except where noted, all files are in ASCII
text format. For further information, contact RSoft.

B.A. Standard RSoft File Format

This is the standard ASCII file format which is used to represent functions of one, two, or three
variables. This is the most common format used by RSoft software and is used by several
different input and output file types such as field profiles, mode profiles, material property
profiles (refractive index, etc), user-defined tapers, user-defined profiles, or with the userdata()
functions described in Appendix C. When using this format, the extension of the file (i.e. .dat,
.fld, etc.) does not matter.

Note that the TDR file format described in Section B.B can also be used as input virtually
anywhere in the software that the Standard RSoft File format can be used. In addition, TDR files
may be output as described in Section B.B.2.

B.A.1. Syntax Parameters

Before discussing the actual syntax, we first define several parameters used in the syntax.

RSoft CAD  209

While these types of files describe a function of X, Y, and/or Z, these coordinates are local to the
data file and do not have to correspond to actual coordinates. This is especially true when using
the userdata() functions where X, Y, and/or Z represent the value of symbols.

Parameter Description
Nx, Ny, Nz Number of data points in the X, Y, and Z directions.
X0, Y0, Z0 The X, Y, and Z coordinate of the first data point.
Xn, Yn, Zn The X, Y, and Z coordinate of the last data point.
Zpos The Z coordinate. The actual numerical value does not affect the field data in
any way and is included for bookkeeping purposes. For a 3D data file, this
should be set to the keyword Z_DEPENDENT.
Optional_Data These parameters are not required, but are used in special cases. For instance,
if the field is a result of a mode calculation, there are two optional fields
containing the real and imaginary parts of the mode effective index. Also, the
wavelength and refractive index of the plane where the data is defined can be
specified here.
Output_Type The sub-format of the data. This field can be one of the following keywords:

Each output type consists of a pair of keywords: the first is used for data files
with one coordinate, the second (_3D) for data files with two coordinates.

• OUTPUT_AMPLITUDE and OUTPUT_AMPLITUDE_3D

Incicates that the data corresponds to the amplitude of a field.
• OUTPUT_REAL and OUTPUT_REAL_3D
Indicates that the data corresponds to the real part of a field.
• OUTPUT_IMAG and OUTPUT_IMAG_3D
Indicates that the data corresponds to the imaginary part of a field.
• OUTPUT_REAL_IMAG and OUTPUT_REAL_IMAG_3D
Indicates that the data corresponds to both the real and imaginary parts of
a field. The data is presented in two columns; the first contains the real
part of the field, and the second the imaginary part of the field. Note that
for data files of two coordinates (_3D), there should be two columns of
data per Y value, one for the real and one for the imaginary.
• OUTPUT_AMP_PHASE and OUTPUT_AMP_PHASE_3D
Indicates that the data contained in the file corresponds to both the

RSoft CAD  210

 amplitude and phase of a field. The data is presented in two columns; the
first contains the amplitude of the field, and the second the phase. Note
that for data files of two coordinates (_3D), there should be two columns
of data per Y value, one for the amplitude and one for the phase.

B.A.2. File Syntax

Since this type of file is used to describe a wide variety of data types, there are several syntax
variations. This section describes the major formats that will be encountered when using the
software. While some variation is possible, it is best to use the formats described here when
manually creating a file.

The keywords used in these definitions are described in the previous section. For example of
each of these file types, see the next section.

1D Uniform Grid Syntax:

The syntax for a 1D data file with a uniform X grid is:
/rn,a,b/nx0
Nx X0 Xn Zpos Output_Type Optional_Data
Data(Xmin)
…
Data(Xmax)

where there are two header lines: one ‘/’ line that indicates to the software how to interpret the
header and one line that contains the X coordinate information and additional information such
as the data type. The data is specified in the remaining rows of the file, one row for each point.

1D Non-Uniform Grid Syntax:

The syntax for a 1D data file with a non-uniform X grid is:
/rn
Nx X0 Xn Zpos Output_Type Optional_Data
X0 Data(X0)
… …
Xn Data(Xn)

where there are two header lines: one ‘/’ line that indicates to the software how to interpret the
header and one line that contains the X coordinate data and additional information such as the
data type. The data is specified along with the X values in the remaining rows of the file.

1D Non-Uniform Grid Syntax with No Header:

The syntax for a 1D data file with a non-uniform X grid and no header is:

RSoft CAD  211

 X0 Data(X0)
… …
Xn Data(Xn)

where there are no header lines, and the Data(X) entries can be complex. This format is useful
for defining the refractive index information for materials where the X data represents
wavelength and the Data(X) entries are the real and imaginary parts of the refractive index.

2D Uniform Grid Syntax:

The syntax for a 2D data file with a uniform X and Y grid is:
/rn,a,b/nx0
/rn,qa,qb
Nx X0 Xn Zpos Output_Type Optional_Data
Ny Y0 Yn
Data(X0,Y0) … Data(X0,Yn)
… …
Data(Xn,Y0) … Data(Xn,Yn)

where there are four header lines: two ‘/’ lines that indicate to the software how to interpret the
header and two lines that contain the X coordinate data, Y coordinate data, and additional
information such as the data type. Note that the data at each X and Y value are defined as a
matrix, each row and column representing a single X and Y value respectively.

2D Non-Uniform Grid Syntax:

The syntax for a 2D data file with a non-uniform X and Y grid is:
/rn
/r
/rq
Nx X0 Xn Zpos Output_Type Optional_Data
Ny Y0 Yn
0 Y0 … Yn
X0 Data(X0,Y0) … Data(X0,Yn)
… … …
Xn Data(Xn,Y0) … Data(Xn,Yn)

where there are five header lines: three ‘/’ lines that indicates to the software how to interpret the
header and two lines that contain the X coordinate data, Y coordinate data, and additional
information such as the data type. Note that the X values are now defined explicitly in the first
column and the Y values are defined explicitly in the first data row. The first 0 in the Y
coordinate row is required as a spacer only as the actual value does not matter. The data itself is a
matrix as in the 2D uniform grid syntax.

3D File Syntax
The syntax for a 3D data file with a uniform X and Y grid is:

RSoft CAD  212

 /rn,a,b/nx0
/r,qa,qb
/r
Nx X0 Xn Z_DEPENDENT Output_Type Optional_Data
Ny Y0 Yn
Nz Z0 Zn
Data(X0,Y0,Z0) … Data(X0,Yn,Z0)
… …
Data(Xn,Y0,Z0) … Data(Xn,Yn,Z0)
Data(X0,Y0,Z1) … Data(X0,Yn,Z1)
… …
Data(Xn,Y0,Z1) … Data(Xn,Yn,Z1)
… …
Data(X0,Y0,Zn) … Data(X0,Yn,Zn)
… …
Data(Xn,Y0,Zn) … Data(Xn,Yn,Zn)

where there are six header lines: three ‘/’ lines that indicates to the software how to interpret the
header and three lines that contain the X coordinate data, Y coordinate data, Z coordinate data,
and additional information such as the data type. For a 3D file, the Z position (Zpos) field should
be set to the keyword Z_DEPENDENT as shown. The data is a series of matrices each
corresponding to multiple Z cross-sections.

A Note about Headers

Various outputs will create data files that have some additional items in the first header line.
These are primarily plotting commands that effect how WinPLOT will plot the file. For example,
the command /ls1 indicates that the first line style (solid) should be used to plot all data series
in the file. When creating files, it is generally not necessary to include these types of options in
the data file.
An important and useful exception to this rule is the /xy command, which will transpose the x
and y coordinates. While this does effect plotting in WinPLOT, this command can also be used
to transpose the data before using the file in the CAD or in a simulation tool as a user-profile,
launch field, etc.

B.A.3. Example Files

In order to better illustrate the field file format, several examples are given. In each case, a
function will be given, and then a corresponding data file.

Example 1:
The following file contains the function f(x) = x2 over the X range [0,1] with an X grid of 0.1:
/rn,a,b/nx0
11 0 1 0 OUTPUT_REAL
0.00
0.01

RSoft CAD  213

 0.04
0.09
0.16
0.25
0.36
0.49
0.64
0.89
1.00

Example 2:
The following file contains the function f(x) = x2 + ix3 over the X range [0,1] with an X grid of
0.1:

In this type of format, the second column contains imaginary/phase data.

/rn,a,b/nx0
11 0 1 0 OUTPUT_REAL_IMAG
0.00 0.000
0.01 0.001
0.04 0.008
0.09 0.027
0.16 0.064
0.25 0.125
0.36 0.216
0.49 0.343
0.64 0.512
0.81 0.729
1.00 1.000

Example 3:
The following file contains the function f(x) = x2 over the X range [0,1] with an X grid of 0.1 and
with no header.
0.00 0.000
0.10 0.010
0.20 0.040
0.30 0.090
0.40 0.160
0.50 0.250
0.60 0.360
0.70 0.490
0.80 0.640
0.90 0.810
1.00 1.000

RSoft CAD  214

Example 4:
The following file contains the function f(x) = x2 + ix3 over the X range [0,1] with an X grid of
0.1 and no header:

In this type of format, the third column represents imaginary/phase data.

0.00 0.000 0.000

0.10 0.010 0.001
0.20 0.040 0.008
0.30 0.090 0.027
0.40 0.16 0.064
0.50 0.25 0.125
0.60 0.36 0.216
0.70 0.49 0.343
0.80 0.64 0.512
0.90 0.81 0.729
1.00 0.00 1.000

Example 5:
The following file contains the function f(x,y) = xy over the X range [1,2] and Y range of [0,1]
with an x grid of 0.25 and a y grid of 0.5:
/rn,a,b/nx0
/r,qa,qb
5 1 2 0 OUTPUT_REAL_3D
3 0 1
0.000 0.500 1.000
0.000 0.625 1.250
0.000 0.750 1.500
0.000 0.875 1.750
0.000 1.000 2.000

Example 6:
The following file contains the function f(x,y) = x+iy over the X range of [1,2] and Y range of
[0,1] with an X grid of 0.25 and a Y grid of =0.5:

In this type of format, every other column represents imaginary/phase data.

/rn,a,b/nx0
/r,qa,qb
5 1 2 0 OUTPUT_REAL_IMAG_3D
3 0 1
1.00 0.0 1.00 0.5 1.00 1.0
1.25 0.0 1.25 0.5 1.25 1.0
1.50 0.0 1.50 0.5 1.50 1.0

RSoft CAD  215

 1.75 0.0 1.75 0.5 1.75 1.0
2.00 0.0 2.00 0.5 2.00 1.0

Example 7:
The following file contains the function f(x) = x2 on a non-uniform x grid with 5 points (0.00,
0.10, 0.50, 0.55, and 1.00):

In a 1D non-uniform grid format, the first column contains the X values and the second contains
the data.

/rn
5 0 1 0 OUTPUT_REAL
0.00 0.0000
0.10 0.0100
0.50 0.2500
0.55 0.3025
1.00 1.0000

Example 8:
The following file contains the function f(x,y) = 2x+i3y on a non-uniform x grid with 5 points
(1.00, 1.10, 1.50, 1.60, and 2.00) and a non-uniform y grid with points (100, 125, 200):

In a 2D non-uniform grid format, the first column contains the X values and the first data row
contains the Y values. Also, since this is a complex-valued function, every other column
(excluding the first) contains imaginary data.

/rn
/r
/rq
5 1 2 0 OUTPUT_REAL_IMAG_3D
3 1 2
0 1.00 1.25 2.00
1.00 2.00 3.00 2.00 3.75 2.00 6.00
1.10 2.20 3.00 2.20 3.75 2.20 6.00
1.50 3.00 3.00 3.00 3.75 3.00 6.00
1.60 3.20 3.00 3.20 3.75 3.20 6.00
2.00 4.00 3.00 4.00 3.75 4.00 6.00

Example 9:
The following file contains the function f(x,y,z) = x+ y + z over the x range [-1,1], y range of [-
1,1], and z range [0,1] with an x grid of 1, a y grid of 1, and a z grid of 0.5

This file contains three slices, each 3x3, which contain the field data at each Z point.

RSoft CAD  216

 /rn,a,b/nx0
/r,qa,qb
/r
3 -1 1 Z_DEPENDENT OUTPUT_REAL_3D
3 -1 1
3 0 1
-2.000 1.000 0.000
-1.000 0.000 1.000
0.000 1.000 2.000
-1.500 -0.500 0.500
-0.500 0.500 1.500
0.500 1.500 2.500
-1.000 0.000 1.000
0.000 1.000 2.000
1.000 2.000 3.000

B.B. TDR File Format

Note that the Standard RSoft File format described in Section B.A can also be used as input
virtually anywhere in the software that the TDR File format can be used.

TDR files are the basic input and output used by Synopsys’ TCAD Sentaurus product. This
section only covers the relevant information for using TDR files in the RSoft software. See the
TCAD Sentaurus documentation for a complete description of the TDR format.

B.B.1. Using TDR Files as Inputs to the RSoft Software

TDR files can be used as inputs in the software anywhere a data file can be entered. This
includes user profiles, launch fields, etc. The syntax is:
<tdr-file>[:<quantity>]

where <tdr-file> is the name of the TDR file and quantity is optional (and case-sensitive!)
and sets the name of the desired quantity within the TDR to be used. For example, entering
[Link]:doping

will use the doping quantity from the [Link] file. If no quantity is specified, the first quantity
in the file will be used.

Manipulating TDR Files for Use as RSoft Input

In some cases the coordinate system and how the structure has been drawn in TCAD Sentaurus
will be different than the RSoft CAD. It these cases it will be necessary to manipulate the TDR
data in order to move between the two systems.

RSoft CAD  217

The required manipulation typically consists of one or more data transformations including
scaling, flipping, or transposing one or more coordinates. These operations, which can also be
performed on data in the RSoft Standard File format described in Section B.A, are:

More information about the bdutil utility can be found in Appendix E. Note that some of these
options are documented as part of bdconv, but bdutil acts like bdconv if no function is
specified. Note that the tdrutil utility can also be used to work with TDR files.

• Translating TDR Data:

To translate data along an axis, use the –trx# or –try# options. For example, to translate the
X axis in the file [Link] by 4.2 and save it as [Link], use this command:
bdutil –trx4.2 [Link] [Link]

• Scaling TDR Data:

To scale data along an axis, use the –scx# or –sxy# options. For example, to scale the X axis
in the file [Link] by 2.5 and save it as [Link], use this command:
bdutil –scx2.5 [Link] [Link]

• Flipping TDR Data:

To flip data along an axis, just scale by -1. For example, to flip the file [Link] along the Y
axis and save it as [Link], use this command:
bdutil –scy-1 [Link] [Link]

• Transposing TDR Data:

To transpose the X/Y data use the –t option. For example, to transpose the X/Y data in the
[Link] file and save it to the file [Link], use this command:
bdutil –t [Link] [Link]

B.B.2. Enabling TDR File Output from the RSoft Software

By default, when any input to the RSoft software is in the TDR format, all data outputs will also
be in a TDR format. To explicitly enable the output of TDR file, set the variable output_as_tdr
to 1. Note that any TDR output will use the same filenames as non-TDR output but the file
format will of course be different.

B.C. User-Profile Functions

A user-profile can be defined by either an expression or a data file. Data files should be in the
standard file format described in Section B.A. User profile files can be in 1D, 2D, and/or 3D
formats as described in Section 6.A.3.

RSoft CAD  218

B.D. User-Taper Functions
A user-taper can be defined by either an expression or a data file. Data files should be in the
standard file format described in Section B.A. User taper files should be in a 1D format as
described in Section 6.B.4.

B.E. Polygon Files

Polygons provide a customizable approach to creating components. A polygon is defined by a
series of data points in the (X,Z) plane contained in a data file. These points should be on
separate lines, and the first and last points should be the same.
For example, a square could be defined by the file:
0,0
0,1
1,1
1,0
0,0

Polygons can be scaled in both the X and Z coordinates, as well as rotated at an arbitrary degree.
Also, these files can either be statically loaded once, or dynamically loaded at runtime. See
Section 5.D.

RSoft CAD  219

Appendix C

RSoft Expressions

Virtually any numeric input field in any RSoft product can accept an analytical expression
involving pre-defined and user-defined variables that are defined in the symbol table. This
appendix discusses the form that these expressions can take, the allowed operators, and the built-
in functions that are available.
Expressions are written in the common form accepted by most programming languages (e.g. C,
Fortran, etc.), and can include numbers, variables, arithmetic operators, and function calls.

C.A. Arithmetic Operators & Order of Operations

Valid operators for RSoft expressions, listed in order or precedence, are:

Operator Description
^ exponentiation
* multiplication
/ division
+ addition
- subtraction

Parenthesis can be used as usual to group operations and override this precedence.

RSoft CAD  220

C.B. Built-in Constants
RSoft supports several built-in constants. These variables may be used in the symbol table, or in
any numeric field within the software. These variable names should be avoided when creating
your own variables to avoid conflict.

Variable Value
inf infinity

c 2.9979246e014 µm/s

e 2.718281828

pi 3.141592653

true 1

false 0

yes 1

no 0

$id This parameter equals the segment number of the segment it is used from (if used
in Segment 1, its value will be 1. This is useful to provide different seed numbers
for multiple segments when using the random number generator functions
described below.

C.C. Using File Names in Variables/Expressions

The CAD allows file-name fields to contain a variable which represent the actual file name. This
can be useful when performing scans that use a different data file for each step. If the text in the
field begins with a $ character, the remaining text is interpreted as a symbol that is defined as the
file name. It is possible to use these parameters for any data file in the CAD, such as for files for
launch conditions, monitor files, polygon files, user tapers, user profiles, material names etc.
Using MOST, these variables can be scanned easily.

C.D. Defining Default Value for Variable when

used in Expression
In most cases, variables that are used in an expression are defined in the symbol table. In some
cases, especially for built-in symbols, a variable may not be defined in the Symbol Table if it is
set to its default value. In these cases, it is possible to define a default value for such a variable in
the expression where it is used using the ‘variable:value’ syntax.

RSoft CAD  221

For example, consider the following expression based on the variable dimension, the built-in
variable corresponding to the Model Dimension setting that controls whether the CAD design is
2D or 3D:
My_symbol = if(eq(dimension:3,2), Value2D, Value3D)

This expression sets the symbol My_symbol to Value2D for 2D (dimension equals 2), and
Value3D for 3D (dimension not equal 2). If dimension is not defined in the Symbol Table, this
expression uses a default value of dimension=3 by using the syntax ‘dimension:3’.

C.E. Square Bracket Notation for Hierarchy

Export Symbols
The Hierarchy Export option described in Section 6.F.5 uses a square bracket notation [xx] to
denote that the value of the symbol should be taken from the child file instead of defined in the
parent file. The value displayed inside the square brackets is dynamically updated based on the
value in the child file.

C.F. Built-In Functions

RSoft expressions support the standard function calls found in programming languages as well as
several other functions useful for optical design.

C.F.1. Basic Functions

Function Description
abs(x) absolute value
exp(x) exponential
log(x) natural logarithm
log2(x) logarithm in base 2
log10(x) logarithm in base 10
sqrt(x) square root
cos(x) cosine
sin(x) sine
tan(x) tangent
acos(x) inverse cosine

RSoft CAD  222

 asin(x) inverse sine
atan(x) inverse tangent
atan2(y,x) inverse tangent
sinh(x) hyperbolic sine
cosh(x) hyperbolic cosine
tanh(x) hyperbolic tangent
sinc(x) sinc function [sin(x)/x]
erf(x) error function
erfc(x) complementary error function
ceil(x) truncate up to next higher integer
floor(x) truncate down to next lower integer
round(x[,y]) When used with one argument, this function rounds x to nearest
integer (y=1); when used with two arguments, it rounds x to the
nearest multiple of y
min(x1,x2,…) minimum of arguments in list
max(x1,x2,…) maximum of arguments in list
hypot(x,y) length of hypotenuse of right triangle with sides x and y
sign(x) sign of argument (-1 or +1)
clip(x) clip function defined as 0 for x less than 0, 1 for x greater than 1,
and x otherwise.
list(x,y,…) function to pass a list of parameters

C.F.2. Comparison Functions

These functions allow two arguments to be compared:

Function/Definition Description

1 x = y Determine if two arguments are equal.

eq ( x, y ) = 
0 x  y
1 x − y  z Determine if two arguments (x,y) are equal
eq ( x, y, z ) = 
0 x − y  z within a tolerance (z).

RSoft CAD  223

 1 x  y Determine if two arguments are unequal.
ne( x, y ) = 
0 x = y
1 x − y  z Determine if two arguments (x,y) are unequal
ne( x, y, z ) = 
0 x − y  z
within a tolerance (z).

1 x  y Determine if one argument (x) is greater than

gt ( x, y ) = 
0 x  y
another (y).

1 x  y Determine if one argument (x) is greater than or

ge( x, y ) = 
0 x  y
equal to another (y).

1 x  y Determine if one argument (x) is less than

lt ( x, y ) = 
0 x  y
another (y).

1 x  y Determine if one argument (x) is less than or

le( x, y ) = 
0 x  y
equal to another (y).

C.F.3. Logical Functions:

These logical functions allow arguments to be logically compared:

Function/Definition Description

 a x  0.5 This is the equivalent of a standard logical ‘if’

if ( x, a, b) = 
b x  0.5 function.

1 x  0.5 This is the equivalent of a standard logical ‘not’

not ( x) = 
0 x  0.5 function.

1 x  0.5 and y  0.5 This is the equivalent of a standard logical ‘and’

and ( x, y ) =  function.
0 otherwise

1 x  0.5 or y  0.5 This is the equivalent of a standard logical ‘or’

or ( x, y ) =  function.
0 otherwise

 v1 x = 1 This is the equivalent of a standard ‘switch’

 statement where the first argument is the index
case( x, v1, v2,...) = v2 x = 2
 ... x = ... and the remaining arguments are the potential
 values. For example, case(2,4,8,4) will return
8.

0 if 'x' does not exist This function checks if a symbol is defined and
defined ( x) =  returns 0 (undefined) or 1 (defined).
1 if 'x' exists

RSoft CAD  224

C.F.4. Step Functions
These functions are step functions that are especially useful for defining tapers:

Function/Definition Description

 0 x0 This is a standard non-repeating step function

 that equals 0.5 at the step.
u ( x) = 0.5 x = 0
 1 x0


0 x  0 This is a variation of the function u(x) that

 equals 0 at the step.
um( x) = 0 x = 0
1 x  0

0 x  0 This is a variation of the function u(x) that
 equals 1 at the step.
up( x) = 1 x = 0
1 x  0

1 0  x  0.5 This is a periodic step function that is
step ( x) =   x  [0,1)
0 0.5  x  1
periodically extended outside the x range [0,1 ).

1 0  x  a This is a variation of the function step(x) with a

step 2( x, a) =   x  [0,1)variable duty cycle controlled by the second
0 a  x  1
argument a (which must be between 0 and 1).

C.F.5. Random Functions

These deterministic functions allow the generation a random number or a random function of a
continuous variable.

Random Numbers [ranseq()]

The deterministic function ranseq() returns a “random” result for a given pair of arguments. Its
syntax is:
ranseq(int seed, int i)

where seed is the random seed, and i is an integer useful for correlating several “random”
parameters together. The distribution is linear between -0.5 and 0.5, and several special seeds are
recognized: 0 returns 0.0, -1 returns 0.5, and -2 returns -0.5.
As an example, consider a case where several parameters within a component are to have
random values. To do this, define one or more seed variables in the symbol table, and then use
the same index variable i within the component. In this way two or more fields within the
component are correlated. Changing the seeds within the symbol table will result in different

RSoft CAD  225

random numbers. If this correlation is unnecessary, both of the arguments can be treated as
separate seed parameters.

Examples:

Example Value Description

ranseq(5,4) 0.3715 Returns a random number between -0.5 and
0.5.
0.5+ranseq(1,1) 0.3964 Returns a random number between 0 and 1.
0.5+ranseq(1317,1942) 0.1452 Returns another random number between 0
and 1.

Random Function [ranline()]

The deterministic function ranline() returns a “random” function of a continuous variable. Its
syntax is:
ranline(int seed, double corlen, double z)

where seed is the random seed, corlen is the correlation length of the random function, and z is
the continuous variable. The seed variable behaves the same as the function ranseq() described
above and is not repeated here. This function has the same statistical behavior as the built-in
roughness option described in Section 6.D.4, and may be used in tapers. When using this
function, be careful with the normalizations of tapers with respect to the units of z and corlen.
For example, if L is the true physical length of the taper, and CorLen is the true physical
correlation length, then the taper function should use ranline(iseed,Corlen/L,z).

C.F.6. Data-File Functions

These functions allow data contained in data files to be used in an expression. Three functions
(userdata(), userreal(), and userimag()) allow the value of an expression to be defined via
an ASCII data file as a function of up to three other variables and one function (userderiv())
allows the value of an expression to be defined via the derivative of data contained in an ASCII
data file as a function of one variable. The names of data files used with these functions should
not contain spaces.

While these functions can be used anywhere an expression allowed, it is better to employ them
only in the symbol table and assign their result to another symbol to be used anywhere since their
evaluation can be time-consuming.

• Standard Data-File Functions

RSoft CAD  226

 The userdata() and userreal() functions are synonymous, and return the real part of
values in a data file (which can be real or complex). The userimag() function returns the
imaginary part (or 0 if the value is real). All functions have the same syntax; only userdata()
is described here.
The syntax for these functions is:
userdata("filename",var1)
userdata("filename",var1,var2)
userdata("filename",var1,var2,var3)

where the first argument is the file name in quotes, and the remaining arguments (up to three)
are the variables the data is a function of and must be defined in the Symbol Table. This
function will evaluate the variable(s), then look up the corresponding value in the data file.
• User Derivative Function
The userderiv() function returns the first derivative of data contained in an ASCII data file.
The syntax for this function is:
userderic("filename",var)

where the first argument is the file name in quotes and the second argument is the variable
that the data is a function of and must be defined in the Symbol Table. This function will
evaluate the variable(s), then look up the corresponding first-derivative based on the data in
the file.

Additional Notes:
• When using these functions to define materials stored in the Group Library discussed in
Section 8.A.1, you can use the keyword <groupmat> to refer to the PATH defined. See note
in Section 8.A.7.
• You can use a symbol to specify the filename using the ‘$’ prefix, for example:
filename = [Link]
userdata($filename, free_space_wavelength)

• The format of the file is currently limited to the formats described in Appendix B, see a
simple example below.
• The RSoft CAD will not give an error if a blank string is used for the filename. This is useful
when using the ‘:’ syntax to define default values. Consider the following series of symbols:
V_exp = 7
V_file = userdata($F:,free_space_wavelength)
Parameter = if(defined(F),V_exp,V_file)
F = [Link]

In the above example, if the symbol F is not defined, then the V_file symbol will be set to a
null value (set via the ‘:’ default option described in Section C.D). In this case, the Parameter
value will be set to the value V_exp.

RSoft CAD  227

Example 1:
As an example, consider the case of defining the background index as a function of wavelength.
Assume the refractive index information is contained in a data file named [Link] and has the
following contents:
/rn,a,b/nx0/ls1
5 1.50 1.60 0 OUTPUT_REAL
1.4430
1.4435
1.4440
1.4445
1.4450

This data file defines the refractive index over a wavelength range of 1.50 to 1.60 µm using 5
points. See Appendix B for more details about this file format. It is also possible to use a header-
less format like this:
1.500 1.4430
1.525 1.4435
1.550 1.4440
1.575 1.4445
1.600 1.4450

with two columns containing the wavelength and real index data. A three column file with the
imaginary index can also be used.
The proper syntax to define the background index as a function of the wavelength would then be:
background_index = userdata("[Link]",free_space_wavelength)

The second and third numeric arguments could be used to conveniently add dependence on other
variables, such as temperature or concentration, if the data file contained this information. In this
case, setting the variable free_space_wavelength to 1.55 will result in a background_index
value of 1.4440. Results are interpolated from the data in the data file if needed.

Example 2: Using a Data File to Define Refractive Index in the RSoft CAD
It is important to use refractive index data that corresponds to the actual materials being
simulated to ensure meaningful simulation results. Refractive index data, which can be found in
reference materials, on the Internet, or from experiment, can be used to create a new material in
the RSoft CAD’s Material Library. This new material can then be used with any of RSoft’s
passive device simulation tools.
• The refractive index data should be in a simple text format: the first column contains the
wavelength, the second column contains the real index, and the third optional column
contains the imaginary index. It is important that the file contains data for the entire
wavelength range to be simulated. The data can be at non-uniformly spaced wavelengths and
no header is required, though a simple text header can be used.

RSoft CAD  228

• For this example, the data file will be put in the same directory as the .ind design file in
which the material will be used. If needed, you can create a centrally stored Group Library of
materials which can be shared among design files; see Section 8.A for details.
• Open the Material Editor with the left CAD toolbar, create a new material, and give the
material a descriptive name.
• Click the Import NK Data button and select the material data file. Expressions using the
userreal() and userimag() functions will automatically used to load the data into the
material.
• You can use the CTRL-doubleclick feature of the RSoft CAD to see the current value of the
index. Hold the CTRL key and double-click the mouse on the Index (real) text field to see the
value which should correspond to the value in the data file at the current wavelength.

C.F.7. Material Functions

There are two types of material functions supported by the RSoft CAD: functions that can return
the refractive indices of materials that are defined in the Material Library, and functions that
return the refractive index of special materials such as Graphene. This section describes both
types of material functions.

Functions to Obtain Refractive Index of Defined Materials

There are six functions that allow for the refractive indices of materials to be used in symbol
table expressions. The first two functions lookup the real and imaginary values for the refractive
index at the global wavelength (free_space_wavelength) defined in the .ind file:
nreal(“material”)
nimag(“material”)

where material is the name of the material as defined in the Material Editor. To calculate the
refractive index at a specific wavelength use the form:
nreal(“material”, lambda)
nimag(“material”, lambda)

where lambda is the wavelength [μm]. The other four function lookup the real and imaginary
values of the refractive index at a specific wavelength for ternary and quaternary materials:
nkinterp2_real(“materialA”,”materialB”,x,lambda,dir)
nkinterp2_imag(“materialA”,”materialB”,x,lambda,dir)
nkinterp4_real(“materialA”,“materialB”,“materialC”,“materialD”,x,y,lamb
da,dir)
nkinterp4_imag(“materialA”,“materialB”,“materialC”,“materialD”,x,y,lamb
da,dir)

where material[A|B|C|D] is the name o fthe material(s) as defined in the Material Editor, x
and y are the ternary and quaternary alloy parameters (set in ‘Semiconductor Tab’ in the Material
Editor), lambda is the wavelength [μm], and dir is the directory code for the material file

RSoft CAD  229

location. The reason the directory is specified is so that the basis material(s) need not be in the
Project Materials portion of the Material Editor. Possible directory codes are 0 (the working
directory), 1 (RSoft Material library), or 2 (Group Material Library). These functions return the
following value:
nkinterp2_[real|imag]() = x*A + (1-x)*B
nkinterp4_[real|imag]() = x*y*A + x*(1-y)*B + (1-x)*y*C + (1-x)*(1-y)*D

where A, B, C, D are the n (real)or k (imag) values from the respective materials evaluated at
lambda.

Special Material Functions

The RSoft CAD supports functions that return the refractive index of special materials:
• Graphene
The properties of Graphene depend on the material properties and environment. The
graphene_real and graphene_imag functions calculate the real and imaginary refractive
index of Graphene as a function of several parameters:
graphene_real(lambda,u,g1,g2,t,h,e)
graphene_imag(lambda,u,g1,g2,t,h,e)

where

Parameter Description and Notes

lambda The wavelength [μm]
u The chemical potential μ [eV]
g1 The intraband Γ1 [eV]. Note:
• Γ [eV] = hbar* Γ [1/sec]
• To convert a time constant τ, typically Γ = 1/(2τ)
g2 The interband Γ2 [eV]
• See notes above for Γ units and conversions
• Setting g2 to -1 makes g2 = g1.
t The temperature T [K]
h The Graphene thickness H [μm], typically set to 0.0007
e The reference epsilon εref

The graphene_real() and graphene_imag() functions are used by the built-in Graphene
models in the Material Library but can also be used in any expression throughout the
software. When using these functions make sure to set the appropriate parameters that

RSoft CAD  230

 correspond to your exact situation. Note that older versions of the software had fewer
arguments and need to be manually updated to the form described above.
The model used by these functions calculates the real and imaginary refractive index as
follow:

i ( ,  , 1 ,  2 , T )
 ( ) =  ref +
 0 H
where ε0 is the permittivity of free space and σ(ω,μ,Γ1,Γ2,T) is calculated as described in Eqn.
1 of the reference below. Note the following exceptions:
• Γ1 is used for the intraband term and Γ2 is used for the interband term
• The sign convention for ‘i’ is changed (i is replaced with –i)
• The integral is solved exactly for all cases, the approximation used in Eqn. 4 is not used
Reference: Hanson et al, “Dyadic Green’s functions and guided surface waves for a surface
conductivity model of graphene”, Journal of Applied Physics 103, (2008).

C.F.8. Effective Index Functions

There are several functions that can be used to analytically compute effective and group indices
of certain 2D and 3D geometries.
• Effective index of a 3-layer slab waveguide mode - slabneff()
The function slabneff() computes the effective index of a 3-layer slab waveguide mode;
it’s syntax is:
slabneff(p,m,lambda,nc,ns,nf,h[,g])

where p is the polarization (0=TE, 1=TM), m is the mode number (0 is fundamental), lambda
is the free space wavelength in µm, nc is the cover index, ns is the substrate index, nf is the
film or guiding layer index, and h is the thickness of the film or guiding layer in µm. The last
argument g is optional and, if set to 1, calculates the group index instead of the effective
index.
• Effective index of rib waveguide geometries – eimneff()
The function eimneff() computes the effective index via the effective index method for rib
waveguide geometries; it’s syntax is:
eimneff(p,mx,my,lambda,nc,ns,nf,w,h,hside[,g])

where p is the polarization (0=TE, 1=TM), mx and my are the X and Y mode numbers (0 is
fundamental), lambda is the free space wavelength in µm, nc is the cover index, ns is the

RSoft CAD  231

 substrate index, nf is the film or guiding layer index, w and h is the width and height of the
rib in µm, and hside is the slab height. The last argument g is optional and, if set to 1,
calculates the group index instead of the effective index.
• Effective index of LP fiber modes – lpneff()
The function lpneff() computes the effective index of LP fiber modes; it’s syntax is:
lpneff(l,m,lambda,nclad,ncore,d[,g])

where l and m are the mode numbers, lambda is the free space wavelength in µm, nclad is
the cladding index, ncore is the core index, and d is the core diameter in μm. The last
argument g is optional and, if set to 1, calculates the group index instead of the effective
index.

C.F.9. Polygon Functions

These functions can be used to define a 2D polygon function which can be used to make custom
user-profiles, etc:
polygon(n, x, y)
polygon(“file”, x, y)

The first form generates a regular polygon with n sides. The second form takes an arbitrary set of
points contained in a file which must be in quotes. The file should be in the polygon format in
Appendix B.E.

C.F.10. PDK Functions

The pdkinfo() function returns design parameters from the PDK currently set by the PDK Info
File option in the Global Settings dialog. A generic PDK is included and used by default, PDKs
from specific foundries can be installed if you have permission from the foundry (contact support
for details). User-defined custom PDKs can be added. The syntax is:

Function/Syntax Description
pdkinfo(“parameter”) Returns a numeric value from a pdkinfo file.
$pdkinfo(“parameter”) Returns a string value from a pdkinfo file.

As an example, consider the following sample pdkinfo file, and then example function calls.
WG_width = 0.5
WG_height = 0.22
WG_material = “Si”

Example usage would be pdkinfo(“WG_width”) which would return 0.5, and

$pdkinfo(“WG_material”) which would return “Si”. Note that it can be useful in some cases
to also define the parameters using the layer number, see discussion in the Custom PDK Utility
manual.

RSoft CAD  232

C.F.11. String Functions
There are several functions that help manipulate and use strings:

Function Description
strlen(x) Returns the length of string x.
$cat(x,y,..) Concatenates string arguments. For example, cat(“AA”,”BB”)
returns “AABB”.
$itoa(x) Returns a string from an integer argument x.
$ftoa(x) Returns a string from a float argument x.
atoi() Returns an integer from a string argument x.
atof() Returns a float from a string argument x.

C.F.12. Unit Conversion Functions

There are several functions that help convert units:

Function Description
eng2sci(x) Converts engineering format to scientific format. It accepts one
string argument. For example, eng2sci(“2u”) returns 2e-6. The
accepted codes and factors are:
• Y, a factor of 1e24
• Z, a factor of 1e21
• E, a factor of 1e18
• P, a factor of 1e15
• T, a factor of 1e12
• G, a factor of 1e9
• M, a factor of 1e6
• K, a factor of 1e3
• M, a factor of 1e-3
• u, a factor of 1e-6
• n, a factor of 1e-9
• p, a factor of 1e-12
• f, a factor of 1e-15

RSoft CAD  233

 • a, a factor of 1e-18
• z, a factor of 1e-21
• y, a factor of 1e-24

lm2um(x) Returns a length metric in microns, and returns 1e-

6*eng2sci(x). The x argument is a string, but if a number
argument is used, that number is returned.

C.F.12. Sentaurus Workbench (TCAD) Functions

There are functions that enable and simplify the use of the RSoft Photonic Device Tools within
TCAD Sentaurus. The swb() and $swb() functions enable seamlessly switching between a
default value before Sentaurus Workbench pre-processes and the actual parameter value after
pre-processing. The $swb() function should be used with string values. The syntax is:
swb(“@text_to_be_preprocessed@”, default_value)
$swb(“@text_to_be_preprocessed@”, “default_value”)

Before SWB pre-processes, the second argument is returned. When the text is pre-processed,
then the first argument is returned. See the TCAD Sentaurus documentation for more
information on pre-processing, including supported syntax and options.

Example Value Returned

swb(“@wavelength@”, 1.55) 1.55

swb(“1.31”, 1.55) 1.31

$swb(“n@node@_fps.tdr”, “saved_dnk.tdr”) “saved_dnk.tdr”

$swb(“n8_fps.tdr”, “saved_dnk.tdr”) “n8_fps.tdr”

RSoft CAD  234

Appendix D

Symbol Table Variables

In this appendix we document the symbol table variable names which control the internal
operation of the RSoft CAD. Most of these variables correspond to dialog fields which have
already been documented.
The following sections list the variable names corresponding to these dialog fields:

D.A. Startup Window/Global Settings

The following variables correspond to the fields founds in the Startup/Global Settings dialog
box:

Field Name Variable Name

Waveguide Model dimension
Dimension
Radial Calculation radial

Effective Index Calculation eim

Polarization polarization

Simulation Tool sim_tool

Free Space Wavelength free_space_wavelength

Background Index background_index

RSoft CAD  235

Background Material background_material

Index Difference delta

Component Width width

Component Height height

Profile Type profile_type

3D Structure Type structure

Cover Index cover_index

Cover Material cover_material

Slab Index slab_index

Slab Height slab_height

Slab Material slab_material

D.B. Special Effects

The following variables correspond to the fields in the Index Generation Options dialog box:

Field Name Variable

Simulated Bend simulated_bend

Simulated Bend Radius simulated_bend_radius

Litho Resolution litho_res

Litho Roughness litho_sigma

Litho Correlation Length litho_corlen

Sidewall Angle Ref (0-1) sidewall_reference

Absolute User Profile user_profile_absolute

D.C. Advanced Grid Parameters

The following variables correspond to the fields found in the Advanced Grid Parameters dialog:

Field Name Variable Name

X Domain Min boundary_min

X Domain Max boundary_max

RSoft CAD  236

 X Grid Size (Bulk) grid_size

X Grid Size (Edge) grid_edge_x

X Griding Ratio grid_ratio_x

X Minimum Divisions grid_mindiv_x

X Grid Grading grid_bulk_nonuniform_x

X Interface Alignment grid_align_x

Y Domain Min boundary_min_y

Y Domain Max boundary_max_y

Y Grid Size (Bulk) grid_size_y

Y Grid Size (Edge) grid_edge_y

Y Griding Ratio grid_ratio_y

Y Minimum Divisions grid_mindiv_y

Y Grid Grading grid_bulk_nonuniform_y

Y Interface Alignment grid_align_y

Z Domain Min domain_min

Z Domain Max domain_max

Z Grid Size (Bulk) step_size

Z Grid Size (Edge) grid_edge_z

Z Griding Ratio grid_ratio_z

Z Minimum Divisions grid_mindiv_z

Z Grid Grading grid_bulk_nonuniform_z

Z Interface Alignment grid_align_z

D.D. Advanced Features

The following variables correspond to advanced features discussed throughout this manual:

Variable Description
alpha Sets the default imaginary index for waveguide segments.
background_alpha Sets the imaginary index for the background.
cover_alpha Sets the imaginary index of the cover for rib/ridge/multilayer
structure types.

RSoft CAD  237

default_material Sets the default material for components that have Material
Properties set to Locally Defined, Index Difference set to delta,
and Index (Imag part) set to alpha.
diffusion_shape Sets the diffusion shape for diffused waveguides.
diffusion_gamma Sets the nonlinear relation between the diffusion concentration and
the index.
diffusion_ratio Sets the diffusion width/height ratio.
change_profile Enables the use of a Z-dependent user profile.
index_profile_all Outputs all index tensors for anisotropic calculation when
computing the index profile.
import_symbol_file Controls symbol import options.
import_symbol_file_auto
import_list

D.E. Display Material Profile

The following variables correspond to Material Profile Calculations:

Variable Description
boundary_min X Domain Min
boundary_max X Domain Max
grid_size X Grid Size
slice_grid_size X Slice Grid
boundary_min_y Y Domain Min
boundary_max_y Y Domain Max
grid_size_y Y Grid Size
slice_grid_size_y Y Slice Grid
domain_min Z Domain Min
domain_max Z Domain Max
step_size Z Grid Size
slice_step_size Z Slice Grid
slice_display_mode Display Mode
prefix Output File Prefix

RSoft CAD  238

index_profile = 1 Compute Index Profile
index_profile = 2 Compute Loss Profile

D.F. CAD Preferences:

The options controlled from the CAD Preferences dialog box are not set via variables in the
Symbol Table, but rather through the initialization file [Link]. This file can be found in
your users directory.

Preferences File Options

The [Link] file location is HOMEDRIVE+HOMEPATH on Windows, and HOME on Linux, where
HOMEDRIVE, HOMEPATH, and HOME are standard environment variables. The RSOFT_INI_PATH
environment variable enables the use of a customer [Link] file. Set it to the folder that
contains the custom file.

RSoft CAD  239

Appendix E

Utilities

RSoft software includes a number of utilities that can be used for a variety of tasks, including,
but not limited to:

These utilities are documented in this appendix in alphabetical order.

• Simple Data File Operations

o Cropping/regridding/slicing data files (bdconv)
o Reordering data columns (shufflemat)
o Arithmetic operations with data columns/files (mathmat and bdutil –o)
• Simple Post-Processing
o Far-field calculation (bdutil –f)
o Overlap integrals (bdutil –i)
o Spot size calculations (bdutil –s)
o Dispersion parameter calculation (disperse)
o Peak-extraction from spectra (fhakit)
o Analysis of FullWAVE state files (fwmodekit)
o Combine results from individual frequencies (rscombine)
• Data Conversion

RSoft CAD  240

 o Bare-matrix to RSoft format (mat2bp)
o Complex matrix conversion (bdconv –v)
o Bitmap image to RSoft data format (bmp2ind)
o Conversion of NK material data file to Lorenzian coefficients (fitdisp)
• Customized Simulation Output
o Specialized simulation outputs that are not automatically created by a simulation can be
calculating using these utilities.
• Field Conversion Interfaces
o Convert to Synopsys’ CODE V software (Ray-Tracing Converter, bp2codev, codev2bp)
o Convert to Synopsys’ LightTools software (bdutil –ray, BSDF Utilities)
• Field Scattering (BSDF) Interfaces
o Use non-periodic excitation for periodic DiffractMOD and FullWAVE simulations
(BSDF Utilities)
o Interfaces with Synopsys’ LightTools and CODE V products (BSDF Utilities)
• Interfaces with Other Tools
o Many tools have text inputs and outputs. Many of these utilities can be used to convert
data to/from the RSoft format.
• Custom Post-Processing
o Combining several of these utilities to create custom output!
These utilities provide a convenient way to perform and automate common tasks. Many of these
utilities can also be spawned directly from MOST, the RSoft scanning tool, making scan post-
processing essentially transparent.

RSoft CAD  241

bdconv – Matrix Manipulation
This utility performs matrix transformations on data files. These transformations include
extracting cross-sections (or “slices”) from a data file, changing the grid sizes in a file, and
changing the file format of an arbitrary data file. This utility is designed to work file formats
discussed in Appendix B.A but can be used with any arbitrary data file as well.

Note that bdconv and bdutil share the same options, either utility name can be used.

Syntax:
bdconv supports the following syntax and options:
usage: bdconv [options] input-file-name output-file-name(s)

options:
-? help - prints this message
-x<range> select range in x (<range> = # or #,#)
-y<range> select range in y (<range> = # or #,#)
-dx# set grid size in x
-dy# set grid size in y
-rp# rotate in phi
-rt# rotate in theta
-tx# translate in x (shift data – keep domain)
-ty# translate in y (shift data – keep domain)
-sc{x|y|z}# scale x,y,z axes
-tr{x|y|z}# translate x,y,z axes
-sym{x|y|z}[#] mirror about axis (+-1=min, +-2=max,<0=antisym,def=1)
-sc# scale data values
-tr# translate data values
-c conjugate data values
-unwrap unwrap phase of data values
-v<type> select output value type (<type> = a,r,i,z,q,n,p)
-t transpose matrix (3D only)
-a alternate output fmt with explicit x/y coordinates(2D/3D)
-abare alternate bare matrix output format (3D only)

Options:
-x<range> These two options are used to extract data either from a certain
-y<range>
range in the data file or at a given value of one coordinate. This is
useful to either reduce the domain of a 2D file, or to take a cross
section from a 2D file and thus create a 1D file.

RSoft CAD  242

 eg. Change the domain of a file input.m00 from [xmin, xmax,
ymin, ymax] = [-5,5,-7,7] to [-2,3,-7,7] and save the output as
[Link]:
bdconv –x-2,3 input.m00 [Link]

eg. Calculate the cross section from input.m00 at Y=2 to create a

file [Link] with a single column of data:
bdconv –y2 input.m00 [Link]

-dx# These two options are used to set the grid size in either the X or Y
-dy#
directions to a desired setting. The points are interpolated between
given data points so that the file can be set to either have more or
less data points than before.
eg. To change the x grid size of input.m00 from 0.1 µm to 0.5 µm
in order to decrease the size of the file:
bdconv –dx.5 input.m00 [Link]

The created file, [Link], contains less grid points than the file
input.m00, and therefore will be smaller and have a decreased
resolution. The grid size can be decreased, and thus the size of the
file would be increased. This option cannot create any new data,
but it can be useful to increase or decrease the number of grid
points.

-rp# These switches either rotate a field or translate the center

-rt#
coordinate of a field.
-tx#
-ty# The -rp and -rt commands tilt the input field by the angle phi
given in degrees so that it will propagate at an angle of phi in the
XZ plane. Phi is measured from the Z axis.
The -tx and -ty commands translate the center of the input field in
X or Y by the given number but maintains the same data domain.
eg. To tilt the file mode.m00 in the XZ plane by 30 degrees and
save the result in the file mode_tile.m00:
bdutil –rp30 mode.m00 mode_tile.m00

eg. To translate the file mode.m00 by 3 microns in the X direction,

and 2 microns in the Y direction and save the result in the file
mode2.m00:
bdutil –trx3 –try2 mode.m00 mode2.m00

RSoft CAD  243

-sc{x|y|z}# This command scales the x, y, or z axes.
-tr{x|y|z}# This command translates the x, y, or z axes.
-sym{x|y|z}[#] This option mirrors data around the axis specified to construct
‘full’ data from data calculated with symmetry. The x, y, or z axis
can be selected along with an optional # number:
• The number # is 1 corresponds to the minimum side along the
selected axis, 2 for the maximum.
• A positive # indicates symmetry, negative indicates anti-
symmetry.
• The default # is 1 (symmetry on the minimum side).
eg. To mirror a file [Link] which contains the positive half of
the data along the X axis:
bdconv -symx [Link] [Link]

eg. To mirror a file [Link] which contains the first quadrant ot

data (positive X/Y):
bdconv -symx [Link] [Link]

-sc# This command scales the data values.

-tr# This command translates the data values

-c This option takes the conjugate of the input data file and saves it to
another data file. This is the same as manually taking a data file of
type Real/Imaginary, and changing the sign of the columns of
imaginary data.

-unwrap This command outputs a ‘real’ data file which contains unwrapped
phase. If the input file is real, it assumes the phase is in the real part
in degrees. If the file is complex, it extracts the phase.

-v<type> This option allows the user to convert a data file in one of the
RSoft data formats into another RSoft format. The types are:
a amplitude

RSoft CAD  244

 r real
i imaginary
z real and imaginary
q amplitude and phase
n normalized
p phase
The utility will take the input file, extract the data, and convert it to
the requested format so the user can get the desired information.
eg. To convert input.m00 from a data format of Real/Imaginary
(3D) to Amplitude (3D):
bdconv –va input.m00 [Link]

-t This option will transpose the data in the input file and save it in
the output file. This is useful when performing data manipulation
or matrix calculations with the data file.

-a This option will output a file that has the absolute coordinates in x
and y of the data points. Typically, a file using the RSoft file
formats has several lines on the top which dictate the domain min
and max as well as the number of points that are in the file for each
row or column. This option will convert a file so that the absolute
coordinates are given.

-abare This command will strip the header information from a data file in
an RSoft file format. By using this, all domain and number of point
information will be removed from the data file.

RSoft CAD  245

bdutil – Overlap Integrals, Far Fields, Spot Sizes,
and LightTools Ray Files
The bdutil utility performs several calculations within several RSoft programs and is included
so the user can perform these calculations separately. The utility can calculate the overlap
integral between two fields, grating coefficients, the field size, the far field projection, simple file
arithmetic, and convert a far-field into a LightTools ray file. This utility is designed to work with
the file formats as outlined in Appendix B.
Depending on the type of calculation to be performed, the utility accepts one or more data files,
calculates the requested information, and then outputs the new data either as a set of numbers or
a new data file. The options for each switch will be examined, and then several examples will be
given to illustrate these options.

Note that bdconv and bdutil share the same options, either utility name can be used.

Syntax
bdutil supports the following syntax and options:
usage: bdutil -i [options] input-file-name1 [input-fle-name2 [weight-file]]
bdutil -k [options] mode-file1 mode-file2 index-file1 index-file2
bdutil -s [options] input-file-name
bdutil -f [options] input-file-name(s) output-file-name
bdutil -ray [options] input-file-name output-file-name
bdutil -o$ [options] input-file-name1 input-fle-name2 outpt-file-name
bdutil without a function operates as bdconv.

options:

-? help - prints this message

-i power or overlap integral calculation (1 or 2 files)
-wx<range> select window in x (<range> = #,#)
-wy<range> select window in y (<range> = #,#)
-xz<range> select window in z (<range> = #,#)
-sq1 square field 1 in overlap integral
-k coupling coefficient calculation
-s field size calculation
-f# far field calculation: #=distance
-fa far field calculation: intensity versus angle
-fp far field calculation: periodic
-f far field calculation: k-space
-fi far field calculation: k-space (inverse)
-flens# field image through ideal lens (#=focal length in um)
-r automatically resize domain

RSoft CAD  246

 -da# override default angular spacing for above
-l# set wavelength (default=1.0)
-nu# set frequency (default=1.0)
-n# set far field refractive index (default=1.0)
-nn# set near field refractive index (default=auto)
-nl set launch angle for periodic far field
-za zero order angle for periodic far field
-zt zero order theta for periodic far field
-ray ray output from far field (angles only)
-rayp ray output from far field (angles and positions)
must specify extent with /wx /wy /wz
-nrays# approximate # of rays (default is # of angles in far field)
-nraysx force nrays to be exact
-o$ perform operation on data files ($=+|-|*|/)
-v<type> Select output value (<type> = a,r,u,z,q,n,p)

Overlap Integrals: -i
The –i switch indicates that an overlap should be performed. The syntax for this option is:
bdutil –i [options] input-file1-name1 [input-file-name2 [weight-file]]

Note that the -ir switch can be used to calculate overlap integrals in radial coordinates.

Options:
The following options control how an overlap integral is calculated:

-wx<range> These options control the limits of integration. By default, the utility
uses the limits of integration which are given in the data files as
coordinate ranges. If the files do not lie in the same range, the utility
calculates the integral over the area where the files overlap.

-sq1 This option squares the first field before calculating an overlap
integral. See example below for details.

-bare This option restricts output to only numbers which can help simplify
scripting and other tasks.
-ir Enables radial overlap integrals. The -irout=<filename> option will
-irout
output the radial integral to a file <filename>_theta.dat.

Usage:
There are several combinations of input data files possible and the utility performs different
calculations in each case:

RSoft CAD  247

These examples assume that the files contain only one coordinate. The definitions can be
generalized for files which contain two coordinates. Note that these integrals are normalized.

One Input File:

If one data file is given which contains a function f(x), the utility calculates three integrals of the
form:
x2

S =  h ( x )dx
x1

where h(x) is equal to f(x), |f(x)|2, and |f(x)|4. Since f(x) can be a complex function, the value of
the integral of h(x) = f(x) contains both a real and imaginary part.
Note that using the switches –i1, -i2, or –i3 will output only the numerical result without any
accompanying text which can aid in using the overlap results in scripts. The switch -i1 outputs
the f(x) result, -i2 outputs |f(x)|2, and –i3 outputs |f(x)|4.

Two Input Files:

If two data files are given which contain functions f(x) and g(x) respectively, the utility calculates
the overlap integral of these functions as:
x2

 f ( x ) g ( x )dx
*

S=
x1
1 1
 x2  2
 x2  2

  f ( x ) f * ( x )dx    g ( x ) g * ( x )dx 
x  x 
 1   1 
where the asterisk denotes the complex conjugate. The program returns values for both S and |S|2.
Since a complex conjugate is taken the order in which the files are input is important. Note that
using the –iu switch will calculate the overlap integral without the normalization, i.e. just the
numerator shown in the equation above.

Two Input Files and a Weight Function:

If a weight function σ(x) is given in addition to two data files, the utility calculates the overlap
integral using the form:

RSoft CAD  248

 x2

 f ( x )  ( x ) g ( x )dx
*

S=
x1
1 1
 x2  2
 x2  2

  f ( x ) f * ( x )dx    g ( x ) g * ( x )dx 
x  x 
 1   1 
where the asterisk denotes the complex conjugate. The program returns values for both S and |S|2.
Since a complex conjugate is taken the order in which the files are input is important. Note that
using the –iu switch will calculate the overlap integral without the normalization, i.e. just the
numerator shown in the equation above.

Using the –sq1 Option:

As an example of using the –sq1 option, consider an overlap with two data files:
x2

 ( f ( x )) g * ( x )dx
2

S=
x1
1
 x2  x2  2

  f ( x ) f ( x )dx   g ( x ) g * ( x )dx 
*
x  x 
1  1 
This is the same form as shown above except the first field f(x) is squared in the numerator and
the normalization term in the denominator for the first field does not have a square root (and
therefore is effectively squared). The –sq1 option can also be used with the weight function as
needed.

Examples:
To calculate the integral of a function f(x), contained in a file [Link] over a range of (-3,2):
bdutil –i –wx-3,2 [Link]

To calculate the overlap integral between two data files, [Link] and [Link], with the
default limits of integration and also squaring the first field:
bdutil –i –sq1 [Link] [Link]

To calculate the overlap integral between two data files, [Link] and [Link], with a
weight file, [Link], over the range (-2.3, 6.1), use the command:
bdutil –i –wx-2.3,6.1 [Link] [Link] [Link]

Coupling Coefficients: -k
The –k switch calculates the coupling coefficient, kappa, between two modes traveling in
opposite directions in a grating. The correct syntax for this calculation is:
bdutil -k [options] mode-file1 mode-file2 index-file1 index-file2

RSoft CAD  249

This switch requires four files to be input: two index profiles corresponding to the peak and
mean index perturbation of the grating and two mode files which correspond to the two modes
being studied. These files need to be in the file formats specified in Appendix B. They can be
produced directly with RSoft software or through any other means as long as the file format is
correct. The -bare option can be used to restrict the output to only numbers which can help
simplify scripting and other tasks.

Usage:
The utility will produce both the real and imaginary parts of kappa, as well as the effective index
of the grating. Kappa is defined as:

  (n − n12 ) vt *t dxdy

+ +
2
k
= 0 − − 2
+ +
4neff , 
 
− −
 t *t dxdy

where n1 is the mean index and n2 is the peak index. The calculated number is a dimensionless
form of the above expression in which the k0 factor is not present.
This is consistent with the book: Raman Kashyap, “Fiber Bragg Gratings,” Academic press,
1998, pp.143, equation (4.3.6). It is also consistent with the paper: P. Correc, “Coupling
Coefficients for Trapezoidal Gratings,” IEEE Journal of Quantum Electronics, Vol. QE-24, No.
1, 1988, pp.8, equation (1), (2) and (3).

Example:
To calculate the coupling coefficient between a forward mode, mode1.m00, and a backward
mode, mode2.m00, in a grating with a peak index profile of [Link] and a mean index
profile of [Link]:
bdutil -k mode1.m00 mode2.m00 [Link] [Link]

Field Size Calculation: -s

The –s switch calculates the 1/e height and width of a field in microns. Its syntax is
bdutil -s [options] input-file-name

For 2D data files, the 1/e width is clearly defined. For 3D data files, bdutil calculates the
effective width/height. For example, to calculate the width of a field a(x,y), bdutil first
calculates the average amplitude b(x) by integrating a(x,y) along y, then finds the 1/e width of
b(x).

Example:
For example, the 1/e width and height of a mode file mode.m00 can be found using the
command:

RSoft CAD  250

 bdutil –s mode.m00

Far Field Calculations: -f

The –f, –f#, -fa[u], -fp, –fi, and –flens# switches calculate the spatial-far field projection of
a field file in a homogeneous dielectric region (free space) via a Fourier transform. The syntax
for this option is (-f is used as an example but any of the previous switches can be used):
bdutil –f [options] input-file-name(s) output-file-name

If multiple input files are specified, they will be interpreted as vector components of the same
field (i.e. Ex, Ey, and/or Ez). Note that when non-uniform grids are used, the near-field is
converted to a uniform grid (with average grid size of non-uniform grid) before the far-field is
calculated.

Usage:
To calculate a far-field with bdutil you must use one of these switches:

-fa The –fa switch produces a plot of intensity vs. angle along with
values for the beam divergence of the full angle at both the 1/e as well
as the half intensity. This computation starts with the spatial near-
field, E(x) and computes the Fourier transform, F(kx). Next, F(kx) is
normalized such that

 F ( x ) dx =  F (k x )dk x

The angular distribution of the far field I(θx) is then computed as

2
 2 
I ( x ) = cos ( x ) F  sin ( x ) 
  

and normalized by its maximum value, resulting in a cumulative form

of:
 2

cos ( x )  E ( x ) exp ( j sin ( ) k x ) dx

2

I ( x ) E ( x )
2 x 0
−
= 2
= 2
I0 
E0
 E ( x ) dx
−
0

RSoft CAD  251

 This derivation illustrates a 1D version of the calculation; the 2D is
directly analogous.

By default, the angular resolution of the Fourier transform is

calculated directly from the range of data in the input file. To override
this angular resolution, the option -da# can be used to specify a
custom angular spacing. The normalization can be skipped using the -
fpu option.

The files output by this option are functions of the angles ϕ and θ (in
3D) and are meant to be plotted in polar coordinates (use the /polar
WinPLOT option).

-f The –f switch calculates the intensity of the far-field in k-space via a

Fourier integral. This computation is equivalent to the first steps of the
method to compute the far field as a function of angle described above
to compute the normalized value of F(kx).

-f# The –f# switch calculates the field at a distance of # microns. This
calculation is performed via the FFT-BPM method and propagates for
one step which is only valid for homogenous or free-space regions.
See the BeamPROP manual for further discussion of this method. The
option –r can be used to increase the boundary size if the field
diverges outside of the boundary specified in the input file.

-fp The -fp switch calculations the far-field projection of a periodic field
from a single plane wave launch. This method assumes that the data
file contains one period of the field, decomposes it into plane waves,
and then plots the amplitude in each diffraction order. The options –za
and –zp can be used to set the launch angles of the plane wave and the
–nl option can be used to set the index of the launch material.

-fi The –fi switch calculates the far-field in inverse k-space.

-flens# The –flens# option calculates the field image through an ideal lens
where # is the focal length in μm.

RSoft CAD  252

Options:
These options can be used to further control the far-field calculation:

-da This option overrides the default angular spacing for the –fa switch
which calculates the far-field intensity vs. angle.

-l By default, bdutil will use the wavelength defined in the header of

-nu
the data file(s) used to calculate the far-field. These options allow the
user to specify the wavelength (or frequency) that should be used in
the calculation: only one of these options should be specified at once.
The option –l sets the wavelength [µm] and defaults to a value of 1
µm and the option –nu# sets the frequency [µm-1] and defaults to a
value of 1 µm-1.

-n This sets the far-field refractive index that the field is propagated
through. By default, it is assumed the field propagates through air, and
therefore this value defaults to 1. A value of -1 indicates that the near-
field index should be extended to the far-field. This option is useful,
for example, when computing the farfield in a substrate or another
material than air.

-nn This sets the near-field refractive index. This should be set to the
index where the near-field was measured. By default, this value is
taken from the header of the data file(s) used. If not specified in the
data file, the far-field index value (-n) is used.

-nl This sets the launch index when using the –fp switch.

-r This option can be used to automatically resize the domain when

using the –f# option.

-v<type> This option allows the user to set the format of the far-field output.
The options are the same as the bdconv utility.

-za This option sets the launch angle ϕ for the –fp option (1D and 2D data
files).

RSoft CAD  253

 -zt This option sets the launch angle θ for the –fp option (2D data files
only).

Examples:
The far field projection of a field file [Link] given as intensity as a function of angle with an
angular resolution of 0.1 degrees at a wavelength of 1.55 µm in a material with a refractive index
of 1.5 can be saved in a file [Link]:
bdutil –fa –da0.1 –l1.55 –n1.5 [Link] [Link]

The field profile for the above case after a distance of 100 µm:
bdutil –f100 –r –l1.55 –n1.5 [Link] [Link]

The far-field from a periodic near-field [Link] calculated with a plane wave launch incident
in a material with refractive index of 1.5, at a launch angle of 10 degrees, at a wavelength of 1.55
µm, and calculated into a material with a refractive index of 2 can be calculated with this
command:
bdutil –fp –l1.55 –nn1.5 –n2 –za10 [Link] [Link]

LightTools Interface
The –rays switch converts RSoft far-field into a LightTools ray file for use as a Ray Data File in
LightTools. Its syntax is:
bdutil –ray [options] input-file-name output-file-name

where the filenames correspond to the RSoft far-field file and the desired LightTools ray data
file. This process can be automated when using the LED Utility by using the options described in
the LED Utility manual. Note that the polarization data in the RSoft BSDF file will be averaged
when converted.

Options:
-ray[p] Required for ray data file conversion. The –ray option outputs a point
source ray data file, the –rayp option outputs the ray data file over an
extended plane.

-nrays# This sets the number of rays to use. The default is based on the field
resolution in the far-field data file.

RSoft CAD  254

 -nraysx If the number of rays requested is more than the resolution the data
file supports, this option will force the converter to use the number of
rays requested.

-wx#,# Sets the dimensions of the extended plane.

-wy#,#
-wz#,#

Example:
To convert RSoft far-field run1_ffoutput.dat into the LightTools ray data file lt_rdf.txt
with approximately 4000 rays, use the command:
bdutil –ray –nrays4000 run1_ffoutput.dat lt_rdf.txt

Simple File Arithmetic and Data Transformations: -o, -sc, -tr

The –o switch perform simple mathematical operations on one or two fields and outputs the
result. Its syntax is:
bdutil –o$ [options] input-file-name1 input-file-name2 output-file

where $ can be one of four mathematical operations: +, -, /, or * for addition, subtraction,

division, and multiplication. Note that the –sc and –tr options can be used to scale and translate
the data or axes by a factor (see bdconv documentation). Note that the –sc and –tr options can
be used together, the scale operation will be done first.

Example:
The file [Link] and [Link] can be added together and saved as [Link] using the
command:
bdutil –o+ [Link] [Link] [Link]

The file [Link] can be scaled by a factor of 2.4 and saved as [Link] using the command:
bdutil –sc2.4 [Link] [Link]

The X data in the file [Link] can be translated by 5 and saved to the file [Link] using
the command:
bdutil –tx5 [Link] [Link]

RSoft CAD  255

bmp2ind – Convert Bitmap Image to a User
Profile
The bmp2ind utility converts an indexed-color bitmap image file into an RSoft data file. This
allows, for example, the use of an SEM image to define the cross-section of a waveguide or fiber
via a user-defined index profile. The utility supports monochrome (1-bit), 16-color (4-bit), or
256-color (8-bit) indexed-color bitmaps.

This utility does not support 24-bit bitmaps. See ‘Background’ section for details about
converting bitmaps to a supported format.

Syntax:
bmp2ind supports the following syntax and options:
usage: bmp2ind [options] <input-file> [<output-file>]

options:
-? help – print this message
-w# the unit width of the output profile
-h# the unit height of the output profile
-b# the lower index that the profile will scale to
-f# the upper index that the profile will scale to
-s# 0-std format, 1-scientific format
-m=$ specifies the index map to be used

Usage:
-w# Sets the width (-w) and height (-h) of the output profile. The default
-h#
value for both of these options is 2.

-b# Sets the data value range that will be mapped to the color palette in the
-f#
bitmap image:
• The –b option sets data value assigned to the lowest color index in
the color palette. The default value for this parameter is 0.
• The –f option sets the data value assigned to the highest color index
in the bitmap file. The default value for this parameter is 1.
The data value for each pixel is calculated as:

RSoft CAD  256

 f ( x, y ) =
ci
cm
( v f − vb ) + vb
Where vb is the data value set by the –b option, vf is data value set by the
–f option, ci is the color index of the pixel, and cm is the maximum color
index in the color palette.
The default data range of (0,1) is designed to create normalized index
profile functions as described in Section 6.A. These options, when used
with a grayscale bitmap, use the gray level to set the data value.
Typically, this means that black (color index of 0) maps to the minimum
data value, and white (color index of 255) maps to the maximum data
value.
The use of these options is illustrated in Example 2 below. Note that the
use of the –m option supersedes these options.

-s# Controls the use of scientific notation in the output data file: a value of 1
uses it, a value of 0 does not.

-m=$ Selects an optional index map file. This file allows you to map specific
color indices in the palette to specific data values. The use of this option
supersedes the –b and –f options.
The index map file is a simple text file that contains multiple lines, each
line which maps a color index or range of color indices to a data value.
For example, a 16-color bitmap might have an index map file like:
0 = 1.0
1-5 = 3.5
6-7 = 1.5
8-9 = 1.8
10-13 = 2.4
14-15 = 1.3333

where, for example, any pixels in the bitmap with a color index of 11
will be assigned a refractive index of 2.4. The use of this feature is
illustrated in Example 1 below.

RSoft CAD  257

Indexed-Color Bitmap Background
Note that our use of the term ‘color’ here refers to an arbitrary RGB (red/green/blue) value,
including white, black, shades of gray, etc.

The bmp2ind utility supports monochrome (1-bit), 16-color (4-bit), or 256-color (8-bit) indexed-
color bitmap image files. An indexed-color bitmap file is a data matrix where each pixel, or
matrix value, is assigned a color index from a color palette, or lookup table, as illustrated in Fig.
1. There are several programs that can be used to save these types of bitmaps as well as
view/modify the color palette. These include Adobe® Photoshop® ([Link]), Microsoft
Paint (saving bmp files only), and IrfanView (free for non-commercial use,
[Link]).

Figure 1: Example 16-color bitmap. a) The bitmap as it would appear on-screen with color
indexes for each point. b) The color palette used to display this bitmap. Each color index
corresponds to a RGB color, for example 9 = red = RGB(255, 000, 000). This is the standard
Microsoft Windows default 16-color palette.

RSoft CAD  258

 Figure 2: Example 8-bit grayscale (256-color) bitmap. a) The bitmap as it would appear on
screen with color indexes for each point. b) The 256-level grayscale color palette used.

The bmp2ind utility converts the color index of each data point, or pixel, to a refractive index
data value. While it is necessary to understand the relationship between color index and RGB
color to have a meaningful data conversion, the actual color map used does not matter. When
used with the –m option, the utility will convert each color index to a particular data value, and
when used with the –f and –m options, the color index will be scaled between the defined
min/max values.

Examples
Consider the two bitmaps illustrated in Figs. 1 and 2; we will convert these into index profile
data files to illustrate the use of the bmp2ind utility:

Note that these bitmaps are very low resolution for illustrative purposes, when converting your
own images it is better to use high-resolution bitmap images.

Example 1: 16-color Bitmap

The bitmap in Fig. 1 represents the cross-section of a typical SOI (Silicon-on-Insular) structure.
For the purposes of this example, assume that the bitmap image represents a 5 μm by 5 μm
region, and that the colors map to actual refractive index values as:

Color Color Desired Desired

Index Material Refractive Index
Red 9 Silicon 3.5
Yellow 11 Silica 1.5

RSoft CAD  259

 Magenta 13 Air 1.0

Based on the information in the above table, we create a simple text file that maps the color
indexes to the desired refractive indexes. One such file is:
0-9 = 3.5
10-11 = 1.5
12-15 = 1.0

Assuming that the bitmap is named [Link] and the index map is named [Link], the
conversion can be done with the following command:
bmp2ind -w5 -h5 =m=[Link] [Link] [Link]

where –w5 and –h5 set the dataset’s dimensions, and [Link] is the output file name. The
resulting data file is shown in Fig. 3 and should be used with the Absolute index in User Profile
option described in Section 6.D.5.

Figure 3: Data file a) as plotted in WinPLOT (with interpolation disabled), and b) the
converted data as seen in a text editor (produced with the –s0 option and slightly edited for
visual clarity).

Example 2: 8-bit Grayscale Bitmap

The bitmap in Fig. 2 also represents the cross-section of a typical SOI (Silicon-on-Insular)
structure. For this example we will use the default data range (0, 1) and the default X and Y
coordinate range (-1, 1) so that it can more easily be used as an index profile in the RSoft CAD.
Assuming the bitmap is named [Link], the conversion can be done with the following
command:
bmp2ind [Link] [Link]

where [Link] is the output file name. The insulator region, which has a color index of 128,
will, according to the equation f(x,y) = (ci/cm)*(vf-vb)+vb, have a data value of (128/255)*(1-
0)+0, or 0.501961. The resulting data file is shown in Fig. 4.

RSoft CAD  260

 Figure 4: Data file a) as plotted in WinPLOT (with interpolation disabled), and b) the
converted data as seen in a text editor (produced with the –s0 option for visual clarity).

RSoft CAD  261

bp2codev and codev2bp – Ray Tracing
Converter for CODE V
The bp2codev and codev2bp utilities convert RSoft field data to Synopsys’ CODE V data and
vice versa.

Converting from RSoft to CODE V (bp2codev)

The bp2codev utility can be used to convert RSoft field files to the CODE V format.

Syntax:
bp2codev supports the following syntax and options:
usage:
bp2codev [options] input-file output-file
bp2codev -V [options] input-file [input-file2] [input-file3] output-file
bp2codev -I [options] input-file output-file [output-file2]

options:
-? help - prints this message
-R CODEV "Real" output format (default is "Complex")
-V CODEV "Vector" output format (Ex and/or Ey and/or Ez input)
-I CODEV "INT" output format (outfile1=FIL [outfile2=WFR])
-L# wavelength (um) [N/A for INT format]
-N# refractive index [N/A for INT format]

Usage:
-? This option prints the help for this utility.

-R This option enables the output of a “Real” formatted CODE V file.

The default, if this option is unset, is “Complex”.

-V This option enables the output of a “Complex” formatted CODE V

file. In this mode, 1, 2, or 3 RSoft field files for different vector
components can be used.

-I This option enables the output of the CODE V “INT” format. In this
mode, the first output file is the “FIL” file, and the second optional
output file is the “WFR” file.

RSoft CAD  262

 -L# This option sets the wavelength in microns of the RSoft field files.
The default takes the wavelength value from the header of the data
files, and if not provided in the header, uses a value of 1.

-N# This option sets the refractive index of the plane where the RSoft field
files are defined. The default takes the refractive index value from the
header of the data files, and if nor provided in the header, uses a value
of 1.

Examples:
To convert the vector RSoft field contained in the files field_ex.dat, field_ey.dat,
field_ez.dat, where the wavelength is defined in the file header, to the CODE V file
[Link] use this command:
bp2codev field_ex.dat field_ey.dat field_ez.dat [Link]

To convert the scalar RSoft field contained in the file [Link] where the wavelength (1.55
μm) is not defined in the file header, to the CODE V file [Link] use this command:
bp2codev –l1.55 [Link] [Link]

Note that the phase in the WFR output file is always wrapped.

Converting from CODE V to RSoft (codev2bp)

The codev2bp utility can be used to convert CODE V files to the RSoft field format.

Syntax:
codev2bp supports the following syntax and options:
usage:
codev2bp input-file-name output-file-name ("Complex" input)
codev2bp FIL-file-name [WFR-file-name] output-file-name ("INT" input)

options:
-? help - prints this message
-A interpret FIL data as amplitude instead of intensity
-P# output plot scripts (1=write script, 2=spawn script)

Usage:

RSoft CAD  263

 -? This option prints the help for this utility.

-A Interprets the FIL dataset as amplitude instead of intensity.

-P This option enables the output of WinPLOT plot files.

Examples:
To convert the vector CODEV field contained in the file [Link] to the RSoft file format with a
prefix of field, use this command:
codev2bp [Link] field

RSoft CAD  264

BSDF Utilities (bsdfgen and bsdfutil)
The RSoft BSDF Utilities bsdfgen and bsdfutil can be used to generate and use BSDF (Bi-
Directional Scattering Function) files.

These utilities are fully documented in the RSoft BSDF Utilities manual.

• The BSDF Generation Utility uses an RSoft product (FullWAVE or DiffractMOD) to

generate an RSoft BSDF file that contains the scattering information of a periodic structure
as a function of input angle(s), polarization, and/or wavelength. The simulations only require
one unit cell of the periodic structure.
• The BSDF Viewer can be used to visualize the contents of an RSoft BSDF file.
• RSoft BSDF files can be used via the RSoft BSDF UDOP can be used to define surface
properties in LightTools.
• The bsdfutil utility can be used to ‘pass’ an arbitrary finite optical field ‘through’ the
BSDF database and reconstruct the transmitted and reflected near-field. This effectively
allows the use of an arbitrary finite optical field with single unit-cell FullWAVE or
DiffractMOD simulations. This would allow, for example, a finite Gaussian beam to be
launched at a periodic surface grating.
• These utilities, along with the Ray Tracing Converter Utility, enable interfaces with ray-
tracing tools such as Synopsys’ CODE V package.

RSoft CAD  265

Custom PDK Utility (smatgen)
The generation of custom PDK models, including s-matrix, effective index, or other data, is a
common task for the optical designer. The RSoft tools include the Custom PDK Utility, see the
Custom PDK Utility manual. The utility automates the calculation of simulation data, mask files,
and creates several PDK-related files including a compact model that can be simulated in
Synopsys’ PIC Solution tools.

RSoft CAD  266

coatuf – Create Coating for User Profile
When used as a height profile on a multilayer component, a user profile data file can be used to
define a surface in the RSoft CAD. This data file can be obtained from Atomic Force
Microscopy (AFM) or other means. The coatuf utility allows users to add a coating on top of
this surface by creating a second user profile data file that can then be used in the RSoft CAD to
define the coating.

Syntax:
coatuf supports the following syntax and options:
usage: coatuf [options] <userprofile-file> <output-file>
options:
-? help - prints this message
-t# set thickness in all directions
-tx# set thickness in X
-ty# set thickness in Y
-tf# set thickness in F (function value)
-sx# set scale in X
-sy# set scale in Y
-sf# set scale in F (function value)

Options:
-? This option display help for this utility.

-t# Sets the coating thickness. The –t option can be used to set all
-tx#
directions equally, alternatively the individual options can also be
-ty#
-tf# used.
-s# Sets the scale factor for applying the coating. In many cases the
-sx#
data in the user function is normalized and will be scaled in one or
-sy#
-sf# more ways in the RSoft CAD. These options set the scale between
the coordinate system in the data file and the final desired
coordinate system. If these options are not set correctly, the
thickness will not have the correct thickness. The data in the file
typically has coordinates in the header to specify the X/Y extent of
the data and the data will have its own extent.
There are many cases, here are two simplest:
• For the case where the data in the file is in real coordinates, i.e.

RSoft CAD  267

 the X/Y coordinates in the header and the data values all
represent the actual data coordinates, then the scale factor
should be 1 for all axes.
• For the case where the data in the header is normalized from -1
to 1 for both X and Y, and the data values are normalized from
0 to 1, the X and Y scale factors should be half of their
respective size in real coordinates and the function scale
coordinate should be the actual total thickness of the uncoated
surface.

Example:
To add a thickness of 0.3 μm in all directions to the user profile contained in the file
[Link] and save it in the file profile_coat.dat where the data in the profile uses the
normalized coordinate system (X/Y from -1 to 1, data form 0 to 1) but will have a final size of 10
μm in X, 4.5 μm in Y, and a uncoated thickness of 2.3 μm, use this command:
coatuf –t0.3 –sx5 –sy2.25 –sf2.3 [Link] profile_coat.dat

The profile_coat.dat file can then be used as a height profile for a second segment in the
CAD that is positioned at the same location as the uncoated segment.

RSoft CAD  268

disperse – Calculate Dispersion Relations
The disperse utility takes a file containing effective index values and produces a dispersion
relation.

Syntax
disperse supports the following syntax and options:
usage: disperse [options] neffdatafile
options:
-v view derivative data in Notepad
-p plot derivative data in WinPlot
-b compute derivatives of beta vs k instead of neff vs x
-g compute group velocity and dispersion vs lambda
-x# x data is (0=lambda,1=1/lambda,2=k) (default=0)
-c y data is complex (i.e. "nk" data)
-k output nimag if present
-fp# enable polynomial fit (default=disabled, # is order
[default=3])

The input file neffdatafile must contain effective index data as a series of columns, with the
first column containing wavelength/frequency data as indicated by the –x option.

RSoft CAD  269

dxf2ind – Convert DXF Mask to .ind File
See Mask Conversion Utilities section.

RSoft CAD  270

ffutil – Far-Field Utility
This program calculates multi-plane far-fields.

Syntax
ffutil supports the following syntax and options:
usage: ffutil [options] [<ind-file>]

options:
--prefix $ prefix for near field files
--nu# set frequency (default=1.0)
--nphi# set number of phi points (default=72)
--ntheta# set number of theta points (default=36)
--nf-index# index of near field media (default=1.0)
--ff-use-planes$ near field planes to use (default=xp,xm,yp,ym,zp,zm)
--ff-preserve preserve near field files (default=0)

Usage
ffutil calculates multi-plane far-field with these options:

--prefix $ Sets the Output Prefix of the near-field files.

-nu# Sets the frequency in μm-1, set to 1 by default.

-nphi# Sets the number of phi and theta points, defaults to 72

-ntheta#
and 23 by default.

-nf-index# Sets the near-field index, defaults to 1.0.

-ff-use-planes$ Select the planes to use, by default all 6 are used.

--ff-preserve Save the near field files after the calculation.

RSoft CAD  271

fhakit – Fast Harmonic Analysis of Complex
Exponential Time Series
The analysis of cavity modes and Q-factors by FullWAVE’s Q-Finder tool is based on the Fast
Harmonic Analysis (FHA) technique discussed in Chapter 9 of the FullWAVE manual. Given a
time (or distance) series that is expected to be a sum of complex exponential terms (or
approximately so):

f (t ) =  ck exp ( i 2 vk t ) + c.c.
k

FHA can efficiently extract the complex frequency and amplitude of those terms which fall
within a narrow frequency range.
This analysis can be performed directly at the command line using the utility fhakit. This can
be useful when you need to analyze a series that was not produced by an automated Q-Finder
calculation, or if you wish to analyze a signal in a different way to that performed by Q-Finder.
For example, from a single propagation, you could extract the complex frequencies of cavity
modes that occur in quite different frequency bands by running fhakit several times on the
same .tmn file. In contrast, because it searches dynamically for a resonance, Q-Finder focuses on
a single frequency band in each run.

Syntax
fhakit supports the following syntax and options:
• Extract complex frequencies and amplitudes using Fast Harmonic Analysis method:
fhakit -fha [options] filename prefix

• Perform FFT and extract real peak frequencies:

fhakit -fft [options] filename prefix
options:
-help [show this message]
-v <int> [verbosity (default=0)]
-q [quiet mode (default=0). Don’t write to
standard output.]
-dt <float> [time (or distance) step. [default=value
from column zero]]
-cols <string> [comma-separated list of columns to
analyze [default=column one]]
-complex [interpret columns as complex pairs
[default=no]]
-J <int> [number of local basis states for FHA
expansion]

RSoft CAD  272

 -numin <float> [minimum search frequency (required for
FHA)]
-numax <float> [maximum search frequency (required for
FHA)]
-toff <int> [time offset (number of points to drop at
start of time series). [default=0]]
-tmax <int> [maximum number of time steps of time
series to use [default=all]]
-n [don't suppress negative frequencies]
-t <float> [threshold: fraction of total energy to
accept a peak [default=1e-5]]
-rt <float> [relative threshold: fraction of energy
of maximum accepted peak to accept peak
[default=1e-4]]
-tol <float> [tolerance for consistency check of FHA
peaks [default=1e-3]]
-spec [interpolate spectrum [default=0]]
-si [sort in order of decreasing intensity
instead of increasing frequency]

Usage:
fhakit can be used in two modes, FHA or FFT, which are selected by the flag –fha or –fft. In
each case, fhakit reads data from a file <filename> and writes a list of detected peaks and
properties to a file named <prefix>_m.nufha or <prefix>_m.nufft, where m is an integer from
0 indicating the particular set of columns being analyzed. Intermediate output is written to a file
named <prefix>_fha.log.
In each mode, fhakit looks for all resonances in the bandwidth between numin and numax
whose amplitudes satisfy certain criteria discussed in the explanation of options below.
The output files contain the following data: resonance index, complex angular frequency,
complex frequency, complex amplitude, quality factor, an internal consistency error, the mean
“energy” of the resonance, and the fraction of total energy in the signal represented by this
resonance. The mean energy is simply the mean value of the norm of the function f:

1 N −1
 f j (t )
2
E=
N j =0
In FFT mode, the imaginary parts of the frequency and the Q factor are not available and are
recorded as zeros. Further, the values of peak and fractional energy are not very meaningful.
However, in FFT mode only, the total energy in the band between numin and numax is recorded
in the header as the Bandwidth energy.
By default, fhakit interprets the zeroth column of a data file as the time variable (which must be
equally spaced) and reads real signal data from the first column. However, multiple columns
including the zeroth column may be analyzed using the –cols option and interpreted as complex

RSoft CAD  273

data using the –complex flag. If the zeroth column is used as signal data, then the time step must
be specified using the –dt option.

Options:
-help This option prints the help for this utility.

-v <int> Verbosity. A value of 0 means low output, 1 means verbose

output. In particular, in verbose mode, a list of candidate
eigenvalues is output to the screen. The candidate values list all
detected resonances that lie within the requested frequency band
regardless of their strength. This list of candidates is then culled
according to the –tol, –t and –rt options. Thus this mode can
be useful for identifying small peaks which are not showing up in
the final answer.

-q Sets quiet mode. In this mode, all output is written only to log
files. There is no text written to standard output.

-dt <float> Sets the value of the increment between time steps. This option is
required if the -cols option specifies that the zeroth column is to
be interpreted as data. Otherwise, it is optional and can be used
to scale frequencies differently to the default time step extracted
from the zeroth column.

-cols <string> This option takes a comma-separated list of positive integers that
specifies which columns of the file (counted from zero,) should
be analyzed. The first requested column produces data in the file
<prefix>_0.nufha, the second column produces
<prefix>_1.nufha, etc. If the -complex option is set, the
columns are treated as real and imaginary pairs and there must be
an even number of columns specified.

-complex Indicates that the input file columns should be interpreted as

real/imaginary pairs. If the -cols option is not specified, then
columns 1 and 2 are analyzed as a complex pair.

RSoft CAD  274

-J <int> Number of basis states used for FHA calculation. Values above
400 or so will run increasingly slowly. Normally this should not
need to be set.

-nummin <float> Minimum/maximum value of frequency range in which to look

-nummax <float>
for peaks. The argument is a plain frequency, not an angular
frequency.

-toff <int> Number of time steps to ignore at the start of the time series. The
default is 0. This can be useful if the start of the time series is
noisier than the rest of the signal and gives rise to spurious weak
resonant peaks.

-tmax <int> Maximum number of time steps to include. This can be used to
truncate a signal which degenerates into noise at large values of
time. If both –toff and –tmax options are specified the range of
time steps sampled is[toff, tmax+toff).

-n Include negative frequency peaks in analysis. By default, fhakit

ignores negative frequencies. For a “clean” real signal, every
positive frequency peak should be accompanied by a negative
frequency of the same amplitude. For complex signals, the
positive and negative frequency peaks are in general different.
With this option, the allowed frequency range is effectively the
union of [-numax, -numin] and [numin, numax].

-t <float> Specifies the threshold peak energy as a fraction of total energy

in the signal. Any peak accounting for less than this fraction of
energy is discarded. The default is 10e-5. Use a smaller value if
you find peaks of interest are being ignored.

-et <float> Specifies the relative threshold peak energy as fraction of the
energy of the largest accepted peak. Any peak accounting for less
than this fraction of the energy of the largest peak in the
frequency search range is discarded. The default is 10e-3. Use a
smaller value if you find peaks of interest are being ignored.

RSoft CAD  275

-tol <float> Specifies the tolerance of the eigenvalue consistency error. A
consistency check is performed on each candidate peak to detect
spurious peaks. The acceptable error can vary greatly depending
on the quality of the signal. The default value is 10e-3, but for a
low-noise signal, the achievable consistency error may be much
lower. On the other hand, for a noisy signal, it may be necessary
to set values as large as 10 or more. Use the following strategy
to set this option. At first, use the default value. If fhakit fails to
find the expected peaks, run it in verbose mode and check the
consistency error reported in the list of candidate resonances. If
all these errors are larger than the default 10e-3, then use this
option to set a suitable value.

-spec In FHA mode, generates a plot named <prefix>_spec_fha.pcs

that represents the complex spectrum corresponding to the
detected resonances. In FFT mode, generates a plot named
<prefix>_spec_fft.pcs that contains the complex FFT in the
specified frequency range.

-si Causes resonant peaks to be sorted in order of decreasing peak

intensity instead of increasing real frequency.

RSoft CAD  276

findpks – Find Peaks Utility
The findpks utility can be used to find peaks in a data file.

Syntax:
findpks supports the following syntax and options:
usage: findpks [options] <infile >outfile
pptions:

-n# set maximum number of peaks (default=5)

-t# set threshold (fraction of maximum, default=0.01)
-m# only use first # peaks for determining maximum (default=all)
-mx# only use peaks with x<# for determining maximum (default=all)
-minx# output only peaks with x>=# (default=all)
-maxx# output only peaks with x<=# (default=all)
-x exclude peak at x=0
-s# set scale factor by which to multiply peak positions
-v# set optional value to output before peaks
-c<cols> set col number(s) to search for peaks (default=1,noargs=all)
(<cols> = comma-separated list of #'s or #-# ranges)
-cr# set relative tol for multicolumn search (default=0.5*dx)
-ca# set absolute tol for multicolumn search (default=N/A)

Options:
-n# Sets the maximum number of peaks to find. By default, 5 peaks are
found.
-t# Sets the threshold used to define a peak. This is set as a fraction of
the maximum peak value in the file, by default it is 0.01.
-m# Sets the number of peaks to use when determining the maximum
peak value.
-mx# Use only peaks with x > # when determining the maximum peak
value.
-minx# Output peaks with x > min or x < max
-maxx#
-x Exclude peak at x = 0.
-s# Sets a scale factor by which to multiple peak positions.
-v# Sets optional value to output before peak, useful for building data
files to be plotted.

RSoft CAD  277

-c<cols> Sets the columns within which to search for peaks. Default = all.
-cr# Sets relative tolerance for multi-column search (default = 0.5*dx)
-ca# Sets absolute tolerance for multi-column search.

RSoft CAD  278

fitdisp – Dispersion Fitting Utility
The fitdisp utility can be used via a graphical interface in the Material Editor as described in
Section 8.B.4.

The fitdisp utility allows users to create Lorentzian dispersion models from data files that can
be used with the Material Editor in the RSoft CAD. The data file should have two or three
columns: the first with frequency or wavelength data, the second with the real refractive index
data (epsilon or index), and a third optional column with the imaginary index data (epsilon or
index).
The fitdisp utility will display the coefficients for the material fit as well as create a file name
[Link] which will have the material defined in the Material Editor. This file can be opened in
the RSoft CAD, and using the Group Library described in Section 8.A.1, this material can be
transferred to other .ind design files.

The fitdisp utility is primarily useful for FullWAVE users who want to use a Pulse simulation
which requires the use of Lorentzian dispersion model. Users of other tools such as
DiffractMOD, BeamPROP, and/or FemSIM, of FullWAVE users using CW simulations, can use
the raw data directly in the Material Editor without having to use a curve fit.

Syntax:
fitdisp supports the following syntax and options:

usage: fitdisp [options] input-file

(input-file is 'wavelength(um) n [k]' by default)

options:
-? help - prints this message
-xtype# x data type, 0=Frequency(1/um), 1=Wavelength(um) (default=1)
-ytype# y data type, 0=Eps, 1=Index (default=1)
-xytype# sets xtype and ytype together
-o# order, i.e. # of Lorentzians to be used (default=1)
-epsinf# epsilon at infinite frequency (default=1)
-drude force first term to be Drude (c=0)
-x#,# set x range to fit (default=all data)
-v# verbosity (default=3)
-p# plot results, 0=none 1=.pcs 2=.pcs+spawn (default=1)
-out=$ prefix for output files (default="out")

RSoft CAD  279

Usage:
-? This option prints the help for this utility.

-xtype# This option sets the type of X data in the data file. A value of 0
indicates frequency data in μm-1, and a value of 1 (the default)
indicates wavelength data in μm.

-ytype# This option sets the type of Y data in the data file. A value of 0
indicates epsilon data, and a value of 1 (the default) indicates index
data.

-xytype# This option provides a convenient way to set the X and Y data types
together.

-o# This option sets the number of Lorentzians to use for the curve-fit. By
default, one Lorentzian is used.

-epsinf# This option sets the epsilon value at infinite frequency and defaults to
a value of 1.

-x#,# This option sets the X data range to use for the fit. By default, all the
actual data in the file is used.

-v This sets the verbosity. A value of 3 is used by default.

-p This option controls the output of a WinPLOT plot file. A value of 0

disables plot output, a value of 1 creates a WinPLOT plot file, and a
value of 2 creates a WinPLOT plot file and opens it.

-out=$ This option sets the output prefix. Note that you must use an equal
sign for this option, e.g. –out=run3.

RSoft CAD  280

Example:
To attempt a 4 Lorentzian curve-fit of the refractive index data contained in a file [Link],
(which has three columns, wavelength, real index, and imaginary index), use this command:
fitdisp –o4 –out=fit [Link]

To attempt a 4 Lorentzian curve-fit of the refractive index data contained in a file [Link],
(which has three columns, frequency, real epsilon, and imaginary epsilon), use this command:
fitdisp –o4 –xytype0 –out=fit [Link]

Another example of material data curve fitting can be found in Tutorial 5 in the FullWAVE
manual.

RSoft CAD  281

fwmodekit – Extract Data From FullWAVE State
File
fwmodekit allows the user to extract data from state files created via the Continue Simulation
option. This utility is unsupported and its behavior is subject to change.

Syntax
fwmodekit supports the following syntax and options:
• fwmodekit (-h|--help)

Prints help message.

• fwmodekit --version

Prints version.
• fwmodekit (-s|--slice) (-x x0|-y y0|-z z0) --prefix <prefix> [--nobcs]
(ex|ey|ez|hx|hy|hz|eden|hden|tden) <fstfile> <indfile>

For a 3D simulation, extracts and plots a 2D slice of field component to a plot file.
For a 2D simulation, extracts and plots the entire domain to a plot file.
• fwmodekit (-f|--fslice) (-x x0|-y y0|-z z0) --prefix <prefix> [--nobcs] [-
-lightcone <lightcone k-radius>] (ex|ey|ez|hx|hy|hz|eden|hden|tden)
<fstfile> <indfile>

Extracts and plots fourier transform of slice of field component to a plot file.
• fwmodekit (--view3d) --prefix <prefix> (ex|ey|ez|hx|hy|hz|eden|hden|tden)
<fstfile> <indfile>

For a 3D simulation, extracts and plots a 3D view of the field components. No symmetry
boundary processing is applied.
• fwmodekit (--fileslice) --prefix <prefix> (-x x0|-y y0|-z z0) <slicefile>
<indfile>

Takes an existing slice file on disk and uses the boundary conditions to extend the file to the
full domain. The slice direction option (-x, -y or -z) is only required for slices taken from 3D
calculations and the value (x0, y0, z0) is not important and can be set to 0.
• fwmodekit (-c|--compare) <file1>.fst <file2>.fst

Compares fields in two fst files and writes their fractional difference to standard output.
• fwmodekit (-a|--analyze) <file1>.fst

Summarizes properties of .fst file.

• fwmodekit (-v|--volume) --prefix <prefix> [--nu nu0] <fstfile> <indfile>

RSoft CAD  282

 Finds the mode volume of the mode stored in the .fst file. The mode volume is defined as

 U ( x ) dx
3
E
V=
max U E ( x ) 

where UE(x) is the electric energy density. The resonant frequency can be specified. This
helps to determine the required simulation length and is recommended though not required.
When fwmodekit detects a symmetry-based subdomain calculation, the volume is multiplied
by 2 for each direction in which symmetry conditions have been applied. This can be
suppressed using the option --nobcs.

Options:
-q This option enables ‘Quiet mode’ which suppresses display of
the generated file in a plot window.

-g Run FullWAVE calls in a GUI window.

--nobcs Suppress use of boundary conditions to unwrap field profiles on

subdomains.

--np <int> Number of machines required for cluster calculation.

--machinefile File of machine names required for cluster calculation.

<machfile>
--timesteps <nsteps> Number of timesteps before output (default=1).

--nu Resonant frequency of mode under analysis.

RSoft CAD  283

ind2gds, ind2dxf – Convert .ind File to Mask
See Mask Conversion Utilities section.

RSoft CAD  284

gds2ind – Convert GDS Mask to .ind File
See Mask Conversion Utilities section.

RSoft CAD  285

killall – Utility to stop processes
The killall utility can be used to stop processes with a particular name.

WARNING: This utility can be used to terminate any running process, not just Synopsys
processes. Use at your own risk.

Note that this utility is only included on Windows operating systems. Linux systems typically
have a similar utility. Note that the Linux syntax may be different what is documented here,
please check the killall man page on your system.

Syntax
fwmodekit supports the following syntax and options:
killall <process_name>

where <process_name> is the name of the process.

Example:
To stop all RSoft CAD processes on your system, use this command:
killall bcadw32

RSoft CAD  286

Mask Conversion Utilities
The RSoft tools include various utilities to convert to and from common mask formats:

GDS Utilities
RSoft .ind files can be converted to and from GDS masks via the gds2ind and ind2gds
utilities.

GDS to .ind File: gds2ind

This utility creates a .ind file from a GDS mask file. Note that these options can also be used
when loading a GDS file via a circuit reference as described in Section 6.F.2.

Syntax:
gds2ind supports the following syntax and options:
usage: gds2ind [options] gds-input-file-name [ind-output-file-name]
options:
-? help - prints this message
-i# show structure info only (#=hierarchy-level, default=1)
-n$ set structure name to process (default=last)
-l$ set layer name to process (default=all)
-h# set default layer thickness (default=1)
-xy import to XY plane

Options:

-i# Display structure information, # sets hierarchy level.

-n# Sets the structure or cell name to process.
-l$ Select the layer to process.
-h# Sets the default layer thickness (can be changed in .ind file later).
-xy Import to the XY plane instead of the XZ plane.
-pv,-pw,-ps Enables the automatic placement of vertices at ‘port’ locations. See
Section 6.F.8 for more details.

Ind file to GDS: ind2gds

This utility creates a .gds file from an RSoft design file (.ind). Note that this utility is the same
used by the GDS export option described in Section 6.E.1.

RSoft CAD  287

Syntax:
ind2gds supports the following syntax and options:
usage: ind2gds [options] ind-file-name [name=value ...]
options:
-? help - prints this message
-o=$ set output file (default=<infile>.gds)
-r# set resolution (CAD preferences set default)
-l# map layer number 0 to #
-d# map data type 0 to #

Options:

-0=$ Sets the output file, by default the input file name will be used.
-r# Sets the output resolution, defaults to the CAD preference.
-l$ Maps layer number 0 to #.
-d# Maps data type to #

Notes:
Many options can be set via symbols (name = value), including:

Symbol Description
gds_export_dir Sets the cross-section direction to export. A value of 1 = X
(YZ plane), 2 = Y (XZ plane), and 3 = Z (XY plane).
gds_export_pos Sets the export cross-section position.
gds_clipping Enables clipping the structure.
gds_clip_xmin Sets the clipping ranges. If unset or set to default, the
gds_clip_xmax
whole structure will output.
gds_clip_ymin
gds_clip_ymax
gds_clip_zxin
gds_clip_zmax

DXF Utilities
RSoft .ind files can be converted to and from DXF masks via the dxf2ind and ind2dxf
utilities. Use the -? option to view the options supported by each utility.

RSoft CAD  288

mat2bp – Convert Matrix Data to RSoft Format
This program converts raw matrix data into the RSoft file format.

Syntax
mat2bp supports the following syntax and options:
usage: mat2bp input-file-name output-file-name
options:
-? help - prints this message
-x<range> set range in x (<range> = #,#)
-y<range> set range in y (<range> = #,#)

Usage
mat2bp converts a raw matrix of data contained in a data file into a file in the RSoft format
which could then be used as an index profile, input field, etc. Since the RSoft file format contains
information which relates the domain of the data, the user needs to provide this information in
order to convert the matrix into the RSoft format. The mat2bp utility supports these options:

-x#,# This option sets the X domain for the data file. The range should be
input as #,#, where the first number is the domain min, and the
second number is the domain max. These numbers will be used to
build the header for the new RSoft Data file that will be output.

-y#,# This option sets the y domain for the data file. The range should be
input as #,#, where the first number is the domain min, and the
second number is the domain max. These numbers will be used to
build the header for the new RSoft Data file that will be output.

The utility will automatically add the appropriate header lines to the file.
e.g. to covert the file [Link] into a RSoft file [Link], where the input file is defined over
an X range of (-10,10) and a Y range of (-1,6):
mat2bp –x-10,10 –y-1,6 [Link] [Link]

RSoft CAD  289

mathmat – Do Arithmetic with Data Files
This program accepts multiple input files and combines the columns using mathematical
expressions specified by the user. Any expression that would be valid in the CAD symbol table
may be used. A circuit file may be specified to take advantage of existing symbols and new
symbols may be defined in the command itself.
mathmat is often useful with BandSOLVE band structure output, since BandSOLVE results have
the wavevector as the independent variable when often we wish to express results as a function
of the frequency. Tutorial 6 in the BandSOLVE manual contains demonstrations of this
technique.

Syntax
mathmat supports the following syntax and options:
usage: mathmat <symbol_table_expr> [input_files]

options:
-help [show this message]
-longhelp [show full help]
-p <int> [output precision (default=10)]
-f [use fixed point format]
-nohead [disable header]
-trunc [truncate at shortest file]
-bp <string> [use symbol table from RSoft .ind file]
-sym <string> [define new symbols]
-maxy [in all columns (except 0), set output_val to ‘.’
if abs(output_val) > maxy]

Usage:
mathmat reads from standard input or a list of input files, performs operations on the columns
defined by a list of symbol table expressions and writes the answer to standard output. The
operations are performed on each row in the column separately. The columns in the first input
file are named 'a0', 'a1', 'a2',... The columns in the second input file are named 'b0', 'b1', 'b2',...
and so on. As a quick example, this command will take the average of columns 1 and 2 in a file
and write it as a function of column 0:
mathmat 'a0,(a1+a2)/2' [Link]

It is also possible to create commands that perform operations on entire files that have the same
shape. For example, this command will take the average of two files and write the results:
mathmat '(__a+__b)/2' [Link] [Link]

RSoft CAD  290

mathmat write output to standard output (usually the terminal or command prompt). To save data
in a new file, use the redirection operator ‘>’. For example, the simple command above to take
the average of two columns can be saved in a file [Link] with this command:
mathmat ‘a0, (a1+a2)/2’ [Link] > [Link]

Referencing a column which does not exist or using more letter prefixes than there are input files
causes a fatal error.
These names are used in the symbol table expression which must be a comma separated list of
normal math expressions using these variable names.

Options:
-help These options display help for this utility.
-longhelp

-p <int> This option sets the output precision used. The default is 10 digits.

-f This option enables the use of a fixed point format.

-nohead This option disables the use of mathmat headers.

-trunc This option truncates the output data at the shortest file.

-bp <string> This option enables the use of symbols from an existing .ind file.

-sym <string> This option allows the definition of new symbols.

Examples:
eg. Take the average of columns 1 and 2, and write as a function of column 0:
mathmat 'a0,(a1+a2)/2' [Link]

eg. Take the average of columns 1 in two files, and write as a function of column 0 of first file:
mathmat 'a0,(a1+b1)/2' [Link] [Link]

eg. Calculate a complicated expression:

RSoft CAD  291

 mathmat 'a0,sqrt(a1/pi)*(a2*3+b1*exp(b2))' [Link] [Link]

eg. Use symbols Period and delta defined in a file [Link]:

mathmat –bp [Link] ‘a0*Period, delta*a1’ [Link]

eg. Define and use symbol scale:

mathmat 'a0*scale' -sym'scale=3.1' [Link]

RSoft CAD  292

parutil – Generate nk data files from TCAD
Sentaurus parameter files
The parutil utility performs operations using the TCAD Sentaurus parameter files and/or the
complex refractive index (CRI) library from Sentaurus. It extracts wavelength dependent CRI
information into a n/k data file which can be used in the RSoft tools.

Syntax
parutil supports the following syntax and options:
usage: parutil [options] <material> [<out-file>]
options:
-h help - prints this message
-prefix=$ output prefix (<out-file>=<prefix>_nk.dat)
-parfile=$ TCAD parameter file
-swbnode# TCAD node number for preprocessing parfile
-checkmat checks if material was found

parutil takes as input the material name, <material>, and creates a corresponding nk data file
containing wavelength dependent CRI information. The CRI information is either obtained from
the explicitly defined parameter file, -parfile, or from the material definition in the Sentaurus
database. If the material definition is found in the explicitly specified parfile it takes
precedence over the default definition in the Sentaurus database. If the parfile has
preprocessing or #include directives a valid Sentaurus Workbench node number, associated with
the RSoft tool, needs to be specified via the -swbnode option to preprocess the parfile. The
name of the output file, <out-file>, to be generated can either be explicitly specified as a
command line argument after <material>, or the output prefix to be used can be specified via
the -prefix option. If neither option is specified, then the name of the output file would be
<matname>_nk.dat. If the -checkmat option is selected it checks if the material is found either
in the explicitly specified parfile or in Sentaurus Device CRI library and prints it to standard
output; no n/k file gets created.

RSoft CAD  293

Ray-Tracing Conversion Utility
RSoft provides a utility to convert between popular Ray-Tracing file formats, including
Synopsys’ CODE V, and RSoft’s field files. A description of the usage of this utility can be
found in Tutorial 9 of the BeamPROP manual.

The Ray-Tracing Conversion Utility uses the bp2codev, codev2bp, and bdconv utilities to
perform the conversion. These utilities are documented in Appendix E. When using this utility,
note that some options such as the wavelength, index, etc. can be set via the Options button.
These options correspond to the same options that the command line utilities use.

RSoft CAD  294

rsanimate – Create Animations from Plot Files
This program is designed to create animations from a series of (up to 1000) WinPLOT plot files.

Animations can be created using the utility described in this section or using the Export
Animation DataBROWSER feature described in Chapter 2 of the DataBROWSER manual.

Syntax:
rsanimate supports the following syntax and options:
usage: rsanimate [options] <file-list>
options:
-? help - prints this message
-o <file> set output file name (must include .gif or .mpeg
extension)
-p <opts> set additional plot options (could use @ to include
a script)

Options:
-? This option display help for this utility.

-o Sets the output file name. GIF and MPEG files are supported; the
names must include the .gif or .mpeg extension.
-p Add additional WinPLOT plot options to each plot.

Example:
To create an animation of the WinPLOT files [Link], [Link], and [Link] and save
it as [Link], use this command:
rsanimate –o [Link] [Link] [Link] [Link]

To create the same animation, but using the /zw WinPLOT command to automatically scale the
color scale to the data, use the command:
rsanimate –o [Link] –p /zw [Link] [Link] [Link]

RSoft CAD  295

rscombine – Combine Frequency Results
rscombine is a utility which combines absorption monitor profiles obtained at multiple
frequencies into a single absorbed photon density for use in an electronic simulator like
Sentaurus Device.
It supports FullWAVE FDTD pulsed output, FullWAVE FDTD CW scan output, and
DiffractMOD RCWA wavelength scan output, using the monitor information in the .ind file to
get the names of the files. It can also apply weights through the -srcspecfile option and has an
-apd option to produce absorbed photon density (see discussion for apd calculation in tdrutil in
Section E.S). See full list of options below.

Syntax:
rscombine supports the following syntax and options:
usage: rscombine [options] <output-file-prefix>
general options:
-? help - prints this message
-prefix=$ prefix used for data generation
-indfile=$ indfile used for data generation (default='prefix'.ind)
-srcspecfile=$ source spectrum filefrequency only)(default=whitelight)

extra options for apd usage for absorption monitors:

-apd convert absorption to absorbed photon density before
combining
-int# set plane wave intensity (W/cm^2, default=1)
-pow# set guided wave power (W)
-outvar=$ name of output dataset (default=AbsorbedPhotonDensity)
-outunits=$ set units of output dataset (default=cm^-3*s^-1)

extra options for CW scan usage:

-spec# specification 0-frequency, 1-wavelength (default=0)
-min# minimum frequency/wavelength depending on spec
-max# maximum frequency/wavelength depending on spec
-num# # of frequency/wavelength points

Example:
To combine frequency simulation results calculated by running file [Link] with an Output
Prefix of run4, calculating the Absorbed Photon Density, an input intensity of 4, and over a
wavelength range of 0.3 to 2.2 μm, saving results with a prefix new_results, use this command:
rscombine –prefix=run4 -indfile=[Link] -apd -spec1 -int4 -min0.3
-max2.2 new_results

RSoft CAD  296

rspython – Python Environment
The rspython utility is a built-in Python interpreter capable of running Python scripts.

See Appendix H for details on supported Python versions and usage scenarios.

This utility can be run standalone and is also automatically used when processing Circuit
References that contain .py files or running MOST User Simulators. See Appendix H for details
on supported Python versions.

Syntax:
rspython supports the following syntax and options:
usage: rspython <python-file>

Note, running without a <python-file> will open an interactive Python environment.

RSoft CAD  297

shufflemat – Rearrange Data Files
This program is designed for reordering or extracting the columns in one or more data files. An
online manual page can be obtained by typing “shufflemat - help” or
“shufflemat - longhelp” at a command prompt.

Syntax:
shufflemat supports the following syntax and options:
usage: shufflemat [options] [output_files]

options:
-help [show this message]
-longhelp [show full help]
-t [transpose input data]
-ec <string> [extract named columns]
-pc <string> [permute named columns]
-er <string> [extract named rows]
-pc <string> [permute named rows]
-elt <string> [extract named element]
-p <int> [output precision (default=10)]
-f [use fixed point format]
-h [keep column titles]
-numsym <string> [treat this string in data files as a null value]
-nullval <float> [set internal value for null symbols (default 0)

shufflemat reads from standard input, rearranges or extracts the columns or rows and writes to
specified output files, or to standard output if no output files are named. The columns and rows
in the input file are denoted by integers starting from 0. As a quick example, this command will
write columns 1, 2, and 3 of the file [Link] to standard output:
shufflemat -ec 0,2,3 < [Link]

shufflemat write output to standard output (usually the terminal or command prompt). To save
data in a new file, use the redirection operator ‘>’. For example, the simple command above can
be saved in a file [Link] with this command:
shufflemat -ec 0,2,3 < [Link] > [Link]

Options:
-help This option displays the short help for this utility.

-longhelp This option displays the long help for this utility.

RSoft CAD  298

-t The -t (transpose) option takes the whole matrix and writes the
transpose matrix to output. Any header lines in the input are
ignored. Only one output file is allowed.

-ec <string> The ‘extract’ options take a string argument specifying which
-er <string>
columns (-ec) or rows (-er) should be output. All other
columns/rows are dropped. The argument consists of a comma
separated list of column numbers. Ranges may be specified using a
hyphen (-) and multiple lists may be separated by a semicolon (;)
Each range is output to a different filename specified on the
command line. There can be no white space in the range
specifications. If semicolons are used, the extraction string must be
protected from the shell by quotes (double-quotes for
DOS/Windows).

-pc <string> The ‘permute’ options write all input to output but performs
-pr <string>
permutations or cyclic rotations of specified columns (-pc) or rows
(-er). Permutations are expressed as a space-separated list of
integers inside parentheses. Several permutations can be expressed
together, but may have no common column numbers. Permutations
lists separated by semicolons will be directed to different output
files. The permutation string must be always protected from the
shell by quotes (double-quotes for DOS/Windows).

-elt <string> This option extracts a specific element from a file. The string
argument should be the row/column of the element.

-p <int> This option sets the output precision used. The default is 10 digits.

-f This option enables the use of a fixed point format.

-h This option keeps the column titles.

-numsym Sets the string used in the data file(s) for null values.

RSoft CAD  299

 -nullvall Sets the value used for null values in data files (default 0).

Examples:
eg. Place the transpose of the file [Link] in the file [Link].
shufflemat -t < [Link] > [Link]

eg. Write columns 0, 2 and 3 of [Link] to standard output:

shufflemat -ec 0,2,3 < [Link]

eg. Write columns 1,2,3 and 5 of [Link] to standard output:

shufflemat -ec 1-3,5 < [Link]

eg. Write columns 1,2,3 to [Link] and columns 4 and 6 to [Link]:

shufflemat -ec "1-3;4,6" [Link] [Link] < [Link]

eg. Swap columns 2 and 3, writing to standard output:

shufflemat -pc "(2 3)" <[Link]

eg. Rotate columns 2, 3 and 6 writing to standard output:

shufflemat -pc "(2 3 6)" <[Link]

eg. Rotate columns 2, 3 and 6, and columns 4 and 5, writing to standard output:
shufflemat -pc "(2 3 6)(4 5)" <[Link]

eg. Rotate columns 2, 3 and 6 in one output file and columns 4 and 5 in another file:
shufflemat -pc "(2 3 6);(4 5)" <[Link] [Link] [Link]

eg. Extract element 2,3 from one file, writing to standard output:
shufflemat –elt2,3 <[Link]

RSoft CAD  300

smatgen - Custom PDK Utility
See Custom PDK Utility section.

RSoft CAD  301

tailmon – Extract Final Value Utility
The tailmon utility extracts the final value from a text output file, useful when combing the
result of multiple simulations. It is analogous to the Linux tail command.

Syntax:
tailmon supports the following syntax and options:
usage: tailmon [options] <monfile
options:
-v<first-value>

Options:
-v# Replace the value in the first column with #

Notes:
This utility uses typical redirection inputs/outputs, and takes input via the < operator, and will
write output to the screen or a file via the > operator (write new file) or >> operator (append to
file).

RSoft CAD  302

tdrutil – TDR File Utility
The TDR Utility (tdrutil) performs several useful operations using Sentaurus TDR files. The
syntax for this utility is described in this section along with each of the 5 use scenarios and
appropriate options.

Syntax:
tdrutil supports the following syntax and options:
usage: 1. Display Information:
tdrutil [options] <[Link]>
2. Extract/Convert Datasets to a Tensor Mesh:
tdrutil [options] <[Link]>[:<dataset>] <[Link]>
3. Extract Datasets to a Compatible Mesh:
tdrutil [options] <[Link]> <[Link]> <[Link]>
4. Generate Absorbed Photon Density from RSoft Monitor:
tdrutil -apd [options] <[Link]> <[Link]>
5. Generate an RSoft CAD file with dynamic reference to a TDR file:
tdrutil [options] <[Link]> <[Link]>

general options:
-? help - prints this message
-info# set information level (from 1 to 5, default=2)

options for usage 2:

[:<dataset>] set name of dataset to extract (default=first applicable)
-var=$ set name of dataset to extract (default=first applicable)
-getdnk extract index perturbation (if present)
-sc# scale dataset by argument (default=1)
-tgf# set tensor grid factor (all dirs)
-tgs# set tensor grid size (all dirs) (um)
-tg[x|y|z]# set tensor grid size (indicated dir) (um)

options for usage 4:

-lam# set wavelength (um)
-int# set plane wave intensity (W/cm^2, default=1)
-pow# set guided wave power (W)
-outvar=$ set name of output dataset (default=AbsorbedPhotonDensity)
-outunits=$ set units of output dataset (default=cm^-3s^-1)

Options:
The options for tdrutil are broken up into the 5 use scenarios. Each use is described here along
with an example.

RSoft CAD  303

General Options:
-? This option display help for this utility.

-info# Sets the information level used in utility log. Values are from 1 to
5, the highest level of detail. The default is 2.

Usage 1 - Display TDR Information:

This command will display information about the TDR file, including details on the datasets in
the file, statistics on the regions in the file, etc. The -info# option can be used to set the level of
detail output. The syntax for this usage is:
tdrutil [options] <[Link]>

For example, the command:

tdrutil [Link]

displays information about the data in the TDR file [Link].

Usage 2 - Extract/Convert Datasets to a Tensor Mesh:

This usage will extract a dataset from one TDR file and save it as a new TDR file. Several
options are given to scale and/or resample the data. The syntax for this usage and associated
options are:
tdrutil [options] <[Link]>[:<dataset>] <[Link]>

[:<dataset>] Sets the name of the dataset to extract. If no name is specified, the
-var=$
first applicable set will be extracted. The -var=* option will
extract all datasets.
-getdnk Extracts the n/k index perturbation from the carrier densities in the
input TDR file if present. The model used for the extraction is
documented in Section 8.I. Note that the options -lambda, -fca,
and -fca_file can be used to set the wavelength, fca_method,
and fca_file respectively. See Section 8.I for complete details.
-sc# Scales the dataset by a value.
-tgf# These options set the output tensor grid resolution:
-tgs#
-tg[x|y|z] • The -tgf# option sets the new tensor grid relative to the
original grid by a factor for all directions.
• The -tgs# option sets the new tensor grid in μm for all
directions.
• The -tgx#, -tgy#, and -tgz# options set the new tensor grid

RSoft CAD  304

 in μm for the specified direction.

For example, the command:

tdrutil -tfs0.01 [Link]:absorption [Link]

will extract the ‘absorption’ dataset from the file [Link], set the tensor grid to 0.01 μm
for all coordinates, and save the results in the new TDR file [Link].

Usage 3 - Extract Datasets to a Compatible Mesh:

This usage will extract a dataset from one TDR file on the mesh contained in another TDR file
and save the result as a new TDR file. The syntax for this usage is:
tdrutil [options] <[Link]> <[Link]> <[Link]>

For example, the command:

tdrutil [Link] [Link] [Link]

will extract the dataset from the file [Link], put it on the mesh contained in the file
[Link], and save it as the new TDR file [Link].

Usage 4 - Generate Absorbed Photon Density from RSoft Monitor:

This option generates the Absorbed Photon Density (APD) from an RSoft monitor and saves it as
a new TDR file. The APD is in units of 1/(cm3·s) defined as:

 P 
APD =  in   Abs 1012
 hc 
 
Where Abs is the absorption calculated by the simulator, and Pin is the input power in Watts.
Note that the Launch Power should be set to 1 and Launch Normalization should be set to Unit
Power.
The syntax for this usage and associated options are:
tdrutil -apd [options] <[Link]> <[Link]>

-lam# Sets the wavelength in μm.

-int# Sets the plane wave intensity in W/cm2. The default is 1.

-pow# Sets the guided wave power W.

-outvar=$ Sets the name of the output dataset. The default is

AbsorbedPhotonDensity.

RSoft CAD  305

For example, the command:
tdrutil -lam0.6 [Link] [Link]

will compute the absorbed photon density from the absorption profile in the file
[Link] at a wavelength of 0.6 μm and save it in the file [Link].

Usage 5 - Generate an RSoft CAD file with Dynamic Reference to a TDR

File:

Note that this process is automated when adding an RSoft step to an existing SWB project.
Simply Edit the RSoft CAD file for the RSoft step and you will be prompted to select a boundary
TDR file and the TDR Utility will create the RSoft CAD .ind file.

This usage performs the bootstrap process required to generate an RSoft .ind design file from a
boundary TDR file. This process is only required once when adding an RSoft simulator to a
Sentaurus Workbench workflow. The geometry of the structure will be automatically and
dynamically read from the TDR file. Note that it may be necessary to add additional CAD
objects to the design for measurement/output, etc. See the TCAD Tutorials in the MultiPhysics
Utility and TCAD Sentaurus Interface manual for examples of this usage.
The syntax for this usage is:
tdrutil [options] <[Link]> <[Link]>

For example, this command:

tdrutil myTDRFile_bnd.tdr [Link]

will create the file [Link] that is dynamically linked to the boundary TDR file
myTDRFile_bnd.tdr.

RSoft CAD  306

wgmode – Waveguide Mode Utility
This program calculates modal effective indices for slab waveguides, fibers, and diffused
structures. It is the command line analog of the slab_neff() symbol table function described in
Section C.F.8.
This is an interactive program; you must answer the question at each step. The defaults for each
question are shown in square brackets [], hit enter to use the default. Once all questions are
answered, the program will output the computed modal properties. The program will
automatically run again except the default values will be set to the most previous answers. This
allows the properties of similar cases to be easily iterated through. To stop running the utility,
type CTRL-C.

Example:
For example, the properties of the fundamental mode of a typical single mode fiber at λ = 1.55
μm can be found by running wgmode and answering the questions with 2, 0, 1, 1.55, 1.44, 1.45,
8.2 as shown here (the answers used for this example are highlighted below in bold):
Waveguide Structure [1=Slab,2=Fiber,3=Diffused] [ 1 ] : 2
Azimuthal Mode [m=0,...] [ 0 ] : 0
Radial Mode [n=1,...] [ 1 ] : 1
Wavelength [microns] [ 1.000000 ] : 1.55
Index of Clad (nclad) [ 1.000000 ] : 1.44
Index of Core (ncore) [ 1.010000 ] : 1.45
Diameter (d) [ 1.000000 ] : 8.2
The waveguide parameters for mode LP(0,1) are:
normalized frequency (V) : 2.8254066
normalized mode index (b) : 0.62111807
mode effective index (Neff) : 1.4462193
longitudinal propagation constant (Beta) : 5.8624929
transverse propagation constant (Kappa) : 0.42417888
transverse decay constant (Gamma) : 0.54310547 (1/G=1.841)

The output shown here shows, for example, that the computed mode effective index (Neff) is
1.4462193.

RSoft CAD  307

Appendix F

Color Scales

Color scale files provide a means to customize data presentation using RSoft products. All of the
simulation products that use the RSoft CAD are capable of producing large amounts of data.
Choosing which data to present and how to present it according to needs and taste is an
important issue. Scale files are located in the directory <rsoft_dir>\RLOTDIR. Several default
scale files are included, as well as several example files. You can also create and add your own
custom files.

F.A. What are Scale Files?

A scale file is a color lookup table which is used to graphically display 2D data sets. It has three
main components: a data range, a color lookup table, and a color shades value.

Figure 1: A simple fiber mode profile as seen in WinPLOT, along with the default scale file
used by BeamPROP to display field profiles.

RSoft CAD  308

• Data Range
Each color scale is defined for a specific range. For example, most default color scales in
BeamPROP have a range of [0,1], and FullWAVE a range of [-1,1]. The scale file used in
Figure 1 is defined over [0,1]. The scale is set by the highest and lowest values in the first
column; the rest of the numbers serve as documentation of the value range of each color.
A numerical value which lies outside the defined range of a scale file will be plotted with a
default color. For values which are greater than the defined range, White is used, and for
values which are less than the defined range, Black is used. The /zw WinPLOT command
can be used to automatically scale a color scale to the data limits; see the WinPLOT manual
for details.
• Color Lookup Table
A color scale is defined by a series of color steps. The color scale shown in Figure 1 is
defined by 6 colors, and assigns the color Magenta to the range 0.0-0.2, Blue to the range
0.2-0.4, and Red to the value 1.0. The maximum number of color steps is 101.
• Color Shades
Upon close examination of the colors shown in Fig. 1, it may seem that they do not agree
with the definition of the Color Lookup Table above, but, in fact, the program is blending the
colors together. This effect is achieved through the use of Color Shades. This setting is not
contained in the scale file, but can be set for each plot individually.
The Color Shades number sets the number of interpolated points between each defined
color. For example, setting this option equal to 4 results in 4 color points interpolated
between each defined color point. This option can be very useful for reducing display related
CPU usage during simulations, especially for slower systems. Fig. 2 shows the effect of
changing the color scale number.

Figure 2: Effect of the color shade number. a) Color shades = 1, b) 4, and c) 32.

F.B. Using Color Scales

There are three ways to change the scale file:

RSoft CAD  309

F.B.1. Setting the scale file during a simulation
You can change the scale file used to render contour plots shown in a simulation window by
opening the Simulation Parameters dialog, clicking the Display… button, clicking the Scale
File… button, and then choosing the desired scale file. To set the number of color shades used,
use the Color Shades field in the Display Properties box.

F.B.2. Changing the scale file after a simulation has completed

You can change the scale file used to display computed results in a completed simulation
window by choosing View/Display/Options/Color Scale from the pull-down menu, and then
selecting the desired scale file. You cannot change the number of color shades used after a
simulation has completed.

F.B.3. Changing the scale file in a WinPLOT window.

To change the scale file used to display data in a WinPLOT window, first open a saved plot. This
can be done via the View Graphs icon in the CAD interface or through your computers OS
(Windows, Linux, etc). Once the plot is opened, select Options/3D Data Display/Options/Color
Scale from the pull-down menu, and choose the desired .scl file. To make this change
permanent, you should save the file after making the change.
You can change the number of color shades used by selecting Options/3D Data
Display/Options/Color Shades.
Note: For advanced users, you can make these changes directly in the Edit Window. Use the
‘/cscl“FILE”’ command to se the color scale, and the ‘/cnum#’ command to set the color
shades. If you want more information about WinPLOT commands, please consult the WinPLOT
manual.

F.C. Scale File Format

Several color scale files are included in the RSoft distribution. The user can utilize these files for
data presentation, or create customized scale files. Scale files are text based files which can be
created/modified in any text editor, and should be saved in directory <rsoft_dir>\RPLOTDIR.
A scale file consists of several rows of number/color combinations. The number of rows
determines the number of color steps in the scale. The scale file format is:
Min Value Color
.
.
Max Value Color

The first column contains the range information, and the second column contains the color
lookup information. As stated before, the range information is only specified by the first and last

RSoft CAD  310

line, or min and max value. The other numbers only serve to document the color ranges. There
are two ways to specify the color information in a scale file:

F.C.1. Using Predefined Color Keywords

You can use one of several predefined keywords to specify a color in a scale file. The color
keywords currently recognized are shown in Fig. 3.

Figure 3: The predefined color keywords currently recognized with their corresponding RGB
values.

An example of this type of color scale is the BeamPROP default scale shown in Fig. 1.

F.C.2. Using RGB Values

If a color you wish to use is not defined be a color keyword, you can directly input RGB values
for the desired color. The values accepted range from 0 to 255. The first column is the R value,
the second the G value, and the third the B value.
Example: The scale file [Link], which is included in the distribution, is defined via RGB
values.

RSoft CAD  311

Figure 4: The color scale file [Link] which is defined through the use of RGB values.
This file can be found in the directory <rsoft_dir>/RPLOTDIR.

RSoft CAD  312

Appendix G

Polarization Conventions

While the concept of polarization gives rise to many of the most interesting effects in
electromagnetism, the labeling of polarizations is a common cause of confusion since there is no
universal agreement across different branches of photonics. This section describes the definitions
used by RSoft software.

G.A. Background
First, some background information.

1D Structures:
In 1D, that is for structures in which the refractive index varies along only one dimension,
Maxwell’s equations are always separable into two sets of three equations. In one set of
equations, the electric field is oriented along one of the invariant directions (i.e. the direction
along which the refractive index does not vary); in the other the magnetic field lies along the
invariant direction.
The simplest example of such a structure is the slab waveguide in which the field propagates
along the z direction say, and the refractive index n(x) is a function of only x. For this case, it is
universally agreed that TE polarization refers to the electric field lying along the invariant
direction y, and TM polarization refers to the magnetic field lying along y. Note that we could

RSoft CAD  313

equivalently refer to these states as y-polarization and x-polarization respectively, where we
label the states by the direction in which the electric field points.
Such problems are separable in all RSoft tools.

2D Structures:
In 2D structures, i.e. structures in which the refractive index varies as a function of two
dimensions, Maxwell’s equations are again separable, but only if the wave vector of the
propagating solution lies in the plane of variation. This means that for RSoft tools that model
propagation along a waveguide/fiber (BeamPROP, DiffractMOD, ModePROP, and FemSIM) the
equations are in fact not strictly separable. For the tools that permit propagation in the plane
(FullWAVE and BandSOLVE) the equations are separable. The literature conventionally uses
the labels TE to refer to states in which the electric field lies within the plane of variation, and
TM to refer to states in which the electric field lies perpendicular to the plane of variation. As we
see below, however, in the RSoft CAD these notations are reversed for consistency.

3D Structures:
In 3D structures, i.e. structures in which the refractive index varies in all three dimensions,
Maxwell’s equations are completely coupled and as a result, there is never a separation into TE
and TM states.

G.B. The RSoft Convention

Optical scientists and engineers apply the labels ‘TE’ and ‘TM’ to several different types of
electromagnetic modes. One class of modes is those that propagate along waveguides such as
optical fibers or on-chip waveguides. Another class of modes are the eigenmodes of infinite
structures such as the Bloch modes, which exist in photonic crystals. The usage of the terms TE
and TM can be quite different between the various types of modes and a completely natural,
uniform usage across all problems is both almost impossible to achieve and would be
inconsistent with the most common usages in the literature.
Additionally, another possible source of confusion is the use of the terms 2D and 3D. In
documenting the RSoft CAD environment and simulation tools, we count the propagation
direction as a dimension, so a 2D structure is one whose cross-sectional profile varies as a
function of one variable, usually x, while a 3D structure is one whose cross-sectional profile
varies as a function of two variables, usually x and y. If we wish to indicate that a structure varies
as a function of all three variables, we normally use the term 3D structure also. This convention
occasionally gives rise to some confusing combinations of terminology. For example, a CAD
representation of photonic crystal fiber along the z direction is regarded as a 3D structure.
Propagating a field down the fiber in BeamPROP is performed in BeamPROP’s 3D simulation
mode, but the problem of finding the same fiber’s modes in BandSOLVE would be solved in

RSoft CAD  314

BandSOLVE’s 2D-XY mode. While not ideal, such variations in terminology are inevitable given
the large number of tools that share the RSoft CAD. In most cases, there is in practice little
confusion.
Now let us examine how polarization is labeled in RSoft tools.

2D CAD Designs:
The most basic structures that can be drawn in the RSoft CAD are waveguides that vary as a
function of only x. The classic slab waveguide is the simplest example of this class. As there is
no variation in the y-direction, the previous discussion indicates that these waveguides fall into
the category of 2D CAD designs. For such structures, it is familiar from elementary waveguide
optics that in a TE mode, the electric field points along y and in a TM mode, the magnetic field
points along y. In this case the interpretation of ‘TE’ is the literal meaning: the electric field is
transverse to the direction of propagation (the wave vector k,) and implicitly the magnetic field is
not transverse. Similarly, ‘TM’ literally means that the magnetic field is transverse to the
direction of propagation and the electric field is not.
For 2D CAD designs, this is the precise definition used in all RSoft tools except BandSOLVE:
‘TE’ means that the electric field lies along Y; ‘TM’ means that the magnetic field lies along Y.
(The 2D rule for BandSOLVE is discussed below.) Observe that we could also label these states
by the direction in which the electric field points. That is, TE polarization could equivalently be
called y-polarization and TM polarization called x-polarization. This is important when we turn
to 3D problems.

3D CAD Designs:
Now consider 3D CAD designs: those for which the waveguides vary in two transverse
dimensions. For modal propagation along these waveguides, Maxwell’s equations are not
separable and the literal meaning of TE and TM modes cannot be applied—virtually all the
modes are hybrid and neither field is transverse to the direction of propagation. This point is
crucial so let us restate it more forcefully:

For all 3D dielectric waveguides, there are no modes that are truly TE or TM.

The only counter-examples to this statement are the TE0m and TM0m modes of azimuthally-
invariant waveguides such as the step-index fiber or concentric Bragg (“Omniguide”) fibers, and
even then, most of the observed modes, including the fundamental mode, are hybrid LP-like
states.
Nevertheless, in weakly guided structures (and even in quite strongly guided structures,) the
modes can be close to transverse and the terms TE and TM are widely applied to modes of such
structures. We therefore need to be precise about the way the labels are to be applied. The
argument runs as follows:

RSoft CAD  315

Typically, the longitudinal component of one field is much larger than the longitudinal
component of the other. Thus it can be useful to think of the modes as ‘quasi-TE’ and ‘quasi-
TM’. Moreover, in many commonly encountered waveguides such as ribs, channels or diffused
guides, the cross section varies more rapidly in one direction than the other, and as a result the
strengths of the transverse components differ markedly and can be designated as major and
minor components. So consider a rib waveguide with propagation along Z and where the layers
of the guide are oriented along X:

This is the classic rib waveguide that is easily produced in the RSoft CAD. Then for a ‘quasi-TE’
mode, the longitudinal field Ez is small, and the major field Ex is much larger than the minor field
Ey. For a ‘quasi-TM’ mode, the longitudinal field Ez is still small, but the major field is now Ey
and the minor field is Ex. So in 3D simulations, the labels TE and TM should strictly be
interpreted as quasi-TE and quasi-TM. This provides the explanation for one of the more
confusing aspects of the RSoft CAD environment: for 3D structures, TE means the electric field
is oriented predominantly along x, and TM means the field is oriented predominantly along y.
While this appears to be in conflict with the 2D definitions, for which TE means the electric field
points along Y, the rules are natural in the sense that in each case, TE refers to the polarization
for which the electric field predominantly points along the transverse direction in which the
index profile is most slowly varying. This is ‘natural’ only because most standard layered 3D
waveguides drawn in the RSoft CAD tend to vary slowly along x. If for perverse reasons, you
manage to construct a rib waveguide in which the layers are oriented along the y direction, then
to find the mode that would conventionally be labeled quasi-TE, you will need to use the TM
polarization setting.

3D CAD Designs Simulated via EIM:

When using the Effective Index Method, use the same conventions as for a 3D structure. The
software will use the appropriate polarization when simulating the structure in 2D.

RSoft CAD  316

G.C. Simple rules to remember
The above discussion provides a detailed motivation for the behavior of the CAD with respect to
polarization but is a lot to remember. For day-to-day work the following simple rules are easier:

2D Designs:
In 2D designs, TE means that the electric field lies strictly along Y; TM means that the magnetic
field lies strictly along Y. In terms of the electric field, we can also label these as ‘y-polarization’
and ‘x-polarization’ respectively.

3D Designs:
In 3D designs there are (virtually) no true transverse modes, so we interpret the TE and TM
buttons as meaning quasi-TE and quasi-TM for structures that have a natural orientation within
the CAD. Their operational meaning is simply:
• For TE, the electric field is launched predominantly along x. This can be equivalently labeled
as x-polarization.
• For TM, the electric field is launched predominantly along y. This can be equivalently
labeled as y-polarization.
Remember also that in FullWAVE and full-vector BeamPROP, the field components are all
coupled so that for certain polarization-rotating structures, the fields will not necessarily
maintain these orientations during propagation.

G.D. Complications
There are some variations on these rules for certain tools:

BeamPROP:
For BeamPROP, please note that the handling of polarization is controlled by the Vector Mode
setting. Please see the BeamPROP manual for more details.

3D FullWAVE:
In 3D FullWAVE, the definition of the launch polarization is complicated by the ability to rotate
the propagation direction of the incoming field to any angle. In this case, the rules stated above
are to be interpreted within the rotated launch frame.

RSoft CAD  317

BandSOLVE:
The polarization rules in BandSOLVE are quite different. The details are explained fully in the
BandSOLVE manual but the pertinent facts are as follows:
• In the subset of 2D problems for which the Bloch vector lies in the plane of index variation,
TE and TM are strictly interpreted in the sense of transverse fields. TE polarization means
that the electric field is perpendicular to the Bloch vector (and to the plane of index
variation). TM polarization means that the magnetic field is perpendicular to the Bloch
vector. For standard 2D photonic crystal layouts in the XZ plane, this definition is consistent
with the other tools in that for TE, the electric field lies along Y. However, the definition is
the opposite of the normal usage in the photonic crystal literature.
• In all other cases, the only allowed polarization state is hybrid. However, a weaker form of
separation into even and odd parity states is allowed for certain structures with symmetry
planes. Please refer to the BandSOLVE manual for further details.

3D DiffractMOD:
In 3D DiffractMOD, the definition of the launch polarization is complicated by the ability to
rotate the propagation direction of the incoming field to any angle as well as the definition of the
primary direction. The rules stated above are to be interpreted within the rotated frame. In
addition for 3D DiffractMOD computations, the polarization is defined by value of the
Polarization Angle field. An angle of 0 corresponds to p polarization while 90 corresponds to s
polarization. For normal incidence and a primary direction along +Z, this implies that with
Polarization Angle=0 the electric field lies along X, while for Polarization Angle=90 the
electric field lies along Y. Please refer to the DiffractMOD manual for further details.

RSoft CAD  318

Appendix H

Python Usage

The RSoft tools can utilize Python in various contexts, including:

• Via the rspython utility:
• Manually running rspython (see Appendix E)
• Using a .py file within Circuit References (see Section 6.F)
• Using a Python-based MOST User Simulator (see MOST manual)
• Using a Python-based MOST User Metrics (see MOST manual)
• Using a Python-based MOST User Optimization Algorithms (see MOST manual)
This appendix documents supported Python versions, as well as several Python classes for layout
scripting (RSoftCircuit() and supporting classes) as well as data manipulation
(RSoftUserfunction() class). See Chapter 10 for an overview of scripting, including several
examples.

H.A. Python Version Support

The version of Python supported is 3.11 on both Windows and Linux.

RSoft CAD  319

Please note that as of the 2024.09 release, Python 2 support has been removed in both the API’s
described in this appendix, and in MOST. Refer to the MOST documentation for additional
information. Users with existing Python 2 scripts or MOST metrics, etc. will need to update them
to Python 3. In general, this requires changes in two areas. The first is directly related to Python
2/3 differences, the most common one being the print command, which are documented
publicly on various websites. The second is related to the required switch from numarray to
numpy and is described in the next section.

In addition to rspython, note that it is possible to use the API’s described in subsequent sections
with compatible distributions of Python from another source.
Please contact photonics_support@[Link] with questions.

Note: Numeric Array Support

The transition from using Python 2 to Python 3 in rspython is in part related to the language
differences themselves as noted above, but is also impacted by the use of different numeric array
packages. The numeric array package previously supported by rspython in Python 2.x was
numarray, while Python version 3.x supports the newer numpy. In many cases, the user does not
need to worry about this, as the RSoftCircuit() and RSoftUserFunction() classes are either
agnostic to the choice or produce the corresponding arrays which are used in generally the same
manner (e.g. when indexing). However, when the user is creating an array as an input to a
member of these classes, numpy arrays must be used.
Please contact photonics_support@[Link] with questions.

H.B. Layout API: RSoftCircuit() class

There Python API can be used to script the layout of a .ind design file. It contains the main
RSoftCircuit() class as well as the RSoftComponent() and RSoftVertex() classes. It is
recommended that you use the API in conjunction with the circuit reference feature described in
Section 6.F. See Chapter 10 for an overview of scripting, including several examples.

H.B.1. RSoftCircuit() class

Create a new RSoftCircuit object using this code:
from rstools import RSoftCircuit
c = RSoftCircuit()

The following class functions are supported:

read(filename)

Reads an existing .ind design file (string filename), for example:

RSoft CAD  320

 [Link](‘[Link]’)

write(filename)

Writes an .ind design file to the specified filename (string), for example:
[Link](‘[Link]’)

set_symbol(name, value)

Set a symbol with the specified name and value. The name is a string, the value can be a
number or a standard string expression that can use the built-in functions and/or other
symbols. Examples:
c.set_symbol(‘Period’, 0.5)
c.set_symbol(‘Radius’, ‘0.35*Period’)

delete_symbol(name)

Deletes a symbol with specified name (string), for example:

c.delete_symbol(‘Period’)

load_settings(file)

Load settings from specified file (string). Equivalent to the Load Settings option in the
Global Settings dialog (see Section 3.B). Example:
c.load_settings(‘[Link]’)

set_exports(exports)

Set hierarchy exports; exports is a space-separated list of symbols to include. This

feature enables the efficient passing of symbols from the parent to child files in cases
where the symbol does not need to be an argument. See Section 6.F.5 for details about
hierarchy exports. Example:
c.set_exports(‘Period Angle’)

set_default tags(tags)

Set default tags; tags is a space-separated list of symbols to include. See Section 6.K.5
for details about default tags. Example:
c.set_default_tags(‘grid_ppw_x=35 grid_size_y=0.001’)

add_material( matname=’’ , libname=’’, groupname=’’, datafile=’’, nr=’’,

ni=’’)

Adds a material to the design file from one of several sources, depending on the
arguments.

Parameter Description

RSoft CAD  321

 matname The material name
libname The library name (optional)
groupname The group name (optional)
datafile Set the material properties; set either the name of an nk
nr
datafile or the real (nr) and optional imaginary (ni)
ni
values. If neither datafile nor nr/ni are present, the
material is taken from the Material Library.

add_monitor( position=(), dimensions=() )

Adds a monitor to the design file.

Parameter Description
position A tuple that sets the starting (X,Y,Z) position of the
segment. This is the starting vertex. The values can be
numbers or standard expression strings.
dimensions A tuple that sets the dimensions (width/height) of the
starting vertex. If not set, a default value of
(‘width’,’height’) is used. The values can be
numbers or standard expression strings.

add_portmonitor( position=(), dimensions=() )

Adds a port monitor to the design file.

Parameter Description
position A tuple that sets the starting (X,Y,Z) position of the
segment. This is the starting vertex. The values can be
numbers or standard expression strings.
dimensions A tuple that sets the dimensions (width/height) of the
starting vertex. If not set, a default value of
(‘width’,’height’) is used. The values can be
numbers or standard expression strings.

add_segment( position=(), offset=(), dimensions=(), dimensions_end=() )

Adds a segment (see Chapter 4) to the design with the following properties:

Parameter Description
position A tuple that sets the starting (X,Y,Z) position of the

RSoft CAD  322

 segment. This is the starting vertex. The values can be
numbers or standard expression strings.
offset A tuple that sets the (X,Y,Z) offset of the ending vertex.
This is the offset between the starting vertex and ending
vertex. The values can be numbers or standard
expression strings.
dimensions A tuple that sets the dimensions (width/height) of the
starting vertex. If not set, a default value of
(‘width’,’height’) is used. The values can be
numbers or standard expression strings.
dimensions_end A tuple that sets the dimensions (width/height) of the
ending vertex. If not set, the default value is equal to
dimensions. The values can be numbers or standard
expression strings.

This function returns a RSoftComponent() object which can be further configured, see
Section H.B.2.
Examples:
seg1 = c.add_segment(position=(0,0,0),offset=(0,0,‘L’))
c.add_segment(position=(0,0,0),offset=(0,0,‘L’),dimensions=(0.5,0.2)

add_sbend( position=(), offset=(), dimensions=(), dimensions_end=()

s_bend_type=0 )

Adds an s-bend segment (see Section 6.B.3) to the design with the following properties:

Parameter Description
position A tuple that sets the starting (X,Y,Z) position of the
sbend. This is the starting vertex. The values can be
numbers or standard expression strings.
offset A tuple that sets the (X,Y,Z) offset of the ending vertex.
This is the offset between the starting vertex and ending
vertex. The values can be numbers or standard
expression strings.
dimensions A tuple that sets the dimensions (width/height) of the
starting vertex. If not set, a default value of
(‘width’,’height’) is used. The values can be
numbers or standard expression strings.
dimensions_end A tuple that sets the dimensions (width/height) of the
ending vertex. If not set, the default value is equal to

RSoft CAD  323

 dimensions. The values can be numbers or standard
expression strings.
sbend_type An integer that sets the sbend type. A value of 0
(default) is Circular-Arc, 1 is Cosine, and 2 is Raised-
Sine.

This function returns a RSoftComponent() object which can be further configured, see
Section H.B.2.
Examples:
sbend1 = c.add_sbend(position=(0,0,0),offset=(0,0,‘L’),sbend_type=2)
c.add_sbend(position=(0,0,0),offset=(0,0,‘L’),dimensions=(0.5,0.2)

add_arc( position=(), arcinfo=(), dimensions=(), dimensions_end=() )

Adds an arc segment (see Section 6.B.3) to the design with the following properties:

Parameter Description
position A tuple that sets the starting (X,Y,Z) position of the arc.
This is the starting vertex. The values can be numbers or
standard expression strings.
arcinfo A tuple that sets the radius, initial angle, and final angle
as (radius, angle-initial, angle-final). The
values can be numbers or standard expression strings;
angles should be in degrees.
dimensions A tuple that sets the dimensions (width/height) of the
starting vertex. If not set, a default value of
(‘width’,’height’) is used. The values can be
numbers or standard expression strings.
dimensions_end A tuple that sets the dimensions (width/height) of the
ending vertex. If not set, the default value is equal to
dimensions. The values can be numbers or standard
expression strings.

This function returns a RSoftComponent() object which can be further configured, see
Section H.B.2.
Examples:
arc1 = c.add_arc(position=(0,0,0),arcinfo=(‘R’,0,90))
c.add_arc(position=(0,0,0),arcinfo=(‘R’,0,45),dimensions=(0.5,0.2)

add_lens( position=(), dimensions=(), shape=0 )

RSoft CAD  324

 Adds a lens (see Section 5.B) to the design with the following properties:

Parameter Description
position A tuple that sets the (X,Y,Z) position of the lens. The
values can be numbers or standard expression strings.
dimensions A tuple that sets the dimensions (width/height) of the
lens. If not set, a default value of (‘width’,’height’)
is used. The values can be numbers or standard
expression strings.
shape An integer that sets the lens shape: a value of 0 (default)
correspond to a cylinder, a value of 1 corresponds to a
sphere.

This function returns a RSoftComponent() object which can be further configured, see
Section H.B.2.
Examples:
sphere1 = c.add_lens(position=(0,0,0), dimensions=(‘D’,‘D’) shape=1)

add_circle( position=(), dimensions=(), shape=0 )

Adds a circle (see Section 5.C) to the design with the following properties:

Parameter Description
position A tuple that sets the (X,Y,Z) position of the circle. The
values can be numbers or standard expression strings.
dimensions A tuple that sets the dimensions (width/height) of the
circle. If not set, a default value of
(‘width’,’height’) is used. The values can be
numbers or standard expression strings.
shape An integer that sets the circle shape: a value of 0
(default) correspond to a cylinder, a value of 1
corresponds to a sphere.

This function returns a RSoftComponent() object which can be further configured, see
Section H.B.2.
Examples:
circle1 = c.add_circle(position=(0,0,0), dimensions=(‘D’,‘D’)
shape=1)

add_polygon( position=(), x=None, y=None, file=’’, angle=0 )

RSoft CAD  325

 Adds a polygon (see Section 5.D) to the design with the following properties:

Parameter Description
position A tuple that sets the (X,Y,Z) position of the polygon.
The values can be numbers or standard expression
strings.
X Defines the points of the polygon via x/y arrays or a data
y
file. The x/y arrays should be 1D arrays of the same
file
length.
angle Sets the rotation angle for the polygon in degrees.

This function returns a RSoftComponent() object which can be further configured, see
Section H.B.2.
Examples:
polygon1 = c.add_polygon(position=(0,0,0), file=’[Link]’,
angle=45)

add_mark( position=(), dimensions=() )

Adds a mark (See Section 5.E) to the design with the following properties:

Parameter Description
position A tuple that sets the (X,Y,Z) position of the mark. The
values can be numbers or standard expression strings.
dimensions A tuple that sets the dimensions (width/height) of the
mark. If not set, a default value of
(‘width’,’height’) is used. The values can be
numbers or standard expression strings.

This function returns a RSoftComponent() object which can be further configured, see
Section H.B.2.
Examples:
mark1 = c.add_mark(position=(0,0,0))

add_textblock( name, text=’’ )

Adds a text block to the design with the following properties:

Parameter Description
name The name of the text block.

RSoft CAD  326

 text The body of the text block.

This is an advanced option; contact technical support with questions.

add_vertex( position=(), dimensions=() )

Adds a vertex (See Section 6.F.7) to the design with the following properties:

Parameter Description
position A tuple that sets the (X,Y,Z) position of the vertex. The
values can be numbers or standard expression strings.
dimensions A tuple that sets the dimensions (width/height) of the
vertex. If not set, a default value of
(‘width’,’height’) is used. The values can be
numbers or standard expression strings.

This function returns a RSoftComponent() object which can be further configured, see
Section H.B.2.
Examples:
vertex1 = c.add_vertex(position=(0,0,0))

add_circuitref( position=(), file=’’, angle=0 )

Adds a circuit reference (See Section 6.F) to the design with the following properties:

Parameter Description
position A tuple that sets the (X,Y,Z) position of the circuit
reference. The values can be numbers or standard
expression strings.
file Sets the file to be used for the circuit reference.
angle Sets the rotation angle for the polygon in degrees.

This function returns a RSoftComponent() object which can be further configured, see
Section H.B.2.
Examples:
ref1 = c.add_circuitref(position=(0,0,0))

add_polygon( position=(), x=None, y=None, file=’’, angle=0 )

Adds a circuit reference (See Section 6.F) to the design with the following properties:

RSoft CAD  327

 Parameter Description
position A tuple that sets the (X,Y,Z) position of the circuit
reference. The values can be numbers or standard
expression strings.
X Double 1D numeric array (same length as y)
y Double 1D numeric array (same length as x)
file Sets the file to be used for the circuit reference.
angle Sets the rotation angle for the polygon in degrees.

This function returns a RSoftComponent() object which can be further configured, see
Section H.B.2.

attach( from_component, to_component, to_vertex, from_vertex,

attach_angles=0, attach_dimensions=0 )

This function enables users to connect components/vertices using the reference options
described in Section 4.B.1.

Parameter Description
from_component The RSoftComponent() object to attach to.
to_component The RSoftComponent() object to attach.
to_vertex Sets the vertex number to attach to.
from_vertex Sets the vertex number to attach.
attach_angles Set to 1 to dynamically pass the segment angle.
attach_dimentions Set to 1 to dynamically pass the segment dimensions.

Example:
seg1 = c.add_segment()
seg2 = c.add_segment()
[Link](seg2,seg1)

H.B.2. RSoftComponent() class

An RSoftComponent is created when adding components to an RSoftCircuit() object. The
following class functions allow various component properties to be set:

attach_portmonitors()

Adds port monitors to unterminated vertices of an RSoftComponent() object.

RSoft CAD  328

 For example:
cr = c.add_circuitreference(position=(0,0,0), file=’[Link]/0’)
cr.attach_portmonitors()

attach_portvertices()

Automatically detect vertex positions and add vertices.

For example:
cr = c.add_circuitreference(position=(0,0,0), file=’[Link]/0’)
cr.attach_portvertices()
cr.attach_portmonitors()

color()

Set the component color.

For example:
circle1 = c.add_circle(position=(0,0,0), shape=1)
[Link]( ‘4’ )

dimensions( dimensions=() )

Sets the dimensions of the component vertex, specifically a tuple that sets the dimensions
(width/height). If the component has multiple vertices, this option sets the starting vertex
dimensions. See vertex() function to set properties of other vertices. The values can be
numbers or standard expression strings.
For example:
circle1 = c.add_circle(position=(0,0,0), shape=1)
[Link]( (‘D’,’D’) )

draw_priority()

Set the draw priority.

For example:
circle1 = c.add_circle(position=(0,0,0), shape=1)
circle1.draw_priority( 4 )

mask_layer()

Set the mask layer.

For example:
circle1 = c.add_circle(position=(0,0,0), shape=1)
circle1.mask_layer( 88 )

material()

RSoft CAD  329

 Sets the material of the component via a string matname. Note that this option is designed
to be used within a circuit reference. This option only assigns the material to the
component, the .ind file generated must then be used as a circuit reference in a parent
file where the material properties are defined in that parent file.
For example:
seg1 = c.add_segment( position=(0,0,0) )
[Link](‘SiO2’)

orientation()

Sets the segment orientation of the component. Set to 0 for Z Axis, for Seg Axis, and 2
for Extended.
For example:
seg1 = c.add_segment( position=(0,0,0) )
[Link](0)

polyinfo(x=None, y=None, file=’’, sides=0)

Sets the polygon information for polygon segments. The x and y arguments are double
1D numeric arrays of the same length.
For example:
seg1 = c.add_segment( position=(0,0,0) )
[Link](‘POLYGON’)
[Link](sides=8)

position( position=(), ref_type=-1 )

Sets the position of the component vertex, specifically a tuple that sets the (X,Y,Z)
coordinates. If the component has multiple vertices, this option sets the starting vertex
dimensions. See vertex() function to set properties of other vertices. The optional
ref_type option sets whether the position is absolute (0) or relative (-1). The values can
be numbers or standard expression strings.
For example:
circle1 = c.add_circle(shape=1)
[Link]( (1,1,1) )

priority()

Set the priority.

For example:
circle1 = c.add_circle(position=(0,0,0), shape=1)
[Link]( 4 )

RSoft CAD  330

rotation( angles=() )

Sets the rotation angles of the component in degrees, specially a tuple with the three
rotation angles phi, theta, and psi that are used to rotate extended segments (see Section
6.K.2). The values can be numbers or standard expression strings.
For example:
seg1 = c.add_segment( position=(0,0,0) )
[Link]( (0,90,0) )

set_name()

Sets the component name of the component. The argument is a string.

For example:
seg1 = c.add_segment( position=(0,0,0) )
seg1.set_name( “MySegment” )

set_property(name=’’, value=’’, sub=’’)

An advanced generic function to set any component property whose keyword in a .ind
file is known. The optional sub argument sets a nested property. Note: you must use a
valid property name; setting an invalid property name will result in a parse error when the
design file is used.
For example:
seg1 = c.add_segment( position=(0,0,0) )
seg1.set_property(‘sidewall_angle’, 80)
set1.set_property(‘combine_mode’, ‘COMBINE_ADD’)

mon3 = c.add_monitor( position=(0,0,0) )

mon3.set_property(‘monitoroutputmask’, 8)

set_symbol(name, value)

Set a symbol with the specified name and value in the local symbols of a circuit
reference component. The name is a string, the value can be a number or a standard
string expression that can use the built-in functions and/or other symbols. Examples:
c.set_symbol(‘Period’, 0.5)
c.set_symbol(‘L’, ‘2*width’)

delete_symbol(name)

Deletes a symbol with specified name (string) from the local symbols of a circuit
reference component, for example:
c.delete_symbol(‘Period’)

set_tags()

RSoft CAD  331

 Sets a component tag. The argument is a string of form “name=value”.
For example:
seg1 = c.add_segment( position=(0,0,0) )
seg1.set_tags( “tag=value” )

set_comment() )

Sets the component comment. The argument is a string.

For example:
seg1 = c.add_segment( position=(0,0,0) )
seg1.set_comment( “This is a comment!” )

structure()

Set the 3D Structure Type. Options include ‘CHANNEL’, ‘FIBER’, ‘POLYGON’,

‘MULTILAYER’, ‘RIBRIDGE’, and ‘DIFFUSED’.

For example:
seg1 = c.add_segment( position=(0,0,0) )
[Link]( ‘FIBER’ )

vertex( vindex=0 )

Returns an RSoftVertex() object for the requested vertex index vindex. See Section
H.B.3 for details on this type of object and how to set the dimensions and position of the
vertex.
For example:
seg1 = c.add_segment( position=(0,0,0) )
vertex = [Link](1)

H.B.3. RSoftVertex() class

An RSoftVertex object can be extracted from an RSoftComponent() object, allowing additional
control over these vertices. The following class functions allow various vertex properties to be
set:

dimensions( dimensions=() )

Sets the dimensions of the vertex, specifically a tuple that sets the dimensions
(width/height). The values can be numbers or standard expression strings.
For example:
seg = c.add_segment(position=(0,0,0))
vertex2 = [Link](1)
[Link]( (‘D’,’D’) )

RSoft CAD  332

position( position=(), ref_type=-1 )

Sets the position of the component vertex, specifically a tuple that sets the (X,Y,Z)
coordinates. The optional ref_type option sets whether the position is absolute (0) or
relative (-1). The values can be numbers or standard expression strings.
For example:
seg = c.add_segment()
vertex2 = [Link](1)
[Link]( (0,0,0) )

H.B.4. RSoftMaterial() class

An RSoftMaterial object contains the definition of a material. The following class functions
allow various material properties to be set:

set_property(name=’’, value=’’, sub=’’)

An advanced generic function to set any material property whose keyword in a .ind file
is known. The optional sub argument sets a nested property. Note: you must use a valid
property name; setting an invalid property name will result in a parse error when the
design file is used.
For example:
mat = c.add_material( matname=’MyMat’)
mat.set_property(‘anisotropic’,1,’optical’)
mat.set_property(‘nr_xx’,1.5,’optical’)
mat.set_property(‘nr_yy’,1.6,’optical’)
mat.set_property(‘nr_zz’,1.5,’optical’)

H.B.5. Using Hierarchy Exports

In some cases, it is not necessary to pass all parameters to the Python script as arguments,
hierarchy exports can also be used. Use the set_exports() function, see Section 10.A.1 for an
example.

H.B.6. Best Practices

It is recommended to follow these best practices when using the RSoftCircuit() class:
• See Section 10.A for several examples that illustrate possible design approaches
• When used within a circuit reference block, use with the rsoft_layout() function described
in Section 6.F.6. This simplifies the passing of arguments to the script. Any output will be
written to a log file to assist in debugging.
• Use Hierarchy Exports, see Section H.B.5.

RSoft CAD  333

• A material can be set via the see RSoftComponent() class, see details the material()
function in Section H.B.2.

H.B.7. Example Usage

Here are several simple example scripts to illustrate the usage of the RSoftCircuit() class:

See examples in Section 10.A for more examples

Example 1:
This example creates an 8x8 array of segments in the XY plane and saves as [Link]:
from rstools import RSoftCircuit
#Constants
N = 8
file = ‘[Link]’
#Create new RSoftCircuit object
c = RSoftCircuit ()
#Set some symbols
c.set_symbol(Period, 0.5)
c.set_symbol(Radius, ‘Period*0.3’)
c.set_symbol(D, ‘2*Radius’)
c.set_symbol(Length, 10)
for i in range(N):
for j in range(N):
x0 = (‘{}*Period’.format(i))
y0 = (‘{}*Period’.format(j))
z0 = 0
c.add_segment( (x0,y0,z0), (0,0,’Length’), (‘D’,’D’), (‘D,‘D’) )
#Save .ind file
[Link](file)

Example 2:
This example creates an NxN array of segments in the XY plane. Note that this example must be
run via a circuit reference, if set, the Local Symbol Arg1_num will be used to set the array size
(N).
from rstools import RSoftCircuit
def rsoft_main(ind_file=’’,
N=int(10)):
#Create new RSoftCircuit object
c = RSoftCircuit ()

RSoft CAD  334

 #Set some symbols
c.set_symbol(Period, 0.5)
c.set_symbol(Radius, ‘Period*0.3’)
c.set_symbol(D, ‘2*Radius’)
c.set_symbol(Length, 10)
for i in range(N):
for j in range(N):
x0 = (‘{}*Period’.format(i))
y0 = (‘{}*Period’.format(j))
z0 = 0
c.add_segment( (x0,y0,z0), (0,0,’Length’), (‘D’,’D’), (‘D,‘D’) )
#Save .ind file
[Link](ind_file)

Example 3:
This example implements the same structure in Example 2, but sets the Period, Radius, and
Length as Hierarchy Exports which can then easily be set in the Local Symbols table of the
circuit reference.
from rstools import RSoftCircuit
def rsoft_main(ind_file=’’,
N=int(10)):
#Create new RSoftCircuit object
c = RSoftCircuit ()
#Set some symbols
c.set_symbol(Period, 0.5)
c.set_symbol(Radius, ‘Period*0.3’)
c.set_symbol(D, ‘2*Radius’)
c.set_symbol(Length, 10)
#Set Hierarchy Exports
c.set_exports(‘Period’, ‘Radius’, ‘Length’)
for i in range(N):
for j in range(N):
x0 = (‘{}*Period’.format(i))
y0 = (‘{}*Period’.format(j))
z0 = 0
c.add_segment( (x0,y0,z0), (0,0,’Length’), (‘D’,’D’), (‘D,‘D’) )
#Save .ind file
[Link](ind_file)

Example 4:
See This example creates a tapered waveguide, including port monitors.
from rstools import RSoftCircuit

# inputs

RSoft CAD  335

name = 'wg_taper'

# structure properties
opts = {}
opts['Lin'] = 0.5
opts['Ltaper'] = 2
opts['W1'] = 0.5
opts['W2'] = 1.0
opts[‘H’] = ‘height’

# simulation settings
opts['Dx'] = 0.02
opts['Dy'] = opts['Dx']
opts['Dz'] = opts['Dx']
opts['boundary_gap_x'] = 0.5
opts['boundary_gap_z'] = 0
opts['color_outline'] = -1
opts['domain_max'] = 'Lin*2+Ltaper'
opts['bpm_output_monitors'] = 1
opts['bpm_output_monitors_warned'] = 1
opts['free_space_wavelength'] = 1.55
opts['sim_tool'] = 'ST_FULLWAVE'
opts['grid_nonuniform'] = 1
opts['grid_size'] = 'Dx'
opts['grid_size_y'] = 'Dy'
opts['step_size'] = 'Dz'
opts['eim'] = 1
opts['launch_port'] = 1
opts['polarization'] = 0
opts['plot_aspectratio'] = 1

# create design file, load settings, add symbols

c = RSoftCircuit()
c.load_settings('[Link]')
for key in opts:
c.set_symbol(key, opts[key])

# add/attach components
wg_in = c.add_segment(offset=(0,0,'Lin'), dimensions=('W1', 'H'))
wg_taper = c.add_segment(offset=(0,0,'Ltaper'), dimensions=('W1',
'H'), dimensions_end=('W2', 'H'))
wg_out = c.add_segment(offset=(0,0,'Lin'), dimensions=('W2', 'H'))

[Link](wg_taper, wg_in, 1,0,0)

[Link](wg_out, wg_taper, 1,0,0)

# add/attach port positions

port1 = c.add_portmonitor(dimensions=('W1', 'height'))
port2 = c.add_portmonitor(dimensions=('W2', 'height'))

[Link](port1, wg_in, 0,0,1)

RSoft CAD  336

 [Link](port2, wg_out, 1,0,1)

[Link]('%[Link]'%name)

H.C. RSoft Data API: RSoftUserFunction()

The RSoftUserFunction()class can be used to pragmatically interact with RSoft data files. All
standard RSoft file formats described in Appendix B are supported, including 1D, 2D, 3D, real,
complex, uniform, and non-uniform data formats. Note that files in the TCAD Sentaurus TDR
format are not currently supported. See Chapter 10 for an overview of scripting, including
examples.

H.C.1. Reference
Create a new RSoftUserFunction object using this code:
from rstools import RSoftUserFunction
uf = RSoftUserFunction()

The following class functions are supported:

read(filename)

Reads a data file (string filename) into the object, for example:
[Link](‘[Link]’)
write(filename)

Writes the object data to the specified filename, for example:

[Link](‘[Link]’)
type()

Returns the data type in the object, for example:

t = [Link]()

The types are in the data header of the file, they are 1 for amplitude, 2 for real, 3 for
imaginary, 4 for real/imag, and 5 for amp/phase.
isreal()

Returns 1 if the data in the object is real (i.e. not complex), for example:
if ([Link]()):
print(‘Data is real’)
iscomplex()

Returns 1 if the data in the object is complex-valued, for example:

if ([Link]()):
print(‘Data is complex’)

RSoft CAD  337

get_arrays()

Returns a tuple of the arrays within the object, returns multiple arrays depending on rank
of the data object. For example:
[Link](‘[Link]’)
(x,y,f) = uf.get_arrays()

where [Link] is a 2D data file and x, y, and f will contain the x, y, and data arrays
from the file.

set_arrays(x,f)
set_arrays(x,y,f)
set_arrays(x,y,z,f)

Sets the arrays in the object, for example:

x = array([1,2,3,4.5])
f = array([1,4,9,16,25])
uf.set_arrays(x,f)

eval(x)
eval(x,y)
eval(x,y,z)

Returns the real value of the function at the given x, y, and/or z coordinate value. The
data in the object will be linearly interpolated, similar to the userdata() functions
described in Section C.F.6. Note: this function returns 0 if the x, y, or z coordinates
outside the range defined in the data object. Example:

value = [Link](1.5)

eval_complex(x)
eval_complex(x,y)
eval_complex(x,y,z)

Returns the complex value of the function at the given x, y, and/or z coordinate value.
The data in the object will be linearly interpolated, similar to the userdata() functions
described in Section C.F.6. Note: this function returns 0 if the x, y, or z coordinates
outside the range defined in the data object. Example:

value = [Link](1.5) #value is a complex number, i.e. (1+7j)

get(i)
get(i,j)
get(i,j,k)

Returns the value of the data at the given i, j, and/or k position in the function array, for
example:
value = [Link](10) #value is the 10th point in the function

RSoft CAD  338

rank()

Returns the rank (dimension) of the data in the object, for example:
if ([Link]() == 2):
#do something if the data is 2D

nx()
ny()
nz()

Returns the number of X, Y, or Z points in the object, for example:

NX = [Link]()

x(i)
y(i)
z(i)

Returns the X, Y, or Z coordinate for the ith position in the coordinate array, for example
x_coordinate = uf.x(4)

unwrap_phase(array)

Unwraps phase data in the array. See Section H.C.2 for an example.

H.C.2. Example Usage

Here are several simple example scripts to illustrate the usage of the RSoftUserFunction()
class:

See examples in Section 10.C for more examples

Example 1:
This example opens the data file [Link] and evaluates the function at x=0.25:
from rstools import *
#Create new RSoftUserFunction object
uf = RSoftUserFunction()
#Open data file
[Link](‘[Link]’)
#Evaluate data at x = 0.25 (using interpolation as necessary)
value = [Link](0.25)
#Print value
print(value)

RSoft CAD  339

Example 2:
This example opens the data file [Link] squares the data values, and saves the file:
from rstools import *
#Create new RSoftUserFunction object
uf = RSoftUserFunction()
#Open data file
[Link](‘[Link]’)
#Extract arrays and save in x (coordinate) and f (data value) arrays
(x,f) = uf.get_arrays()
#Square data values
f = f**2
#Store arrays
uf.set_arrays(x,f)
#Save file as data2_squared.dat
[Link](‘data2_squared.dat’)

Example 3:
This example illustrates how to unwrap phase in file [Link] and write the unwrapped data to
the file data_unwrapped.dat.
from rstools import *
#Create new RSoftUserFunction object
uf = RSoftUserFunction()
#Open data file
[Link](‘[Link]’)
#Get arrays
(x,f) = uf.get_arrays()
#Unwrap phase in f array
unwrap_phase(f)
#Write arrays back to object
uf.set_arrays(x,f)
#Save unwrapped data to new file
[Link](‘data_unwrapped.dat’)

RSoft CAD  340

Appendix I

RSoft CAD Release Notes

This appendix summarizes the changes from previous versions of the RSoft CAD Environment.
Notes for versions not listed here can be found in the BeamPROP manual.

Versions 2015.06 and Later

The release notes for version 2015.06 and later versions can be found here:
<rsoft_dir>\readme\releasenotes

Version 2014.09
• With this release, the RSoft Component Suite will no longer officially support 32-bit
operating systems, Windows XP, Windows Vista, or Red Hat Enterprise Linux (RHEL) 4. If
you have any questions, please contact rsoft_support@[Link].
• A new Configure Licensing utility (Windows only) provides an easy way to reconfigure
licensing options. This includes installing a new license key file and restarting SCL. This
option can be opened from the Start Menu and requires administrative rights.
• Several improvements to SCL licensing implementation.
• Updated Linux installation routine.
• Added a new Tree Control to the RSoft CAD window. The tree lists all elements in a design,
including components, circuit references, monitors, pathways, launch fields, user profiles and

RSoft CAD  341

 tapers, layer tables, materials, and embedded circuits. All elements can now be selected and
edited via the tree, in addition to previous methods.
• Added an expiration warning prior to expiration of a user’s license key file.
• Expanded the circuit hierarchy feature to allow child design files to be embedded, or stored,
within the parent file. A list of the embedded files in a parent design file can be viewed using
the new Tree Control in the RSoft CAD.
• Added a Dynamic Array Wizard that simplifies the creation of dynamic arrays. A library of
common unit cell elements is now included. The wizard will automatically embed the desired
unit cell within the parent file.
• Added context-sensitive program usage tips designed to help new users learn to use the
software and teach advanced users about new features.
• Added two new features for the automatic import of refractive index data from text files into
the Material Editor. The Import NK Data option imports the data file into a nondispersive-
type material using the userdata() functions. This approach is sufficient for accurately
modeling dispersion in all simulation tools except FullWAVE pulse or impulse calculations.
The Fit NK Data option attempts to fit the data to Lorentizan coefficients that can be used to
define a dispersive-type material suitable for use with FullWAVE pulse or impulse
calculations.
• Added pan/zoom control via keyboard and mouse.
• Added a new options toolbar to WinPLOT for easy access to commonly used plot options.
This includes plot type, color scales, automatic data scaling, aspect ratio, coordinate system,
axes scale type, complex data display, and symbol/line styles.
• Simulation tools (BandSOLVE and GratingMOD currently) that do not use the same
simulation domain as the Compute Material Profile domain (the domain used when
computing material profiles via the button in the left CAD toolbar) now display the
simulation domain in the CAD as a dashed line to visually denote that the actual simulation
domain is not displayed.
• Improved support for combined arc and width tapers for seg-axis and extended segments.
• Added several new preferences for the Tree Control and several new simulation tool options.

Version 2013.12
• All products in RSoft’s Component Design Suite have been increased to version 2013.12
which is the Synopsys standard.
• Updating Licensing to use Synopsys Common Licensing (SCL).
• Updated Ray-Tracing interface to handle CODE V’s complex and vector field formats.

RSoft CAD  342

• The RSoft field file header has been expanded to include named parameters in the header,
including wavelength and refractive index.
• Added option to bdutil utility to convert an RSoft far-field into a Ray Data File.
• Added option to bdutil utility perform simple arithmetic operations on data files such as
adding files.
• Improved far-field output of the bdutil utility. The –fa option now outputs the same data as
the built-in far-field options of BeamPROP, FullWAVE, DiffractMOD, etc. Contact us with
questions.
• New command line utilities to convert .ind files to DXF and GDS files.
• Enhanced Semiconductor options in the Material Editor.
• New rotatable 3D color-coded plots in WinPLOT.

Changes in Program Behavior

• Changed default file selection mode in DataBROWSER.
• Changed the default parameters and units for the Zemax portion of the Raytracing Converter.
The new units are in microns.
• Changed case-sensitivity of data files on Linux. Previously, all data files referenced in an .ind
design file were case-sensitive. Now, the program will first look for the file as specified in
the design file, and if it is not found, then look an entirely lowercase name.

Changes form Version 8.2 to Version 9.0

The change log which details the history of changes from Version 8.2 to Version 9.0 is available
in the file [Link].
• All products in RSoft’s Component Design Suite have been increased to version 9.0 with this
release.
• Released LED Utility.
• Import/Export features in the Preferences dialog now effect the index generation.
• New PEC material in Material database.
• Components can now be selected by Component Name, if defined.
• New feature to offset a group of components by a fixed value.
• Maximum number of dispersion terms in the Material Editor increased from 6 to 24.
• Dynamically sized arrays of objects can be created with the hierarchy feature.
• Panning options with the mouse, scroll bars, and a keyboard shortcut.

RSoft CAD  343

• New features to simply the drawing of common 3D objects.
• New ‘polygon’ structure type to easily assign a polygon cross-section to a component.
• Improved extended segments that allow for arbitrary position, tapering, and profiles.
• New Z Position Taper for extended segments.
• New ‘twisting’ taper which twists a segment’s profile along the segment axis.
• Arc reference point can now be at the center of the arc.

Changes from Version 8.1 to Version 8.2

The change log which details the history of changes from Version 8.2 to Version 8.2 is available
in the file [Link].
• New group editing features.
• Improved undo functionality.
• New option to interpret user-defined profiles absolutely.
• Released the Solar Cell Utility and Multi-Physics Utility.
• Added new options for far field calculations of periodic fields.
• Added new symbol table functions that allow users to access the real and imaginary
refractive index of a material at a specific wavelength (nreal() and nimag()).
• Many updates to specific simulation engines. See appropriate manual for details.

Changes from Version 8.0 to Version 8.1

The change log which details the history of changes from Version 8.0 to Version 8.1 is available
in the file [Link].
• Updated Material Editor that allows materials to be shared between design files.
• New stress modeling capabilities.
• New 1D mode solver TmmSIM.

Changes from Version 7.0 to Version 8.0

The change log which details the history of changes from Version 7.0 to Version 8.0 is available
in the file [Link].
• The CAD window has been updated with new graphics, a new 'view' toolbar, and new
buttons.

RSoft CAD  344

• Several options which used to be in the Global Settings (Polarization, BPM Vector Mode,
FDTD dispersion/non-linearity, and anisotropy) have been relocated to an appropriate
simulation parameters window.
• Several items in the Additional Segment Properties window have been moved to the
Component Properties window. These include the Display Color and Merge Priority.
• The BPM-based mode solvers included with the CAD are now controlled via the Mode
Options window which can be opened via the Mode Calculation Parameters window (which
can be opened via the button in the left toolbar). This includes the mode method and mode
numbers to find.
• Multipane Mode: This allows the user to view different axis views individually or all at
once. Additionally, a 3D view is available. These options are only available in 3D (in 2D,
the XZ plane is displayed). Multipane view has two modes, 1P and 4P which can be chosen
from the view toolbar. In 1P mode, the user can select one cut direction from the view
toolbar to display the appropriate cross section. In 4P view, all three cross-sections can be
seen at once, as well as a 3D view.
• Cut Mode: This allows the user to view a particular cut of a structure when in an axis view
mode. For example, when viewing a Y cut (the XZ plane), the structure at a particular Y
position can be seen. This is done via the +/- buttons on the view toolbar. The current cut
position, as well as the cut step size are shown and can be changed by the user. To return to
the normal view, click the 'All' button.
• Added a new drawing mode for out of plane segments. This allows the user to place
segments that are perpendicular to the drawing window. For example, segments can be
placed along the Z axis when viewing the XY plane. This is not enabled for all axis views.
• Added the Y coordinate to all component properties windows to parallel X and Z.
• Added an "End equals Start" feature to segments for index and geometry.
• The Compute Index Profile window now has control for cut position as well as a choice of a
either real or imaginary index.
• The Compute Index Profile window now has a Output Options window.
• Added a new mode for defining the refractive index absolutely. This feature should be used
with caution.
• Expanded the CAD preferences window with additional options.
• Launch Fields are now displayed in the CAD window.
• The simulation domain is now displayed in the CAD window.

RSoft CAD  345

Changes from Version 6.0 to Version 7.0
The change log which details the history of changes from Version 6.0 to Version 7.0 is available
in the file [Link].
• Added support to use a variable in file name fields to represent a data file. This allows for
greater ease in using different data files for each step of a scan.
• Introduction of an entirely new index/grid/mesh generation algorithm which improves
accuracy and allows more flexible layout.
• Added support for non-uniform grids. This can significantly improve accuracy while
improving simulation speed.
• Added support for the Material Editor for all simulation tools.
• Added a status line for all simulation tools. The status line now shows the cursor coodinates,
as well as other information such as the BeamPROP Z step or the FullWAVE time step.
• Enabled a cross-cut feature in Winplot.
• Added support for arbitrary polgyons.
• Added abilty to control the angular range of plots for th far field output options.
• Added three new fields to the Advanced Components Properties dialog, which can be
accessed through the More… button. These allow users to name each component, add
comments, and provide special tags for advanced use.
• Expanded the sidewall angle option to include multilayer structure types.

Changes from Version 5.1 to Version 6.0

The change log which details the history of changes from Version 5.1 to Version 6.0 is available
in the file [Link].
• Added a toolbar icon to bring up help for the currently active simulation program. The
CAD/Simulation help icons are distinguished with the words CAD/SIM at the lower right of
the icons, respectively.
• Implemented a new capability for z-dependent user-defined index profiles. To enable
recognition, you must set the variable change_profile = 1. Do not set this variable
unnecessarily as it can affect performance.
• Implemented a new capability for data files to be used in expressions to represent a function
of up to three variables.
• Major updates to FullWAVE, including new features for dispersion, nonlinearity, and
magnetic materials (including negative refractive index materials). More information in the
FullWAVE manual.

RSoft CAD  346

• Improved consistency and functionality of index display options for showing XY, XZ, and
YZ cross-sections. For FullWAVE, BandSOLVE, and DiffractMOD, using the
fdtd_index_profile = 1 option will give a more representative plot for XZ. This option
will be made automatic and consistent with the other options in a future version.
• A new option for displaying the index profile of 3D geometries. The new option is called
"3D Volume", and will generate a volumetric plot of the dielectric constant (i.e., N^2, not N).
• A major new feature for hierarchical CAD layout. This feature lets you include references to
circuits (.ind files) inside another circuit.
• A new, unified array layout generator.
• Corrected an error in the "disperse" utility which produced incorrect results for some
combinations of options.
• Adds a feature, applicable to all products, which allows the imaginary part of the reference
background index to be specified in the additional component properties dialog. This can be
important when defining some structures to improve index averaging of the imaginary index
at interfaces.
• During an anisotropic calculation, all index tensors can be saved via the index_profile_all
option.
• General speed improvements.

RSoft CAD  347

 Accept Fit, 157 Circle Mode, 77
Accept Symbol, 14, 35 Circuit Reference, 25, 37, 105
Add, 96 Circuit Reference Box, 108, 120
Add to Project, 147 Clear Log, 22
All, 18 Clipping, 102
Allow Mixed Properties, 120 Close, 22
Allow Mixed Types, 120 Close All, 22

Index Allow Redraw 2D, 48

Allow Redraw 3D, 48
Alloy X, 167
Alloy Y, 167
Close All Browsers, 22
Close All Plots, 22
Cluster Settings, 22
color scale, 48, 120
Angle, 55, 79 Color Shades, 48, 308, 309
Angle Value, 55 Combine Mode, 96
anisotropic, 152, 154 Comment, 123, 148
Arc, 25, 37 Component Color, 120
Arc (XZ), 91, 92 Component Delta-N, 12, 31
Arc Data…, 71, 92 Component Diameter, 77
Arc Final Angle, 71, 92 Component Fields, 130
Arc Initial Angle, 71, 92 Component Height, 31, 57, 58, 59,
Arc Radius, 92 60, 62, 64, 66, 67, 74, 77, 79,
Arc Reference, 71, 92 83, 181
Arc Type, 92 Component Material, 31
Arrange Icons, 22 Component Name, 123
Array Layout, 22 Component Tree, 120
Aspect Ratio, 25 Component Width, 18, 31, 50, 57,
( Attach On Edit, 120 58, 59, 60, 62, 64, 67, 74, 83,
Attach Tolerance, 120 181
(Acceptors cm-3), 167
Auto, 41 Compute Modes, 22, 25, 202
AutoFit, 157 Contents, 139
. AutoFit Method, 157 Continue Simulation, 282
Automatically Regrid after Zoom, Convert a GDS-II
.ind file, 140, 141
120 DXF
Azimuthal Mode #, 204 or CIF file into an RSoft .ind
1 design file:, 102
Convert to Polygons, 22
1:1, 15 B Copy (Ctrl-C), 22
1P, 12, 15, 40 Copy Material, 146
Back Radius, 74
Background Index, 12, 17, 31, 56, Copy Selection, 25
2 58, 59, 60, 62, 64, 67, 83 Correlation Length, 99, 100
Background Material, 31, 59, 60, 62, Cosine S-Bend, 92
2.5D, 31 64, 67 Cover Index, 31, 60, 62, 64
2D, 31 BCC, 115 Cover Material, 31, 60, 62, 64
Bend Origin, 98 Crystal Axes, 125, 154
3 Bounding Shape, 115 Cubic, 115
Custom, 148
3D, 12, 31 cut, 178
3D Shape, 77 C Cut (Ctrl-X), 22
3D Structure Type, 25, 31, 39, 56, Cascade, 22 Cut Direction, 102
57, 58, 69, 71, 74, 79, 83, 130 Center, 71 Cut Position, 102
3D Structure Types, 128 Center Color, 120 Cut Selection, 25
Center Thickness, 74 Cut View Controls, 25
4 Centered Cells, 174 Cylinder, 77
Change Order, 22
4P, 12, 15, 40 channel, 128 D
Chi2, 159
A Chi3, 159 Data File, 88, 95
Chi3 Saturation Coef:, 159 Data File..., 79
Absolute index in User Profile, 101, Choose Simulation Tool, 25 Data Mode, 79
259 Circle, 25, 37 Date Mode, 67
Default, 83, 96, 148, 174 Edit Table, 171 GDS file:, 106
Default (Fixed Z), 47 Edit User Profiles, 25 Global Settings, 22
Defect Size, 115 Edit User Tapers, 25 Global Symbols…, 108
Definition, 130 Edit X Axis View, 25 Go, 22
Delete, 116 Edit Y Axis View, 25 GPU Settings, 22
Delete (Del), 22 Edit Z Axis View, 25 Grading Ratio, 174
Delete Filter, 133 Effective Index Calculation, 31, 128 GratingMOD Grating Layout, 22
Delete Group, 133 Electrode/Heater Type, 123 Grid Edge, 177
Delete Material or Library, 146 Electro-Optic, 164 Grid Edge Factor, 174, 177
Delete Symbol, 35 Embedded, 106 Grid Edge Size, 174
Diamond lattices, 115 Embedded Circuits, 106 Grid Grading, 174, 177
diffused, 83 Embedded File, 106 Grid Minimum Divisions, 177
Dimensions, 136 Enable Auto Clustering, 120 Grid Options…, 174
Direction Indicators, 120 Enable Free-Carrier Index Effects, Grid PPW, 177
Dispersion/Nonlinearity, 155, 159 167 Grid Ratio, 177
Dispersive, 152, 155, 157 Enable Incomplete Ionization, 167 grid size, 174, 177
display, 148 Enable Nonuniform, 174 Grid Size (Bulk), 174
Display Axes, 148 Enable Radial, 204 Group Libraries, 145
Display Color, 54, 130, 148, 181 Enable Scroll Bars, 120 Group Properties, 133
Display Material Profile, 20, 22, 25, End, 125
45, 202, 207 Eps Linear, 151
Display Mode, 47, 48 Eps Nonlinear, 159
H
Display Opts…, 148 Eps_xx, 164 Height, 81
Display…, 48, 309 Eps_yy, 164 Height Taper, 91
dN/dT, 165 Eps_zz, 164 HeightCoded, 48
Domain Max, 174, 204 Epsilon (imag), 152 Help for Active Sim Program, 25
Domain Min, 20, 47, 174, 204 Epsilon (real), 152 Help for CAD program, 25
Donors (cm-3), 167 Exit, 22 Hexagonal, 115
Draw Center Line, 120 exponential, 91 Hexagonal (HCP), 115
Draw Edge, 120 Export Circuit, 22 Hide Group, 133
Draw Interior, 120 Export Mask Layers, 103
Draw Priority, 123, 130 Export Polygon Mode, 103
Drawing Preferences, 39 Export:, 102 I
Drawing Tool Options, 25 Expression, 14, 35 Import Circuit, 22
Duplicate, 22 Extended, 69 Import NK Data…, 152
Duplicate Selection, 25 Extended Segments, 25, 37, 69, 71 Import Symbols, 22
Dynamic Array, 25, 115 External File, 106, 181 Import/Export Resolution, 69, 103
Dynamic Color Shades, 120 Import:, 102
Dynamic Load, 67, 79 In, 22
Dynamically reference a GDS-II
F
Inactive, 83, 117
IND FCC, 115 Inactive Color, 120
or SPT file using a Circuit File Name, 102 Index (imag part), 56, 66
Reference component:, 102 Final Angle, 92 Index (imag), 152
Fit NK Data…, 157 Index (Imag. part), 59, 60, 62, 67
E Fit Tol, 157 Index (real), 152
Flatten Circuit, 22 Index Difference, 17, 56, 58, 59, 60,
E/H Control Parameter, 123 Flip Horizontal, 22 62, 66, 67, 77, 83, 87
Edge Color, 120 Flip Section Horizontally, 25 Index Display Max, 48
Edit Default Tags, 177 Flip Section Vertically, 25 Index Display Min, 48
Edit Default Tags…, 22 Flip Vertical, 22 Index Generation, 22
Edit Filter, 133 Free, 92 Index Input Mode, 120
Edit Global Settings, 25 Free Space Wavelength, 31, 157 Index Profile Type, 31, 74, 83, 87
Edit Hierarchy Exports…, 22 Front Radius, 74 Index Taper, 91
Edit Launch Field, 25 Full, 22 Initial Angle, 92
Edit Layers, 25, 65 Input Mode, 152, 154
Edit Materials, 25 Insert, 22
Edit Pathway Monitors, 25
G
Interface Alignment, 174
Edit Pathways, 25, 116 Gain Term, 155, 159
Edit Symbols, 14, 25, 35, 108 gaussian, 83
K MPI Display Whole Domain, 120 Perform Simulation, 9
Mu Linear, 162 Period, 115
Kappa_xx, 165 Mu Nonlinear, 163 Phi, 74, 125
Kappa_yy, 165 multilayer, 31 Plot Aspect Ratio, 48
Kappa_zz, 165 Multi-Pane View, 120 Plot Range/Resolution, 148
Keep these Settings Next Time, 143 Multi-Physics, 22 Points, 87, 88, 94
Multi-Threading Settings, 22 Poisson’s Ratio (nu), 166
L Pol (3D EIM), 128
polarization, 204
Lambda Min Lambda Max, 157 N Polarization Angle, 317
Last, 22 Name, 14, 35, 130 Poly Info…, 67
Lattice Info…, 106, 115 Neff (imag), 204 polygon, 25, 37
Lattice Size, 115, 136, 137, 138 Neff (Real), 204 Polygon File, 67
Lattice Type, 115, 141 New, 116 Polygon Mode, 79
Launch DataBROWSER, 22, 25 New Circuit, 12, 25, 31 Preferences, 22
Launch MOST, 25 New Library, 145, 146 Print (Ctrl-P), 22
Launch Normalization, 303 New Material, 144, 146 Printer Setup, 22
Launch Power, 159, 303 New Profile…or User #, 83 Priority, 77
Layer Table, 64, 66 New Symbol, 14, 35 profile type, 57, 117
LED, 22 New Taper…, 91 Profiles…, 87
lens, 25, 37 New… (Ctrl-N), 22 Psi, 125
Lens Mode, 74 None, 55, 91, 159, 174
Lens Shape, 74 Normalization, 159
Line Thickness, 81 Q
Number of Sides, 67
linear, 91 Q-Finder, 22
Linear Eps, 157 quadratic, 91
Litho Resolution, 99 O
Litho Roughness, 99, 100 Offset, 17, 55
Load Settings, 12, 31, 102 R
Offset Value, 17, 55
Local Symbols, 181 OK, 9, 12, 18, 20, 202 R_x, 164
Local Symbols…, 108, 115 Open Circuit, 25 R_y, 164
Locally Defined, 147, 151 Open Folder In Working Directory, R_z, 164
Lorentzian, 157 25 Radial Calculation, 31
Open Log, 22 Radius, 92
M Open Terminal in Working Raised Sine S-Bend, 92
Directory, 25 Randomness, 142
Map File, 120 Open… (Ctrl-O), 22 Raster, 103
Mark Type, 81 Origin, 148 Raw, 103
Mask Layer, 103, 123, 130 Out, 22 Raytracing-Converters, 22
Material, 58 Outline Color, 48 Rectangular, 115
Material Properties, 144, 147, 151, Output All Index Tensor Elements, Redo (Ctrl-Y), 22
181 49 Redo Last Change, 25
Material Settings, 102 Output Prefix, 49, 143, 148, 202, 204 ReDraw, 22
Material System, 167 Output…, 49 Redraw Circuit, 25
Materials…, 144 Overlay Grid, 178 Reference Phi, 69, 124
Max, 87, 88, 94 Reference Psi, 69, 124
Max # Lorentzians, 157 Reference Theta, 69, 124
Max Level Undo, 120 P Reference Type, 17, 55
Measurement Data, 148 P1, 166 Reference X, 124
Merge, 96 P2, 166 Reference Y, 124
Merge Priority, 96, 130, 181 P3, 166 Reference Z, 124
Min, 87, 88, 94 Pan Mode, 25, 41 ReGrid, 22
Min/Max, 89 Paste (Ctrl-V), 22 Reject Symbol, 35
Minimum Divisions, 174 Paste Clipboard Contents, 25 Relative To, 55
Mode Options…, 202 Paste Material, 146 Rib/Ridge, 31
Model Dimension, 12, 31, 83 Paste Offset, 120 Rotate Selection, 25
Monitor, 25, 37 Pathway Color, 120 Rotate…, 22
More…, 98, 122, 340 pathway edit mode, 116 RSoft .ind file:, 106
MOST Optimizer/Scanner, 22 PDK Info file, 31, 120, 232 Rubberband Color, 120
Run Script, 22 Structure Preference, 25, 58 Warnings, 22
Surface Color, 47, 48 WDM Router Layout…, 22
Width Taper, 91
S Width/Length, 81
T
Save (Ctrl-S), 22
Save As…, 22 Tables, 22 X
Save Circuit, 25 Tag, 123
Save Library, 146 Tags, 177 X, 12, 40, 55
s-bend, 25, 37, 92 Tapered-Laser, 22 X Alloy, 167
Scale Factor, 115 Tapers…, 93 X Cut, 47
Scale File…, 309 Target Value, 130 X Pos Taper, 91
Scale Files…, 48 TE, 204 X Scale, 79
Seg Orientation, 69 Test, 87 X’Y’ Rotation, 69
Seg-Axis, 69 Text Editor, 35, 50 X’Y’ Rotation End, 124
Segment (In Plane), 15, 25, 37 Thermal Exp. (alpha), 166 X’Y’ Rotation Start, 124
Segment (Out of Plane), 25, 37 Thermo-Optic, 165 X’Y’ Rotation Taper, 124
Select All, 22, 133 Theta, 125 xx(im), 154
select mode, 25, 42 Tile, 22 xx(re), 154
Select Tolerance, 120 TM, 204 XZ Rotation Angle, 106
Select/Edit Components…, 22 TMM-Mode-Solver, 22
Semiconductor, 167 Toggle Component Tree, 15, 25, 30
Send Forward/Send Backward, 25 Try Fit, 157
Y
Send to Front/Send to Back, 25 Y, 12, 40, 55
Set View, 22 Y Alloy, 167
Show 3D Pane, 120
U
Y Cut, 47
Show 3D View, 25 Undo, 15 Y Pos Taper, 91
Show All Draw Tools, 120 Undo Last Change, 25 Yes, 98
Show Color Scale, 48 Undo(Ctrl-Z), 22 Young’s Modulus (E), 166
Show Default Comp as Comp Color, Unit Peak, 159
120 Unit Power, 159
Show Group, 133 Use Exact Geometry data in a
Z
Show Launch Fields, 120 Sentaurus TCAD TDR file in Z, 12, 40, 55
Show Simulation Domain, 120 an RSoft .ind design file:, 102 Z Cut, 47
sidewall angle, 59, 99, 123 Use Magnetic Materials, 162, 163 Z Pos Taper, 91
Sidewall Angle Ref., 99, 123 Use Material (Add to Project), 146 Z Position, 204
Simple, 157 Use Material Colors, 120 Z Scale, 79
simulated bend, 98, 123 Use Perpendicular Launch Width z-Axis, 69
Simulated Bend Radius, 98, 123 Convention, 120 Zoom Factor (>1.0), 120
Simulation Region, 25, 37 Use Single Precision, 120 Zoom Full, 25, 41
Simulation Tool, 12, 31 User #, 91 Zoom In, 25, 41
Single-Pane/Four-Pane, 25 zoom mode, 25, 41
Slab Height, 31, 62, 64, 66 Zoom Out, 25, 41
Slab Index, 31, 62 V
Zoom/Full Affects All Panes, 120
Slab Material, 31, 62 Vector Mode, 317 Zoom/Full Border, 120
Slice Grid, 47 Via an Index Difference:, 56 Zoom/Full Uses Simulation Domain,
Slice Output, 120 Via Materials:, 56 120
Slice Position Y, 47 View BSDF File, 22
Solar-Cell, 22 View Graphs, 25, 309
SolidModel, 48 View Grid, 178
Source Type, 87, 88, 94, 95 View Phi, 48
Special Effects, 142 View Single Plot, 22
Sphere, 77 View S-Matrix File, 22
Start, 125 View Theta, 48
Start with Tree Expanded, 120
Static Color Shades, 120
Static Load, 79 W
step index, 83 Warn when automatically resizing
Straddled Cells, 174 grid, 120
Stress-Optic, 166