Ze Max Manual

ZEMAX
Optical Design Program
User's Guide
February 3, 2005
ZEMAX Development Corporation

[email protected]
www.zemax.com
Important notice
ZEMAX is a registered trademark of ZEMAX Development Corporation Copyright ZEMAX Devel-

opment Corporation 1990-2005. All rights reserved.
All other product names or trademarks are property of their respective owners.
Information in this manual is subject to change without notice and does not represent a commitment on
the part of the vendor. The software described in this manual is furnished under a license agreement
and may be used or copied only in accordance with the terms of the agreement.
The material in this manual is copyrighted and may not be reproduced or electronically distributed
without written permission from the copyright holder.
ZEMAX Development Corporation (ZDC) provides this publication "as is" without warranty of any kind,
either express or implied, including but not limited to the implied warranties or conditions of merchant-
ability or fitness for a particular purpose. In no event shall ZDC be liable for any loss of profits, loss of
business, loss of use or data, interruption of business, or for indirect, special, incidental, or consequential
damages of any kind, even if ZDC has been advised of the possibility of such damages arising from any
defect or error in this publication or in the Software.
TABLE OF CONTENTS
INTRODUCTION Chapter 1
About this document ............................................................................................................................... 31
What does ZEMAX do? .......................................................................................................................... 31
What doesn't ZEMAX do?....................................................................................................................... 31
Learning to use ZEMAX.......................................................................................................................... 32
System requirements .............................................................................................................................. 32
Multiple processor computers ..................................................................................................... 32
Installation procedure.............................................................................................................................. 32
Installing the key ......................................................................................................................... 32
Running ZEMAX ......................................................................................................................... 33
Policy on the key..................................................................................................................................... 33
Definition of support ................................................................................................................................ 33
Getting technical support ........................................................................................................................ 33
Policy on bug fixes .................................................................................................................................. 33
USER INTERFACE Chapter 2
Introduction ............................................................................................................................................. 35
Types of windows ................................................................................................................................... 35
Main window operations ......................................................................................................................... 35
Editor windows operations ...................................................................................................................... 36
Graphic windows operations................................................................................................................... 36
Using the annotation feature ....................................................................................................... 38
Using pan and zoom ................................................................................................................... 39
Text windows operations ........................................................................................................................ 39
Dialog operations .................................................................................................................................... 39
Aborting long computations .................................................................................................................... 40
Summary of useful shortcuts .................................................................................................................. 40
Using the Windows clipboard.................................................................................................................. 42
ZEMAX file types by extension ............................................................................................................... 42
CONVENTIONS AND DEFINITIONS Chapter 3
Introduction ............................................................................................................................................. 45
Active configuration................................................................................................................................. 45
Angular magnification ............................................................................................................................. 45
Apodization ............................................................................................................................................. 45
Back focal length..................................................................................................................................... 45
Cardinal planes ....................................................................................................................................... 45
Chief ray.................................................................................................................................................. 45
Coordinate axes...................................................................................................................................... 46
Diffraction limited .................................................................................................................................... 46
Edge thickness........................................................................................................................................ 46
Effective focal length............................................................................................................................... 47
Entrance pupil diameter .......................................................................................................................... 47
Entrance pupil position............................................................................................................................ 47
Exit pupil diameter .................................................................................................................................. 47
Exit pupil position .................................................................................................................................... 47
Extra data................................................................................................................................................ 47
Field angles and heights ......................................................................................................................... 47
Float by stop size .................................................................................................................................... 47
3
Ghost reflections ..................................................................................................................................... 48
Glasses ................................................................................................................................................... 48
Hexapolar rings....................................................................................................................................... 48
Image space F/# ..................................................................................................................................... 48
Image space numerical aperture (NA) .................................................................................................... 48
Lens units................................................................................................................................................ 48
Marginal ray ............................................................................................................................................ 48
Maximum field......................................................................................................................................... 48
Non-paraxial systems ............................................................................................................................. 49
Non-sequential ray tracing ...................................................................................................................... 49
Normalized field and pupil coordinates ................................................................................................... 49
Object space numerical aperture ............................................................................................................ 50
Parameter data ....................................................................................................................................... 50
Paraxial and parabasal rays ................................................................................................................... 50
Paraxial image height ............................................................................................................................. 51
Paraxial magnification............................................................................................................................. 51
Paraxial working F/# ............................................................................................................................... 51
Primary wavelength ................................................................................................................................ 51
Radii........................................................................................................................................................ 51
Real propagation..................................................................................................................................... 51
Sagittal and Tangential ........................................................................................................................... 51
Semi-diameters....................................................................................................................................... 52
Sequential ray tracing ............................................................................................................................. 52
Strehl ratio............................................................................................................................................... 52
Surface apertures ................................................................................................................................... 52
System aperture...................................................................................................................................... 53
Tangential ............................................................................................................................................... 53
Thicknesses ............................................................................................................................................ 53
Total internal reflection (TIR) .................................................................................................................. 53
Total track ............................................................................................................................................... 53
Units........................................................................................................................................................ 53
Vignetting factors .................................................................................................................................... 53
Virtual propagation.................................................................................................................................. 55
Wavelength data ..................................................................................................................................... 55
Working F/#............................................................................................................................................. 55
FILE MENU Chapter 4
New......................................................................................................................................................... 57
Open ....................................................................................................................................................... 57
Save........................................................................................................................................................ 57
Save As................................................................................................................................................... 57
Use Session Files ................................................................................................................................... 57
Program Mode ........................................................................................................................................ 57
Insert Lens .............................................................................................................................................. 58
Preferences............................................................................................................................................. 58
Address ....................................................................................................................................... 58
Directories ................................................................................................................................... 59
Graphics...................................................................................................................................... 59
Miscellaneous ............................................................................................................................. 60
Editors ......................................................................................................................................... 61
Printing ........................................................................................................................................ 62
Colors.......................................................................................................................................... 63
Buttons 1-16, Buttons 17-32, Buttons 33-48 ............................................................................... 63
Status Bar ................................................................................................................................... 64
4
Exit .......................................................................................................................................................... 64
Recently used files.................................................................................................................................. 64
EDITORS MENU Chapter 5
Lens Data................................................................................................................................................ 65
Inserting and deleting surfaces ............................................................................................................... 65
The surface number display........................................................................................................ 65
Cutting, copying, and pasting surface data ................................................................................. 65
Entering surface comments ........................................................................................................ 66
Entering radii data ....................................................................................................................... 66
Entering thickness data............................................................................................................... 66
Entering glass data ..................................................................................................................... 66
Entering semi-diameter data ....................................................................................................... 66
Entering conic data ..................................................................................................................... 66
Entering parameter data ............................................................................................................. 66
The Surface Properties dialog box.............................................................................................. 66
Surface properties type tab ......................................................................................................... 67
Surface Type ................................................................................................................... 67
Surface DLL .................................................................................................................... 67
Surface Color .................................................................................................................. 67
Surface Opacity............................................................................................................... 67
Row Color........................................................................................................................ 67
Make Surface Stop.......................................................................................................... 67
Make Surface Global Coordinate Reference................................................................... 67
Surface Cannot Be Hyperhemispheric ............................................................................ 67
Surface properties draw tab ........................................................................................................ 68
Hide Rays To This Surface ............................................................................................. 68
Skip Rays To This Surface.............................................................................................. 68
Do Not Draw This Surface............................................................................................... 68
Do Not Draw Edges From This Surface .......................................................................... 68
Draw Edges As................................................................................................................ 68
Surface properties aperture tab .................................................................................................. 68
Aperture type and other aperture controls....................................................................... 68
User defined apertures and obscurations ....................................................................... 69
The UDA file format......................................................................................................... 70
UDA Examples ................................................................................................................ 71
Surface properties scattering tab ................................................................................................ 72
Surface scattering settings .............................................................................................. 73
Surface tilt/decenter tab .............................................................................................................. 73
Surface physical optics tab ......................................................................................................... 74
Surface coating tab ..................................................................................................................... 74
Setting and removing solves ....................................................................................................... 74
Setting and removing variables................................................................................................... 74
Menu options............................................................................................................................... 74
Edit .............................................................................................................................................. 74
Solves ......................................................................................................................................... 75
Options........................................................................................................................................ 75
Help............................................................................................................................................. 75
Merit Function ......................................................................................................................................... 76
Edit .............................................................................................................................................. 76
Tools ........................................................................................................................................... 76
Help............................................................................................................................................. 77
Multi-Configuration.................................................................................................................................. 77
Edit .............................................................................................................................................. 77
5
Solves ......................................................................................................................................... 77
Tools ........................................................................................................................................... 78
Help............................................................................................................................................. 78
Tolerance Data ....................................................................................................................................... 78
Edit .............................................................................................................................................. 78
Tools ........................................................................................................................................... 79
Help............................................................................................................................................. 79
Extra Data ............................................................................................................................................... 79
Edit .............................................................................................................................................. 80
Solves ......................................................................................................................................... 80
Tools ........................................................................................................................................... 80
Help............................................................................................................................................. 80
Non-Sequential Components.................................................................................................................. 80
Menu options............................................................................................................................... 80
Edit .............................................................................................................................................. 80
Solves ......................................................................................................................................... 81
Errors .......................................................................................................................................... 82
Detectors..................................................................................................................................... 82
Database..................................................................................................................................... 82
Help............................................................................................................................................. 82
Undo, Redo, and Recover ...................................................................................................................... 82
Undo: None ................................................................................................................................. 82
Undo: Memory 1 Step ................................................................................................................. 82
Undo: Disk Multi Step.................................................................................................................. 82
SYSTEM MENU Chapter 6
Update .................................................................................................................................................... 85
Update All ............................................................................................................................................... 85
General ................................................................................................................................................... 85
Aperture ...................................................................................................................................... 85
Aperture Type.................................................................................................................. 85
Aperture Value ................................................................................................................ 85
Apodization Type............................................................................................................. 86
Uniform apodization ........................................................................................................ 86
Gaussian apodization...................................................................................................... 86
Cosine cubed apodization ............................................................................................... 86
User defined apodization................................................................................................. 86
Apodization Factor .......................................................................................................... 87
Telecentric Object Space ................................................................................................ 87
Units ............................................................................................................................................ 87
Lens Units ....................................................................................................................... 87
Source Units.................................................................................................................... 87
Analysis Units.................................................................................................................. 87
Title/Notes ................................................................................................................................... 88
Lens Title......................................................................................................................... 88
Notes ............................................................................................................................... 88
Glass Catalogs............................................................................................................................ 88
Ray Aiming.................................................................................................................................. 88
Ray Aiming ...................................................................................................................... 88
Use Ray Aiming Cache ................................................................................................... 89
Robust Ray Aiming (slow) ............................................................................................... 89
Pupil Shift: X, Y, and Z .................................................................................................... 89
Environment ................................................................................................................................ 90
Use Temperature, Pressure ............................................................................................ 90
6
Temperature in degrees C .............................................................................................. 90
Pressure in ATM.............................................................................................................. 90
Polarization ................................................................................................................................. 90
Unpolarized ..................................................................................................................... 90
Jx, Jy, X-Phase, Y-Phase................................................................................................ 90
Convert thin film phase to ray equivalent ........................................................................ 91
Files............................................................................................................................................. 91
Coating File ..................................................................................................................... 91
Scatter Profile.................................................................................................................. 91
ABg Data File .................................................................................................................. 91
GRADIUM Profile ............................................................................................................ 91
Miscellaneous ............................................................................................................................. 91
Reference OPD ............................................................................................................... 91
Paraxial Rays .................................................................................................................. 92
Fast Asphere Trace......................................................................................................... 92
Fast Semi-Diameters....................................................................................................... 92
Check GRIN Apertures.................................................................................................... 93
Semi Diameter Margin (lens units) .................................................................................. 93
Semi Diameter Margin % ................................................................................................ 93
Global Coordinate Reference Surface ............................................................................ 93
Method To Compute F/# ................................................................................................. 94
Non-Sequential ........................................................................................................................... 94
Maximum Intersections Per Ray ..................................................................................... 94
Maximum Segments Per Ray.......................................................................................... 95
Maximum Nested/Touching Objects ............................................................................... 95
Minimum Relative Ray Intensity ...................................................................................... 95
Minimum Absolute Ray Intensity ..................................................................................... 95
Glue Distance In Lens Units............................................................................................ 95
Missed Ray Draw Distance in Lens Units ....................................................................... 96
Retrace source rays upon file open................................................................................. 96
Fields ...................................................................................................................................................... 96
Vignetting factors ........................................................................................................................ 96
Saving and loading field data ...................................................................................................... 96
Wavelengths ........................................................................................................................................... 96
Saving and loading wavelength data .......................................................................................... 97
Next Configuration .................................................................................................................................. 97
Last Configuration................................................................................................................................... 97
ANALYSIS MENU Chapter 7
Introduction ............................................................................................................................................. 99
Layout ..................................................................................................................................................... 99
2D Layout.................................................................................................................................... 99
3D Layout.................................................................................................................................. 100
The orientation indicator................................................................................................ 101
Ray errors...................................................................................................................... 101
Configuration data ......................................................................................................... 102
Raylist file format........................................................................................................... 102
Wireframe.................................................................................................................................. 102
Solid Model ............................................................................................................................... 103
Shaded Model ........................................................................................................................... 103
ZEMAX Element Drawing ......................................................................................................... 104
Special characters......................................................................................................... 105
ISO Element Drawing................................................................................................................ 105
Summary of ISO 10110 symbols and codes ................................................................. 106
7
Diameters and bevels.................................................................................................... 107
Tolerance data .............................................................................................................. 107
NSC 3D Layout ......................................................................................................................... 107
NSC Shaded Model .................................................................................................................. 108
Fans ...................................................................................................................................................... 108
Ray Aberration .......................................................................................................................... 108
Evaluating results at intermediate surfaces................................................................... 109
Optical Path............................................................................................................................... 110
Pupil Aberration......................................................................................................................... 110
Spot Diagrams ...................................................................................................................................... 111
Standard.................................................................................................................................... 111
Through Focus .......................................................................................................................... 113
Full Field.................................................................................................................................... 113
Matrix ........................................................................................................................................ 113
Configuration Matrix .................................................................................................................. 113
MTF....................................................................................................................................................... 114
FFT MTF ................................................................................................................................... 114
FFT Through Focus MTF .......................................................................................................... 115
FFT Surface MTF...................................................................................................................... 116
FFT MTF vs. Field..................................................................................................................... 116
Comment about vignetting factors................................................................................. 117
FFT MTF Map ........................................................................................................................... 117
Huygens MTF............................................................................................................................ 118
Huygens Through Focus MTF .................................................................................................. 119
Huygens Surface MTF .............................................................................................................. 119
Geometric MTF ......................................................................................................................... 119
Geometric Through Focus MTF ................................................................................................ 120
Geometric MTF vs. Field........................................................................................................... 121
Geometric MTF Map ................................................................................................................. 121
PSF ....................................................................................................................................................... 122
FFT PSF.................................................................................................................................... 122
Assumptions used in the FFT PSF calculation.............................................................. 123
Discussion of the FFT method and sampling issues..................................................... 123
FFT PSF Cross Section ............................................................................................................ 125
FFT Line/Edge Spread.............................................................................................................. 125
Huygens PSF ............................................................................................................................ 126
Huygens PSF Cross Section..................................................................................................... 127
Wavefront.............................................................................................................................................. 128
Wavefront Map.......................................................................................................................... 128
Interferogram............................................................................................................................. 128
Foucault Analysis ...................................................................................................................... 129
Surface.................................................................................................................................................. 130
Surface Sag .............................................................................................................................. 130
The Contour Format String............................................................................................ 131
Surface Phase........................................................................................................................... 131
RMS ...................................................................................................................................................... 132
RMS vs. Field............................................................................................................................ 132
Comments about RMS wavefront computations ........................................................... 133
RMS vs. Wavelength................................................................................................................. 133
RMS vs. Focus.......................................................................................................................... 134
Encircled energy ................................................................................................................................... 135
Diffraction .................................................................................................................................. 135
Geometric.................................................................................................................................. 136
Geometric Line/Edge Spread.................................................................................................... 137
8
Extended Source....................................................................................................................... 137
Illumination............................................................................................................................................ 139
Relative Illumination .................................................................................................................. 139
Effective F/# .................................................................................................................. 140
Vignetting Plot ........................................................................................................................... 140
Illumination XY Scan ................................................................................................................. 140
Illumination 2D Surface ............................................................................................................. 142
Image Analysis...................................................................................................................................... 142
Geometric Image Analysis ........................................................................................................ 142
The IMA format.............................................................................................................. 144
The BIM format.............................................................................................................. 144
How rays are chosen for analysis ................................................................................. 144
Calculating efficiency of multi-mode fibers .................................................................... 145
Text output .................................................................................................................... 146
Geometric Bitmap Image Analysis ............................................................................................ 146
Diffraction Image Analysis......................................................................................................... 148
Comment about the coherent optical transfer function.................................................. 151
Extended Diffraction Image Analysis ........................................................................................ 151
IMA/BIM File Viewer.................................................................................................................. 153
Biocular Analysis................................................................................................................................... 153
Field of View.............................................................................................................................. 153
Dipvergence/Convergence........................................................................................................ 154
Miscellaneous ....................................................................................................................................... 156
Field Curvature/Distortion ......................................................................................................... 156
Grid Distortion ........................................................................................................................... 158
Footprint Diagram ..................................................................................................................... 160
Longitudinal Aberration ............................................................................................................. 160
Lateral Color.............................................................................................................................. 161
Y-Ybar Diagram ........................................................................................................................ 161
Chromatic Focal Shift................................................................................................................ 162
Dispersion Diagram................................................................................................................... 162
Glass Map ................................................................................................................................. 163
Internal Transmittance vs. Wavelength..................................................................................... 163
System Summary Graphic ........................................................................................................ 163
Aberration coefficients .......................................................................................................................... 164
Seidel Coefficients .................................................................................................................... 164
Zernike Fringe Coefficients ....................................................................................................... 165
Strehl ratio approximation ............................................................................................. 167
Zernike Standard Coefficients................................................................................................... 168
Zernike Annular Coefficients ..................................................................................................... 171
Calculations .......................................................................................................................................... 174
Ray Trace.................................................................................................................................. 174
Fiber Coupling Efficiency .......................................................................................................... 174
YNI Contributions ...................................................................................................................... 176
Sag Table.................................................................................................................................. 177
Cardinal Points.......................................................................................................................... 178
Glass and Gradient Index ..................................................................................................................... 178
Dispersion Diagram................................................................................................................... 178
Glass Map ................................................................................................................................. 178
Internal Transmittance vs. Wavelength..................................................................................... 179
Grin Profile ................................................................................................................................ 179
Gradium Profile...................................................................................................................... 180
Universal Plot........................................................................................................................................ 180
New Universal Plot.................................................................................................................... 180
9
Polarization ........................................................................................................................................... 182
Polarization Ray Trace.............................................................................................................. 182
Polarization Pupil Map .............................................................................................................. 182
Transmission............................................................................................................................. 183
Phase Aberration ...................................................................................................................... 183
Transmission Fan...................................................................................................................... 184
Coatings................................................................................................................................................ 184
Reflection vs. Angle .................................................................................................................. 184
Transmission vs. Angle ............................................................................................................. 185
Absorption vs. Angle ................................................................................................................. 185
Diattenuation vs. Angle ............................................................................................................. 185
Phase vs. Angle ........................................................................................................................ 185
Retardance vs. Angle................................................................................................................ 186
Reflection vs. Wavelength......................................................................................................... 186
Transmission vs. Wavelength ................................................................................................... 186
Absorption vs. Wavelength ....................................................................................................... 186
Diattenuation vs. Wavelength ................................................................................................... 187
Phase vs. Wavelength .............................................................................................................. 187
Retardance vs. Wavelength ...................................................................................................... 187
Physical Optics ..................................................................................................................................... 187
Paraxial Gaussian Beam........................................................................................................... 187
Limitations of the analysis ............................................................................................. 188
Overview of Gaussian beams ....................................................................................... 188
Default beam parameters.............................................................................................. 188
Propagating the embedded beam ................................................................................. 188
The quality factor........................................................................................................... 189
Interactive analysis........................................................................................................ 189
Skew Gaussian Beam............................................................................................................... 189
Physical Optics Propagation ..................................................................................................... 190
Comment about beam projection .................................................................................. 192
Beam File Viewer ...................................................................................................................... 193
TOOLS MENU Chapter 8
Optimization .......................................................................................................................................... 195
Optimization .............................................................................................................................. 195
Global Search ........................................................................................................................... 195
Hammer Optimization ............................................................................................................... 195
Merit Function Listing ................................................................................................................ 195
Remove All Variables................................................................................................................ 195
Glass Substitution Template ..................................................................................................... 195
Tolerancing ........................................................................................................................................... 196
Tolerancing ............................................................................................................................... 196
Tolerance Listing....................................................................................................................... 196
Tolerance Summary.................................................................................................................. 196
Test Plates ............................................................................................................................................ 197
Test Plate Fitting ....................................................................................................................... 197
Test Plate Lists.......................................................................................................................... 198
Catalogs................................................................................................................................................ 198
Glass Catalogs.......................................................................................................................... 198
Glass Compare ......................................................................................................................... 199
Lens Catalogs ........................................................................................................................... 199
Coatings................................................................................................................................................ 202
Edit Coating File........................................................................................................................ 202
Reload Coating File................................................................................................................... 203
10
Add Coatings to All Surfaces .................................................................................................... 203
Coating Listing .......................................................................................................................... 203
Scattering.............................................................................................................................................. 203
ABg Scatter Data Catalogs ....................................................................................................... 203
Wavelength scaling of ABg data ................................................................................... 204
Scatter Function Viewer ............................................................................................................ 204
Apertures .............................................................................................................................................. 206
Convert Semi-Diameters to Circular Apertures......................................................................... 206
Convert Semi-Diameters to Floating Apertures ........................................................................ 206
Remove All Apertures ............................................................................................................... 206
Replace Vignetting With Apertures ........................................................................................... 206
Fold Mirrors........................................................................................................................................... 207
Add Fold Mirror ......................................................................................................................... 207
Limitations of reversing surfaces following the fold mirror............................................. 207
Delete Fold Mirror ..................................................................................................................... 207
Export Data ........................................................................................................................................... 208
Export IGES/SAT/STEP Solid................................................................................................... 208
Export IGES Line Work ............................................................................................................. 210
Export 2D DXF File ................................................................................................................... 211
Miscellaneous ....................................................................................................................................... 212
Reverse Elements..................................................................................................................... 212
Tilt/Decenter Elements.............................................................................................................. 212
Scale Lens ................................................................................................................................ 212
Make Focal................................................................................................................................ 213
Quick Focus .............................................................................................................................. 213
Convert to NSC Group .............................................................................................................. 214
Replicate Object........................................................................................................................ 215
Ghost Focus Generator............................................................................................................. 215
Performance Test...................................................................................................................... 217
Lock/Unlock All Windows .......................................................................................................... 217
Create Polygon Object .............................................................................................................. 217
Slider ......................................................................................................................................... 218
REPORTS MENU Chapter 9
Introduction ........................................................................................................................................... 219
Surface Data ......................................................................................................................................... 219
System Data ......................................................................................................................................... 219
Prescription Data .................................................................................................................................. 219
Comments on computing element volumes .............................................................................. 220
Report Graphics 4/6.............................................................................................................................. 220
MACROS, EXTENSIONS, AND WINDOWS MENUS Chapter 10
Edit/Run ZPL Macros............................................................................................................................ 223
Refresh Macro List................................................................................................................................ 223
Macro Names........................................................................................................................................ 223
Extensions ............................................................................................................................................ 223
Refresh Extensions List ........................................................................................................................ 224
Extension Names.................................................................................................................................. 224
Listing of open windows........................................................................................................................ 224
SURFACE TYPES Chapter 11
Introduction ........................................................................................................................................... 225
Parameter data ..................................................................................................................................... 225
11
Extra data.............................................................................................................................................. 225
Summary of surface types .................................................................................................................... 225
User defined surfaces ............................................................................................................... 225
Surface types ............................................................................................................................ 226
ABCD .................................................................................................................................................... 228
Alternate Even ...................................................................................................................................... 228
Alternate Odd........................................................................................................................................ 229
Atmospheric Refraction......................................................................................................................... 229
Biconic .................................................................................................................................................. 229
Biconic Zernike ..................................................................................................................................... 230
Binary Optic 1 ....................................................................................................................................... 230
Binary optic coefficients sign conventions................................................................................. 232
Binary Optic 2 ....................................................................................................................................... 232
Binary Optic 3 ....................................................................................................................................... 233
Birefringent In and Birefringent Out ...................................................................................................... 235
Defining the index ..................................................................................................................... 236
Defining the crystal axis orientation .......................................................................................... 236
Determining which ray is traced ................................................................................................ 237
Accounting for phase rotation ................................................................................................... 237
Transmission and other properties of birefringent surfaces ...................................................... 237
Sample ZEMAX files ................................................................................................................. 237
Conjugate.............................................................................................................................................. 238
Coordinate Break .................................................................................................................................. 239
Cubic Spline.......................................................................................................................................... 240
Comments about spline surfaces.............................................................................................. 240
Cylinder Fresnel.................................................................................................................................... 240
Diffraction Grating ................................................................................................................................. 241
Elliptical Grating 1 ................................................................................................................................. 242
Elliptical Grating surface shape................................................................................................. 242
Elliptical Grating 2 ................................................................................................................................. 243
Even Asphere ....................................................................................................................................... 243
Extended Asphere ................................................................................................................................ 244
Extended Cubic Spline.......................................................................................................................... 245
Extended Fresnel.................................................................................................................................. 245
Extended Odd Asphere......................................................................................................................... 246
Extended Polynomial ............................................................................................................................ 247
Extended Toroidal Grating .................................................................................................................... 248
Fresnel .................................................................................................................................................. 249
Generalized Fresnel.............................................................................................................................. 249
Gradient 1 ............................................................................................................................................. 250
Discussion on maximum step size for GRIN surfaces .............................................................. 250
Restrictions on surfaces following GRIN surfaces .................................................................... 251
Gradient 2 ............................................................................................................................................. 251
Gradient 3 ............................................................................................................................................. 251
Gradient 4 ............................................................................................................................................. 252
Gradient 5 ............................................................................................................................................. 252
Gradient 6 ............................................................................................................................................. 253
Gradient 7 ............................................................................................................................................. 254
GRADIUM ......................................................................................................................................... 255
GRADIUM profile file format...................................................................................................... 255
Gradient 9 ............................................................................................................................................. 257
Gradient 10 ........................................................................................................................................... 258
12
Grid Phase ............................................................................................................................................ 258
Using the shear distance........................................................................................................... 258
Phase coefficients sign conventions ......................................................................................... 259
Grid Sag................................................................................................................................................ 259
Importing grid data .................................................................................................................... 259
Bicubic spline vs. linear interpolation ........................................................................................ 260
Suggestions for using the Grid Sag surface.............................................................................. 260
Hologram 1 ........................................................................................................................................... 261
Hologram 2 ........................................................................................................................................... 262
Irregular................................................................................................................................................. 262
Jones Matrix.......................................................................................................................................... 263
Non-Sequential Components................................................................................................................ 263
Odd Asphere......................................................................................................................................... 263
Odd Cosine ........................................................................................................................................... 263
Optically Fabricated Hologram.............................................................................................................. 264
Defining the substrate shape .................................................................................................... 264
Defining the construction optics ................................................................................................ 265
Optimizing the construction and playback optics ...................................................................... 265
Selecting the hologram type...................................................................................................... 265
Paraxial ................................................................................................................................................. 266
Paraxial XY ........................................................................................................................................... 267
Periodic ................................................................................................................................................. 267
Polynomial ............................................................................................................................................ 268
Radial Grating ....................................................................................................................................... 268
Radial NURBS ...................................................................................................................................... 269
Retro Reflect ......................................................................................................................................... 270
Standard ............................................................................................................................................... 270
Modeling an ellipse with the standard surface .......................................................................... 270
Modeling an axicon with the standard surface .......................................................................... 270
Superconic ............................................................................................................................................ 271
Tilted ..................................................................................................................................................... 272
Toroidal ................................................................................................................................................. 272
Toroidal Grating .................................................................................................................................... 273
Toroidal Hologram ................................................................................................................................ 273
Toroidal NURBS ................................................................................................................................... 274
User Defined ......................................................................................................................................... 275
The UDS DLL............................................................................................................................ 275
Refractive and reflective UDS DLLs.......................................................................................... 276
Gradient index UDS DLLs......................................................................................................... 277
Diffractive UDS DLLs ................................................................................................................ 277
Lenslet arrays using UDS DLLs ................................................................................................ 277
User defined surface apodization using DLLs........................................................................... 277
Polarization and coating data using DLLs................................................................................. 277
Error handling and UDS ............................................................................................................ 278
Sample DLLs............................................................................................................................. 278
Variable Line Space Grating ................................................................................................................. 282
Zernike Fringe Phase............................................................................................................................ 282
Zernike Fringe Phase coefficients sign conventions ................................................................. 283
Zernike Fringe Sag ............................................................................................................................... 283
Zernike Standard Phase ....................................................................................................................... 284
Zernike Standard Phase coefficients sign conventions............................................................. 285
Zernike Standard Sag ........................................................................................................................... 285
Zone Plate............................................................................................................................................. 286
13
NON-SEQUENTIAL COMPONENTS Chapter 12
Introduction ........................................................................................................................................... 289
Paraxial data and ray tracing with NSC ................................................................................................ 289
Optimization with NSC .......................................................................................................................... 290
The two methods of using NSC ray tracing .......................................................................................... 290
NSC with ports .......................................................................................................................... 290
NSC without ports ..................................................................................................................... 290
Combining NSC with and without ports..................................................................................... 290
Overview of NSC ray tracing with ports ................................................................................................ 291
The entry port ................................................................................................................ 291
The exit port .................................................................................................................. 291
Getting rays in ............................................................................................................... 292
Tracing rays within the NSC.......................................................................................... 292
Getting rays out ............................................................................................................. 292
Overview of NSC ray tracing without ports ........................................................................................... 292
Getting rays in ............................................................................................................... 293
Tracing rays within the NSC.......................................................................................... 293
Detecting rays ............................................................................................................... 293
Launching rays for analysis........................................................................................... 293
NSC Objects ......................................................................................................................................... 293
Annular Aspheric Lens .............................................................................................................. 297
Annular Volume......................................................................................................................... 297
Annulus ..................................................................................................................................... 298
Aspheric Surface....................................................................................................................... 298
Aspheric Surface 2.................................................................................................................... 299
Axicon Surface .......................................................................................................................... 299
Biconic Lens.............................................................................................................................. 300
Biconic Surface ......................................................................................................................... 301
Making a hyperhemispheric surface.............................................................................. 302
Binary 1 ..................................................................................................................................... 302
Binary 2 ..................................................................................................................................... 303
Binary 2A................................................................................................................................... 303
Cone.......................................................................................................................................... 304
Compound Parabolic Concentrator (CPC)................................................................................ 305
Cylinder Pipe............................................................................................................................. 305
Cylinder Volume........................................................................................................................ 306
Cylinder 2 Pipe.......................................................................................................................... 306
Cylinder 2 Volume..................................................................................................................... 306
Diffraction Grating ..................................................................................................................... 307
Ellipse........................................................................................................................................ 307
Elliptical Volume........................................................................................................................ 307
Even Asphere Lens................................................................................................................... 308
Extended Polynomial Lens........................................................................................................ 308
Extended Polynomial Surface ................................................................................................... 309
Fresnel 1 ................................................................................................................................... 309
Fresnel 2 ................................................................................................................................... 310
Hexagonal Lenslet .................................................................................................................... 312
Hologram Lens.......................................................................................................................... 312
Hologram Surface ..................................................................................................................... 313
Imported .................................................................................................................................... 314
Comments about imported objects................................................................................ 314
Ray tracing accuracy for imported objects .................................................................... 314
Ray tracing speed for imported objects......................................................................... 315
14
Limitations on imported objects..................................................................................... 315
Jones Matrix.............................................................................................................................. 315
Lenslet 1.................................................................................................................................... 315
Lenslet 2.................................................................................................................................... 317
Micro Electro Mechanical System (MEMS)............................................................................... 317
Null Object................................................................................................................................. 318
Odd Asphere Lens .................................................................................................................... 318
Polygon Object.......................................................................................................................... 319
Rectangular Corner................................................................................................................... 319
Rectangle .................................................................................................................................. 320
Rectangular Pipe....................................................................................................................... 320
Rectangular Pipe Grating.......................................................................................................... 320
Rectangular Roof ...................................................................................................................... 321
Rectangular Torus Surface ....................................................................................................... 321
Rectangular Torus Volume ....................................................................................................... 321
Rectangular Volume.................................................................................................................. 322
Rectangular Volume Grating..................................................................................................... 322
Slide .......................................................................................................................................... 322
Sphere....................................................................................................................................... 323
Standard Lens........................................................................................................................... 323
Standard Surface ...................................................................................................................... 324
STL Object ................................................................................................................................ 324
Fast Load/Fast Trace STL objects ................................................................................ 325
Tabulated Faceted Radial ......................................................................................................... 325
Tabulated Faceted Toroid ......................................................................................................... 326
Tabulated Fresnel Radial .......................................................................................................... 327
Toroidal Lens ............................................................................................................................ 327
Toroidal Surface........................................................................................................................ 327
Torus Surface............................................................................................................................ 328
Torus Volume............................................................................................................................ 328
Triangular Corner ...................................................................................................................... 329
Triangle ..................................................................................................................................... 329
User Defined Object.................................................................................................................. 329
Sample Object DLLs ..................................................................................................... 330
Zernike Surface......................................................................................................................... 331
The use of facets .................................................................................................................................. 331
Detectors............................................................................................................................................... 332
Detector Rectangle object......................................................................................................... 332
Comments on coherent data computations................................................................... 333
Detector Surface object............................................................................................................. 334
Comments about detector triangles .............................................................................. 335
Detector Volume object............................................................................................................. 335
Objects as detectors ................................................................................................................. 335
Sources................................................................................................................................................. 336
Placing sources inside objects .................................................................................................. 336
Adding new source types .......................................................................................................... 336
Parameters common to all source objects ................................................................................ 337
Source Diode ............................................................................................................................ 337
Source DLL ............................................................................................................................... 338
Source DLL parameters ................................................................................................ 338
Creating a new Source DLL .......................................................................................... 338
Sample Source DLLs .................................................................................................... 339
Source Ellipse ........................................................................................................................... 339
Source Filament ........................................................................................................................ 340
15
Source File ................................................................................................................................ 340
ASCII vs. binary format files .......................................................................................... 340
Memory requirements for source file objects ................................................................ 341
Restrictions on the number of rays selected ................................................................. 341
Binary Source File format.............................................................................................. 341
ASCII Source File format............................................................................................... 342
Intensity normalization in source files............................................................................ 342
Source Gaussian....................................................................................................................... 342
Source Point.............................................................................................................................. 342
Source Radial............................................................................................................................ 342
Source Ray ............................................................................................................................... 343
Source Rectangle...................................................................................................................... 343
Source Tube.............................................................................................................................. 343
Source Two Angle..................................................................................................................... 343
Source Volume Cylinder ........................................................................................................... 343
Source Volume Ellipse .............................................................................................................. 344
Source Volume Rectangle ........................................................................................................ 344
Object Placement.................................................................................................................................. 344
The object coordinate system ................................................................................................... 344
Reference objects ..................................................................................................................... 345
Placing objects inside of, adjacent to, or overlapping other objects.......................................... 345
Glue distance ............................................................................................................................ 345
Nesting object limits .................................................................................................................. 345
Nesting volumes........................................................................................................................ 346
Nesting surfaces ....................................................................................................................... 346
Refraction and reflection from NSC objects.......................................................................................... 346
Diffraction from NSC objects................................................................................................................. 347
The object properties dialog box........................................................................................................... 348
Type tab .................................................................................................................................... 348
Coating/Scattering tab............................................................................................................... 349
Bulk Scattering tab .................................................................................................................... 349
GRIN tab ................................................................................................................................... 349
Diffraction tab ............................................................................................................................ 350
Sources tab ............................................................................................................................... 350
Defining paths using the Consider Objects list ..................................................................................... 351
Defining a Scatter Towards list ............................................................................................................. 351
User defined apertures ......................................................................................................................... 352
Coating and scatter groups................................................................................................................... 352
Polarization and thin film coatings ........................................................................................................ 352
Coatings on surfaces in contact ................................................................................................ 352
Scattering.............................................................................................................................................. 352
Fraction to scatter and number of scatter rays.......................................................................... 352
Scatter models .......................................................................................................................... 353
No scattering ............................................................................................................................. 354
Lambertian scattering................................................................................................................ 354
Gaussian scattering .................................................................................................................. 354
ABg model scattering ................................................................................................................ 354
BSDF properties............................................................................................................ 355
The ABg BSDF model ................................................................................................... 355
Defining ABg data ......................................................................................................... 356
Scattering rays using the ABg model ............................................................................ 356
User defined scattering ............................................................................................................. 356
Defining an object to use DLL defined surface scattering ............................................. 357
DLL parameters............................................................................................................. 357
16
Creating a new DLL....................................................................................................... 357
Sample surface scattering: TwoGaussian.DLL ............................................................. 357
Bulk scattering ...................................................................................................................................... 357
No Bulk Scattering .................................................................................................................... 357
Angle Scattering........................................................................................................................ 358
DLL Defined Scattering ............................................................................................................. 358
Defining an object to use DLL defined bulk scattering .................................................. 358
DLL parameters............................................................................................................. 358
Creating a new DLL....................................................................................................... 358
Coherence length modeling .................................................................................................................. 359
Defining GRIN media for non-sequential ray tracing ............................................................................ 359
Defining an object to be of gradient index material ................................................................... 359
Discussion on maximum step size for GRIN objects ................................................................ 359
GRIN DLL parameters .............................................................................................................. 359
Creating a new GRIN DLL ........................................................................................................ 359
Sample GRIN DLLs................................................................................................................... 360
Defining DLLs for ray splitting at diffractive surfaces ............................................................................ 361
Defining an object to use the diffraction DLL ............................................................................ 361
Diffraction DLL parameters ....................................................................................................... 361
Creating a new Diffraction DLL ................................................................................................. 361
Ray splitting .......................................................................................................................................... 362
Ray splitting and polarization .................................................................................................... 362
Putting it all together ............................................................................................................................. 363
Performing a Monte Carlo ray trace...................................................................................................... 363
The Ray Trace/Detector Control ............................................................................................... 363
Choosing the number of analysis rays ...................................................................................... 365
Lost energy ............................................................................................................................... 365
Ray database files..................................................................................................................... 366
Ray Database Viewer ........................................................................................................................... 367
The Detector Viewer ............................................................................................................................. 369
The filter string ...................................................................................................................................... 371
Filter string examples ................................................................................................................ 373
Defining Polygon Objects...................................................................................................................... 373
The comment symbol: !............................................................................................................. 374
The coating group name symbol: C .......................................................................................... 374
The invisible symbol: I............................................................................................................... 374
The vertex symbol: V ................................................................................................................ 374
The triangle symbol: T............................................................................................................... 374
The rectangle symbol: R ........................................................................................................... 375
Maximum triangles in the polygon object .................................................................................. 375
Example POB file ...................................................................................................................... 375
Defining STL Objects ............................................................................................................................ 376
Maximum triangles in STL objects ............................................................................................ 376
Example STL files ..................................................................................................................... 376
Special considerations for faceted objects............................................................................................ 376
Comments about DLLs ......................................................................................................................... 377
SOLVES Chapter 13
Introduction ........................................................................................................................................... 379
Curvature: Marginal ray angle............................................................................................................... 380
Curvature: Chief ray angle .................................................................................................................... 380
Curvature: Pick up ................................................................................................................................ 380
Curvature: Marginal ray normal ............................................................................................................ 381
Curvature: Chief ray normal.................................................................................................................. 381
17
Curvature: Aplanatic ............................................................................................................................. 381
Curvature: Element power .................................................................................................................... 381
Curvature: Concentric with surface....................................................................................................... 381
Curvature: Concentric with radius......................................................................................................... 381
Curvature: F number............................................................................................................................. 381
Thickness: Marginal ray height ............................................................................................................. 381
Thickness: Chief ray height................................................................................................................... 382
Thickness: Edge thickness ................................................................................................................... 382
Thickness: Pick up ................................................................................................................................ 382
Thickness: Optical path difference........................................................................................................ 382
Thickness: Position ............................................................................................................................... 382
Thickness: Compensator ...................................................................................................................... 382
Thickness: Center of curvature ............................................................................................................. 383
Glass: Model ......................................................................................................................................... 383
Glass: Pick up ....................................................................................................................................... 383
Glass: Substitute................................................................................................................................... 383
Glass: Offset ......................................................................................................................................... 383
Semi-Diameter: Pick up ........................................................................................................................ 384
Semi-Diameter: Maximum .................................................................................................................... 384
Conic: Pick up ....................................................................................................................................... 384
Parameter: Pick up ............................................................................................................................... 384
Parameter: Chief ray............................................................................................................................. 384
Suggestions for use .............................................................................................................................. 384
OPTIMIZATION Chapter 14
Introduction ........................................................................................................................................... 385
Selecting variables................................................................................................................................ 385
Defining the default merit function ........................................................................................................ 385
Selecting the type of optimization ............................................................................................. 385
Physically significant merit functions......................................................................................... 386
Selecting the pupil integration method.................................................................................................. 387
Rings ......................................................................................................................................... 387
Arms.......................................................................................................................................... 388
Grid ........................................................................................................................................... 388
Delete Vignetted........................................................................................................................ 388
Setting thickness boundary values............................................................................................ 388
Start At ...................................................................................................................................... 389
Assume Axial Symmetry ........................................................................................................... 389
Ignore Lateral Color .................................................................................................................. 389
Relative X Weight...................................................................................................................... 389
Overall Weight........................................................................................................................... 389
Pitfalls with the default merit function.................................................................................................... 389
Optimization with apodized beams ....................................................................................................... 390
Modifying the merit function .................................................................................................................. 390
Int1 and Int2 .............................................................................................................................. 390
Hx, Hy, Px, Py ........................................................................................................................... 390
Operand weights less than zero................................................................................................ 390
Operand weights equal to zero ................................................................................................. 391
Operand weights greater than zero........................................................................................... 391
Merit function definition ............................................................................................................. 391
Optimization operands .......................................................................................................................... 391
Understanding boundary operands....................................................................................................... 421
Using MTF operands ............................................................................................................................ 421
Performing an optimization ................................................................................................................... 422
18
Defining complex operands .................................................................................................................. 423
Optimizing glass selection .................................................................................................................... 424
Using model glasses ................................................................................................................. 424
Using glass substitution ............................................................................................................ 426
Optimizing zoom and multi-configuration lenses .................................................................................. 426
Optimizing extra data ............................................................................................................................ 426
Optimizing tolerance sensitivity............................................................................................................. 426
Optimizing objects in a non-sequential group with sequential rays ...................................................... 427
Optimizing with sources and detectors in non-sequential mode........................................................... 427
Comments about random numbers and NSTR......................................................................... 428
Pixel numbering for detectors ................................................................................................... 428
Optimizing with the IMAE operand........................................................................................................ 428
Using gradient index operands ............................................................................................................. 428
DLTN......................................................................................................................................... 428
LPTD ......................................................................................................................................... 429
User defined operands.......................................................................................................................... 429
Optimizing with ZPL macros ..................................................................................................... 429
Optimizing with externally compiled programs.......................................................................... 431
The global optimum .............................................................................................................................. 433
GLOBAL OPTIMIZATION Chapter 15
Introduction ........................................................................................................................................... 435
Capabilities of ZEMAX .......................................................................................................................... 435
The Global Search algorithm ................................................................................................................ 436
The Hammer algorithm ......................................................................................................................... 437
Optimizing glass selection .................................................................................................................... 437
Using glass substitution ............................................................................................................ 437
Restricting selected glasses...................................................................................................... 438
Summary............................................................................................................................................... 439
TOLERANCING Chapter 16
Introduction ........................................................................................................................................... 441
The basic procedure ............................................................................................................................. 441
Tolerance operands .............................................................................................................................. 442
TRAD: Tolerance on radius....................................................................................................... 443
TCUR: Tolerance on curvature ................................................................................................. 444
TFRN: Tolerance on fringes...................................................................................................... 444
TTHI: Tolerance on thickness ................................................................................................... 444
TCON: Tolerance on conic........................................................................................................ 444
TSDX, TSDY: Tolerance on surface decenters ........................................................................ 445
TSTX, TSTY: Tolerance on surface tilts.................................................................................... 445
TIRX, TIRY: Tolerance on surface TIR ..................................................................................... 445
TIRR: Tolerance on surface irregularity .................................................................................... 445
TEXI: Tolerance on surface irregularity using the Fringe Zernike model .................................. 446
TEZI: Tolerance on surface irregularity using the Standard Zernike model.............................. 447
TPAR: Tolerance on parameter data ........................................................................................ 448
TEDV: Tolerance on extra data values ..................................................................................... 448
TIND: Tolerance on index ......................................................................................................... 448
TABB: Tolerance on Abbe ........................................................................................................ 448
TCMU: Tolerance on coating multiplier..................................................................................... 448
TEDX, TEDY: Tolerance on element decenters ....................................................................... 448
19
TETX, TETY, TETZ: Tolerance on element tilts........................................................................ 449
TOFF: Tolerance off (can be used for comments) .................................................................... 449
TUDX, TUDY, TUTX, TUTY, TUTZ: Tolerance on user defined tilts and decenters................. 449
TNPS, TNPA: Tolerances on non-sequential data.................................................................... 449
TMCO: Tolerance on multi-configuration data .......................................................................... 450
Tolerancing with the Irregular surface type........................................................................................... 450
Tolerance control operands .................................................................................................................. 450
CEDV: Define extra data value compensator ........................................................................... 451
General comments about min and max values on compensators ................................ 451
CMCO: Define multi-configuration operand compensator ........................................................ 451
COMP: Define compensator ..................................................................................................... 451
CPAR: Define parameter compensator..................................................................................... 451
SAVE: Save sensitivity analysis lenses .................................................................................... 451
SEED: Seed the random number generator ............................................................................. 452
STAT: Define statistics.............................................................................................................. 452
TWAV: Test wavelength............................................................................................................ 452
Defining default tolerances ................................................................................................................... 452
Surface tolerances .................................................................................................................... 452
Element tolerances ................................................................................................................... 453
Defining compensators ......................................................................................................................... 453
Preparing the lens for tolerancing ......................................................................................................... 454
Performing the tolerance analysis......................................................................................................... 454
How ZEMAX computes the tolerance analysis ..................................................................................... 457
Evaluating compensators.......................................................................................................... 458
Sensitivity analysis................................................................................................................................ 458
Inverse sensitivity analysis.................................................................................................................... 459
Monte Carlo analysis ............................................................................................................................ 460
Normal statistical distribution .................................................................................................... 460
Uniform statistical distribution ................................................................................................... 460
Parabolic statistical distribution ................................................................................................. 461
User defined statistical distribution............................................................................................ 461
Discussion of Monte Carlo analysis method ............................................................................. 462
Nesting rules for Monte Carlo analysis ................................................................................................. 462
Using Tolerance Scripts........................................................................................................................ 463
Tolerance Script overview......................................................................................................... 463
The Tolerance Script commands .............................................................................................. 463
! ..................................................................................................................................... 463
CEDV ............................................................................................................................ 463
CLEARCOMP................................................................................................................ 463
CLOSEFILE................................................................................................................... 464
CMCO ........................................................................................................................... 464
COMP............................................................................................................................ 464
CPAR ............................................................................................................................ 464
FORMAT ....................................................................................................................... 464
LOADMERIT ................................................................................................................. 464
NOMINAL ...................................................................................................................... 465
OPENFILE..................................................................................................................... 465
OPTIMIZE ..................................................................................................................... 465
PERTURB ..................................................................................................................... 465
REPORT ....................................................................................................................... 466
SAVE............................................................................................................................. 466
UPDATE........................................................................................................................ 467
Tolerance Script example ......................................................................................................... 467
Tolerancing multi-configuration (zoom) lenses ..................................................................................... 467
20
Tolerancing with solves......................................................................................................................... 468
Trouble shooting the tolerance results.................................................................................................. 468
Optimizing for tolerance sensitivity ....................................................................................................... 468
Pitfalls when tolerancing ....................................................................................................................... 468
Summary............................................................................................................................................... 468
MULTI-CONFIGURATIONS Chapter 17
Introduction ........................................................................................................................................... 471
The first step ......................................................................................................................................... 471
Summary of multi-configuration operands ............................................................................................ 471
Defining the number of configurations .................................................................................................. 474
Defining each configuration .................................................................................................................. 475
Adding and removing elements ............................................................................................................ 475
Changing configurations ....................................................................................................................... 475
Optimization with multi-configurations .................................................................................................. 475
Suggestions for organizing multiple-configuration merit functions ............................................ 475
Using solves with multi-configuration data............................................................................................ 476
MCE Pick up solves .................................................................................................................. 476
MCE Thermal pick up solves .................................................................................................... 476
USING GLASS CATALOGS Chapter 18
Introduction ........................................................................................................................................... 477
Specifying which glass catalogs to use................................................................................................. 477
Editing and reviewing glass catalogs .................................................................................................... 477
Description of catalog data ................................................................................................................... 478
Creating a new catalog ......................................................................................................................... 479
Copying or moving glass catalog files................................................................................................... 479
The glass dispersion formulas .............................................................................................................. 480
The Schott formula.................................................................................................................... 480
The Sellmeier 1 formula ............................................................................................................ 480
The Herzberger formula ............................................................................................................ 481
The Conrady formula ................................................................................................................ 481
The Handbook of Optics 1 formula ........................................................................................... 481
The Handbook of Optics 2 formula ........................................................................................... 481
The Extended formula............................................................................................................... 482
General comments of using dispersion formulas ...................................................................... 482
Fitting index data................................................................................................................................... 482
Fitting melt data .................................................................................................................................... 482
Discussion of melt fitting method: ............................................................................................. 483
Defining Transmission Data.................................................................................................................. 484
Modeling gases and liquids................................................................................................................... 484
Finding a glass quickly.......................................................................................................................... 485
Using MIL number glasses ................................................................................................................... 485
Using model glasses............................................................................................................................. 485
Glass catalog sources........................................................................................................................... 486
Obsolete catalog data ........................................................................................................................... 490
THERMAL ANALYSIS Chapter 19
Introduction ........................................................................................................................................... 491
21
Defining temperature and pressure ...................................................................................................... 491
Defining wavelengths............................................................................................................................ 491
Index of refraction computation............................................................................................................. 491
Environmental effects on index for gradient index, MIL number, and model glasses ............... 493
Defining multiple temperature and pressure values.............................................................................. 493
Defining which parameters consider thermal effects ............................................................................ 493
Radius of curvature (CRVT) values .......................................................................................... 494
Thickness (THIC) values........................................................................................................... 494
Parameter values ...................................................................................................................... 494
Extra data values ...................................................................................................................... 494
All other values.......................................................................................................................... 495
Defining multiple environments within a single configuration................................................................ 495
Automatic thermal setup ....................................................................................................................... 495
Adding TCE data................................................................................................................................... 495
Modeling gases and liquids................................................................................................................... 495
Adding thermal index variation data...................................................................................................... 495
Optimizing athermal lenses................................................................................................................... 496
Limitations of thermal analysis.............................................................................................................. 496
POLARIZATION ANALYSIS Chapter 20
Introduction ........................................................................................................................................... 497
Review of polarization concepts ........................................................................................................... 497
The electric field vector ............................................................................................................. 497
Field vs. ray phase conventions................................................................................................ 498
The polarization ellipse ............................................................................................................. 498
Definition of terms ..................................................................................................................... 498
Intensity, I ...................................................................................................................... 498
Phase, P........................................................................................................................ 498
Path length through medium, ....................................................................................... 499
Internal absorption per lens unit, .................................................................................. 499
Internal Transmittance, IT ............................................................................................. 499
Propagation Phase Factors, pc, ps ............................................................................... 499
Amplitude Reflectance, ................................................................................................ 499
Amplitude Transmittance, ............................................................................................. 499
Intensity Reflectance, Rs, Rp ........................................................................................ 499
Intensity Transmittance, Ts, Tp ..................................................................................... 499
Intensity Absorption, As, Ap .......................................................................................... 500
Diattenuation, D............................................................................................................. 500
Coating Phase, Ps, Pp .................................................................................................. 500
Retardance, S ............................................................................................................... 500
Phase Difference Between X and Y, Pxy ...................................................................... 500
Major and Minor semi axis of the Polarization Ellipse, EM, Em .................................... 500
Angle of XY Polarization Ellipse, Ap.............................................................................. 500
Properties of uncoated surfaces ........................................................................................................... 500
Defining coatings in ZEMAX ................................................................................................................. 501
Coating file data syntax............................................................................................................. 501
The MATE data section................................................................................................. 503
The TAPR data section ................................................................................................. 503
The COAT data section................................................................................................. 505
Defining replicated groups of coating layers ................................................................. 506
Defining simple ideal coatings....................................................................................... 506
The IDEAL data section ................................................................................................ 506
The IDEAL2 data section .............................................................................................. 507
The TABLE data section ............................................................................................... 507
22
Adding comments to the coating file ............................................................................. 508
Limits on the amount of coating data ............................................................................ 508
Editing the coating file ................................................................................................... 508
Optimizing coatings with ZEMAX.......................................................................................................... 508
Default materials and coatings supplied with ZEMAX .......................................................................... 509
Specifying coatings on surfaces ........................................................................................................... 510
What ZEMAX does if no coating is specified ........................................................................................ 510
Defining the incident polarization .......................................................................................................... 510
Defining polarizing components............................................................................................................ 511
What ZEMAX can compute using polarization analysis........................................................................ 512
Bulk absorption and transmission ............................................................................................. 512
Modeling birefringent materials............................................................................................................. 512
Modeling frustrated total internal reflection ........................................................................................... 512
Limitations of polarization analysis ....................................................................................................... 514
PHYSICAL OPTICS PROPAGATION Chapter 21
Introduction ........................................................................................................................................... 515
Support for multiple processors ............................................................................................................ 516
Diffraction propagation.......................................................................................................................... 516
Representation of the electric field............................................................................................ 516
The Fresnel number.................................................................................................................. 516
Near and far field........................................................................................................... 517
Angular spectrum propagation .................................................................................................. 517
Fresnel diffraction...................................................................................................................... 519
Selecting the correct propagator ............................................................................................... 520
Fraunhofer diffraction ................................................................................................................ 520
The pilot beam ...................................................................................................................................... 520
Sign conventions for phase data........................................................................................................... 521
Propagating in and out of the Rayleigh range....................................................................................... 522
Separation of X and Y propagation ........................................................................................... 523
Comments about point spacing and sampling ...................................................................................... 524
Propagation through arbitrary optical surfaces ..................................................................................... 524
Propagating through non-sequential surfaces .......................................................................... 525
Accounting for polarization.................................................................................................................... 525
Memory requirements ........................................................................................................................... 525
Defining the initial beam........................................................................................................................ 526
Gaussian Waist ......................................................................................................................... 526
Gaussian Angle......................................................................................................................... 527
Gaussian Size+Angle................................................................................................................ 527
Top Hat ..................................................................................................................................... 527
User defined table ..................................................................................................................... 528
ZEMAX Beam File (ZBF) binary format......................................................................... 528
ZEMAX Beam File (ZBF) ASCII format ......................................................................... 528
User defined DLL program........................................................................................................ 529
Beam DLL parameters .................................................................................................. 529
Creating a new Beam DLL ........................................................................................................ 529
Surface specific settings ....................................................................................................................... 529
Computing Fiber Coupling .................................................................................................................... 530
Where the integral is computed ................................................................................................ 531
Defining the fiber mode ............................................................................................................. 531
Decenters and tilts .................................................................................................................... 531
Choosing the location for the receiving fiber ............................................................................. 531
Algorithm assumptions.......................................................................................................................... 533
23
Samples ................................................................................................................................................ 533
Free space propagation ............................................................................................................ 533
A pinhole aperture..................................................................................................................... 534
A lens array ............................................................................................................................... 534
Talbot imaging........................................................................................................................... 534
Fresnel lens............................................................................................................................... 534
ZEMAX PROGRAMMING LANGUAGE Chapter 22
Introduction ........................................................................................................................................... 535
Creating ZPL macros ............................................................................................................................ 535
Running ZPL macros ............................................................................................................................ 535
Assigning ZPL macros to buttons ............................................................................................. 536
An overview of ZPL............................................................................................................................... 536
ZPL variables ........................................................................................................................................ 536
ZPL functions ........................................................................................................................................ 537
Using the FICL() function ...................................................................................................................... 546
ZPL math operations............................................................................................................................. 546
ZPL logical operators ............................................................................................................................ 546
ZPL string variables and operations ..................................................................................................... 547
ZPL string logical operators .................................................................................................................. 549
ZPL keywords ....................................................................................................................................... 549
APMN, APMX............................................................................................................................ 549
APXD, APYD............................................................................................................................. 550
APTP......................................................................................................................................... 550
ATYP, AVAL.............................................................................................................................. 551
BEEP......................................................................................................................................... 552
COAT ........................................................................................................................................ 552
CLOSE ...................................................................................................................................... 552
CLOSEWINDOW ...................................................................................................................... 552
COLOR ..................................................................................................................................... 553
COMMENT................................................................................................................................ 553
CONI ......................................................................................................................................... 553
COPYFILE ................................................................................................................................ 553
CURV ........................................................................................................................................ 554
DELETE .................................................................................................................................... 554
DELETECONFIG ...................................................................................................................... 554
DELETEFILE............................................................................................................................. 554
DELETEMCO............................................................................................................................ 555
DELETEMFO ............................................................................................................................ 555
DELETEOBJECT ...................................................................................................................... 555
EDVA ........................................................................................................................................ 555
END........................................................................................................................................... 556
EXPORTBMP............................................................................................................................ 556
EXPORTCAD............................................................................................................................ 556
EXPORTJPG ............................................................................................................................ 557
EXPORTWMF........................................................................................................................... 557
FINDFILE .................................................................................................................................. 557
FLDX, FLDY, FWGT, FVDX, FVDY, FVCX, FVCY, FVAN ....................................................... 558
FOR, NEXT ............................................................................................................................... 558
FORMAT ................................................................................................................................... 559
FTYP ......................................................................................................................................... 559
GCRS........................................................................................................................................ 560
GDATE...................................................................................................................................... 560
GETEXTRADATA ..................................................................................................................... 560
24
GETGLASSDATA ..................................................................................................................... 560
GETLSF .................................................................................................................................... 561
GETMTF ................................................................................................................................... 562
GETPSF.................................................................................................................................... 562
GETSYSTEMDATA .................................................................................................................. 563
GETTEXTFILE .......................................................................................................................... 564
GETVARDATA.......................................................................................................................... 565
GETZERNIKE ........................................................................................................................... 566
GLAS......................................................................................................................................... 567
GLENSNAME............................................................................................................................ 567
GOSUB, SUB, RETURN, and END .......................................................................................... 568
GOTO........................................................................................................................................ 568
GRAPHICS ............................................................................................................................... 568
GTEXT ...................................................................................................................................... 569
GTEXTCENT ............................................................................................................................ 569
GTITLE...................................................................................................................................... 570
HAMMER .................................................................................................................................. 570
IF-THEN-ELSE-ENDIF.............................................................................................................. 570
IMA............................................................................................................................................ 571
IMASHOW................................................................................................................................. 571
IMASUM.................................................................................................................................... 572
IMPORT_ED ............................................................................................................................. 572
INPUT ....................................................................................................................................... 572
INSERT ..................................................................................................................................... 572
INSERTCONFIG ....................................................................................................................... 573
INSERTMCO............................................................................................................................. 573
INSERTMFO ............................................................................................................................. 573
INSERTOBJECT....................................................................................................................... 573
LABEL ....................................................................................................................................... 574
LINE .......................................................................................................................................... 574
LOADCATALOG ....................................................................................................................... 574
LOADLENS ............................................................................................................................... 574
LOADMERIT ............................................................................................................................. 575
LOCKWINDOW......................................................................................................................... 575
NEXT......................................................................................................................................... 575
NSTR ........................................................................................................................................ 575
NUMFIELD................................................................................................................................ 576
NUMWAVE ............................................................................................................................... 576
OPEN ........................................................................................................................................ 576
OPTIMIZE ................................................................................................................................. 577
OPTRETURN............................................................................................................................ 577
OUTPUT ................................................................................................................................... 577
PARM........................................................................................................................................ 578
PARn......................................................................................................................................... 578
PARAXIAL................................................................................................................................. 578
PAUSE ...................................................................................................................................... 578
PIXEL ........................................................................................................................................ 579
POLDEFINE.............................................................................................................................. 579
POLTRACE............................................................................................................................... 579
POP........................................................................................................................................... 580
PRINT ....................................................................................................................................... 580
PRINTFILE................................................................................................................................ 581
PRINTWINDOW........................................................................................................................ 581
PWAV........................................................................................................................................ 581
25
QUICKFOCUS .......................................................................................................................... 582
RADI.......................................................................................................................................... 582
RANDOMIZE............................................................................................................................. 582
RAYTRACE............................................................................................................................... 583
RAYTRACEX ............................................................................................................................ 583
READ ........................................................................................................................................ 584
READNEXT............................................................................................................................... 585
READSTRING........................................................................................................................... 585
RELOADOBJECTS................................................................................................................... 585
REM, ! ....................................................................................................................................... 586
REMOVEVARIABLES............................................................................................................... 586
RENAMEFILE ........................................................................................................................... 586
RETURN ................................................................................................................................... 586
REWIND.................................................................................................................................... 586
SAVELENS ............................................................................................................................... 587
SAVEMERIT ............................................................................................................................. 587
SAVEWINDOW......................................................................................................................... 587
SCATTER ................................................................................................................................. 588
SDIA.......................................................................................................................................... 588
SETAIM..................................................................................................................................... 588
SETAIMDATA ........................................................................................................................... 588
SETAPODIZATION................................................................................................................... 589
SETCONFIG ............................................................................................................................. 589
SETDETECTOR ....................................................................................................................... 589
SETMCOPERAND.................................................................................................................... 590
SETNSCPARAMETER ............................................................................................................. 590
SETNSCPOSITION .................................................................................................................. 591
SETNSCPROPERTY................................................................................................................ 591
SETOPERAND ......................................................................................................................... 593
SETSTDD ................................................................................................................................. 593
SETSURFACEPROPERTY ...................................................................................................... 593
SETTEXTSIZE .......................................................................................................................... 595
SETTITLE ................................................................................................................................. 595
SETUNITS ................................................................................................................................ 595
SETVAR.................................................................................................................................... 596
SETVECSIZE............................................................................................................................ 596
SETVIG ..................................................................................................................................... 596
SHOWFILE ............................................................................................................................... 596
SOLVETYPE............................................................................................................................. 597
STOPSURF............................................................................................................................... 598
SUB........................................................................................................................................... 598
SURFTYPE ............................................................................................................................... 598
TELECENTRIC ......................................................................................................................... 599
TESTPLATEFIT ........................................................................................................................ 599
THIC.......................................................................................................................................... 599
TIMER ....................................................................................................................................... 599
UNLOCKWINDOW ................................................................................................................... 600
UPDATE.................................................................................................................................... 600
VEC1, VEC2, VEC3, VEC4....................................................................................................... 600
WAVL, WWGT .......................................................................................................................... 601
XDIFFIA .................................................................................................................................... 601
ZBFCLR .................................................................................................................................... 602
General comments about all ZBF related keywords ..................................................... 602
ZBFMULT.................................................................................................................................. 602
26
ZBFPROPERTIES .................................................................................................................... 602
ZBFRESAMPLE........................................................................................................................ 603
ZBFSHOW ................................................................................................................................ 603
ZBFSUM ................................................................................................................................... 603
ZBFTILT .................................................................................................................................... 604
Example 1 ............................................................................................................................................. 604
Example 2 ............................................................................................................................................. 605
ZEMAX EXTENSIONS Chapter 23
Introduction ........................................................................................................................................... 607
Requirements for writing Extensions .................................................................................................... 607
Establishing the link .............................................................................................................................. 608
Terminating the link............................................................................................................................... 608
Extracting data from ZEMAX ................................................................................................................ 608
The data items ...................................................................................................................................... 609
CloseUDOData ......................................................................................................................... 609
DeleteConfig ............................................................................................................................. 609
DeleteObject ............................................................................................................................. 609
DeleteSurface ........................................................................................................................... 609
ExportCAD ................................................................................................................................ 609
ExportCheck.............................................................................................................................. 610
FindLabel .................................................................................................................................. 610
GetAddress ............................................................................................................................... 610
GetAperture............................................................................................................................... 610
GetAspect ................................................................................................................................. 610
GetBuffer................................................................................................................................... 610
GetComment............................................................................................................................. 611
GetConfig .................................................................................................................................. 611
GetDate..................................................................................................................................... 611
GetExtra .................................................................................................................................... 611
GetField..................................................................................................................................... 611
GetFile....................................................................................................................................... 611
GetFirst ..................................................................................................................................... 611
GetGlass ................................................................................................................................... 611
GetGlobalMatrix ........................................................................................................................ 612
GetIndex.................................................................................................................................... 612
GetLabel.................................................................................................................................... 612
GetMetaFile............................................................................................................................... 612
GetMulticon ............................................................................................................................... 613
GetName................................................................................................................................... 613
GetNSCData ............................................................................................................................. 613
GetNSCMatrix ........................................................................................................................... 613
GetNSCObjectData................................................................................................................... 613
GetNSCParameter .................................................................................................................... 614
GetNSCPosition ........................................................................................................................ 614
GetNSCSettings........................................................................................................................ 614
GetOperand .............................................................................................................................. 614
GetPath ..................................................................................................................................... 614
GetPolState............................................................................................................................... 614
GetPupil .................................................................................................................................... 614
GetPolTrace .............................................................................................................................. 615
GetPolTraceDirect..................................................................................................................... 615
GetRefresh................................................................................................................................ 615
GetSag ...................................................................................................................................... 616
27
GetSequence ............................................................................................................................ 616
GetSerial ................................................................................................................................... 616
GetSettingsData........................................................................................................................ 616
GetSolve ................................................................................................................................... 616
GetSurfaceData ........................................................................................................................ 617
GetSurfaceDLL ......................................................................................................................... 617
GetSurfaceParameter ............................................................................................................... 617
GetSystem ................................................................................................................................ 617
GetSystemAper......................................................................................................................... 618
GetTextFile................................................................................................................................ 618
GetTol ....................................................................................................................................... 618
GetTrace ................................................................................................................................... 619
GetTraceDirect.......................................................................................................................... 619
GetUDOSystem ........................................................................................................................ 620
GetUpdate................................................................................................................................. 620
GetVersion ................................................................................................................................ 620
GetWave ................................................................................................................................... 620
ImportExtraData ........................................................................................................................ 620
InsertConfig............................................................................................................................... 620
InsertObject............................................................................................................................... 621
InsertSurface............................................................................................................................. 621
LoadFile .................................................................................................................................... 621
LoadMerit .................................................................................................................................. 621
MakeGraphicWindow ................................................................................................................ 621
MakeTextWindow...................................................................................................................... 622
NewLens ................................................................................................................................... 622
OpenWindow............................................................................................................................. 622
Optimize .................................................................................................................................... 623
PushLens .................................................................................................................................. 623
PushLensPermission ................................................................................................................ 623
ReleaseWindow ........................................................................................................................ 623
SaveFile .................................................................................................................................... 623
SetAperture ............................................................................................................................... 624
SetBuffer ................................................................................................................................... 624
SetConfig .................................................................................................................................. 624
SetExtra .................................................................................................................................... 624
SetField ..................................................................................................................................... 624
SetFloat..................................................................................................................................... 625
SetLabel .................................................................................................................................... 625
SetMulticon ............................................................................................................................... 625
SetNSCObjectData ................................................................................................................... 625
SetNSCPosition ........................................................................................................................ 625
SetNSCParameter .................................................................................................................... 625
SetNSCSettings ........................................................................................................................ 625
SetOperand............................................................................................................................... 626
SetPolState ............................................................................................................................... 626
SetSettingsData ........................................................................................................................ 626
SetSolve.................................................................................................................................... 626
SetSurfaceData......................................................................................................................... 626
SetSurfaceParameter................................................................................................................ 626
SetSystem................................................................................................................................. 626
SetSystemAper ......................................................................................................................... 627
SetUDOData ............................................................................................................................. 627
SetUDOItem.............................................................................................................................. 627
28
SetVig........................................................................................................................................ 627
SetWave.................................................................................................................................... 627
WindowMaximize ...................................................................................................................... 627
WindowMinimize ....................................................................................................................... 628
WindowRestore......................................................................................................................... 628
Tracing large numbers of rays .............................................................................................................. 628
Step 1: Placing the ray data in the array ....................................................................... 628
Mode 0: Similar to GetTrace ......................................................................................... 628
Mode 1: Similar to GetTraceDirect ................................................................................ 630
Mode 2: Similar to GetPolTrace .................................................................................... 630
Mode 3: Similar to GetPolTraceDirect........................................................................... 631
Mode 5: For tracing non-sequential rays....................................................................... 632
Step 2: Pass the array to ZEMAX ................................................................................. 633
How ZEMAX calls the client.................................................................................................................. 634
Generating a text window ..................................................................................................................... 634
Generating a graphic window ............................................................................................................... 635
ADDRESS................................................................................................................................. 635
BOX........................................................................................................................................... 635
DATA......................................................................................................................................... 635
DATE......................................................................................................................................... 635
FRAME...................................................................................................................................... 635
LENSNAME .............................................................................................................................. 635
LINE .......................................................................................................................................... 636
NOFRAME ................................................................................................................................ 636
PENSTYLE ............................................................................................................................... 636
TEXT ......................................................................................................................................... 636
TITLE ........................................................................................................................................ 636
A sample Extension program................................................................................................................ 636
The DDE_DEMO code.............................................................................................................. 637
A simplified technique for implementing Extensions............................................................................. 637
The ZCLIENT program.............................................................................................................. 637
A sample Extension program using ZCLIENT .......................................................................... 638
Writing Extensions from the PhasePlot template ...................................................................... 638
Getting analysis data using GetTextFile and GetMetaFile........................................................ 639
29
30
Chapter 1 INTRODUCTION
About this document

ZEMAX is available in two different editions: ZEMAX-SE (Standard Edition) and ZEMAX-EE (Engineering
Edition). This manual covers both editions, however certain features are unique to ZEMAX-EE, as noted in the
text. If a feature is available in ZEMAX-EE, but not ZEMAX-SE, the following message will usually be shown in
the manual where the feature is described:
!This feature is only available in the EE edition of ZEMAX.

Note that ZEMAX-EE does everything ZEMAX-SE does, plus additional features. This manual covers both
editions of ZEMAX running on Microsoft operating systems. ZEMAX is a registered trademark of ZEMAX
Development Corporation.
What does ZEMAX do?
ZEMAX is a program which can model, analyze, and assist in the design of optical systems. The interface to
ZEMAX has been designed to be easy to use, and with a little practice it can allow very rapid interactive design.
Most ZEMAX features are accessed by selecting options from either dialog boxes or pull-down menus. Keyboard
shortcuts are provided for quickly navigating or bypassing the menu structure. This manual provides explanations
of the conventions used in ZEMAX, descriptions of procedures, and a description of available features.
What doesn't ZEMAX do?
Neither the ZEMAX program nor the ZEMAX documentation will teach you how to design lenses or optical
systems. Although the program will do many things to assist you in designing and analyzing optical systems, you
are still the designer. The ZEMAX documentation is not a tutorial on optical design, terminology, or methodology.
Technical support available to ZEMAX users includes assistance in using the program, but does not include
tutoring on fundamental optical design principles. If you have little or no experience in optical design, you may
want to read up on any of the many good books available on the subject. The following table lists some (but by
no means all) of the books which will aid in your education.
REFERENCES ON LENS DESIGN
Author Title Publisher

Bass Handbook of Optics McGraw-Hill
Born & Wolf Principles of Optics Pergamon Press
Fischer & Tadic-Galeb Optical System Design McGraw-Hill
Geary, Joseph M. Introduction to Lens Design: With Willmann-Bell
Practical ZEMAX Examples
Hecht Optics Addison Wesley

Kingslake, Rudolph Lens Design Fundamentals Academic Press
Laikin, Milton Lens Design, Third Edition Marcel Dekker
Mahajan, Virenda Aberration Theory Made Simple SPIE Optical Engineering Press
O' Shea, Donald Elements of Modern Optical Design John Wiley and Sons
Rutten and van Venrooij Telescope Optics Willmann-Bell
Shannon, Robert The Art and Science of Optical Cambridge University Press
Design
31 Chapter 1: INTRODUCTION
Author Title Publisher
Smith, Gregory Hallock Practical Computer-Aided Lens Willmann-Bell, Inc.
Design
Smith, Warren Modern Optical Engineering McGraw-Hill
Smith, Warren Modern Lens Design McGraw-Hill
Welford Aberrations of Optical Systems Adam Hilger Ltd.
Welford Useful Optics University of Chicago Press
Most importantly, ZEMAX is no substitute for good engineering practices. No design should ever be considered
finished until a qualified engineer has checked the calculations performed by the software to see if the results are
reasonable. This is particularly important when a design is to be fabricated and significant costs are involved. It
is the engineers responsibility to check the results of ZEMAX, not the other way around.
Learning to use ZEMAX
A tutorial that covers the basic use of ZEMAX and other topics is available as a help file. To use the tutorial,
install ZEMAX, then choose Help, Tutorial from the main menu. The latest tutorial may be downloaded from our
web site, www.zemax.com. Short courses which cover beginner to advanced use of ZEMAX are offered
throughout the year, please check our web site, www.zemax.com, for details.
System requirements
ZEMAX requires the current version of the operating system, 200 Mb (or more) of available hard disk space,
a CD-ROM drive, a minimum display resolution of 1024 x 768 pixels, and Internet and e-mail access for technical
support and program updates. Either a parallel port or a USB port is also required for the key (see below).
The amount of RAM ZEMAX requires varies substantially depending upon the nature of the optical system
being modeled and the type of analysis being done. For conventional imaging systems, 256 Mb of RAM is
adequate and should be considered a minimum requirement. For modeling of very complex objects, extensive
physical optics, or scattering and illumination analysis, substantially more memory may be required for optimal
performance. For these types of analysis, 512 Mb of RAM should be considered the minimum, and better
performance will be achieved with 1 Gb or more of RAM. It is possible to perform physical optics analysis with
ZEMAX using very large arrays that require substantially more RAM than this, see Memory requirements on
page 525 for details.
Multiple processor computers
Many ZEMAX features are designed to run on multiple processors simultaneously. If more than one CPU is
installed in the computer system, ZEMAX will automatically detect and use the additional processors.
Installation procedure
Before you install ZEMAX, please verify that you have the required hardware, defined in the previous section.
Installing the key
ZEMAX is copy-protected using a hardware device called a key. The key only permits one copy of ZEMAX to
run at a time. Attempting to run ZEMAX on more than one computer at a time, or attempting to defeat the intention
of the key is a violation of the owner's copyright, and is punishable by severe criminal and civil penalties.
Attach the key supplied with the program on the parallel printer port at the rear of the computer if you have
been supplied a parallel port type key, or to the USB port if you have been supplied a USB type key. For parallel
port keys, if you do not have a printer, place the key in the port by itself. If you have a printer, disconnect the printer
cable, then attach the key to the port, then attach the printer cable to the key. Your printer, and all other computer
operations, will be unaffected by the presence of the key. You do not need to hook up a printer to the key, but if
you do, the printer may have to be turned on. If you do not have the key installed correctly, ZEMAX will issue an
error message and not allow the program to run.
Chapter 1: INTRODUCTION 32
Running ZEMAX
The ZEMAX installation program will create a new program group. To run ZEMAX, double click on the ZEMAX
icon within this program group.
Policy on the key
ZEMAX is multiple-use protected by a device called a key. The key must be attached to the computer on which
ZEMAX is running. The key prevents more than one copy of ZEMAX from running at one time. The registered
user of the ZEMAX software is allowed to install the ZEMAX software on more than one computer, as long as
there is no possibility that more than one copy of ZEMAX will be running at one time. For example, the user is
allowed to install the ZEMAX software on a computer at work and another computer at home, as long as the two
installations of ZEMAX are never used at the same time. The key must be physically moved from one computer
to another, so that only one copy of ZEMAX can run at one time.
The key supplied with the ZEMAX software is worth the full purchase price of the software. If the key is lost or
stolen, it will not be replaced without payment of the full purchase price. Insure the key as you would any other
business or personal asset of comparable value. Defective keys, or keys that become inoperable, will be replaced
at no charge if the key has a current support contract. Defective keys must be returned before a replacement key
will be provided.
!If the key is lost or stolen, it will not be replaced without payment of the full purchase price.
Definition of support
The purchase of a new license to use ZEMAX includes a period of one year of support. Additional years of
support may be purchased at the time the new license is purchased or at a later date. The support period is
associated with a specific ZEMAX key. To "have support" for a given ZEMAX key means the support period for
that key has not expired. Support consists of two benefits: technical support and program updates.
Technical support is defined as ZEMAX Development Corporation responding to any question that requires
an engineer to resolve. This includes answering telephone calls, e-mail, faxes, or letters regarding how to use
ZEMAX, understanding ZEMAX features, problems with your key, program installation, or reporting suspected
bugs.
Program updates include new features as they are developed, along with any bug fixes recently incorporated
in the program. Updates also include new electronic documentation. Program updates are available for download
from the ZEMAX Development Corporation web site, www.zemax.com, for all users who have support.
Getting technical support
If you have any questions about installing or using ZEMAX, please try the following suggestions to find the
information you require:
1) Look in the Table of Contents to see if there is a chapter or section on the subject.
2) Check the index (at the back of this document) to see if it is referenced.
3) Look for a sample file of the type of lens you are trying to set up, if appropriate.
If you still cannot find the information you require, you may need to call, FAX, e-mail, or write for technical
support. The phone numbers and address are on the title page of this manual. If you telephone, try to be in front
of your computer when you call. Note that technical support is available only if you have support for the ZEMAX
key you are using, as described in the previous section.
Policy on bug fixes
All non-trivial computer programs have bugs. Every effort is made to find and fix all bugs before a new version
is released. However, the program is sufficiently complex such that even a talented team of pre-release program
testers will not find every bug. ZEMAX Development Corporation therefore may provide periodic bug fix releases.
If you find an apparent bug, please report it to us. Try to isolate the exact actions which cause or lead up to
the bug. Try to see if the error is unique to the lens file you are working with. Bugs are usually easier to find if they
also appear on the example files provided with the program.
33 Chapter 1: INTRODUCTION
If you find a bug which causes significant degradation of program performance while you have support, the
bug fix will be provided free of charge. ZEMAX Development Corporation reserves the right to determine what
bugs constitute significant degradation. If you do not have support, you may be required to renew your support
before receiving the bug fix; in this case you will receive the latest program version.
Bug fixes may not be provided if the problem is traced to incompatible, non-standard, or insufficient hardware,
or if you are using an outdated version of the operating system.
Chapter 1: INTRODUCTION 34
Chapter 2 USER INTERFACE
Introduction
This chapter describes the conventions used by the ZEMAX user interface, and describes some keyboard
shortcuts for common window operations. ZEMAX is easy to use once you learn the conventions used throughout
the program. For step by step sample sessions using ZEMAX, see the tutorial (choose Help, Tutorial from the
main ZEMAX menu).
Types of windows
ZEMAX has different types of windows, each of which serves a different purpose. The types of windows are:
The main window: This window has a large blank area, with a title bar, menu bar, and toolbar at the top. The
commands available on this menu bar apply in general to the current optical system as a whole.
Editor windows: There are six different editors: the Lens Data Editor, the Merit Function Editor, the Multi-
Configuration Editor, the Tolerance Data editor, and in ZEMAX-EE only, the Extra Data Editor and the Non-
Sequential Components Editor.
Graphic windows: These windows are used to display graphic data such as layouts, ray fans, and MTF plots.
Text windows: Text windows are used to display text data such as prescription data, aberration coefficients,
and numerical data.
Dialogs. Dialogs are pop-up windows which cannot be resized. Dialogs are used to change options or data
such as field angles, wavelengths, apertures, and surface types. They are also used extensively to change
options on graphic and text windows, such as for changing the number of rays on a layout plot.
All windows can be moved or resized (except dialogs) using the standard mouse or keyboard commands. If
you are not familiar with these procedures, see any good book on using Windows or the Windows documentation.
Main window operations
The main window bar has several menu headings. Most menu headings correspond to a chapter of the same
name later in this manual. See the specific chapter for detailed instructions on using the features available from
each menu option. These are the menu bar headings:
File: Used to open, close, save, and rename (save as) lens files.
Editors: Used to invoke (make appear) any of the editor windows.
System: Used to define properties of the optical system as a whole.
Analysis: Analysis groups together features which do not change the lens data, but compute numerical or
graphical results from the lens data. These features include layouts, ray fans, spot diagrams, diffraction
calculations, and more.
Tools: Tools are features which can change the lens data or perform advanced computations on the system
as a whole. These include optimization, tolerancing, test plate fitting, and more.
Reports: Reports are used to document a lens design. These features include system data summaries, surface
data summaries, and report graphics.
Macros: Used to edit and run ZPL macros.
Extensions: Provides access to ZEMAX Extensions, which are compiled features added to ZEMAX.
Window: Selects from the list of currently open windows which window to move to the front of the display.
Help: Provides access to the program documentation.
Most of the commonly used menu options have keyboard equivalents which may be faster to use. For example,
typing Ctrl-Q will quit ZEMAX. The keyboard shortcuts are listed along side the menu options.
A handy shortcut key for switching between windows within the main window is Control-Tab. This will advance
control to the next window in the list of windows maintained by ZEMAX.
35 Chapter 2: USER INTERFACE

The main window also displays a row of buttons just below the menu bar. This row of buttons is used to quickly
select common operations. All of the buttons represent functions available on the menus. The button definitions
may be changed on the File, Preferences dialog box, on the button bar tabs. The three letter mnemonics used
by the buttons are defined on that dialog box tab. A screen resolution of 1024 x 768 or higher is recommended
for displaying buttons.
Editor windows operations
The editor windows are used primarily for entering lens and merit function data. Each editor is similar to a
spreadsheet, with rows and columns. The intersection of a row and a column forms a cell. If the editor is the active
window (the one with a highlighted title bar) then one cell will be shown highlighted or in reverse colors. This cell
is called the active cell, and it has what is called the input "focus". The inverted color of the cell is called the cursor,
although it is not a cursor in the usual sense.
Having the input focus means that any data typed on the keyboard will be sent to the active cell. The exceptions
are control commands such as cursor keys or control key combinations, which are sent directly to the main
window. To modify the data in the active cell, type in the new data and press the Enter key when finished.
To add an incremental value to a cell, type a plus sign and then the increment, then Enter. For example, to
change a 12 to 17, type "+5" and Enter. The "*" multiply and "/" divide symbols also work. To subtract a value,
type a minus sign and a space followed by the value to subtract. The space is required to distinguish between
subtraction and entering a negative number.
To modify a portion of the contents of a cell without retyping the entire value, first highlight the cell, then press
the back space or the F2 function key. The left cursor, right cursor, home, and end keys can then be used to
navigate within the cell for editing. The mouse may also be used to select and replace portions of text. Once
changes have been made to the data in the cell, pressing Enter will complete editing and leave the cursor on that
cell. Pressing up or down cursors will also complete editing, and move the cursor accordingly. Pressing Tab or
Shift-Tab will complete editing and move the cursor right or left.
To abort editing of any cell, press the escape key.
The left, right, up and down cursor keys will move the cursor accordingly. Pressing the control key and the left,
right, up and down keys simultaneously will move the editor display one page at a time in the appropriate direction.
The Tab and Shift-Tab keys also move the cursor right and left.
The page up and page down keys move the cursor one screen at a time. Control-page up and control-page
down move the cursor to the top and bottom of the current column. The home and end keys will move the cursor
to the first column first row and first column last row, respectively. Control-home and control-end will move the
cursor to first and last row of the last column, respectively.
Clicking once in any cell will move the cursor to that cell. Double clicking on a cell will invoke the solve dialog
box for that cell if one exists. Right mouse clicking will also invoke the solve dialog box for that cell if one exists.
Graphic windows operations
Graphic windows have the following menu items:
Update: This recomputes the data displayed in the window with the current settings.
Settings: Invokes a dialog box which controls the options for that window.
Print: Prints the contents of the window.
Window: The submenus under Window are:
Annotate: See Using the annotation feature on page 38 for details. The items under Annotate are:
Line: Draws a single line on the graphic window.
Arrow: Draws a line segment with a directional arrow at the end point.
Text: Prompts for and then draws text on the graphic window.
Box: Draws a box on the graphic window.
Edit: Allows extended editing of annotations.
Chapter 2: USER INTERFACE 36

Copy Clipboard: Copies the window contents to the Windows clipboard. See Using the Windows clipboard
on page 42 for details.
Export: Exports the displayed graphic as a Windows Metafile, BMP, or JPG file. JPG files support high,
medium, and low quality. Medium quality typically produces acceptable image quality with a significant
reduction in file size.
Lock: If selected, the window will be converted into a "static" window whose data cannot change. The locked
window contents may be printed, copied to the clipboard, or saved in a file. The application for this feature
is for comparing the results of different lens files. Once a window is locked, it cannot be updated, and so
any new lens files which are subsequently loaded may be analyzed and compared to the locked window
results. Once the window is locked, it cannot be updated until unlocked. See also Lock/Unlock All
Windows on page 217.
Unlock: Unlocks a window previously locked. See also Lock/Unlock All Windows on page 217.
Clone: This selection will open a new window whose settings and displayed data are initially identical to the
current window. This feature is useful for creating a new window based upon the settings in the original
window. The cloned window acts like any other window after it is created, so it may be updated or have its
settings changed independent of the first window.
Aspect Ratio: The aspect ratio may be selected to be 3 x 4 (height x width) which is the default, or 3 x 5, 4
x 3, or 5 x 3. The latter two are taller than they are wide. The default aspect ratio may be set on the Graphics
tab of the File, Preferences dialog box.
Active Cursor: The active cursor displays the value of the coordinates the cursor is currently pointing to in
the title bar of the window when the cursor covers an "active" region of the displayed graphic. On most X
vs. Y type graphs, the meaning of the displayed values is obvious. On some graphics, such as the 3D
Layout, the displayed image is a projection of a 3D object on a 2D plane. The projection of the image renders
the coordinate data displayed by the active cursor less meaningful if the image has been rotated. Not all
graphics support the active cursor. The active cursor is by default "off" but can be turned on and off by
choosing this menu option. The active cursor can be set to automatically be on or off when a new graphic
window is created on the Graphics tab of the File, Preferences dialog box.
Configuration: Selects either the current or any specific configuration for the data to be displayed. The
default is current which means the data in the window is displayed for the active configuration. Some
analysis windows, such as the 3D layout, Report Graphics, and Spot Diagrams, independently allow
selection of one or more configurations to be displayed on the settings dialog box. That option, if present
on the settings box, will override this menu option, and this option will be grayed out.
Overlay: Provides a list of all open graphical windows; any of these may be selected for overlaying with the
currently displayed data. The overlay feature is useful for comparing two similar graphs or layouts to detect
small changes.
Text: Displays a text listing of the data in a new window. Not all graphics windows support this option.
Zoom: Controls the "zooming" in on smaller areas of the graphic. See "Using pan and zoom" below. The
submenus under the Zoom menu are:
In: Zooms in by 2X centered on the current center position.
Out: Zooms out by 2X.
Last: Restores the previous zoom setting.
Unzoom: Restores the complete view of the graphic.
There are two mouse shortcuts available when using graphic windows:
Double clicking anywhere in the body of graphic window will update the contents. This is the same as selecting
Update.
Clicking with the right mouse button anywhere in the body of a graphic window will invoke the Settings dialog
box.

Using the annotation feature
There are several ways to annotate graphics windows with custom lines, arrows, boxes, and text notes. One
way is to choose Window, Annotate from any graphic window menu bar, then select Line, Arrow, Text, or Box. To
draw a line, choose the Line menu item, then click at the starting position of the line, holding down the left mouse
button, then drag the crosshairs to the ending position of the line, then release the mouse button. A similar
procedure will draw an arrow or box on the screen.
To add text to the window, select Window, Annotate, Text. A text entry dialog box will appear. Type in the
desired text, then click "OK", then click on the desired location for the text on the window display area.
For more precise control over the exact locations of lines and text, as well as control over the text font, and
the ability to add more complex annotations, choose Annotate, Edit from the graphic menu. This will invoke the
annotation editor, which consists of a text editor and several buttons. There is also a single checkbox for enabling
or disabling annotations for the graph.
The text edit field is used to define the annotations to be applied to the graph. To insert a new line, use the
keyboard command Ctrl-Enter.
There are several supported commands, each with a specific syntax:
TEXT "string" x y angle fontx fonty
The TEXT command will write any text within the double quotes at the location specified by x and y, at an angle
in degrees given by angle, using a fixed font whose width and height are given by fontx and fonty. The coordinates
are in normalized units, where the left edge of the graph has a coordinate of x = 0.0, the right edge x = 100.0, the
bottom edge is y = 0.0, and the top edge is y = 100.0. The origin is the lower left hand corner of the screen. The
"angle" value is in degrees. The fontx and fonty units are in arbitrary units. The angle, fontx, and fonty values may
all be left undefined, and then default values will be used.
LINE x1 y1 x2 y2
The LINE command draws a straight line from x1, y1 to x2, y2. The units and coordinate system are as
described for x and y in the description of the TEXT command.
ARROW x1 y1 x2 y2 SIZE
The ARROW command draws a one-headed arrow pointing from x1, y1 to x2, y2. The units and coordinate
system are as described for x and y in the description of the TEXT command. If Size is 1.0 or is omitted, the arrow
head is drawn at the default size. To scale the default arrow head size, use any other floating point value for size.
For example, a size value of 2.0 will make the head twice as large as the default, a value of 0.5 will make it half
the default size, etc.
BOX x1 y1 x2 y2
The BOX command draws a box with opposite corners from x1, y1 to x2, y2. The units and coordinate system
are as described for x and y in the description of the TEXT command.
ELLIPSE x y rx ry
ELLIPSE draws an ellipse centered at x and y with x half width of rx and half height of ry. If ry is the same as
rx or if ry is omitted, a circle of radius rx results.
There are also several buttons on the annotation dialog box:
OK: Accepts the annotations as displayed and exits.
Cancel: Reverts back to the last annotations and exits.
Save: Opens a "Save As" type box where the file name may be specified. The annotations are saved in the
user named file.
Load: Opens a "Load: type box where the file name to load may be selected. The loaded file contains the
annotations to use.
Reset: Clears the edit buffer.
Help: Invokes the help function.

Using pan and zoom
Any graphic window may be panned (scrolled left, right, up and down) or zoomed (magnified). To activate the
pan and zoom, choose any ZEMAX graphic window, then click the left mouse button and hold the button down
for 1/2 second anywhere in the window. The cursor will change from an arrow to a cross. Now drag the mouse
down and right to define a rectangle of the desired size which covers the area to zoom in on. Now, let go of the
left mouse button. The selected area will be magnified to fill the window, while maintaining the aspect ratio of the
plot.
To pan, drag the scroll bars on the sides of the window. Panning is only allowed after a graphic has been
zoomed.
To restore the graphic to the original size, select Zoom, Unzoom from the Graphic window menu bar.
Keyboard shortcuts are supported when using pan and zoom, see Summary of useful shortcuts on page 40
for details.
Text windows operations
Text windows have the following menu items:
Update: This recomputes the data displayed in the window with the current settings.
Settings: Invokes a dialog box which controls the options for that window.
Print: Prints the contents of the window.
Window: There are five submenus under Window:
Copy Clipboard: Copies the window contents to the Windows clipboard. See Using the Windows clipboard
on page 42 for details.
Save Text: Saves the displayed text data in an ASCII file.
Lock: If selected, the window will be converted into a "static" window whose data cannot change. The locked
window contents may be printed, copied to the clipboard, or saved in a file. The application for this feature
is for comparing the results of different lens files. Once a window is locked, it cannot be updated, and so
any new lens files which are subsequently loaded may be analyzed and compared to the locked window
results. Once the window is locked, it cannot be updated until unlocked. See also Lock/Unlock All
Windows on page 217.
Unlock: Unlocks a window previously locked. See also Lock/Unlock All Windows on page 217.
Clone: This selection will open a new window whose settings and displayed data are initially identical to the
current window. This feature is useful for creating a new window based upon the settings in the original
window. The cloned window acts like any other window after it is created, so it may be updated or have its
settings changed independent of the first window.
Configuration: Selects either the current or any specific configuration for the data to be displayed. The
default is current which means the data in the window is displayed for the active configuration.
There are two mouse shortcuts available when using text windows:
Double clicking anywhere in the body of graphic window will update the contents. This is the same as selecting
Update.
Clicking with the right mouse button anywhere in the body of a graphic window will invoke the Settings dialog
box.
Dialog operations
Most dialog boxes are self explanatory. Usually there are buttons for OK and Cancel which are common in
Windows dialog boxes.
Analysis features, such as ray fan plots, have dialog boxes which allow selection of various options. All of these
dialog boxes have six buttons:
OK: Causes the window to recompute and redisplay the data with the currently selected options.

Cancel: Reverts to those options selected before the dialog was invoked, and does not recompute the data.
Save: Saves the currently selected options for use as the default settings in the future. See below.
Load: Loads the default values that were previously saved. See below.
Reset: Resets the default values to the "factory default" settings.
Help: Calls the ZEMAX help system. The help page displayed will contain information about the options on the
active dialog box.
The Save and Load buttons have a dual functionality. When Save is pressed, the settings are saved for the
current lens file, as well as all lenses without specific settings of their own. For example, if lens "A" is loaded, and
the number of rays on the layout plot for "A" is set to 15, and then Save is pressed, the new default number of
rays for "A" will be 15. Also, any new lenses created, or old lenses without their own settings, will also use 15 as
the new default. Now suppose that a subsequent lens "B" is loaded, and the number of rays is changed to 9, and
then Save is pressed again. For "B", and all lenses that have no specific settings of their own, 9 will be the new
default. However, the original "A" lens will retain the setting of 15 because it has it's own private settings.
The Load button works in a similar way. When Load is pressed, ZEMAX checks to see if that specific lens has
any previous settings saved. If it does, then those settings are loaded. If not, then ZEMAX loads the last saved
settings for all lenses. From the previous example, a new lens "C" will load up the settings of 9 rays, since that
was the last setting saved; while lenses "A" and "B" would load up settings of 15 and 9, respectively, because
they have private settings.
The Save and Load private settings are stored in a file with the same name as the lens file, except the extension
is CFG rather than ZMX. No lens data is stored in the CFG file, just the user defined settings for each analysis
feature.
The other options that appear on dialog boxes can be selected using either the keyboard or the mouse. For
keyboard control, use the Tab and Shift-Tab key combinations to move from option to option. The space bar can
be used to toggle the currently selected checkbox. The cursor keys can be used select items within a drop down
list. Pressing the first letter of an entry in a drop down box will also select that entry.
Aborting long computations
Certain ZEMAX tools may require relatively long computation times. For example, the optimization, global
optimization, and tolerancing tools may run from several seconds to many days. To terminate the execution of
these tools, there is a "Terminate" button displayed which can be clicked. After the terminate button is clicked,
ZEMAX gracefully exits the computation and returns control to the main program. Usually, the results of the
computation are not available, and are not displayed.
Some analysis features, such as MTF and the image analysis feature, will run for long periods of time in some
circumstances. For example, large MTF grids or large image analysis ray densities require longer computation
times. However, analysis features do not display a status box or terminate button, because analysis features
display their output directly in a window. For this reason, the keyboard command "Escape" is used to terminate
lengthy analysis computations. There is no mouse equivalent for this feature; only the Escape key is used.
The Escape key will terminate MTF, PSF, encircled energy, and other diffraction computations. If the Escape
key is pressed, control will return to the main program (it may take 1 or 2 seconds) and the data displayed in the
window will be invalid. For the image analysis feature, the escape key will terminate the tracing of new rays,
however, rays that have already been traced will be displayed, and data from those rays is accurate, if incomplete.
Summary of useful shortcuts
The following tables summarize useful shortcuts, using both the keyboard and the mouse.
ZEMAX SHORTCUTS
Action Result
Ctrl + Tab Moves the input focus from window to window.

Shift + Ctrl + Tab Moves the input focus from window to window (backwards).

Action Result
Ctrl + letter Shortcut for many ZEMAX tools and functions. For example, Ctrl-L invokes the 2D
layout plot. All the keyboard shortcuts are listed alongside the menu items.
F1..F10 Function keys are also used as shortcuts for many functions. All the keyboard
shortcuts are listed alongside the menu items.
Backspace When an editor window has the input focus, the highlighted cell can be edited using
the backspace key. Once the backspace key is pressed, the mouse and left-right
cursors can be used for editing.
Left double click If the mouse is positioned over any graphic or text window, double clicking will
cause the window contents to be recalculated and redrawn. This is the same as
selecting Update. For editor windows, invokes the solves dialog box.
Right click If the mouse is positioned over any graphic or text window, right clicking will invoke
the settings box for that window. This is the same as selecting Settings. For editor
windows, invokes the solves dialog box.
Tab Moves to the next cell in the editor windows, or the next field in the dialog boxes.
Shift + Tab Moves to the previous cell in the editor windows, or the previous field in the dialog
boxes.
Home/End On spreadsheet editors, moves to the top left/bottom left of the current editor.
On text windows, moves to the top/bottom of the window.
On graphic windows, Zoom in/out.
Ctrl + Home/End On spreadsheet editors, moves to the top right/bottom right of the current editor.
On graphic windows, Last Zoom/Unzoom.
Cursor Key (Left, Right, Up, On spreadsheet editors, moves one cell at a time.
Down) On 3D graphic windows, rotates the view around X and Y.
Ctrl + Cursor Key (Left, On spreadsheet editors, moves one screen at a time.
Right, Up, Down) On graphic windows, Pan left, right, up, down.
Page Up/Down On spreadsheet editors, moves up or down one screen at a time.
On 3D graphic windows, rotates the view around Z.
Ctrl + Page Up/Down On spreadsheet editors, moves to top or bottom of column.
WINDOWS SHORTCUTS
Action Result
Alt + Tab Toggles between all currently running applications. Extremely useful for quickly moving
between ZEMAX and other running applications.
Ctrl + Esc Invokes the Windows task list. The task list allows selection of another running
application.
Alt Selects the top menu bar of the current application.
Alt + letter Selects the menu option with that letter. For example, Alt + F selects the File menu.
Tab Moves to the next option or field.
Shift + Tab Moves to the previous option or field.
Space bar Toggles selected checkboxes on or off.

Action Result
Enter Equivalent to pressing the highlighted or default button in a dialog box.

letter Pressing the underlined letter of a word in a drop down box will select that entry in the
drop down box.
Using the Windows clipboard

One of the most useful features in Windows is the clipboard. The clipboard is "holding area" for graphics and
text. The advantage to using the clipboard is that virtually all Windows programs can either import or export to
the clipboard.
Since ZEMAX is primarily used to generate graphical and text data, ZEMAX only supports exporting to the
clipboard. Once the desired data has been copied to the clipboard, it is easy for another application, such as a
word processor, graphics editor, or desktop publishing system to retrieve the data. For example, the graphics
within this manual were generated within ZEMAX, copied to the clipboard, and then pasted into a desktop
publishing program from the clipboard.
To get ZEMAX graphics and text into the clipboard, choose the graphic or text window desired, then select
Window, Copy Clipboard. Nothing will appear to happen (the data transfer is extremely fast), however the data
will now be available to other applications.
To now get the clipboard data into a document processing application, run that application, and choose Paste,
usually from the Edit menu of that application. See the documentation for that application for details.
Some Windows applications cannot import ZEMAX graphics, even if they appear correctly in the Windows
Clipboard Viewer. In this case, the work around is to use the "Export Metafile" option described in Graphic
windows operations on page 36. Once the metafile is created, most Windows applications can import the
graphic.
Another way of getting ZEMAX graphics into other applications is to perform a screen capture which creates
a bitmap image of either the entire screen or any single window. To capture the entire screen as a bitmap image,
press Ctrl-Print Screen on the keyboard. To capture a single window, select that window and press Alt-Print
Screen. Once the screen bitmap has been captured, the image can usually be pasted into another application
using Ctrl-V or Edit, Paste, depending upon the application.
ZEMAX file types by extension
The following table lists some of the extensions used by ZEMAX to indicate different types of data files.
ZEMAX FILE TYPES BY EXTENSION
Extension Description
AGF ASCII Glass Format. These are ASCII files which contain the data for the glass catalogs. An
ASCII format is used so that additional data may be incorporated in future releases of ZEMAX.
ANN Annotation files. These binary files store user defined annotations for ZEMAX graphics.
BGF Binary Glass Format. To accelerate the loading of AGF files, ZEMAX converts AGF files to BGF,
which are the version specific binary images of the glass catalogs.
C C language source code files.
CFG Configuration file. ZEMAX.CFG is the main configuration file which describes the user-selected
options on the environment dialog box. There are numerous other files that may be present
which end in CFG, such as RAY.CFG, which contains the user-defined defaults for the ray fan
plot control screen.
DLL Dynamic Link Library files. These files are externally compiled programs linked into ZEMAX at
run time. Typically, DLLs are used to implement user-defined surfaces.

Extension Description
EXE Executable file. This includes the main program, ZEMAX.EXE and other executables and
utilities used by ZEMAX.
IGS, IGES Initial Graphics Exchange Specification (IGES) file.
IMA Image file, used by the image analysis feature.
NOT Element drawing note files.
POB Polygon object definition file for NSC object.
SES These files are the "session" files, which define the screen configuration for each saved ZMX
lens file.
STP, STEP STEP files are CAD format files for describing solid model data.
TPD Test Plate Data files. These contain test plate lists supplied by various optical fabricators, and
are used in the test plate fit feature.
ZBF ZEMAX Beam Files. These are used by the Physical Optics Propagation feature.
ZMF ZMF are compilations of ZMX format files which are used to define lenses in the stock lens
catalogs.
ZMX ZMX files are the files used to store lens data. The ZMX file is an ASCII file which contains the
complete description of the lens, including apertures, wavelengths, prescription data, and the
merit function.
ZPL ZEMAX Programming Language macros.

Chapter 3 CONVENTIONS AND DEFINITIONS
Introduction
This chapter describes conventions and defines terminology used throughout this manual. Most of the
conventions and terms ZEMAX uses are common in the optics industry, however there may be some important
differences.
Active configuration
The active configuration is the configuration currently being displayed in the lens data editor. For details see
the Chapter "Multi-Configurations".
Angular magnification
The ratio of the paraxial image space chief ray angle to the paraxial object space chief ray angle. The angles
are measured with respect to the paraxial entrance and exit pupil locations.
Apodization
Apodization refers to the uniformity of illumination in the entrance pupil of the system. By default, the pupil is
always illuminated uniformly. However, there are times when the pupil should have a non-uniform illumination.
For this purpose, ZEMAX supports pupil apodization, which is a variation of amplitude over the pupil.
Three types of pupil apodization are supported: uniform, Gaussian, and tangential. For each type (except
uniform), an apodization factor determines the rate of variation of amplitude in the pupil. See the discussion on
apodization types and factors in the chapter "System Menu".
ZEMAX also supports user defined apodizations, which may be placed on any surface. Surface apodizations
behave differently than pupil apodizations, because surfaces need not be located at a pupil. For more information
on surface apodizations, see User Defined on page 275.
Back focal length
ZEMAX defines the back focal length as the distance along the Z axis from the last surface made of glass to
the paraxial image plane for the object at infinite conjugates. If no surfaces are made of glass, the back focal
length is the distance from surface 1 to the paraxial image plane for the object at infinite conjugates.
Cardinal planes
The term cardinal planes (sometimes called cardinal points) refers to those special conjugate positions where
the object and image surfaces have a specific magnification. The cardinal planes include the principal planes,
where the lateral magnification is +1, the anti-principal planes, where the lateral magnification is -1, the nodal
planes, where the angular magnification is +1, the anti-nodal planes, where the angular magnification is -1, and
the focal planes, where the magnification is 0 for the image space focal plane and infinite for the object space
focal plane.
Except for the focal planes, the cardinal planes are conjugates with each other, that is, the image space
principal plane is conjugate with the object space principal plane, etc. If the lens has the same index in both object
space and image space, the nodal planes are identical to the principal planes.
ZEMAX lists the distance from the image surface to the various image space planes, and lists the distance
from the first surface to the various object space planes.
Chief ray
If there is no vignetting, and there are no aberrations, the chief ray is defined to be the ray that travels from a
specific field point, through the center of the entrance pupil, and on to the image plane. Note that without vignetting
or aberrations, any ray passing through the center of the entrance pupil will also pass through the center of the
stop and the exit pupil.
When vignetting factors are used, the chief ray is then considered to be the ray that passes through the center
of the vignetted pupil, which means the chief ray may not necessarily pass through the center of the stop.
45 Chapter 3: CONVENTIONS AND DEFINITIONS

If there are pupil aberrations, and there virtually always are, then the chief ray may pass through the center of
the paraxial entrance pupil (if ray aiming is off) or the center of the stop (if ray aiming is on), but generally, not both.
If there are vignetting factors which decenter the pupil, then the chief ray will pass through the center of the
vignetted entrance pupil (if ray aiming is off) or the vignetted stop surface (if ray aiming is on).
The common convention used is that the chief ray passes through the center of the vignetted pupil, while the
principal ray passes through the center of the unvignetted stop. ZEMAX never uses the principal ray. Most
calculations are referenced to the chief ray or the centroid. Note the centroid reference is generally superior
because it is based upon the aggregate effect of all the rays that actually illuminate the image plane, and not on
the arbitrary selection of one ray which is "special".
Coordinate axes
The optical axis is the Z axis, with the initial direction of propagation from the object being the positive Z
direction. Mirrors can subsequently reverse the direction of propagation. The coordinate system is right handed,
with the sagittal X axis being oriented "into" the monitor on a standard layout diagram. The tangential Y axis is
vertical.
The direction of propagation is initially left-to-right, down the positive Z axis. After an odd number of mirrors
the beam physically propagates in a negative Z direction. Therefore, all thicknesses after an odd number of
mirrors should be negative.
Diffraction limited
The term diffraction limited implies that the performance of an optical system is limited by the physical effects
of diffraction rather than imperfections in either the design or fabrication. A common means of determining if a
system is diffraction limited is to compute or measure the optical path difference. If the peak to valley OPD is less
than one quarter wave, then the system is said to be diffraction limited.
There are many other ways of determining if a system is diffraction limited, such as Strehl ratio, RMS OPD,
standard deviation, maximum slope error, and others. It is possible for a system to be considered diffraction limited
by one method and not diffraction limited by another method.
On some ZEMAX plots, such as the MTF or Diffraction Encircled Energy, the diffraction limited response is
optionally shown. This data is usually computed by tracing rays from a reference point in the field of view. Pupil
apodization, vignetting, F/#s, surface apertures, and transmission may be accounted for, but the optical path
difference is set to zero regardless of the actual (aberrated) optical path.
For systems which include a field point at 0.0 in both x and y field specifications (such as 0.0 x angle and 0.0
y angle), the reference field position is this axial field point. If no (0, 0) field point is defined, then the field
coordinates of field position 1 are used as the reference coordinates instead.
Edge thickness
ZEMAX uses two different definitions for the term "edge thickness". Usually, the edge thickness is computed
for a specific surface by:
Ei = Zi + 1 Zi + Ti
where Z i is the sag of the surface at the +y semi-diameter of the surface, Z i + 1 is the sag of the next surface
at the +y semi-diameter of the next surface, and T i is the axial thickness of the surface. Note that the edge
thicknesses are computed accounting for the sag at the respective semi-diameter of each surface, which in
general are different.
Note also that edge thickness is normally computed for the +y radial aperture, which may be inadequate if the
surface is not rotationally symmetric, or if surface apertures have been placed upon either of the surfaces.
The exception to this rule is when computing edge thickness solves. Because the edge thickness solve can
change the center thickness, the edge thickness solve can change where rays strike the following surface, which
in turn means the semi-diameter of the next surface may change. If the semi-diameter of the next surface is used
in the edge thickness computation, an "infinite loop" or circular definition may occur.
Chapter 3: CONVENTIONS AND DEFINITIONS 46

For this reason, edge thickness solves compute the edge thickness strictly at the semi-diameter of the first
surface, for both surfaces. The semi-diameter of the second surface is never used, although the curvature or
shape of the surface is used.
Effective focal length
The distance from the rear principal plane to the paraxial image plane. This is calculated for infinite conjugates.
Principal plane calculations are always based upon paraxial ray data. The effective focal length is always
referenced to an index of refraction of 1.0, even if the image space index is not unity.
Entrance pupil diameter
The diameter in lens units of the paraxial image of the stop in object space.
Entrance pupil position
The paraxial position of the entrance pupil with respect to the first surface in the system. The first surface is
always surface 1, not the object surface, which is surface 0.
Exit pupil diameter
The diameter in lens units of the paraxial image of the stop in image space.
Exit pupil position
The paraxial position of the exit pupil with respect to the image surface.
Extra data
Extra data values are used to define certain non-standard surface types. For example, extra data values are
used to define the phase of diffractive optic surfaces such as the Binary 1 surface type. For a complete discussion
of the extra data values see Extra data on page 225.
Field angles and heights
Field points may be specified as angles, object heights (for systems with finite conjugates), paraxial image
heights, or real image heights. Field angles are always in degrees. The angles are measured with respect to the
object space z axis and the paraxial entrance pupil position on the object space z axis. Positive field angles imply
positive slope for the ray in that direction, and thus refer to negative object coordinates. ZEMAX converts x field
angles ( x ) and y field angles ( y ) to ray direction cosines using the following formulas:
l
tan x = ---
n
m
tan y = ----
n
2 2 2
l +m +n = 1
where l, m, and n are the x, y, and z direction cosines.

If object or image heights are used to define the field points, the heights are measured in lens units. When
paraxial image heights are used as the field definition, the heights are the paraxial image coordinates of the chief
ray on the image surface, and if the optical system has distortion, then the real chief rays will be at different
locations. When real image heights are used as the field definition, the heights are the real ray coordinates of the
chief ray on the image surface.
To set the field type and values, see Fields on page 96.
Float by stop size
Float by stop size is one of the system aperture types supported by ZEMAX. This phrase refers to the fact that
the entrance pupil position, object space numerical aperture, image space F/#, and stop surface radius all are

specified if just one of them is specified. Therefore, setting the stop radius, and then allowing the other values to
be whatever they are, is a perfectly valid way of defining the system aperture. It is particularly handy when the
stop surface is a real, unchangeable aperture buried in the system, such as when designing null corrector optics.
Ghost reflections
Ghost reflections are spurious, unwanted images formed by the small amount of light which reflects off of,
rather than refracts through a lens face. For example, the multiple images of the aperture stop visible in
photographs taken with the sun in the field of view are caused by ghost reflections. Ghost images can be
problematic in imaging systems and in high power laser systems.
Glasses
Glasses are entered by name in the glass column. Available glasses may be reviewed, and new ones entered
using the glass catalog tool. See the Chapter "Using Glass Catalogs" for details.
Blanks are treated as air, with unity index. Mirrors can be specified by entering "MIRROR" for the glass type,
although this name will not appear in the glass catalog. The index of refraction of the mirror space is always equal
to the index of refraction of the media before the mirror.
For information on the affect of temperature and pressure on index of refraction data, see Defining tempera-
ture and pressure on page 491.
Hexapolar rings
ZEMAX usually selects a ray pattern for you when performing common calculations such as spot diagrams.
The ray pattern refers to how a set of rays is arranged on the entrance pupil. The hexapolar pattern is a rotationally
symmetric means of distributing a set of rays. The hexapolar pattern is described by the number of rings of rays
around the central ray. The first ring contains 6 rays, oriented every 60 degrees around the entrance pupil with
the first ray starting at 0 degrees (on the x-axis of the pupil). The second ring has 12 rays (for a total of 19, including
the center ray in ring "0"). The third ring has 18 rays. Each subsequent ring has 6 more rays than the previous ring.
Many features which require a sampling parameter to be specified (such as the spot diagram) use the number
of hexapolar rings as a convenient means of specifying the number of rays. If the hexapolar sampling density is
5, it does not mean that 5 rays will be used. A sampling of 5 means 1 + 6 + 12 + 18 + 24 + 30 = 91 rays will be used.
Image space F/#
Image space F/# is the ratio of the paraxial effective focal length calculated at infinite conjugates over the
paraxial entrance pupil diameter. Note that infinite conjugates are used to define this quantity even if the lens is
not used at infinite conjugates.
Image space numerical aperture (NA)
Image space NA is the index of image space times the sine of the angle between the paraxial on-axis chief
ray and the paraxial on-axis +y marginal ray calculated at the defined conjugates for the primary wavelength.
Lens units
Lens units are the primary unit of measure for the lens system. Lens units apply to radii, thicknesses, apertures,
and other quantities, and may be millimeters, centimeters, inches, or meters.
Marginal ray
The marginal ray is the ray that travels from the center of the object, to the edge of the entrance pupil, and on
to the image plane.
If there is vignetting, ZEMAX extends this definition by defining the marginal ray to be at the edge of the
vignetted entrance pupil. If ray aiming is on, then the marginal ray is at the edge of the vignetted stop.
See also the definition of the chief ray.
Maximum field
The maximum field is the minimum radial coordinate that would enclose all the defined field points if the x and
y values of each field point were plotted on an Cartesian XY plot. The maximum field is measured in degrees if
the field type is angles, or in lens units for object height, paraxial image height, or real image height. The field

type is described in Field angles and heights on page 47. To set the field type and values, see Fields on
page 96.
Non-paraxial systems
The term non-paraxial system refers to any optical system which cannot be adequately represented by paraxial
ray data. This generally includes any system with tilts or decenters (implemented with coordinate break surfaces),
holograms, gratings, cubic splines, ABCD matrices, gradient index, diffractive components, or non-sequential
surfaces.
A great deal of optical aberration theory has been developed for systems with conventional refractive and
reflective components in rotationally symmetric configurations. This includes Seidel aberrations, distortion,
Gaussian beam data, and virtually all first order properties such as focal length, F/#, and pupil sizes and locations.
All of these values are calculated from paraxial ray data.
If the system being analyzed contains any of the non-paraxial components described, then any data computed
based upon paraxial ray tracing cannot be trusted. ZEMAX will generally use exact real rays rather than paraxial
rays for ray tracing through these surfaces and components.
Non-sequential ray tracing
Non-sequential ray tracing means rays are traced only along a physically realizable path until they intercept
an object. The ray then refracts, reflects, or is absorbed, depending upon the properties of the object struck. The
ray then continues on a new path. In non-sequential ray tracing, rays may strike any group of objects in any order,
or may strike the same object repeatedly; depending upon the geometry and properties of the objects.
See also Sequential ray tracing.
Normalized field and pupil coordinates
Normalized field and pupil coordinates are often used in both the ZEMAX program and documentation. There
are four normalized coordinates: Hx, Hy, Px, and Py. The Hx and Hy values are the normalized field coordinates;
Px and Py are the normalized pupil coordinates.
The normalized field and pupil coordinates represent points on a unit circle. The radial size of the field is then
used to scale the normalized field coordinates, and the entrance pupil radius is used to scale the normalized pupil
coordinates. For example, suppose the maximum object height is 10 mm. This would be the case if there are 3
defined object height fields, at 0, 7, and 10 mm. The coordinate (Hx = 0, Hy = 1) would refer to a ray which starts
at the top of the object (x = 0 mm, y = 10 mm). The coordinate (Hx = -1, Hy = 0) would refer to a ray which starts
at (x = -10 mm, y = 0 mm) on the object surface.
The Hx and Hy values always refer to points on a circle in object angle space whose radius is determined by
the maximum radial field. The radial field for each field point individually is determined by computing the square
root of the sum of the squared x-field and y-field values. The maximum radial field is the largest of these individual
radial fields. For example, if a single field point is defined whose x-field is 10 and y-field is -6, then the maximum
radial field is 11.66, and Hx and Hy will then be normalized to this radius. Note that if there are two separate field
points, one with an x-field of 10 and a y-field of 0; and a second field point with an x-field of 0 and a y-field of -6,
then the maximum radial field is 10, not 11.66. The maximum radial field is computed this way for all field types,
whether height or angle. The maximum radial field is always a positive value.
The pupil coordinates work the same way. Suppose the entrance pupil radius (Not diameter) is 8 mm. Then
(Px = 0, Py = 1) refers to a ray which is aimed to the top of the entrance pupil. On the entrance pupil surface, the
ray will have a coordinate of (x = 0 mm, y = 8 mm).
Note that the normalized coordinates should always be between -1 and 1, and that
2 2 2 2
H x + H y 1, P x + P y 1 .
The advantage of using normalized coordinates is that certain rays always have the same coordinates,
independent of the size or location of either the object or the entrance pupil. For example, the marginal ray is the
ray which travels from the center of the object to the top of the entrance pupil, or (Hx = 0, Hy = 0, Px = 0, Py = 1).
The chief ray travels from the top of the field to the center of the pupil, or (0,1,0,0). Note that "top of the field" in

this context means the maximum positive field value in the y direction. If the field type is object height, the
maximum y field value is the maximum y coordinate on the object surface. If the field type is angle, then the
maximum y field value is the maximum y angle in object space; this ray actually comes from negative coordinate
space. For more discussion, see Field angles and heights on page 47.
Another significant advantage of this system is that rays defined in normalized coordinates remain meaningful
as the pupil size and position changes. Suppose prior to optimizing a lens design, you define a ray set to compute
the system merit function. By using normalized coordinates, the same ray set will work unaltered if the entrance
pupil size or position or object size or position is changed later, or perhaps even during the optimization procedure.
The normalized field coordinates work even if the field positions are defined in terms of angles. For example,
suppose you have selected y-field angles of 0, 7, and 10 degrees. This implies your maximum field "radius" in
angular space is 10 degrees. Normalized field coordinates of Hx = 0 and Hy = 1 select the x-field of 0 and the y-
field of 10 degrees. The coordinates Hx = -0.5 and Hy = 0.4 select the x-field angle of -5 degrees, and a y-field
angle of 4 degrees. Note that even if no x-angles are defined, you may trace rays from these fields using non-
zero values for Hx. Notice that the coordinates are normalized field angles if you define your object in terms of
field angles. If you are using object height then the Hx and Hy are normalized object heights.
Object space numerical aperture
Object space numerical aperture is a measure of the rate of divergence of rays emanating from the object
surface. The numerical aperture is defined as the index of refraction times the sine of the paraxial marginal ray
angle, measured in object space. The marginal ray defines the boundary of the cone of light diverging from an
object point.
Parameter data
Parameter data values are used to define certain non-standard surface types. For example, parameter data
may include aspheric coefficients, grating spacings, or tilt and decenter data. For a complete discussion of the
parameter data values see Parameter data on page 225.
Paraxial and parabasal rays
The term paraxial means "near the axis". Paraxial optics are optics that are well described by the linear form
of Snells law. Snells law is:
n sin = n' sin ' .
For small angles this becomes
n = n'' .
Many definitions in optics are based upon this assumption of linearity. Aberrations are deviations from this
linearity, and so the paraxial properties of optical systems are often considered the properties the system has in
the absence of aberrations. Paraxial rays are traced using formulas which assume the optical surface power is
based only upon the vertex radius of curvature, ignoring local linear tilts and higher order curvature of the surface.
Paraxial data is computed on a plane tangent to the surface vertex, assuming the vertex radius of curvature
is an acceptable approximation to the surface power over the entire aperture of the surface. Certain unusual
surface types do not have a paraxial analog, so the real ray tracing calculations are made at these surfaces, even
for a paraxial ray.
ZEMAX computes many paraxial entities, such as focal length, F/#, focal position, entrance pupil diameter,
and others. These values should be used with caution when the optical system has components which violate
the assumption that the vertex curvature is an acceptable approximation to the surface power over the entire
aperture of the surface.
For many analysis features, paraxial data is required, typically as a reference against which real rays are
measured. To ensure these features work properly, even for optical systems that do not meet the paraxial
assumption above, ZEMAX traces "parabasal" rays which are real (real means using Snell's law explicitly) that
make small angles with respect to a reference ray, which is usually an axis or chief ray. The parabasal rays are

used to compute the limiting properties of the system as the stop size is decreased, which provides a good
estimate of the paraxial properties.
The reason ZEMAX uses parabasal rays rather than paraxial formulas is because many optical systems
include non-paraxial components. Non-paraxial means these components are not well described by conventional
axial first-order theory. This includes tilted or decentered systems, systems using holograms, diffractive optics,
general aspheres, and gradient index lenses.
In summary, paraxial ray data is computed using first order approximations to the surface power for tracing
rays, while parabasal rays are real, exact ray traces close to a chief or reference ray. Most paraxial data, such as
EFL, F/#, and magnification, use paraxial rays and the data is invalid if the optical system is not well described
by the vertex power of every surface. Most analysis features in ZEMAX use parabasal rays, to allow these
features to work with a greater range of optical systems, including those with optical surfaces not well described
solely by their vertex surface power.
Paraxial image height
The paraxial radial size of the image in lens units of the full field image at the paraxial image plane.
Paraxial magnification
The radial magnification, being the ratio of paraxial image height to object height. The paraxial magnification
is measured at the paraxial image plane. The value is always zero for infinite conjugate systems.
Paraxial working F/#
The paraxial working F/# is defined as
1
W = ------------------ ,
2n tan
where is the paraxial marginal ray angle in image space and n is the index of refraction of image space. The
paraxial marginal ray is traced at the specified conjugates. For non-axial systems, this parameter is referenced
to the axis ray and is averaged over the pupil. The paraxial working F/# is the effective F/# ignoring aberrations.
See also the definition for working F/#.
Primary wavelength
The primary wavelength in micrometers is displayed. This value is used for calculating most other paraxial or
system values, such as pupil positions.
Radii
The radius of curvature of each surface is measured in lens units. The convention is that a radius is positive
if the center of curvature is to the right (a positive distance along the local z axis) from the surface vertex, and
negative if the center of curvature is to the left (a negative distance along the local z axis) from the surface vertex.
This is true independent of the number of mirrors in the system.
Real propagation
A real propagation means the rays are propagated in the direction energy would actually flow. See also Virtual
propagation on page 55 and Thicknesses on page 53.
Sagittal and Tangential
The term "tangential" refers to data computed in the tangential plane, which is the plane defined by a line and
one point: the line is the axis of symmetry, and the point is the field point in object space. The sagittal plane is the
plane orthogonal to the tangential plane, which also intersects the axis of symmetry at the entrance pupil position.
For typical rotationally symmetric systems with field points lying along the Y axis, the tangential plane is the
YZ plane and the sagittal plane is the plane orthogonal to the YZ plane which intersects the center of the entrance
pupil.
The problem with this definition is that it is not readily extended to non-rotationally symmetric systems. For this
reason, ZEMAX instead defines the tangential plane to be the YZ plane regardless of where the field point is, and

tangential data is always computed along the local y axis in object space. The sagittal plane is the orthogonal to
the YZ plane, and intersects the center of the entrance pupil in the usual way, and sagittal data is always computed
along the x axis in object space.
The philosophy behind this convention is as follows. If the system is rotationally symmetric, then field points
along the Y axis alone define the system imaging properties, and these points should be used. In this case, the
two different definitions of the reference planes are redundant and identical. If the system is not rotationally
symmetric, then there is no axis of symmetry, and the choice of reference plane is arbitrary.
One feature, the computation of Fast Semi-Diameters (see Fast Semi-Diameters on page 92), does use the
"true" tangential plane, which ZEMAX defines as the plane that contains the actual field point and the z axis in
object space.
Semi-diameters
The size of each surface is described by the semi-diameter setting. The default setting is the radial distance
to the aperture required to pass all real rays without clipping any of the rays. Typing a value for the semi-diameter
column results in the character "U" being displayed next to the value. The "U" indicates that the semi-diameter is
user defined. When a user-defined semi-diameter is placed on a surface with refractive power (which is done by
typing in a value in the appropriate column), and no surface aperture has been defined, ZEMAX automatically
applies a "floating" aperture to the surface. A floating aperture is a circular aperture whose radial maximum
coordinate is always equal to the semi-diameter of the surface. For more information on surface aperture types,
see Surface properties aperture tab on page 68.
Semi-diameters on any surface for axial symmetric systems are computed very accurately, as long as the
surface does not lie within the caustic of the ray bundle (note this usually occurs at or near the image surface).
ZEMAX estimates semi-diameters for axial systems by tracing a few marginal pupil rays. For non-axial systems,
ZEMAX estimates the required semi-diameters using either a fixed number of rays or by an iterative technique,
which is slower but more accurate. See Fast Semi-Diameters on page 92 for details. It is important to note that
the "automatic" semi-diameter computed by ZEMAX is an estimate, although it is generally a very good one.
Some surfaces may become so large in aperture that the surface z coordinate becomes multiple valued; for
example, a very deep ellipse may have more than one z coordinate for the same x and y coordinates on the
surface. For the case of spherical surfaces, this condition is called "hyperhemispheric" and ZEMAX uses this term
even if the surface is not a sphere. Hyperhemispheric surfaces are denoted by an asterisk "*" in the semi-diameter
column. The indicated semi-diameter is of the outer edge of the surface, which will have a smaller radial aperture
than the maximum radial aperture.
Sequential ray tracing
Sequential ray tracing means rays are traced from surface to surface in a predefined sequence. ZEMAX
numbers surfaces sequentially, starting with zero for the object surface. The first surface after the object surface
is 1, then 2, then 3, and so on, until the image surface is reached. Tracing rays sequentially means a ray will start
at surface 0, then be traced to surface 1, then to surface 2, etc. No ray will trace from surface 5 to 3; even if the
physical locations of these surfaces would make this the correct path.
See also Non-sequential ray tracing.
Strehl ratio
The Strehl ratio is one commonly used measure of optical image quality for very high quality imaging systems.
The Strehl ratio is defined as the peak intensity of the diffraction point spread function (PSF) divided by the peak
intensity of the diffraction point spread function (PSF) in the absence of aberrations. ZEMAX computes the Strehl
ratio by computing the PSF with and without considering aberrations, and taking the ratio of the peak intensity.
The Strehl ratio is not useful when the aberrations are large enough to make the peak of the PSF ambiguous, or
for Strehl ratios smaller than about 0.1.
Surface apertures
Surface apertures include circular, rectangular, elliptical, and spider shaped apertures which can vignette rays.
There are also user defined shapes for surface apertures and obscurations; and a "floating" aperture that is based
upon the current semi-diameter value. Surface apertures do not effect ray launching or tracing, except for the

termination of a ray if it does not pass the surface aperture. Surface apertures have no effect on the system
aperture. For more information, see page 68.
System aperture
The system aperture is the overall system F/#, Entrance Pupil Diameter, Numerical Aperture, or Stop Size.
Any of these 4 quantities is sufficient to define the other 3 for a particular optical system. The system aperture is
used to define the object space entrance pupil diameter, which in turn is used to launch all rays. The system
aperture is always circular. Rays may be vignetted after being launched by various surface apertures. There is
only one system aperture, although there may be many surface apertures.
Tangential
See Sagittal and Tangential on page 51.
Thicknesses
Thicknesses are the relative distance to the next surface vertex in lens units. Thicknesses are not cumulative,
each one is only the offset from the previous vertex along the local z axis. The orientation of the local z axis can
change using coordinate breaks (see Coordinate Break on page 239) or surface tilts and decenters (see
Surface tilt/decenter tab on page 73).
Thicknesses corresponding to real propagation (see Real propagation on page 51) always change sign after
a mirror. After an even number of mirrors (including zero mirrors), thickness are positive for real propagations and
negative for virtual propagations (see Virtual propagation on page 55). After an odd number of mirrors,
thicknesses are negative for real propagations and positive for virtual propagations. This sign convention is
independent of the number of mirrors, or the presence of coordinate breaks. This fundamental convention cannot
be circumvented through the use of coordinate rotations of 180 degrees.
Total internal reflection (TIR)
TIR refers to the condition where a ray makes too large an angle with respect to the normal of a surface to
meet the refraction condition as specified by Snell's Law. This usually occurs when a ray with a large angle of
incidence is refracting from a high index media to a lower index media, such as from glass to air. When doing
sequential ray tracing, rays which TIR are considered errors, and are terminated. Physically, the ray would reflect
rather than refract from the boundary, but ZEMAX does not consider this effect when doing sequential ray tracing.
For non-sequential ray tracing, rays which TIR are properly reflected.
Total track
Total track is the length of the optical system as measured by the vertex separations between the "left most"
and "right most" surfaces. The computation begins at surface 1. The thickness of each surface between surface
1 and the image plane is considered, ignoring any coordinate rotations. The surface which lies at the greatest z
coordinate defines the "right most" surface, while the surface with the minimum z coordinate defines the "left most"
surface. Total track has little value in non-axial systems.
Units
See.Units on page 87
Vignetting factors
Vignetting factors are coefficients which describe the apparent entrance pupil size and location for different
field positions. ZEMAX uses five vignetting factors: VDX, VDY, VCX, VCY, and VAN. These factors represent
decenter x, decenter y, compression x, compression y, and angle, respectively. The default values of all five
factors are zero, which indicates no vignetting.
Both the field of view and the entrance pupil of an optical system can be thought of as unit circles. The
normalized field and pupil coordinates, defined in Normalized field and pupil coordinates on page 49, are the
coordinates on these two unit circles. For example, the pupil coordinates (px = 0, py = 1) refer to the ray which is
traced from some point in the field to the top of the entrance pupil. If there is no vignetting in the system, ZEMAX
will trace rays to fill the entire entrance pupil during most computations.
Many optical systems employ deliberate vignetting. This means a portion of the rays are intentionally "clipped"
by apertures other than the stop surface. There are two common reasons for introducing vignetting in an optical

system. First, vignetting decreases the size of the lenses, particularly in wide angle lenses. Second, vignetting
may remove a portion of the beam which would be excessively aberrated. Vignetting usually increases the F/#
as a function field angle (which darkens the image), but the image quality may improve if the most severely
aberrated rays are clipped.
Vignetting factors redefine the entrance pupil for a specific field position. The normalized pupil coordinates are
modified using two successive transformations. First, the coordinates are scaled and shifted:
P x = VDX + Px ( 1 VCX ) , and
P y = VDY + P y ( 1 VCY ) .
The scaled and shifted coordinates are then rotated by the vignetting angle:
P x = P x cos P y sin , and
P y = P x sin + P y cos ,
where is the vignetting angle VAN. The VDX term can shift the apparent pupil left and right, while VCX makes
the pupil larger or smaller in the x direction. Similar results hold for the VDY and VCY values. Note that if the
vignetting factors are all zero, the pupil coordinates are left unmodified. Vignetting factors provide a convenient
way of designing optics which employ vignetting. However, there are restrictions to using vignetting factors that
must be understood.
Some ZEMAX features are capable of tracing rays from arbitrary field positions where no vignetting factors
have been assigned. These features may not provide completely accurate results for data computed at field
positions other than the defined fields. Some features will remove the vignetting factors for these computations
by placing a clear aperture on each surface that vignettes the rays an equivalent amount. Features that
automatically remove the vignetting factors are described in the Analysis chapter.
Some features in ZEMAX do not automatically remove vignetting factors for intermediate field positions, such
as ray operands in the merit function (operands like REAX that can launch a single ray, for example) or ZPL
macros. If the vignetting factors are not removed, ZEMAX will attempt to interpolate the vignetting factors. For
rotationally symmetric systems, or systems with field points entirely along the y axis, ZEMAX interpolates between
adjacent field points to estimate the vignetting factors to use at intermediate field points. For more general optical
systems with X field values, ZEMAX uses the closest defined field point for determining the vignetting factors for
an arbitrary field point.
Once a vignetting factor is defined, it is up to the designer to ensure that rays beyond the apparent pupil are
in fact vignetted! If the vignetting factor is used to shrink the size of the lenses, then the lenses should be made
no larger than is required to pass the rays which are at the edge of the apparent pupil. If rays from beyond the
vignetted aperture are allowed to pass in the real optical system, then the lens performance will not correlate with
the computer model.
Identical or nearly identical field coordinates may not be defined with different vignetting factors. Field
coordinates must be different by roughly 1E-06 times the maximum field coordinate if two neighboring fields use
different vignetting factors. This is required because ZEMAX must determine vignetting factors for any field
coordinates, not just those at defined field positions; and identical field coordinates with different vignetting factors
have no physical interpretation. The proper way to set up this sort of system is to use multiple configurations, and
change the vignetting factors via the multi-configuration editor.
The vignetting factors work with and without ray aiming turned on. If ray aiming is off, then the paraxial entrance
pupil is remapped according to the equations given earlier. If ray aiming is turned on, then the remapping is done
at the stop surface.
One possible application of vignetting factors is to account for pupil aberration without using the ray aiming
feature. This is an advanced trick which can be used to speed up ray tracing in wide angle systems.

Vignetting factors may be defined on the "Field Data" dialog box. See the Chapter "System Menu" for more
information. Vignetting factors may also be zoomable parameters; see the Chapter "Multi-Configurations". For
more information on the use of vignetting as a design tool, see any of the good books referenced in the first
chapter.
Virtual propagation
A virtual propagation means the rays are propagated in a direction opposite to the direction energy would
actually flow. Virtual propagations are often useful for placing virtual sources or pupils. See also Real propaga-
tion on page 51 and Thicknesses on page 53.
Wavelength data
Wavelength data are always measured in micrometers referenced to "air" at the current system temperature
and pressure. The default system temperature is 20 degrees Celsius, and the default air pressure is 1.0
atmospheres. If the system temperature and/or pressure is modified, or under the control of multi-configuration
operands, care must be taken to adjust the wavelengths to the new air temperature and pressure.
Wavelength data is entered on the "Wavelength Data" dialog box; see the Wavelengths on page 96 for
details.
!Wavelength data are always measured in micrometers referenced to "air" at the system
temperature and pressure.
Working F/#
Working F/# is defined as
1
W = ----------------- ,
2n sin
where is the real marginal ray angle in image space and n is the index of refraction of image space. The
marginal ray is traced at the specified conjugates. Working F/# ignores surface apertures but considers vignetting
factors.
For off axis field points or non-axial systems, working F/# is determined by the average of the square of the
numerical aperture between the axis ray and four marginal rays, at the top, bottom, left, and right side of the
vignetted pupil. The average of the square of the numerical aperture of the four rays is converted back to
equivalent F/#.
Working F/# is generally much more useful than image space F/# because it is based upon real ray data at
the actual conjugates of the lens. See also Paraxial working F/# on page 51.
If the marginal rays cannot be traced (due to ray errors) then a smaller pupil is temporarily used to estimate
the working F/#. In this case, ZEMAX scales the data to estimate the working F/# at the full pupil size, even though
rays are not be traceable at the full aperture.
If the marginal rays are nearly parallel to the chief ray, the resulting F/# may become so large as to be
inaccurate. ZEMAX will automatically "cap" the F/# at 10,000 if the calculated F/# becomes larger than 10,000.
This result simply means the F/# cannot be calculated accurately using rays. Such a large F/# means the output
beam is nearly collimated, and many assumptions ZEMAX makes are not valid in this case. The two solutions
are to bring the nearly collimated beam to a focus using a paraxial lens, or to use the exit pupil size and position
to estimate the F/#. For more information, see Method To Compute F/# on page 94.
For a generalized definition of F/# useful for evaluating image brightness, see Effective F/# on page 140.

Chapter 4 FILE MENU
New
Purpose:
Clears the current lens.
Discussion:
This option restores ZEMAX back to the start-up condition. If the current lens has not been saved, ZEMAX will
prompt to save the lens before it is destroyed.
Open
Purpose:
Opens an existing ZEMAX lens file.
Discussion:
This option opens a previously saved lens file.
Save
Purpose:
Saves lens file.
Discussion:
This option saves the current lens file. To save the lens under a different name or directory, use Save As.
Save As
Purpose:
Saves lens file, optionally with another name.
Discussion:
This option saves the current lens file with the option to change the current file name or directory.
Use Session Files
Purpose:
Toggles the environment setting between using and not using session files. If the menu option is checked, then
session files are used.
Discussion:
For a complete discussion of session files, see Editors on page 61.
Program Mode
Purpose:
Selects the user interface and program mode.
Discussion:
There are two distinct modes to the ZEMAX User Interface:
Sequential or Mixed Sequential/Non-Sequential Design ("Sequential mode")
Non-Sequential Design ("Non-Sequential mode" or "NSC mode")
In Sequential mode, all program features are available, including the ability to place a non-sequential group
upon any surface. In Sequential mode, ZEMAX will trace rays which leave the object surface through all defined
surfaces and non-sequential groups. This is the preferred mode for designing imaging systems, or any system
requiring optimization, tolerancing, and detailed image analysis.
In Non-Sequential mode, multiple changes are made to the user interface and to the lens data to simplify the
analysis of these systems:
57 Chapter 4: FILE MENU

Only surface 1 is used, and all NSC objects are placed in a single non-sequential group placed on surface 1.
The object and image surfaces are not used.
The Lens Data Editor and Extra Data Editors are not used. The Non-Sequential Components Editor (NSCE)
is used instead as the primary editor.
The field dialog box is not used. The wavelengths dialog box is still used to define the wavelengths to be traced.
Only rays which originate from sources placed within the NSCE are traced.
Features which are not used in Non-Sequential mode are removed from the menus and button bars to simplify
the user interface. These features include ray fans, MTF plots, spot diagrams, and many other features unique
to sequential ray tracing. The primary means of analysis in Non-Sequential mode is tracing rays to one or more
detector objects.
Using Non-Sequential mode makes the operation of the program easier for systems which are non-imaging.
It is always possible to switch from Non-Sequential mode to Sequential mode without loss of data. Switching
from Sequential mode to Non-Sequential mode will cause all the sequential data to be deleted.
Insert Lens
Purpose:
Inserts surface data from a previously stored lens data to the current lens file.
Discussion:
This option is similar to the "Open" command, but the current lens data is not disturbed. After the file to insert
is selected, ZEMAX will prompt for the surface number where the new lens data should be inserted. New surfaces
will be inserted into the lens prescription to make room for the new data. The dialog box also has an "Ignore
Object" checkbox which by default will ignore the object thickness of the new lens. Therefore, the new lens data
to be inserted will start at surface 1 rather than surface 0.
Although this feature can save a lot of typing, the resulting lens may have some extra surfaces, or may need
a little hand editing to achieve the desired result.
Preferences
ZEMAX allows users a number of options that can be set and saved so that these options are selected
automatically when ZEMAX is run. The main configuration file is ZEMAX.CFG; this file can be deleted to return
the start-up configuration to the default state. The preference options are divided into groups as follows.
Address
Purpose:
These settings determine how the "address" box will appear. The address box can be used to display user
defined text such as a company name or drawing number. The address box appears on the bottom right corner
of most graphics.
Settings:
Item Description
Address Line 1 The first line of text to appear in the address box.
Address Line 2 The second line of text to appear in the address box.
Address Line 3 The third line of text to appear in the address box.
Address Line 4 The fourth line of text to appear in the address box, unless file name or zoom position
is chosen instead.
Address Line 5 The fifth line of text to appear in the address box, unless file name or zoom position is
chosen instead.
Show Line 4 As Choose either the entered text, the lens file name, or the zoom position.
Chapter 4: FILE MENU 58

Item Description
Show Line 5 As Choose either the entered text, the lens file name, or the zoom position.
Hide Address If checked the address box will not appear.
Directories
Purpose:
These settings determine where ZEMAX will place or search for certain types of files. The "?" button may be
used to select the desired path.
Settings:
Item Description
ZEMAX Path The directory where ZEMAX will look for files used by the program, such as glass
catalogs and stock lens catalogs.
Output Path The directory where text and graphics output will be sent.
Lens Path The directory for lens files.
ZPL Path The directory for ZPL Macros. Macros may also be placed in subdirectories within the
specified macro.
Undo Path The directory for storing lens files for the undo/redo feature.
Stock Path The directory for the stock lens files.
Objects Path The directory for Non-Sequential Component object files and DLLs.
Glass Path The directory for glass catalog files.
Coating Path The directory for coating definition files.
Graphics
Purpose:
These settings determine the size, color, and behavior of most ZEMAX graphics windows.
Settings:
Item Description
B/W Display By default, ZEMAX draws color graphics to the screen. If this checkbox is selected, then
all graphics on screen will be black and white.
B/W Plots By default, ZEMAX prints color graphics to printers. If this checkbox is selected, then all
printed graphics will be black and white. Only printers which support color will print in
color, otherwise, gray scales will be used to simulate color.
Show Options First If this checkbox is selected, then the settings box will appear before any analysis graphic
or text is computed and displayed.
Win X,Y Size The default x and y size for graphics and text windows in pixels. This can be adjusted
to suit the monitor size and resolution.
Background The background color on graphics windows can be selected from this drop down list.

Item Description
Metafiles ZEMAX generates several different types of Windows Metafile formats. Metafiles are
used to copy graphics to the clipboard, or to copy graphics to files on disk, where the
graphics may then be pasted or imported into other Windows programs. Most 16 bit
Windows 3.1 applications used the "16 Bit Standard" format. However, some Windows
3.1 applications use a variation called the "16 Bit Placeable" format. Newer, 32 bit
applications use the "32 Bit Enhanced" format. When a 32-bit format is used, the
extension used is EMF, for enhanced metafile format.
Metafile Pen Width Determines the pen width in device dependent pixels for graphics which are exported
as Metafiles, via either the clipboard or as disk files. Larger values yield thicker lines.
Aspect Ratio The default aspect ratio for ZEMAX graphic windows is 3 x 4, which fits well on a
standard 8.5 x 11 inch paper printer. For 11 x 17 inch printing, a 3 x 5 aspect ratio fits
better. The 4 x 3 and 5 x 3 aspect ratios are taller than they are wide. This option selects
the default aspect ratio for both printing and screen graphics. Each graphic screen may
have it's own aspect ratio modified by using the Window, Aspect Ratio setting.
Use Active Cursor This setting determines whether or not the active cursor is initially on when creating a
new window. Each window may be set to display or not to display the active cursor
individually once the window is opened using the Window menu pull down.
Frame Zoomed If selected, zoomed graphics which are pasted to the clipboard will be framed; otherwise,
Graphics the frame will not be shown.
Layouts Rotate Z, If selected, all 3D Layout type plots will rotate the displayed lens first in Z, then Y, then
Y, X X about the windows coordinates. This selection results in rotations that are similar to
ZEMAX Coordinate Break conventions if the order flag is zero.
If this option is not selected, then the displayed lens will be rotated first in X, then Y, then
Z about the windows coordinates. This selection results in rotations that are similar to
ZEMAX Coordinate Break conventions if the order flag is not zero.
Highlight Layouts If selected, then the current surface (in the LDE or EDE) or object (in the NSCE) will be
highlighted in all open layout type plots. Moving the cursor from row to row will cause
the layout plots to redraw with the selected surface or object highlighted. The highlight
color is defined on the Colors tab. Highlighting does slow down navigation of the editors
somewhat, because the graphics are continuously redrawn.
Highlight While If selected, then highlighting (see Highlight Layouts above) will occur while scrolling with
Scrolling the vertical scroll bars on the right side of the editors. On complex lens drawings this
may interfere with proper updating of the display because the lens drawing updates are
not fast enough to keep up with the scroll bar controls.
Fletch Size Determines the relative size of fletch marks drawn on the various layout plots. The
default value of 1.0 draws fletches at the default size; a value of 2.0 would draw the
fletches twice as large as normal, and 0.5 would draw fletches at half size. Any value
between 0.0 and 100.0 may be used.
Orientation Determines the relative size of the orientation indicator drawn on the various layout
Indicator Size plots. The default value of 1.0 draws the indicator at the default size; a value of 2.0 would
draw the indicator twice as large as normal, and 0.5 would draw the indicator at half size.
Any value between 0.0 and 10.0 may be used. A value of zero will remove the indicator.
Miscellaneous
Purpose:
Miscellaneous settings.

Settings:
Item Description
Text Window Font The font size in points to be used for text displayed in an analysis window, such as a ray
fan text listing. To change the size of text displayed in the editors see "Editors" below.
Date/Time Select either None for no date or time, Date for the date only, and Date/Time for both
the date and the time to be printed on plots and graphics.
Check For Updates Select how often ZEMAX should check to see if a new version of ZEMAX is available.
Updates may be downloaded from www.zemax.com/downloads. An active Internet
connection is required for ZEMAX to be able to check for updates. The "Check For
Updates" option listed under the Help menu can be used to check for updates anytime.
Remote Desktop The ZEMAX key driver by default does not allow ZEMAX to run under Windows Remote
Desktop feature. To allow ZEMAX to run under Remote Desktop, select this option to
"On", then exit and restart ZEMAX. This feature requires version 7.0.0 or later of the
Sentinel driver, available from the ZEMAX website www.zemax.com. When Remote
Desktop is set to "On" only one instance of ZEMAX at a time can run.
The reason this is not the default setting is that if ZEMAX terminates abnormally
(crashes) the remote desktop enabled driver does not allow ZEMAX to run again for
several minutes, until the driver resets itself. Also, the remote driver only allows one
application at a time to connect to the key, preventing multiple versions of ZEMAX from
running.
To modify the setting without running ZEMAX, edit the ASCII text file ZEMAX.CFG found
in the same directory as the ZEMAX.exe executable file. Look for the line in the file that
starts with RDSK. The syntax is RDSK 0 for Remote Desktop off and RDSK 1 for Remote
Desktop on.
Editors
Purpose:
These settings determine characteristics of the spreadsheet editors. If the editor cell sizes are too narrow to
display all the data in each cell, an asterisk "*" symbol will be printed instead of the truncated data.
Settings:
Item Description
Decimals Decimals will alter the number of decimal digits displayed on the lens data editor.
Selecting "Compact" will vary the number of decimals displayed to minimize the space
required to display numbers.
Exponential Above/ ZEMAX displays numeric data in either fixed decimal or exponential (scientific) notation.
Below If the absolute value of the number is either a) greater than or equal to the "Exponential
Above" value, or b) not zero but less than or equal to the "Exponential Below" value,
then the number is displayed in exponential notation. Otherwise fixed notation is used.
Font Size The font size in points to be used for text. The default is an 8 point text.
Auto Update Controls how and when ZEMAX performs updates on the editor data. The "None" setting
means pupil positions, solves, and other lens data editor data are not updated until
"Update" is selected from the "System" menu. The "Update" setting will cause an update
to be performed whenever new data is typed in to the lens, extra, or multi-configuration
editors. "Update All" causes all windows to be updated whenever new data is typed in
the editors. See the "System menu" chapter for details on Update and Update All.
Undo The Undo feature has three selectable states: None, Memory 1 Step, and Disk Multi
Step. See the Chapter "Editors Menu" for details on the Undo feature.
LDE Cell Size The width in characters of a single edit cell in the Lens Data Editor. Larger cells will mean
fewer columns can be displayed, but the data is generally more readable.

Item Description
MFE Cell Size The width in characters of a single edit cell in the Merit Function Editor.
MCE Cell Size The width in characters of a single edit cell in the Multi-Configuration Editor.
EDE Cell Size The width in characters of a single edit cell in the Extra Data Editor.
TDE Cell Size The width in characters of a single edit cell in the Tolerance Data Editor.
NSC Cell Size The width in characters of a single edit cell in the Non-Sequential Components Editor.
Default Catalog The name of the default glass catalog to use. See Glass Catalogs on page 88.
Show Comments If checked, then the surface comments column is displayed in the Lens Data Editor.
Otherwise, the column is hidden from view.
Use Session Files If checked, all open windows will be closed, and new windows will be opened and
restored to their original locations when new lens files are loaded. If unchecked, files will
be loaded without changing the arrangement of windows on the screen.
Allow Extensions If checked, then any ZEMAX extension can replace the lens data in the Lens Data Editor
To Push Lenses with new data, without warning or prompt. This option should not be checked unless it
is intended that an extension control ZEMAX. See Extensions on page 223 for more
information on extensions.
Color Editor Rows If checked, then rows in the spreadsheet are color coded. See Row Color on page 67.
Printing
Purpose:
These settings determine characteristics of the printed output.
Settings:
Item Description
Skip Print Dialog If this box is checked, then ZEMAX will not present the "Print" dialog box
which allows selection of printer type and other options when "Print" is
selected from any window. If the box is unchecked, then the default printer
settings will be used as specified by the Windows Print Manager.
Rotate Plots If selected, this will cause all printed graphics to be rotated by 90 degrees.
This allows a landscape type orientation while printing in portrait mode. See
the discussion below.
Plot Width See the discussion below.
Pen Width The width of the pen in pixels. Values of 0 result in fine lines, larger values
yield thicker lines.
Print Font Size The font size in points to be used for text when a text window is printed.
Left Graphic Margin % The left margin of the plot as a percentage of the maximum possible plot
width. Only affects graphics printing.
Right Graphic Margin % The right margin of the plot as a percentage of the maximum possible plot
width. Only affects graphics printing.
Top Graphic Margin % The top margin of the plot as a percentage of the maximum possible plot
height. Only affects graphics printing.
Bottom Graphic Margin % The bottom margin of the plot as a percentage of the maximum possible plot
height. Only affects graphics printing.

Item Description
Left Text Margin The left margin to be used when printing text files, measured in average
character widths. Note that when printing text files in portrait mode, this is the
left margin, allowing room for hole punching and binding of the printed page.
Top Text Margin The top margin to be used when printing text files, measured in lines. Note
that when printing text files in landscape mode, this is the left margin,
allowing room for hole punching and binding of the printed page.
Discussion:
The "Plot Width" control is different from most other settings in ZEMAX, because it is actually used to tell
ZEMAX how wide the plots are rather than how wide they should be. Every printer will print ZEMAX graphics at
a different size. To get accurate scaling on layout plots and scale bars, ZEMAX must be informed as to how big
the plots are when printed. With this information, ZEMAX can print exactly at 1:1, or 2:1, and so on.
Pressing the "Plot Width" button will invoke the Print dialog box. This is the same box that is displayed when
printing graphics from ZEMAX. This box allows selection of the printer driver, and usually allows selection of
printer specific items such as resolution, orientation, and perhaps other options which vary from printer to printer.
Using the Print dialog box, select the driver and mode you normally like to print in. Note ZEMAX prints graphics
in "landscape mode" by rotating the plot 90 degrees and then using the portrait setting. This is done because
virtually all printers default to portrait mode. It is therefore better to stay in portrait mode all the time, and use the
rotate plots feature. ZEMAX will use these settings to determine how wide the actual plot will be when printed,
and place this width, in inches, in the "plot width" edit box. Note that the plot width is automatically recalculated
for the default printer settings whenever the print orientation or margins are adjusted, or the "Reset" button is
pressed.
Once the actual plot width has been computed, the scaling on layout plots will be very accurate. However, it
will only be accurate if the same printer driver and mode settings are actually used for printing. If a different printer
driver or mode is selected at print time, then the plot width scaling will not automatically be recalculated. To get
correct scaling when printing to a new printer or using a different printer mode, the plot width environment variable
must be reset using the procedure described above.
Lastly, there are times when it may be required to override the default settings for the plot width. For example,
if the final printed output is to be reduced to a certain size for inclusion in another document, then the known final
size can be used to get exact scaling at the final image size. To accomplish this, enter the known final image width
in inches in the plot width edit box, and press Save. All subsequent printing will be done assuming the final image
size specified.
Note that the precise scaling controls specified only affect the Layout and Element Drawing features, since all
other plots are scale independent.
Colors
This dialog box is used to define the colors used for drawing graphics and for the background color of rows in
the various spreadsheet editors. Each color is used by the corresponding wavelength or field position when
graphing ray fan data, spot diagrams, and other data. For example, typically wavelength or field number 1 uses
color 1, field or wavelength 2 uses color 2, and so on. The Red, Green, and Blue values define the color. Each
entry must be between 0 and 255. Colors are defined using the RGB values for 24 bit color (16 million colors),
but will only display up to the resolution provided by the graphics hardware present. The resulting colors are
shown to the right of each RGB value. When using the defined colors in the spreadsheet editors, ZEMAX
automatically blends the color with white to make the foreground text more readable. The Highlight Color is the
color ZEMAX uses to draw lens surfaces and NSC objects currently selected in the LDE and NSCE, respectively.
Buttons 1-16, Buttons 17-32, Buttons 33-48
Purpose:
These settings determine which functions are available from the button bar at the top of the ZEMAX main
screen. The buttons are only user-definable in sequential mode. In NSC mode, ZEMAX uses a fixed set of button

definitions because many of the sequential button features are unavailable in NSC mode. See Program Mode
on page 57.
Discussion:
There are 48 buttons which can be set to invoke any of the ZEMAX main menu features. Each button has an
identical list of menu options which can be associated with the button. Selecting "Off" will cause the button to
disappear. To make the interface more manageable, the available buttons are arranged in groups of 16 on
separate tabs.
ZPL macros may be assigned to buttons. The name on the button will be the first 3 letters of the macro name.
For more information on ZPL macros, see ZEMAX PROGRAMMING LANGUAGE on page 535. ZPL macros
must be placed in the appropriate directory or sub directory, see Directories on page 59.
Status Bar
Purpose:
These settings determine which values are displayed in the status bar at the bottom of the ZEMAX main
screen.
Discussion:
There are 4 regions which can be set to display various data, such as EFL, EPD, F/#, etc.
Exit
Purpose:
Exits ZEMAX.
Discussion:
If the lens has been modified, ZEMAX will prompt to save the lens. If not, the program terminates.
Recently used files
The most recently used lens files are listed at the bottom of the file menu. Selecting one of these files from the
list will cause the file to load. This is a shortcut for using the File, Open command.

Chapter 5 EDITORS MENU
Lens Data
The Lens Data Editor is the primary spreadsheet where the majority of the lens data will be entered. This data
includes the radius of curvature, thickness, and glass for each surface in the system. Single lenses are defined
by two surfaces (a front and a back), and the object and image also each require one surface. This fundamental
data can be entered directly on the spreadsheet. When the Lens Data Editor is displayed, data can be entered
on the spreadsheet by typing in the required values at the highlighted bar. Each column is labeled by the data
type, and each row represents a single optical (or perhaps a dummy) surface. The cursor keys move the
highlighted bar to whichever column is desired. Moving the cursor continuously to the right or left causes the
display to scroll, which provides access to the other columns of data such as semi-diameters, conic constants,
and "parameter" terms whose values depend upon the surface type. The display will scroll left to right or right to
left. The page up and page down keys will move the bar to the top and bottom of the display. The display will also
scroll up and down if required, when the number of lens surfaces is sufficiently large.
Inserting and deleting surfaces
Note that initially (unless a lens has been loaded) three surfaces are shown; the object, stop, and image
surfaces. The object and image surfaces are permanent, and cannot be deleted. However, other surfaces can
be inserted and deleted by using the Insert and Delete keys. No surface can be inserted in front of the object, or
behind the image. In this context "in front of" means a smaller surface number, and "behind" means a larger
surface number, in the sense that the light reaches the various surfaces sequentially. ZEMAX numbers surfaces
from the object, being surface zero, upwards to the last surface, being the image surface.
To enter values in the spreadsheet, move the cursor to the correct cell, and start typing. To edit values currently
displayed, press the backspace key. Once you are editing the cell contents, you can use the left/right cursor keys
and the home and end keys to navigate around in the text. When the data has been changed, press any cursor
key, or click on any other cell, or press enter.
There are also a few shortcuts available. To add a value to the current value, type a plus sign before the
number. For example, if the displayed number is 10, typing "+5" and pressing enter will change it to 15. The
symbols "*" and "/" also work. To subtract a value, use the minus sign followed by a space. For example entering
"- 5" would change a 17 to a 12. Note the space between the "-" and the "5". If no space was entered, the program
assumes you are entering a new value which happens to be negative. Entering "*-1" will change the sign of the
number.
The surface number display
The left most column of the Lens Data Editor displays the surface number and type for each surface. The
surface number is zero for the object surface, 1 for the first surface, and so on to the image surface. Three
"special" surfaces, the object, stop, and image, display "OBJ", "STO", and "IMA", respectively, instead of the
surface number. The object surface is always surface zero, and the image surface is always the last surface,
however the stop surface may be any surface number in between.
Some additional information is displayed adjacent to the surface number. When an aperture is defined on a
surface, ZEMAX will display an asterisk "*" symbol next to the surface number. If surface tilt/decenter data is
defined on a surface, the "+" symbol will be displayed. If both aperture and tilt/decenter data are defined the "#"
symbol is displayed.
To define the surface type and other properties, see The Surface Properties dialog box on page 66. To define
the stop surface, see Make Surface Stop on page 67. For information on defining surface aperture data, see
Aperture type and other aperture controls on page 68. For information on defining tilt/decenter data, see
Surface tilt/decenter tab on page 73.
Cutting, copying, and pasting surface data
See the discussion on the "Edit" menu below.
65 Chapter 5: EDITORS MENU

Entering surface comments
Each surface has a comment field which can be used to enter text. The comment column is used for improving
the readability of the lens prescription, and has no affect on ray tracing. Some analysis features also display the
surface comments. The entire comment column can be hidden, see the "Options" menu description below.
Entering radii data
To enter or change the radius of curvature of a surface, move the cursor to the desired cell, and type in the
new value. Radii data is always entered and displayed in lens units, which have dimensions of length.
Entering thickness data
To enter or change the thickness of a surface, move the cursor to the desired cell, and type in the new value.
Thickness data is always entered and displayed in lens units, which have dimensions of length. The thickness of
a surface is the distance to the next surface. The only thickness that is not used is the thickness of the image
surface.
Thicknesses always change sign after a mirror. After an odd number of mirrors, all thicknesses should be
negative. This sign convention is independent of the number of mirrors, or the presence of coordinate breaks.
This fundamental convention cannot be circumvented through the use of coordinate rotations of 180 degrees.
Entering glass data
The material used for each surface is usually specified by entering the name of the glass in the "Glass" column
on the Lens Data Editor. The glass name entered must be in one of the currently loaded glass catalogs. The
default catalog is the "Schott" catalog; others are available. To use multiple glass catalogs, or to review, edit, or
append the glass catalogs, see the Chapter "Using Glass Catalogs". To specify that a particular surface be a
mirror, use the glass name "mirror".
There is an optional "/P" command that may be appended to the glass name when entering a new glass. This
option will cause ZEMAX to alter the curvatures before and after the surface to maintain a constant power at the
front and back vertices of the surface. For example, if the glass type is already BK7, entering a new glass type
of "SF1/P" will change the glass type to SF1 while adjusting the radii of the surface and the surface after, to
maintain constant power. ZEMAX keeps the power at the vertices constant, however, the overall lens power will
change slightly due to the change in the optical thickness of the glass. This effect is generally very small for thin
lenses.
Entering semi-diameter data
The default semi-diameter is automatically calculated to be the radial clear aperture required to pass all rays
from all field points. If any value is entered for the semi-diameter, then the new number will be retained and a "U"
will be placed next to the value. This indicates the semi-diameter is user defined. For information on the use of
semi-diameters, see Semi-diameters on page 52. For information on vignetting rays with apertures, see
Surface properties aperture tab on page 68.
Entering conic data
Conic data is permitted on many different surface types. To enter or change the conic constant of a surface,
move the cursor to the desired cell, and type in the new value. Conic constants are always dimensionless. See
Standard on page 270 for a discussion of how conic surfaces are defined.
Entering parameter data
Parameter data consists of additional numeric values which define the properties of certain surface types. For
more information on parameter data, See the Chapter "Surface Types".
The Surface Properties dialog box
Double clicking on the left most column (the one with the surface numbers listed) on the surface will invoke
the surface properties dialog box. The previous/next surface buttons allow rapid navigation through all surfaces.
The following surface properties may be defined on this dialog box.
Chapter 5: EDITORS MENU 66

Surface properties type tab
Surface Type
ZEMAX models planes, spheres, and conics; all of these surface types are grouped under the category of
standard surface. On the surface properties dialog box is a list of surface types. Select the appropriate surface
type from the drop down list. ZEMAX supports many different types of surfaces in addition to the standard surface.
The types are discussed in detail in the Chapter "Surface Types". Many optical designs only use the standard
surface type.
Surface DLL
If the surface type is "User Defined" then the surface shape and ray trace properties are defined in an external
program linked into ZEMAX called a Dynamic Link Library, or DLL. This control selects which DLL the surface
uses. See the Chapter "Surface Types" under "User Defined" for more information.
Surface Color
By default, on the Shaded Model Layout plot, ZEMAX draws mirrors in green, and refractive and dummy
surfaces in blue. The color of the surface as drawn on shaded models may alternatively be selected to be any of
the colors defined on the Colors tab of the File, Preferences dialog box.
Surface Opacity
If the opacity is set at 100%, then the surface will be rendered on the shaded model plot as a solid color, and
the surface may fully obscure other surfaces from view. If the opacity is less than 100%, then the surface is
partially transparent, which allows other surfaces to be visible through the partially obscure surface.
Row Color
This control chooses the color of the row in the Lens Data Editor for the surface. By default, glass surfaces,
coordinate breaks, mirrors, and paraxial lenses are color coded. Any surface may use either no color, the default
color, or a user defined color. The user defined colors are described in Colors on page 63. The coloring of rows
may be disabled, see Editors on page 61. The entire spreadsheet may be set to show default or no colors, see
Menu options on page 74.
Make Surface Stop
The stop surface may be any surface in the system, except for the object and image surfaces. To change the
stop surface, select the "Type" tab and then click on the checkbox labeled "Make surface stop" and press OK.
The dialog box will disappear and the surface now will display the "STO" label instead of the surface number. This
control is grayed out if the surface is the object, image, or is already the stop surface.
It is important to define the stop surface such that the entrance pupil is on the same axis as the object surface.
You can ensure this condition by placing the stop surface of the optical system before any coordinate breaks,
obscuration decenters, holograms, gratings, or other components which can alter the optical axis. If your system
is symmetric for rotations about the optical axis, then this limitation does not apply. Only systems that use surfaces
which can tilt or decenter the optical axis should have the stop placed before any such surfaces. If coordinate
breaks are used, but only to implement fold mirrors in an otherwise axial system, then the pupil locations will be
correctly computed even if the stop is placed after the fold mirrors.
In certain systems it is not possible to place the aperture stop before coordinate breaks. In this case, ray aiming
must be used. Ray aiming is discussed in Ray Aiming on page 88.
Make Surface Global Coordinate Reference
The global coordinate reference surface may be any surface in the system. To make the currently selected
surface the global coordinate reference surface, choose the "Type" tab and check this box on. See Global
Coordinate Reference Surface on page 93.
Surface Cannot Be Hyperhemispheric
ZEMAX normally detects if a surface must be hyperhemispheric (the surface continues past the maximum
possible radial aperture and curves back toward the vertex to fill more than a hemisphere) to pass all rays. If this
option is checked, the surface is not allowed to become hyperhemispheric. This switch should be used in
conjunction with a floating or circular aperture to vignette any rays which intercept beyond the desired aperture.

Surface properties draw tab
Hide Rays To This Surface
If checked, no rays will be drawn to the surface in the various types of layout plots. See Skip Rays below for
an alternate means of hiding rays to dummy surfaces.
Skip Rays To This Surface
If checked, rays will skip the surface. This means ray segments will be drawn from the last surface that was
not a coordinate break and did not have the Skip Rays option checked, all the way to the next surface that is not
a coordinate break and does not have the Skip Rays option checked. For example, if Skip Rays is selected on
surfaces 3, 4, and 5, rays will be drawn directly from the coordinates on surface 2 to their coordinates on surface
6, as long as surfaces 2 and 6 are not coordinate breaks.
The purpose of the Skip Rays option is to suppress the drawing of rays to and from dummy surfaces. Although
this can be useful in some cases, this option can cause strange, misleading layout plots if the skipped surfaces
can refract or diffract the rays. Therefore, the option should only be used on dummy (optically ineffective)
surfaces. ZEMAX does not check or verify that surfaces with the Skip Rays option are in fact dummy surfaces,
so care is required when using this feature.
Do Not Draw This Surface
If checked, the surface will not be drawn on layout plots.
Do Not Draw Edges From This Surface
If this option is checked, then no edges will be drawn from this surface to the next surface. This feature makes
for cleaner drawings in some systems, especially those using liquids or other non-air fillers between optical
surfaces.
Draw Edges As
This setting controls how edges that connect a surface to the next surface are drawn on the various layout and
element drawings. Usually this setting is only used to connect edges on surfaces with circular apertures defined
by the radial semi-diameter. Other cases may use another, usually simpler algorithm than the options described
here, or may not draw the edge at all. The available options are described below:
Squared To Next Surface: A flat face is drawn radially outward from the smaller of the current and next surfaces
to match the semi-diameter of the larger surface. The outer edge of the surface is then formed by a cylinder
extending to connect the surface edges.
Tapered To Next Surface: The current and next surface are connected by a single tapered cylinder.
Flat To Next Surface: The current surface edge is drawn as a cylinder at the current semi-diameter value
extending to the contact point on the next surface.
The 3D Layout, Wireframe, Shaded Model, and Solid Model plots currently do not support the "Flat To Next
Surface" option. Surfaces with this setting selected will be drawn using the "Squared To Next Surface" method.
Surface properties aperture tab
Aperture type and other aperture controls
Surface apertures are used to account for the effects of vignetting. The supported types of surface apertures
in ZEMAX are: no aperture, circular aperture, circular obscuration, rectangular aperture, rectangular obscuration,
elliptical aperture, elliptical obscuration, spider obscuration, user defined aperture, user defined obscuration, and
floating aperture. The aperture and obscuration types define regions which pass or stop rays, respectively. More
than one aperture may be described at a given optical element by inserting a dummy surface with zero thickness
at the desired location, and then setting additional apertures on that surface. This is useful for constructing
complex apertures. Multiple simultaneous apertures and obscurations may also be defined on a single surface
using the user defined apertures and obscurations feature.
Setting aperture data for each surface is done on the surface data dialog box. The surface data dialog box can
be reached by double clicking on the left most column in the Lens Data Editor. When "none" is selected for the
aperture type (the default), all rays which can be refracted or reflected by this surface are allowed to proceed. To

clear any aperture to this default state, or to change the current aperture type, select another aperture type on
the surface data dialog box.
The individual aperture types are described below.
Circular Aperture/Obscuration: A Circular Aperture defines an annular region which vignettes all rays which
strike the surface inside of the minimum radius, and outside of the maximum radius. If the ray is between the
minimum and maximum radii, then the ray will be allowed to proceed. The Circular Obscuration is the
complement of the Circular Aperture.
Rectangular Aperture/Obscuration: Rays are vignetted which intercept the surface outside the rectangular
region defined by the half widths in x and y. The Rectangular Obscuration is the complement to the Rectangular
Aperture.
Elliptical Aperture/Obscuration: Rays are vignetted which intercept the surface outside the elliptical region
defined by the half widths in x and y. The Elliptical Obscuration is the complement to the Elliptical Aperture.
Spider: The spider is defined by the width of each arm, and by the number of arms. ZEMAX assumes the arms
are all the same width, and that they are spaced in equal radial angles. The first arm starts at a radial position
of zero degrees, which is along the local positive x-axis. More complex spiders, which contain arms of different
widths at unequal angles can be constructed using several spiders on adjacent dummy surfaces. The
coordinate break surfaces can be used to rotate the spider(s) to any desired angle.
User Defined Aperture/Obscuration: See the discussion in the next section.
Floating Aperture: A floating aperture is very similar to the circular aperture, except the minimum radius is
always zero, and the maximum radius is always equal to the semi-diameter of the surface. Since the semi-
diameter value may be adjusted by ZEMAX (when in automatic mode) the aperture value "floats" as the semi-
diameter value. The floating aperture is useful when macros or external programs use ZEMAX to trace rays
that may lie outside of the default semi-diameters, and these rays are to be vignetted.
All of the preceding apertures are modeled as a surface projected from the vertex tangent plane onto the
optical surface. The actual ray-surface x and y intercept coordinates are used to determine vignetting, the z
coordinate is ignored. Different results may be calculated for steep optical surfaces if the aperture is placed on a
dummy surface in front of the optic, instead of being directly entered on the curved surface. This will only occur
if the ray incidence angles are steep. Usually it is best to place the apertures directly on the optical surface, unless
of course the dummy surface more accurately represents your situation.
All types of apertures may be decentered from the current optical axis by specifying either X- or Y-decenters,
or both. The decenters are given in lens units. It is important to remember that decenters do not move the chief
ray; the stop must be on the same axis as the object. To design an off axis telescope, for example, set the stop
on axis and decenter the system.
If the identical aperture is used on more than one surface, the aperture "Pick Up From" is a useful feature. Any
aperture may be picked up from any previous surface.
User defined apertures and obscurations
The circular, rectangular, and elliptical apertures and obscurations are easy to use and cover common cases.
However, there are times when a more general aperture model is required.
ZEMAX supports a general user specified aperture defined by a series of line segments, arcs, circles,
polygons, and rectangles. The aperture may be closed in a simple or complex way, and multiple aperture regions
may be defined which are either nested or not nested. To define a user defined aperture or obscuration, select
the desired type (aperture or obscuration) from the list of aperture types. The "Aperture File" control will list all
available user defined aperture (UDA) data files. The UDA files are text files that may be created and edited using
any text editor. The UDA files are stored in the \Objects subdirectory. The UDA file format is described in the next
section.
The button "Edit Aperture File" will invoke a text editor to allow user editing of the selected UDA file. The UDA
file needs to saved and the lens file updated to make the changes effective. The UDA Scale is a dimensionless
multiplier that scales the aperture defined in the UDA file. This control allows scaling of the UDA without need to
modify the UDA file.

The UDA file format
The UDA file consists of a series of simple text commands that define pieces of the boundary shape of the
aperture. The individual pieces are called entities. All UDA entities are defined as they lie on the XY plane. The
Z coordinate is determined by the sag equation of the surface to which the UDA is applied. The commands which
define the entities consists of three letter mnemonics followed by numerical values in ASCII format, stored in a
text file whose extension ends in UDA. The file name may not include any spaces.
In the entity definitions, all spatial coordinates are in lens units and all angles are in degrees. The available
entity types are described in the following table.
USER DEFINED APERTURE (UDA) ENTITIES
Entity Name/Syntax Description

ARC cx cy angle n Arc. Appends an arc to the current end point. The starting point of the arc is the
current end point, defined by a previous ARC or LIN command. The cx and cy
values are the center of the circle on which the arc lies. Note that the current end
point, along with cx and cy, define the radius of the arc. The angle (in degrees)
defines the length of the arc. Positive angles are clockwise on the XY plane.
ZEMAX internally represents the arc as a series of line segments. The parameter
n determines how many line segments are used to approximate the arc. A
recommended value is roughly 1 segment for each 6 degrees of arc. The
maximum allowed value is 64 segments.
BRK Break. Ends the current aperture definition, closing any previously defined
aperture. Entering the coordinates:
0.0 0.0
or
LIN 0.0 0.0
will also be interpreted as a break.
CIR cx cy radius n Circle. Defines a circle centered on cx and cy, with the specified radius. ZEMAX
draws the circle as a series of n line segments. The minimum value of n is 8. The
maximum value of n is 64, which is the default value if n is omitted. The actual
aperture applied to the rays is an exact circle, independent of the value for n.
ELI cx cy rx ry angle n Ellipse. Defines an ellipse centered on cx and cy, with the specified radial (half)
widths in the x and y directions. ZEMAX draws the ellipse as a series of n line
segments. The minimum value of n is 8. The maximum value of n is 64, which
is the default value if n is omitted. The actual aperture applied to the rays is an
exact ellipse, independent of the value for n.
The parameter angle (in degrees) may be used to rotate the ellipse clockwise to
any desired position.
LIN x y n Line. Appends a line segment from the current end point to the specified x, y
or coordinates. Use BRK to close a polygon formed by LIN commands. ZEMAX
xy optionally represents the line as a series of shorter line segments; this is useful
when the line lies on a steeply contoured part. The parameter n determines how
many line segments are used to construct the total line. If n is omitted then 1 line
segment is used.
The mnemonic LIN is not required; if two coordinates are listed with no
mnemonic, LIN is assumed. This is for backward compatibility with older
versions of ZEMAX.
POL cx cy radius n angle Polygon. Defines an "n" sided regular polygon centered on cx and cy, with each
vertex the specified radial distance from the center point. The POL command
may be used to defined triangles, squares, pentagons, hexagons, and other
polygons with equal length sides. The first point on the polygon starts at the angle
0.0 (along the local +x axis) and proceeds clockwise from there. The parameter
angle (in degrees) may be used to rotate the polygon to any desired position.
The maximum allowed number of segments is 64, the minimum is 3.

Entity Name/Syntax Description
REC cx cy xhw yhw angle Rectangle. Defines a rectangle centered on cx and cy with the specified half
nx ny widths in the x and y directions. The parameter angle (in degrees) may be used
to rotate the rectangle clockwise to any desired position. The optional arguments
nx and ny define how many segments to use in the x and y directions. Using a
larger number of segments will yield better rendering of NSC object surfaces that
are steeply curved over the rectangle aperture.
! comment Comment. Any line starting with the ! symbol is ignored, and may be used for
placing comments in the UDA.
The end of a polygon is indicated by either a BRK entity or a LIN entity with both X and Y being set to zero.
For this reason, a polygon may not be defined which has a vertex at (0,0). If a vertex must be defined at (0,0), a
work around is to instead use some very small value for one of the points, such as (1e-6, 0). As long as at least
one of the coordinates is not zero, the point will be considered a vertex rather than indicating the end of a polygon.
The last listed vertex in the polygon is assumed to then connect to the first point.
Note that the UDA entities REC and CIR define "stand alone" apertures not connected to any other aperture
entities, while LIN and ARC define pieces of a polygon whose definition ends with a BRK. When using LIN and
ARC entities, the first entity defined should be an LIN, which defines the first point of the polygon. Multiple
apertures may be defined separated by BRK entities.
The maximum number of line segments defined within the UDA file is limited only by available memory.
However, ray tracing and rendering of surfaces and objects is slower when using very complex UDA data files.
See the following sections for UDA examples.
UDA Examples
To define a square 20 lens units on the side, the UDA file is:
LIN -10, -10
LIN -10, 10
LIN 10, 10
LIN 10, -10
BRK
Note the BRK indicates the end of the polygon definition, and
the last point is assumed to join back to the first point, thus defining
the last side of the square. This same aperture may be defined with
a single REC command:
REC 0 0 10 10 0
Note the REC command does not require a BRK command to
define the end of the aperture.
Multiple polygons may be defined by separating them with a
single BRK. For example, to define an aperture consisting of two
slits, each 5 lens units wide, with an inner separation of 10 lens
units, the UDA file is:
LIN -10, -10
LIN -10, 10
LIN -5, 10
LIN -5, -10
BRK
LIN 10, -10
LIN 10, 10
LIN 5, 10
LIN 5, -10
BRK
Or the equivalent
REC -7.5 0 2.5 10 0
REC +7.5 0 2.5 10 0

To create a square with rounded edges, use the ARC and LIN
commands together. Here is a UDA that creates a square, 4 lens
units on a side, with rounded corners. Each rounded corner is
formed by one ARC with a radius of 1 lens unit and 12 segments
along the arc:
LIN -1 2
LIN 1 2
ARC 1 1 90 12
LIN 2 -1
ARC 1 -1 90 12
LIN -1 -2
ARC -1 -1 90 12
LIN -2 1
ARC -1 1 90 12
The first LIN command defines the starting point near the upper left corner. The subsequent commands define
each of the 8 parts of the aperture. Note a BRK command is not needed if only a single aperture region is defined.
Multiple polygons may also be nested. If a ray intercepts a point
within a polygon that is within another polygon, then the point is
considered to be outside the aperture. This convention allows
"islands" to be defined within apertures that become obscurations,
and vice-a-versa. Any number of nesting levels is allowed, and
each level toggles the inside/outside status of the point.
Here is a UDA with the rounded corner square from the previous
example, with 5 small sub apertures nested within; a pentagon, a
hexagon, an ellipse, a rectangle, and a circle. Note the BRK
command is not needed to separate the REC, ELI, POL, and CIR
entities from the ARC and LIN entities because REC, ELI, POL,
and CIR all define stand alone aperture regions:
LIN -1 2
LIN 1 2
ARC 1 1 90 12
LIN 2 -1
ARC 1 -1 90 12
LIN -1 -2
ARC -1 -1 90 12
LIN -2 1
ARC -1 1 90 12
REC 1 1 .3 .5 -30
ELI -1 -1 .6 .2 15
POL 1 -1 .5 5 0
POL -1 1 .5 6 0
CIR 0 0 .3
Surface properties scattering tab

Highly polished optical surfaces typically have a small amount of scattered light over a small, roughly angular
cone around the specular refracted or reflected ray direction. Surface scattering properties are useful for modeling
this small amount of scattered light, to analyze the effects of the scattering on MTF or other quality measures.
Scattering is modeled in ZEMAX by randomly deviating the refraction or reflection angle of some or all of the rays
leaving a surface.
There are two fundamentally different ways of modeling scattering in ZEMAX: sequential and non-sequential.
For the sequential model, the scattering is assumed to deviate the ray angle only very slightly. The primary
effect is a "blurring" of the spot diagram or ray pattern at the image surface. Sequential scattering is not

intended for modeling the effects of back scatter, wide angle scatter (such as Lambertian) or other scatter
where the scattered rays may not follow a well behaved sequential path to the image surface.
For the non-sequential model, scattering may occur over any range of angles, and rays will be traced correctly
no matter in what direction they propagate. The non-sequential model is the better choice if it is necessary to
model scatter from surfaces that are not part of the sequential ray path, such as from lens mounts, baffles, or
other objects. See the Chapter "Non-Sequential Components" for details on this method of modeling scatter.
It is not necessary to use the non-sequential components feature of ZEMAX to model scattering, if the system
is otherwise sequential and the only scattering of interest is small angle scatter from the small scale roughness
of the optical surfaces, and the only desired analysis is the degradation in image quality due to scattering.
Surface scattering settings
ZEMAX uses the same terminology and scattering models for both sequential surfaces and non-sequential
components. See the Chapter "Non-Sequential Components" on page 353 for a detailed technical description of
the scattering models. For information on defining ABg data, see ABg Scatter Data Catalogs on page 203.
The only differences between the non-sequential scattering and the sequential scattering models are:
The non-sequential ABg scattering model supports both reflection and refraction coefficients, while the
sequential ABg scattering model uses a single set of coefficients; since sequential surfaces either reflect or
refract, but not both. Whether a sequential surface reflects or refracts depends only upon the whether the
surface is a mirror or not.
Although Lambertian is a valid option on the sequential surface dialog, rays which scatter using this model
may go in any forward direction, which may cause rays to scatter at large enough angles so that they do not
correctly propagate through the rest of the optical system.
Only some ZEMAX analysis features use the scattering data. Most features, such as optimization and layout
diagrams, ignore surface scattering. Features that use the scattering data, such as the spot diagrams have a
"Scatter Rays" checkbox.
Surface tilt/decenter tab
Surface tilts and decenters allow a change in the coordinate system to be implemented both before and after
ray tracing to the surface. Applications include decentering a surface and returning to the original coordinate
system, tilting a fold mirror and tilting again to follow the beam, tilting a surface to model a wedge, and many
others. Surface tilts and decenters are redundant with and very similar to coordinate breaks. For detailed
information on coordinate breaks, see Coordinate Break on page 239. A surface tilt/decenter can be thought of
as a coordinate break, followed by the surface, followed by another coordinate break.
The advantage to using surface tilts and decenters is the elimination of "dummy" coordinate break surfaces in
the Lens Data Editor. This allows for a somewhat less cluttered display some users prefer. The disadvantage of
using the surface tilts and decenters is the current implementation does not support optimization of surface tilt
and decenter data.
The operation of surface tilts and decenters is done in a sequence:
Before the surface, the coordinate system is decentered in X, decentered in Y, and then tilted around X, tilted
around Y, tilted around Z. Decenters are measured in lens units, tilts in degrees about the respective axis in a
right hand direction. Alternatively, the order may be chosen to tilt then decenter. In this case, the coordinate
system is tilted around Z, tilted around Y, tilted around X, decentered in Y then finally decentered in X.
!The order of the decenters and tilts matters!

After the surface, the same set of operations may be performed in either order. The values for the before and
after decenters and tilts, and the order in which they are done, may be independent. However, it is frequently
useful to have the after tilts and decenter values be related to previous before values. The options available are:
a) Explicitly define the tilts and decenters.
b) Pick-up the values from the before data of the current surface (typically used for fold mirrors)
c) Reverse the values from the before data of the current surface (sometimes called a decenter/ tilt and return)

d) Pick up the values from the before data of a prior surface
e) Reverse the values from the before data of a prior surface (typically used for decentering a range of
surfaces)
All of these options are supported by the "After Surface" settings. Reversing values from a prior surface
involves changing the order and picking up the tilt and decenter values from the target surface and reversing the
sign.
The coordinate system resulting from the before and after tilt and decenters will define the coordinate system
for the next surface. The thickness of a surface is the thickness in the new coordinate system after all the tilt and
decenters are applied, measured along the resulting Z axis.
For more information on coordinate transformations, see Coordinate Break on page 239.
Surface physical optics tab
See the chapter PHYSICAL OPTICS PROPAGATION on page 515 and Surface specific settings on
page 529 for a description of these features.
Surface coating tab
The coating tab allows selection of the optical coating to be applied to the surface. For information on defining
coatings, see Defining coatings in ZEMAX on page 501 and Optimizing coatings with ZEMAX on page 508.
This tab supports an additional capability to individually modify the thickness of any layer of a defined coating
applied to a particular surface without changing the underlying definition of the coating. Individual layer thickness-
es may be scaled by a "coating multiplier", a dimensionless parameter. This parameter may be unique for every
layer and every surface, even if the same coating is applied to more than one surface. The available controls are:
Use Layer Multipliers: If unchecked, all coating layer multipliers are disabled for this surface. The coating
thickness will be as defined by the coating data file. If checked, the coating multipliers are considered.
Layer: Chooses the particular layer to edit or review the coating multiplier for.
Multiplier: The dimensionless multiplier for the selected layer.
Status: The multiplier may be a fixed value, a variable for optimization, or a value picked up from a prior layer.
Setting and removing solves
Most data columns (such as radii and thickness) support one or more solves. To set a solve on a cell, double
click with the left mouse button on that cell. Alternatively, single click with the right mouse button, or select the
menu option from the Lens Data Editor menu. Solves are described in SOLVES on page 379.
Setting and removing variables
To make a parameter variable, click on the cell containing the parameter, and press Ctrl-Z on the keyboard.
Ctrl-Z also removes the variable status, and acts like a toggle.
Menu options
The Lens Data Editor menu options are used for inserting and deleting surfaces, selecting surface types, and
setting solves and variables.
Edit
The edit menu offers these options:
Surface Type: This option allows the type of a surface to be changed.
Insert Surface: This inserts a new surface row in the spreadsheet at the current row. The Insert key is a shortcut
for this menu option.
Insert After: This inserts a new surface row in the spreadsheet after the current row. The Ctrl-Insert key
combination is a shortcut for this menu option.
Delete Surface: This deletes the current row in the spreadsheet. The Delete key is a shortcut for this menu
option.

Cut Surfaces: Copies all the data for a single surface or a range of surfaces to the Windows Clipboard, then
deletes the surfaces. The surface or range of surfaces must first be selected using either of the following
techniques:
Using the mouse: Click on the first surface to be selected. Hold down the left mouse button, and drag the
mouse cursor to cover the range of surfaces desired. The selected surfaces will be highlighted in inverse
colors. To select only one surface, drag the mouse up or down from the surface desired until two surfaces
are selected, then move the mouse back to the desired surface.
Using the keyboard: Move the cursor to any cell on the desired surface. Then hold down the shift key, while
moving the cursor up or down until the desired range of surfaces is selected. The selected surfaces will be
highlighted in inverse colors. To select only one surface, move the cursor up or down from the surface
desired until two surfaces are selected, then move the cursor back to the desired surface.
Copy Surfaces: Copies all the data for a single surface or a range of surfaces to the Windows Clipboard. To
select a single surface or a range of surfaces, see the discussion under "Cut Surfaces" above. Non-Sequential
Component surfaces may not be copied.
Paste Surfaces: Copies all the data for a single surface or a range of surfaces from the Windows Clipboard to
the current cursor location in the Lens Data Editor. The surface data must first have been copied to the
Windows Clipboard using either "Cut Surfaces" or "Copy Surfaces" described above.
Copy Cell: Copies a single cell's data to the Windows Clipboard.
Paste Cell: Pastes the data for a single cell from the Windows Clipboard to the current cell. The data must first
have been copied to the Windows Clipboard using "Copy Cell" described above.
Edit Cell: Puts the cell in edit mode.
Copy Spreadsheet: Copies either the highlighted range of surfaces, or the entire spreadsheet (if no range of
surfaces is selected) to the Windows Clipboard in a text format suitable for being pasted into another Windows
application, such as a spreadsheet or word processor. The format is tab delimited text. When using this feature
ZEMAX uses curvature rather than radius to retain maximum numerical precision.
Set Row Colors: Selects either default or none for the color for each row in the spreadsheet. Individual row
colors may be set using the row color setting, see Row Color on page 67.
Solves
Solves and variables may be placed on many Lens Data Editor values:
Radius: Sets solves on the radius of curvature.
Thickness: Sets solves on the thickness.
Glass: Sets solves on the glass.
Semi-Diameter: Sets solves on the semi-diameter.
Conic: Sets solves on the conic constant.
Parameter: Sets solves on the parameter columns.
Variable Toggle: This will toggle the variable status of the currently selected cell. Ctrl-Z is a shortcut for this
operation.
Options
Show Comments: If this menu option is checked, then the comment column is displayed. If it is unchecked,
then the comment column is hidden. The hide/show status of the comment column is only used for the current
session. To set this option automatically at the start of a ZEMAX session, see the Chapter "File Menu" under
"Preferences".
Help
Opens the help system for using the Lens Data Editor.

Merit Function
The Merit Function Editor is used to define, modify, and review the system merit function. The system merit
function is used for optimization, and is described in the Chapter "Optimization".
Edit
Operand Type: This option allows the type of operand and other data to be changed. For a complete
description of merit function operands, see OPTIMIZATION on page 385.
Insert Operand: Inserts a new row in the spreadsheet at the current row. The Insert key will perform this same
function.
Insert After: Inserts a new row in the spreadsheet after the current row. The Ctrl-Insert key combination will
perform this same function.
Delete Operand: Deletes the row at the current cursor location. The Delete key will perform this same function.
Delete All: Deletes all operands in the merit function.
Cut Operands: Copies all the data for a single operand or a range of operands to the Windows Clipboard, then
deletes the operands. The operand or range of operands must first be selected using either of the following
techniques:
Using the mouse: Click on the first operand to be selected. Hold down the left mouse button, and drag the
mouse cursor to cover the range of operands desired. The selected operands will be highlighted in inverse
colors. To select only one operand, drag the mouse up or down from the operand desired until two operands
are selected, then move the mouse back to the desired operand.
Using the keyboard: Move the cursor to any cell on the desired operand row. Then hold down the shift key,
while moving the cursor up or down until the desired range of operands is selected. The selected operands
will be highlighted in inverse colors. To select only one operand, move the cursor up or down from the
operand desired until two operands are selected, then move the cursor back to the desired operand.
Copy Operands: Copies all the data for a single operand or a range of operands to the Windows Clipboard.
To select a single operand or a range of operands, see the discussion under "Cut Operands" above.
Paste Operands: Copies all the data for a single operand or a range of operands from the Windows Clipboard
to the current cursor location in the Merit Function Editor. The operand data must first have been copied to the
Windows Clipboard using either "Cut Operands" or "Copy Operands" described above.
Copy Spreadsheet: Copies either the highlighted range of operands, or the entire spreadsheet (if no range of
operands is selected) to the Windows Clipboard in a text format suitable for being pasted into another Windows
application, such as a spreadsheet or word processor. The format is tab delimited text.
colors may be set using the row color setting on the operand type dialog box.
Tools
Update: This option recomputes the merit function. All operands are evaluated, and the new values are
displayed.
Default Merit Function: Invokes the dialog box which used to define one of the default merit functions. See the
Chapter "Optimization".
Save: Saves the current merit function in a *.MF file. This step is only required if the merit function is to be
subsequently loaded into another lens. ZEMAX automatically stores the merit function with the lens when the
entire lens is saved.
Load: Loads a merit function previously stored in a *.MF file or in a *.ZMX file. Either file type may be selected;
only the merit function portion of the file will be loaded into the spreadsheet. The current merit function is

destroyed.
Help
Opens the help system.
Multi-Configuration
The Multi-Configuration editor is very similar to the Lens Data Editor. To edit the contents of a cell, move the
cursor to that cell and type in the new data. To set a solve on a cell, double click with the left mouse button, or
select the solve type menu option.
Edit
Operand Type: This option allows the type of multi-configuration operand to be changed. For a complete
description of multi-configuration operands, see MULTI-CONFIGURATIONS on page 471.
Insert Operand: This inserts a new row in the spreadsheet at the current row. The new operand type is "OFF"
which means the operand is ignored. The Insert key is a shortcut for this menu option.
Insert After: This inserts a new row in the spreadsheet after the current row. The new operand type is "OFF"
which means the operand is ignored. The Ctrl-Insert key combination is a shortcut for this menu option.
Delete Operand: This deletes the current row in the spreadsheet. The Delete key is a shortcut for this menu
option.
techniques:
to the current cursor location in the Multi-Configuration Editor. The operand data must first have been copied
to the Windows Clipboard using either "Cut Operands" or "Copy Operands" described above.
Insert Configuration: Selecting this item will insert a new column, which corresponds to a new configuration.
Delete Configuration: Deletes the configuration indicated by the current position of the cursor. This eliminates
the column and all the data it contained.
Solves
Solve Type: Selecting this option will invoke the solve dialog box for the cell which is currently indicated by the

cursor.
Variable Toggle: This will toggle the variable status of the currently selected cell.
Tools
Auto Thermal: This tool is used to perform the bulk of the tedious work in setting up a multi-configuration
thermal analysis. A dialog box will appear which allows setting of the number of configurations, as well as the
minimum and maximum temperatures. There are also options to delete or retain the existing multi-configura-
tion data, and to sort the data by surfaces rather than the default sort by operand type. Be sure to read and
understand how the thermal analysis works in ZEMAX. For more information, see THERMAL ANALYSIS on
page 491 and Limitations of thermal analysis on page 496.
If the "Delete existing configuration data" option is selected, then 1 nominal configuration at the current
temperature and pressure will be created. Additional configurations are then defined to cover the temperature
range specified. If 3 configurations are requested, then there will be 1 nominal configuration (configuration 1),
and 3 configurations that span the specified temperature range in equal increments, for a total of 4 configura-
tions. The air pressure will be assumed to be the same as the nominal pressure.
If the "Use existing n configurations as nominal" option is selected, then the existing MCE data will be used to
create the nominal configurations. The tool will automatically add new operands for other data typically
required for thermal modeling, such as glass curvature, semi-diameters, parameter, and extra data editor
values. In this mode, the total number of configurations will be determined by the number of nominal
configurations times the number of new configurations, plus the original nominal configurations.
For each radius, thickness, glass, semi-diameter, parameter, and extra data value affected by temperature
effects, appropriate operands will be inserted with TCE pickup solves. It is always good engineering practice
to check the results of this automatic thermal set up tool carefully to make sure no important parameters have
been omitted.
!Check the results of the automatic thermal set-up tool carefully!
Make Single Config: This tool will delete all the multi-configuration data, leaving the lens as a single
configuration system in whatever configuration was active at the time this tool was selected.
Add All Data: This tool will add to the MCE all data currently in the system. This eliminates the need to add
many data items manually. For some systems, it is easier to delete unwanted data than to insert desired data.
Help
Tolerance Data
The Tolerance Data editor is used to define, modify, and review the system tolerance values. See the Chapter
"Tolerancing" for details.
Edit
Operand Type: This option allows the type of operand and other data to be changed. For a complete
description of tolerance operands, see TOLERANCING on page 441.
Insert Operand: Inserts a new row in the spreadsheet at the current row. The Insert key will perform this same
function.
Insert After: Inserts a new row in the spreadsheet after the current row. The Ctrl-Insert key combination will
perform this same function.
Delete Operand: Deletes the row at the current cursor location. The Delete key will perform this same function.
techniques:

to the current cursor location in the Tolerance Data Editor. The operand data must first have been copied to
the Windows Clipboard using either "Cut Operands" or "Copy Operands" described above.
Tools
Default Tolerances: Invokes the default tolerances dialog box. See Defining default tolerances on page 452
for details.
Loosen 2X: Increases all tolerance ranges by a factor of 2. This is a quick way of loosening the tolerances if
they are all too tight.
Tighten 2X: Decreases all tolerance ranges by a factor of 2. This is a quick way of tightening the tolerances if
they are all too loose.
Sort by Surface: Sorts all operands in ascending order first by surface number, then by type. The operands
COMP and CPAR will always be placed at the top of the list. The SAVE operand will automatically be moved
to stay just beneath the same operand it was beneath before sorting, because the SAVE operand is associated
with the previous operand in the list. The STAT operand, if present, will be placed at the top of the list, and this
operand must manually be moved or reinserted. Since STAT affects all operands which follow in the list, sorting
the list will invalidate the STAT operand. Whenever STAT is used in the body of the tolerance list (to change
the statistics on the fly) then editing of the list to correctly replace the STAT operand will be required whenever
sorting. Note multiple STAT operands may be required if the operands that originally followed the STAT
operand are dispersed through the list by the sort operation.
Sort by Type: Sorts all operands in ascending order by type, then by surface number. See Sort by Surface.
Save: Saves the current tolerance data in a *.TOL file. This step is only required if the data is to be subsequently
loaded into another lens or stored for archival purposes. ZEMAX automatically stores the tolerance data with
the lens when the entire lens is saved.
Load: Loads tolerance data previously stored in a *.TOL file or in a *.ZMX file. Either file type may be selected;
only the tolerance data portion of the file will be loaded into the spreadsheet. The current tolerance data is
destroyed.
Help
Extra Data

The extra data editor is very similar to the lens data spreadsheet editor, except only the extra data values are
displayed and edited. See Extra data on page 225 for a discussion of the surface types which use extra data
values. Surfaces cannot be inserted or deleted on the extra data editor.
Edit
Solves
Solve Type: Selecting this option will invoke the solve dialog box for the cell which is currently indicated by the
cursor.
Variable Toggle: This will toggle the variable status of the currently selected cell.
Tools
Import: The import tool is used to load extra data values for extra data surfaces from an ASCII file rather than
by typing the numbers in directly. This menu option will invoke a dialog box which presents a list of ASCII data
files which end in the extension .DAT. The dialog box also permits specification of which surface number should
receive the data. Numerical data must be in the ASCII file exactly as it appears in the extra data spreadsheet.
The format of the ASCII file is a single column of free-format numbers, and the file must end in the DAT
extension. ZEMAX will look for the file in the default directory for ZEMAX specified on the Directories tab of
the File, Preferences dialog box. Data for grid sag and phase surfaces may also be imported using this tool.
For a description of the proper file format, see Importing grid data on page 259.
Help
Opens the help system for using the Extra Data Editor.
Non-Sequential Components
The NSC Editor is only used by the Non-Sequential Components surface type supported by ZEMAX-EE. The
editor is very similar to the lens data spreadsheet editor. See NON-SEQUENTIAL COMPONENTS on page 289
for a detailed discussion of non-sequential components and ray tracing. The editor can only be activated if a Non-
Sequential Components surface is present in the Lens Data Editor, or if the program mode is set to Non-
Sequential Design. For information on changing the program mode, see Program Mode on page 57.
Menu options
The Non-Sequential Components Editor menu options are used for inserting and deleting objects, selecting
object types and properties, and setting solves and variables.
Edit
Object Properties: Activates a dialog box from which the object type, surface and bulk properties may be
selected. For complete documentation on the extensive settings available, see The object properties dialog
box on page 348.
Next Group: If there is more than one Non-Sequential Components surfaces defined in the Lens Data Editor,
selecting this option will advance to the group of objects for the next non-sequential surface.
Edit Object: If the object is defined by an ASCII file, such as a Polygon Object (POB), then this menu option
will be available. If selected, a text editor will be invoked so that the defining data file may be edited. After
editing and saving the file; choose Reload Object to update the new information in ZEMAX.
Reload Object: This reloads and recreates an object that was defined by an external data file that was changed
since the lens file was last loaded. Typically this option is used to refresh a POB or STL object if another
application has modified the defining data file.

Reload All Objects: This reloads and recreates all objects listed in the editor, see "Reload Object" above.
Insert Object: This inserts a new object row in the spreadsheet at the current row. The new object type is "Null
Object". The Insert key is a shortcut for this menu option.
Insert After: This inserts a new object row in the spreadsheet after the current row. The new object type is "Null
Object". The Ctrl-Insert key combination is a shortcut for this menu option.
Delete Object: This deletes the current object row in the spreadsheet. The Delete key is a shortcut for this
menu option.
Cut Objects: Copies all the data for a single object or a range of objects to the Windows Clipboard, then deletes
the objects. The object or range of objects must first be selected using either of the following techniques:
Using the mouse: Click on the first object to be selected. Hold down the left mouse button, and drag the
mouse cursor to cover the range of objects desired. The selected objects will be highlighted in inverse
colors. To select only one object, drag the mouse up or down from the object desired until two objects are
selected, then move the mouse back to the desired object.
Using the keyboard: Move the cursor to any cell on the desired object. Then hold down the shift key, while
moving the cursor up or down until the desired range of objects is selected. The selected objects will be
highlighted in inverse colors. To select only one object, move the cursor up or down from the object desired
until two objects are selected, then move the cursor back to the desired object.
Copy Objects: Copies all the data for a single object or a range of objects to the Windows Clipboard. To select
a single object or a range of objects, see the discussion under "Cut Objects" above.
Paste Objects: Copies all the data for a single object or a range of objects from the Windows Clipboard to the
current cursor location in the NSC Editor. The object data must first have been copied to the Windows
Clipboard using either "Cut Object" or "Copy Objects" described above.
Copy Spreadsheet: Copies either the highlighted range of objects, or the entire spreadsheet (if no range of
objects is selected) to the Windows Clipboard in a text format suitable for being pasted into another Windows
colors may be set using the row color setting on the object properties dialog box.
Solves
Solves and variables may be placed on X, Y, and Z positions, tilts about the X, Y, and Z axis, and any of the
parameter data:
X Position: Sets fixed, variable, or pickup solves on the X position of an object.
Y Position: Sets fixed, variable, or pickup solves on the Y position of an object.
Z Position: Sets fixed, variable, or pickup solves on the Z position of an object.
Tilt about X: Sets fixed, variable, or pickup solves on the tilt about the X axis of an object.
Tilt about Y: Sets fixed, variable, or pickup solves on the tilt about the Y axis of an object.
Tilt about Z: Sets fixed, variable, or pickup solves on the tilt about the Z axis of an object.
Material: Sets fixed or pickup solves on the material of an object.
Parameter: Sets fixed, variable, or pickup solves on any parameter of an object. Pickup solves on parameters
are only allowed if the source object is a preceding object, or if the source object is the same object and the
source parameter column is to the left of the destination column.
Variable Toggle: This will toggle the variable status of the currently selected cell. Ctrl-Z is a shortcut for this
operation.

Errors
Ignore Errors: If checked, ray trace errors that occur inside of a non-sequential group will not be reported if
detected. These errors involve possibly improper placement of objects in a non-sequential group. However, in
some systems it is possible to have properly defined object placement and still occasionally an error message
will be generated. Checking this option will suppress reporting of these error messages.
Detectors
Ray Trace/Detector Control: Opens up the dialog box which controls clearing the detectors and tracing rays
from NSC sources. See the chapter "Non-Sequential Components" for details.
Detector Viewer: Opens up a detector viewer window. A detector viewer can show the data recorded by any
defined detector.
Database
Ray Database Viewer: Opens up a ray database viewer.
Help
Opens the help system for using the NSC Editor.
Undo, Redo, and Recover
There are three different states to the Undo capability in ZEMAX: None, Memory 1 Step, and Disk Multi Step.
The Undo state is set on the Editors tab of the Preferences dialog box described in the Chapter "File Menu".
Undo: None
If the Undo feature state is set to None, then no Undo capability is supported. Use this option on computers
which do not have sufficient system memory or disk space to support the Undo feature.
Undo: Memory 1 Step
ZEMAX stores a copy of the current lens in memory before and after every edit or optimization. If Undo is
selected, then the current lens is swapped with the previous lens. If Redo is then selected, the lenses are swapped
again, which results in the edit being restored.
Memory 1 Step Undo is useful for restoring a lens when an accidental edit is made, or for restoring a lens to
the prior state after an optimization. However, only a single Undo step is supported. The advantage of this option
is speed; the saving of the prior lens in memory is so fast as to be unnoticeable.
Undo: Disk Multi Step
ZEMAX stores a copy of the current lens in a ZMX file on disk after every edit or optimization. These stored
lens files are used to implement an infinite multi-step Undo feature, which allows the reversal of any change or
series of changes made to the lens. Undo is useful for restoring a lens when an accidental edit is made, or for
restoring a lens to the prior state after an optimization, or even several changes.
To reverse the changes made to a lens, select Undo from the Editors Menu. Any number of Undo's may be
executed, back to the first edit made after loading the lens file. The Redo function reverses the last Undo.
ZEMAX maintains a directory of Undo files, which by default is a subdirectory within the main ZEMAX directory
called \UNDO. The Undo files are automatically deleted whenever the file is saved, a new file is opened, or
ZEMAX is normally terminated. If ZEMAX abnormally terminates, the operating system fails, the computer power
is disrupted, or for any other reason the lens data is lost, ZEMAX may be able to recover the lost data by restoring
the last Undo file. When ZEMAX starts, a check is made to see if any of the Undo files exist. Since these files are
deleted during a normal termination, the presence of Undo files indicates a previous abnormal termination.
ZEMAX will issue a warning message with the option of restoring the last Undo file. If restored, this file should
immediately be saved in a new file name, since the old file name is not stored with the lens.
The Undo feature does slow down the operation of the Editors slightly, since every edit is followed by a save
operation. The save does not slow down ray tracing or optimization speed, only lens data editing.
If more than one ZEMAX session is running simultaneously, each session will have its own undo files. However,
to recover from a crash or abnormal program termination, the same number of ZEMAX sessions will need to be

run to recover all the files. For example, if 2 sessions of ZEMAX are running, and the power fails, the first new
session of ZEMAX will be able to recover the former first session file. A second session of ZEMAX will need to
be launched to recover the former second session file.

Chapter 6 SYSTEM MENU
Update
This option only updates data in the Lens Data and Extra Data editors. Update is used to recompute first-order
properties, pupil positions, semi-diameters, index data, and solves. Only data presented in the Lens Data and
Extra Data editors are affected. See "Update All" in this chapter for more information.
Update All
This option updates all windows to reflect the most recent lens data. ZEMAX does not automatically change
all data presented in graphic and text windows to reflect the latest lens data. The reason is that if ZEMAX
constantly recalculated MTF, ray fan, spot diagram, and other data while new data is being typed in the lens data
editor, then program response would be very slow. Instead, make all the required changes to the lens, and then
select Update All to refresh and recompute all the data windows.
Individual graphic and text windows (not editors) can also be updated by double clicking anywhere within the
window.
General
This option invokes the General System Data dialog box, which is used to define commonly used data that
pertains to the lens as a whole system, rather than data associated with a single surface. The general data is
divided into groups as follows.
Aperture
The aperture group consists of the following settings.
Aperture Type
The system aperture defines the size of the beam through the system on axis. To set the system aperture, you
need to define the system aperture type and the system aperture value. Use the cursor keys to select from the
drop down list the desired system aperture type. The system aperture type can be any one of the following:
Entrance Pupil Diameter: The diameter of the pupil in lens units as seen from object space
Image Space F/#: The infinite conjugate paraxial F/# in image space
Object Space Numerical Aperture: The numerical aperture ( n sin m ) of the marginal ray in object space
Float By Stop Size: Defined by the semi-diameter of the stop surface
Paraxial Working F/#: The defined conjugate paraxial F/# in image space.
Object Cone Angle: The half angle in degrees of the marginal ray in object space, which may exceed 90
degrees. Object cone angle may not be used if the entrance pupil is virtual, that is, if the distance from the object
to the entrance pupil is negative. When using object cone angle, the normal apodization of rays in the pupil is
uniform in angle space, rather than uniform on a plane. This change is required to support angles greater than
90 degrees. This ray distribution may be significantly different than other aperture type settings for large cone
angles.
These terms are further defined in the Chapter "Conventions and Definitions". If you select "Object Space NA"
or "Object Cone Angle" for a system aperture type, the object thickness must be less than infinity. Only one type
of system aperture may be defined; for example, once the Entrance Pupil Diameter is specified, all the other
aperture definitions are determined by the lens data prescription.
Aperture Value
The system aperture value meaning depends upon the system aperture type selected. For example, if
"Entrance Pupil Diameter" is selected as the system aperture type, the system aperture value is the entrance
pupil diameter in lens units. ZEMAX uses the system aperture type and system aperture value together to
determine such fundamental quantities as the entrance pupil size, and clear apertures for all components.
85 Chapter 6: SYSTEM MENU

Selecting "Float by Stop Size" for a system aperture type is the only exception to this rule. The semi-diameter
of the stop surface (set on the lens data editor) is used to define the system aperture if "Float by Stop Size" is
selected for a system aperture type.
Apodization Type
By default, the pupil is always illuminated uniformly. However, there are times when the pupil should have a
non-uniform illumination. For this purpose, ZEMAX supports pupil apodization, which is a variation of amplitude
over the pupil. Three types of pupil apodization are supported: uniform, Gaussian, and cosine cubed. User defined
apodizations are also supported.
Uniform apodization
Uniform means rays are distributed uniformly over the entrance pupil, simulating uniform illumination. This is
usually the case for distant objects.
Gaussian apodization
Gaussian apodization imparts an amplitude variation over the pupil that is Gaussian in form. The apodization
factor refers to the rate of decrease of the beam amplitude as a function of radial pupil coordinate. The beam
amplitude is normalized to unity at the center of the pupil. The amplitude at other points in the entrance pupil is
given by
2
G
A() = e ,
where G is the apodization factor and is the normalized pupil coordinate. If the apodization factor is zero,
then the pupil illumination is uniform. If the apodization factor is 1.0, then the beam amplitude has fallen to the 1
over e point at the edge of the entrance pupil (which means the intensity has fallen to the 1 over e squared point,
about 13% of the peak). The apodization factor can be any number greater than or equal to 0.0. Values larger
than about 4.0 are not recommended. This is because most computations will sample too few rays to produce
meaningful results if the beam amplitude falls too quickly at the edges of the pupil.
Cosine cubed apodization
Cosine cubed apodization is appropriate to simulate the intensity fall off characteristic of a point source
illuminating a flat plane (such as the entrance pupil, which is always a plane). For a point source, the intensity of
a ray illuminating a differential area on a plane is given by
3
I ( ) = ( cos ) ,
where is the angle between the z axis and the ray intersecting the entrance pupil, and the relative intensity
at the center of the pupil is 1.0. Converting to normalized pupil coordinates and taking the square root yields the
pupil coordinate amplitude apodization
1
A ( ) = --------------------------------------------
-,
2 34
( 1 + ( tan ) )
where tan is the tangent of the angle between the z axis and the marginal ray. ZEMAX uses the entrance
pupil position and size to compute tan . The apodization factor is not used by the cosine cubed apodization.
Note cosine cubed apodization is only useful for point sources or for field points close to the axis when compared
to the entrance pupil diameter.
User defined apodization
ZEMAX also supports user defined apodizations on any surface, rather than just the entrance pupil. User
defined surface apodizations are implemented using the user defined surface type described in the chapter
"Surface Types".
Chapter 6: SYSTEM MENU 86

Apodization Factor
The apodization factor determines how fast the amplitude decays in the pupil. See the previous section on the
Apodization Type for details.
Telecentric Object Space
If this checkbox is selected, ZEMAX will assume the entrance pupil is located at infinity, regardless of the
location of the stop surface. All chief rays leaving the object surface will be parallel to the local Z axis. Ray aiming
and field points defined using angles may not be used when this option is selected.
Units
Lens Units
Lens units defines the units of measure for dimensions in most of the spreadsheet editors. These dimensions
apply to data such as radii, thickness, entrance pupil diameters, non-sequential position coordinates, and most
other parameters in ZEMAX.
There are four choices for lens units: millimeters, centimeters, inches, or meters.
For most imaging analysis features, such as ray fans and spot diagrams, the displayed units are micrometers.
The symbol for micrometers is m. The term microns is also commonly used in optics. The unit micron is identical
to micrometer. Wavelengths are also always defined in micrometers. The choice of lens units has no affect on
these units.
Source Units
Source units define the unit of measure for the flux (power) or energy emitted by non-sequential sources. This
setting only affects sources defined in the non-sequential component editor. Source units may be either watts,
lumens, or joules, with the additional choice of the prefixes femto, pico, nano, micro, milli, kilo, mega, giga, or
tera. Watts are used for radiometric analysis, lumens for photometric analysis, and joules for energy analysis.
The key difference between radiometric and photometric units is that photometric units are wavelength weighted
to the response of the human eye. See the table in the next section for more information on units.
Analysis Units
Analysis units define the unit of measure for irradiance (radiometric) or illuminance (photometric). This setting
only affects data as displayed on detectors collecting light from sources defined in the non-sequential components
editor. Irradiance units are watts/area, where area is in square meters, centimeters, millimeters, feet, or inches.
Illuminance units are lumens/area. Energy density units are joules/area. The unit prefixes femto, pico, nano,
micro, milli, kilo, mega, giga, or tera are all supported.
Here is a summary of radiometric, photometric, and energy units by ZEMAX.
RADIOMETRIC, PHOTOMETRIC, AND ENERGY UNITS
Radiometric Unit Photometric Unit Energy Unit

Flux Flux joule
watt lumen
Radiant Intensity Luminous Intensity Intensity
watt/steradian lumen/steradian = candela joule/steradian
Irradiance Illuminance Irradiance
watt/meter^2 lumen/meter^2 = lux = meter- joule/meter^2
watt/centimeter^2 candle joule/centimeter^2
watt/millimeter^2 lumen/centimeter^2 = phot joule/millimeter^2
watt/foot^2 lumen/foot^2 = footcandle joule/foot^2
watt/inch^2 joule/inch^2
Radiance Luminance Radiance
watt/steradian-meter^2 lumen/steradian-meter^2 joule/steradian-meter^2

One unfortunate convention in the optics industry (and in ZEMAX) is to use the term "Intensity" for two distinctly
different things. Intensity is commonly used in ray tracing applications and the optical design community to define
the flux represented by a single ray, measured in watts or lumens. Intensity is also used in the illumination and
radiometry fields, but the definition used there is flux per solid angle, measured in watts per steradian or lumens
per steradian (candela).
ZEMAX keeps track of energy by associating with each ray a flux value, which is equal to the square of the
electric field vector also associated with the ray. When this ray strikes a detector object, ZEMAX computes the
area the pixel represents to compute irradiance (flux/area) and the solid angle the pixel represents to compute
the radiometric intensity (flux/solid angle). The difference in these two uses of the word intensity can always be
resolved by considering the nature of the analysis ZEMAX is performing.
For more information on radiometric and photometric units, see the Handbook of Optics referenced in What
doesn't ZEMAX do? on page 31.
Title/Notes
Lens Title
The lens title will appear on graphic and text output. The title is specified by typing the title in the required
space. Additional text data may be placed on most graphic output, see the Chapter "File Menu" under
"Preferences" for details.
Notes
The notes section allows entry of a few lines of text which is stored with the lens file.
Glass Catalogs
This control group has an edit field which lists the names of the currently used glass catalogs (without file
extensions). Multiple catalogs may be specified by separating the names with a space. To add a catalog to the
list, either type the name of the catalog in the list, or choose the catalog name from the drop down list, then click
on "Use This Catalog". To remove a catalog from the list, either delete the name of the catalog from the list, or
choose the catalog name from the drop down list and click on "Dont Use Catalog".
Glass catalogs must be placed in the directory path \ZEMAX\GLASSCAT, unless the glass catalog directory
path has been changed (see Directories on page 59). For more information on changing glass catalogs, see
USING GLASS CATALOGS on page 477. To set a default catalog, see Editors on page 61.
Ray Aiming
The ray aiming group has the following controls.
Ray Aiming
If ray aiming is off, ZEMAX will use the paraxial entrance pupil size and location determined by the aperture
settings and calculated at the primary wavelength on axis for launching rays from the object surface. This means
ZEMAX will ignore entrance pupil aberration. For slow systems with modest field angles, this is perfectly
acceptable. However, certain systems, such as those with low F/#'s or large field angles, may have significant
entrance pupil aberration. The two primary effects of pupil aberration are the shift in location of the pupil with field
angle, and the anamorphic scaling of the edges of the pupil.
ZEMAX can be instructed to account for the aberration if ray aiming is selected on. With ray aiming on, every
ray trace is performed iteratively, with the program adjusting the ray aiming so that the ray crosses the correct
location on the stop surface.
The correct location on the stop surface is determined by first computing the stop surface radius. The correct
stop surface coordinates are then computed using a linear scaling of the pupil coordinates. For example, the
marginal ray has a normalized pupil coordinate of Py = 1.0. The correct coordinate on the stop surface for the
marginal ray is the stop surface radius times Py.
The stop surface radius is computed by tracing paraxial rays from the center of the object to the stop surface
at the primary wavelength. This is the preferred method of computing the stop radius, since paraxial rays are well
behaved and paraxial definitions are used for most optical parameters such as focal length, F/#, and magnifica-
tion. However, for systems that have significantly aberrated pupils, there will be a difference between the paraxial
and real ray stop radius. These systems will exhibit a difference between the real ray system aperture size and

the defined paraxial size. For example, the paraxial object space numerical aperture may be defined as 0.4, but
the actual numerical aperture of real rays may be a different value. To use real rays instead of paraxial rays to
determine the stop radius, choose the "Aberrated" option. Note that for most systems aberrated ray aiming is
neither required nor desired. Aberrated ray aiming will not work in systems where the stop lies in a caustic or
where the real rays cannot be traced at the full entrance pupil diameter or numerical aperture. If aberrated ray
aiming causes any of these problems, change back to "On" rather than "Aberrated". Note that once the stop radius
is determined, ALL rays, both real and paraxial, are aimed to the correct location on the stop, regardless of
whether paraxial or real rays were used to determine the stop radius.
For systems with virtual stops, such as some eyepieces, the effective stop location and size may be a function
of wavelength. For these systems, use the multi-configuration capability to treat each wavelength and system
aperture definition separately.
To eliminate any ambiguity in the calculation of the actual stop size, set ray aiming on, and then set the system
aperture type to "float by stop size". This eliminates the need for any ray tracing at all to determine the stop size,
and both real and paraxial rays will be aimed to the real stop exactly.
Although ray aiming is more accurate than paraxial entrance pupil aiming, most ray traces will take from two
to eight times as long to perform. Therefore, ray aiming should only be used when required. To determine the
amount of entrance pupil aberration in your system, select ray aiming off, and then look at the pupil aberration
plot (see Pupil Aberration on page 110). Pupil aberration of less than a few percent is generally insignificant. If
your system has significant pupil aberration, select ray aiming on and repeat the calculation. The aberration will
decrease to zero, or very nearly so.
Use Ray Aiming Cache
If checked, ZEMAX caches ray aiming coordinates so that new ray traces take advantage of previous iterations
of the ray tracing algorithm. Using the cache can speed up ray tracing dramatically. However, use of the cache
does require that the chief ray can be traced accurately. For some systems, the chief ray cannot be traced, and
for these systems, the cache should be turned off.
Robust Ray Aiming (slow)
If checked, ZEMAX uses a more reliable, but slower algorithm for aiming rays. This switch should only be set
if the ray aiming algorithm is failing even with the cache turned on. This switch has no affect unless the ray aiming
cache is checked on. Robust mode goes through an additional check to make sure that if multiple ray paths to
the same stop surface location exist, only the correct one is chosen. This is typically a problem in very fast, very
wide angle systems where off axis fields may find a virtual path to the stop that confuses the ray aiming iteration.
Pupil Shift: X, Y, and Z
For most systems, selecting ray aiming on will eliminate the effects of pupil aberration, as least as far as getting
the rays to correctly trace through the system. It does not, of course, actually eliminate the aberration, it merely
accounts for it. For some very wide angle or highly tilted or decentered systems, the ray aiming feature will fail if
unassisted. The problem is that the paraxial entrance pupil is used as a first guess to trace the ray. If the pupil
aberration is severe, it is possible that even this first guess cannot be traced, which prevents the algorithm from
taking a second, more refined guess.
The solution is to provide a rough guess as to how much the pupil has shifted with respect to the paraxial pupil.
This is called "Pupil shift", and there are three components; x, y, and z. The default value of zero for all three may
be modified to assist the algorithm in finding a successful first guess for the ray aiming. Positive values for the z
shift indicate that the real pupil is behind (to the right of in the conventional optical coordinate system) the paraxial
pupil, negative values indicate the pupil is shifted forward. Most wide angle systems have forward-shifted pupils.
The z pupil shift value provided is scaled linearly with the field angle of the ray being traced, so the pupil shift
refers to the offset of the pupil at full field. The x and y pupil shift values account for the change in location of
pupils when the object plane is tilted or when the stop is decentered. If "Scale pupil shift factors by field" is
selected, the x and y pupil shift values are also scaled with field, otherwise, the shift value is used for all fields
without any scaling. All shifts are in lens units.
It is important to understand that the exact values of the pupil shift is unimportant. Once the first guess ray can
be traced, the algorithm will robustly find the exact pupil location. The pupil shift values are just to get the ray
aiming started. Generally, guessing at the pupil shift values is an acceptable way of determining a suitable value.

Environment
Use Temperature, Pressure
This checkbox should be checked if the temperature and pressure values are different from 20 and 1,
respectively. If this box is unchecked, then all temperature and pressure effects are ignored. This speeds up the
computation of index data, and so the box should be left unchecked if the nominal temperature and pressure are
to be used.
Note that the system wavelengths are referenced to whatever the current system temperature and pressure is.
!Wavelength data are always measured in micrometers referenced to "air" at the current
ambient system air pressure.
If the system temperature and air pressure change; care should be taken to adjust the wavelength definitions
to match the new environment. See Wavelength data on page 55 and THERMAL ANALYSIS on page 491 for
more information.
!Thermal analysis is only available in the EE edition of ZEMAX.

Temperature in degrees C
The ambient temperature in degrees Celsius.
Pressure in ATM
The air pressure in atmospheres. A value of 0.0 implies vacuum, 1.0 implies sea level.
Polarization
The polarization tab is used to set the default input polarization state for many sequential analysis computa-
tions which use polarization ray tracing. For sequential analysis features such as spot diagrams and RMS vs.
field that use the "Use Polarization" switch to enable polarization ray tracing and apodization, this tab is the only
means for setting the initial polarization state. For these features, the polarization ray trace is only used to
determine the transmitted intensity of the ray while accounting for Fresnel, thin film, and bulk absorption effects.
The vectorial nature of the electric field is ignored, and scalar theory is still assumed. The rays are attenuated in
intensity and a weighted computation is performed. To define the polarization state of non-sequential sources,
see Sources tab on page 350.
!Polarization ray tracing is only available in the EE edition of ZEMAX.

Unpolarized
If checked, then the polarization values Jx, Jy, X-Phase, and Y-Phase are ignored, and an unpolarized
computation will be performed. An unpolarized computation is performed by tracing two rays with orthogonal
polarization and averaging the resulting transmission. Note that unpolarized computations take longer than
polarized computations, which take longer than computations that ignore polarization entirely.
Some ZEMAX features, such as the optimization operand CODA, require a single specific polarization state
to be defined. These features will use the defined Jx, Jy, X-Phase, and Y-Phase values defined below even if the
"Unpolarized" option is checked.
Jx, Jy, X-Phase, Y-Phase
The polarization is defined by four numbers: Jx, and Jy, which are the magnitudes of the electric field in X and
Y, and the X-Phase and Y-Phase, which are the phase angles in degrees. Internally, ZEMAX normalizes the
electric field vectors to have unity intensity. For an important discussion of the conversion from Jones vectors to
3D electric field, see Defining the incident polarization on page 510.
To edit these values, uncheck the "Unpolarized" box first.

Convert thin film phase to ray equivalent
If selected, ZEMAX converts the polarization phase computed using thin film conventions to phase along the
ray. If unselected, the ray coefficients will not be converted from the field coefficients. For more details on the
difference between these conventions, see Field vs. ray phase conventions on page 498. The recommended
and default setting is to convert the field thin film phase to ray phase.
Files
Coating File
The name of the file, in the \ZEMAX\COATINGS directory, that contains the coating material and layer
definitions used by this lens. The default name is COATING.DAT. Each lens file may use a separate coating file
if desired.
Scatter Profile
The name of the file, in the \ZEMAX\PROFILES directory, that contains the scattering profiles used by this
lens. New scatter profiles may be added or old ones deleted on the Scattering tab of the NSC Objects property
dialog box. See Coating/Scattering tab on page 349 for details. The default name is SCATTER_PROFILE.DAT.
Each lens file may use a separate scatter profile file if desired.
ABg Data File
The name of the file, in the \ZEMAX\ABG_DATA directory, that contains the ABg data definitions used by this
lens. The default name is ABG_DATA.DAT. Each lens file may use a separate ABg data file if desired.
GRADIUM Profile
The name of the file, in the \ZEMAX\GLASSCAT directory, that contains the GRADIUM surface profile data.
The default name is PROFILE.GRD. GRADIUM profile files must end in the extension GRD. See GRADIUM
on page 255.
Miscellaneous
Reference OPD
The Optical Path Difference, or OPD, is of value in optical design computations because the OPD represents
the phase error of the wavefront forming an image. Any deviations from zero OPD contribute to a degradation of
the diffraction image formed by the optical system.
Because the exit pupil is the image of the stop in image space, the exit pupil represents the only location in
image space where the beam has a clearly defined edge. The illumination at the exit pupil is generally smoothly
varying in amplitude and phase, and there is a clearly defined boundary between regions of zero amplitude and
non zero amplitude. In other words, it is reasonable to assume that there are no diffraction effects apparent in the
wavefront when viewed in the exit pupil. This is asymptotically true if all apertures in the optical system are large
compared to the stop limited beam size incident upon each aperture. Even if the exit pupil is virtual, which is often
the case, the exit pupil still defines the only location in image space where the beam is diffraction free. For more
information on diffraction image formation and the importance of the exit pupil, see any of the good books
referenced in the first chapter.
As the wavefront propagates from the exit pupil towards the image plane, the beam profile becomes complex
in amplitude and phase, and the wavefront extends over all space due to the effects of diffraction. For this reason,
the phase error as measured in the exit pupil is uniquely and critically important to the description of the wavefront
and image quality.
ZEMAX by default uses the exit pupil as a reference for OPD computations. Therefore, when the OPD is
computed for a given ray, the ray is traced through the optical system, all the way to the image plane, and then
is traced backward to the "reference sphere" which lies in the exit pupil. The OPD as measured back on this
surface is the physically significant phase error important to diffraction computations, such as MTF, PSF, and
encircled energy. The additional path length due to the tracing of the ray backwards to the exit pupil, subtracted
from the radius of the reference sphere, yields a slight adjustment of the OPD called the "correction term". This
computation is correct and is the desired method for all cases of practical interest.
However, ZEMAX does permit selection of two other reference methods.

The reference to "Infinity" makes the assumption that the exit pupil is very far away (even if it may not be) and
that the OPD correction term is given strictly by the angular error in the ray. There is only one possible reason to
use this setting: in the unlikely event that ZEMAX is unable to correctly compute the exit pupil location. This can
occur with some very unusual optics which do not form images (real or virtual) of the stop surface. ZEMAX has
special code to handle all known cases where this might occur, and therefore, this setting should not be used
unless ZEMAX Development Corporation technical support specifically recommends it. There are currently no
known cases where this would be the recommended setting.
The reference to "Absolute" means that ZEMAX does not add any correction term at all to the OPD
computation, but adds up the total optical path length of the ray and subtracts it from the chief ray. This method
is never physically significant, and the only intended purpose is for debugging and testing of the OPD algorithms
by ZEMAX Development Corporation.
In summary, always use the "exit pupil" reference, unless specifically instructed to change the setting by a
ZEMAX Development Corporation Engineer. Erroneous data may easily result if "exit pupil" is not selected.
Paraxial Rays
Paraxial ray properties are generally not defined for non rotationally symmetric systems. For this reason,
ZEMAX by default ignores all tilts and decenters due to coordinate breaks when tracing paraxial rays. By ignoring
tilts and decenters, ZEMAX can compute the paraxial properties of an equivalent centered system, which is
generally the correct approach even for systems without symmetry.
The default setting "Ignore Coordinate Breaks" is therefore highly recommended. Choosing a different setting
may result in the failure of ZEMAX to accurately compute all paraxial data, ray aiming, and OPD computations.
There is only one known case where "Consider Coordinate Breaks" may be required. For ray tracing through
steeply tilted gratings, coordinate breaks may be required even for paraxial rays, otherwise, the rays may not be
able to satisfy the grating equation. This is because diffraction gratings bend rays by an amount that depends
critically on the incident angle.
Fast Asphere Trace
When tracing rays through certain types of aspheric surfaces, iteration is required if no closed form solution
exists to the ray-surface intercept equations. If this box is checked (the default condition) ZEMAX makes an initial
guess for the ray-surface intercept solution in an attempt to accelerate convergence of the iteration. However,
some wildly curved aspheric surfaces will not converge if this "fast guess" is used. For systems which use these
surfaces, this box may need to be unchecked, in which case ZEMAX will use a slower but more robust algorithm
for finding the solution. Whether the box is checked or unchecked, ZEMAX will find an accurate ray intercept
solution, or an error message will be issued.
Fast Semi-Diameters
ZEMAX computes "automatic" semi-diameters to estimate the clear aperture required on each surface to pass
all rays at all field points and wavelengths. For axial systems, this computation can be done accurately by tracing
just two rays, the top and bottom marginal rays, for each field and wavelength. By default, for axial systems
ZEMAX will trace just these two rays in the "true" tangential plane (for a definition of this term see Sagittal and
Tangential on page 51) of the vignetted pupil for each field and wavelength, then use the radial coordinates of
each ray at each surface to determine the semi-diameter required.
For some non-axial systems, the resulting estimate is not accurate enough. This typically includes systems
which have tight edge and clear aperture constraints. For these systems, there is no general way of accurately
computing the semi-diameters other than by tracing a large number of rays in the vignetted pupil. ZEMAX will
iteratively trace as many marginal rays around the perimeter of the pupil as required to determine every semi-
diameter to an accuracy of about 0.01% (5 significant figures).
Iteration can be slow, since ZEMAX needs to update the semi-diameters frequently, especially during
optimization. There is a trade-off between speed and accuracy. For axial systems, ZEMAX will use the slower
iterative algorithm only if the "Fast Semi-Diameters" option is checked "off"
For non-axial systems, ZEMAX will automatically use the slower iterative algorithm. If the "Fast Semi-
Diameters" option is checked "on", ZEMAX will trace only as many marginal rays as required to estimate the
automatic semi-diameters to about 0.01%. The algorithm starts by tracing 2, then 4, then 8 rays, and so on, until
the semi-diameter values converge to within 0.01% of the previous estimate. If the "Fast Semi-Diameters" option

is checked "off", ZEMAX will trace at least 32 rays around the vignetted pupil at each field and wavelength, and
more if required to estimate the automatic semi-diameters to about 0.01%. Tracing at least 32 rays is safer, but
slower, and should only be used when very high precision is required in the automatic semi-diameter calculations.
Both methods used by ZEMAX to determine semi-diameters are accurate only for surfaces whose maximum
size is set by the radial coordinate of a marginal ray. The methods do not work for surfaces that lie in a caustic.
For surfaces in a caustic, the exact ray-based semi-diameter could be determined using a spot diagram or
footprint analysis, however this level of precision is likely never required since diffraction effects, which the rays
do not model, would be significant.
Check GRIN Apertures
If checked, this setting instructs ZEMAX to check all gradient index ray traces for surface aperture vignetting.
Each GRIN trace within the media is checked to see if the ray ever passes outside the front surface aperture
boundary, and if it does, then the ray is vignetted. If this setting is not checked, then the ray may travel outside
the boundary defined on the front surface, as long as the ray passes the aperture at the surface.
Semi Diameter Margin (lens units)
The semi-diameter of every surface in "automatic" mode is computed to be the radial aperture required to pass
all rays without clipping. For systems with closely spaced elements in or near edge contact, this yields surface
apertures that provide no clearance for finishing or mounting. Often, optical surfaces are only well finished within
some fraction of the full radial aperture, typically between 90% and 98%, depending upon the part size.
The semi diameter margin control allows specification of an additional amount of radial aperture as a fixed
number in lens units. The default value of zero leaves no margin, while a margin of 2.0 lens units would add 2.0
lens units to the semi-diameters of all surfaces under "automatic" control.
See "Semi Diameter Margin %" below. If both "percent" and "lens units" margin values are non-zero, the
percent is added first, then the lens unit margin.
Semi Diameter Margin %
The semi-diameter of every surface in "automatic" mode is computed to be the radial aperture required to pass
all rays without clipping. For systems with closely spaced elements in or near edge contact, this yields surface
apertures that provide no clearance for finishing or mounting. Often, optical surfaces are only well finished within
some fraction of the full radial aperture, typically between 90% and 98%, depending upon the part size.
The semi diameter margin control allows specification of an additional amount of radial aperture as a
percentage. The default value of zero leaves no margin, while a margin of 5% would add 5% to the semi-diameters
of all surfaces under "automatic" control. The maximum allowed margin is 50%.
See "Semi Diameter Margin (lens units)" above. If both "percent" and "lens units" margin values are non-zero,
the percent is added first, then the lens unit margin.
Global Coordinate Reference Surface
Global coordinates are defined by a rotation and translation from the local coordinates on each surface. The
conversion can be written as
xg xo R 11 R 12 R 13 x l
y g = y o + R 21 R 22 R 23 y l ,
zg zo R 31 R 32 R 33 z l
where the "g" subscript indicates global coordinate, "o" indicates an offset coordinate (the translation) and "l"
indicates local coordinate. The rotation matrix R and the offset vector can be computed for any surface using any
other surface as a global reference.
The rotation matrix yields valuable insight into the orientation of the surface coordinate system with respect to
the global reference surface. On the local surface, the unit vector which is oriented along the local x direction is

(1, 0, 0 ). This vector may be rotated using the R matrix to yield the orientation of the local x axis in the global
coordinate system. A matrix multiplication for each of the three local axis unit vectors yields
R 11 R 12 R 13

x l = R 21 , y l = R 22 , z l = R 23 .
R 31 R 32 R 33
Note that the global coordinate system orientation of the unit vectors oriented along the local coordinate axes
are the columns of the R matrix. The global vertex listing on the prescription report lists the R matrix components
and offset vector for each surface, all referenced to the global reference surface. If the surface is a coordinate
break, then the R matrix includes the effects of the coordinate translation and rotation. If the reference surface is
a coordinate break, the reference coordinate system is defined after the decentration and rotation of the local
coordinate system. If the surface for which R is computed precedes the reference surface, and it is a coordinate
break, the R matrix is for the system prior to the coordinate transformation. When in doubt, insert a dummy surface
that is not a coordinate break at the location of interest to check the global orientation.
The reference surface will be incremented or decremented automatically as surfaces above the reference
surface are inserted or deleted. Any surface may be chosen as the reference surface, unless the object is at
infinity, then 0 may not be the reference surface. The reference surface is also used to define the point of overlap
for multiple zoom positions on the 3D layout plot.
Method To Compute F/#
By default, ZEMAX computes the paraxial and working F/# of a system using ray tracing. For details, see
Paraxial working F/# on page 51 and Working F/# on page 55. For systems with very large F/#s, the ray tracing
method may be inaccurate, because the very small angles between the marginal and chief rays leads to round-
off error, and even small amounts of aberrations such as spherical aberration can significantly effect the
calculation of F/#. For these reasons, ZEMAX will "cap" the maximum F/# to a value of 10,000. It is strongly
recommended that rays be used to compute the F/# for all systems whose F/# is less than 10,000.
The preferred method of modeling systems with very large F/#s is to use a paraxial lens in the optical space
preceding the image surface with any convenient focal length selected to "convert" the nearly collimated beam
to a focusing beam with a more moderate F/#. This method avoids round-off error problems associated with nearly
collimated rays traveling great distances. The only disadvantage to this approach is that the effective magnifica-
tion of the paraxial lens must separately be accounted for. For example, if the true F/# of the system is 20,000, a
paraxial lens may be selected to convert the beam to a F/20 system. However, the diffraction results ZEMAX
computes based upon the F/#, such as the Point Spread Function width, need to be interpreted as being scaled
by the additional factor of 1000.
Optionally, the F/# can be computed using the exit pupil distance divided by the exit pupil diameter. This
method can be selected using this "Method To Compute F/#" switch. There are some important points to consider
when using this option. First, ZEMAX makes no attempt to scale the F/# with either pupil orientation or field angle.
Therefore, this option should only be used when the chief ray at all field position makes a small angle with respect
to the Z axis in image space. Second, the exit pupil diameter and position used are based upon the paraxial values
for these parameters (see Exit pupil diameter on page 47). In the presence of significant pupil aberration, the
paraxial values may not agree with the real, aberrated pupil sizes. The best way to avoid this possibility is to place
the stop at the end of the optical system, prior to the image surface, and use ray aiming (see Ray Aiming on
page 88) and float by stop size for the aperture type (see Aperture Type on page 85).
Non-Sequential
The controls on this tab define how rays trace within a NSC group.
Maximum Intersections Per Ray
This control defines how many times a single ray may intersect objects along any single path from the original
source parent ray to the final object intersection. When ray splitting is used, this parameter controls the maximum
number of generations of child rays split off from the parent ray.

In some systems, such as a source placed inside a reflective sphere with perfect reflectivity and no bulk
absorption, the rays will bounce around until this limit is reached, and the energy of the ray finally will be discarded.
These systems are of course not physically possible, since no reflection or propagation through anything but a
vacuum is perfectly lossless.
Maximum Segments Per Ray
This is the maximum number of segments per ray launched, not the total number of segments that ZEMAX
can keep track of. A segment is that portion of a ray path from one intersection to the next. When a ray is launched
from the source, it travels to the first object. That is 1 segment. If the ray then splits into 2 rays, each of those are
another segment (for a total of 3). If each of those rays split again, there will be 7 segments. Generally, if ray
splitting is being used, the number of segments grows far faster, and needs to be set much larger, than the number
of ray-object intersections does.
It is critical to understand that each ray launched from a source can have up to the maximum number of
segments. If the maximum number of segments is 1,000, and 5,000 individual rays are launched, ZEMAX can
store up to 5,000,000 segments. The total RAM requirement is about 140 bytes times the maximum number of
segments. Setting the limit at 100,000 segments would require 14 Mb of RAM per ray. For this reason, do not
arbitrarily (and needlessly) set the maximum to very large numbers; there is a limit of 2,000,000 ZEMAX will allow.
Note that with even a modest number of segments, ZEMAX can still trace millions of rays.... the maximum number
of segments is only a limit on how many segments are allowed in one single ray. Most ZEMAX features only use
one ray at a time, so the total RAM requirement is driven by the maximum number of segments rather than the
number of rays traced. The exception is the Shaded Model display, which must be able to allocate enough RAM
to hold all segments for all source rays drawn.
Maximum Nested/Touching Objects
This defines an upper limit on how many objects can be inside, straddled, or in direct contact with each other.
For example, if object 3 is inside of 2 which is inside of object 1, the maximum number of nested objects is 3.
There may be any number of groups of objects each nested 3 deep in this case. The limit applies to the total
nesting in any one collection of objects, however, there may be any number of such collections within the NSC
group. If several objects share a common boundary, such as multiple volumes touching at one face, then the
maximum nested objects must be set at least as high as the total number of objects that share a common point
in space.
Minimum Relative Ray Intensity
As each ray splits, the energy decreases. The relative ray intensity is a lower limit on how much energy the
ray can carry and still be traced. This parameter is a fraction, such as 0.001, relative to the starting ray intensity
from the source. Once a child ray falls below this relative energy, the ray is terminated.
Minimum Absolute Ray Intensity
This parameter is very similar to the minimum relative ray intensity, except it is absolute in source units rather
than relative to the starting intensity. If this is zero, the absolute ray intensity threshold is ignored. The initial
intensity of each ray is always given by the source intensity divided by the total number of analysis rays for that
source. The number of layout rays is not used to determine the initial intensity of rays, even for those rays drawn
on layout plots.
Glue Distance In Lens Units
When two NSC objects are placed in contact, such as a lens touching one face of a prism, numerical roundoff
will cause the ray tracing algorithm to sometimes detect a very tiny distance between the two objects. This can
also occur when objects are rotated in 3D space and placed close, but not exactly, next to one another because
of the finite number of digits entered in the spreadsheet editor.
The glue distance is the distance below which the objects are considered in contact. It is important that objects
not be separated by distances very close to the glue distance. If it is intended that two objects be in contact, then
the maximum spacing between the objects should be several times smaller than the glue distance. If it is intended
that the objects be separated, then the distance between the object should be several times larger than the glue
distance. Object spacings very close to the glue distance will yield inconsistent ray tracing or geometry errors.
This should be avoided by adjusting either the object spacing or the glue distance.

The glue distance also determines the minimum propagation length for ray tracing. If a ray-object intersection
is less than the glue distance away from the previous intercept, the intercept is ignored. This can affect ray tracing
results when tracing rays from faceted reflectors, and a ray strikes within the glue distance of the edge between
two angled facets.
In the majority of cases, no adjustment should be made to the glue distance parameter. The glue distance
must be no smaller than 1E-10 and no larger than 1.0.
Missed Ray Draw Distance in Lens Units
This parameter is the distance to draw the ray segments that miss all objects. ZEMAX will draw a short ray
segment to indicate the direction the ray was traveling in. This parameter also controls how large a source
indicator arrow to draw. If zero, ZEMAX will select a default value for this parameter when drawing miss rays and
some sources.
Retrace source rays upon file open
If checked, then NSC source rays will be retraced when the file is opened. This allows refreshing of detector
windows automatically.
Fields
The field dialog box allows specification of field points. Field points may be specified as angles, object heights
(for systems with finite conjugates), paraxial image heights, or real image heights. The buttons can be used to
activate or deactivate field positions, as well as to sort, save and load the data. For information about field
conventions, see Field angles and heights on page 47.
Vignetting factors
ZEMAX also provides data fields for defining vignetting factors. The vignetting factors are VDX, VDY, VCX,
VCY, and VAN. These vignetting factors should be left at zero if there is no vignetting in the system. The factors
are described in the Chapter "Conventions and Definitions" under Vignetting factors on page 53.
There is also a button labeled "Set Vig" on the fields dialog box. Clicking on this button will recompute the
vignetting factors for each field based upon the current lens data. The set vignetting algorithm estimates the
vignetting decenter and compression factors so that the four marginal rays in the top, bottom, left, and right edges
of the pupil pass within the apertures of each surface. Only the primary wavelength is used. To clear the vignetting
factors back to the default values of zero, click on "Clr Vig". The set vignetting algorithm may not work properly
if there are central or non-circular obscurations or apertures in the system. In this case, the vignetting factors must
be user defined.
The algorithm starts by launching a grid of rays through the pupil. At each surface with a surface aperture, the
ray is tested to see if it passes within the specified aperture. All rays which pass all surfaces are then used to
compute the unvignetted pupil centroid. The edge of the unvignetted pupil is then computed using an iterative
method accurate to about 0.001%.
The algorithm may not work in all cases. For systems where the set vignetting algorithm fails, the vignetting
factors will need to be adjusted manually. The accuracy of the set vignetting algorithm can be tested by tracing a
few marginal rays.
!Identical or nearly identical field coordinates may not be defined with different vignetting
factors; see the Chapter Conventions and Definitions for details.
Saving and loading field data

The Save and Load buttons on the field dialog box are used to store and retrieve the field data independently
from the lens data. The file format is text that can be edited or created outside of ZEMAX.
Wavelengths
The wavelengths dialog box is used to set wavelengths, weights, and the primary wavelength. The buttons
can be used to activate and deactivate wavelengths, and to sort, save, and load the data. A list of commonly used
wavelengths is also included. To use the entries on the list, select the desired wavelength set, and click on the
"Select" button.

For more information, see the Wavelength data on page 55.
Saving and loading wavelength data
The Save and Load buttons on the wavelength data dialog box are used to store and retrieve the wavelength
data independently from the lens data. The file format is text that can be edited or created outside of ZEMAX.
Next Configuration
This menu option provides a quick way to change all graphics to reflect the next configuration (or zoom
position). All spreadsheet, text, and graphics data will be updated.
Last Configuration
This menu option provides a quick way to change all graphics to reflect the last configuration (or zoom
position). All spreadsheet, text, and graphics data will be updated.

Chapter 7 ANALYSIS MENU
Introduction
This chapter provides detailed descriptions of each of the analysis features ZEMAX supports. Analysis in this
context means any graphical or text data computed from data defining the lens. This includes aberrations, MTF,
spot diagrams, and many other computations. Program features which modify the lens data or which manipulate
other data (such as glass catalog data) are described in the chapter Tools Menu.
Selecting a menu option will immediately perform the requested calculation. Once the graph or text window is
displayed, you may select the Settings menu option to modify the default settings for that window. Once you have
made the appropriate changes, click on OK and the program will recalculate and redisplay the data presented
in the window. If you prefer to change the settings before the graphic or text data is displayed, use the Show
Options First checkbox on the Graphics tab of the File, Preferences dialog box.
For a description of the OK, Cancel, Save, Load, Reset, and Help buttons present on most of the "Settings"
dialog windows, see the chapter "User Interface".
Each analysis window has an "Update" menu item. The update function forces ZEMAX to recompute and
redisplay the data presented in the window. This is useful if the lens data has changed and the graph now displays
obsolete data. Double clicking within the window has the same effect as selecting Update. Clicking with the right
mouse button is equivalent to clicking on "Settings". For more information, see USER INTERFACE on page 35.
Layout
2D Layout
Purpose:
Layout diagram. This is a YZ cross section through the lens.
Settings:
Item Description
First Surface The first surface to be drawn.
Last Surface The last surface to be drawn.
Wavelength Either any one or all wavelengths may be shown.
Field Either any one or all field positions may be shown.
Number of Rays The number of rays specifies the number of tangential rays to be drawn for each defined
field. The rays will be evenly distributed along the fan of the pupil, unless apodization
has been specified. This parameter may be set to zero.
Scale Factor If the scale factor is set to zero, then "Fill Frame" will be selected, which will scale the
range of surfaces drawn to fill the graphic page. If a numeric value is entered, then the
plot will be drawn in "real" scale, times the scale factor. For example, a scale factor of
1.0 will plot the lens actual size on the printer (not the screen). A factor of 0.5 will plot
the lens at half scale.
Y Stretch The relative exaggeration of the Y direction. This control is useful for drawing systems
with very large aspect ratios. Ignored if negative, zero, or unity.
Upper Pupil Limit The maximum pupil coordinate to draw rays to.
Lower Pupil Limit The minimum pupil coordinate to draw rays to.
Marginal and Chief Draws only the marginal and chief rays, overriding the other ray settings.
Only
99 Chapter 7: ANALYSIS MENU

Item Description
Color Rays By Select "Fields" to use color to distinguish between each field position, or "Waves" to
distinguish between each wavelength.
Suppress Frame Suppresses drawing of the frame on the bottom of the screen, which leaves more room
for the layout plot itself. No scale bar, address block, or other data will be displayed.
Delete Vignetted If checked, rays are not drawn if they will be vignetted by any surface.
Fletch Rays If checked, small arrows are drawn on each ray to indicate the direction of propagation.
Discussion:
This feature is not available if you use coordinate breaks, spider obscurations, obscuration decenters, X-
angles, holograms, or other attributes which spoil the rotational symmetry of the lens. Use the 3D layout instead.
If rays miss a surface, then the rays will not be drawn to the surface where the error occurred. If the ray is total
internal reflected, then the ray will be drawn up to but not past the surface where the error occurred. Ray failures
can be evaluated in detail by using Ray Trace on page 174.
3D Layout
Purpose:
Draws 3D layout plots of the lens system. The algorithm draws a wireframe style representation of the lens.
Settings:
Item Description
First Surface The first surface to be drawn.
Last Surface The last surface to be drawn.
Wavelength Either any one or all wavelengths may be shown.
Field Either any one or all field positions may be shown.
Number of Rays The number of rays specifies the number of rays to be drawn for each selected field and
wavelength. The rays will be evenly distributed along the fan of the pupil, or around the
perimeter if Ring is the selected Ray Pattern, or unless apodization has been specified.
This parameter may be set to zero. It is ignored if the Ray Pattern is set to List.
Ray Pattern Choose XY Fan, X Fan, Y Fan, Ring, List, Random, or Grid to indicate what the pattern
of rays to be traced should be. The List option indicates that the rays to be traced are
user defined and listed in a file, see the discussion below for information on the ray list
format. If List is selected the Number of Rays setting is ignored.
Hide Lens Faces If checked, this option will suppress drawing of the lens faces, and only the lens edges
will be drawn. This is useful because some complicated systems look cluttered with the
faces drawn.
Hide Lens Edges If checked, this option will suppress drawing of the outer aperture of the lens. This is
useful for giving the 3D layout a 2D "cross section" appearance.
Hide X Bars If checked, this option will suppress drawing of the X portions of the lens faces. This
option is useful when "Hide Lens Edges" is checked and "Hide Lens Faces" is not
checked.
Chapter 7: ANALYSIS MENU 100

Item Description
Rotation About X The angle in degrees by which the lens appears to be rotated about the X axis.
Rotation About Y The angle in degrees by which the lens appears to be rotated about the Y axis.
Rotation About Z The angle in degrees by which the lens appears to be rotated about the Z axis.
Color Rays By For sequential rays, "Fields" will use color to distinguish between each field position,
"Waves" to distinguish between each wavelength, or "Config" to distinguish between
configurations.
For non-sequential rays "Fields" will use color to distinguish between each source
number, unless a source has a user defined color assigned to that source.
Delete Vignetted If checked, rays are not drawn if they will be vignetted by any surface.
Configuration Select "All" to draw all configurations at once, or select the one configuration to draw, or
select "Current" to show the active configuration.
Offset X, Y, Z The X, Y, and Z direction offset between configurations in lens units. Only has an affect
on the drawing if "All" configurations are being drawn.
Split NSC Rays If checked, rays from NSC sources will be statistically split at ray-surface intercepts.
Rays entering from the entry port are not affected by this setting.
Scatter NSC Rays If checked, rays from NSC sources will be statistically scattered at ray-surface
intercepts. Rays entering from the entry port are not affected by this setting.
Discussion:
Pressing the left, right, up, down, Page Up, or Page Down keys will rotate the displayed image for a different
perspective.
The orientation indicator
To create the layout, ZEMAX projects the 3D coordinates of the rotated optical model onto the 2D plane of the
plot. Conceptually, the optical model is first rendered in an unrotated 3D space, with +z oriented to the right, +y
up, and +x away from the viewer into the page. The coordinates and orientations of all surfaces and objects are
defined by the Global Coordinate Reference Surface (see Global Coordinate Reference Surface on page 93).
This 3D model is then rotated by the Rotation About X, Y, and Z values specified in the settings (see also
Graphics on page 59 for information about the rotation order). The resulting 3D model is then projected onto a
2D plane by plotting the z coordinate along the horizontal direction and the y coordinate along the vertical
direction; the x coordinate is ignored.
To help visually define the orientation of the model, an orientation indicator is shown in the bottom left hand
corner of the drawing. This indicator consists of 3 lines extending along the directions of the +x, +y, and +z axes
of the global coordinate reference surface. These 3 lines rotate as the drawing is rotated. If the coordinates of the
3D orientation indicator lines lie in the plane of the drawing, or would extend out (toward the viewer), they are
drawn in black. If the orientation indicator lines would extend in (away from the viewer) then the lines are drawn
in grey.
Ray errors
can be evaluated in detail by using Ray Trace on page 174. These comments only apply to rays from the
sequential object surface, and not to rays from NSC sources.

Configuration data
When drawing all configurations, an offset may be added to each configuration in the x, y, and z directions
independently. The offsets may all be zero if desired. If the offsets are zero, then all the configurations are
superimposed; otherwise, the configurations are all displaced from one another by the specified amount. Note
that all offsets are defined from the global coordinate reference surface position. The global coordinate reference
surface is defined on the Miscellaneous tab of the System, General dialog box. If all offsets are zero, the multiple
configurations are all overlapped at the global coordinate reference surface.
Raylist file format
If List is chosen for the ray pattern, the rays to be traced are defined in a file. The file must be called
RAYLIST.TXT and be placed in the main ZEMAX directory. The file format is ASCII, with two distinct methods for
defining the rays supported, implicit and explicit. The implicit format file consists of two numbers on each line, one
for the px and one for the py normalized pupil coordinates. The specified rays are traced at each defined field
and wavelength selected.
Example: Four marginal rays are defined by:
0.0 -1.0
0.0 1.0
-1.0 0.0
1.0 0.0
The explicit format file consists of the word EXPLICIT followed by the values x, y, z, l, m, n, and wavenumber;
where x, y, and z are the ray starting coordinates, l, m, and n are the direction cosines, and wavenumber is an
integer indicating the wavelength to use. All coordinates are in object space. If the object thickness is infinity, then
the spatial coordinates are relative to surface 1. If the object is not at infinity, then the coordinates are relative to
surface 0. In both cases the ray itself is in the object space medium, prior to refraction into surface 1. If explicit
format is used, then the field and wavelength settings are ignored, and only those rays listed in the file are traced.
Example: Three rays at wavelengths 1, 2, and 3 along the Y axis parallel to the Z axis are defined as follows:
EXPLICIT
0.0 -5.0 0.0 0.0 0.0 1.0 1
0.0 +0.0 0.0 0.0 0.0 1.0 2
0.0 +5.0 0.0 0.0 0.0 1.0 3
Wireframe
Purpose:
Draws a representation of the lens.
Settings:
The options are similar to those available for the 3D layout feature. The "Hide Lens Edges" and "Hide X Bars"
checkboxes are not available, and several new controls are added as described below.
Item Description
Draw Section Select "Full" to draw each lens element completely. The 3/4, 1/2, and 1/4 options draw
just that much of the element, yielding a cut away perspective of the lens interior.
Radial Segments The number of radial segments used to approximate the lens shapes. Larger numbers
require more processing time.
Angular Segments The number of angular segments used to approximate the lens shapes. Larger numbers
Discussion:
The wireframe algorithm describes the lens as a collection of polygon facets. The number of facets used to
display the lens elements can be modified using the radial and angular segment options. The wireframe model

is identical to the solid model, except hidden lines are not removed. This representation may cause the screen
to become cluttered with lines. The "Hide Lens Faces" option can be used to clean up the display. The advantage
to this display method is speed; it is faster than the solid model.
perspective.
can be evaluated in detail, see Ray Trace on page 174.
See also The orientation indicator on page 101.
Solid Model
Purpose:
Draws a hidden-line representation of the lens.
Settings:
The options are similar to those available for the 3D layout feature. The "Hide Lens Edges" and "Hide X Bars"
checkboxes are not available, and several new controls are added as described below.
Item Description
Draw Section Select "Full" to draw each lens element completely. The 3/4, 1/2, and 1/4 options draw
just that much of the element, yielding a cut away perspective of the lens interior.
Radial Segments The number of radial segments used to approximate the lens shapes. Larger numbers
Angular Segments The number of angular segments used to approximate the lens shapes. Larger numbers
Discussion:
The solid model algorithm describes the lens as a collection of polygon facets. The lines and facets which are
hidden from view are removed, which gives the lens a solid appearance. This algorithm is slower than the other
layout plots. The number of facets used to display the lens elements can be modified using the radial and angular
segment options.
perspective.
can be evaluated in detail, see Ray Trace on page 174.
See also The orientation indicator on page 101.
Shaded Model
Purpose:
Draws a shaded solid model representation of the lens using OpenGL graphics.
Settings:
The options are almost identical to those available for the Solid Model feature, except there are additional
controls for setting the lighting level and background color.
There is an additional control labeled "opacity". The options are to ignore, which ignores the surface and object
opacity values and draws all surfaces and objects as opaque; and method 1 or 2, which use different algorithms
to render the scene. Some systems are difficult for OpenGL to render when there are a mixture of opaque and
transparent surfaces and objects; choose the method which yields the preferred rendering. For information on
setting the opacity of surfaces and objects see Surface Opacity on page 67 and The object properties dialog
box on page 348.

ZEMAX Element Drawing
Purpose:
This feature creates a mechanical drawing of surface, singlet, cemented doublet, or cemented triplet elements
suitable for use in optical shop fabrication.
Settings:
Item Description
Surface The first surface of the element to be drawn.
Show As Select either "Surface", "Singlet", "Doublet", or "Triplet".
Note File Name The name of the ASCII file which contains the notes to be appended to the notes section
of the element drawing. Notes should always start at number 2, since number 1 is
reserved for the units specification.
Edit Note File Clicking on this button will invoke the Windows NOTEPAD.EXE editor, which can then
be used to modify the selected note file.
Radius The radius tolerance value for the specified surface.
Irregularity The irregularity tolerance value for the specified surface.
Clear Aper The clear aperture diameter of the specified surface.
Thickness The center thickness tolerance of the specified surface.
element to fill the right half of the element drawing. If a numeric value is entered, then
the plot will be drawn in "real" scale, times the scale factor. For example, a scale factor
of 1.0 will plot the element at actual size on the printer (not the screen). A factor of 0.5
will plot the element at half scale.
Decimals The number of decimals to use in numeric values. Press the "Reset all but surface, show
as, and titles" button to reformat all displayed numbers.
Drawing Title This field is for any user defined text. The default is the lens title.
Date If left blank, the default date format is used. If any text is supplied, the exact text is used.
Drawing Name All of these fields are for user defined text. Any text may be entered. No default value.
Approved
Revision
Drawn By
Project
Note Font Size Choose Standard, Medium, Small, or Fine. These are in order of decreasing font size.
The Note Font Size setting only affects the size of the note file that is annotated on the
drawing. Smaller fonts permit larger note files to be displayed.
Save As/Load From These two buttons allow saving and reloading of all settings displayed on the element
drawing settings box. The data is stored in a binary format with the extension ELE.
Reset all but... If selected, this button will reset all the default tolerances and apertures for the specified
surfaces, but the current surface, show as, and text titles will remain as they are.
Discussion:
The element drawing settings may be stored for the specific lens file by pressing the Save button. Unlike most
analysis features, the element drawing feature saves all the settings for each surface separately. For example,

the notes and tolerances for surface 1 may be saved, and then new notes and tolerances for surface 3 may be
entered and then saved. To recall the settings for any specific surface, change the surface number to the desired
surface, and then press the Load button. If a match is found with a previously saved surface, the settings for that
surface will be displayed. This feature makes it easy to regenerate complex drawings for multiple element
systems.
An important feature of the element drawing capability is the ability to load different note files and place them
on the drawing. The default note file "DEFAULT.NOT" is a generic set of notes which will rarely be useful as is.
However, the user can modify the note files (they are ASCII files which any word processor or text editor can
modify) and store them under different names. For example, you may want to have a .NOT file for each type of
optic you design, and then load the most appropriate note file when the element drawing is generated.
If the first line of the note files starts with the number 1, then ZEMAX will print the note file as provided, with
no default notes. If the note file starts with any other symbol, the first default note will be "1) All dimensions in
millimeters" or whatever the current lens units are. To use the default note, start the user supplied notes at note
number 2. The line breaks and spacings in the note file will be replicated exactly on the element drawing.
Whenever a new element drawing is generated, or the "Reset" button is pressed, the default settings will be
regenerated. The default tolerances are taken from the tolerance data editor. The maximum of the min/max
tolerance range is used as the default. For example, if the TTHI thickness tolerance is -.03, +.05, the tolerance
value will be 0.05. Only TTHI, TRAD, and TIRR tolerances are considered. If a suitable default cannot be
generated, the tolerance is set to zero. Note all tolerance fields are text; and may be edited to suit any
requirement.
A handy conversion between radius tolerance and the power tolerance in fringes for a Newton's rings type
optical test against a test plate is given by
2
R
#fringes = ------- -----2- ,
R
where R is the radius error, is the test wavelength, is the radial aperture, and R is the radius of
curvature. This formula is an approximation for shallow curvatures. For more information, see Malacara, Optical
Shop Testing, J. Wiley & Sons, Inc.
Special characters
There are useful special characters that may be inserted and edited in the text fields of the tolerances. The
symbols are inserted by pressing and holding Alt on the keyboard and typing a 4 digit number on the numeric
keypad. Some common characters are - Alt 0177, - Alt 0181, - Alt 0174, - Alt 0169.
ISO Element Drawing
Purpose:
This feature creates an ISO 10110 type drawing of surface, singlet, or doublet elements suitable for use in
optical shop fabrication.
Settings:
Item Description
First Surface The first surface of the element to be drawn.
Show As Choose surface, singlet, or doublet.
Other settings There are many other settings available for this feature that correspond to the tolerances
defined in the ISO 10110 specification. See the discussion for details.
Discussion:
The ISO 10110 Element Drawing is an interpretation of the drawing specification "ISO 10110 Optics and
Optical Instruments -- Preparation of drawings for optical elements and systems: A User's Guide", by Ronald K.

Kimmel and Robert E. Parks, eds., published by the Optical Society of America. For more information see OSAs
web site at www.osa.org.
Although the ISO specification only covers the case of single elements, the ZEMAX ISO drawing also supports
surfaces and doublets. However, not all the ISO drawing particulars are incorporated on a drawing of a cemented
doublet. If the doublet drawing results are not satisfactory, draw the doublet as two separate singlet drawings
instead.
Summary of ISO 10110 symbols and codes
ZEMAX does not automatically include default values for all ISO tolerances; however text fields are provided
for user specification of these tolerances. The following table summarizes the ISO 10110 symbols and codes
used by ZEMAX.
Summary of ISO 10110 symbols and drawing codes
Symbol or Code Number Description

R Radius of curvature in lens units. CC indicates concave, CX indicates
convex.
Diameter in lens units.
e Effective diameter in lens units.
n refractive index
v Abbe number
() Coating specification
P; Rq; Lmin, Lmax Indicates a polished surface on the lens drawing. The details of the polish
specification are not provided on the drawing, but are specified in the
surface data. Rq = maximum permissible rms surface roughness in
micrometers, Lmin/Lmax = minimum and maximum length of sampling
interval for rms calculation in mm.
0/A Stress Birefringence; A = maximum optical path difference in nm/cm.
1/NxA Bubbles & Inclusions; N = number of bubbles and/or inclusions, A =
bubble grade number (size)
2/A; B Inhomogeneity and Striae; A = homogeneity class number, B = striae
class number
3/A(B/C) Surface Form Tolerance; A = maximum sagitta error in fringes or - if
or tolerance is part of radius tolerance, B = peak-to-valley irregularity in
3/A(B/C) RMSx < D fringes or - if no tolerance is given, C = non-spherical rotationally
symmetric error in fringes. If (B/C) is replaced by A(B), no tolerance is
or given.
3/- RMSx < D For RMS tolerances, D = maximum rms error in fringes, t = total rms
where x is either t, i, or a deviation from nominal, i = rms irregularity, a = rms asymmetry remaining
after asphericity is subtracted from irregularity.
4/S(L) Centering Tolerance; S = surface tilt angle in minutes of arc, L = lateral
displacement
5/NxA; CNxA; LN" x A"; EA Surface Imperfection Method 1; N = number of imperfections, A = grade
number (square root of imperfection area), C = coating imperfection
designation, N = number of coating imperfections, A = grade number, E
= edge chips designation, A = chip protrusion from edge, L = long
scratch designation, N" = number of scratches, A" = grade number
(scratch width (mm))

Symbol or Code Number Description
5/TV or Surface Imperfection Method 2; T = transmissive test, V = visibility class
/TV; EA" number, R = reflective test, E,A" = edge chips, as in Method 1 above.
5/RV or
RV; EA"
6/Hth;L;pdg;fp;nts x np Laser Irradiation Damage Threshold; Hth = energy density threshold, L
or = laser wavelength (nm), pdg = pulse duration group, fp = pulse repetition
6/Eth;L;nts rate (Hz), nts = number of test sites, np = number of pulses per test site,
Eth = power density threshold in watt-cm^2.
Diameters and bevels

The semi-diameter value is used to determine the drawn radial aperture of each surface. A flat bevel is added
to any surface whose semi-diameter is less than the largest semi-diameter being drawn. To place flat bevels on
all surfaces, use the "Dia (flat) optional" setting available on the "L Surf Codes 3-4" tab. This optional value must
exceed twice the semi-diameter of every surface drawn, otherwise the value is ignored.
Tolerance data
When the "Reset from TDE" button is pressed, radius, thickness, index, surface tilt, and surface decenter
tolerances are taken from the tolerance data editor. The tolerances are automatically reset after changing the
surface number or show as settings. Only TRAD, TCUR, TFRN, TTHI, TIND, TABB, TSTX, TSTY, TSDX, and
TSDY tolerances are considered. If a tolerance is not provided, the tolerance is set to zero. The TFRN tolerance
is applied to the surface form tolerance code 3/A. Note all tolerance fields may be edited to suit any requirement.
NSC 3D Layout
Purpose:
Draws 3D layout plots of the sources and objects in a single NSC group.
Settings:
Item Description
Split Rays If checked, rays from NSC sources will be statistically split at ray-surface intercepts.
Rays entering from the entry port are not affected by this setting.
Scatter Rays If checked, rays from NSC sources will be statistically scattered at ray-surface
intercepts. Rays entering from the entry port are not affected by this setting.
Use Polarization If checked, polarized rays will be traced for each ray required, and the resulting
transmitted intensity through the system will be accounted for. See the System Menu
chapter under Polarization for information on defining the polarization state and other
details.
Color Rays By Select "Sources" to use color to distinguish rays traced from each source, unless a
source has a user defined color assigned to that source, or "Waves" to distinguish
between each wavelength, or "Config" to distinguish between configurations.

Item Description
Rotation About X The angle in degrees by which the lens appears to be rotated about the X axis.
Rotation About Y The angle in degrees by which the lens appears to be rotated about the Y axis.
Rotation About Z The angle in degrees by which the lens appears to be rotated about the Z axis.
Offset X, Y, Z The X, Y, and Z direction offset between configurations in lens units. Only has an affect
on the drawing if "All" configurations are being drawn.
Filter If blank, all rays are drawn. Otherwise, only rays which meet the criteria defined by the
filter string will be drawn. See The filter string on page 371 for information about the
filter string syntax.
Ray Database If "none" is selected, new rays will be traced and displayed. If a ZRD data base file is
selected, then rays contained within the database will be displayed. In either case the
filter, if any, is applied. Generally, reading rays from a large database is faster than
retracing them. The other advantage to using a database of rays is that the rays are
always the same, until the database is replaced. ZEMAX cannot tell if the selected
database is for rays from the current lens being displayed; so care should be taken when
selecting the ZRD file that it corresponds to the lens data being displayed. For more
information on the ZRD file, see Ray database files on page 366.
Discussion:
The settings in the above table are very much like their counterparts on the 3D Layout feature. However, this
feature only draws objects in a single NSC group, and only traces and draws rays from sources defined within
the group.
To make surfaces partially transparent, see Surface properties type tab on page 67.
NSC Shaded Model
Purpose:
Draws a shaded solid model representation of the lens using OpenGL graphics.
Settings:
The options are very similar to those available for the NSC 3D Layout and Shaded Model features. One
additional feature is the option to color detector objects by either the energy incident on the detector in the last
analysis or by just the rays traced in the layout view. The detector shows false color or black and white displays,
using either coherent intensity or any of the other options supported by the detector viewer. See Detectors on
page 332 for details.
There is an additional control labeled "opacity". The options are to ignore, which ignores the surface and object
opacity values and draws all surfaces and objects as opaque; and method 1 or 2, which use different algorithms
to render the scene. Some systems are difficult for OpenGL to render when there are a mixture of opaque and
transparent surfaces and objects; choose the method which yields the preferred rendering. For information on
setting the opacity of surfaces and objects see Surface Opacity on page 67 and The object properties dialog
box on page 348.
Fans
Ray Aberration
Purpose:
Shows ray aberrations as a function of pupil coordinate.

Settings:
Item Description
Plot Scale Sets the maximum vertical scale for the plots. The maximum scale is in micrometers for
ray fans, waves for OPD plots, or percent for entrance pupil aberration plots. This
overrides the automatic selection of scale for the plots. Enter zero for automatic scaling.
Number of Rays This is the number of rays traced on each side of the origin of the plot.
Wavelength The wavelength number for which the calculation should be performed.
Field The field number for which the calculation should be performed.
Tangential Selects which aberration component to plot for the tangential fan. Since tangential fans
are functions of the y pupil coordinate, the default is to plot the y component of the
aberration.
Sagittal Selects which aberration component to plot for the sagittal fan. Since sagittal fans are
functions of the x pupil coordinate, the default is to plot the x component of the
aberration.
Use Dashes Selects either solid lines or dashed lines to differentiate the various curves.
Check Apertures Specifies whether or not to check if rays pass all surface apertures. If this is selected,
rays which do not pass surface apertures will not be drawn.
Vignetted Pupil If checked, the pupil axis will be scaled to the unvignetted pupil, in which case the data
will reflect the vignetting in the system. If unchecked, the pupil axis will be scaled to fit
the vignetted pupil.
Surface Selects the surface at which the fan is to be evaluated. This is useful for evaluating
intermediate images. See Evaluating results at intermediate surfaces on page 109.
Discussion:
The tangential fans show either the x or the y component of the transverse ray aberration as a function of the
y pupil coordinate of the ray. The default option is to plot the y component of the aberration. However, since
transverse ray aberrations are vectors, this is an incomplete description of the aberration. When ZEMAX plots
the y component, the plot is labeled EY, when plotting the x component of the aberration, the plot is labeled EX.
The vertical axis scale is given at the bottom of the graph. The data being plotted is the difference between
the ray intercept coordinate and the chief ray intercept coordinate. The tangential fan is the plot of the difference
between the ray x or y coordinate and the chief ray x or y coordinate at the primary wavelength, as a function of
the y pupil coordinate. The sagittal plot is the difference between the ray x or y coordinate and the chief ray x or
y coordinate as a function of the x pupil coordinate. The horizontal scale for each graph is the normalized entrance
pupil coordinate, either PX or PY.
If "All" wavelengths are shown, then the plot is referenced to the primary wavelength chief ray. If monochro-
matic, then the chief ray for the selected wavelength is used as a reference. For this reason, the data for non-
primary wavelengths will in general change when switching between monochromatic and polychromatic displays.
Because ray aberrations are vectors, with both an x and a y component, the ray aberration fan is an incomplete
description of the aberrations, especially when the image plane is rotated or the system is otherwise non-
rotationally symmetric. Also, the fans only indicate aberrations along two "slices" through the pupil, rather than
over the entire entrance pupil. The primary purpose of the ray fan plot is to determine what aberrations are present
in the system; it is not a complete description of the system performance, especially for systems without rotational
symmetry.
Evaluating results at intermediate surfaces
ZEMAX can compute analysis results for surfaces other than the image surface by applying assumptions.
These assumptions work in most cases, however, there may be cases where the methods described here for
analyzing intermediate surfaces is not appropriate, including those systems which require significant use of ray

aiming. All of the changes made to the lens described are made on a copy of the original lens data made for
analysis purposes so no changes are made to the original lens data.
If the field type is either real or paraxial image height, the field type is changed to angle or object height for
infinite or finite conjugate systems, respectively. The angles and heights used correspond to the primary
wavelength chief ray angles and heights as computed for the unaltered system.
If the selected surface precedes the stop surface, ZEMAX moves the stop surface to a (possibly virtual) dummy
space prior to the existing surface 1, and changes the system aperture to an entrance pupil diameter equal to the
original paraxial entrance pupil diameter computed for the original stop position. If the selected surface follows
the stop surface, no changes are made to the stop definition.
Then, surfaces which follow the selected surface are deleted. The glass of the new image surface is set to be
the same as the selected surface.
Most analysis features that compute results for intermediate surfaces make more sense if the rays are allowed
to come to a focus after refraction from the desired surface. For example, the OPD plot is generally a useful
diagnostic only when the OPD is measured at the (possibly virtual) focus of that surface. Other features which
require a temporary image to be formed include PSF, MTF, and diffraction encircled energy. For these features
requiring a temporary image, the new image surface is set to be a standard plane surface. A paraxial marginal
ray height solve is placed on the selected surface thickness to place the image surface at paraxial focus for the
selected surface. The analysis computation then proceeds at this newly created intermediate image surface. Note
the analysis occurs at the paraxial focus formed by the rays after refracting through the selected surface.
Some analysis features make more sense if the data is evaluated directly on the surface itself, without allowing
the rays to focus to an image. These features include the various spot diagrams, footprint analysis, geometric
encircled energy, line/edge spread function, and extended source encircled energy.
Optical Path
Purpose:
Shows optical path difference as a function of pupil coordinate.
Settings:
The options are identical to those for ray aberration fans, except the only option for "Tangential" and "Sagittal"
is OPD, since OPD is a scalar quantity.
Discussion:
The vertical axis scale is given at the bottom of the graph. The data being plotted is the optical path difference,
or OPD, which is the difference between the optical path length of the ray and the optical path length of the chief
ray. Usually, the calculation is referenced back to the difference between the ray path lengths at the system exit
pupil. The horizontal scale for each graph is the normalized entrance pupil coordinate.
If "All" wavelengths are shown, then the plot is referenced to the primary wavelength based reference sphere
and chief ray. If monochromatic, then the reference sphere and chief ray for the selected wavelength is used as
a reference. For this reason, the data for non-primary wavelengths will in general change when switching between
monochromatic and polychromatic displays.
Pupil Aberration
Purpose:
Shows entrance pupil distortion as a function of pupil coordinate.
Settings:
The options are identical to those for ray aberration fans, except the only option for "Tangential" and "Sagittal"
is pupil aberration, since pupil aberration is a scalar quantity. The "Surface" only allows "Image" because this data
is always computed at the stop surface.
Discussion:
Entrance pupil aberration is defined as the difference between the real ray intercept on the stop surface and
the on axis primary wavelength paraxial ray intercept as a percentage of the paraxial stop radius. If the maximum
aberration exceeds a few percent, ray aiming (see Ray Aiming on page 88) may be required to get the rays in
object space to correctly fill the stop surface. If ray aiming is switched on, the entrance pupil aberration will appear

to be zero (or a very small residual value), because the distortion is accounted for by the ray tracing algorithm.
This is used as a check that ray aiming is working correctly. The definition used here for pupil aberration is not
intended to be complete or in agreement with other definitions. The sole purpose of this feature is to provide
guidance as to whether or not ray aiming is required.
Spot Diagrams
Standard
Purpose:
Show spot diagrams.
Settings:
Item Description
Pattern The pattern may be either hexapolar, square, or dithered. These terms refer to the
pattern of rays as they appear in a pupil plane. Defocus the lens significantly to see the
pattern, if desired. Dithered spot diagrams are generated by pseudo-random rays which
eliminate the symmetrical artifacts in the spot diagram typical of rectangular or hexapolar
patterns. The pattern is distorted to give the correct distribution of rays if pupil
apodization is specified. There is no best pattern to use, each shows a different
character of the spot diagram.
Refer To The spot diagrams by default are referenced to the real chief ray. The RMS and GEO
spot radii listed at the bottom of the diagram (and defined in the discussion section) are
calculated assuming the chief ray is the "zero aberration" point. However, this option
allows selection of other reference points. The centroid is defined by the average
position of the rays traced. The middle is defined so that the maximum ray errors are
equal in the plus and minus x and y directions. The vertex is defined by the local
coordinates 0,0 on the selected surface.
Show Scale Scale bar is the default. Selecting "Airy Disk" will draw an elliptical ring around each spot
showing the size of the Airy ellipse. The Airy disk radius is 1.22 times the wavelength
(primary wavelength is used if polychromatic) times the F/# of the system; which in
general depends upon field position and pupil orientation. If the Airy disk is larger than
the spot, the Airy disk will set the scale size, otherwise the spot radius will set the scale.
Selecting "Square" will draw a box, centered on the reference point, whose width is twice
the distance from the reference point to the outermost ray. Selecting "Cross" will draw
a cross through the reference point location. The "Circle" setting will draw a circle
centered on the reference point.
Wavelength The wavelength number for which the calculation should be performed.
Surface Selects the surface at which the spot diagram is to be evaluated. This is useful for
evaluating intermediate images or vignetting.
Plot Scale Sets the maximum scale size of the display in micrometers. A setting of zero will
generate an appropriate scale.
Delta Focus The delta focus option is only used if through-focus spot diagrams are selected. This is
the Z-axis spacing between spot diagram planes. Five spot diagrams will be shown for
each field angle. The defocus will be -2, -1, 0, 1, and 2 times the delta focus provided.
The delta focus units are in micrometers. For some systems the default delta focus may
be too small to cause a change in the spot diagram structure.
Ray Density The ray density specifies the number of hexapolar rings to be traced if a hexapolar or
dithered pattern is selected, or the number of rays across the width and height if a
rectangular pattern is selected. The more rays traced, the greater the accuracy of the
RMS spot radius, although the computation time increases. There are 6 rays in the first
hexapolar ring, 12 in the second, 18 in the third, and so on.

Item Description
Use Symbols If checked, this option will draw different symbols rather than dots for each wavelength.
This helps distinguish the various wavelengths.
details. Only ZEMAX-EE supports this capability.
Scatter Rays If checked, rays will be statistically scattered at ray-surface intercepts that have defined
scattering properties. Only ZEMAX-EE supports this capability.
Direction Cosines If checked, the data presented will be the direction cosines of the rays rather than the
spatial coordinates of the rays. The x direction data will be the x direction cosine of the
ray, the y direction data will be the y direction cosine. The image coordinates will also
be given as the reference point direction cosines. Direction cosines are dimensionless.
distinguish between each wavelength, or "Config" to distinguish between configurations.
Discussion:
The ray density has a maximum value based upon the number of fields displayed, the number of wavelengths
defined, and available memory. Through-focus spot diagrams will trace half of the maximum number of rays
possible on standard spot diagrams.
The GEO spot radius listed on the plot for each field point is the distance from the reference point (which is
either the chief ray at the primary wavelength, the centroid of all the rays traced, or the middle of the spot cluster)
to the ray which is farthest away from the reference point. In other words, the GEO spot radius is the radius of
the circle centered at the reference point which encloses all the rays.
The RMS spot radius is the root-mean-square radial size. The distance between each ray and the reference
point is squared, and averaged over all the rays, and then the square root is taken. The RMS spot radius gives
a rough idea of the spread of the rays, since it depends upon every ray. The GEO spot radius only gives
information about the one ray which is farthest from the reference point.
For information on the X and Y RMS spot radii, see the "text" listing for the spot diagram.
The Airy disk radius is given by 1.22 times the wavelength (primary wavelength is used if polychromatic) times
the F/# of the beam, which in general depends upon field position and pupil orientation. This is the radius to the
first dark ring of the Airy disk for a circular, uniformly illuminated entrance pupil. The Airy disk may be optionally
drawn to give an idea of the scale of the plot. For example, if all the rays are well within the Airy disk, then the
system is often said to be "diffraction limited". If the RMS spot radius is significantly larger than the Airy disk radius,
then the system is not diffraction limited. The threshold for diffraction limited performance depends upon which
criterion is used. There is no absolute boundary at which the system becomes diffraction limited. The Airy disk
shown is not an accurate representation of the diffraction dark ring shape or size if the system does not have
uniform illumination or if vignetting is used to eliminate some of the rays. ZEMAX does not plot vignetted rays on
spot diagrams, nor are they used in computing the RMS or GEO spot radii.
ZEMAX generates grids of rays based upon the wavelength weighting factors and the pupil apodization, if any.
The wavelength with the largest weight uses the maximum grid size set by the "Ray Density" option. Wavelengths
with lower weights use grids with fewer rays to maintain the correct representation in the diagram. Ray grids are
also distorted to maintain the correct ray distribution, if apodization is specified. The RMS spot radius stated on
the spot diagram considers the wavelength weighting and apodization factors. However, it is only an estimate of
the RMS spot radius based on the rays actually traced. It is not a very accurate estimate for some systems.
The image plane intercept coordinates of the reference point are shown underneath each spot diagram. If a
surface other than the image plane is specified, then the coordinates are the intercept coordinates of the reference

point on that surface. Since the reference point may be selected to be the centroid, this provides a convenient
way of determining the centroid coordinates.
Through Focus
Purpose:
Show spot diagrams as they change through focal plane shifts.
Settings:
The options are identical to the standard spot diagram.
Discussion:
The through focus spot diagrams are useful for estimating astigmatism, or for analyzing best focus or depth
of focus.
Full Field
Purpose:
Shows spot diagram with all field points on a common scale.
Settings:
The options are identical to the standard spot diagram.
Discussion:
The "Full Field" spot diagram type is similar to the "Standard" type, except all of the spots are plotted with
respect to the same reference point, as opposed to a separate reference point for each field position. This
provides some idea of how the spot would look relative to the other field points. For example, this can be used to
determine if two closely spaced image points can be resolved. The "Full Field" spot diagram type is useless if the
spot radius is small compared to the total field size, because in this case the spots for each field will appear as
"dots". If "chief ray" is selected as the reference point, then the chief ray for field position 1 will be used.
Matrix
Purpose:
Show spot diagram as a matrix of individual diagrams, with each field along a row and each wavelength down
a column.
Settings:
The options for the matrix spot diagram are similar to those for the standard spot diagram, except for the
addition of the following option.
Item Description
Ignore Lateral Color If checked, this option will reference each spot diagram to the reference point for each
field and wavelength independently. This in effect ignores the effects of lateral color
which can displace the reference points for each wavelength.
Discussion:
The matrix representation is a convenient way of distinguishing the components of wavelength dependent
aberrations.
Configuration Matrix
Purpose:
Show spot diagram as a matrix of individual diagrams, with each field along a row and each configuration down
a column.
Settings:
The options for the configuration matrix spot diagram are similar to those for the standard spot diagram.
Discussion:

The configuration matrix representation is a convenient way of distinguishing the components of configuration
dependent aberrations. Along the left side of the plot the field positions are listed; only the field positions for the
last configuration are listed if more than one configuration is shown and the field definitions are part of the multi-
configuration data which changes.
MTF
FFT MTF
Purpose:
Computes the diffraction modulation transfer function (MTF) data for all field positions using an FFT algorithm.
Settings:
Item Description
Sampling The size of the ray grid used to sample the pupil. The sampling may be 32x32, 64x64,
etc. Although higher sampling yields more accurate data, calculation times increase.
Show Diffraction Select whether or not the diffraction limited data should be displayed.
Limit
Max Frequency Specify the maximum spatial frequency (in cycles per millimeter) plotted.
Wavelength The wavelength number to be used in the calculation.
Field The field number to be used in the calculation.
Type Select either modulation, real, imaginary, phase, or square wave response.
Surface Selects the surface at which the MTF is to be evaluated. This is useful for evaluating
Discussion:
!See the discussion sections of the FFT and Huygens PSF. Those comments also apply to this
feature.
The diffraction MTF computation is based upon an FFT of the pupil data. The resulting MTF is the modulation
as a function of spatial frequency for a sine wave object, although optionally the real, imaginary, phase, or square
wave response is available. The Square wave MTF is the modulation response for a square wave target of the
specific frequency, as opposed to the response to a sine wave target for the other plots. The square wave
response is computed from the MTF data using the following formula:
4 M ( v ) M ( 3v ) M ( 5v ) M ( 7v )
( v ) = --- ------------- ---------------- + ---------------- ---------------- +
1 3 5 7
where S(v) is the square wave response, M(v) is the sinusoidal modulation response, and v is the spatial
frequency.
The cutoff frequency at any one wavelength is given by one over the wavelength times the working F/#. ZEMAX
computes the working F/# at each wavelength for each field for the sagittal and tangential response separately.
This yields accurate MTF data even for systems with anamorphic and chromatic distortion, such as those
incorporating cylinders or gratings.

The diffraction calculations are more accurate as the sampling increases, the peak-to-valley and maximum
slope of the OPD decrease, and the transverse ray aberrations decrease. If the peak-to-valley OPD in the pupil
is too great, then the wavefront sampling is too coarse and aliasing occurs. Aliasing will result in inaccurate data.
ZEMAX will attempt to detect when aliasing occurs, and issue an appropriate error message. However, ZEMAX
cannot automatically detect when the sampling is too low in all cases, especially in the presence of very steep
slopes on the wavefront phase.
The FFT based MTF assumes a (reasonably) uniform distribution of rays on the exit pupil in cosine space to
be accurate. Some systems, such as extremely fast off-axis reflectors, have dramatic stretching of the exit pupil
and the FFT based MTF will thus be inaccurate. For these systems, the Huygens MTF should be used instead.
For more information, see Huygens MTF on page 118.
When the OPD in waves is large, such as more than 10 waves, it is probably a good idea to switch to geometric
MTF instead of diffraction MTF. For these highly aberrated systems, the geometric MTF is very accurate,
especially at low spatial frequencies (the higher frequency MTF falls of rapidly when aberrations are large). If
shown, the diffraction limit curve is for the aberration free response at the reference field position; see Diffraction
limited on page 46.
The spatial frequency scale of the MTF plot is always in cycles per mm in image space, which is the correct
term for sinusoidal MTF response. The term line pairs per mm is often used, but strictly speaking line pairs per
mm only applies to bar, as opposed to sinusoidal targets. ZEMAX uses these terms interchangeably, as is
common in the industry. MTF is always measured in image space, so any magnification of the system needs to
be considered when determining spatial frequency response for object space.
The nature of the FFT algorithm is that the computation is done in pupil space coordinates. For this reason,
rotating the image surface will have no effect on the orientation of the computed MTF. The tangential response
corresponds to a periodic target oriented with lines along the object space X axis, and the sagittal response
corresponds to a periodic target oriented with lines along the object space Y axis. This is different from the
conventions of the Geometric MTF on page 119 and the Huygens MTF on page 118.
FFT Through Focus MTF
Purpose:
Computes the FFT modulation transfer function data as a function of focus shift at a specific spatial frequency.
Settings:
Item Description
Delta Focus The range of defocus used.
Frequency The spatial frequency (in cycles per millimeter) for which data is plotted.
# Steps The number of focal planes at which the data is computed. A smooth curve is drawn
through the computed points. More steps yield higher accuracy and longer computation
times.
Type Select either modulation, real, imaginary, phase, or square wave response.

Discussion:
See FFT MTF on page 114 for details.
FFT Surface MTF
Purpose:
Displays the FFT computed MTF data as a 3D surface, contour, grey scale or false color map. This plot is
useful for visualizing the MTF response for object orientations other than purely sagittal or tangential.
Settings:
Item Description
Rotation Rotation specifies how the surface plots are rotated for viewing; either 0, 90, 180, or 270
degrees.
Scale The scale factor overrides the automatic vertical scaling set by the program on the
surface plots. Generally, this value should be set to unity. The scale factor can be greater
than unity to vertically stretch the plot, or less than unity to compress it.
Field The field number determines for which defined field position the calculation should be
performed.
Show As Choose surface plot, contour map, grey scale, or false color map as the display option.
Discussion:
The regular MTF plot is just two orthogonal cross sections through the surface MTF plot. This plot is primarily
qualitative. See FFT MTF on page 114 for details.
FFT MTF vs. Field
Purpose:
Computes the FFT MTF data as a function of field position, and displays the data in a graph.
Settings:
Item Description
Frequency 1-6 The spatial frequencies (in cycles per millimeter) for which data is plotted.
Remove Vignetting If checked, vignetting factors are automatically removed. See the comments about
Factors vignetting factors in the discussion section below.

Item Description
Field Density The field density is the number of points between zero degrees and the maximum field
at which the MTF is calculated, intermediate values are interpolated. A maximum of 100
field points is allowed.
Scan Type Choose +y, +x, -y, or -x field scan direction.
Discussion:
See FFT MTF on page 114 for details. This feature plots MTF vs. field height up to the maximum defined
radial field coordinate.
Comment about vignetting factors
Vignetting factors determine the size and shape of the pupil as seen from different field points (see Vignetting
factors on page 96 for a full discussion). Because this analysis feature needs to trace rays at arbitrary
intermediate field points where no specific vignetting factors are defined, the use of vignetting factors is not
recommended. If "Remove Vignetting Factors" is checked on (the default), any defined vignetting factors will
automatically be replaced with surface apertures for this computation. The surface aperture method is generally
more accurate than the vignetting factor method when the pupil is overfilled with light. The resulting data may be
different between the two methods. In some cases, particularly where the vignetting factors are being used to
define the shape of the source beam rather than the apertures of the optics, it may be required to use the defined
vignetting factors. In this case, check the "Remove Vignetting Factors" box off. ZEMAX will then use the closest
defined field to determine the vignetting factors to use for an arbitrary field point.
FFT MTF Map
Purpose:
Computes the FFT MTF as a function of field position, and displays the data over a rectangular region of field.
Settings:
Item Description
X or Y Field Width The X or Y field width in field units. This is the total width or height, not the half width or
height. Field units are degrees in object space if field angle is used, otherwise field units
are the same as lens units.
Frequency The spatial frequency at which to compute the MTF.
Wavelength The wavelength number to be used in the calculation, or All for polychromatic MTF.
X or Y Pixels The number of pixels at which to compute the MTF in each respective direction. Note
the size of the pixels is determined by both the number of pixels and the width of the
field; the pixels are not required to be square. The MTF is computed at the center of the
pixel and the MTF is assumed to have that value over the entire region of the pixel for
display purposes.
MTF Data Choose Tangential, Sagittal, or average MTF to be displayed.
Reference Field This control selects the field number that corresponds to the center of the map. If zero
is selected the (0, 0) field coordinate is used as the center of the map.

Item Description
Show As Choose grey scale or false color map as the display option.
Remove Vignetting If checked, vignetting factors are automatically removed. See Comment about
Factors vignetting factors on page 117.
Discussion:
See FFT MTF on page 114 for details.
This feature computes the MTF at each field point on a 2D grid. If the total number of points is large, the
computation time may become quite large. See also the Geometric MTF Map feature on page 121.
Huygens MTF
Purpose:
Computes the diffraction modulation transfer function (MTF) data using a Huygens direct integration algorithm.
Settings:
Item Description
Pupil Sampling Selects the size of the grid of rays to trace to perform the computation. Higher sampling
densities yield more accurate results at the expense of longer computation times.
Image Sampling The size of the grid of points on which to compute the diffraction image intensity. This
number, combined with the image delta, determine the size of the PSF used to compute
the MTF. See also Huygens PSF on page 126.
Image Delta The distance in micrometers between points in the image grid.
Zero Padding Adds a zero-value guard band around the computed PSF before performing the Fourier
transform.
Configuration Select "All" to perform a calculation of the MTF based upon a coherent sum of the PSF
at each wavelength across all configurations, or select the "Current" or any single
configuration. Note the MTF is computed from the PSF, which is a coherent sum for the
same wavelength in each configuration, followed by an incoherent sum of the resulting
PSFs for different wavelengths. For this reason, each defined wavelength must be the
same in all configurations. Wavelength and configuration weights may be used but the
wavelength values must be identical. See CWGT and WLWT in Summary of multi-
configuration operands on page 471. This coherent sum also assumes that the image
surface is located in the identical position in all configurations.
Type Select the data to display, currently modulation is the only option.
Max Frequency The maximum spatial frequency in cycles per millimeter to display.
Discussion:
!See the discussion sections of the Huygens PSF. Those comments also apply to this feature.
The Huygens MTF computes an FFT of the Huygens PSF. The initial PSF sampling settings for Image
Sampling and Image Delta settings are the same as for the Huygens PSF, therefore it is instructive to do a
Huygens PSF first (see Huygens PSF on page 126). The zero padding setting adds zero intensity values around
the PSF and has the visual effect of increasing the pixel density in the transform. Since the transform is done on
the PSF in image space coordinates, the tangential response corresponds to spatial frequencies in the y direction
in local image surface coordinates, and the sagittal response corresponds to spatial frequencies in the x direction.
The Huygens MTF also has no dependence on the location of rays in the paraxial pupils. The MTF can therefore
be computed for any system that the Huygens PSF can be computed for including many non-sequential systems
using ports where reference rays required by other diffraction algorithms would not make it through, or for systems
where pupils or images formed by multiple non-sequential sub-apertures are overlapped. Systems with extreme
exit pupil distortion, such as very fast off-axis reflectors, are also handled correctly with the Huygens technique.
The spatial frequency scale of the MTF is always in cycles per mm in image space.
The nature of the Huygens algorithm is that the computation is done in image space coordinates. For this
reason, rotating the image surface will affect the orientation of the computed MTF. The tangential response
corresponds to the image of a periodic target oriented with lines along the image space X axis, and the sagittal
response corresponds to the image of a periodic target oriented with lines along the image space Y axis. This is
different from the conventions of the FFT MTF on page 114.
Huygens Through Focus MTF
Purpose:
Computes the diffraction modulation transfer function (MTF) data using a Huygens direct integration algorithm
and displays the data as a function of delta focus.
Settings:
Item Description
times.
The other settings are the same as for the Huygens MTF feature, as described on page 118.
Discussion:
This feature is very similar to the Huygens MTF feature, as described on page 118.
Huygens Surface MTF
Purpose:
Computes the diffraction modulation transfer function (MTF) data using a Huygens direct integration algorithm
and displays the data as a surface, grey scale, false color, or contour plot.
Discussion:
This feature is very similar to the Huygens MTF feature, as described on page 118.
Geometric MTF
Purpose:
Computes the geometric MTF, which is an approximation to the diffraction MTF based upon ray aberration
data.

Settings:
Item Description
Max Frequency The maximum spatial frequency (in cycles per millimeter) for which data is plotted.
Multiply by When checked, will scale the geometric MTF by the diffraction limited MTF to yield a
Diffraction Limit more realistic result for systems with small aberrations. Should always be used.
Discussion:
The geometric MTF is a useful approximation to the diffraction MTF if the system is not close to the diffraction
limit. The primary advantage to using the geometric MTF is for systems which have too many waves of aberration
to permit accurate calculation of the diffraction MTF. The geometric MTF is also very accurate at low spatial
frequencies for systems with large aberrations.
The nature of the geometric algorithm is that the computation is done in image space coordinates. For this
reason, rotating the image surface will affect the orientation of the computed MTF. The tangential response
corresponds to the image of a periodic target oriented with lines along the image space X axis, and the sagittal
response corresponds to the image of a periodic target oriented with lines along the image space Y axis. This is
different from the conventions of the FFT MTF on page 114.
Geometric Through Focus MTF
Purpose:
Computes the geometric MTF data through focus at a specific spatial frequency.
Settings:
Item Description
times.
Multiply by When checked, will scale the geometric MTF by the diffraction limited MTF to yield a
Diffraction Limit more realistic result for systems with small aberrations. Should always be used.

Item Description
Discussion:
See the Geometric Transfer Function section on page 119 for details.
Geometric MTF vs. Field
Purpose:
Computes the geometric modulation transfer function data as a function of field position.
Settings:
The settings are identical to those for the FFT MTF vs. Field feature, with the added ability to scatter rays, and
a choice to scale by the diffraction limit or not.
Discussion:
This feature is nearly identical to the (diffraction) MTF vs. Field feature, except the geometric MTF is used
rather than the diffraction based MTF.
Geometric MTF Map
Purpose:
Computes the geometric modulation transfer function data as a function of field position, and displays the data
over a rectangular region of field.
Settings:
Item Description
X or Y Field Width The X or Y field width in field units. This is the total width or height, not the half width or
height. Field units are degrees in object space if field angle is used, otherwise field units
are the same as lens units.
Frequency The spatial frequency at which to compute the GMTF.
Wavelength The wavelength number to be used in the calculation, or All for polychromatic GMTF.
X or Y Pixels The number of pixels at which to compute the GMTF in each respective direction. Note
the size of the pixels is determined by both the number of pixels and the width of the
field; the pixels are not required to be square. The GMTF is computed at the center of
the pixel and the GMTF is assumed to have that value over the entire region of the pixel
for display purposes.
MTF Data Choose Tangential, Sagittal, or average GMTF to be displayed.

Item Description
Reference Field This control selects the field number that corresponds to the center of the map. The
value zero refers to the center of the object (hx = 0, hy = 0), even if no field point is
defined there.
Show As Choose grey scale or false color map as the display option.
Discussion:
See the Geometric Transfer Function on page 119 for details. This feature computes the GMTF at each field
point on a 2D grid. If the total number of points is large, the computation time may become quite large. See also
the (Diffraction) MTF Map feature on page 117.
PSF
FFT PSF
Purpose:
Computes the diffraction point spread function (PSF) using the Fast Fourier Transform (FFT) method.
Settings:
Item Description
Display The display size indicates what portion of the computed data will be drawn when a
graphic display is generated. The display grid can be any size from 32 x 32 up to twice
the sampling grid size. Smaller display sizes will show less data, but at higher
magnification for better visibility. This control has no effect on the text display.
degrees.
Type Select linear (intensity), logarithmic (intensity), or phase.
Image Delta The delta distance between points in image space, measured in micrometers. If zero, a
default spacing is used. If negative, the image delta is set to the maximum allowed value
and the full sampling grid is used. See the discussion for details.
Normalize If checked, the peak intensity will be normalized to unity. Otherwise, the peak intensity
is normalized to the peak of the unaberrated PSF (the Strehl ratio).
Surface Selects the surface at which the PSF is to be evaluated. This is useful for evaluating

Discussion:
The FFT method of computing the PSF is very fast, however, a few assumptions are made which are not
always valid. The slower, but more general Huygens method makes fewer assumptions, and is described in
Huygens PSF on page 126.
Assumptions used in the FFT PSF calculation
The FFT PSF computes the intensity of the diffraction image formed by the optical system for a single point
source in the field. The intensity is computed on an imaginary plane which is centered on and lies perpendicular
to the incident chief ray at the reference wavelength. The reference wavelength is the primary wavelength for
polychromatic computations, or the wavelength being used for monochromatic calculations. Because the
imaginary plane lies normal to the chief ray, and not the image surface, the FFT PSF computes overly optimistic
(a smaller PSF) results when the chief ray angle of incidence is not zero. This is often the case for systems with
tilted image planes, wide angle systems, systems with aberrated exit pupils, or systems far from the telecentric
condition.
The other main assumption the FFT method makes is that the image surface lies in the far field of the optical
beam. This means the computed PSF is only accurate if the image surface is fairly close to the geometric focus
for all rays; or put another way, that the transverse ray aberrations are not too large. There is no hard and fast
limit, however if the transverse aberrations exceed a few hundred wavelengths, the computation is likely not
accurate. Note that even systems with very little wavefront aberration can have large transverse ray aberrations;
for example, a cylinder lens which only focuses rays along one direction. In this case, the transverse aberrations
along the unfocused direction will be on the order of the beam diameter. The Huygens PSF method may provide
more accurate results in these cases as well.
For most lenses, a less important assumption is that scalar diffraction theory applies. The vectorial nature of
the light is not accounted for. This is significant in systems that are very fast, around F/1.5 (in air) or faster. The
scalar theory predicts overly optimistic (a smaller PSF) results when the F/# is very fast.
For systems where the chief ray is nearly normal (less than perhaps 20 degrees), the exit pupil aberrations
are negligible, and the transverse ray aberrations are reasonable, then the FFT PSF is accurate and generally
much faster than the Huygens PSF method.
When in doubt, both PSF methods should be employed for comparison. A solid understanding on the part of
the user of these assumptions and the method of computation is essential to recognize cases where the accuracy
may be compromised.
Discussion of the FFT method and sampling issues
The FFT PSF algorithm exploits the fact that the diffraction PSF is related to the Fourier transform of the
complex amplitude of the wavefront in the exit pupil of the optical system. The amplitude and phase in the exit
pupil are computed for a grid of rays, an FFT is performed, and the diffraction image intensity is computed.
There is a tradeoff between the sampling grid size in the pupil, and the sampling period in the diffraction image.
For example, to decrease the sampling period in the diffraction image, the sampling period in the pupil must
increase. This is done by "stretching" the pupil sampling grid so that it overfills the pupil. This process means
fewer points actually lie within the pupil.
As the sampling grid size is increased, ZEMAX scales the grid on the pupil to yield an increase in the number
of points that lie on the pupil, while simultaneously yielding closer sampling in the diffraction image. Each time
the grid size is doubled, the pupil sampling period (the distance between points in the pupil) decreases by the
square root of 2 in each dimension, the image plane sampling period also decreases by the square root of 2 in
each dimension, and the width of the diffraction image grid increases by a factor of the square root of 2 (since
there are twice as many points in each dimension). All ratios are approximate, and asymptotically correct for large
grids.
The stretching is referenced to a grid size of 32 x 32. The 32 x 32 grid of points is placed over the pupil, and
the points that lie within the pupil are actually traced. For this grid size, the default distance between points in the
diffraction image plane is given by
= F ------------ ,
n2
2n

where F is the working F/# (not the same as the image space F/#), is the shortest defined wavelength, and
n is the number of points across the grid. The -2 factor is due to the fact the pupil is not centered on the grid (since
n is even), but is offset at n/2 + 1. The 2n in the denominator is due to the zero-padding described later.
For grids larger than 32 x 32, the grid is by default stretched in pupil space by a factor of 2 each time the
sampling density doubles. The general formula for the sampling in image space is then
12
= F ------------ ------
n 2 32
,
2n n
and the total width of the image data grid is
W = 2n .
Since the stretching of the pupil grid decreases the number of sample points in the pupil, the effective grid size
(the size of the grid that actually represents traced rays) is smaller than the sampling grid. The effective grid size
increases as the sampling increases, but not as quickly. The following table summarizes the approximate effective
grid size for various sampling density values.
DEFAULT EFFECTIVE GRID SIZES FOR PSF CALCULATIONS
Sampling Grid Size Approximate Effective Pupil Sampling
32 x 32 32 x 32
64 x 64 45 x 45
128 x 128 64 x 64
256 x 256 90 x 90
512 x 512 128 x 128
1024 x 1024 181 x 181
2048 x 2048 256 x 256
4096 x 4096 362 x 362
8192 x 8192 512 x 512
The sampling is also a function of wavelength. The discussion above is only valid for the shortest wavelength
used in the calculation. If the computation is polychromatic, then the longer wavelengths will be scaled to have
smaller effective grids. The scale factor used is the ratio of the wavelengths. This should be considered when
selecting sampling grids for systems with broad wavelength bands. For polychromatic computations, the data for
shorter wavelengths is more accurate than for longer wavelengths.
The image delta, , can be selected manually if a different sampling distance is required. If the image delta
is zero, ZEMAX uses the default spacing and sampling grids described above. If the image delta is greater than
zero, then ZEMAX scales the pupil sampling to yield the desired image delta size. The actual amount of stretching
depends upon the grid size, the image delta, the defined wavelengths, the F/#s at each field and wavelength,
and the aspect ratio of the exit pupil. If the image delta is set too small, then not enough points will be left to sample
the pupil; if the image delta is too big, then the pupil grid will not extend over the full width of the exit pupil. Both
of these cases are trapped by ZEMAX and an error message will be issued if they occur. If the image delta is less
than zero, then ZEMAX does not stretch the pupil at all. This maximizes the extent of the PSF in image
coordinates and uses the full grid of rays in the pupil, at the expense of not increasing the spatial resolution with
sampling, as the image delta will be nearly constant with increased sampling.

Once the sampling is specified, ZEMAX doubles the array size in a process called "zero padding". This means
for a 32 x 32 sampling, ZEMAX uses the center portion of a 64 x 64 grid. Therefore, the diffraction PSF will be
distributed over a 64 x 64 size grid. The sampling in the image space is always twice the pupil sampling. Zero
padding is performed to reduce aliasing.
The nature of the FFT algorithm is that the computation is done in pupil space coordinates. For this reason,
rotating the image surface will have no effect on the orientation of the computed PSF. The X and Y orientation of
the PSF corresponds to the X and Y orientation of rays in the entrance pupil, which is not always the same as
the spatial X and Y orientation of the image. To compute the PSF in the image space coordinates, see Huygens
PSF on page 126.
FFT PSF Cross Section
Purpose:
This feature plots cross sections through the diffraction PSF.
Settings:
Item Description
Row/Col The row or column to display. For a sampling of 32 x 32, there are 64 rows and 64
columns (see the discussion section of the FFT PSF feature). Whether a row or a column
is used depends upon the "Type" setting. The default "Center" displays the center of the
PSF, which is not always the peak of the PSF.
Plot Scale Sets the maximum horizontal scale in micrometers. Enter zero for automatic scaling.
Type Select X or Y cross sections, either linear, logarithmic, or phase. X cross sections are
called rows, and Y cross sections are called columns, but this is arbitrary
Discussion:
!See the discussion section of the FFT PSF. Those comments also apply to this feature.
The cross sections are taken directly from the PSF data. Because the PSF is computed directly from the phase
in the exit pupil, the orientation of the coordinate system may not be correct in all cases. What constitutes positive
x or y may not agree with data presented in image space coordinates such as the spot diagram.
FFT Line/Edge Spread
Purpose:
This feature plots the edge or line spread functions based upon a computation and integration of the diffraction
FFT PSF.

Settings:
Item Description
Spread Choose either line or edge spread function.
Plot Scale Sets the maximum horizontal scale in micrometers. Enter zero for automatic scaling.
Type Select X or Y orientation and either linear or logarithmic scale.
Discussion:
!See the discussion section of the FFT PSF. Those comments also apply to this feature.
The Line Spread Function is computed by integration of the PSF along either rows (for Y orientation) or
columns (for X orientation). The Edge Spread Function is the subsequent running integral of the Line Spread
Function. The X orientation refers to the line or edge being oriented parallel to the X axis of the PSF. The Y
orientation refers to the line or edge being oriented parallel to the Y axis of the PSF. See also Geometric Line/
Edge Spread on page 137.
Huygens PSF
Purpose:
Computes the diffraction PSF using direct integration of Huygens wavelets method. The Strehl ratio is also
computed.
Settings:
Item Description
Pupil Sampling Selects the size of the grid of rays to trace to perform the computation. Higher sampling
densities yield more accurate results at the expense of longer computation times.
Image Sampling The size of the grid of points on which to compute the diffraction image intensity. This
number, combined with the image delta, determine the size of the area displayed.
Image Delta The distance in micrometers between points in the image grid.
Rotation Rotation specifies how the surface plots are rotated; either 0, 90, 180, or 270 degrees.
Type Select linear (intensity), or logarithmic (intensity).

Item Description
Configuration Select "All" to perform a coherent sum of the PSF at each wavelength across all
configurations, or select the "Current" or any single configuration. Note this is a coherent
sum for the same wavelength in each configuration, followed by an incoherent sum of
the resulting PSFs for different wavelengths. For this reason, each defined wavelength
must be the same in all configurations. Wavelength and configuration weights may be
used but the wavelength values must be identical. See CWGT and WLWT in Summary
of multi-configuration operands on page 471. This coherent sum also assumes that the
image surface is located in the identical position in all configurations.
The True Color option creates an RGB color representation of the PSF by converting
the wavelengths to the closest RGB equivalent and summing over all wavelengths. The
accuracy of the True Color presentation is limited by the RGB method of rendering color
on a computer display; and it is not possible to exactly represent monochromatic colors.
Use Centroid If checked, the plot will be centered on the geometric image centroid. If unchecked, the
plot will be centered on the chief ray.
Discussion:
One way of considering the effects of diffraction is to imagine each point on a wavefront as a perfect point
source with an amplitude and phase. Each of these point sources radiates a spherical "wavelet", sometimes
called a "Huygens wavelet" after Huygens, who first proposed the model. The diffraction of the wavefront as it
propagates through space is given by the interference, or complex sum, of all the spherical wavelets radiated.
To compute the Huygens PSF, a grid of rays is launched through the optical system, and each ray represents
a particular amplitude and phase wavelet. The diffraction intensity at any point on the image surface is the
complex sum of all these wavelets, squared. Unlike the FFT PSF, ZEMAX computes the Huygens PSF on an
imaginary plane tangent to the image surface at the chief ray intercept. Note the imaginary plane is normal to the
normal of the surface, not the chief ray. Therefore, the Huygens PSF accounts for any local tilt in the image surface
caused by either the image plane slope, the chief ray incidence angle, or both.
The Huygens method accounts for the evolving shape of the diffraction image as the beam propagates along
the image surface. This is an important effect if the image surface is tilted with respect to the incoming beam.
Another advantage to the Huygens PSF method is that any grid size and spacing, may be selected by the user.
This allows direct comparison between PSF's from two different lenses, even if the F/#'s or wavelengths are
different.
The only disadvantage of the Huygens PSF is speed. Direct integration is slow when compared to the FFT
method (see the previous section for details). The computation time depends upon the pupil grid size squared
times the image grid size squared, times the number of wavelengths. ZEMAX accounts for any symmetry the
system has.
Huygens PSF Cross Section
Purpose:
Computes the diffraction PSF using direct integration of Huygens wavelets method. The Strehl ratio is also
computed. This feature is very similar to the Huygens PSF on page 126, with the difference being the data is
plotted as a cross section.

Wavefront
Wavefront Map
Purpose:
Displays the wavefront aberration.
Settings:
Item Description
degrees.
Scale The scale factor is used to override the automatic vertical scaling set by the program on
the surface plots. The scale factor can be greater than unity to vertically stretch the plot,
or less than unity to compress it.
Reference To By default, the wavefront aberration is referenced to the reference sphere for the
Primary wavelength being used. If this box is checked, then the primary wavelength reference
sphere will be used instead. In other words, checking this box will cause the data to
exhibit the effects of lateral color.
Use Exit Pupil By default, the shape of the pupil is distorted to show the approximate shape of the exit
Shape pupil as seen from the on axis chief ray image point. If this box is unchecked, then
instead the plot will be scaled to circular entrance pupil coordinates, no matter how
distorted the exit pupil may actually be.
Surface Selects the surface at which the data is to be evaluated. This is useful for evaluating
Remove Tilt If checked, the linear X- and Y- tilt is removed from the data. This is equivalent to
referencing the OPD data to the centroid.
Discussion:
See also the Interferogram feature described below.
Interferogram
Purpose:
Generates and displays interferograms.
Settings:
Item Description
Scale Factor Determines the number of fringes per wave of OPD. Useful for modeling double pass
interferometers (i.e. use a scale factor of two).

Item Description
Show As Choose contour map, grey scale, or false color map as the display option.
Wavelength The wavelength is the wavelength number to be used in the calculation.
X-Tilt The number of fringes of tilt to add in the x-direction after applying the scale factor.
Y-Tilt The number of fringes of tilt to add in the y-direction after applying the scale factor.
Beam 1 Selects the first beam for the interferogram. See the discussion.
Beam 2 Selects the second beam for the interferogram. See the discussion.
Ref Beam To Vertex Normally, ZEMAX references the OPD to the chief ray; which in effect subtracts out tilt
from the wavefront phase. For interferometry, it is sometimes desirable to retain the
wavefront tilt. Checking this option on will add tilt to the beam based upon the deviation
of the chief ray from the image surface vertex. This option is only useful for field positions
whose chief ray is reasonably close to the surface vertex, where the assumption that tilt
is described by the deviation of the chief ray is valid.
Discussion:
This feature works by computing two pupil maps, one each from beams 1 and 2. The phase (or OPD) of these
two pupil maps is subtracted, and then optionally some linear phase is added as a function of the x and y pupil
coordinate to simulate tilt fringes. The individual beams may be OPD as computed for any one configuration, or
a "reference" beam which has identically zero OPD may be selected.
Interferometers may be simulated by modeling the two paths through the system using two configurations, and
then computing the interferogram of the two resulting beams. The accuracy of this approach is limited by some
simplifying assumptions:
-Any lateral shift or magnification difference between the two beams is ignored; it is assumed that the pupils
perfectly overlap at the exit pupil.
-Any differences in transmission are ignored; so the two OPD values at any one point in the pupil are assumed
to be of equal value in intensity and the phase can be subtracted to yield the net phase difference.
Foucault Analysis
Purpose:
Generates and displays Foucault knife-edge shadowgrams.
Settings:
Item Description
Type Selects either linear or logarithmic display of the data.
Show As Choose surface, contour map, grey scale, false color map, or cross section as the
display option.
Wavelength The wavelength is the wavelength number to be used in the calculation.
Row/Column When "Show As" is selected to be a cross section, this control defines the row or column
number to display.

Item Description
Knife Choose Horizontal Above, Horizontal Below, Vertical Left, or Vertical Right. The Vertical
Left knife blocks out all the light near focus from the knife position coordinate left; that
is, towards negative x coordinates. The Vertical Right blocks all light from the knife
position rightward. The Horizontal Above blocks all light from the knife position up, and
Horizontal Below blocks all light from the knife position down. The terms left, right, up
and down refer to the -x, +x, +y, and -y directions in the local coordinate system of the
image surface.
Position The position in micrometers relative to the chief ray of the knife. The coordinate is
assumed to be in X or Y depending upon whether an X or Y knife is selected.
Data The computed shadowgram, the reference shadowgram, or the difference between the
two may be selected. See the discussion for details.
Reference The name of the bitmap reference image file.
Decenter X/Y The decenter in X or Y of the reference shadowgram image relative to the computed
shadowgram image. The units are relative to the full width or height of the reference
shadowgram image. For example, an X Decenter of 0.25 will shift the reference image
relative to the computed image by 25% of the full width of the reference image.
Scale X/Y The scale factor in X or Y of the reference shadowgram image pixels relative to the
computed shadowgram image pixels.
Discussion:
This feature simulates the placement of either an X- or Y- oriented knife edge at any position near focus; then
computes the resulting shadowgram after propagating the vignetted beam back to the near field. The method of
calculation involves computing the diffraction based complex amplitude PSF at focus via the FFT method; then
a portion of the complex amplitude is vignetted by the simulated knife edge, and the remaining complex amplitude
is propagated back to the near field. The shadowgram calculated this way is called the "computed" shadowgram
for this feature.
This feature also allows the import of either a BMP or JPG bitmap file of a reference or measured shadowgram.
The reference shadowgram may be displayed for convenient check of orientation.
The difference between the computed and reference shadowgram may be displayed. ZEMAX computes the
RMS difference between the computed and reference shadowgrams, and this RMS difference may be optimized
using the FOUC operand described in the chapter "Optimization". Optimizing the RMS difference permits
quantitative determination of the aberrations present in the beam that created the measured shadowgram. When
calculating the difference between the computed and reference shadowgram, the two images must be registered
together to overlap correctly. The decenter x/y and scale x/y controls are used for registering the two images.
Surface
Surface Sag
Purpose:
Displays the sag of a surface as a 2D color or contour map, or as a 3D surface plot.

Settings:
Item Description
Sampling The size of the grid used. The sampling may be 32x32, 64x64, etc. Although higher
sampling yields more accurate data, calculation times increase. ZEMAX actually
increases the sampling by 1, so that an odd sampling number is used. This allows for a
more symmetric display of the data.
Contour Format The contour format string. For a discussion of the contour format string syntax, see "The
Contour Format String" below. The contours are defined in lens units.
Surface The surface number to compute the sag display for.
Discussion:
This feature accounts for the size and shape of any aperture present on the surface; even if the aperture is
decentered. The sag is computed on a uniform grid of points in XY plane, and the Z value of the sag is the
displayed data.
See also the Surface Phase feature described below.
The Contour Format String
The Contour Format string allows some control over the appearance of the contour map.
If left blank, the default contour spacing will be selected. If a single value is entered, then the contour spacing
will be set to this value. For example, if 0.05 is entered, the contour interval will be 0.05. If multiple values
separated by spaces are entered, only those contours will be drawn. For example, if "0.8 0.5 0.2" is entered, only
those three contours will be drawn. If the specified contour settings result in too many contours to be drawn
properly, the default number of contours will be selected.
Surface Phase
Purpose:
Displays the phase of a surface as a 2D color or contour map, or as a 3D surface plot.
Settings:
Item Description
Sampling The size of the grid used. The sampling may be 32x32, 64x64, etc. Although higher
sampling yields more accurate data, calculation times increase. ZEMAX actually
increases the sampling by 1, so that an odd sampling number is used. This allows for a
more symmetric display of the data.
Contour Format The contour format string. For a discussion of the contour format string syntax, see The
Contour Format String on page 131. The contours are defined in periods; each period
is a phase change of 2 .
Surface The surface number to compute the phase display for.

Discussion:
This feature accounts for the size and shape of any aperture present on the surface; even if the aperture is
decentered. The phase is computed on a uniform grid of points in XY plane, and the phase value is the displayed
data. This feature defines phase in units of periods; so one period represents a phase change of 2 . Surfaces
which do not impart a phase change to the ray, such as the Standard surface, will display a phase of zero
everywhere on the surface phase display. See also the Surface Sag feature described above.

RMS
RMS vs. Field
Purpose:
Plots RMS radial, x, and y spot radius, RMS wavefront error, or Strehl ratio as a function of field angle.
Settings:
Item Description
Ray Density If the method is Gaussian quadrature, then the ray density specifies the number of radial
rays to be traced. The more rays traced, the greater the accuracy, although the
computation time increases. The maximum number is 20 which is sufficient for pupil
aberrations up to order 40. If the method is rectangular array, then the ray density
indicates the grid size. Rays outside the circular entrance pupil are ignored. See the
"Discussion" section for details.
angle specified at which the RMS/Strehl ratio is calculated, intermediate values are
interpolated. A maximum of 100 field points is allowed.
Plot Scale Sets the maximum vertical scale for the plot. Zero results in automatic scaling.
Method Selects either Gaussian quadrature or rectangular array. The Gaussian quadrature
method is very fast and accurate, but only works if there is no vignetting. If any rays are
to be vignetted, then rectangular array is more accurate.
Data Selects either wavefront error, spot radius, spot x-direction, spot y-direction, or Strehl
ratio. Strehl ratio is only available for monochromatic calculations, and is computed
using the method described in Strehl ratio approximation on page 167. For a
polychromatic Strehl, see the Huygens PSF on page 126.
Refer To Select either chief ray or centroid. For monochromatic calculations, the specified
wavelength is used for reference. For polychromatic calculations, the primary
wavelength is used for reference. Both reference points subtract out wavefront piston.
The centroid reference mode also subtracts out the tilt of the wavefront, which yields
smaller RMS values.
Orientation Select +y, -y, +x, or -x field direction. Note the data will only be computed to the limits of
the defined fields in the selected direction.
Wavelength Select "All" to display data for each wavelength and a polychromatic computation, select
any one wavelength to plot monochromatic data, or select "Poly Only" which only plots
the polychromatic data. Strehl ratio can only be computed for a single wavelength.
Show Diffraction If checked, then a horizontal line indicating the diffraction limited response will be drawn
Limit on the plot. For RMS radius, x, or y; the diffraction limit is assumed to be 1.22 times the
working F/# on axis times the wavelength (primary wavelength if polychromatic). The
change in the diffraction limit due to changes in F/# with field are ignored; a single value
is used across the range of the plot. For Strehl ratio, 0.8 is used, and for RMS wavefront,
0.072 waves is used. These are all approximate indicators for convenience only; the
actual meaning of "diffraction limited" may be open to interpretation.
Discussion:

This feature calculates the RMS error or Strehl ratio as a function of field angle for each wavelength.
Two different methods of calculation are used; either a Gaussian quadrature method, or a rectangular array
of rays. For the Gaussian quadrature method, the rays traced are arranged in a radial pattern with an optimal
weighting to estimate the RMS with a minimum number of rays. The method is described in G. W. Forbes, "Optical
system assessment for design: numerical ray tracing in the Gaussian pupil", J. Opt. Soc. Am. A, Vol. 5, No. 11,
p1943 (1988). Although the method is very efficient, the algorithm is not accurate if some of the rays are clipped
due to surface apertures. To compute the RMS wavefront in systems with surface apertures requires the use of
the rectangular array method, and a larger number of rays for sufficient accuracy.
Comments about RMS wavefront computations
For RMS wavefront computations, ZEMAX always subtracts out the mean OPD. RMS wavefront, when
referenced to the centroid, also requires computation of a shifted, titled reference sphere. The raw, uncorrected
OPD values do not account for the possible shift in image centroid location due to any coma that may be present.
The correction method is described in M. Rimmer, "Analysis of Perturbed Lens Systems", Applied Optics Vol. 9,
No. 3, p533 (1970). This same method is used when computing RMS wavefront referenced to the centroid by the
various RMS wavefront merit function optimization operands. The method is more accurate than simple direct
integration of the "raw" OPD values, which are referenced to a sphere centered on the chief ray - image surface
intercept point.
RMS vs. Wavelength
Purpose:
Plots RMS radial, x, and y spot radius, RMS wavefront error, or Strehl ratio as a function of wavelength.
Settings:
Item Description
Wave Density The wave density is the number of points between the minimum and maximum defined
wavelength at which the RMS/Strehl ratio is calculated, intermediate values are
interpolated. A maximum of 100 points is allowed.
Data Selects wavefront error, spot radius, spot x-direction, spot y-direction, or Strehl ratio.
Refer To Select either chief ray or centroid. Both reference points subtract out wavefront piston.
smaller RMS values.
Field Select "All" to display data for all fields, select one field to display data for a single field.

Item Description
Show Diffraction If checked, then a curve indicating the diffraction limited response will be drawn on the
Limit plot. For RMS radius, x, or y; the diffraction limit is assumed to be 1.22 times the working
F/# on axis times the wavelength. The change in the diffraction limit due to changes in
F/# with field are ignored. For Strehl ratio, 0.8 is used, and for RMS wavefront, 0.072
waves is used. These are all approximate indicators for convenience only; the actual
meaning of "diffraction limited" may be open to interpretation.
Discussion:
This feature calculates the RMS error or Strehl ratio as a function of wavelength for each field position. The
method of calculation is identical to that described in RMS vs. Field on page 132; see that section for a detailed
discussion. The wavelength range is determined by the minimum and maximum wavelengths defined on the
wavelengths dialog box.
RMS vs. Focus
Purpose:
Plots RMS radial, x, and y spot radius, RMS wavefront error, or Strehl ratio as a function of focus change.
Settings:
Item Description
Focus Density The focus density is the number of points between the minimum and maximum focus
shift specified at which the RMS/Strehl ratio is calculated, intermediate values are
interpolated. A maximum of 100 points is allowed.
Data Selects either RMS wavefront error, RMS spot radius, RMS X-direction, RMS Y-
direction, or Strehl ratio. Strehl ratio is only available for monochromatic calculations,
and is computed using the method described in Strehl ratio approximation on
page 167. For a polychromatic Strehl, see the Huygens PSF on page 126.
Refer To Select either chief ray or centroid. For monochromatic calculations, the specified
wavelength is used for reference. For polychromatic calculations, the primary
wavelength is used for reference. Both reference points subtract out wavefront piston.
smaller RMS values.
Wavelength Select "All" to display data for a polychromatic computation, select any one wavelength
to plot monochromatic data.
Min Focus The minimum value of the defocus to plot. The units are lens units.

Item Description
Max Focus The maximum value of the defocus to plot. The units are lens units.
Show Diffraction If checked, then a horizontal line indicating the diffraction limited response will be drawn
Limit on the plot. For RMS radius, x, or y; the diffraction limit is assumed to be 1.22 times the
working F/# times the wavelength (primary wavelength if polychromatic) on axis. The
change in the diffraction limit due to changes in F/# with field are ignored; a single value
is used across the range of the plot. For Strehl ratio, 0.8 is used, and for RMS wavefront,
0.072 waves is used. These are all approximate indicators for convenience only; the
actual meaning of "diffraction limited" may be open to interpretation.
Discussion:
This feature calculates the RMS error or Strehl ratio as a function of a change in focus position for each field
position. The method of calculation is identical to that described in RMS vs. Field on page 132; see that section
for a detailed discussion. ZEMAX adds the specified focus shift to the value of the thickness of the surface prior
to the image plane. If the optical system has an odd number of mirrors, this surface normally would have a
negative thickness, and therefore negative focus values move the image plane farther away from the last
component. For systems with an even number of mirrors, then negative focus values move the image plane closer
to the last component.
Encircled energy
Diffraction
Purpose:
Encircled energy diagram. This is the percentage of total energy enclosed as a function of distance from either
the chief ray or the image centroid at the image of a point object.
Settings:
Item Description
Type The analysis type option specifies how the encircled energy is calculated; either
encircled (radial), X-only, Y-only, or ensquared.
Maximum Distance This setting overrides the default scaling. The units are micrometers. To choose the
default scaling option, enter zero.
Refer To Select chief ray, centroid, or vertex as the reference point. Vertex refers to the
coordinates (0, 0) on the image surface. This option will only return meaningful data if
the diffraction image at all selected fields is within the maximum distance of the vertex.
When vertex is selected, ZEMAX is unable to detect if the sampling is sufficient, so some
care should be taken to set the sampling high enough for accurate results.

Item Description
Show Diffraction If checked, the diffraction limited results are computed and displayed. See the
Limit discussion below.
Use Huygens PSF If checked, the more accurate but slower Huygens PSF method is used to compute the
PSF. This option should always be used if the image surface is tilted, or if the chief ray
is not close to normal to the image surface.
Discussion:
!See the discussion sections of the FFT and Huygens PSF. Those comments also apply to this
feature.
The accuracy of the diffraction encircled energy calculation is limited by the magnitude and slope of the OPD
error and the sampling density used. If the sampling density is insufficient, ZEMAX will issue an error message
indicating that the data is inaccurate. To increase the accuracy, increase the sampling density or decrease the
OPD error. If shown, the diffraction limit curve is for the aberration free response at the reference field position
(see Diffraction Limited in the chapter Conventions and Definitions).
Geometric
Purpose:
Computes encircled energy using ray-image plane intercepts.
Settings:
Item Description
encircled (radial), X-only, Y-only, or ensquared.
Max Distance This setting overrides the default scaling. The units are micrometers. To choose the
Refer To Select either chief ray, centroid, vertex, or middle as the reference point. The middle is
the coordinate upon which the smallest circle enclosing all the ray intercepts is centered.
Multiply by If checked, ZEMAX approximates the diffraction encircled energy by scaling the
Diffraction Limit geometric data by the theoretical diffraction limit curve computed for a rotationally
symmetric Airy disk. The only way to compute an obscured or asymmetric pupil
diffraction limit function would be to perform an exact diffraction calculation, in which
case the diffraction encircled energy feature should be used instead. The diffraction limit
approximation is only useful for systems with unobscured pupils, reasonably rotationally
symmetric images, and modest field angles since the approximation ignores the change
in F/# with field.

Item Description
Discussion:
The X- and Y-only options will compute the fraction of rays which are contained with plus or minus the specified
distance from either the chief ray or the image centroid. If a scale of 10 micrometers is shown, then the region
enclosed is 20 micrometers across (and infinite in the other direction). The geometric encircled energy is not a
good indicator of performance if the system is close to diffraction limited.
Geometric Line/Edge Spread
Purpose:
Computes the geometric response to a line object and an edge object.
Settings:
Item Description
Max Radius The maximum radius setting overrides the default scaling. The units are micrometers.
To choose the default scaling option, enter zero.
Type The type option specifies which data is displayed on the graph; line and edge, line only,
or edge only.
Discussion:
The line response function (or line spread function, LSF) is the cross section of the intensity pattern of the
image of a line object. The edge spread function (ESF) is the cross section of the intensity pattern of the image
of an edge (a semi-infinite plane). The tangential and sagittal data refer to the orientation of the line or edge. The
zero coordinate is referenced to the chief ray. This is a geometric calculation. See also FFT Line/Edge Spread
on page 125.
Extended Source
Purpose:
Computes encircled energy using an extended source similar to the geometric image analysis feature.

Settings:
Item Description
Field Size This value defines the full width of the square image file in field coordinates, which may
be either lens units or degrees, depending upon the current field definition (heights or
angles, respectively).
Rays x 1000 This setting determines approximately how many rays will be traced. The number of rays
traced is approximately 1000 times the specified value. The reason the number of rays
is only approximate is because the distribution of rays over the pixels in the image must
be uniform. For example, if there are 1500 pixels in an image file, then at least 1500 rays
will be traced, even if a value of 1 is selected. The distribution of rays at each wavelength
is in proportion to the wavelength weights.
encircled (radial), X-only, Y-only, or ensquared. The X-only and Y-only options are
sometimes called "enslitted" and correspond to the total fraction of energy contained
within an expanding slit.
There are also options for X- or Y-distributions, which show the energy distribution in
either x or y directions. These latter two options also report the geometric full width at
half max. The X- or Y- distributions are the amount of energy falling on a pixel which is
narrow in one direction and infinite in the other direction.
Refer To Select either chief ray or centroid as the reference point.
Multiply by If checked, ZEMAX approximates the diffraction encircled energy by scaling the
Diffraction Limit geometric data by the theoretical diffraction limit curve. The diffraction limit curve
ZEMAX uses is based upon the unobscured circular pupil. The only way to compute the
obscured pupil diffraction limit function would be to perform the exact diffraction
calculation, in which case the diffraction encircled energy feature should be used
instead. The diffraction limit approximation is only useful for systems with unobscured
pupils and modest field angles, since the approximation ignores the change in F/# with
field.
Field The image file may be centered on any defined field position. This permits a small target
such as a bar chart to be moved to any location in the field of view.
File The name of the .IMA image file. This file must reside in the \ImaFiles directory. See the
discussion section in the Geometric Image Analysis feature for a full description of the
IMA file format.
Max Distance This setting overrides the default scaling. The units are micrometers. To choose the
Discussion:
The X- and Y-only options will compute the fraction of rays which are contained with plus or minus the specified
distance from either the chief ray or the image centroid. If a scale of 10 micrometers is shown, then the region

enclosed is 20 micrometers across (and infinite in the other direction). The geometric encircled energy is not a
good indicator of performance if the system is close to diffraction limited.
See the Geometric Image Analysis feature discussion for details about extended source modeling and the IMA
file format.
Illumination
Relative Illumination
Purpose:
Computes the relative illumination as a function of radial field coordinate for a uniform Lambertian scene. This
feature also computes the Effective F/#.
Settings:
Item Description
Ray Density The number of rays on one side of an array of rays used to integrate the illumination of
the exit pupil. A value of 10 will trace about 10 x 10 x pi / 4 or 78 rays. Higher ray densities
yield more accurate results at the expense of longer computation times.
Field Density The number of points along the radial field coordinate to compute the relative
illumination for. Larger field densities yield smoother curves.
Wavelength Selects the wavelength for computation. Relative illumination is a monochromatic entity.
Log Scale If checked, a logarithmic rather than linear scale will be shown.
Discussion:
This feature computes the relative illumination (RI) as a function of radial y field coordinate. RI is defined as
the intensity of illumination per unit area of image surface normalized to the illumination at the point in the field
that has maximum illumination (which may not be on axis). The computation considers apodization, vignetting,
apertures, aberrations of both the image and pupils, variations in F/#, chromatic aberrations, image surface
shape, angle of incidence, and optionally, polarization effects assuming unpolarized light. The method is based
upon one described in M. Rimmer, "Relative illumination calculations", Proc. SPIE Vol. 655, p99 (1986). The
published method was extended to include apodization, transmission, polarization, and non-planar image surface
effects. The computation method assumes the following are all true:
1. The object scene is plane, uniform, and Lambertian.
2. The image surface is a reasonably good conjugate (that is, an image) of the object surface, so that rays
coming from small patches of light on the object surface are imaged to patches of light on the image surface.
Aberrations are fine, but the rays should be reasonably localized on the image surface.
3. The exit pupil is not too close to the image surface. This condition will be satisfied if the F/# is larger than
about 0.1 and the ray aberrations are small compared to the exit pupil distance.
The relative illumination is computed by integration of the effective area of the exit pupil as seen from the image
point(s). The integration is carried out in direction cosine space using a uniform grid in image cosine space.
Note that the RI computation will not in general yield a cosine-fourth law curve, because the cosine-fourth law
is an approximation based upon a thin, aberration free lens with the stop at the lens illuminating a plane image
surface. For general systems including telecentric systems, systems with pupil or image aberrations, or vignetting,

the RI can be computed using an integration of the projected solid angle or effective area of the exit pupil as seen
from the image location.
For systems with very high amounts of vignetting, or for systems with non-linear cosine space aberrations that
would violate the assumptions of the computation, the relative illumination cannot be calculated, and an error
message will be displayed. Cosine space aberrations may be displayed using the spot diagram feature.
Effective F/#
The text listing of the relative illumination data also includes data for the Effective F/#. The Effective F/# is the
F/# required for a perfect optical system with 100% transmission and a circular exit pupil to have the same image
illumination as the system being evaluated. The Effective F/# is computed by:
- ,
E = ------
4A
where A is the area of the projected solid angle of the pupil in cosine space weighted for system transmission.
The Effective F/# is a useful metric for comparing the brightness of the image formed by different optical systems,
because it accounts for RI and is independent of the aperture shape.
Vignetting Plot
Purpose:
Calculates fractional vignetting as a function of field angle.
Settings:
Item Description
Ray Density The ray density specifies the number of rays to be traced. The more rays traced, the
greater the accuracy, although the computation time increases. For a ray density of n,
ZEMAX traces a grid of (2n+1) x (2n + 1) rays at each field point.
angle specified at which vignetting is calculated, intermediate values are interpolated.
Discussion:
Fractional vignetting is the percentage of rays incident upon the entrance pupil which pass all obscurations
and apertures in the system and survive to the image plane, normalized to relative pupil area. The graphic
generated by this function shows fractional vignetting as a function of field position. If too few rays are used, the
results may be inaccurate. This is especially true in systems with many apertures and large field angles.
Only the primary wavelength is used in this calculation. This is a geometric calculation. Only positive y field
positions are used, therefore this feature is only appropriate for rotationally symmetric lenses and fields. Rays
which cause errors such as missing a surface or those which are TIR are considered vignetted.
See also the relative illumination feature.
Illumination XY Scan
Purpose:
Computes relative illumination for an extended source along a line across any surface.

Settings:
Item Description
Sampling Selects the grid size used to sum the illumination. The sampling determines how many
pixels are used to collect the ray data.
Rays x 1000 Determines the approximate total number of rays to be traced in computing the
illumination due to the extended source.
File The name of the IMA file used to define the shape of the extended source. See the
Geometric Image Analysis feature for details.
Source Size The full width in field units of the extended source.
Rotation The angle to rotate the extended source in object space about the normal to the center
of the extended source.
Show As Select either X or Y direction scan.
Smoothing Smoothing helps remove jagged lines due to low ray sampling by applying a smoothing
operator to average the data in adjacent pixels.
Field The field number indicates which field position is used as the center reference point for
the extended source.
Surface The scan may be computed at any surface, however, the relative illumination
computation is only used at the image surface.
Detector Size The total width of the detector in lens units. The detector size is divided into pixels
according to the "Sampling" setting.
Use Relative If checked, then the RI computation described in the feature "Relative illumination" is
Illumination used to weight the rays from various points in the field of view to accurately account for
the effects of exit pupil radiance and solid angle. The computation is generally more
accurate, but slower, if this feature is used. RI may only be used if the selected surface
is the image surface, and ZEMAX assumes the image surface is a reasonable conjugate
of the object surface. See the RI feature for details.
If unchecked, the RI is estimated from the system distortion.
RI is only considered if the surface is the image surface.
Assume If checked, rays are weighted according to the cosine of the angle they make with
Lambertian Source respect to the object surface. This cosine factor is not appropriate in all cases. This
option is not available if "Use Relative Illumination" is checked, since the RI calculation
already assumes the object is Lambertian.
Discussion:
The illumination XY scan is similar to the relative illumination (RI) feature, with the added capability to estimate
the RI for non-uniform extended sources. For uniform extended Lambertian sources, the RI feature is faster and
more accurate. However, for systems with complex source properties, the illumination XY scan can estimate
illumination by Monte Carlo ray tracing combined with the conventional RI computation.
The extended sources are defined in the same way as described in the Image Analysis feature.

Illumination 2D Surface
Purpose:
Computes relative illumination for an extended source over a 2D surface.
Settings:
The options are identical to the XY scan, except that a 2D surface is drawn as either an isometric surface or
contour plot, or as a grey scale or false color map.
Discussion:
See the illumination XY scan feature.
Image Analysis
Geometric Image Analysis
Purpose:
The geometric image analysis feature has many applications. It can be used to model extended sources,
analyze useful resolution, represent the appearance of imaged objects, and provide intuition as to image rotation.
Image analysis is also useful for estimating multi-mode fiber coupling efficiency.
This feature is based strictly upon geometrical ray tracing; see the Diffraction Image Analysis for a diffraction
based feature with similar functionality.
The ability of this feature to render color images is limited, for analyzing color bitmap images see Geometric
Bitmap Image Analysis on page 146.
The image analysis feature uses special IMA or BIM files to describe the object to be imaged. The IMA and
BIM file formats are described in the discussion section which follows.
Settings:
Item Description
Field Size This value defines the full width of the square image file in field coordinates, which may
be either lens units or degrees, depending upon the current field definition (heights or
angles, respectively). Note that when using "field angle" as the field type the square IMA
file is inherently distorted; see the discussion below for details.
Image Size If "Show" is selected as a spot diagram, this value sets the size of the scale bar which
is superimposed on the image. It has no effect on the actual size of the image. The image
size is set by the object scale and the magnification and aberrations of the system. The
default may not be acceptable to see the desired portion of the image.
If "Show" is selected as any other option, this value sets the size of the detector used to
capture rays. Rays landing outside of the image size are ignored, and are not included
in the total detected rays, which will decrease the computed efficiency.
Parity The "Even" setting leaves the object as it would appear when viewed looking down the
negative Z axis in object space. The parity can be set to "Odd" which reverses the object
from top to bottom.
Rotation The rotation can be set to any angle in degrees. The algorithm actually rotates the object
before tracing the rays, so this feature can be used to switch from tangential to sagittal
orientation of bar targets, for example.
Rays x 1000 This setting determines approximately how many rays will be traced. The number of rays
traced is approximately 1000 times the specified value. The reason the number of rays
is only approximate is because the distribution of rays over the pixels in the image must
be uniform. For example, if there are 1500 pixels in an image file, then at least 1500 rays
will be traced, even if a value of 1 is selected. The distribution of rays at each wavelength
is in proportion to the wavelength weights.
Show Choose surface plot, contour map, grey scale, false color map, or spot diagram as the
display option.

Item Description
Source The source may be uniform or Lambertian. The uniform setting weights all rays equally.
Lambertian weights all rays by the cosine of the angle the ray makes with the axis of the
object surface.
Field The image file may be centered on any defined field position. This permits a small target
such as a bar chart to be moved to any location in the field of view. The resulting image
is then centered on the chief ray coordinate of this field position.
File The name of the .IMA or .BIM image file. This file must reside in the \ImaFiles directory.
See the discussion section for a full description of the IMA and BIM file formats.
Edit IMA File Pressing this button will invoke the Windows Notepad editor which allows modification
of the currently selected IMA file. This button is disabled if the file type is BIM.
Surface The surface number at which to evaluate the rays. The default is the image surface.
Other surfaces may be selected, for example, to visualize the beam footprint on an
optical surface.
# Pixels The number of pixels across the width of the selected image size. This value is not used
if "spot diagram" is the method of displaying the image data.
NA The numerical aperture (NA) cut-off. If zero, this feature is ignored. If a number greater
than 0 is entered, then all rays with a numerical aperture greater than the specified
number are ignored.
Total Watts The total power in watts radiated by the source into the entrance pupil of the optical
system. This flux is then used to normalize the detected power according to the relative
pixel values and the total efficiency.
This helps distinguish the various wavelengths. This value is only used if "spot diagram"
is the method of displaying the image data.
Reference Selects the reference coordinate for the center of the plot, either the chief ray or the
surface vertex.
Save As BIM File If a file name ending in the BIM extension is provided, and "Show" is not set to "Spot
Diagram", then the output image will be saved in the specified file and be placed in the
\ImaFiles subdirectory.
Discussion:
ZEMAX supports three different file formats. Two of these formats end in the IMA extension, one in the BIM
extension.

The IMA format
There are two different IMA file formats, one ASCII and one binary. Whichever file format is used, the file must
end in the extension IMA. ZEMAX will distinguish between the two types of file formats automatically.
The ASCII image file is a text file which ends in the extension .IMA. At the top of the file is one number which
indicates the size of the file in pixels. The remaining rows and columns contain the pixel data, with one character
to each pixel. All IMA files must be square, with n x n pixels defined. For example, a 7 x 7 representation of the
letter "F" could be described by the following IMA file:
7
0111110
0100000
0100000
0111100
0100000
0100000
0100000
Note that the single entry "7" starts the file, and it is followed by a carriage return. Then there are 7 rows of 7
columns each, and each row is terminated with a carriage return. The columns are not separated by a space or
any other character. The image file must be square. ZEMAX will attempt to allocate enough memory to hold the
image file and will report an error if there is not enough memory.
The "intensity" at each pixel can be any digit between 0 and 9. The number of rays each pixel will generate is
proportional to this value. Pixels with a value of 0 do not radiate any rays.
The binary IMA file format is more complicated than the ASCII format, and binary IMA files cannot be edited
with a text editor. However, the binary IMA files are dramatically more powerful. Each pixel in the binary IMA file
is represented by an unsigned byte, which means there are 256 "gray-scale" levels of intensity. Furthermore,
each wavelength can be assigned a separate pixel map. Therefore, very realistic photograph like extended
sources can be modeled.
The binary IMA file format requires 3 16-bit header values. The first 16-bit value is a signed integer that must
be equal to zero. The second 16-bit signed integer is the width of the pixel map in pixels, which can be any number
from 1 to 8000. The third 16-bit signed integer is the number of pixel maps, which correspond to the number of
colors (or wavelengths) represented in the file.
For example, a 3-color binary pixel map of a 50 by 50 image would have 6 bytes of header (0, 50, and 3),
followed by 2500 bytes for color 1, then 2500 bytes for color 2, then 2500 bytes for color 3, for a total of 7506
bytes. The data for each color is stored by columns for each row (the column index changes faster than the row
index).
The BIM format
The drawback to the IMA format is that a maximum of 256 grey scale levels are supported. The BIM format is
a binary double precision floating point file format which effectively makes the number of grey scales many
trillions. The BIM format consists of the following binary values:
1 32 bit integer representing the number of x pixels, nx.
1 32 bit integer representing the number of y pixels, ny.
followed by n*n 64 bit double precision floating point values representing the relative intensity. The first pixel
is the bottom left corner, and the remaining pixels are listed by rows along x.
Currently, the nx and ny values must be identical or an error message will be issued.
How rays are chosen for analysis
The rays generated by each pixel are chosen randomly from coordinates within the pixel cell. The entrance
pupil coordinates are also chosen randomly for each ray. The distribution of rays is uniform over the pixel and
over the circular paraxial entrance pupil (if ray aiming is used, then there may be some pupil distortion). For the
ASCII IMA files, the number of rays generated by each pixel is equal to the pixel intensity times the number of
wavelengths times the ray density. The wavelength used for each ray is selected randomly in proportion to the
wavelength weights provided on the wavelength data screen. For the binary IMA files, the number of rays
generated from each pixel is proportional to the ray density times the fractional intensity relative to 256.

The field size determines the physical size of the image file as seen by the optical system. For example, if a
30 x 30 pixel size image file is used, and the field size is 2.0 mm (this assumes the fields are defined in terms of
object or image heights), then each pixel represents a 66.67 micrometer region. If the same image file is later
used with a system with a 40 degree full field of view (using field angles), then the field size can be set to 40 to
cover the entire field. Each pixel will now represent a 1.33 degree square. The difficulty in using field angle for
defining the object field of view is that field angle units are inherently anamorphic. X-direction angles represent a
different subtended angle at a Y angle of 80 degrees than at an a Y angle of 10 degrees. If field angles are being
used, and the field of view is fairly wide (more than about 40 degrees in any direction) then great care should be
taken in interpreting the results for an extended object. For a precise definition of the field angles ZEMAX uses,
see Field angles and heights on page 47.
By separating the form of the object from the scale, the same image file can be used for many applications.
For example, the sample image file "letterf.ima" contains a 7x7 grid of pixels defining the capital letter F. The
object scale can be set to 1 mm, then 0.1 mm, then 0.01 mm to get a feel for how small a character F the optical
system can resolve, without the need to change the IMA file.
Note that if fields are defined by image height, then field size determines the size of the object in image space,
not object space. The field size is always in whatever units the fields are defined in, and so for image height the
field size determines image height. The size of the object is then determined by the field size divided by the
magnification of the lens.
The choice of field position also permits great flexibility in analyzing image quality. For example, the letter F
image file can be tested at several field points to see if the resolution is strongly affected by field aberrations. The
object scale is set to the height of the letter, but the image will be centered about the chief ray intercept of the
selected field point.
The source is by default a uniform radiator of rays. Uniform here means uniform in the entrance pupil. All rays
generated fall within the entrance pupil, and they are all weighted equally. Since ray wavelengths are selected
randomly in proportion to the wavelength weights, no explicit wavelength weighting is required. The uniform
setting is usually preferred for large object distance systems with small fields of view. The source may also be
defined to be Lambertian, which weights all rays by a cosine factor.
The percent efficiency is defined by
W
%E = ----------i
W j
where the sum i is over all rays which were unvignetted, and the sum over j is over all rays which were
launched. The efficiency calculation considers vignetting, source distribution, wavelength weights, and reflection
and transmission losses in the optical system if the "Use Polarization" checkbox is selected. If "spot diagram" is
selected for the "show" option, then the percent efficiency includes all unvignetted rays. The other display options
vignetted rays that are beyond the extent of the detector size. Therefore, the percent efficiency will be different
between spot diagram displays and other displays if rays fall outside the region of the detector.
Calculating efficiency of multi-mode fibers
ZEMAX has an algorithm for accurately computing fiber coupling into single-mode fibers; for details see Fiber
Coupling Efficiency on page 174. Also see Computing Fiber Coupling on page 530.
To estimate the coupling efficiency for multi-mode fibers, a geometric approach may be used. Place a circular
aperture at or just before the image surface with the appropriate maximum radial aperture representing the core
size. Then set the NA (see the table above) to the maximum acceptance NA of the fiber. The percent efficiency
will then be calculated by summing all the rays that pass the core aperture within the specified NA. The NA of a
typical multi-mode fiber with an inner core of index n i and an outer cladding of index n o is given by
2 2
NA = n i no .

Text output
Selecting the "Text" option on the image analysis window menu bar will generate and display an ASCII file
listing the ray data. If the "Show" option is set to "Spot Diagram", the file will have 9 columns. The first column is
the sequential ray number. The second and third columns are the x and y field coordinates (either degrees or
object height). The fourth and fifth columns are the normalized pupil coordinates, Px and Py. The sixth column is
the integer wavelength number. The seventh column is the weight of the ray, which depends upon the source
properties. The eighth and ninth columns are the image coordinates in lens units, relative to the reference ray.
If the "Show" option is not set to "Spot Diagram", then the text display will list the weighted ray count in each
pixel. The pixel listing starts at the -x, -y (bottom left) corner pixel, and proceeds up the y columns. Use the
"Escape" key to terminate a lengthy image analysis computation.
Geometric Bitmap Image Analysis
Purpose:
This feature creates an RGB color image based upon ray tracing data using an RGB bitmap file as the source.
This feature has many applications. It can be used to model extended sources, analyze useful resolution, display
distortion, represent the appearance of imaged objects, provide intuition as to image rotation, display beam
footprints, and show surface plots of illumination on any surface, to name just a few.
This feature is based strictly upon geometrical ray tracing. The bitmap image analysis feature uses standard
Windows BMP and JPG files as the source image, see the discussion for details.
Settings:
Item Description
Field Y Size This value defines the full y direction size of the source bitmap in field coordinates, which
may be either lens units or degrees, depending upon the current field definition (heights
or angles, respectively).
Parity The "Even" setting leaves the object as it would appear when viewed looking down the
negative Z axis in object space. The parity can be set to "Odd" which reverses the object
from top to bottom.
Rotation The rotation can be set to any angle in degrees. The algorithm actually rotates the object
before tracing the rays, so this feature can be used to switch from tangential to sagittal
orientation of bar targets, for example.
Rays x 1000 This setting determines how many rays will be traced. The distribution of rays at each
wavelength is in proportion to the relative RGB intensities of each pixel.
X-Pixels The number of pixels across the X direction on the detector.
Y-Pixels The number of pixels across the Y direction on the detector.
Grey Scale If checked, the RGB intensities will be averaged at each detector pixel to yield a grey
scale detected image. Rays will still be traced according to the relative RGB intensities
of the source bitmap, but all color information will be lost when the detected image is
displayed.

Item Description
Wavelength If RGB is selected, then 3 wavelengths will be defined, 0.656, 0.587. and 0.486
micrometers for red green, and blue, respectively; no matter what the current
wavelength definitions are. If 1+2+3 is selected, then wavelengths 1, 2, and 3 as
currently defined on the wavelength data box will be used. The red channel of the source
bitmap will be used for wavelength 3, green channel for wavelength 2, and the blue
channel for wavelength 1. The displayed image will be in RGB format no matter what
wavelengths are defined using this option. For selection of a specific wavelength, such
as 1, 2, 3, etc... the B, G, or R channel image will be used; for wavelengths higher than
3 the B channel is always used, for wavelengths selected which are not defined the
highest defined wavelength will be used.
Field The source image may be centered on any defined field position. This permits a small
target such as a bar chart to be moved to any location in the field of view. The resulting
image is then centered on the chief ray coordinate of this field position.
Input The name of the .BMP or .JPG source image file. This file must reside in the \ImaFiles
directory.
Surface The surface number at which to evaluate the rays.
X-Pixel Size The size in lens units of each detector pixel measured in the X direction.
Y-Pixel Size The size in lens units of each detector pixel measured in the Y direction.
Show Select either "object or image". If object is selected, then the source bitmap will be
drawn. If image is selected then rays will be traced and the detected image will be
displayed. Note the number of rays, pixels, and pixel sizes are ignored when drawing
the object bitmap.
Output The name of the .BMP or .JPG to write the detected bitmap to. The detected bitmap size
is determined by the number of x and y pixels defined; but the pixels size must be the
same in x and y for the aspect ratio to be correct in the output bitmap file. The file name
must end in either a BMP or JPG extension, with no path name supplied. This file will
be created or overwritten without warning and will be placed in the \ImaFiles directory.
Discussion:
See also the Geometric Image Analysis feature discussion, as this feature is very similar.
The BMP files used as source bitmaps must be standard Windows format, uncompressed, 24-bit color RGB
bitmaps. Not all files ending in the .BMP extension meet this definition. Contact technical support if you have a
BMP file that does not work correctly for advice on converting the file to the proper format. Any properly formatted
JPG file will work.
The rays generated by each pixel are chosen randomly from coordinates within the source pixel cell. The
entrance pupil coordinates are also chosen randomly for each ray. The distribution of rays is uniform over the
pixel, the ray distribution over the entrance pupil is uniform unless pupil apodization is defined.
The Field Y Height determines the physical size of the source file as seen by the optical system. For example,
if a 50H x 100W pixel source file is used, and the Field Y Height is 2.0 mm (this assumes the fields are defined
in terms of object or image heights), then each pixel represents a 0.040mm x 0.040mm region, and the source
bitmap covers a 2.0mm high x 4.0 mm wide area.
Note that if fields are defined by image height, then Field Y Height determines the size of the object in image
space, not object space. The Field Y Height is always in whatever units the fields are defined in, and so when
using image height as a field type the Field Y Height determines the source bitmap height in image space. The
size of the source in object space is then determined by the Field Y Height divided by the magnification of the lens.

The pixels from which rays are traced are chosen randomly according to the RGB intensity of each pixel in
each color channel. Rays are generated randomly from anywhere within the bitmap until the specified number of
rays have been launched, although not all these rays may make it through the lens.
Once a ray is generated, it is traced through to the selected surface. If the ray is vignetted or an error occurs,
the ray is ignored. Otherwise, the pixel on the receiving detector that the ray struck is determined, and the intensity
of the ray is added to the detectors bin count in the appropriate color channel.
After all the rays are traced, an RGB image is created from the normalized counts in each detector pixel.
The percent efficiency is defined by
W i
%E = ----------
W j
where the sum i is over all rays which were unvignetted, and the sum over j is over all rays which were
launched. The efficiency calculation also includes consideration of reflection and transmission losses in the
optical system if the "Use Polarization" checkbox is selected, as well as ray errors. Rays that are beyond the
extent of the detector are considered to be vignetted.
Selecting the "Text" option on the menu bar will generate and display an ASCII file listing the detected pixel
data.
Use the "Escape" key to terminate a lengthy image analysis computation.
Diffraction Image Analysis
Purpose:
The diffraction image analysis feature is similar to the geometric image analysis feature, except the complex
system Optical Transfer Function (OTF) is used to compute the image appearance. This method accounts for the
finite pass band and other diffraction related effects real optical systems have on image formation.
The image analysis feature uses IMA/BIM files to describe the object to be imaged. See The IMA format on
page 144 and The BIM format on page 144.
Settings:
Item Description
File Size The full width in lens units of the region defined by the IMA file measured in image space.
Note IMA files are always square.
Oversampling Sets the factor by which the IMA file pixels are oversampled. This increases the effective
resolution of the IMA file without the need to define a new IMA file. If the original IMA file
has an odd number of pixels, oversampling will make the number of pixels even because
the oversampling values are all even.
Zero Padding Determines the actual size of the region to compute the diffraction image of by adding
zero intensity values around the IMA file pixels. This increases the size of the displayed
diffraction image without changing the size of the unaberrated image; this allows study
of the energy diffracted well away from the perfect image location.
To keep the IMA file data centered, zero padding will yield an odd number of pixels if the
original IMA file has an odd number of pixels. The number of pixels will be 2*n-1, where
n is the original number of pixels. If the original IMA file has an odd number of pixels,
the modified image will have 2*n pixels.
OTF Sampling The grid size of the sampling in the pupil; larger grids yield a more accurate
representation of the system OTF. This has no affect on the size of the diffracted image,
just the accuracy of the predicted frequency response.

Item Description
Data Type Choose either the incoherent image, coherent image, raw image, incoherent transfer
function, coherent transfer function, or the transform of the raw image. See the
discussion below for information about the coherent transfer function and the limitations
of this computation.
Diffraction Limited If checked, aberrations are ignored. Apertures are still considered.
Field The field number indicates for which field position the optical transfer function is
computed.
File The name of the .IMA image file. This file must reside in the \ImaFiles directory. See the
discussion section for a full description of the IMA file format.
of the currently selected IMA file.
Contour Format String on page 131. The contours are defined in units of relative
intensity.
Discussion:
This feature can compute complex diffraction image properties from extended sources. The method involved
in the computation is based upon Fourier Optics, which is described in clear and insightful detail in "Introduction
to Fourier Optics" by Joseph Goodman, McGraw-Hill 1968. A more detailed treatment of coherent imaging theory
may be found in "Linear Systems, Fourier Transforms, and Optics" by Jack Gaskill, John Wiley 1978. See those
references for more information on coherent vs. incoherent imaging and other Fourier optics theory. There are
several important assumptions in the method that must be understood by the user before this feature may be
used to draw important conclusions.
The IMA file defines the relative intensity at each of an arbitrary number of pixels arranged on a square grid.
For example, the letter "F" may be described by a 7 x 7 grid such as:
0111110
0100000
0100000
0111100
0100000
0100000
0100000
The file size parameter determines how big each pixel is in the image space of the optical system. Note this
is different from the geometric image analysis feature, where the IMA file defines the size and shape in "field"
space, which may be either object or image space. For the diffraction image analysis, the IMA file defines the
ideal image shape in image space. If the file size is 0.1 mm, then each pixel is 14.286 micrometers wide for this
7 x 7 pixel image.
Although the IMA file is convenient for defining simple shapes, the resolution of files created by hand is
generally too low to see sufficient detail in the diffracted image. The oversampling option remedies this problem
by increasing the resolution; the number of pixels is increased, and the data from each pixel is replicated as
required to yield the same shape at higher resolution. With an oversampling of 2X; the letter "F" file IMA image
becomes a 14 x 14 grid:

00111111111100
00111111111100
00110000000000
00110000000000
00110000000000
00110000000000
00111111110000
00111111110000
00110000000000
00110000000000
00110000000000
00110000000000
00110000000000
00110000000000
Note the shape is the same, there is just twice the sampling in each direction. The width of the image is
unaffected; the pixels are half the width and there are twice as many in each direction now.
Because the effects of diffraction tend to blur and extend the ideal image, it is desirable to increase the size
of the displayed area beyond the limits of the IMA file. This can be done with the zero padding option. This option
increases the IMA file size by adding zero intensity values around the defined IMA file pixels. The letter "F" file
would look like this with a zero padding of 2X:
0000000000000
0000000000000
0000000000000
0000111110000
0000100000000
0000100000000
0000111100000
0000100000000
0000100000000
0000100000000
0000000000000
0000000000000
0000000000000
The size and shape of the image is the same, but additional area has been defined that some energy may
now diffract into. Note the width of the new image has increased by the zero padding factor, but the width of the
"F" part of the image is the same. If the original file has an odd number of pixels (as this sample letter F does)
then the number of pixels after zero padding will remain odd to preserve the notion of a "center" pixel. The number
of total pixels is thus 2*n-1 if n is odd, or 2*n if n is even, where n is the number of original pixels. Zero padding
is not available when using the binary IMA file format.
The oversampling and zero padding may be used together, however the array sizes quickly become quite
large. For an oversampling of 8X, a zero padding of 4X, and an original pixel size of 12 x 12, the resulting array
becomes 384 x 384, which ZEMAX internally further zero pads up to 512 x 512 for transform purposes.
Once the input image is defined, the image is transformed into frequency space, multiplied by the OTF, and
transformed back into position space. The resulting image is filtered by the complex OTF to yield the diffraction
image. The primary assumption made by this implementation of the Fourier method is that the OTF does not
change over the extent of the image region (to avoid this assumption see Extended Diffraction Image Analysis
on page 151). This means the field of view defined by the size of the image is small enough so that the OTF is
the same over all points in the image. The user must take care to be sure the image region defined is small with
respect to the rate of change of the field aberrations. ZEMAX computes the OTF for the field point selected, and
assumes this OTF is valid over the entire region covered by the image.
Because of this assumption, distortion will not be visible in the predicted image, since only the variation in OTF
over the field will introduce distortion. To see the effects of distortion or other "large field" effects, see Geometric
Image Analysis on page 142 or Extended Diffraction Image Analysis on page 151.
Note that the diffraction image analysis feature is good at computing detailed image data for small images,
while the geometric image analysis feature is good at computing image data for larger scale images.

Comment about the coherent optical transfer function
The other assumption this calculation makes is in the method used to compute the coherent optical transfer
function. The coherent optical transfer function is assumed to be the complex pupil function:
iKW ( d i f x, d i f y )
H ( f x, f y ) = P ( d i f x, d i f y )e
where H is the complex OTF, d i is the pupil to image distance, f x and f y are the spatial frequencies, P is the
pupil function (which determines the relative transmission over the pupil, and is zero outside the pupil), and W is
the wavefront aberration function. This approximation is valid for a wide range of optical systems. For more
information on the development of coherent imaging theory, see the references by Gaskill or Goodman.
This feature also accounts for any vignetting at the image surface due to surface apertures (apertures at other
surfaces are accounted for via the effects of those apertures on the optical transfer function). After the diffraction
image is computed, any energy that lies outside of the aperture defined on the image surface is eliminated. The
fraction of energy within the unvignetted portion of the image surface aperture is reported on the text version of
the analysis, if the image surface has a defined aperture.
Extended Diffraction Image Analysis
Purpose:
The extended diffraction image analysis feature is similar to the diffraction image analysis feature, except the
Optical Transfer Function (OTF) may vary over the field of view of the image. This feature uses IMA/BIM files to
describe the object to be imaged. See The IMA format on page 144 and The BIM format on page 144.
Settings:
Item Description
File Size The full width in lens units of the region defined by the IMA file. Note IMA files are always
square. This control does not set the size of the resulting image; that is defined by the
Display Size and the OTF sampling (see the discussion below).
Oversampling Sets the factor by which the IMA file pixels are oversampled. This increases the effective
resolution of the IMA file without the need to define a new IMA file. If the original IMA file
has an odd number of pixels, oversampling will make the number of pixels even because
the oversampling values are all even.
Display Size If smaller than the maximum image size available (see the discussion below) then this
parameter will define the width of the displayed image in lens units.
OTF Sampling The grid size of the sampling in the pupil; larger grids yield a more accurate
representation of the system OTF. Larger grids will also yield a larger maximum display
size (see the discussion below).
OTF Grid The grid size of the OTF computation. A denser OTF grid yields a more accurate
computation of the variation of OTF over the field of view of the image at the expense
of more memory usage and slower computation.
Resolution This multiplier yields more points in the final image while keeping the display size fixed,
at the expense of more memory usage.
Data Type Choose either the incoherent image or coherent image. See Comment about the
coherent optical transfer function on page 151 for information about the coherent
transfer function and the limitations of this computation.
Diffraction Limited If checked, aberrations are ignored. Apertures are still considered.

Item Description
Use Delta If checked, each pixel in the IMA file will be assumed to represent a delta function. This
Functions is useful for checking the imaging of point sources like stars. If unchecked, the whole
pixel is assumed to be a luminous square area.
File The name of the image file. This file must reside in the \ImaFiles directory. See The IMA
format on page 144 and The BIM format on page 144 for a description of file formats.
of the currently selected IMA file. BIM files cannot be edited this way.
Field The field number indicates for which field position the image will be centered on.
Contour Format String on page 131. The contours are defined in units of relative
intensity.
Consider Distortion If checked, the real vs. paraxial ray distortion will be considered in forming the image
appearance.
Output File If a name is provided, the complex amplitude of the resulting image will be saved in a
ZBF file with the specified name. The file will be saved in the \POP\BeamFiles
subdirectory. See ZEMAX Beam File (ZBF) binary format on page 528.
Discussion:
This feature can compute complex diffraction image properties from extended sources while accounting for
the variation in the optical transfer function (OTF) over the field of view. Most of the discussion in the Diffraction
Image Analysis feature applies, see Diffraction Image Analysis on page 148 for details. The differences between
the Diffraction Image Analysis feature and the Extended Diffraction Image Analysis feature are described below.
Diffraction image formation can be thought of as a filtering or as a convolution process. Suppose the ideal,
unaberrated, undiffracted image is described by a function "A" which describes image amplitude as a function of
spatial coordinates in the image space of an optical system. Convolving this function with the system PSF (see
FFT PSF on page 122) here denoted by "P" yields the final image "I":
I ( x, y ) = A ( x, y ) P ( x, y ) ,
where the notation A P denotes the convolution of A and P. Taking the Fourier transform of this equation
yields the spatial filtering perspective of the image forming process:
i ( f x, f y ) = a ( fx, f y ) o ( f x, f y ) ,
where i, a, and o are the transforms of I, A, and P into the spatial frequency domain. The function o is called
the optical transfer function (OTF); which acts as a filter scaling the amplitude and phase of the spatial frequency
components of the image.
The Extended Diffraction Image Analysis eliminates one major assumption of the Diffraction Image Analysis
feature: that the OTF is constant over the field of view represented by the function A. This is accomplished by
considering the source IMA file one pixel at a time, and computing the Fourier transform of the one pixel. The one
pixel transform is multiplied by the OTF corresponding to that pixel. The sum over all pixels is then computed in
spatial frequency space, and finally the sum of the filtered pixel transforms is Fourier transformed back to form
the final image.

Computing the OTF for every pixel in the IMA file is slow and not practically required, as the OTF over any
reasonably small field of view does not change rapidly. As an alternative, this feature computes a grid of OTF
which spans over the field of view, and then interpolation is used to compute the effective OTF at any single pixel.
The greater the size of the OTF grid, the more accurate the results will be, at the expense of longer computation
times and more memory usage.
This feature can require large amounts of RAM. For an OTF grid size of n x n, ZEMAX will actually store n*n+1
OTFs: the n*n OTFs on the grid plus one to use as a running sum of the image frequency components. The OTF
itself is a complex value array of double precision points. ZEMAX automatically zero pads the OTF grid to achieve
accurate sampling. If the OTF sampling is 64 x 64, a single OTF will be 128 x 128, using up 256k of RAM (128 x
128 x 2 x 8 bytes). For an OTF grid of 5 x 5, the total comes to 26*256k = 6.5 Mb. For 128 x 128 sampling and a
9 x 9 grid, the RAM required is 82 Mb.
The file size parameter determines how big each pixel in the IMA file is in the image space of the optical system.
Note this is different from the geometric image analysis feature, where the IMA file defines the size and shape in
"field" space, which may be either object or image space. For the diffraction image analysis, the IMA file defines
the ideal image shape in image space. If the file size is 0.1 mm, then each pixel is 14.286 micrometers wide for
this 7 x 7 pixel image.
Because the integral over all pixels is done in frequency space, the maximum display size is determined by
the smallest increment of spatial frequency in the grid of OTFs. Since the OTF varies over the field of view,
ZEMAX computes all the OTFs, then scales and interpolates all of them to a common scale. The maximum
display size can be increased by increasing the OTF sampling. The maximum display size is approximately given
by Fn , where is the wavelength, F is the smallest F number in the field of view, and n is the number of points
in the OTF sampling. For any given optical system, the maximum display size increases only with OTF sampling.
IMA/BIM File Viewer
Purpose:
This feature displays IMA/BIM files without processing. See Geometric Image Analysis on page 142 for a
discussion of IMA/BIM files.
Settings:
Item Description
File The name of the .IMA or .BIM image file. This file must reside in the \ImaFiles directory.
See the discussion section for a full description of the IMA and BIM file formats.
Show Choose surface plot, contour map, grey scale, or false color map as the display option.
Color If the selected file is a binary IMA, this control will select the color channel to display.
of the currently selected IMA file. This button is disabled if the file type is BIM.
Discussion:
See Geometric Image Analysis on page 142 for a discussion of IMA/BIM files.
Biocular Analysis
Field of View
Purpose:
Displays the field of view for up to four configurations. See the discussion for important assumptions before
using this feature.

Settings:
Item Description
Config 1, 2, 3, 4 Selects which configurations to display the field of view data for.
X Points, Y Points Defines the number of points to use in each direction.
Left/Right X The left and right limits in field of view units.
Top/Bottom Y The upper and lower limits in field of view units.
Use Angles If checked, the field of view units are degrees, otherwise, the units are direction cosines.
Discussion:
This feature is intended to show the field of view for up to four configurations. Field of view in this context
means the angle of rays emanating from the stop surface, not the object surface, which are unvignetted all the
way to the image surface. The primary usage is the analysis of biocular systems, where both eyes look through
the same optical path to see a projected image. This feature makes all of the following assumptions:
-The field of view is either angles in degrees or in direction cosines directly. The angles/cosines are
measured along the chief ray from the stop position relative to the local Z axis at the stop surface. Surfaces
prior to the stop are ignored.
-The image surface is assumed to be the location of the image the eyeball is looking at. Each selected
configuration should represented a single (usually decentered) eyeball position.
-The eyeball decenter should be set up so that the X, Y coordinates on the image surface represent the
same point on the image source in all configurations. For example, if the image source is a CRT, then the
point with image coordinates (x = 1, y = 2) should correspond to the same physical location on the CRT in
all configurations.
-All surfaces should have fixed apertures for the purposes of vignetting rays that are outside of the field of
view.
The field of view is computed by tracing chief rays at various angles from each configuration. If the ray passes
all apertures, then the point on the plot corresponding to that angle will be inside of the closed curve for that
configuration. Each configuration will generate a closed curve representing the limits of the field of view for that
configuration. The ray tracing is done iteratively to determine the exact angles at which the chief rays become
vignetted.
If the field of view units are cosines, then the plot is linear in cosine space, with the Z direction cosine
determined from the X and Y direction cosines. If the field of view units are angles, then the direction cosines of
the chief rays are determined from the X and Y field angles using these formulas:
l m 2 2 2
tan ( x ) = --- , tan ( y ) = ---- , l + m + n = 1 ,
n n
where l, m, and n are the X, Y, and Z direction cosines, respectively.

Dipvergence/Convergence
Purpose:
Displays the dipvergence and convergence for biocular analysis.

Settings:
Item Description
Left Eye Config Defines the configuration to use as the left reference eye.
Right Eye Config Defines the configuration to use as the right eye, whose dipvergence and convergence
is determined relative to the left eye.
Do X Scan If checked, the plot scan is across the X direction field of view, otherwise, the scan is
across the Y direction field of view.
Use Angles If checked, the field of view units are degrees, otherwise, the units are direction cosines.
Min/Max Angle/ Sets the limits of the scan in X or Y, in either cosine space or in degrees. Note the limits
Cosine do not need to be symmetric, and the min value need not be less than the max value.
X/Y Angle/Cosine If "Do X Scan" is checked, then up to 6 Y angles/cosines offsets may be selected for
1-6 display. If "Do X Scan" is unchecked, then up to 6 X angles/cosines offsets may be
selected for display. To turn off one or more of the offset values, set the value to -99.
# Points The number of points to use across the full scan.
Convergence This value is subtracted from the computed convergence value. This allows only the
Offset deviation from the desired value to be displayed.
Discussion:
The assumptions made for this feature are the same as for Field of View on page 153.
When two eyes are used to look through a biocular lens, there is usually a small angular difference between
the direction the two eyes must look to see the same image point. The vertical (up/down) angle is called
dipvergence. The horizontal left-right angle is called convergence if the eyes gaze toward a common point in front
of the viewers head, so the chief rays from the two eye positions would converge as they move toward the lens
and away from the viewers head. If the two chief rays diverge as the rays propagate toward the lens, as if to see
a virtual image point behind the head, then the angle is called divergence. Convergence and divergence are really
the same thing from a computational standpoint. ZEMAX uses the common convention that convergence is a
positive value and divergence is a negative value; for this reason only the term convergence will be used in the
subsequent discussion, with the understanding that if the convergence is negative then it is properly called
divergence. Usually, convergence is more tolerable than divergence and the two aberrations may have different
specifications. Dipvergence and convergence are both measured in milliradians, and typical limits are on the order
of 1.0 milliradians for visual systems.
The computation proceeds for a given point in the field of view by tracing a reference ray in the left eye
configuration. The same angle chief ray is then traced in the right eye configuration. In general, the right eye ray
will not land on the exact same X, Y coordinates on the image surface as the left eye reference ray did. ZEMAX
iterates the right eye trace until the chief ray is found that matches the left eye reference ray intercept coordinates.
The resulting right eye chief ray will in general make some angle with respect to the left eye in both the vertical
and horizontal directions, and it is these angles that are the dipvergence and convergence, respectively.
Note that both the left eye and the right eye chief rays must pass all surfaces without errors and without
vignetting for the computation to be valid. If both rays do not trace, no data will be returned for that field of view.
The field of view overlap between the configurations is very useful for setting appropriate min/max scan values,
see the Field of View on page 153 for determining the field of view overlap.
It is possible for the iteration required for the right eye to fail. This typically happens if either no solution exists
or if the dipvergence/convergence is so large that algorithm becomes unstable. Failed iteration will usually show
up as gaps in the plot.

Miscellaneous
Field Curvature/Distortion
Purpose:
Displays the field curvature and distortion plots.
Settings:
Item Description
Max Curvature Defines the maximum scale for the field curvature plot in lens units. Enter zero for
automatic.
Max Distortion Defines the maximum scale for the distortion plot in percent. Enter zero for automatic.
Ignore Vignetting See the discussion section.
Factors
Distortion Select Standard, F-Theta, or Calibrated. See the discussion section for details.
Discussion:
The field curvature plot shows the distance from the currently defined focal (image) plane to the paraxial focal
plane as a function of field coordinate. The tangential data are the distances measured along the Z-axis from the
currently defined focal plane to the paraxial focal plane measured in the tangential (YZ) plane. The sagittal data
are the distances measured in the plane orthogonal to the tangential plane. The base of the plot is on axis, and
the top of the plot represents the maximum field (angle or height). There are no units on the vertical scale because
the plot is always normalized to the maximum radial field.
The field curvature for the tangential and sagittal rays is defined as the distance from the defined image plane
to the paraxial focus for that ray. In non-rotational systems, the real ray and the chief ray may never intersect, and
so the value presented is at the point of closest approach.
The field scan is along the +Y field by default. If +X or -X is selected, then the field scan is along the X direction,
in which case the tangential curve is for the XZ plane, and the sagittal curve is in the YZ plane.
Users often ask why the field curvature plot does not always start at zero for zero field. The reason is that the
plot shows the distance from the currently defined image plane to the paraxial focal plane, and the currently
defined image plane need not be coincident with the paraxial focal plane. If there is any defocus, then the two
planes are offset, and so is the field curvature data.
"Standard" distortion in percent is defined as the real chief ray height, minus the reference ray height, divided
by the reference ray height, times 100:
y chief y ref
Distortion = 100 --------------------------- ,
y ref
where all heights are taken to be the image surface radial coordinate, at whatever image surface is defined
(the data is not referred to the paraxial image plane). The reference ray height is computed by tracing a real ray
from a very small field height, and then scaling the results as required. This generalization permits the
computation of reasonable distortion values even for systems not well described by paraxial ray tracing. The
reference height for the undistorted ray in a rotationally symmetric system at paraxial focus is given by
y ref = f tan ,

where f is the focal length and is the angle in object space. ZEMAX actually uses the generalization
described above, rather than this equation, however the idea is the undistorted height goes as the tangent of the
field angle.
"F-Theta" distortion does not use the tangent relationship, but instead uses the height given by the focal length
of the lens multiplied by the angle the chief ray makes in object space. This so called "F-Theta" height is only
meaningful in systems with the object at infinity, when field heights are measured in angles. The reference height
for the undistorted ray at the paraxial focal plane is given by
y ref = f ,
where f is the focal length and is the angle in object space. F-Theta is used in scanning systems where the
image height needs to be linear with scan angle.
"Calibrated" distortion is similar to "F-Theta" distortion, except that the "best-fit" focal length is used rather than
the system focal length. Calibrated distortion measures the deviation from linearity between the image height and
the field angle, without the restriction that the proportion of linearity be defined by the focal length of the system.
A focal length is chosen that best fits the data, rather than the system focal length, although in general the best
fit focal length is close to the system focal length. The calibrated focal length used is given on the text listing for
this feature. The reference height for the undistorted ray is given by
y ref = f' ,
where f is the calibrated focal length and is the angle in object space. One somewhat surprising effect of
this definition of calibrated distortion is the non-zero distortion at zero field angle. The reason for this non-zero
distortion is best explained by looking at the limiting behavior of the definition of calibrated distortion. Calibrated
distortion in percent is defined as
y chief y ref
Distortion = 100 --------------------------- .
y ref
For small angles, the real y chief ray coordinate in any reasonable optical system is well described by
y chief = f , and the reference ray coordinate by y ref = f' , so the distortion near zero field is
y chief y ref f f' f f'

Distortion = 100 --------------------------- = 100 ----------------- = 100 ---------- ,
y ref f' f'
which is not zero, unless f = f' . Therefore, the calibrated distortion is not generally zero on axis. This does
not mean the image height is not zero, it is an artifact of the different choice of focal lengths for the reference and
actual ray coordinates near the optical axis. Note that a percentage distortion near the axis is not significant
because the field itself approaches zero at zero field angle.
For non-rotationally symmetric systems, distortion is poorly defined and the data presented is probably
meaningless. The reason is that no single number adequately describes distortion at a single field point if the
system is not rotationally symmetric. Instead, see Grid Distortion on page 158.
Strictly speaking, the field curvature and distortion plots are only valid for rotationally symmetric systems with
plane object and image surfaces. However, ZEMAX uses a generalization of the field curvature and distortion
concepts to give reasonable results for some, but not all, non-rotationally symmetric systems. Caution should be
used when interpreting these data for non-rotationally symmetric systems or systems with object and/or image
surfaces that are not planes.

By default, ZEMAX ignores vignetting factors when computing the field curvature and distortion plots.
Vignetting factors can change the chief ray location on the stop surface, such that the chief ray no longer goes
through the center of the stop.
ZEMAX cannot compute the distortion in the manner described here if the field type is real image height. The
reason is that when using real image height, ZEMAX iterates each ray trace to find the exact field coordinate to
hit the desired image coordinate; implicitly removing the distortion. As a workaround, ZEMAX automatically
changes the field type from real to paraxial image height for the purposes of this computation, and this may affect
the accuracy of the results for some systems. Generally the results will be accurate if the distortion is small.
Grid Distortion
Purpose:
Displays a grid of chief ray intercept points to indicate distortion.
Settings:
Item Description
Display Select either "Cross" to mark each chief ray intercept with a cross, or "Vector" to
plot a vector from the ideal image point to the actual chief ray image point.
Grid Size The size of the grid.
Ref. Field The reference field position. See the "Discussion".
Scale If the scale is different than 1.0, then the "x" points on the distortion grid will be
exaggerated by the selected scale factor. The scale factor may be negative to
change positive to negative distortion on the plot.
H/W Aspect If unity, then a square field will be selected. The output image may not be square
if the system is not symmetric, but the object field will be square. If the H/W aspect
is greater than 1, then the "height" or "y" field will be expanded by the aspect ratio.
If the aspect is less than 1, then the "height" or "y" field will be compressed by the
aspect ratio. The aspect ratio is the "y" field height divided by the "x" field width.
The aspect ratio only affects the input field; the image aspect ratio is determined
by the optical system imaging properties.
Symmetric Magnification If checked, then the X magnification is required to be identical to the Y
magnification. This causes distortion to be referenced to a symmetric predicted
grid rather than an anamorphic predicted grid.
Field Width The full width, in field units, of the "x" field of view.
Discussion:
This feature displays or computes the coordinates of a grid of chief rays. In a system without distortion, the
chief ray coordinates on the image plane follow a linear relationship with the field coordinate:
xp f
= A B x ,
yp C D fy
where xp and yp are the predicted image coordinates relative to a reference image point and fx and fy are the
linear coordinates on the object surface relative to a reference point. For optical systems using angles as a field
definition, fx and fy are the tangents of the field angles (the field coordinates must be linear, therefore, tangents
of angles rather than angles are used). To compute the ABCD matrix, ZEMAX traces rays over a very small region
centered upon the reference field position. Usually, this is the center of the field of view. ZEMAX allows selection
of which field position to use for reference.

By default, ZEMAX sets the corner of the field grid in object space to be at the maximum radial field distance.
Because the object height is linear with field angle and not angle directly, the full width of the field when angles
are used to define the field is given by
1
wide = 2 tan ------2- tan r ,
2
where r is the maximum radial field angle at the corner of the field.
The ray coordinates in image space for the very small field of view are used to determine the ABCD matrix
components. The use of an ABCD matrix allows for coordinate rotations. If the image plane is rotated, such that
a y object coordinate images to both an x and a y image coordinate, the ABCD matrix will automatically account
for the rotation. The grid distortion plot shows the linear grid, and then marks the actual chief ray intercept for a
ray with the same linear field coordinates with an "X" for each point on the grid.
If the optical system is not rotationally symmetric, then the distortion is not generally radial. The distortion is a
vector, and the magnitude of the distorted vector must be used to compute the total distortion.
The text listing available tabulates the predicted image coordinate, the actual image coordinate, and the
"percent distortion" defined by
R distorted
P = 100% -------------------- , where
R predicted
2 2
R real = ( xr ) + ( yr ) ,
2 2
R predicted = ( x p ) + ( y p ) , and
2 2
R distorted = ( xp xr ) + ( yp yr ) ,
and the subscripts r and p refer to the real and predicted coordinates on the image plane relative to the
reference field position image location, respectively. Because the values for R real and R predicted are always
positive, this definition will always yield a positive value for P . However, it is still a frequently useful concept to
distinguish between "positive" and "negative" distortion. To support this, if R real is less than R predicted , then the
sign of P is changed to be negative.
!This definition of generalized distortion may not be applicable in all cases, and the results
should be used with caution.
ZEMAX cannot compute the distortion in the manner described here if the field type is real image height. The
reason is that when using real image height, ZEMAX iterates each ray trace to find the exact field coordinate to
hit the desired image coordinate; implicitly removing the distortion. As a workaround, ZEMAX automatically
changes the field type from real to paraxial image height for the purposes of this computation, and this may affect
the accuracy of the results for some systems. Generally the results will be accurate if the distortion is small.
Grid distortion cannot be calculated if the field units are angles and the maximum angle equals or exceeds 90
degrees. This limitation is due to the assumption that the predicted image height is proportional to the tangent of
the field angle in object space. When the field angle exceeds 90 degrees, the tangent does not predict the linear
image height correctly.

The grid distortion plot is oriented in field units, so that the +y direction corresponds to the +y field coordinate
in field units (see Fields on page 96). This may or may not be different than the orientation of a spot diagram or
other plot plotted in image space units.
Footprint Diagram
Purpose:
Displays the footprint of the beam superimposed on any surface. Used for showing the effects of vignetting
and for checking surface apertures.
Settings:
Item Description
Ray Density Determines the number of rays traced across the half pupil; a setting of 10 will trace a
grid of 21 x 21 rays.
Surface The surface to show the beam footprint on.
Delete Vignetted If checked, then rays which are vignetted by subsequent surfaces will not be drawn.
Rays which are vignetted by prior surfaces are never drawn.
This helps distinguish the various wavelengths.
distinguish between each wavelength, or "Config" to distinguish between configurations.
Discussion:
This feature will draw the shape of the surface, and then superimpose on that surface a grid of rays. If there
is no aperture on the surface, then a circular surface shape with a radial clear aperture of the semi-diameter value
is shown. Otherwise, the shape of the aperture is shown. The surface aperture is always shown as centered in
the frame; even if the aperture is decentered on the actual surface. If there is an obscuration on the surface, then
the obscuration will be drawn along with the circular aperture defined by the semi-diameter.
The ray grid size is specified by the ray density parameter, and rays may be from any or all fields, at any or all
wavelengths. Rays which are vignetted by surfaces prior to the surface shown are not drawn. Rays which are
vignetted by the surface or subsequent surfaces are not drawn if "delete vignetted" is checked, otherwise, they
are drawn. The ray set is apodized if any system pupil apodization is selected. The number of rays shown divided
by the total number of rays launched is shown as a percentage.
Longitudinal Aberration
Purpose:
Displays the longitudinal aberration as a function of pupil height at each wavelength.
Settings:
Item Description
Plot Scale Defines the maximum scale for the plot in lens units. Enter zero for automatic.

Discussion:
This feature computes the distance from the image surface to where a zonal marginal ray "focuses", or crosses
the optical axis. The computation is performed only for the on axis field point, and only for zonal marginal
tangential rays as a function of pupil zone. The base of the plot is on axis, and the top of the plot represents the
maximum entrance pupil radius. There are no units on the vertical scale because the plot is always normalized
to the maximum entrance pupil radius. The horizontal scale is in lens units, and represents the distance from the
image surface to the point where the ray crosses the optical axis.
Because longitudinal aberration is defined in terms of the distance to the ray-axis crossing point, this feature
may produce meaningless data for non-rotationally symmetric systems. Great care should be exercised in
interpreting this plot for non-rotationally symmetric systems.
Lateral Color
Purpose:
Displays the lateral color as a function of field height.
Settings:
Item Description
Plot Scale Defines the maximum scale for the plot in lens units. Enter zero for automatic.
Use Real Rays If checked, real rays are used, otherwise paraxial rays are used.
All Wavelengths If checked, then data for all defined wavelengths will be displayed. Each wavelength will
be referenced to the primary wavelength. If not checked, then the difference between
the shortest and longest wavelength rays will be used. See the discussion.
Show Airy Disk If checked, then the Airy disk radius at the primary wavelength will be plotted on either
side of the reference line to indicate the extent of the Airy disk.
Discussion:
This feature computes the lateral color either of two ways:
If "All wavelengths" is checked: The data computed is the distance on the image surface from the intercept of
the chief ray at each wavelength to the primary wavelength chief ray intercept.
If "All Wavelengths" is not checked: The data computed is the distance on the image surface from the shortest
wavelength chief ray intercept to the longest wavelength chief ray intercept.
The base of the plot is on axis, and the top of the plot represents the maximum field radius. Only positive field
angles or heights in the Y direction are used. The vertical scale is always normalized to the maximum field angle
or height. The horizontal scale is in micrometers. Either real or paraxial rays may be used.
If "Show Airy Disk" is check on, then the Airy disk radius will be approximated as the square root of the sum
of the squares of the X and Y direction Airy disk radii.
This feature may produce meaningless data for non-rotationally symmetric systems. Great care should be
exercised in interpreting this plot for non-rotationally symmetric systems.
Y-Ybar Diagram
Purpose:
Y-Ybar diagram.
Settings:
Item Description
First Surface The first surface for which data will be plotted.
Last Surface The last surface for which data will be plotted.

Item Description
Plot Scale Sets the maximum scale for the plot. The plot is always shown in a square box; the
default scale is the maximum transverse ray coordinate. Enter zero for automatic.
Discussion:
The Y-Ybar diagram is a plot of marginal ray height as a function of chief ray height for a paraxial skew ray at
every surface in the lens.
Chromatic Focal Shift
Purpose:
Chromatic focal shift plot.
Settings:
Item Description
Maximum Shift The maximum extent in lens units for the horizontal axis. The vertical axis scale is set
by the range of wavelengths defined. Enter zero for automatic.
Pupil Zone The radial zone in the pupil used to compute the back focus. The default value of zero
means a paraxial ray will be used. Values between 0 and 1 mean real marginal rays are
used in the appropriate zone in the entrance pupil. A 1 is at the edge of the pupil, or full
aperture.
Discussion:
This is a plot of the shift in back focal length with respect to the primary wavelength. At each plotted
wavelength, the shift in the image plane required to reach focus for that color marginal ray is computed. The shift
distance is computed in the same media (glass or air) as the surface prior to the image surface. This plot may
not be meaningful for non-paraxial systems.
The maximum shift setting overrides the default scaling. Units are in lens units. The entire plot is always
referenced to the primary wavelength paraxial focus. The diffraction limited depth of focus listed is given by the
2
formula 4F , where F is the working F/# and is the primary wavelength.
Dispersion Diagram
Purpose:
Plots index of refraction as a function of wavelength for any material in the glass catalog.
Settings:
Item Description
Min Wavelength Defines the left X- axis of the dispersion plot.
Max Wavelength Defines the right X- axis of the dispersion plot.
Minimum Index Defines the bottom Y- axis of the dispersion plot. Enter zero for automatic scaling.
Maximum Index Defines the top Y- axis of the dispersion plot. Enter zero for automatic scaling.
Glass The name of the material.
Use Temperature, If checked, then the change in index due to temperature and pressure effects will be
Pressure considered.
Discussion:

This is useful for checking if constants of dispersion or other formula data were entered correctly.
Glass Map
Purpose:
Plots names of glasses on the glass map according to the d-light index of refraction and Abbe V-number. The
index and Abbe number are computed from the data in the glass catalog. All currently loaded glass catalogs are
searched for glasses within the boundary values specified in the following table.
Settings:
Item Description
Min Abbe Defines the left X- axis of the plot.
Max Abbe Defines the right X- axis of the plot.
Min Index Defines the bottom Y- axis of the plot. Enter zero for automatic scaling.
Max Index Defines the top Y- axis of the plot. Enter zero for automatic scaling.
Discussion:
This is useful for locating a glass with particular refractive and dispersive properties. By convention, the glass
map is shown with the Abbe number decreasing while going from left to right, which is why the default min and
max Abbe numbers seem to be reversed.
Internal Transmittance vs. Wavelength
Purpose:
Plots the internal transmittance for any thickness as a function of wavelength for any material in the glass
catalog.
Settings:
Item Description
Min Wavelength Defines the left X- axis of the plot.
Max Wavelength Defines the right X- axis of the plot.
Minimum Defines the bottom Y- axis of the plot.
Transmission
Maximum Defines the top Y- axis of the plot. Enter zero for automatic scaling.
Transmission
Thickness The thickness of the glass in millimeters.
Discussion:
This is useful for checking the transmittance of a particular glass. See the Chapter Polarization Analysis.
System Summary Graphic
Purpose:
Displays in a graphic window a summary of the system data, similar to the text based system data report.
Settings:
None.
Discussion:

This graphic is primarily used to display a summary of system data within the printed page of the report
graphics 4 or 6 feature; described in the Reports Menu chapter.
Aberration coefficients
Seidel Coefficients
Purpose:
Displays Seidel (unconverted, transverse, and longitudinal), and wavefront aberration coefficients.
Settings:
Item Description
Wavelength The wavelength number to use for the calculation.
Discussion:
ZEMAX will compute the unconverted Seidel, transverse, longitudinal, and some wavefront coefficients. The
Seidel coefficients are listed surface by surface, as well as a sum for the entire system. The coefficients listed are
for spherical aberration (SPHA, S1), coma (COMA, S2), astigmatism (ASTI, S3), field curvature (FCUR, S4),
distortion (DIST, S5), longitudinal color (CLA, CL), and transverse color (CTR, CT). The units are always the same
as the system lens units, except of course for the coefficients measured in waves.
These calculations are only valid and accurate for systems consisting entirely of standard surfaces. Any
systems which contain coordinate breaks, gratings, paraxial, or other non-standard surfaces are not adequately
described by the paraxial rays which are used to compute the coefficients.
Transverse aberration coefficients are listed for each surface and for the system as a whole. The coefficients
given are transverse spherical (TSPH), transverse sagittal coma (TSCO), transverse tangential coma (TTCO),
transverse astigmatism (TAST), transverse Petzval field curvature (TPFC), transverse sagittal field curvature
(TSFC), transverse tangential field curvature (TTFC), transverse distortion (TDIS), transverse axial color (TAXC),
and transverse lateral color (TLAC). The transverse aberrations are in the system lens units. The transverse
aberration coefficients may be very large in optical spaces where the light is nearly collimated, and have little
meaning in these optical spaces.
Longitudinal aberration coefficients are computed for longitudinal spherical aberration (LSPH), longitudinal
sagittal coma (LSCO), longitudinal tangential coma (LTCO), longitudinal astigmatism (LAST), longitudinal Petzval
field curvature (LPFC), longitudinal sagittal field curvature (LSFC), longitudinal tangential field curvature (LTFC),
longitudinal distortion (LDIS), longitudinal axial color (LAXC), and longitudinal lateral color (LLAC). The longitu-
dinal aberrations are in the system lens units. The longitudinal aberration coefficients may be very large in optical
spaces where the light is nearly collimated, and have little meaning in these optical spaces.
The wavefront coefficients given are spherical aberration (W040), coma (W131), astigmatism (W222), field
curvature Petzval (W220P), distortion (W311), axial color defocus term (W020), lateral color tilt term (W111), field
curvature sagittal (W220S), field curvature medial (W220M), and field curvature tangential (W220T). All the
wavefront coefficients are in units of wavelengths at the edge of the exit pupil. The various aberration coefficients
are interrelated according to the following table. The symbols n and u refer to the index of refraction and the
paraxial marginal ray angle in the object space side of each surface. The primes above the n and u symbols
indicate that these are the values on the image space side of the surface. For a discussion of the meaning and
derivation of the Seidel aberration coefficients, see Welford, Aberrations of Optical Systems, Smith, Modern Lens
Design, or O' Shea, Elements of Modern Optical Design. A list of good references can be found in REFERENCES
ON LENS DESIGN on page 31.

INTERRELATIONSHIP OF ABERRATION COEFFICIENTS
Name Seidel Wave Description Transverse Longitudinal
Spherical Spherical S1
S1 S1
S1 ----- ----------- -------------
-
2
8 2n'u' 2n'u'
Coma Sagittal S2
S2
----------- -------------
-
2
2n'u' 2n'u'
S2
S2 -----
2 Tangential
3S 2 3S 2
----------- -------------
-
2n'u' 2
2n'u'
Astigmatism From tangential S3
S3 to sagittal foci S3
S3 ----- -------- ----------
-
2 n'u' 2
n'u'
Field Curvature Gaussian to S4
S4 S4 Petzval S4
----- ----------- -------------
-
4 2n'u' 2
2n'u'
Gaussian to
S3 + S4 sagittal ( S3 + S4 ) S3 + S 4
S3 + S4 ----------------- ---------------------- -----------------
4 2n'u' 2
2n'u'
Gaussian to
2S 3 + S 4 medial ( 2S 3 + S 4 ) 2S 3 + S 4
2S 3 + S 4 -------------------- ------------------------- --------------------
4 2n'u' 2
2n'u'
Gaussian to
3S 3 + S 4 tangential ( 3S 3 + S 4 ) 3S 3 + S 4
3S 3 + S 4 -------------------- ------------------------- --------------------
4 2n'u' 2
2n'u'
Distortion Distortion S5
S5 S5
S5 ----- ----------- -------------
-
2
2 2n'u' 2n'u'
Axial Color Chromatic CL
CL aberrations are C
CL ------ measured
-------L- ----------
-
2
2 n'u' n'u'
between the
extreme defined
Lateral Color wavelengths,
C CT
CT CT referenced to the -------T- ----------
-
selected n'u' 2
wavelength. n'u'
The Petzval radius of curvature shown on the Seidel screen is in system lens units, as is the Lagrange (or
optical) invariant.
Zernike Fringe Coefficients
Purpose:
Calculates the Zernike coefficients using the Fringe (also called the "University of Arizona") polynomials.

Settings:
Item Description
Sampling Specify the density in the pupil to use for coefficient fitting. Larger grid sizes are more
accurate, although the computation time increases.
Max Term Specify the maximum Zernike coefficient to compute. Any value up to 37 may be
specified.
Field The field number to use for the calculation.
Ref OPD To Vertex Normally, ZEMAX references the OPD to the chief ray; which in effect subtracts out tilt
Discussion:
The Zernike screen displays the individual coefficients as well as the peak-to-valley, RMS, variance, Strehl
ratio, residual RMS fit error, and maximum fit error.
The RMS of the wavefront error, , when referenced to the mean, is defined as
2 2 2
= W W
2
where W is the wavefront error, W is the mean square error, and W is the average wavefront error. The RMS
can actually be computed several different ways. If the mean wavefront term is ignored, then the RMS "referenced
2
to zero" results. This computation yields the square root of W directly, and is rarely used.
If the mean wavefront is subtracted from all the wavefront phase values (the absolute phase reference has no
physical meaning), then the RMS is "referenced to the mean".
Typically, the RMS is further referenced to the tilted and shifted reference sphere which minimizes the RMS.
This is equivalent to subtracting out not only the mean (which is piston) but the average tilt in x and y as well. This
is justified because tilt shifts the location of the diffraction image centroid, but otherwise has no affect on image
quality. For brevity, ZEMAX calls this reference point the "centroid", although it is a reference point which is usually
close to but not exactly at the diffraction image centroid. Most of the time, the RMS is taken to mean the RMS
referenced to the centroid, which is always the lowest of the three numbers.
The W values used by the Zernike RMS computations are the "raw" OPD values measured by computing the
phase of each ray as it intercepts the reference sphere. The reference sphere is centered on the chief ray - image
surface intercept point, and has a radius equal to the "exit pupil position" length (see Exit pupil position on
page 47). This method does not consider a subtle effect due to the change in the location of the center of the
reference sphere in the presence of coma. For this reason, the Zernike based estimate of the RMS wavefront
referenced to the centroid may differ slightly from that computed by the RMS analysis feature; see Comments
about RMS wavefront computations on page 133 for more information.

Strehl ratio approximation
The Strehl ratio is computed using the RMS referenced to the centroid by the following approximation:
( 2 )
2
S = e .
This approximation is valid for monochromatic calculations resulting in a Strehl ratio higher than about 0.10.
This feature computes a maximum of 37 Zernike terms. The particular Zernike terms used are not orthonormal,
but are instead all normalized to have unity magnitude at the edge of the pupil. Some of the higher order terms
in the expansion were dropped to keep the total number of terms small, and the terms remaining were selected
to favor accurate fitting of higher order spherical aberration. This particular set of Zernike polynomials is
sometimes called the "Fringe" or "University of Arizona" notation. The more formal, and more general polynomial
set is the Standard notation, sometimes called the "Born & Wolf" or the similar "Noll" notation, which is described
under the "Zernike Standard Coefficients" feature.
The Zernike Fringe polynomials are defined in the following table. The angle is measured counter clockwise
from the local +x axis. The radial coordinate is the normalized dimensionless parameter .
ZERNIKE FRINGE POLYNOMIALS
Term Z( , )
1 1
2 cos
3 sin
4 2
2 1
5 2
cos 2
6 2
sin 2
7 2
( 3 2 ) cos
8 2
( 3 2 ) sin
9 4 2
6 6 + 1
10 3
cos 3
11 3
sin 3
12 2 2
( 4 3 ) cos 2
13 2 2
( 4 3 ) sin 2
14 4 2
( 10 12 + 3 ) cos
15 4 2
( 10 12 + 3 ) sin

Term Z( , )
16 6 4 2
20 30 + 12 1
17 4
cos 4
18 4
sin 4
19 2 3
( 5 4 ) cos 3
20 2 3
( 5 4 ) sin 3
21 4 2 2
( 15 20 + 6 ) cos 2
22 4 2 2
( 15 20 + 6 ) sin 2
23 6 4 2
( 35 60 + 30 4 ) cos
24 6 4 2
( 35 60 + 30 4 ) sin
25 8 6 4 2
70 140 + 90 20 + 1
26 5
cos 5
27 5
sin 5
28 2 4
( 6 5 ) cos 4
29 2 4
( 6 5 ) sin 4
30 4 2 3
( 21 30 + 10 ) cos 3
31 4 2 3
( 21 30 + 10 ) sin 3
32 6 4 2 2
( 56 105 + 60 10 ) cos 2
33 6 4 2 2
( 56 105 + 60 10 ) sin 2
34 8 6 4 2
( 126 280 + 210 60 + 5 ) cos
35 8 6 4 2
( 126 280 + 210 60 + 5 ) sin
36 10 8 6 4 2
252 630 + 560 210 + 30 1
37 12 10 8 6 4 2
924 2772 + 3150 1680 + 420 42 + 1
Zernike Standard Coefficients

Purpose:

Calculates the orthonormal Zernike coefficients computed using the notation defined in R. Noll, "Zernike
polynomials and atmospheric turbulence", J. Opt. Soc. Am., Vol. 66, No. 3, p207 (1976). This notation allows
terms to be consistently defined up to (almost) arbitrary order, rather than the limited 37 term Fringe polynomials.
Settings:
Item Description
specified.
Discussion:
This feature is almost identical to the "Zernike Fringe Coefficients" feature, except a somewhat different
numbering scheme is used, more terms in the expansion are available for fitting, the terms are orthogonal and
normalized, and none of the terms are skipped. See the description of Zernike Fringe Coefficients on page 165
for some important details.
The most general way to express the Zernike polynomials is in the form
m
m im Rn ( ) cos m
R n ( )e =
Rm ( ) sin m
n
The radial portion of the polynomial is defined by two indices, n and m. The n index defines the order of the
5
radial power; so an n value of 5 would indicate all polynomials whose maximum radial power was . Only certain
values for m are allowed once n is chosen; n + m must be even, and 0 m n . For details on Zernike Standard
polynomials, see "Principles of Optics", by Born & Wolf, referenced in Chapter 1, or the Noll reference listed
above. The terms are orthonormal such that the magnitude of the coefficient of each term is the RMS contribution
of the term.
This feature lists the formulas next to the fitted coefficients; the entire 231 term table is too long to include here.
The first 28 terms are given below. The angle is measured counter clockwise from the local +x axis. The radial
coordinate is the normalized dimensionless parameter .

ZERNIKE STANDARD POLYNOMIALS
Term Z( , )
1 1
2
4 cos
3
4 sin
4 2
3 ( 2 1 )
5 2
6 ( sin 2 )
6 2
6 ( cos 2 )
7 3
8 ( 3 2 ) sin
8 3
8 ( 3 2 ) cos
9 3
8 sin 3
10 3
8 cos 3
11 4 2
5 ( 6 6 + 1 )
12 4 2
10 ( 4 3 ) cos 2
13 4 2
10 ( 4 3 ) sin 2
14 4
10 cos 4
15 4
10 sin 4
16 5 3
12 ( 10 12 + 3 ) cos
17 5 3
12 ( 10 12 + 3 ) sin
18 5 3
12 ( 5 4 ) cos 3
19 5 3
12 ( 5 4 ) sin 3
20 5
12 cos 5
21 5
12 sin 5
22 6 4 2
7 ( 20 30 + 12 1 )

Term Z( , )
23 6 4 2
14 ( 15 20 + 6 ) sin 2
24 6 4 2
14 ( 15 20 + 6 ) cos 2
25 6 4
14 ( 6 5 ) sin 4
26 6 4
14 ( 6 5 ) cos 4
27 6
14 sin 6
28 6
14 cos 6
Zernike Annular Coefficients

Purpose:
Calculates the orthonormal Zernike coefficients computed using the notation defined in R. Noll, "Zernike
polynomials and atmospheric turbulence", J. Opt. Soc. Am., Vol. 66, No. 3, p207 (1976), with extensions for
annular pupils, as described in V. N. Mahajan, "Zernike annular polynomials for imaging systems with annular
pupils", J. Opt. Soc. Am., Vol. 71, No. 1, p75 (1981). Thanks are due to Brett Patterson of the UK Astronomy
Technology Centre who assisted in the development of this feature.
Settings:
Item Description
specified.
Obscuration The obscuration ratio, of the annular pupil.
Discussion:
This feature is similar to the "Zernike Standard Coefficients" feature, except the pupil may be annular rather
than circular. See that discussion, and the references above, for important information. The annular nature of the
pupil is defined by the single parameter , defined as
R inner
= -------------- ,
R outer

where Rinner and Router are the inner and outer radii of the annular pupil. When is zero, the annulus is a
circle, and the Annular and Standard Zernike polynomials are identical. When is greater than zero, the annular
Zernike polynomials are distinct from the Standard Zernike polynomials. ZEMAX only traces rays in the annular
region when fitting annular polynomials, and unvignetted wavefront outside the annular region is ignored.
The annular polynomials are too complex in general to list on the text output window, so the first 22 terms are
listed here. The angle is measured counter clockwise from the local +x axis. The radial coordinate is the
normalized dimensionless parameter .
ZERNIKE ANNULAR POLYNOMIALS
Term Z( , , )
1 1
2 2 cos -
------------------
2
1+
3 2 sin -
-----------------
2
1+
4 2 2
3 ( 1 + 2 )
------------------------------------------
2
1
5 2
6 sin 2-
-----------------------------
2 4
1+ +
6 2
6 cos 2-
-----------------------------
2 4
1+ +
7 4 2 2 2
2 2 ( 2 2 + 3 + ( 2 + 3 ) ) sin -
-----------------------------------------------------------------------------------------------------------
2 2 2 2 4
( 1 ) ( 1 + ) ( 1 + 4 + )
8 4 2 2 2
2 2 ( 2 2 + 3 + ( 2 + 3 ) ) cos -
------------------------------------------------------------------------------------------------------------
2 2 2 2 4
( 1 ) ( 1 + ) ( 1 + 4 + )
9 3
2 2 sin 3 -
----------------------------------------
2 4 6
1+ + +
10 3
2 2 cos 3 -
----------------------------------------
2 4 6
1+ + +
11 4 2 4 2 2
5 ( 1 + 6 + 6 + ( 4 6 ) )-
-------------------------------------------------------------------------------------------
2 2
( 1)

Term Z( , , )
12 4
2 2 4
( 1 ) ( 1 + + ) - 2 ( 3 + 3 6 4 2 + 2 ( 3 4 2 ) + 4 ( 3 4 2 ) ) cos 2
10 -------------------------------------------------------------
2 4 6 8
1 + 4 + 10 + 4 +
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
2 3 2 4
( 1) (1 + + )
13 4
2 2 4
( 1) (1 + + ) 2 6 2 2 2 4 2
10 -------------------------------------------------------------- ( 3 + 3 4 + ( 3 4 ) + ( 3 4 ) ) sin 2
2 4 6 8
1 + 4 + 10 + 4 +
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
2 3 2 4
( 1) (1 + + )
14 4
10 cos 4 -
----------------------------------------------------
2 4 6 8
1+ + + +
15 4
10 sin 4 -
----------------------------------------------------
2 4 6 8
1+ + + +
16
( 2 3 )EQ cos -
------------------------------------------------------- , where
2 3 2 4
( 1 ) ( 1 + 4 + )
2 2 2 4
( 1 ) ( 1 + 4 + )
E = -------------------------------------------------------- , and
2 4 6
1 + 9 + 9 +
8 2 4 6 2 4 2 4 2 2 4
Q = ( 3 + 3 12 + 10 12 ( 1 ) + 2 ( 15 24 + 5 ) + 4 ( 3 12 + 10 ) ) .
17
( 2 3 )EQ sin -
------------------------------------------------------- , where E and Q are defined as in term 16 above.
2 3 2 4
( 1 ) ( 1 + 4 + )
18 3
( 2 3 )EQ cos 3
------------------------------------------------------------ , where
2 4 2 4
( 1) ( 1 + )( 1 + )
2 6 2 4
( 1 ) (1 + )(1 + )
E = -----------------------------------------------------------------------------------------------------
- , and
2 4 6 8 10 12
1 + 4 + 10 + 20 + 10 + 4 +
8 2 2 2 4 2 6 2
Q = 4 + 4 5 + ( 4 5 ) + ( 4 5 ) + ( 4 5 ) .
19 3
( 2 3 )EQ sin 3
------------------------------------------------------------ , where E and Q are defined as in term 18 above.
2 4 2 4
( 1) ( 1 + )( 1 + )
20 5
2 3 cos 5
------------------------------------------------------------------
2 4 6 8 10
1+ + + + +

Term Z( , , )
21 5
2 3 sin 5
------------------------------------------------------------------
2 4 6 8 10
1+ + + + +
22 6 2 4 6 4 2 2 2 4
7 ( 1 + 12 + 30 20 3 ( 3 + 4 ) + 3 ( 3 12 + 10 ) )-
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
2 3
( 1)
Calculations
Ray Trace
Purpose:
Paraxial and real trace of a single ray.
Settings:
Item Description
Hx Normalized x-field coordinate. The value should be between -1 and 1.
Hy Normalized y-field coordinate. The value should be between -1 and 1.
Field Select either a specific field or "Arbitrary" to enter in Hx and Hy. If a specific field is
selected; the Hx and Hy controls are grayed out.
Wavelength The wavelength number of the ray to trace.
Px Normalized x-pupil coordinate. The value should be between -1 and 1.
Py Normalized y-pupil coordinate. The value should be between -1 and 1.
Global Coordinates If checked, all ray trace data is given in global coordinates rather than local coordinates,
except tangent angles.
Type Select "Dir Cosines" to display the ray direction cosines at each surface, "Tangent Ang"
to display the tangent of the angle the ray makes at each surface, or "Ym,Um,Yc,Uc" to
display the paraxial marginal and chief ray intercept/tangent angle display. The tangent
angle is the ratio of the x (or y) direction cosine to the z direction cosine.
Discussion:
The Hx, Hy, Px, Py, and Global Coordinates settings are ignored if the "Ym, Um, Yc, Uc" option is selected.
For a description of normalized coordinates, see the chapter "Conventions and Definitions".
For the other options, this feature allows the user to specify the normalized object coordinates, the normalized
pupil coordinates, and wavelength number and then view the real and paraxial ray coordinates at every surface.
The first set of data presented is for a real ray. The values presented are the coordinates (in the surface local
or the global coordinate system) of the ray intercept point. The direction cosines (or tangent angles) are the values
of the ray in the surface after refraction. The direction cosine value is the cosine of the angle the ray makes with
respect to the specified axis (the x-direction cosine is the cosine of the angle the ray makes with respect to the
x-axis, etc.) The second set of data is just like the first set, except it is for a paraxial ray. Tangent angles are always
computed relative to the local Z axis; regardless of the Global Coordinates setting.
Fiber Coupling Efficiency
Purpose:

This feature computes the coupling efficiency for single mode fiber coupling systems. For multi-mode fiber
coupling, see Calculating efficiency of multi-mode fibers on page 145. Also see Computing Fiber Coupling on
page 530.
Settings:
Item Description
Source Fiber
NA x/y Numerical Aperture of the source fiber in object space in the xz and yz planes
respectively. This is n times the sin of the half angle to the 1 over e squared intensity
point.
X Angle, Y Angle The angular rotation of the source fiber in object space, in degrees, measured from the
nominal orientation of the source fiber. The X angle is the angle measured in the XZ
plane; the Y angle is measured in the YZ plane.
Sampling The grid size to use for the numerical integration.
A lign So u rce to If unchecked, the fiber is oriented along the object z axis, centered on the field point. If
Chief Ray checked, the center of the fiber points along the chief ray for that field point.
Ignore Source Fiber If checked, then illumination of the pupil as specified by the System, General,
Apodization setting will be used. All computations are then referenced to the total energy
incident upon the entrance pupil.
Receiving Fiber
NA x/y Numerical Aperture of the receiving fiber in image space in the xz and yz planes
respectively. This is n times the sin of the half angle to the 1 over e squared intensity
point.
Tilt About X, Y The angular rotation of the receiving fiber in image space, in degrees, measured from
the nominal orientation of the receiving fiber. If both X and Y rotations are specified, the
rotation is first around X, then around Y. Note the rotations are about the specified axes;
so Tilt About X rotates the fiber toward or away from the Y axis. Rotations are done after
the XYZ decenters below.
Decenter XYZ The linear position shift in the x/y/z direction in lens units of the receiving fiber, measured
from the nominal position of the receiving fiber. Shifts are done before the XY tilts above.
Align Receiver to If unchecked, the fiber is positioned at XY coordinates (0.0, 0.0) on the image surface.
Chief Ray If checked, the fiber is repositioned in x and y to be centered at the point the chief ray
intercepts the image surface. In either case the receiving fiber is oriented along the
image space z axis, unless rotated using the controls above.
Discussion:
This feature computes fiber coupling for single-mode fibers with a Gaussian shaped mode. For multi-mode
fiber coupling, see Calculating efficiency of multi-mode fibers on page 145. For a more detailed and flexible
physical optics treatment of fiber coupling see Computing Fiber Coupling on page 530.
Fiber coupling efficiency is computed based upon a two fiber or a one fiber model. In the two fiber model, light
emerges from a source fiber to fill (or partially overfill) the entrance pupil of an optical system. Energy not collected
by the entrance pupil is lost, reducing the overall efficiency. If desired, the source fiber may be ignored for a one

fiber model; and in this case the efficiency is computed relative to the energy entering the entrance pupil; which
in turn is a function of the system apodization (see Apodization Type on page 86).
The system efficiency (S) is the sum of the energy collected by the entrance pupil which passes through the
optical system, accounting for both the vignetting and transmission of the optics (if polarization is used), divided
by the sum of all the energy which radiates from the source fiber:
2
t ( x, y )Fs ( x, y ) dx dy
S = E----------------------------------------------------- ,
Fs ( x, y ) dx dy
where F s is the source fiber amplitude function and the integral in the numerator is only done over the entrance
pupil of the optical system, and t ( x, y ) is the amplitude transmission function of the optics. The transmission is
affected by bulk absorption and optical coatings (if use polarization is checked on). If the source fiber is ignored,
then the integral in the denominator is only done over the entrance pupil, and the F s function is determined by
the system apodization, if any.
Aberrations in the optical system introduce phase errors which will affect the coupling into the fiber. Maximum
coupling efficiency is achieved when the mode of the wavefront converging towards the receiving fiber perfectly
matches the mode of the fiber in both amplitude and phase at all points in the wavefront. This is defined
mathematically as a normalized overlap integral between the fiber and wavefront amplitude:
T =
F r ( x, y )W ( x, y ) dx dy
-------------------------------------------------------------------------------------------------------------------
-,
Fr ( x, y )Fr ( x, y ) dx dy W ( x, y )W ( x, y ) dx dy
where F r ( x, y ) is the function describing the receiving fiber complex amplitude, W ( x, y ) is the function
describing the complex amplitude of the wavefront from the exit pupil of the optical system, and the symbol
represents complex conjugate. Note that these functions are complex valued, so this is a coherent overlap
integral.
Large phase errors in the wavefront function are difficult to numerically integrate accurately. Fortunately, rays
with huge phase aberrations are not going to contribute to the fiber coupling anyway, as these rays move through
2 phase cycles so quickly that they tend to all average to zero. To improve the accuracy of the fiber coupling
computation for highly aberrated systems, the integration algorithm will ignore rays with more than n/8 waves of
OPD error, where n is the number of rays across the pupil, which is determined by the sampling setting.
T has a maximum possible value of 1.0, and will decrease if there is any mismatch between the fiber amplitude
and phase and the wavefront amplitude and phase.
ZEMAX computes the values S and T. The total power coupling efficiency is the product of these numbers. A
theoretical maximum coupling efficiency is also computed; this value is based upon ignoring the aberrations but
accounting for all vignetting, transmission, and other amplitude mismatches between the modes.
To model the reflection loss from the front face of the receiving fiber, set the image space glass to the material
the fiber is made of, and check the box "use polarization". The reflectivity of the receiving fiber (and other surfaces)
will be accounted for.
YNI Contributions
Purpose:
This feature lists for each surface the paraxial YNI value, which is proportional to the Narcissus contribution
of that surface.

Settings:
Item Description
Discussion:
See Narcissus: reflections on retroreflections in thermal imaging systems, Applied Optics, Vol. 21, No. 18,
p3393 (1982) for a discussion.
Sag Table
Purpose:
This feature lists for the selected surface the sag (z-coordinate) at various Y-coordinate distances from the
vertex. The best fit spherical (BFS) radius is computed, and the sag of the BFS and the difference between the
BFS and the surface is also tabulated. The depth of material that would need to be removed from the BFS to yield
the asphere is also listed. Only the Y-coordinate of the surface is considered, therefore this feature is not useful,
and should not be used, if the surface is non-rotationally symmetric.
Settings:
Item Description
Surface The surface number to use for the calculation.
Step Size The distance between steps measured from the vertex at which the sag is computed.
The default value of zero will automatically choose a reasonable step size.
Reverse Direction See "Discussion".
Discussion:
This feature can be used to determine the maximum aspheric deviation of a surface. The sag table is also
useful for lens fabrication.
This feature lists out five columns:
Y-Coord: The y coordinate of the point on the surface being computed.
Sag: The sag of the surface at the y coordinate.
BFS Sag: The sag of the best fit sphere. The BFS is determined by the radius of the sphere that minimizes
the volume of material that would need to be removed from a spherical surface to yield the aspheric surface.
Deviation: The difference between the BFS Sag and the Sag, ignoring any vertex offset (see Remove below).
Remove: The depth of material at the specified y coordinate to remove assuming that the surface was first
generated to the best fit sphere. Note that for some aspheric surfaces, material may need to be removed from
the vertex; for other types of aspheres the minimum volume to remove is achieved by removing more material
from the central annular zones of the BFS.
The BFS radius and position (vertex offset) relative to the aspheric surface depends on the properties of the
aspheric surface, and whether the surface defines a boundary between air and glass, or glass and air, and
whether the surface is concave or convex. ZEMAX always attempts to choose a BFS that will "add" material on
the air side of the surface. The vertex offset and BFS radius will be selected to minimize the volume of material
that needs to be removed from the BFS to yield the asphere surface. For some surfaces, such as those between
glass in a cemented doublet, ZEMAX cannot determine which side is "air". For these surfaces, if the desired data
is not generated by default, selecting "Reverse Direction" will choose the other side of the surface as the air
surface.
It is always advisable to check the data and sign convention carefully before any fabrication decisions are
made.

Cardinal Points
Purpose:
This feature lists for the selected range of surfaces and wavelengths the locations of the focal, principal, anti-
principal, nodal, anti-nodal planes. The calculation is done for any defined wavelength and either the X-Z or Y-Z
orientation.
Settings:
Item Description
First Surface The starting surface number of the group to compute the cardinal points for.
Last Surface The ending surface number of the group to compute the cardinal points for.
Wavelength The wavelength number to use for the computation.
Orientation The orientation to use for computing the cardinal plane locations.
Discussion:
This feature may not return reliable results if coordinate breaks or non-centered optics are included within the
specified surface range.
Glass and Gradient Index
Dispersion Diagram
Purpose:
Plots index of refraction as a function of wavelength for any material in the glass catalog.
Settings:
Item Description
Min Wavelength Defines the left X- axis of the dispersion plot.
Max Wavelength Defines the right X- axis of the dispersion plot.
Minimum Index Defines the bottom Y- axis of the dispersion plot. Enter zero for automatic scaling.
Maximum Index Defines the top Y- axis of the dispersion plot. Enter zero for automatic scaling.
Use Temperature, If checked, then the change in index due to temperature and pressure effects will be
Pressure considered.
Discussion:
This is useful for checking if constants of dispersion or other formula data were entered correctly.
Glass Map
Purpose:
Plots names of glasses on the glass map according to the index of refraction and Abbe V-number. The index
and Abbe number are taken directly from the entries in the glass catalog, and are not computed based upon the
wavelength data or dispersion coefficients. All currently loaded glass catalogs are searched for glasses within the
boundary values specified in the following table.

Settings:
Item Description
Min Abbe Defines the left X- axis of the plot.
Max Abbe Defines the right X- axis of the plot.
Min Index Defines the bottom Y- axis of the plot. Enter zero for automatic scaling.
Max Index Defines the top Y- axis of the plot. Enter zero for automatic scaling.
Discussion:
This is useful for locating a glass with particular refractive and dispersive properties. By convention, the glass
map is shown with the Abbe number decreasing while going from left to right, which is why the default min and
max Abbe numbers seem to be reversed.
Internal Transmittance vs. Wavelength
Purpose:
Plots the internal transmittance for any thickness as a function of wavelength for any material in the glass
catalog.
Settings:
Item Description
Min Wavelength Defines the left X- axis of the plot.
Max Wavelength Defines the right X- axis of the plot.
Minimum Defines the bottom Y- axis of the plot.
Transmission
Maximum Defines the top Y- axis of the plot. Enter zero for automatic scaling.
Transmission
Thickness The thickness of the glass in millimeters.
Discussion:
This is useful for checking the transmittance of a particular glass. See the Chapter Polarization Analysis.
Grin Profile
Purpose:
Plots the index of refraction along any axis for gradient index surfaces.
Settings:
Item Description
Surface The surface number of the gradient index surface.
Wavelength The wavelength number to use.
Show Index vs. Choose X, Y, or Z as the independent variable.
Min X, Y, or Z Defines the left X- axis of the plot, the minimum value of the independent variable.
Max X, Y, or Z Defines the right X- axis of the plot, the maximum value of the independent variable.

Item Description
X, Y, or Z Value The value of the two fixed coordinates.
Minimum Index Defines the bottom Y- axis of the plot. Use zero for the default value.
Maximum Index Defines the top Y- axis of the plot. Enter zero for the default value.
Discussion:
This plot shows index of refraction along either the X, Y, or Z axis for gradient index surfaces.
Gradium Profile
Purpose:
Plots the axial index profile of a Gradium surface.
Settings:
Item Description
Profile The name of the profile. See the chapter on surface types for details.
Surface The surface number to plot.
Discussion:
If the surface is selected to "None", then the reference wavelength for the profile glass family is used, no matter
what wavelength is selected. If a Gradium surface is selected, then any defined wavelength, or the reference
wavelength may be selected. Also, if a surface number is selected, then the starting and ending points for the
glass blank will be indicated using an "X" on the plot. The starting and ending positions include consideration of
the sag of the surface at the defined semi-diameter.
Universal Plot
New Universal Plot...
Purpose:
Displays as either a plot or as a text listing the value of any optimization operand as a function of some other
parameter.
Settings:
Item Description
Independent variable settings
Select either surface, system, multi-configuration, or NSC data as the independent
variable.
Select the parameter to use as the independent variable. The available selections
include radius, curvature, thickness, conic, parameter, and extra data values if surface
parameters are chosen. For system parameters, the selections include the system
aperture, field and wavelength data, apodization factor, temperature, and pressure. For
configuration data, all multi-configuration operands are listed. For NSC data, object
position and parameter data are listed.

Item Description
The surface number for the independent variable parameter if the parameter is
associated with a surface. This field is used for the configuration number if configuration
data is selected. If NSC data is selected, this field is used for the object number.
Start/Stop Value The beginning and ending range of the independent variable.
# of Steps The number of values between the start and stop values, inclusive, to compute the
dependent variable function value.
Dependent variable settings
Operand The optimization operand to use as the dependent variable function. The whole system
merit function is also included at the top of the list. If the system merit function is
selected, then either the entire merit function or any specific operands value may be
selected using the "Line" control below.
Line The optimization operand # to use as the dependent variable function. The whole
system merit function is also included at the top of the list.
Int1/Int2 The Int1 and Int2 values for the selected operand. The dialog box will display the short
name for the Int1 and Int2 values to aid in identifying the data that needs to be supplied.
If the edit control is grayed out; then the selected operand does not use that value.
Hx/Hy/Px/Py The Hx, Hy, Px, Py values for the selected operand. The dialog box will display the short
name for these values to aid in identifying the data that needs to be supplied. If the edit
control is grayed out; then the selected operand does not use that value.
Min/Max Plot Value The min and max value to plot for the dependent variable (Y) function. If both of these
values are zero, then a default scale will be chosen. If either one is not zero, then the y
range of the plot will extend from the minimum to the maximum. If the data does not fit
within the specified range, then the data may plot beyond the borders of the data box.
Plot Title Any text may be entered which will be used as the title for the plot or text listing.
Save As New This will save the current settings in a Universal Plot (UPL) file specified by the user.
Universal Plot Any name which may be used as a valid file name may be entered. Although the
"browse" window allows the file to be saved in any directory, ZEMAX will only recognize
the file if it resides in the main ZEMAX directory.
Once the settings are saved, the file name will appear as a menu option under the
Universal Plot menu option for quick regeneration of the plot.
Load From This will load the settings from a previously saved Universal Plot (UPL) file specified by
the drop down list to the right of the button.
Discussion:
This feature will either plot as a graph or create a text listing of the value of any optimization operand as a
function of any number of system, surface, or multi-configuration data.
For example, suppose a plot of the sagittal MTF at 30 lp/mm was needed as a function of the decenter of a
lens group (which may be a useful diagnostic for tolerancing analysis). Since the MTFS operand computes the
sagittal MTF, the Universal Plot can generate such a graphic or text listing. See the Chapter "Optimization" for a
list of available optimization operands. A decenter of a lens group is defined by Parameter 1 or 2 on the relevant
coordinate break surface, and Parameter 1 and 2 are both listed in the available surface group parameters.
Because of the number of different plots this feature can generate, there are no intelligent defaults for either
the independent or dependent settings. These values need to be carefully provided in the settings dialog box. If
the optimization operand cannot be computed, an error message will be displayed and the plot will not be
generated.
Because many optimization operands accept Hx and Hy values to define the field point for the calculation,
these operands may require that the number of fields be set to 1, then set Hx = 0 and Hy = 1, then finally choose

Y Field 1 as the independent variable for a plot of the operand as a function of field. A similar trick works for getting
plots as a function of wavelength.
Some individual operands, like DIFF and SUMM, have no meaning if selected from the "Operand" column
because these operands are only defined when used as part of a larger merit function. If results from a calculation
like DIFF are required, then the merit function should be selected from the Operand list and the individual DIFFs
within the merit function selected for the "Line".
The escape key will terminate the analysis if the computation is taking too long.
Polarization
!These features are only available in the EE editions of ZEMAX.
For more information on polarization calculations, see POLARIZATION ANALYSIS on page 497.
Polarization Ray Trace
Purpose:
The polarization ray trace feature generates a text window which displays all the polarization data for a single
ray.
Settings:
Item Description
Jx, Jy Jones electric field. see Defining the incident polarization on page 510.
X, Y Phase Phase of the X, Y component of the Jones electric field in degrees.
Hx Normalized x-field coordinate. The value should be between -1 and 1.
Hy Normalized y-field coordinate. The value should be between -1 and 1.
Px Normalized x-pupil coordinate. The value should be between -1 and 1.
Py Normalized y-pupil coordinate. The value should be between -1 and 1.
Wavelength The wavelength number of the ray to trace.
Discussion:
This feature tabulates all of the data computed by ZEMAX to perform polarization ray tracing. See POLAR-
IZATION ANALYSIS on page 497.
Polarization Pupil Map
Purpose:
Generates a graph of the polarization ellipse as a function of pupil position. This aids in visualizing the change
in polarization over the pupil.
Settings:
Item Description
Wavelength The wavelength number of the rays to trace.
Field The field position number of the rays to trace.

Item Description
Surface The surface at which the data is shown. Data is after refraction through the specified
surface.
Sampling The grid size of the sampling in the pupil.
Discussion:
The polarization ellipse is a representation of the figure traced out by the electric field vector as the wave
propagates during one cycle. The magnitude of the ellipse is determined by the transmission of the ray, which is
generally a function of pupil position. See POLARIZATION ANALYSIS on page 497.
Transmission
Purpose:
Computes the integrated and surface by surface transmission through the optical system considering
polarization effects.
Settings:
Item Description
Sampling Specify the density in the pupil. Larger grid sizes are more accurate, although the
computation time increases.
Unpolarized If checked, the electric field definition is ignored and the unpolarized computation is
performed.
Discussion:
This feature tabulates for each field position and wavelength the integrated transmission of the optical system
for the specified polarization. The transmission is computed as a fraction of 100%, with 100% being that
transmitted if there were no absorption, reflection, or vignetting losses. The transmission calculation accounts for
vignetting factors, vignetting due to apertures or obscurations, ray clipping due to ray trace errors, surface Fresnel
or coating losses, and bulk internal transmittance due to absorption.
Also tabulated for each field and wavelength is the relative and total transmission of the chief ray. This allows
identification of where significant surface losses originate.
See also Relative Illumination on page 139.
Phase Aberration
Purpose:
Computes the polarization induced phase aberration of an optical system.
Settings:
Item Description
Discussion:

This feature computes for the specified field position and wavelength the phase aberration in the image space
X and Y orientations. Polarization phase aberrations are induced by the effects of refraction through dielectric
media, and by reflection from metallic or dielectric mirrors.
Like the usual OPD plots, the polarization phase aberration is referenced to the chief ray. However, there are
cases where the chief ray phase cannot be determined in both orientations. For example, in an axial symmetric
system, if the incident polarization is linear in the Y direction, there is zero intensity in the X direction for the chief
ray, and therefore the X phase is indeterminate. For other rays in the pupil, there is generally a slight rotation of
the polarization, and therefore the resulting electric field in the X direction yields a valid phase angle. To avoid
this discontinuity in the phase, ZEMAX averages two rays on either side of the chief ray to interpolate the chief
ray phase. The problem may still appear under certain circumstances, even with this averaging technique. In all
cases, the phase data is still valid, because the phase aberration has no effect on image quality if the intensity is
zero.
Transmission Fan
Purpose:
Generates a graph of the transmitted intensity for each field and wavelength as a function of either tangential
or sagittal pupil fans.
Settings:
Item Description
Unpolarized If checked, the electric field definition is ignored and the unpolarized computation is
performed.
Discussion:
The polarization transmission fan is useful for determining the transmission variation over the pupil as a
function of field and wavelength. See POLARIZATION ANALYSIS on page 497.
Coatings
!These features are only available in the EE editions of ZEMAX.
For more information on coating calculations, see POLARIZATION ANALYSIS on page 497.
Reflection vs. Angle
Purpose:
Computes the S, P, and average polarization intensity coefficients for reflection for the specified surface as a
function of incident angle.
Settings:
Item Description
Min Angle The minimum angle of incidence to plot. This defines the left edge of the plot.
Max Angle The maximum angle of incidence to plot. This defines the right edge of the plot.
Min Y The minimum y value to plot. This defines the bottom edge of the plot.

Item Description
Max Y The maximum y value to plot. This defines the top edge of the plot.
Surface The surface number which defines the interface to use.
Object The object number if the surface is a Non-Sequential Components surface.
Coat Group The coat/scatter group (CSG) number which defines the interface to use, if the surface
is a Non-Sequential Components surface.
Wavelength The wavelength number to use.
Reverse Direction If the selected surface is a Non-Sequential Components surface, choosing this option
will reverse the incident media and substrate.
Discussion:
The incident angle is measured in the medium prior to the specified surface.
Transmission vs. Angle
Purpose:
Computes the S, P and average polarization intensity coefficients for transmission for the specified surface as
a function of incident angle.
Settings:
See "Reflectivity vs. Angle" above.
Discussion:
Absorption vs. Angle
Purpose:
Computes the S, P, and average polarization intensity coefficients for absorption for the specified surface as
a function of incident angle.
Settings:
Discussion:
Diattenuation vs. Angle
Purpose:
Computes the R (reflected) and T (transmitted) diattenuation for the specified surface as a function of incident
angle.
Settings:
Discussion:
Phase vs. Angle
Purpose:
Computes the S and P polarization phase for reflection (if the surface is a mirror) or for transmission (if the
surface is not a mirror) for the specified surface as a function of incident angle.
Settings:

Discussion:
Retardance vs. Angle
Purpose:
Computes the retardance for the specified surface as a function of incident angle.
Settings:
Discussion:
Reflection vs. Wavelength
Purpose:
Computes the S, P, and average polarization intensity coefficients for reflection for the specified surface as a
function of incident wavelength.
Settings:
Item Description
Min Wave The minimum wavelength to plot. This defines the left edge of the plot.
Max Wave The maximum wavelength to plot. This defines the right edge of the plot.
Min Y The minimum y value to plot. This defines the bottom edge of the plot.
Max Y The maximum y value to plot. This defines the top edge of the plot.
Surface The surface number which defines the interface to use.
Object The object number if the surface is a Non-Sequential Components surface.
Coat Group The coat/scatter group (CSG) number which defines the interface to use, if the surface
is a Non-Sequential Components surface.
Angle The angle of incidence, in degrees, to use.
Reverse Direction If the selected surface is a Non-Sequential Components surface, choosing this option
will reverse the incident media and substrate.
Discussion:
See POLARIZATION ANALYSIS on page 497.
Transmission vs. Wavelength
Purpose:
Computes the S, P, and average polarization intensity coefficients for transmission for the specified surface
as a function of incident wavelength.
Settings:
See "Reflectivity vs. Wavelength" above.
Discussion:
See POLARIZATION ANALYSIS on page 497.
Absorption vs. Wavelength
Purpose:
Computes the S, P, and average polarization intensity coefficients for absorption for the specified surface as
a function of incident wavelength.

Settings:
Diattenuation vs. Wavelength
Purpose:
Computes the R (reflected) and T (transmitted) diattenuation for the specified surface as a function of incident
wavelength.
Settings:
Phase vs. Wavelength
Purpose:
Computes the S and P polarization phase for reflection (if the surface is a mirror) or for transmission (if the
surface is not a mirror) for the specified surface as a function of incident wavelength.
Settings:
Retardance vs. Wavelength
Purpose:
Computes the retardance for the specified surface as a function of incident wavelength.
Settings:
Physical Optics
Paraxial Gaussian Beam
Purpose:
Computes Paraxial Gaussian Beam parameters. Paraxial Gaussian Beams are limited to axial analysis of
rotationally symmetric systems. See also Skew Gaussian Beam on page 189 and Physical Optics Propagation
on page 190.
Settings:
Item Description
M2 Factor The M2 quality factor used to simulate mixed mode beams. See the Discussion.
Waist Size The radial size of the embedded (perfect TEM00 mode) beam waist in object space in
lens units.
Surf 1 to Waist The distance from surface 1 (NOT the object surface) to the beam waist location. This
parameter will be negative if the waist lies to the left of surface 1.
Update See below.
Orient
Surface
Discussion:
This feature computes ideal and mixed mode M2 Gaussian beam data, such as beam size, beam divergence,
and waist locations, as a given input beam propagates through the lens system. This discussion is not meant to
be a complete tutorial on laser beam propagation theory. For more information on Gaussian beam propagation,

see one of the following references: A. E. Siegman, Lasers, University Science Books (1986), R. Herloski, S.
Marshall, and R. Antos, "Gaussian beam ray-equivalent modeling and optical design", Applied Optics Vol. 22, No.
8 p1168 (1983), M. W. Sasnett and T. F. Johnston, Jr., "Beam characterization and measurement of propagation
attributes, Proc. SPIE Vol. 1414, p21 (1991), and A. E. Siegman, "New developments in laser resonators", Proc.
SPIE Vol. 1224, p2 (1990).
Limitations of the analysis
Gaussian beams are idealized, "perfect" beams, and this limits the range of systems for which Gaussian beam
analysis is useful. Because the calculation of Gaussian beam parameters is based upon paraxial ray data the
results cannot be trusted for systems which have large aberrations, or those poorly described by paraxial optics,
such as non-rotationally symmetric systems. Flat fold mirrors will not invalidate the results, but decentered or tilted
components generally will. This feature ignores all apertures, and assumes the Gaussian beam propagates well
within the apertures of all the lenses in the system.
For a more general beam propagation analysis feature, see Physical Optics Propagation on page 190.
Overview of Gaussian beams
A Gaussian laser beam is described by a beam waist size, a wavelength, and a location in object space. The
Gaussian beam is an idealization that can be approached but never attained in practice. However, real laser
beams can be well described by an embedded Gaussian beam with ideal characteristics, and a quality factor,
called M2, which defines the relative beam size and divergence with respect to the embedded Gaussian mode.
The ideal M2 value is unity, but real lasers will always have an M2 value greater than unity.
This feature requires the definition of the initial input embedded beam properties, and the M2 value. The input
embedded beam is defined by the location of the input beam waist relative to surface 1 (note this is not the object
surface, but the first surface after the object) and the waist radial size at this location. ZEMAX then propagates
this embedded beam through the lens system, and at each surface the beam data is computed and displayed in
the output window. ZEMAX computes the Gaussian beam parameters for both X and Y orientations.
Default beam parameters
ZEMAX defaults to a waist size of 0.05 lens units (no matter what the lens units are) and a surface 1 to waist
distance of zero; this of course means the waist is at surface 1. The default values may be reset by clicking on
the "Reset" button. After the default values are computed and displayed, any alternate beam waist size and
location may be entered and that Gaussian beam will be traced instead.
Propagating the embedded beam
Once the initial beam waist and location parameters are established, ZEMAX traces the embedded beam
through the system and computes the radial beam size, the narrowest radial waist, the surface coordinate relative
to the beam waist, the phase radius of curvature of the beam, the semi divergence angle, and the Rayleigh range
for every surface in the system. ZEMAX calls these parameters the Size, Waist, Waist Z, Radius, Divergence,
and Rayleigh, respectively, on the text listing that is generated.
The dimensions for all parameters are lens units except for the semi divergence angle, which is in units of
radians. The embedded Gaussian beam parameters can be computed using the following standard formulas.
The Rayleigh range, z r , is defined as
2
0
z r = ---------- .

The phase radius of curvature is given by
2
zr
R ( z ) = z + ---- ,
z

where z is the distance from the beam waist. The radial beam size at any z is computed from
12
( z ) = 0 1 + ----
z 2
,
z r
where 0 is the beam waist. The divergence angle of the beam is given by
1
= tan -----0- .
zr
Finally, the Gouy shift is the phase of the center of the Gaussian beam:
1 z
= tan ----
zr
The quality factor

All of the preceding results are correct for the ideal embedded Gaussian beam. For aberrated, mixed-mode
beams, an extension to the fundamental Gaussian beam model has been developed by Siegman. The method
uses a term called the beam quality factor, usually denoted by M2. The factor M2 can be thought of as "times
diffraction limited" number, and is always greater than unity. The M2 factor determines the size of the real,
aberrated Gaussian beam by scaling the size and divergence of the embedded Gaussian mode by M. It is
common practice to specify M2 for a laser beam, rather than M, although the factor M is used to scale the beam
size. The M2 factor must be measured to be determined correctly. If the M2 factor is set to unity, the default value,
ZEMAX computes the TEM00 data described above. If M2 is greater than unity, then ZEMAX computes both the
embedded Gaussian beam parameters as well as the scaled data.
Interactive analysis
The Settings dialog box for this feature also supports an interactive mode. After defining the various input beam
parameters, clicking on "Update" will immediately trace the specified Gaussian beam, and display the usual
results in the dialog box. The parameters for any surface may be viewed, and the surface number selected from
the drop down list. The orientation may also be selected using the provided control.
The interactive feature does not in any way modify the lens or the system data; it is a handy calculator for
displaying paraxial Gaussian beam data.
Skew Gaussian Beam
Purpose:
Computes Skew Gaussian Beam parameters. Skew beams may enter an optical system at any surface from
any field position, and may travel through the optical system off axis. The Skew Gaussian Beam parameters are
computed using real rays and account for astigmatism but not higher order aberrations. See also Paraxial
Gaussian Beam on page 187 and Physical Optics Propagation on page 190.
Settings:
Item Description
X Waist Size The x direction radial size of the beam waist in the space prior to the "start surface" in
lens units.
Y Waist Size The y direction radial size of the beam waist in the space prior to the "start surface" in
lens units.

Item Description
Surf to Waist The distance from the "start surface" to the beam waist location. This parameter will be
negative if the waist lies to the left of the start surface.
Start Surface The surface to begin the beam propagation.
End Surface The surface to end the beam propagation.
Discussion:
For important background information on Gaussian beams, see Paraxial Gaussian Beam on page 187.
This feature computes the ideal skew Gaussian beam data, such as beam size, beam divergence, and waist
locations, as a given input beam propagates through the lens system. The beam is not limited to being on axis in
a symmetric optical system, and beam at any angle may be traced anywhere in the optical system.
The beam is initially defined by a waist size in the x and y directions in the optical space prior to the starting
surface. The field and wavelength settings are used to define a chief ray which is traced through the optical
system. This chief ray becomes the axis for the skew Gaussian beam throughout the propagation.
The x and y directions are initially defined by the entrance pupil orientation. The +y direction relative to the
chief ray is defined as a vector pointing from the chief ray coordinate at the entrance pupil to the coordinate of
the px = 0, py = 1 marginal ray coordinate at the entrance pupil. The px = 1, py = 0 marginal ray defines the +x
direction. For a discussion of the normalized pupil coordinates, see Normalized field and pupil coordinates on
page 49.
As the chief ray propagates through the optical system, the two marginal rays are propagated as well, and are
used to define the +x and +y directions in each optical space. All the data listed for x and y directions are measured
in this moving coordinate system.
The Skew Gaussian Beam feature results are generally not accurate for systems including non-sequential
component groups. For the most accurate results, replace non-sequential component groups with sequential
equivalents.
Physical Optics Propagation
!This feature is only available in the EE editions of ZEMAX.
Purpose:
This feature propagates an arbitrary coherent optical beam through the optical system. For a complete
description of this feature see PHYSICAL OPTICS PROPAGATION on page 515. The Physical Optics Propa-
gation feature results are generally not accurate for systems including non-sequential component groups; see
Propagating through non-sequential surfaces on page 525 for more information. For the most accurate results,
replace non-sequential component groups with sequential equivalents.
Settings:
Beam Definition Tab:
Item Description
X- Y- Sampling The number of points used to sample the beam.
X- Y- Width The initial width in lens units of the region represented by the array. See Auto below.
Auto Pressing this button will calculate the optimum X- and Y- Width values to maintain the
approximately the same number of pixels across the beam both within and outside the
Rayleigh range. See Comments about point spacing and sampling on page 524.

Item Description
Wavelength The wavelength number to use for the beam.
Field The field number of the chief ray used to align the initial beam.
Start Surface The starting surface for the initial beam. The beam will begin in the optical space just
prior to the starting surface.
The entrance pupil may be used as the starting surface. If the system is telecentric in
object space, then surface 1 will be used as the starting surface even if the entrance
pupil is selected.
End Surface The surface at which to terminate the propagation. The beam will stop just after entering
the optical space of the end surface.
Type See the discussion and Defining the initial beam on page 526.
Surf to Beam The distance from the starting surface to the starting beam position in lens units. This
value is negative if the beam starts to the left of the surface.
File See the discussion and Defining the initial beam on page 526.
Use Polarization If checked, 2 beam arrays will be propagated, one each for the X and Y polarizations of
the beam.
Separate X, Y If checked, the X and Y directions are propagated independently. This allows much more
efficient sampling if the beam is astigmatic or the optics are cylindrical or toroidal.
Peak Irradiance The initial beam peak irradiance in power per area. This is an alternative to Total Power.
Total Power The initial beam total power. This is an alternative to Peak Irradiance.
Parameters Different initial beam types require various parameters to define the distribution. These
parameter names and values will change depending upon the beam type setting. For
details, see Defining the initial beam on page 526.
Display Tab:
Item Description
Show As Choose surface, contour, false color, grey scale, cross sections, encircled, ensquared,
or enslitted displays. Note that the encircled energy is a function of radius; and the
ensquared and enslitted widths are the half widths.
Row/Column If cross section is chosen for Show As, this control selects the row or column to view.
Data Choose irradiance (beam power per area), phase (of the beam), transfer function
intensity, or transfer function phase. The Ex and Ey refer to the X and Y polarization
components of the beam. The transfer function values are the intensity and phase of
transmission for the final surface only; these are primarily used for debugging and
analysis of the POP results.
For text POP windows only, a propagation report is also available. The propagation
report lists various data about the propagation algorithm parameters used, and contains
warnings that indicate a possible inaccuracy in the results.
Scale Choose linear or logarithmic scaling of the data. Logarithmic scaling only applies to
irradiance data on graphical (not text) displays.
Project The beam may be viewed from any one of these perspectives: along the beam (this is
along the chief ray used by the pilot beam), along the surface normal, or along the local
Z axis of the surface. Note ZBF beam file data is always "along the beam". See the
comments about beam projection below.

Item Description
Save Output Beam If checked and a file name is provided, the complex amplitude of the beam at the end
To surface will be written to a file. This file can be read back in as a starting beam using the
Beam Type control. ZEMAX will add the extension "ZBF" to the provided file name. ZBF
files are stored in the directory \POP\BeamFiles.
Save Beam At All If checked, the beam file will be saved at every surface from the start surface to the end
Surfaces surface that uses the physical optics propagator. The "Save Output Beam To" option
must also be checked and a file name provided. The individual surface beam files will
be named according to the provided file name, with a suffix indicating the surface
number. For example, if the output file name is "MyBeamData.ZBF", then the beam file
name for surface 12 will be "MyBeamData_12.ZBF". ZBF files are stored in the directory
\POP\BeamFiles.
Zero Ph ase For Phase values are meaningless if the irradiance is extremely low. Computing the phase
Relative Irradiance angle of data points with nearly zero irradiance (relative to the peak irradiance in the
Below beam) will result in noisy plots and text listing of meaningless data. This value sets the
lower limit on the relative irradiance of data points for which the phase is computed. Data
points with relative irradiance lower than this threshold will have a phase value of zero.
Normalize Cross If zero, the vertical scale on cross section plots will be set by the maximum data value.
Sections To Otherwise, the maximum vertical scale is set by the normalization value or the maximum
data value, whichever is larger. Phase plots always scale from -pi to pi.
Contour Format String on page 131. The contours are defined in the same units as the
data being displayed.
Zoom In Magnifies the displayed area of the beam. This is useful for eliminating some of the
guard band from the displayed beam data.
Fiber Data Tab:
Item Description
Compute Fiber If checked, the fiber data on this tab will be used and the fiber coupling computed;
Coupling Integral otherwise, no fiber coupling computations will be performed. See Computing Fiber
Coupling on page 530.
Tilt About X/Y (deg) The tilt about each axis in degrees of the fiber mode with respect to the beam. The phase
of the fiber mode will be modified with a linear tilt proportional to the tilt angles.
Fiber Type Selects the mode for the fiber. See Computing Fiber Coupling on page 530.
File Selects the name of the DLL or data file describing the fiber mode. See Computing Fiber
Coupling on page 530.
Parameters Different fiber mode types require various parameters to define the mode. These
parameter names and values will change depending upon the fiber type setting.
Fiber Position The center of the receiving fiber may be referenced to the surface vertex or to the chief
ray. The incoming POP beam is referenced to the chief ray position and angle.
Discussion:
For a full discussion of physical optics propagation, see PHYSICAL OPTICS PROPAGATION on page 515.
Comment about beam projection
The analysis computes the beam irradiance or phase on a plane tangent to the chief ray at the point where
the chief ray intercepts the surface. The data is shown after the beam refracts into, or reflects from, the end
surface. The chief ray representing the center of the beam will generally intercept the surface at some angle of
incidence other than zero, and the surface normal will generally be at some angle to the local Z axis. Using the
"project" option on the "display" tab, the beam data may be viewed along the beam, along the surface normal, or

along the local Z axis. The latter two options are implemented by elongating the beam by the cosine of the angle
between the beam and the desired projection.
The beam itself is always represented by the data array in a coordinate system normal to the chief ray. For
this reason, the beam projected onto a surface by this feature may be elongated compared to the same beam
stored as a file and viewed with the Beam File Viewer, see Beam File Viewer on page 193.
Beams in ZEMAX are always centered on the chief ray for the selected field and wavelength. Therefore, the
data in the beam file is positioned relative to the chief ray. The center point in the beam file is at the coordinate
(nx/2+1, ny/2+1).
The analysis window may include some of the following data:
Display X Width/ Y Height: The width and height of the beam array in lens units at the end surface.
Peak Irradiance: The maximum irradiance in power per area. To set the power and area units see Units
on page 87.
Total Power: The total power in the beam. To set the power units see Units on page 87.
Pilot Beam Data: The Pilot Beam data includes the radial beam size, beam waist, position, and Rayleigh
range. For detailed information on the Pilot Beam see The pilot beam on page 520.
Fiber Coupling: If "Compute Fiber Coupling Integral" is checked, then the fiber coupling is computed and
displayed. The system efficiency is the fraction of beam power that transmits from the starting surface to
the surface at which the fiber coupling is computed. The receiver efficiency is the fraction of the power in
the beam mode that couples into the fiber mode. The total coupling efficiency is the product of these two
numbers. For more information see Computing Fiber Coupling on page 530.
Beam File Viewer

!This feature is only available in the EE editions of ZEMAX.
Purpose:
This feature allows viewing and analysis on previously stored ZEMAX Beam Files (ZBF). These files may be
user defined or may be generated by the Physical Optics Propagation feature, see the feature Physical Optics
Propagation on page 190. See also the Chapter PHYSICAL OPTICS PROPAGATION on page 515. For
information on the ZBF file format, see ZEMAX Beam File (ZBF) binary format on page 528.
Settings:
The available settings are a subset of what is available for the Physical Optics Propagation feature described
above.
Description:
The primary advantage to using this feature is that the beam may be propagated once through the optical
system, using the Physical Optics Propagation feature described above, and then the beam files may be analyzed
and viewed without the need to re-propagate the beam. Beam files for every surface may be generated by
choosing "Save Beam At All Surfaces" on the Physical Optics Propagation feature settings.
!For rapid viewing through a range of files that differ only by the surface number, use the left
and right cursor keys.
See the Comment about beam projection on page 192.

ZBF files are stored in the directory \POP\BeamFiles.

Chapter 8 TOOLS MENU
Optimization
Optimization
Purpose:
Optimization is used to improve or modify a design to meet specific conditions.
Discussion:
See OPTIMIZATION on page 385 for a full discussion of building merit functions, setting variables, and using
the optimization feature.
Global Search
Purpose:
This feature initiates a search for the global optimum, which is the best possible design for a given merit
function and set of variables.
Discussion:
See GLOBAL OPTIMIZATION on page 435.
Hammer Optimization
Purpose:
This feature automates the repetitive optimization of a design to escape local minima in the merit function.
Discussion:
See GLOBAL OPTIMIZATION on page 435.
Merit Function Listing
Purpose:
This function generates a text listing of the merit function which can be saved or printed. The overall merit
function numerical value is listed, as well as the numerical merit function of all user added operands (defined as
those operands placed before the DMFS operand) and the default merit function (all operands after the DMFS
operand). See also Merit function definition on page 391.
Remove All Variables
Purpose:
This is tool provides a quick way to remove all of the variables defined in the current lens.
Discussion:
Variables are removed by setting each of the values to "fixed" at the current value.
Glass Substitution Template
Purpose:
The template is used to define the maximum limits of the glass cost as well as the mechanical resistance
properties of glasses. The template is used to select acceptable glasses during global optimization, when
converting from model to real glasses (see Using model glasses on page 485), and when computing the value
of the RGLA operand (RGLA on page 415).
195 Chapter 8: TOOLS MENU

Settings:
Item Description
Use Glass Substitution Template If checked, the defined parameters are used to restrict the glasses
available for substitution by the global optimizer.
Exclude Glasses With No Data If checked, glasses with incomplete template data will not be selected
by the global optimizer. Only data corresponding to non-zero
template values are tested before excluding a glass.
Standard, Preferred, Obsolete, Special These checkboxes indicate the acceptable status settings of glasses.
These options are additive. For example, if both Standard and
Obsolete are checked, then either Standard or Obsolete glasses are
acceptable, Note a single glass may have only one status. See
Description of catalog data on page 478.
Maximum Relative Cost The maximum relative cost. The cost of Schott BK7 is 1.0, all other
glasses are more expensive, and have larger relative costs. Use zero
to ignore relative cost.
Maximum Climatic Resistance (CR) The maximum allowed value of the CR code. Use zero to ignore CR.
Maximum Stain Resistance (FR) The maximum allowed value of the FR code. Use zero to ignore FR.
Maximum Acid Resistance (SR) The maximum allowed value of the SR code. Use zero to ignore SR.
Maximum Alkali Resistance (AR) The maximum allowed value of the AR code. Use zero to ignore AR.
Maximum Phosphate Resistance (PR) The maximum allowed value of the PR code. Use zero to ignore PR.
OK Accepts the current settings.
Cancel Reverts to the prior settings.
Reset Restores the default settings.
Discussion:
If "Use Glass Substitution Template" is unchecked, any glass defined in the current catalogs may be selected
if not explicitly excluded (by having "exclude substitution" checked in the glass catalog) and if the wavelength
range is suitable for the current lens. For glasses that do not include the cost or resistance codes, a "?" will be
displayed in the glass catalog. If no cost or resistance data is provided for a glass, the glass will not be selected
if "Exclude Glasses With No Data" is checked. Template data is stored along with the lens file; so the template
parameters are lens file specific.
Tolerancing
Tolerancing
Purpose:
Tolerancing.
Discussion:
See the chapter "Tolerancing" for a complete discussion.
Tolerance Listing
Purpose:
This function generates a text listing of the tolerances which can be saved or printed.
Tolerance Summary
Purpose:
Chapter 8: TOOLS MENU 196

This function generates a text listing of the tolerances which can be saved or printed. The format is somewhat
easier to read than the Tolerance Listing, and it uses no mnemonics which makes the tolerances easier to
understand for fabricators and others not familiar with ZEMAX terminology.
Test Plates
Test Plate Fitting
Purpose:
Provides automatic fitting of radii to a vendor test plate list.
Settings:
Item Description
File Name Selects different test plate lists
Method of Fit Chooses the fitting order, which may affect the quality of the fit.
# Opt Cycles Chooses the number of optimization cycles to run, or Automatic.
Discussion:
This feature automates the task of matching radii of lens elements to the existing tooling of a particular vendor.
The current merit function is used as a figure of merit during the fitting process.
To match a particular radius, make it variable on the lens data editor. As many radii as desired can be matched
at one time. ZEMAX will attempt to fit all radii with variable status in the Lens Data Editor, even if the radii
correspond to surfaces that are not spheres, such as toroids, cylinders, or other aspheric surfaces. To prevent
fitting of the radii on these surfaces, remove the variable flag before performing the fitting.
Now select the test plate fitting tool. Select which vendor test plate file is to be used. The fitting method may
be selected as follows:
Try All Methods: Try all of the following methods, and uses whichever one yields the lowest merit function.
Best To Worst: Fits the radii with the closest test plate match (measured in fringes) first.
Worst To Best: Fits the worst fitting radii first.
Long To Short: Fits the longest radii first.
Short To Long: Fits the shortest radii first.
Now press OK and the test plate fitting will begin. ZEMAX starts by searching the test plate list for the closest
match (in fringes) between all of the radii and all of the test plates. Each test plate must be of the correct shape
(convex or concave or both, as required) and must have sufficient diameter to test the clear aperture of the lens
surface (as determined by the semi-diameter value on the Lens Data Editor). The test plate has sufficient diameter
if the diameter is at least 3/4 of the clear aperture diameter of the lens surface. The surface being fitted must also
have an edge sag no more than 0.2 mm different than the test plate at the semi-diameter of the surface. In practice
this is hundreds of fringes of power and test plates that do not meet this test would likely increase the merit
function unacceptably.
The radius of the test plate that best fits one of the radii is then substituted in for the actual radius. The variability
of the chosen radius is removed, and the lens is reoptimized. For this reason it is important to allow compensators
such as spacings to be variable as well as the radii to be fit. The reoptimization will adjust all remaining variables,
including the remaining radii. Note the optimization will use the current merit function. After optimization, if there
are any variable radii remaining, the procedure repeats. Note that the radii are not fit in general in the order they
appear in the lens data editor.
During the fitting process, the number of radii remaining and the current merit function will be presented. After
all of the radii have been fit, a report will be presented on the screen. The report will provide the vendor
identification information, and a list of the radii that were changed and which vendor test plate ID numbers were
selected.

There is no way to know if the test plates selected are optimal. If there are a large number of test plates in the
test plate list, and the values are reasonably continuous without large gaps, the fitting will usually be quite good.
If the merit function increases unacceptably during the fitting process, either a different vendor test plate list should
be used, the design needs to be modified, or some of the lenses may have to have custom test plates made.
Usually, the last radii to be fit are the likely candidates for custom test plates. The report shows the order in which
the radii were fit.
Using the "Try All Methods" option will yield the lowest overall merit function, since all 4 methods are tried and
the only the one yielding the lowest merit function is retained. However, there may be some algorithm other than
the 4 listed that might yield a better fit.
There are occasions where the algorithm is unable to find any suitable test plate match for a given radius. This
can occur when no test plate with a large enough diameter is sufficiently close to the desired radii. If this occurs,
a message stating "NO MATCH FOUND!" will be printed in the report file, and that radii will subsequently be
ignored. Usually this means test plates for this radii will have to be custom made.
All test plate data is provided by the respective vendors, and no warranty is provided as to the accuracy or
completeness of the data. For details on the latest vendor test plate lists, contact the vendor. New test plate files
may be added to ZEMAX. The test plate data files end in the extension .TPD, and must be placed in the
\TESTPLAT directory. The files are in ASCII with the following format:
! Header line 1
! Header line 2
! etc... up to 15 lines of header info is supported
partname radius diameter code
etc...
where partname is an ID number or name for the test plate, radius is the radius of curvature in millimeters,
diameter is the actual plate diameter in millimeters, and code is an integer value. The code should be -1 for
concave only, 0 for both concave and convex, and 1 for convex only. All four values must be on one line, separated
by spaces or tabs. No spaces or tabs are allowed in the partname parameter. Each test plate (or pair) should be
on one line, and the lines separated by carriage returns. The numeric values can be in free format, but must be
in millimeters. The maximum number of test plates allowed per file is 30,000. The header lines can be used for
any purpose, and typically indicate the vendors name, address, phone, e-mail, and other contact data. The ! is
replaced by a space when ZEMAX lists the data out on test plate lists and fitting reports.
Test Plate Lists
Purpose:
Displays in a text window a list of test plates from a specific vendor.
Settings:
Item Description
File Name The name of the test plate file. The file must be in the \TESTPLAT directory.
Discussion:
All units on the report are in mm. The CC and CX columns indicate the availability of concave and convex test
plates, respectively.
Catalogs
Glass Catalogs
Purpose:
Provides access to the glass catalogs.
Discussion:
See the chapter "Using Glass Catalogs" for an extended discussion of this feature.

Glass Compare
Purpose:
Compares 2 or 3 glass catalogs to assist in finding glasses used in visible light with similar index and dispersion
characteristics.
Settings:
Item Description
Primary The name of the primary glass catalog. This catalog is used as the reference catalog.
Secondary The name of the secondary glass catalog. Only glasses which are considered equivalent
to those in the primary catalog will be listed.
Tertiary The name of the tertiary glass catalog. Only glasses which are equivalent to those in the
primary catalog will be listed.
Index Tolerance The maximum allowed difference in index at d-light for a glass from the secondary or
tertiary catalog to be considered equivalent to a glass from the primary catalog.
Abbe Tolerance The maximum allowed difference in Abbe number at d-light for a glass from the
secondary or tertiary catalog to be considered equivalent to a glass from the primary
catalog.
Show Either all glasses in the primary catalog may be listed, or just those used in the current
lens file and configuration.
Sort Results may be sorted by index at d-light, name, or Abbe number at d-light.
Discussion:
This feature only lists glasses that are used in the F-C wavelength range. If "Show Used Glasses" is selected,
the name of the glass displayed in the glass column of the lens data editor is used to determine if the glass from
the primary catalog is used by the current lens and configuration. However, there is no check made that the glass
from the primary catalog is actually used by the lens. For example, if the current lens uses glass A1 from catalog
X, the primary catalog is Y, and Y also contains a glass called A1, then the data for A1 from catalog Y will be
displayed, even though the lens actually uses the A1 as defined in catalog X. If more than one catalog uses the
same name for different glasses, great care should be taken to verify all data is for the intended glass.
Lens Catalogs
Purpose:
Used to search for a particular lens from or browse through the stock lens catalogs.
Settings:
Item Description
Vendor Lists the available stock lens vendors. Each vendor name is the name of the file
which contains the stock lenses available from that vendor. The vendor files must
be placed in the directory for stock lenses, which is set on the Directories tab of
the File, Preferences dialog box.
Use Focal Length If checked, then the specified focal length range is used as part of the search
criteria, otherwise, any value will be accepted.
Focal Length Min/Max Used to define the range of acceptable focal lengths in millimeters.
Use Diameter If checked, then the specified entrance pupil diameter range is used as part of the
search criteria, otherwise, any value will be accepted.
Diameter Min/Max Used to define the range of acceptable entrance pupil diameters in millimeters.

Equi-, Bi-, Plano-, If any of these options are checked, then the search will be limited to lenses
Meniscus meeting at least one of the selected criteria. For example, if both Equi- and Bi- are
selected, then both equiconvex/equiconcave and biconvex/biconcave lenses will
be included.
Sp h e r i c a l , GRIN, If any of these options are checked, then the search will be limited to lenses
Aspheric, Toroidal meeting at least one of the selected criteria. The "Spherical" category is also used
as "other" for lenses which are not gradient index, aspheric, or toroidal.
Max elements Selects the maximum number of elements allowed in lenses that meet the search
criteria. Choose "Any #" to include all lenses in the search.
Search Results Lists all lens files in the specified directory which match the current search criteria.
Search Causes a new search to be conducted for lenses in the specified vendor file based
upon the current search criteria.
Load Loads the currently selected lens file into the Lens Data Editor, and updates all
windows to display data for the newly loaded lens.
Insert Inserts the currently selected lens file to the existing lens in the Lens Data Editor.
A prompt will be issued for the insert file options, such as at which surface the lens
should be inserted.
Prescription Displays a prescription report of the currently selected lens file.
Layout Displays a layout plot (or 3D Layout for systems without axial symmetry) of the
currently selected lens file.
Exit Closes the dialog box.
Discussion:
The "Search Results" table will list the part name, focal length, and diameter of all the lenses from the selected
vendor which meet the specified search criteria. After the diameter, a 3 part code is listed, such as (P,S, 1). The
first entry is the shape code, which is E, B, P, M, or "?", for Equi-, Bi-, Plano-, Meniscus, or other. Other is used
if the lens is multi-element. The second entry is S, G, A, or T, for Spherical, GRIN, Aspheric, or Toroidal. The third
entry is the number of elements.
Upon installation, ZEMAX creates a subdirectory within the ZEMAX directory called STOCKCAT. Within
STOCKCAT, many individual files are stored with the extension .ZMF. Each of these files contains a large number
of individual ZMX lens files, each of which represent stock lenses available from various vendors. The following
table lists the represented vendors and contact data.
STOCK LENS VENDORS
File Name Contact Data

ARCHEROPTX.ZMF Archer OpTx
3412 Enterprise Drive, Rowlett, TX 75088
972-463-8001
www.archeroptx.com
COHERENT.ZMF Coherent
3845 Atherton Road Suite #1, Rocklin, CA 95765.
Tel: 916-435-4230 Fax: 916-435-4234
Web: www.ealingcatalog.com
CORNING.ZMF Corning
3997 McMann Road, Cincinnati, OH 45245
Tel: 513-752-7000
Web: www.corningprecisionlens.com

CVI.ZMF CVI
200 Dorado Place SE, Albuquerque, New Mexico 87123
Tel: 800-296-9541 Fax: 505-298-9908
Web: www.cvilaser.com
EDMUND OPTICS.ZMF Edmund Industrial Optics
101 East Gloucester Pike, Barrington, NJ 08007
Tel: 856-573-6250 Fax: 856-573-6295
Web: www.edmundoptics.com
ESCO.ZMF Esco Products
171 Oak Ridge Road, Oak Ridge, New Jersey 07438
Tel: 800-922-3726 Fax: 973-697-3011
Web: www.escoproducts.com
GELTECH.ZMF LightPath Technologies, Inc.
Polycarbonate (Geltech) lenses from LightPath Technologies
(see below for contact data).
ISP OPTICS.ZMF ISP Optics
1 Bridge Street, Irvington, NY 10533
Tel: 914-591-3070 Fax: 914-591-3715
Web: www.ispoptics.com
JML.ZMF JML Optical Industries, Inc.
76 Fernwood Ave., Rochester, NY 14621
Tel: 585-342-8900 Fax: 585-342-6125
Web: www.jmloptical.com
LIGHTPATH.ZMF LightPath Technologies, Inc.
2603 Challenger Tech Ct, Suite 100, Orlando, FL 32826
Tel: 407-382-4003 Fax: 407-382-4007
Web: www.lightpath.com
LINOS PHOTONICS.ZMF LINOS Photonics, Inc.
459 Fortune Boulevard, Milford, MA 01757-1723
Tel: 508-478-6200 Fax: 508-478-5980
Web: www.linos-photonics.com
MELLES GRIOT.ZMF Melles Griot
2051 Palomar Airport Rd., 200, Carlsbad, California 92009
Tel: 800-835-2626 Fax: 760-804-0049
Web: www.mellesgriot.com
NEWPORT CORP.ZMF Newport Corporation
1791 Deere Avenue, Irvine, CA 92606
Tel: 800-222-6440 Fax: 949-253-1680
Web: www.newport.com
NSG.ZMF NSG America, Inc.
28 World's Fair Drive, Somerset, NJ 08873
Tel: 732-469-9650 Fax: 732-469-9654
Web: www.nsgamerica.com

OPTICS FOR RESEARCH.ZMF Optics For Research
P.O. Box 82 Caldwell, NJ 07006
Tel: 973-228-4480 Fax: 973-228-0915
Web: www.ofr.com
OPTOSIGMA.ZMF OptoSigma Corporation
2001 Deere Avenue, Santa Ana, California 92705
Tel: 949-851-5881 Fax: 949-851-5058
Web: www.optosigma.com
PHILIPS.ZMF Philips Competence Centre Plastic B.V.
PO Box 218-5600 MD, Eindhoven, The Netherlands
Tel: 31-40-273-2394
ROLYN OPTICS.ZMF Rolyn Optics Company
706 Arrowgrand Circle, Covina, California 91722
Tel: 626-915-5707 Fax: 626-915-1379
Web: www.rolyn.com
ROSS OPTICAL.ZMF Ross Optical Industries
1410 Gail Borden, El Paso, TX 79935
Tel: 800-880-5417 Fax: 915-595-5466
Web: www.rossoptical.com
SPECIAL OPTICS.ZMF Special Optics
315 Richard Mine Road, Wharton, NJ 07885
Tel: 973-366-7289 Fax: 973-366-7407
Web: www.specialoptics.com
THORLABS.ZMF Thorlabs Inc.
435 Route 206 North, Newton, NJ 07860
Tel: 973-300-3007 Fax: 973-300-3607
Web: www.thorlabs.com
To find a catalog lens that closely matches the properties of one or more surfaces in the Lens Data Editor,
place the cursor on the first surface of the surface group in the Lens Data Editor, then select Tools, Lens Catalogs.
The default focal length and diameter search ranges will be set to appropriate values for that lens. ZEMAX
computes the lens focal length and diameter and adds plus/minus 5% to define the default search ranges.
The Lens Search tool provides the capability to search through the available lenses to find an appropriate
choice. Once selected, the lens file may be loaded or appended to an existing design.
Coatings
Edit Coating File
Purpose:
Invokes the Windows NOTEPAD editor to edit the coating file for the current lens. This file contains the material
and coating definitions.
Discussion:
See the chapter "Polarization Analysis". If the coating file is edited, the file must be reloaded manually by
selecting Reload Coating File as described below, or the file may be saved and reloaded to update the new
coating data.

Reload Coating File
Purpose:
Reloads the coating file for the current lens. This is required if the file has been edited since the lens was
loaded.
Add Coatings to All Surfaces
Purpose:
Adds any specified coating to all surfaces with air-glass boundaries.
Discussion:
When selected, this tool will prompt for the name of the coating to apply. Any defined coating name may be
selected, or choose none to remove any existing coatings from surfaces. All surfaces which define a transition
from glass to air will have the coating applied (or removed); so this feature is primarily for applying anti-reflection
coatings.
Coating Listing
Purpose:
This function generates a text listing of the definitions of materials and coatings contained in the coating file
for the current lens.
Scattering
ABg Scatter Data Catalogs
Purpose:
Provides access to the ABg scatter data catalogs.
Settings:
Item Description
File The current ABg scatter data file name. This file must be placed in the \ABG_DATA
directory; the file name may be selected on the System, General dialog box on the Files
tab, as described on page 91.
Name The name of the ABg data being edited.
Wave The wavelength in micrometers at which the ABg data is defined (or was measured). If
this value is zero, no scaling with wavelength will be performed. Otherwise, the A and B
values will be scaled with wavelength as defined in the discussion below.
Use If checked, then the displayed row will be used. Up to 10 data rows may be defined for
each ABg data name.
Angle The angle of incidence in degrees. The value should be between 0.0 and 90.0, inclusive.
Intermediate angles of incidence are linearly interpolated in cosine space. If data is
required at angles less than the smallest angle defined, data for the smallest angle
defined will be used. Data for the largest angle defined will be used for any incidence
angles greater than the largest angle defined.
A, B, g The A, B, and g values at the specified angle of incidence.

Item Description
TIS Pressing this button will compute the total integrated scatter (TIS) for the displayed data.
This value must be less than 1.0 for accurate scattering results. See the discussion.
Save Saves the modified data to the catalog.
Save As Saves the modified data to a catalog of a different name.
Insert Inserts a new ABg data entry into the catalog; and prompts for the name of the new entry.
Delete Deletes the displayed ABg data entry.
Reload Loads from the last saved data file the ABg data; eliminating any changes made since
the last save.
Sort Sorts in ascending order by angle the displayed data. The data is automatically sorted
whenever the catalog is saved or loaded.
Report Sends the data to a text window which may then be printed.
Rename Prompts for a new name for the ABg data.
Discussion:
ABg data values are used to define scattering properties of both sequential and non-sequential surfaces. Each
ZEMAX lens files may use a separate ABg data catalog if desired. The ABg data file used is specified on the
System, General dialog box on the Files tab, as described on page 91. The selected ABg catalog only may be
edited from within ZEMAX using the ABg Scatter Data Catalog dialog box.
For a theoretical discussion of ABg scattering; see ABg model scattering on page 354.
The integral of the ABg distribution function over the unit vector circle that represents physically possible
scattering ray directions determines the total integrated scatter, or fraction to scatter. This integral is automatically
computed by ZEMAX when required; to view this integral press the TIS button. The TIS must be less than 1.0;
otherwise conservation of energy may not be preserved and the scattering will not be correct.
Wavelength scaling of ABg data
If a reference wavelength other than zero is defined, then the A and B coefficients are scaled according to the
following relations:
g4
A' = A --------- , and
ref
g
B' = B --------- ,
ref
where is the wavelength being traced and the unsubscripted A, B, and g parameters are measured at the
reference wavelength, ref .
Scatter Function Viewer

Purpose:
Displays BSDF as a function of the magnitude of the scatter vector x.

Settings:
Item Description
Lambertian Selects Lambertian scattering.
Gaussian Selects Gaussian scattering.
ABg Selects ABg scattering.
User Defined Selects User Defined scattering.
Wavelength The wavelength number to use. ABg scattering uses the specified wavelength, along
with the reference wavelength defined in the ABg catalog to scale the ABg coefficients.
Fraction The fraction of total energy to scatter. This value must be between 0 and 1. The fraction
to scatter linearly scales the BSDF. This fraction is ignored for ABg scatter functions.
Sigma The sigma value to use for Gaussian scattering.
ABg Name The name of the ABg scattering data to use. ABg scatter data names are defined in the
ABg Scatter Data Catalogs; see ABg Scatter Data Catalogs on page 203.
DLL Name The name of the user defined scattering DLL to use. See User defined scattering on
page 356.
(parameter names) These parameters are for user defined scattering DLL.
Bins The number of histogram bins to use in the computation and plot of the BSDF for user
defined scattering. Using a larger number of bins increases the resolution of the plot, at
the expense of requiring more rays to generate smooth data.
Rays The number of rays to scatter to determine the shape of the BSDF curve for user defined
scattering.
Alpha X, Y The angle in the X-Z and Y-Z planes respectively, of the specular ray. See the discussion
for details.
X, Y Minimum The minimum value of the scatter vector x, and BSDF (y), to plot, respectively. Note
these are powers of 10 as the BSDF plot is a log-log plot.
X, Y Decades The number of log scale decades to plot in the x and BSDF (y) directions, respectively.
Discussion:
This feature plots or lists BSDF data for any of the surface scattering models supported by ZEMAX. The BSDF
plot is a log - log plot, with BSDF on the vertical axis, and the magnitude of the scattering vector x on the horizontal
axis. The scattering vector x is a 2D vector that defines the change in the x and y direction cosines between the
specular ray and the scattered ray, when the local surface normal is oriented along the z direction. The vector x
is therefore the difference between the specular and scattered ray when these rays are projected onto the plane
of incidence. For a detailed discussion and drawing illustrating the scattering conventions used by ZEMAX, see
Scatter models on page 353.
For User Defined and ABg scattering, the Theta X and Theta Y values are used to define the specular ray
direction cosines. The defining equations which convert angles to cosines are as follows:
2 2
n = 1 ( 1 + ( tan x ) + ( tan y ) ) ,
l = n tan ( x ) , and
m = n tan ( y ) .

The comments below apply to each of the available scattering models:
Lambertian: Lambertian scattering is independent of incident angle and wavelength. The total integrated
scatter is 1.0. The resulting BSDF "curve" is a flat line with a value of 1 .
Gaussian: Gaussian scattering is symmetric about the specular ray projection. The total integrated scatter is
always 1.0.
ABg: The ABg values are defined in the ABg Scatter Data Catalogs; see ABg Scatter Data Catalogs on
page 203 for more information. The total integrated scatter is not generally 1.0; the actual value is listed in the
output plot or listing. ABg data may be scaled with wavelength; if a reference wavelength is defined in the catalog.
ABg data is also generally a function of incident angle. ZEMAX uses the catalog data with the angle of incidence
closest to that of the defined specular ray, however, no scaling or interpolation with angle is done.
User Defined: ZEMAX cannot determine the exact curve of the BSDF for user defined scattering. As an
alternative, a number of identical specular rays are scattered by the DLL, and the resulting scatter vector x values
are binned in a histogram, normalized, and plotted. BSDF plots made in this way are typically noisy in a complex
way that depends upon the nature of the scattering function defined by the DLL.
Apertures
Convert Semi-Diameters to Circular Apertures
Purpose:
Converts all surfaces without surface apertures to fixed semi-diameters with circular apertures set to the semi-
diameter value.
Discussion:
The primary purpose of this feature is to simplify the analysis of vignetting effects. For most optical designs, it
is best to use vignetting factors (see the chapter "System Menu" for a description of vignetting factors) during
optimization. However, vignetting factors are an approximation. This feature converts all semi-diameters to
surface apertures. The vignetting factors may then be deleted (this is not done automatically by this feature) and
the pupil can be overfilled to see where the rays really make it through the system.
Convert Semi-Diameters to Floating Apertures
Purpose:
Converts all surfaces without surface apertures to floating apertures which vignette at the semi-diameter value.
Discussion:
This feature is very similar to "Convert Semi-Diameters to Circular Apertures" except floating apertures are
used instead of fixed circular apertures. Floating apertures leave the semi-diameter value of the surface in
"automatic" mode, and dynamically adjust the vignetting aperture to match the semi-diameter value. Note if any
semi-diameters are "fixed" they remain fixed, and all vignetting will occur at the specified semi-diameter on each
surface.
Remove All Apertures
Purpose:
Turns off all surface apertures.
Discussion:
All surface aperture types are set to none by this tool.
Replace Vignetting With Apertures
Purpose:
Turns off vignetting factors and places surface apertures.
Discussion:
This tool uses the identical algorithm as many ZEMAX analysis features do to replace vignetting factors with
surface apertures. Analysis features that need to trace rays from field points in between defined field points need

to do this because vignetting is only defined for discrete field points. This tool is useful as a check that ZEMAX
does this conversion properly; and is also handy for removing vignetting factors.
Fold Mirrors
Add Fold Mirror
Purpose:
Inserts a fold mirror, including the required coordinate breaks, to bend a beam.
Settings:
Item Description
Fold surface Selects which surface will become the fold mirror. The selected surface should be a
dummy surface already located at the desired fold position.
Reflect angle The angle between the incident and reflected beam.
Tilt type Select either tilt about the local x or y axis.
Discussion:
This feature inserts two dummy surfaces, one before and one after the selected fold surface. The fold surface
then becomes a mirror, and the two newly inserted adjacent surfaces are set to be coordinate breaks with the
appropriate tilt angles. The second tilt angle is set as a pickup from the first tilt angle. Finally, all subsequent
surface thicknesses and curvatures are changed sign to account for the new mirror.
This feature may not provide useful or even correct alterations if the selected fold surface is not a plane,
standard type dummy surface in air. The dummy surface should be located at the desired fold mirror position
before using this feature. For example, to insert a fold mirror midway between two lenses 100 mm apart, a dummy
surface should be inserted between the two lenses, with the thickness before and after the dummy being set to
50 mm. The dummy surface then should be used as the fold surface.
Limitations of reversing surfaces following the fold mirror
After inserting a mirror, all subsequent surfaces must be reversed in the sense that +z becomes -z for the rays
to trace the same. For most ZEMAX surfaces, this means changing the sign of the radius of curvature, or any
other parameters that determine the sign of the surface sag. For user defined surfaces, ZEMAX cannot
automatically make this adjustment. For non-sequential component (NSC) surfaces, it may not be possible.
ZEMAX attempts to reverse NSC surfaces by changing the sign of the Z coordinate of the exit port, and rotating
all components around the y axis 180 degrees. However, this method fails if the objects inside the NSC are not
symmetric in their local XZ and YZ planes. Because the z parity of the objects must be modified, there is no
general way to do this for all NSC objects. This feature may also not work for lenses with multi-configuration data
that alters the thicknesses or glass types of any surfaces from the fold surface onward.
Delete Fold Mirror
Purpose:
Deletes an existing fold mirror, including any neighboring coordinate breaks.
Settings:
Item Description
Fold surface Selects which mirror surface will be deleted.
Discussion:
This feature deletes a fold mirror. If the mirror surface is preceded by a coordinate break with zero thickness,
the coordinate break will be deleted. If the mirror surface is followed by a coordinate break, and the mirror surface
has zero thickness, the following coordinate break will be deleted. The thicknesses of all deleted surfaces will be
added to preceding surfaces as required. All following surfaces and thicknesses will be automatically reversed

as required to follow the correct sign convention. See Limitations of reversing surfaces following the fold mirror
on page 207 for details. This feature may not provide useful or even correct alterations if the selected surface is
not a plane, standard type mirror surface in air.
Export Data
Export IGES/SAT/STEP Solid
Purpose:
Exports the current lens data to a file as a collection of 3D solids in IGES, SAT, or STEP format, with various
options. See also the Export IGES Line Work feature.
Settings:
Item Description
First/Last Surface The range of surfaces to include in the exported data.
Wavelength The wavelength number to use for any traced rays to be exported.
Field The field number to use for any traced rays to be exported.
Number of Rays The number of rays to trace; the exact meaning depends upon the "Ray Pattern".
Split NSC Rays, These settings only affect rays from NSC sources. For more information, see the
Scatter NSC Rays, discussion on the similar controls described in NSC 3D Layout on page 107.
Use Polarization
Ray Pattern Selects the type of ray set to export. This control is very similar to that defined for the
3D layout plot; see 3D Layout on page 100. The solid beam option exports a solid
volume representing the envelope of the rays. The solid beam option only works
properly if the marginal rays are all traceable and unvignetted and if these rays do not
form a caustic at any surface. The number of rays must be at least 8 to export a solid
beam, and a smoother solid results if more rays are used.
Lens/Ray Layer Selects the layers of the output file to place lens and ray data.
Config One or more configurations may be exported at once using this control.
Current will export only data from the current configuration.
Selecting a configuration number, such as 1, 2, or 3, will export data only from that
configuration.
"All By File" will export all configurations, but each to a separate file name. The file name
for each configuration is the specified save file name (the save file name is selected after
OK is pressed) with the name "_config000x" appended to indicate the data is for
configuration "x".
"All By Layer" will export all configurations, but each to a separate layer within the same
file. The first configuration will be placed in the layer specified by the lens layer and ray
layer settings. Subsequent configurations are placed in the lens and ray layers
incremented by 1 for each configuration. For example, if the lens layer is 1 and the ray
layer is 10 and there are 3 configurations, then configuration 1 will have lens data in
layer 1 and ray data in layer 10. Configuration 2 will have lens and ray data in layers 2
and 11, and configuration 3 will have lens and ray data in layers 3 and 12, respectively.
"All At Once" will export all configurations at once in a single layer. No attempt is made
by ZEMAX to avoid overlapping or redundant data export.
File Type Selects the output file format type.
Delete Vignetted If checked, rays are not included in the data export if they will be vignetted by any
surface.
Surfaces as Solids If checked, surfaces on either side of a glass material will be combined to form a single
solid object, where possible. Not all ZEMAX surface types may be combined to form a
single solid.

Item Description
Export Dummy If checked, dummy surfaces will be exported.
Surfaces
Dummy Thickness The distance in lens units for the thickness of dummy surfaces if they are automatically
converted to solids. Both "Export Dummy Surfaces" and "Surfaces as Solids" need to
be checked for this control to have any affect.
Tolerance The approximate accuracy of the exported objects in lens units. Smaller tolerances yield
more accurate representations at the expense of larger exported files and longer
computation times.
Spline Segments The number of segments to use when spline entities are exported.
Discussion:
For information on the accuracy of exported surfaces see Comments about imported objects on page 314.
The Initial Graphics Exchange Specification (IGES) is an American National Standard whose intended purpose
is to facilitate transfer of data between CAD programs. ZEMAX currently supports version 5.3 of the IGES
standard. For more information on IGES, contact U.S. Product Data Association, P. O. Box 3310, Gaithersburg,
MD 20885-3310. The SAT format is used by the ACIS geometry modeling engine developed by Spatial
Technologies. The STEP format is AP203 as defined by ANS US PRO/IPO-200-203-1994, also available from
the U.S. Product Data Association.
ZEMAX exports five types of data:
Lines: Lines are used to indicate the paths of rays traveling through the optical system. Within GRIN media,
rays are exported as a series of line entities.
Surfaces: Surfaces may be of arbitrary shape, including user defined surfaces; with any aperture shape
ZEMAX supports, including user defined apertures. Data for some types of surfaces (particularly aspheres and
toroids) which are exported as splines have an accuracy that depends on the number of spline points used;
more points yields more accuracy at the expense of larger, slower exported data files. For more information
on the limitations of the accuracy of the exported objects see the discussion below.
Lens Solid: Lens Solids include a non-zero volume enclosed by two surfaces, with the edge region formed by
the extrusion of the edges of the two surfaces. Most surfaces surrounding glass with similar shaped apertures
(e.g. both surfaces have rectangular apertures) may be exported as solids.
Faceted Solid: A Faceted Solid is defined by a collection of triangles that form a volume, such as a prism or
Fresnel lens. Non-sequential STL and POB objects, among others, are exported as Faceted Solids.
Parametric Solid: Some NSC objects, such as the Torus Volume, are exported as exact solid NURBS objects.
ZEMAX can export the surface or solid data for all sequential surfaces and non-sequential objects. ZEMAX
exports all surfaces and rays in a 3D coordinate system referenced to the global coordinate reference surface
which is described in Global Coordinate Reference Surface on page 93.
If exporting surfaces as solids, ZEMAX will export mirror and dummy surfaces as thin solids. The front and
back face will be of the shape of the surface, with a thickness equal to the dummy thickness in lens units.
Most NSC objects are exported exactly, or by using one or more spline based surfaces similar to those used
for sequential surface export. However, some objects, such as user defined objects, are exported using a faceted
approximation to the smooth surfaces. Increasing the number of facets used for rendering these objects will also
increase the resolution of the exported representation.
There is always the possibility that the conversion from the ZEMAX representation of surfaces and objects to
the CAD file format representation will cause a loss of precision. For more information, see Ray tracing accuracy
for imported objects on page 314.
This feature will prompt for the file name and path to save the exported data to. The default file is EXPORT.IGS
in the current output directory. If the file name provided ends in the extension IGS; then the output file will be in
IGES format. If the file name extension is SAT, the output will be in SAT format. If the file ends in the STP
extension, the output will be in STEP format. Otherwise, the output will be in IGES format.

To import IGES, SAT, and STEP files see Imported on page 314.
Export IGES Line Work
Purpose:
Exports the current lens data as an IGES line work format file, with various options. See also the Export IGES/
STEP Solid feature.
Settings:
Item Description
Number of Rays The number of rays to trace; the exact meaning depends upon the "Ray Pattern".
Ray Pattern Selects the type of ray set to export. This control is very similar to that defined for the
3D layout plot; see 3D Layout on page 100 for details.
Lens/Ray Layer Selects the layers of the output file to place lens and ray data.
surface.
Spline Segments The number of segments to use when spline entities are exported.
Discussion:
For information on the accuracy of exported surfaces see Comments about imported objects on page 314.
The Initial Graphics Exchange Specification (IGES) is an American National Standard whose intended purpose
is to facilitate transfer of data between CAD programs. ZEMAX currently supports version 5.2 of the IGES
standard. For more information on IGES, contact U.S. Product Data Association, P. O. Box 3310, Gaithersburg,
MD 20885-3310.
ZEMAX exports lines, spherical arcs, and splines to represent the shape and position of each surface. ZEMAX
does not export the "edge" of the lens, and makes no distinction between dummy surfaces, glass surfaces, and
mirrors. The type and number of each IGES entity exported for a given surface depends upon the surface
aperture, if any, and the symmetry, if any of the surface. The following table lists how ZEMAX determines the best
representation of the surface. ZEMAX exports all surfaces and rays in a 3D coordinate system referenced to the
global coordinate reference surface which is described in the System Menu chapter.
This feature will prompt for the file name and path to save the exported data to. The default file is EXPORT.IGS
in the current output directory. The output will be in IGES format no matter what extension is chosen.
ZEMAX makes no attempt to export objects defined as part of a Non-Sequential Components surface. See
the Export IGES/STEP Solid feature if this capability is required.

IGES ENTITIES EXPORTED BY SURFACE APERTURE AND SHAPE
Surface Aperture Surface Shape Entities Used
None, Circular Aperture or Obscuration Plano One line along the X axis, one line along
the Y axis, one circle in the XY plane at the
semi-diameter radius.
Spherical One arc in the XZ plane, one arc in the YZ
plane, one circle in the XY plane at the
semi-diameter radius.
Other One spline in the XZ plane, one spline in

the YZ plane, one circle in the XY plane at
the semi-diameter radius.
Rectangular Aperture or Obscuration Plano Three lines parallel to the X axis, three
lines parallel to the Y axis, arranged at the
edges and center of the surface to form a
grid.
Other Three splines parallel to the X axis, three

splines parallel to the Y axis, arranged at
the edges and center of the surface to form
a grid when projected upon the XY plane.
Elliptical Aperture or Obscuration Plano One line along the X axis, one line along
the Y axis, one spline defining the edge of
the surface in the XY plane.
Other One spline along the X axis, one spline

along the Y axis, one spline defining the
edge of the surface.
User Defined Aperture or Obscuration Any One line for each user defined segment.
The line entity is IGES entity 110. The circular arc entity is IGES entity 100. The spline entity is IGES entity
112. Rays are exported as line entities. Within GRIN media, rays are exported as a series of line entities. A
warning will be issued if not all surfaces within the specified range were exported.
For surfaces where both an obscuration and another shape need to be exported, use a dummy surface
collocated with the surface that has the second aperture on it. For example, to export a rectangular lens with a
rectangular obscuration on the surface, use two surfaces with a zero thickness between them; one with the
aperture and one with the obscuration. Make sure the surfaces have the same shape by setting the surface types
the same and using pick up solves where appropriate.
Export 2D DXF File
Purpose:
Exports the current lens data as a 2D DXD format file.
Settings:
Item Description
Number of Rays The number of rays to trace.

Item Description
surface.
Marginal and Chief Draws only the marginal and chief rays, overriding the other ray settings.
Only
Discussion:
This feature will prompt for the file name and path to save the exported data to. The default file is EXPORT.DXF
in the current output directory. The output will be in DXF format no matter what extension is chosen. This feature
is only available for rotationally symmetric lenses.
Miscellaneous
Reverse Elements
Purpose:
Reverses a lens element or group.
Settings:
Item Description
First Surface This is the first surface of the lens group to be reversed.
Last Surface The last surface of the lens group to be reversed.
Discussion:
The feature may not work correctly in all cases if mirrors, coordinate breaks, multi-configuration controlled
data, surface tilts and decenters, or non-standard surfaces are included in the range of surfaces. Check the
reversed system results carefully to verify that the desired result was achieved.
Tilt/Decenter Elements
Purpose:
Tilts or decenters a single surface or a group of surfaces forming an element.
Settings:
Item Description
First Surface This is the first surface of the lens group to be tilted/decentered.
Last Surface The last surface of the lens group to be tilted/decentered.
Decenter X, Y The X and Y decenters in lens units.
Tilt X, Y, Z The rotation about X, Y, and Z in degrees.
Order Either tilt then decenter, or decenter then tilt.
Discussion:
The feature inserts coordinate break and dummy surfaces as required to tilt and/or decenter a single surface
or a range of surfaces. For more information on coordinate breaks see Coordinate Break on page 239.
Scale Lens
Purpose:

Scale will scale the entire lens by the specified factor. This is useful for scaling an existing design to a new
focal length, for example. Wavelengths are not scaled. The scale lens feature may also be used to change the
units from mm to inches, or other combinations of unit types.
Settings:
Item Description
Scale by factor If selected, then a scale factor may be entered directly.
Scale by units If selected, then the lens will be converted by the selected units.
Discussion:
Scaling of data is performed based upon the units of measure for the data. If the scale factor is X, then data
measured in lens units of length will be scaled by the factor X. Data measured in units of lens units squared (such
as millimeters squared) will be scaled by X squared. Some polynomial coefficients, such as those on the Even
Aspheric surface, have units that change from term to term, and ZEMAX accounts for this when scaling the data.
Other parameters, such as the conic constant, are dimensionless, and are therefore not scaled.
ZEMAX will generally scale all lens data in the Lens Data Editor, the Extra Data Editor, and the Non-Sequential
Components Editor correctly. ZEMAX also will scale almost all of the data in the Merit Function Editor, the Multi-
Configuration Editor, and the Tolerance Data Editor correctly as well.
ZEMAX will not attempt to scale some types of data, even though scaling would be appropriate for those types
of data. These exceptions are for data items such as the multi-configuration operand PAR3, that modify general
parameter data of surfaces or objects. These data items should be scaled manually, if required.
Data items that are not scaled properly should be reported to ZEMAX technical support.
Make Focal
Purpose:
Make focal length is similar to scale lens, except the desired focal length is typed in directly. The entire lens is
then scaled to make the focal length the specified value.
Quick Focus
Purpose:
Quickly focuses an optical system by adjusting the back focal distance.
Settings:
Item Description
Spot Size Radial Focus to best RMS spot radius image plane.
Spot Size X Only Focus to best RMS x direction spot size image plane.
Spot Size Y Only Focus to best RMS y direction spot size image plane.
Wavefront Error Focus to best RMS wavefront error image plane.
Use Centroid Reference all calculations to the image centroid rather than the chief ray. This option is
slightly slower, but more appropriate for systems dominated by coma.
Discussion:
This feature adjusts the thickness of the surface prior to the image plane. The thickness chosen is that which
minimizes the RMS aberration. There are several different RMS computations that may be performed as
described in the "options" table above. The "best focus" position depends upon which criterion is selected. The
RMS is always computed as a polychromatic average over the field, using the defined fields, wavelengths, and
weights.

Convert to NSC Group
Purpose:
Converts a range of surfaces in the Lens Data Editor into a group of components in a Non-Sequential
Components surface.
Settings:
Item Description
First Surface This is the first surface of the lens group to be converted to non-sequential components.
Last Surface The last surface of the lens group to be converted.
Discussion:
This feature converts a range of sequential surfaces in the Lens Data Editor into a group of non-sequential
components.
!Dont assume this feature perfectly converts your sequential surface data; carefully check the
conversion results before doing any important analysis.
There are some restrictions on the ability of ZEMAX to convert sequential surfaces to NSC objects:
The surfaces must all be of type Standard, Even Aspheric, Toroidal, or Coordinate Break. This restriction will
be relaxed as more NSC object types and more conversion logic is added to ZEMAX.
All coordinate break surfaces must be between elements, in "air" and not imbedded within the lenses.
The surface prior to the first surface and the last surface in the selected range must both be made of air with
unity index.
The stop must precede the first surface.
ZEMAX can only convert rectangular, circular, and elliptical apertures on standard surface mirrors. Most other
combinations of non-circular apertures and surface types may not convert properly. Non-circular apertures on
standard surfaces are modeled using user defined aperture (UDA) files that ZEMAX automatically creates. For
more information on UDAs see User defined apertures and obscurations on page 69.
There are times when the ray tracing after conversion will not yield exactly the same results as before. The
possible explanations for this include the following:
Some rays may not trace the same inside a non-sequential group if some of the surfaces have central
obscurations or annular apertures. This is especially likely to occur in telescopes with obscured apertures. In
the non-sequential representation, the chief ray will either not make it through the group or will take a radically
different path.
Rays that are exactly at the edge of a sequential surface may miss the surface, or strike the outer cylinder wall
of a lens, in the non-sequential trace. For this reason, ZEMAX adds a very small increase in the diameter of
each surface when converting to a NSC group.
For systems with concave surfaces at the ends of the selected range, the edges of the surface may fall outside
of the entry and/or exit ports, which will cause ray tracing errors. The solution is to manually edit the entry/exit
port locations, or, insert dummy surfaces before and after the selected range prior to conversion.
The feature works by the following algorithm:
First a Non-Sequential Component surface and a dummy exit port surface are inserted.
The exit port is positioned to coincide with the last surface in the defined range.
The range of surfaces is converted into either NSC standard surfaces (for mirrors) or NSC standard or toroidal
lenses (for pairs of surfaces with glass in between).
Coordinate breaks are replaced with null objects with equivalent rotations; and subsequent objects will
reference the null objects.

The original surfaces are then deleted.
The feature may not work correctly in all cases. Some additional dummy surfaces with small spacings may
need to be inserted prior to and after the range of lens surfaces to be converted. The feature may also not work
for lenses with multi-configuration data that alters the thicknesses or glass types of any surfaces within the range
of surfaces to convert.
The most important point: Dont assume this feature perfectly converts your sequential surface data; carefully
check the conversion results before doing any important analysis.
Replicate Object
Purpose:
Replicates an existing object into an array of identical objects.
Settings:
Item Description
Object The object to be replicated.
Number X, Y, Z The total number of objects in the array in the X, Y, and X directions.
Delta X, Y, Z The spacing interval along each direction.
Add Pickup Solves If checked, all replicated objects will pick up parameter data from the original object.
Relative If checked, all replicated objects will use relative references to the original object,
References otherwise, absolute references are used See Reference objects on page 345.
Discussion:
This feature makes multiple copies of an existing object. Each copy is offset by a different amount to create
an array of identical objects. The original object is placed at the center of the array, and all replicated objects are
referenced to the original objects position and orientation.
Ghost Focus Generator
Purpose:
Ghost focus analysis. For a description of ghosts, see Ghost reflections on page 48.
Settings:
Item Description
Bounces Select either single-bounce or double-bounce analysis.
First Surface The first surface to consider reflections from.
Last Surface The last surface to consider reflections from.
Save Files If checked, the files used to compute the ghost ray trace are saved to disk.

Item Description
Image Plane Only If checked, only data for the image surface will be shown when computing double
bounce ghosts.
G h o s t R e f le c t o r The name of the coating, if any, to apply to surfaces that will be changed from refractive
Coating to reflective to simulate ghosting. For example, if a 1% reflective coating is applied to all
optical surfaces; then enter a coating name of "I.99". The result will be a surface that
reflects 1% of the energy; and this will aid in accurate computation of total ghost energy
using the polarization transmission analysis, if desired. This example uses an ideal
coating, but real coatings may also be used as ghost coatings if the substrates are
properly defined in the coating model.
This feature is intended for computing detailed transmission data when looking at the
generated and saved ghost files using polarization ray tracing and the transmission
feature, see Transmission on page 183.
Discussion:
!This feature may not work correctly for systems which contain coordinate breaks, multi-
configuration data, or some special surface types such as the non-sequential components
surface.
This feature generates lens files which are derived from the current lens prescription data. The generated files
are set up to reflect light at a given surface, rather than refract it. The portion of the optical system prior to the
new reflective surface is then duplicated so the rays can be traced back through. The purpose of this analysis is
to check whether or not rays reflected from any optical surface can form "ghost" images on other components or
near the focal plane. These effects are significant in high power laser systems, where focused reflections can
damage optics. Ghost images also reduce contrast. Both single and double reflections are supported.
!An alternative to this type of ghost analysis is to use the non-sequential capabilities of
ZEMAX. For details, see NON-SEQUENTIAL COMPONENTS on page 289.
For each ghost system, the marginal ray height, paraxial ray F/#, and RMS spot radius on axis are listed. Glass
surfaces which may have an internal focus are also indicated. For the image surface, when doing double-bounce
ghost analysis, the distance from the image surface to the ghost and the EFL of the ghost system are also
provided.
The lens files generated are stored in the file GHfffsss.ZMX, and these lens files can be opened for more
analysis. The fff is the number of the first ghost surface, and sss is the number of the second ghost surface. For
example GH007002.ZMX is a ghost focus file of the double bounce off surface 7 then 2. For single bounce ghosts,
only the fff number is non-zero.
Dummy surfaces (those that do not define an index boundary) and mirror surfaces are ignored as candidate
bounce surfaces. The only exception is the image surface, which is considered a potential first bounce surface
even if there is no index change on that surface.
This feature may not work correctly for systems that have coordinate breaks, or if the lens is under multi-
configuration control. This latter problem is circumvented by activating the configuration of interest, and then
deleting all multi-configuration data before executing this feature. Not all ghost reflection configurations can be
traced; there are occasionally problems with total internal reflection or rays completely missing surfaces. The
GHfffsss.ZMX file may need to be opened and modified to analyze it in detail.
If the first reflected surface is before the stop, then the entrance pupil position is incorrectly calculated as well.
This problem is easily overcome (for the purposes of analyzing the ghosts only) by using the following procedure
before you begin the ghost focus generation:
1) Record the entrance pupil position and diameter.
2) Create a dummy surface at surface one.

3) Give this new dummy surface a thickness equal to the negative of the entrance pupil position value
recorded.
4) Make the dummy surface the stop, and use an entrance pupil diameter equal to the entrance pupil diameter
recorded.
5) For finite conjugates only, increase the object surface thickness by an amount equal to the thickness of the
dummy surface.
These steps will give you a real entrance pupil in object space, and the rays will be correctly traced when
reflected off surfaces before the stop. Ghost focus analysis can be very complex for moderately complicated
systems, and care is required in interpreting the results of this analysis.
Performance Test
Purpose:
System ray trace and update performance test.
Discussion:
Performance test runs a check on the number of ray-surfaces per second and the number of system updates
per second the computer hardware/lens combination is capable of.
The ray surfaces per second performance number is measured by tracing a large number of random skew
rays through the current optical system, and then dividing the number of rays times the number of surfaces traced
through by the elapsed time in seconds.
The system updates per second is calculated by performing many system updates and then dividing the
number of system updates performed by the elapsed time in seconds. System updating includes recomputing
the pupil positions, field data (such as ray aiming coordinates), lens apertures, index of refraction, solves, and
other fundamental checks on the lens that must be performed prior to any ray tracing.
The speed will vary tremendously depending upon the system processor, clock speed, and lens complexity.
Lock/Unlock All Windows
Purpose:
Locks or Unlocks all open analysis windows at once.
Discussion:
This tool locks or unlocks all windows. For a description of the lock function, see Graphic windows operations
on page 36.
Create Polygon Object
Purpose:
Creates POB files of basic geometrical shapes which may be used as NSC detector objects.
Discussion:
Creating simple shapes as POB files is generally straightforward using a text editor. However it is often useful
to make a detector object with a large number of facets, such as a cube with hundreds of facets on each side.
When the object has many facets, it is easier to create the POB using this tool.
To create the POB file, choose the object type from the list, set the appropriate parameters, enter the name of
the file to create, and then click OK. Note the file name should not include a path, but should include the extension
POB. The tool will automatically place the POB file in the \OBJECTS directory, ready to use by the NSC editor.
!Suggestions for additional basic POB shapes should be directed to technical support.
For information on POB objects, see Polygon Object on page 319. For information on using objects as
detectors, see Objects as detectors on page 335. For information on the POB format, see Defining Polygon
Objects on page 373.

Slider
Purpose:
The slider control is used to interactively adjust any system or surface parameter while viewing any analysis
window.
Settings:
Item Description
The data to be modified is defined by the following three settings:
Group (item name Select either the surface, system, configuration, or NSC data group.
not shown)
Data (item name Select the data to modify. The available selections include radius, curvature, thickness,
not shown) conic, parameter, and extra data values if surface data are chosen. For system
parameters, the selections include the system aperture, field and wavelength data,
apodization factor, temperature, and pressure. For configuration data, all multi-
configuration operands are listed. For NSC data, object position and parameter data are
listed.
Number (item name The surface number for the data to be modified if the data is associated with a surface.
not shown) This field is used for the configuration number if configuration data is selected. If NSC
data is selected, this field is used for the object number.
Window Select either "All" or any specific analysis window to update when the slider is adjusted.
See the discussion.
Start/Stop Value The beginning and ending range of the data corresponding to the extreme limits of the
slider control.
Animate Automatically increments the data over the defined range, and updates the selected
windows in a continuous loop. Press Stop (the animate button changes to a Stop
when animation is running) to terminate the animation.
Save Saves the current data, but leaves the slider control box open.
Exit Restores the original (or last saved) data for the modified parameter and closes the
slider control.
Discussion:
The slider control can be used to adjust any surface or system parameter while monitoring how the changed
value affects the data displayed in any open analysis window or in all windows.
When the slider control is activated, a copy is made of the data to be modified. If a new type of data is selected
(for example, the surface number or parameter or surface/system setting is altered) the old data is automatically
restored, unless the "Save" button has been pressed. The Save button replaces the saved copy of the original
data with the newly modified data.
If any or all of the analysis windows require a long time to update, the slider control may act erratically. This is
due to the Windows operating system handling of the slider control requiring a very fast update while dragging
or scrolling the slider bar control. If this occurs, select a single window for updating rather than "all" or choose
analysis options to reduce computation time.

Chapter 9 REPORTS MENU
Introduction
This chapter provides detailed descriptions of each of the report features ZEMAX supports. The Settings menu
option allows the default parameters for the calculation to be changed. Selecting this option will cause a feature
dialog box to appear on the screen. The dialog box typically has five buttons: OK, Cancel, Save, Load, and Reset.
OK causes the window to recalculate and redisplay the data with the currently selected options. Cancel reverts
the options to those selected before the dialog box appeared, and does not update the data in the window. Save
stores the currently selected options as defaults for future use, and then causes the window to recalculate and
redisplay the data. Load loads the most recently saved default options, but does not exit the dialog box. Reset
returns the options to the factory default values, but does not exit the dialog box. A report window can be updated
by double clicking with the left mouse button within the window. The settings dialog box is invoked by clicking with
the right mouse button.
Surface Data
Purpose:
Display surface data.
Settings:
Item Description
Surface The surface number to display the data for.
Discussion:
This feature generates a text window which displays surface-specific data. The data includes surface and
element powers and focal lengths, edge thickness, index of refraction, and other data for the surface.
If the glass type of the surface is a model glass, then ZEMAX will list out the index of refraction at each defined
wavelength calculated from the model glass parameters. Also listed is the name of the Best Fit Glass, which is
the name of the glass in the currently loaded catalog(s) which has the closest index of refraction to the model
glass in an RMS sense. Specifically, ZEMAX computes the index error as the sum of the squares of the difference
between the model index and the actual glass index using the dispersion formulas. The sum is over the defined
wavelengths. The index error is computed for every glass in the current catalogs, and the glass with the lowest
RMS index deviation is designated as the best fit glass.
Note that the best fit glass may have a different V number than the model glass, however, this is due to the
approximations made in the model glass dispersion. Since index of refraction is the physically significant
parameter, only the index is used is making the glass selection. When changing from model glass to a real fixed
glass, the same algorithm is used to make the selection. For more information on model glasses, see Using
model glasses on page 485.
System Data
Purpose:
Display system data.
Settings:
None.
Discussion:
This feature generates a text window which lists many system-related parameters, such as pupil positions and
sizes, magnification, F/#, etc.
Prescription Data
Purpose:
219 Chapter 9: REPORTS MENU

This function generates a list of all surface data, and summarizes the lens system. This is the feature to use
for printing the contents of the Lens Data Editor.
Settings:
Item Description
General Data Includes F/#, pupil positions, magnification, etc.
Surface Data Surface type, radii, thickness, glass, semi-diameter, conic.
Surface Detail Surface Parameter data. All NSC object data is also listed in this section for NSC
groups.
Edge Thickness The X and Y edge thickness of each surface.
Multi-Config Data A table of multi-configuration operand data.
Solves/Variables Solve types and data, variables.
Index Data Index of refraction and TCE data for each wavelength/surface.
Global Vertex Global coordinates of the vertex and the rotation matrix for each surface. See
Global Coordinate Reference Surface on page 93 for a discussion of this data.
Element Volume Volume in cc, density, and mass. See below.
F/ Numbers Lists working F/# for each field position and wavelength combination.
Cardinal Points Lists the locations of focal, principal, anti-principal, nodal, and anti-nodal points.
POP Settings Lists the surface by surface settings used by the Physical Optics Propagation
feature.
Discussion:
This report lists specification data, indices of refraction, global coordinates, element volumes, and more. It is
suitable for describing the lens prescription.
Comments on computing element volumes
When ZEMAX computes element volumes for spherical or plane standard surfaces with circular edges, the
edge of the surface with the smaller diameter is assumed to be squared up to the semi-diameter of the larger
surface. The volume are exact for this common special case.
For other elements, with possibly aspheric faces, elliptical, annular, or rectangular apertures and/or decentered
apertures, a numerical integration technique is used that yields approximate volumes to an accuracy of about
0.1%. If zero volume is reported than the algorithm is unable to compute the volume to sufficient accuracy. This
happens when the aperture types are not the same, or have different sizes or decenters.
When computing the density of elements, the density in grams per cubic centimeter for a catalog glass is
retrieved from the glass catalog. For gradient index surfaces, ZEMAX assumes the density is 3.6 g/cc, which may
not be a good estimate.
Report Graphics 4/6
Purpose:
This function generates a graphical window that simultaneously displays either 4 or 6 analysis graphics. The
primary advantage of this feature is that multiple analysis graphs may be printed on a single page, making a
suitable summary for reports, archival documentation, or promotional literature.
Settings:
The report graphics window works somewhat differently from other analysis windows. If the Settings option
is selected from the window's menu bar, then a dialog box will appear which allows selection of the type of graphic
Chapter 9: REPORTS MENU 220

to be displayed in each position within the window. The selected graphs to display can be saved as default
selections like any other window. The configuration number for each window may also be selected.
Additionally, the settings may be saved as a custom report for later re-use. The Save As New Report button
will save the current settings in a file. Once the settings are saved, the name will appear as a menu option for
quick regeneration of the plot. To change the settings for an individual graph within the window, the right mouse
button must be used. First, unzoom the display (if the graph has been zoomed), then right mouse click anywhere
in the window within the region of the plot whose settings need changing.
221 Chapter 9: REPORTS MENU

Chapter 9: REPORTS MENU 222
Chapter 10 MACROS, EXTENSIONS, AND WINDOWS MENUS
Edit/Run ZPL Macros

Purpose:
Runs ZEMAX Programming Language macros. This option invokes a dialog box which allows editing, viewing,
and executing of macros.
Discussion:
See the chapter ZEMAX Programming Language for details on the ZPL macro language.
This dialog box is most useful for developing and debugging new macros. It also displays a terminate button,
which allows the macro execution to be terminated.
Refresh Macro List
Purpose:
Updates the macro list.
Discussion:
This feature updates the macro list; which may be required if any macros have been added or deleted since
the last time the list was refreshed.
Macro Names
Purpose:
A list of all ZPL macros in the default directory for macros is shown. By clicking on the macro name, it
immediately executes.
Discussion:
It is much faster to execute a macro from the list of macro names than to use the ZPL Macros dialog box. To
terminate a ZPL macro, press the escape key.
Extensions
Purpose:
Runs ZEMAX extensions.
Discussion:
See ZEMAX EXTENSIONS on page 607 for details on creating ZEMAX extensions. The extensions supplied
with ZEMAX are described in the following table.
EXTENSIONS INCLUDED WITH ZEMAX
Extension Name Description

ArrayDemo Creates a table of rays traced for the current lens. This example program includes
source code and illustrates the use of array ray tracing commands.
DDE_DEMO Creates a text listing of ray and system data. This example includes source code
and illustrates the basic use of DDE features.
PhasePlot Creates a graph showing the phase and the inverse of the phase slope for Binary
2 surfaces. The plot indicates the phase over the semi-diameter of the surface.
Pupil Map Maps an x-y grid of rays from the entrance pupil or a ray-aimed stop surface to
another surface in the system.
223 Chapter 10: MACROS, EXTENSIONS, AND WIN-

Extension Name Description
SagCalculation Creates an extended listing of sag data for non-radial symmetric surfaces.
TransmissionPlot Computes transmission vs. Wavelength.
Refresh Extensions List

Purpose:
Updates the list of extensions.
Discussion:
This feature updates the list of extensions that appears in the Extensions menu; which may be required if any
extensions have been added or deleted since the last time the list was refreshed. Any new extensions must be
placed in the \Extend subdirectory of the main ZEMAX directory.
Extension Names
Purpose:
A list of all ZEMAX extensions in the \ZEMAX\Extend directory is shown. Clicking on the extension name will
execute that extension.
Discussion:
See ZEMAX EXTENSIONS on page 607 for details on creating and executing ZEMAX extensions.
Listing of open windows
Purpose:
This menu lists all currently open windows by window title. Clicking on any window title listed will bring that
window to the front of the display and give that window the input focus.
Discussion:
Clicking on any window title listed will bring that window to the front of the display and give that window the
input focus.
Chapter 10: MACROS, EXTENSIONS, AND WIN- 224

Chapter 11 SURFACE TYPES
Introduction
ZEMAX models many types of optical components. These include conventional spherical glass surfaces, plus
aspheres, toroids, cylinders, and others, ZEMAX can also model components such as diffraction gratings, binary
optics, Fresnel lenses, holograms, and others.
Because of the large number of surface types ZEMAX supports, a conventional spreadsheet arrangement for
the user interface would be difficult to use. For example, there is no need to have a column for diffraction order
for a surface that has no diffractive capability. To make the user interface as uncluttered as possible, ZEMAX uses
different surface types to indicate what kinds of data are needed to define that type of surface.
Parameter data
A Standard surface can be a plane, spherical, or conic aspheric surface which is followed by a homogeneous
material (such as air, mirror, or glass). The only parameters required are a radius (which may be infinity to yield
a plane), a thickness, a conic constant (the default zero value indicates a sphere), and the name of the glass type.
Other surface types use these same basic data, as well as other values. For example, the Even Asphere
surface uses all the Standard surface column data plus additional values which describe coefficients on a
polynomial. These values are called parameters. The most important property of the parameter values to
understand is that their meaning changes depending upon the surface type selected. For example, the Even
Asphere surface uses parameter 1 to specify the coefficient on the parabolic aspheric term. However, the Paraxial
surface uses parameter 1 to specify the focal length of the surface. Both surface types use parameter 1, but for
different purposes, since both surface types are never used on the same surface at the same time.
This sharing of data storage simplifies the ZEMAX interface, as well as reduces the total memory required to
run the program. After you change a surface from Standard to some other surface type, ZEMAX will automatically
change the column headings on the parameter columns to reflect what each parameter does on that surface. As
you move the cursor from cell to cell, the column headings will always show you what that cell is used for. If the
current surface does not use the parameter column, the column heading will display "Unused". For more
information see Lens Data on page 65.
Extra data
Some surfaces cannot be described by only a few parameter values. For example, the Binary Optic 1 surface
type requires nine parameters as well as up to several hundred additional numbers. This would make for a very
large spreadsheet, and so a separate editor is used for the extra data values. However, the concept is the same.
The extra data values are shared by each surface type, and change meaning depending upon the surface type
selected. The column headings in the Extra Data Editor also change as the cursor is moved from surface to
surface.
Summary of surface types
ZEMAX models planes, spheres, and conics; all of these surface types are grouped under the category of
standard surface. By double clicking on the Surface Type column other surface models can be selected. A pop-
up screen will appear which lists all of the available surface types which can be selected. ZEMAX supports many
different types of surfaces in addition to the standard surface.
User defined surfaces
No matter how many surfaces are added to ZEMAX, there always seems to be the need to add another surface
type to solve a particular design, modeling, or tolerancing problem. If the surface type needed for a problem is
not already included with ZEMAX, it is fairly easy to add new surface types using the User Defined surface
described on page User Defined on page 275. User defined surfaces are created by writing software that defines
the properties about the surface, and then dynamically linking the software into ZEMAX.
If you have a need for a custom surface type, and you do not wish to write the software for it yourself, please
contact ZEMAX Technical Support for a quote on developing a custom surface to meet your requirements. We
225 Chapter 11: SURFACE TYPES

have considerable experience in developing ray tracing algorithms, and can generally write user defined surface
code for a small fee, often on very short notice.
Surface types
The types of surfaces supported by ZEMAX are summarized in the following table. The most common surface
type is the Standard surface described on page 270. For types not listed, see User Defined on page 275.
SUMMARY OF SURFACE TYPES
Surface Model Description Page

ABCD Uses ABCD matrix to model a black box. 228
Alternate Even Even Asphere surface with alternate solution selected. 228
Alternate Odd Odd Asphere surface with alternate solution selected. 229
Atmospheric Refraction Refraction caused by looking through Earth's atmosphere. 229
Biconic A conic asphere in X and Y independently. 229
Biconic Zernike A biconic surface with x, y, and Zernike polynomial terms added. 230
Binary Optic 1 Uses 230 term polynomial to define phase. 230
Binary Optic 2 Uses radial polynomial to define phase. 232
Binary Optic 3 Dual zone aspheric and diffractive surface.
Birefringent In/Out For modeling uniaxial crystals; supports tracing of rays in ordinary 235
or extraordinary mode.
Conjugate Defines surface with perfect imaging at two points. 238
Coordinate Break Permits rotation and decentration. 239
Cubic Spline Rotationally symmetric fit to 8 points. 240
Cylindrical Fresnel Polynomial cylindrical Fresnel on a polynomial cylindrical 240
surface.
Diffraction Grating Ruled grating on standard surface. 241
Elliptical Grating 1 Elliptical grating with aspheric terms and polynomial grooves. 242
Elliptical Grating 2 Elliptical grating with aspheric terms and grooves formed by tilted 243
planes.
Even Asphere Standard surface plus polynomial asphere terms. 243
Extended Asphere Uses radial polynomial to define sag. 244
Extended Cubic Spline Rotationally symmetric fit of up to 250 points. 245
Extended Fresnel Polynomial Fresnel on a polynomial surface. 245
Extended Odd Asphere Uses odd terms of radial powers. 246
Extended Polynomial Uses 230 term polynomial expansion to define sag. 247
Extended Toroidal Grating An aspheric toroidal grating with extended polynomial terms 248
Fresnel Plane surface with refractive power. 249
Generalized Fresnel XY polynomial Fresnel on an aspheric substrate. 249
Chapter 11: SURFACE TYPES 226

Gradient 1 Radial gradient index material surface. 250
Gradient 2 Radial gradient index material surface. 251
Gradient 3 Radial and axial gradient index material surface. 251
Gradient 4 X, Y, and Z gradient index material surface. 252
Gradient 5 Radial and axial gradient index material surface with dispersion 252
model.
Gradient 6 Radial gradient index material surface with Gradient Lens Corp. 253
dispersion model.
Gradient 7 Spherical gradient profile. 254
Gradium Axial gradient index material surface with dispersion model. 255
Gradient 9 Radial gradient index material surface with NSG SELFOC lens 257
dispersion model.
Gradient 10 Y gradient index material surface with dispersion model. 258
Grid Phase A phase surface described by a grid of points. 258
Grid Sag A surface shape described by a grid of points. 259
Hologram 1 Two-point optically fabricated hologram. 261
Hologram 2 Two-point optically fabricated hologram. 262
Irregular A standard surface with decenter, tilt and other deformations. 262
Jones Matrix General Jones Matrix for modifying polarization state. 263
Non-Sequential Components For tracing rays non-sequentially through a collection of 3D 275
surfaces and objects.
Odd Asphere Standard surface plus polynomial asphere terms. 263
Odd Cosine Odd Asphere surface plus cosine polynomial terms. 263
Optically Fabricated Hologram Optically fabricated hologram with arbitrary construction optics 264
and elliptical substrate.
Paraxial Thin lens surface, has ideal behavior. 266
Paraxial XY Thin lens with separate specification in X,Y. 267
Periodic Cosine shaped surface. 267
Polynomial Polynomial expansion in x and y. 268
Radial Grating A diffraction grating with radial phase profile 268
Radial NURBS Uses NURBS curve to define a rotationally symmetric surface. 269
Retro Reflect Retro reflects rays back along incident path. 240
Standard Includes planes, spheres, and conics. 270
Superconic Superconic asphere with fast convergence. 271
Tilted Defines a tilted surface without changing coordinate systems. 239

Toroidal Conic and aspheric toroids and cylinders. 272
Toroidal Grating Ruled grating on a conic toroid. 240
Toroidal Hologram Toroidal substrate with two point optically fabricated hologram. 273
Toroidal NURBS Uses NURBS curve to define a toroidally symmetric surface. 274
User Defined A general surface which uses an arbitrary user defined function 275
to describe the refractive, reflective, diffractive, transmissive, or
gradient properties of the surface.
Variable Line Space Grating Variable line space grating surface. 282
Zernike Fringe Phase Uses 37 Zernike Fringe polynomials to define phase. 282
Zernike Fringe Sag Uses 37 Zernike Fringe polynomials to define sag. 283
Zernike Standard Phase Uses up to 231 Zernike Standard polynomials to define surface 284
phase.
Zernike Standard Sag Uses up to 231 Zernike Standard polynomials to define surface 285
sag.
Zone Plate Fresnel Zone Plate model using annular rings of varying depth. 286
ABCD
The ABCD surface provides a powerful method for modeling "black box" optical systems. If you have a lens
(or a complete optical system) which is only a subsection of what you wish to model, and you do not have
prescription data for the individual components, you can still model the behavior to first order.
The ABCD surface accepts eight parameters: Ax, Bx, Cx, Dx, Ay, By, Cy, and Dy. These are used to form two-
by-two matrices (one for the x-direction and one for the y) which are used to alter the ray as it crosses the surface.
The exit ray is related to the incident ray by
x' Ax Bx x
= ,
'x Cx Dx
with a similar expression for the y components. Because there is no reliable way to compute the phase through
an ABCD surface, any calculation that requires OPD data, such as OPD fans, MTF, and Zernike coefficients, will
not be supported if an ABCD surface is present in the lens description.
PARAMETER DEFINITIONS FOR ABCD SURFACES
Parameter 1 Parameter 2 Parameter 3 Parameter 4 Parameter 5 Parameter 6 Parameter 7 Parameter 8
Ax Bx Cx Dx Ay By Cy Dy
Alternate Even
There are two solutions to the ray-surface intercept equations used in tracing a ray to the next optical surface,
if that surface is of the form of the standard or even aspheric surface types. ZEMAX selects the correct solution
in the vast majority of cases. However, in certain systems, so-called "strange" rays are actually intended to
intersect the next surface at the other, "alternate" solution. Strange rays are most likely to occur after a grazing
incidence reflection in which the ray is still traveling in the same direction (the Z component of the ray vector does
not change sign). The alternate even surface model is identical to the even asphere surface model, except the

alternate solution is used. ZEMAX may not be able to correctly compute the optical path difference when the
alternate even surface is being used.
Alternate Odd
There are two solutions to the ray-surface intercept equations used in tracing a ray to the next optical surface,
if that surface is of the form of the standard or even aspheric surface types. ZEMAX selects the correct solution
in the vast majority of cases. However, in certain systems, so-called "strange" rays are actually intended to
intersect the next surface at the other, "alternate" solution. Strange rays are most likely to occur after a grazing
incidence reflection in which the ray is still traveling in the same direction (the Z component of the ray vector does
not change sign). The alternate odd surface model is identical to the odd asphere surface model, except the
alternate solution is used. ZEMAX may not be able to correctly compute the optical path difference when the
alternate odd surface is being used.
Atmospheric Refraction
This surface is used to simulate the effects of refraction through the Earth's atmosphere when viewing a star
or point source. The atmosphere has a small but non-zero dispersion which introduces a tilt term to the incoming
wavefront which depends upon wavelength. ZEMAX uses a model based upon those described in the following
publications:
P. K. Seidelmann, Ed., "Refraction - Numerical Integration", Section 3.281, Explanatory Supplement to the
Astronomical Almanac, pp. 141-143, University Science Books, Mill Valley, 1992.
C. Y. Hohenkerk and A. T. Sinclair, NAO Technical Note 63, Royal Greenwich Observatory Science and
Engineering Research Council, 1985.
Six parameters are supplied to the model: The observed zenith angle of the source in degrees, the height of
the observer above sea level in meters, the ambient temperature at the observer in Kelvin, the atmospheric
pressure at the observer in millibars, the relative humidity (a number between 0.0 and 1.0), and the latitude of the
observer in degrees.
ZEMAX computes the atmospheric refraction angle in radians for all defined wavelengths, then subtracts from
all wavelengths the amount of refraction at the primary wavelength; thereby using the primary wavelength as a
reference. To disable the referencing to the primary wavelength, set the "Absolute" flag to 1. Atmospheric
refraction manifests itself as a small tilt in the OPD fan plot, or a slight chief ray offset similar to lateral color in the
ray fan plot. All refraction is assumed to occur in the y direction only.
This surface models color separation. To model atmospheric turbulence, considering the Grid Phase surface
with an externally provided data file. For information on the Grid Phase surface, see Grid Phase on page 258.
PARAMETER DEFINITIONS FOR ATMOSPHERIC REFRACTION SURFACES
Parameter 1 Parameter 2 Parameter 3 Parameter 4 Parameter 5 Parameter 6 Parameter 7
Zenith Height Temperature Pressure Humidity Latitude Absolute
Biconic
The biconic surface is similar to a toroidal surface, except the conic constant and base radius may be different
in the X and Y directions. The biconic surface allows specification of Rx, Ry, Kx, and Ky directly. The sag of a
biconic is given by:
2 2
cx x + cy y
z = ---------------------------------------------------------------------------------------- ,
2 2 2 2
1 + 1 ( 1 + k x )c x x ( 1 + k y )c y y
where
1 1
c x = ----- , c y = ----- .
Rx Ry

The radius in the x direction is set in the parameter 1 column. If set to zero, the x radius is interpreted to be
infinity.
PARAMETER DEFINITIONS FOR BICONIC SURFACES
Parameter 1 Parameter 2
Rx kx
Biconic Zernike
The biconic Zernike surface is similar to a biconic surface, with the added capability to add X, Y, and Zernike
polynomial deformations. The sag of a biconic Zernike is given by:
2 2 16 16 N
cx x + cy y
z = ---------------------------------------------------------------------------------------- + i x + i y + A i Z i ( , ) ,
i i
2 2 2 2
1 + 1 ( 1 + k x )c x x ( 1 + k y )c y y i = 1 i=1 i=1
where
1 1
c x = ----- , c y = ----- ,
Rx Ry
and the Z i terms are the Zernike standard terms as described in Zernike Standard Sag on page 285.
Parameter 0 is the extrapolate flag. If the extrapolate flag is set to 0, the Zernike terms are ignored outside of the
normalization radius. If the extrapolate flag is set to 1, then the Zernike terms are considered no matter where
the ray lands on the surface; even if the ray lands beyond the normalization radius. The radius and conic in the
x direction are set in the parameter 1 and 2 columns. If the radius is set to zero, the x radius is interpreted to be
infinity. Parameter 3 is the maximum number of Zernike terms, this value may be between 0 and 210, inclusive.
Parameter 4 is the normalization radius used in the Zernike expansion.
PARAMETER DEFINITIONS FOR BICONIC ZERNIKE SURFACES
Parameter 0 Parameter 1 Parameter 2 Parameter 3 Parameter 4
Extrapolate Rx kx N Normalization Radius
EXTRA DATA DEFINITIONS FOR BICONIC ZERNIKE SURFACES
Extra Data Number Description

1-16 Coefficients on powers of x
17-32 Coefficients on powers of y
33-242 Coefficient on the Zernike Standard polynomials
Binary Optic 1
Binary optics, also called kinoforms, are similar to holograms and diffraction gratings in that small grooves or
lines across the optical surface impart a change in phase of the wavefront passing through the surface. ZEMAX
does not model the wavelength-scale grooves directly. Instead, ZEMAX uses the phase advance or delay
represented by the surface locally to change the direction of propagation of the ray. Other effects, such as
efficiency or multiple order diffraction are ignored. Binary surfaces can have zero thickness, with no index change
across the surface, or may have different media on either side of the surface. In this case, the refraction due to

the material change will be accounted for as well as the diffraction effects. Full exact polychromatic ray and OPD
tracing is supported.
The binary optic 1 surface is similar to the extended polynomial surface, except the polynomial terms represent
the variation in phase (rather than surface height) across the optic surface. The coefficients therefore have units
of radians rather than lens units. The shape of the binary optic 1 surface is identical to the even asphere surface
type; planes, spheres, conics, and polynomial aspheres up to order 16 are supported. The sag of the surface is
given by the following expression:
2
cr 2 4 6 8 10 12 14 16
z = -------------------------------------------------- + 1 r + 2 r + 3 r + 4 r + 5 r + 6 r + 7 r + 8 r ,
2 2
1 + 1 ( 1 + k )c r
where the terms are identical to those in the even asphere surface model. See Even Asphere on page 243
for a complete discussion. The binary optic 1 surface adds phase to the ray according to the following polynomial
expansion:
= M A i E i ( x, y )
i=1
th
where N is the number of polynomial coefficients in the series, A i is the coefficient on the i extended
polynomial term, and M is the diffraction order. The polynomials are a power series in x and y, as described in
Extended Polynomial on page 247. The coefficients A i all have units of radians ( 2 radians is one wave). If
Parameter 9 is any value other than zero, then the absolute value of the x and y coordinates are used for
computing the phase in the above expression. This "Absolute" option yields a different, generally discontinuous
phase profile.
PARAMETER DEFINITIONS FOR BINARY OPTIC 1 SURFACES
Param 0 Param 1-8 Param 9
M, the diffraction order 1 - 8 Absolute? (see discussion)
EXTRA DATA DEFINITIONS FOR BINARY OPTIC 1 SURFACES

1 Maximum term number.
2 Normalization radius. All ray-intercept points are divided by this number to determine
the x and y coordinates for polynomial evaluation.
3-232 Polynomial terms.
233 and above Not used.
The "Maximum term number" is used to specify the maximum polynomial term to be used in calculating the
surface phase. This number is provided to speed the ray tracing calculation, as terms beyond this number are
3
ignored. For example, if the last term in the series you wish to use is xy , which is term number 13, then specify
"13" in the maximum term number column. Note that the term number is 13, because it is the 13th term in the
polynomial expansion, and NOT 15, which is the "extra data number", which is the position in the extra data list
of parameters. The "extra data number" is always two greater than the term number!

!The "extra data number" is always two greater than the term number!
The normalization radius scales the X and Y intercept coordinates of the ray so that the polynomial terms are
all dimensionless and the coefficients are all in radians. For modeling phase surfaces that are not well described
by polynomials, see Grid Phase on page 258.
Binary optic coefficients sign conventions
It is important to understand the binary optic surface phase coefficients sign convention. Imagine a collimated
beam being focused by an infinitesimal thin binary optic surface. If the collimated beam becomes a focused,
converging beam on the other side of the binary optic, then the path length of a marginal ray is longer than the
path length of the axis ray. Therefore, the binary optic must add a negative path length to the marginal ray. Using
this convention, a positive power binary optic has a negative quadratic phase coefficient. Although the sign
convention chosen by ZEMAX is arbitrary, it may be important to know what the convention is as far as fabrication
decisions are concerned. It is always a good idea to test a few cases to verify the sign convention being used by
the software before committing a design to fabrication.
Binary Optic 2
Binary optics are similar to holograms and diffraction gratings in that small grooves or lines across the optical
surface impart a change in phase of the wavefront passing through the surface. ZEMAX does not model the
wavelength-scale grooves directly. Instead, ZEMAX uses the phase advance or delay represented by the surface
locally to change the direction of propagation of the ray. Other effects, such as scattering, efficiency, or multiple
order diffraction are ignored. Binary surfaces can have zero thickness, with no index change across the surface,
or may have different media on either side of the surface. In this case, the refraction due to the material change
will be accounted for as well as the diffraction effects. Full exact polychromatic ray and OPD tracing is supported.
The binary optic 2 surface is similar to the extended asphere surface, except there are additional polynomial
terms which represent the variation in phase (rather than surface height) across the optic surface. The coefficients
therefore have units of radians rather than lens units. The shape of the binary optic 2 surface is identical to the
even asphere surface type; planes, spheres, conics, and polynomial aspheres up to order 16 are supported. The
sag of the surface is given by the following expression:
2
cr 2 4 6 8 10 12 14 16
z = -------------------------------------------------- + 1 r + 2 r + 3 r + 4 r + 5 r + 6 r + 7 r + 8 r ,
2 2
1 + 1 ( 1 + k )c r
where the terms are identical to those in the Even Asphere Surface model. See Even Asphere on page 243
for a complete discussion. The Binary Optic 2 surface adds phase to the ray according to the following polynomial
expansion:
= M Ai
2i
i=1
th
where N is the number of polynomial coefficients in the series, A i is the coefficient on the 2i power of ,
which is the normalized radial aperture coordinate, and M is the diffraction order. The first extra data term is N ,
480
the number of terms, which can be zero to exclude all binary effects, or any integer up to 240 (or the term).
Extra data value number 2 is the normalization radius. Extra data values 3 through 242 are the coefficients.
Param 0 Param 1 Param 2 Param 3 Param 4 Param 5 Param 6 Param 7 Param 8
M 1 2 3 4 5 6 7 8


3 2
Coefficient in radians on
4 4
n+2 2n
surface phase. This number is provided to speed the ray tracing calculation, as terms beyond this number are
14
ignored. For example, if the last term in the series you wish to use is , which is term number 7, then specify
polynomial expansion, and NOT 9, which is the "extra data number", which is the position in the extra data list of
parameters. The "extra data number" is always two greater than the term number!
all dimensionless and the coefficients are all in radians. For modeling phase surfaces that are not well described
by polynomials, see Grid Phase on page 258.
See Binary Optic 1 on page 230 for a discussion of sign conventions.
Binary Optic 3
The Binary Optic 3 surface is very similar to the Binary Optic 2 surface. The key difference is that the Binary
Optic 3 supports two concentric radial zones, with independent radius, conic, and polynomial aspheric deforma-
tion and diffractive phase data for each zone. The surface is divided into two zones by two radial coordinates,
A 1 , and A 2 . The inner radial zone extends from the center of the surface to the radial coordinate A 1 . The outer
radial zone extends from A 1 outward. The radial coordinate A 2 is used for normalizing the phase coefficients in
the outer zone, even though the surface may extend past the coordinate A 2 . The outer zone is offset from the
inner zone to make the surface sag continuous across the zone boundary. ZEMAX requires that 0 < A 1 < A 2 .
The sag of the surface in the inner zone is given by the following expression:
2 N
c1 r
- + 1i r 2i , for r A 1 ,
z 1 = ------------------------------------------------------
2 2
1 + 1 ( 1 + k 1 )c 1 r i=1
where N is the number of aspheric terms which may be set by the user as described below. The value for the
inner zone curvature, c 1 , is the reciprocal of the radius of curvature specified in the Lens Data Editor. The inner

zone conic constant is also set in the Lens Data Editor, in the usual conic column. A similar expression with
different coefficients and an offset value is used to define the sag of the surface in the outer zone:
2 N
c2 r
- + 2i r 2i
z 2 = z o + ------------------------------------------------------ ,for r > A 1 ,
2 2
1 + 1 ( 1 + k 2 )c 2 r i=1
where the term z o is chosen to make the surface continuous across the boundary between the inner and outer
zones at the radial coordinate A 1 , or z o = z 1 ( A 1 ) z 2 ( A 1 ) (the value of z o is temporarily set to zero while
evaluating z 2 for this calculation). The outer zone radius of curvature (from which the value c 2 is computed) and
the outer zone conic are set in the parameter data as described below. For a flat outer zone radius, use zero.
Both the inner and outer zones have a diffractive phase profile, with independent coefficients. The phase of
the inner zone is given by:
N
r
1 = M 1 1i 1 , and 1 = ------ ,
2i
A1
i=1
th
where N is the number of polynomial coefficients in the series, 1i is the coefficient on the 2i power of 1 ,
which is the normalized radial aperture coordinate, and M 1 is the diffraction order. A similar expression describes
the phase for the outer zone:
N
r
2 = o + M 2 2i 2 , where 2 = ------ .
2i
A2
i=1
The phase offset serves a similar purpose to the sag offset, and is defined by o = 1 ( 1 ) 2 ( A 1 A 2 ) (the
value of o is temporarily set to zero while evaluating 2 for this calculation). Note that the normalization radius
A 2 is used only for defining the radial aperture to normalize the phase coefficients for the outer zone. The outer
zone of the surface, and the associated phase profile, may extend beyond this value.
The dual zone nature of this surface creates a complication when computing the phase of the surface as the
zone boundary is crossed. The phase offset value o makes certain the phase is continuous across the zone
boundary. This is desirable for design and analysis purposes, because phase jumps of hundreds of waves make
interpretation and analysis difficult. However, the phase offset is artificial, and this must be accounted for in the
actual design. For optimum imaging properties, the outer zone should be in phase with the inner zone. This
condition is met when the inner and outer zones differ in phase by an integral number of wavelengths at the
boundary, or more to the point, if o = J2 , where J is some arbitrary integer. This condition can usually be met
by a small change in A 1 , as long as there is some difference in the slope of the phase on either side of the
boundary. To make this boundary condition simple to meet, ZEMAX computes sin o and places this value in
parameter 7. The merit function boundary operand PMVA can then be used to target this value to be zero. Note
this value is computed from the phase data, and should not be user defined or made variable.

Param 1 Param 2 Param 3 Param 4 Param 5 Param 6 Param 7
R2 k2 A1 A2 M1 M2 sin o

1 Maximum term number, N.
2, 3, 4, 5 The coefficients 11 , 11 , 21 , and 21 if N >= 1.
6, 7, 8, 9 The coefficients 12 , 12 , 22 , and 22 if N >= 2.
10, 11, 12, 13, etc. The pattern continues in groups of 4 coefficients for N groups.

Birefringent In and Birefringent Out
This pair of surfaces models uniaxial crystals, such as calcite. These types of crystals are described by a
crystal axis; which defines the axis of symmetry for the material, and two dispersion curves; one defining the
"ordinary" index and the other the "extraordinary" index. Such materials refract rays differently depending upon
the polarization state of the ray and the angle the ray makes with respect to the crystal axis. These different types
of refraction yield two different possible refraction angles for any specific ray, and so they are called birefringent
or double refraction materials. For details on birefringent materials, see Saleh and Teich, Fundamentals of
Photonics, Wiley Interscience. For information on ray tracing details, see Quan-Ting Liang, "Simple ray tracing
formulas for uniaxial optical crystals", Applied Optics Vol. 29, No. 7, (1990). Only a brief description of these
materials and their ray tracing properties is given here. Note that when rays propagate close to parallel to the
crystal axis, some crystal materials, such as quartz, may exhibit an effect called optical activity that ZEMAX does
not model. Optical activity may be accounted for using the Jones Matrix surface, see Jones Matrix on page 263.
Birefringent materials refract rays according to Snells law, but the effective index of refraction in the media
depends upon the input polarization state and the angle the refracted ray makes with respect to the crystal axis.
Ordinary rays are refracted according to
n sin = n o sin ' ,
where the subscript "o" indicates the ordinary index. This is just Snells law, of course. For extraordinary rays,
the refraction law is
n sin = n ( w ) sin ' .
This is also Snells law, but the effective index of refraction in the birefringent material is a function of the angle
w , which is the angle between the crystal axis vector a and the refracted wave vector k . Additionally, the ray
vector s , which is the vector pointing in the direction of energy propagation (also called the Poynting vector), does
not follow the wave vector k , but makes a small angle with respect to k . In isotropic media k and s are the same,

so for most optical design we keep track of k ; here we must consider the ray and the wave vector being different.
The angle w is defined by
cos w = k a .
The effective index of refraction is defined by
-------------
1 - 2 cos w 2 sin w 2
= --------------- + -------------- ,
n ( w ) no ne
where n o is the ordinary and n e is the extraordinary index of refraction.
The angle between k and s is defined by
cos = k s , where
2 2
( n e n o ) tan w
tan = --------------------------------------
2
-,
2
n e + ( n o tan w )
and the vectors k and s are both coplanar with the crystal axis vector a . The wave vector k points along the
normal to the wavefront, while s points along the direction of energy propagation. For ray tracing purposes,
ZEMAX uses the components of s as the ray direction cosines.
Because the birefringence effects both the refraction into the media, as well as the refraction out of the media,
two surfaces are required to model a birefringent element: the "birefringent in" and the "birefringent out" surfaces.
Each "in" must be followed by an "out" surface; if this is violated ZEMAX will issue an error message and not trace
any rays. The only exception is that any number of coordinate break surfaces (or surface tilts and decenters) may
be placed between the "in" and "out" surfaces. The ray is propagated to the next surface using the s vector
components; however the refraction out of the birefringent media is governed again by the k vector components.
Defining the index
To define the ordinary index, ZEMAX uses the glass catalog and glass name in the usual way. For example,
to use Calcite, define the surface to be a birefringent "in" type, and enter Calcite for the glass name. ZEMAX uses
this name to compute the ordinary index at any defined wavelength.
To compute the extraordinary index, ZEMAX appends "-E" to the glass name, and then looks for this material
in the glass catalog. For the case of Calcite, ZEMAX searches for "Calcite-E" in the glass catalog, and if found,
uses that for the extraordinary index. If not found, an error message is issued. This technique allows definition of
any material with any ordinary and any extraordinary index to be defined and used.
Most ZEMAX calculations, such as EFL, EPD, F/#, etc. use the ordinary index of refraction. The extraordinary
index is only used to trace extraordinary rays. The Prescription report will list both the ordinary and the
extraordinary index, and this data should be carefully checked before using this feature for any critical analysis.
Defining the crystal axis orientation
The crystal axis orientation is defined using parameter columns 2, 3, and 4 for the x-, y-, and z- direction
cosines of the crystal axis ray. For example, to define the crystal axis along the x axis, the vector values are (1,
0, 0). For the axis aligned to the surface vertex normal, which is along the z axis, the vector is (0, 0, 1). Note the
coordinates are local to the surface; and ZEMAX converts to unit vector components.

Determining which ray is traced
ZEMAX will trace the ordinary ray if the mode is set to 0 or 2. If the mode is set to 1 or 3, the extraordinary ray
will be traced. The mode is parameter 1 in the lens data editor. ZEMAX cannot trace both rays at once, however
it is easy to create a multi-configuration lens with the mode set to 0 in configuration 1, and to 1 in configuration
2; this allows inspection of both possible paths as well as simultaneous optimization and layouts of the traced
rays. The difference between modes 0 and 2 for ordinary rays and modes 1 and 3 for extraordinary rays is
described in the next paragraph.
Accounting for phase rotation
To account for phase rotation when performing polarization ray tracing, there are two limiting cases to consider.
Conceptually, the beam splits into an ordinary beam and an extraordinary beam with a small angle between the
two beams. If the beams then propagate a large distance, the ordinary and extraordinary beams will separate
and become two distinct beams. If however the propagation distance is short, the ordinary and extraordinary
beams coherently interfere and recombine, resulting in a single beam with a (usually) rotated polarization vector.
Conceptually this is like tracing both the ordinary and extraordinary rays, then coherently recombining them after
propagation through the birefringent media. Which model is used to determine the transmission and polarization
properties may be selected by the mode.
If the mode is 0 or 1, then only the ordinary or extraordinary part, respectively, is traced. The fraction of ray
energy in the part of the ray that is not traced is discarded. No polarization rotation will be modeled using mode
0 or 1. For transmission computations, both rays need to be traced separately and the total energy computed. It
is difficult to compute the effects of phase rotation using mode 0 or 1.
If the mode flag is 2, then the ordinary ray is traced, however the phase rotation due to the extraordinary ray
is accounted for. If the mode is 3, the extraordinary ray is traced, and the phase rotation due to the ordinary ray
is accounted for. For modes 2 and 3, no energy is discarded, and the polarization of the ray is properly rotated
by the birefringent media.
Transmission and other properties of birefringent surfaces
When light propagates through birefringent media, the index of refraction of the glass is different for the S and
P polarizations. The ordinary index is what is seen by the perpendicular, or S-polarized light, while the effective
index is seen by the parallel or P-polarized light. Note the S and P polarization directions used in this context are
not in general the same as those used by the coating and Fresnel surface effects computation. Here S and P
refer to the perpendicular and parallel orientations relative to the crystal axis rather than the surface normal vector.
The plane that contains the refracted ray and the crystal axis vector is the parallel plane; and the P vector lies in
this plane normal to the ray vector. The S vector is perpendicular to both P and the ray vector.
For polarization analysis of birefringent materials, a few assumptions are needed. If the mode is 0, the ordinary
ray is traced, which only has an S component, so the P component transmission is set to zero. If the mode is 1,
the extraordinary ray is traced, and the S component is therefore set to zero. This technique yields the correct
transmission results for each possible path separately. However, to get the total transmission requires analysis
of each possible combination of modes for every pair of birefringent surfaces. If there are 2 pairs of birefringent
surfaces in the system, 4 separate ray traces are required; and if there are 3 pairs of birefringent surfaces, 8
traces required, etc.
If the mode is 2, then the ray follows the ordinary path, but the ordinary index is used to phase rotate the S
component of the field, and the effective index (properly referenced to the ordinary ray direction) is used to rotate
the P component of the electric field. Only 1 trace is needed to properly model both transmission and phase
rotation. If the mode is 3, then the ray follows the extraordinary path, with a similar phase rotation as in the case
of mode 2.
Sample ZEMAX files
For some samples of the birefringent surface type, see the "\Samples\Sequential\Birefringent prisms" directo-
ry. Shown below is the "Wollaston prism" sample file showing two configurations superimposed. The two halves
of the prism are both birefringent, with the crystal axis oriented at 90 degrees to each other.

PARAMETER DEFINITIONS FOR BIREFRINGENT "IN" SURFACES
Parameter 0 Parameter 1 Parameter 2 Parameter 3 Parameter 4 Parameter 5
Length to draw Mode, 0, 1, 2, or X-cosine Y-cosine Z-cosine Parax Ignore. If

crystal axis, 0 to 3. anything other
not draw. than 0, paraxial
rays ignore the
extraordinary
index.
The birefringent "out" surface does not use any parameters. The surface shape of the birefringent in and out
surfaces is the same as a standard surface; which includes planes, spheres, and conic aspheres; all symmetric
about the local Z axis.
Conjugate
The conjugate surface is defined by two user-specified points. ZEMAX always uses the surface vertex as the
reference point; the two points required to define the conjugate surface are specified relative to this vertex. The
conjugate surface will always perfectly image one point to the other point, assuming the surface is a mirror.
Although the conjugate surface can have any material type, it is useful to think of it as being defined by its reflective
properties.
If the z-coordinates of the two points are either both positive or both negative, then the image formed from one
of the points to the other is real. In this case, the distance from one of the points to an arbitrary point on the surface,
plus the distance from the arbitrary point on the surface to the second point, is constant for all points on the
surface. One additional constraint is needed to make the surface unique: the surface must pass through the vertex
of the local coordinate system. If the surface is reflective, then one point is the conjugate of the other, hence the
name.
The surface generated by these two points satisfies the following expression if both z1 and z2 have the same
sign:
2 2 2 2 2 2 2 2 2 2 2 2
( x x1 ) + ( y y1 ) + ( z z1 ) + ( x x2 ) + ( y y2 ) + ( z z2 ) = x1 + y1 + z1 + x2 + y2 + z 2 .
Note that the surface must intersect the point (0,0,0). Several types of surfaces can be formed with this model.
For example, a sphere can be formed by setting the x and y values to zero, and the two z values each to the
radius of the sphere. An elliptical surface of arbitrary orientation can be formed by specifying non-zero values for
either the x or y values.
If z1 and z2 have opposite signs, then the image formed from one of the points to the other is virtual. In this
case, the distance from one of the points to an arbitrary point on the surface, minus the distance from the arbitrary

point on the surface to the second point, is constant for all points on the surface. Like the real imaging case, the
surface must pass through the vertex of the local coordinate system.
The surface generated by these two points satisfies the following expression if z1 and z2 have opposite signs:
2 2 2 2 2 2 2 2 2 2 2 2
( x x1 ) + ( y y 1 ) + ( z z 1 ) ( x x 2 ) + ( y y2 ) + ( z z 2 ) = x1 + y1 + z1 x2 + y2 + z2 .
Note that the surface must intersect the point (0,0,0). Several types of surfaces can be formed with this model.
For example, a hyperbola can be formed by setting the x and y values to zero, and the two z values to opposite
values. If the z values are equal but opposite, then a plane will be generated.
The coordinates of the two construction points are specified in the parameter columns, as shown in the
following table. Neither the z1 nor the z2 value can be zero.
PARAMETER DEFINITIONS FOR CONJUGATE SURFACES
x1 y1 z1 x2 y2 z2
Coordinate Break
The coordinate break surface is used to define a new coordinate system in terms of the current system. It is
always considered a "dummy" surface for ray tracing purposes. There are six parameters used to describe the
new coordinate system: x-decenter, y-decenter, tilt about x, tilt about y, tilt about z, and a flag to indicate the order
of tilting and decentration. Coordinate break decenters are always specified in lens units. Coordinate tilts are
specified in degrees, and are right-handed with respect to the positive axes. Coordinate breaks are always relative
to the coordinate system of the previous surface.
An alternate way of implementing coordinate breaks is to use surface tilts and decenters, see Surface tilt/
decenter tab on page 73 for details.
!The order of the decenters and tilts matters!

If the "order" flag is set to zero, ZEMAX first decenters in x, and then in y (because these coordinates are
orthogonal which is first does not matter). Then ZEMAX tilts about the local z axis, then the local y axis, then the
local x axis. Although the order of rotations seems reversed from "normal" this convention is used because it is
equivalent to rotating around the x axis, then rotating around the "new" y axis resulting from the x rotation; then
finally rotating around the new z axis which resulted from the previous x and y rotations. Either perspective is valid.
If the "order" flag is any other value (such as unity), then the tilts are done first, in the order x, y, and z, and
then the decenters are done. This "order" flag is extremely useful because a single coordinate break can undo
an earlier coordinate break, even for compound tilts and decenters.
The coordinate break acts like a plane surface oriented in the coordinate system after the decenters and tilts
have been applied. However, the surface is never drawn, and cannot be used to define the boundary between
two media. The glass type will always be the same as the prior surface, and ZEMAX will display "-" for the glass
name, which is meant to indicate that a glass type cannot be entered there. Coordinate breaks themselves can
never be mirrors, nor can the object surface be a coordinate break.
The coordinate rotations are described by a series of three rotation matrices. If the "order" flag is not zero the
coordinate rotation is described by:
x' cos ( z ) sin ( z ) 0 cos ( y ) 0 sin ( y ) 1 0 0 x

y' = sin ( z ) cos ( z ) 0 0 1 0 0 cos ( x ) sin ( x ) y .
z' 0 0 1 sin ( y ) 0 cos ( y ) 0 sin ( x ) cos ( x ) z

If the order flag is zero, then the z and x matrices are exchanged, so the z rotation is first, followed by y, then x.
PARAMETER DEFINITIONS FOR COORDINATE BREAK SURFACES

Decenter X Decenter Y Tilt About X Tilt About Y Tilt About Z Order
Cubic Spline
The cubic spline surface is described by sag values which are the distances between the vertex tangent plane
and the surface. Spline surfaces are used to describe unusual correctors, headlamps, and other non-standard
optical surfaces, but rarely for imaging applications because of the fundamental properties of splines. See
"Comments about spline surfaces" below for more discussion.
The Cubic Spline surface uses eight values to represent the sag at one-eighth, two-eighths, and so on to eight-
eighths of the semi-diameter of that surface. Cubic spline surfaces are rotationally symmetric. All eight points
must be defined. A subset cannot be used, although the semi-diameter may be defined to exceed the useful
aperture of the surface. This is often required because of the steep curvatures occasionally introduced by spline
fitting. If eight points provides an overly coarse sampling, see Extended Cubic Spline on page 245. For a more
general non-rotationally symmetric surface, see Grid Sag on page 259.
Comments about spline surfaces
Cubic splines are formed by a piece-wise concatenation of curved segments. Within the bounds of each
segment, the curve is defined by a third order polynomial. The polynomial coefficients describing each segment
are determined from the sag values of the defined segment boundaries. The determination of the coefficients is
driven by the boundary requirements that (a) the curve goes through the defined points, (b) the curve be
continuous across segment boundaries, and (c), the first derivatives are also continuous across the segment
boundaries. For a third order spline, it is not possible to require higher order derivatives, such as second or third,
to be continuous across segment boundaries. For this reason, splines are of limited accuracy and usefulness in
high precision optical design. Note that fundamental optical properties, such as surface power, are determined
by second order derivatives, and basic aberrations such as coma and spherical are controlled by third and fourth
order derivatives.
A common characteristic of tracing rays through spline surfaces is rough or noisy looking ray data, with
discontinuities in some results. These ray trace discontinuities are a fundamental limitation of splines, and they
are not due to a flaw in ZEMAX, or a lack of numerical precision.
Higher order splines of course exist, and one way to eliminate the discontinuities is to use a higher order spline
and fewer segments. In the limit, this is essentially the same as using a single high order polynomial for the whole
surface, see for example the Even Asphere on page 243. This is why high order polynomials, and not splines
or NURBS, dominate in precision optical design; they are continuously smooth and differentiable to all orders.
For an excellent discussion of spline theory, properties and algorithms, see Numerical Recipes in C, by Press
et al., Cambridge University Press.
PARAMETER DEFINITIONS FOR CUBIC SPLINE SURFACES
Sag at 1/8 Sag at 2/8 Sag at 3/8 Sag at 4/8 Sag at 5/8 Sag at 6/8 Sag at 7/8 Sag at 8/8
Cylinder Fresnel
The Cylinder Fresnel surface has a polynomial aspheric cylindrical substrate with an independent polynomial
aspheric cylindrical function defining the Fresnel lens properties. The surface substrate sag is given by:
2
cs y 2 4 6 8 10 12 14 16
z s = ---------------------------------------------------- + 1 y + 2 y + 3 y + 4 y + 5 y + 6 y + 7 y + 8 y .
2 2
1 + 1 ( 1 + k s )c s y

The previous expression is used to compute the ray-surface intercept. Once the intercept is found, the
refraction (or reflection) of the surface is determined by the local slope of the Fresnel facets. The Fresnel facet
shape is described by an identical expression, with independent coefficients:
2
cf y 2 4 6 8 10 12 14 16
z f = --------------------------------------------------- + 1 y + 2 y + 3 y + 4 y + 5 y + 6 y + 7 y + 8 y .
2 2
1 + 1 ( 1 + k f )c f y
Note that the curvature (symbol c), the conic constant (symbol k), and all of the polynomial coefficients are
independent for the substrate sag and Fresnel portions of the surface. The refraction at the surface accounts for
only the Fresnel sag, while the ray-surface intercept depends only upon the substrate sag.
The substrate sag radius, conic, and polynomial terms are all specified in the Lens Data Editor. The Fresnel
sag terms are specified in the Extra Data Editor. However, the Extra Data Editor uses curvature (the reciprocal
of radius) rather than radius for the Fresnel sag. The Extra Data Values used are summarized in the following
table.
EXTRA DATA DEFINITIONS FOR CYLINDER FRESNEL SURFACES

1 Maximum aspheric polynomial term number, n. The maximum is 8.
2 The Y direction curvature (NOT RADIUS) of the Fresnel surface. This parameter will
effect the refraction of the surface, but not the shape of the substrate.
3 The Y direction conic constant of the Fresnel surface. This parameter will effect the
refraction of the surface, but not the shape of the substrate.
4 2
Coefficient on y
n+3 2n
Coefficient on y
The "Maximum aspheric polynomial term number" is used to specify the maximum polynomial term to be used
in calculating the Fresnel sag. This number is provided to speed the ray tracing calculation, as terms beyond this
number are ignored. As with any complex surface model, extreme care should be taken to evaluate the accuracy
and appropriateness of this model, especially where fabrication decisions are concerned.
Because there is no reliable way to compute the phase through a Fresnel surface which is not a plane, any
calculation that requires OPD data, such as OPD fans, MTF, and Zernike coefficients, will not be supported if a
non-plane substrate Fresnel surface is present in the lens description.
Diffraction Grating
The diffraction grating surface can be used to model straight-line ruled gratings. The grating lines are parallel
to the local x-axis. Other orientations can be simulated by using a coordinate break surface before and after the
grating surface. For a plane grating, rays traced to the grating are refracted according to the equation
M
n 2 sin 2 n 1 sin 1 = -------- = MT ,
d
where d is the grating spacing (always in micrometers), 2 is the refracted angle, 1 is the incident angle, M
is the diffraction order, is the wavelength (always in micrometers), n 1 and n 2 are the indices of refraction before
and after the grating, and T is the grating frequency in lines per micrometer. Note that the sign convention for M
is completely arbitrary. ZEMAX uses the definition for T (lines per micrometer) rather than d (micrometers per
line). The grating surface can be plane, spherical, or conical, and the medium before the grating, as well as the

grating itself, can be air, glass, "MIRROR" or any other valid glass type. The grating is described by the y-spacing
of the grating lines measured in lines per micrometer (independent of the system units) and the diffraction order.
ZEMAX only models gratings to the extent of deviating ray paths. Other properties, such as efficiency, and relative
transmission are not supported. If the grating spacing is too small (or if T is too large) to satisfy the grating relation,
then a "Ray missed surface" error will be reported.
PARAMETER DEFINITIONS FOR DIFFRACTION GRATING SURFACES
Grating lines per micrometer Diffraction order
Elliptical Grating 1
The elliptical grating 1 surface is a combination of an elliptical surface with polynomial aspheric terms and a
grating with grooves that are straight when projected on to the vertex tangent plane (although the surface itself
is usually not planar). The grating lines are parallel to the local x-axis, and may be equally spaced or have a
variable spacing defined by a 4 term polynomial (up to the cubic term in y). Other orientations can be simulated
by using a coordinate break surface before and after the grating surface. For a plane grating, rays traced to the
grating are refracted according to the equation
M
n 2 sin 2 n 1 sin 1 = -------- = MT ,
d
where d is the effective grating spacing (always in micrometers), 2 is the refracted angle, 1 is the incident
angle, M is the diffraction order, is the wavelength (always in micrometers), n 1 and n 2 are the indices of
refraction before and after the grating, and T is the effective grating frequency in lines per micrometer. ZEMAX
uses the definition for T (lines per micrometer) rather than d (micrometers per line). The effective grating
frequency T is defined by:
--1- = d eff = ----

1- 2 3
+ y + y + y .
T T0
Note the constant spacing term is defined in reciprocal units to be compatible with other grating surfaces in
ZEMAX; while the higher order terms are defined in a polynomial in the y coordinate. If T evaluates to zero, the
grating component is ignored, which is useful for modeling elliptical aspheres which are not gratings.
The medium behind the grating surface may be air, glass, or a mirror. ZEMAX only models gratings to the
extent of deviating ray paths. Other properties, such as efficiency and relative transmission are not supported. If
the grating spacing is too small (or if T is too large) to satisfy the grating relation, then a "Ray missed surface"
error will be reported.
Elliptical Grating surface shape
The surface shape is an ellipse with added polynomial aspheric terms. The sag of the surface is given by the
following expression:
N
2
cu
z = ---------------------------- + A i E i ( x, y ) ,
2
1+ 1u i=1
where
2 2 2 2 2
u = a x +b y .

In these expressions, x, y, and z are the ray-surface intercept coordinates, a, b, and c are the coefficients which
define the shape of the ellipse, and the extended polynomials represented by E are defined in the section
Extended Polynomial on page 247.
Note the symbols a, b, and c do not have the same meaning in this sag expression as these same symbol
names do in the description of the standard surface. To make a spherical elliptical surface, the values a and b
should be identical and equal to one over the radius of curvature of the surface, and the value c should be equal
to the radius of curvature:
1 1
a = b = --- = --- .
R c
PARAMETER DEFINITIONS FOR ELLIPTICAL GRATING 1 SURFACES
Lines per Diffraction a b c

micrometer, order
T0
Elliptical Grating 2
The elliptical grating 2 surface is very similar to the elliptical grating 1 surface. The distinction is that the grating
lines are formed by the intersection of tilted, parallel, evenly spaced planes with a substrate defined by an elliptical
surface with polynomial aspheric terms. The intersection of the surface and tilted planes yield a grating with
grooves that are straight when projected on to a plane tilted by an angle y , measured in the local y direction.
When projected onto a surface which is not plane, the grooves are curved, not straight.
The grating is defined by the parameter T, which is the lines per micrometer (the grating frequency) when the
groove lines are projected onto the tilted XY plane. This is the reciprocal of the spacing between the planes which
form the grooves. The diffraction order is M, and the elliptical substrate is defined by the parameters a, b, c, and
extra data terms, exactly as for the elliptical grating.
The medium behind the grating surface may be air, glass, or a mirror. ZEMAX only models gratings to the
extent of deviating ray paths. Other properties, such as efficiency and relative transmission are not supported. If
the grating spacing is too small (or if T is too large) to satisfy the grating relation, then a "Ray missed surface"
error will be reported. The surface shape is an ellipse with added polynomial aspheric terms, see the Elliptical
Grating surface shape on page 242.
PARAMETER DEFINITIONS FOR ELLIPTICAL GRATING 2 SURFACES
Lines per Diffraction order a b c y

micrometer, T
Even Asphere
Rotationally symmetric polynomial aspheric surfaces are described by a polynomial expansion of the deviation
from a spherical (or aspheric described by a conic) surface. The even asphere surface model uses only the even
powers of the radial coordinate to describe the asphericity. The model uses the base radius of curvature and the
conic constant. The surface sag is given by
2
cr 2 4 6 8 10 12 14 16
z = -------------------------------------------------- + 1 r + 2 r + 3 r + 4 r + 5 r + 6 r + 7 r + 8 r .
2 2
1 + 1 ( 1 + k )c r

Note that the coefficients have units. The coefficients are entered in the corresponding parameter columns, as
shown in the following table.
PARAMETER DEFINITIONS FOR EVEN ASPHERE SURFACES
1 2 3 4 5 6 7 8
Extended Asphere
The extended asphere surface is similar to the Even Asphere surface; see Even Asphere on page 243.
However, the extended asphere can support aspheric coefficients up to order 480, whereas the even asphere is
limited to 16. Also, a slightly different method is used for computing the coefficients of the polynomial terms. The
sag of the surface is given by the following expression:
N
2
cr
z = -------------------------------------------------- + i ,
2i
2 2
1 + 1 ( 1 + k )c r i=1
where the first expression is identical to the standard surface, and the second term is a power series sum over
a normalized radial coordinate. The normalized radial coordinate is used because then the coefficients i all
have units of lens units. The extra data terms are defined in the following table.
EXTRA DATA DEFINITIONS FOR EXTENDED ASPHERE SURFACES

1 Maximum term number. The maximum is 240.
3 2
Coefficient in lens units on
4 4
n+2 2n
surface sag. This number is provided to speed the ray tracing calculation, as terms beyond this number are
14
ignored. For example, if the last term in the series you wish to use is , which is term number 7, then specify
polynomial expansion, and NOT 9, which is the "extra data number", which is the position in the extra data list of
parameters. The "extra data number" is always two greater than the term number!
all dimensionless and the coefficients are all in lens units. For modeling surfaces that are not well described by
polynomials, see Grid Sag on page 259 or User Defined on page 275.

Extended Cubic Spline
The Extended Cubic Spline surface is similar to the Cubic Spline surface (see Cubic Spline on page 240),
with the key difference being more terms are allowed. The Extended Cubic spline supports between 4 and 240
terms. The surface sag is defined by a list of sag points evenly spaced along the radius of a rotationally symmetric
shape. The sag must always be zero at zero radius, and so this point is not entered into the spreadsheet. For
important information about the limitations of splines, see Comments about spline surfaces on page 240. For a
non-rotationally symmetric spline surface, see Grid Sag on page 259.
If you have a list of sag points that describes your optical surface (as might be generated from a computer
program other than ZEMAX) and you wish to enter them into ZEMAX, use the load capability of the extra data
editor. This feature is described in Extra Data on page 79.
EXTRA DATA DEFINITIONS FOR EXT. CUBIC SPLINE SURFACES

1 Maximum number of spline points.
2 Step size in lens units between points along the radius.
3 The sag at a distance of one step size away from the vertex.
4...n The sag at a distance of 2... (n-2) step sizes away from the vertex.
The "Maximum number of spline points" is used to specify how many points define the surface. This number
may be between 4 and 240. At least 4 points are required to define the curved surface.
Spline surfaces can cause rough or ragged ray tracing results. A more general and much smoother solution
is to use the grid sag surface, which is not restricted to rotational symmetry. See Grid Sag on page 259.
Extended Fresnel
The Fresnel surface described in Fresnel on page 249 models a plane surface which has the refractive (or
reflective) power of a curved surface. The Extended Fresnel surface increases the flexibility of this model by
supporting a plane, spherical, conic, or polynomial aspheric substrate on which a plane, spherical, conic, or
polynomial Fresnel pattern is etched.
The surface sag is identical to the even aspheric surface:
2
cs r 2 4 6 8 10 12 14 16
z s = ---------------------------------------------------- + 1 r + 2 r + 3 r + 4 r + 5 r + 6 r + 7 r + 8 r .
2 2
1 + 1 ( 1 + k s )c s r
See the section Even Asphere on page 243 for details. The previous expression is used to compute the ray-
surface intercept. Once the intercept is found, the refraction (or reflection) of the surface is determined by the
local slope of the Fresnel facets, which depends upon both the Fresnel facet shape expression for Zf (below) and
the substrate shape expression for Zs (above). The Fresnel facet shape is described by an expression virtually
identical to the even asphere sag expression:
2 n
cF r
zF - + i r 2i .
= ----------------------------------------------------
2 2
1 + 1 ( 1 + k F )c F r i=0
The only difference is that the latter expression need not use all 8 terms if n is less than 8, the maximum
accepted value. Note that the curvature (symbol c), the conic constant (symbol k), and all of the polynomial
coefficients (symbol ) are independent for the substrate sag and Fresnel portions of the surface. The refraction
at the surface accounts for both the substrate sag and the Fresnel sag, while the ray-surface intercept depends

only upon the substrate sag. The intention was to model a Fresnel lens molded on a plane which is curved or
warped into a new substrate shape after it is fabricated.
The substrate sag radius, conic, and polynomial terms are all specified in the Lens Data Editor, just like the
even asphere surface. The Fresnel sag terms are specified in the Extra Data Editor. However, the Extra Data
Editor uses curvature (the reciprocal of radius) rather than radius for the Fresnel sag. The Extra Data Values used
are summarized in the following table.
EXTRA DATA DEFINITIONS FOR EXTENDED FRESNEL SURFACES

1 Maximum aspheric polynomial term number, n. The maximum is 8.
2 The curvature (NOT RADIUS) of the Fresnel surface. This parameter will effect the
refraction of the surface, but not the shape of the substrate.
3 The conic constant of the Fresnel surface. This parameter will effect the refraction of the
surface, but not the shape of the substrate.
4 2
Coefficient on r
n+3 2n
Coefficient on r
The "Maximum aspheric polynomial term number" is used to specify the maximum polynomial term to be used
in calculating the surface sag. This number is provided to speed the ray tracing calculation, as terms beyond this
number are ignored. As with any complex surface model, extreme care should be taken to evaluate the accuracy
and appropriateness of this model, especially where fabrication decisions are concerned.
Extended Odd Asphere
The Extended Odd Asphere surface is similar to the Odd Asphere surface described in Odd Asphere on
page 263. However, the Extended Odd Asphere can support aspheric coefficients up to order 240, whereas the
Odd Asphere is limited to 8. Also, a slightly different method is used for computing the coefficients of the
polynomial terms. The sag of the surface is given by the following expression:
N
2
cr
z = -------------------------------------------------- + i ,
i
2 2
1 + 1 ( 1 + k )c r i=1
where the first expression is identical to the standard surface, and the second term is a power series sum over
a normalized radial coordinate. The normalized radial coordinate is used because then the coefficients i all
have units of lens units. The extra data terms are defined in the following table.
EXTRA DATA DEFINITIONS FOR EXTENDED ODD ASPHERE SURFACES

1 Maximum term number. The maximum is 240.
3 1

4 2
n+2 n
The "Maximum term number" is used to specify the maximum term to be used in calculating the surface sag.
This number is provided to speed the ray tracing calculation, as terms beyond this number are ignored.
Extended Polynomial
The extended polynomial surface is similar to the polynomial surface, except more terms are allowed. The
surface also supports a base conic asphere surface upon which the polynomial aspheric terms are added. The
surface sag is of the form:
N
2
cr
z = -------------------------------------------------- + A i E i ( x, y ) .
2 2
1 + 1 ( 1 + k )c r i=1
th
where N is the number of polynomial coefficients in the series, and A i is the coefficient on the i extended
polynomial term. The polynomials are a power series in x and y. The first term is x, then y, then x*x, x*y, y*y, etc.
There are 2 terms of order 1, 3 terms of order 2, 4 terms of order 3, etc. The maximum order is 20, which makes
a maximum of 230 polynomial aspheric coefficients. The position values x and y are divided by a normalization
radius so the polynomial coefficients are dimensionless.
For example, the 12th term in the polynomial expansion, which is extra data term number 14, is the coefficient
2 2
on the term in x y . The coefficients A i all have units which are the same as the lens units, such as millimeters
or inches.
EXTRA DATA DEFINITIONS FOR EXT. POLYNOMIAL SURFACES

3
of parameters. The "extra data number" is always two greater than the term number! The maximum term number
is the same as "N" in the surface sag equation.
all dimensionless and the coefficients are all in lens units.

Extended Toroidal Grating
Extended toroidal grating surfaces are a combination of the toroidal (page 272), toroidal grating (page 240),
and extended polynomial (page 247) surfaces. Toroidal surfaces are described by defining a curve in the Y-Z
plane, and then rotating this curve about an axis parallel to the Y axis and intersecting the Z axis. The resulting
shape is then deformed by the addition of extended polynomials. Finally, a linear diffraction grating is placed upon
the surface. The curve in the Y-Z plane is defined by:
2
cy 2 4 6 8 10 12 14 16
z = -------------------------------------------------- + 1 y + 2 y + 3 y + 4 y + 5 y + 6 y + 7 y + 8 y .
2 2
1 + 1 ( 1 + k )c y
This curve is similar to the even asphere (page 243) surface sag formula, except the coordinate argument is
y, not r . This curve is then rotated about an axis a distance R from the vertex. This distance R is referred to as
the radius of rotation, and may be positive or negative. The Y-Z radius of curvature is specified in the same column
on the spreadsheet editor as the radius for standard surfaces. The radius of rotation is set on parameter column
number 1. To model a cylinder lens which is flat in the X direction use zero, which ZEMAX interprets as infinite
radius.
The resulting surface is then modified by the addition of the extended polynomial terms:
z = z toroid + A i E i ( x, y ) .
i=1
th
For example, the 12th term in the polynomial expansion, which is extra data term number 14, is the coefficient
2 2
on the term in x y . The coefficients A i all have units which are the same as the lens units, such as millimeters
or inches.
The diffraction grating is defined in terms of the number of lines per micrometer and the diffraction order. These
values are specified in parameter columns 2 and 3, respectively. The grating lines are parallel to the local x axis,
and are evenly spaced when projected onto a plane.
PARAMETER DEFINITIONS FOR EXTENDED TOROIDAL GRATING SURFACES
Parameter 1 Parameter 2 Parameter 3 Parameters 4-11
Radius of Rotation Grating Lines per Diffraction order 1 8

micrometer
EXTRA DATA DEFINITIONS FOR EXTENDED TOROIDAL GRATING SURFACES


3
of parameters. The "extra data number" is always two greater than the term number! The maximum term number
is the same as "N" in the surface sag equation. The normalization radius scales the X and Y intercept coordinates
of the ray so that the polynomial terms are all dimensionless and the coefficients are all in lens units.
Fresnel
The Fresnel surface model is used to simulate flat surfaces which have been etched to have a spherical (or
optionally aspheric) profile on a small scale (for Fresnel lenses which are not flat see the extended Fresnel surface
type description). The surface intercept is determined by computing the intersection of the incoming ray with a
plane. Once the plane intercept points are found, the surface is then treated as spherical (or aspherical) for the
purposes of refraction into the next medium. This is only an approximation to a real Fresnel lens, however. The
real Fresnel lens has grooves which may alter the exact intercept point. The model used here is adequate for
Fresnel lens which have fine grooves (the groove depth is very shallow compared to the aperture). Extreme
Fresnel lenses, such as those used in lighthouses, are not modeled well. The radius of curvature and conic
constant, if any, are specified in the same manner as a standard surface. The parameter values are exactly the
same as for the even asphere model; polynomial aspherics are supported to 16th order.
PARAMETER DEFINITIONS FOR FRESNEL SURFACES
1 2 3 4 5 6 7 8
Generalized Fresnel
This surface represents a generalization of the Fresnel surface described in Fresnel on page 249. The basic
concept of a Fresnel surface, as far as ray tracing is concerned, is that the overall shape of the surface substrate
is independent of the local slope. The generalized Fresnel uses a polynomial aspheric substrate model, identical
to the Even Aspheric surface described in Even Asphere on page 243. The sag of the surface is given by the
following expression:
2
cr 2 4 6 8 10 12 14 16
z = -------------------------------------------------- + 1 r + 2 r + 3 r + 4 r + 5 r + 6 r + 7 r + 8 r ,
2 2
1 + 1 ( 1 + k )c r
where the terms are identical to those in the even asphere surface model. See that section for a complete
discussion.
Once the ray has intercepted the surface, the ray reflects or refracts as if the surface had a shape described
by a polynomial expansion of the form:
N
z = Ai Ei ( x, y ) ,
i=1

th
polynomial term. The polynomial terms are identical to those described in Extended Polynomial on page 247.
It is the gradient of the extended polynomial terms alone that determines the reflective or refractive properties of
the surface, not the surface substrate shape.
One application for this surface type is modeling faceted surfaces. For example, a flat substrate may consist
of a series of small faceted planes, which would reflect or refract the light as though the surface was tilted. This
can be simulated using a flat substrate and a linear x or y tilt term in the polynomial coefficients.
PARAMETER DEFINITIONS FOR GENERALIZED FRESNEL SURFACES
1 2 3 4 5 6 7 8
EXTRA DATA DEFINITIONS FOR EXT. POLYNOMIAL SURFACES

ignored.
Gradient 1
Media whose index of refraction is described by
2
n = n 0 + n r2 r + n r1 r ,
2 2 2
where r
= x + y , can be modeled using the gradient 1 surface. Four parameters are required: the
maximum step size, t , the base index, n 0 , the radial quadratic index, n r2 , and the radial linear index, n r1 . Note
that n r2 and n r1 have units.
Discussion on maximum step size for GRIN surfaces

The maximum step size t determines the trade-off between ray tracing speed and accuracy. The exact value
required depends upon the numerical aperture of the system and the magnitude of the coefficients. To determine
a suitable step size, start at a large value (on the order of the thickness of the surface), and then perform a spot
diagram. Note the RMS spot radius. Now decrease the step size by a factor of two. If the RMS spot radius changes
by less than a few percent, the new step size is probably small enough. Otherwise, decrease the step again. For
the final design phase, you may want to decrease the step size again. Using too small a step size needlessly

slows the ray tracing speed without enhancement of accuracy. OPD tracing generally converges more slowly than
ray tracing, so you may want to repeat the above procedure while inspecting the OPD fan. Occasionally check
to ensure that the step size is appropriate as the design progresses.
Restrictions on surfaces following GRIN surfaces
The ray trace through a gradient index media requires iteration to determine the point of intersection of the ray
with the surface following the gradient index surface. Because of this, not all surface types are allowed to follow
gradient index surface types. If the type of the surface after a gradient index surface is not supported, a message
will be presented indicating the error. Support for additional surface types to follow a GRIN surface may be added
upon request; contact ZEMAX Technical Support for details.
PARAMETER DEFINITIONS FOR GRADIENT 1 SURFACES
Parameter 1 Parameter 2 Parameter 3
t n0 nr2
Gradient 2
2 2 4 6 8 10 12
n = n 0 + n r2 r + n r4 r + n r6 r + n r8 r + n r10 r + n r12 r ,
2 2 2
where r = x + y , can be modeled using the gradient 2 surface. Eight parameters are required: the
maximum step size, t , the base index squared, n 0 , and the remaining six coefficients of the preceding equation.
Note that some of the coefficients have units.
The maximum step size t determines the trade-off between ray tracing speed and accuracy. For details see
Discussion on maximum step size for GRIN surfaces on page 250. Also see Restrictions on surfaces following
GRIN surfaces on page 251.
t n0 n r2 n r4 n r6 n r8 n r10 nr12
Gradient 3
2 4 6 2 3
n = n 0 + n r2 r + n r4 r + n r6 r + n z1 z + n z2 z + n z3 z ,
2 2 2
where r = x + y , can be modeled using the gradient 3 surface. Eight parameters are required: the
maximum step size, t , the base index, n 0 , and the remaining six coefficients of the preceding equation. Note
that some of the coefficients have units.

t n0 n r2 n r4 n r6 n z1 n z2 n z3
Gradient 4
2 2 2
n = n 0 + n x1 x + n x2 x + n y1 y + n y2 y + n z1 z + n z2 z ,
can be modeled using the gradient 4 surface model. There are eight parameters required: the maximum step
size, t , the base index, n 0 , and the remaining six coefficients of the preceding equation. Note that some of the
coefficients have units. This particular GRIN model is useful for gradient index optics with a cylindrical power
profile. It is also useful for modeling thermal gradients in optical elements, if you have sufficient data to compute
the coefficients.
The linear transverse terms, n x1 and n y1 , are ignored when performing paraxial ray tracing.
t n0 nx1 n x2 ny1 n y2 n z1 n z2
Gradient 5
The gradient 5 surface has the gradient form:
2 4 2 3 4
n ref = n 0 + n r2 r + nr4 r + n z1 z + n z2 z + n z3 z + n z4 z ,
2 2 2
where r = x + y . Eight parameters are required: the maximum step size, t , the base index, n 0 , and the
remaining six coefficients of the preceding equation. Note that some of the coefficients have units.
The important feature of the gradient 5 model is that the model allows specification of dispersion properties of
the media. The dispersion data is user defined and is stored in an ASCII file called SGRIN.DAT. The format for
SGRIN.DAT will be described shortly.
The material name is entered in the glass column of the gradient 5 surface type. If the glass column is blank,
then dispersion effects are ignored.

To perform ray tracing, ZEMAX first computes the index at a "reference" wavelength using the preceding
formula for n ref . The index at any other wavelength is then computed using the following method based upon a
general expansion of the Sellmeier formula:
3 2 2
K i ( ref )
n ( ) = n ( ref ) + -------------------------------
2 2
, where
2
i=1
L i
K_MAX L_MAX

j1 j1
Ki = K ij [ n ref ] , Li = L ij [ n ref ] , and n ref = n ( ref ) .
j=1 j=1
The coefficients K ij and L ij define the dispersion of the material, whereas the gradient coefficients specified
by parameters 2-8 (see the following table) define the index gradient profile at the reference wavelength. This
very general dispersion model permits modeling of nearly arbitrary gradient index dispersion over a wide
wavelength band. The parameters K_MAX and L_MAX may be as little as 1 term, or up to 8 terms for more
accurate modeling.
The dispersion data is stored in the ASCII file SGRIN.DAT which must reside in the \GLASSCAT directory.
The SGRIN.DAT file contains blocks of 10 lines each. The file format has the following structure:
MATERIALNAME
MIN_WAVELENGTH MAX_WAVELENGTH
REF_WAVELENGTH
K_MAX L_MAX
K11 K12 K13 ... K1K_MAX
K21 K22 K23 ... K2K_MAX
K31 K32 K33 ... K3K_MAX
L11 L12 L13 ... L1L_MAX
L21 L22 L23 ... L2L_MAX
L31 L32 L33 ... L3L_MAX
Multiple materials may be defined in the same file by placing additional blocks of 10 lines one after the other,
with no blank lines between. The supplied file SGRIN.DAT contains coefficients to describe some gradient
materials offered by LightPath Technologies.
t n0 n r2 n r4 n z1 n z2 n z3 n z4
Gradient 6
The gradient 6 surface has the following gradient profile:
2 4
n = n0 + n1 r + n 2 r .
The difference between the gradient 6 and gradient 1 surface models is that the gradient 6 surface uses
dispersion formulas to compute n 0 , n 1 , and n 2 automatically rather than using entries from the lens data
editor.

The n x values are computed from the following formula:
2 Cx Dx
n x = A x + B x + -----2- + -----4- ,

with identical expressions (but different values for A, B, C, and D) for n 0 , n 1 and n 2 . This expression uses
nanometers rather than micrometers for .
The dispersion data is user defined and is stored in an ASCII file called GLC.DAT. The GLC.DAT file contains
blocks of 13 lines each. The first line in the file is the name of the material, which can be any name (without special
characters such as spaces or quotes) less than 10 characters long. The next 12 lines are the values for A, B, C,
and D for n 0 , n 1 , and then n 2 , respectively. There are no blank lines allowed between blocks. ZEMAX can
read in data for up to 25 different materials in the GLC.DAT file.
The dispersion data in the supplied GLC.DAT file was provided by the Gradient Lens Corporation (GLC) in
Rochester, NY, (716) 235-2620. Contact GLC for detailed information on material properties. Not all of the
materials offered by GLC are included in the GLC.DAT file. The following materials are included: ARS10, ARS20,
ARS27, and ARS31.
To use the gradient 6 surface materials, change the surface type to gradient 6, and then enter the appropriate
material name in the glass column of the Lens Data Editor.
Parameter 1
Gradient 7
The gradient 7 surface has a gradient profile with spherical symmetry. The index of refraction is described by
2
n = n 0 + ( r R ) + ( r R ) , where
R 2 2 2
r = ------ x + y + ( R z ) .
R
The coordinates x, y, and z are the usual coordinates measured with respect to the vertex tangent plane, and
R is the radius of the isoindex contours measured at the vertex. The isoindex contours are spherical shells
centered about the point z = R. The starting index, n 0 , is measured at the vertex of the surface, not at the center
of the isoindex contours.
There are five parameters required: the maximum step size, t , the base index, n 0 , R, , and . Note that
and have units.
The isoindex contour radius R may be specified independently from the front or rear radius of curvature of the
lens. However, if R is set to be zero, then ZEMAX assumes the isoindex radius and the front radius of curvature
are equal.

t no R
GRADIUM
This surface type models lenses made from stock gradient index blanks available from LightPath Technolo-
gies, Inc. The blanks have an axial gradient index profile, which describes a reference index of refraction as a
function of axial position within the glass blank. All that is required to define the lens is the starting position within
the blank, the name of the stock blank profile, and of course the radii and thickness. The GRADIUM surface has
a gradient index profile which is described by a polynomial of the form:
11
z + z i
n = n i --------------- .
z max
i=0
The z coordinate is the distance from the front vertex of the surface, zmax is the maximum z coordinate of the
blank (also known as the boule thickness), and the value z is the "offset" distance along the profile. Unlike most
other gradient glass models in ZEMAX, the GRADIUM surface uses only fixed, predefined axial profile coeffi-
cients. The only design parameter required is the offset value, z . The available profiles provided with ZEMAX
are defined in the ASCII file PROFILE.GRD. Other profile files may be defined, if they end in the extension GRD
and are placed in the \ZEMAX\GLASSCAT directory. To select a different profile file, see GRADIUM Profile on
page 91. For a list of the available profiles, see Gradium Profile on page 180.
GRADIUM profile file format
The file format is a series of blocks of 13 lines of data defined as follows:
PROFILE_NAME GLASS_FAMILY MAX_Z DENSITY UNUSED
n0
n1
...
n11
Each block of data starts with the name of the profile, which can be any valid ASCII name less than 20
characters. On the same line, the glass family name follows, which must be a gradient index material as defined
in the SGRIN.DAT file, which is described in the section on the gradient 5 surface type. The glass family name
defines the reference wavelength which the profile describes. The last entry on the line is the maximum z
coordinate of the blank. The 12 polynomial coefficients follow, from n0 to n11. The maximum allowed number of
profiles is 100.
When ZEMAX performs ray tracing, the local z coordinate on the surface (which may be negative) is computed.
The offset value is then added to determine where in the profile the coordinate lies.
Normally, the resulting value must always be positive and less than or equal to the maximum z, otherwise an
error is generated (See the discussion of "Capping" below). The reference index is then evaluated. Once the
reference index is computed, the index at the wavelength being traced is computed using the technique described
in the section on the gradient 5 surface.
The reference wavelength index at the front vertex is displayed as parameter 3. This value may be changed
in the Lens Data Editor. When a new value is entered, ZEMAX computes the appropriate z to yield the specified
reference index. However, z is the value that matters. The reference index is just displayed for convenience,
and should not be made variable, or a multi-configuration operand. Note the reference index is the vertex index

at the reference wavelength, which is the wavelength defined in the glass family definition file SGRIN.DAT. It is
not necessarily the primary wavelength.
The GRADIUM surface model also supports 4 additional parameter terms that are intended for use in
tolerancing: Decenter X, Decenter Y, Tilt X, and Tilt Y. These four terms model an axial gradient that is not perfectly
centered and not perfectly parallel to the local Z axis. The tolerance terms modify the axial profile by redefining
the axial coordinate z as follows:
z' = t x ( x d x ) + t y ( y d y ) + t z z , where
1
---
2 2 2
t z = [ 1.0 tx ty ] ,
and tx, ty, tz are the coefficients of the unit vector which points along the axial gradient axis and dx and dy are
the decenters in lens units of the start of the profile. If tx and ty are both zero, then the dx and dy values do not
matter (since the gradient is only along the z axis) and the tz value is unity. The terms tx and ty determine the
slope of the profile axis in x and y, which is intended to model the tolerance of axial alignment between the gradient
axis and the mechanical axis of the lens. This expression is a linear approximation which is only valid for
asymptotically small tx and ty.
The tolerance terms, d x , d y , t x , and t y , are ignored when performing paraxial ray tracing.
Usually, only the defined range of the profile is used. However, in some cases the profile may be extended in
one or both directions to add additional glass to the ends of the profile, which allows the use of GRADIUM in
thicker lenses. This technique is called "capping". By default, ZEMAX turns capping off, so that any ray trace
which requires glass beyond the profile limits is flagged as an error. This automatically enforces boundary
constraints during optimization. To remove this restriction, the capping flag can be set to be 1, 2, or 3. The default
value of zero indicates the blank is bounded to the profile length at both ends. If the capping flag is 1, then only
the left edge is bounded (the right edge is allowed to pass beyond the profile limit). If the capping flag is 2, then
only the right edge is bounded. And if the capping is 3, then neither the left nor right edges of the blank are
bounded, and both the thickness and the offset may take on any values. The additional material added to the
beginning and end of the profile is assumed to be a homogeneous glass with the same index and dispersion as
the respective end points of the profile. This assumption may be inaccurate if there is some slope to the profile
at the defined endpoints, which is usually the case. Contact LightPath Technologies for more detailed information
on designing with GRADIUM capping.
All GRADIUM profiles are defined in pairs; there is a "positive" and "negative" profile. These are identical
profiles that are reversed in the axial direction. When modeling lenses in double pass, it is necessary to use one
profile for the forward pass, and the other (reversed) profile for the backward pass.
PARAMETER DEFINITIONS FOR GRADIUM SURFACES
Parameter Parameter Parameter Parameter Parameter Parameter Parameter Parameter Parameter

0 1 2 3 4 5 6 7 8
boule t z n ref dx dy tx ty Capping

thickness

Gradient 9
The gradient 9 surface may be used to model SELFOC materials available from NSG America, Inc., or any
gradient glass with a similar index variation. The "sag" or z-coordinate of the gradient 9 surface is the same as
the standard surface plus a "tilt" term in both x and y directions:
2
cr
z = -------------------------------------------------- + x tan ( ) + y tan ( ) ,
2 2
1 + 1 ( 1 + k )c r
where c is the curvature (the reciprocal of the radius), r is the radial coordinate in lens units, k is the conic
constant, and tan and tan are the tangents of the tilt angles in x and y. Note this is not the same surface shape
as a tilted standard surface, but it is a close approximation as long as the curvature is small, or if the tilt angles
are small. The gradient 9 surface has the following gradient profile:
A 2
n = n 0 1.0 --- r .
2
Both A and n 0 are functions of wavelength:
2
K K
A ( ) = K 0 + -----21- + -----42- ,

C
n0 = B + ----2- ,

where the wavelength is measured in micrometers. The dispersion data is user defined and is stored in an
ASCII file called GRADIENT_9.DAT. The GRADIENT_9.DAT file contains blocks of 6 lines each. The first line in
the file is the name of the material, which can be any name (without special characters such as spaces or quotes)
less than 20 characters long. The next 5 lines are the values for B, C, K0, K1, and K2. There are no blank lines
allowed between blocks. ZEMAX can read in data for up to 25 different materials in the GRADIENT_9.DAT file.
The dispersion data in the supplied GRADIENT_9.DAT file was provided by NSG America, Inc., in Somerset,
NJ, (732) 469-9650, or by Corning, Inc., Corning NY, (607) 974-4755. Contact the respective manufacturers for
detailed information on material properties. Not all of the materials offered by these manufacturers are necessarily
included in the GRADIENT_9.DAT file. The included materials are: SLS-1.0, SLS-2.0, SLW-1.0, SLW-1.8, SLW-
2.0, SLW-3.0, SLW-4.0, and SLH-1.8 for NSG and Corning-Grin for Corning.
To use the gradient 9 surface materials, change the surface type to gradient 9, and then enter the appropriate
material name in the glass column of the Lens Data Editor.
t tan tan

Gradient 10
The gradient 10 surface models glass with the following gradient profile:
2 3 4 5 6
n = n 0 + n y1 y a + ny2 y a + n y3 y a + ny4 y a + n y5 y a + n y6 y a ,
where y a = y and the || symbols indicate the absolute value. This form of gradient has a discontinuity at the
plane y = 0, and the gradient is bisymmetric about the y = 0 plane. If the glass type is left blank, then there is no
dispersion. If a glass type is entered in the glass column, then it must be one of the gradient 5 materials defined
in the section on the gradient 5 surface type on page 252. The above formula then defines the index profile at the
reference wavelength for the material, and the index at other wavelengths is computed according to the dispersion
model defined for the gradient 5 surface. The linear transverse term, n y1 , is ignored when performing paraxial
ray tracing.
t n0 ny1 n y2 ny3 n y4 ny5 n y6
Grid Phase
This surface is nearly identical to the Grid Sag surface (see Grid Sag on page 259). The key differences are:
- The units of sag are radians of phase instead of units of length.
- The unitflag data is only used to scale the delx, dely, and derivative values.
- The surface shape is a plane.
- A diffraction order is supported. The diffraction order is a multiplier on the phase values. A value of zero
will turn "off" the phase effects. Setting the order to -1 will reverse the sign of all defined phase values.
- A "shear distance" is supported. See "Using the shear distance" below.
The file format and general information provided in the grid sag description is otherwise valid.
Using the shear distance
One method of modeling atmospheric turbulence is to use a Grid Phase surface with data generated by an
external atmospheric modeling program. Accurate modeling may require multiple surfaces with different phase
data modeling atmospheric layers separated by long distances. Because of the separation between these layers,
different field angles pass through different parts of the atmosphere, depending upon the distance between the
layers and the ray angle. The separation causes a field dependent lateral shearing between the surfaces.
To model this shearing, one could in principle place multiple Grid Phase surfaces separated by large distances
in front of the optical system. In practice, this works poorly, because the bending of the rays as they pass through
a Grid Phase surface may cause the rays to miss the entrance pupil entirely. To avoid this problem, the Grid Phase
surface may be placed at the entrance pupil of the system, and then the "shear distance" corresponding to the
actual position of the phase layer may be defined. When ray tracing to the Grid Phase surface, for phase purposes
only, ZEMAX will adjust the ray coordinates to correspond to the coordinates the incoming ray would have had if
the Grid Phase surface had been placed at the shear distance. Normally, shear distances are negative as the
distance is measured from the surface to the effective position of the phase layer. If the shear distance is zero,
no lateral shifting occurs.

Phase coefficients sign conventions
PARAMETER DEFINITIONS FOR GRID PHASE SURFACES

Diffraction order Shear Distance Interpolation method. Use 0 for bicubic
spline, 1 for linear. See Bicubic spline
vs. linear interpolation on page 260.
Grid Sag
The grid sag surface has a shape defined by a base plane, sphere, conic asphere, or polynomial asphere plus
additional sag terms defined by a rectangular array of sag values. The surface shape is determined by either a
linear or a bicubic spline interpolation of the sag values. The sag can be described at discrete points by
8
2
cr
= -------------------------------------------------- + i r .
2i
z base
2 2
1 + 1 ( 1 + k )c r i = 1
Z ij = Z base + Z ( x i, y j ), where
x i = xdec + delx ( i i c ), y j = ydec + dely ( j j c ), and
nx + 1 ny + 1
i c = --------------, j c = --------------, 1 i nx, 1 j ny,
2 2
where delx and dely are the spacings between grid points in the x and y directions, xdec and ydec are optional
decentration coordinates, and nx and ny are the number of points in the grid in each direction. The nx and ny
values must be odd and no smaller than 5. The terms defining Z base are identical to those defined for the even
asphere surface. Note that the sag defined by the grid points may be decentered relative to the sag defined by
the base aspheric shape.
Importing grid data
All of this data must be computed and tabulated outside of ZEMAX, arranged in the proper file format, and
then read using the Extra Data Editor "Import" feature (see Extra Data on page 79). The proper file format is as
follows:
nx ny delx dely unitflag xdec ydec
z dz/dx dz/dy d2z/dxdy nodata
.
.
.
The first line in the file contains seven values, which define the (integer) number of points in the x and y
directions, the (floating point) increment in the x and y directions, an (integer) flag indicating the units of the data:
0 for mm, 1 for cm, 2 for in, and 3 for meters, and the (floating point) decenter of the grid points relative to the
base surface in x and y. Any required scaling to the current lens units will be performed. Note that sag and cross
derivative values have dimensions, and are therefore scaled, but first derivative values are dimensionless, and
are not scaled.
The remaining nx*ny lines of the file contain four (floating point) numbers and (optionally) one integer each.
The four floating point values are the sag, the x derivative of the sag, the y derivative of the sag, and the cross
derivative d/dxdy. The optional fifth data entry is an integer flag that indicates if the data is invalid. Valid measured

data should either have a zero or blank space for the nodata flag. Points for which the data is not valid should
have a nodata value of unity. Currently, ZEMAX reads in the nodata flag value but does not use it; therefore, the
sag data and derivatives should be set to zero or some other suitable value for points with no valid data.
The first data line in the file corresponds to the upper left corner of the surface, that is, the corner defined by
-x and +y limits. Each point that follows is read across the face of the surface from left to right. After nx points,
the nx+1 point is read in as the first value in row 2, and so on until nx*ny points have been read in. The file must
be in ASCII and end in the extension .DAT.
The derivative values are required for smooth bicubic interpolation of the sag between the data points. The
derivative values are not used by the linear interpolation algorithm. If all of the derivative values (dz/dx, dz/dy, and
d2z/dxdy) are zero for every point in the file, then ZEMAX will automatically estimate the derivatives using a finite
difference method.
Bicubic spline vs. linear interpolation
The sag of the surface at any arbitrary point is computed using either of two interpolation algorithms. The
bicubic spline algorithm is used by default, and provides an interpolation that is smooth in both sag and the first
derivative of the sag. The bicubic spline is the preferred algorithm for reasonably smooth data. For noisy pseudo-
random like data, or grid data with sharp discontinuities, bicubic spline can cause "overshoot", which means the
interpolation of the surface data can yield sag points which are far from the grid points. For these cases, a linear
interpolation algorithm may provide more useful results. The choice of bicubic spline or linear interpolation is user
selectable as one of the parameters.
Suggestions for using the Grid Sag surface
For important information about the limitations of splines, see Comments about spline surfaces on page 240.
To use the grid sag surface, first change the surface type to grid sag. Then from the Extra Data Editor, choose
Tools, Import, and specify the correct surface number and DAT file name. The data will be loaded into memory,
but will not be displayed on the Extra Data Editor. The reason for not displaying the data is the potentially large
number of points. Once the grid file is loaded, the values can be checked using the Surface Data Report. The
data is thereafter stored with the lens file, and the original DAT file is ignored. If the DAT file must be changed,
the file needs to be reloaded using the import feature in the Extra Data Editor.
The grid size is only limited by available memory. Each point in the grid requires 4 8-byte double precision
values and 1 1-byte value, or 33 bytes. A 255 by 255 grid file would require approximately 2 Megabytes of
memory. Surfaces using Grid Sag surfaces with many points can be slow to edit and read and write from disk.
One way to speed up editing is to turn off the "disk multi step" undo feature in ZEMAX. See Preferences on
page 58.
The bicubic interpolation algorithm is smooth to third order, with exact results at the grid points, and large grid
files are generally not required for reasonably smooth surface shapes. However, like all low order spline models
of surfaces, the Grid Sag surface can never closely model a high order asphere surface. The reason is that no
reasonable number of piece-wise third order polynomials can accurately follow a higher order shape. The Grid
Sag surface is intended to model arbitrary low order shapes without higher order undulations.
The Grid Sag surface is not defined outside of the bounds of the grid. Rays traced outside the grid area are
treated as a ray miss error. It is a good idea to make the valid data portion of the grid slightly larger than the
maximum area illuminated by rays; specifically, do not define the grid to be exactly the same size as the beam
print. Rays nearly exactly at the edge of the grid may not trace as intended.
For Grid Sag surfaces that model steeply curved optics, such as fast mirrors, it is better to use the base radius
and conic to define the parent shape, and then use the Grid Sag data to define the deviations from the base
shape; rather than to imbed the entire sag definition for the surface in the grid data. The reason is ZEMAX can
use the base radius and conic to find a good "first guess" to the ray-surface intercept, then use the additional grid
sag data to iterate to the exact solution. It is difficult and slow for ZEMAX to find the correct intercept between a
steeply incident ray and a steeply curved Grid Sag surface if the base spherical surface is flat.
An example C language program, GRIDSAMP.C, shows the correct method for creating a grid DAT file; the
example used is a spherical surface. The GRIDSAMP program also shows how to use finite differences to
compute the derivatives when only the sag formula of a surface is known analytically.
No optimization or tolerancing of grid data is possible.

PARAMETER DEFINITIONS FOR GRID SAG SURFACES
Parameter 0 Parameter 1-8
Interpolation method. Use 0 for bicubic spline, 1 for linear. 1 - 8 , see Even Asphere on page 243
See Bicubic spline vs. linear interpolation on page 260.
Hologram 1
The hologram 1 surface can be used to model optically constructed holograms. The hologram surface can be
plane, spherical, or conical, and the medium behind the hologram can be air or glass. The glass can also be
"MIRROR" which indicates the hologram is constructed and used in reflection. The hologram itself is described
by the x, y, and z coordinates of two different construction points, a construction wavelength, and the diffraction
order. The hologram deviates ray paths according to the equation
m
n ( r' o r' r ) = ---------p- n ( r o r r ) ,
c
where n is the unit vector normal to the surface of the hologram at ray intersection point, r o is the unit vector
along the first construction beam, r r is the unit vector along the second construction beam, r' r is the unit vector
along the incident readout beam, r' o is the refracted ray, c and p are the construction and playback
wavelengths, and m is the diffraction order. A value of m = 0 means the ray is undeviated, while other integral
values of m refer to higher diffracted orders. The notation used here is from the book "Aberrations of Optical
Systems" by Welford, Adam Hilger (1986). Modeling holograms requires an understanding of their behavior which
is beyond the scope of this manual, and the user is advised to see the discussion in Welford, or some other
reference, before using this feature.
Most holograms are constructed and used in transmission or reflection. There are occasions where the
hologram is constructed in transmission, and then the substrate is aluminized and used in reflection. This special
case can be simulated with the hologram surface by specifying a negative construction wavelength. Although ray
tracing will be correct for this special case, OPD tracing will not work.
ZEMAX only models holograms to the extent of deviating ray paths. Other properties, such as efficiency and
relative transmission are not supported.
The two construction beams are defined in terms of their source points. The x, y, and z coordinates of the
source points are measured relative to the hologram vertex coordinate are defined in current lens units. ZEMAX
computes the unit vector at the ray-surface intersection point using the local coordinate data and the construction
point data for the two construction beams. The construction wavelength always has units of micrometers.
The hologram is defined by the interference between the two defined construction beams, with no aberrations
assumed on the construction beams. Optically fabricated holograms with aberrated construction beams may be
modeled in a very general way, see Optically Fabricated Hologram on page 264.
!Optically fabricated holograms with aberrated construction beams may be modeled using the
"Optically Fabricated Hologram" surface.
The hologram 1 surface assumes that both construction beams diverge from the specified construction points.
Because of the reciprocity of the construction beams, this is identical to the case where both construction beams
converge toward the construction points. Some hologram fabrication methods require one beam to be converging
while the other beam is diverging. See "The hologram 2 surface" for information on this latter type of hologram.

PARAMETER DEFINITIONS FOR HOLOGRAM SURFACES
X1 Y1 Z1 X2 Y2 Z2 c m
Hologram 2
The hologram 2 surface is very similar to the hologram 1 surface. The key difference is that the hologram 1
surface assumes both construction beams either diverge from or converge to the construction points, whereas
the hologram 2 surface assumes one construction beam converges to one construction point, and the other
construction beam diverges from the other construction point. Which beam is first or second does not matter due
to reciprocity. The parameter data is the same for both hologram 1 and hologram 2 surfaces.
Irregular
The irregular surface is a standard surface shape (plane, spherical, or conic) that has additional aspheric
deviations in terms of decenter, tilt, spherical aberration, astigmatism, and coma. This surface type is primarily
used by the tolerancing algorithm to model irregularities in a standard shape surface. The surface sag is given by:
2
cr 4 2 2
z = -------------------------------------------------- + Z s + Z a ' y + Z ' y ,
2 2
1 + 1 ( 1 + k )c r
where
x y 2 2
x = ----------, y = ----------, = x + y , ' y = y cos x sin ,
r max r max
and rmax is the maximum radial aperture of the lens, defined by the semi-diameter value for the surface. The
coefficients Zs, Za, and Zc represent the amount of spherical aberration, astigmatism, and coma, respectively, in
lens units at the maximum radial aperture. The astigmatism and coma are oriented along a line that makes an
angle in degrees with respect to the y axis.
The x and y coordinates of the previous equations are in a decentered and tilted coordinate system defined
by the decenter x, decenter y, tilt about x, and tilt about y values. The decenters are in lens units, and the tilt is in
degrees. The tilt and decenter values work exactly like the coordinate break surface defined in this chapter,
however, the tilts and decenters are undone after the ray is traced to the surface. Ray tracing is done according
to this algorithm:
The surface is decentered, tilted about x, then about y.
The ray is traced to the surface.
The surface is untilted about y, untilted about x, then undecentered.
The irregular surface uses the first seven parameters to define the decenter, tilt, and Z coefficients, and the
eighth parameter to define the angle. All the coefficients are measured in lens units, except the tilt angles which
are in degrees.
PARAMETER DEFINITIONS FOR IRREGULAR SURFACES

Decenter X Decenter Y Tilt About X Tilt About Y Zs Za Zc

Jones Matrix
This surface is used to define an arbitrary polarizing component. The surface shape is always plane. The Jones
matrix modifies a Jones vector (which describes the electric field) according to
E' x Ex
= A B ,
E' y C D Ey
where A, B, C, D, Ex, and Ey are all complex numbers. See Defining polarizing components on page 511 for
a complete discussion of the Jones matrix. Only polarization analysis features consider the effects of this surface
type. This surface uses eight parameter values, and none of the extra data values.
PARAMETER DEFINITIONS FOR JONES MATRIX SURFACES
Ar Ai Br Bi Cr Ci Dr Di
Non-Sequential Components
The sag of the Non-Sequential Components surface is the same as the Standard surface.
This surface enables ray tracing through one or more non-sequential objects. See the chapter "Non-Sequential
Components" for details.
Odd Asphere
The odd asphere model deviation is similar to the even asphere, except both even and odd powers of r are
used. The sag is given by
2
cr 1 2 3 4 5 6 7 8
z = -------------------------------------------------- + 1 r + 2 r + 3 r + 4 r + 5 r + 6 r + 7 r + 8 r .
2 2
1 + 1 ( 1 + k )c r
PARAMETER DEFINITIONS FOR ODD ASPHERE SURFACES
1 2 3 4 5 6 7 8
Odd Cosine

The odd cosine surface is an extension of the Odd Asphere surface, with 16 radial terms plus up to 6 additional
cosine terms. The sag is given by
16 m
2
cr P
z = -------------------------------------------------- + i r + Ai s i cos ( B i + C i ) .
i
2 2
1 + 1 ( 1 + k )c r i=1 i=1
The first term is the sag for a standard surface (plane, sphere, or conic). The second term is similar to the Odd
Asphere surface, but the number of coefficients is fixed at 16. The third term supports m cosine terms, where the
integer m must be between 0 and 6, inclusive. The coordinate s is the normalized radial coordinate given by s =
r / R, where R is the user defined normalization radius. If R is zero or negative, the cosine terms are ignored. The
surface may not be continuous in the first derivative if any B value is not an integer. The i coefficients have
units which depend upon the index i, and the A i coefficients have units of length in lens units. The angle is
measured in radians, and so B is dimensionless and C is in units of radians. The coefficients are entered in the
corresponding extra data columns, as shown in the following table.
EXTRA DATA DEFINITIONS FOR ODD COSINE SURFACES
Parameters 1-16 Parameter 17 Parameter 18 Parameters 19-42
The odd aspheric The number of cosine terms The normalization radius R The A, P, B, and C terms in
coefficients i . m, must be between 0 and for the coordinate s in the the cosine portion of the sag
6. cosine portion of the sag formula.
formula.
Optically Fabricated Hologram

The Hologram 1, Hologram 2, and Toroidal Hologram surfaces described elsewhere in this chapter are used
to define optically fabricated holograms assuming the construction optics are "perfect" and no aberrations are
imposed on the construction beams.
The Optically Fabricated Hologram is far more general. The hologram is defined using 3 ZEMAX lens files:
1) The playback file, in which the Optically Fabricated Hologram surface is placed.
2) The construction file for beam #1
3) The construction file for beam #2
When rays are traced in the playback system, ZEMAX automatically calls the two construction systems to
determine the vectors defining the interference of the two construction beams at the interception point in the
playback system. There are significant advantages to this method of defining the hologram:
1) The construction optics may be completely arbitrary. Each construction file may consist of multiple lenses,
mirrors, or even other holograms; anything ZEMAX can model. The aberrations introduced in the construction
beams are therefore fully considered.
2) The two construction optics files may include completely different optics.
3) Any variables set in the construction system automatically become variable in the playback system; allowing
simultaneous closed loop optimization of the construction and playback systems.
Defining the substrate shape
The shape of the Optically Fabricated Hologram surface substrate is identical to the Elliptical Grating Surface,
see the Elliptical Grating surface shape on page 242 for details.

Defining the construction optics
To define an Optically Fabricated Hologram, first create two lens files that each define the optics used to
illuminate and record the hologram. There are several important rules which MUST be followed when defining
the construction optics files:
1) The files must have the same name, with a "_1" and "_2" appended to the end. For example, the file names
"FAB_1" and "FAB_2" are valid file names for a pair construction optics files.
2) The construction files must reside in the same directory as the playback lens file.
3) The surface at which the two construction beams will interfere must be the stop surface in each file. Only
the ray intercept vectors at the respective stop surfaces will determine the hologram properties. The stop
surface may be a different surface number in each file.
4) The stop surface in each construction file must be of the elliptical grating surface type.
5) Only 1 configuration is allowed in each construction file. The playback system may use multiple configura-
tions.
6) Only 1 field point is allowed in each construction file, although the field definition is arbitrary.
7) Only 1 wavelength is allowed in each construction file, and the wavelengths in each construction file must
be identical. The playback wavelength is arbitrary, and need not be the same as in the construction optics.
8) Ray aiming must be turned on in each construction file.
9) The system aperture type must be "float by stop size" in each construction file.
10) No virtual propagations are allowed in the construction files; the rays must all take physically significant
paths.
With the construction optics files defined, the name of the files is given to ZEMAX via the comments column.
In the playback system, the hologram surface is set to "Optically Fabricated Hologram" and the comment for that
surface is set to the name that defines the construction optics file names. For example, if "FAB_1" and "FAB_2"
are used for the construction optics file names, enter "FAB" in the comment column. ZEMAX will append the "_1"
and "_2" and automatically read in the construction optics files. If no name is given, the surface will ignore the
effects of the hologram.
Optimizing the construction and playback optics
As ZEMAX traces rays in the playback system, any variables which affect the shape of the playback hologram
(the semi-diameter, a, b, and c parameters, or extra data parameters) are automatically copied into the
construction optics stop surfaces. This is why the stop surfaces in the construction optics must be of the elliptical
grating type, so the construction beams overlap on a surface with the same shape as the playback surface.
Any variables set in the construction optics automatically become variable in the playback system, so the
construction optics may be optimized simultaneously with any and all variables in the playback system.
Whenever the playback lens file is saved; the construction optics files are automatically saved as well.
To constrain design parameters in the construction files while optimizing both the playback and construction
files simultaneously, use the CMFV operand in the playback systems merit function to call the merit functions
defined in each of the construction beams. See the Chapter "Optimization" for information about the CMFV
operand.
Selecting the hologram type
There are two ways of interfering the hologram construction beams:
1) By assuming that both construction beams either diverge or converge from their respective sources, or
2) By assuming that one source beam converges and one diverges from their respective sources.
Assumption #1 is analogous to the model for the Hologram 1 surface, while assumption #2 is analogous to
the Hologram 2 surface. The Optically Fabricated Hologram surface supports both modes of interference. To
choose the hologram 1 type, set the "Holo Type" in parameter 1 to be 1, otherwise, set the "Holo Type" in
parameter 1 to be 2. The type must be either 1 or 2.

In the majority of cases, ZEMAX can correctly compute the OPD through the hologram using the default "OPD
Mode" of zero. However, there are some special cases where the OPD is not computed correctly. For these cases,
the user must manually determine the proper OPD algorithm by setting the OPD Mode equal to 1, 2, 3, or 4. The
correct OPD Mode is a function of the construction and playback geometry and the total number of mirrors in the
three files. There is no reliable algorithm for determining the proper OPD Mode automatically in all cases. If the
OPD is obviously wrong, try each of the OPD mode values 0 - 4 until the OPD values are computed correctly.
PARAMETER DEFINITIONS FOR OPTICALLY FABRICATED HOLOGRAM SURFACES
Holo Type Diffraction order a b c OPD Mode
Paraxial
The paraxial surface acts as an ideal thin lens. The paraxial surface is particularly useful for analyzing and
optimizing systems whose output is collimated, or nearly collimated light. Afocal systems can be modeled by
placing a paraxial surface before the image plane, and then setting the thickness of the paraxial surface (the
distance to the image plane) to be the same as the focal length. If the focal length is chosen to be one meter, then
all aberration data presented in micrometers (such as ray fans) can be interpreted to be in units of microradians.
Two parameters are required to model the paraxial surface: the focal length and the OPD mode. The focal
length is that which would be measured in air (unity index) although the paraxial model will support imaging into
a non-unity index medium. The OPD mode indicates how ZEMAX should calculate the optical path difference for
rays refracted by the paraxial lens. Although the ray tracing through paraxial lenses is well defined (see below),
computing the OPD can be more difficult, especially in the presence of significant aberrations, as discussed
below.
If the OPD mode is set to zero (the default), then the OPD computation is based upon the unaberrated
conjugate positions. This is preferred for optical systems with modest aberrations which are well described by
first-order optics.
If the aberrations are large (more than 10 waves), or the system is not well described by first-order optics, then
the paraxial lens OPD computation cannot assume that the lens is working at fixed conjugates for all incoming
rays. For these systems, the mode should be set to 1. In this mode, the OPD is computed based upon the ray
bending of the aberrated input beam. Using mode = 1 requires ZEMAX to iteratively integrate the actual phase
introduced by the surface. This results in mode = 1 being substantially slower than mode = 0, and mode = 1 is
not more accurate if the aberrations are reasonably small (less than 5-10 waves). When using mode = 1, for
maximum OPD accuracy, the working F/# of the paraxial lens should be no faster than about F/4.
!Mode = 1 is substantially slower than mode = 0!

The paraxial surface refracts rays using the following equations:
n'u x' = nux x
n'u y' = nuy y

where is the surface power, n is the index of refraction, primes indicate values on the image side of the
surface, and the angles are slopes which are computed from the ray direction cosines:
l
u x = ---
n
m
u y = ----
n
The paraxial surface shape is plane.

PARAMETER DEFINITIONS FOR PARAXIAL SURFACES
Focal length OPD Mode
Paraxial XY
The paraxial XY surface type is similar to the paraxial surface except the optical power can be specified in the
X and Y directions separately. This surface can therefore be used as a paraxial cylindrical or toroidal lens. Two
parameters are required to define the paraxial XY surface: the X power, and the Y power. OPD through paraxial
XY surfaces is computed using mode = 1 described in the paraxial surface description. Optical power is measured
in inverse length lens units, such as inverse millimeters if the lens units are millimeters. See Units on page 87.
The paraxial XY surface shape is plane.
PARAMETER DEFINITIONS FOR PARAXIAL XY SURFACES
X Power Y Power
Periodic
This surface shape is described by the following expression:
2
cr 1
z = -------------------------------------------------- A --- [ 1 + cos ( 2x ) ] [ 1 + cos ( 2y ) ] 1 ,
2 2 4
1 + 1 ( 1 + k )c r
where A is the peak to valley height in lens units of the modulation, and the values and are the spatial
frequencies of the oscillations in the x and y directions, respectively. Note this expression is the sag of a sphere
plus the cosine modulation. The sag of the surface is exactly zero at the vertex, and the amplitude is the peak to
valley amplitude. The frequencies are measured in inverse lens units.
This surface uses three parameter values, and none of the extra data values.
PARAMETER DEFINITIONS FOR PERIODIC SURFACES

A

Polynomial
The base radius of curvature and the conic constant are not used by this surface model. The sag of the
polynomial surface is given by
2 4 6 8 2 4 6 8
z = 1 x + 2 x + 3 x + 4 x + 5 y + 6 y + 7 y + 8 y .
A more general polynomial surface is also available, see Extended Polynomial on page 247.
PARAMETER DEFINITIONS FOR POLYNOMIAL SURFACES
1 2 3 4 5 6 7 8
Radial Grating
The radial grating surface is similar to the diffraction grating surface (see Diffraction Grating on page 241),
except the grating lines have radial symmetry, the grating line spacing is variable over the surface, and the
substrate is an even aspheric surface shape (see Even Asphere on page 243). For a plane grating, rays traced
to the grating are refracted according to the equation
M
n 2 sin 2 n 1 sin 1 = -------- ,
d
where d is the grating spacing (always in micrometers), 2 is the refracted angle, 1 is the incident angle, M
is the diffraction order, is the wavelength (always in micrometers), and n 1 and n 2 are the indices of refraction
before and after the grating. The radial grating surface allows d to vary over the surface according to the equation:
1 1 2 2
d ( p ) = A0 + A1 p + A2 p + A3 p + A4 p + ,
where the Ai coefficients all have units of micrometers, and p is the normalized radial coordinate defined by
r
p = --- ,
R
where r is the radial coordinate on the surface in lens units and R is the user defined normalization radius. The
grating spacing d can be interpreted in two different ways. The normal ZEMAX convention is to measure d along
the projection of the grating on the XY plane, ignoring any sag or curvature of the underlying surface. The Radial
Grating surface supports an additional "Grating Mode", where d is interpreted as being measured along the local
surface tangent. The Grating Mode can be set to 0 or 1 in parameter 9 in the Lens Data Editor.
Note that the sign convention for M is arbitrary. The grating surface can be plane, spherical, conical, or even
aspheric, and the medium before the grating, as well as the grating itself, can be air, glass, "MIRROR" or any
other valid glass type. ZEMAX only models gratings to the extent of deviating ray paths. Other properties, such
as efficiency, and relative transmission are not supported. If the grating spacing is too small to satisfy the grating
relation, then a "Ray missed surface" error will be reported.
PARAMETER DEFINITIONS FOR RADIAL GRATING SURFACES
Parameter 0 Parameter 1-8 Parameter 9
Diffraction Order Even aspheric coefficients 1 8 Grating Mode

EXTRA DATA DEFINITIONS FOR RADIAL GRATING SURFACES

1 Maximum number of terms
2 Normalization Radius
3 0
Coefficient on p
4 1
Coefficient on p
n (n 3)
Coefficient on p
Radial NURBS
The acronym NURBS stands for Non-Uniform Rational B-Spline. NURBS are a very general class of curves
and surfaces. For a complete discussion of NURBS, which is well beyond the scope of this discussion, see The
NURBS Book, Second Edition, by Les Piegl and Wayne Tiller, Springer-Verlag, ISBN 3-540-61545-8.
The radial NURBS surface is defined by a series of weighted control points. The control points define a curve
which starts at the origin and lies in the YZ plane along the +Y direction. Once this curve is defined, a figure of
revolution is formed by rotating the curve a full 360 degrees around the Z axis. Unlike a spline surface, a NURBS
curve does not actually go through the control points, except for the first and last control points.
Each control point has a positive "y" coordinate, a "z" coordinate (which may be positive or negative), and a
weight "w". The first point is always at y = 0 and z = 0, so the curve begins at the vertex origin of the surface.
Subsequent points, numbered from 1 to the maximum number of points desired, extend monotonically outward
in the +Y direction. The y values must be spaced at least 1.0E-3 lens units apart for numerical stability in the
spline fit.
The weights should all be set to 1.0 initially. The higher the weight, the closer the curve will be to the actual
control point. Lower weights mean the curve is less constrained to be near the control point. See the above
reference for details on the influence of weights on a NURBS curve.
The y value of the last point defined determines the maximum radial clear aperture of the surface. This value
should generally be fixed, and not be a variable. Any rays which do not intercept the surface within the defined
areas of the curve will be terminated with a "ray miss" error.
The advantage to the NURBS description is that any shape may be defined and traced reliably. Unusual
aspheric correctors which cannot be described by polynomials may be modeled as NURBS.
The disadvantages to the NURBS description include very slow ray tracing, and sometimes great difficulty in
finding suitable starting values for the control curve points and weights.
This surface does not use any of the parameter columns.
EXTRA DATA DEFINITIONS FOR RADIAL NURBS SURFACES

1 Number of control points. At least 4 points are required, and no more than 60 allowed.
2 y coordinate for control point 1.
3 z coordinate (sag) for control point 1.

4 w (weight) value for control point 1.
3n-1, 3n, 3n+1 y, z, and w for control point n.
Retro Reflect
The Retro Reflect surface is a planar shape surface. If the surface is made of glass or air, rays refract according
to Snells law in the usual way. If the surface is made of "MIRROR", rays do not reflect in the usual way; instead
the rays are undeviated. This means under subsequent propagation in the opposite direction the rays will follow
the incident path; that is, the rays retro-reflect.
This surface uses no parameter or extra data values.
Standard
The most commonly used optical surface is a spherical surface. The sphere is centered on the current optical
axis, with the vertex located at the current axis position. ZEMAX treats planes as a special case of the sphere (a
sphere with infinite radius of curvature) and conics as a special case as well. The "sag" or z-coordinate of the
standard surface is given by
2
cr
z = --------------------------------------------------
2 2
1 + 1 ( 1 + k )c r
where c is the curvature (the reciprocal of the radius), r is the radial coordinate in lens units and k is the conic
constant. The conic constant is less than -1 for hyperbolas, -1 for parabolas, between -1 and 0 for ellipses, 0 for
spheres, and greater than 0 for oblate ellipsoids. For more information on conic constants, see REFERENCES
ON LENS DESIGN on page 31. The standard surface does not use any of the parameter values.
Modeling an ellipse with the standard surface
There are a few handy formulas for converting the semi major and semi minor axis lengths of an elliptical
surface to a radius and conic description. If "a" is the semi major axis length, and "b" is the semi minor axis length,
then
2
1 b
--- = R = ----- ,
c a
2 2
a b
2
k = = ----------------
2
.
a
Modeling an axicon with the standard surface

2 2
The standard surface can be used to make an almost perfect axicon. If ( 1 + k )c r 1 , the standard surface
reduces to
r 1
z = ----------------------------- or z = r tan , where tan = ----------------------------- ,
( ( 1 + k ) ) ( ( 1 + k ) )
and is the axicon angle, measured from the XY plane to the axicon surface. To create an axicon, calculate
the conic constant value (k) from the angle ( ) desired, and use any small value for the radius of curvature. The
resulting value of k must be negative. The exact value of the radius or curvature is not important, as long as it is

roughly three or more orders of magnitude smaller than the radial aperture of the axicon. The axicon is not perfect
in the sense that there is no cusp at the origin; the region around the surface vertex will be rounded off over a
size given approximately by the radius value. This is actually a desirable property for ray tracing, as the surface
is everywhere smooth.
Superconic
The most common form of a polynomial aspheric surface uses a power series expansion in the radial
coordinate r to define the surface sag, where r is defined by
2 2 2
r = x +y .
For example, the Even Aspheric surface described in Even Asphere on page 243 uses such an expansion.
Since r does not depend upon z, the expansion term is the distance from the vertex to the point on the surface
as projected on to the tangent plane. Generally, the departure of the asphere from the tangent plane increases
with radial aperture. As the departure increases, the power series expansion parameter r corresponds to a point
on the tangent plane which is farther from the point on the surface. This causes the expansion to have poor
convergence.
A novel solution proposed by Alan Greynolds of Breault Research Organization is to instead expand in powers
of the distance from the vertex to the point on the surface. The expansion is then in terms of
2 2 2 2
s = x +y +z .
Starting with the conic equation for a surface
2 2 2 2
kz 2Rz + x + y + z = 0 ,
where k is the conic constant and R is the radius of curvature, a general power series expansion can be made
of the form
2
Az 2Bz + C = 0 .
The constants are defined as
k
A = --- ,
R
2 4
B = 1 + U 1 s + U 2 s + , and
2
s 4 6
C = ---- + V 1 s + V 2 s + ,
R
where U and V are coefficients which define the aspheric shape. Note that if all the U and V terms are zero, a
standard conic results. If A is also zero, then the superconic becomes a sphere. The coefficients A, U 1 , and V 1
together form a Cartesian oval. These properties make the superconic stable when optimizing for the coefficients.
The superconic can be used to model surfaces which otherwise would require aspheric terms of very high order.
ZEMAX models superconics with up to 240 terms, in practice designs rarely use more than 5 terms.

EXTRA DATA DEFINITIONS FOR SUPERCONIC SURFACES

1 Maximum term number. The maximum is 250, but 4-10 is typical.
2 U1
3 V1
even n n---
2
U
odd n n1
------------
2
V
Tilted
The tilted surface is a plane that makes an angle with respect to the x and y axes. The surface is easily defined
in terms of the tangent angle between the plane and the X and Y axes:
z = x tan x + y tan y .
The tilted surface uses the first two parameters to define the tangents of the x and y angles. This surface is
very useful for implementing tilted object and image surfaces, as well as tilted faces on prisms. It should not be
used for implementing fold mirrors; use the coordinate break surface instead.
PARAMETER DEFINITIONS FOR TILTED SURFACES
tan x tan y
Toroidal
Toroidal surfaces are formed by defining a curve in the Y-Z plane, and then rotating this curve about an axis
parallel to the Y axis and intersecting the Z axis. Toroids are defined using a base radius of curvature in the Y-Z
plane, as well as a conic constant and polynomial aspheric coefficients. The curve in the Y-Z plane is defined by:
2
cy 2 4 6 8 10 12 14
z = -------------------------------------------------- + 1 y + 2 y + 3 y + 4 y + 5 y + 6 y + 7 y .
2 2
1 + 1 ( 1 + k )c y
This curve is similar to the even aspheric surface sag formula, except the sixteenth order term has been
omitted, and the coordinate argument is y, not r . This curve is then rotated about an axis a distance R from the
vertex. This distance R is referred to as the radius of rotation, and may be positive or negative. The Y-Z radius of
curvature is specified in the same column on the spreadsheet editor as the radius for standard surfaces. The
radius of rotation is set on parameter column number 1. To model a cylinder lens which is flat in the X direction
use zero, which ZEMAX interprets as infinite radius.
Note if the Y-Z radius is set to infinity, a surface with power in x but not in y can be described, therefore, the
cylinder may be oriented in either direction. The other parameter columns are used for the optional aspheric
coefficients, as specified in the following table. If aspheric coefficients are required in the X direction, then rotate
the toroid with a pair of coordinate break surfaces and rotate about Z. If different aspheric surfaces are required

in both the X and Y directions, see the "biconic", "polynomial" and "extended polynomial" surfaces described
elsewhere in this chapter.
PARAMETER DEFINITIONS FOR TOROIDAL SURFACES
Radius of 1 2 3 4 5 6 7
Rotation
Toroidal Grating
Toroidal grating surfaces are similar to regular toroidal surfaces, except no aspheric sag terms are supported,
and a diffraction grating may be placed on the toroidal surface. Toroidal gratings are described by defining a curve
in the Y-Z plane, and then rotating this curve about an axis parallel to the Y axis and intersecting the Z axis.
Toroidal gratings are defined using a base radius of curvature in the Y-Z plane, as well as a conic constant. The
curve in the Y-Z plane is defined by:
2
cy
z = -------------------------------------------------- .
2 2
1 + 1 ( 1 + k )c y
This curve is similar to the standard surface sag formula, except the coordinate argument is y, not r . This curve
is then rotated about an axis a distance R from the vertex. This distance R is referred to as the radius of rotation,
and may be positive or negative. The Y-Z radius of curvature is specified in the same column on the spreadsheet
editor as the radius for standard surfaces. The radius of rotation is set on parameter column number 1. To model
a cylinder lens which is flat in the X direction use zero, which ZEMAX interprets as infinite radius.
Note if the Y-Z radius is set to infinity, a surface with power in x but not in y can be described, therefore, the
cylinder may be oriented in either direction.
The diffraction grating is defined in terms of the number of lines per micrometer and the diffraction order. These
values are specified in parameter columns 2 and 3, respectively. The grating lines are parallel to the local x axis,
and are evenly spaced when projected onto a plane.
For a toroidal grating with aspheric deformation terms see Extended Toroidal Grating on page 248.
PARAMETER DEFINITIONS FOR TOROIDAL GRATING SURFACES
Radius of Rotation Grating Lines per micrometer Diffraction order
Toroidal Hologram
The Toroidal Hologram surface shape is identical to the Toroidal surface described in Toroidal on page 272.
The curve in the Y-Z plane is defined by:
2
cy 2 4 6 8 10 12 14
z = -------------------------------------------------- + 1 y + 2 y + 3 y + 4 y + 5 y + 6 y + 7 y .
2 2
1 + 1 ( 1 + k )c y
The radius and conic values, along with the parameter values, are used to define the surface shape. The extra
data values are used to define the holographic properties of the surface. The terms used to define the holographic
construction beams are identical to those described in the earlier sections on the Hologram 1 and Hologram 2
surface types. Extra data values 1-6 are the x, y, and z construction points for sources 1 and 2, value 7 is the
construction wavelength, and value 8 is the playback order M.

The hologram is defined by the interference between the two defined construction beams, with no aberrations
assumed on the construction beams. Optically fabricated holograms with aberrated construction beams may be
modeled in a very general way using the Optically Fabricated Hologram on page 264.
!Optically fabricated holograms with aberrated construction beams may be modeled using the
"Optically Fabricated Hologram" surface.
There is also a ninth extra data value which is used as a "flag" to indicate if the two construction beams are
converging or diverging. If both beams are converging or diverging to or from the construction points, then the
flag value should be +1. If one beam is converging and one is diverging, then the flag should be -1. Any value
less than zero is the same as -1, and any value greater than zero is the same as +1. This flag value serves to
distinguish these two cases, which is the only difference between the Hologram 1 and 2 surface types.
PARAMETER DEFINITIONS FOR TOROIDAL HOLOGRAM SURFACES
Radius of 1 2 3 4 5 6 7
Rotation
EXTRA DATA DEFINITIONS FOR TOROIDAL HOLOGRAM SURFACES

1 Construction coordinate X1 for first construction beam.
2 Construction coordinate Y1 for first construction beam.
3 Construction coordinate Z1 for first construction beam.
4 Construction coordinate X2 for second construction beam.
5 Construction coordinate Y2 for second construction beam.
6 Construction coordinate Z2 for second construction beam.
7 Construction wavelength.
8 Playback order.
9 Converge/Diverge flag.
Toroidal NURBS
The toroidal NURBS surface is similar to the radial NURBS surface (page 269). The key difference is that
instead of rotating the curve about the Z axis to form a surface; the curve is instead mirror-imaged to the -Y axis,
then rotated about an offset Y axis to form a toric shape.
The offset axis may be in the +Z or -Z direction, depending upon the sign of the radius of rotation. If the radius
of rotation is zero, then the radius is assumed to be infinite and the surface has cylindrical symmetry. The X width
of the surface is determined by either the range of rotation angle or by direct specification of the -x and +x limits;
the latter being used if the radius of rotation is infinite (which is defined by a zero value).

PARAMETER DEFINITIONS FOR TOROIDAL NURBS SURFACES
Radius of Rotation Min X or angle Max X or angle
EXTRA DATA DEFINITIONS FOR TOROIDAL NURBS SURFACES

1 Number of control points. At least 4 points are required, and no more than 60 allowed.
2 y coordinate for control point 1.
3 z coordinate (sag) for control point 1.
4 w (weight) value for control point 1.
3n-1, 3n, 3n+1 y, z, and w for control point n.
User Defined
The user defined surface (UDS) is a powerful, flexible, and fast way of implementing surfaces not already built
into ZEMAX. A UDS may be any shape, have any refractive, reflective, or diffractive properties, may impart any
arbitrary phase to the beam, and may be followed by either homogeneous media or by a gradient index media of
arbitrary form. A UDS may also apodize or attenuate the beam arbitrarily, or be used to define any electric field
or coating data for polarization analysis. This latter capability allows the beam to be partially transmitted according
to an arbitrary formula or table as defined by the user at any surface in the optical system.
The secret to this great flexibility is that all the properties of the surface are defined by the user in a separate
C or C++ program compiled and linked into ZEMAX dynamically using the Windows Dynamic Link Library (DLL)
capability. The DLL must contain functions which return to ZEMAX all the data required to draw the surface, trace
rays, compute refraction angles, and for gradient index, to determine the index as a function of position within the
media following the surface.
Because the DLL is compiled code, and because ZEMAX passes to the DLL a pointer to the data that needs
to be computed, UDS are very fast; nearly as fast as native ZEMAX code (there is a small amount of overhead
due to the function call).
The power of the UDS comes at a price, although it is a reasonable one. Use of the UDS does require that the
user have a suitable compiler or development tool that can generate 32 bit Windows compatible DLLs. It is also
assumed that the user can write the required code, and most importantly, ensure that the code is reliable and
bug-free. To maximize speed, ZEMAX performs very little error checking on data returned by the DLL, and so
buggy UDS DLLs are quite capable of bringing ZEMAX to a crash.
For this reason, technical support on the implementation of UDS is strictly limited to illustrating that the
provided sample files work correctly. If you need a UDS DLL, and do not possess the desire or ability to write
them yourself, please feel free to contact ZEMAX Technical Support for a quote on developing a custom DLL to
meet your requirements. We have considerable experience in developing ray tracing algorithms, and can
generally write DLLs at very competitive rates on very short notice.
The UDS DLL
The best way to learn how to create a UDS DLL is to start by copying the source code file (the one that ends
in .C) of a sample DLL to a new file name, edit the sample to suit your requirements, and then recompile to make
a new DLL. A good portion of the DLL source code is "boiler plate" which is common to all DLLs.
There are two files that are required to define a UDS DLL: a C (or C++) source code file such as MY_SURF.C,
and a header file called USERSURF.H. Only the C file needs to be modified. The file is essentially 2 functions;

the first is "DLLMain" which is used to (optionally) initialize data used by the DLL. The other function is
"UserDefinedSurface", and this is the function that is generally modified for each surface type. The function name
cannot be changed; ZEMAX looks for a function of this name. The function may also be defined as
"UserDefinedSurface2"; this function definition uses a newer, larger data structure that supports additional
features not supported in the original definition of "UserDefined Surface".
ZEMAX passes two structures to the function, and these structures are defined in the USERSURF.H file. The
USERSURF.H file defines two separate "fixed data" structures, one used by UserDefinedSurface and the other
by UserDefinedSurface2. Inside the UserDefinedSurface or UserDefinedSurface2 function is a C "switch-case"
construct. ZEMAX passes a "type" parameter, which is a number between 0 and 10, inclusive, that indicates what
data ZEMAX needs computed and returned to ZEMAX. The type codes are described in the following table:
UDS DLL Type Codes
TYPE What ZEMAX wants the DLL to compute

0 The surface name, radial symmetry status, and grin status
1 The name of the parameter columns
2 The name of the extra data columns
3 The sag and alternate sag of the surface
4 A paraxial ray trace to and refraction through the surface
5 A real ray trace to and refraction through the surface, including transmittance and
polarization data if any
6 The index and first derivatives for gradient index propagation
7 The default data when the user first selects the surface type
8 The initialization of the DLL, if required. This includes allocating static memory the DLL will
require, loading data files, or other one-time calculations or initializations. This type code is
called once each time the DLL is loaded. However, ZEMAX may load the same DLL multiple
times. For example, each analysis window in ZEMAX gets its own copy of the lens data,
and each copy will load and initialize the DLL every time the analysis window is updated.
ZEMAX may update multiple analysis windows at the same time (parallel execution) and so
multiple DLLs may be running simultaneously. For this reason, using fixed file names to
store temporary data or to make log files is not recommended; the same file will be created
or written to multiple times as ZEMAX loads the DLL multiple times.
The DLL only needs to handle this type code if the DLL requires one-time initialization before
the DLL can trace rays.
9 The termination of the DLL, if required. This includes releasing allocated memory or closing
data files. This type value is called when the DLL is removed from the surface or the lens is
closed. See the comments for type code 8 above for details on how ZEMAX may load,
initialize, and terminate multiple DLLs at once.
The DLL only needs to handle this type code if the DLL must release memory or perform
some other "clean-up" before terminating.
10 The scaling of parameter and extra data values used by the DLL. The scale factor is stored
in the USER_DATA structure in the path argument.
Extensive comments are provided in the sample DLL source code files. The DLL files can be found in the
directory \ZEMAX\DLL, and any new DLL files must be placed there as well.
Refractive and reflective UDS DLLs
For conventional homogeneous surfaces which are reflective or refractive, but not gradient or diffractive, start
with the file US_STAND.C. That file is the source code for US_STAND.DLL, which is a clone of the built-in ZEMAX
standard surface. It is not exactly the same code, and is somewhat slower than the form of equations used by
ZEMAX, however, it is functionally equivalent. US_STAND.C includes a generic Snells' law refraction function

which also works for reflection. The DLL, like all UDS DLLs, provides the sag, ray intercept equations, surface
normals, optical path, and paraxial refraction functionality.
Gradient index UDS DLLs
Gradient surfaces should use US_GRIN1.C as a starting point. The code is very similar to US_STAND.C,
except a flag is returned indicating that the surface is a gradient type, and data type "6" is implemented to return
the index of refraction given X, Y, Z, and all the parameter and extra data. The first derivative of the index in the
X, Y, and Z directions must also be provided.
Diffractive UDS DLLs
Diffractive optics are very similar to standard optics, except the rays are further deviated by the derivative of
the phase as a function of X and Y, and the optical path length needs to be modified to include the phase change.
Generally, "refraction" for diffractive optics is defined by:

l' = l + ------ ------
2 x

m' = m + ------ ------
2 y
where l and m are the direction cosines, and the z direction cosine, n, is computed to make the magnitude of
the direction vector unity, and is the phase in radians.
A sample DLL, US_GRATE.DLL, illustrates the diffractive computation by cloning the ZEMAX grating surface.
Lenslet arrays using UDS DLLs
Lenslet arrays are easily modeled using the user defined surface. Basically, the ray trace determines which
segment of the array is struck, then uses the local lens curvature to determine the refraction. The sample source
code and DLL are provided as US_ARRAY.C and US_ARRAY.DLL, respectively.
User defined surface apodization using DLLs
The real ray trace portion of the DLL, type code 5, allows definition of the surface transmittance. The surface
transmittance must be a number between 0.0 and 1.0, and indicates the relative fraction of intensity that the ray
transmits through the surface. In this context, transmit means "continues on", so for reflective surfaces, the
transmitted portion is that which normally reflects to the next surface.
The surface transmittance can be used to define arbitrary surface apodizations. The transmittance function
can be any formula based upon the ray coordinates, direction cosines, surface parameters, or other data; or may
be derived from a look up table, or any other method that can be implemented within the DLL.
The surface transmittance need not be defined. If it is not defined by the DLL, ZEMAX assumes it is 1.0.
Whatever the surface transmittance is, the ray intensity will still be modified by ZEMAX to account for the Fresnel
surface, thin film, and bulk absorption affects that are normally accounted for when doing polarization ray tracing.
If ZEMAX is not doing polarization ray tracing, then the DLL defined transmittance is the only attenuating effect.
Unlike pupil apodization, surface apodizations defined using UDS may be placed on any surface anywhere in
the optical system. Note that for apodizations which are a function of ray position (and most are) different fields
will then see a different effective apodization. Note that this technique can be used to model arbitrary neutral
density filters, or filters whose transmission are wavelength dependent.
Polarization and coating data using DLLs
The real ray trace portion of the DLL, type code 5, allows definition of either the transmitted or reflected electric
field directly, or the definition of the s and p orientation complex reflection and transmission coefficients. The field
or coating data can be based upon any data available within the DLL, including ray cosines, normal vectors, index,
or other user defined data, or any other method that can be implemented within the DLL. The field and coating
data need not be defined. If it is not defined by the DLL, ZEMAX uses the default algorithm for the surface. For a
source code example of how to define the electric field or coating data see the sample DLL US_POLARIZATION.

Error handling and UDS
The convention ZEMAX uses internally is to return a zero value if the DLL computed a meaningful result, and
no error occurred. Otherwise, the DLL should return -1. The exception is when ray tracing, either paraxial or real.
If the ray misses the surface, it should return the surface number. If the ray total internally reflects (TIR's) then
the return value should be the negative of the surface number. ZEMAX uses these error codes to help provide
meaningful diagnostics to both the user and various ZEMAX features.
Sample DLLs
Numerous sample UDS DLLs have been written and are provided as both a ready to use DLL and as C source
code for study and modification. The easiest way to write a DLL is to find one most similar to the one you need,
and copy and edit the C source code file as required. The following table lists the DLLs available and a brief
description of each. Note that although the sample DLLs have been tested and are generally considered reliable,
they are provided "as is".
SAMPLE UDS DLL'S
DLL name Description

US_ANAMR Anamorphic asphere. This aspheric surface is defined by the following expression:
2 2
CxX + CyY
Z = -----------------------------------------------------------------------------------------------------------------
2 2 2 2
1 + 1 ( 1 + Kx ) ( Cx X ) ( 1 + Ky ) ( Cy Y )
2 2 2 2 2 3
+ AR [ ( 1 AP )X + ( 1 + AP )Y ] + BR [ ( 1 BP )X + ( 1 + BP )Y ]
2 2 4 2 2 5
+ CR [ ( 1 CP )X + ( 1 + CP )Y ] + DR [ ( 1 DP )X + ( 1 + DP )Y ]
US_ANASP Annular aspheric surface. This surface is described in the paper "Annular surfaces
in annular field systems" by Jose Sasian, Opt. Eng. 36(12) 3401-3403 (December
1997). This surface is similar to an odd aspheric polynomial, with a radial offset in
the polynomial expansion to better model surfaces used in annular field systems.
US_APGXY This DLL is a clone of the ZEMAX "Standard" surface type, with the added capability
to modify the surface transmission. This surface defines an apodization function of
the form
x 2 y 2
T ( x, y ) = Exp 2G x --- 2G y --- ,
R R
where R is the semi-diameter of the surface and x and y are the coordinates at which
the ray intercepts the surface.
US_ARRAY Models an n x m array of lenses, each of dimension H x W; all 4 parameters may be
specified. The lenses are rectangular in outline but are "standard" in shape; that is,
plane, spherical, or conic asphere.
US_ARRAYEVEN Models an n x m array of lenses, each of dimension H x W; all 4 parameters may be
specified. The lenses are rectangular in outline but are "even aspheric" in shape;
that is, plane, spherical, conic asphere, and radial polynomial asphere.
US_CYLAR Models a vertical array of n cylindrical lenses, each H high. The lenses are cylindrical
or conic aspheres in the YZ plane.
US_DGCYL Models a diffraction grating on a cylindrical surface, where the grating lines are
equally spaced along the arc of the surface rather than equally spaced along the y
coordinate of the tangent plane. The grating lines are parallel to the axis of the
cylinder.

US_FILT1 2
This surface defines an apodization function of the form T = 1.0 Exp ( G )
where p is the radial coordinate normalized to the semi-diameter of the surface.
US_FILT2 This surface defines an apodization function where the optical density is Dmax from
zero to a normalized radial coordinate p1, then linearly goes to D=0 at p2, and D= 0
thereafter. The density is D = -Log10(T) where T is the transmission.
US_FILT3 This surface defines
an apodization function of the form
D = 0.5D max ( 1.0 + cos ) where p is the radial coordinate normalized to the
semi-diameter of the surface.
US_FILT4 This surface has an apodization function defined by:
T ( r ) = 1 if r < R R ,
T ( r ) = 0 if r > R + R ,
( r ( R R ) )
1 + cos ---------------------------------------
2R
T ( r ) = --------------------------------------------------------------- if R R < r < R + R ,
2
where r is the radial coordinate, R is the aperture radius, and is a dimensionless

fraction of the aperture radius. The transmission varies smoothly from 1 to 0 over
twice the distance defined by R . The value for is defined as 0.001 if the provided
value for is zero. This function makes a convenient "soft-edged" aperture which
is useful for optimization of transmission through a circular aperture.
T ( na ) = 1.0 if na < NA - ,
T ( na ) = 0.0 if na > NA + , otherwise
( na ( NA NA ) )
T ( na ) = 0.5 1 + cos --------------------------------------------------- ,
( 2NA )
where na is the numerical aperture of the incoming ray, NA is a user defined

parameter, and is fixed at 0.001 (although this value may be edited in the DLL
source code). This type of filter demonstrates apodization of a ray based upon its
incident angle, rather than position. Rays whose na exceeds the NA- value will be
attenuated smoothly to zero. This permits easier, more continuous optimization.
T ( x, y ) = ( 1 4 ) ( 1 + sin ( 2x ) ) ( 1 + sin ( 2y ) ) ,
where and are the spatial frequencies of the sinusoidal transmission in x and
y, respectively.

US_FILT7 This surface apodizes the surface transmission via linear interpolation of values
specified on the extra data editor along the radial coordinate.
US_FILT8 This surface models a soft edged rectangular aperture. The MaxX and MaxY
parameters are the X and Y half widths of the rectangular aperture. The transmission
varies smoothly from 1 to 0 over twice the distance defined by the DelX and DelY
distances from the aperture boundaries. This function makes a convenient "soft-
edged" aperture which is useful for optimization of transmission through a
rectangular aperture.
US_GCYL Models a cylindrical surface with a gradient index medium. The GRIN medium has
concentric cylindrical shells of constant index, and a polynomial GRIN profile
centered on the center of curvature. This DLL was intended to model light
transversing a fiber perpendicular to the axis of the fiber (through the side).
US_GRATE Models a standard surface with grating lines parallel to the x axis, equally spaced
along y in the tangent plane. This is a clone of the ZEMAX "Diffraction Grating"
surface. The purpose is to show how diffractive surfaces are modeled in ZEMAX
UDS DLLs.
US_GRIN1 Models a quadratic GRIN medium. This is a clone of the ZEMAX "Gradient 1"
surface. The purpose is to show how GRIN surfaces are modeled in ZEMAX UDS
DLLs.
US_IGRIN An example of a GRIN with dispersion.
US_ITERA This DLL is a clone of the ZEMAX "Standard" surface, except it uses "dumb" iteration
to find the intercept point rather than a closed form expression. The point of this
example is to show how to find the surface intercept point if only the sag expression
is known. Most polynomial aspheres require this type of iteration because the ray-
surface intercept formulas cannot be determined in closed form.
US_LUNE Models a perfect Luneberg lens GRIN surface.

US_MEMS Models a Micro-Electromechanical System (MEMS), such as a Digital Mirror Device
(DMD). The MEMS consists of a 2D array of small rectangular mirrors. The mirrors
may tip at any of three angles, each rotated about an axis to point the mirror in any
direction. The mirrors may be set to any of the three states by addressing rows,
columns, or by individual mirrors if desired to model any state the MEMS can be in.
Although the model is strictly geometric, it effectively models where rays are
reflected by such a device for an arbitrary setting of the individual mirrors. The
parameters on the model are:
Nx: Number of X direction mirrors
Ny: Number of Y direction mirrors
Wx: Total width in lens units in X direction
Wy: Total width in lens units in Y direction
A0, A1, A2: Angle of mirror in state 0, 1, or 2
Rotation Angle: Angle of rotation about the Z axis to tip the mirror
Extra Data Value 1: If 0, mirrors are addressed by rows, if 1, by columns, if 2, by
individual mirrors.
Extra Data Value 2: State of rows/columns/mirrors 1 - 15
Extra Data Value 3: State of rows/columns/mirrors 16 - 30
Extra Data Value n: State of rows/columns/mirrors 1+15*(n-2) - 15*(n-1)...
The rotation angle effectively rotates the plane of tip of the mirror around the local Z
axis; with the initial tip plane being around the local x axis. The rotation angle is
measured clockwise from the +y axis. The mirror angles are then tipped about the
rotated tip direction.
The Extra Data Values (EDV) are used to define the state of the rows/columns/
mirrors using a base 3 integer value. To determine the values for any logic state of
the MEMS, construct a table similar to the one below, which shows the values for 3
rows or columns or mirrors:
r/c/m: 3 2 1 EDV
0 0 0 0
0 0 1 1
0 0 2 2
0 1 0 3
0 1 1 4
0 1 2 5
0 2 0 6
etc..
Note the EDVA is given by:

EDV = M1*(3^0) + M2*(3^1) + M3*(3^2) + ....
where M1 is the logic state (0, 1, or 2) of the first row/column/mirror and M2 is the
logic state of the second row/column/mirror, etc. Up to 15 row/column/mirror values
are defined by each EDV.
US_OFFST The offset surface. This surface simulates an additional propagation of distance that
is hidden within the surface. The propagation distance is independent for each
wavelength. This allows modeling of systems that have independent focusing of
each color channel, such as a digital projector.
US_OGIVE Ogive surface shape. An ogive is identical to the standard surface, except the axis
of rotation of the surface is offset by an amount ro. The surface sag is given by
2
cr g
z = ---------------------------------------------------
- , where
2 2
1 + 1 ( 1 + k )c r g
2 2
rg = ro + x + y .

US_POLARIZATION This DLL shows how to define the electric field or coating data.
US_STAND This DLL is a clone of the ZEMAX "Standard" surface type. This is the simplest DLL
surface and is a good place to start the study of UDS DLLs.
US_STAND2 This DLL is a clone of the ZEMAX "Standard" surface type using the
UserDefinedSurface2 function and structure definition.
Variable Line Space Grating

The variable line space grating surface models a special grating with straight, but unevenly spaced lines. The
grating is described in M. Hettrick and S. Bowyer, "Variable line-space gratings: new designs for use in grazing
incidence spectrometers", Applied Optics, Vol. 22, No. 24, p3921 (1983). The grating is designed to form images
upon a concave spherical surface. A converging beam is used to illuminate the grating in reflection. The grating
lines are parallel to the local x axis, and lie in the x-z plane. Only plane gratings are supported, and the feature
has only been tested when the grating is used in reflection (the glass type must be "MIRROR").
Although the hologram surface type could be used to model the ideal grating, in practice holographic
construction is not always possible. A compromise is made in the fabrication, where the grating lines are straight
and parallel (as opposed to curved for the hologram case). This introduces aberrations into the beam, which this
surface model accounts for.
Five parameters are used to describe the surface: M, the diffraction order upon playback; L, the (positive)
radius of the concave focal surface, cos ( 0 ) , the cosine of the angle the incident beam makes with the plane of
the grating, cos ( 0 ) , the cosine of the angle the reflected beam makes with the plane of the grating, and 0 , the
construction wavelength. This is the wavelength at which the grating will reflect the axis ray along the angle 0 .
The grating frequency in lines per micrometer is given by
1 L cos 0 y L cos 0 y
T = ----- ---------------------------------------------------- ---------------------------------------------------- ,
0 2 2 2 2
y + L 2Ly cos 0 y + L 2Ly cos 0
where y varies across the grating surface. The variable line space grating surface uses five of the parameter
surfaces, but none of the extra data surfaces.
PARAMETER DEFINITIONS FOR VARIABLE LINE SPACE GRATING SURFACES
M L cos 0 cos 0 0
Zernike Fringe Phase

The Zernike Fringe Phase surface has a substrate shape identical to the Standard surface (which supports
planes, spheres, and conics) plus additional phase terms defined by the Zernike Fringe coefficients. The surface
sag is identical to the standard surface formula. The additional phase terms deviate and add optical path to the
rays as they cross the surface. This surface model is well suited to modeling system aberrations for which
measured interferometer data is available. The Zernike Fringe phase surface can also be used to model some
holograms and binary optics surfaces. The phase of the surface is given by
= M 2A i Z i ( , )
i=1

th
where N is the number of Zernike coefficients in the series, A i is the coefficient on the i Zernike Fringe
polynomial, is the normalized radial ray coordinate, is the angular ray coordinate, and M is the diffraction
order. The Zernike Fringe polynomials are defined in the table given in Zernike Fringe Coefficients on page 165.
ZEMAX supports up to 37 terms. The coefficients A i all have units of waves. One wave is 2 radians. Note that
the phase of the surface is independent of the wavelength; ZEMAX accounts for the deviation in the ray path
based upon the wavelength being traced. If the "Extrapolate" flag is set to 0, the Zernike terms are ignored outside
of the normalization radius. If the "Extrapolate" flag is set to 1, then the Zernike terms are considered no matter
where the ray lands on the surface; even if the ray lands beyond the normalization radius.
Note that the Zernike Fringe Phase surface describes phase variations (or wavefront error), not surface
deformations directly. If you have Zernike coefficient data in terms of surface deformations, as might have been
measured with a profilometer, see Zernike Fringe Sag on page 283. See also Zernike Standard Phase on
page 284.
PARAMETER DEFINITIONS FOR ZERNIKE FRINGE PHASE SURFACES
Diffraction order Extrapolate
EXTRA DATA DEFINITIONS FOR ZERNIKE FRINGE PHASE SURFACES

1 Number of terms.
2 Normalization radius. Coordinates are normalized by this value.
3 - 39 Coefficients on Zernike polynomials 1 - 37, respectively, in units of primary waves.
The "Number of terms" is used to specify the maximum Zernike polynomial term to be used in calculating the
phase. This number is provided to speed the ray tracing calculation; terms beyond this number are ignored.
Zernike polynomials are orthogonal over the unit circle, and so the normalization radius should be set to the
radius over which the coefficient data was normalized. Zernike polynomials tend to diverge quite rapidly beyond
the normalization radius, and so care should be taken that rays do not strike the surface beyond this radius.
Although the ray tracing algorithm may work, the data may be inaccurate. The extrapolate flag may be set to zero
to ignore the Zernike terms for rays that land outside the normalization radius.
Zernike Fringe Phase coefficients sign conventions
Zernike Fringe Sag
The Zernike Fringe Sag surface is defined by the same polynomial as the Even Aspheric surface (which
supports planes, spheres, conics, and polynomial aspheres) plus additional aspheric terms defined by the Zernike
Fringe coefficients. The surface sag is of the form:
8 N
2
cr
z = -------------------------------------------------- + i r + A i Z i ( , )
2i
2 2
1 + 1 ( 1 + k )c r i = 1 i=1
th
where N is the number of Zernike coefficients in the series, A i is the coefficient on the i Zernike Fringe
polynomial, r is the radial ray coordinate in lens units, is the normalized radial ray coordinate, and is the

angular ray coordinate. The Zernike Fringe polynomials are defined in the table Zernike Fringe Coefficients on
page 165. ZEMAX supports up to 37 Zernike Fringe terms. The coefficients A i all have units which are the same
as the lens units, such as millimeters or inches. The coefficients i have units, and are as defined in the section
Even Asphere on page 243. If the "Extrapolate" flag is set to 0, the Zernike terms are ignored outside of the
normalization radius. If the "Extrapolate" flag is set to 1, then the Zernike terms are considered no matter where
the ray lands on the surface; even if the ray lands beyond the normalization radius.
Note that the Zernike Fringe Sag surface describes surface deformations, not wavefront error directly. If you
have Zernike coefficient data in terms of waves of OPD, as might have been measured with an interferometer,
see Zernike Fringe Phase on page 282. Also see Zernike Standard Sag on page 285.
PARAMETER DEFINITIONS FOR ZERNIKE FRINGE SAG SURFACES

0 1 2 3 4 5 6 7 8
Extrapolate 1 2 3 4 5 6 7 8
EXTRA DATA DEFINITIONS FOR ZERNIKE FRINGE SAG SURFACES

1 Number of terms.
3 - 39 Coefficients on Zernike polynomials 1 - 37, respectively, in lens units.
surface sag. This number is provided to speed the ray tracing calculation; terms beyond this number are ignored.
Zernike Standard Phase
The Zernike Standard phase surface has a substrate shape identical to the Standard surface (which supports
planes, spheres, and conics) plus additional phase terms defined by the Zernike Standard coefficients. The
surface sag is identical to the standard surface formula. The additional phase terms deviate and add optical path
to the rays as they cross the surface. This surface model is well suited to modeling system aberrations for which
measured interferometer data is available. The Zernike Standard Phase surface can also be used to model some
holograms and binary optics surfaces. The phase of the surface is given by
= M 2A i Z i ( , )
i=1
th
where N is the number of Zernike coefficients in the series, A i is the coefficient on the i Zernike Standard
polynomial, is the normalized radial ray coordinate, is the angular ray coordinate, and M is the diffraction
order. The Zernike Standard polynomials are defined in the table given in Zernike Standard Coefficients on
page 168. ZEMAX supports up to 231 terms. The coefficients A i all have units of waves. One wave is 2 radians.

Note that the phase of the surface is independent of the wavelength; ZEMAX accounts for the deviation in the
ray path based upon the wavelength being traced. If the "Extrapolate" flag is set to 0, the Zernike terms are
ignored outside of the normalization radius. If the "Extrapolate" flag is set to 1, then the Zernike terms are
considered no matter where the ray lands on the surface; even if the ray lands beyond the normalization radius.
Note that the Zernike Standard Phase surface describes phase variations (or wavefront error), not surface
deformations directly. If you have Zernike coefficient data in terms of surface deformations, as might have been
measured with a profilometer, see Zernike Standard Sag on page 285.
PARAMETER DEFINITIONS FOR ZERNIKE STANDARD PHASE SURFACES
Diffraction order Extrapolate
EXTRA DATA DEFINITIONS FOR ZERNIKE STANDARD PHASE SURFACES

1 Number of terms.
3 - 233 Coefficients on Zernike polynomials 1 - 231, respectively, in units of primary waves.
phase. This number is provided to speed the ray tracing calculation; terms beyond this number are ignored.
Zernike Standard Phase coefficients sign conventions
Zernike Standard Sag
The Zernike Standard Sag surface is defined by the same polynomial as the Even Aspheric surface (which
supports planes, spheres, conics, and polynomial aspheres) plus additional aspheric terms defined by the Zernike
Standard coefficients. The surface sag is of the form:
8 N
2
cr
z = -------------------------------------------------- + i r + A i Z i ( , )
2i
2 2
1 + 1 ( 1 + k )c r i = 1 i=1
th
where N is the number of Zernike coefficients in the series, A i is the coefficient on the i Zernike Standard
polynomial, r is the radial ray coordinate in lens units, is the normalized radial ray coordinate, and is the
angular ray coordinate. The Zernike Standard polynomials are defined in the table given in Zernike Standard
Coefficients on page 168. ZEMAX supports the first 231 Zernike Standard terms. The coefficients A i all have
units which are the same as the lens units, such as millimeters or inches. The coefficients i have units, and are
as defined in the section on the "Even Asphere" surface type. If the "Extrapolate" flag is set to 0, the Zernike terms

are ignored outside of the normalization radius. If the "Extrapolate" flag is set to 1, then the Zernike terms are
considered no matter where the ray lands on the surface; even if the ray lands beyond the normalization radius.
Note that the Zernike Standard Sag surface describes surface deformations, not wavefront error directly. If you
have Zernike coefficient data in terms of waves of OPD, as might have been measured with an interferometer,
use the Zernike Standard Phase surface instead.
PARAMETER DEFINITIONS FOR ZERNIKE STANDARD SAG SURFACES

0 1 2 3 4 5 6 7 8
Extrapolate 1 2 3 4 5 6 7 8
EXTRA DATA DEFINITIONS FOR ZERNIKE STANDARD SAG SURFACES

1 Number of terms.
3 - 233 Coefficients on Zernike polynomials 1 - 231, respectively, in lens units.
surface sag. This number is provided to speed the ray tracing calculation; terms beyond this number are ignored.
Zone Plate
This surface is used to model a Fresnel Zone Plate (FZP). A FZP consists of a refractive or reflective surface
upon which annular zones of varying depth are cut or etched. Typically, the spacing between grooves is large
compared to the wavelength, and therefore FZP's are entirely refractive, and not diffractive, devices. The varying
depth of the annular zones imparts a slight phase change to different zones of the beam, due to the change in
path length through the material.
The FZP is modeled by altering the local depth of the surface. Edge effects, such as rays striking the side of
the zone, and diffraction from the edges, are ignored. The FZP is defined by a normalized optical thickness of
each zone. Each zone is spaced a fixed radius apart, and up to 250 zones may be defined. Three parameter
values are required: the mode, the r , and the reference wavelength, 0 .
If the mode is zero, the steps are not interpolated, and each zone has a constant optical thickness. This results
in discontinuous phase steps, like a staircase. If the mode is unity, then the optical thickness is interpolated
between adjacent zones. Although the phase will be continuous, the first derivative of the phase will generally be
discontinuous at the zone boundaries. The " r " is the step in radius between adjacent zones in lens units. The
"reference wavelength" is the wavelength in micrometers used to normalize the optical thickness. For example,
if the FZP defines the boundary between a material with index 1.55 and air, with index 1.00, and the reference
wavelength is 0.45 micrometers, then an optical thickness of 1 wave corresponds to a groove depth of
0 0.45
T = ---------------- = -------------- = 0.818182 .
n2 n1 0.55
This thickness is the actual groove depth if the optical thickness is given to be 1.0. ZEMAX uses this normalized
optical thickness for convenience. To specify a zone that introduces one-half wave of phase at the reference

wavelength, enter an optical thickness of 0.5. At other wavelengths, the phase will be greater or less depending
upon the dispersion of the two adjacent media.
The optical thickness is defined at n points, separated by r : r 1, r 2, r 3, r n . The number of points, "n", is
specified in the first column of the Extra Data Editor. The remaining terms are defined in columns 2 through n+1
in the Extra Data Editor.
PARAMETER DEFINITIONS FOR ZONE PLATE SURFACES

Mode r Ref. Wave
EXTRA DATA DEFINITIONS FOR ZONE PLATE SURFACES

1 Number of points
2 Optical thickness at r
3 Optical thickness at 2 r
4 Optical thickness at 3 r
n+1 Optical thickness at n r

Chapter 12 NON-SEQUENTIAL COMPONENTS
Introduction
Most imaging systems are well described by optical surfaces which are sequential, which means that rays
always trace from the object surface to surface 1, then 2, then 3, etc... in a strict sequence. Each ray "hits" each
surface once and only once in this predetermined sequence. Rays in sequential mode never trace from surface
3 to 9 then back to 1, for example. The sequential model is simple, numerically fast, and extremely useful and
complete for many important cases.
However, there are times when a non-sequential trace is required. Non-sequential means the rays trace in the
actual physical order they hit various objects or surfaces, and not necessarily in the order the objects are listed
in the software user interface. Note rays in a non-sequential trace may hit the same object repeatedly, and entirely
miss other objects. Generally, the order in which objects are hit by rays depends upon the object geometry and
the angle and position of the input ray. Objects which require or at least benefit from non-sequential ray tracing
include faceted objects, prisms, light pipes, lens arrays, reflectors, and Fresnel lenses. Certain types of analysis,
such as stray or scattered light effects, are only practical in a completely non-sequential environment.
Traditionally, lens design programs that supported surfaces (rather than 3D objects) for sequential ray tracing
would implement non-sequential ray tracing using the same surface model; the rays would simply intersect
surfaces in a possibly out of sequence order.
The disadvantage of using surfaces in a more general non-sequential approach is that surfaces do not
adequately describe most optical components. For example, lenses not only have a front and back surface, they
also have edges and perhaps flattened outer faces for mounting. Light may intercept, and then refract or reflect
from these additional surfaces normally ignored by sequential surface ray tracing codes. Complex prisms, such
as a dove or roof prism, contain many faces, and the rays may intersect these faces in a complex order depending
upon the input angle and position of the ray.
To support these types of components in a very general and accurate way requires the use of full 3D solid
object models instead of just 2D surfaces. ZEMAX calls this type of ray tracing non-sequential components, or
NSC, which is different from non-sequential surfaces or NSS. NSC ray tracing in ZEMAX supports all of the
following features:
Definition and placement of multiple sources, objects and detectors.
Real radiometric and photometric units; including watts, lumens, lux, phot, footcandles, and others.
Automatic determination of ray-object intersection order.
Automatic detection of reflection, refraction, and total internal reflection (TIR).
Support for a very wide range of 3D objects, including diffractive optics.
Polarization ray tracing and arbitrary thin film coatings.
Statistical models for scattering, including Lambertian, Gaussian, and ABg.
Automatic ray splitting for efficient analysis.
This chapter provides information on setting up a NSC group, defining objects, and tracing rays through NSC.
Paraxial data and ray tracing with NSC
There is no paraxial ray tracing in NSC. When paraxial rays are traced to a non-sequential surface, the
equivalent real rays are traced instead. Nearly all paraxial data, such as focal length and F/#, are meaningless
for non-sequential systems. Some NSC systems are well behaved in the sense that real rays close to the axis
behave much as idealized paraxial rays would. However, NSC systems with faceted parts, obscurations, and
objects only partially in the path of the beam will not be able to correctly predict any paraxial data, and paraxial
data should be considered meaningless for these systems.
289 Chapter 12: NON-SEQUENTIAL COMPONENTS

Optimization with NSC
See Optimizing objects in a non-sequential group with sequential rays on page 427 and Optimizing with
sources and detectors in non-sequential mode on page 427.
The two methods of using NSC ray tracing
ZEMAX supports two distinct methods of using NSC ray tracing:
Tracing rays through an NSC group that is part of an otherwise sequential system (NSC with ports).
Tracing rays through an NSC group that contains all the objects of interest (NSC without ports).
Although the method of defining and placing objects within a NSC group is the same for both methods, the
details of how rays are launched, what analysis may be performed, how energy distributions are determined, and
what types of systems are best modeled by each method, are considerably different. NSC with and NSC without
ports are described below.
NSC with ports
When non-sequential object or objects are part of an otherwise sequential system, NSC with ports is generally
the best approach to use. An example of this system would be a point or extended surface object represented by
rays which follow a sequential path through one or more conventional lenses, then follow a non-sequential path
through a prism or light pipe before illuminating the image surface.
This method requires the use of ports for rays to enter and leave the NSC group. Ports are described in detail
in the section "Overview of NSC ray tracing with ports" below. When using ports, rays are launched from defined
field positions on the object surface, and these rays ignore source and detector objects within the NSC group.
Rays must leave the NSC group via the exit port; then continue through the remainder of the sequential system.
When using analysis features such as ray fans, spot diagrams, and MTF, only rays which enter and leave the
NSC group through ports are considered. All the usual sequential ZEMAX system data, such as field positions
and pupil sizes, determine the properties of rays entering the NSC group. All normal ZEMAX analysis, such as
ray fans and spot diagrams, are still available (although the data may be meaningless depending upon the
properties of the NSC system).
NSC without ports
Some systems have no sequential paths or portions at all; such as headlamp reflectors, complex light pipes,
or general illumination systems. It is also possible to analyze ghost, stray, and scattered light properties of
nominally sequential systems (such as a camera lens or telescope) by placing the entire system in a non-
sequential group and performing non-sequential ray tracing on the entire model. These types of systems and
analysis are better suited to NSC without ports.
This method requires the use of NSC sources to launch rays. When not using ports, rays are launched only
from defined NSC sources, and these rays consider detector objects within the NSC group.
Since rays launched from sources within the NSC group cannot leave through the exit port; the only available
analysis is the ray distribution and energy as determined using detector objects.
Combining NSC with and without ports
The 3D Layout, Wireframe, Solid Model, and Shaded Model analysis features are all capable of simultaneously
displaying BOTH rays coming from the sequential entry port as well as any sources defined for use in NSC without
ports. Rays entering from the entry port do not interact with any NSC detectors; and rays launched from NSC
sources do not interact with the entry port, the exit port, or any optics defined outside the NSC group.
Therefore it is advised, though not required, that when using NSC without ports (i.e. you have defined sources
among your NSC objects) that you set the number of rays in these layout features to zero. This will make for a
less confusing display. There are times when showing the sequential rays coming from the entry port and non-
sequential rays on the same plot is useful, especially when placing baffles in a stray light analysis.
Alternatively, there is a 3D Layout feature which is specific to NSC, which shows only objects and rays defined
within a single NSC group.
Chapter 12: NON-SEQUENTIAL COMPONENTS 290

Overview of NSC ray tracing with ports
Ray tracing through a group of NSC objects using ports is accomplished with the following basic steps:
1) A Non-Sequential Components surface is inserted in the Lens Data Editor. This surface becomes the entry
port for the non-sequential group.
2) The Non-Sequential Components surface parameters are used to define the location of the exit port for the
non-sequential group.
3) Objects are defined in a list associated with the Non-Sequential Components surface.
4) ZEMAX traces a ray sequentially to the entry port, then non-sequentially within the NSC group, until the ray
strikes the exit port.
5) Rays entering the NSC group through the entry port cannot split.
The Non-Sequential Components surface has parameters which determine where the rays will exit the NSC
group as described below.
The entry port
The Non-Sequential Components surface acts like a plane, sphere, or conic aspheric surface whose location
is determined by the previous surfaces in the Lens Data Editor, in the usual way. The surface shape may be
hyperhemispheric to allow acceptance of rays over a full 4 steradians. The Non-Sequential Components
surface is the entry port into a group of objects which will be traced non-sequentially. The entry port is how rays
get into the NSC group.
The exit port
There are 9 parameters used in the definition of the Non-Sequential Components surface:
Draw Ports?: If 0, no ports are drawn, if 1, draw entry, 2, draw exit, 3 draw both.
Exit Location X: The x position of the exit port relative to the entry port.
Exit Location Y: The y position of the exit port relative to the entry port.
Exit Location Z: The z position of the exit port relative to the entry port.
Exit Tilt About X: The rotation about the local X axis of the exit port.
Exit Tilt About Y: The rotation about the local Y axis of the exit port.
Exit Tilt About Z: The rotation about the local Z axis of the exit port.
Order: If the order flag is zero, then the above locations and tilts are done in the following order: decenter x,
decenter y, decenter z, rotate around global z, rotate around global y, rotate around global x. If the flag is any
value other than zero, then the order is reversed. This follows the same convention as the sequential
coordinate break surface when using order flag = 0 (see page 239).
Reverse Rays: If this flag is 0, then ZEMAX assumes the non-sequential group acts like a refractive lens. If
this flag is 1, then ZEMAX assumes the non-sequential group acts like a mirror. For example, if rays entering
the non-sequential group travel in a positive direction with respect to the local z axis, and leave the exit port
still traveling in the positive local z direction, the flag should be zero. If the rays reverse direction relative to the
incoming direction, then the reverse rays flag should be 1.
The diameter of the circular exit port in lens units. This value is defined by the semi-diameter of the surface
following the Non-Sequential Components surface. Note any additional aperture may be placed on the exit
surface if another aperture shape is required.
These parameters define the location and size of the exit port relative to the entry port. If the exit port is located
exactly at the entry port, then rays will immediately exit the non-sequential group without striking any objects. This
generally means the exit location z parameter must not be zero; although this is the default value when the non-
sequential surface is first created.
The glass column of the Non-Sequential Components surface is also used to define the "background" material
and index of refraction of the media in which NSC objects are placed. The surface after the Non-Sequential
Components surface acts like a plane surface oriented in the coordinate system after the decenters and tilts have
been applied. However, the surface is never drawn, and cannot be used to define the boundary between two

media. The glass type will always be the same as the prior surface, and ZEMAX will display "-" for the glass name,
which is meant to indicate that a glass type cannot be entered there.
Note the exit port position is the same as the surface following the Non-Sequential Components surface, and
its location in 3D space is determined by the parameters of the Non-Sequential Components surface. The
thickness of the Non-Sequential Components surface is not used; only the location and tilt parameter values. The
exit port should not be placed at the same location as, or within the glue distance (Glue Distance In Lens Units
on page 95) of the entrance port; otherwise rays entering the NSC group will immediately strike the exit port and
not intersect any of the objects within.
Getting rays in
A ray leaves the object surface, and traces through the lens in the usual sequential fashion until it reaches the
Non-Sequential Components surface. The ray is then sent into the group of components associated with that
surface, and the non-sequential tracing begins.
Tracing rays within the NSC
Once inside the NSC group, 3 things can happen to a ray:
1) It can hit the exit port.
2) It can hit nothing at all.
3) It can hit one of the objects within the group.
If the ray hits the exit port, the ray coordinates and the direction cosines are computed on the exit port, and
the ray then traces sequentially again through the remaining surfaces of the lens.
If the ray hits nothing at all, then the ray tracing is terminated, and the ray trace function returns a "ray missed"
error on the following surface (since the ray never struck the exit port, which is always the next surface in the
sequential portion of the ray trace).
If the ray strikes an object in the NSC group, then the ray will either reflect, refract, total internal reflect (TIR),
or be absorbed, depending upon the properties of the object struck. Rays entering the NSC group through the
entry port cannot split. If the ray is absorbed, the ray trace is terminated and a ray miss error is returned, otherwise
the new ray coordinates and direction cosines are computed, and the process repeats until one of the following
conditions is met:
1) The ray hits the exit port.
2) The ray hits no object.
3) The ray is absorbed.
4) The ray has intercepted more than the maximum allowed number of objects (see Maximum Intersections
Per Ray on page 94).
Cases 1, 2, and 3 are handled exactly as described above. In case 4, even though the ray technically still can
be traced, it is terminated to prevent infinite loops from occurring. In this case, the ray trace returns a ray miss
error.
Getting rays out
When a ray strikes the exit port, the coordinates and direction cosines of the ray in the coordinate system of
the exit port are computed, and then the ray traces sequentially through any remaining surfaces. If one of the
following surfaces is another Non-Sequential Components surface, then the process begins again for the
components defined for that group. Note rays within one NSC group cannot "see" objects defined in another
group, even if they physically share the same location in space; nor can the rays "see" surfaces outside the
present NSC group.
Overview of NSC ray tracing without ports
Ray tracing through a group of NSC objects without using ports is accomplished with the following basic steps:
1) Choose File, Non-Sequential Mode from the main menu.
2) Insert sources, objects, and detectors in the Non-Sequential Components editor.

There are no entry or exit ports to consider; these are ignored when using NSC without ports. The only data
that needs to be defined outside the NSC Editor are:
The wavelengths to be used for ray tracing (on the wavelength data editor)
The glass catalogs to be used (on the general system data dialog box)
The coating definitions (in the coating file associated with the current lens)
Getting rays in
To get rays into the NSC group, define one or more sources. ZEMAX supports point, rectangular, elliptical,
user defined, and other source models. Each source object has the following parameters defined (and perhaps
other parameters as well):
# Layout Rays: Defines how many random rays to launch from the source when creating layout plots.
# Analysis Rays: Defines how many random rays to launch from the source when performing analysis. See
Choosing the number of analysis rays on page 365.
Power (units): Power is the total power over the defined range of the source. The power units are specified by
the system source units. See Source Units on page 87 for details.
Wavenumber: The wavenumber to use when tracing random rays. Zero means polychromatic; which chooses
ray wavelengths randomly with the weighting defined on the wavelength data dialog box.
Note multiple sources may be superimposed with different powers and wavenumbers to create correct
polychromatic sources. Sources may be placed anywhere without restriction (even inside objects).
Once the ray is launched, non-sequential tracing begins.
Tracing rays within the NSC
Once inside the NSC group, 2 things can happen to a ray:
1) It can hit nothing at all.
2) It can hit one of the objects within the group.
If the ray hits nothing at all, then the ray tracing for this ray is terminated.
If the ray strikes an object, then the ray will either reflect, refract, total internal reflect (TIR), scatter, split, diffract,
or be absorbed; or a combination of these, depending upon the properties of the object struck.
Detecting rays
If the ray strikes a detector object; the pixel which the ray struck is determined, and the total pixel energy is
incremented by the ray energy. Detectors may be absorbing, reflecting, transmissive, or refractive.
The process repeats until one of the following conditions is met:
1) The ray hits no object.
2) The ray is absorbed.
3) The ray has intercepted more than the maximum allowed number of objects.
4) The total number of ray segments exceeds the maximum allowed number.
5) The relative or absolute energy of the ray falls below the minimum threshold.
The limitations on number of objects, ray segments, and ray energy are defined on the System dialog box, see
Non-Sequential on page 94. Cases 1 and 2 are handled exactly as described above. In case 3, 4, and 5, even
though the ray technically still can be traced, it is terminated to prevent infinite loops from occurring.
Launching rays for analysis
Rays may be launched, and detectors reset, from the Detectors, Ray Trace/Detector Control.
NSC Objects
ZEMAX NSC object types include ellipses, triangles, rectangles, spheres, cylinders, and other basic shapes.
Complex objects such as arbitrary prisms, aspheric lenses, torics, toruses, and other optical components are also

available. The reflective, refractive, and absorptive properties of these objects are determined by the material
assigned to the objects. For details on reflective, refractive, and absorptive properties, see a following section.
Each NSC object type is described in the following summary table, and in greater detail in the following
sections. Note these basic objects may be combined to form more complex objects. See Object Placement on
page 344 for information on placing objects inside or adjacent to one another.
!Detectors are described on page 332. Sources are described on page 336.
If an object type is required that is not listed, please contact technical support to suggest the new object type
be added to ZEMAX.
SUMMARY OF NSC OBJECTS
Object Name Description Page

Annular Aspheric Lens An annular volume with an aspheric surface on one face. 297
Annular Volume A volume formed by two tilted, elliptical annular planar faces. 297
Annulus A planar ellipse with an elliptical hole in the middle. 298
Aspheric Surface A surface with a conic substrate plus even and odd radial 298
polynomial power aspheric terms.
Aspheric Surface 2 Identical to the Aspheric Surface, with support for more complex 299
aperture types.
Axicon Surface An axicon formed by rotating a spherical arc around the z axis. 299
Biconic Lens A circular or rectangular lens with separate curvature and conic in 300
the X and Y directions on both front and back faces.
Biconic Surface A surface with separate curvature and conic in the X and Y 301
directions. May be partially hyperhemispheric under special
circumstances.
Binary 1 A standard lens with a binary 1 phase profile on the front face. The 302
binary 1 phase profile is an X-Y polynomial.
Binary 2 A standard lens with a binary 2 phase profile on the front face. The 303
binary 2 phase profile is a radial polynomial in even powers of r.
Binary 2A An even aspheric lens (front and back faces) with a binary 2 phase 303
profile on the front face. The binary 2 phase profile is a radial
polynomial in even powers of r.
Cone A section of a cone defined by two points, r and z, which forms a 304
line segment, which is then rotated around the local z axis.
Compound Parabolic A compound parabolic concentrator (CPC). The CPC may be 305
Concentrator (CPC) hollow or solid.
Cylinder Pipe A tapered cylindrical surface shape. 305
Cylinder Volume A tapered cylindrical volume, with end caps. 306
Cylinder 2 Pipe A cylindrical surface shape with tilted ends. 306
Cylinder 2 Volume A cylindrical volume, with tilted end caps. 306
Diffraction Grating A standard lens with a diffraction grating with constant line spacing 307
on the front face.
Ellipse A planar ellipse. 307

Elliptical Volume An elliptical volume (or shell). The shape is a tapered cylinder with 307
an elliptical cross section.
Even Asphere Lens A rotationally symmetric lens with up to 16th order aspheres on 308
both the front and back faces.
Extended Polynomial Lens A lens with extended polynomial surfaces on front and back faces. 308
Extended Polynomial Surface A surface with an extended polynomial shape. 309
Fresnel 1 A true exact (not faceted) Fresnel lens, either radial or cylindrical, 309
defined in terms of a number of construction parameters.
Fresnel 2 An idealized Fresnel lens object having a circular or rectangular 310
shape, with one face an ideal Fresnel lens with either radial or
cylindrical polynomial aspheric terms over a base conic asphere.
Hexagonal Lenslet A lenslet array with hexagonal symmetry. 312
Hologram Lens A circular or rectangular solid with an optically fabricated 312
hologram on the front face.
Hologram Surface A circular or user defined plane, conic, spherical or aspherical 313
surface with an optically fabricated hologram on the face.
Imported An object or objects imported from a CAD program in either IGES 314
or STEP format.
Jones Matrix An elliptical flat object which supports 8 ABCD parameters which 315
can be use to model neutral density filters, polarizers, rotators,
and other arbitrary Jones Matrix type objects.
Lenslet 1 A rectangular shaped lens object with a curved back face. This 315
object can be used to model an array element, and may be
diffractive.
Lenslet 2 A rectangular shaped lens object with curved front and back faces. 317
This object can be used to model an array element.
MicroElectroMechanical An array of flat facets (intended to be made of material "MIRROR") 317
System (MEMS) where the individual facets may be tipped about any of 3 angles.
The facets may be addressed either by rows, by columns, or by
individual pixels.
Null Object A non-existent object. This is the default when new objects are 318
inserted into the NSC Editor, and is handy to use as a place holder.
Odd Asphere Lens A rotationally symmetric lens with up to 12th order odd and even 318
power aspheres on both the front and back faces.
Polygon Object A user defined general polygon object which may be open or 319
closed, and may have refractive, reflective, or absorptive faces, or
a mixture of these types of faces.
Rectangular Corner A corner composed of 3 rectangular surfaces. 319
Rectangle A rectangle defined by a width and height. 320
Rectangular Pipe A four sided box. 320
Rectangular Pipe Grating A four sided box with a grating on the side faces. 320
Rectangular Roof A roof composed of two rectangular surfaces meeting at an angle. 321

Rectangular Torus Surface A section of a rectangular torus which defines a surface; open at 321
the ends.
Rectangular Torus Volume A section of a rectangular torus with closed ends to define a 321
volume.
Rectangular Volume A six sided box enclosing a volume. 322
Rectangular Volume Grating A six sided box with a diffraction grating on side faces. 322
Slide A color RGB file, in either BMP or JPG format. The slide may be 322
scaled to any size and aspect ratio, and may be used to filter a
source to create colored image sources.
Sphere A sphere. 323
Standard Lens A 5 surface lens object with two Standard faces, two flat edges, 323
and a cylindrical outer edge.
Standard Surface A surface with the same shape as the ZEMAX Standard surface 324
(a plane, sphere, conic asphere, or hyperhemisphere).
STL Object A general polygon object defined in STL format, typically exported 324
from a CAD program.
Tabulated Faceted Radial A faceted object generated from a table of coordinates rotated 325
about the local z axis. This object type can be used to define
surfaces or volumes.
Tabulated Faceted Toroid A faceted object generated from a table of coordinates rotated 326
about an axis parallel to the local y axis. This object type can be
used to define surfaces which are cylindrical or toroidal.
Tabulated Fresnel Radial A smooth Fresnel lens generated from a table of coordinates. This 327
is very similar to the Tabulated Faceted Radial object except the
surfaces are smooth, not faceted. See also the Fresnel object.
Toroidal Lens A circular, elliptical, or rectangular lens with aspheric toroidal 327
surfaces on the front and back sides.
Toroidal Surface A rectangular surface with an aspheric toroidal shape. 327
Torus Surface A section of a torus which defines a surface; open at the ends. 328
Torus Volume A section of a volume torus; like a surface but with end caps to 328
enclose a volume.
Triangular Corner A corner composed of 3 triangles. 329
Triangle A triangle defined by 3 points in a plane. 329
User Defined Object An object defined by a user provided DLL. 329
Zernike Surface A surface defined by radial aspheric and standard Zernike 331
polynomial sag terms.
Most objects require specification of various parameters. The parameters required by each object type and
detailed descriptions of each object follow.

Annular Aspheric Lens
The annular aspheric lens is an annular solid with a plano front
face. The rear face is defined by a conic asphere with even
aspheric polynomial terms to 16th order, identical to the Even
Asphere Lens object (see Even Asphere Lens on page 308). The
annular shape is defined by a min and max radial aperture. The
aspheric rear face is offset from the front by a thickness measured
at a specified radial aperture. The object is defined by the following
parameters:
1. The number of angular facets
2. The number of radial facets
3. The Minimum Radial Aperture
4. The Maximum Radial Aperture
5. The Thickness Aperture, which is the radial aperture to measure the object thickness
6. Radius
7. Conic
8. Thickness (measured at the Thickness Aperture)
9-16. The Even Aspheric coefficients alpha2 through alpha16.
Annular Volume
The annular volume consists of tilted annular elliptical faces front and rear, separated by an axial distance.
The defining parameters are:
1: The number of angular facets. See The use of facets on page 331.
2: The number of radial facets. See The use of facets on page 331.
3: The front face minimum x coordinate.
4: The front face minimum y coordinate.
5: The rear face minimum x coordinate.
6: The rear face minimum y coordinate.
7: The front face maximum x coordinate.
8: The front face maximum y coordinate.
9: The rear face maximum x coordinate.
10: The rear face maximum y coordinate.
11: The length.
12: The front face x tilt angle in degrees.
13: The front face y tilt angle in degrees.
14: The rear face x tilt angle in degrees.
15: The rear face y tilt angle in degrees.
Coating/Scatter Groups: Outside face connecting the "maximum" apertures CSG #0, front face CSG #1, back
face CSG #2, inside face connecting the "minimum" apertures on the front and back CSG #3.

Annulus
The annulus is a plane elliptical surface shape defined by 4
parameters:
1: The X Maximum Half Width.
2: The Y Maximum Half Width.
3: The X Minimum Half Width.
4: The Y Minimum Half Width.
The annulus resides entirely within the local XY plane. If the
minimum half widths are set to zero, then no "hole" exists and rays
will hit anywhere within the maximum half widths. In this case, the
annulus is identical to the ellipse.
The reference coordinate is the center of the annulus. Coating/
Scatter Groups: All faces CSG #0.
Aspheric Surface
An aspheric surface is defined by the following sag equation:
N
2
cr
z = -------------------------------------------------- + i r ,
i
2 2
1 + 1 ( 1 + k )c r i=1
where c is the curvature of the surface, k is the conic constant, r is the radial coordinate, and the terms are
aspheric coefficients. The surface supports specification of both a minimum and a maximum radial aperture; so
annular surfaces may be defined. Note that both even and odd terms are defined; up to approximately 240
coefficients may be used.
The following parameters are used to define the aspheric sur-
face:
1: The radius of curvature. If this value is zero, then the curva-
ture is assumed to be zero.
2: The conic constant k.
3: The maximum radial aperture in lens units.
4: The minimum radial aperture in lens units. This value may be
zero.
5: The number of angular facets. See The use of facets on
page 331.
6: The number of radial facets. See The use of facets on
page 331.
7: The number of terms to use in the aspheric expansion. Ray tracing will be faster if this term is no larger than
the highest order non-zero coefficient.
8-250: The coefficients on the polynomial expansion.
Coating/Scatter Groups: All faces CSG #0. This object supports user defined apertures, see User defined
apertures on page 352. For other aperture shape options, see Aspheric Surface 2 on page 299.

Aspheric Surface 2
This surface is similar to the Aspheric Surface above, with
added support for elliptical or rectangular decentered annular ap-
ertures. The following parameters are used to define the aspheric
surface:
3: The maximum X direction aperture in lens units.
4: The maximum Y direction aperture in lens units.
5: The minimum X direction aperture in lens units. This value
may be zero.
6: The minimum Y direction aperture in lens units. This value
may be zero.
7: The X direction aperture decenter in lens units.
8: The Y direction aperture decenter in lens units.
9: The "Is Rectangle" flag. Use 0 for elliptical symmetry, 1 for rectangular symmetry.
10: The number of angular or X facets. See The use of facets on page 331.
11: The number of radial or Y facets. See The use of facets on page 331.
12: The number of terms to use in the aspheric expansion. Ray tracing will be faster if this term is no larger
than the highest order non-zero coefficient.
Coating/Scatter Groups: All faces CSG #0.
Axicon Surface
There are many types of axicons. Common polynomial types
can be modeled using the Aspheric Surface (see Aspheric Sur-
face on page 298).
The Axicon Surface object uses a different method for defining
the object shape. The Axicon surface is defined by a section of a
circle (an arc) that lies in the YZ plane. The arc is defined by three
points. The start of the arc is at (z=0, y=0). The center of the arc is
at the user defined coordinates (cz, cy). The end of the arc is at the
point where z = L, where L is a user defined parameter correspond-
ing to the length of the axicon along the Z axis. The surface is
formed by revolving the arc around the Z axis.
The following parameters are used to define the axicon surface:
1: The length of the axicon along the local z axis.
2: The z coordinate of the center of the arc measured in the YZ plane.
3: The y coordinate of the center of the arc measured in the YZ plane.
4: The number of angular facets. See The use of facets on page 331.
5: The number of z direction facets. See The use of facets on page 331.

Biconic Lens
A biconic lens is similar to a toroidal lens, except the conic constant and base radius may be different in the X
and Y directions. The biconic lens allows specification of Rx, Ry, Kx, and Ky for the front and back surfaces
independently. The sag of a biconic surface is given by:
2 2
cx x + cy y
z = ---------------------------------------------------------------------------------------- ,
2 2 2 2
1 + 1 ( 1 + k x )c x x ( 1 + k y )c y y
where
1 1
c x = ----- , c y = ----- .
Rx Ry
The biconic lens is defined by these parameters:

1: The radial height of the lens object in lens units. This value is
used for the y direction half height if the lens is rectangular.
2: The x half width of the lens object in lens units. If zero, the
lens is circular; otherwise, the lens is rectangular.
3: The center thickness of the lens in lens units.
4: The maximum number of facets in the X direction if the
surface boundary is rectangular in shape, otherwise this is the
maximum number of angular facets. See The use of facets on
page 331.
5: The maximum number of facets in the Y direction if the
maximum number of radial facets. See The use of facets on page 331.
6: The base radius of curvature in the XZ plane for the front surface. If this value is zero, then the XZ curvature
is assumed to be zero.
7: The base radius of curvature in the YZ plane for the front surface. If this value is zero, then the YZ curvature
8: The X direction conic for the front surface.
9: The Y direction conic for the front surface.
10: The base radius of curvature in the XZ plane for the back surface. If this value is zero, then the XZ curvature
11: The base radius of curvature in the YZ plane for the back surface. If this value is zero, then the YZ curvature
12: The X direction conic for the back surface.
13: The Y direction conic for the back surface.
Coating/Scatter Groups: Front face CSG #1, back face CSG #2, all other faces CSG #0.

Biconic Surface
A biconic surface is similar to a toroidal surface, except the conic constant and base radius may be different
in the X and Y directions. The biconic surface allows specification of Rx, Ry, Kx, and Ky directly. The sag of a
biconic is given by:
2 2
cx x + cy y
z = ---------------------------------------------------------------------------------------- ,
2 2 2 2
1 + 1 ( 1 + k x )c x x ( 1 + k y )c y y
where
1 1
c x = ----- , c y = ----- .
Rx Ry
The biconic surface is defined by 13 parameters:

1: The base radius of curvature in the XZ plane. If this value is
zero, then the XZ curvature is assumed to be zero.
2: The base radius of curvature in the YZ plane. If this value is
zero, then the YZ curvature is assumed to be zero.
3: The X direction conic.
4: The Y direction conic.
5: The maximum X aperture in lens units.
6: The maximum Y aperture in lens units.
7: The minimum X aperture in lens units. This parameter is
ignored if the surface has elliptical symmetry.
8: The minimum Y aperture in lens units. This parameter is ignored if the surface has elliptical symmetry.
9: The maximum number of facets in the X direction if the surface boundary is rectangular in shape, otherwise
this is the maximum number of radial facets. See The use of facets on page 331.
10: The maximum number of facets in the Y direction if the surface boundary is rectangular in shape, otherwise
this is the maximum number of angular facets. See The use of facets on page 331.
11: The "Is Rectangle?" flag. If zero, the resulting surface shape will be elliptical. If non-zero, the surface will
have a rectangular boundary.
12: The "Is Top Hyper?" flag. If zero, the maximum Y aperture will lie on the non-hyperhemispheric portion of
the surface. If non-zero, the maximum Y aperture of the surface will be hyperhemispheric. This parameter is
ignored if the surface is elliptical, if the Y direction conic is less than or equal to -1.0, if the maximum Y aperture
is less than zero, or if the XZ base radius is non-zero.
13: The "Is Bot Hyper?" flag. If zero, the minimum Y aperture will lie on the non-hyperhemispheric portion of
the surface. If non-zero, the minimum Y aperture of the surface will be hyperhemispheric. This parameter is
ignored if the surface is elliptical, if the Y direction conic is less than or equal to -1.0, if the minimum Y aperture
is greater than zero, or if the XZ base radius is non-zero.

Making a hyperhemispheric surface
The biconic surface can also be used to make a hyperhemi-
spheric cylinder surface, but only under these conditions:
1) The Rx and Kx values are set to zero.
2) The "Is Rectangle?" flag is set to 1 (that is, the surface
boundary is a rectangle).
3) The Ky value is greater than -1.
For a surface that is hyperhemispheric on the top half, set the
maximum Y aperture to a positive value and set the "Is Top Hyper?"
flag to unity.
For a surface that is hyperhemispheric on the bottom half, set
the minimum Y aperture to a negative value and set the "Is Bot
Hyper?" flag to unity.
The surface may be hyperhemispheric in both the top and bottom portions at once, but currently only for the
case of Rx = 0.0. If either the top or bottom is hyperhemispheric, the actual maximum Y aperture is given by
2
2 ( Ry )
y = ------------------- .
( 1 + ky )
In this special case, a hyperhemispherical cylinder will result.

Binary 1
The Binary 1 object is a standard lens with a diffractive optic phase profile similar to the Binary 1 surface type
on the front face. The Binary 1 phase profile adds phase to the ray according to the following polynomial
expansion:
= M A i E i ( x, y ) ,
i=1
where N is the number of polynomial coefficients in the series, M is the diffraction order, and A i is the
th
coefficient on the i extended polynomial term. The polynomials are a power series in the normalized
coordinates x and y, as described in Extended Polynomial Lens on page 308. The coefficients A i all have units
of radians ( 2 radians is one wave).
The definitions for the parameters are:
1-9: See the Standard Lens description for information on these
parameters.
10: The diffraction order M.
11: The normalization radius. The x and y coordinates are
normalized to this value. This keeps all coefficients in units of
radians.
12: The maximum term number.
13-242: The values of the coefficients on the polynomial terms.
This object does not diffract rays correctly if the front surface is
made hyperhemispheric. See also the Binary 2 object.

For important information on diffractive objects, see Diffraction from NSC objects on page 347.
Binary 2
The Binary 2 object is a standard lens with a diffractive optic phase profile similar to the Binary 2 surface type
on the front face. The Binary 2 phase profile adds phase to the ray according to the following polynomial
expansion:
= M Ai ,
2i
i=1
th
which is the normalized radial aperture coordinate. The coefficients A i all have units of radians ( 2 radians is
one wave).
parameters.
11: The normalization radius. The x and y coordinates are
normalized to this value. This keeps all coefficients in units of
radians.
12: The maximum term number.
This object does not diffract rays correctly if the front surface is
made hyperhemispheric. See also the Binary 1 object.
Binary 2A
The Binary 2A object has an even polynomial aspheric front and back face, with a diffractive optic phase profile
similar to the Binary 2 surface type on the front face. The even aspheric sag formula is:
12
2
cr
z = -------------------------------------------------- + i r .
2i
2 2
1 + 1 ( 1 + k )c r i=1
The Binary 2A object consists of two of these faces, separated by a thickness.

The Binary 2A phase profile is placed on the front face, and adds phase to a ray according to the following
polynomial expansion:
= M Ai ,
2i
i=1

th
which is the normalized radial aperture coordinate. The coefficients A i all have units of radians ( 2 radians is
one wave).
1: The Maximum Radial aperture.
2: The Thickness of the lens at the center.
3: The number of angular Facets. See The use of facets on
page 331.
4: The number of radial Facets. See The use of facets on
page 331.
5: The front face radius of curvature.
6: The front face conic constant k.
7-18: The front face coefficients 1 12 .
19: The back face radius of curvature.
20: The back face conic constant k.
21-32: The back face coefficients 1 12 .
34: The normalization radius. The x and y coordinates are normalized to this value. This keeps all coefficients
in units of radians.
35: The maximum binary phase term number.
See also the Binary 1 and Binary 2 objects. For important information on diffractive objects, see Diffraction
from NSC objects on page 347.The reference coordinate is the center of the front face. Coating/Scatter Groups:
Front face CSG #1, back face CSG #2, all other faces CSG #0.
Cone
A cone is defined by 4 parameters:
1: The z coordinate of the first point.
2: The radial coordinate of the first point.
3: The z coordinate of the second point.
4: The radial coordinate of the second point.
The line segment defined by the points is rotated about the z
axis to form a section of a cone. This object can be used to make
an annular or circular shape (if the two z coordinates are identical)
or a cylindrical shape (if the two r coordinates are identical). In this
sense the cone is redundant with the annulus and cylinder pipe
objects. The cone is used as a primitive for creating Fresnel lenses.
The reference coordinate is locally (0, 0, 0), and the points that
define the cone may be placed anywhere relative to the reference point. Coating/Scatter Groups: All faces CSG
#0.

Compound Parabolic Concentrator (CPC)
A CPC is defined by 6 parameters:
1: The radial aperture at z = 0.
2: The maximum acceptance angle in degrees.
3: The length along the local Z axis.
page 331.
5: The number of length facets. See The use of facets on
page 331.
6: The "Is Volume?" flag.
A CPC is used to concentrate light entering one end of the CPC
to the other end. Only rays that make an angle less than the
acceptance angle with respect to the local Z axis will pass through the CPC; other rays will reflect back out. This
type of CPC is the "Basic CPC" as described in detail in "High Collection Nonimaging Optics" by W. T. Welford
and R Winston, Academic Press (1989).
If the "Is Volume?" flag is zero, then the object is a hollow shell. Otherwise, the object is a closed solid volume.
The maximum value for the CPC length is given by
a ( 1 + sin )
L = ---------------------------- ,
tan sin
where a is the radial aperture and is the acceptance angle. Longer lengths will be truncated to this value.
Coating/Scatter Groups: Front face CSG #1, rear face CSG #2, outer faces CSG #0.
Cylinder Pipe
A cylinder pipe is a rotationally symmetric surface defined by 3
parameters:
1: The radius of the front circular aperture.
2: The length along the local Z axis of the cylinder.
3: The radius of the rear circular aperture.
This object is normally used to define a reflective light pipe.
The reference coordinate is the center of the front aperture.

Cylinder Volume
A cylinder volume is a rotationally symmetric volume defined by
3 parameters:
1: The radius of the front circular face.
3: The radius of the rear circular face.
This object is very similar to the cylinder pipe, except the front
and rear faces are included to make the shape a closed volume.
Because the object is a closed volume, it may be reflective,
refractive, or absorbing.
The reference coordinate is the center of the front aperture.
Coating/Scatter Groups: Front face CSG #1, back face CSG #2, all
other faces CSG #0.
Cylinder 2 Pipe
A cylinder 2 pipe is a rotationally symmetric surface defined by
4 parameters:
1: The radius of the cylinder.
3: The tilt of the front face along the y direction in degrees.
4: The tilt of the rear face along the y direction in degrees.
5: The tilt of the front face along the x direction in degrees.
6: The tilt of the rear face along the x direction in degrees.
The reference coordinate is the center of the front face. Coating/
Cylinder 2 Volume
A cylinder volume is a rotationally symmetric volume defined by
4 parameters:
1: The radius of the cylinder.
3: The tilt of the front face along the y direction in degrees.
4: The tilt of the rear face along the y direction in degrees.
5: The tilt of the front face along the x direction in degrees.
6: The tilt of the rear face along the x direction in degrees.
This object is very similar to the cylinder 2 pipe, except the front
and rear faces are included to make the shape a closed volume.
The reference coordinate is the center of the front aperture. Coating/Scatter Groups: Front face CSG #1, back
face CSG #2, all other faces CSG #0.

Diffraction Grating
The diffraction grating is very similar to a Standard Lens, with 2
additional parameters:
parameters.
10: The grating line frequency in lines/micrometer on the front
face.
11: The diffraction order for the front face.
The grating is assumed to consist of equally spaced lines
parallel to the local x axis. The grating frequency is the lines per
micrometer along the y direction; projected down on to the surface.
This object does not diffract rays correctly if the front diffractive
surface is made hyperhemispheric.
Ellipse
The ellipse is a plane elliptical surface shape defined by 2
parameters:
The ellipse resides entirely within the local XY plane. This object
is a special case of the more general annulus.
The reference coordinate is the center of the ellipse. Coating/
Elliptical Volume
An elliptical volume is a tapered volume or surface with an
elliptical cross section defined by these parameters:
1: The x half width of the front face.
2. The y half width of the front face.
3: The x half width of the back face.
4. The y half width of the back face.
6: The number of z facets. See The use of facets on page 331.
page 331.
8: The "Is Volume?" flag. Use 1 to make a closed volume, else
use 0 to make a hollow shell.
This object is similar to the cylinder pipe and volume. Note this object may be either a hollow shell or a closed
volume, depending upon the Is Volume? flag.
The reference coordinate is the center of the front face. Coating/Scatter Groups: Front face CSG #1, back face
CSG #2, all other faces CSG #0.

Even Asphere Lens
The Even Asphere surface shape is defined by:
2
cr 2 4 6 8 10 12 14 16
z = -------------------------------------------------- + 1 r + 2 r + 3 r + 4 r + 5 r + 6 r + 7 r + 8 r ,
2 2
1 + 1 ( 1 + k )c r
which is exactly the same sag formula as for the Even Asphere surface. The Even Asphere Lens object
consists of two of these faces, separated by a thickness. The total object shape is defined by 24 parameters:
page 331.
page 331.
16: The back face conic constant k.
Extended Polynomial Lens
The Extended Polynomial surface shape is defined by:
N
2
cr
z = -------------------------------------------------- + A i E i ( x, y ) .
2 2
1 + 1 ( 1 + k )c r i=1
th
The Extended Polynomial Lens object consists of two of these faces, separated by a thickness. The total object
shape is defined by these parameters:

page 331.
maximum number of radial facets. See The use of facets on
page 331.
6: The front face radius of curvature. If this value is zero, then the curvature is assumed to be zero.
8: The front face normalization radius.
9: The front face number of extended polynomial terms.
10: The rear face radius of curvature. If this value is zero, then the curvature is assumed to be zero.
11: The rear face conic constant k.
12: The rear face normalization radius.
13: The rear face number of extended polynomial terms.
14 and on: The front and rear face extended polynomial coefficients. The rear face coefficients follow the front
face coefficients.
Extended Polynomial Surface
This object is very similar to the Extended Polynomial Lens object. Rather than being a solid, the object is a
shell surface with only 1 set of extended asphere coefficients. The object is defined by these parameters:
1: The radial height if circular. This object supports user defined
aperture shapes, see User defined apertures on page 352.
2: The radius of curvature.
page 331.
page 331.
6: The normalization radius.
7: The number of extended polynomial terms.
8 and on: The extended polynomial coefficients.
The reference coordinate is the center of the face. Coating/Scatter Groups: All faces CSG #0.
Fresnel 1
This object is a general radially symmetric or cylindrical solid Fresnel lens constructed by modeling the actual
facets and faces of the Fresnel surface. For an idealized Fresnel lens (which ignores the detailed structure of the
Fresnel surface and thus ray traces much faster) see the Fresnel 2 object.
The substrate shape is a flat disk (if radial) or a rectangle (if cylindrical). One face of the substrate consists of
radial or rectangular facets defining the profile of the Fresnel that yields optical power. The profile is constructed

of radially flat facets (or a series of flat faces if subsegments are used) whose endpoints are defined by a sag
expression identical to the Even Asphere surface:
2
cr 2 4 6 8 10 12 14 16
z = -------------------------------------------------- + 1 r + 2 r + 3 r + 4 r + 5 r + 6 r + 7 r + 8 r .
2 2
1 + 1 ( 1 + k )c r
However, to make a Fresnel lens, each facet is offset just

enough along the z axis so all facets start at the same z coordinate
as the center vertex point. This yields a lens that is "collapsed" into
a relatively small volume. The Fresnel facets are automatically
generated by ZEMAX using these 16 parameters:
1: Radial Height: This is the maximum radial aperture of the lens
if radially symmetric, or the y half height if cylindrically symmetric.
2: X Half Width: This is the half width of the lens if cylindrically
symmetric. If this parameter is zero, then a rotationally symmetric
lens is generated.
3: +Depth/-Frequency: If this parameter is positive, then it
corresponds to the depth of each groove in lens units. If negative,
then it corresponds to the frequency of the grooves. For example,
a value of -2.0 will yield 2 grooves per radial lens unit. If the groove depth is defined, the radial positions of the
grooves will generally vary; if the groove frequency is defined; the groove depth will vary. For the case of a defined
groove depth ZEMAX automatically computes the exact radial coordinate at which the sag has changed by the
specified depth. This is done using an iterative search.
4: Pitch (degrees): The pitch is the angle the "inactive" faces
(those faces nominal parallel to the local z axis) make with respect
to the z axis. The pitch may be positive or negative. A pitch angle
of a few degrees is typically added to Fresnel molds to make
extraction of the molded part easier.
5: Thick: The thickness of the Fresnel in lens units. This value
may be positive or negative; but should be chosen such that the
absolute value of the thickness exceeds the deepest groove depth;
or a non-physical Fresnel may be generated without warning or
error message.
6: Radius: The base radius of curvature. This is one over the
value "c" in the sag expression above.
7: Conic: The conic constant "k" in the sag expression above.
8-15: The coefficients on the even radial powers of r. Note these coefficients have units as r is not normalized.
16: The number of sub segments. The greater the number of sub segments, the smoother the approximation
to a curved surface between the grooves. A value of 1 yields flat grooves, larger values yield progressively more
smooth curved faces at the expense of ray tracing speed.
Because the object is a closed volume, it may be reflective, refractive, or absorbing.
The reference coordinate is the center vertex of the side of the lens with the Fresnel facets. If the radius or
aspheric terms yield negative sag values, then it is important to offset the position of the Fresnel such that the
entire solid resides inside the non-sequential group. If the entry port is placed inside the lens; incorrect ray tracing
will result. Coating/Scatter Groups: All faces CSG #0.
Fresnel 2
This object is an idealized Fresnel lens. Unlike the Fresnel 1, this Fresnel lens uses the approximation that
the Fresnel facets are infinitesimally small, and may be ignored for purposes of computing the ray-object
intersection point.

The substrate shape is either a flat disk or a rectangle. The front face of the substrate consists of a radial or
cylindrical Fresnel lens. The radial profile is defined by a sag expression identical to the Even Asphere surface:
2
cr 2 4 6 8 10 12 14 16
z = -------------------------------------------------- + 1 r + 2 r + 3 r + 4 r + 5 r + 6 r + 7 r + 8 r .
2 2
1 + 1 ( 1 + k )c r
If the Fresnel surface is cylindrical, then the profile is described by an identical expression in the y coordinate
alone:
2
cy 2 4 6 8 10 12 14 16
z = -------------------------------------------------- + 1 y + 2 y + 3 y + 4 y + 5 y + 6 y + 7 y + 8 y .
2 2
1 + 1 ( 1 + k )c y
Note that the Fresnel lens may be either radial or cylindrical

independent of the substrate shape. It is possible to define a radial
Fresnel on a rectangular substrate or a cylindrical Fresnel on a
radial substrate if desired.
The Fresnel 2 object is defined using these 14 parameters:
1: Radial Height: This is the maximum radial aperture of the lens
if radially symmetric, or the y half height if cylindrically symmetric.
2: X Half Width: This is the half width of the lens if cylindrically
symmetric. If this parameter is zero, then a rotationally symmetric
lens is generated.
3: Thick: The thickness of the Fresnel in lens units. This value
must be positive.
4: Is Cylinder: If zero, the Fresnel surface will be radial, otherwise, the surface will be cylindrical.
5: Radius: The base radius of curvature. This is one over the
value "c" in the sag expressions above.
6: Conic: The conic constant "k" in the sag expressions above.
7-14: The coefficients on the even radial powers of r if "Is
Cylinder" is zero, otherwise these are the coefficients on the even
radial powers of y. Note these coefficients have units as y and r are
not normalized.
The reference coordinate is the center vertex of the side of the
lens with the Fresnel facets. Coating/Scatter Groups: All faces
CSG #0.

Hexagonal Lenslet
The hexagonal lenslet object models a rectangular volume with
a front face consisting of a hexagonal array of even aspheric
lenslets. To make the part rectangular, partial hexagonal lenses are
added along the sides as required. The even asphere surface sag
expression is identical to that of the Even Asphere Lens object (see
Even Asphere Lens on page 308). The back face is plano. The
object uses the following parameters:
1: The number of columns (must be an odd positive integer).
2: The number of rows (must be an odd positive integer).
3: # of facets. If this value is set to 0, 1, 2, 3, or 4, then 6, 24, 96,
384, or 1536 triangles respectively will be used to render each
lenslet. Higher or lower values are internally set to either 4 or 0.
Higher faceting yields a better looking picture, but has no affect on
ray trace accuracy and may consume large amounts of memory and slow down both ray tracing and rendering.
See The use of facets on page 331.
4:Width: This is the full width of a single hexagon lenslet in lens units.
5: Thickness: The thickness of the lenslet, measured along the +z axis from the center lenslet vertex to the
back plano face.
6: Radius: The radius of curvature of each lenslet in lens units.
7: Conic: The conic constant of each lenslet.
8-15: The aspheric coefficients on the powers of r.
The reference coordinate is the center of the front face. Coating/Scatter Groups: Side faces CSG #0, front face
CSG #1, back face CSG #2.
Hologram Lens
This object is an ideal optically fabricated hologram similar to the Hologram 1 and Hologram 2 sequential
surface models (see Hologram 1 on page 261 for a description of these surfaces). The hologram is a solid, and
may be circular or rectangular in shape. The front and back faces may be plane, spheres, or conic aspheres. The
hologram surface is on the front face. See also Hologram Surface on page 313.
The Hologram object is defined using these 18 parameters:
page 331.
maximum number of radial facets. See The use of facets on page 331.
6: Radius 1: The radius of the front face.
7: Conic 1: The conic constant of the front face.
8: Radius 2: The radius of the back face.
9: Conic 2: The conic constant of the back face.

10: Holo Type: Use 1 for Hologram type 1 (both sources converging/diverging) or 2 for Hologram type 2 (one
source converging, one diverging) See Hologram 1 on page 261.
11: Order: The diffraction order to use. Multiple orders may be specified, see Diffraction tab on page 350.
12: Construction Wavelength: The wavelength in micrometers used to fabricate the hologram.
13-18: The X, Y, and Z coordinates in lens units of the construction points relative to the vertex of the front face
of the Hologram.
Because the object is a closed volume, it may be air, reflective, refractive, or absorbing. If the order is zero or
if the ray totally internally reflects at the hologram boundary, no hologram diffraction is computed.
Hologram Surface
This object is an ideal optically fabricated hologram similar to the Hologram 1 and Hologram 2 sequential
surface models (see Hologram 1 on page 261 for a description of these surfaces). The hologram is a surface,
and may be circular or user defined in shape. The surface shape may be plane, sphere, or a conic and/or
polynomial asphere. See also Hologram Lens on page 312.
The surface shape is defined by the following sag equation:
N
2
cr
z = -------------------------------------------------- + i r ,
i
2 2
1 + 1 ( 1 + k )c r i=1
where c is the curvature of the surface, k is the conic constant, r is the radial coordinate, and the terms are
aspheric coefficients. The surface supports specification of both a minimum and a maximum radial aperture; so
annular surfaces may be defined. Note that both even and odd terms are defined; up to approximately 230
coefficients may be used. This is the same shape as the Aspheric Surface object, see Aspheric Surface on
page 298.
The following parameters are used to define the hologram
surface:
3: The maximum radial aperture in lens units.
zero.
page 331.
page 331.
7: Holo Type: Use 1 for Hologram type 1 (both sources converging/diverging) or 2 for Hologram type 2 (one
source converging, one diverging) See Hologram 1 on page 261.
8: Order: The diffraction order to use. Multiple orders may be specified, see Diffraction tab on page 350.
9: Construction Wavelength: The wavelength in micrometers used to fabricate the hologram.
10-15: The X, Y, and Z coordinates in lens units of the construction points relative to the vertex of the front face
of the Hologram.

Coating/Scatter Groups: All faces CSG #0.This object supports user defined apertures, see User defined
apertures on page 352.
Imported
Objects may be imported in either IGES, SAT, or STEP format. These file formats are commonly supported by
CAD programs. Once imported, an object may have optical properties such as coatings, glass, and scattering
functions applied, and then be ray traced like any other optical component.
To import an object, set the object type to "Imported" and select the file name from the drop-down box, or place
the file name in the comment column. The file must reside in the \OBJECTS directory and end in the extension
IGS, IGES, SAT, STP, or STEP. The extension must be included with the file name if entered on the comment
column.
The following parameters are used by the Import object type.
1) Scale: This dimensionless parameter scales the entire imported solid. Upon import, ZEMAX will automati-
cally attempt to scale the dimensions of the imported solid to match the current dimensions in ZEMAX; this scale
factor is applied after that conversion.
2) Mode: The mode controls the tradeoff between set-up time and ray tracing speed. Use mode 1 for fast set
up time and slower ray tracing, mode 2 for medium set up time and medium ray tracing, and mode 3 for slow set
up time and fast ray tracing. Generally use mode 1 during set-up of the analysis, and mode 3 for analysis tracing
a large number of rays.
3, 4, 5) # X Y, Z Voxels: Voxels is a name derived from "volume pixels". A voxel is a 3D rectangle that defines
some portion of the total volume occupied by imported solids. Voxel technology allows for fast ray tracing by pre-
computing which objects, or portions of objects; lie within a given voxel. A ray entering a voxellated space may
only intersect some subset of the total number of voxels; and therefore only these voxels need to be checked for
possible ray-object intersections. The greater the number of voxels, the longer the set-up time but the faster the
ray tracing. It generally takes some experimenting to determine the optimum number of voxels. Use 10-10-10 if
no other values seem obviously superior.
6) Chord Tolerance: This setting only affects the rendering of the solid. To render the solid, ZEMAX converts
the solid to list of triangles which approximate the shape. The tolerance is the maximum allowed distance in lens
units between a single triangle and the actual surface of the solid. More triangles are added if the tolerance is set
smaller which yields more accurate rendering, at the expense of speed and a larger memory requirement. The
default value of zero will use a chord tolerance related to the size of the object sufficient to generates a coarse
approximation of the object shape that will render quickly.
Comments about imported objects
The advantage to using imported objects is that solids of any shape may be ray traced within ZEMAX. There
are no limits on the shape, complexity, or number of objects that may be imported and ray traced. Multiple objects
may be imported in a single file; however, all optical properties will be identical for all the imported objects within
a single file. To assign different optical properties (such as the material type) to each imported object, the objects
must reside in separate files. The disadvantage to using imported objects is ray tracing speed, and in some cases,
ray tracing accuracy. See the discussion below for important considerations regarding imported objects. An
alternative to imported objects exists, see User Defined Object on page 329.
Ray tracing accuracy for imported objects
Not all types of surface shapes may be ray traced with adequate accuracy using representations supported
by CAD file formats, such as IGES, SAT, and STEP. For planes, spheres, and cylinders, the CAD representation,
if done correctly, is of very high precision suitable for optical accuracy ray tracing. However, higher order shapes
do not usually have a native representation in CAD formats. For example, an aspheric surface with a polynomial
term of the form r^16 may have no equivalent representation in the chosen CAD format. A CAD program will
generally approximate this shape using a segmented spline, which is in general a piece-wise fit of the surface
using multiple lower order polynomials. Typically, multiple third or fourth order polynomials are used to approxi-
mate the surface. This is probably adequate for mechanical design, but not for optical precision ray tracing, where
surfaces must be know to tiny fractions of the wavelength of light.

This problem often arises when a high optical precision surface is modeled in ZEMAX, then exported as a CAD
file, then imported as an CAD file for subsequent ray tracing. The optical precision of the part is lost upon exporting
the native ZEMAX asphere as a CAD spline.
For non-imaging optics, the precision of the CAD representation is usually adequate, but for imaging systems,
great care must be taken to verify that the imported CAD part is a suitably accurate description of the desired
shape. Note ZEMAX uses a relative internal optical precision of about 1E-12 for ray tracing. Most CAD
representations of objects are many orders of magnitude more coarse.
Ray tracing speed for imported objects
Simple objects such as spherical lenses typically ray trace slower when imported in CAD format than the native
ZEMAX object of identical shape. Ray tracing speed for imported objects is critically dependent upon the efficient
representation of the solid shape within the imported file. The identical object may be represented in a nearly
infinite number of ways using the various solid and surface entity types supported by the CAD formats ZEMAX
can import. For example, an efficient representation of an object may use only a few spline surfaces; while an
inefficient representation of the object may use hundreds of smaller spline surfaces. Although from a mechanical
modeling perspective the two representations may both be valid and the resulting solids identical, the represen-
tation with the larger number of spline surfaces will ray trace dramatically slower. The only remedy is to return to
the source of the CAD file and see if a more efficient representation may be generated.
Limitations on imported objects
Not all possible valid CAD format files may be imported, as ZEMAX may only import solids. No lines or surfaces
are allowed. Shells must be converted to thin solids before being imported into ZEMAX. Solids must be simply
closed with a continuous exterior surface without holes or gaps. No internal surfaces or faces are allowed. Solids
composed of multiple solid volumes either in contact or overlapping cannot be ray traced. Multiple volumes not
in contact are allowed. Files containing valid solids which do not import may be sent to technical support for
review; however no guarantee is offered that ZEMAX can be made to accommodate all possible CAD format
entity types and files.
Jones Matrix
The Jones Matrix is a plane elliptical surface whose shape is
defined by 2 parameters, and polarization transmission/reflection
properties by 8 "ABCD" parameters:
3-10: Ar, Ai, Br, Bi, Cr, Ci, Dr, Di
The ellipse resides entirely within the local XY plane. This object
is a special case of the more general annulus.
The Jones Matrix may be used to model neutral density filters,
polarizers, and rotators. For a description of the Jones Matrix
parameters, see Defining polarizing components on page 511.
The reference coordinate is the center of the ellipse. Coating/
Scatter Groups: All faces CSG #0. This object supports user defined apertures, see User defined apertures on
page 352.
Lenslet 1
A lenslet 1 object consists of an array of rectangular volumes, each with a flat front face and a curved back
face. The back face may be plane, sphere, conic, or polynomial asphere; or a spherical, conic, or polynomial
aspheric toroid. The shape may be decentered with respect to the center of the lenslet. A toroidal surface is
defined by a curve in the YZ plane which is then rotated about an axis parallel to the Y axis but displaced by a
distance R; called the radius of rotation.

If the back face is rotationally symmetric, the radial profile is defined by a sag expression identical to the Even
Asphere surface:
2
cr 2 4 6 8 10 12 14 16
z = -------------------------------------------------- + 1 r + 2 r + 3 r + 4 r + 5 r + 6 r + 7 r + 8 r .
2 2
1 + 1 ( 1 + k )c r
If the back face is a toric, then the profile is described by an identical expression in the y coordinate alone:
2
cy 2 4 6 8 10 12 14 16
z = -------------------------------------------------- + 1 y + 2 y + 3 y + 4 y + 5 y + 6 y + 7 y + 8 y ,
2 2
1 + 1 ( 1 + k )c y
where c is the reciprocal of the radius of curvature in YZ plane.

The back surface may also be diffraction grating. The grating is
assumed to consist of equally spaced lines parallel to the local x
axis. The grating frequency is the lines per micrometer along the y
direction; projected down on to the surface.
The lenslet 1 is defined by these parameters:
1: The X Half-Width in lens units of each lenslet.
2: The Y Half-Width in lens units of each lenslet.
3: The thickness along the local Z axis of each lenslet.
4-5: The number of X and Y facets for each lenslet. See The
use of facets on page 331. See parameters 22 and 23 below.
6: The radius of curvature of each lenslet, use zero for a plane.
7: The conic constant of each lenslet.
8: The "Is Toric" flag; if zero the lenslet surface is rotationally symmetric, otherwise it is toroidal.
9: The radius of rotation of each lenslet if the lenslet surface is toric.
10: The grating line frequency in lines/micrometer on the back face.
11: The diffraction order. Multiple orders may be specified, see Diffraction tab on page 350.
12-19: The aspheric coefficients 1-8.
20-21: The decenter x and decenter y of the curved back face.
22-23: The number of lenslet elements in x and y. If the total number of lenslets is large, set the number of X
and Y facets to small numbers (1 or 2) to conserve memory and speed up ray tracing.
Note a cylinder lens results if the radius of rotation is set to zero. The reference coordinate is the center of the
front face. Coating/Scatter Groups: Out edge side faces CSG #0, flat front face CSG #1, curved back face CSG
#2, inside side faces CSG #3 (these are the side faces formed in the interior of an array of lenslets when
decentered lens faces meet).

Lenslet 2
A lenslet 2 object consists of an array of rectangular volumes,
each with curved front and back faces. The faces may be plane,
sphere, or conic asphere.
The lenslet 2 is defined by these parameters:
1: The X Half-Width in lens units of each lenslet.
2: The Y Half-Width in lens units of each lenslet.
3: The thickness along the local Z axis of each lenslet.
4-5: The number of X and Y facets for each lenslet. See The
use of facets on page 331. See parameters 10 and 11 below.
6: The radius of curvature of the front face of each lenslet, use
zero for a plane.
7: The conic constant of the front face of each lenslet.
8: The radius of curvature of the back face of each lenslet, use zero for a plane.
9: The conic constant of the back face of each lenslet.
10-11: The number of lenslet elements in x and y. If the total number of lenslets is large, set the number of X
and Y facets to small numbers (1 or 2) to conserve memory and speed up ray tracing.
Micro Electro Mechanical System (MEMS)
This object simulates a Micro Electro Mechanical System
(MEMS). The MEMS consists of an array of small rectangular
mirrors (commonly called pixels). The mirrors may tip at any of
three angles, each rotated about an axis to point the mirror in any
direction. The mirrors may be turned on and off by rows, by
columns, or by individual mirrors if desired to model any state the
MEMS can be in. This type of device is sometimes called a Digital
Mirror Device (DMD). The following parameters are used to define
the MEMS:
1: The number of X pixels across each row.
2: The number of Y pixels down each column.
3: The full X width of the array in lens units. The X pixel width is
this number divided by the number of X pixels.
4: The full Y width of the array in lens units. The Y pixel width is this number divided by the number of Y pixels.
5, 6, 7: The tip angle of the pixel when in state 0, 1, or 2.
8: The rotation angle around the Z axis about which the pixel will tip. This angle is measured clockwise from
the +Y axis.
9: P-Flag. If this value is 0, then the pixels are addressed by rows. If 1, then the pixels are addressed by
columns, and if 2, the pixels are addressed individually.
10 and above: Integer values which define the state of the rows, columns, or pixels, as described below.
The pixels are set to either state 0, 1, or 2 using a series of base-3 integer values. To determine the values for
any logic state of the MEMS, construct a table similar to the one below, which to save space just shows the values
for 3 rows or columns or pixels (r/c/p):
r/c/p: 3 2 1 X
0 0 0 0
0 0 1 1
0 0 2 2
0 1 0 3

0 1 1 4
0 1 2 5
0 2 0 6
etc..
where the value X is entered into parameter 10 to control the first 15 r/c/p; parameter 11 to control r/c/p 16-30,
and so on for as many r/c/p exist.
Note that X for any given state of the MEMS is given by:
X = M1*(3^0) + M2*(3^1) + M3*(3^2) + ....
where M1 is the logic state (0, 1, or 2) of the first row/column/pixel and M2 is the logic state of the second row/
column/pixel, etc. Up to 15 r/c/p values are defined by each parameter value in the editor.
Null Object
This is a non-existent object. It can be used as a place holder or a reference point for other objects.
The reference coordinate is locally (0, 0, 0).
Odd Asphere Lens
The Odd Asphere surface shape is defined by:
12
2
cr
z = -------------------------------------------------- + i r ,
i
2 2
1 + 1 ( 1 + k )c r i=1
which is very similar to the Odd Asphere surface (there are 4 additional terms). The Odd Asphere Lens object
consists of two of these faces, separated by a thickness. The total object shape is defined by 32 parameters:
page 331.
page 331.
6: The front face conic constant.
20: The back face conic constant.

Polygon Object
The polygon object is a very general user-defined object. It can
be used to define an open polygon reflector, such as a faceted
mirror; or a closed polygon with some portions reflective and others
refractive or absorptive. The Polygon Object is based on a collec-
tion of 3D triangles whose vertices are placed in an ASCII file with
the POB extension. See the Defining Polygon Objects on
page 373 for more details. Any Polygon Object may be used as a
detector as described in Objects as detectors on page 335.
There are no fixed limits to the total number of vertices or
polygons. The POB file name is referenced in the "comments"
column of the Polygon Object row, without the POB extension. For
example, if the POB file myobject.POB is placed in the \Objects
directory, then specify "myobject" in the comment column of the
Polygon Object type row in the NSC Editor. Polygon objects require two parameters:
1) A scale factor. All vertices in the POB file are multiplied by this parameter.
2) A flag to indicate if the POB file defines a volume or a surface. If the "Is Volume?" parameter is zero, then
ZEMAX assumes the POB file defines an open surface. If the "Is Volume?" parameter is any non-zero value, then
ZEMAX assumes the POB file defines a closed volume.
See Special considerations for faceted objects on page 376 for information on limitations of ray tracing
through faceted objects. A tool for creating POB files for use as detectors is described in Create Polygon Object
on page 217.
The reference coordinate is locally (0, 0, 0), and the polygons that compose the object may be placed
anywhere relative to the reference point. Coating/Scatter Groups: Each facet belongs to a CSG as defined in the
POB file, see Defining Polygon Objects on page 373 for details.
Rectangular Corner
Rectangular corners are defined by 1 parameter:
1) The full width of the squares.
Rectangular corners are composed of 3 squares which meet at
90 degrees. The squares are aligned in the positive XY, XZ, and
YZ planes. Each square is of dimension X by X.
See The use of facets on page 331 for information on limita-
tions of ray tracing through faceted objects.
The reference point is the point where all 3 squares touch.

Rectangle
A rectangle is a flat surface defined by 2 parameters:
1) The X half width.
2) The Y half width.
Rectangles are flat and reside entirely within the local XY plane
and are placed at the local Z = 0 coordinate.
The reference point is the center of the rectangle. Coating/
Rectangular Pipe
Rectangular pipes are defined by 9 parameters:
1) The X half width of the open front face.
2) The Y half width of the open front face.
3) The Z length of the pipe along the local Z axis.
4) The X half width of the open rear face.
5) The Y half width of the open rear face.
6-7) The X, Y angle tilt in degrees of the front face.
8-9) The X, Y angle tilt in degrees of the rear face.
Rectangular pipes are 4 sided boxes. The front and rear faces
are open. This object is typically used as a rectangular light pipe.
The reference point is the center of the front open face. Coating/
Rectangular Pipe Grating
This object is the same shape as the Rectangular Pipe and uses the same first 9 parameters. However, the
grating version adds a linear diffraction grating on all four sides. There are two additional parameters:
10: The grating line frequency in lines/micrometer on the side faces.
The grating consists of equally spaced lines perpendicular to the local z axis, lying on each of the four faces.
The grating frequency is the lines per micrometer along the z direction; projected down on to either the XZ or YZ
plane. Note the grating exists on all four sides of the pipe. The reference point is the center of the front open face.

Rectangular Roof
There are 3 parameters used to define a rectangular roof:
1) The X half width.
2) The Y half width.
3) The angle between the two rectangles.
A rectangular roof is composed of two rectangles which meet at
an angle.
The reference point is the midpoint of the line joining the two
rectangles. Coating/Scatter Groups: All faces CSG #0.
Rectangular Torus Surface

A rectangular torus is a surface formed by rotating a rectangle
about a displaced axis. The rotation about the displaced axis may
be over a full 360 degrees; or just some subset of that angular
range. See also the discussion of the Rectangular Torus Volume
for modeling refractive solid torus shapes.
The rectangular torus surface is defined by 6 parameters:
1: The outer radius of the torus, Rout.
2: The inner radius of the torus, Rin.
3: The start angle of the torus, 1 .
4: The stop angle of the torus, 2 .

5: The thickness of the torus, Ty.
6: The # of angular facets; these divide the range 2 1 . See The use of facets on page 331.
The rectangle lies in the YZ plane with the center at x = 0, y = 0, z = (Rout+Rin)/2. This position of the rectangle
corresponds to the rotation angle = 0 . The angles of rotation are about the Y axis and must meet this condition:
0 1 2 360 .
There is also the restriction that Rout > Rin > 0 and Ty > 0. The reference coordinate is the center of the axis
of rotation. Coating/Scatter Groups: All faces CSG #0.
Rectangular Torus Volume
This object is essentially identical to the Rectangular Torus Surface, except the ends of the torus are closed
to make a solid volume. This allows the object to made of a refractive material. See the description of the
Rectangular Torus Surface for details. Coating/Scatter Groups: Start angle end cap CSG #1, stop angle end cap

Rectangular Volume
There are 9 parameters used to define the rectangular volume:
1) The X half width of the front face.
2) The Y half width of the front face.
3) The Z length of the volume along the local Z axis.
4) The X half width of the rear face.
5) The Y half width of the rear face.
6) The front face tilt angle in degrees along X.
7) The front face tilt angle in degrees along Y.
8) The rear face tilt angle in degrees along X.
9) The rear face tilt angle in degrees along Y.
Rectangular volumes are 6 sided solids. The reference point is the center of the front face. Coating/Scatter
Groups: Front face CSG #1, back face CSG #2, all other faces CSG #0.
Rectangular Volume Grating
This object is the same shape as the Rectangular Volume and uses the same first 9 parameters. However, the
grating version adds a linear diffraction grating on four faces of the object: the top, bottom, left, and right faces.
There is no grating on the front or back faces. There are two additional parameters:
10: The grating line frequency in lines/micrometer on the side faces.
The grating consists of equally spaced lines perpendicular to the local z axis, lying on each of the four faces.
The grating frequency is the lines per micrometer along the z direction; projected down on to either the XZ or YZ
plane. Note the grating exists on all four sides of the volume. The reference point is the center of the front face.
Slide
Slides are color RGB transparencies defined by either a BMP or JPG format graphics file.
Slides use the following parameters:
1) The width in lens units of the image.
2) The pixel aspect ratio, defined as height/width. Most bitmaps use a value of 1.0 for the pixel aspect ratio.
The height of the slide will be determined by number of pixels, the slide width, and the pixel aspect ratio.
This object can be used to model radiant scenes by placing a source behind the slide. For example, to make
a Lambertian radiant scene, place a collimated rectangular source (see Source Rectangle on page 343) behind
the slide, then set the slide to have a Lambertian scattering property (see Scattering on page 352).

Sphere
Spheres are defined by two parameters:
1) The radius of the sphere.
2) A flag to indicate if the sphere is a solid volume or a hollow
shell. If the "Is Volume?" parameter is zero, then ZEMAX assumes
the sphere defines a hollow shell. Rays must either reflect or
absorb from such a surface; refraction is not allowed. If the "Is
Volume?" parameter is any non-zero value, then ZEMAX assumes
the sphere is a closed volume. The volume may be reflecting,
refracting, or absorbing.
This object can be used to model bubbles by placing a sphere
within a glass volume and setting the material type to blank (air) or
the name of a defined glass which describes a gas.
The reference point is the center of the sphere. Coating/Scatter Groups: All faces CSG #0.
Standard Lens
The standard lens is a lens composed of standard ZEMAX
surfaces. Standard surfaces may be planes, spheres, conic as-
pheres, or hyperhemispheres. The standard lens is composed of
5 separate sections:
1) A standard shape front face.
2) A standard shape rear face.
3) An annular ring between the clear aperture of the front face
and the edge of the front face.
4) An annular ring between the clear aperture of the rear face
and the edge of the rear face.
5) A possibly tapered cylindrical surface connecting the edges
of the front and rear faces of the lens.
9 parameters are used to define a standard lens:
1) The radius of curvature of the front face. Use zero for infinity (flat).
2) The conic constant of the front face.
3) The semi-diameter to the clear aperture of the front face. Use a negative value to yield the hyperhemispheric
sag point.
4) The radial aperture to the edge of the front side of the lens.
5) The center-to-center thickness of the lens.
6) The radius of curvature of the rear face. Use zero for infinity (flat).
7) The conic constant of the rear face.
8) The semi-diameter to the clear aperture of the rear face. Use a negative value to yield the hyperhemispheric
sag point.
9) The radial aperture to the edge of the rear side of the lens.
All 5 surfaces may refract, reflect, or absorb light, depending upon the material properties.
The reference point is the center of the front face of the lens. Coating/Scatter Groups: Front face CSG #1,
back face CSG #2, all other faces CSG #0.

Standard Surface
The standard surface object is very much like a standard sur-
face in sequential ZEMAX. Standard surfaces include planes,
spheres, conic aspheres, and hyperhemispheric spheres and as-
pheres.
The standard surface requires 4 parameters:
1) The radius of curvature. Use zero for infinity (flat).
2) The conic constant.
3) The semi-diameter to the maximum clear aperture. Use a
negative value to yield the hyperhemispheric sag point.
4) The semi-diameter to the minimum clear aperture. This value
must be positive. If greater than zero, this value creates a "hole" in
the surface.
The surface is rotationally symmetric about the local z axis.
If the radius is negative, then the surface is concave towards the negative z axis. If the radius is positive, the
surface is concave towards the positive z axis.
If the semi-diameter is negative, then the surface will become hyper-hemispheric, with the radial aperture of
the open end equal to the absolute value of the semi-diameter.
Surfaces may reflect or absorb rays.
The reference point is the center of the surface. Coating/Scatter Groups: All faces CSG #0. This object
supports user defined apertures, see User defined apertures on page 352. For a similar surface shape with
more complex aperture shapes supported, see Aspheric Surface 2 on page 299.
STL Object
The STL object is a very general user-defined object. It can be
used to define an open polygon reflector, such as a faceted mirror;
or a closed polygon shape such as a prism or other solid. The STL
Object format is based upon a collection of 3D triangles. This
format is widely supported by mechanical CAD programs. Both the
ASCII and the binary variations of the STL file format are support-
ed. See the section "Defining STL objects" which follows for de-
tails. See also User Defined Object on page 329, Polygon
Object on page 319, and Imported on page 314.
There are no fixed limits to the total number of vertices or
polygons.
The STL file name is referenced in the "comments" column of
the STL Object row, without the STL extension. For example, if the
STL file myobject.STL is placed in the \Objects directory, then specify "myobject" in the comment column of the
STL Object type row in the NSC Editor.
The reference coordinate is locally (0, 0, 0), and the polygons that compose the object may be placed
anywhere relative to the reference point. Some STL export implementations only allow objects to be placed such
that all vertex coordinates are positive. ZEMAX does not require this to be the case, and will import triangle
vertices anywhere in 3D space.
STL objects require two parameters:
1) A scale factor. All vertices in the STL file are multiplied by this parameter.
2) A flag to indicate if the STL file defines a volume or a surface. If the "Is Volume?" parameter is zero, then
ZEMAX assumes the STL file defines an open surface. Rays must either reflect or absorb from such a surface;
refraction is not allowed. If the "Is Volume?" parameter is any non-zero value, then ZEMAX assumes the STL file
defines a closed volume. The volume may be reflecting, refracting, or absorbing.

See The use of facets on page 331 for information on limitations of ray tracing through faceted objects.
Fast Load/Fast Trace STL objects
There are two variations of the STL object with identical parameters as described above. The Fast Load STL
Object reads the triangles defined in the STL data file, and uses the raw triangle list for ray tracing and rendering.
Although loading the STL object in this manner is fast, ray tracing is inefficient because of the very large number
of triangles typically present in STL files. The Fast Trace STL Object reads the triangles defined in the STL data
file, then sorts and pre-processes data on the triangles to make any subsequent ray tracing much faster.
The recommended usage of these two related objects is to use the Fast Load STL Object to check the
accuracy, position, and scale of an STL object as placed in the system. Once the object is appropriately defined
in the model, use the Fast Trace version to do extensive ray tracing. Note that if the number of triangles in the
STL is small, there is no reason to not use the Fast Trace version all the time.
Tabulated Faceted Radial
A tabulated object is based upon coordinates defined in a ASCII
file ending in the extension TOB. The coordinates represent the
starting and ending points of facets. A figure of revolution is
generated by replicating a specified number of facets over some
angular range. The axis of revolution is the local Z axis.
The TOB file format is two columns of data separated by either
one or more spaces or one or more tab characters. A sample TOB
file might look like this:
1.5 3.5
2.2 4.5
3.0 5.5
3.0 6.0
The first number of each pair is the local Y coordinate. This
value must be zero or positive. The second value is the local Z
coordinate. Each pair of values after the first pair represent a
"zone". If there are 6 pairs of numbers, then there are 5 zones to
the object. There is a maximum of 246 zones per object. Multiple
objects may be used if more zones are required.
ZEMAX will generate facets which approximate a smooth sur-
face for each zone. The facets can cover any fraction of a full circle,
defined by a start and stop angle. To make a full figure of revolution,
the start angle should be set to 0.0 degrees, and the stop angle to
360.0 degrees. Both angles must be zero or positive and be less
than or equal to 360.0 degrees.
The number of facets generated over this angular range can be
specified independently for each zone; so there may be 40 facets in the first zone, 80 in the second, 50 in the
third, etc.
The parameters used to define the object are:
1) A scale factor. All vertices in the TOB file are multiplied by this parameter.
2) A flag to indicate if the TOB file defines a volume or a surface. If the "Is Volume?" parameter is zero, then
ZEMAX assumes the TOB file defines an open surface. Rays must either reflect or absorb from such a surface;
refraction is not allowed. If the "Is Volume?" parameter is any non-zero value, then ZEMAX assumes the TOB file
defines a closed volume. The volume may be reflecting, refracting, or absorbing.
3) Start angle. The angle in degrees to begin the revolution of the TOB defined coordinates.
4) Stop angle. The angle in degrees to end the revolution of the TOB defined coordinates.
5) Zone 1 facets. The number of facets between the start and stop angles for the first zone.
4+n) Zone n facets. The number of facets between the start and stop angles for the nth zone.

If the "Is Volume" flag is set, the TOB file must define an object that will be a closed volume upon rotation. This
requires that the object be rotated fully around 360.0 degrees. Fully closed volumes defined by TOB files may be
used to model faceted approximations to true Fresnel lenses.
Tabulated Faceted Toroid
A tabulated object is based upon coordinates defined in a ASCII
file ending in the extension TOB. The coordinates represent the
starting and ending points of facets. A figure of rotation is generat-
ed by replicating a specified number of facets over some angular
range. The axis of rotation is an axis parallel to the local Y axis
offset by a specified radius. If the radius is set to zero, then a
surface with cylindrical symmetry is generated instead of a torus.
The TOB file format is two columns of data separated by either
one or more spaces or one or more tab characters. A sample TOB
file might look like this:
1.5 3.5
2.2 4.5
3.0 5.5
3.0 6.0
The first number of each pair is the local Y coordinate. This
value may be negative, zero, or positive. The second value is the
local Z coordinate. Each pair of values after the first pair represent
a "zone". If there are 6 pairs of numbers, then there are 5 zones to
the object. There is a maximum of 246 zones per object. Multiple
objects may be used if more zones are required.
ZEMAX will generate facets which approximate a smooth sur-
face for each zone. The facets can cover any fraction of a full circle,
defined by a start and stop angle. To make a full figure of revolution,
the start angle should be set to -180.0 degrees, and the stop angle
to 180.0 degrees. Both angles must have an absolute value less
than or equal to 180.0 degrees.
The number of facets generated over this angular range can be
specified independently for each zone; so there may be 40 facets in the first zone, 80 in the second, 50 in the
third, etc.
The parameters used to define the object are:
1) A scale factor. All vertices in the TOB file are multiplied by this parameter.
2) Radius of rotation. If positive, then the axis of rotation is in the positive local Z direction, parallel to the local
Y axis, in the YZ plane. If negative, then the axis of rotation is in the negative local Z direction, parallel to the local
Y axis, in the YZ plane. If zero, then a cylinder results. In the special case of zero radius, the number of facets
parameters are ignored (since a single facet perfectly models a flat plane) and the start and stop angle are
interpreted as start and stop x coordinate in lens units.
3) Start angle. The angle in degrees to begin the rotation of the TOB defined coordinates, unless the radius of
rotation is zero; in this case the start angle defines the starting x coordinate in lens units.
4) Stop angle. The angle in degrees to end the rotation of the TOB defined coordinates, unless the radius of
rotation is zero; in this case the stop angle defines the stopping x coordinate in lens units.
5) Zone 1 facets. The number of facets between the start and stop angles for the first zone.
4+n) Zone n facets. The number of facets between the start and stop angles for the nth zone.

Tabulated Fresnel Radial
This object is nearly identical to the tabulated faceted radial object. The key difference is that the radially
symmetric surfaces are smooth, rather than faceted.
See the description of the tabulated faceted radial object for a description of this type of object and the
tabulated object file format. Coating/Scatter Groups: All faces CSG #0.
Toroidal Lens
A toroidal lens consists of a rectangular, circular, or elliptical lens with possibly aspheric toroidal surfaces on
the front and back faces. A toroidal surface is defined by a curve in the YZ plane which is then rotated about an
axis parallel to the Y axis but displaced by a distance R; the radius of rotation. The curve in the YZ plane is defined
by:
2
cy 2 4 6 8 10 12
z = -------------------------------------------------- + 1 y + 2 y + 3 y + 4 y + 5 y + 6 y ,
2 2
1 + 1 ( 1 + k )c y

The toroidal lens is defined by 23 parameters:
1: The Radial Height of the lens in the y direction.
2: The X Half-Width. If this parameter is zero, then the outer
boundary of the lens will be a circle with a radial size equal to the
Radial Height. If this parameter is positive, the outer boundary of
the lens will be rectangular. If this parameter is negative, the outer
boundary of the lens will be elliptical.
3: The Thickness of the lens along the local Z axis.
4: The number of X facets. If the lens is rotationally symmetric,
then this parameter defines the number of angular facets. See The
use of facets on page 331.
5: The number of Y facets. If the lens is rotationally symmetric, then this parameter defines the number of radial
facets. See The use of facets on page 331.
6, 7, 8: The radius of rotation, radius of curvature, and conic for the front face.
9-14: The coefficients on the powers of y for the front face.
15, 16, 17: The radius of rotation, radius of curvature, and conic for the back face.
18-23: The coefficients on the powers of y for the back face.
To make any of the 4 radii flat; use a value of zero. Note a cylinder lens results if the radius of rotation is set
to zero. The reference coordinate is the center of the front face. Coating/Scatter Groups: Front face CSG #1, back
face CSG #2, all other faces CSG #0.
Toroidal Surface
A toroidal surface consists of a rectangular surface with a possibly aspheric toroidal shape. A toroidal surface
is defined by a curve in the YZ plane which is then rotated about an axis parallel to the Y axis but displaced by a
distance R; the radius of rotation. The curve in the YZ plane is defined by:
2
cy 2 4 6 8 10 12
z = -------------------------------------------------- + 1 y + 2 y + 3 y + 4 y + 5 y + 6 y ,
2 2
1 + 1 ( 1 + k )c y

The toroidal lens is defined by 23 parameters:
1: The X Half-Width in lens units.
2: The Y Half-Width in lens units.
3: The number of X facets. See The use of facets on page 331.
4: The number of Y facets. See The use of facets on page 331.
5, 6, 7: The radius of rotation, radius of curvature, and conic for
the surface.
8-13: The coefficients on the powers of y for the surface.
To make either the radius of rotation or radius of curvature flat;
use a value of zero. Note a cylindrical surface results if the radius
of rotation is set to zero. The reference coordinate is the center of the front face. Coating/Scatter Groups: All faces
CSG #0.
Torus Surface
A torus is a circle rotated about a displaced axis. The rotation
about the displaced axis may be over a full 360 degrees; or just
some subset of that angular range. This object can be used to
model optical fibers or curved light pipes. See also the discussion
of the Torus Volume for modeling refractive solid torus shapes.
The torus surface is defined by 6 parameters:
1: The radius of rotation about the Y axis of the circle, R.
2: The radius of the circle, r.
3: The start angle of the torus, 1 .
4: The stop angle of the torus, 2 .
5: The # of angular facets; these divide the range 2 1 . See

The use of facets on page 331.
6: The # of radial facets around the circle of radius r. See The use of facets on page 331.
The circle lies in the YZ plane with the center of the circle at x = 0, y = 0, and z = R. This position of the circle
corresponds to the rotation angle = 0 . The angles of rotation are about the Y axis and must meet this condition:
0 1 2 360 .
There is also the restriction that R > r; otherwise, a closed volume or a smooth surface will not result.
The reference coordinate is the center of the axis of rotation. Coating/Scatter Groups: All faces CSG #0.
Torus Volume
This object is essentially identical to the Torus Surface, except the ends of the torus are closed to make a solid
volume. This allows the object to made of a refractive material. See the description of the Torus Surface for details.
Coating/Scatter Groups: Start angle end cap CSG #1, stop angle end cap CSG #2, all other faces CSG #0.

Triangular Corner
Triangular corners are defined by 1 parameter:
1) The full width of the short side of the triangle.
Triangular corners are defined by 3 triangles which meet at 90
degrees. The triangles are aligned in the positive XY, XZ, and YZ
planes. Each surface is a 45-45-90 triangle, where the two short
sides are of a length specified by parameter 1.
The reference point is the point where all 3 triangles touch.
Triangle
Triangles are defined by 3 points in the XY plane, which is a total
of 6 parameters:
1) The X coordinate of vertex 1.
2) The Y coordinate of vertex 1.
The reference point is the (0, 0, 0) coordinate, which may not
be part of the triangle at all; depending upon the values give for the
vertex points. Coating/Scatter Groups: All faces CSG #0.
User Defined Object
There are several different ways to create user defined objects. For simple polygon based objects, such as
prisms and other shapes with all flat faces, see Polygon Object on page 319. Objects may also be defined in
an external CAD program and then imported into ZEMAX, see Imported on page 314 and STL Object on
page 324. There are also methods for user definitions for gradient index media (Defining GRIN media for non-
sequential ray tracing on page 359), surface scattering properties (Scattering on page 352), bulk scattering
properties (Bulk scattering on page 357), and diffraction (Defining DLLs for ray splitting at diffractive surfaces
on page 361).
The method of creating a user defined object described in this section involves the use of an external, user
supplied program called a Dynamic Link Library or DLL. The advantages to defining an object using a DLL, rather
than the other methods listed above, are:
-DLL defined objects generally ray trace much faster, and with much higher numerical precision, than
objects imported from CAD programs.
-Any number of complex curved shapes may be combined in a single object, unlike the polygon object which
only has flat faces.
-Objects may have a mixture of reflective and refractive curved faces, with user definable coating and
scatter groups.
-The DLL description is inherently parametric, which means the object is dynamically regenerated when
any defining property is modified. This allows interactive design, modification, and even optimization.
-User defined coating data, including detailed control over the complex amplitude reflection and transmis-

sion coefficients is supported. Coating data may be ray position, cosine, or object coordinate dependent.
The entire definition for the object is contained within the DLL. There are four different parts to the DLL code:
-The computation of basic data about the object, such as whether it is a solid volume or a hollow shell, and
the names of all the used parameters.
-The determination of a list of triangles which approximate the shape of the object. These triangles are used
to render the object as well as provide a "close first guess" for ray tracing purposes.
-The exact computation of the ray-surface intercept. Most of this code is actually inside ZEMAX; the DLL
only needs to find the exact ray-surface intercept point given a very close guess. ZEMAX handles all the
logic for non-sequential tracing, nesting, refraction, reflection, diffraction, optical path, scattering, etc.
-The user defined coating data used by the object, if any. The DLL or ZEMAX can compute coating data.
The DLL must include two functions:
UserObjectDefinition
UserParamNames
The UserObjectDefinition function is called to create a list of triangles which approximate the shape of the
object. ZEMAX traces rays non-sequentially to the triangles to determine close intercepts, if any. Once a potential
close intercept is found, the UserObjectDefinition function is called again to determine the exact intercept. This
is implemented using an iterative routine provided in any of the sample object DLL source files. A single object
may contain multiple complex curved faces, each with its own defining function and iteration routines.
The best way to learn the use of Object DLLs is to study an existing DLL and modify it as needed. The sample
DLLs provided with ZEMAX include extensive documentation and comments on the data formats and DLL
functions; see any of the sample object code files for examples.
!All Object DLLs must be placed in the \OBJECTS\DLL\USEROBJECTS subdirectory off the
main ZEMAX directory.
Coating/Scatter Groups: All faces have user definable CSG #s.
If you require a User Defined Object DLL, but do not wish to write the DLL yourself, contact ZEMAX Technical
Support for a quote on developing a custom DLL to meet your requirements. We have considerable experience
in developing ray tracing algorithms, and can generally write user defined object DLL code for a small fee, often
on very short notice.
Sample Object DLLs
The following sample object DLLs and source code are included with ZEMAX.
SAMPLE OBJECT DLLS

CoatingSample A rectangular surface, with user definable dimensions in x, y. This sample shows how
to implement user defined coating data. The transmission of this rectangle is 1.0 outside
of a circle of radius "a", and a user defined value inside of the circle.
EllipticalVolume An elliptical solid volume, with user definable dimensions in x, y, and z; and selectable
number of facets. This DLL demonstrates the creation of a list of triangles to
approximate the object shape, associating object faces with coating and scatter groups,
and exact ray-surface iteration to arbitrary surface description functions.
HalfCylinder A half cylinder, with user definable radius, length, and number of facets. This DLL
demonstrates the creation of a list of triangles to approximate the object shape,
associating object faces with coating and scatter groups, and exact ray-surface iteration
to arbitrary surface description functions.

Zernike Surface
A Zernike surface is defined by the following sag equation:
8 N
2
cr
z = -------------------------------------------------- + i r + A i Z i ( , ) ,
2i
2 2
1 + 1 ( 1 + k )c r i=1 i=1
where c is the curvature of the surface, k is the conic constant, r is the radial coordinate, the terms are
th
aspheric coefficients, N is the number of Zernike coefficients in the series, A i is the coefficient on the i Zernike
Standard polynomial, r is the radial ray coordinate in lens units, is the normalized radial ray coordinate, and
is the angular ray coordinate. The Zernike Standard polynomials are defined in the table given in Zernike
Standard Coefficients on page 168. The surface supports specification of both a minimum and a maximum radial
aperture; so annular surfaces may be defined.
The following parameters are used to define the Zernike sur-
face:
3: The maximum radial aperture in lens units. This aperture is
also used to normalize the radial coordinates used in the Zernike
polynomial expansion.
zero.
page 331.
6: The number of radial facets. See The use of facets on page 331.
7-14: The even radial aspheric coefficients .
15: The number of terms to use in the Zernike expansion.
16-247: The Zernike coefficients.
Coating/Scatter Groups: All faces CSG #0. This object supports user defined apertures, see User defined
apertures on page 352.
The use of facets
Some objects support a user defined specification for the number or radial, angular, or x/y facets, such as the
Biconic Lens, and many others. It is important to note that the facets are only used for two purposes:
1) To render the object
2) To provide a "first guess" as to the location of the ray-object intercept point
An arbitrary ray may intersect an object in multiple places. For accurate ray tracing, iteration is used to find
the exact intercept point. Iteration requires a good close first guess; otherwise the ray may converge to an
incorrect solution, or may not converge at all. For ray tracing, the faceted representation is used only to get the
rays close enough to the actual surface so that iteration works efficiently. The actual surface shape of the object
is modeled exactly for ray tracing purposes; the accuracy of the ray trace is not limited by the faceted
approximation.
The number of facets needs only to be high enough to render the object with acceptable quality and to preclude
erroneous solutions for the iterative ray intercept computation. Using more facets will cause the ray tracing to be
slower and not increase the accuracy of the ray intercept positions or normals. If rays seem to be missing an
object, try increasing the number of facets in the representation.

Detectors
There are four types of detectors in ZEMAX summarized in the table below.
SUMMARY OF NSC SOURCES AND DETECTORS

Detector Rectangle A flat rectangular detector with an arbitrary number of pixels. This 332
detector can record and display incoherent, coherent, point
spread function, polarization, and other data. This is the most
powerful detector in terms of the analysis it provides, but it is
limited to a flat rectangular shape.
Detector Surf A circular or annular detector with an arbitrary number of pixels in 334
the radial and angular directions. The surface may follow a plane,
sphere, conic asphere, or aspherical shape. The surface detector
can only record incoherent irradiance data.
Detector Volume A rectangular volume with an arbitrary number of voxels in the 335
local x, y, and z directions. The detector volume may be nested
within or straddle any other object. Multiple detector volumes may
also be superimposed and all will be illuminated by rays passing
through the individual voxels.
Faceted objects as detectors Most faceted objects of arbitrary shape may be used as a detector 335
that records incoherent irradiance data.
Detailed descriptions of these detector types follow.

Detector Rectangle object
The detector rectangle object stores energy data from NSC source rays that strike it. The resulting data
distributions may be viewed for incoherent light in spatial or angular domains, or spatially coherent irradiance or
phase, or coherent irradiance as a point spread function. Detector rectangles may be reflective, transparent, or
absorbing if the material is set to "MIRROR", blank, or "ABSORB", respectively.
1) X Half Width: The x width in lens units.
2) Y Half Width: The y width in lens units.
3) # X Pixels: The number of pixels along the x direction.
4) # Y Pixels: The number of pixels along the y direction.
Parameters 5-9 are only used to define how the detector is displayed on shaded model plots:
5) Data type: Select 0 for incoherent irradiance, 1 for coherent irradiance, 2 for coherent phase, 3 for radiant
intensity, 4 for radiance in position space, and 5 for radiance in angle space. If photometric units are selected,
the options are the equivalent photometric quantities. For more information on selecting units, see Analysis
Units on page 87.
6) Color: Select for 0 grey scale, 1 for inverse grey scale, 2 for false color, and 3 for inverse false color.
7) Smoothing: The amount of smoothing. The algorithm is described in The Detector Viewer on page 369.
8) Scale: Choose 0 for linear, 1 for log -5, 2 for log -10, and 3 for log -15.
9) Plot Scale: choose a maximum value to normalize the color display to; this is useful for setting a common
scale across multiple detectors.
10) Front Only. If 0, then rays may strike the detector on the front or the back side. If this flag is 1, then rays
coming from the back will be ignored, and will miss the detector. The "back" side is the side facing towards the
positive local z axis.
11) PSF Mode Wavenumber. If 0, then the coherent irradiance display is the coherent data detected for each
pixel individually. This is the default and most appropriate choice if the detector is large and the intent is to look

at the beam far from focus, such as fringes created by two overlapping beams from an interferometer. If the
wavenumber is equal to a wavelength number, then the coherent data for each pixel is the Huygens integral
of all rays incident upon the detector, resulting in an approximation to the diffraction Huygens Point Spread
Function. When using PSF Mode, the width of the detector should be a small number of wavelengths, to allow
adequate pixel resolution to see the diffraction structure, and to avoid aliasing artifacts inherent to the Huygens
model when applied to large areas.
12-15) The minimum and maximum x and y direction angles in degrees. These settings are used to control
the sensitivity of the detector to rays incident upon the detector at an angle when displaying radiant intensity
or radiance in angle space. The default setting is -90.0 degrees to +90.0 degrees in both x and y directions,
which allows display of angle data for all rays striking the detector. If the angular range is set to cover a subset
of the possible incident angles, rays that strike the detector at angles outside the defined range are ignored
ONLY for radiant intensity and radiance in angle space displays.
16) Polarization flag. If 0, the detector ignores polarization of the incident rays and considers all incident rays.
If the flag is 1, 2, or 3, the detector will only consider the component of the rays polarized along the local X, Y,
and Z directions, respectively. If the flag is 4, 5, or 6, the detector will only consider the component of the rays
polarized along X and Y, X and Z, and Y and Z, respectively. Setting the polarization flag to 1, 2, or 3 will also
consider the phase of the electric field as well as the phase of the ray (due to optical path, index, and phase
objects) when computing the coherent amplitude and phase displayed by the detector.
The information stored on any detector may be viewed by selecting Detectors, Detector Viewer on the NSC
Editor, or when the program mode is Non-sequential, on the Analysis menu. The Detector Viewer also has an
option for displaying data from any previously saved ray data base. If "none" is selected for the ray database,
then the data currently stored in the detector will be displayed, otherwise, the data stored in the selected ray
database will be used to regenerate the displayed image. For more information see The Detector Viewer on
page 369.
Five data items are stored for every ray that strikes the detector:
The incoherent intensity of the ray. This energy is stored by incrementing the counter corresponding to the
pixel the ray struck. The sum of the ray intensity divided by the pixel area yields the irradiance in flux per area.
The incoherent intensity of the ray in angular space. This energy is stored by incrementing the counter
corresponding to the pixel the ray struck in angle space. The sum of the ray intensity divided by the solid angle
the pixel represents yields the intensity in flux per solid angle.
The coherent real part, imaginary part, and amplitude of the ray. The amplitude is the square root of the ray
intensity, while the phase is determined by the total optical path length from the source referenced to the center
of the spatially distributed pixel. By separating the real and imaginary parts of the ray, interference between
many rays may be simulated. Phase due to polarization effects is also considered if the polarization flag is 1,
2, or 3; see the discussion for parameter 16 above.
For more information on the definitions and units for the terms irradiance, intensity, and flux, see Analysis
Units on page 87.
Comments on coherent data computations
The propagation and interference of light generally has properties of both particles and waves. Rays can be
thought of as the particle representation, and diffraction interference (such as for a diffraction PSF) can be thought
of as a wave representation.
For NSC analysis, ZEMAX uses ray tracing to determine optical paths and energy distributions. ZEMAX
accounts for the phase along the ray, and this allows for computation of some interference and diffraction effects.
However, it is important that the user understand what assumptions the model makes and how these assumptions
affect the accuracy of the results.
When a ray strikes a detector, ZEMAX computes the real and imaginary parts of the electric field by using the
intensity and phase of the ray referenced to the center of the pixel struck. The real and imaginary parts may then
be summed for many rays that strike the same pixel. ZEMAX also sums both the amplitude and intensity
(amplitude squared) for each pixel.
Because the phase is accounted for, some rays will constructively interfere with other rays while other rays
will destructively interfere. This allows ZEMAX to simulate effects such as fringes in interferometers (shearing or
otherwise) or interference from various orders of a diffraction grating. However, computing the coherent irradiance

involves some assumptions. Physically, destructive interference means the energy would have propagated
somewhere other than where the ray went. ZEMAX cannot determine where the energy went, and therefore
cannot account for conservation of energy in coherent irradiance calculations.
To compute the coherent irradiance from the ray data, ZEMAX first computes a coherence factor. The
coherence factor is the sum of the real part of the amplitude of all rays squared, plus the sum of the imaginary
part of the amplitude of all rays squared, divided by the sum of the amplitude for all rays squared. Defined this
way, the coherence factor is 1.0 if all rays constructively interfere, and is zero if all rays destructively interfere.
Finally, this coherence factor is multiplied by the sum of the incoherent intensity to yield the coherent intensity.
It is also important to understand that ZEMAX considers ALL sources to be coherent with respect to one
another, and for the phase of the ray to be zero at the starting coordinate of the ray, wherever it may be. This
generally limits the usefulness of the interference analysis to single point or planar collimated sources. The initial
phase of the ray and the coherence length of the source may be defined, see Coherence length modeling on
page 359 and Sources tab on page 350.
Detector Surface object
The detector surface object stores energy data from NSC source rays that strike it. The resulting data
distributions may be viewed as irradiance for incoherent light. Detector surfaces may be reflective, transparent,
or absorbing if the material is set to "MIRROR", blank, or "ABSORB", respectively.
The surface shape of a detector surface object is identical to the Aspheric Surface object. See Aspheric
Surface on page 298 for information on the defining formulas for the surface shape.
1) Radius: The radius of curvature in lens units. If zero, a plane base surface will result.
2) Conic: The conic constant. See Aspheric Surface on page 298 for the surface shape formula.
3) Max Aperture: The maximum radial size in lens units.
4) Min Aperture: The minimum radial size in lens units. If this value is greater than zero, an annular detector
will be created.
5) # A Zones: The number of angular zones, Na. See "Comments about detector triangles" below.
6) # R Zones: The number of radial zones, Nr. See "Comments about detector triangles" below.
Parameters 7-11 are only used to define the default settings for the Detector Viewer:
7) Data type: Select 0 for incoherent irradiance. This is currently the only valid option.
8) Color: Select for 0 grey scale, 1 for inverse grey scale, 2 for false color, and 3 for inverse false color.
9) Smoothing: The amount of smoothing. This value is not supported for drawing on shaded model plots, but
is used to set the default on the Detector Viewer display. The algorithm is described in The Detector Viewer
on page 369.
10) Scale: Choose 0 for linear, 1 for log -5, 2 for log -10, and 3 for log -15.
11) Plot Scale: choose a maximum value to normalize the color display to; this is useful for setting a common
scale across multiple detectors.
12) Front Only. If 0, then rays may strike the detector on the front or the back side. If this flag is 1, then rays
coming from the back will be ignored, and will miss the detector. The "back" side is the side facing towards the
positive local z axis.
14-250: The coefficients on the polynomial expansion. See Aspheric Surface on page 298 for the surface
shape formula.
Editor, or when the program mode is Non-sequential, on the Analysis menu. The Detector Viewer also has an
option for displaying data from any previously saved ray data base. If "none" is selected for the ray database,
then the data currently stored in the detector will be displayed, otherwise, the data stored in the selected ray

database will be used to regenerate the displayed image. For more information see The Detector Viewer on
page 369.
One data item is stored for every ray that strikes the detector: The incoherent intensity of the ray. This energy
is stored by incrementing the counter corresponding to the pixel the ray struck. The sum of the ray intensity divided
by the pixel area yields the irradiance in flux per area.
Units on page 87.
Comments about detector triangles
ZEMAX uses a collection of triangles to conform to the shape of the curved surface. There are Nr radial zones,
and Na angular zones. Together these two parameters define Nr * Na polygonal regions. Each region is generally
divided into two triangles, so the total number of triangular pixels is 2 * Nr * Na. However, if the minimum aperture
(parameter 4 above) is zero, then the innermost radial zone has only Na triangles rather than 2 * Na. In this case
the total number of triangles is 2 * Nr * Na - Na. The algorithm that selects the triangle coordinates attempts to
make the area of each triangle approximately the same, however the area from triangle to triangle can not be
held exactly constant. The corners of each triangle will lie on the aspheric surface defined by the radius, conic,
and aspheric coefficients. The triangles are flat and do not lie exactly on the surface if the surface is not flat.
Detector Volume object
The detector volume object stores incident and absorbed energy data from NSC source rays that pass through
the individual voxels. The resulting data distributions may be viewed as incident flux, absorbed flux, or absorbed
flux per unit volume. Detector volumes may be transparent, or be made of any valid material. Detector volumes
may also be nested within or straddle other objects. Objects within detector volumes may be refractive, reflective,
absorbing, gradient index, or have surface and or bulk scatter properties defined.
The shape of a detector volume object is a rectangular solid.
1-3) The X, Y, and Z half widths of the volume. The center of the volume is at the local coordinate origin.
4-6) The number of voxels in the X, Y, and Z directions. The total number of voxels is nx * ny * nz.
Editor, or when the program mode is Non-sequential, on the Analysis menu. Data for detector volume objects is
not available when using previously saved data from a ray data base. Pressing the left and right arrow keys while
viewing detector volume data on the NSC detector viewer will cycle through all the Z-planes. For more information
see The Detector Viewer on page 369.
Two data items are stored for every ray that passes through each voxel: The incident flux of the ray in source
units, and the absorbed flux in the voxel. The incident flux is the flux of the ray as it enters the voxel. A ray segment
that starts in a particular voxel will not increase the incident flux in that voxel, because the incident flux is only
incremented for rays entering the voxel. The absorbed flux is a function of the path length of the ray through the
voxel, as well as the coefficient of absorption of the material the ray is passing through.
Detector volumes may be nested within other NSC object types. For example, to measure the absorbed flux
in a cylindrical glass rod, place the rod within the detector volume, and set the glass type of the rod on the rod
object. If the glass type has an absorption coefficient associated with the material, the portion of the flux of rays
passing through the glass rod will be recorded by the detector volume.
Units on page 87.
Objects as detectors
Any object type that ZEMAX internally represents as a collection of triangles may act like a detector. This
includes objects that use facets to form flat faces, such as the polygon, STL, and rectangular volume objects.
Objects that approximate the shape of a curved surface or solid for rendering purposes by using facets, such as
the aspheric surface, toroidal lens, biconic lens, and other complex shapes, may also be detectors.
To make any of these objects a detector, choose "Object Is A Detector" from the Type tab of the Object
Properties dialog box. Once the object is set to be a detector, the intensity detected on each triangular facet used

to render the object will be displayed on the Shaded Model display, and on the text listing of the Detector Viewer
window. The color scheme the detector will display on the Shaded Model may also be selected on the Object
Properties dialog box. The colors chosen may represent total flux on the facet or may be normalized to be flux
per unit area the facet represents.
Objects acting as detectors can only display incoherent intensity. Each individual triangle used to draw the
faceted object becomes a single pixel. All rays which strike the pixel will be summed to yield the total intensity.
The faceted detectors are cleared whenever all detectors are cleared from the detector control dialog box.
A tool for creating POB files for use as detectors is described in Create Polygon Object on page 217.
Sources
Sources include points, ellipses, rectangles, volumes, and user defined types. Any source may be placed
inside of any object, or not in any object, but not both (a source may not straddle an object boundary).
Placing sources inside objects
By default, ZEMAX assumes a source is placed in the global medium of the non-sequential group. However,
sources may be placed entirely inside any solid object, or inside any number of nested solids. Two steps are
required to properly place a source inside of a solid object:
1) The source location and size must be correctly set such that all rays initiate from inside the object. This
means the source cannot extend outside of the object in which it is placed.
2) The number of the object in which the source is placed must be provided on the "Inside Of" data column of
the source. If this first object is in turn inside of a second object; then the first object must indicate the second
object number in the "Inside Of" data column for the first object. If there are more levels of nested objects, each
object must define the object it has been placed inside of on the "Inside Of" data column.
The source must be listed after all objects that the source is placed inside of in the NSCE.
Adding new source types
If a source is required that is not listed, please contact technical support to suggest the new type be added to
ZEMAX. Some simple sources, such as Lambertian planes, may be simulated by irradiating a Lambertian
scattering surface with one of the other source types. Many objects may also act as detectors, for example any
polygon object may be a detector. See the discussion for information on the Source DLL and Source File for user
defined sources.
SUMMARY OF NSC SOURCES

Source Diode An array of diodes with separate X/Y distributions 337
Source DLL A source defined by an external user supplied program. 338
Source Ellipse An elliptical surface that emits light from a virtual source point. 339
Source Filament A source in the shape of a helical filament. 340
Source File A user defined source whose rays are listed in a file. 340
Source Gaussian A source with a Gaussian distribution. 342
Source Point A point source that radiates into a cone. The cone may be of zero 342
width or be extended up to a full sphere if desired.
Source Radial A radial symmetric source based upon a spline fit of arbitrary 342
intensity vs. angle data.
Source Ray A point source aligned with direction cosines. 343
Source Rectangle A rectangular surface that emits light from a virtual source point. 343

Source Tube A source in the shape of a cylindrical tube. 343
Source Two Angle A rectangular or elliptical surface that emits light into a cone with 343
distinct angles in the X and Y directions.
Source Volume Cylinder A volume source in the shape of a cylinder with an elliptical cross 343
section.
Source Volume Ellipse A source in the shape of an elliptical volume. 344
Source Volume Rectangle A volume source in the shape of a rectangle. 344
The individual sources are described in the following sections.

Parameters common to all source objects
All source type objects have the same definition for parameters 1 through 5. These parameters are:
1) # Layout Rays: Defines how many random rays to launch from the source when creating layout plots.
2) # Analysis Rays: Defines how many random rays to launch from the source when performing analysis. See
Choosing the number of analysis rays on page 365.
3) Power (units): Power is the total power over the defined range of the source. The power units are specified
by the system source units. See Source Units on page 87 for details.
4) Wavenumber: The wavenumber to use when tracing random rays. Zero means polychromatic; which
chooses ray wavelengths randomly with the weighting defined on the wavelength data editor.
5) Color #: The pen color to use when drawing rays from this source. If zero, the default color will be chosen.
The other parameters have source type specific meanings as described in the follow sections.
Source Diode
The source diode model can be used to define one diode, a 1D array of diodes, or a 2D array of diodes. Each
diode has an intensity distribution given by:
2Gx y 2Gy
2 -----x + -----
x y
I ( x, y ) = I o e ,
where x is the XZ divergence angle in degrees, G x is the "supergaussian" factor for the X direction, with
similar definitions for the Y subscripted values. Note if G x is 1.0, then a typical Gaussian distribution results. If
G x is greater than 1.0, then the distribution becomes more "squared". Both G x and G y must be greater than or
equal to 1.0.
Most laser diode manufacturers specify the far field divergence angles as the full width of the distribution
between the half power points, fwhm . For a true Gaussian distribution ( G x = 1), setting the left hand side of the
1 1
equation to --- I , setting y to zero, substituting for x the value of --- , then solving for x gives
2 o 2 fwhm
fwhm
x = --------------------
- , or
2 ln ( 2 )
x = ( 0.8493218 ) fwhm .

For example, a diode with a fwhm in the x direction of 10 degrees, the value for x would be 8.493218
degrees. A similar conversion applies in the y direction.
An astigmatism term may also be defined. This value must be positive, and represents the distance along the
local -Z axis from which the XZ distribution is measured. At the local XY plane at Z = 0, the resulting ray pattern
is a line oriented along the local X axis.
If more than 1 diode is desired, the number x and number y, along with the delta x and delta y spacings may
be defined.
The parameters used to define this source are:
1-5) See Parameters common to all source objects on page 337.
6) Astigmatism: The distance to offset the XZ distribution in lens units.
7) X divergence in degrees.
8) X supergaussian factor.
9) Y divergence in degrees.
10) Y supergaussian factor.
11-12) Number of diodes in X/Y.
13-14) Spacing of diodes in X/Y in lens units.
Source DLL
Although many types of built-in sources are included with ZEMAX, there are times when the best way to model
a source is to define an algorithm to generate rays with the desired properties.
ZEMAX also supports user-defined sources as tables of rays; see the Source File discussion on page 340.
To define a source using a program, the algorithm which generates random rays must be written and compiled
into a Windows Dynamic Link Library, or DLL. Numerous DLLs are provided with ZEMAX with source code. New
DLLs may be easily created with a suitable compiler. See also Comments about DLLs on page 377.
Source DLL parameters
Each DLL may use between zero and 30 user defined data values as parameters in the computation of the
source properties. These values are defined by the DLL and are only used by the DLL.
Creating a new Source DLL
UserSourceDefinition
UserParamNames
When launching a ray from a source modeled by a DLL, ZEMAX passes to the UserSourceDefinition function
the source parameters, the wavelength, and other data. The UserSourceDefinition function then is required to
compute the following values:
x, y, z: the starting coordinates for the ray
l, m, n: the starting direction cosines for the ray
i: the initial relative intensity of the ray
These values are returned to ZEMAX and are used to initiate the ray trace. The function UserParamNames is
used to define the names of all used parameters. These names appear in the parameter columns of the source
DLL object in the NSC Editor.
The best way to learn the use of Source DLLs is to study an existing DLL and modify it as needed. The sample
DLLs provided with ZEMAX include extensive documentation and comments on the data format; see any of the
sample source code files for examples.

!All Source DLLs must be placed in the \OBJECTS\DLL\SOURCES subdirectory off the main
ZEMAX directory.
Sample Source DLLs

The following sample source DLLs and source code are included with ZEMAX.
SAMPLE SOURCE DLLS

FIBER1 A planar fiber model. The rays emanate from a disk of radius R. The intensity profile is
given by
2 4
I ( r ) = A + Br + Cr ,
where
0.0 r R .
Once the ray starting position is determined, the ray emanates from the surface into a
cone whose numerical aperture depends upon the radial coordinate:
2 4
NA ( r ) = D + Er + Fr .
Within the angular cone defined by the numerical aperture, the ray distribution is
uniform. The parameters R, A, B, C, D, E, and F are all defined on the NSC Editor.
Source Ellipse
The Source Ellipse is a flat elliptical (and optionally annular) surface which emits rays. Although the origin of
each ray launched lies on the surface of the ellipse with a uniform distribution, the distribution of the ray directions
may be any one of these:
a) All rays appear to emit from a point placed anywhere relative to the source. The location of this point is
defined in the parameter list below. When used in this mode, the source acts like an imaging point source.
b) Rays emit in a cosine distribution of the form
Cn
I ( ) I o ( cos ) ,
where the exponent Cn may be any value greater than or equal to unity, and need not be an integer. When
used in this mode, the source acts like a diffuse cosine source. The larger Cn, the narrower the distribution
becomes. Note this distribution is rotationally symmetric about the local Z axis.
c) Rays emit in a Gaussian distribution of the form
( Gx l + Gy m )
2 2
I ( l, m ) I o e ,
where l and m are the direction cosines of the ray in the X and Y axis directions and Gx and Gy are constants.
This form may be used to define a far field pattern that is different in the X or Y directions. The larger Gx and
Gy are, the narrower the distribution becomes in the respective directions.
The nature of the ray distribution is defined by the parameter values. If Cn, Gx, and Gy, are all zero, then all
rays will appear to emit from a virtual point source. If Cn is 1.0 or greater, then the cosine distribution will result
(no matter how the source distance or Gx or Gy are set). If Cn is zero, but either Gx and Gy is non-zero, then a
Gaussian distribution will result.
6) X Half Width: The x half width in lens units.
7) Y Half Width: The y half width in lens units.

8) Source Distance: The distance along the local z axis from the apparent source point to the location of the
source object. This value may be positive or negative. If zero, the rays are collimated. If positive, the apparent
source point is behind the object. Considered only if Cn, Gx, and Gy are all zero.
9) Cosine Exponent: The power on the cosine term. This is Cn in the cosine distribution expression above.
10) Gaussian Gx: The X term in the Gaussian distribution. Ignored if Cn is not zero.
11) Gaussian Gy: The Y term in the Gaussian distribution. Ignored if Cn is not zero.
12) Source X: The X coordinate of the point that emits the rays. If Source Distance is zero, then the Source X
parameter is the X direction cosine of the collimated ray bundle. Considered only if Cn, Gx, and Gy are all zero.
13) Source Y: The Y coordinate of the point that emits the rays. If Source Distance is zero, then the Source Y
parameter is the Y direction cosine of the collimated ray bundle. Considered only if Cn, Gx, and Gy are all zero.
14) Min X Half Width: The minimum x half width in lens units. Use a value greater than zero and less than the
X Half Width to define an annular region.
15) Min Y Half Width: The minimum y half width in lens units. Use a value greater than zero and less than the
Y Half Width to define an annular region.
Source Filament
The Source Filament can be thought of as a thin wire coiled in a helix shape. The wire turns "N" times along
the full Z coordinate length given by "L". The radius of the turns is defined by "R". Rays emanate from a randomly
chosen point along the helix in a random direction.
6) Length "L" in lens units.
7) Radius "R" in lens units.
8) Number of turns "N" (dimensionless). N may be fractional or even negative to reverse the rotation direction
of the helix.
The axis of the helix, and the length L, are oriented along the Z axis.
Source File
The Source File is a source whose ray coordinates, cosines, and intensity are defined in a user supplied file.
This feature allows creation of arbitrary user defined sources. The name of the file containing the ray data must
be placed in the comment column of the object. The file extension must be DAT and the file must be placed in
the \OBJECTS directory. The file format may be either ASCII or binary, both formats are described below.
The parameters are:
6) Randomize?: If set to zero, the rays will normally be traced in the order listed in the file. If non-zero, then
the ray order is randomized once, when the file is read or any parameter in the NSC Editor changes for the source
object and the source is updated. The randomize feature is only available if the total number of rays in the file is
1,000,000 or less. Files with more than 1,000,000 rays are too large to hold in memory while randomizing.
7) Total (units): This value is provided for reference only, and is the total power in source units defined by all
the rays in the file. This value is set by ZEMAX upon reading the file and should not be altered or set by the user.
The actual power of each ray is defined by parameter 3 above and the number of rays being traced.
ASCII vs. binary format files
The maximum number of rays currently allowed for ASCII files is 1,000,000. If there are more than 1,000,000
rays defined in an ASCII source file, ZEMAX will offer to convert the file to binary format automatically. The original
file will be renamed with the ".OLD" extension, and the new binary DAT file will be created using the same name
as the original file. After the conversion is complete, an option will be presented to save or delete the old ASCII
file. This conversion is done to save disk space (binary files are about 30% as large as equivalent ASCII files),
load time (binary files read about 20x faster than ASCII files), and system memory (large binary files are limited

only by available disk space and require very little RAM, even if billions of rays are being traced, because the
data is left on disk).
Memory requirements for source file objects
For both ASCII and binary files, if the total number of rays to be traced is less than 1,000,000, ZEMAX stores
the rays in system memory for maximum speed. This requires about 28 bytes of memory for each ray, or up to
28 Mb of memory for 1,000,000 rays. If the total number of rays is 1,000,000 or greater, ZEMAX leaves the ray
data on disk, and reads the file as required. This requires less memory, and allows ZEMAX to trace a huge number
of rays, limited only by the systems file storage capacity.
Restrictions on the number of rays selected
There are restrictions on the values for the number of layout and analysis rays when using the Source File:
- The number of analysis rays must be equal to or larger than the number of layout rays. The number of
analysis rays will be set to the larger of the two numbers entered in the NSC Editor.
- The number of analysis and layout rays may not exceed the number of rays defined in the file.
Binary Source File format
The Binary Source File consists of a header structure of the form:
typedef struct
{
int Identifier; // Will be set to 8675309 for quick check of proper format
int NbrRays; // The number of rays in the file
char Description[100]; // A text description of the source
float SourceFlux; // The total flux in watts of this source
float RaySetFlux; // The flux in watts represented by this Ray Set
float Wavelength; // The wavelength in micrometers, 0 if a composite
float AzimuthBeg, AzimuthEnd; // Angular range for ray set (Degrees)
float PolarBeg, PolarEnd;// Angular range for ray set (Degrees)
long DimensionUnits; // METERS=0, IN=1, CM=2, FEET=3, MM=4
float LocX, LocY,LocZ; // Coordinate Translation of the source
float RotX,RotY,RotZ; // Source rotation (Radians)
float ScaleX, ScaleY, ScaleZ; // Scale factor to expand/contract source
float unused1, unused2, unused3, unused4;
int reserved1, reserved2, reserved3, reserved4;
} NSC_RAY_DATA_HEADER;
The data types float and int are both 32 bit types.The "Identifier" integer must be the value 8675309. The other
data may or may not be included; ZEMAX only uses the NbrRays and DimensionUnits parameters. After
the header follow NbrRays ray structures. Each ray structure is of the form:
typedef struct
{
float x, y, z;
float l, m, n;
float intensity;
} NSC_RAY_DATA;
The binary format described above is used by ZEMAX when generating source files from ray trace data (see
the discussion of "Save Rays on Object n" in the section Ray Database Viewer on page 367). This format is
also supported by the ProSource program available from Radiant Imaging (www.radimg.com). To generate a
binary source data file for use in ZEMAX from the ProSource program follow these instructions:
Have a source file (*.rsm) open in ProSource. This enables the "Gen Source" button on the main toolbar.
Click on the "Gen Source" button on the main toolbar. This opens the Ray Generation dialog box.
From the "File Format" area of this dialog box, choose "ZEMAX Binary".
Click on "Generate Rays", at the bottom, to close the dialog box and create the binary file.
These instructions may not be complete or current, and the documentation for ProSource should be consulted
for more information.

ASCII Source File format
The ASCII Source File consists of a single line of header data with just two integer numbers of the form:
number_of_rays dimension_flag
The number_of_rays indicates the total number of rays in the file. The dimension_flag is 0 for meters, 1 for
inches, 2 for centimeters, 3 for feet, and 4 for millimeters.
The remaining lines in the file are of the format:
x y z l m n intensity
Any line starting with the "!" symbol is assumed to be a comment line and is ignored. Any number of rays may
be defined in a single file using this ASCII format. However, if the number of rays exceeds 1,000,000, the file will
automatically be converted to binary format by ZEMAX the first time the file is opened for a source file object. For
a discussion of this conversion see ASCII vs. binary format files on page 340.
Intensity normalization in source files
Each ray may have a different relative intensity. If the relative intensity value is not 1.0 for each ray, then
normalization occurs as described below.
When the source file is first loaded into memory, the intensity of each ray is summed and then normalized to
the average intensity. If the total flux of the source is then defined to be some number of watts, a subset of the
rays can be traced and their intensities will yield approximately, but not exactly, the total flux. Normalization is
required for an arbitrary subset of the rays to yield approximately the total desired power.
Source Gaussian
The Source Gaussian has a Gaussian distribution of rays which appear to emanate from a point source. The
parameters are:
6) Beam Size: The beam radius at the 1 over e^2 point in intensity in lens units.
7) Position: The distance from the apparent point of divergence of the rays to the source plane location. If zero,
the rays are collimated.
Source Point
The Source Point is a point which emits rays into a cone. The cone angle may be any value between 0 and
180 degrees (which would radiate into a full sphere). The parameters are:
6) Cone angle: The semi-cone angle in degrees.
Source Radial
The Source Radial is a flat, rectangular or elliptical shape that emits rays into a hemisphere. The distribution
of the rays in this hemisphere is symmetric about the local z axis and is defined by a cubic spline fit to arbitrary
intensity vs. angle data. The number of points can be any integer between 5 and 180, inclusive. The angular range
from 0.0 to 90.0 degrees is evenly divided among the specified number of points. The data for each point is the
relative intensity measured in the far field of the source at the angle corresponding to that point. The provided
data is fit to a cubic spline with a maximum possible resolution of about 0.5 degrees. Rays generated from the
spline fit will follow the correct distribution in the far field of the source.
The parameters are:
6) X Half Width: The x half width in lens units. If less than zero, the emitting region will be elliptical.
7) Y Half Width: The y half width in lens units. If less than zero, the emitting region will be elliptical.
8-9) Unused, reserved for future expansion of this feature.
10) The number of points, must be between 5 and 180, inclusive.
11+) The relative intensity data at each angle from the normal.

Source Ray
The Source Ray is a point which emits rays along specified direction cosines. This is useful for debugging, for
example, the direction cosines may be used to recreate a ray that causes geometry errors. The parameters are:
6-8) Cosines: The local X, Y, and X direction cosines of the ray. These are automatically normalized.
Source Rectangle
The Source Rectangle is a flat rectangular surface which emits rays from a virtual source point. The parameters
are identical to the Source Ellipse, but the shape of the source is a rectangle rather than an ellipse.
Source Tube
The Source Tube is similar to the Source Volume Cylinder, except the rays only emanate from the surface of
the tube rather than the full volume.
6) Length "L" in lens units.
7) Radius "R" in lens units.
The axis of the tube, and the length L, are oriented along the Z axis.
Source Two Angle
The Source Two Angle is a flat surface, with either an elliptical or a rectangular shape, which emits rays. The
origin of each ray launched lies on the surface of the source with a uniform distribution. The angular distribution
is uniform in X and Y, with distinct maximum half angles for each direction. The angular distribution may also be
either rectangular or elliptical.
8) X Half Angle in degrees: The half angle of the cone of rays in the XZ plane.
9) Y Half Angle in degrees: The half angle of the cone of rays in the YZ plane.
10) Spatial Shape: Use 0 for rectangular, 1 for elliptical.
11) Angular Shape: Use 0 for rectangular, 1 for elliptical.
Source Volume Cylinder
The Source Volume Cylinder is a 3D volume formed by an elliptical shape in the XY plane that is symmetrically
extruded along the Z axis. The center of the source volume is at the local origin of the object. Rays are emitted
in random directions from anywhere inside the volume, with uniform probability in both position and ray direction.
Points within the cylinder volume satisfy these relations:
------
X- 2 ------
Y- 2 ------
Z 2
W y 1 W z 1
+ , and ,
W x
where W refers to the half widths in X, Y, and Z. The parameters are:

8) Z Half Width The z half width in lens units.

Source Volume Ellipse
The Source Volume Ellipse is a 3D volume formed by an elliptical shape in the XY, XZ, and YZ planes. The
center of the source volume is at the local origin of the object. Rays are emitted in random directions from
anywhere inside the volume, with uniform probability in both position and ray direction. Points within the ellipse
satisfy this relation:
------
X- 2 ------
Y- 2 ------
Z 2
W z 1
+ + ,
W x W y
where W refers to the half widths in X, Y, and Z. The parameters are otherwise identical to that for the Source
Volume Cylinder.
Source Volume Rectangle
The Source Volume Rectangle is a 3D volume formed by a rectangular shape in the XY, XZ, and YZ planes.
The center of the source volume is at the local origin of the object. Rays are emitted in random directions from
anywhere inside the volume, with uniform probability in both position and ray direction. Points within the rectangle
satisfy these relations:
X W x , Y W y , and Z W z ,
where W refers to the half widths in X, Y, and Z. The parameters are otherwise identical to that for the Source
Volume Cylinder.
Object Placement
The conventions and restrictions on placing objects in the NSC group is critically important. Objects may be
placed anywhere in 3D space; and objects may be placed with respect to any other object. Objects may also be
placed entirely inside of other objects, or may be placed adjacent to other objects.
The object coordinate system
Each objects position is defined by 6 parameters: the X, Y, and Z coordinates, and the rotation about the global
Z, then global Y, then global X axis. Note the coordinate translation is done first, then the tilts, in the order Z, Y,
and X. This is the same convention as for the coordinate break surface when using order flag = 0 (see page 239).
The conversion from object local coordinates to global coordinates can be written in equation form as
xg xo R 11 R 12 R 13 x l
y g = y o + R 21 R 22 R 23 y l ,
zg zo R 31 R 32 R 33 z l
where the g subscript indicates the global coordinate, o is the offset, and l is the local object coordinate. The
matrix R is the rotation matrix, which relates the orientation of local and global coordinates.These equations can
be written more compactly as
G = O + RL ,
Where G is the global coordinate vector, O is the offset vector, R is the rotation matrix, and L is the local
coordinate vector. For more information on properties of the rotation matrix, see Global Coordinate Reference
Surface on page 93.

Reference objects
It is often useful to reference an objects position and rotation relative to another object. This is useful especially
when placing related objects in a group, and then decentering or tilting the entire group.
The object to which the coordinates are relative to is the "reference object". The default reference object is
object 0, which is the vertex of the Non-Sequential Component surface. If a positive number greater than zero is
specified, then the coordinates of the object are referenced to the location and rotation of the specified object.
This is an "absolute" reference object. If the reference object number is negative, then the reference object is
determined by adding the current object number to the negative reference object number. This is a "relative"
reference object. For example, if the reference object is -3 on object 8, the reference object will be 5 because 8
- 3 = 5. Relative reference objects are particularly useful when copying and pasting groups of objects; this is
easiest if all the objects in the group use relative references to the first object in the group.
When a reference object is used, the rotation and offset matrixes then become:
G' = O' + R'G
' = O' + R' [ O + RL ]
' = [ O' + R'O ] + [ R'R ]L
Any number of coordinate reference nesting is supported; so that object 9 could be placed in the coordinate
frame of object 5 which in turn is placed in the frame of object 3. The only restriction is that the reference object
must precede in the object list the object whose coordinates are being referenced.
Placing objects inside of, adjacent to, or overlapping other objects
Placing sources within other objects is described in Placing sources inside objects on page 336.
Objects can be combined to make more complex objects, by placing one object inside of, adjacent to, or
overlapping another object. What determines the ray tracing properties of such a compound object depends upon
the position and type of the various objects and whether or not they touch or overlap. Here the word touch means
that one or more points on the boundary surface of one object is in the same location in 3D space as a point on
a boundary surface of another object. Mirror objects may be placed anywhere, even in contact with or partially or
fully inside any other object without restriction. Rays will always reflect from mirror surfaces back into the medium
the rays had been traveling through.
Glue distance
When two NSC objects are placed in contact, such as a lens touching one face of a prism, numerical round
off will cause the ray tracing algorithm to sometimes detect a very tiny distance between the two objects. This can
also occur when objects are rotated in 3D space and placed close, but not exactly, next to one another because
of the finite number of digits entered in the spreadsheet editor and the finite precision of the computer. Objects
placed within the glue distance of one another are automatically "cemented" together for ray tracing purposes.
The glue distance is the distance below which the objects are considered in contact. This value should only
need adjustment from the default value of 1E-06 lens units (or 0.001 micrometers if the lens units are millimeters)
in very rare cases.
The glue distance also determines the minimum propagation length for ray tracing. If a ray-object intersection
is less than the glue distance away from the previous intercept, the intercept is ignored. This can affect ray tracing
results when tracing rays from faceted reflectors, and a ray strikes within the glue distance of the edge between
two angled facets, see page 376 for a discussion. In most cases, no adjustment should be made to the glue
distance. However, the value may be edited on the Non-Sequential tab of the System, General dialog box.
Nesting object limits
There is a user defined limit on the maximum number of nested objects. This defines an upper limit on how
many objects can be placed inside each other. For example, if the maximum number of nested objects is 3, then

object 3 may be placed inside of 2 which is placed inside of object 1. There may be any number of groups of
objects each nested 3 deep in this case. The limit applies to the total nesting in any one collection of objects,
however, there may be any number of such collections within the NSC group. The maximum number of nested
objects is set on the Non-Sequential tab of the General dialog box. Setting the nesting limit no higher than required
for the system being modeled conserves memory usage, although the minimum setting is 4.
Nesting volumes
Volumes of refracting material are more complicated, because ZEMAX must keep track of the index of
refraction through which the ray is propagating. The rule to remember is: if a ray strikes more than one object at
the exact same point in space; the last object listed in the NSC Editor determines the properties of the surface or
volume at that point.
!If a ray strikes more than one object at the exact same point in space; the last object listed in
the NSC Editor determines the properties of the surface or volume at that point.
For example, if a diffraction grating lens is object #1, and a non-

grating lens of the same thickness and radii made of air or glass is
placed inside the first lens is object #2; then rays that strike in the
zone including both objects #1 and #2 will act as though they just
hit object #2.
This allows defining objects with "holes" and other compound
objects. Objects may be touching at one or more faces, nested or
not, or may partially overlap, to create a wide range of compound
solid shapes.
Nesting surfaces
The nesting rules defined above apply to solid volumes of refractive or reflective material. Some special
surfaces may also be "nested" in the sense that more than one surface may exist at the same point in space. The
rules are similar to those that apply to volumes, but surfaces cannot be refracting.
The same rule applies to surfaces as does for solids: if a ray strikes more than one object at the exact same
point in space; the last object listed in the NSC Editor determines the properties of the surface at that point.
!If a ray strikes more than one object at the exact same point in space; the last object listed in
the NSC Editor determines the properties of the surface at that point.
For surfaces, there are a few rules to consider when more than one surface exists at the ray intercept point:
1) The last surface listed will determine the properties of the surface.
2) If the last surface listed is a mirror, the ray will reflect.
3) If the last surface listed is an absorber, the ray will be absorbed.
4) If the last surface listed is neither a mirror nor an absorber, the ray will ignore the surface.
5) Surface objects may not share boundaries with volume objects unless the surface object is reflective or
absorbing, or unless the volume object is listed after the surface object; in which case the volume defines the
properties of the common boundary.
6) Only standard surface objects may share boundaries, support for nesting of other surface object types may
be added in future versions of the software.
Refraction and reflection from NSC objects
All NSC objects allow specification of a material on the NSC Editor (gradient index media are defined
separately, see Defining GRIN media for non-sequential ray tracing on page 359). The reflective, refractive, and

absorptive properties of an object depend upon the object type, whether or not the object describes a surface or
a volume, and the name of the material specified. The material name may be blank, "MIRROR", "ABSORB", or
a homogeneous glass name as defined in one of the currently selected glass catalogs. Blank materials are
assumed to indicate an index of unity, which may or may not be the background index of the non-sequential space
(which is set by the index of the NSC surface on the Lens Data Editor). The properties of objects are described
by the following table.
OBJECT PROPERTIES BY MATERIAL
Object Type Material is blank or glass Material is "MIRROR" Material is "ABSORB"

Volume All faces are refractive. All faces are reflective. All faces are absorbing.
Surface Object is ignored.
Polygon Object (volume) If a face in the POB file is If a face in the POB file is If a face in the POB file is
marked as reflective or marked as absorbing, marked as reflective,
absorbing, then the face then the face remains then the face remains
remains reflective or absorbing. All other faces reflective. All other faces
absorbing. Refractive become reflective. become absorbing.
faces remain refractive.
Polygon Object (surface) Object is ignored. If a face in the POB file is If a face in the POB file is
marked as absorbing, marked as reflective,
then the face remains then the face remains
absorbing. All other faces reflective. All other faces
become reflective. become absorbing.
Reflective objects are usually surfaces rather than volumes. Examples include flat, curved, and faceted
mirrors. Hollow light pipes can also be modeled as a group of flat reflectors. Some reflectors do enclose a volume;
the interior of which no ray will ever see. Examples include rectangular and spherical volumes with all surfaces
set to be reflective. Any volume with a material name of "MIRROR" will be all reflective.
Refractors must always enclose a volume. Volume refractors include lenses, prisms, solid light pipes, and
closed faceted volumes. Absorbers may be surfaces or volumes with the material name set to "ABSORB".
Polygon Objects may be composed of facets, and the individual facets may be refractive, reflective, or absorptive.
Note that STL Objects act just like other volumes, and not like Polygon Objects, because the STL format does
not support individual facets having refractive, reflective, or absorptive properties.
Diffraction from NSC objects
Some NSC objects have one or more diffractive faces; such as the Diffraction Grating, Binary 1, and Binary 2
objects. These objects refract or reflect rays as well as diffract them, according to the grating period or phase and
the diffraction order and wavelength. For any ray, if the diffraction order being traced does not satisfy the grating
equation, then the energy of that ray will refract or reflect along the zero order path.
Ray splitting is supported on diffractive surfaces, but splitting is only allowed by order and not by reflected and
transmitted Fresnel coefficients. For this reason, no "ghost" rays are generated from diffractive surfaces.
All diffractive NSC objects support a parameter to defined the "order" for diffraction from the object. This order
is called the "primary" diffraction order. Sequential rays which enter the non-sequential group through the entry
port will diffract only along the primary order.
Rays which originate from a non-sequential source will diffract only along the primary order if ray splitting is
off. If ray splitting is on, then the diffraction may be optionally controlled by the settings on the "Diffraction" tab of
the Object Properties dialog box. This tab includes optional settings which will split the ray by order; allowing more
than one diffraction order to be simultaneously traced.
For more information on diffraction, see Diffraction tab on page 350.

The object properties dialog box
For all objects inserted into the NSC Editor, the object properties dialog box is used to define properties such
as the object type, special aperture settings, scattering and coating properties, and gradient index properties. The
previous/next object buttons allow rapid navigation through all objects. The object properties dialog box has
multiple tabs as described below.
Type tab
The Type tab supports the following controls:
Type: Used to select the general object type, for example, sphere, ellipse, rectangle, or other type.
Data File: If the object type is defined by an external file, such as a polygon object, the file name may be
selected here.
User Defined Aperture: If selected, a user defined aperture (UDA) file will be used to define the extent of the
object. Not all objects support this feature. For more information see page 352.
File: If a user defined aperture is selected, the name of the aperture file may be selected from this control.
Edit User Aperture: This button will invoke a text editor to allow user editing of the selected UDA file. The UDA
file needs to saved and the object reloaded to make the changes effective.
UDA Scale: The UDA Scale is a dimensionless multiplier that scales the aperture defined in the UDA file.
Color: This control is used to select the color the object will be drawn with on the shaded model display.
Opacity: If the opacity is set at 100%, then the object will be rendered on the shaded model plot as a solid
color, and the object may fully obscure other objects from view. If the opacity is less than 100%, then the object
is partially transparent, which allows other objects to be visible through the partially opaque surface.
Row Color: This control chooses the Cackler of the row in the NSC Editor for the object. By default, objects
made of glass, "MIRROR", or "ABSORB" material, as well as sources and detectors are color coded. Any
object may use either no color, the default color, or a user defined color. The user defined colors are described
in Colors on page 63. The coloring of rows may be disabled, see Editors on page 61.
Consider Objects: For a detailed discussion of this feature see "Defining paths using the Consider Objects list"
which follows on page 351.
Use Consider Objects When Splitting: If checked, then when rays are split the child rays will only look for
intersections with the objects listed on the consider objects list. If unchecked, all possible object intersections
will be checked. See "Defining paths using the Consider Objects list" which follows on page 351.
Rays Ignore This Object: If an object is included for rendering or reference purposes only, ray tracing will be
faster if this box is checked, as ZEMAX will not bother to check if the ray intercepts the object. Checking this
option on can cause incorrect ray trace results if any rays could intersect the object, as rays will travel through
the object as if it were not there.
Use Global XYZ Rotation Order: If checked, the rotation convention for object tilts is to first rotate about the X
axis, then the Y axis, then the Z axis. It unchecked, the rotations are done about the Z axis, then the Y axis,
then the X axis. This latter ZYX convention is what a sequential coordinate break surface does if the order flag
is zero. Note that rotating using the ZYX convention is exactly the same as rotating about the X axis, then
rotating about the "new" Y axis, then finally the "new" Z axis.
Object Is A Detector: If checked, then rays striking facets on the object will increment a detector associated
with the faceted representation. Only objects that have a faceted representation support this feature; for other
objects, this option will be unavailable.
Show As: This control selects the color scheme used to represent faceted objects that are detectors on the
shaded model plot. Color by flux will color each facet according to the total power incident on that facet. Color
by irradiance will color each facet by the power per area; which is the flux divided by the area the facet
represents. Generally choosing irradiance will be yield more easily interpreted results if the facets are not all
the same size. Only faceted objects that are detectors use this feature.
Do Not Draw Object: If checked, then the object will not be drawn on layout plots. Rays will still act as though
the object were there. This control is useful to remove from the layout an object that encloses other objects,
so the inner objects may be seen more easily.

Draw Local Axis: If checked, the local x, y, and z axis will be drawn on the 3d Layout plot. The length of the x
and y axis drawn is half that of the "missed ray draw distance", see Missed Ray Draw Distance in Lens Units
on page 96. The z axis is twice this length. The vertex of the coordinate indicator is at the local vertex.
Coating/Scattering tab
The Coating/Scattering tab supports the following controls:
Coating/Scatter Group: For a description of coating/scattering groups, see page 352.
Profile: A profile is a collection of settings related to the thin film coatings and scattering model data applied
to a coating scattering group. Suppose that many objects in an optical system are all composed of the same
material and have the same coating and scattering properties. Rather than type this identical data in for each
object; the data may be typed in once, then saved to a profile with a user defined name. Once saved, any other
object may use the same profile. Profiles are stored in the file whose name is specified on the Files tab of the
System, General dialog box; see Scatter Profile on page 91 for details. If unique settings for this object are
desired, then choose "Use definitions below" to edit the scattering parameters. If a profile is selected; then the
coating and scattering data boxes will be disabled, as this data is defined by the selected profile. However, the
settings will be displayed.
Save: Saves the current scattering settings as a new profile. Pressing this button will prompt for the name of
the new profile.
Delete: Removes the currently selected profile from the Scatter Profile file.
Coating: The name of the coating to apply to the surfaces within the selected group. See page 352. Coatings
are defined in the system coating file, see Defining coatings in ZEMAX on page 501 for details.
Scatter Towards: For a detailed discussion of this feature see "Defining a Scatter Towards list" which follows
on 351.
For all other scattering controls, see the discussion Scattering on page 352.
Bulk Scattering tab
The Bulk Scattering tab is used to define bulk (volumetric) scattering of solids. The tab supports the following
controls:
Model: Select either no bulk scattering, angle scattering, or DLL defined scattering. For a detailed description
of bulk scattering, see Bulk scattering on page 357.
Mean path: The mean path between scatter events.
Angle: The angular cone in which the rays scatter.
DLL: The name of the DLL which defines the scattering function.
Remaining controls: The remaining controls are used to define the parameters to be passed to the DLL. To
make these parameters variable or under multi-configuration control, use the NPRO operand; for details see
NPRO on page 473.
GRIN tab
The GRIN tab is used to define the properties of solids made of a gradient index material. The tab supports
the following controls:
Use DLL defined GRIN media: If checked, then the object will use an externally provided DLL to define the
properties of the gradient index media.
DLL: The name of the DLL to use.
Maximum Step Size: The maximum step size to use during piecewise ray tracing.
NPRO on page 473.
For more information on defining GRIN media, see page 359.

Diffraction tab
For important information that applies to all diffractive objects, see Diffraction from NSC objects on page 347.
The Diffraction tab is used to define the properties of diffractive surfaces. The tab supports the following
controls:
Split: Selects how rays are split off from diffractive surfaces. The options are:
Dont split by order: The rays will not split at the surface. Only the order defined by the object parameters
will be traced, and all the transmitted energy goes into this one order.
Split by table below: A user defined number of rays will be traced over a range of integral orders. The fraction
of energy given to each order is defined by the user in a table.
Split by DLL function: An external DLL program is used to define which orders are traced, how much energy
each order is given, and optionally, what the output ray properties are (for user defined diffraction). For more
information on defining Diffraction DLLs, see page 361.
DLL: The name of the DLL to use. This DLL must be placed in the \DLL\DIFFRACT subdirectory of the
\OBJECTS directory. See Directories on page 59.
Start/Stop order: The beginning and ending order number. These numbers determine how many cells in the
table are active; and how many times the DLL will be called to compute the output ray properties.
NPRO on page 473.
Sources tab
The Sources tab is used to define properties of source objects, including polarization state, coherence length,
and initial phase, position, and direction of rays of light emanating from NSC sources. For important information
see Defining the incident polarization on page 510. The tab supports the following controls:
Random Polarization: If checked, the source will emit randomly polarized light. If unchecked, the polarization
state may be defined using other controls on this tab.
Jx, Jy: The magnitude of the electric field in the local x and y directions, respectively.
X-Phase, Y-Phase: The phase in degrees of the electric field in the local x and y directions, respectively.
Initial Phase: The initial phase of the ray in degrees, with 360 degrees being equal to one wave of optical path.
This setting only affects coherent ray computations which depend upon the phase of the ray.
Coherence Length: The length of ray propagation in lens units over which the phase is known. For details on
Coherence Length effects see Coherence length modeling on page 359.
Pre-Propagation: The distance in lens units the ray is propagated before beginning the actual ray trace through
NSC objects. Pre propagation moves the starting point of rays forward or backward along the ray direction
cosines. This feature is useful for defining rays at one position but beginning the ray trace at a different position
along the ray path; such as prior to or after an object near the source. The Pre-Propagation distance will alter
the initial phase and electric field of the ray to account for the propagation length. Pre-Propagation occurs
before reversing the rays if the reverse rays option is selected.
Reverse Rays: Checking this option will reverse the direction cosines of every ray. This is useful for reversing
the initial direction of rays from the source. Reverse rays is done after the Pre-Propagation distance is
considered, if any.
Since rays in general occupy a 3D space, the electric field in the Z direction is determined by the above data
assuming the Ex and Ey values are the projections down on to the local YZ plane of the source. The resulting
electric field will always be perpendicular to the ray propagation vector.
The polarization properties of the source should only be set non-random if the source is reasonably well
collimated. Partially polarized sources may be simulated by superimposing sources with different polarization
properties at the same location.

Defining paths using the Consider Objects list
It is often the case that rays propagating within an NSC group follow a reasonably well defined path. For
example, when several lenses are mounted in a tube, rays striking lens #5 will either hit lens #4 or lens #6 next;
or possibly the tube itself. If the list of possible objects to intersect is known, and the number of objects on this
list is small compared to the total number of objects; then significant ray tracing speed gains may be made by
telling ZEMAX to only consider the subset of objects the ray may hit.
The Object Type tab has a user-definable list labeled "Consider Objects" for doing this. If left blank (the default),
then rays leaving this object may strike any object, and ZEMAX will use its own internal algorithms for determining
which object is actually hit.
To specify the subset of objects the ray may hit when leaving an object, list the object numbers separated by
a space. For example, to specify that rays leaving an object may hit objects 4, 6, or 23, enter the string "4 6 23"
on the Consider Objects data field. Note the objects own object number should normally be included in the list.
ZEMAX will automatically update this table as new objects are inserted or deleted. The maximum number of
objects that may be listed is 10. If a ray may hit more than 10 different objects; leave this field blank. Use zero for
the object number of the exit port.
If you are unsure as to what objects a ray leaving any object will hit, leave this field blank. If any object is listed;
only the listed objects will be tested for ray intercepts! This means incorrect ray tracing results will occur if the list
does not explicitly include all possible objects that may be hit by rays leaving the current object. If the ray can
leave the object, and strike the same object again, the consider objects list for that object must explicitly list itself
as one of the possible objects. For example, if rays striking object 5 can refract or reflect and hit object 5 again,
object number 5 must be listed in the consider objects list for object 5 if a consider objects list is defined.
Defining a Scatter Towards list
When scattering rays for a stray light analysis, a very large number of rays may need to be traced to find a
relatively small number of rays that strike an object of interest, such as a detector. The Scatter Towards feature
speeds up the analysis by ignoring scattered rays which do not propagate towards an object of interest.
The Scatter Towards list is very similar to the Consider Objects list described in the preceding section. The
Scatter Towards list is a string of integer object numbers separated by spaces. A scattered ray will only be traced
if the ray intersects one of the objects listed.
If you are unsure as to what objects a ray leaving any object will hit, leave this field blank. If any object is listed;
only the listed objects will be tested for ray intercepts! This means incorrect ray tracing results will occur if the list
does not explicitly include all possible objects that may be hit by rays leaving the current object. It is not required
for an object to consider itself; ZEMAX will always check to see if a ray strikes the same object multiple times
automatically.
For example, suppose object 2 defines the Scatter Towards list as "2 3 4". If a ray scatters off object 2, it will
only be traced if the ray will intersect objects 2, 3, or 4. Note that possible objects to trace towards may or may
not include the object that scattered the ray.
Note that choosing an object on the Scatter Towards list does not guarantee that a scattered ray will be traced
toward that object. The scattered rays are generated based upon the scattering properties of the object the
incident ray strikes (see Scattering on page 352). ZEMAX will generate the scattered rays, then ignore those
rays that do not intersect any of the listed objects. The method used by ZEMAX does not ensure that a ray will
scattered toward a listed object. However, the number of scattered rays (see Fraction to scatter and number of
scatter rays on page 352) may be made large so that it is more probable some of them will head in the desired
directions.
Although algorithms exist that can guarantee that a ray scatters in a particular direction, for proper energy
normalization they require that the integral of the scatter distribution function over the projected solid angle of the
target object be known. Since this integral is extremely difficult to compute in virtually all cases, the algorithms
require severe approximations be made to the integral which may introduce data inaccuracy. ZEMAX does not
make these solid angle integral approximations, and uses only the exact scatter distribution function in computing
ray energies and scatter directions.

To turn off the Scatter Towards feature, leave the list blank. In this case, all scattered rays will be traced. The
key difference between the Scatter Towards list and the Consider Object list is that the Scatter Towards list only
applies to scattered rays, while the Consider Object list applies to both specular and scattered rays.
User defined apertures
User defined apertures (UDA) may be placed on some NSC surface objects. UDAs are defined exactly as
described for sequential surfaces, see User defined apertures and obscurations on page 69 for a complete
description. To place an aperture on an object surface, open the Object properties dialog box, choose the "Type"
tab, select the "User Defined Aperture" checkbox, and select the UDA data file name from the menu. The UDA
file must be placed in the \Objects directory.
Coating and scatter groups
A coating/scatter group (CSG) is a collection of one or more facets or faces to an object upon which the same
thin film coating and scattering properties may be applied. For example, for a singlet lens, there are 3 CSGs: the
front face, the back face, and all remaining faces (which include the edges and squared faces around the edges).
In this case, 3 different coatings need to be applied; one to each coating group, if all faces of the lens are coated.
There are currently 4 CSGs per object; although most objects use only a single CSG. CSGs are numbered
from 0; so the valid groups are 0 through 3. The group properties are defined with each object. Coatings and
scatter properties are applied to each CSG on the Object Type selection dialog box. The CSG control also
describes which CSG is used for which faces of the object selected.
Polarization and thin film coatings
Rays traced through non-sequential components may be done while accounting for polarization effects, or
polarization may be ignored. The initial polarization state for a ray is determined by the source properties, see
Sources tab on page 350.
If polarization ray tracing is being used, transmission, reflection, and absorption of optical energy is accounted
for at all surfaces. Bulk absorption is also accounted for.
Thin film coatings significantly affect transmission and reflection properties of optical surfaces. Surfaces are
initially uncoated, but coatings may be applied to surfaces or group of surfaces.
See the previous section for a discussion of coating and scatter groups (CSG). Each CSG may have its own
thin film coating applied.
Coatings on surfaces in contact
If two surfaces are in contact, such as two 45-45-90 prisms placed so that one face of each prism is in contact
with the other, then a coating may be applied "between" the surfaces in contact.
This is accomplished using the same convention described above in the section "Object Placement". The rule
is: the LAST object listed in the NSC Editor determines the properties of the interface between two objects.
For example, to place a thin metal coating on the interface between two prisms arranged to form a beam
splitter, the first prism object listed should have the contact surface "uncoated", while the second object listed
should have the contact surface coated with the appropriate thin film coating. Rays striking this interface from
either side will see the correct coating, and the ray transmission and/or reflection will be correctly computed. Note
prisms modeled as POB objects can have different coatings applied to different faces, so some faces may be
anti-reflection coated while others are coated with a reflective coating.
Scattering
Scattering may occur at any ray-surface intersection. Scattering properties are defined for each coating and
scatter group (CSG) on an object. See the section Coating and scatter groups on page 352 for information on
CSGs. The default scatter model is "No scattering", which means no scattering will occur. The resulting ray is
called the unscattered ray or sometimes the "specular" ray (even if the surface is not actually reflective).
Fraction to scatter and number of scatter rays
If a scattering model other than "No scattering" or "ABg" is selected, the "Fraction to Scatter" must be defined.
This fraction must be between zero (no rays will be scattered) and 1.0 (every ray will be scattered). For the ABg

model scattering, the fraction to scatter is determined by the ABg parameters, see Defining ABg data on
page 356.
If ray splitting is off, the decision to scatter or not to scatter is made by the generation of a single random
number between 0.0 and 1.0. If this random number is larger than the fraction to scatter, the ray will not scatter,
otherwise, the ray will scatter. For example, if the fraction to scatter is 1.0, the ray always scatters. If the fraction
to scatter is 0.0, the ray will never scatter. If the fraction to scatter is 0.25, then on average one out of four rays
will scatter. All of the energy of the ray follows the randomly generated scatter path. The number of scatter rays
has no affect if ray splitting is off.
If ray splitting is on, then ZEMAX will split the specular ray into one or more scattered rays, while still possibly
tracing the specular ray. The specular ray will receive a fraction of the original energy equal to (1.0 - f) where f is
the fraction to scatter. The remaining energy will be divided equally among the one or more scattered rays. The
number of scatter rays determines how many scatter rays will be generated. For example, if the fraction to scatter
is 1.0, then the specular ray will receive zero energy and will no longer be traced; and all the energy will be divided
equally among the scattered rays. If the fraction to scatter is 0.0, no scattered rays will be traced, and the specular
ray retains all the original energy. If the fraction to scatter is 0.25 and the number of scatter rays is 5, then the
specular ray will receive a relative energy of 0.75, and each of the 5 scattered rays will have a relative energy of
0.05. If the number of scatter rays is set to zero, then the fraction to scatter is ignored and no scattering occurs.
The fraction to scatter may be interpreted as the "Total Integrated Scatter" or TIS, which accounts for all energy
other than that which is specularly reflected or refracted, or is absorbed.
Scatter models
Scatter models are defined in terms of an intensity probability distribution function. When ZEMAX scatters a
ray, a new direction of propagation is chosen. The direction is chosen using a probability function and one or more
random numbers. The net effect is that if many rays are traced, the resulting scattered ray distribution would
approach the probability distribution function. The following figure defines the vectors used to describe the
scattering models.
S
R
I
v v
x
v
0

The normal vector N defines the orientation of the surface at the ray-surface interception point. The incident

ray vector is I , the specular ray vector is R , and the scattered ray vector is S . The specular vector may be the

reflected or the refracted ray vector; the figure shows this ray as the reflected ray vector for simplicity. Note N ,

I , R , and S are all unit vectors. The projection of the specular and scattered ray vectors down to the surface are
denoted by 0 and , respectively. The projections are not unit vectors; the vector 0 has a magnitude equal to
sin r , while has a magnitude of sin s , where r and s denote the angle between the normal vector and the
specular and scattered ray angles, respectively. The scatter vector is denoted x , and if x tends to zero, the
scattered and specular vector become the same. How x is determined depends upon the scatter model chosen.
There are five scattering models available: none, Lambertian, Gaussian, ABg, and user defined. Each
available scatter model is defined in the following sections.
No scattering
No modification to the specular ray is made, x is zero.
Lambertian scattering
Lambertian scatter means the scattered intensity probability distribution function goes as cos s . Note
Lambertian scattering is independent of both the ray incident and exiting angle. The scatter vector x is chosen
randomly within a unit circle, and the vector 0 is set to zero length. Most diffuse surfaces are very nearly
Lambertian.
Gaussian scattering
Gaussian scattering is of the form:
2
x
----------
-

2
P ( x ) = Ae
where A is a normalizing constant. A random number is used to generate an appropriate value for x . The
resulting distribution is rotationally symmetric in direction cosine space, no matter what angle the specular ray
makes with respect to the surface normal. The value for is defined on the CSG property dialog box, and
corresponds to the 1 over e (36.79%) probability point in direction cosine space for the scatter vector magnitude.
The value for should be between 1E-15 and 4.0, which correspond to nearly specular and Lambertian limits,
respectively.
ABg model scattering
The ABg scattering model is a very powerful and widely used method for defining the Bidirectional Scattering
Distribution Function, or BSDF. BSDF is defined as the scattered radiance per unit incident irradiance, or
dL s ( s, s )
BSDF ( i, i, s, s ) = ---------------------------
dE i ( i, i )
where is measured from the normal, and is the azimuthal angle, and the subscripts i and s refer to incident
and scattered directions, respectively. Note BSDF has units of inverse steradians.

The general term BSDF can refer to two separate functions, the BRDF and BTDF, for reflective and transmitted
distributions, respectively.
BSDF properties
For many optical surfaces, the BSDF is independent upon incident direction if it is plotted as a function of
direction cosines instead of angles. This simplification is described in James E. Harvey, "Light-Scattering
Properties of Optical Surfaces", Ph.D. Dissertation, University of Arizona (1976), and J. Harvey and A. Kotha,
"Scattering Effects from Residual Optical Fabrication Errors", Proc. SPIE, July, (1999).
This representation plots BSDF as a function of 0 , which is x . This quantity is the distance between
the scattered and unscattered ray vectors when projected down to the scattering surface (see the discussion
under "Scattering" above).
This scattering model is generally accepted as a good model to use when the scattering is mainly due to
random isotropic surface roughness, and the scale of the roughness is small compared to wavelength of light
being scattered. These assumptions are generally valid for polished optical surfaces.
The ABg BSDF model
When the BSDF is plotted as a function of x , the resulting data may usually be fit to a function of the form
A
BSDF = -----------------g- .
B+ x
The three parameters of the fit, A, B, and g, give rise to the name ABg BSDF model. For a description, see E.
R. Freniere, G. G. Gregory, and R. C. Chase, "Interactive Software for Optomechanical Modeling," Proc. SPIE
Vol. 3130, p128 (1997). The following restrictions are placed on the coefficients of this model: A must be greater
than or equal to zero, and B must be greater than 1E-12 unless g = 0. If g is zero, B may be zero. If A is zero, no
scattering will occur. If g is zero (note g may be any value, positive or negative, but it typically is between 0 and
3), then the BSDF is constant in direction cosine space:
A
BSDF = ------------- ,
B+1
and the resulting scattering is Lambertian. If A, B, and g are set to provide a relatively flat BSDF curve, it is far
faster numerically to use the Lambertian model instead.

A typical ABg BSDF curve is shown below for the parameters A = 0.002, B = 0.0005, g = 2.0.
BSDF for A=0.002, B=0.0005, g=2.0
1
1 . 10
BSDF( x)
3
1 . 10
4 x 0
1 . 10 1 . 10
The ABg BSDF model has several interesting properties (when g is not zero!):
g
When 0 B , the BSDF curve flattens out to a value of A/B as 0 goes to zero.
g
When 0 B , the BSDF becomes a straight line on the log-log plot, with a slope of -g.
The transition point between the flat and sloped parts of the curve occurs at x = t , where
t = exp 10 ------------------ .
log 10B
g
Defining ABg data
It is frequently the case where the same ABg data will be applied to many objects in the optical system (for
example, when several objects are all painted with the same type of paint). To avoid the need to type in the same
data redundantly, and to simplify the task of editing the ABg data, ZEMAX provides an ABg data catalog. The
catalog may be edited directly in ZEMAX; see ABg Scatter Data Catalogs on page 203. Note the ABg data
defined must integrate over all possible scatter angles to a total value of less than 1.0 to conserve energy.
Scattering rays using the ABg model
ZEMAX allows definition of separate ABg data for reflection and refraction. If the specular ray reflects or
refracts, the ray is subsequently scattered using the appropriate coefficients. The ABg data names which appear
in the dialog box control are those defined by the ABg scatter data file. The ABg data file to be used is selected
from the "Files" tab of the General dialog box, see page 91 for details.
ZEMAX first decides whether or not to scatter the ray, as described in the section "Scattering" above. If the
ray scatters, a scattered ray is randomly generated. The ray direction is generated so that if a large number of
rays were scattered, the appropriate BSDF function would result.
User defined scattering
Completely general surface scattering may be defined via an external program called a Dynamic Link Library
(DLL). Sample DLLs are provided with ZEMAX with source code. New DLLs may be easily created with a suitable
compiler. See also Comments about DLLs on page 377.

Defining an object to use DLL defined surface scattering
To make an object use a DLL defined scatter function, choose the "User Defined" scattering type from the
object properties dialog box, then select from the available DLL scatter functions listed in the "DLL Name" box.
DLL parameters
Each DLL may use up to 6 user defined data values as parameters in the computation of the scattering
properties. These values are defined by the DLL and are only used by the DLL.
Creating a new DLL
UserScatterDefinition
UserParamNames
When a ray is scattered, ZEMAX passes to the UserScatterDefinition function the local x, y, and z coordinates
on the surface, the local normal vector, the local specular ray vector, polarization data, and other parameter data
as defined by the user. UserScatterDefinition then is required to determine the following values:
Whether or not the ray actually scatters
The direction cosines of the scattered ray
The attenuation of the ray, if any
Optionally the new electric field vector (if this is not provided ZEMAX will make a reasonable guess)
These values are returned to ZEMAX and are used to continue the trace. The function UserParamNames is
used to define the names of all used parameters. These names appear in the Coating/Scattering tab of the object
properties dialog box. Surface scattering DLLs must be placed in the \DLL\SurfaceScatter subdirectory of the
\OBJECTS directory. See Directories on page 59.
Note that the "Fraction to scatter" control is active when User Defined scatter is selected. Only the fraction of
energy which is allowed to scatter is sent to the DLL for possible scattering. However, the DLL may choose to not
scatter the ray (see for example the sample DLL in the next section).
More details can be found in the source code of the sample surface scattering DLLs provided with ZEMAX.
Sample surface scattering: TwoGaussian.DLL
The surface scattering sample TwoGaussian.DLL supports a scatter model which is the superposition of 4
scattering models: specular, Lambertian, and two individual Gaussian models. The DLL uses the parameter data
to define the fraction of energy which scatters according to the Gaussian Sigma 1, the fraction which scatters
according to Sigma 2, and the fraction that is Lambertian. The remaining fraction, if any, is specular. To see how
this is accomplished in the DLL, review the source code file Two Gaussian.c provided with ZEMAX.
Bulk scattering
Bulk (volumetric) scattering models random scattering of rays while propagating through a solid object. ZEMAX
supports 3 modes for the bulk scattering of a solid object:
-No Bulk Scattering
-Angle Scattering
-DLL Defined Scattering
Each model is described below.
No Bulk Scattering
If no bulk scattering is selected, rays will propagate through the solid without scattering.

Angle Scattering
Angle scattering uses a simple model for scattering within a solid. Rays traveling a distance x within the media
have an integrated probability of having been scattered given by
x
p ( x ) = 1.0 e , where
1
= ----- ,
M
and the symbol M is the mean free path in lens units. Note that as x increases, the probability that the ray has
scattered asymptotically approaches 1.0. Setting this expression to a randomly selected value between 0.0 and
1.0, then solving for x, yields a randomly generated path length with the correct statistics. If this path length is
greater than the distance the ray propagates to the next ray-object intersection, then no scattering occurs;
otherwise, the ray scatters at the specified position along the propagation direction.
Once the position is determined, the scattering is modeled by a new ray direction being chosen. For angle
scattering, a random angle is chosen so that the new ray direction lies uniformly distributed within a cone that
makes an angle with respect to the current ray direction. The semi-angle of the cone is one of the parameters to
the model, and this parameter should be set to between 0.0 and 180 degrees. The latter value will scatter the ray
randomly in any direction. Once the ray scatters, ZEMAX automatically will adjust the polarization vector and
randomize the phase of the scattered ray.
DLL Defined Scattering
If the angle scattering model is not sufficient, more complex bulk scattering functions may be defined via an
external program called a Dynamic Link Library (DLL). Sample DLLs are provided with ZEMAX with source code.
New DLLs may be easily created with a suitable compiler. See also Comments about DLLs on page 377.
Defining an object to use DLL defined bulk scattering
To make an object use a DLL defined bulk scatter function, choose the Bulk Scattering tab from the object
properties dialog box, then select Model: "DLL Defined Scattering". The available DLL functions are then listed
in the "DLL:" control.
DLL parameters
media properties. These values are defined by the DLL and are only used by the DLL.
Creating a new DLL
UserBulkDefinition
UserParamNames
While ray tracing through a solid using DLL defined bulk scattering, ZEMAX passes to the UserBulkDefinition
function the current propagation length along the ray, various ray data values, and other parameter data as
defined by the user. UserBulkDefinition then is required to determine if the ray will scatter at some point along the
propagation length. If the ray will scatter, then UserBulkDefinition must determine the following values:
The position at which the ray will scatter
The new direction cosines of the ray
The attenuation of the ray, if any
Optionally the new electric field vector (if this is not provided ZEMAX will make a reasonable guess)
These values are returned to ZEMAX and are used to continue the trace. The function UserParamNames is
used to define the names of all used parameters. These names appear in the Bulk Scatter tab of the object

properties dialog box. Bulk scattering DLLs must be placed in the \DLL\BULKSCAT subdirectory of the \OBJECTS
directory. See Directories on page 59.
Coherence length modeling
By default, non-sequential sources in ZEMAX are either purely monochromatic, or simulated to be polychro-
matic by tracing more than one monochromatic wavelength. Physically, no source is perfectly monochromatic. A
more realistic model for a nominally monochromatic source is to account for the range of wavelengths emitted by
the source. A convenient parameter is the coherence length; defined as
c
x = ----- ,
f
where x is the coherence length, c is the speed of light in vacuum and f is the range of frequencies emitted
by the source, also called the bandwidth. A more convenient representation is the range of wavelengths emitted
defined in terms of the coherence length:
2

= ------ .
x
Using this expression, it is easy to compute the variation of optical path length (phase) as the ray propagates.
This is modeled in ZEMAX by randomly choosing a wavelength within ( ) 2 for purposes of computing the
optical path and thus the phase. ZEMAX chooses the random wavelength with a uniform distribution. The greater
the path difference between two rays from the same source, the greater the difference in phase between the rays
will be; which means coherent effects such as fringe visibility will be reduced.
ZEMAX actually traces the ray using the central wavelength. The randomized wavelength is only used to
compute the effects of coherence length on the phase. Therefore, only coherent computations will exhibit the
phase degradation. To turn off coherence degradation, either set the coherence length to a very large number or
to zero. To make sources completely incoherent, set the coherence length to a small but non-zero number.
Defining GRIN media for non-sequential ray tracing
Any solid enclosed volume may be made of a homogeneous material (the default assumption) or the media
may be of a gradient index. All gradient index materials used by non-sequential objects are defined in a separate
program called a Dynamic Link Library (DLL). Numerous DLLs are provided with ZEMAX with source code. New
DLLs may be easily created with a suitable compiler. See also Comments about DLLs on page 377.
Defining an object to be of gradient index material
To make an object media by of a gradient index material, choose the GRIN tab from the object properties dialog
box, then select "Use DLL defined GRIN media". The available gradient index materials are then listed in the
"DLL:" control.
Discussion on maximum step size for GRIN objects
The maximum step size determines the tradeoff between ray tracing speed and accuracy. The exact value
required depends upon the rate at which the index of refraction changes and the desired accuracy of the
computations. If the maximum step size is too large, the ray trace results will have large errors, and some rays
will miss objects they would otherwise hit, or vice-a-versa.
GRIN DLL parameters
media properties. These values are defined by the DLL and are only used by the DLL.
Creating a new GRIN DLL
UserGrinDefinition

UserParamNames
While ray tracing through a GRIN media, ZEMAX passes to the UserGrinDefinition function the current x, y,
and z location of the ray (in the objects local coordinate system), the wavelength, and other parameter data as
defined by the user. UserGrinDefinition then is required to compute the following four values:
n (the index of refraction)
n * dn/dx (the index times the derivative with respect to the x direction)
n * dn/dy (the index times the derivative with respect to the y direction)
n * dn/dz (the index times the derivative with respect to the z direction)
These four values are returned to ZEMAX and are used to trace the ray through the grin media.
The function UserParamNames is used to define the names of all used parameters. These names appear in
the GRIN tab of the object properties dialog box. GRIN DLLs must be placed in the \DLL\GRINS subdirectory of
the \OBJECTS directory. See Directories on page 59.
Sample GRIN DLLs
The following sample GRIN DLLs are provided with ZEMAX.
SAMPLE GRIN DLLS

GRIN1 Defines a GRIN media of the form:
2
n = N 0 + N r1 r + N r2 r ,
similar to the sequential Gradient1 surface type.
2 2 4 6 8 10 12
n = n 0 + n r2 r + n r4 r + n r6 r + n r8 r + n r10 r + n r12 r ,
2 4 6 2 3
n = n 0 + nr2 r + n r4 r + n r6 r + n z1 z + n z2 z + n z3 z ,
2 2 2
n = n 0 + nx1 x + n x2 x + n y1 y + n y2 y + n z1 z + n z2 z ,
2 4 2 3 4
n = n 0 + n r2 r + nr4 r + n z1 z + n z2 z + n z3 z + n z4 z ,
similar, but not identical to, the sequential Gradient5 surface type. Note there is no
dispersion in this grin model (no dependence on wavelength).
2 4
n = n 0 + n1 r + n 2 r , where
2 C D
n x = A x + B x + -----2x- + -----4x- , for all three values of n x ,

which is similar to the sequential Gradient6 surface type.

A g
n = n 0 1.0 --- r , where
2
where both A and n 0 are functions of wavelength:
2
K1 K 2
A ( ) = K 0 + -----2- + -----4- , and

C
n 0 = B + ----2- .

This is similar to the sequential Gradient9 surface type, with the additional parameter
"g" which is fixed at 2.0 for the sequential surface Gradient9.
2 3 4 5 6
n = n 0 + n y1 y a + n y2 y a + n y3 y a + n y4 y a + n y5 y a + n y6 y a ,
where y a = y and the || symbols indicate the absolute value. This form of gradient
has a discontinuity at the plane y = 0, and the gradient is bisymmetric about the y =
0 plane. This is similar to the sequential Gradient10 surface type, but without the
dispersion model.
TORUS Defines a GRIN media of the form:
2
n = N 0 + N r2 t ,
where t is the distance from an axis of rotation defining the toroidal symmetry of the
gradient index. The torus radius is one of the defining parameters. The torus radius
is measured from the coordinate z = 0.0 on the object to the location of the axis of
rotation defining the torus.
Defining DLLs for ray splitting at diffractive surfaces

Any diffractive surface may split rays that started at a non-sequential source, by order, to allow for simultaneous
tracing of multiple diffracted orders. How many rays split, and how much energy is in each order, may be defined
externally in a separate program called a Dynamic Link Library (DLL). Numerous DLLs are provided with ZEMAX
with source code. New DLLs may be easily created with a suitable compiler. See also Comments about DLLs
on page 377.
Defining an object to use the diffraction DLL
To make a diffractive object use the DLL, choose the Diffraction tab from the object properties dialog box, then
select "Split by DLL function". The available DLLs are then listed in the "DLL:" control.
Diffraction DLL parameters
diffractive surface properties. These values are defined by the DLL and are only used by the DLL.
Creating a new Diffraction DLL
UserDiffraction
UserParamNames
While ray tracing through a diffractive surface, ZEMAX passes to the UserDiffraction function the current x, y,
and z location of the ray (in the objects local coordinate system), the wavelength, the present order being traced,
and other ray specific data and parameter data as defined by the user. UserDiffraction then is required to compute

the relative energy in the specified order, and return this value to ZEMAX. ZEMAX will refract or reflect the ray
into the specified diffraction order and give the resulting ray the specified fraction of the input ray energy computed
by the DLL. If the new child ray has sufficient energy, the ray is traced.
The Diffraction DLL format also supports as an option for the DLL to compute ALL output properties of the ray;
including direction cosines, electric field polarization vector, and transmitted energy. If the DLL also computes
these properties, the DLL must set a flag indicating the values were passed back to ZEMAX for further ray tracing.
See the sample DLLs for documentation on this feature.
The function UserParamNames is used to define the names of all used parameters. These names appear in
the Diffract tab of the object properties dialog box. See the sample DLLs for the use of this feature. Diffraction
DLLs must be placed in the \DLL\DIFFRACT subdirectory of the \OBJECTS directory. See Directories on
page 59.
Ray splitting
Generally, when a ray strikes a surface, part of the energy will be reflected, part will be transmitted, and
depending upon the surface properties, part may be absorbed. Ray splitting refers to the ability of ZEMAX to
compute both the reflected and refracted paths, and then continue to trace both rays. Rays reflected from
otherwise refractive interfaces are commonly called ghost reflections.
Of course, once a ray splits, each of the "child" rays will in general strike another object, and the rays may split
again and again. After many ray-object intersections, the total number of rays can become extremely large, so
controls must be placed on the ray tracing to ensure that the computation will eventually end.
There are several ways to restrict the amount of ray splitting that occurs:
Maximum number of ray-object intersections: This control defines how many times a ray along any path, from
the original source parent ray to the final ray-object intersection, may intersect another object.
Maximum number of ray segments: A ray segment is that portion of a ray path from one intersection to the
next. When a ray is launched from the source, it travels to the first object. That is 1 segment. If the ray then
splits into 2 rays, each of those are another segment (for a total of 3). If each of those rays split again, there
will be 7 segments. Generally, the number of ray segments grows far faster, and needs to be set much larger,
than the number of ray-object intersections does.
Minimum relative ray intensity: As each ray splits, the energy decreases. The relative ray intensity is a lower
limit on how much energy the ray can carry and still be traced. This parameter is a fraction, such as 0.001,
relative to the starting ray intensity from the source. Once a child ray falls below this relative energy, the ray is
terminated.
Minimum absolute ray intensity: This parameter is very similar to the minimum relative ray intensity, except it
is absolute in system source units rather than relative to the starting intensity. If this is zero, the absolute ray
intensity threshold is ignored. Source units are defined on the Units tab of the System, General dialog box.
See Source Units on page 87 for details.
All of these settings are defined on the Non-Sequential tab of the System, General dialog box. See Non-
Sequential on page 94 for details.
Ray splitting and polarization
Because accurate reflection and transmission computation requires polarization information, ray splitting is
only allowed when performing polarization ray tracing.

Ray splitting can be turned off, and in this case the transmitted
path is always taken at a refractive interface unless the ray TIRs;
the reflected path is of course always taken if the surface is a
mirror.
The figure at right shows some of the ray paths possible when
rays are split. There is only 1 input ray drawn.
When using ray splitting, the number of rays being traced can
become enormous if the minimum relative ray energy is set too low.
For example, for rays traced inside a cube of glass with 50%
coatings on all surfaces, a relative ray intensity of 0.01 will yield
about 256 ray segments. If the relative ray intensity is set to 1E-8,
about 270 million ray segments will be generated for EACH ray
traced into the cube! Systems with relatively low reflectivity (such
as AR coated optics) will not generate as many ray segments because the intensity of the reflected paths falls so
much more quickly. Still, it is advised to set the relative ray transmission reasonably high, around 0.001, until the
model is working well and more detailed results are needed.
Because the total number of rays gets so large, the 3D Layout plots can become very cluttered. One way to
reduce the amount of rays drawn is to use the "filter" string to suppress drawing of some of the rays, see The
filter string on page 371.
Putting it all together
The real benefit of the non-sequential ray tracing, polarization, scattering, and ray splitting features occurs
when they are all used together to determine ray distributions and detected energies. The order in which these
effects are all accounted for is as follows:
The object interface the ray strikes is computed first.
If the surface is reflective or the ray TIRs, the reflected path is chosen.
Otherwise,
If ray splitting is on, the reflected and the refracted path are both determined.
If ray splitting is off, the refracted path is chosen.
The ray(s) are scattered if scattering is enabled.
The ray(s) intensity and polarization vectors are adjusted.
The ray(s) continue on to the next object interface.
Performing a Monte Carlo ray trace
On 3D Layouts and other analysis features which draw NSC objects and rays, the effects of sources, ray
splitting, and scattering may be seen from the randomly generated rays on these plots. Both ray splitting and ray
scattering may be selected on or off on these plots to isolate their effects. The layout plots are extremely useful
for qualitatively seeing the effects of sources, objects, scattering, and ghost or other reflections.
To accurately compute quantitative energy distributions on the placed detectors generally requires a larger
number of rays than the qualitative analysis of the layout plots. This is why the number of layout and number of
analysis rays are defined separately in the source parameter listing.
The Ray Trace/Detector Control
The "Ray Trace/Detector Control" dialog box listed under "Detectors" on the NSC Editor menu bar controls
the ray tracing for analysis using defined sources, objects, and detectors. The controls on the dialog box are
described below.

RAY TRACE/DETECTOR CONTROL OPTIONS
Item Description
Clear Detectors Select either "All" or any one specific detector to clear. This resets all of the detectors
pixels to zero energy.
Trace Begins tracing rays from all defined sources. Once the trace begins, only the Terminate
and Auto Update controls are available. See the discussion below on other effects of
initiating a ray trace.
Terminate Stops ray tracing. This action may require a brief time.
Auto Update If checked, all detectors will be redrawn approximately every 60 seconds, so progress
may be monitored. See Choosing the number of analysis rays on page 365.
# CPU's Selects the number of CPU's over which to spread the ray tracing task. More than 1 may
be selected, even on a single CPU computer, in which case the single CPU will time
share the multiple simultaneous tasks. The default is the number of processors detected
by the operating system.
Use Polarization If checked, polarization will be used to determine ray energy, reflection, transmission,
absorption, and thin film effects. If Split Rays is checked, Use Polarization will be
checked automatically, as it is required for accurate ray splitting.
Split Rays If checked, rays will split according to the interface properties as described in Ray
splitting on page 362.
Scatter Rays If checked, rays will scatter according to the scatter properties of the interface.
Ignore Errors If checked, errors generated by ray tracing are ignored, and any errors that occur will
not terminate the ray trace. This should be turned on with caution, since errors may
indicate a serious flaw in the system being traced. However, some perfectly good
systems occasionally have a few rays fail because of finite computer precision or other
minor problems. If the "lost energy (errors)" given at the end of the ray trace is small
relative to the total power of all sources defined, then the few rays which trigger errors
can be safely ignored. Increasing the number of facets that represent complex objects
may decrease lost ray energy. See Lost energy on page 365.
Save Rays If checked, all rays traced will be saved to the file name specified. The file name only
should be provided, as the path will be the same as the lens file being traced. The
extension should be ZRD. The ZRD extension is used by ZEMAX for listing files in the
data base viewer. Saving rays slows down ray tracing significantly, and should only be
used when the data is needed for subsequent analysis. For more details, see the
discussion below. See also "Filter" below.
Filter This option is only available if "Save Rays" is checked on. If the filter is not left blank,
then the defined filter will be applied to each ray tree traced before it is saved to disk.
Only if the ray tree passes the filter will the ray tree be saved. This may substantially
reduce the amount of data saved, and therefore will reduce the resulting ZRD file size.
For a description of the filter syntax, see The filter string on page 371.
There are some important considerations to understand when performing a Monte Carlo ray trace. When the
detectors are cleared, all the energy counts in each pixel are reset to zero. Once the Trace button is selected,
ZEMAX will trace exactly the number of specified analysis rays for each source listed in the NSC Editor. Each
source has an associated power in watts or other units (Analysis Units on page 87). The initial intensity of each
ray from a given source is the total power divided by the number of rays to be traced. For example, for a 1 watt
source tracing 1,000,000 rays; each ray initially has 1.0E-06 watts of power associated with it.
If the ray tracing is terminated before all the rays are traced; the total power and peak power as listed on the
detector objects will be inaccurate; although the ray distribution will be accurate for the number of rays traced.

The analysis cannot be paused and resumed; so once a ray trace is terminated, selecting Trace again will start
a new trace.
If the detectors are cleared, then Trace is pressed, the ray trace completes, and Trace is pressed again; all
detectors will show the total power from both traces. ZEMAX has no reliable way of knowing how many times
rays have been traced between detector clearings; since sources may be added, deleted, or modified at any time.
For accurate power computations, the detectors should be cleared before tracing; then a single trace should
be performed without early termination.
Choosing the number of analysis rays
If the total number of rays to be traced is large, it may be important to understand the order in which the rays
are traced. ZEMAX divides the ray tracing task into multiple "threads", where each thread may execute in parallel
on separate CPUs if the computer is so equipped. Even on single CPU computers, the ray tracing is divided into
threads. The reason ZEMAX divides the ray tracing task into threads is to give a proportional representation of
all sources present early in the ray trace. If all the rays from source 1 were traced before any rays from source 2
were traced, a proportional early estimate of the final results would not be possible. The manner in which this
division is done is affected by the number of analysis rays chosen for each source.
ZEMAX divides the total number of rays into an integral number of groups with equal rays, and this integral
number must be a power of either 2 or 5, whichever yields the larger number of threads. Further, the number of
rays in each group must be at least 400, and the total number of threads must be 1,024 or less. Both of these
limits are somewhat arbitrary but are reasonable limits for efficient ray tracing.
For example, suppose a source uses 100,000 analysis rays. The power of 2 division would yield 32 threads
of 3,125 rays each; note 2^5 = 32 and 32 * 3125 = 100,000. The power of 5 division would yield 125 threads of
800 rays each; note 5^3 = 125 and 125 * 800 = 100,000. ZEMAX would therefore choose 125 threads of 800 rays
each, since this is the larger number of threads.
If multiple sources are defined, ZEMAX uses the same logic but applies it to all sources, choosing again the
largest integral number of threads that evenly divides all sources. For example, suppose source 1 uses 10,000
rays and source 2 uses 4,000 rays. The power of 2 division yields 8 threads of 1,250 rays from source 1 and 500
rays from source 2. The power of 5 division yields 5 threads of 2,000 rays from source 1 and 800 rays from source
2. In this case the power of 2 division would be used because the number of threads is larger.
Once the number of threads is selected, ray tracing of the first thread begins. If additional CPUs are available
for ray tracing, the second and subsequent threads are launched simultaneously as well until all CPUs are fully
utilized. When each thread completes, the next thread is launched, until all threads are complete and all ray
tracing is finished.
In summary, for long ray traces where early proportional results are desired, choose the number of analysis
rays to be powers of 2 or 5 times 400. Large numbers that work well include 10,000, 50,000, 250,000, 500,000,
1,000,000, and 5,000,000. Larger numbers also work fine, but the total number of threads will not increase.
Lost energy
When tracing some NSC systems, not every ray can be traced successfully. When a ray trace is terminated,
the energy associated with the ray is called "lost energy". The lost energy is measured in source units (see
Source Units on page 87).There are different causes for terminating the trace of a particular ray:
- If a ray falls below the minimum energy threshold for tracing, the ray will be terminated. The minimum relative
and absolute energy thresholds are described in Non-Sequential on page 94. The ray trace control reports this
type of lost energy as the energy lost due to threshold settings.
- Sometimes geometry errors or round-off errors lead to conditions where the ray can no longer be traced.
Geometry errors are usually caused by poorly formed solids or improper nesting of objects. For some rays, correct
intercept or refraction data cannot be computed. This may happen when iteration to a surface fails, or the coating
data cannot be computed. Rays that hit the maximum number of object intercepts, nesting levels, or segment
levels, are also terminated with an error condition. The ray trace control reports these types of lost energy as the
energy lost due to errors.
- Rays that strike the edge of an object may be terminated if the local refraction and/or reflection cannot be
computed accurately. A ray is considered to strike the edge of an object if the intercept point lies within a threshold

distance to the edge. For most non-faceted objects, the threshold distance is approximately the glue distance in
lens units. For facets along the edge of a faceted object, the threshold distance is approximately the glue distance
(interpreted here as a dimensionless number) multiplied by the dimensions of the facet. Note that rays that strike
very close to the edge of a part have no useful physical interpretation anyway; an actual object would diffract (or
scatter) all energy incident within a few wavelengths of the edge. Rays that miss all objects, or are absorbed by
either coatings, bulk transmission, or detectors, are also terminated. ZEMAX absorbs all these rays without error
or report, as the energy is not considered lost because the energy is known to be absorbed or propagated out of
the system and is thus accounted for.
If the fraction of lost energy is relatively small, lost energy can be ignored. Even well designed and accurate
optical models will occasionally have spurious ray errors. Rays that strike within edges or seams between parts
and are thus absorbed will usually represent an insignificant fraction of the total energy in the system. If the lost
energy is unacceptably high, then the settings (see for example Minimum Relative Ray Intensity on page 95) or
the object geometry needs to be checked.
Ray database files
Ray data may be saved for subsequent analysis by checking on "Save Rays" and providing a file name in the
edit box provided. The file name should end in ZRD, although any valid file name and extension will be accepted.
As the rays are traced, the ray data will be saved to the file specified.
Saving the ray data will slow down the ray tracing significantly, because of the (relatively) slow speed of the
hard disk access. However, it is often convenient to trace the rays once; then do multiple analysis on the resulting
data. Rays contained in a ZRD database may be drawn on layout plots rather than tracing new rays; see NSC
3D Layout on page 107 for details.
The ZRD file format is binary to keep the file as compact as possible.
The first record in the file is a single 32-bit integer indicating the version number of the ZRD file format. ZEMAX
will check this number when reading the file; to be sure the format is what ZEMAX expects. The second record
in the file is a single 32-bit integer indicating the maximum possible number of segments in the remaining records.
The actual number of segments will always be less than or equal to this number.
The remainder of the file consists of a 32-bit integer indicating the number of segments that follows, then that
many segments, then another 32-bit integer indicating the number of segments in the next record, and so on until
the end of the file.
A single ray segment is described by a "C" structure with the following format, packed on 8 byte boundaries
to yield a size of 200 bytes:
typedef struct
{
unsigned int status;
int level;
int hit_object;
int in_object;
int parent;
int storage;
int xybin, lmbin;
double index, starting_phase;
double x, y, z;
double l, m, n;
double l2, m2, n2;
double path_to, intensity;
double phase_of, phase_at;
double exr, exi, eyr, eyi, ezr, ezi;
} RAYPATH;
Each data item is described below.
status: bitwise flags indicating the status of the ray. The flags (in integer notation) are: terminated: 1, reflected:
2, refracted: 4, scattered: 8, diffracted: 16, ghost: 32, diffracted from previous object: 64, scattered from
previous object: 128, ray segment had fatal error: 256, bulk scattered: 512.
level: The number of ray segments between the ray segment and the original source. This is the number of
ray-object intercepts the ray has accumulated.

hit_object: The object number the ray intercepted. If zero, the ray did not hit anything. For the zero segment,
this value will be the source segment object number.
in_object: The object number the ray is propagating inside of. This generally determines the index of refraction
the ray is traveling through. A value of zero means inside the "background media".
parent: The (prior) ray segment from which the ray originated. If zero, the ray came from the original source
point. Note more than 1 child ray may have the same parent; but each ray has only one parent. The parent
segment number will always be less than the child rays segment number.
storage: A temporary buffer used by ZEMAX for indexing. For segment 0 only, this integer is the ray wavelength
number. Other segments may contain meaningless data.
xybin, lmbin: The pixel number on a detector object which the ray struck. The xybin is for spatial data, the lmbin
is for angular (intensity) data.
index: the index of refraction of the media. Not an adequate description for gradient index media.
starting_phase: the initial optical path length the ray starts with. This is usually the contribution from diffraction
gratings when rays are split off.
x, y, z: the global coordinates of the ray intercept.
l, m, n: the global direction cosines of the ray. Not an adequate description for gradient index media.
nx, ny, nz: the global normal vector of the object at the intercept point.
path_to: the physical (not optical) path length of the ray segment.
intensity: the intensity of the ray segment.
phase_of: the phase of the object. This is the optical phase path length equivalent in lens units added by
gratings, holograms, and other phase modifying surfaces.
phase_at: the accumulated total phase of the ray. This value is only computed at detector surfaces.
exr, exi, eyr, eyi, ezr, ezi: the electric field in the global x, y, and z coordinates, real and imaginary parts. These
values will all be zero if the ray trace did not use polarization.
The ZRD format is subject to change, however, ZEMAX will automatically convert older formats to the current
format. Note the structure size is 200 bytes for each ray segment. If 100,000 rays are traced with an average of
50 segments each, the resulting file will be about 100,000 x 50 x 200 bytes = 1Gb in size. The file sizes can quickly
become enormous, so great care must be taken to limit the number of segments before deciding to save the ray
data to a file. Use of a filter string to limit the saved data to the rays of interest may substantially reduce the amount
of data written to disk.
The ZRD file consists of alternating blocks of data. For example, if three rays are traced, with 5, 7, and 9
segments each, then the ZRD file will be formatted as follows:
4 bytes:version
4 bytes:maximum number of segments possible, will be 9 or larger in this example
4 bytes:5 (the number of segments in ray 1)
200 bytes x 5 (the segment data for ray 1)
For information on what ZEMAX can do with the ZRD file, see the next section.
Ray Database Viewer
The Ray Database Viewer reads in a previously saved ZRD file, and displays the data in a text format. For
information on creating a ZRD file, see the previous section.

RAY DATABASE VIEWER OPTIONS
Item Description
File The name of the ZRD file to use. ZEMAX will look in the directory where the current lens
is saved, and list all files with the ZRD extension found there.
Show Unprocessed If checked, the complete data tree for each ray traced will be displayed.
Data
Expand Into If checked, each branch of the tree will be expanded to show a single complete path
Branches from source to ray termination.
XYZ If checked, the global X, Y, and Z coordinate data of the ray segment-object intercept
point will be displayed.
LMN If checked, the global L, M, and N direction cosine data of the segment will be displayed.
Normal If checked, the global Nx, Ny, and Nz object normal vector components at the ray
segment-object intercept point will be displayed.
Path If checked, the path length in lens units from the parent segment is displayed. This is
the physical, not the optical path length. ZEMAX will also show the phase in radians, but
only for segments striking a detector surface.
Exyz If checked, the global electric field in the X, Y, and Z directions will be displayed.
First/Last Ray The first and last number of the rays to display. These controls are very useful for limiting
the length of the displayed data. If the subset data is to be saved in a new ZRD file, only
rays within this range will be included.
Apply Filter If checked, the filter string entered will be used to reduce the amount of data displayed.
The filter string is also used to determine the subset of the segments which are saved
to a new ZRD file. For information on the filter string syntax, see The filter string on
page 371.
Save Subset Data If checked, the data displayed which passes the filter string will be saved in ZRD format
As in the file name specified. The file name should end in the extension ZRD, with no path
provided. The path used will be the same as the current lens file. To prevent accidental
overwrite of the subset data file, this control is automatically "unchecked" after every
save.
Save Rays on If checked, ray segments which hit the specified object number and which pass the
Object "n" As defined filter (if any) will be saved as source rays in a binary source file format of the
specified name. Note the ray position data is saved in global coordinates. This file may
subsequently be used as a source file. For information on the ray source file format, see
Source File on page 340. The file name should end in the extension DAT, with no path
provided. The path used will be the same as for ray source files, typically
ZEMAX\OBJECTS. To prevent accidental overwrite of a previously created data file, this
control is automatically "unchecked" after every save.
Only rays within the range of the first/last ray setting are saved. To save all rays that hit
the specified object from the ZRD file, choose 1 for the first ray and any integer greater
than the number of rays in the ZRD file for the last ray. If the total number of rays exceeds
1000, then printing of the individual ray data in the ray database report will be
suppressed to speed up the process. To see the text listing, choose "Update" again from
the text window menu bar. The second update will show the text data because the "Save
Rays On Object" box is automatically unchecked after every save.
The text display for the ray database viewer can be quite long, so it is best to limit the amount of data displayed
by using the first/last controls and the filter string.
A typical unprocessed ray listing looks like this, with some columns removed for brevity:
Ray 1, Wavelength 1:
Seg# Prnt Levl In Hit XRTS DGEF BZ X Y Z

0 0 0 0 0 ---- ---- -- 0.0000E+000 5.0000E+000 0.0000E+000
1 0 1 0 2 -**- ---- -- 0.0000E+000 3.0870E+000 9.0000E+000
2 1 2 0 0 *--- -*-- -- 0.0000E+000 2.7044E+000 7.2002E+000
3 1 2 2 2 -**- ---- -- 0.0000E+000 2.5468E+000 1.2838E+001
4 3 3 2 2 -**- -*-- -- 0.0000E+000 9.5506E-001 9.0000E+000
5 4 4 2 2 -**- -*-- -- 0.0000E+000 -6.9886E-001 1.2988E+001
6 5 5 2 2 -**- -*-- -- 0.0000E+000 -2.0350E+000 9.0000E+000
7 6 6 2 2 --*- -*-- -- 0.0000E+000 -3.2848E+000 1.2730E+001
8 7 7 0 3 *--- ---- -- 0.0000E+000 -5.7054E+000 1.8400E+001
9 6 6 0 0 *--- ---- -- 0.0000E+000 -2.9069E+000 7.3797E+000
10 5 5 0 3 *--- ---- -- 0.0000E+000 -4.2858E+000 1.8400E+001
11 4 4 0 0 *--- ---- -- 0.0000E+000 -9.6429E-002 7.4900E+000
12 3 3 0 3 *--- ---- -- 0.0000E+000 9.6510E-001 1.8400E+001
"Seg#" is the segment #, and is sequential from 0 to the total number of segments the ray generated.
"Prnt" is the parent number, which indicates which prior segment generated that particular segment. For
example, the parent of segment 11 is segment 4, whose parent is 3, whose parent is 1, whose parent is 0.
Segment 0 is the source segment, and the starting coordinates indicate the location of the source point. Any
segment can be traced back to the source in this manner.
"Levl" is how many segments away from the source ray the segment is. For segment 11, there are 4 segments
back to the source (4, 3, 1, and 0).
"In" indicates what object number the segment is traveling through.
"Hit" indicates the object number struck at the end of the segment. The source segment has a hit number of
0, since the XYZ coordinates listed are the starting and not the terminating coordinates of the segment. For all
other segments, the XYZ coordinates and other data are at the segment end on the "hit" object. If the hit object
is 0, the ray missed all objects and is terminated.
"XRTS DGEF BZ" is a quick indicator of what happened to the segment. X means terminated, R reflected, T
transmitted, S scattered, D diffracted, and B bulk scattered. All these values indicate what happened at the end
of the segment, when the segment struck the object "hit". The values G, E, and F indicate the segment was a
ghost, diffracted, or scattered, respectively, from the parent segments "hit" object number, so these events
occurred at the start of the segment. The symbol Z indicates a ray error occurred and the ray was terminated.
Note more than one event may occur for any ray segment.
The other data displayed includes the global XYZ coordinates of the end of the segment, the global normal
vector coordinates of the object struck at the end of the segment, the ray intensity, the object comments, and
possible other data as this feature is expanded.
The Detector Viewer
The Detector Viewer displays data from detectors in either a graphic or text format. For information on tracing
rays to record data on detectors, see Performing a Monte Carlo ray trace on page 363.
DETECTOR VIEWER OPTIONS
Item Description
Surface Selects the NSC surface number. This is always surface 1 when in NSC mode.
Detector Selects the detector object. If the window is a text window, there is also an option to
summarize all detectors.
Show As Choose grey scale or false color display, or cross section plots of rows or columns. Cross
sections are not available in text displays.
Row/Column Selects the row or column to display for cross section plots.

Z-Plane This control is only used by detector volume objects. The Z-Plane selects which cross
section of a volume detector to display. The number of Z-Planes in a detector volume is
set by the number of voxels in the Z direction. The text version of the Detector Viewer
also supports an "All option to display all cross sections as well as a table of all voxel
center coordinates and data. Pressing the left and right arrow keys while viewing
detector volume data will cycle through all the Z-Planes.
Smoothing Smoothing reduces pixel noise due to low sampling by averaging the data in adjacent
pixels. The data in every pixel is replaced by the average of the data in the pixel and the
neighboring 8 pixels. This 3 x 3 pixel averaging algorithm is applied the number of times
specified by the smoothing parameter.
Show Data See discussion below for important information.
Ray Database If none is selected, the ray data currently stored on the detectors is used. Otherwise, the
previously saved ray database data is used.
Filter A filter string may be applied to select only rays with particular properties. This option is
only available when using a ray database. For information on the filter string syntax, see
The filter string on page 371.
Scale Choose linear or logarithmic scales. See "Max/Min Plot Scale" below.
Max/Min Plot Scale If non-zero values are provided, the plot will be scaled to the specified min and max data
range. If these values are both zero, the default scaling will be applied. If the maximum
value of the data is larger than the Max Plot Scale, the Max Plot Scale is ignored and
default scaling will be applied. The Max and Min plot scale values are ignored when the
Scale (see above) is Logarithmic.
Discussion:
Once ray data has been recorded on the detectors, or from rays stored in a ray database, the Show Data option
determines what aspect of the data is displayed as follows:
Incoherent Irradiance: This is incoherent power per area as a function of spatial position on the detector. The
power of each ray that strikes the same pixel is summed without regard to the phase of the ray. This option
becomes Incoherent Illuminance when using photometric rather than radiometric units.
Coherent Irradiance: This is coherent power per area as a function of spatial position on the detector. The
complex amplitude of each ray that strikes the same pixel is summed, keeping track of the real and imaginary
parts separately considering the phase of each ray. The final amplitude is then squared to yield the coherent
power. This option becomes Coherent Illuminance when using photometric rather than radiometric units.
Coherent Phase: The phase angle of the complex amplitude sum used in Coherent Irradiance.
Radiant Intensity: The power per solid angle in steradians as a function of incident angle upon the detector.
This option becomes Luminous Intensity when using photometric rather than radiometric units.
Radiance (Position Space): The power per area per solid angle in steradians as a function of spatial position
on the detector. Since it is not possible to display the variation in angle space and position space at the same
time, this plot is effectively identical to the Incoherent Irradiance display, with the data values divided by two pi.
Note this scaling assumes the detector is collecting data over a half hemisphere, no matter what angular range
the detector actually collects rays over. This option becomes Luminance when using photometric rather than
radiometric units.
Radiance (Angle Space): The power per area per solid angle in steradians as a function of incident angle upon
the detector. Since it is not possible to display the variation in angle space and position space at the same time,
this plot is similar to the Radiant Intensity display, with the data values divided by the product of the area of the
detector and the cosine of the angle at the center of the pixel. This option becomes Luminance when using
photometric rather than radiometric units.
Incident Flux: For detector volumes, the incident flux is the flux entering each voxel, measured in source units
(watts, lumens, or joules).

Absorbed Flux: For detector volumes, the absorbed flux is the flux absorbed within each voxel, measured in
source units (watts, lumens, or joules).
Absorbed Flux/Volume: For detector volumes, the absorbed flux/volume is the flux absorbed within each voxel,
measured in source units (watts, lumens, or joules) per unit volume measured in cubic lens units.
The filter string
It is frequently convenient to display only NSC rays or data which have certain properties. For example, when
scattering and splitting are turned on, the layout diagram will become very confusing if many rays are traced. The
filter string allows definition of a "test" rays must pass before they are drawn or displayed. The filter string syntax
consists of logical operations between flags that indicate if one segment within a ray hit, miss, reflected, refracted,
scattered, diffracted, or ghost reflected from an object within the NSC group. There are also "Extended" flags
which support numerical arguments. The supported flags are described in the table below.
FILTER STRING FLAGS
Flag Description
Bn Ray bulk scattered inside of object n. If the n value is 0, then bulk scattered segments
from any object will return true for this test.
Dn Ray diffracted after striking object n. See En.
En Diffracted from parent segments object n. This flag only gets set for ray segments split
from diffractive elements, for order numbers other than zero, when ray splitting is on.
Fn Scattered from parent segments object n. This flag only gets set for ray segments split
from scattering surfaces when ray splitting is on. The specular segment does not get
this flag, only scattered segments. If the n value is 0, then scattered segments from any
object will return true for this test.
Gn Ghost reflected from parent segments object n. This flag only gets set for ray segments
reflected from refractive objects when ray splitting is on. If the n value is 0, then ghost
segments from any object will return true for this test.
Hn Ray hit object n. To test whether a ray hit an object, the flag is of the form Hn. For
example, to test if a ray hit object 5, the flag would be H5.
Jn Similar to Gn, except that all segments prior to the ghost reflection point are set to have
zero intensity. This allows Detector Viewers to look only at ghost energy, not direct
incident energy, even if the ray later ghosted off another object. The zero intensity values
will only affect the Detector Viewer, not the ray database viewer or layouts.
Mn Ray missed object n. To test whether a ray missed an object, the flag is of the form Mn.
For example, to test if a ray missed object 15, the flag would be M15.
On Ray originated at source number n. O0 (that is "O" as in Origin and "0" as in zero) will
select all sources.
Rn Ray reflected after striking object n. The flag R7 would test if the ray reflected after
striking object 7. See Gn.
Sn Ray scattered after striking object n. See Fn.
Tn Ray transmitted (refracted) in to or out of object n. The flag T4 would test if the ray
refracted in or out of object 4 after striking the object.
Wn Ray uses wavelength n. If the n value is 0, then rays with any wavelength will return true
for this test.

X_GHOST(n,b) Ray segment has ghosted exactly b times, and has hit object n at least once. If n is zero,
any ray segment that has ghosted b times will pass the test. For example, to consider
only all second generation ghosts (ghost rays from ghost parents), use X_GHOST(0, 2).
X_GHOST does not consider ghost ray segments that end in a TIR condition; although
rays that TIR are considered ghosts. For example, if a third generation ghost ray leaves
one surface, strikes another surface, and then TIRs from this second surface,
X_GHOST(0, 3) will not include this segment because the segment ended in a TIR and
not a ray termination (the ray reflected and continued). This same segment will however
be included in the filter X_GHOST(0, 4) because the ray ghosted a fourth time (at the
TIR point). This is an artifact of how ZEMAX defines segments and counts ghost rays.
In all cases, all ghost rays can be found if sufficiently high values of b are tested. Note
rays which TIR from refractive surfaces are considered ghosts, but rays reflected from
mirror surfaces are not.
See also Gn.
X_IAGT(n,v) Ray has absolute intensity greater than value v on object n. If the ray never strikes object
n, this flag is false.
X_IALT(n,v) Ray has absolute intensity less than value v on object n. If the ray never strikes object
n, this flag is false.
X_IRGT(n,v) Ray has intensity relative to initial intensity greater than value v on object n. If the ray
never strikes object n, this flag is false.
X_IRLT(n,v) Ray has intensity relative to initial intensity less than value v on object n. If the ray never
strikes object n, this flag is false.
X_LGT(n,v) Ray has local incident x ray direction cosine greater than value v at point on object n. If
the ray never strikes object n, this flag is false.
X_LLT(n,v) Ray has local incident x ray direction cosine less than value v at point on object n. If the
ray never strikes object n, this flag is false.
X_MGT(n,v) Ray has local incident y ray direction cosine greater than value v at point on object n. If
X_MLT(n,v) Ray has local incident y ray direction cosine less than value v at point on object n. If the
X_NGT(n,v) Ray has local incident z ray direction cosine greater than value v at point on object n. If
X_NLT(n,v) Ray has local incident z ray direction cosine less than value v at point on object n. If the
X_SCATTER(n,b) Ray segment has scattered exactly b times, and has hit object n at least once. If n is
zero, any ray segment that has scattered b times will pass the test. For example, to
consider only first generation scatter rays, use X_SCATTER(0, 1). See also Sn.
X_XGT(n,v) Ray has local x coordinate greater than value v at point on object n. If the ray never
X_XLT(n,v) Ray has local x coordinate less than value v at point on object n. If the ray never strikes
object n, this flag is false.
X_YGT(n,v) Ray has local y coordinate greater than value v at point on object n. If the ray never
X_YLT(n,v) Ray has local y coordinate less than value v at point on object n. If the ray never strikes
X_ZGT(n,v) Ray has local z coordinate greater than value v at point on object n. If the ray never

X_ZLT(n,v) Ray has local z coordinate less than value v at point on object n. If the ray never strikes
Z Ray has fatal error.
All flags except the O, G, E, and F flag indicate what happened at the end of the ray segment, at the point
where the segment intercepted the object indicated by the "n" value. The G, E, and F flags indicate that the ray
segments parent struck object "n" and then launched a child ray that was a ghost, diffracted, or scattered
segment, respectively. The O flag indicates the ray originated from source object "n". The W flag indicates the
ray is of wavelength number "n". The flags of the form X_***(n,v) require two numerical arguments within the
parentheses separated by a comma, or a syntax error will be issued.
Each flag is evaluated for each ray traced, and the flag is assigned a status of TRUE or FALSE. The flags may
be used alone, or may be combined using logical operands. Logical operands generally act on two other logical
flags (the exception is the NOT operand which acts only on the flag to the right of the operand). The supported
logical operands are:
&: Logical AND. Both flags on either side of the & symbol must be TRUE for the AND operation to return TRUE.
|: Logical OR. If either of the flags are TRUE, OR returns TRUE.
^ Exclusive OR (XOR). If either of the flags are TRUE, but not both, XOR returns TRUE.
! Logical NOT. Returns TRUE if the flag to the right was FALSE and vice-a -versa.
The parentheses symbols may also be used to set operator precedence.
Filter string examples
To select only rays that hit object 7, the filter string would be:
H7
To select only rays that hit both objects 7 and 9, the filter string would be:
H7 & H9
To select rays that must have either a) hit object 7 and object 9, but did not reflect off object 6, or b) missed
object 15, the filter string would be:
(H7 & H9 & !R6) | M15
If only ghost rays generated by striking object 3, 4, or 7 are desired, the filter string would be:
G3 | G4 | G7
To select only rays incident at an angle of less than 10 degrees on a flat object number 5 whose normal vector
points along local +z, the filter string would be (note cos(10) = 0.984808):
X_NGT(5, 0.984808)
Note only rays which hit the specified object in the X_NGT flag can possibly make the flag true.
The filter string is checked for basic syntax errors, such as mismatched parentheses, but not all possible syntax
errors are checked and reported.The number of rays which meet the filter string test may be very small, and
perhaps even zero. ZEMAX still traces the number of rays defined for the source (by the # of layout rays
parameter), but only that fraction of these rays which pass the filter will actually be saved, displayed, or drawn.
The filter string may also be applied to ray data base files as the ray data is being saved, see Performing a Monte
Carlo ray trace on page 363.
Defining Polygon Objects
Polygon objects are user-defined objects that are composed of a collection of triangles in 3D space. The
objects are defined in a text file ending in the extension POB (for polygon object).
The file format may be created and edited with any text editor. The file consists of a series of rows of data.
Each row begins with a single letter or symbol, followed by data for that symbol. The supported symbols and their
meanings are defined as follows.

The comment symbol: !
The symbol "!" is used to define a comment line.
Syntax:
! Any comment here
Example:
! A dove prism
The coating group name symbol: C
Names for the coating and scattering groups (CSG) may be defined using the C command.
Syntax:
C csgnumber "any name here"
Example:
C 0 "All faces"
The CSG names are used to display in the surface scattering tab of the object dialog box the names for the
various coating and scatter groups. Note the name must be in quotes. This command is optional.
The invisible symbol: I
Lines which connect vertices that are not to be drawn (such as lines between two adjacent triangles that lie in
the same plane) can be marked as invisible. To mark two vertices as invisible, use the symbol "I" followed by the
two vertex numbers. Any line connecting the marked vertices will not be drawn.
This only affects the 3D Layout plot, and is strictly a cosmetic enhancement to the drawing of polygon objects.
Syntax:
I v1 v2
Example:
I 7 9
Note the vertex numbers must be integers. The numbers are separated by spaces. The I command must
precede any triangles or rectangles that reference the invisible vertices.
The vertex symbol: V
Vertices are defined by the symbol "V" followed by the vertex number and the x, y, and z coordinates of the
vertex.
Syntax:
V number x y z
Example:
V 1 -1.0 -1.0 0.0
Note the vertex number must be an integer, and the x, y, and z coordinates are floating point numbers. The
numbers are separated by spaces.
The triangle symbol: T
Triangles are defined by the connection of 3 vertices.
Syntax:
T vertex1 vertex2 vertex3 isreflective coatinggroup
The vertex numbers must be integer vertices previously defined in the file.
The "isreflective" flag is -1 if the surface absorbs, 1 if the surface reflects, or 0 if the surface refracts. Note using
this flag allows some triangles to be reflective, and others to be refractive or absorptive, within the same Polygon
Object.
The coatinggroup is an integer defining which coating group the facet belongs to. See the section "Coating
Groups" above for details. If the coatinggroup flag is omitted; coating group 0 is assumed.
Example:

T 1 2 3 0 2
This would define a refractive triangle connecting vertices 1, 2, and 3, coated with the material of coating group
2.
The rectangle symbol: R
Rectangles are defined by the connection of 4 vertices, otherwise, they are very similar to triangles. Internally,
ZEMAX converts rectangles into 2 triangles.
Syntax:
R vertex1 vertex2 vertex3 vertex4 isreflective coatinggroup
The vertex numbers must be integer vertices previously defined in the file. The order of the vertices is not
arbitrary, they must be listed in a continuously clockwise or counter clockwise order, not a "bow tie" order. Not all
four vertex polygons form a closed rectangle; for more complex shapes use multiple triangle commands instead.
The "isreflective" flag is -1 if the surface absorbs, 1 if the surface reflects, or 0 if the surface refracts.
The coatinggroup is an integer defining which coating group the facet belongs to. See the section "Coating
Groups" above for details. If the coatinggroup flag is omitted; coating group 0 is assumed.
Example:
R 1 2 3 4 1 0
This would define a reflective rectangle connecting vertices 1, 2, 3, and 4, coated with the material of coating
group 0.
Maximum triangles in the polygon object
There is no fixed upper limit on the number of triangles a polygon object may contain. The ultimate limit is
determined by the amount of real or virtual RAM the computer has available. Each triangle takes up about 100
bytes of memory. However, ZEMAX often maintains multiple copies of the lens data simultaneously, and so a
good rule of thumb is that ZEMAX will need 500 bytes of RAM for each triangle. A 2000 triangle object will require
about 1 Mb of free RAM. A more practical limit is computer speed; ZEMAX will slow down noticeably if the number
of triangles becomes very large.
Example POB file
Here is the complete text of a POB file defining a cube of refractive material. The file is included as the example
CUBE.POB. All 8 faces of the cube belong to coating group 0.
! A cube
! ZDC Dec 14, 1998
! front face vertices
V 1 -1 -1 0
V 2 1 -1 0
V 3 1 1 0
V 4 -1 1 0
! back face vertices
V 5 -1 -1 2
V 6 1 -1 2
V 7 1 1 2
V 8 -1 1 2
! Front
R 1 2 3 4 0 0
! Back
R 5 6 7 8 0 0
! Top
R 4 3 7 8 0 0
! Bottom
R 1 2 6 5 0 0
! Left side
R 1 4 8 5 0 0
! Right side
R 2 3 7 6 0 0

The 8 "V" commands define the vertex coordinates of the 8 corners of the cube. The 6 "R" commands define
each of the 6 faces of the cube. Note the width of the faces is 2 units, and the z coordinates of the back face
vertices is defined to be 2 units; so the shape is a perfect cube. All the coordinates are relative to the object
reference point; which in this case will be the center of the front face. To place the reference point at the center
of the cube, change the vertex definitions to:
V 1 -1 -1 -1
V 2 1 -1 -1
V 3 1 1 -1
V 4 -1 1 -1
V 5 -1 -1 1
V 6 1 -1 1
V 7 1 1 1
V 8 -1 1 1
Defining STL Objects
The STL format is commonly supported by mechanical CAD programs to describe arbitrary objects. Objects
are modeled as a collection of triangles, and the global vertices of the triangles are written out to a file. STL is a
good format for faceted objects. For smoothly curved objects, such as lenses, STL is an approximation which
may be of acceptable accuracy for some non-imaging systems.
There are binary and ASCII formats for STL files; and ZEMAX supports both formats.
To use an STL object, choose the object type to be an "STL Object" and enter the name of the file (without the
STL extension) in the comments column of the STL Object row. The STL file must be placed in the \OBJECTS
directory.
Maximum triangles in STL objects
There is no fixed upper limit on the number of triangles an STL object may contain. The ultimate limit is
determined by the amount of real or virtual RAM the computer has available. Each triangle takes up about 100
bytes of memory. However, ZEMAX often maintains multiple copies of the lens data simultaneously, and so a
good rule of thumb is that ZEMAX will need 500 bytes of RAM for each triangle. A 2000 triangle object will require
about 1 Mb of free RAM. A more practical limit is computer speed; ZEMAX will slow down noticeably if the number
of triangles becomes very large.
Example STL files
Several example STL files may be found in the \OBJECTS directory.
Special considerations for faceted objects
Ray tracing through faceted objects is generally straightforward in a mathematical sense, with one major
exception: how to handle the special case of a ray intercepting "exactly" on the common edge of two adjacent
facets. In this context, "exactly" means within the numerical precision limit of the computer. ZEMAX uses double
precision 64 bit numbers for all ray tracing algorithms, and this yields about 12 decimal places of accuracy.
However, because of the frequent use of the square root and other math operations, the practical accuracy of
many computations is somewhat lower.
The most common difficulty is tracing rays through a roof prism, corner cube, or faceted Fresnel lens where
one of the common edges is placed exactly along the x or y axis of the incoming beam. Since the rays hit exactly
on the edge of two facets, each with a different normal vector, the reflection or refraction properties become
ambiguous.
Physically, infinitely sharp edges cannot exist. Therefore, the solution ZEMAX uses is to absorb any ray which
strikes within the glue distance (see page 345) of an edge of a facet that borders another facet with a different
normal vector. An absorbed ray is terminated, and its energy is discarded.
This problem may be avoided by either of these techniques:
For sequential ray tracing, offset the object axis a very small amount in the X or Y directions; so the rays will
"see" one facet at a time. An offset of about 1e-6 lens units (which is only 1 nanometer if the lens units are
mm) is generally sufficient.
For non-sequential ray tracing, either offset the object as for sequential ray tracing, or use sources with random

ray distributions. Occasionally a ray will strike within the glue distance of the facet edge, and the ray energy
will be absorbed; but the total amount of energy should be very small.
Generally, using randomly generated ray sets, such as those in the image analysis or dithered spot diagram,
will eliminate this problem.
Note that any real roof prism or corner will physically not "correctly" reflect or refract rays within very small
distances of the edge due to manufacturing limitations and physical optics interference.
Comments about DLLs
ZEMAX supports several different types of DLLs for extending the capabilities of the non-sequential features,
including DLLs for sources, gradient index media, diffraction, and scattering.
To create a user defined DLL source, a suitable compiler or development tool that can generate 32 bit Windows
compatible DLL's is required. It is also assumed that the user can write the required code, and most importantly,
ensure that the code is reliable and bug-free. To maximize speed, ZEMAX performs very little error checking on
data returned by the DLL, and so buggy DLL's are quite capable of bringing ZEMAX to a crash.
For this reason, technical support on the implementation of DLL's is strictly limited to illustrating that the
provided sample files work correctly. If you need a DLL, and do not possess the desire or ability to write them
yourself, please feel free to contact ZEMAX Development Corporation for a quote on developing a custom DLL
to meet your requirements. ZEMAX Development Corporation has considerable experience in developing
algorithms, and can generally write DLLs at competitive rates on short notice.
The best way to learn the use of the various DLL types is to study an existing DLL. The sample DLLs provided
with ZEMAX include extensive documentation and comments on the data format; see any of the sample source
code files for examples.

Chapter 13 SOLVES
Introduction
Solves are functions which actively adjust specific values. Solves can be specified on curvatures, thicknesses,
glasses, semi-diameters, conics, and parameters. Solves are set by double clicking on the cell you wish to place
the solve on. The solve type control screen appears. Select the solve type from the drop-down list. Some solves
require additional parameters. The available solves and required parameters are summarized in the following
table. Parameters may also have "fixed" or "variable" status; these are not solves, however, they are listed for
completeness. To turn solves off, select the "FIXED" option from the drop-down list. Note that this turns off the
solve, it does not change the last value. Pressing Ctrl-Z twice with the cursor on the solved parameter will also
remove any solve, and this is the quicker way to do so.
SUMMARY OF SOLVES
Solve type First parameter Second parameter Third parameter Code Integer
Curvature: Fixed 0
Curvature: Variable V 1
Curvature: Marginal ray Angle M 2

angle
Curvature: Chief ray angle Angle C 3
Curvature: Pick up Surface Scale factor P 4
Curvature: Marginal ray N 5

normal
Curvature: Chief ray normal N 6
Curvature: Aplanatic A 7
Curvature: Element power Power X 8
Curvature: Concentric with Surface to be S 9

surface concentric to
Curvature: Concentric with Surface to be R 10
radius concentric with
Curvature: F/# Paraxial F/# F 11
Thickness: Fixed 0
Thickness: Variable V 1
Thickness: Marginal ray Height Pupil zone M 2

height
Thickness: Chief ray height Height C 3
Thickness: Edge thickness Thickness Radial height E 4

(use zero for
semi-diameter)
Thickness: Pick up Surface Scale factor Offset P 5
Thickness: Optical path OPD Pupil zone O 6

difference
379 Chapter 13: SOLVES

Solve type First parameter Second parameter Third parameter Code Integer
Thickness: Position Surface Length from T 7

surface
Thickness: Compensator Surface Sum of surface S 8
thicknesses
Thickness: Center of Surface to be at X 9
curvature the COC of
Glass: Fixed 0
Glass: Model Index Nd Abbe Vd Dpgf 1
Glass: Pick up Surface P 2
Glass: Substitute Catalog name S 3
Glass: Offset Index Nd offset Abbe Vd offset O 4
Semi-Diameter: Automatic 0
Semi-Diameter: Fixed U 1
Semi-Diameter: Pick up Surface P 2
Semi-Diameter: Maximum M 3
Conic: Fixed 0
Conic: Variable V 1
Conic: Pick up Surface Scale factor P 2
Parameter: Fixed 0
Parameter: Variable V 1
Parameter: Pick up Surface Scale factor Offset P 2
Parameter: Chief ray Field Wavelength C 3
Curvature: Marginal ray angle

The effective focal length of the lens can be controlled by placing a paraxial marginal ray angle solve on the
curvature of the last surface before the image. For example, suppose a lens has a 20 mm entrance pupil diameter,
and it is desired to constrain the effective focal length to 100 mm. This requires a marginal ray exit angle from the
last surface of -.1 (this number is 20 divided by 2 divided by 100, the minus sign is because the ray is converging,
or going down toward the image plane). See the F number solve.
Curvature: Chief ray angle
The chief ray angle solves work the same way as the marginal ray angle, except the paraxial chief ray is used
for the calculations instead. The chief ray solve is useful for maintaining a particular magnification or for
maintaining collimation. The letter "C" will appear in the curvature cell to indicate that a chief ray angle solve is
active on that surface.
Curvature: Pick up
Another useful curvature solve is the 'pick up' solve. Specifying a surface number tells ZEMAX to insert at this
surface the value from some other surface. An optional scale factor allows the value to be modified. This is useful
for example in systems where the light passes through the same optics twice, such as a null lens. This pickup
feature allows variables to be coupled together, and they will change under the influence of other solves, editing,
or optimization, in accordance with the hierarchy described below.
Chapter 13: SOLVES 380

Pickup solves can only be used when picking up data from previous surfaces; picking up data from surfaces
which follow the solve would lead to unpredictable results. The letter "P" will appear in the curvature cell to indicate
that a pickup solve is active on that surface.
Curvature: Marginal ray normal
This solve will force the surface to be normal to the paraxial marginal ray. This is also called an image-centered
surface. These special surfaces introduce no spherical aberration or coma. The letter "N" will appear in the
curvature cell to indicate that a normal ray angle solve is active on that surface.
Curvature: Chief ray normal
This solve will force the surface to be normal to the paraxial chief ray. This is also called a pupil-centered
surface. These special surfaces introduce no coma, astigmatism, or distortion. The letter "N" will appear in the
curvature cell to indicate that a normal ray angle solve is active on that surface.
Curvature: Aplanatic
This solve will force the surface to be aplanatic with respect to the paraxial marginal ray. These special surfaces
introduce no spherical aberration, coma, or astigmatism. The letter "A" will appear in the curvature cell to indicate
that an aplanatic solve is active on that surface.
Curvature: Element power
The power of an element is given by:
t2
= c 1 ( n 2 n 1 ) + c 2 ( n 3 n 2 ) c 1 ( n 2 n 1 ) c 2 ( n3 n 2 ) ----- .
n2
This solve adjusts the value of c2, to maintain the specified element power. The solve is assumed to be on the
second of two adjacent surfaces. This solve is ignored if the surface number is less than 2, or if n3 and n2 are
identical at the primary wavelength.
Curvature: Concentric with surface
This solve will force the curvature of the surface such that the surface is concentric about the specified surface.
The specified surface must precede the surface on which the solve is placed. The letter "S" will appear in the cell
to indicate that the surface concentric solve is active on that surface.
Curvature: Concentric with radius
This solve will force the curvature of the surface such that the surface is concentric about the same point that
the specified surface is concentric to. The specified surface must precede the surface on which the solve is
placed. The letter "R" will appear in the cell to indicate that the radius concentric solve is active on that surface.
Curvature: F number
This solve will force the curvature of the surface such that the marginal ray angle exiting the surface is -1/2F
where F is the paraxial F/#. The letter "F" will appear in the cell to indicate that the F/# solve is active on that
surface.
Thickness: Marginal ray height
The most common thickness solve is the marginal ray height solve, which can be used to constrain the image
plane to the paraxial focus. To move the image plane to the paraxial focus, double click on the thickness of the
last surface before the image surface, and select "Marginal Ray Height" from the drop-down list which appears.
Since the marginal ray has a height of zero (it crosses the axis) at the paraxial image plane (assuming a
rotationally symmetric system), we would like to set this last thickness to bring the image surface to the point
where the marginal ray height is zero. The default height value is zero, which is desired, and so click on OK to
exit the screen, and the solve immediately adjusts the thickness appropriately.
In principle, the solve could have been set for some other marginal ray height by entering a value other than
zero for height, the first optional parameter. Note that any surface can have a marginal ray height solve, not just
the thickness before the image plane. The "Height" is the height of the marginal ray on the tangent plane of the

next surface (again, not necessarily the image surface). Note the height is measured at the point the ray intercepts
the tangent plane, not the actual curved surface of the next surface.
The third value, "Pupil Zone", allows the ray pupil coordinate to be defined. The default is zero, which indicates
that a paraxial ray should be used. Any non-zero value indicates that a real marginal ray is to be used. The zone
value must be between -1 and 1. This is the Py coordinate, or the normalized entrance pupil coordinate in the y
direction. This solve can be used to constrain that a particular ray, such as the .7 zonal ray, to have zero transverse
aberration on axis. The letter "M" will appear in the thickness cell to indicate that a marginal ray height solve is
active on that surface.
Thickness: Chief ray height
This is similar to the marginal ray height solve, except the paraxial chief ray is used. This solve is useful for
locating a surface at a pupil plane. The letter "C" will appear in the thickness cell to indicate that a chief ray height
solve is active on that surface.
Thickness: Edge thickness
This solve dynamically adjusts the spacing between two surfaces to maintain a specified distance between
the surfaces at a specified radial aperture. This is useful for preventing negative, or overly sharp edges on
elements. The letter "E" will appear in the thickness cell to indicate that an edge thickness solve is active on that
surface. If the radial aperture is set to zero, then the current semi-diameter will be used. See the chapter
"Conventions and Definitions" for an important discussion on the definition of edge thickness.
Thickness: Pick up
Pickup solves also work with thicknesses. The pickup thickness is given by:
T' = O + S T
where T is the thickness of the surface to pick up from, S is the scale factor, and O is the offset. See the
"Curvature: Pick up" section for more information on pickup solves.
Thickness: Optical path difference
This solve will actively adjust the thickness to maintain a specific optical path difference at a specific pupil
coordinate. The OPD is measured at the exit pupil, not at the surface the solve is placed on. The two parameters
to be set are the OPD in primary waves, and the pupil zone at which to evaluate the OPD. For example, to maintain
the focal position so that the real marginal ray has the same optical path length as the real chief ray, define an
OPD solve on the last thickness before the image plane. Set the OPD parameter to zero, and the pupil zone to
1.0. The letter "O" will appear to indicate the OPD solve is active. Perform an OPD plot and verify that the OPD
is in fact zero at the pupil edge. Only the primary wavelength is used, and only the on-axis field is considered.
More complicated OPD constraints are available by using optimization, described in a separate chapter. The letter
"O" will appear in the thickness cell to indicate that an OPD solve is active on that surface.
Thickness: Position
The position solve maintains the distance from a specified reference surface. If the reference surface precedes
the surface on which the solve is placed, then the sum of the thicknesses from the reference surface to the surface
following the solve surface will be maintained at the specified value. If the reference surface follows the surface
on which the solve is placed, then the sum of the thicknesses from the solve surface to the reference surface will
be maintained at the specified value. If the reference surface is the same as the solve surface, then the thickness
of the solve surface will be set at the solve length value.
The position length solve is particularly useful for maintaining the length of a portion of a zoom lens to a fixed
value. The solve can also be used to meet a total lens length constraint. In either case, the solve may eliminate
optimization variables and operands, enhancing optimization convergence and speed. The letter "T" will appear
in the thickness cell to indicate that a total length solve is active on that surface.
Thickness: Compensator
The thickness compensator solve is very similar to the position solve. This solve maintains the thickness on a
surface so that the sum of the surface thickness and the thickness on another "reference" surface equals a

constant. In equation form, this solve maintains the condition T = S - R, where S is the sum of the thicknesses of
the two surfaces, and R is the thickness of the reference surface. The reference surface must precede the surface
with the solve.
Thickness: Center of curvature
This solve will adjust the thickness of the surface to place the following surface at the center of curvature of
any prior surface. The letter "X" will appear in the cell to indicate that the center of curvature solve is active on
that surface.
Glass: Model
The model glass state is not a solve. The model glass is used to idealize a glass by describing the dispersion
in the visible wavelength region using three parameters: the index of refraction at d-light, the Abbe number at d-
light, and a term describing the partial dispersion. For more information on model glasses, see the Chapter "Using
Glass Catalogs".
Glass: Pick up
Pickup solves also work with glasses, see the "Curvature: Pick up" section for a complete description.
Glass: Substitute
The substitute state is not a solve. If the glass solve type is set to "Substitute", then the global optimization
algorithms are permitted to change the glass type during optimization. This is similar to a variable status in that
the glass type can change.
If no catalog name is given (i.e. the catalog field is blank) then glasses may be selected from all catalogs
selected for use on the General System Data dialog box. If a catalog name is given (i.e. Hoya) then only glasses
from that one catalog will be selected.
To prevent only certain glasses from being chosen during optimization, choose "Exclude Substitution" for the
glass that should be avoided on the glass catalog dialog box, or use a glass substitution template; see the Chapter
"Using Glass Catalogs" for details.
See the Chapter "Global Optimization" for more details on glass substitution.
Glass: Offset
The offset solve allows a small change in index and/or Abbe number to be added to the index as computed
by the dispersion formulas and the glass catalog dispersion data. The primary use of this solve is for tolerancing.
There are several conditions for using the index and Abbe numbers in computing the change in index as a
function of wavelength:
-The minimum wavelength in use is greater than 0.3 micrometers and less than 2.5 micrometers
-The dispersion data in the glass catalog must span the wavelength range of 0.4861327 to 0.6562725
micrometers.
If these conditions are met, then the change in index is given by the difference in the model glass index
computed at the base Nd and Vd as compared to the offset Nd and Vd values. See the discussion of the "Model"
glass in this chapter for a discussion of the model glass properties. In equation form the index is:
n = n base + n ( N d + N d, V d + V d, P d ) n ( N d, V d, P d )
where the function n() is the model glass function. Note this model will add a different index offset to the index
at each wavelength. If the wavelength band criteria listed above is not met, then the index offset is added to all
the base index values:
n = n base + N d
The Abbe offset is ignored in this case.

Semi-Diameter: Pick up
Pickup solves also work with semi-diameters, see the "Curvature: Pick up" section for a complete description.
Semi-Diameter: Maximum
The maximum solve is used for setting the semi-diameter value to the maximum required across multiple
configurations. For example, a zoom lens which has three configurations will generally have three different semi-
diameter values for every surface that is using "automatic" semi-diameters. The maximum solve will compute the
semi-diameter for each configuration, and then use the largest of the values.
Conic: Pick up
Pickup solves also work with conics, see the "Curvature: Pick up" section for a complete description.
Parameter: Pick up
Pickup solves also work with all 8 parameters. The pickup parameter is given by:
P' = O + S P
where P is the parameter value of the surface to pick up from, S is the scale factor, and O is the offset. See
the "Curvature: Pick up" section for more information on pickup solves.
Parameter: Chief ray
The chief ray solve works only on coordinate break surfaces, and only on the first four parameters. When
placed on a decenter x or y parameter, the solve will set the decenter to place the selected wavelength (use zero
for primary wavelength) real chief ray from the selected field position at an x or y coordinate of zero on the
coordinate break surface, respectively. If set on the tilt x and tilt y parameters, the tilt angles are adjusted to make
the exiting chief ray angle zero in the y and x directions, respectively. The order flag on the coordinate break
surface will determine the order in which the parameters are fit, and the solution may not be unique. The output
chief ray coordinates can be checked on the Analysis, Calculations, Ray Trace feature; and it is a good idea to
verify the solve works correctly. This solve is not allowed when the stop surface follows the surface the solve is
placed on and ray aiming (see Ray Aiming on page 88) is on.
There are times when the resulting values will yield an output chief ray angle or coordinate that is small, but
not zero. This is caused by the non-orthogonal rotations not always being able to completely "undo" an arbitrary
rotation in a single set of matrix rotations. There are two solutions to this problem:
1) Try changing the "order" flag from 1 to 0 or 0 to 1, then update the system, to see which set of coordinate
transforms works better.
2) Use two adjacent coordinate break surfaces with identical chief ray solves. This is easy to do by copying
and pasting the entire coordinate break surface. The second coordinate break will usually bring the chief ray
many orders of magnitude closer to the desired zero coordinates or angles.
Suggestions for use
There is a hierarchy by which solves are computed. The solves are computed sequentially from the first surface
to the image surface, in this order: curvature, thickness, glass, semi-diameter, conic, and then parameters.
Because curvature and thickness solves can affect the entrance pupil position, ZEMAX does not allow solves
which rely upon ray tracing, such as marginal ray height solves, to be placed before the stop surface. This will
avoid ambiguous or incorrect settings for the solved parameters. There is no way to predict the results of solves
placed before the stop. A similar problem arises when a marginal ray solve is used on the last surface to control
the focal length - the aperture definition must be entrance pupil diameter (rather than F/#) to keep the solution
unique.
Solves are highly efficient, and should be used when possible instead of variables during optimization. For
example, it is usually better to place a curvature solve on the last surface to control effective focal length than to
optimize for focal length explicitly.

Chapter 14 OPTIMIZATION
Introduction
The optimization feature provided by ZEMAX is quite powerful, and is capable of improving lens designs given
a reasonable starting point and a set of variable parameters. Variables can be curvatures, thicknesses, glasses,
conics, parameter data, extra data, and any of the numeric multi-configuration data. ZEMAX uses an actively
damped least squares method. The algorithm is capable of optimizing a merit function composed of weighted
target values; these target values are called "operands". ZEMAX has several different default merit functions,
described in a subsequent section. These merit functions can be changed easily using the Merit Function Editor.
For details on this procedure see the section "Modifying the merit function".
Optimization requires three steps: 1) a reasonable system which can be traced, 2) specification of the
variables, and 3) specification of the merit function. A reasonable system is a rather loose concept which means
that poorly conceived designs are not likely to be transformed into exceptional designs by the optimization
algorithm (although there are exceptions). The variables, and there must be at least one for the optimization
algorithm to be able to make any progress, are specified on the various editors, as described in the next section.
To reach the optimization screen, select Tools, Optimization. You must specify all variables before using
optimization.
The algorithm used by the optimization feature described in this chapter is designed to find the "local" minimum
of the specified merit function. However, ZEMAX also has the capability to search for a "global" minimum of the
merit function. The global minimum is the lowest possible value for the merit function, and if the merit function is
selected appropriately, this implies the best possible solution to the problem. The global optimization feature is
not for novice users, and is not appropriate for interactive designing. For details see the chapter "Global
Optimization".
Selecting variables
Variables for optimization are specified by pressing Ctrl-Z when the highlighted bar is on the parameter to be
varied in the Lens Data Editor. Note that Ctrl-Z is a toggle. The Multi-Configuration and Extra Data Editors also
contain numeric data that may be made variable by using Ctrl-Z. Glasses cannot be made variable directly
because they are discrete. To optimize glasses, see Optimizing glass selection on page 424.
Defining the default merit function
The merit function is a numerical representation of how closely an optical system meets a specified set of
goals. ZEMAX uses a list of operands which individually represent different constraints or goals for the system.
Operands represent goals such as image quality, focal length, magnification, and many others.
The merit function is proportional to the square root of the weighted sum of the squares of the difference
between the actual and target value of each operand in the list. The merit function is defined this way so a value
of zero is ideal. The optimization algorithm will attempt to make the value of this function as small as possible,
and so the merit function should be a representation of what you want the system to achieve. You do not have to
use the default merit function, you may construct your own as described in a later section.
The easiest way to define a merit function is to select the Tools, Default Merit Function option on the Merit
Function Editor menu bar. A dialog box will appear which allows selection of options for the default merit function.
Each option is explained in the following paragraphs.
Selecting the type of optimization
Several different types of merit functions are available. The default merit function is constructed using four key
choices: The optimization type, data type, reference point, and integration method. The choices are described in
the following tables.
385 Chapter 14: OPTIMIZATION

DEFAULT OPTIMIZATION TYPES
Name Description
RMS RMS is an abbreviation for Root-Mean-Square. This type is by far the most
commonly used. The RMS is the square root of the average value of the squares
of all the individual errors.
PTV PTV is an abbreviation for Peak-To-Valley. There are rare cases where the RMS
is not as important as the maximum extent of the aberrations. For example, if all
the rays need to land within a circular region on a detector or fiber. In these cases,
the Peak-To-Valley (PTV) may be a better indicator of performance. This merit
function type attempts to minimize the PTV extent of the errors.
DEFAULT OPTIMIZATION DATA
Name Description
Wavefront Wavefront is the aberration measured in waves. See Comments about RMS
wavefront computations on page 133.
Spot Radius The radial extent of the transverse ray aberrations in the image plane.
Spot X The x extent of the transverse ray aberrations in the image plane.
Spot Y The y extent of the transverse ray aberrations in the image plane.
Spot X and Y Both the x and y extent of the transverse ray aberrations in the image plane. The
x and y components are considered separately, and both are optimized together.
This is similar to Spot Radius, except the signs of the aberrations are retained,
which yields better derivatives. Note that in computing the radius of an aberration,
the sign information is lost.
DEFAULT OPTIMIZATION REFERENCE POINTS
Name Description
Centroid The RMS or PTV computation of the data is referenced to the centroid of all the
data coming from that field point. Centroid reference is generally preferred,
especially for wavefront optimization. For wavefront optimization, reference to
the centroid subtracts out piston, x-tilt, and y-tilt of the wavefront, none of which
degrade image quality. Centroid reference also yields more meaningful results
when coma is present, since coma shifts the image centroid away from the chief
ray location. Historically, the chief ray has been used because it is simpler to
compute, but this complexity is dealt with by ZEMAX with virtually no performance
loss.
Chief The RMS or PTV computation of the "data" is referenced to the chief ray at the
primary wavelength.
Mean This option is only available if the selected optimization data is wavefront. The
mean reference is very similar to the centroid reference, except only the piston
(mean wavefront) is subtracted out; not the x- or y-tilt. Since the exact point at
which the OPD is defined to be zero is arbitrary; referencing to the mean is
generally preferred to the chief ray reference for those cases where the centroid
is not preferred.
Physically significant merit functions

Note that the numerical value of the merit function is physically significant. If the merit function is RMS-
Wavefront-Centroid, then the numerical value of the merit function is the RMS wavefront error in waves. If the
Chapter 14: OPTIMIZATION 386

merit function is RMS-Spot Radius-Chief, then a value of 0.145 means the RMS spot radius is 0.145 lens units.
If the lens units were millimeters, this would correspond to 145 micrometers RMS.
Novice designers often ask why the RMS spot radius merit function should yield a different optimum design
than the RMS wavefront merit function. The basic reason they are different is that ray aberrations are proportional
to the derivative of the wave aberrations. Therefore, it is unreasonable to expect that the minimum of one
corresponds to the minimum of the other. A general rule of thumb to use is that if the system is close to diffraction
limited (say a PTV wavefront error of less than two waves) then use wavefront error. Otherwise, use spot radius.
Generally speaking, the merit functions which use the centroid as a reference are superior to those that use
the chief ray. Most diffraction based performance measures, such as MTF or encircled energy, improve when the
RMS wavefront error referenced to the centroid decreases. However, it is always best to reoptimize a final design
with the various merit functions to verify which one provides the best performance for the system being designed.
For example, the RMS wavefront centroid reference often yields better low frequency MTF response, but worse
medial frequency response, than the RMS chief ray reference optimization does.
Selecting the pupil integration method
There are two different pupil integration methods used to construct the merit function: Gaussian quadrature
(GQ) or rectangular array (RA). The GQ algorithm is VASTLY superior for almost all cases of practical interest.
The GQ algorithm uses a carefully selected and weighted ray set to accurately compute the RMS or PTV error
over the entrance pupil (strictly speaking, the PTV algorithm is not a GQ algorithm, but it is very similar). The
weighting for all rays is applied according to the weights set on the wavelength and field data dialog boxes, any
pupil apodization function, and by the GQ merit function algorithm. For RMS merit functions, the weighting and
ray set selection used is based on a method described in G. W. Forbes, "Optical system assessment for design:
numerical ray tracing in the Gaussian pupil", J. Opt. Soc. Am. A, Vol. 5, No. 11, p1943 (1988). For the PTV merit
functions, the ray set is based on solutions to the Chebyshev polynomials, described in Numerical Recipes,
Cambridge University Press (1989). If you are interested in detailed information on the basis and accuracy of
these methods, see these references. GQ is much, much more accurate than any other known method, and
requires fewer rays. Therefore, you get the best of both worlds: greater speed and greater accuracy. The GQ
algorithm requires specification of the number of "Rings" and the number of "Arms", and these terms are defined
in subsequent paragraphs. The only drawback to GQ is that it does not work if there are surface apertures in the
optical system. For these systems, GQ is probably no longer a better choice than RA. GQ does work fine when
used with vignetting factors, since the ray pattern is redistributed.
The RA algorithm traces a grid of rays through the pupil. The "Grid" size determines the number of rays traced,
and is described in a subsequent paragraph. The "Delete Vignetted" option (also described later) allows the
vignetted rays to be deleted from the ray set. Vignetted rays in this context are those rays clipped by surface
apertures, not rays which have been altered by the use of vignetting factors (see the chapter "Conventions and
Definitions"). The advantage to the RA algorithm is the ability to accurately account for the effects of vignetting in
the merit function. This is useful in systems such as obscured telescopes and camera lenses which intentionally
clip troublesome rays. The disadvantage to the RA algorithm is speed and accuracy. Usually, more rays are
required to achieve a given degree of accuracy than the GQ algorithm. The bottom line: don't use RA unless you
are using surface apertures.
Rings
The "Rings" setting is only used in the GQ algorithm. It determines how many rays are traced at each field and
at each wavelength. For on-axis fields (zero degrees field angle in a rotationally symmetric system), the number
of rays is equal to the number of rings. For all other fields in symmetric systems, the number of rays traced per
ring is equal to half the number of "arms" (defined in the next paragraph). Only half the rays are traced because
the left-right symmetry of the system is exploited. Each set of rays is traced for each defined wavelength. For
example, if you have one on-axis field, two off-axis fields, three wavelengths, and four rings selected, the number
of rays traced is 3 * (4 + 4*3 + 4*3) = 84. For systems without rotational symmetry, the number of rays per ring is
the number of "arms" independent of field. In the prior example, this means 3 * 3 * 4 * 6 = 216 rays. ZEMAX
automatically calculates these numbers for you; the only reason it is described here is so you will understand how
the default merit function is defined. Optimization runs are longer if more rays are traced.

Arms
The "Arms" setting is also only used in the GQ algorithm. It determines how many radial arms of rays in the
pupil are traced. By default six equally spaced (in angle) arms are traced (or three if the system is rotationally
symmetric). This number may be changed to eight, ten, or twelve. For most common optical systems, six is
sufficient.
You should select the number of rings and the number of arms according to the order of aberrations present
in your system. A simple way of determining the correct number of rings is to select the minimum number, one.
Then go to the optimization dialog box and note the merit function. Now go back to the default merit function tool,
and select two rings. If the merit function changes by more than a few percent, go back and select three, and so
on until the merit function does not change significantly (perhaps 1%). Repeat the procedure for the number of
arms (six arms is almost always plenty). Selecting more rings or arms than required will not improve the
optimization performance, it will only slow the algorithm down needlessly. Tracing more rays than required will
not help you find better solutions!
!Selecting more rings or arms than required will not improve the optimization performance, it
will only slow the algorithm down needlessly.
Grid
The "Grid" is only used by the RA algorithm, and the value determines the number of rays to be used. The grid
size can be 4x4 (16 rays per field per wavelength), 6x6 (36 rays per field per wavelength) etc. Rays on the grid
are automatically omitted if they fall outside the entrance pupil, so the actual number of rays used will be lower
than the grid size squared. Selecting a larger grid size generally yields more accurate results at the expense of
slower execution. However, there may be an advantage in choosing a large grid density, and then selecting the
"Delete Vignetted" checkbox (described in the next paragraph). The reason is that a large grid density will fill the
pupil with rays, and then the operands which are vignetted will be deleted. The result is a reasonable number of
rays which accurately reflect the aperture of the system.
Delete Vignetted
The "Delete Vignetted" checkbox option is only used by the RA algorithm. If selected, then each ray in the
merit function will be traced through the system, and if it is vignetted by a surface aperture, if it misses any surface,
or if it is total internal reflected at any surface, the ray is deleted from the merit function. This keeps the total
number of rays in the merit function to a minimum. The disadvantage is that if the vignetting changes as the design
is optimized, then the merit function may have to be regenerated. It is always a better choice to use the vignetting
factors and then use the GQ algorithm than to delete vignetted rays if possible. Vignetting factors can be adjusted,
if required, during optimization using SVIG in the merit function.
Note that ZEMAX will attempt to trace any ray defined in the merit function regardless if that ray is vignetted
or not. For example, if the chief ray height is targeted using REAY, and there is a central obscuration that vignettes
the chief ray, ZEMAX will still trace the ray and use the operand results as long as the ray can be traced. ZEMAX
does not check to see if defined rays are vignetted, because this introduces substantial overhead during
optimization.
In general, avoid vignetting of rays by surface apertures, and use vignetting factors to shape the beam size
when possible. To optimize on the fraction of unvignetted rays, a macro must be defined to perform the required
computations. However, this method is very prone to stagnation during optimization because small changes in
lens parameters lead to discrete changes in the merit function as rays abruptly jump from being vignetted to not
being vignetted.
Setting thickness boundary values
Boundary constraints may be automatically generated and included in the default merit function by checking
the air and/or glass boundary values on. If selected, then MNCG, MXCG, and MNEG operands will be added to
the merit function to constrain the minimum center thickness, maximum center thickness, and minimum edge
thickness for glass surfaces, respectively. MNCA, MXCA, and MNEA operands will be added to the merit function
to constrain the minimum center thickness, maximum center thickness, and minimum edge thickness for air
spaces, respectively.

The automatic boundary constraint feature is meant to save some manual entry of routine boundary
constraints on optical systems with or without mirrors. More complex lenses, such as those with complex
coordinate breaks, or multi-configurations usually require additional boundary constraints to be added to the merit
function manually.
Start At
The "Start At" option is used to add the default merit function at a specific position within the Merit Function
Editor operand list. ZEMAX will attempt to place the default merit function after the existing targets. The algorithm
used to determine the starting pointing may be ineffective if the default merit function has been edited. To control
where the default start at value will be, see the DFMS operand definition.
Assume Axial Symmetry
If selected, then the default merit function will exploit the left-right and rotational symmetry of the lens when
constructing and evaluating the merit function. Fewer rays will be traced, accelerating the optimization with no
loss of accuracy. In systems with coordinate breaks or non-rotationally symmetric systems, the default is
unselected, which means symmetry will not be exploited. Overriding the default system symmetry is useful,
however, if you are designing a lens that ZEMAX thinks is non-symmetric, but the lack of symmetry does not affect
the aberrations. For example, if tilted but flat fold mirrors are present, these mirrors do not eliminate the left-right
symmetry of the system, but ZEMAX will by default assume symmetry does not exist. Some gradient index
surfaces also use non-symmetric index variation terms which are often zero (they are used only for tolerancing).
Select the checkbox to accelerate the optimization in these cases. See also the "USYM" operand description.
If the merit function will be used for tolerancing, check this option off, because even symmetric lenses generally
become non-symmetric during tolerancing.
Ignore Lateral Color
By default, ZEMAX references all RMS or PTV computations to a common reference point for each field. All
of the rays are traced for all wavelengths for each field point, and the primary wavelength chief ray or the centroid
of all the rays is used as the reference point. If "Ignore Lateral color" is selected, then ZEMAX computes an
independent reference point for each wavelength as well. This is useful for designing systems that intentionally
divide the beam by wavelength, such as a prism or spectrometer system. This option will cause the merit function
to optimize each color spot independently.
Relative X Weight
The relative X weight is the additional weighting to be placed on the X component of the transverse aberrations
when computing the PTV or RMS SPOT X + Y merit function. This setting has no affect on the other merit
functions. If the relative X weight is less than unity, then the Y components are weighted more heavily; if the
relative weight is greater than unity, then the X components are weighted more heavily. If left at the default value
of unity; then the components are equally weighted. This control is useful for systems which intentionally form slit
images, such as spectrometers.
Overall Weight
When switching between different default merit function types, such as RMS spot radius and RMS wavefront
error, the numerical magnitude of the default merit function can change dramatically. This may make it tedious to
manually adjust the weights of operands that are not part of the default merit function. The overall weight is a
factor that scales all the weights in the default merit function. Under most circumstances, this weight may be left
at one.
Pitfalls with the default merit function
The default merit function is easy to set up, numerically efficient, and suitable for a large number of optimization
problems. However, most optical designs require extensions or modifications to the default as the design
progresses. ZEMAX offers significant flexibility in the definition of the merit function, as described in the following
sections.
Note that if the field or wavelength values or weights are changed, you must reconstruct the default merit
function. If you are using the RA algorithm, reconstruct the default merit function if the vignetting influence
changes appreciably during optimization.

!If field or wavelength values or weights are changed, reconstruct the merit function.
Optimization with apodized beams
If no pupil apodization has been specified (see the "System" chapter for details on specifying the pupil
apodization) then ZEMAX assumes uniform illumination when constructing the default merit function. If the
illumination is not uniform, then the rays in the default merit function are weighted according to the apodization
factor. Since the rays selected may be insufficient to adequately represent an apodized beam, use a larger
number of rays (described previously) when using apodization factors. See the chapter "Conventions and
Definitions" for more information about apodization.
Modifying the merit function
The merit function can be modified by the user. To change the merit function, select Editors, Merit Function
from the main menu bar. New operands can be added to the list, or others deleted, using the insert and delete
keys. The current merit function value and the value of each operand can be updated by selecting Tools, Update.
Operands are set by typing the name in the first column and then filling in the remaining data fields. There are
eight fields that may be required to define an operand: Int1, Int2, Hx, Hy, Px, Py, target, and weight. The Int values
are integer parameters whose meaning depends upon the operand selected. Usually, Int1 is the surface indicator,
and Int2 is the wavelength indicator, but not always. Not all of the operands use all of the fields provided.
Int1 and Int2
For those operands that use Int1 to indicate the surface number, the parameter specifies at which surface the
target should be evaluated. Similarly, the Int2 value, when used as a wavelength specifier, describes which
wavelength to use. Int2 must be an integer value equal to the wavelength number. The Int1 and Int2 parameters
have other uses as described later.
Hx, Hy, Px, Py
Many of the operands use Hx, Hy, Px, and Py; these are the normalized field and pupil coordinates (see
"Normalized field and pupil coordinates" in the chapter "Conventions and definitions"). Note that ZEMAX does
not check to see if the specified Hx, Hy, Px, and Py coordinates are within the unit circle. For example a pupil
coordinate of (1, 1) is actually outside the entrance pupil, but you will not get an error message when tracing these
rays unless the rays cannot physically be traced. Some operands use the Hx, Hy, Px, and Py data fields for other
values, depending upon the operand type.
The target is the desired value of the specified parameter. The difference between the target and the value of
the operand is squared, and summed over all operands to yield the merit function. The value of the target and
the operand itself is unimportant in optimization, only the difference between the two. The larger the difference,
the greater the contribution to the merit function.
The weight is the relative importance of that parameter. The weight can be any number, positive or negative.
However, the optimizer will act somewhat differently if the weight is negative, zero, or positive.
Operand weights less than zero
When the weight is negative, the operand will be treated as a Lagrangian multiplier. The Lagrangian multipliers
force the optimization algorithm to find a solution which exactly meets the specified constraint, regardless of the
effect on the other operands. This is sometimes useful to exactly meet an optimization target, such as focal length
or magnification. In some respects, this is similar to a weight of "infinity", however it is implemented in a way that
is numerically more stable.
Because there is generally a non-linear relationship between the variables and the operand targets, ZEMAX
may not converge to the exact target value in a single optimization cycle; however, multiple cycles will usually
converge to the Lagrangian targets with extremely high precision in a few cycles if a solution exists. It is possible
to define Lagrangian targets that cannot be met with the variables provided, especially if there is more than one
Lagrangian target defined. For purposes of computing the overall merit function value, ZEMAX will use the
absolute value of the weight.
For best results, use Lagrangian multipliers sparingly, if at all. Better optimization and adequate accuracy is
usually just as easily achieved using heavier weights on those operands which require exact (or nearly so) values.

If a Lagrangian multiplier operand does not exactly converge with a modest weight, such as -1, try a larger
magnitude negative weight, such as -1000. This will allow the merit function to decrease while meeting the target
exactly. Lagrangian multipliers should not be used with boundary operands, such CTGT, because appropriate
derivatives cannot always be computed for these types of operands.
Operand weights equal to zero
When the weight is zero, the optimization algorithm calculates but ignores the operand. This is very useful for
computing a result that does not have a specific target, but might be used elsewhere in the merit function; or if
the value is used as a check or monitored parameter.
Operand weights greater than zero
If the weight is greater than zero, then the operand will be treated as an "aberration" to be minimized along
with the merit function. The vast majority of operands should have positive weights.
Merit function definition
The merit function is defined as:

2
2
Wi ( Vi Ti )
MF = ------------------------------------- ,
Wi
where W is the absolute value of the weight of the operand, V is the current value, T is the target value, and
the subscript i indicates the operand number (row number in the spreadsheet). The sum index "i" is normally over
all operands in the merit function, however the merit function listing feature (see Merit Function Listing on
page 195) sums the user defined and default operands separately.
Optimization operands
The following tables describe the available operands. The first table is a "quick reference" guide which
categorizes the operands by general subject. The second table provides a detailed description of each operand
(listed alphabetically) and states which operands use which data fields. Note in particular that some of the
operands (such as SUMM) use the Int1 and Int2 values to indicate something other than surface and wavelength.
If an operand does not use a data field, a "-" is shown.
OPTIMIZATION OPERANDS BY CATEGORY
Category Related Operands
First-order optical properties AMAG, ENPP, EFFL, EFLX, EFLY, EPDI, EXPP, ISFN, LINV, OBSN,
PIMH, PMAG, POWR, SFNO, TFNO, WFNO
Aberrations ANAR, ASTI, AXCL, BIOC, BIOD, BSER, COMA, DIMX, DISC, DISG,
DIST, FCGS, FCGT, FCUR, LACL, LONA, OPDC, OPDM, OPDX,
OSCD, PETC, PETZ, RSCE, RSCH, RSRE, RSRH, RWCE, RWCH,
RWRE, RWRH, SPCH, SPHA, TRAC, TRAD, TRAE, TRAI, TRAR,
TRAX, TRAY, TRCX, TRCY, ZERN
MTF data GMTA, GMTS, GMTT
MSWA, MSWS, MSWT
MTFA, MTFS, MTFT
Encircled energy DENC, DENF, GENC, GENF, XENC, XENF

Constraints on lens data COGT, COLT, COVA

CTGT, CTLT, CTVA
CVGT, CVLT, CVVA
DMGT, DMLT, DMVA
ETGT, ETLT, ETVA
MNCA, MNCG, MNCT, MNCV
MNEA, MNEG, MNET
MNPD
MXCA, MXCG, MXCT, MXCV
MXEA, MXEG, MXET
MNSD, MXSD
TTGT, TTHI, TTLT, TTVA
XNEA, XNET, XNEG, XXEA, XXEG, XXET, ZTHI
Constraints on lens properties CVOL, MNDT, MXDT, SAGX, SAGY, SSAG, TMAS, TOTR, VOLU,
NORX, NORY, NORZ, NORD
Constraints on parameter data PMGT, PMLT, PMVA.
Constraints on extra data XDGT, XDLT, XDVA
Constraints on glass data GCOS, GTCE, INDX, MNAB, MNIN, MNPD, MXAB, MXIN, MXPD,
RGLA
Constraints on paraxial ray data PANA, PANB, PANC
PARA, PARB, PARC
PARR, PARX, PARY, PARZ
PATX, PATY
YNIP
Constraints on real ray data CENX, CENY, DXDX, DXDY, DYDX, DYDY, HHCN, IMAE, OPTH,
PLEN, RAED, RAEN, RAGA, RAGB, RAGC, RAGX, RAGY, RAGZ,
RAID, RAIN, RANG, REAA, REAB, REAC, REAR, REAX, REAY, REAZ,
RENA, RENB, RENC, RETX, RETY
Constraints on element positions GLCA, GLCB, GLCC, GLCR, GLCX, GLCY, GLCZ
Changing system data CONF, IMSF, PRIM, SVIG, WLEN
General math operands ABSO, ACOS, ASIN, ATAN, CONS, COSI, DIFF, DIVB, DIVI, EQUA,
LOGE, LOGT, MAXX, MINN, OPGT, OPLT, OPVA, OSUM, PROB,
PROD, QSUM, SQRT, SUMM, SINE, TANG
Multi-configuration (zoom) data CONF, MCOL, MCOG, MCOV, ZTHI
Gaussian beam data GBPD, GBPP, GBPR, GBPS, GBPW
GBSD, GBSP, GBSR, GBSS, GBSW
Gradient index control operands DLTN, GRMN, GRMX
InGT, InLT, InVA
LPTD
Foucalt analysis FOUC
Ghost focus control GPIM
Fiber coupling operands FICL, FICP, POPD
Relative illumination operand RELI, EFNO

Optimization with ZPL macros ZPLM

User defined operands UDOP
Merit function control operands BLNK, DMFS, ENDX, GOTO, OFF, SKIN, SKIS, USYM
Constraints on non-sequential NPGT, NPLT, NPVA
object data. NPXG, NPXL, NPXV
NPYG, NPYL, NPYV
NPZG, NPZL, NPZV
NTXG, NTXL, NTXV
NTYG, NTYL, NTYV
NTZG, NTZL, NTZV
Non-sequential ray tracing and NSDC, NSDD, NSRA, NSTR
detector operands.
Constraints on construction optics CMFV
for optically fabricated holograms
Constraints on optical coatings, CMGT, CMLT, CMVA, CODA
polarization ray trace data
Physical Optics Propagation POPD
(POP) results
Tolerance sensitivity data TOLR
Thermal Coefficient of Expansion TCGT, TCLT, TCVA
data
OPTIMIZATION OPERANDS AND DATA FIELD USAGE
NAME Description Int1 Int2 Hxy, Pxy

ABSO Absolute value. Op # - -
ACOS Arccosine of the value of the specified operand number. If flag is Op # Flag -
0, then the units are radians, otherwise, degrees.
AMAG Angular magnification. This is the ratio of the image to object - Wave -
space paraxial chief ray angles. Not valid for non-paraxial
systems.
ANAR Angular aberration radius measured in the image plane with - Wave Yes
respect to the primary wavelength chief ray. This quantity is
defined as 1 cos , where is the angle between the chief ray
and the ray being traced. See TRAR.
ASIN Arcsine of the value of the specified operand number. If flag is 0, Op # Flag -
then the units are radians, otherwise, degrees.
ASTI Astigmatism in waves contributed by the specified surface. If the Surf Wave -
surface value is zero, the sum for the entire system is used. This
is the third order astigmatism calculated from the Seidel
coefficients, and is not valid for non-paraxial systems.
ATAN Arctangent of the value of the specified operand number. If flag Op # Flag -
is 0, then the units are radians, otherwise, degrees.

AXCL Axial color in lens units. This is the focal plane separation Wave1 Wave2 Pupil
between the two most extreme wavelengths defined. The Zone
distance is measured along the Z axis. If the pupil zone is zero,
paraxial rays are used to determine the paraxial focal plane
locations. If the pupil zone is greater than 0.0 and less than or
equal to 1.0, real marginal rays are used to determine the focal
plane locations.
Not valid for non-paraxial systems.
BIOC Biocular Convergence. Returns the convergence between two Left Eye Right (see left)
eye configurations in milliradians. The left and right eye Config Eye
configurations are selected using the Int1 and Int2 values. The Config
other parameters are:
Wave: The wavelength number to use.
Usea: Use angles, if 0 field units are degrees otherwise radians.
Xang/Yang: The X and Y angle or cosines at which to compute
the convergence.
If the chief rays from both configurations at the specified angles
do not pass through to the image without vignetting, an error is
reported. See Dipvergence/Convergence on page 154 for more
information and important assumptions.
BIOD Biocular Dipvergence. Returns the dipvergence between two eye Left Eye Right (see left)
configurations in milliradians. See BIOC above for details. Config Eye
Config
BLNK Does nothing. Used for separating portions of the operand list. A - - -
comment line may optionally be typed in the space to the right of
the operand name; this comment will be displayed in the editor
as well as in the merit function listing.
BSER Boresight error. Boresight error is defined as the radial chief ray - Wave -
coordinate traced for the on axis field divided by the effective focal
length. This definition yields a measure of the angular deviation
of the image.
CENX Centroid X position. This operand uses a grid of rays to determine Surf Wave (see left)
the x coordinate of the centroid of all rays from a single field point.
The centroid accounts for apodization and apertures, and
optionally polarization.
Surf is the surface number, use 0 for the image surface. Wave is
0 for polychromatic, otherwise use the monochromatic
wavelength number. Hx is used to set the integer field number,
Hy is 0 to ignore polarization and 1 to consider it. Px is the grid
size. A value of 10 would yield a 10 x 10 grid of rays. When CENX
is followed by a CENY with identical settings, both are computed
with the same ray set to save time.
CENY Centroid Y position. See CENX. Surf Wave (see left)

CMFV Construction merit function value. This operand calls the merit Con# Opr# -
functions defined in either of the two construction systems used
to define an optically fabricated hologram. The Con# is either 1
or 2, for the first or second construction system, respectively. The
Opr# is either 0, which will return the entire merit function value
from the construction system; or it is an integer which defines the
operand row # from which to extract the value from. For example,
if Con# is 2 and Opr# is 7, CMFV will return the value of merit
function operand 7 in construction file 2.
If there are more than one optically fabricated hologram surfaces
in the playback system being optimized, the Con# may be
incremented by 2 to specify the second surface parameters be
used, or by 4 to indicate the third hologram surface construction
optics be used, and so on. For example, a Con# of 7 would
indicate construction system 1 on the fourth optically fabricated
hologram surface present.
CMGT Boundary operand that constrains the coating multiplier of a Surf Layr -
surface or all surfaces to be greater than the target value. Use
Surf = 0 for all surfaces, Layr = 0 for all layers.
CMLT Boundary operand that constrains the coating multiplier of a Surf Layr -
surface or all surfaces to be less than the target value. Use Surf
= 0 for all surfaces, Layr = 0 for all layers.
CMVA Boundary operand that constrains the coating multiplier of a Surf Layr -
surface to be equal to the specified target value.
CODA Coating Data. This feature traces a polarized ray using the Surf Wave (see left)
system global polarization state (see Polarization on page 90).
Even if the global polarization state is set to "unpolarized", the
defined polarization state is used because the data computed by
CODA is polarization state specific.
The polarized ray may be traced from any field point in object
space to any point in the pupil, to the specified surface. If the
surface is zero, the ray is traced to the image surface. Use Hx to
select the field#, Hy to select the data#, and Px and Py to define
the pupil coordinate. The absolute value of the data number
determines what data is returned as follows:
0: The relative transmitted polarized intensity (see 8)
1, 2, 3: The intensity reflectance R, transmittance T, absorption A
4,5: The field amplitude transmittance real, imaginary
6,7: The field amplitude reflectance real, imaginary
8: The relative transmitted unpolarized intensity (see 0)
101, 102: E field out X real, imaginary
103, 104: E field out Y real, imaginary
105, 106: E field out Z real, imaginary
110: The phase difference between Ex and Ey; Pxy
111, 112, 113: The E field phase Px, Py, Pz
If the data is related to the coating (1-7), and the data number is
negative, the data is for the "S" polarization, otherwise the data is
for the "P" polarization.
Other coating and polarization data may be added to this list upon
request.
COGT Boundary operand that constrains the conic of surface "Surf" to Surf - -
be greater than the specified target value.
COLT Boundary operand that constrains the conic of surface "Surf" to Surf - -
be less than the specified target value.

COMA Coma in waves contributed by the specified surface. If the Surf Wave -
surface value is zero, the sum for the entire system is used. This
is the third order coma calculated from the Seidel coefficients,
and is not valid for non-paraxial systems.
CONF Configuration. This operand is used to change the configuration New # - -
number during merit function evaluation, which permits
optimization across multiple configurations. This operand does
not use the target or weight columns.
CONS Constant value. This is used to enter in constant values for use - - -
in other operand computations. The value will be identical to the
target value.
COSI Cosine of the value of the specified operand number. If flag is 0, Op # Flag -
COVA Conic value. Returns the conic constant of a surface. Surf - -
CTGT Center thickness greater than. This boundary operand constrains Surf - -
the center thickness of surface "Surf" to be greater than the
specified target value. See also MNCT.
CTLT Center thickness less than. This boundary operand constrains Surf - -
the center thickness of surface "Surf" to be less than the specified
target value. See also MXCT.
CTVA Center thickness value. Constrains the center thickness of Surf - -
surface "Surf" to be equal to the specified target value.
CVGT Curvature greater than. This boundary operand constrains the Surf - -
curvature of surface "Surf" to be greater than the target value.
CVLT Curvature less than. This boundary operand constrains the Surf - -
curvature of surface "Surf" to be less than the target value.
CVOL Cylinder volume. This operand computes the volume in cubic First surf Last surf -
lens units of the smallest cylinder that will contain the specified
range of surfaces. Only the vertex positions and semi-diameters
are used in the calculation, not the sag. The range of surfaces
should not include any coordinate breaks.
CVVA Curvature value. This operand constrains the curvature of Surf - -
surface "Surf" to be equal to the specified target value.
DENC Diffraction Encircled Energy (distance). This operand computes Sam- Wave (see left)
the distance in micrometers to the specified fraction of diffraction pling
encircled, ensquared, x only, or y only (enslitted) energy.
Int1 specifies the pupil sampling, where 1 yields 32 x 32, 2 yields
64 x 64 etc.
Int2 is the integer wavelength number; 0 for polychromatic.
Hx specifies the field number.
Hy is the fraction of energy desired, and must be between zero
and 1.
Px is the type: 1 for encircled, 2 for x only, 3 for y only, and 4 for
ensquared.
Py is the reference point: 0 for chief ray, 1 for centroid, 2 for
vertex.
If the sampling is too low, the radius returned is a 1e+10. See also
DENF, GENC and XENC.

DENF Diffraction Encircled Energy (fraction). This operand computes Sam- Wave (see left)
the fraction of diffraction encircled, ensquared, x only, or y only pling
(enslitted) energy at a given distance from the reference point.
The options and settings are identical to DENC, except Hy, which
here is used as the distance at which the fraction of energy is
desired. See also DENC, GENC, GENF, and XENC.
DIFF Difference of two operands (OP#1 - OP#2). The two arguments Op #1 Op #2 -
are the row numbers of the operands to subtract.
DIMX Distortion maximum. This is similar to DIST, except it specifies Field Wave (see left)
only an upper bound for the absolute value of the distortion. The
Field integer can be zero, which specifies the maximum field
coordinate be used, or any valid field number. Note the maximum
distortion does not always occur at the maximum field coordinate.
If the absolute flag is 0, the value returned is in units of
percentage. If the absolute flag is 1, the distortion is given as an
absolute length rather than a percentage.
This operand may not be valid for non-rotationally symmetric
systems.
DISC Distortion, calibrated. This operand computes the calibrated - Wave (see left)
distortion across the field of view, and returns the absolute value
of the maximum deviation from linearity of the f-theta condition.
If the absolute flag is 0, the value returned is in units of
percentage. If the absolute flag is 1, the distortion is given as an
absolute length rather than a percentage.
This operand is extremely useful for designing f-theta lenses.
DISG Generalized distortion, either in percent or as an absolute Ref Fld Wave Yes
distance. This operand computes the distortion for any ray in the
pupil, from anywhere in the field, at any wavelength, using any
field point as a reference. The method used and assumptions
made are the same as for the grid distortion plot, see Grid
Distortion on page 158. DISG cannot be calculated if the field
units are angles and the maximum angle equals or exceeds 90
degrees. DISG assumes the predicted magnification is not
symmetric.
If the field is defined in terms of angles, the normalized field
coordinates Hx and Hy are defined differently than for other
ZEMAX features. In this case only, Hx and Hy correspond to
divisions on a uniform projected grid in object space, centered on
the reference field coordinate. These normalized field
coordinates are defined as
tan ( )
H = --------------------
tan ( M )
where is the angle relative to the reference field chief ray and
M is the maximum field angle (see Maximum field on
page 48). These definitions are needed to support the reference
field feature.
If the wavelength is a positive number, DISG returns the distortion
as a percentage. If the wavelength is a negative number, the
absolute value of the wavelength number is used to define the
wavelength and the returned value is in units of absolute length
rather than percentage.
As with all distortion concepts, the best way to avoid confusion
and misleading results is to use finite object distances and object
heights to define fields rather than field angles.

DIST Distortion in waves contributed by surface "Surf". This is the third Surf Wave (see left)
order distortion calculated from the Seidel coefficients (see
Seidel Coefficients on page 164), and is not valid for non-
paraxial systems.
If the surface value is zero, the distortion is given in percent
instead (see Field Curvature/Distortion on page 156 for a
detailed definition).
If the absolute flag is set to 1, and the surface number is zero, the
distortion is given as an absolute length rather than a percentage.
See also DISG.
DIVB Divides the value of any prior operand by any factor. Oper# - factor
DIVI Division of first by second operand. The two arguments are the Op #1 Op #2 -
row numbers of the operands to divide.
DLTN Delta N. Computes the difference between the maximum and Surf Wave -
minimum index of refraction on axis for a gradient index surface.
The minimum and maximum z coordinates used are computed
by accounting for the sag of both ends of the surface. See the
section "Using gradient index operands".
DMFS Default merit function start. This operand is a flag to indicate - - -
where the default merit function should be appended, if one is
subsequently created. The row number after this operand will
appear as the default "Start At" row on the default merit function
dialog box.
DMGT Diameter greater than. This boundary operand constrains the Surf - -
diameter of surface "Surf" to be greater than the specified target
value. The diameter is twice the semi-diameter value displayed
on the main spreadsheet.
DMLT Diameter less than. This boundary operand constrains the Surf - -
diameter of surface "Surf" to be less than the specified target
value. The diameter is twice the semi-diameter value displayed
on the main spreadsheet.
DMVA Diameter value. This operand constrains the diameter of surface Surf - -
"Surf" to be equal to the specified target value. The diameter is
twice the semi-diameter value displayed on the main
spreadsheet.
DXDX Derivative of transverse x-aberration with respect to x-pupil - Wave Yes
coordinate. This is the slope of the ray fan plot at the specified
pupil coordinate.
DXDY Derivative of transverse x-aberration with respect to y-pupil - Wave Yes
pupil coordinate.
DYDX Derivative of transverse y-aberration with respect to x-pupil - Wave Yes
pupil coordinate.
DYDY Derivative of transverse y-aberration with respect to y-pupil - Wave Yes
pupil coordinate.
EFFL Effective focal length in lens units. This is paraxial, and may not - Wave -
be accurate for non-paraxial systems.

EFLX Effective focal length in the local x plane of the specified range of First surf Last surf -
surfaces at the primary wavelength.
EFLY Effective focal length in the local y plane of the specified range of First surf Last surf -
surfaces at the primary wavelength.
EFNO Effective F/#. This operand computes the Effective F/# of any field Samp Wave See Text
point. For a discussion of Effective F/# see Effective F/# on
page 140.
Samp is the grid size, where 10 would yield a 10 x 10 grid of rays.
Wave is the integer wavelength number, Hx is used to define the
field number, and Hy specifies if polarization is used (0 for no, 1
for yes). See also RELI.
ENDX End execution. Terminates the computation of the merit function. - - -
All remaining operands are ignored.
ENPP Entrance pupil position in lens units, with respect to the first - - -
surface. This is the paraxial pupil position, valid only for centered
systems.
EPDI Entrance pupil diameter in lens units. - - -
EQUA Equal operand. This operand constrains all operands within the First Last -
specified range of operands to have the same value within the
tolerance specified by the target. The value of this operand is
computed by finding the average of the range of values, and then
summing the absolute value of the errors between each operand
and the average if the error exceeds the target value. See SUMM
and OSUM.
ETGT Edge thickness greater than. This boundary operand constrains Surf Code -
the edge thickness of surface "Surf" to be greater than the
specified target value. The edge thickness is calculated at the
semi-diameter radius along the +y axis if code is zero, the +x axis
if code is 1, the -y axis if code is 2, and the -x axis if code is 3.
See also MNET.
ETLT Edge thickness less than. This boundary operand constrains the Surf Code -
edge thickness of surface "Surf" to be less than the specified
target value. The edge thickness is always calculated at the semi-
diameter radius along the +y axis if code is zero, the +x axis if
code is 1, the -y axis if code is 2, and the -x axis if code is 3. See
also MXET.
ETVA Edge thickness value. Constrains the edge thickness of surface Surf Code -
"Surf" to be equal to the specified target value. The edge
thickness is always calculated at the semi-diameter radius along
the +y axis if code is zero, the +x axis if code is 1, the -y axis if
code is 2, and the -x axis if code is 3. See also MNET.
EXPP Exit pupil position in lens units, with respect to the image surface. - - -
This is the paraxial pupil position, valid only for centered systems.
FCGS Generalized field curvature, sagittal. The field curvature value for - Wave Hx,Hy
any field point, at any wavelength. The value is generalized to
return reasonable results even for non-rotationally symmetric
systems; see Field Curvature/Distortion on page 156.
FCGT Generalized field curvature, tangential; see FCGS. - Wave Hx,Hy

FCUR Field curvature in waves contributed by the specified surface. If Surf Wave -
the surface value is zero, the sum for the entire system is used.
This is the third order field curvature calculated from the Seidel
coefficients, and is not valid for non-paraxial systems.
FICL Fiber coupling efficiency for single mode fibers. The sampling Sam- Wave (see left)
defines the grid size used for the integration; with 1 being 32 x 32, pling
2 being 64 x 64, etc. The wavelength must be monochromatic,
and the wavelength number specified in the Int2 column. The Hx
value is the integer field position number. If Hy is zero, then the
object source fiber is considered; if Hy is non zero, the object
source fiber is ignored. Px and Py are used to define the source
and receiver fiber NA's respectively. The calculated value is the
total coupled energy efficiency, relative to unity. See Fiber
Coupling Efficiency on page 174 for details. See also FICP.
FICP Fiber coupling as computed using the Physical Optics Surf - (see left)
Propagation (POP) algorithm, using whatever the current default
settings are for the POP feature.
To use this operand, first define the settings on the POP analysis
feature as desired, then press Save on the settings box. The
operand FICP will return the same efficiency as computed by the
POP feature.
This operand is redundant with the more general POPD.
See Computing Fiber Coupling on page 530. See also FICL.
FOUC Foucault analysis. This operand returns the RMS difference - - -
between the computed and reference shadowgram as computed
by the Foucault analysis feature, using whatever the current
default settings are. To use this operand, first define the settings
on the Foucault analysis feature as desired, then press Save on
the settings box. The data option "difference" MUST be selected
to return valid data. The operand FOUC will return the RMS
difference between the computed and reference shadowgrams.
Using this operand, the optical system wavefront aberrations may
be optimized to produce the reference shadowgram.
GBPD Gaussian beam (paraxial) divergence in the optical space Surf Wave (see left)
following the specified surface. If Hx is non-zero, then the
computation is for the x-direction beam, otherwise, it is for the y-
direction. The Hy value is used to define the input beam waist,
and Px is used to define the distance from Surface 1 to the waist
location. See the Gaussian beam feature for details.
GBPP Gaussian beam (paraxial) position, which is the distance from the Surf Wave (see left)
waist to the surface, in the optical space following the specified
surface. See GBPD for usage of the Hx, Hy, Px, and Py values.
GBPR Gaussian beam (paraxial) radius of curvature in the optical space Surf Wave (see left)
following the specified surface. See GBPD for usage of the Hx,
Hy, Px, and Py values.
GBPS Gaussian beam (paraxial) size in the optical space following the Surf Wave (see left)
specified surface. See GBPD for usage of the Hx, Hy, Px, and Py
values.
GBPW Gaussian beam (paraxial) waist in the optical space following the Surf Wave (see left)
specified surface. See GBPD for usage of the Hx, Hy, Px, and Py
values.

GBSD Gaussian beam (skew) divergence in the optical space following In# Out# (see left)
the specified surface. The In# is the surface number to start the
propagation. The Out# is the surface at which to compute the
Skew Gaussian Beam data. Hx is the wavelength number, Hy is
the field number, Px is the starting surface to waist distance in
lens units, and Py is the waist size in lens units. If Py is positive,
then the computation is for the y-direction beam, otherwise, it is
for the x-direction.
The input beam is aligned along the chief ray of the specified field.
For details on the Skew Gaussian Beam feature, see Skew
Gaussian Beam on page 189.
GBSP Gaussian beam (skew) position, which is the distance from the Surf Wave (see left)
waist to the surface, in the optical space following the specified
surface. See GBSD for usage of the Hx, Hy, Px, and Py values.
GBSR Gaussian beam (skew) radius in the optical space following the Surf Wave (see left)
specified surface. See GBSD for usage of the Hx, Hy, Px, and Py
values.
GBSS Gaussian beam (skew) size in the optical space following the Surf Wave (see left)
values.
GBSW Gaussian beam (skew) waist in the optical space following the Surf Wave (see left)
values.
GCOS Glass cost. This operand returns the relative cost factor as listed Surf - -
in the glass catalog for the glass on the specified surface.
GENC Geometric Encircled Energy (distance). This operand computes Sam- Wave (see left)
the distance in micrometers to the specified fraction of geometric pling
encircled, ensquared, x only, or y only (enslitted) energy.
Int1 specifies the pupil sampling, where 1 yields 32 x 32, 2 yields
64 x 64 etc.
Int2 is the integer wavelength number; 0 for polychromatic.
Hx specifies the field number.
Hy is the fraction of energy desired, and must be between zero
and 1, exclusive.
Px is the type: 1 for encircled, 2 for x only, 3 for y only, and 4 for
ensquared.
Py is the reference point: 0 for chief ray, 1 for centroid, 2 for
vertex, 3 for middle of the spot.
GENC always scales by the diffraction limited data. See also
GENF, DENC, DENF, and XENC.
GENF Geometric Encircled Energy (fraction). This operand computes Sam- Wave (see left)
the fraction of geometric encircled, ensquared, x only, or y only pling
(enslitted) energy at a given distance from the reference point.
The options and settings are identical to GENC, except Hy, which
desired. See also GENC, DENC, DENF, and XENC.
GLCA Global x-direction orientation vector component of surface "Surf". Surf - -
GLCB Global y-direction orientation vector component of surface "Surf". Surf - -
GLCC Global z-direction orientation vector component of surface "Surf". Surf - -

GLCR Global Coordinate Rotation Matrix component at surface "Surf". Surf Data# -
The 3 x 3 R matrix has 9 components. If the Data# is 1, GLCR
returns R[1][1], Data# 2 returns R[1][2], etc... through Data# 9
returning R[3][3].
GLCX Global vertex x-coordinate of surface "Surf". Surf - -
GLCY Global vertex y-coordinate of surface "Surf". Surf - -
GLCZ Global vertex z-coordinate of surface "Surf". Surf - -
GMTA Geometric MTF average of sagittal and tangential response. The (see + +
Samp value must be an integer (1, 2, ...) where 1 yields 32 x 32 left).
sampling, 2 yields 64 x 64 sampling, etc. Wave can be a valid
wavelength number, or 0 for polychromatic. The Field value must
be a valid field number (1, 2, ...). Freq is the spatial frequency in
cycles per mm. The parameter !Scl is a flag; if zero, then the
diffraction limit will be used to scale the results (recommended)
otherwise, no scaling is done. See Using MTF operands on
page 421.
GMTS Geometric MTF sagittal response. See GMTA for details. (see + +
left).
GMTT Geometric MTF tangential response. See GMTA for details. (see + +
left).
GOTO Skips all operands between the GOTO operand line and the Op# - -
target operand number. Execution of the merit function will start
again at the target line.
GPIM Ghost pupil image. GPIM controls the location of ghost pupils First Second (see left)
(and optionally ghost images) relative to the image plane. surface surface
Double-bounce ghosts form images of the pupil, and if these
images are formed near the focal plane will contaminate the
image with unwanted light. This is the cause of the familiar sun
flare images of the pupil seen through camera lenses pointed
near the sun.
The operand computes any one specific or all possible ghost
pupil image locations and returns one over the absolute value of
the distance from the image plane to the closest pupil ghost. The
operand is defined in this manner so it can be targeted to zero
and weighted and optimized to reduce ghost pupil affects. If the
int1 and int2 parameters are set to specific surface numbers, that
specific ghost path is computed, if either or both of the int values
are -1, then all possible surface combinations are considered. For
example, if Int1 is 12 and int2 is -1, then all double bounces that
first bounce off surface 12 and then 11, 10, 9, etc. are considered,
if both numbers are negative, all possible ghosts are considered.
This same operand also can be used for detecting and controlling
image ghosts (which are distinct from pupil ghosts) by changing
the mode flag in the Hx column from 0 to 1, or to control ghost
pupil magnification, by setting the mode to 2.
The WFB and WSB columns will list the worst combination found
for reference and possible further analysis. Only surfaces with
index changes are considered as possible ghost generators. First
bounces off mirrors are ignored.

GRMN Gradient index minimum index. This boundary operand sets the Surf Wave -
minimum allowable index of refraction value for the gradient index
surface "Surf" at the wavelength number "Wave". The index is
checked at six places: the front vertex, front +y top, front +x side,
rear vertex, rear +y top, and the rear +x side. See also InGT, InLT
and GRMX.
GRMX Gradient index maximum index. This boundary operand sets the Surf Wave -
maximum allowable index of refraction value for the gradient
index surface "Surf" at the wavelength number "Wave". The index
is checked at six places: the front vertex, front +y top, front +x
side, rear vertex, rear +y top, and the rear +x side. See also InGT,
InLT and GRMN.
GTCE Glass TCE. This operand returns the Thermal Coefficient of Surf - -
Expansion Alpha1 as listed in the glass catalog for the glass on
the specified surface.
HHCN Test for the hyperhemisphere condition. ZEMAX traces the Surf Wave Yes
specified ray to the specified surface, and computes the x, y, and
z intercept coordinates. Then, the x and y coordinates only are
used in the sag expression for that surface to see what z
coordinate results. If the z coordinates are not the same, then
HHCN returns 1, otherwise it returns zero. This operand can be
used to prevent optimizations from reaching solutions that require
hyperhemispheric surface shapes.
IMAE Image analysis efficiency. This operand returns the fractional Surf Field (see left)
efficiency as computed by the geometric image analysis feature,
using whatever the current default settings are, except for "Show"
which is always set to spot diagram for this computation,
"Surface", which may be selected using the Int1 field, "Field",
which may be selected using the Int2 field, and "Field Size", which
may be selected using the hx field.
To use this operand, first define the settings on the geometric
image analysis feature as desired, then press Save on the
settings box. The operand IMAE will return the efficiency
(normalized to unity) as computed by the image analysis feature.
If Surf is 0, the surface specified by the saved settings will be
used. If Surf is greater than zero, then the efficiency will be
computed at the specified surface.
If Field is 0, the field number specified by the saved settings will
be used. If Field is greater than zero, then the efficiency will be
computed at the specified field.
If FldSz is 0, the Field Size specified by the saved settings will be
used. If FldSz is greater than zero, then the efficiency will be
computed using the specified Field Size.
See the discussion Optimizing with the IMAE operand on
page 428.

IMSF Image Surface. This operand dynamically changes which surface Surface Refocus -
is interpreted as the "image surface" for all subsequent operands.
The primary usage for this operand is to optimize image quality
at intermediate surfaces. Select surface 0 to restore the image
surface back to the original surface.
If the refocus flag is 0, the specified surface becomes the new
image surface and is not modified in any way. If refocus is set to
1, then a dummy plane surface will follow the specified surface,
and a paraxial focus solve will be used to place this dummy
surface at the focus of the specified surface. This option is most
useful for optimizing aberrations at a virtual focus for intermediate
surfaces that are not already at focus.
Note this operand only temporarily changes a copy of the lens
data use for evaluating the merit function and has no impact on
the original lens data. Care should be taken when the selected
surface precedes the original stop surface. For information on
how this operand is implemented see Evaluating results at
intermediate surfaces on page 109.
INDX Index of refraction. Returns the current index at any surface and Surf Wave
defined wavelength.
InGT Index "n" greater than. This boundary operand constrains the Surf Wave -
index of refraction at wavelength number "Wave" of gradient
index surface "Surf" at one of six points inside the gradient index
lens. For n=1, the point is the front vertex, n=2 is the front +y top,
n=3 is the front +x side, n=4 is the rear vertex, n=5 is the rear +y
top, and n=6 is the rear +x side. In all cases the operand bounds
the index at the specified point to be greater than the specified
target value. For example, "I4GT" constrains the minimum index
at the rear vertex of surface (i.e. the vertex of the next surface) of
the gradient index lens. In all cases the +y top and +x side
distance is defined by the larger of the front and rear semi-
diameters set on the main spreadsheet. See also GRMN and
GRMX, which are similar operands that are easier to use.
InLT Index "n" less than. This operand is similar to InGT except it Surf Wave -
constrains the maximum value of the index of refraction rather
than the minimum. See InGT for a complete description of the
parameter "n".
InVA This operand is similar to InGT except it constrains the current Surf Wave -
value of the index of refraction. See InGT for a complete
description of the parameter "n".
ISFN Image space F/#. This operand is the paraxial infinite conjugate - - -
F/#. See WFNO.
ISNA Image space numerical aperture. This operand is the paraxial - - -
image space na at the defined conjugates. See ISFN.
LACL Lateral color. This is the y-distance between the paraxial chief ray Wave1 Wave2 -
intercepts of the two extreme wavelengths defined. Not valid for
non-paraxial systems.
LINV Lagrange (or optical) invariant of system in lens units. The - Wave -
paraxial marginal and chief ray data are used to compute this
value.
LOGE Log base e of an operand. The argument is the row number of Op # - -
the operand value to take the log of. If the value is less than or
equal to zero, the value zero is returned.

LOGT Log base 10 of an operand. The argument is the row number of Op # - -
the operand value to take the log of. If the value is less than or
equal to zero, the value zero is returned.
LONA Longitudinal aberration in lens units. This is the distance from the Wave - Pupil
current to the focal plane defined by the wavelength and pupil Zone
zone. The distance is measured along the Z axis. If the pupil zone
is zero, paraxial rays are used to determine the paraxial focal
plane locations. If the pupil zone is greater than 0.0 and less than
or equal to 1.0, real marginal rays are used to determine the focal
plane locations. See AXCL.
LPTD This boundary operand constrains the slope of the axial gradient Surf - -
index profile from changing signs within a gradient index
component. See the section "Using gradient index operands".
MAXX Returns the largest value within the indicated range of operands. First Last -
See MINN.
MCOG Multi-configuration operand greater than. This is used to MC op- Config # -
constrain values in the multi-configuration editor. erand #
MCOL Multi-configuration operand less than. This is used to constrain MC op- Config # -
values in the multi-configuration editor. erand #
MCOV Multi-configuration operand value. This is used to directly target MC op- Config # -
or compute values in the multi-configuration editor. erand #
MINN Returns the smallest value within the indicated range of First Last -
operands. See MAXX.
MNAB Minimum Abbe number. This boundary operand constrains the First surf Last surf -
Abbe number of surfaces between "First Surf" and "Last Surf" to
be greater than the specified target value. See also MXAB. This
operand only considers surfaces using model or substitute status
catalog glasses.
MNCA Minimum center thickness air. This boundary operand constrains First surf Last surf -
each of the center thicknesses of surfaces from "First surf" to
"Last surf" which have air (i.e. no glass) as the glass type to be
greater than the specified target value. See also MNCT and
MNCG. This operand controls multiple surfaces simultaneously.
MNCG Minimum center thickness glass. This boundary operand First surf Last surf -
constrains each of the thicknesses of surfaces from "First surf" to
"Last surf" which have a non-air glass type to be greater than the
target value. See also MNCT and MNCA. This operand controls
multiple surfaces simultaneously.
MNCT Minimum center thickness. This boundary operand constrains First surf Last surf -
"Last surf" to be greater than the specified target value. See also
MNCG and MNCA. This operand controls multiple surfaces
simultaneously.
MNCV Minimum curvature. This boundary operand constrains each of First surf Last surf -
the curvatures of surfaces from "First surf" to "Last surf" to be
greater than the specified target value. See also MXCV. This
operand controls multiple surfaces simultaneously.

MNDT Minimum diameter to thickness ratio. Controls the minimum First surf Last surf -
allowable value on the ratio of surface diameter to center
thickness. Only surfaces with non-unity index of refraction are
considered. See also MXDT. This operand controls multiple
surfaces simultaneously.
MNEA Minimum edge thickness air. This boundary operand constrains First surf Last surf -
each of the edge thicknesses of surfaces from "First surf" to "Last
surf" which have air (i.e. no glass) as the glass type to be greater
than the specified target value. See also MNET, MNEG, ETGT,
and XNEA. This operand controls multiple surfaces
simultaneously. The boundary applies to the top "+y" edge of the
surface only; see XNEA for constraining non-rotationally
symmetric surfaces.
MNEG Minimum edge thickness glass. This boundary operand First surf Last surf -
constrains each of the edge thicknesses of surfaces from "First
surf" to "Last surf" which have a non-air glass type to be greater
than the specified target value. See also MNET, MNEA, ETGT,
and XNEG. This operand controls multiple surfaces
simultaneously. The boundary applies to the top "+y" edge of the
surface only; see XNEG for constraining non-rotationally
symmetric surfaces.
MNET Minimum edge thickness. This boundary operand constrains First surf Last surf -
surf" to be greater than the specified target value. See also
MNEG, MNEA, ETGT, and XNET. This operand controls multiple
surfaces simultaneously. The boundary applies to the top "+y"
edge of the surface only; see XNET for constraining non-
rotationally symmetric surfaces.
MNIN Minimum index at d-light. This boundary operand constrains the First surf Last surf -
Nd value of surfaces between "First Surf" and "Last Surf" to be
greater than the specified target value. See also MXIN. This
catalog glasses.
MNPD Minimum P g, F . This boundary operand constrains the First surf Last surf -
deviation of the partial dispersion of surfaces between "First Surf"
and "Last Surf" to be greater than the specified target value. See
also MXPD. This operand only considers surfaces using model
or substitute status catalog glasses.
MNSD Minimum semi-diameter. Constrains the semi-diameter to be First surf Last surf -
larger than the specified target over the surface range.
MSWA Modulation square-wave transfer function, average of sagittal Sam- Wave (see left)
and tangential. See MTFT for details. pling
MSWS Modulation square-wave transfer function, sagittal. See MTFT for Sam- Wave (see left)
details. pling
MSWT Modulation square-wave transfer function, tangential. See MTFT Sam- Wave (see left)
for details. pling
MTFA Modulation transfer function, average of sagittal and tangential. Sam- Wave (see left)
See MTFT for details. pling
MTFS Modulation transfer function, sagittal. See MTFT for details. Sam- Wave (see left)
pling

MTFT Modulation transfer function, tangential. This computes the Sam- Wave (see left)
diffraction MTF. Samp must be an integer (1, 2, ...) where 1 yields pling
32 x 32 sampling, 2 yields 64 x 64 sampling, etc. Wave can be a
valid wavelength number, or 0 for polychromatic. Field must be a
valid field number (1, 2, ...). Freq is the spatial frequency in cycles
per mm. If the sampling is set too low for accurate computation of
the MTF, then the MTF operands all return zero. If both the
tangential and sagittal MTF are needed; place the MTFT and
MTFS operands on adjacent lines and they will be computed
simultaneously. See the discussion "Using MTF operands" in this
chapter for details.
MXAB Maximum Abbe number. This boundary operand constrains the First surf Last surf -
Abbe number of surfaces between "First Surf" and "Last Surf" to
be less than the specified target value. See also MNAB. This
catalog glasses.
MXCA Maximum center thickness air. This boundary operand constrains First surf Last surf -
each of the thicknesses of surfaces from "First surf" to "Last surf"
which have air (i.e. no glass) as the glass type to be less than the
target value. See also MXCT and MXCG. This operand controls
multiple surfaces simultaneously.
MXCG Maximum center thickness glass. This boundary operand First surf Last surf -
constrains each of the center thicknesses of surfaces from "First
surf" to "Last surf" which have a non-air glass type to be less than
the target value. See also MXCT and MXCA. This operand
controls multiple surfaces simultaneously.
MXCT Maximum center thickness. This boundary operand constrains First surf Last surf -
"Last surf" to be less than the specified target value. See also
MXCG and MXCA. This operand controls multiple surfaces
simultaneously.
MXCV Maximum curvature. This boundary operand constrains each of First surf Last surf -
the curvatures of surfaces from "First surf" to "Last surf" to be less
than the specified target value. See also MNCV. This operand
controls multiple surfaces simultaneously.
MXDT Maximum diameter to thickness ratio. Controls the maximum First surf Last surf -
allowable value on the ratio of surface diameter to center
thickness. Only surfaces with non-unity index of refraction are
considered. See also MNDT. This operand controls multiple
surfaces simultaneously.
MXEA Maximum edge thickness air. This boundary operand constrains First surf Last surf -
surf" which have air (i.e. no glass) as the glass type to be less
than the specified target value. See also MXET, MXEG, ETLT, and
XXEA. This operand controls multiple surfaces simultaneously.
The boundary applies to the top "+y" edge of the surface only; see
XXEA for constraining non-rotationally symmetric surfaces.
MXEG Maximum edge thickness glass. This boundary operand First surf Last surf -
constrains each of the edge thicknesses of surfaces from "First
surf" to "Last surf" which have a non-air glass type to be less than
the target value. See also MXET, MXEA, ETLT, and XXEG. This
operand controls multiple surfaces simultaneously. The boundary
applies to the top "+y" edge of the surface only; see XXEG for
constraining non-rotationally symmetric surfaces.

MXET Maximum edge thickness. This boundary operand constrains First surf Last surf -
surf" to be less than the specified target value. See also "MXEG",
"MXEA", "ETLT", and "XXET". This operand controls multiple
surfaces simultaneously. The boundary applies to the top "+y"
edge of the surface only; see XXET for constraining non-
rotationally symmetric surfaces.
MXIN Maximum index at d-light. This boundary operand constrains the First surf Last surf -
Nd value of surfaces between "First Surf" and "Last Surf" to be
less than the specified target value. See also MNIN. This operand
only considers surfaces using model or substitute status catalog
glasses.
MXPD Maximum P g, F . This boundary operand constrains the First surf Last surf -
deviation of the partial dispersion of surfaces between "First Surf"
and "Last Surf" to be less than the specified target value. See also
MNPD. This operand only considers surfaces using model or
substitute status catalog glasses.
MXSD Maximum semi-diameter. Constrains the semi-diameter to be First surf Last surf -
less than the specified target over the surface range.
NORD Normal distance to the next surface. This operand computes the Surf - X, Y
surface normal vector at any coordinate, then returns the
distance to the next surface measured along the normal vector.
NORX Normal vector x component. This operand returns the x Surf - X, Y
component of the surface normal vector at any coordinate.
NORY Normal vector y component. This operand returns the y Surf - X, Y
NORZ Normal vector z component. This operand returns the z Surf - X, Y
NPGT Non-sequential parameter greater than. The Hx value is used to Surf Object (see
define the parameter number. left).
NPLT Non-sequential parameter less than. The Hx value is used to Surf Object (see
define the parameter number. left).
NPVA Non-sequential parameter value. The Hx value is used to define Surf Object (see
the parameter number. left).
NPXG Non-sequential object position x greater than. Surf Object (see left)
If "Ref?" is 0, the coordinates are relative to the reference object.
If "Ref?" is 1, the coordinates are relative to the origin of the NSC
coordinate system or entry port.
If "Ref?" is 2, then the coordinates are relative to the global
coordinate reference surface.
NPXL Non-sequential object position x less than. See NPXG. Surf Object (see left)
NPXV Non-sequential object position x value. See NPXG. Surf Object (see left)
NPYG Non-sequential object position y greater than. See NPXG. Surf Object (see left)
NPYL Non-sequential object position y less than. See NPXG. Surf Object (see left)
NPYV Non-sequential object position y value. See NPXG. Surf Object (see left)
NPZG Non-sequential object position z greater than. See NPXG. Surf Object (see left)

NPZL Non-sequential object position z less than. See NPXG. Surf Object (see left)
NPZV Non-sequential object position z value. See NPXG. Surf Object (see left)
NSDC Non-sequential coherent intensity data. Detector refers to the Surf Detector (see left)
object number of the desired detector.
If the object number corresponds to a detector object, then the
data from the specified pixel is returned.
Data is 0 for real, 1 for imaginary data, 2 for the amplitude sum,
and 3 for the incoherent intensity. See Optimizing with sources
and detectors in non-sequential mode on page 427 for complete
details.
NSDD Non-sequential incoherent intensity data. Detector refers to the Surf Detector (see left)
object number of the desired detector. If the object number is
zero, then all detectors are cleared. The detector object may be
a detector rectangle or a faceted detector object.
If the object number corresponds to a detector object, then the
data from the specified pixel is returned. If the pixel number is
zero, then the sum of the flux or average flux/area for all pixels
for that detector object is returned. If the pixel number is -1, then
the maximum flux or flux/area is returned. If the pixel number is -
2, then the minimum flux or flux/area is returned. If the pixel
number is -3, the number of rays striking the detector is returned.
If the pixel number is -4, the standard deviation (RMS from the
mean) of all the pixel data is returned.
Data is 0 for flux, 1 for flux/area, and 2 for flux/solid angle pixel.
Only data values 0 and 1 (for flux and flux/area) are supported for
faceted detectors. See Optimizing with sources and detectors in
non-sequential mode on page 427 for complete details.
NSRA Non-sequential single ray trace. Source refers to the object Surf Source (see left)
number of the desired source. This source must be defined to
trace just a single analysis ray. If Splt is non-zero, then splitting is
on. If Upol is non-zero then polarization will be used. If splitting is
on polarization is automatically selected.
Scattering is always turned off for this feature, since scattering
introduces random paths for the child rays, which is not suitable
for optimization. Errors are always considered.
If multiple NSRA operands trace the same ray are adjacent in the
merit function editor, the ray will only be traced once for efficiency.
Seg# refers to the segment number that contains the data to be
optimized. The Dat# refers to the data for the specified segment.
Use Dat# 1-9 for x-coordinate, y-coordinate, z-coordinate, x-
cosine, y-cosine, z-cosine, x-normal, y-normal, and z-normal,
respectively. These values are relative to the entry port, see 31-
39 below.
Use Dat# 10-15 for path-to, intensity, phase of, phase at, index,
and starting phase, respectively.
Use Dat# 21-26 for Ex real, Ex imaginary, Ey real, Ey imaginary,
Ez real, and Ez imaginary, respectively. Polarization must be on
for real and imaginary data to be returned.
Use Dat# 31-39 for the coordinate data defined by Dat# 1-9
converted to coordinates relative to the global coordinate
reference surface.
For more information on these data items see Ray database
files on page 366.

NSTR Non-sequential trace. Source refers to the object number of the Surf Source (see left)
desired source. If source is zero, all sources will be traced. If Splt
is non-zero, then splitting is on. If Scat is non-zero, then scattering
is on. If Upol is non-zero then polarization will be used. If splitting
is on polarization is automatically selected. If IgEr is non-zero,
then errors will be ignored. See Optimizing with sources and
detectors in non-sequential mode on page 427 for complete
details.
NTXG Non-sequential object tilt about x greater than. Surf Object -
NTXL Non-sequential object tilt about x less than. Surf Object -
NTXV Non-sequential object tilt about x value. Surf Object -
NTYG Non-sequential object tilt about y greater than. Surf Object -
NTYL Non-sequential object tilt about y less than. Surf Object -
NTYV Non-sequential object tilt about y value. Surf Object -
NTZG Non-sequential object tilt about z greater than. Surf Object -
NTZL Non-sequential object tilt about z less than. Surf Object -
NTZV Non-sequential object tilt about z value. Surf Object -
OBSN Object space numerical aperture. This is only useful for finite - - -
conjugate systems, and is calculated on axis at the primary
wavelength.
OFF This operand indicates an unused entry in the operand list. OFF - - -
operands are automatically converted to BLNK operands upon
evaluation of the merit function. OFF is only used to indicate that
the merit function operand type was not recognized.
OPDC Optical path difference with respect to chief ray in waves. - Wave Yes
OPDM Optical path difference with respect to the mean OPD; this - Wave Yes
operand computes the OPD referenced to the mean OPD of all
rays in the pupil. OPDM has the same restrictions that TRAC
does; see TRAC for a detailed discussion.
OPDX Optical path difference with respect to the shifted and tilted - Wave Yes
reference sphere that minimizes the RMS wavefront error; which
ZEMAX calls the centroid reference. OPDX has the same
restrictions that TRAC does; see TRAC for a detailed discussion.
OPGT Operand greater than. This is used to make any operand an Op # - -
inequality constraint.
OPLT Operand less than. This is used to make any operand an Op # - -
inequality constraint.
OPVA Operand value. This operand constrains the value of a prior Op # - -
operand to be equal to the target value.
OPTH Optical path length. This is the distance, in lens units, the Surf Wave Yes
specified ray travels to the surface "Surf". The distance is
measured from the object for finite conjugates; otherwise the
distance is referenced to the first surface. The optical path
accounts for the index of refraction of the media, and for phase
adding surfaces such as gratings and binary optics. See PLEN.

OSCD Offense against the sine condition (OSC). There are two - Wave Zone
definitions for OSC supported. The first definition is as described
in Welford, Aberrations of Optical Systems (see REFERENCES
ON LENS DESIGN on page 31). This definition is used if the
zone parameter is zero. An alternate definition due to Prof.
Roland Shack which supports computation of OSC as a function
of pupil zone and uses only real rays is available. This definition
is used if the zone parameter is not zero. The two methods will
give very similar results for systems with modest F/#s and
aberrations when the zone parameter is 1.0 for the alternate
definition. This operand has no meaning if the system is not
axially symmetric.
OSUM Sums the values of all operands between the two specified First Last -
operands. See SUMM.
PANA Paraxial ray x-direction surface normal at the ray-surface Surf Wave Yes
intercept. This is the x component of the surface normal vector at
the intersection point of the specified paraxial ray and the surface
"Surf", in the local coordinate system.
PANB Paraxial ray y-direction surface normal at the ray-surface Surf Wave Yes
intercept. This is the y component of the surface normal vector at
PANC Paraxial ray z-direction surface normal at the ray-surface Surf Wave Yes
intercept. This is the z component of the surface normal vector at
PARA Paraxial ray x-direction cosine of the ray after refraction from the Surf Wave Yes
surface "Surf".
PARB Paraxial ray y-direction cosine of the ray after refraction from the Surf Wave Yes
surface "Surf".
PARC Paraxial ray z-direction cosine of the ray after refraction from the Surf Wave Yes
specified surface.
PARR Paraxial ray radial coordinate in lens units at the specified Surf Wave Yes
surface. This is the radial distance from the local axis to the
intersection of the surface "Surf" and the specified paraxial ray, in
the local coordinate system.
PARX Paraxial ray x-coordinate in lens units at the surface "Surf". Surf Wave Yes
PARY Paraxial ray y-coordinate in lens units at the surface "Surf". Surf Wave Yes
PARZ Paraxial ray z-coordinate in lens units at the surface "Surf". Surf Wave Yes
PATX Paraxial ray x-direction ray tangent. This is the tangent of the Surf Wave Yes
angle the paraxial ray makes in the X-Z plane after refraction from
surface "Surf".
PATY Paraxial ray y-direction ray tangent. This is the tangent of the Surf Wave Yes
angle the paraxial ray makes in the Y-Z plane after refraction from
surface "Surf".
PETC Petzval curvature in inverse lens units. Not valid for non-paraxial - Wave -
systems.
PETZ Petzval radius of curvature in lens units. Not valid for non-paraxial - Wave -
systems.

PIMH Paraxial image height at the paraxial image plane for the - Wave -
specified wavelength. Not valid for non-paraxial systems.
PLEN Path length. This operand computes the total optical path length Surf 1 Surf 2 Yes
(including index of refraction and phase surfaces) between
surfaces 1 and 2 for the specified ray, which is always traced at
the primary wavelength. PLEN is essentially the difference
between two OPTH operands. See OPTH.
PMAG Paraxial magnification. This is the ratio of the paraxial chief ray - Wave -
height on the paraxial image plane to the object height. Only
useful for finite conjugate systems. Note the paraxial image plane
is used even if the system is not at paraxial focus.
PMGT Parameter greater than. This boundary operand constrains the Surf Param -
value of a parameter to be greater than the target value. The
parameter values have different meanings depending upon the
surface type. See the Chapter "Surface Types" for a description
of the parameter values.
PMLT Parameter less than. This boundary operand constrains the value Surf Param -
of a parameter to be less than the target value. The parameter
values have different meanings depending upon the surface type.
See the Chapter "Surface Types" for a description of the
parameter values.
PMVA Parameter value. This boundary operand constrains the value of Surf Param -
a parameter to be equal to the target value. The parameter values
have different meanings depending upon the surface type. See
the Chapter "Surface Types" for a description of the parameter
values.
PnGT This operand is obsolete, use PMGT instead. Surf - -
PnLT This operand is obsolete, use PMLT instead. Surf - -
PnVA This operand is obsolete, use PMVA instead. Surf - -

POPD Physical Optics Propagation Data. For important details see Surf Wave (see left)
PHYSICAL OPTICS PROPAGATION on page 515.
To use this operand, first define the settings on the POP analysis
feature as desired, then press Save on the settings box. The
operand will return data based upon the selected settings.
If Surf is zero, then the saved ending surface number will be used;
otherwise, the specified surface will be used as the ending
surface. If Wave# is zero, then the saved wavelength number will
be used; otherwise, the specified wavelength number will be
used. If Field# is zero, then the saved field number will be used;
otherwise, the specified field number will be used.
The data number determines what data the POP feature will
compute and return as follows:
0: The total fiber coupling.
1: The system efficiency for fiber coupling.
2: The receiver efficiency for fiber coupling.
3: The total power.
4: The peak irradiance.
5, 6, 7: The pilot beam position, Rayleigh range, beam waist (x).
8, 9, 10: The pilot beam position, Rayleigh range, beam waist (y).
11, 12, 13: The local X, Y, Z coordinates of the center of the beam
array on the end surface (this is a reference point and is not
related to the amplitude of the beam).
21, 22: The X, Y coordinates of the centroid of the intensity
distribution in local coordinates relative to the center of the beam.
Undefined data numbers will return 0.
30, 31, 32: The mean, RMS and PTV irradiance variation,
respectively, of the non-zero amplitude portion of the beam.
These operands should only be used when the beam has nearly
uniform irradiance and has just been clipped by a surface
aperture.
33, 34, 35: The mean, RMS and PTV phase variation in radians,
respectively, of the non-zero amplitude portion of the beam.
These operands should only be used when the beam has nearly
uniform irradiance and has just been clipped by a surface
aperture.
40, 41, 42: The fraction of the total power enclosed within a circle
of radius specified by the Extra1 value in lens units, referenced to
the beam centroid (40), chief ray (41), or surface vertex (42).
50, 51, 52: The radius in lens units of the circle at which the
fraction of the total power enclosed is equal to the value specified
by the Extra1 value. The circle is centered on the beam centroid
(50), chief ray (51), or surface vertex (52).
The Extra1 and Extra2 values are only used by selected data
numbers which are reserved for future expansion of this feature.
If adjacent POPD operands all have the same surface #, wave #,
field # , Extra1, and Extra2 values, then the POP analysis is done
just once and all data returned at one time. Note the POPD
operands must be on adjacent rows in the MFE for this efficiency
to be implemented.
POWR The surface power (in inverse lens units) of the surface "Surf". Surf Wave -
This operand only works for standard surfaces.
PRIM Primary wavelength. This is used to change the primary - Wav # -
wavelength number during merit function evaluation. This
operand does not use the target or weight columns.
PROB Multiplies the value of any prior operand by any factor. Oper# - factor

PROD Product of two operands. The two arguments are the row Op #1 Op #2 -
numbers of the operands to multiply. See PROB.
QSUM Quadratic sum. This operand squares and then adds all operands First Last -
between the first and last operand (inclusive), then takes the
square root of the sum. See also SUMM, OSUM, EQUA.
RAED Real ray angle of exitance. This is the angle in degrees between Surf Wave Yes
the surface normal and the ray after refraction or reflection. See
also RAID.
RAEN Real ray angle of exitance. This is the cosine of the angle Surf Wave Yes
between the surface normal and the ray after refraction or
reflection at that surface. Does not return correct results if the
glass prior to the surface is a gradient index medium. See also
RAIN.
RAGA Global ray x-direction cosine. This is the direction cosine of the Surf Wave Yes
ray in the global coordinate system. The origin of the global
coordinate system is at the global reference surface.
RAGB Global ray y-direction cosine. See RAGA. Surf Wave Yes
RAGC Global ray z-direction cosine. See RAGA. Surf Wave Yes
RAGX Global ray x-coordinate. This is the coordinate in lens units in the Surf Wave Yes
global coordinate system. The origin of the global coordinate
system is at the global reference surface.
RAGY Global ray y-coordinate. See RAGX. Surf Wave Yes
RAGZ Global ray z-coordinate. See RAGX. Surf Wave Yes
RAID Real ray angle of incidence. This is the angle in degrees between Surf Wave Yes
the surface normal and the incident ray. Note the angle of
incidence is always positive. See also RAED.
RAIN Real ray angle of incidence. This is the cosine of the angle Surf Wave Yes
between the surface normal and the ray before refraction at that
surface. Does not return correct results if the surface prior to surf
is a gradient index surface. See also RAEN.
RANG Ray angle in radians with respect to z axis. The angle is Surf Wave Yes
measured with respect to the local Z axis.
REAA Real ray x-direction cosine of the ray after refraction from the Surf Wave Yes
surface "Surf".
REAB Real ray y-direction cosine of the ray after refraction from the Surf Wave Yes
surface "Surf".
REAC Real ray z-direction cosine of the ray after refraction from the Surf Wave Yes
surface "Surf".
REAR Real ray radial coordinate in lens units at the surface "Surf". Surf Wave Yes
REAX Real ray x-coordinate in lens units at the surface "Surf". Surf Wave Yes
REAY Real ray y-coordinate in lens units at the surface "Surf". Surf Wave Yes
REAZ Real ray z-coordinate in lens units at the surface "Surf". Surf Wave Yes

RELI Relative illumination. This operand computes the relative Samp Wave See Text
illumination of any field point relative to the (0, 0) field point. Note
that in some systems, the illumination increases off axis, and for
these systems the RI may be greater than one since the RELI
operand uses (0, 0) as a reference field point.
Samp is the grid size, where 10 would yield a 10 x 10 grid of rays.
Wave is the integer wavelength number, Hx is used to define the
field number, and Hy specifies if polarization is used (0 for no, 1
for yes). See also EFNO.
RENA Real ray x-direction surface normal at the ray-surface intercept. Surf Wave Yes
RENB Real ray y-direction surface normal at the ray-surface intercept. Surf Wave Yes
RENC Real ray z-direction surface normal at the ray-surface intercept. Surf Wave Yes
RETX Real ray x-direction ray tangent (slope). Surf Wave Yes
RETY Real ray y-direction ray tangent (slope). Surf Wave Yes
RGLA Reasonable glass. This operand restricts the deviation the index, First Last See Text
Abbe, and deviation of the partial dispersion values may take Surf Surf
from actual glasses in the currently loaded glass catalogs. See
Optimizing glass selection on page 424 for a complete
discussion. The constraint is active over the surface range
specified.
RSCE RMS spot radius (ray aberrations) with respect to the geometric Rings Wave Hx, Hy
image centroid, measured in lens units. This operand is similar to
RSCH, except the reference point is the image centroid instead
of the chief ray. See RSCH for details.
RSCH RMS spot radius (ray aberrations) with respect to chief ray. This Rings Wave Hx, Hy
operand uses a Gaussian quadrature method to estimate the
RMS spot radius at a specified field coordinate and wavelength.
The number returned is in lens units. The method used is only
accurate for systems with circular pupils. The Int1 column is used
to specify the number of rings of rays traced (use no more than
required to converge the result). Only Hx and Hy are used to
define the field point, Px and Py are not used. If the "wave" value
is zero, then a wavelength weighted polychromatic calculation is
performed.
RSRE RMS spot radius (ray aberrations) with respect to the geometric Grid Wave Hx, Hy
image centroid, measured in lens units. This operand is similar to
RSCE, except a rectangular grid of rays is used instead of the
Gaussian quadrature method. This operand always considers
vignetting. A grid value of 1 will trace 4 rays, 2 will trace a 2 x 2
grid per quadrant (16 rays), 3 will trace 3 x 3 rays per quadrant
(36 rays), and so on. Symmetry is considered.
RSRH Similar to RSRE, except the reference point is the chief ray. Grid Wave Hx, Hy
RWCE RMS wavefront error with respect to the diffraction centroid. This Rings Wave Hx, Hy
operand is useful for minimizing the wavefront variance, which is
proportional the Strehl ratio and the area under the MTF curve.
The units are waves. See RWCH. See RSCH for details.
RWCH RMS wavefront error with respect to the chief ray. The units are Rings Wave Hx, Hy
waves. This RMS really refers to the standard deviation of the
wavefront, since the mean OPD is subtracted out. See RWCE.
See RSCH for details.

RWRE Like RSRE, except for wavefront error rather than spot radius. Grid Wave Hx, Hy
RWRH Like RSRH, except for wavefront error rather than spot radius. Grid Wave Hx, Hy
SAGX The sag in lens units of the surface "Surf" in the XZ plane at the Surf - -
semi-diameter distance. See also SSAG.
SAGY The sag in lens units of the surface "Surf" in the YZ plane at the Surf - -
semi-diameter distance. See also SSAG.
SFNO Sagittal working F/#, computed at any defined field and Field Wave -
wavelength. See TFNO.
SINE Sine of the value of the specified operand number. If flag is 0, then Op # Flag -
the units are radians, otherwise, degrees.
SKIS Skip if symmetric. If the lens is rotationally symmetric, then Op #
computation of the merit function continues at the specified
operand number.
SKIN Skip if not symmetric. See SKIS. Op #
SPCH Spherochromatism in lens units. This is the difference between Wave1 Wave2 Pupil
the real marginal axial color and the paraxial axial color. The Zone
distance is measured along the Z axis. The pupil zone defines the
zone for which the real marginal axial color is computed.
SPHA Spherical aberration in waves contributed by the specified Surf Wave -
surface. If the surface value is zero, the sum for the entire system
is used.
SQRT Square root of operand. The argument is the row number of the Op # - -
operand value to take the square root of.
SSAG The sag in lens units of the surface "Surf" in the XZ plane at any Surf - x, y
(x, y) coordinate. The x and y values are defined in the Hx and Hy
columns. See also SAGX, SAGY.
SUMM Sum of two operands. The two arguments are the row numbers Op #1 Op #2 -
of the operands to add. See OSUM.
SVIG Sets the vignetting factors. When included, updates the - - -
vignetting factors for the current configuration.
TANG Tangent of the value of the specified operand number. If flag is 0, Op # Flag -
TCGT Thermal Coefficient of expansion greater than. This boundary Surf - -
operand constrains the TCE of surface "Surf" to be greater than
the specified target value.
TCLT Thermal Coefficient of expansion less than. This boundary Surf - -
operand constrains the TCE of surface "Surf" to be less than the
specified target value.
TCVA Thermal Coefficient of expansion value. This boundary operand Surf - -
constrains the TCE of surface "Surf" to be equal to the specified
target value.
TFNO Tangential working F/#, computed at any defined field and Field Wave -
wavelength. See SFNO.

TGTH Sum of glass thicknesses from first to last specified surface. Note First surf Last surf -
that the sum is inclusive, it is not the thickness between the two
surfaces. See TTHI.
TMAS Total mass. Computes the mass of the glass lenses within the First surf Last surf -
specified range of surfaces. The mass of a surface considers the
volume enclosed up to the following surface; therefore to
compute the mass of a single element the first and last surface
numbers should be the same. See Comments on computing
element volumes on page 220 for a discussion of how element
masses and volumes are computed.
TOLR Tolerance data. The "Data" value is 0 for RSS estimated change Data - -
in performance, 1 for nominal performance, and 2 for estimated
performance (nominal plus estimated change). For details, see
Optimizing tolerance sensitivity on page 426.
TOTR Total track (length) of lens in lens units. - - -
TRAC Transverse aberration radial direction measured in the image - Wave Yes
plane with respect to the centroid. Unlike most other operands,
TRAC critically depends upon the placement of other TRAC
operands within the Merit Function Editor to work correctly. TRAC
operands must be grouped together by field position and
wavelength. ZEMAX traces all TRAC rays with a common field
point together, and then uses the collective data to compute the
centroid of all the rays. Each ray individually is then referenced to
the computed centroid. This operand should only be entered into
the Merit Function Editor by the Default Merit Function tool, and
is not recommended for use directly by the user.
TRAD The x component of the TRAR only. TRAD has the same Wave Yes
TRAE The y component of the TRAR only. TRAE has the same Wave Yes
TRAI Transverse aberration radius measured at the specified surface Surf Wave Yes
with respect to the chief ray. Similar to TRAR, except a surface
other than the image surface may be specified.
TRAR Transverse aberration radial direction measured in the image - Wave Yes
plane with respect to the chief ray. See ANAR.
TRAX Transverse aberration x direction measured in the image plane - Wave Yes
with respect to the chief ray.
TRAY Transverse aberration y direction measured in the image plane - Wave Yes
with respect to the chief ray.
TRCX Transverse aberration x direction measured in the image plane - Wave Yes
with respect to the centroid. See TRAC. This operand should only
be entered into the Merit Function Editor by the Default Merit
Function tool, and is not recommended for use directly by the
user.
TRCY Transverse aberration y direction measured in the image plane - Wave Yes
with respect to the centroid. See TRAC. This operand should only
be entered into the Merit Function Editor by the Default Merit
Function tool, and is not recommended for use directly by the
user.

TTGT Total thickness greater than. This boundary operand constrains Surf Code -
the total thickness (including both front and back surface sags) of
surface "Surf" to be greater than the specified target value. The
thickness is calculated at the semi-diameter radius along the +y
axis if code is zero, the +x axis if code is 1, the -y axis if code is
2, and the -x axis if code is 3. This operand automatically changes
the sign on thicknesses in mirror spaces to always yield a positive
value for physically possible lenses. See TTLT and TTVA.
TTHI Sum of thicknesses from first to last specified surface. Note that First surf Last surf -
the sum is inclusive, it is not the thickness between the two
surfaces. See TGTH.
TTLT Total thickness less than. See TTGT. Surf Code -
TTVA Total thickness value. See TTGT. Surf Code -
UDOP User defined operand. Used for optimizing numerical results Macro # Data# Yes
computed in externally compiled programs. See User defined
operands on page 429. See also ZPLM.
USYM If present in the merit function, this operand instructs ZEMAX to - - -
assume radial symmetry exists in the lens even if ZEMAX detects
symmetry does not exist. This speeds execution of the merit
function in some special cases. See Assume Axial Symmetry
on page 389.
VOLU Volume of element(s) in cubic cm. Computes the volume of the First surf Last surf -
lenses and air spaces for the specified range of surfaces. The
volume of a surface includes the volume enclosed up to the
following surface; therefore to compute the volume of a single
element the first and last surface numbers should be the same.
See Comments on computing element volumes on page 220 for
a discussion of how element masses and volumes are computed.
WFNO Working F/#. See Working F/# on page 55, and ISFN, SFNO, - - -
and TFNO.
WLEN Wavelength. This operand returns the specified wavelength in - Wave# -
micrometers.
XDGT Extra data value greater than. The Int2 number must be between Surf Number -
1 and 200 to indicate which of the extra data values is selected.
XDLT Extra data value less. The Int2 number must be between 1 and Surf Number -
200 to indicate which of the extra data values is selected.
XDVA Extra data value. The Int2 number must be between 1 and 200 to Surf Number -
indicate which of the extra data values is selected.
XENC Extended source encircled energy (distance). This operand Type - (see left)
computes the distance in micrometers to the specified fraction of
extended source geometric encircled energy, using whatever the
current default settings are. To use this operand, first define the
settings on the extended source encircled energy feature as
desired, then press Save on the settings box. The only setting
that is overwritten is the type, which is 1 for encircled, 2 for x only,
3 for y only, 4 for ensquared, 5 for x distribution, and 6 for y
distribution.
Hx is the fraction of energy desired, and must be between zero
and 1, exclusive. Hx is ignored for type 5 or 6; for these types the
returned valued is the full width half max independent of Hx.
See also XENF, DENC, DENF, GENC, and GENF.

XENF Extended source encircled energy (fraction). This operand Sam- Wave (see left)
computes the fraction of extended source geometric encircled, pling
ensquared, x only, or y only (enslitted) energy at a given distance
from the reference point.
The options and settings are identical to XENC, except Hx, which
desired. See also XENC, GENC, GENF, DENC, and DENF.
XNEA Minimum edge thickness for air surfaces. This operand checks First surf Last surf -
the edge thickness at numerous points around the perimeter of
the surface, and tests if all points are at least the minimum
specified thickness. See MNEA.
XNEG Minimum edge thickness for glass surfaces. This operand checks First surf Last surf -
the surface, and tests if all points are at least the minimum
specified thickness. See MNEG.
XNET Minimum edge thickness. This operand checks the edge First surf Last surf -
thickness at numerous points around the perimeter of the
surface, and tests if all points are at least the minimum specified
thickness. See MNET.
XXEA Maximum edge thickness for air surfaces. This operand checks First surf Last surf -
the surface, and tests if all points are no more than the maximum
specified thickness. See MXEA.
XXEG Maximum edge thickness for glass surfaces. This operand First surf Last surf -
checks the edge thickness at numerous points around the
perimeter of the surface, and tests if all points are no more than
the maximum specified thickness. See MXEG.
XXET Maximum edge thickness. This operand checks the edge First surf Last surf -
thickness at numerous points around the perimeter of the
surface, and tests if all points are no more than the maximum
specified thickness. See MXET.
YNIP YNI-paraxial. This number is the product of the paraxial marginal Surf Wave -
ray height times the index times the angle of incidence. This
quantity is proportional to the narcissus contribution of the
specified surface. See Applied Optics, Vol. 21, 18, p3393.

ZERN Zernike Fringe coefficient. The Int1, Int2, Hx, Hy, and Px data Term Wave (see left)
values are used to specify the Zernike term number (1-37 for
fringe, 1- 231 for standard), the wavelength number, the sampling
density (1 = 32 x 32, 2 = 64 x 64, etc.), the field position, and the
Zernike type (0 for fringe, 1 for standard), respectively.
The term number, if negative or zero, may also be used to return
other data from the Zernike fitting as follows:
-8: Peak to Valley OPD (to centroid)
-7: Peak to Valley OPD (to chief)
-6: RMS to zero reference (unused by ZEMAX)
-5: RMS to chief ray
-4: RMS to centroid
-3: Variance
-2: Strehl Ratio
-1: RMS fit error
0: Maximum single point fit error
Note that if you use multiple ZERN operands which only differ in
the term number, they should be placed on adjacent lines in the
editor so ZEMAX only does the fitting once; otherwise, the
computation is slower.
ZPLM Used for optimizing numerical results computed in ZPL macros. Macro # Data# Yes
See User defined operands on page 429. See also UDOP.
ZTHI This operand controls the variation in the total thickness of a First surf Last surf -
range of surfaces over multiple configurations. It is similar to the
TTHI operand, except it is an inequality operator. The target value
specified is the maximum allowed difference between the TTHI at
each defined configuration. For example, if there are 3
configurations where TTHI 3 8 would evaluate to 17, 19, and 18.5,
respectively, ZTHI will return 2 (i.e. 19-17) if the target is less than
2. Otherwise, ZTHI returns the target value. To keep all zoom
configurations the same length, use a target of 0.
The operational operands (SUMM, OSUM, DIFF, PROD, DIVI, SQRT) along with the parametric operands
(CVGT, CVLT, CTGT, CTLT, etc..) can be used to define very general and complex optimization operands, as
discussed in the section Defining complex operands on page 423.
Because of the dimensional differences between parameters such as effective focal length (tens of millimeters
or more) and RMS spot radius (micrometers), usually a weighting of one is sufficient for quantities measured in
lens units. However, the residual value of the effective focal length with this weighting is not likely to be zero.
Increasing the weighting will bring the resulting system closer to the desired effective focal length. This effect is
often noticeable when defining ETGT (edge thickness greater than) operands. Usually ETGT with a target of zero
will often yield a value just slightly less than zero. Rather than increase the weight, it is much simpler to provide
a target value of 0.1, or some such number.
After making changes to the operand list, the current values of each operand can be updated by selecting
Tools, Update. This is also useful for checking to see what the current values of each operand are, and which has
the greatest contribution to the merit function. The percent contribution is defined as
2
Wi ( Vi Ti )
% contrib i = 100 ------------------------------------- .
Wj ( Vj Tj )
2
where the index j indicates the sum over all operands.

The merit function is automatically saved with the lens file.

Understanding boundary operands
The boundary operands such as MNCT, CTGT, DIMX, and others behave somewhat differently than specific
target operands such as TRAR and REAY. When you specify a boundary on a parameter, you specify the target
value as the definition of the boundary. For example, to maintain a minimum center thickness on surface 5 of 10
mm, you might use a command such as CTGT 5 10 (where the 5 is in the Int1 column and 10 is in the target
column). If you update the merit function and then observe the "value" column of that operand, there are two
possibilities for the value: 1) if the boundary is violated, that is, the center thickness is less than 10, then the actual
value of the thickness will be displayed, or 2) if the boundary is not violated, that is, the center thickness is greater
than 10, then the value 10 will be displayed.
The rule is if the boundary is violated, the actual value is shown; if it is not violated, the value is set to the target
and is therefore ignored by the optimization algorithm. If during optimization the boundary becomes violated, then
the value will automatically be updated and the optimization algorithm will attempt to correct the offending
parameter.
The boundary operands which constrain a range of surfaces are slightly more complicated. These multiple
surface operands return values which represent the total effect of all violated boundaries within the specified
surface range. For example, the operand MNCT 1 10 will constrain the minimum center thickness of surfaces 1
through 10. If the target is 3.0, which defines the boundary, then the difference between the value of the operand
and the target is the sum of the difference between 3.0 and the thicknesses of all surfaces between 1 and 10
whose center thickness is less than 3.0. If only one surface in the range has a center thickness less than 3.0, say
2.5, then the operand has a value of 2.5. If a second surface is added that has a thickness less than 3.0, say 2.2,
then the operand will have a value of 1.7 (2.5 minus .8; the .8 is 3.0 - 2.2). The total difference between the target
of the operand and the value is 3.0 - 1.7 or 1.3. This difference of 1.3 is due to the violation of 0.5 by the first
surface and another 0.8 by the second surface.
If calculating the value of these boundary operands seems confusing, don't worry; ZEMAX does all the
calculations for you. All you need to do is to specify the boundary type (such as MNCT or MNET) the boundary
range (surface 1 through 10, or whatever) and the desired value (3 mm or whatever). If all the boundary
constraints are met, then the operand value is equal to the target, otherwise, the value will be different and the
merit function will increase. The increased merit function will cause the optimization algorithm to seek a reduction
of the operand contribution.
If a boundary operand does not seem to work, there are several things to check:
1) Make sure the variables you have defined can have some effect on the boundary operands. A common
mistake is to specify MNCT and have some "frozen" thickness within the surface range. If the thickness violates
the boundary and it is not variable, ZEMAX can't fix it. The operands DO NOT ignore violated but frozen
boundaries.
2) If there is a small residual error, try increasing the boundary value. For example, if MNCT is used with a
target of 0.0, and the value is a small number (like -.001) the problem is not that the operand doesn't work, it
is that the residual error is too small to increase the merit function significantly. It is usually better to increase
the target to 0.1, or some other number, rather than to increase the weight. Increasing the weight will only lead
to a smaller violation (like -.0000001) rather than meeting the boundary.
3) Check to see if there is a reasonable contribution to the merit function. You can easily check this with the
percent contribution column. By looking at the percent contribution column, verify that the operand in question
has enough influence on the total merit function. If it does not, increase the weight, or see the preceding
paragraph for advice on changing the target.
Understanding the boundary operands is a crucial part of mastering ZEMAX optimization, and with a little
practice you will find them to offer excellent control and flexibility.
Using MTF operands
The MTF operands such as MTFT, MTFS, and MTFA provide a capability to directly optimize the diffraction
MTF. This is a powerful capability, however, using the MTF operands requires some care on the part of the user.
For systems not close to the diffraction limit, geometric equivalent MTF operands are provided: GMTT, GMTS,
and GMTA. These should be used in place of the diffraction operands for systems with greater than about 2-5
waves of aberration.

The MTF operands calculate the complete diffraction or geometric MTF exactly like the plots available on the
Analysis, Diffraction menu option. Therefore, any optical system for which the MTF plot produces invalid data
(due to excessive OPD in the pupil, see the chapter "Analysis" for details) will also produce meaningless data
during optimization. For example, it is unrealistic to optimize a lens for MTF starting from plane parallel plates,
because of course the MTF generally cannot be accurately computed for such a system. Also, MTF optimization
is considerably slower than RMS spot radius or RMS wavefront error optimization, typically 5 to 50 times slower.
Note that if you use both a MTFT and a MTFS operand for the same field and wavelength data, they should be
placed on adjacent lines in the editor; otherwise, the MTF is computed twice. If the sampling is too low for accurate
computation of the MTF, then the MTF operands return zero rather than a meaningless number.
The slow execution speed may be noticeable when updating the Merit Function Editor display, and entering
and exiting the optimization dialog box. At these times the merit function is updated. If several MTF operands
have been entered, ZEMAX may take several minutes on a slow computer just to refresh the screen.
A good approach is to interactively design your system using RMS wavefront error. Generally speaking,
systems with low RMS wavefront error will have reasonable MTF performance. After the design is very nearly in
the final form, then try switching over to MTF optimization for a "touch-up". It is also a good idea to completely
erase the default merit function before entering any MTF operands, except of course for any boundary constraints
needed (such as MNCT).
The MTF operands use the operand data columns such as Int1, Int2, and Hx differently than most operands.
The Int1 column determines the sampling density to be used. A value of 1 specifies a 32 x 32 grid should be used,
2 specifies a 64 x 64 grid, and so on. Use the smallest grid size for which the data are computed accurately; again
see the "Analysis" chapter for details.
The Int2 column selects the wavelength, like most operands. However, a value of zero may be entered to
specify a polychromatic calculation. In this case, the wavelength-weighted polychromatic MTF is computed.
Obviously, this is slower than a monochromatic MTF calculation.
The Hx column is used to specify the field position, and it must be an integer between 1 and the number of
defined fields.
The Hy column is the spatial frequency in cycles per millimeter, independent of the lens system units. Any
value may be entered; the operand will return 0 if it is past the incoherent cut-off frequency. The value does not
need to be an integral multiple of some fraction; the exact value of the MTF is computed using a cubic spline fit
to the neighboring data points, just like the MTF data plots.
The target and weight columns are used just like any other operand. Of course, if the target is set to 1.0, and
the frequency is non-zero, then the value of the operand can never reach the target.
Performing an optimization
To begin optimization, choose Tools, Optimization from the main menu bar. The optimization control dialog box
will appear with the following options.
OPTIMIZATION OPTIONS
Item Description
Automatic Executes until ZEMAX determines the system is no longer improving significantly.
1 Cycle Executes a single optimization cycle.
5 Cycles Executes 5 optimization cycles.
Inf. Cycles Executes optimization cycles in an infinite, continuous loop until "Terminate" is pressed.
Terminate Terminates a running optimization, and returns control back to the dialog box.
Exit Closes the optimization dialog box.

Auto Update If checked, ZEMAX will automatically update and redraw all open windows at the end of
each optimization cycle. This allows monitoring of the optimization progress using any
analysis feature.
# CPU's Selects the number of CPU's over which to spread the optimization task. More than 1
may be selected, even on a single CPU computer, in which case the single CPU will time
share the multiple simultaneous tasks. The default is the number of processors detected
by the operating system.
Selecting automatic will cause the optimizer to run until no progress is being made. The other options will run
the specified number of cycles. Automatic mode is highly recommended. The time required to run a given
optimization cycle varies enormously with the number of variables, the complexity of the system, the number of
solves, the number of operands, and of course the computer speed. If the cycle is taking too long, or if it appears
to be hung up, or if you feel the design is not progressing adequately, click on Terminate to end the optimization
run.
When the optimization begins, ZEMAX first updates the system merit function. If any of the operands cannot
be computed, the optimization cannot begin, and an error message will be displayed. Operands cannot be
computed if they require the tracing of rays which miss surfaces or which undergo total internal reflection (TIR)
at an index boundary. If such an error message appears, usually the starting lens prescription is in error, or the
ray targets are incorrectly defined (this will not happen with the default merit functions, but might happen with
user defined rays). ZEMAX can automatically recover if the merit function cannot be evaluated during the course
of optimization; only the starting system need be adequate to compute all operands in the merit function.
Defining complex operands
Although the default merit function, coupled with a few predefined operands, is perfectly suitable for the
majority of optical designs, there are times when an unusual constraint needs to be added to the merit function.
Rather than define a very large number of very specific operands, ZEMAX allows you to build your own operands
out of simple building blocks.
ZEMAX allows very general operand definitions. There are two tricks to creating these operands. First, use
certain operands with zero-weighting to define the parameters you need, and second, use the operational
operands to define relationships between them. For example, suppose you require that the thickness of surface
3 and the thickness of surface 4 sum to 10. There is an operand that does this, TTHI. The command structure
would look like this:
Number Type Int1 Int2 Target Weight

1 TTHI 3 4 10 1
However, for illustration only, note that there is an alternate way of calculating the same thing:

1 CTVA 3 0 0
2 CTVA 4 0 0
3 SUMM 1 2 10 1
Operand 1 uses the Center Thickness VAlue (CTVA) command to extract the value of the thickness of surface
3. Similarly, operand 2 is used to extract the thickness of surface 4. The zero weighting on both operands means
the optimization algorithm ignores the constraint; it is only used as an intermediate step. Operand 3 now sums
two operands: number 1 and number 2. The result is the sum of the thicknesses of surfaces 3 and 4 is the value
of operand 3, and this has a non-zero weight. The optimization algorithm will attempt to drive the sum to 10.
Why go to all the trouble of this three-step process if a single TTHI command would do the same thing? The
reason is that this approach can be extended to develop very general operands. For example, suppose you

wanted the radius of curvature of surface 5 to be centered on the vertex of surface 8. Study the following
commands to see if you understand how this is done:

1 CVVA 5 0 0
2 TTHI 5 7 0 0
3 PROD 1 2 1 1
The CVVA command extracts the curvature of surface 5, the curvature we want to control. TTHI 5 7 calculates
the distance from surface 5 to surface 8 (note we only sum to surface 7 to get to surface 8, since the thickness
of surface 8 gives the distance to surface 9). Since the curvature of the surface is the reciprocal of the radius, the
product of the curvature and the distance must be one; hence the target is 1 for operand 3. Operand 3 is also the
only weighted operand in the sequence.
Now consider the requirement that the thickness of surface 5 must be greater than twice the radius of curvature
of surface 4 plus the conic constant of surface 2 (this is nonsensical, but illustrative of the flexibility in the
approach):

1 CTVA 5 0 0
2 CVVA 4 0 0
3 CONS 2 0
4 DIVI 3 2 0 0
5 COVA 2 0 0
6 SUMM 4 5 0 0
7 DIFF 1 6 0 0
8 OPGT 7 0 1
Operand 1 extracts the (center) thickness of surface 5. Operand 2 extracts the value of the curvature of surface
4. Operand 3 sets a constant of two, and operand 4 divides the value 2 by the curvature (yielding twice the radius
of curvature). COVA extracts the conic, and SUMM ads operands 2 and 4. Operand 7 takes the difference of the
thickness and twice the radius plus the conic. Since we want the former to exceed the latter, we set an operand
greater than constraint; the only one to have a non-zero weighting.
Optimizing glass selection
Optimization of glasses is handled somewhat differently than other data. Optimizing the glass choice directly
is a difficult and unpredictable process because there does not exist a continuum of glasses on the glass map.
There are two methods for dealing with this problem: by using model glasses or by using glass substitution. Glass
substitution is usually far superior, but is only supported in the EE edition of ZEMAX.
Using model glasses
The model glass method is to idealize the glass dispersion using a few numerical parameters, and then
optimize these parameters while constraining either the parameter values or the computed index values to be
similar to available glasses. This is the "model" glass method. Model glasses are described in detail in the chapter
"Using Glass Catalogs". One disadvantage of the model glass method is that the optimized parameters and
resulting index values may not correspond to any physically existing glass. Another disadvantage is that model
glasses are only sufficiently accurate in the visible spectrum. This method is used by the conventional optimizer
described in this chapter.

To optimize glasses then requires several steps. First, change the glass of the appropriate surface to a "Model"
glass using the glass solve dialog box in the Lens Data Editor. For information on model glasses, see Using
model glasses on page 485. When you change the glass from "Fixed" to "Model", ZEMAX will make a suitable
guess for the index, Abbe number, and partial dispersion; you only need to change the values if you wish. The
three values can be made variable by clicking on the "Vary" box next to each entry.
The Ctrl-Z shortcut on the glass column will also work; it makes the index, Abbe, and partial dispersion variable
automatically. The model glass data values can now be optimized using the optimization feature in the usual way.
Unconstrained glass optimization usually will lead to very high index materials being selected. This is because
surfaces with high refractivity (a large difference in index across the boundary) need less curvature than low
refractivity surfaces to have the same optical power. Lower curvature surfaces introduce less aberration.
Unfortunately, high index materials are expensive, heavy, harder to fabricate, and may be brittle, delicate, or
susceptible to stains and scratches. Also, very high index materials do not always exist; there are few glasses
(for the visible spectrum) available with an Nd higher than about 1.9. The Vd value also is limited to the range of
roughly 20 to 80. Therefore, it is essential to limit the Nd and Vd values to reasonable ranges during optimization.
The partial dispersion deviation also must be limited in range.
There are two ways to limit the Nd, Vd, and P g, F values. The simplest way is to add the RGLA operand
somewhere in the operand list. The RGLA operand measures the "distance" on the glass map between the index,
Abbe number, and partial dispersion of the model glass to the closest glass in the currently loaded catalogs. For
example, if you are optimizing the index and Abbe and you have specified that the Schott and Hoya catalogs are
used (these are specified on the general data window), the RGLA operand computes the "distance" to each glass
in these catalogs. If the smallest "distance" is less than the target value specified for the RGLA operand, then the
boundary condition is met, and the value of the operand is equal to the target. If the closest glass is farther than
the target value, then the RGLA value is the actual "distance". The "distance" is defined by the square root of the
weighted sum of the squares of the difference between the index, Abbe, and partial dispersion terms for two
glasses. The "distance" between any two glasses is given by
1---
2 2 2 2
d = [ W n ( Nd 1 Nd 2 ) + W a ( Vd 1 Vd 2 ) + W p ( P g, F1 P g, F2 ) ] ,
where the factors Wn, Wa, and Wp weight the various terms. The weighting factors may be user defined on
the RGLA operand parameter list, or if left at zero, will default to 1.0, 1E-04, and 1E+02, respectively.
The best way to use RGLA is to specify the surface range that covers all of the surfaces you are optimizing.
For a target value, start off with 0.05. This will allow the glasses to easily move all over the glass map, since the
spacing between various glasses is usually less than 0.05. After optimization, decrease the target to roughly 0.02
and reoptimize. This will encourage the optimized system to choose index and Abbe numbers reasonably near
actual glasses.
The other method for constraining index and Abbe values is to use the MNIN, MXIN, MNAB, and MXAB
controls. These operands are mnemonics for Minimum and Maximum Index and Abbe values, and they are
documented in the preceding tables. These operands can be used to restrict the optimization to specific
rectangles on the glass map. It may be useful to use RGLA along with MXIN, for example, to restrict the glass
selection to existing glasses with an index lower than some value.
At some point you will want to convert your variable index data back to a real glass. To convert from a model
glass to the closest real glass, remove the model glass status using Ctlrl-Z, or choose the glass solve type as
"Fixed". There will usually not be a perfect match between the optimized Nd and Abbe values and those of an
actual glass in the current catalog. However, ZEMAX will search through the catalog and find the "best fit" glass
using a least-squares criterion similar to the RGLA definition above (the partial dispersion term is omitted). If a
glass substitution template is being used, then the template will also be considered, and only glasses meeting
the template specifications will be considered. For information on templates, see Glass Substitution Template
on page 195. The glass in the catalog which differs the least from the variable index parameters is the glass
selected. This glass is also reported on the "Surface Data Summary" feature (select Reports, Surface Data). The
index of refraction data shown is that calculated from the Nd and Abbe values, not the best fit glass. After

converting from a model glass to a real glass, another optimization run is generally required. For systems with
delicate chromatic aberration balancing, the best glass choice might never be found using variable glasses,
because the model glass dispersion is never identical to the dispersion of a real glass.
Using glass substitution
The glass substitution method is to directly alter the glass types, and then reoptimize to see if the new glasses
yield a better solution. This method can be used manually, by changing the glass type and then reoptimizing, or
the process can be automated using the global optimization techniques described in the next chapter, "Global
Optimization". The global optimization method uses only actual glass catalog materials, and in this sense is the
superior method. See Using glass substitution on page 437 for a discussion on this method for optimizing glass
selection.
Optimizing zoom and multi-configuration lenses
Optimizing zoom lenses is virtually identical to optimizing conventional single-configuration lenses. See the
Chapter "Multi-Configurations" for details.
Optimizing extra data
!This discussion is only relevant for users of ZEMAX-EE.
Certain surface types supported by ZEMAX-EE, such as the Zernike, Zernike phase, extended polynomial,
and binary optic surface use the extra data values. These extra data values can be edited, loaded from ASCII
files, and may be used as variables for optimization. For details on the editor, see the chapter "Editors menu".
To make an extra data value a variable, open up the Extra Data Editor. When the extra data editor appears,
move the cursor to the row and column with the value you want to optimize, and press Ctrl-Z (the same command
used to set variables on the main screen). The variable will now be optimized when the optimization is run.
There are also several boundary constraints for use with the extra data values. XDVA, XDGT, and XDLT are
extra data value, greater than, or less than, respectively. The Int1 column on the spreadsheet display indicates
the surface number to which the operand applies, and the Int2 column is used to specify which of the extra data
values is to be used.
Optimizing tolerance sensitivity
The idea of optimizing a lens for reduced tolerance sensitivity is that a very low merit function is not useful if
the as-built performance is significantly degraded by tolerance sensitivity. Therefore, the optimal design has a
reasonable nominal performance but is relatively insensitive to manufacturing defects. The theory is there may
be a potential trade-off between performance and tolerance sensitivity.
Tolerance sensitivity in an optical design comes from many sources, including angles of incidence of rays on
surfaces, aberration balancing, and the nature of the potential fabrication defects. The interaction of multiple
defects makes accurate tolerance prediction a difficult statistical problem. For a complete discussion of tolerance
analysis, see the chapter TOLERANCING on page 441.
The optimization operand TOLR can in principle be used to optimize for reduced sensitivity to tolerances. To
use TOLR, first optimize a design for reasonable starting performance. Then, define the relevant tolerance
operands, limits, compensators, and criteria as described in the chapter on Tolerancing. Save the options of the
tolerance dialog box; ZEMAX uses these saved options to compute the data returned by TOLR. The only criteria
that is not allowed is "Merit Function" since that would create an infinite loop. If a script tolerance criteria is being
used, do not load any merit functions within the script that contain TOLR operands, as this may also create an
infinite loop that ZEMAX cannot detect.
As part of the tolerance sensitivity analysis, ZEMAX computes a nominal performance estimate, and predicts
an RSS estimated change. The predicted total performance is the sum of nominal and estimated change. These
values are computed and retuned by the TOLR operand for optimization. TOLR values may be targeted and
weighted as any other optimization operand.
TOLR takes a single "data" integer. If data is 0, TOLR returns the estimated change in performance. If data is
1, TOLR returns the nominal performance. If data is 2, TOLR returns the predicted performance, which is the sum
of the nominal and estimated performance change. The first instance of TOLR (the first row in which TOLR

appears) is the only instance for which the tolerance analysis is actually run, so that TOLR only runs once for
each merit function evaluation. The other values computed by TOLR are saved and can be retrieved by using
data values other than 0 on subsequent calls to TOLR. Other values of data are currently unused and may be
used for future expansion of this feature.
The practical difficulty of optimization of tolerance sensitivity lies in the computation time. Complicated lenses
may have hundreds of tolerance operands, and complex criteria may take considerable time to compute. When
using TOLR, the sensitivity analysis may be computed thousands or even millions of times, which can make the
use of TOLR impractical. The following tips should be used to improve the efficiency of the tolerance analysis and
make TOLR efficient:
-Test and verify the tolerance sensitivity analysis outside of optimization before using TOLR.
-Use "Sensitivity" and not "Inverse Sensitivity". Inverse sensitivity is slower and meaningless for optimization.
-Use "Paraxial Focus" for the "Comp" control if possible... optimization of the compensator values requires an
"optimization within an optimization" that is slow and memory intensive; although it is supported if required.
-ZEMAX will automatically ignore the Monte Carlo portion of the tolerance analysis.
-Do not use "Merit Function" as the tolerance criteria, as this will create an infinite loop.
The faster the tolerance sensitivity runs, the faster TOLR and thus the optimization will execute.
Optimizing objects in a non-sequential group with sequential rays
Optimizing variables within a non-sequential group is fundamentally no different from optimizing other
numerical parameters. Variables are set in the same way as for parameters in the Lens Data Editor. However,
optimizing non-sequential object properties is difficult because of the unpredictable way in which rays may (or
may not) propagate through a non-sequential group. For non-sequential objects such as prisms, usually a small
change in position or size of the prism does not dramatically affect the ray path. However, for objects such as light
pipes, a small change in the object definition can dramatically affect the ray path. Rays that once propagated
through an object may miss the object completely if the object position or angle changes slightly. This usually
causes severe errors in the computation of derivatives, and the optimization performs either poorly or not at all.
Another problem with some non-sequential systems is the exit pupil may not be a reasonable image of the
entrance pupil. For this reason, rectangular array rather than Gaussian Quadrature should be used if the system
is a non-imaging system that does not form an image of the entrance pupil at the exit pupil.
For these systems, optimization may proceed more effectively using the global optimization algorithms, which
do not rely exclusively on derivative computation.
Optimizing with sources and detectors in non-sequential mode
Optimizing illumination systems or other optical systems that use non-sequential sources and detectors is
supported using the NSDD and NSTR operands.
A typical merit function would consist of three groups of operands:
First, NSDD operands would be used to clear the data in the current detectors. Use NSDD with the detector
number set to zero to clear all energy in all detectors. Usually a single NSDD at the top of the merit function
is all that is needed. NSDD returns a value of zero and has no effect on the merit function value when used to
clear detectors.
Second, NSTR operands are used to trace rays from NSC sources. NSTR i traces analysis rays from source
i; NSTR 0 traces all analysis rays from all sources. Note the number of analysis rays on the NSC editor
determines how many rays are traced... and how long the evaluation of the NSTR operand will take. NSTR
always returns a value of zero and has no effect on the merit function value. The NSTR operand supports
options to split, scatter, and use polarization.
Third, a new group of NSDD or NSDC operands are used to read out the detector data. NSDD has four
arguments: surface, detector, pixel, and data. Surface is the surface number of the NSC group (use 1 if the
program mode is non-sequential). Detector is the object number of the detector. Both detector objects and
faceted detectors may be used as detectors. If pixel is an integer greater than zero, the flux, flux/area, or flux/
solid angle is returned for that pixel. Which of the three is determined by the data argument, which should be
0, 1, or 2 for flux, irradiance, or intensity, respectively. If pixel is 0, the sum of all the flux or flux/area in the
detector is returned. If pixel is -1, the maximum flux or flux/area for all pixels is returned. If pixel is -2, the

minimum flux or flux/area for all pixels is returned. If pixel is -3, the number of rays striking the detector is
returned. If pixel is -4, the RMS deviation from the mean for all pixels is returned. The units of the returned
data is determined by the system units, see Analysis Units on page 87. If the object is a faceted detector, or
if the pixel number is -1 or 0, only data options 0 and 1 are supported. Similar capability exists for coherent
data using NSDC.
The practical difficulty in optimizing these systems is the difficulty of computing derivatives of detected energy
with respect to variable parameters because of the relatively large uncertainty in computing detected energy.
Many rays must be traced to determine illumination patterns approximately.
Comments about random numbers and NSTR
Note that when launching rays with the NSTR operand, ZEMAX seeds the random number generator with the
identical value every time the merit function is evaluated. This means the identical set of random numbers are
used on every evaluation of the merit function. If no changes are made to the optical system, the merit function
will always evaluate to a consistent value; a desirable property for the optimization algorithm to function. However,
if a change is made to the system, rays may take different paths and the identical set of random numbers may
be used in different ways; for example, in the computation of scattering paths.
Pixel numbering for detectors
For rectangular detectors, the pixels are numbered starting at pixel number 1 in the (-X, -Y) corner. The pixel
numbers increase across the rows, which run in the +X direction. If there are Nx pixels in the X direction, then
pixel number Nx will be located at the end of the first row, which is the (+X, -Y) corner. The next pixel number,
Nx+1, will be the first pixel of the second row, located at (-X, -Y+dY), where dY is the distance between rows (i.e.
the distance between pixels in the y-direction). Pixels continue to be numbered in this manner across successive
rows. The (-X, +Y) corner will be pixel number ((Nx*(Ny-1))+1. The (+X, +Y) corner will be the last pixel, pixel
number (Nx*Ny).
For other detectors, such as polygon objects used as detectors, the pixels are numbered in the order the
polygons are defined, and there is no general way to describe the pixel ordering. For these objects, some
experimentation may be required to determine the exact pixel ordering method.
Optimizing with the IMAE operand
The IMAE operand estimates the efficiency of an optical system by launching many rays into the entrance
pupil; computing the fraction of rays that pass through all surface apertures to any surface. Optimization with this
operand may not proceed smoothly if only hard-edged surface apertures, such as the circular aperture, are used.
This is because ZEMAX estimates the derivative of operand values by making very small differential changes in
the value of each variable, then computes a finite difference of the operand value. For the IMAE operand, a small
change in the value of a variable may not change the efficiency estimate; if no rays are close enough to an
aperture to change from being vignetted to unvignetted or vice-a-versa.
The solution is to replace hard edged apertures with soft-edged apertures placed on a user-defined surface.
A soft-edged aperture has a transmission that is unity over most of the clear aperture, but near the edge the
transmission drops to zero gradually over a small region, rather than abruptly.
Filter functions for doing this are included with ZEMAX as sample DLL files; see the Chapter "Surface Types"
under "User Defined Surface" for details. See in particular the discussion of the US_FILT4.DLL sample.
IMAE uses the current saved settings for the image analysis feature; except for "show" which is always set to
spot diagram for this computation. See Geometric Image Analysis on page 142.
Using gradient index operands
There are several optimization operands which are used to control the properties of gradient index materials
during optimization. Some of them are described below.
DLTN
DLTN is used to control the maximum total change in index within a gradient index lens. Int1 is used to define
the surface number, and Int2 is used to define the wavelength number. DLTN is defined as:
DLTN = n max n min .

The min and max index values are computed at the extreme z coordinates, Z min and Z max. Z min and Z
max are the Z coordinates of the minimum and maximum axial positions of the blank used to make the lens,
before the shaping begins. For a convex surface, they correspond to the vertex. For a concave surface, they
correspond to the maximum sag at that surface.
LPTD
LPTD is used to control the profile of the gradient within the material. Only the Int1 value is used to define the
surface number of the gradient index surface. LPTD is an acronym for LightPath Technology Delta, and the
constraint is used to keep a nonlinear profile monotonically increasing or decreasing. It only needs to be used
when the quadratic or cubic term of the axial gradient is variable. This operand only affects GRIN 5 surface types.
The LPTD operand should be used with a target of 0. The boundary constraint enforces the following
conditions:
n - n n n
------------ > 0 and ------------- > 0, or ------------- < 0 and ------------- < 0 .
z min z max z min z max
Z min and Z max are the Z coordinates of the minimum and maximum axial positions of the blank used to make
the lens, before the shaping begins. For a convex surface, they correspond to the vertex. For a concave surface,
they correspond to the maximum sag at that surface. If the residual value of the operand is less than zero, then
the target may be decreased slightly (try 0.1). Changing the target is usually more effective than increasing the
weight. The value of the LPTD operand must be zero for the blank to be fabricated. Always check the gradient
profile to make sure the slope does not change sign.
User defined operands
There are times when very complex calculations need to be performed, and the results of the computation
need to be optimized. ZEMAX supports some of these calculations already, such as the MTFA operand which
traces many rays, computes the MTF, and then returns a single resulting number to the Merit Function Editor
"value" column. Some limited calculations can be performed within the merit function itself; see for example the
discussion in the section Defining complex operands on page 423.
However, there are problems for which only the flexibility of a user defined program is sufficient for defining
the data computed by an operand. There are two ways of achieving this:
1) Through the use of a ZPL macro
2) Through the use of an externally defined and compiled program
The use of ZPL macros is simpler, well integrated with ZEMAX, and requires very little programming
experience. However, it is limited by the capabilities of the ZPL macro language, and ZPL macros are interpreted,
which means slow execution for complex calculations. ZPL macro optimization is generally a better choice for
simpler macros that execute fairly quickly.
Externally defined programs are more complex to program, require an external C or other language compiler,
and at least some programming experience. However, externally defined programs can be vastly more complex
than what is supported by the ZPL macro language, and since external programs are compiled, they run
significantly faster. The speed difference can be dramatic, and generally more complex calculations will benefit
more from being externally compiled. Indeed, externally defined operand programs may be very complex, tracing
millions of rays or doing other lengthy calculations before returning control to ZEMAX. Note this interface can be
used to optimize lenses based upon data computed by other analysis programs.
Both the ZPL and the externally compiled methods of implementing UDO's are described in detail below.
Optimizing with ZPL macros
If the ZPL macro language (see ZEMAX PROGRAMMING LANGUAGE on page 535) is sufficient to perform
the required computations, then the operand ZPLM may be used to call a ZPL macro from within the merit
function. The macro performs the required computations, then returns the result using the ZPL OPTRETURN
keyword.

ZPLM is simple to use. The Int1 and Int2 values are used to specify the macro number and data field number,
respectively. The macro number is used to indicate which ZPL macro should be executed, while the data field
number indicates which value computed by the macro should be optimized.
The macro number must be an integer between 0 and 99. If the Int1 value for a ZPLM operand is set to 17,
for example, then the macro number is 17, and the macro to be executed must be named ZPL17.ZPL. The macro
name must always use a two digit representation of the macro number. If the macro number was 6, then the macro
to be executed would be ZPL06.ZPL. The ZPL macro file must reside in the default directory for ZPL macros; see
Directories on page 59.
The data field number may be any number between 0 and 50, inclusive. This number refers to a position in a
global array associated with the lens in memory. During execution of the macro, the macro keyword OPTRETURN
specifies which data field number stores the results of the macro calculation. There are 51 different data fields,
so that a single macro call can be used to optimize up to 51 different values simultaneously. For example, suppose
you needed a macro which computed the total length of the lens from surface 1 to the image plane (this is in effect
a user-defined version of the TOTR operand). The macro might look like this:
n = nsur()
x = 0
for i = 1, n, 1
x = x + thic(i)
next
optreturn 0 = x
Note the use of the OPTRETURN keyword. This keyword stores the resulting value for "x" in the global array
position 0. Suppose this macro was named ZPL15.ZPL. To optimize the resulting value for x, the ZPLM merit
function operand would be added to the Merit Function Editor, with Int1 = 15 and Int2 = 0. After updating the merit
function, the "value" would be the same as that returned by TOTR, and it can be optimized in the same way.
ZPLM also permits the use of the Hx, Hy, Px, and Py data fields. These data fields can be read by the ZPL
macro using the PVHX, PVHY, PVPX, and PVPX ZPL functions, respectively. "PV" is a mnemonic for "Pass
Value".
There is one very important thing to know about the data field number. If it is zero, then the macro is executed
and the value from OPTRETURN 0 is returned. However, if the data field number is not zero, then the macro is
not executed, but any previous value stored from an earlier call to the macro is used instead. The advantage to
this convention is substantial. If the macro computes many values, all of which need to be optimized, the macro
only needs to be called once, yet multiple ZPLM operands can access the data. This is much more efficient than
calling the macro multiple times.
For example, suppose a macro named ZPL11.ZPL computes three values, all of which require optimization.
In the macro, the values are stored using OPTRETURN:
OPTRETURN 0 = x
OPTRETURN 1 = y
OPTRETURN 2 = z
Then three ZPLM operands in the merit function can extract the data and perform the optimization with a single
call to the macro:
ZPLM 11 0
ZPLM 11 1
ZPLM 11 2
The macro ZPL11.ZPL is only called during the evaluation of the ZPLM 11 0 operand. Note the Hx, Hy, Px,
and Py values can only be used if the Int2 value is zero, since only in this case is the macro evaluated.
Lastly, it is very important that no changes be made to the lens data during the macro execution. These
changes may interfere with the subsequent evaluation of other operands. ZEMAX does not restore the lens being
evaluated to the state it was in prior to the evaluation of the ZPLM specified macro. Also, ZPLM should not be
used in the middle of a default merit function, but should instead be placed either prior to or after the portion of
the merit function that ZEMAX defined by default. If lens data is changed during the operation of the macro,
ZEMAX has no way of knowing what data was altered, and cannot restore the lens to its original unaltered state.
This could be avoided by allowing the ZPL macro to execute only on a copy of the lens being optimized, rather
than the actual lens, however this is not currently supported. The reason is there may be times where a macro

needs to alter the lens data prior to evaluation of subsequent operands. In this case, two macros should be
executed. The first should modify the data as required, and the second should restore the data to the original
condition. Both macros can be listed in the merit function editor, with the intervening operands executing on the
altered lens data.
Optimizing with externally compiled programs
The second method of creating a user defined operand (UDO) is to write an external Windows program which
computes the data, then use Dynamic Data Exchange (DDE) to pass the data to and from ZEMAX. The DDE
interface in ZEMAX is documented and described in ZEMAX EXTENSIONS on page 607. The material
presented there is not duplicated here; this discussion assumes the material in that chapter is understood.
The operand UDOP is used to call an external client program from within the merit function. The client program
performs the required computations, possibly by making multiple DDE calls back to the ZEMAX server, then
returns the result to ZEMAX using the DDE interface. The computed data is then placed in the "value" column of
the Merit Function Editor and thus may be optimized in the usual way.
UDOP is simple to use. The Int1 and Int2 values are used to specify the client program number and data field
number, respectively. The client program number is used to indicate which client program should be executed,
while the data field number indicates which value computed by the client program should be optimized.
The client program number must be an integer between 0 and 99. If the Int1 value for a UDOP operand is set
to 17, for example, then the client program number is 17, and the client program to be executed must be named
UDO17.EXE. The client program name must always use a two digit representation of the client program number.
If the client program number was 6, then the client program to be executed would be UDO06.EXE. The client
program file must reside in the \UDO directory off the main ZEMAX directory.
When reaching a UDOP operand with a data field number of zero (more on the data field number shortly),
ZEMAX will call the client program. The client program is called with the following syntax, assuming the client
program number was 17:
UDO17.EXE buffercode hx hy px py
The buffercode is an integer value provided by ZEMAX to the client that uniquely identifies the correct lens.
Because ZEMAX is capable of evaluating multiple lenses simultaneously, the buffercode is used as an identifier
so that when the client requests or returns data, it is associated with the correct lens. Note that while optimizing
ZEMAX may be evaluating dozens of lenses simultaneously, each slightly different as derivatives are evaluated
and the optimization progresses. The client must compute data for the indicated lens.
Once execution in the client has begun, the client program must do the following critical steps:
1) Establish a DDE link with the ZEMAX server.
2) Load the correct lens in the ZEMAX server's memory.
3) Compute the required data.
4) Pass the data back to ZEMAX.
5) Clear the ZEMAX server's memory.
6) Terminate the DDE link and exit.
The DDE link is typically maintained by the ZCLIENT code, which is described in the DDE chapter (the user
is of course free to write their own, if preferred). ZCLIENT calls the UserFunction defined by the user to compute
the operand data.
To load the correct lens into the ZEMAX server's memory, a single item must be sent to the ZEMAX server,
GetUDOSystem. The syntax is "GetUDOSystem, buffercode". This will cause ZEMAX to retrieve the correct lens
from system memory, and all subsequent DDE calls will be for actions (such as ray tracing) on this lens.
The data is then computed, optionally using any of the DDE item calls defined in the DDE chapter. Once the
data is computed, up to 51 values are placed in a long formatted string. The string is sent back to the server, and
ultimately to the optimizer within ZEMAX, with the SetUDOData item. The syntax is "SetUDOData, buffercode,
data0, data1, data2, data3, ....,data50". Any data items omitted are assumed to be zero. All data items must be
free formatted integer, exponential, or floating point numbers.

It is critical that the client program pass back a string using SetUDOData, even if it only includes the buffercode
with the remaining fields blank. ZEMAX will wait for the client program to pass this string back. ZEMAX has no
way of knowing how long the computation may take; and so ZEMAX will "hang" until it receives the data. If the
client program crashes, or never returns the data, ZEMAX will never complete execution of the operand, and will
hang forever. Pressing the escape key from within ZEMAX will "break" the evaluation and cause ZEMAX to skip
the operand evaluation.
Terminating the link is achieved by the client program UserFunction returning control to ZCLIENT.
A sample code that computes 3 data items, called a, b, and c, might look like this:
void UserFunction(char *szCommandLine)
{
double a, b, c;
char szBuffer[5000], szSub[256];
int buffer_code;
/* get the buffer code that identifies the lens */

buffer_code = atoi(GetString(szCommandLine, 0, szSub));
/* set the correct lens in the server's memory */

sprintf(szBuffer,"GetUDOSystem,%i", buffer_code);
PostRequestMessage(szBuffer, szBuffer);
/* Here is where we compute the data... these lines are omitted */
/* now return the data for dataflag is 0, 1, and 2 */

/* DO NOT use SetUDOData if the total line length will exceed 255 characters, */
/* use SetUDOItem instead. See the chapter on ZEMAX Extensions for details. */
sprintf(szBuffer,"SetUDOData,%i,%.7f,%.7f,%.7f", buffer_code, a, b, c);
PostRequestMessage(szBuffer, szBuffer);
}
Note that multiple data values may be returned with a single call to the client program. This is where the data
field number comes into play. The data field number may be any number between 0 and 50, inclusive. This
number refers to a position in a global array associated with the lens in memory. During execution of the client
program, the client DDE item "SetUDOData" is used to return a long string of 51 numbers (or less; blank or omitted
values are assumed to be zero). These values, which must be numeric, are stored in the lens buffer for later use
by UDOP. In practice, the 255 character limitation of DDE item names limits SetUDOData to passing back far
fewer than 51 items; but the DDE item "SetUDOItem" works around this limitation by passing back data one at a
time.
!There is a limitation of 255 characters in any DDE item name; and so SetUDOData is limited
to passing back a small number of values. To work around this limitation, use SetUDOItem,
described in the chapter "ZEMAX Extensions".
There are 51 different data fields, so that a single client program call can be used to optimize up to 51 different
values simultaneously. The data field number indicates which of the returned values should be placed in the
"value" column for that UDOP operand.
UDOP also permits the use of the Hx, Hy, Px, and Py data fields. These data fields can be read by the client
program, because they are passed on the command line after the buffercode.
There is one very important thing to know about the data field number. If it is zero, then the client program is
executed and the value from data position 0 is placed in the value column. However, if the data field number is
not zero, then the client program is not executed, but any previous value stored from an earlier call to the client
program is used instead. The advantage to this convention is substantial. If the client program computes many
values, all of which need to be optimized, the client program only needs to be called once, yet multiple UDOP
operands can access the data. This is much more efficient than calling the client program multiple times.
For example, suppose a client program named UDO25.EXE computes three values, all of which require
optimization. In the client program, the values are passes back using "SetUDOData,buffercode,x,y,z". Then three

UDOP operands in the merit function can extract the data and perform the optimization with a single call to the
client program:
UDOP 25 0
UDOP 25 1
UDOP 25 2
The client program UDO25.EXE is only called during the evaluation of the UDOP 25 0 operand. Note the Hx,
Hy, Px, and Py values can only be used if the data field value is zero, since only in this case is the client program
evaluated.
Unlike using ZPL macros, UDO's may freely change the lens data during evaluation since all DDE commands
are executed on a copy of the lens, not the actual lens being optimized.
There is a sample UDO source code file called UDO_DEMO.C; it may be compiled and linked in with ZCLIENT.
The executable needs to be renamed UDOxx.EXE where xx is a two digit integer. The sample UDO returns 6
values: Hx, Hy, Px, Py, and two dummy constants, in data positions 0 through 5, respectively.
Suggestions for use
In preliminary design stages, it is rarely required to trace all of the rays for all of the wavelengths at each field
position during optimization. For this reason, execution times may be substantially decreased by limiting the
number of fields and wavelengths used during optimization. If the weight of selected fields and wavelengths is
set to zero, then the default merit function algorithm will skip the zero weighted fields or wavelengths when
constructing the merit function. This results in fewer rays being traced, speeding execution.
For example, if the lens is being evaluated at five field points, it is possible that only the first, third, and fifth
field need be included in the merit function. Of course, later in the design process all fields may need to be
included and the default merit function reconstructed.
There are a few other tricks to improve performance. Avoid setting boundary operands on variables unless the
optimized solution persists on implausible designs. Boundary operands add computational overhead. Use solves
instead of explicit operands whenever possible. For example, use a curvature solve to control the focal length
rather than an operand if possible.
Optimization is inseparable from the art of modern lens design, and only practice will make a designer a
proficient user of optimization algorithms. Users who are expert at other software optimization algorithms will
probably find ZEMAX easier to use, and with a little practice, the mechanics of using the interface will slip into the
subconscious, and the designer can concentrate upon the design itself. If you are new to computerized
optimization of lenses, there is no better way to learn than to practice.
The global optimum
The design that yields the lowest possible value of the merit function is called the "global" optimum and is by
definition the best possible design. However, there is no known optimization algorithm that can universally find
the global optimum for an arbitrary design problem, unless you consider "direct search" an optimization algorithm
(in other words, try all of the infinite number of possible solutions to see which is best). The art of optical design
with computer assistance has two basic components. First, the designer must be able to determine a suitable
starting design, and second, he or she must play the role of supervisor during the optimization process. A good
supervisor knows when and how to back up and coax the program into a more fruitful direction.
Unfortunately, this often requires considerable experience, and even more often, excessive tedium. An
experienced designer uses a combination of intuition, analysis, and luck in searching for new, better, design
forms. ZEMAX provides an automated capability for performing this global optimum search; the feature is
described in the next chapter.

Chapter 15 GLOBAL OPTIMIZATION
Introduction
!The material presented in this chapter critically depends upon the user having read and
understood the material presented in the previous chapter on "Optimization".
The conventional means of optimizing lenses has for decades been the use of the damped least squares
algorithm (DLS). DLS has many attractive features; it is efficient, and it is very good at finding the "local" minimum
of the merit function. In this context, the word local means the lowest value of the merit function that can be
reached from the current position in solution space without ever increasing the merit function (this is something
of an idealization, in reality DLS can hop over small regions of increased merit function).
To visualize this problem, imagine you are on a hike, and are trying to find the bottom of a valley from a starting
point on the side of a hill. You would like to find the lowest point in the valley. Suppose you cannot see the valley;
it is very foggy and all you can see is the terrain very close to where you now stand. You can determine which
way is downhill, and proceed in that direction until the slope becomes uphill again; at which point you find a new
downhill direction. You might repeat this procedure over and over until a point is reached where all directions are
uphill. This lowest point is then a local minimum, and the bottom of at least this valley.
The problem with this method is that once you have arrived at the local minimum, there is no known way (for
the general optimization problem) to determine if there is not a better, lower minimum somewhere else. For
example, if you were to walk uphill in any direction from this point until you reached a local peak, and then were
to proceed onward downhill into the next valley, you will eventually reach a new local minimum. Is this new
minimum lower or higher than the previous valley? The only way to find out is to take the hike!
You might ask, since computers are so fast, why not just try out every possible configuration to see which is
best? To get a feel for the scope of the problem, consider a cemented doublet lens with six degrees of freedom
(degrees of freedom manifest themselves as variables for optimization). If you assume that each variable can
take on 100 possible values (a coarse sampling), then there are 1E+12 different possible systems. If each system
evaluation requires the tracing of 20 rays (a low estimate) and you can trace 1,000,000 rays per surface per
second, then the time required is about 8E+07 seconds, or about 2.5 years. For a four-element lens (16 variables)
evaluated at three fields and three wavelengths, using 100 rays for evaluation would require 1E+32 system
evaluations or many billions of times the age of the universe.
There are approaches to the global optimization problem that (thankfully) do not require unreasonable
computational effort. These algorithms include simulated annealing, multistart, expert systems, neural networks,
and others. All of these algorithms have strengths and weaknesses which are beyond the scope of this chapter.
Capabilities of ZEMAX
There are two separate global optimization algorithms in ZEMAX, each with a different purpose. The first
algorithm you are likely to use is called "Global Search", and it is used to find new design forms given only the
merit function and a starting design. Global search uses a combination of genetic algorithms, multistart,
conventional damped least squares, and some expert system heuristics to search for new design forms. The
Global Search algorithm is very good at finding promising design forms, however, it does not usually produce
"finished" designs. The second algorithm is used for this purpose.
The second algorithm is called "Hammer" optimization (lens designers often talk about hammering on a design
to squeeze out the last bit of performance). The Hammer algorithm is used for exhaustively searching for the
optimum solution once a reasonably good starting point is found, presumably by prior experience or the Global
Search algorithm. The Hammer algorithm only requires a partially optimized lens and merit function in the form
of a ZMX file.
Although the global optimization algorithms are extremely useful, it is important to realize that there is no
guarantee that the true global optimum will always, or even occasionally be found. Of course, there is no way to
even determine if any solution is the global optimum, even if it is the best you have ever found (remember the
universe-lifetimes scale of the problem).
435 Chapter 15: GLOBAL OPTIMIZATION

!It is important to realize that there is no guarantee that the true global optimum will always,
or even occasionally be found.
Both the Global Search and Hammer algorithms require extensive computational effort to be effective. These
algorithms are not intended to be used interactively! (that is what DLS optimization is for). If you set up global
optimization, and watch while the computer works, you are bound to be disappointed. Global optimization is highly
effective when you set the problem up and let the computer run for many hours, or even several days, but not for
ten minutes. The ideal situation is to set up the problem before stopping work for the night, and let Global Search
(or Hammer, depending upon your requirements) work overnight. In the morning, you should have useful results
to look at.
The Global Search algorithm
Before you begin the Global Search, you must come up with a very rough starting point. "Very rough" means
the design has the correct number of surfaces, a defined stop surface, and initial glasses selected. The fields and
wavelengths must be defined. You also need to define a merit function; see OPTIMIZATION on page 385 for
details on this procedure. The very rough design can be parallel plates of glass with a curvature solve on the last
surface to control the focal length. If a solve is not used to control the focal length, then the system should have
at least the approximate focal length desired. Also, the variable parameters must all be defined. The lens must
be saved before initiating Global Search. ZEMAX uses the starting focal length as a scaling parameter, so the
initial design should have at least the approximately correct focal length!
From the main screen, select Tools, Global Search. There are four buttons on the dialog box labeled: Start,
Stop, Resume, and Exit. There is also an option to select the number of "best" files to save. Select Start. ZEMAX
copies the initial file into new ZEMAX format files called GLOPT_001.ZMX through GLOPT_nnn.ZMX where nnn
is the maximum number of files to save. ZEMAX will then begin looking at various combinations of lens
parameters extracted from the ranges you have defined. The optimization will proceed on a newly generated lens
until ZEMAX concludes the new lens has been sufficiently optimized.
As each new lens is generated, ZEMAX will compare the merit function of the new lens to the best lenses
found so far, and will place it in the correct location in the best lenses list, renaming other lens files as required.
If the lens has a higher merit function than all of the lenses on the best list, then it is discarded. The cycle repeats
indefinitely. Each time a new lens is found which is better than the worst in the list of best lenses, it is placed in
the correct place in the list. After hundreds of lenses have been replaced (which may require many tens of
thousands of lenses to be evaluated) the resulting set will hopefully contain some very good designs, or at least
some promising forms. The global search dialog box will also display the merit functions of the ten best lenses
found so far. If the number of lenses to save is greater than ten, these files are stored on disk, but the merit
functions are not displayed.
The algorithm also periodically returns to lenses in the best list to see if they can be improved upon.
Occasionally some lenses will be improved and placed back in the list. If this happens, the older design being
replaced is rejected if it has the same basic form as the new lens. This is done to keep some diversity in the best
list, otherwise all the lenses will be of nearly identical form.
To terminate the search, select Stop. Depending upon what the algorithm is doing, it may exit immediately, or
it may require several seconds. Once the algorithm has terminated, you can click on Exit. You can now open any
of the GLOPT_xxx.ZMX files for further analysis.
The Resume button is very similar to the Start button, however, Resume will first load the existing GLOPT_xxx
files and place their current merit functions in the best list. Thus, Resume begins the search from where a previous
run ended. Resume does not erase the existing best files, whereas Start erases the files and begins the search
anew entirely based upon the lens currently in the Lens Data Editor. If Resume is selected when a completely
unrelated file is in the Lens Data Editor, ZEMAX will attempt to optimize the lens using the current data in the
Lens Data and Merit Function Editors, but will use the old GLOPT files respective merit functions for comparison
purposes.
Global search will rarely find the global optimum by itself. The reason is that the Global Search concentrates
the effort on finding new, promising design forms rather than converging exactly on the best possible solution for
each form. This latter job is left to a separate algorithm, called "Hammer Optimization", described in the following
section.
Chapter 15: GLOBAL OPTIMIZATION 436

The Hammer algorithm
After reviewing the design forms generated by Global Search, you will probably want to investigate one or two
of them. The one with the lowest merit function is not always the best (although it should be if you designed your
merit function well). For example, the second-best solution may be easier to fabricate. Whatever criterion you use
to determine the most promising solution, you now want to find the best design possible using this selected lens
as a starting point.
The following controls are on the Hammer dialog box:
Hammer: Begins the Hammer optimization loop. Hammer optimization will take the lens and exhaustively
attempt to refine it by making adjustments and optimizations. Each time the lens is improved, it will be saved
to disk in a temporary file.
Auto DLS: Calls the conventional (local) damped least squares (DLS) optimizer and runs it in Automatic mode.
Sometimes it is useful to optimize a lens before calling Hammer if the lens has not already been optimized.
Stop: Terminates the Hammer search.
Exit: Closes the dialog box.
Auto Update: If checked, all open windows will be updated every time a lower merit function is found.
#CPU: The number of CPUs to use in implementing the Hammer search.
Hammer optimization only requires a ZMX file with variables and a merit function to get started. The Hammer
optimization screen shows the starting merit function, and the best merit function found so far. Although good
results can occasionally be had in several minutes, the algorithm should be allowed to run for several hours, and
preferably overnight. To terminate the search, select Stop, then Exit.
If ZEMAX abnormally terminates, the last saved Hammer file can be found in the temporary file. The temporary
file name is constructed from the starting lens file name. If the lens being optimized is stored in the file
C:\ZEMAX\SAMPLES\MYFILE.ZMX
then the temporary file will be called
C:\ZEMAX\SAMPLES\MYFILE_HAMMER.ZMX.
The Hammer algorithm can also be used effectively on partially optimized designs that were not generated
with Global Search. Feel free to use the Hammer on any design!
Optimizing glass selection
If a glass is made into a "model" glass described by the index, Abbe, and partial dispersion deviation, then the
parameters to the model may be made variable and optimized just like any other numerical parameters. However,
the model glass method has one serious drawback. After a good solution is found using model glasses, a
conversion must be made back from the model glass to a real glass. The design must then be reoptimized using
the new glass selection.
Unfortunately, for many systems the newly optimized design will perform worse than the model glass design.
Even more frustrating is that the optimal design using real glasses may have a different form than that found using
the model glass.
Traditionally, to find a better combination of glass types, the designer would locate alternate glasses on the
glass map, substitute in the new selections, and reoptimize. If the new solution was better, the glass choices
would be retained; otherwise, a new set of glasses would be evaluated. The procedure would continue as long
as the designer could bear to continue the search.
Using glass substitution
ZEMAX automates this procedure by allowing glasses to have a "substitute" status associated with them. If a
glass is marked (using the glass solve dialog box) as a substitute glass, then the global optimization algorithms
(both Hammer and Global Search) automatically perform iterative substitution of similar glass types during
optimization. This allows ZEMAX to optimize not only the numerical prescription values such as radius and
thickness, but also allows direct optimization of real glass selection, without resorting to idealizing the glass
dispersion.

To use the substitution feature, set the status of each glass free to change to "substitute" on the glass solve
dialog box (double clicking on any glass name is the quickest way to reach this dialog box).
Once the glass substitution status is defined, then invoke either the Hammer or Global Search optimizers.
ZEMAX will automatically alter the glass types during the search for better designs. By default, ZEMAX will
choose any glass from any of the currently active catalogs, which are specified on the System, General dialog box.
Restricting selected glasses
In practice, it is usually necessary to restrict the available glasses for substitution for these reasons:
Not all "glasses" in the catalogs are actually glasses; some are liquids, gases, crystals, or plastics that may
not be acceptable for use.
Many glasses are very expensive, heavy, brittle, or have other undesirable mechanical properties.
Perhaps only certain catalogs are desired for the newly chosen glasses; while other catalogs must still be used
for other surfaces in the optical system.
For these reasons, ZEMAX provides several optional methods to restrict the pool of available glasses used
for substitution:
By defining a specific catalog only be used for any given surface.
By defining a "template" which places limits on the cost, acid resistance, stain resistance, and other properties
of substituted glasses.
By "excluding" glasses on a case-by-case basis in the glass catalog.
By defining "penalties" in the merit function which discourage selection of glasses with undesirable properties.
The glass solve dialog box permits specification of a catalog name to use for candidate substituted glasses.
If no catalog name is given (i.e. the catalog field is blank) then glasses may be selected from all catalogs selected
for use on the System, General dialog box. If a catalog name is given (i.e. Hoya) then only glasses from that one
catalog will be selected. This allows different surfaces to select from different catalogs, if required. It is also useful
for restricting the number of glasses considered for substitution in a very general way. Note that user defined
catalogs may be specified. No extension should be specified, ZEMAX automatically appends the .AGF extension.
A Glass Substitution Template may be defined to restrict which glasses from all catalogs are chosen for
substitution based upon their cost, AR, SR, FR, CR, and PR code values. See Glass Substitution Template on
page 195 for information on defining a template.
To prevent only certain glasses from being chosen during optimization, choose "Exclude Substitution" for the
glass that should be avoided on the glass catalog dialog box; see the Chapter "Using Glass Catalogs" for details.
The advantage in excluding glasses is that other glasses in the catalog may still be used, without having to define
a separate catalog.
Penalties are operands in the merit function such as GCOS, GTCE, and INDX which are used to increase the
merit function if glasses have unacceptable properties. This is the least efficient method, because glasses may
still be selected, and the subsequent lens optimized, even if the resulting merit function ends up being too high
to be considered a good solution. However, it is useful for defining relationships between glasses, such as
minimizing the difference of GTCE between two glasses which form a cemented surface (to prevent the lens from
breaking under thermal stress).
Suggestions for use
There are several techniques for maximizing the performance of the global optimization algorithms (both
Global Search and Hammer):
1) If possible, place the stop on the first surface. If your entrance pupil is imbedded in the system, you can
model this by using a dummy first surface for the stop and then use a negative thickness to get to surface 2.
This enhances performance because ZEMAX does not need to calculate where the entrance pupil is. You can
make this thickness variable if appropriate. This technique does not work well for systems which inherently
have entrance pupil distortion, such as wide angle lenses.
2) Use a marginal ray angle solve on the last glass surface curvature to control the effective focal length instead
of the EFFL operand. Dropping one variable decreases the dimensionality of the problem significantly, and the

EFFL operator does require one additional ray trace, slowing down the merit function evaluation.
3) Use a marginal ray height solve (for zero ray height) on the last thickness before the image plane. Most
lenses are well corrected for the 0.7 pupil zone on axis. You may want to use another pupil zone if your intuition
guides you to. This solve will ensure that every design generated by Global Search is in focus, just as the
curvature solve ensures the correct focal length. These two solves together can improve the performance of
the Global Search algorithm by several orders of magnitude. For Hammer optimization, replace the solve with
a variable to allow for optimal defocus.
4) Use the MNCT and MNET operands. These are essential for avoiding negative center and edge thickness-
es. ZEMAX uses these boundary constraints to determine the appropriate range of each variable in the search.
The solutions will wander into these unacceptable regions of solution space unless you specifically forbid it.
5) Keep your merit function as simple as possible. Using 2 or 3 rings in the default merit function construction
(see the chapter "Optimization") is often a good idea. You can always go to more rings for the Hammer
optimization.
6) Be willing to run several long runs; perhaps one before starting a design problem to get ideas for design
forms, a few more to explore various promising designs, and a last run to ensure the design you have settled
on cannot be improved (Hammer is particularly good at this last task).
7) Use substitute glasses rather than model glasses, especially when using Hammer. There is not much point
in using Hammer on a design that uses model glasses, because ultimately the glasses will need to be
converted back to real glasses and then optimized again. Substitute glasses are highly useful, especially late
in the design process when the design form has been determined.
8) You may want to dim or turn off your monitor, or use a screen saver to protect the phosphor from burn-in if
you plan on running the search for many days.
9) The Hammer algorithm may be terminated and then restarted without significant loss of information. As the
algorithm proceeds, it automatically saves any improved designs.
10) If a Global Search is terminated, the search may be resumed at a later time by choosing resume instead
of start.
11) While the optimizer is running, it will dominate the computers resources. Other applications will still run,
but they will be less responsive, Under Windows NT/2000, the resources given to ZEMAX may be reduced by
opening the Task Manager; then choosing the ZEMAX.EXE Process; right mouse click; then choose Set
Priority -> LOW.
12) Some computes may be programmed to go into "sleep" mode if there is no user activity for a set amount
of time. This feature may need to be disabled to allow ZEMAX to continue running, even if there is no input
from the keyboard or mouse for long periods.
Summary
Finally, it is again stressed that the global optimum search feature may not ever find the global optimum to any
particular problem. It is very good at finding alternative design forms that would have been tedious to discover by
hand. It is a very powerful tool to add to your toolbox; it is not a substitute for optical design skills.
The Global Search algorithm does have a strong random component, and therefore no two runs will yield the
exact same solutions every time. Sometimes the solutions will be worse, sometimes better, but usually they are
just different for runs of similar duration.

Chapter 16 TOLERANCING
Introduction
ZEMAX provides a flexible and powerful tolerance development and sensitivity analysis capability. The
tolerances available for analysis include variations in construction parameters such as curvature, thickness,
position, index of refraction, Abbe number, aspheric constants, and much more. ZEMAX also supports analysis
of decentration of surfaces and lens groups, tilts of surfaces or lens groups about any arbitrary point, irregularity
of surface shape, and variations in the values of any of the parameter or extra data. Since the parameter and
extra data terms may describe aspheric coefficients, gradient index coefficients, and more, any of these values
may also be made part of the tolerance analysis. The various tolerances may be used in any combination to
estimate alignment and fabrication error effects on system performance.
!ZEMAX always uses exact ray tracing for tolerance analysis; there are no approximations or
extrapolations of first order results in the ZEMAX tolerance algorithms.
Tolerances are defined using operands, such as TRAD, which defines a tolerance on a radius. The tolerance
operands are automatically saved with the lens file. Tolerance operands are edited on the Tolerance Data Editor
available from the Editors group on the main menu bar.
Tolerances may be evaluated by several different criteria, including RMS spot radius, RMS wavefront error,
MTF response, boresight error, user defined merit function, or a script which defines a complex alignment and
evaluation procedure. Additionally, compensators may be defined to model allowable adjustments made to the
lens after fabrication. ZEMAX also allows limits to be placed on the change of a compensator.
Tolerances may be computed and analyzed three ways:
Sensitivity Analysis: For a given set of tolerances, the change in the criteria is determined for each tolerance
individually. Optionally, the criteria for each field and configuration individually may be computed.
Inverse Sensitivity: For a given permissible change in criteria, the limit for each tolerance is individually
computed. Inverse sensitivity may be computed by placing a limit on the change in the criteria from nominal,
or by a limit on the criteria directly. The criteria may be computed as an average over all fields and
configurations, or on each field in each configuration individually.
Monte Carlo Analysis: The sensitivity and inverse sensitivity analysis considers the effects on system
performance for each tolerance individually. The aggregate performance is estimated by a root-sum-square
calculation. As an alternative way of estimating aggregate effects of all tolerances, a Monte Carlo simulation
is provided. This simulation generates a series of random lenses which meets the specified tolerances, then
evaluates the criteria. No approximations are made other than the range and magnitude of defects considered.
By considering all applicable tolerances simultaneously and exactly, highly accurate simulation of expected
performance is possible. The Monte Carlo simulation can generate any number of designs, using normal,
uniform, parabolic, or user defined statistics.
The basic procedure
Tolerancing a lens consists of these steps:
1) Define an appropriate set of tolerances for the lens. Usually, the default tolerance generation feature
described in this chapter is a good place to start. Tolerances are defined and modified on the Tolerance Data
Editor, available on the Editors menu of the main menu bar.
2) Modify the default tolerances or add new ones to suit the system requirements.
3) Add compensators and set allowable ranges for the compensators. The default compensator is the back
focal distance, which controls the position of the image surface. Other compensators, such as image plane tilt,
may be defined. There is no limit to the number of compensators that may be defined.
4) Select an appropriate criteria, such as RMS spot radius, wavefront error, MTF, or boresight error. More
complex criteria may be defined using a user defined merit function, or for comprehensive flexibility, a tolerance
script, which is described later.
441 Chapter 16: TOLERANCING

5) Select the desired mode, either sensitivity or inverse sensitivity. For inverse sensitivity, choose criteria limits
or increments, and whether to use averages or compute each field individually.
6) Perform an analysis of the tolerances.
7) Review the data generated by the tolerance analysis, and consider the budgeting of tolerances. If required,
modify the tolerances and repeat the analysis.
Details on this basic procedure will be provided in subsequent sections.
Tolerance operands
A tolerance operand has a four letter mnemonic, such as TRAD for Tolerance Radius. Three integer values,
abbreviated Int1, Int2, and Int3, are associated with the mnemonic to identify the surface or surfaces of the lens
to which the tolerance applies. Some tolerance operands use the Int numbers for purposes other than defining
surface numbers as indicated in the following table.
Each tolerance operand also has a minimum and maximum value. These values refer to the maximum
acceptable change from the nominal value. Each operand also has space for an optional comment to make the
tolerance set easier to read. The available tolerance operands are listed in the following table, and are described
in detail below.
TOLERANCE OPERANDS
Name Int1 Int2 Int3 Description Page

Surface Tolerances
TRAD Surf # - - Tolerance on radius in lens units 443
TCUR Surf # - - Tolerance on curvature in inverse lens 444
units
TFRN Surf # - - Tolerance on radius in fringes 444
TTHI Surf # Adjust # - Tolerance on thickness or position in 444
lens units, with optional adjuster.
TCON Surf # - - To l e r a n c e o n conic constant 444
(dimensionless)
TSDX Surf # - - Tolerance on Standard surface x- 445
decenter in lens units
TSDY Surf # - - Tolerance on Standard surface y- 445
decenter in lens units
TSTX Surf # - - Tolerance on Standard surface tilt (TIR) 445
in x in degrees
TSTY Surf # - - Tolerance on Standard surface tilt (TIR) 445
in y in degrees
TIRX Surf # - - Tolerance on Standard surface tilt (TIR) 445
in x in lens units
TIRY Surf # - - Tolerance on Standard surface tilt (TIR) 445
in y in lens units
TIRR Surf # - - To l e ra n c e o n Sta n d a r d s u r f a c e 445
irregularity
TEXI Surf # # of Terms - Tolerance on surface irregularity using 446
Fringe Zernikes
TEZI Surf # # of Terms - Tolerance on surface irregularity using 446
Standard Zernikes
Chapter 16: TOLERANCING 442

Name Int1 Int2 Int3 Description Page
TPAR Surf # Parameter - Tolerance on parameter value of 448
# surface
TEDV Surf # Extra Data - Tolerance on extra data value of surface 448
#
TIND Surf # - - Tolerance on index of refraction at d 448
light (see notes).
TABB Surf # - - Tolerance on Abbe Vd number (see 448
notes)
TCMU Surf # Layer # - Tolerance on coating multiplier 448
Element Tolerances
TEDX First Surf Last Surf - Tolerance on element x- decenter in 448
lens units
TEDY First Surf Last Surf - Tolerance on element y- decenter in 448
lens units
TETX First Surf Last Surf - Tolerance on element tilt about x in 449
degrees
TETY First Surf Last Surf - Tolerance on element tilt about y in 449
degrees
TETZ First Surf Last Surf - Tolerance on element tilt about z in 449
degrees
User Defined Tolerances
TUDX Surf # - - Tolerance on user defined decenter x 449
TUDY Surf # - - Tolerance on user defined decenter y 449
TUTX Surf # - - Tolerance on user defined tilt x 449
TUTY Surf # - - Tolerance on user defined tilt y 449
TUTZ Surf # - - Tolerance on user defined tilt z 449
Non-Sequential Component Tolerances
TNPS Surf # Object # Data Tolerance on NSC object position. The 449
data code is 1-6 for x, y, z, tilt x, tilt y, tilt
z, respectively
TNPA Surf # Object # Parameter Tolerance on NSC object parameter 449
Multi-Configuration Value Tolerances
TMCO Row # Config# - Tolerance on multi-configuration editor 450
value
For each tolerance, a minimum and a maximum value are specified on the Tolerance Data Editor. The
tolerances are described in detail below.
TRAD: Tolerance on radius
Used to tolerance directly on the radius of curvature. Min and max are the extreme errors in lens units.

For example, if the nominal radius of a surface is 100 mm, and the min and max TRAD values for that surface
are -.50 mm and +.50 mm, then the tolerance analysis will be performed with the radius of the surface set to 99.50
mm and 100.50 mm.
TCUR: Tolerance on curvature
Used to tolerance in units of curvature, which is directly related to power. Min and max are the extreme errors
in inverse lens units.
For example, if the nominal radius of a surface is 100 mm, then the nominal curvature is 0.01 inverse mm. If
the min and max TCUR values for that surface are -.001 and +.001 inverse mm, then the tolerance analysis will
be performed with the radius of the surface set to 111.11 mm and 90.909 mm.
TFRN: Tolerance on fringes
Fringe tolerances are very useful when tolerancing flat or large radii surfaces. Min and max are the extreme
errors in (dimensionless) fringes. The TWAV operand is used to define the test wavelength. The change in the
sag of a surface for small changes in curvature is given approximately by
2
r
Z = ---- C ,
2
where r is the semi-diameter of the surface. The change in the sag is related to the error in fringes by

Z = --- N ,
2
where N is the number of fringes. The factor of one half assumes a double pass Newton's rings type test. For
more information see Malacara, Optical Shop Testing.
TTHI: Tolerance on thickness
TTHI is used to tolerance both absolute positions of elements as well as thicknesses of lenses within element
groups.
When ZEMAX creates the default tolerances, it is assumed that all variations in thickness affect only that
surface and any surfaces in contact with that element. For example, if the first lens in a contact doublet has a
+1.0 mm change in thickness, the front and rear vertex of the second lens both shift by +1.0 mm.
However, since ZEMAX defines the position of all surfaces by using an offset from the previous surface, adding
1.0 mm to the surface will shift all subsequent lenses in the system by +1.0 mm. What is more likely to occur in
fabrication is that the +1.0 mm offset would be absorbed by the first air space after the lens group. TTHI can
handle this case by allowing a "adjustment" surface to be specified. When ZEMAX creates the default tolerances,
the adjustment surface is specified as the first air space which follows the surface being toleranced.
To illustrate, imagine a lens where surface 3 was made of BK7, and surface 4 was made of F2, and surface 5
was air. The nominal thicknesses are 3, 4, and 6 mm, respectively. If a TTHI operand was defined by the default
tolerance algorithm for surface 3, an adjustment would be defined for surface 5. If the tolerance value was +.1
mm, then during analysis the thicknesses would be changed to 3.1, 4.0, and 5.9, respectively. Thus, the absolute
positions of surfaces 6 through the image surface are unaffected by the change in thickness on surface 3.
The adjustment is optional; to disable it, set the adjustment to the same surface number as the tolerance, such
as TTHI 3 3. For some lens systems, such as those that are assembled by stacking spacers in a tube, the
adjustment may not be desired.
Int1 is used to define the surface number, Int2 in the adjustment surface number, unless Int2 is equal to Int1.
Min and max are the extreme errors in lens units.
TCON: Tolerance on conic
TCON is used to define a tolerance on a conic constant. Min and max are the extreme errors, dimensionless.

TSDX, TSDY: Tolerance on surface decenters
TSDX and TSDY are used to analyze decentration in X and Y respectively of a Standard surface type. The
Int1 value indicates the number of the surface, and this surface must be a Standard surface type. Surfaces other
than Standard surface types may be toleranced using the TEDX and TEDY operands described later. The min
and max values are the decentering in lens units.
The analysis of TSDX and TSDY uses the Irregular surface type. See the discussion Tolerancing with the
Irregular surface type on page 450.
TSTX, TSTY: Tolerance on surface tilts
TSTX and TSTY are used to analyze tilts of a Standard surface about the X and Y axes, respectively. The Int1
value indicates the number of the surface, and this surface must be a Standard surface type. Surfaces other than
Standard surface types may be toleranced using the TETX and TETY operands described later.
The min and max values are the tilt in degrees about the local X and Y axis of the lens.
See the description of the related TIR operands TIRX and TIRY.
The analysis of TSTX and TSTY uses the Irregular surface type. See the discussion Tolerancing with the
TIRX, TIRY: Tolerance on surface TIR
TIRX and TIRY are used to analyze tilts of a Standard surface along the X and Y axes, respectively. The Int1
value indicates the number of the surface, and this surface must be a Standard surface type. Surfaces other than
Standard surface types may be toleranced using the TETX and TETY operands described later.
TIRX and TIRY are used to specify a tolerance on total indicator runout, or TIR, which measures the amount
of "wedge" in a lens.
The min and max values are twice the amount of "sag" in lens units measured at the maximum radial aperture
of the surface where the maximum radial aperture is defined by the semi-diameter of the surface. The change is
sag as a function of the x or y normalized coordinate for a TIRX or TIRY value is given by:
TIRX TIRY
Z x = ------------- x ; Z y = ------------- y .
2 2
For example, if the TIRX tolerance is 0.10 mm, then the change in sag at the maximum +x aperture of the lens
is 0.05 mm, and the deviation at the maximum -x aperture is -.05 mm, for a "total" TIR of 0.10 mm. A similar
discussion applies to TIRY. The min and max values are used to model the tilt of the surface in each direction.
The tilt angle that is actually placed on the surface is given by
1 Z 1 Z
x = tan --------- ; y = tan --------- ,
y x
S S
where S is the semi-diameter of the surface. Note that a sag along Y implies a rotation about the X axis, and
a sag along X implies a rotation about the Y axis.
The analysis of TIRX and TIRY uses the Irregular surface type. See the discussion Tolerancing with the
TIRR: Tolerance on surface irregularity
!For a more detailed treatment of irregularity, see the discussion on the TEXI and TEZI
operands which follows.
TIRR is used to analyze irregularity of a Standard surface. The Int1 value indicates the number of the surface,
and this surface must be a Standard surface type. Analysis of irregularity on surface types other than Standard
is not directly supported.

Modeling irregularity is somewhat more problematic than other types of tolerances. This is primarily because
irregularity by nature is random, and not deterministic such as a change in radius. Therefore, some assumptions
about the nature of the irregularity need to be made in order to perform the analysis. The assumption ZEMAX
makes when using TIRR is that the irregularity is half spherical aberration, and half astigmatism. This is less
restrictive model than assuming 100% astigmatism, because astigmatism cannot be compensated by focus, and
is therefore a more serious defect in the lens.
The min and max values are the irregularity in units of fringes measured at the maximum radial aperture of the
surface where the maximum radial aperture is defined by the semi-diameter of the surface. The TWAV operand
is used to define the test wavelength.
ZEMAX assumes fringes are measured in a double pass Newton's rings type test. For example, a TIRR of "W"
fringes would yield a change in sag of the surface of
t W 4 2
z = ---------- ( + y ) ,
4
where t is the test wavelength (defined by the TWAV operand), is the normalized radial coordinate, and
y is the normalized radial coordinate in the y direction. The change in wavefront optical path is related to the
change in sag and the index of refraction of the two media the surface separates:
n2
OP = z ----- 1 .
n1
The analysis of TIRR uses the Irregular surface type.

When computing the Monte Carlo analysis, the angle of the astigmatism is chosen randomly between 0 and
360 degrees. This allows simulation of randomly oriented astigmatic error, which is less severe and more realistic
than placing all the astigmatism along the y axis of each element.
See the discussion Tolerancing with the Irregular surface type on page 450 for more information.
TEXI: Tolerance on surface irregularity using the Fringe Zernike model
!See also the discussion for TEZI on page 447. TEZI is a superior alternative to TEXI.
TEXI is used to analyze random irregular deviations of small amplitude on a surface that is either a Standard,
Even Aspheric, or Zernike Fringe Sag surface. The Int1 value indicates the number of the surface, and the Int2
value defines the number of Fringe Zernike terms used to simulate the irregularity on the surface. Analysis of
irregularity on surface types other than Standard, Even Aspheric, or Zernike Fringe Sag is not directly supported.
TEXI uses the Zernike Fringe Sag surface (see Zernike Fringe Sag on page 283) to model the irregularity
rather than using the third order aberration formulas used by TIRR. When using TEXI, the min and max tolerance
values are interpreted to be the approximate magnitude of the zero to peak error of the surface in double-pass
fringes at the test wavelength. The zero to peak is only a very rough measure of the irregularity. Whether the zero
to peak and peak to valley are the same depends upon the particular Zernike term used. The min tolerance value
is automatically set to the negative of the max value; this is done to yield both positive and negative coefficients
on the Zernike Fringe Sag surface. The vagueness of defining irregularity by the zero to peak measure is the
prime disadvantage to TEXI, and the reason why the newer TEZI operand is the preferred alternative.
ZEMAX computes the coefficients on the individual terms of the Zernike polynomial that defines the deviation
in the shape of the surface using the following formula:
f
c = --- ------- ,
2 n

where f is the number of double pass fringes, n is the number of Zernike terms used in the irregularity model,
and is the test wavelength. The coefficients are scaled by one over root n to account for the fact that a random
collection of Zernike terms will generally sum in an RSS sense; so the PTV error is not linear with the number of
terms. Since it is convenient to specify the approximate overall PTV, the terms on each Zernike are computed
according to the formula above. Note there is a "c" value for both the min and the max tolerance in fringes.
For the sensitivity analysis, the surface is converted to a Zernike Fringe Sag surface and all the coefficients of
the Zernike polynomial are set to either the min or the max "c" value from the above equation. Note that since the
surface sag deformation is the same for all polynomial terms at the edge of the aperture, a "c" value of 0.001
when using 20 Zernike terms will yield a maximum sag deviation of 20c.
For the Monte Carlo analysis, the surface is converted as for the sensitivity analysis, but each polynomial term
is assigned a coefficient randomly chosen which lies between the min and max "c" tolerance values. The random
value is chosen using the statistical model selected for the operand; see the STAT command for a discussion.
The default number of Zernike terms is 37, the maximum. Generally speaking, if fewer terms are used, the
irregularity will be of low frequency, with fewer "bumps" across the surface. If more terms are used, there will be
higher frequency irregularity, with more "bumps" across the surface. Note the TIRR irregularity operand models
the lowest frequency form of irregularity, with just a quadratic and quartic deviation across the surface. TEXI can
model much more irregular surfaces, and with 37 terms used, about 5-15 "bumps" will typically be seen over the
surface.
Because the Zernike Fringe Sag surface sag expression contains portions of both the Standard and Even
Aspheric surfaces, either of these surface types may be modeled by the Zernike Fringe Sag surface created with
the TEXI operand. If the surface already is a Zernike Fringe Sag surface, then the deviations are added to the
polynomial terms already there. If the surface is either a Standard or Even Aspheric surface, the normalization
radius of the Zernike Fringe Sag surface is set to the semi-diameter of the surface. If the surface was already a
Zernike Fringe Sag surface, then the min and max tolerances are assumed to be measured at the normalization
radius already defined.
TEXI always ignores Zernike term 0, the piston term, and sets this value to zero.
TEZI: Tolerance on surface irregularity using the Standard Zernike model
!See also the discussion for TEXI on page 446. TEZI is a superior alternative to TEXI.
TEZI is used to analyze random irregular deviations of small amplitude on a surface that is either a Standard
or Even Aspheric surface. The Int1 value indicates the number of the surface, and the Int2 value defines the
number of Standard Zernike terms used to simulate the irregularity on the surface. Analysis of irregularity on
surface types other than Standard or Even Aspheric surfaces is not directly supported.
TEZI uses the Zernike Standard Sag surface (see Zernike Standard Sag on page 285) to model the
irregularity. When using TEZI, the max tolerance value is the exact RMS error of the surface in lens units. The
min tolerance value is automatically set to the negative of the max value; this is done to yield both positive and
negative coefficients on the Zernike Standard Sag surface. The resulting RMS is of course always a positive
number whose magnitude is equal to the max tolerance value.
For the sensitivity analysis, the surface is converted to a Zernike Standard Sag surface and all the coefficients
of the Zernike polynomial for terms greater than #1 (the "piston" term) are set to a value so that the sum of the
squares of the coefficients yields the specified RMS value. All coefficients are set to the same value.
For the Monte Carlo analysis, the surface is converted as for the sensitivity analysis, but each polynomial term
is assigned a coefficient randomly chosen between zero and 1, and the resulting coefficients are then normalized
to yield the exact RMS tolerance. The random value is chosen using the statistical model selected for the operand;
see the STAT command for a discussion.
The number of Zernike terms may be between 0 and 231, the maximum. Generally speaking, if fewer terms
are used, the irregularity will be of low frequency, with fewer "bumps" across the surface. If more terms are used,
there will be higher frequency irregularity, with more "bumps" across the surface. Note the TIRR irregularity
operand models the lowest frequency form of irregularity, with just a quadratic and quartic deviation across the
surface. TEZI can model much more irregular surfaces.

Because the Zernike Fringe Sag surface sag expression contains portions of both the Standard and Even
Aspheric surfaces, either of these surface types may be modeled by the Zernike Fringe Sag surface created with
the TEZI operand. The normalization radius of the Zernike Fringe Sag surface is set to the semi-diameter of the
surface.
TEZI always ignores Zernike term 0, the piston term, and sets this value to zero.
TPAR: Tolerance on parameter data
TPAR is used to tolerance parameter data. The Int1 number is the surface number. The Int2 number indicates
the parameter number. See SURFACE TYPES on page 225 for tables listing which parameter numbers are
used for each surface model. TPAR is useful for determining tolerances on aspheric coefficients, and more. Min
and max are the changes in whatever units are used by the parameter.
TEDV: Tolerance on extra data values
TEDV is the tolerance for extra data values. Extra data values are used by the extra data editor in ZEMAX-EE
to define aspheric terms, diffraction coefficients, or other data. Int1 identifies the surface, Int2 identifies the extra
data number, and min and max are the extreme errors in whatever units are used by the extra data value. See
the Chapter "Surface Types" for tables listing which extra data numbers are used for each surface model.
TIND: Tolerance on index
TIND is the tolerance on index of refraction. Int1 identifies the surface, and min and max are the extreme errors
in index of refraction. The index error is modeled as an "offset" which is independent of wavelength, unless either
the glass of the surface is a "model" glass defined by the index at d-light, Abbe number, and dPgF, or the glass
of the surface is a catalog glass and all defined wavelengths are between 0.3 and 2.5 micrometers. In either of
these latter cases, the TIND is interpreted to be a change in the d-light index, which will change the index at all
wavelengths in a nonlinear fashion.
TABB: Tolerance on Abbe
TABB is the tolerance on the Abbe or Vd number. Int1 identifies the surface, and min and max are the extreme
errors in the Abbe number. If either the glass of the surface is a "model" glass defined by the index at d-light, Abbe
number, and dPgF, or the glass of the surface is a catalog glass and all defined wavelengths are between 0.3 and
2.5 micrometers, TABB is interpreted to be a change in the d-light Abbe number, which will change the index at
all wavelengths in a nonlinear fashion. Otherwise, TABB is ignored.
The change in index is given by an estimate of the differential change in the index at any wavelength as a
function of the d-light index, Abbe, and dPgF.
TCMU: Tolerance on coating multiplier
Coating multipliers are described elsewhere, see Optimizing coatings with ZEMAX on page 508 and Surface
coating tab on page 74. The TCMU operand allows tolerancing of the coating multipliers, which in effect places
a tolerance on the thickness of individual layers in the optical coating on a surface.
TEDX, TEDY: Tolerance on element decenters
TEDX and TEDY are used to analyze decentration in X and Y respectively. The two surface numbers defined
by Int1 and Int2 indicate the beginning and ending surfaces of a lens group. The min and max values are the
decentering in lens units.
These tolerances require ZEMAX to insert a coordinate break surface before and after the group, and then
decenter the entire group as a unit. For this reason, TEDX and TEDY generally should not be used on either side
of an existing coordinate break. If you require this capability, see the discussion on TUDX and TUDY.
TEDX and TEDY may also be used to model decenter of surfaces, like TSDX and TSDY. TEDX and TEDY will
work for surfaces of any type, including Standard surfaces and non-Standard surfaces, while TSDX and TSDY
only work for Standard surfaces. To decenter a single surface using TEDX or TEDY,
set Int1 and Int2 to the same surface number.
To check to see if ZEMAX is actually doing what you want, see the SAVE command discussed in the section
"Trouble shooting the tolerance results".

TEDX and TEDY operands may be nested. For example it is possible to analyze the decentration of one
surface group defined by surfaces 5 through 20; while simultaneously analyzing decenters of elements defined
by surfaces 5-8, 8-12, 14-20, etc. This capability can simulate the alignment errors of elements within an
assembly, as well as the alignment error of the assembly as a whole. The nesting rules are fully described in the
section Nesting rules for Monte Carlo analysis on page 462.
TETX, TETY, TETZ: Tolerance on element tilts
TETX, TETY, and TETZ are used to analyze tilts of either a surface or a lens group about the X, Y, or Z axes,
respectively. The two surface numbers defined by Int1 and Int2 indicate the beginning and ending surfaces of a
lens group. The min and max values are the tipping angles in degrees.
These tolerances require ZEMAX to insert a coordinate break surface before and after the lens group, and to
use a dummy surface at the end of the group to return to the front vertex. The entire group may then be pivoted
about a point as a unit. For this reason, TETX/Y/Z should not be used when the range of surfaces defined includes
a coordinate break which is related to a following coordinate break outside the surface range controlled by the
TETX/Y/Z by pickup solves. The hazard is if the two resulting tilt ranges overlap, then the position of the elements
may not be what is intended. If you require this capability, see the discussion on TUTX, TUTY, and TUTZ.
TETX and TETY may also be used to model tilt of single surfaces, sometimes called "wedge", like TSTX and
TSTY. TETX and TETY will work for surfaces of any type, including Standard surfaces and non-Standard surfaces,
while TSTX and TSTY only work for Standard surfaces. To tilt a single surface using TETX or TETY, set Int1 and
Int2 to the same surface number.
TETX and TETY by default pivot about the front vertex of a lens group, however it is often advantageous to
pivot about some other point. For example, well-designed lens mounts will pivot about the nodal point of a lens
to maintain focus during alignment. This case is easily modeled by ZEMAX through the use of a dummy surface
at the nodal point. Change the starting surf (Int1) to be at the nodal point dummy surface, and the pivoting will be
about that point. The first surface may be located anywhere with respect to the rest of the lens group; and so the
tilt may be about any point.
TETX, TETY, and TETZ operands may be nested. For example it is possible to analyze the tilt of one surface
group defined by surfaces 5 through 20; while simultaneously analyzing tilts of elements defined by surfaces 5-
8, 8-12, 14-20, etc. This capability can simulate the alignment errors of elements within an assembly, as well as
the alignment error of the assembly as a whole. The nesting rules are fully described in the section Nesting rules
for Monte Carlo analysis on page 462.
TOFF: Tolerance off (can be used for comments)
This operand is a place holder that has no affect on the tolerance analysis. The TOFF operand may be used
to enter comments.
TUDX, TUDY, TUTX, TUTY, TUTZ: Tolerance on user defined tilts and decenters
These five tolerances, TUDX, TUDY, TUTX, TUTY, and TUTZ, are used for more general user defined tilts and
decenters. The names are mnemonics for Tolerance User Decenter/Tilt X, Y, and Z. These are very similar to the
TEDX, TEDY, TETX, and TETY tolerances. The difference is that ZEMAX does not automatically insert the
required coordinate break surfaces to achieve the specified decenters and tilts in the tolerances. To use the
TUDX, TUDY, TUTX, TUTY, and TUTZ commands, you must have already defined the surface specified by Int1
to be a coordinate break surface. Normally, but not always, there will be a second coordinate break later in the
Lens Data Editor that has a pickup solve on the toleranced parameter. The pickup solve may be positive or
negative, as the case warrants. This permits some complex pivoting and decentering about arbitrary points, with
a certain finesse required. It is important that the coordinate breaks inserted for use with the TUxx tolerances
have nominal values of zero for all tilts and decenters. Min and max are in lens units for TUDX and TUDY, and in
degrees for TUTX, TUTY, and TUTZ, just like the coordinate break surface.
TNPS, TNPA: Tolerances on non-sequential data
TNPS defines tolerances on Non-Sequential Component (NSC) data. Three integer parameters are required:
the NSC surface number (use 1 if the program mode is NSC), the object number, and a code. The code is 1, 2,
or 3 for the objects X, Y, or Z position, respectively; the code is 4, 5, or 6 for the X-Tilt, Y-Tilt, or Z-Tilt, respectively.

TNPA defines tolerances on NSC parameter data. The surface and object integers are defined in the same
way as for TNPS; the third integer is for the parameter number. See NSC Objects on page 293 for a detailed
explanation of what parameter numbers are used by which object types.
TMCO: Tolerance on multi-configuration data
TMCO specifies the tolerance on any numerical value in the Multi-Configuration Editor. The two integer
arguments are the operand row number and configuration number. This operand is ignored unless all configura-
tions are toleranced at once.
Tolerancing with the Irregular surface type
The tolerance operands TSDX, TSDY, TSTX, TSTY, and TIRR all use the Irregular surface type to model the
perturbation to the lens surface. For a detailed description of the Irregular surface type, see page 262.
The overwhelming advantages of using the Irregular surface type are speed, simplicity, and flexibility. Any
Standard surface type can be converted to an Irregular surface type without the need for dummy surfaces or
coordinate breaks. Furthermore, the aggregate effects of tilt, decenter, and irregularity may all be modeled
simultaneously using the Monte Carlo analysis.
When ZEMAX computes the tolerance analysis using the operands TSDX, TSDY, TSTX, TSTY, or TIRR, the
surface is first converted from a Standard to an Irregular surface type. This is why only Standard surface types
are supported when using these operands.
Tolerance control operands
There are also a few tolerance control operands which can be entered in the Tolerance Data Editor. These
operands are not tolerances, but are used to define compensators, to save intermediate results for further
evaluation, to define statistical properties, and to define the test wavelength for fringe tolerances.
TOLERANCE CONTROL OPERANDS
Name Int1 Int2 Description

CEDV Surface # Extra Data # Sets an extra data value as a compensator.
CMCO Operand # Config # Sets a multi-configuration operand value as a
compensator. See the discussion on CMCO below for
important information.
COMP Surface # Code Sets a compensator. Code is 0 for thickness, 1 for
curvature (1 over the radius), 2 for conic.
CPAR Surface # Parameter # Sets a parameter as a compensator. For a list of
parameters that control surface attributes, see SURFACE
TYPES on page 225.
SAVE File Number - Saves the file used to evaluate the tolerance on the prior
row in the editor. See the discussion below.
SEED Seed # - Seeds the random number generator for Monte Carlo
analysis. Use a value of 0 to choose a random seed.
STAT Type # of standard Sets the type of statistical distribution for randomly
deviations selected Monte Carlo parameter analysis. See the
discussion below.
TWAV - - This operand sets the test wavelength. The "Min" column
is used to edit and display the test wavelength.
The tolerance control operands are described in detail below.

CEDV: Define extra data value compensator
CEDV is used to define the surface number and extra data number of the compensator to use for tolerance
analysis. Int1 is used to specify the surface, and Int2 is used to define the extra data number. For example, to
specify extra data value 17 of surface 9 is a compensator, use CEDV with Int1 = 9, Int2 = 17.
See SURFACE TYPES on page 225 for tables defining which extra data values are used by which surface
types.
General comments about min and max values on compensators
The min and max values of compensator operands such as CEDV, CMCO, COMP, and CPAR indicate the
maximum change allowed for the compensator. For example, if the nominal value of a compensator is 50, and
the min and max values are -1.0 and 1.0, then the compensator will be restricted to lie between 49.0 and 51.0.
This bounding of the compensators is ignored if "merit function" or "user script" is selected as the tolerance
criteria, although the compensators will still be used.
There is no limit to the number of compensators that may be defined.
CMCO: Define multi-configuration operand compensator
CMCO is used to define the operand number and configuration number of the compensator to use for tolerance
analysis. Int1 is used to specify the operand number, and Int2 is used to define the configuration number. For
example, to specify multi-configuration operand 6 in configuration 4 is a compensator, use CMCO with Int1 = 6,
Int2 = 4. The order of the CMCO operands matters. The CMCO operands must be defined in order of operand
number then configuration. All CMCO operands for MCE operand 1 must precede CMCO operands for MCE
operand 2, etc. Within each group of CMCO operands using the same MCE operand number, the CMCO
operands must be listed in order of configuration number.
See MULTI-CONFIGURATIONS on page 471 for information on using multi-configuration operands.
See also General comments about min and max values on compensators on page 451.
COMP: Define compensator
COMP is used to define the surface number and type of compensator to use for tolerance analysis. Int1 is
used to specify the surface, and Int2 is used to define the type. Int2 uses a "code" defined as follows:
Int2 = 0, compensator is thickness
Int2 = 1, compensator is curvature (1 over the radius)
Int 2 = 2, compensator is conic
CPAR: Define parameter compensator
CPAR is used to define the surface number and parameter number of the compensator to use for tolerance
analysis. Int1 is used to specify the surface, and Int2 is used to define the parameter number. For example, to
specify parameter 2 of surface 5 is a compensator, use CPAR with Int1 = 5, Int2 = 2.
See SURFACE TYPES on page 225 for tables defining which parameters are used by which surface types.
SAVE: Save sensitivity analysis lenses
The operation of the tolerance feature is not always transparent. To get a closer look at what the tolerance
routine is actually doing, use the SAVE operand. SAVE is inserted after any tolerance you would like to inspect
in more detail. For example, suppose you had the tolerance operand TEDX in the Tolerance Data Editor. After
reviewing the resulting sensitivities, the results do not appear to make sense. Edit the tolerance operands in the
Tolerance Data Editor to add the SAVE command after the TEDX command. For the Int1 value, enter in 0 or any
other digit. The next time the tolerance analysis is run, ZEMAX will save the file used to compute the TEDX
tolerance in a file named TSAV000n.ZMX where n is the integer number specified in the Int1 column.
The lens file will be saved in the same directory as the current lens. ZEMAX will save the lens used to evaluate
the preceding tolerance in a ZMX format file. Run the tolerance file in the usual way, and then load the filename

you specified. It will show you the prescription data, the coordinate breaks, pickups, variables (which are optimally
adjusted compensators), and merit function used to determine the criteria data (usually for the maximum
tolerance value). This procedure will let you verify the configuration is in fact what you are trying to model, and
will give you insight into how ZEMAX adjusts your lens prescription to determine the tolerances.
SEED: Seed the random number generator
By default, every Monte Carlo analysis will produce new, randomly generated systems. There are times when
it is useful to have multiple Monte Carlo analysis runs produce the same perturbations every time. The SEED
operand may be used for this purpose. If the SEED is any value other than zero, the random number generator
will be seeded with the specified value. This value is automatically incremented for each generated system with
a single Monte Carlo analysis so that each generated system is unique, yet random and reproducible. Using a
value of 0 will select a random seed based upon the system clock, which will generally not be the same each time
the Monte Carlo analysis is run.
The placement of the SEED operand matters. All tolerance operands listed under a SEED operand will be
affected by the seed value provided. Multiple SEED operands may be placed in the tolerance data editor, so that
different operands may use the same seed every time, while others may use random seeds.
STAT: Define statistics
STAT is used to define the statistics "on the fly" during the Monte Carlo analysis. The STAT command takes
two integer arguments. Int1 specifies the statistics type: 0 for normal distribution, 1 for uniform distribution, 2 for
parabolic distribution, and 3 for user defined distribution statistics. The Int2 value is used only by the normal
distribution, and it defines "n", the number of standard deviations between the mean and the tolerance extremes.
The statistics types are defined in detail in Monte Carlo analysis on page 460.
TWAV: Test wavelength
This operand sets the test wavelength. When setting the default tolerances, ZEMAX adds a TWAV operand
with a value of 0.6328 micrometers (HeNe) for the test wavelength. If no TWAV operand is defined, then ZEMAX
defaults to the primary wavelength. More than one TWAV may be placed in the operand list; each operand defines
the test wavelength for the operands that follow. Only operands whose min and max values are measured in
fringes are affected by this setting. The "Min" column in the Tolerance Data Editor is used to edit and display the
test wavelength.
Defining default tolerances
The default tolerances may be defined by selecting Tools, Default Tolerances from the Tolerance Data Editor
menu bar. The Tolerance Data Editor may be activated by selecting Editors, Tolerance Data from the main menu.
The default tolerances dialog box consists of several sections grouped by tolerance type:
Surface tolerances
Radius: If this box is checked, then the default radius tolerances will be included. The default tolerances may
be specified by a fixed distance in lens units or by fringes of power at the test wavelength (defined by the TWAV
operand). This tolerance is only placed on surfaces which have optical power, which excludes dummy surfaces
which have the same index on both sides. If the surface is plano, then the default tolerance is specified as a
variation in fringes, even if another option is selected.
Thickness: If checked, a thickness tolerance is specified on each vertex separation. It is assumed that all
variations in thickness affect only that surface and any surfaces in contact with that element; therefore, the first
air space after the thickness is used as an adjuster. See the detailed discussion in TTHI: Tolerance on
thickness on page 444.
Decenter X/Y: If checked, decenter tolerances are added to each individual lens surface. Tolerances are
defined as a fixed decenter amount in lens units. ZEMAX uses TSDX and TSDY for Standard surface
decenters, and TEDX and TEDY for non-Standard surfaces.
Tilt (TIR) X/Y: If checked, a tilt or "total indicator runout" tolerance in either lens units or degrees is added to
each lens surface. ZEMAX uses TSTX and TSTY for Standard surface tilts in degrees, TIRX and TIRY for
Standard surface total indicator runout (tilt or wedge) in lens units, and TETX and TETY for non-Standard
surface tilts in degrees. If lens units are selected, ZEMAX automatically converts the dimension to degrees of

tilt if required for the TSTX/TSTY or TETX/TETY operands. The conversion from total indicator runout in lens
units to tilt in degrees is given by
y
x = ------ , where
2S
S is the semi-diameter of the surface and is the total indicator runout, with a similar expression for y . This
expression assumes the angles are relatively small. Note that a length of total indicator runout in the x direction
corresponds to a tilt about the y axis, and a length of total indicator runout in the y direction corresponds to a
tilt about the x axis.
S + A Irreg: If checked, a spherical and astigmatism irregularity is specified on each Standard surface type.
For details, see the TIRR description given earlier.
Zern Irreg: If checked, a Zernike irregularity is specified on each Standard surface type. For details, see the
TEXI description given earlier.
Index: TIND is used to model changes in the index of refraction. The units are change in relative index.
Abbe: TABB is used to model changes in the Abbe number. The units are percentage of the catalog Abbe
value.
Element tolerances
Decenter X/Y: If checked, decenter tolerances are added to each lens group. Tolerances may be defined as
a fixed decenter amount in lens units.
Tilt X/Y: If checked, a tilt tolerance in degrees is added to each lens group and surface. It is important to note
that lens groups are by default tipped about the vertex of the first surface in the group. See the section TETX,
TETY, TETZ: Tolerance on element tilts on page 449 for information on tipping about some other point.
In addition to the tolerance definitions there are two other options on the dialog box:
Start At Row: This control indicates where in the Tolerance Data Editor the default tolerances should be placed.
If the row number is greater than 1, then the new default tolerances will be appended starting at the specified
row number.
Use Focus Comp: If checked, then a default compensator of the back focus (the thickness prior to the image
surface) will be defined. Using at least one compensator can greatly relax certain tolerances, however whether
or not compensators should be used depends upon the specifics of the design. Other compensators may be
defined. See the Section "Defining compensators" for more information.
There are also six buttons:
OK: Accept these settings and generate the default tolerances.
Cancel: Close the dialog box without changing the default tolerances.
Save: Save these settings for future use.
Load: Restore the previously saved settings.
Reset: Restore the settings to the default values.
Help: Invoke the help system.
By default, the Monte Carlo analysis that ZEMAX performs draws random values from a Gaussian "normal"
distribution. Once the default tolerances are defined, they are saved with the lens file automatically. If additional
surfaces are inserted in the Lens Data Editor, the tolerance surfaces are renumbered automatically.
Defining compensators
Many different types of compensators may be defined; thicknesses (most commonly used), curvature, conic
constants, any parameter or extra data value, of any surface or surfaces. Multi-configuration operands may also
be defined as compensators. The parameter values are useful for using tilts and decenters of particular
components for compensation. The surface to be tilted must already be defined as a coordinate break (or perhaps
a tilted surface), with any appropriate pick ups as required.

The image surface focus is specified as a compensator by default. You may add or delete compensators to
customize the analysis to your particular situation, and may use as many compensators as you like. Generally,
using more compensators will loosen tolerances, and complicate the actual alignment of the system.
All compensators are defined using the COMP, CPAR, CEDV, and CMCO tolerance operands. Defining
compensators is described in the section "Tolerance control operands".
When using a script to define the tolerance procedure, the compensators may be altered and different sets of
compensators may be defined at different stages of the simulated alignment procedure. For details on this option,
see Using Tolerance Scripts on page 463.
Preparing the lens for tolerancing
Generally speaking, no changes are required to the lens before starting the tolerance procedure. There are
however a few exceptions.
The tolerance algorithm works most reliably if the criteria being used consists only of image quality evaluation
operands; such as the RMS spot radius, RMS wavefront, or MTF. If a user-defined merit function is used, some
care must be taken in the operands selected for use in the merit function.
Merit functions that contain construction constraints that reference individual surface numbers can cause
incorrect results. The problem arises because ZEMAX must insert dummy surfaces to automatically tilt and
decenter surfaces and groups of surfaces. This insertion causes a renumbering of the operands in the merit
function which may or may not be correct.
For example, to tilt a range of surfaces, ZEMAX must insert 3 dummy surfaces, and then copy the thickness
of the previous last surface in the range to the new dummy surface which follows the range. This can cause the
operands which previously constrained the thickness on one surface to no longer constrain the correct value or
surface.
A similar problem arises when negative or small thickness values are not allowed by the merit function
constraints; since the insertion of the dummy surfaces may violate these boundary constraints.
The recommended solution is to use only merit functions that do not include surface-specific constraints. If
these operands must be used; then the surfaces targeted should be dummy surfaces in between lens elements;
rather than surfaces which are part of a lens group.
Note that for tolerancing purposes it is not usually required to constrain negative edge or center thicknesses,
or other mechanical packaging constraints. These unphysical conditions, which must be prevented during
optimization, are not a concern after the lens is physically fabricated.
Performing the tolerance analysis
Once all the tolerance operands and compensators have been defined, the tolerance analysis may be
performed. To perform a tolerance analysis, select Tolerancing from the Tools menu in the main program menu
bar.
The dialog box which appears has several controls which are described below.
Mode: Four modes are supported: Sensitivity, Inverse Limit, Inverse Increment, and Skip Sensitivity.
Sensitivity mode computes the change in the criteria for each of the extreme values of the tolerances.
Inverse Limit computes the value of each tolerance that will yield a criteria being equal to the value specified
by the Limit parameter. The Limit parameter is only available if the mode is Inverse Limit. Inverse mode will
change the min and max values of the tolerance operands. See Inverse sensitivity analysis on page 459.
Inverse Increment computes the value of each tolerance that will yield a change in the criteria equal to the
value specified by the Increment parameter. The Increment parameter is only available if the mode is
Inverse Increment. Inverse mode will change the min and max values of the tolerance operands. See
Inverse sensitivity analysis on page 459.
Skip Sensitivity will bypass the sensitivity analysis and proceed to the Monte Carlo analysis.
Limit: When using Inverse Limit mode, this control is active and is used to define the limit on the criteria for
computing inverse tolerances. For example, suppose the Criteria is RMS Spot Radius, and the nominal RMS

of a system is 0.035. If Limit is set to 0.050, then ZEMAX will compute the min and max value of each tolerance
that degrades the performance to an RMS of 0.050. The Limit value must represent worse performance than
the nominal system has. When using MTF as a merit, then Limit is the lower bound on MTF since lower
numbers indicate worse performance. See Inverse sensitivity analysis on page 459. The nominal value for
the currently selected criteria may be computed by pressing the "?" button adjacent to the Limit edit window.
Increment: When using Inverse Increment mode, this control is active and is used to define the limit on the
change in the criteria for computing inverse tolerances. For example, suppose the Merit is RMS Spot Radius,
and the nominal RMS of a system is 0.035. If Increment is set to 0.01, then ZEMAX will compute the min and
max value of each tolerance that degrades the performance to an RMS spot radius of 0.045. The Increment
value must be positive to represent a degradation in performance. When using MTF as a merit, the Increment
is still positive, and ZEMAX automatically interprets this number as a decrease in MTF from the nominal. See
Inverse sensitivity analysis on page 459. The nominal value for the currently selected criteria may be
computed by pressing the "?" button adjacent to the Increment edit window.
# Monte Carlo Runs: This control is used to specify how many Monte Carlo simulations should be performed.
The default setting of 20 will generate 20 random lenses which meet the tolerances specified. See the section
"Monte Carlo simulations" for more details. The number of Monte Carlo runs may be set to zero, which will
omit the Monte Carlo analysis from the summary report.
Save MC Runs: This option is used to save a specific number of lens files generated during the Monte Carlo
analysis. The value specifies the maximum number of lens files to be saved. For example, suppose 20 is
selected. After the first Monte Carlo lens is generated, the lens file will be saved in the file MC_T0001.ZMX.
The second Monte Carlo lens file will be generated, then saved in MC_T0002.ZMX, and so on. Only the first
20 Monte Carlo lenses will be saved (the last will be MC_T0020.ZMX). If fewer than 20 Monte Carlo runs were
requested, then fewer than 20 lenses would be saved. Be sure you have no lens files with the names
MC_Txxxx.ZMX, as ZEMAX will overwrite these files without warning as the lenses are saved. The purpose
of this feature is to allow further study on the lenses being generated by the Monte Carlo feature.
Criteria: This control is used to specify what shall be used as the criteria for tolerancing. The options are:
RMS spot size (radius, x, or y): The best choice for systems which are not close to the diffraction limit; for
example, systems with more than one wave of aberration. This is the fastest option. ZEMAX always uses
a centroid reference for tolerancing.
RMS wavefront: The best choice for systems which are close to the diffraction limit; for example, systems
with less than one wave of aberration. This is nearly as fast as RMS spot radius. ZEMAX always uses a
centroid reference for tolerancing.
Merit Function: Uses whatever merit function has been defined for the lens. This is useful for user-defined
tolerancing criteria. User-defined merit functions may also be required for systems with non-symmetric
fields, or with significant surface apertures which remove rays. If the user-defined merit function is used,
no boundary constraints on the compensators are automatically added to the merit function. If the merit
function was generated by ZEMAX as one of the default merit functions, make sure the "Assume Axial
Symmetry" option was checked OFF, see Assume Axial Symmetry on page 389 for a discussion.
Geometric or Diffraction MTF (average, tangential, or sagittal): The best choice for systems which require
an MTF specification. If average is selected, the average of the tangential and sagittal responses is used.
Diffraction based MTF tolerancing can be problematic if the tolerances are loose, because the diffraction
MTF may not be computable or meaningful if the OPD errors are too large. This is especially true if the
spatial frequency is high enough and the performance poor enough for the MTF to go to zero at some
frequency below the frequency being analyzed. MTF is the slowest of the default criteria. The frequency at
which the MTF is computed is specified in the "MTF Frequency" control.
Boresight error: Boresight error is defined as the radial chief ray coordinate traced for the on axis field
divided by the effective focal length. This definition yields a measure of the angular deviation of the image.
ZEMAX models boresight error by using just one BSER operand (see the Optimization chapter for details
on BSER). Any element or surface decenters or tilts will tend to deviate the chief ray and increase the values
of the BSER operand. Boresight error is always computed at the primary wavelength in radians. Boresight
error should only be used with radially symmetric systems. Note boresight error gives no indication of image
quality; it is a measure of the deviation of the axis ray.
User Script: A user script is a macro-like command file which defines the procedure to be used for alignment

and evaluation of the lens during tolerancing. For details on this option, see the section Using Tolerance
Scripts on page 463.
MTF Frequency: If MTF is selected as the Merit, then this control is active and is used for defining the MTF
frequency. MTF frequency is always measured in line pairs per mm in the image space.
Sampling: Sampling is used to set how many rays are traced when computing the tolerance criteria. Higher
sampling traces more rays, and gives more accurate results. However, the execution time increases. If the
selected criteria is RMS spot or RMS wavefront, then the sampling value is an integer that refers to the number
of rays traced along a radial arm of the pupil in the Gaussian quadrature technique (see Selecting the pupil
integration method on page 387 for a description of this technique). The number of arms is always twice the
number of rays along each arm. If MTF is the selected criteria, then the sampling refers to the pupil grid size,
with a sampling of 1 yielding a 32 x 32 grid, sampling of 2 yielding a 64 x 64 grid, etc. Usually, a sampling of
3 or 4 is sufficient for quality optical systems. Systems with high amounts of aberration require higher sampling
than systems with low aberration. The most reliable method for determining the best sampling setting is to run
the tolerance at a sampling of 3, then again for a sampling of 4. If the results change moderately, then use the
higher setting. If they change substantially, check the next higher sampling setting. If the results change little,
go back to the lower sampling. Setting the sampling higher than required increases computation time without
increasing accuracy of the results.
Config: For multi-configuration lenses, indicates which configuration should be used for tolerancing. The
selected configuration only will be considered, and the configuration number will be printed on the final report.
If "All" is selected, then all configurations will be considered at once.
Comp: This control determines how the compensators are evaluated. "Optimize All" will use the optimization
capability of ZEMAX to determine the optimum values of all defined compensators. Although optimization is
accurate, it is slow to execute. If "Paraxial Focus" is selected, only the change in paraxial back focus error is
considered as a compensator; all other compensators are ignored. Using Paraxial Focus is very useful for
rough tolerancing, and is significantly faster than using "Optimize All".
# Opt Cycles: This determines how rigorously ZEMAX will attempt to optimize compensator values. If set to
Auto, then ZEMAX will call the optimizer in "Auto" mode, which will run the optimizer until the optimization of
the compensators has converged. For rough tolerancing, a low number, such as 1, 2, or 3, may be used. If the
compensators are difficult to optimize, a higher setting may increase accuracy. If too few optimization cycles
are selected, then the tolerances will be pessimistic; the predicted performance will be worse than the actual
performance. The "Auto" setting is the safest to use. Higher settings increase accuracy at the expense of run
time. This setting is only used if Comp is set to "Optimize All".
Fields: Generally speaking, the field definitions used for optimization and analysis are inadequate for
tolerancing. for example, a rotationally symmetric lens may use field definitions of 0, 7, and 10 degrees. For
tolerancing purposes, the lack of symmetry in the field definitions may cause inaccurate results when analyzing
tilt or decenter tolerances. When constructing a merit function to use for tolerancing ZEMAX can use three
different field settings:
Y-Symmetric: ZEMAX computes the maximum field coordinate, then defines new field points at +1.0, +0.7,
0.0, -0.7, and -1.0 times the maximum field coordinate, in the Y direction only. All X field values are set to
zero. This is the default for rotationally symmetric lenses.
XY-Symmetric: Similar to Y-Symmetric, except there are 9 field points used. The 5 Y-Symmetric points are
used, and -1.0, -0.7, +0.7, and +1.0 are added in the X axis direction only.
User Defined: Use whatever field definitions exist in the current lens file. This option is required when using
vignetting factors, tolerancing multiple configuration lenses, or using tolerance scripts. It is also highly
recommended when tolerancing non rotationally symmetric lenses or lenses with complex field weighting
that user defined fields be used.
If user defined fields are used, no adjustment of the weights is performed. For the Y-Symmetric case, the center
point has a weight of 2.0, all others have a weight of 1.0. For the XY-Symmetric case, the center point has a
weight of 4.0, all others unity.
Separate Fields/Configs: If checked, the criteria will be computed and displayed for all field positions in all
configurations individually. If unchecked, the criteria will be computed as an average over all field positions in
all configurations. During inverse tolerancing, if Separate Fields/Configs is unchecked, the inverse analysis is
done on the overall tolerance criteria, which is usually an average of performance over all fields in all

configurations. The problem with using the average performance is that some fields or configurations may be
significantly degraded by tolerance defects while other fields or configurations are not, and the average may
not reveal the severity of the loss of performance at a few fields or configurations. If this option is checked,
then ZEMAX computes the criteria at each field in each configuration individually, and verifies that each field
meets either the Limit or Increment value. For Inverse Increment mode, ZEMAX computes the nominal
performance at every field position and reduces the tolerances until the criteria at every field is degraded no
more than the increment value. See Inverse sensitivity analysis on page 459.
Script: The name of the script, if using the User Script criteria. User scripts must be ASCI files ending in the
extension TSC and be in the same directory as the ZEMAX.EXE program executable.
Force Ray Aiming On: If the lens being toleranced is already using ray aiming, then ray aiming will be used
when evaluating the tolerances. If ray aiming is not already on, then ray aiming will only be used if this box is
checked. Generally, using ray aiming yields more accurate results but slower computation speed. For
preliminary or rough tolerance work, leave the switch at the default "off", but for final or precise work, set the
switch "on".
Show Descriptions: If checked, a full description of the meaning of each tolerance operator will be provided in
the analysis report. If unchecked, only the tolerance operator abbreviations will be listed.
Show Compensators: By default, the compensator values are not printed out during the sensitivity analysis. If
this box is checked, then each compensator value will be printed along with the change in the criteria for each
tolerance.
Statistics: Choose either a Gaussian "normal" distribution, "uniform", or "parabolic" distribution. This setting is
only used by the Monte Carlo analysis; see that discussion for details on the statistics mode as well as the
"STAT" command which provides detailed control over the statistics model used.
Overlay MC Graphics: If checked, each open analysis graphical window (such as a ray fan or MTF graph) will
be updated and overlayed for each Monte Carlo generated lens. The resulting plots are useful for showing the
total range of performance for the simulated lenses. Analysis graphs which do not automatically change scale
and do not depend upon surface numbers are the most useful, such as MTF, MTF vs. Height, Encircled Energy,
and other plots that allow a user defined fixed scale, such as ray fan plots. Plots which change scale
dynamically or require a fixed range of surfaces, such as layout plots, will not work reliably and are not
supported. Static, Text, and Editor windows are not updated. Overlayed graphic windows will be flagged as
static after the tolerance analysis is complete. The time to compute each of the analysis graphics for each MC
lens obviously will slow down the tolerance analysis.
Hide All But Worst: If checked, printing of all the sensitivity data will be turned off. This is useful for decreasing
the size of the output report. The "hide" check box is normally used in conjunction with the "Show worst"
control. The "Show worst" control can be set to sort and display only some number of tolerance operands; this
permits limiting the printout to the most severe tolerances only.
Show worst: See "Hide all but worst" above.
Status: This control is used by the tolerance algorithm to provide status messages during computation of the
tolerance analysis.
There are also six buttons on the bottom of the dialog box:
OK: Performs the tolerance analysis using the current options.
Cancel: Exits the dialog box without performing the tolerance analysis.
Terminate: Terminates the tolerance analysis.
Save: Save the currently selected options for future use.
Load: Restore the previously saved settings.
Reset: Restore the settings to the default values.
Once all of the options have been selected, press OK to begin the tolerance analysis. Details on the method
of calculation are provided in the next section.
How ZEMAX computes the tolerance analysis
ZEMAX begins the tolerance analysis by saving the lens in a temporary file which will be used to restore the
lens after the tolerancing is complete. All of the changes made during tolerancing will ultimately be discarded,

and the original lens file will be restored unmodified. The exception is during inverse sensitivity analysis, where
the tolerance data min and max limits may be altered.
ZEMAX then removes all of the variables. Solves are left in place, however, solves may cause problems during
tolerancing; see Tolerancing with solves on page 468 for more information.
The tolerance operators are read, and the compensating parameters defined by COMP and CPAR specified
are set as variables. The compensators defined may be altered during the analysis if a tolerance script is being
used. If ray aiming is on in the lens being toleranced, or if the "Force Ray Aiming On" switch is checked, then ray
aiming will be used in evaluating the tolerances, otherwise, ray aiming is left off. Tolerances computed using ray
aiming are more accurate, but execution speed is slower. See Ray Aiming on page 88 for details on ray aiming.
ZEMAX then uses the Criteria, Field, MTF Frequency, and Sampling settings on the tolerance dialog box to
define an appropriate merit function for the tolerancing. Since this is done to the temporary file only, the original
merit function defined for the lens is not disturbed.
Boundary constraints are added to the merit function to limit the compensators to the min and max boundaries
specified using the COMP and CPAR commands. If using a user defined merit function, or if the "Comp" control
is set to "Paraxial Focus", boundary constraints on the compensators are ignored.
If "Comp" is set to "Optimize All", ZEMAX then calls the optimization function to find the best values for the
defined compensators. Otherwise, the back focal position is set to paraxial focus. The resulting lens file is then
saved for subsequent use by the tolerancing algorithm.
The criteria of this lens is considered to be the "nominal". Note that the nominal criteria value generally will not
be the same as the criteria value that is reported on the optimization or merit function editor windows, because
ZEMAX constructs a new merit function just for use in tolerancing. ZEMAX then proceeds with either sensitivity,
inverse, or Monte Carlo analysis as described in subsequent sections.
Evaluating compensators
ZEMAX has two methods for evaluating compensators selected by the "Comp" control: either "Optimize All"
or "Paraxial Focus". If "Paraxial Focus" is selected, then several assumptions are made which greatly speeds up
the tolerance evaluation. First, all defined compensators and compensator boundary constraints are ignored.
Back focus is implemented as a compensator using a paraxial focus error solve. This means the focus is adjusted
to maintain the amount of defocus in the nominal lens design, without attempting to reoptimize the back focus
precisely. These assumptions greatly speed up the tolerance evaluation. If "Optimize All" is selected, then ZEMAX
uses the slower but more accurate optimization algorithm to find the best value for all the compensators.
"Paraxial Focus" mode is fast, accurate, and should be used if back focus is the only compensator, and the
system is well described by paraxial rays, such as systems with rotational symmetry. "Paraxial Focus" should not
be used when the system is highly non-symmetric or there are multiple compensators. When in doubt, run both
modes and compare the results. Since "Paraxial Focus" does not precisely optimize the compensators, "Paraxial
Focus" results are generally more pessimistic than "Optimize All" results.
Sensitivity analysis
For the sensitivity analysis, each tolerance is evaluated independently using the following algorithm:
The temporary lens is restored.
The parameter whose tolerance is being evaluated is adjusted to the extreme minimum value. For example,
if the tolerance being evaluated is TRAD, and the nominal value is 100 mm, with a minimum tolerance of -0.1
mm, the radius is set to 99.9. If the tolerance is an element tilt or decenter tolerance, dummy coordinate breaks
are inserted as required to model the perturbation. For surface tilts and decenters, such as TSDX, TSDY,
TSTX, TSTY, TIRX, or TIRY, the Irregular surface type is used if the surface is initially of type Standard.
The compensators are adjusted.
The resulting criteria are printed on the report.
The procedure is repeated for the maximum tolerance.
This basic algorithm is repeated for each tolerance operand.
The value of the sensitivity analysis is that the tolerances which are too loose will, in general, have greater
contributions to the increase in the criteria than other tolerances. This technique allows the designer to identify

surfaces which are highly sensitive to certain errors, such as tilts or decenters. Different surfaces also will, in
general, have very different sensitivities to the various errors. The sensitivity analysis aids in identification of which
tolerances need to be tightened, and which might be loosened. It is also valuable for finding the optimum (and
minimum) number of compensators, and the required range of adjustment. There are in fact many more
applications for this feature; for example, designing lens mounts to maximize compensation leverage.
!The sensitivity analysis aids in identification of which tolerances need to be tightened, and
which might be loosened.
The amount of output can be overwhelming, especially for lenses with many elements and a corresponding
large number of tolerances. Often, the sensitivity to the tolerances varies widely over all the possible tolerances.
The "Show worst" control is extremely useful for summarizing the worst offenders, because it sorts the tolerances
by contribution to the criteria and then prints them in descending order. The "Hide all but worst" control turns off
the bulk of the printing if only the worst offenders are of interest.
After all the individual tolerances are computed, ZEMAX then computes a variety of statistics, the most
important of which is the estimated change in the criteria and associated estimated performance. ZEMAX uses
a Root Sum Square (RSS) assumption for computing the estimated changes in the performance. For each
tolerance, the change in performance from the nominal is squared and then averaged between the min and max
tolerance values. The resulting averaged squared values are then summed for all the tolerances, and the square
root of the result is taken. The average of the min and max tolerances is taken because the min and the max
tolerance cannot both occur simultaneously, and so summing the squares would result in an overly pessimistic
prediction. The resulting RSS is the estimated change in performance.
Inverse sensitivity analysis
If an inverse sensitivity is being performed, the tolerances are computed the same way as for the sensitivity
analysis. However, the calculation is performed iteratively inside a loop while adjustments are made to the min
and max tolerances. For Inverse Limit mode, the adjustments are made until the resulting criteria is approximately
equal to the Max Criteria. For Inverse Increment mode, the adjustments are made until the resulting change in
criteria is approximately equal to the Increment value.
For example, if the mode is Inverse Limit, the Criteria is RMS spot radius, the nominal criteria is 0.035, and
the Max Criteria is 0.050, ZEMAX will adjust the tolerances until the criteria is 0.050. If the mode is Inverse
Increment, the Criteria is RMS spot radius, the nominal criteria is 0.035, and the Increment is 0.010, ZEMAX will
adjust the tolerances until the merit is 0.045. For Inverse Limit, the Max Criteria must be greater than the nominal,
except for MTF Criteria, in which case the Max Criteria must be less than the nominal, or an error message will
be generated and the analysis terminated. For Inverse Increment, the Increment must always be positive.
If the option to "Separate Fields/Configs" is not selected, the criteria ZEMAX uses to perform the inverse
analysis is the overall criteria, which is typically an average over all fields and configurations. The problem with
using the average performance is that some fields or configurations may be significantly degraded by tolerance
defects while other fields or configurations are not, and the average may not reveal the severity of the loss of
performance at a few fields or configurations. If "Separate Fields/Configs" is checked, then ZEMAX computes the
criteria at each field in each configuration individually, and verifies that each field meets either the Max Criteria or
Increment value. For Inverse Increment mode, ZEMAX computes the nominal performance at every field position
and reduces the tolerances until the criteria at every field is degraded no more than the increment value.
If the starting value of the tolerance yields performance better than the Max Criteria or Increment value would
dictate, the tolerance is not adjusted. This means the tolerance will never be loosened; it can only be tightened
during inverse sensitivity analysis. For example, if the nominal is 0.035, and the Max Criteria is 0.050, and the
initial tolerance yields a criteria of 0.040, the tolerance will not increase. To compute the actual limit, the tolerance
must first be loosened on the Tolerance Data Editor, and then inverse sensitivity repeated. This is done to prevent
looser than necessary tolerances. Generally, tolerances that are looser than some reasonable value do not
decrease manufacturing costs.
The estimated change in performance is computed in the same way as for the sensitivity analysis, using the
newly adjusted tolerances. The inverse sensitivity analysis aids in tightening individual tolerances so no one
defect contributes too much to performance degradation.

Note that the inverse sensitivity is computed for each tolerance individually. The overall degradation in
performance estimate will still be given by the RSS of all the individual increases. Inverse tolerancing can be
disabled for individual operands; which prevents the inverse tolerance algorithm from tightening tolerances even
if the estimated performance exceeds the specified limits. The option to disable inverse tolerancing is on the
"Operand Type" dialog box listed under the "Edit" menu of the Tolerance Data Editor.
Monte Carlo analysis
Unlike the sensitivity and inverse sensitivity analysis, the Monte Carlo analysis simulates the effect of all
perturbations simultaneously.
!The Monte Carlo analysis simulates the effect of all perturbations simultaneously.
For each Monte Carlo cycle, all of the parameters which have specified tolerances are randomly set using the
defined range of the parameter and a statistical model of the distribution of that parameter over the specified
range. By default, all parameters are assumed to follow the same normal distribution with a total width of four
standard deviations between the extreme minimum and maximum allowed values. For example, a radius of
100.00 mm with a tolerance of +4.0/ - 0.0 mm will be assigned a random radius between 100.00 and 104.00 mm,
with a normal distribution centered on 102.00 mm and a standard deviation of 1.0 mm.
This default model may be changed using the STAT command. Each tolerance operand may have a separate
definition for the statistics, or operands with the same statistical distribution form may be grouped together. All
tolerance operands which follow a STAT command use the statistical distribution defined by that STAT command.
As many STAT commands as desired may be placed in the tolerance data editor.
The STAT command accepts two arguments, Int1 and Int2. Int1 should be set equal to 0 for normal, 1 for
uniform, 2 for parabolic, and 3 for user defined statistics. For normal statistics only, the Int2 value should be set
to the number of standard deviations between the mean and extreme values of the parameter.
The available statistical distributions are described below.
Normal statistical distribution
The default distribution is a modified Gaussian "normal" distribution of the form
1 x2
p ( x ) = -------------- exp --------2- , n x n .
2 2
The modification is that the randomly selected value x (measured as an offset from the midpoint between the
two extreme tolerances) is restricted to lie within "n" standard deviations of zero. The default value of "n" is two,
however "n" may be changed using the Int2 argument of the STAT command defined earlier. This is done to
ensure that no selected value will exceed the specified tolerances. The standard deviation is set to one over "n"
times half the maximum range of the tolerances. For example, if "n" is 2, and a thickness is nominally 100 mm
with a tolerance of plus 3 or minus 1 mm, then the selected value will be drawn from a normal distribution with a
101 mm mean and a range of plus or minus 2 mm, with a standard deviation of 1.0. If "n" is 5, then the standard
deviation will be 0.4. The larger "n" is, the more likely the selected value will be closer to the mean value of the
tolerance extremes. The smaller "n" is, the more the normal looks like the uniform distribution.
Uniform statistical distribution
The uniform distribution is of the form
1
p ( x ) = -------, x .
2
The value is one half of the difference between the max and min tolerance values. Note that the randomly
selected value will lie somewhere between the specified extreme tolerances with uniform probability.

Parabolic statistical distribution
The parabolic distribution is of the form
3x 2
p ( x ) = --------3- , ( x ) ,
2
where is defined exactly the same as for the uniform distribution. The parabolic distribution yields selected
values that are more likely to be at the extreme ends of the tolerance range, rather than near the middle, as for
the normal distribution.
User defined statistical distribution
The user defined statistical distribution is defined by an ASCII file with tabulated distribution data. A general
probability function may be defined as
p ( x i ) = T i, 0.0 x i 1.0 ,
where the T values are tabulated for some number of discrete X values. This general distribution may be
numerically integrated, and from the integral of the tabulated values an estimated X may be randomly generated
with the statistics matching that of the tabulated distribution.
The format of the file is two columns of data as follows:
X1 T1
X1 T2
X3 T3
...etc.
where the X values are monotonically increasing floating point numbers between 0.0 and 1.0, inclusively, and
the T values are the probabilities of getting that X value. Note ZEMAX uses a probability distribution that covers
the range from 0.0 to 1.0, and so the first X1 value defined MUST be equal to 0.0 (it may have any probability T1,
including zero), and the last value defined MUST have an Xn value of 1.0. Up to 200 points may be used to define
the distribution between X = 0.0 and X = 1.0; a warning will be issued if too many points are listed.
For each tolerance operand subsequently defined (until another STAT command is reached) the defined min
and max values will determine the actual range of the random variable X. For example, if a value of 100.0 has a
tolerance of -0.0 and +2.0, the probability distribution will extend over the range of 100.0 to 102.0.
Once the data is defined in a file, the file must be placed in the same directory as the ZEMAX program, and
the file name (with extension) must be placed in the comments column of the tolerance data editor on the same
line as the STAT command. The STAT type must be set to "3".
!The file name of the user defined probability distribution must be placed in the comments
column of the tolerance data editor for the STAT operand.
One possible distribution could be:

0.0 0.0
0.1 0.5
0.2 1.0
0.3 0.5
0.4 0.0
0.5 0.5
0.8 4.0
1.0 5.0
Note the X data values do not need to be evenly spaced; finer spacing may be used in regions of rapid change
in probability. This distribution has two peaks, and the higher peak is highly skewed to the maximum side of the
distribution.

The user defined statistical distribution is very flexible, and may be used to model any probability distribution,
including skewed, multiply peaked, or measured statistical probability data. Multiple distributions may be defined
and used in the same tolerance analysis.
Discussion of Monte Carlo analysis method
Note that going from normal to uniform to parabolic statistics yields successively a more pessimistic analysis,
and thus more conservative tolerances.
For each cycle, the compensators are adjusted and then the criteria and compensator values are printed. After
all of the Monte Carlo trials a statistical summary is provided.
The value of the Monte Carlo analysis is estimating the performance of the lens considering all of the
tolerances simultaneously. Unlike the sensitivity analysis, which identifies the "worst offenders" in the system, the
Monte Carlo analysis estimates the real-world performance of a system which meets the tolerances specified.
The statistical summary provided may be highly useful for lens systems which are mass-produced. Lenses which
are one of a kind, of course do not follow these statistics because of the inadequate sampling. However, the Monte
Carlo analysis is still useful because it indicates the probability that a single lens will meet the required
specification.
Nesting rules for Monte Carlo analysis
When performing the Monte Carlo analysis, all tolerances are considered simultaneously. It is possible to
define element tilt and decenter operands that conflict or are ambiguous if certain rules are not followed.
Tolerances such as TEDX, TEDY, TETX, and TETY require ZEMAX to insert coordinate breaks before and
after the surface group, and then tilt or decenter the group as a whole. The tilts and decenters performed by the
first coordinate break must be "undone" by the second coordinate break. This can only be accomplished if the
vertices of the first and second coordinate breaks are at the same location in 3D space. ZEMAX ensures this
condition through the use of pickup and position solves.
This method will fail if the surface ranges specified by multiple tolerance operands overlap. For example, if
there is a TETX on surfaces 3-8, and then a subsequent TETX on surfaces 5-12, the first coordinate break which
tilts the 5-12 group will change the location of the surfaces 5-8, and the second coordinate break of the first group
will be moved. In this case, ZEMAX cannot ensure the coordinate breaks will work as intended. In fact,
overlapping coordinate breaks do not have a unique interpretation, and it is difficult to imagine an optical system
where they even have a meaningful physical interpretation.
Tolerances may be nested however, because nesting implies an unambiguous order to the tilts and decenters.
For example, a TETX on 5-12 followed by a TETX on 5-9 and a TETX on 10-12 is perfectly reasonable. This order
would simulate the tilting of an assembly, composed of multiple elements, each of which may be tilted within the
assembly itself.
The nesting rules are:
1) All element tilt and decenter tolerances must be nested.
2) The outermost pair of surfaces in each nested group must be first.
Here is an example of a valid set of operands:
TETX 5 12
TETX 5 10
TETX 11 12
Here is an invalid set of operands:
TETX 5 12
TETX 9 15
TETX 5 15
The second operand is invalid because it partially overlaps the first (this violates rule 1). The third operand is
invalid because although nested with operand 1, it is the outermost range of the two (this violates rule 2). The
second operand must be deleted or modified, but the third operand could be placed in front of the first operand
to make a legal operand list:
TETX 5 15
TETX 5 12

Note that an operand is considered nested even if it shares one or both surface limits with a previous operand,
so that TETX 5-15 may be followed by another TETX 5-15 or by TETX 5-12 or TETX 13-15, but not TETX 4-13.
Using Tolerance Scripts
Tolerance Script overview
Tolerance scripts are macro-like command files that define a procedure to follow for evaluating the perfor-
mance of a lens during tolerancing. Scripts allow simulation of a complex alignment and evaluation procedure for
a lens. With scripts, the following actions can be taken to evaluate a perturbed lens:
-Add or remove compensators.
-Load new merit functions.
-Optimize defined compensators using any merit function.
-Monitor and report any value ZEMAX can compute via a merit function (which includes essentially any value
ZEMAX can compute since the merit function may call ZPL macros). Other data, such as Zernike coefficients,
may also be computed.
-Write data out to ASCII or binary files for later analysis.
-Save lenses at any stage of the analysis to a ZMX file.
Any number of these operations may be combined in the script; so multiple merit functions and compensator
groups may be defined. The script is executed many times during the evaluation of the tolerance analysis,
including:
-Once to compute the nominal data.
-Twice for each tolerance operand in the sensitivity analysis, (once for the min and once for the max tolerance).
-Multiple times for each tolerance operand in the inverse sensitivity analysis (inverse sensitivity may require
iteration).
-Once for each generated Monte Carlo lens.
The tolerance script files must be ASCII files ending in the extension TSC, and be placed in the same directory
as the ZEMAX program.
Note that when using tolerance scripts, the min and max boundary values of compensators are not enforced;
see General comments about min and max values on compensators on page 451 for more information.
The Tolerance Script commands
The tolerance script commands are defined and described below.
!
Syntax:
! A comment line!
The "!" symbol is used to define comments in the script that are ignored during execution of the script.
CEDV
Syntax:
CEDV surf param
CEDV defines a new extra data value compensator. The value "surf" is the surface number in the original lens
file. ZEMAX automatically renumbers the surface if additional coordinate break or other dummy surfaces were
inserted by the tolerance program. The value for "param" corresponds to the extra data number for the surface.
CLEARCOMP
Syntax:
CLEARCOMP

CLEARCOMP removes all current compensators. New compensators must be defined before calling OPTI-
MIZE.
CLOSEFILE
Syntax:
CLOSEFILE
CLOSEFILE closes the current output file, and no more data will be written to the output file until the next
OPENFILE is executed. See OPENFILE below.
CMCO
Syntax:
CMCO operand config
CMCO defines a new multi-configuration operand compensator. The value "operand" is the operand (row)
number in the original lens file. The value for "config" corresponds to the configuration number.
COMP
Syntax:
COMP surf code
COMP defines a new compensator. The value "surf" is the surface number in the original lens file. ZEMAX
automatically renumbers the surface if additional coordinate break or other dummy surfaces were inserted by the
tolerance program. The value for "code" is 0 for thickness, 1 for curvature, 2 for conic.
CPAR
Syntax:
CPAR surf param
CPAR defines a new parameter compensator. The value "surf" is the surface number in the original lens file.
ZEMAX automatically renumbers the surface if additional coordinate break or other dummy surfaces were
inserted by the tolerance program. The value for "param" corresponds to the parameter number for the surface.
FORMAT
Purpose:
Specifies the numerical precision format for subsequent REPORT statements.
Syntax:
FORMAT m.n [EXP]
Discussion:
The integers m and n are separated by a decimal point. The value m refers to the total number of characters
to be printed, even though some of them may be blank. The value n refers to the number of places to display
after the decimal point. Therefore, FORMAT 8.4 will cause subsequent REPORT statements to print 8 characters,
with 4 numbers after the decimal point. FORMAT .5 will cause REPORT to show 5 decimal places, and as many
total places as needed. FORMAT only affects numeric output from REPORT. If a number is too large to fit within
the m decimal places, then the m portion of the FORMAT statement will be ignored. The optional keyword EXP
after the m.n expression indicates exponential notation should be used. The default format is 16.8 EXP.
Example:
FORMAT 18.9 EXP
LOADMERIT
Syntax:
LOADMERIT filename.mf
LOADMERIT "C:\SOMEPATH\SOME MERIT FILE NAME.MF"
The MF file must reside in the same directory as ZEMAX if no path is specified. The file should be in the proper
format as saved using the Tools, Save option of the Merit Function Editor.
When the merit function is loaded, it replaces any existing merit function. Any operands in the merit function
which reference surface numbers should be the surface numbers for the original, unaltered lens. ZEMAX

automatically adjusts the surface numbers as required if additional coordinate break or other dummy surfaces
were inserted by the tolerance program. The new merit function is then evaluated. The resulting merit function
value is what will be returned to the tolerance analysis program as the "criteria" unless a subsequent LOADMERIT
or OPTIMIZE command is executed.
NOMINAL
Syntax:
NOMINAL keyword
The NOMINAL script command allows some control over the evaluation of the script for the computation of
the nominal performance only. The supported keywords are:
OPT_OFF: Turns off optimization of the compensators. Any subsequent OPTIMIZE commands are ignored.
OPT_ON: Turns on optimization of the compensators. Any subsequent OPTIMIZE commands are executed.
The NOMINAL command is only executed during computation of the nominal system performance, and not
during the analysis of individual tolerances.
OPENFILE
Syntax:
OPENFILE "FILENAME" mode
FILENAME may be any valid file name, with the full path and extension included, such as "C:\DATA\My
File.DAT". The mode parameter should be BIN for a binary file, or ASC for an ASCII file.
OPENFILE will open a new data file the first time the script is run for a tolerance analysis run. Each subsequent
call to the script will append data to the same file.
Data is written to the file whenever the REPORT command is executed. However, the text string is not included;
just the numerical data the REPORT command prints to the tolerance output file. If mode is ASC, the file will be
ASCII with each REPORT value written on a separate line. If mode is BIN, each value will be written out as a 64-
bit double precision number.
Because data is written out each time the script is called, the amount of data in the file depends upon how
many tolerance operands and Monte Carlo runs are being performed. The first call to the script is to compute the
nominal criteria, then the sensitivity analysis, then the Monte Carlo analysis. OPENFILE should not be used
during an inverse sensitivity run, because there are multiple calls to execute the script and no practical way to
discern which data belongs to which operand.
The file is automatically closed after the script is finished executing. Multiple OPENFILE commands may be
used within the same file to write different data to different files. See also CLOSEFILE.
OPTIMIZE
Syntax:
OPTIMIZE n
OPTIMIZE calls the damped least squared optimizer and executes "n" cycles of optimization. If n is zero or is
omitted the optimizer runs in automatic mode, terminating when convergence is detected.
PERTURB
Syntax:
PERTURB type int1 int2 int3 stat nstd min max
PERTURB is used to randomly change a numerical parameter in the lens. The integer parameter type must
be 0 for surface data, 1 for parameter data, 2 for extra data, 3 for multi-configuration data, 4 for non-sequential
component position data, and 5 for non-sequential parameter data. The arguments to PERTURB change
meaning depending upon the value of type as described below.
Type = 0, surface data: int1 is used for the surface number, int2 is 0 for thickness, 1 for radius, 2 for conic,
and 3 for semi-diameter. The value of int3 must be set to zero.
Type = 1, parameter data: int1 is the surface number, int2 is the parameter number, int3 must be set to zero.
Type = 2, extra data: int1 is the surface number, int2 is the parameter number, int3 must be set to zero.

Type = 3, multi-configuration data: int1 is the configuration number, int2 is the operand number, int3 must
be set to zero.
Type = 4, NSC position data: int1 is the surface number, int2 is the object number, int3 must be 1 for x, 2
for y, 3 for z, 4 for tilt x, 5 for tilt y, or 6 for tilt z.
Type = 5, NSC parameter data: int1 is the surface number, int2 is the object number, int3 is the parameter
number.
The stat integer indicates the statistical model to use: 0 for normal (Gaussian), 1 for uniform, and 2 for
parabolic. Only these 3 distributions are currently supported. The nstd is a floating point value that indicates the
number of standard deviations from the center of the distribution to either edge. A small value for nstd (less than
1.0) will yield a nearly uniform distribution. Large values of nstd will yield a sharply peaked distribution, with most
of the randomly generated values being clustered close to the center of the range. Only the normal distribution
uses the nstd value, however, it must be provided and should be set to zero if the other stat types are used.
The min and max values indicate the range over which the random values are selected.
For example, to randomly perturb the thickness of surface 6 by -0 to +1 with a normal distribution covering
plus/minus 3 standard deviations, the syntax would be:
PERTURB 0 6 0 0 0 3.0 0.0 1.0
The resulting random value will be between 0.0 and 1.0, and will likely be close to 0.5 because the number of
standard deviations is moderately large. This random value will then be added to the current value for the
thickness of surface 6. Note that PERTURB works on all values, whether they are compensators or not. To perturb
values controlled by multi-configuration operands, use type 3 to perturb the operand directly rather than the
parameter under control.
The PERTURB command is ignored while evaluating the nominal criteria.
REPORT
Syntax:
REPORT "text" operand
REPORT "text" SZERNIKE term field wave sampling maxorder
There are two different ways to use the REPORT command:
Syntax 1: REPORT will print any user defined text to the tolerance output window, along with the value of any
operand in the currently loaded merit function. The value for operand is an integer corresponding to the operand
number (row) of the value to print (note that any value may be computed in a merit function, and if it is not needed
for optimization, it may be weighted to zero and is still available for reporting). If operand is zero, then the value
of the entire merit function will be printed out. The numeric format is set by the FORMAT command.
Syntax 2: REPORT, when followed by a text string and then the keyword SZERNIKE, can be used to compute
and report Standard Zernike coefficients (See Zernike Standard Coefficients on page 168). If the term number
is zero, then the other values for field, wave, sampling, and maxorder are used to define the Zernikes to be
computed. The computed values are then saved in a buffer, and no output is generated. If term is between 1 and
the most recent maxorder value used in a previous SZERNIKE call, then that Zernike term is reported. A sample
session follows:
REPORT "Compute Only!" SZERNIKE 0 1 1 1 4
REPORT "Zernike 1 = " SZERNIKE 1
Note that only when the term is zero are the Zernikes actually computed; the following calls retrieve the
computed Zernikes one at a time.
SAVE
Syntax:
SAVE n

SAVE will save the current lens file in a ZMX file; with the name TSAVnnnn.ZMX where nnnn is the four digit
integer representation of the number "n". For example, if "n" is 6, the file will be saved as TSAV0006.ZMX.
UPDATE
Syntax:
UPDATE
Updates the current lens file. Use after PERTURB commands.
Tolerance Script example
As an example, suppose an optical system is assembled from multiple lens elements. As part of the alignment
and evaluation process the system is adjusted as follows:
Element number 2 is decentered until an axial test beam is centered on the image.
Element number 4 is then shifted along the axis until the proper magnification is achieved.
Finally, the back focus is adjusted to maximize the on-axis MTF only.
The distortion is then measured and logged.
The MTF at 5 field points is evaluated.
Assume that element number 2 is decentered using parameters 1 and 2 of surface 3, which is an existing
coordinate break, the position of element 4 is thickness 10, and the back focus is thickness 15. Furthermore,
assume a merit function which centers the axis ray on the image surface is defined in CENTER.MF, the
magnification enforcing merit function is MAGNIFY.MF, the MTF merit function is MTF.MF, and the evaluation
merit function is EVALUATION.MF. The corresponding script might look like this:
! clear any existing compensators for a clean start
CLEARCOMP
! load the centering merit function
LOADMERIT CENTER.MF
! define the two compensators to decenter element 2
CPAR 3 1
CPAR 3 2
! optimize 4 cycles
OPTIMIZE 4
! clear the decenters, load the magnification merit, and adjust thickness 10
CLEARCOMP
LOADMERIT MAGNIFY.MF
COMP 10 0
OPTIMIZE 4
! now load the MTF merit function, and adjust back focus
CLEARCOMP
LOADMERIT MTF.MF
COMP 15 0
OPTIMIZE 4
! finally, load the evaluation merit function, and report the distortion and 5 MTF
values
! these should be the first 6 operands in EVALUATION.MF.
CLEARCOMP
LOADMERIT EVALUATION.MF
REPORT "Distortion = " 1
REPORT "MTF at field 1 = " 2
At the end of the script, the entire merit function value resulting from the last LOADMERIT or OPTIMIZE
command is returned as the criteria which ZEMAX reports and uses.
Tolerancing multi-configuration (zoom) lenses
Tolerance analysis can be performed on each configuration of a multiple configuration lens. Select the desired
configuration from the "Configuration #" drop down box. Generally, it is not required to tolerance all configurations,

but that is certainly an option. When using inverse tolerances, successive tolerance analysis in each configuration
will yield the tightest tolerances which apply to all configurations collectively.
In some cases, ZEMAX cannot perform tolerances correctly on surfaces controlled by multi-configuration
operands. For example, consider an element decenter defined by the tolerance operand TEDX 3 4, while the radii
and thicknesses of these surfaces are also being controlled by thermal operands in the multi-configuration editor.
In a case like this, ZEMAX will issue an error message. The solution is to separate the two effects by adding
dummy surfaces before and after the element. The new dummy surfaces (3 and 6) are added before and after
the real lens surfaces (now numbered 4 and 5). Now the decenter can be specified using TEDX 3 6, and the
thermal controls may still be applied to surfaces 4 and 5.
Tolerancing with solves
Generally speaking, you should delete all solves and variables before tolerancing a lens. It may be more
convenient to save the lens in a temporary file for tolerancing if you have many solves or variables to delete that
you may need later. Although the tolerancing will still work if you leave solves in your lens file, the results may not
be what you expect. The reason is that certain solves, such as paraxial marginal ray height solves, no longer
have meaning when the optical system contains tilted or decentered elements. Even if the system is initially
rotationally symmetric, most tolerance commands such as TETX and TETY force the system to be non-
rotationally symmetric. Pickup solves also can cause strange results because the tolerancing algorithm some-
times will manipulate the surface description data (such as thicknesses) when inserting and deleting coordinate
breaks. The same limitation applies to position solves, which may no longer work correctly when the lens data is
altered by the tolerancing algorithm.
There are however occasions where a pickup solve is really what you want. For example, if you are using a
lens in double-pass, and a tilt on one element implies a tilt on a subsequent element, then a pickup solve can be
used to pickup the tilts from the first element. For this special case, you will need to use the TUTX, and TUTY
commands, and the default tolerances will not work. If you have these types of solves in the system, ignore the
warning "Solves should be deleted before tolerancing".
Trouble shooting the tolerance results
See the discussion in the section on the SAVE operand.
If any of the calculated tolerance data show a value of "Infinity", this means the criteria could not be evaluated
for the specified tolerance. Usually the criteria cannot be evaluated because of total internal reflection of some of
the ray targets. The statistical data which follows the sensitivity analysis is usually meaningless if any of the
tolerances have a criteria value of infinity. One solution is to decrease by a factor of two or more the tolerance
value, and then repeat the analysis.
Optimizing for tolerance sensitivity
See Optimizing tolerance sensitivity on page 426.
Pitfalls when tolerancing
One possible error is non-physical propagation of rays when using tilt tolerances such as TETX and TETY. If
two elements are separated by a very small air space or dummy surface, the default tolerances will include tilting
of each element independently. If the element spacing is small, it is possible for one of the elements to be tipped
to the point where it "collides" with the other element. This could not happen in practice, however, for small
amounts of tilt the tolerance values are still a good indication of performance.
Because the tolerance routine tilts and decenters lenses, the algorithm automatically turns ray aiming "on"
(see Ray Aiming on page 88 for details about ray aiming). If the nominal criteria the tolerance algorithm reports
is different from the expected value, check to see if ray aiming is "off" in the lens file. Try using ray aiming "on"
and re-optimizing. Generally, if there is a big difference between criteria when ray aiming is turned on and off,
then it should be left on.
Summary
The tolerance routine is quite flexible and powerful. ZEMAX uses no approximations, extrapolations, or
estimations to compute the tolerances. For this reason, it gives useful results for both conventional and complex
systems. It is very important to appreciate that tolerancing is a complex procedure, and the algorithms used by

ZEMAX to manipulate the lens data are not infallible. Therefore, it is the designers responsibility to verify that the
program is computing reasonable results.

Chapter 17 MULTI-CONFIGURATIONS
Introduction
ZEMAX supports a very general capability for defining, analyzing, and optimizing optical systems which are
used in multiple configurations. Multi-configurations are used to design zoom lenses, or to optimize lenses tested
and used at different wavelengths, or optical systems used in various configurations, to name a few. Like other
ZEMAX features, multi-configuration is well integrated. However, like tolerancing, it requires a little more care and
practice to become proficient.
ZEMAX uses a substitution procedure for defining multiple configurations. The configurations are differentiated
by different values for the same parameter. For example, in a zoom lens, the spacings between various elements
may take on more than one value. Each set of values used together forms one configuration.
The first step
By far the most important step is to define one configuration using the normal ZEMAX mode first. It is a good
idea to start with the most complex configuration first. If all of the configurations have the same number of
elements, pick any one of them. Once you have the basic configuration defined, it is time to define new
configurations which are variations of the first. The first configuration does not need to be optimized yet, you can
optimize across configurations later.
Select Editors, Multi-Configuration from the main menu bar. The spreadsheet which appears is the multi-
configuration editor (MCE). Using the menu bar on the MCE, both configurations (columns) and operands (rows)
can be inserted and deleted. The insert and delete keys will also add or remove new operand rows. The data
entered on the MCE will be saved automatically whenever you save your lens file.
Summary of multi-configuration operands
To change the operand type, double click on the type column. A dialog box will appear where the type and
number of the multi-configuration operand can be changed. The operands are also summarized in the following
table.
SUMMARY OF MULTI-CONFIGURATION OPERANDS
Type Numbers 1,2,3 Description

APER Ignored System aperture value (whatever the current aperture definition is,
such as entrance pupil diameter or F/#). See also SATP.
APDX Surface # Surface aperture X- decenter. The surface must have a defined
aperture (NOT semi-diameter).
APDT Ignored System apodization type. Use 0 for none, 1 for Gaussian, 2 for
tangential. See also APDF.
APDF Ignored System apodization factor. See also APDT.
APDY Surface # Surface aperture Y- decenter. The surface must have a defined
aperture (NOT semi-diameter).
APMN Surface # Surface aperture minimum value. The surface must have a defined
aperture (NOT semi-diameter). This same operand also works to
control the first parameter of all surface aperture types, such as the
X-Half Width on rectangular and elliptical apertures.
APMX Surface # Surface aperture maximum value. The surface must have a defined
aperture (NOT semi-diameter). This same operand also works to
control the second parameter of all surface aperture types, such as
the Y-Half Width on rectangular and elliptical apertures.
471 Chapter 17: MULTI-CONFIGURATIONS

APTP Surface # Surface aperture type. The integer values indicating the aperture type
are 0-10 for none, circular aperture, circular obscuration, spider,
rectangular aperture, rectangular obscuration, elliptical aperture,
elliptical obscuration, user aperture, user obscuration, and floating
aperture; respectively.
CONN Surface # Conic constant.
COTN Surface # The name of the coating, if any, to be applied to the surface.
CRVT Surface # Curvature of surface.
CSP1 Surface # Curvature solve parameter 1.
CSP2 Surface # Curvature solve parameter 2.
CWGT Ignored The overall weight for the configuration. This number only has
meaning relative to the weights in other configurations. The
configuration weight is only used by the default merit function
algorithm to set up a merit function which favors configurations via
relative weighting. If the configuration weight is zero, then the
configuration is ignored during construction of the default merit
function.
EDVA Surface, Extra The EDVA operand is used to assign multiple values to the extra data
Data Number values. This operand requires 2 numerical arguments: the surface
number and the extra data value number.
FLTP Ignored Field type. Use 0 for angle in degrees, 1 for object height, 2 for
paraxial image height, 3 for real image height.
FLWT Field # Field weight.
FVAN Field # Vignetting factor VAN.
FVCX Field # Vignetting factor VCX.
FVCY Field # Vignetting factor VCY.
FVDX Field # Vignetting factor VDX.
FVDY Field # Vignetting factor VDY.
GCRS Ignored The global coordinate reference surface.
GLSS Surface # Glass.
GPJX Ignored Global Jones polarization vector component Jx.
GPJY Ignored Global Jones polarization vector component Jy.
GPIU Ignored Global polarization state "is unpolarized", 1 if polarization state is
unpolarized, otherwise state is polarized.
GPPX Ignored Global polarization state phase x.
GPPY Ignored Global polarization state phase y.
HOLD Ignored Holds data in the multi-configuration buffer, but has no other effect.
Useful for temporarily turning off one operand without losing the
associated data.
LTTL Ignored Lens title. When using LTTL, the lens title is limited to 20 characters.
Chapter 17: MULTI-CONFIGURATIONS 472

MABB Surface # Model glass Abbe.
MCOM Surface # Surface comment.
MDPG Surface # Model glass dPgF.
MIND Surface # Model glass index.
MOFF Ignored An unused operand, may be used for entering comments.
NCOM Surface, Object Modifies the comment for non-sequential objects in the NSC Editor.
NCOT Surface, Object, Modifies the coating on each coating scatter group for non-sequential
CSG# objects in the NSC Editor.
NGLS Surface, Object The material type for non-sequential objects in the NSC Editor.
NPAR Surface, Object, Modifies the parameter columns for non-sequential objects in the
Parameter NSC Editor.
NPOS Surface, Object, Modifies the x, y, z, tilt x, tilt y, and tilt z position values for non-
Position sequential objects in the NSC Editor. The position flag is an integer
between 1 and 6 for x, y, z, tilt x, tilt y, and tilt z, respectively.
NPRO Surface, Object, Modifies various properties of NSC objects. Property is an integer
Property value indicating what data is controlled:
1 - Inside of object number
2 - Reference object number
3 - Do Not Draw Object flag
4 - Rays Ignore Object flag
201-212 - User defined gradient index parameters
301-312 - User defined diffraction parameters
401-412 - User defined bulk scatter parameters
PAR1 Surface # Parameter 1. Obsolete, use PRAM instead.
PRAM Surface, Parameter value. This operand controls any of the parameters. See
Parameter Parameter data on page 225.
PRES Ignored Air pressure in atmospheres. Zero means vacuum, 1 means normal
air pressure. See Environment on page 90.
PRWV Ignored Primary wavelength number.
PSP1 Surface # Parameter solve parameter 1 (the pickup surface). This operand
requires 2 numerical arguments: the surface number and the
parameter number.

PSP2 Surface # Parameter solve parameter 2 (the scale factor).This operand requires
2 numerical arguments: the surface number and the parameter
number.
PSP3 Surface # Parameter solve parameter 3 (the offset).This operand requires 2
numerical arguments: the surface number and the parameter
number.
PUCN Ignored Used for picking up a range of values from a previous configuration.
If a positive integer configuration number is provided, then all values
below the PUCN operand will be picked up from the configuration
number specified. If the configuration value is negative, then a
negative pickup while be used. If the configuration number is zero,
then the values below the PUCN operand will not have pickup solves
applied. Note two PUCN operands can be used to define the
beginning and end of a range of values to be picked up. All specified
configuration numbers must be less than the configuration the PUCN
data is provided for.
RAAM Ignored Ray aiming. Use 0 for off, 1 for on.
SATP Ignored System aperture type. Use 0 for Entrance Pupil Diameter, 1 for Image
Space F/#, 2 for Object Space NA, 3 for Float By Stop Size, 4 for
Paraxial Working F/#, 5 Object Cone Angle. See also APER.
SDIA Surface # Semi-diameter.
STPS Ignored Stop surface number. The stop can be moved to any valid surface
number (excluding the object and image surfaces) by specifying an
integer argument for each configuration.
TELE Ignored Telecentric in object space, 0 for no, 1 for yes.
TEMP Ignored Temperature in degrees Celsius. See Environment on page 90.
THIC Surface # Thickness of surface.
TSP1 Surface # Thickness solve parameter 1. See SUMMARY OF SOLVES on
page 379.
page 379.
page 379.
UDAF Surface # User defined aperture file. Surface must use either a user defined
aperture or user defined obscuration aperture type. See Surface
properties aperture tab on page 68.
WAVE Wave # Wavelength.
WLWT Wave # Wavelength weight.
XFIE Field # X-field value.
YFIE Field # Y-field value.
Defining the number of configurations

The number of configurations (also called the number of zoom positions) is changed by inserting or deleting
configurations using the Edit menu option.

Defining each configuration
To define a multi-configuration operand, double click on the name of the operand you wish to change (the
names are listed in the left most column). For example, suppose you want to enter in a multiple value for the
thickness of surface 5. Assume you want three different configurations. Insert 2 new configurations using the Edit
menu. Double click on the left most column in row 1. Select "THIC" from the drop-down list on the dialog box for
the Operand Type. Select "5" for the surface number, then click on OK. Now in each configuration column, enter
the desired thickness in that configuration.
Adding and removing elements
One application for the multi-configuration feature is to design systems that can be operated with or without
certain elements being present. This can easily be handled by making "phantom" elements in some of the
configurations. The trick is to define the element (really the first surface of the element) to have a glass type which
changes with the configuration. Suppose in configuration 1 the element is present and is made of BK7. In
configuration 2, the element is absent. Using the GLSS operand, enter in "BK7" for configuration 1 and leave the
field for configuration 2 blank. Although the surfaces which define the element are still present in the lens
prescription, they have no effect because the glass has been removed (the element becomes a pair of dummy
surfaces). Note that the number of surface must remain constant in all configurations.
Changing configurations
To perform any analysis, proceed as usual. The program will use the current configuration for all calculations
and graphics. To change configurations, double click on the column header at the top of the configuration you
wish to change to on the MCE. The shortcut keys Ctrl-A and Shift-Ctrl-A also work.
Optimization with multi-configurations
ZEMAX will optimize multi-configuration data as readily as the conventional spreadsheet data. To make a
multi-configuration parameter variable, place the cursor on the parameter, and press Ctrl-Z. This is a toggle;
pressing Ctrl-Z again would eliminate the variability. When optimization is invoked, the optimization will automat-
ically count the new variable. As many multi configuration variables as you like may be defined.
To optimize across configurations, select "Default Merit Function" from the Tools menu of the merit function
editor. ZEMAX will build an appropriate merit function for you. Optimization across configurations is performed
using the CONF operand. This special operand changes the current configuration during the evaluation of the
merit function. This means all operands defined after CONF will be relevant to the new configuration. CONF may
be used multiple times in the operand sequence to evaluate various parameters.
Boundary constraints and other user entered optimization operands in multiple-configuration merit functions
are evaluated, and therefore enforced, only in the configuration in which they are defined. For example, if a CONF
1 operand is followed by various operands such as EFFL or REAY, these will only be evaluated in configuration
1. To enforce these same operands in configuration 2, the same operands need to be repeated under the CONF
2 operand.
The advantage to this system is that the entered operands, or their respective targets or weights, may be
different in each configuration. The disadvantage is the need to copy the operands that apply in more than one
configuration to each configuration.
Suggestions for organizing multiple-configuration merit functions
There are two common different ways to organize a multiple-configuration merit function. The first way is to
add the user defined operands within each CONF group of a default merit function:
CONF 1
User operands for configuration 1...
Default operands for configuration 1...
CONF 2:
CONF 3:
...etc.
The other method is to build the default merit function, then add all the user operands at the top of the merit
function to keep them all in one place:

CONF 1
CONF 2
CONF 3
etc...
DMFS
CONF 1
CONF 2
CONF 3
etc...
Both merit functions will accomplish the same job, but the first method executes faster because the overhead
in changing between configurations is reduced. The second method is easier to edit and maintain. Note the use
of the DMFS operand after the user defined operands. This operand serves as a marker, so that when the default
merit function is reconstructed, it will be appended after the DMFS operand, and the user entered operands are
not lost.
Note that if you change the field angles, heights, or weights, or the wavelength values or weights on the multi-
configuration screen you should rebuild the default merit function. When the default merit function is constructed,
it uses the data for each configuration to determine the rays traced, and the appropriate weighting.
!Note that if you change the field angles, heights, or weights, or the wavelength values or
weights on the multi-configuration screen you should rebuild the default merit function.
Using solves with multi-configuration data

Two types of solves are supported on the multi-configuration screen: a pick up from another operand and
configuration, with an optional scale factor and offset, and a thermal pick up solve.
MCE Pick up solves
Suppose there are three configurations, and in two of the configurations the values for one of the rows (say a
THIC or GLSS) need to be the same value. A pick-up solve on one of the configurations for that operand can be
used to ensure they are always the same.
To set a pickup solve, double click on the row and configuration on which the solve is to be placed. A dialog
box will appear which allows definition of the solve type, the target configuration and operand, and a scale factor
and offset. The new cell value is defined by: new_value = target_value*scale + offset. Note the target_value may
be from any other cell in the MCE, as long as the target configuration and operand number is less than or equal
to the current configuration and operand number, respectively.
MCE Thermal pick up solves
ZEMAX-EE also supports a "Thermal Pick Up" solve which incorporates thermal effects; see the chapter
"Thermal Analysis" for details.

Chapter 18 USING GLASS CATALOGS
Introduction
There are several standard catalogs supplied with ZEMAX, and you may create your own custom catalogs.
The standard catalogs may be edited to suit your requirements. This chapter describes how to add glasses to the
existing or new catalogs, and how to use the catalogs in your optical designs. There is no limit to the number of
glass catalogs you may create.
ZEMAX computes indices of refraction from formulas and coefficients entered into the glass catalogs. When
you specify a glass name such as "BK7" in the glass column of the LDE, ZEMAX looks for the name in each of
the currently loaded glass catalogs. If the glass is found, ZEMAX uses the coefficients for that glass, and then
using the formula for the glass selected in the catalog, computes the indices at each of the defined wavelengths.
It is important to note that all ZEMAX glass catalogs assume that the index computed by the dispersion
formulas is the relative index of refraction computed as a function of the relative wavelength. Relative means
relative to air at 1.0 atmosphere and a temperature defined in the glass catalog.
!ZEMAX assumes that the index computed by the dispersion formulas is the relative index of
refraction computed as a function of the relative wavelength.
This method may seem more complex than directly entering in the indices of refraction, but the advantages
are numerous. For one, the formulas are generally more accurate than user-entered data. Catalog data is more
convenient as well, requiring the user to only supply the glass name. This is a particular advantage during the
glass selection phase of the design. Additionally, any wavelength may be used, even if no explicit index data at
that wavelength is available. The primary disadvantage is that the coefficients must be calculated, although this
data is either readily available in catalogs or easily calculated. If you have the index data for some material not
in the catalog, or if you feel your data is better than the catalog data, ZEMAX will compute the coefficients for you
automatically; see the section "Fitting index data".
!For important comments about obsolete glasses and differences between glasses with
identical names, see Obsolete catalog data on page 490.
Specifying which glass catalogs to use

This chapter describes the loading, editing, and managing of the glass catalogs. To specify that a particular
catalog be used for a particular lens, use the Glass Catalog tab of the System, General dialog box. The default
entry is "schott" which refers to the SCHOTT.AGF glass catalog. You may specify an alternative, or additional
catalogs on this line. To specify an alternative catalog, enter the name of the catalog (with no extension). You may
also specify multiple simultaneous catalogs by listing them with spaces between the names. For example, to use
the "schott" and the "hoya" catalogs, enter "schott hoya". To use the "ohara", "schott", and "infrared" catalogs,
enter "ohara schott infrared". Because ZEMAX uses spaces as delimiters between glass catalog names, spaces
are not allowed within the names of the glass catalog files.
All the listed catalogs must be in the \ZEMAX\GLASSCAT subdirectory of the main ZEMAX directory. The
reason that glass catalog names are specified on this screen is that the catalog choice is stored with each lens
separately. If you now save the lens, and reload it at a later time, the correct glass catalogs, and only those
catalogs, will automatically be loaded. You can specify a different combination of catalogs for every lens you
design, if you like.
Editing and reviewing glass catalogs
To edit or review data in an existing glass catalog, select Tools, Glass Catalogs. Select the catalog name from
those listed in the drop-down list on the dialog box. Once the catalog is selected, you may insert, cut, copy, paste,
or modify data in the catalog, as is described in the following sections. You can save the newly modified catalog
to either the same name or a new name. When editing the glass catalogs supplied with ZEMAX, be sure to save
the modified data to a new name using the "Save Catalog As" button. This is important because future releases
477 Chapter 18: USING GLASS CATALOGS

of ZEMAX may include an updated catalog which will be installed over the existing catalog, and any changes that
had been made to the existing catalog will be lost.
!Be extremely careful when editing the glass catalogs; erroneous ray trace data will result if
the catalog data is incorrectly modified.
Description of catalog data

The glass catalog dialog box displays a great deal of data about each individual glass. The data fields are
described in the following table.
GLASS CATALOG DATA FIELDS
Item Description
Catalog Used to specify which of the .AGF format catalogs to display. The comment string
to the right of the catalog name may be used to describe the catalog.
Glass Used to specify which glass within the catalog to display data for.
Rename In the event that a glass needs to be given a different name, the name of the
currently selected glass can be modified in this field.
Formula The dispersion of each glass is described by a formula. This control allows
selection of which formula is used. If this setting is changed, then the dispersion
data becomes invalid unless the appropriate coefficients are also entered. See
The glass dispersion formulas on page 480.
Status The status indicates the general availability of the glass. The available settings are
Standard, Preferred, Obsolete, Special, and Melt. The status values are generally
specified by the manufacturer. Standard glasses are generally available for
purchase. Preferred glasses are usually frequently melted glasses, and more likely
to be available upon demand. Obsolete glasses are no longer manufactured, but
may be available. Special is a general category used to indicate a glass that does
not fall into one of the other categories. Melt is a flag used by ZEMAX to indicate
glasses that have been created in the catalog by the melt fit feature, see Fitting
melt data on page 482.
Index Nd The index at d-light, or 0.587 micrometers. ZEMAX does not use this number when
computing the index of refraction. It is displayed solely for reference. The entry
may be meaningless for some materials which do not transmit well in the region
around 0.587 micrometers.
Abbe Vd The Abbe value at d-light. This number is not used by ZEMAX when computing
the index of refraction. It is displayed solely for reference.
Exclude Substitution If checked, then this glass will not be selected during global optimization,
conversion from model to real glasses, or be considered by the RGLA optimization
operand.
Ignore Thermal This switch allows accurate thermal modeling of non-solid materials, such as
Expansion gasses and liquids, by allowing direct specification of the thermal coefficient of
expansion in the lens data editor rather than in the glass catalog. Only the edge
effects are considered; the radius of curvature and other thermal pickup solves will
all use the adjacent material TCE, rather than the gas or liquid TCE.
K1, L1... A0, A1..., A, B, The first eight rows in the center column of the dialog box display the dispersion
C, etc. coefficient data. The names of these coefficients changes depending upon the
glass formula.
Chapter 18: USING GLASS CATALOGS 478

Item Description
TCE The TCE value is the thermal coefficient of expansion. This is a dimensionless
parameter. The value displayed or entered needs to be multiplied by 1e-6 to yield
the actual values.
ZEMAX currently uses the TCE value to model linear thermal expansion
independent of the temperature range being used. The catalogs values supplied
by the glass vendor are usually the TCE for the temperature range from -30 to +70
degrees celsius, but ZEMAX will use the TCE value for any temperature range.
Temp The reference temperature in degrees Celsius for which the index data is defined.
See also Defining temperature and pressure on page 491.
D0, D1, D2, E0, E1, Ltk These are the thermal coefficients used by the thermal analysis model. See the
chapter "Thermal Analysis" for details.
p p is the material density in grams per cubic centimeter.
dPgF The value is the deviation of the relative partial dispersion from the normal line.
Min Wave, Max Wave The minimum and maximum wavelengths in micrometers over which the
dispersion formula returns valid index data.
Melt Frequency This integer value indicates the relative frequency with which the glass is melted
by the manufacturer. The convention used is 1 means the glass is melted very
frequently, 2 means less frequently, etc. up to a value of 5 to indicate a very
infrequently melted glass.
Comment This is an optional comment specific to the individual glass.
Relative Cost This number is intended to indicate the approximate relative cost of the glass as
compared to BK7. For example, a value of 3.5 would indicate the glass costs about
3.5 times as much as BK7 per pound. The number is only intended as a rough
guide. The exact glass cost may vary with the quality, amount, form, and
availability of the glass being purchased. Contact the glass vendor for more
information.
CR, FR, SR, AR, PR These are the glass codes that indicate how resistant the glass is to various
environmental effects. The codes correspond to Climate Resistance (CR), Stain
Resistance (FR), Acid Resistance (SR), Alkali Resistance (AR), and Phosphate
Resistance (PR). Generally, the lower the number (zero being best) the more
durable the glass is. For a complete description of these codes and the test
conditions used to measure them, contact the glass vendor.
Creating a new catalog

To create a new glass catalog, choose "Save Catalog As" and specify a new name. After the new catalog is
created, any unwanted data may be removed using the "Cut Glass" button.
!Note that ZEMAX does not allow any spaces in the names of glasses or glass catalogs.
Copying or moving glass catalog files
To copy or move a glass catalog file, copy or move only the file ending with the AGF extension. When ZEMAX
reads the catalog with the AGF extension, a file with the same name will be created with a BGF extension. For
example, ZEMAX will read in the file SCHOTT.AGF and create a file named SCHOTT.BGF. The BGF version is
a binary format version of the same data stored as ASCII in the AGF file. ZEMAX will subsequently read the BGF
version of the glass catalog because the binary version can be read much faster than the ASCII version of the
file. ZEMAX will automatically recreate the BGF version if the date stamp of the AGF file changes, or if the BGF
file is deleted or missing, or if the ZEMAX version number changes.

The BGF file is of no value without the AGF file; so if a glass catalog must be copied, moved, or sent via email
to another directory or location, copy the AGF file and not the BGF file. ZEMAX will automatically recreate the
BGF whenever it is needed.
The glass dispersion formulas
The coefficients in the catalog are used in any one of several polynomial formulas that ZEMAX recognizes.
There is also a dispersion formula described by the six-digit MIL number, but those indices are calculated directly
from the MIL number entered on the spreadsheet. The MIL number formula is not part of the glass catalog, and
so it will not appear. See Using MIL number glasses on page 485 for a discussion of the MIL number glass
formula. You may add new glasses to the currently loaded catalog if you have index data given in the form of one
of the following equations. In all of the equations is in micrometers.
The Schott formula

The Schott constants of dispersion formula is
2 2 2 4 6 8
n = a0 + a 1 + a 2 + a3 + a4 + a5 .
The required coefficients are available in most manufacturers glass catalogs. Schott no longer uses this
formula, but it is widely used by other glass manufacturers. See also The Extended formula on page 482.
The Sellmeier 1 formula
The Sellmeier 1 formula is
2 2 2
2 K1 K2 K3
n 1 = ----------------
2
- + ----------------
2
- + ----------------
2
-.
L1 L2 L3
Coefficients for all three terms may be entered to describe the material, although fewer terms may be used.
See also the Sellmeier 3 and Sellmeier 5 formulas.
The Sellmeier 2 formula is
2
2 B1 B2
n 1 = A + ----------------
2
- + ----------------
2 2
-.
2
1 2
Only two terms are used, there is no wavelength dependence in the numerator of the second term, and there
is a constant term.
The Sellmeier 3 formula is just like the Sellmeier 1 formula, with one additional term added:
2 2 2 2
2 K1 K2 K3 K4
n 1 = ----------------
2
- + ----------------
2
- + ----------------
2
- + ----------------
2
-.
L1 L2 L3 L4

The Sellmeier 4 formula is:
2 2
2 B - --------------
D
n = A + --------------
2
+ 2 -.
C E

The Sellmeier 5 formula is just like the Sellmeier 3 formula, with one additional term added:
2 2 2 2 2
2 K1 K2 K3 K4 K5
n 1 = ----------------
2
- + ----------------
2
- + ----------------
2
- + ----------------
2
- + ----------------
2
-.
L1 L2 L3 L4 L5
The Herzberger formula

The Herzberger expression is
2 2 4 6
n = A + BL + CL + D + E + F ,
1
L = ------------------------
2
-.
0.028
The Herzberger formula is used mainly in the infrared spectrum.

The Conrady formula
The Conrady formula is
A B-
n = n 0 + --- + -------- .
3.5
The Conrady formula is extremely useful for fitting to sparse data. For example, if you have only three index-
wavelength pairs of data, fitting to the six-term Schott formula would yield meaningless data at the intermediate
wavelengths.
The Handbook of Optics 1 formula
There are two similar formulas from the Handbook of Optics. The "Handbook 1" formula is:
2 B - 2
n = A + -------------------
2
D ,
( C)
The Handbook of Optics 2 formula

The "Handbook 2" formula is:
2
2 B - 2
n = A + -------------------
2
D .
( C)

The Extended formula
The Extended constants of dispersion formula is
2 2 2 4 6 8 10 12
n = a 0 + a 1 + a2 + a3 + a4 + a5 + a6 + a7 .
This is similar to the Schott formula, with two additional terms added.
General comments of using dispersion formulas
It is important to note that some publications use equations similar, but not identical to any of these
expressions. It is often possible to rearrange the expressions to get them into the required form, and then
recompute the required coefficients.
It is also a good idea to check the coefficients against a tabular listing of the index data available in many
handbooks and publications. Use the dispersion plot or listing feature, or the prescription data report, which lists
index data at each surface. If there are discrepancies, check the data you have entered carefully, and verify that
the correct units and formula are being used.
Fitting index data
!See also the discussion "Fitting melt data" below.
It is often the case that the materials you are designing with are in the catalog already. If they are not, you can
enter the coefficients for the formulas described previously. As an alternative, ZEMAX will compute either the
Schott, Herzberger, Conrady, or Sellmeier 1 dispersion formula coefficients for you. With the glass catalog dialog
box displayed, click on "Fit Index Data" and the Fit Index Data dialog box appears.
On the left side of this screen is a two-column spreadsheet editor. Using the mouse, enter in the wavelength
(in micrometers) and index data you have. The more data you enter, the more accurate will be the fit.
If you have more data than will fit in the spreadsheet, use the data most closely representing the wavelength
region of interest. At least three points are required to get a good fit if you are using the Conrady formula, six or
more (and preferably twelve to fifteen) for the Schott, Herzberger, or Sellmeier 1 formulas. Select the formula to
use by selecting the formula name from the drop-down list. You may want to try each of the models in turn, to see
which gives the lowest residual.
The RMS error is the RMS fit error between the given data, and the index data generated using the resulting
fit coefficients. The max error is the largest error between the fit and any one data point. Both numbers can be
compared to the magnitude of the index of refraction; which is of course dimensionless. Because the Sellmeier
1 formula has non-linear coefficients, the fitting is iterative, and this formula takes much more computer time to
fit the data than the other formulas do.
Now move the cursor over to the "Name" field, and enter the material name for the catalog. Select "Fit" and
ZEMAX will compute the optimal coefficients. The residual RMS error, and the maximum single-point error, are
listed on the bottom of the display. To enter this data in the currently loaded catalog, select "Add to catalog".
ZEMAX will issue a message verifying that the glass was saved.
!When a glass is added to the catalog, the transmission data, if any, needs to be added as
described in the next section. Otherwise, the default internal transmission of 1.0 is used at all
wavelengths.
The index and wavelength data may also be saved to an ASCII file for later use, and loaded for fitting again
by selecting the appropriate buttons. The ASCII file may also be edited outside of ZEMAX, and then loaded for
fitting.
Fitting melt data
!See also the discussion "Fitting index data" above.
It is important to understand that the index of refraction values computed by ZEMAX, or listed in the catalog
of the glass manufacturer, are average values for the index of refraction over a large number of "melts" or batches
of the glass. A specific piece of glass from one melt will deviate from the catalog or nominal values slightly. The
deviation is typically small, but the difference between the nominal and actual index values may be important for
some systems.
Usually, when quality optical glass is shipped from the manufacturer, a data sheet will accompany the glass
which indicates the index of refraction for the supplied glass at a few wavelengths, either as an offset from the
nominal catalog values or as the measured index directly. Typically 3-5 wavelength-index data points are
provided. This data is called "melt" data because it is specific to the batch of glass melted at one time.
The Melt Data tool available from the glass catalog dialog box is a handy utility for converting the limited melt
data provided into a usable new glass type in the glass catalog.
There are a maximum of 8 wavelength-index points allowed for melt data. If you have more than 8 points, use
the "Fit Index" tool described in the previous section. The minimum number of points allowed is 3, however, at
least 4 and preferably 5 points should be used to get a good melt fit. The wavelength range defined by the melt
index data should be as broad as possible, and should at a minimum cover the intended range of wavelengths
to be used for ray tracing through the melt glass. In all cases, the fitted data should be inspected carefully for
accuracy before it is used.
The Melt Data tool supports the following controls:
Glass: The name of the nominal glass in the selected glass catalog.
Melt Name: The name of the new glass to be created. The default is the nominal glass name with "_MELT"
appended on. The name length may not exceed 20 characters.
Fit only these wavelengths: If checked, the melt index fit will only be done over the wavelength range defined
by the minimum and maximum wavelengths of the provided melt data points. This allows much more precise
fitting of the glass data, however, the melt glass cannot be used outside of this wavelength range. If unchecked,
ZEMAX will attempt to extrapolate the data (see Discussion of melt fitting method: on page 483) to create a
melt fit that is valid over the entire wavelength spectrum of the original glass data. This is the key difference
between fitting index data (see Fitting index data on page 482) and fitting melt data.
Formula: The dispersion formula to use for the new melt glass. Choose either Schott, Herzberger, Conrady,
or Sellmeier 1. The default value is the Schott formula unless the nominal glass uses one of these formulas;
in which case the same formula as the nominal glass will be used.
Use: This box turns "on" and "off" each row of data.
Wavelength: The wavelength in micrometers for the index values to be entered.
Nominal: The index of refraction at the defined wavelengths using the nominal glass dispersion.
Actual: The actual measured index from the melt data. Note if the actual value is edited, the "delta" value is
automatically adjusted to keep the data consistent.
Delta: The difference between the actual and nominal index of refraction. Note if the delta value is edited, the
"actual" value is automatically adjusted to keep the data consistent.
Fit/Insert: Choosing this button starts the fitting process as described below.
Cancel: Aborts the melt fitting process.
After the fitting is finished, the new melt glass will be inserted in the catalog, the catalog will be saved, and a
report summarizing the fit will be presented.
Discussion of melt fitting method:
The problem with fitting melt data is the generally low number of points available; typically 3-5. Most fitting
routines need at least 8 points for good accuracy. So, the problem is to extrapolate from a few points the variation
in index over a large enough number of points to fit the resulting dispersion accurately. The accuracy of the
resulting melt fit at the defined data points depends largely upon the extent of the wavelength range. Greater
accuracy is achieved if "Fit only these wavelengths" (see discussion above) is checked, at the expense of a
reduced spectral range over which the fit is valid.
ZEMAX does melt fitting using the following algorithm:

First, a fit of the actual dispersion data is computed using the Conrady formula. The Conrady formula is used
because it is stable and reasonable when as few as three points are defined.
Then a Conrady fit of the nominal data is computed using only the defined wavelength points.
A large number of index points covering either the entire usable wavelength range (if "Fit only these
wavelengths" is unchecked) or the wavelength range defined by the melt points (if "Fit only these wavelengths"
is checked) of the nominal glass is generated. To each nominal index value, an offset is added which is the
difference between the two Conrady fits that were generated using only the melt data wavelengths.
Finally, the resulting data is fit using the selected formula (not necessarily the Conrady). This is the final fit
inserted into the catalog.
After the melt fitting is finished, a report summarizing the melt fitting process is presented. Check this report
carefully before using the new melt glass!
ZEMAX automatically copies over all transmission, density, cost factor, and other data from the nominal glass
to the melt glass.
!Check the generated melt fitting report for index accuracy carefully before using the new melt
glass!
Defining Transmission Data

Selecting the "Transmission" button invokes the transmission data editor within the glass catalog. Transmis-
sion refers to the intensity transmittance of light that depends upon the thickness of the glass as well as the
wavelength. ZEMAX models the transmitted intensity using Beer's law:

t = e ,
where is the absorption coefficient and is the path length through the glass. The parameter generally
depends upon wavelength and has units of inverse length. See the Chapter "Polarization Analysis" for information
on polarization ray tracing and transmission.
The transmission is defined in the glass catalog by a series of 3 numbers: the wavelength in micrometers, the
intensity transmittance, and the reference thickness in mm. For example, the transmission of a glass at 0.35
micrometers may be 0.65 for a thickness of 25 mm. Multiple data points may be defined in the transmission data
editor. Internally, ZEMAX converts the data to "per mm", and interpolates between defined wavelengths. If ray
tracing is being performed at a wavelength outside of the defined wavelengths, then the data for the closest
wavelength is used; otherwise, ZEMAX performs a linear interpolation.
!Not all of the glasses listed in the supplied catalogs have valid transmission data, especially
for infrared materials and other non-commercial glass types. If transmission data is supplied
by the manufacturer, it is usually included. If no reliable data is available, or if the data has
been omitted, the default internal transmission of 1.0 is used at all wavelengths.
Modeling gases and liquids

Once a material is defined in the glass catalog, ZEMAX uses the TCE specified for that material to determine
the thermal expansion of the radius, center thickness, and other data for the surfaces using the material. However,
if the material is not a solid, but is instead a gas or a liquid, then generally the thermal expansion is not governed
by the material properties, but is instead determined by the edge thickness of the mounting material.
In this special case, ZEMAX needs to use the TCE supplied in the Lens Data Editor to define the mounting
material properties, rather than the TCE supplied in the glass catalog. This can be accomplished by setting the
"Ignore Thermal Expansion" switch for the material in the glass catalog.

Finding a glass quickly
The fastest way to view the data for any glass is to click once on the name of the glass in the Lens Data Editor,
then select Tools, Glass Catalogs (or select the GLA button if displayed). The correct catalog and glass will be
displayed.
Using MIL number glasses
MIL number glasses are those described by a six-digit number, such as 517640 for BK7. The first three digits
in the MIL number is the d-light index minus one, without the decimal place. The last three digits is the Abbe V-
number times 10. You can enter a six-digit number directly in for the glass rather than the glass name if you desire.
ZEMAX uses a formula for computing the index at each defined wavelength based upon the index and Abbe
number defined by the MIL number. The formula is based upon a least-squares fit of coefficient data of many
typical glasses. Typically, the index data calculated is accurate to roughly .001. At wavelengths in the deep UV
or infrared the index value becomes less reliable. MIL number glasses are generally inferior substitutes for the
constants of dispersion (or other) models for the glass, however they are useful if no other data is available.
!Note MIL number glasses are an approximation, although usually a very good approximation
in the visible range. Outside the visible wavelength range, such as in the ultraviolet or infrared,
the MIL number glass is not accurate and should not be used.
It is important to note that the indices calculated from the six-digit MIL number are not the same as those
calculated from the glass catalog, even if the MIL number you are using corresponds to a glass in the catalog.
Index data is calculated directly from the MIL number entered on the main screen; not from the glass catalog data,
even if a glass with that MIL number is in the catalog.
Because any glass with a six digit name is assumed to be a MIL number glass defined by ZEMAXs internal
equations, six digit numbers may not be used for glass names defined in the glass catalogs.
Using model glasses
ZEMAX can idealize the dispersion of glass using the index at d-light (.5875618 micrometers), the Abbe
number, and a term which describes the deviation of the partial dispersion from the "Normal Line". The index at
d-light is given the symbol Nd. The Abbe number (also called the V-number) is given the symbol Vd and is defined
by
Nd 1
V d = -------------------- ,
NF NC
where N F and N C are the indices of refraction at 0.4861327 and 0.6562725 micrometers, respectively. The
partial dispersion term is P g, F .
ZEMAX uses a formula based upon the typical dispersion of standard glasses in the visible range to estimate
the index at any defined wavelength within the visible range as a function of the Nd and Vd values. This formula
is accurate to roughly 0.0001 for typical glasses.
The Nd, Vd, and P g, F values are specified on the glass solve dialog box, which can be reached from the
Lens Data Editor.
See also Using model glasses on page 424 and Optimizing glass selection on page 437.
!Note model glasses are an approximation, although usually a very good approximation in the
visible range. Outside the visible wavelength range, such as in the ultraviolet or infrared, the
model glass is not accurate and should not be used.

Glass catalog sources
The data for the glass catalogs included with ZEMAX are generally provided by the manufacturers. The data
has been converted to ZEMAX format. The following vendors have supplied data for the ZEMAX glass catalogs:
GLASS VENDORS
Catalog Vendor Address Phone Website
ARCHER Archer OpTx 3412 Enterprise Drive, 972-463-8001 www.archeroptx.com

Rowlett, TX 75088
CORNING Corning France Optical 7 bis avenue de Valvins, F- 33-1-6469-7590 www.corning.com

Division 77211 Avon Cedex, France
CHENGDU Chengdu Guangming No. 720 Huan-Chen Rd., 86-28-4332218 www.xinhuglass.com

Jiading District, Shanghai
China
HEARAEUS Heraeus Quarzglass Quarzstrabe, D-63450 49-0-618-1350 www.heraeus.de

GmbH & Co. KG Hanau
HIKARI Hikari Glass USA, Inc. 2711 Aliso Creek Road, 949-831-4908 www.hikariglass.com
Suite 170, Aliso Viejo, CA
92656
HOYA Hoya Optics, Inc. 3400 Edison Way, Fremont, 510-252-8370 www.hoya.co.jp
CA 94538
LZOS Russian optical glass 095-552-1486 www.lzos.ru/indexe.htm

fabricator
OHARA Ohara Corporation 23141 Arroyo Visa, Suite 949-858-5700 www.oharacorp.com

200, Rancho Santa
Margarita, CA 92688
PILKINGTON Pilkington Special Glass Glascoed Road, St Asaph, 44-1-7455-83301 www.pilkington.com

Limited Clwyd, LL18, OLL, United
Kingdom
SCHOTT Schott Glass 400 York Avenue, Duryea, 570-457-7485 www.schott.com

Technologies PA 18642
SUMITA Sumita 4-7-25 Harigaya, Urawa- 81-48-832-3165 www.sumita-opt.co.jp

Ku,Saitama-City, Saitama,
330-8565 Japan
UMICORE Umicore Electro-Optic Watertorenstraat 33, 2250 +32 14 24 55 97 www.optics.umicore.com

Materials Olen, BELGIUM
ZEON Zeon Chemicals L.P. Furukawa Sogo Bldg., 2-6-1 33-216-2335 www.zeonchemicals.com
Marunouchi, Chiyoda-ku,
Tokyo 100-8323, Japan
Although the data contained in the glass catalogs is generally reliable, it is always possible for errors to be
made during translation or editing of the data. It is absolutely crucial that all index data be verified for accuracy
by the end user! This is especially true where fabrication of the optics is being considered.
!The catalog data may be in error, and it needs to be verified before it can be trusted.
The data contained in other catalogs provided with ZEMAX, such as the BIREFRINGENT, INFRARED, and
MISC catalogs have been compiled from published sources as described in the following table. Some materials,
such as calcite, are defined in more than one catalog. Materials may be included in the catalog which are not
referenced in the table, or the source is specified as "unknown". These materials should be especially scrutinized
before trusting the data accuracy.

BIREFRINGENT CATALOG DATA SOURCES
Material Source
ADP Handbook of Optics Vol. II
AG3ASS3 Handbook of Optics Vol. II
AGGAS2 Handbook of Optics Vol. II
AGGASE2 Handbook of Optics Vol. II
AL2O3 Handbook of Optics Vol. II
ALN Handbook of Optics Vol. II
BATIO3 Handbook of Optics Vol. II
BBO Handbook of Optics Vol. II
BEO Handbook of Optics Vol. II
CALCITE Handbook of Optics Vol. II
CAMOO4 Handbook of Optics Vol. II
CAWO4 Handbook of Optics Vol. II
CDGEAS2 unknown
CDS Handbook of Optics Vol. II
CDSE Handbook of Optics Vol. II
CUGAS2 Handbook of Optics Vol. II
KDP Handbook of Optics Vol. II
LAF3 Handbook of Optics Vol. II
LINBO3 Handbook of Optics Vol. II
LIO3 Handbook of Optics Vol. II
LIYF4 unknown
MGF2 M.J. Dodge, "Refractive Properties of Magnesium Flouride," Applied Optics, Vol. 23, No.
11: p1980 (1985)
PBMOO4 Handbook of Optics Vol. II
QUARTZ Handbook of Optics Vol. II
SRMOO4 Handbook of Optics Vol. II
TAS Handbook of Optics Vol. II
TE Handbook of Optics Vol. II
TEO2 Handbook of Optics Vol. II
YV04 Handbook of Optics Vol. I
ZNGEP2 Handbook of Optics Vol. II
ZNO Handbook of Optics Vol. II

INFRARED CATALOG DATA SOURCES
Material Source
ALN Handbook of Optics Vol. II
ALON Handbook of Optics Vol. II
AMTIR1 Amtir Spec Sheet
AMTIR3 Amtir Spec Sheet
BAF2 Handbook of Optics Vol. II
BEO Handbook of Optics Vol. II
CAF2 Handbook of Optics Vol. II
CALCITE Handbook of Optics Vol. II
CDSE Handbook of Optics Vol. II
CDTE Handbook of Optics Vol. II
CLEARTRAN Spec Sheet (Rohm & Haas)
CLEARTRAN_OLD Spec Sheet (Morton)
CSBR Handbook of Optics Vol. II
F_SILICA The Infrared & Electro-Optical Systems Handbook V. III
GAAS Amtir Spec Sheet
GEO2 Handbook of Optics Vol. II
GE_LONG JOSA, 47, 244 (57): 48, 579 (58)
GE_OLD JOSA, 47, 244 (57): 48, 579 (58)
GERMANIUM Handbook of Optical Constants of Solids
IRG2 Schott Spec Sheet
IRGN6 Schott Spec Sheet
KBR ISP Optics data sheet (www.ispoptics.com)
KCL ISP Optics data sheet (www.ispoptics.com)
KRS5 Handbook of Optics Vol. II
LIF Handbook of Optics Vol. II

Material Source
MGF2 M.J. Dodge, "Refractive Properties of Magnesium Flouride," Applied Optics, Vol. 23, No.
11: p1980 (1985)
MGO Handbook of Optics Vol. II
NACL The Infrared & Electro-Optical Systems Handbook V. III
PBF2 Handbook of Optics Vol. II
SAPPHIRE The Infrared & Electro-Optical Systems Handbook V. III
SILICON Handbook of Optics Vol. II
SRF2 Handbook of Optics Vol. II
SRTIO3 Handbook of Optics Vol. II
ZBLA Handbook of Optics Vol. II
ZBLAN Handbook of Optics Vol. II
ZNGEP2 Handbook of Optics Vol. II
ZNSE Handbook of Optics Vol. II
ZNS The Infrared & Electro-Optical Systems Handbook, V.III
MISC CATALOG DATA SOURCES
Material Source
ACRYLIC Handbook of Optics Vol. II
BASF5 Laikin, Lens Design
BASF55 Laikin, Lens Design
CAF2 Handbook of Optics Vol. II
CDS Handbook of Optics Vol. II
COC Hoechst Celanese Spec sheet
CR39 The Photonics Design and Applications Handbook, Optical Plastics: Properties and
Tolerances, H. D. Wolpert, pp. H-300 - H-307, (1991).
KDP Handbook of Optics Vol. II
LAF3 Handbook of Optics Vol. II
LIYF3 Handbook of Optics Vol. II
PMMA Handbook of Optics Vol. II
POLYCARB Handbook of Optics Vol. II
POLYSTYR Handbook of Optics Vol. II
PYREX Laikin, Lens Design
QUARTZ Handbook of Optics Vol. II
SAN Handbook of Optics Vol. II

Material Source
SEAWATER Laikin, Lens Design
SILICA Handbook of Optics Vol. II
TEO2 Handbook of Optics Vol. II
TYPEA Laikin, Lens Design
VACUUM Laikin, Lens Design
WATER Handbook of Optics Vol. I
Obsolete catalog data

Changes in environmental protection laws have required the discontinuation of the manufacture of many
optical glasses between the years 1990 - 2000. Roughly 2 out of 3 glasses which were listed in earlier catalogs
of the various glass manufacturers have been discontinued. In some cases, glasses were reformulated to comply
with new environmental restrictions. These new glasses may have identical or similar names as the old glasses,
depending upon the glass vendor. The reformulated glasses may have different index data than the glasses they
replace.
Obsolete glasses still exist in optical shops and may be used for new designs, if a supply can be located. Also,
many older designs using the old glasses need to be modeled and ray traced, especially when designing new
optics to work in harmony with the existing optical system. For these reasons, data for obsolete glasses are still
provided with the catalogs distributed with ZEMAX. Obsolete glasses are indicated by the "obsolete" status
displayed in the glass catalog dialog box.
Because some new glasses may have the same name as old glasses, although the exact composition may
have changed, optical engineers need to be especially vigilant about checking the index data predicted by the
software against the melt sheets of the glass which will actually be used.
Never blindly trust the accuracy of the index computed using catalog coefficients. There are numerous sources
of potential error, such as the measurement of the original sample, the fitting of the data, the typing in of the data
in the manufacturers catalog, and then retyping into ZEMAX, and finally, the ZEMAX code itself.

Chapter 19 THERMAL ANALYSIS
Introduction
Environmental factors such as ambient temperature and pressure can affect the performance of optical
systems. There are three primary factors to consider. First, the index of refraction of glass depends upon both
temperature and wavelength; relative indices which are measured with respect to air also change with pressure.
Second, glass expands and contracts with temperature, which can change the radius and thickness of a lens.
Third, the spacings between lens elements can change due to the expansion and contraction of the mounting
material.
The thermal analysis features of ZEMAX can account for all these effects. By accounting for thermal effects,
ZEMAX can be used to analyze and optimize a design for any specific temperature or for a range of temperatures.
The index of refraction data given by the dispersion formulas are referenced to a specific temperature and
pressure, which is typically 20 or 25 degrees Celsius (depending upon the manufacturer) and 1 atmosphere. Also,
index of refraction data is by convention referenced to that of air, which means thereafter the air has an index of
unity. The index referenced to air is called the relative index of refraction. When the index is referenced to vacuum
(which truly has an index of unity) then the index is called the absolute index of refraction. The difference between
these two references for any glass is a function of wavelength, temperature, and pressure.
Defining temperature and pressure
Two user provided values define the environment: the ambient temperature in degrees Celsius and the
ambient air pressure measured in atmospheres. These values are set on the dialog box which can be found under
the Environment tab of the System, General dialog box. There is also a checkbox on this dialog box with the
caption "Use Temperature, Pressure". If this checkbox is unchecked, then the temperature and pressure data are
ignored. If "Use Temperature, Pressure" is checked, then the temperature and pressure values are considered
during computation of the index of refraction.
By default, any defined temperature and pressure applies to all surfaces in the optical system. However, optical
systems which require multiple temperatures to be defined in the same system are also supported. This is
required for optical systems that have some lenses in a vacuum or in a heated area; while others are not.
Defining wavelengths
Wavelength data is entered on the "Wavelength Data" dialog box; see the Wavelengths on page 96 for
details.
!Wavelength data are always measured in micrometers referenced to "air" at the system
temperature and pressure.
Index of refraction computation

ZEMAX always uses relative, not absolute index data. One way to think about relative index is that the index
of "air" (indicated by a blank entry in the glass column) is always defined to be exactly 1.0 at all wavelengths at
the system temperature and pressure. Air and glass spaces at other temperatures and pressures are normalized
relative to 1.0.
There are several steps required for ZEMAX to compute the relative index data. The basic steps for computing
the index for each glass type at each wavelength are:
-Scale the wavelength to be in air at the reference temperature of the glass.
491 Chapter 19: THERMAL ANALYSIS

-Compute the relative index of refraction of the glass from the catalog dispersion formulas.
-Compute the index of air at the reference temperature of the glass.
-Compute the absolute index of the glass (relative to vacuum) at the reference temperature of the glass.
-Compute the change in absolute index of refraction of the glass at the specified temperature.
-Compute the index of air at the system temperature and pressure.
-Compute the index of the glass relative to the air at the system temperature and pressure.
The end result, which is the index of the glass at the specified temperature and pressure relative to air at the
system temperature and pressure, is what is used by ZEMAX for ray tracing. The index of a surface made of air
is treated in the same way. This means that air surfaces at different temperatures and pressures from the system
temperature and pressure will have slightly lower or higher index. For example, if the system pressure is 1.0 and
the pressure of one air surface is set to 0.0, the index of this one surface will be approximately 0.99973. If the
system pressure is 0.0 and the air space has a pressure of 1.0, the index for that surface will be approximately
1.00027. Remember, air at the system temperature and pressure is defined to be 1.0, all other indices are relative.
!Air at the system temperature and pressure is defined to be 1.0, all other indices are relative.
Note also that ZEMAX can easily model systems used in a vacuum by changing the ambient air pressure to
zero. If only some surfaces are at vacuum, this can be set up using the TEMP and PRES commands on the multi-
configuration editor.
For the index of refraction of the glass, ZEMAX uses the dispersion formulas and data stored in the glass
catalog. For details, see Description of catalog data on page 478. For the index of air, ZEMAX uses the following
formula:
( n ref 1 )P
n air = 1 + -------------------------------------------------------------------------
3
-
1.0 + ( T 15 ) ( 3.4785 10 )
where
2 2
2949810 - --------------------
25540 - 8
n ref = 1 + 6432.8 + --------------------------
2
+ 2
1.0 10 ,
146 1 41 1
T is the temperature in Celsius, P is the relative air pressure (dimensionless) and is measured in
micrometers. This formula for the index of air is from F. Kohlrausch, Praktische Physik, 1968, Vol 1, page 408.
The change in absolute index of the glass with temperature is given by the following expression:
2
3 E 0 T + E 1 T
2
n 1 2
n abs = -------------- D 0 T + D 1 T + D 2 T + -----------------------------------
2 2
- ,
2n tk
where n is the relative index at the reference temperature of the glass, T is the change in temperature relative
to the reference temperature of the glass ( T is positive if the temperature is greater than the reference
temperature of the glass), and the six other constants are provided by the glass manufacturer to describe the
thermal behavior of the glass. This model for the change in the glass index was developed by Schott Glass
Technologies, Inc.
The six constants must all be provided in the glass catalog for the computation to be valid. The default values
of zero for all six constants yields zero for the change in index; therefore if no thermal data has been added to
the catalog, no thermal effects are considered. ZEMAX cannot compute thermal index variation for any glass type
Chapter 19: THERMAL ANALYSIS 492

without these six constants provided. However, some approximations are available if the six coefficients are not
available. See the section Adding thermal index variation data on page 495 for details.
Environmental effects on index for gradient index, MIL number, and model glasses
ZEMAX only partially considers environmental effects on the index of refraction of gradient index surfaces, MIL
number glasses, or model glasses which are described only by index and Abbe number. The relative index of
refraction computed using these methods is adjusted to account for the change in the index of surrounding air at
the system temperature and pressure. However, the change in the absolute index of these glasses with
temperature and pressure is not considered, because no "dn/dt" data is defined for these glass types. However,
it is possible to manually define the variations in properties across multiple environments using the multi-
configuration editor.
Defining multiple temperature and pressure values
To analyze or optimize an entire lens at a specific temperature and pressure, all that is required is to define
the relevant data on the environment tab of the general data dialog box described in Environment on page 90.
All the radii and thickness data are then assumed to be measured at that temperature, and ZEMAX computes
the index data accordingly. However, the real power of the thermal analysis feature comes into play when a lens
must be analyzed at or optimized for multiple environments, such as a broad temperature range or varying
altitudes (or both). There are several new issues these lens systems introduce:
A means of specifying the nominal temperature at which radii and thickness are measured must be provided.
The change in index, radii, and thickness must be accounted for as the environment changes.
The thermal effects on mounting material must be considered.
Some surfaces may be at one temperature or pressure, while other surfaces are at a different temperature or
pressure.
ZEMAX is up to the task in all particulars. The basic approach to setting up a multiple environment lens is:
Define the lens at some nominal temperature and pressure. This should be the environment under which the
lens will be fabricated. All radii and thickness data will be specified only at this temperature and pressure.
Now define additional configurations using the multi-configuration feature (see the chapter Multi-configura-
tions). In each additional configuration, the temperature and pressure will be specified, and special solves
called "thermal pickup solves" will be used to adjust the radii and thickness data for each configuration. The
multi-configuration operands that control temperature and pressure are TEMP and PRES.
TCE stands for thermal coefficient of expansion. When a glass element changes temperature, the linear
change in size is given by the expression
L' = L ( 1 + T ) ,
where L is a linear dimension, is the TCE, and T is the change in temperature. As the material expands,
the radius of curvature also expands. Therefore, both the thickness and the radii of a glass surface change linearly
with temperature. The assumption of linearity is only an approximation, but a reasonably good one for most
materials and temperature ranges.
The TCE coefficient is defined along with the glass dispersion data in the glass catalog. See the chapter "Using
Glass Catalogs" for details on the TCE data.
Defining which parameters consider thermal effects
On the multi-configuration editor, there is a special solve called "Thermal Pick Up". This pick up solve is used
to compute a new value for a multi-configuration parameter based up on the temperature and pressure of the
new configuration as compared to the "reference" configuration. Thermal pick up solves only affect data for certain
types of multi-configuration values, as described below.

Radius of curvature (CRVT) values
If the multi-configuration operand is a CRVT, then the curvature of that surface for that configuration is
computed from the curvature in the nominal configuration, the difference in temperature between the two
configurations, and the TCE for that material.
If the glass type for a surface is specified in the catalog (such as BK7 or F2) then the TCE from the catalog is
used. If the glass type is air (a blank entry) or "MIRROR" then there are two possibilities: if the previous surface
glass type is a catalog glass, then the TCE for that glass is used, otherwise, the TCE value entered in the TCE
column will be used. These rules have one important ramification: if an element is a cemented doublet, the TCE
for the first surface will be that of the first glass, while the TCE for the second and third surfaces will be that of the
second glass. ZEMAX ignores the "stress" induced in the common surface. This assumption may not be accurate
for a large temperature range. Mirrors surrounded by air use the TCE from the TCE column, mirrors preceded by
glass are assumed to be made of that same glass, and therefore use the TCE for the preceding glass.
Thickness (THIC) values
If the operand is a THIC, then there are two possibilities. If the surface is composed of a catalog glass, then
the TCE for that glass is used. Otherwise, the TCE specified in the TCE column of the surface is used. The TCE
column is used for entering user-defined TCE data for the material used to manufacture the mount. There is one
other important point: if the surface is not composed of a catalog glass, then the thermal expansion is computed
along a length of material which extends from the edge of the surface to the edge of the next surface. Since the
material expands along the length of the edges, not at the center thickness, this is a more accurate computation.
For example, suppose two lenses were separated by aluminum with a center thickness spacing of 80 mm. If
the sag of the first surface (the rear of the first lens) lens was -5 mm, and the sag of the second surface (the front
face of the second lens) was 8 mm, the total edge thickness is then 93 mm. If the TCE for aluminum is 23.50 E-
6, then under a temperature change of +20 degrees the edge thickness would change from 93 to 93.0437 mm.
Ignoring the change in diameter of the spacer and the change in the sag of the two lens faces (ZEMAX actually
accounts for both of these effects, but for simplicity they are ignored here), the center thickness would then change
to 80.0437. Note that this is a different amount of expansion than if the computation had only considered the
center thickness.
Because the expansion along the edge accounts for the change in radii of the adjacent surfaces, even a TCE
of 0.0 will result in a change of thickness if the radii change. To turn off thermal expansion of a thickness, do not
use a TCE of 0.0, rather, delete the thermal pickup solve entirely.
Parameter values
The thermal pick up solve properties for parameter values depends upon the parameter number and surface
type. If the parameter is interpreted as having units of length, then the appropriate scaling is performed just like
the radius of curvature described earlier. If the units are in powers of length, such as length squared or inverse
length, then appropriate scaling is also performed. Otherwise, the thermal pick up solve ignores thermal effects
and picks up the value from the nominal configuration.
Special surface types, such as polynomial aspheres, binary optics, holograms, and other surfaces that use
parameter or extra data need to have operands and thermal pickup solves added manually to the multi-
configuration editor; the automatic thermal setup tool does not add these operands automatically.
Extra data values
The thermal pick up solve properties for extra data values depends upon the surface type. Generally, only the
"normalization radius" needs to be scaled. This convenience is due to the fact that thermal expansion can be
considered a scaling of length. Since most extra data surfaces use dimensionless coefficients, only the
normalization radius needs to be scaled. This works effectively for binary optic and polynomial aspheric surfaces
equally well. For extra data surface types which do not use a normalization radius, the thermal pick up solve
ignores thermal effects and picks up the value from the nominal configuration.
Special surface types, such as polynomial aspheres, binary optics, holograms, and other surfaces that use
parameter or extra data need to have operands and thermal pickup solves added manually to the multi-
configuration editor; the automatic thermal setup tool does not add these operands automatically.

All other values
All other values are straight pick ups, and the value will be identical to that in the nominal configuration. Thermal
effects are ignored.
Defining multiple environments within a single configuration
Sometimes it is required to have different parts of the optical system at different temperatures and pressures.
Note this is different from an entire system being at different environments across multiple configurations.
Groups of surfaces can be assigned their own temperature and pressure using the TEMP and PRES multi-
configuration operands, even if there is only 1 configuration defined. The key is that each TEMP and PRES
operand defines the environment for all operands that follow in the multi-configuration editor. The last TEMP and
PRES operands listed in the editor define the "global" environment, which will apply to all data not listed in the
multi-configuration editor.
For example, suppose an optical system model requires surfaces 1-5 to be at a temperature of 20 degrees C,
while surfaces 6-10 are at 50 degrees C. The first operand listed should be TEMP (this same discussion applies
to PRES), defining the initial environment of 50 degrees C. All the curvatures, thicknesses, semi-diameters,
glasses, and other values for the surfaces 6-10 should be listed after the TEMP operand. Then the list should
end with another TEMP, this one defining the 20 degree "global" environment. The resulting system will be
evaluated at the respective temperatures (and/or pressures) for each surface.
It is extremely important to understand the two basic rules:
!1. All multi-configuration editor operands that follow a TEMP or PRES operand are evaluated
at that temperature or pressure.
!2. The last TEMP and PRES listed in the multi-configuration editor define the temperature and
pressure for all other lens data, on the multi-configuration editor or not.
By far the most important step in setting up a complicated multiple environment lens is to check the set up
carefully. Two excellent tools for doing this are the index of refraction data and multi-configuration data table on
the prescription report. This table lists the temperature and pressure for each glass type, and the thermal pick up
relationships.
Checking the thermal pick up solves on each parameter is also a good idea; the data should be at least partially
verified by hand to verify the correct temperature range and expansion are being used.
Automatic thermal setup
A tool for automatically setting up a lens for thermal analysis is described in Tools on page 78.
Adding TCE data
There are two kinds of TCE data. For surfaces which use a glass type named in one of the catalogs (such as
BK7 from the Schott catalog), ZEMAX uses the TCE data specified in the catalog. See the chapter "Using Glass
Catalogs" for the description of , the thermal coefficient of expansion value.
If the surface does not use a catalog glass, then the TCE value is extracted directly from the TCE column in
the Lens Data Editor. The TCE column is the last column in the spreadsheet, to the right of the parameter columns.
TCE is measured in units of 1E-06 per degree C. Therefore, a value of 23.50 E-06 per degree C would be
entered as 23.5. ZEMAX automatically considers the 1E-6 factor when computing thermal effects.
Modeling gases and liquids
See Modeling gases and liquids on page 484 for important information on thermal modeling of non-solid
materials.
Adding thermal index variation data
The variation of index of refraction with temperature, ambient air pressure, and wavelength is modeled for any
glass using the polynomial expression given earlier. The expression requires six coefficients to define the

temperature and wavelength dependence on the change in absolute index. It is often the case that for materials
added by the user, the six coefficients describing the variation are not available. However, most glass catalogs
contain at least a single linear approximation of the ratio of index change with temperature change. This value is
called dn/dt. If the only data available is a single dn/dt, then an approximation to the general expression can be
made assuming all the coefficients except D 0 are zero:
2
n 1
n abs = -------------- [ D 0 T ] ,
2n
which implies that a reasonable approximation to D 0 is given by
2n - dn
D 0 = -------------
2
------ .
n 1 dt
The D 0 needs to be calculated and entered into the glass catalog. The relative index at the center wavelength
measured at the reference temperature for the glass is an adequate value for the index n. Extreme care should
be taken to ensure this is an adequate approximation by subsequently checking the computed index values at
various wavelengths and temperatures. Note that the dn/dt data should be the absolute, not the relative dn/dt.
!Extreme caution and suspicion is warranted when using only a single dn/dt value.
Using a single dn/dt value to estimate the D 0 term is only a rough approximation. The actual index variation
with temperature is not likely to be linear over any extended wavelength or temperature range. Therefore, extreme
caution and suspicion is warranted when using only a single dn/dt value.
Optimizing athermal lenses
To optimize an athermal lens, first define the multiple configurations required to simulate the lens at each
temperature using the methods described in the previous sections. Then define the variables for optimization only
in the nominal configuration. For example, suppose the nominal configuration is number 1, and configurations 2,
3, and 4 are defined using thermal pick up solves for every curvature and thickness. Make only the curvatures
and thicknesses in the nominal configuration variable.
It is also possible to optimize the TCE of the spacing material between lens groups. To do so, set a variable
on the value in the TCE column of the Lens Data Editor.
Limitations of thermal analysis
There are several limitations to the accuracy of the thermal analysis capabilities of ZEMAX. First of all, the
TCE data should always be checked for accuracy over the temperature range of use. The index data coefficients
should also be verified with the manufacturer of the glass being used.
The thermal analysis does not necessarily work correctly with tilted, decentered, or otherwise unconventional
optical systems. The difficulty arises when computing edge thickness displacements on components that are not
symmetric; for example, between two lenses tilted with respect to one another.
The thermal index and TCE data provided for Schott glasses comes from Schott, and they state the data is
accurate over the temperature range from -40 to +80 degrees Celsius, and over the wavelength range 0.435 to
0.644 micrometers. The data may be extrapolated out to 1.06 micrometers with reduced accuracy. Data for other
glasses is provided as is and the range of accuracy is unknown.
Because of the complexity of thermal effects modeling, none of the data should be trusted in critical
applications and all computations, index values, and TCE data should be verified independently of ZEMAX. This
is true even when working with Schott glasses within the ranges specified above.

Chapter 20 POLARIZATION ANALYSIS
Introduction
Ray tracing programs generally treat rays as purely geometric entities, which have only a position, orientation,
and phase. For example, a ray is completely described at a surface by the ray intercept coordinates, the direction
cosines which define the angles the ray makes with respect to the local coordinate axes, and the phase, which
determines the optical path length or difference along the ray.
At the boundary between two media, such as glass and air, refraction occurs according to Snell's law. Usually,
the effects at the interface which do not affect beam direction are ignored. These effects include amplitude and
phase variations of the electric field which depend upon the angle of incidence, the incident polarization, and the
properties of the two media and any optical coatings at the interface.
Polarization analysis is an extension to conventional ray tracing which considers the effects that optical
coatings and reflection and absorption losses have on the propagation of light through an optical system.
Review of polarization concepts
The ZEMAX User's Guide is not intended to be a tutorial on polarization theory. The subject is extensive and
is treated well elsewhere. See for example "Polarization" by Jean M. Bennett, and "Optical Properties of Films
and Coatings" by John A. Dobrowski, both in The Handbook of Optics Volume I, McGraw Hill, 1995.
The electric field vector
The amplitude and polarization state of the electric field is described by a vector:
Ex
E = Ey ,
Ez
where Ex, Ey, and Ez are complex valued. The electric field vector must be orthogonal to the propagation
vector of the ray. At a boundary between two media, the transmittance, reflectance, and phase of the electric field
is different for the S and P components of the field. The S component of the field is the projection of E that lies
along the axis orthogonal to the plane of incidence, while the P projection lies within the plane of incidence. The
plane of incidence contains both the ray propagation vector and the surface normal vector at the intercept point.
The electric field is then divided into Es and Ep components, both of which are complex valued.
The electric field after the boundary is computed by the consideration of the S and P transmittance (if the
surface is refractive) or reflectance (if the surface is reflective). ZEMAX computes the reflectance and transmit-
tance coefficients, but only uses the one appropriate to propagation of the ray. For brevity, the rest of this
discussion will use the terms transmittance and refraction to indicate the terms of interest. If the surface is a
reflector (i.e. the glass type is MIRROR) then the reflectance coefficients will be used by the program automati-
cally. The electric field after refraction is then
E s' = Es s, E p' = Ep p ,
where the transmittance coefficients, s and p , are complex valued. After the Es and Ep projections are
computed, they are recomposed into the Ex, Ey, and Ez electric field vector and propagation continues to the next
surface. The computation of the transmittance coefficients depends upon the index of refraction of the incident
media, the (possibly complex) index of refraction and thickness of each layer in the optical coating (if any), and
497 Chapter 20: POLARIZATION ANALYSIS

the (possibly complex) index of refraction of the substrate. The details of this computation are provided in The
Handbook of Optics Volume I, McGraw Hill, 1995, and will not be repeated here.
Field vs. ray phase conventions
The thin film industry uses a different phase convention than is required for ray tracing. The thin film convention
is to measure the phase shift along the normal vector as an imaginary plane wave propagates from the outermost
coating layer to the substrate. This convention implies the phase shift is largest at normal incidence, and tapers
off to zero by approximately the cosine of the angle of incidence for larger angles of incidence. ZEMAX calls the
s and p values "Field" coefficients when using this reference. Although these values are computed and
displayed for comparison to thin film programs, they are not used by ZEMAX for ray tracing.
For ray tracing, the optical phase advance or delay is measured along the ray. ZEMAX traces rays directly to
the substrate, ignoring the coating thickness. This is because the thickness between surfaces specified in the
Lens Data Editor is the thickness between the substrates, ignoring coatings. The coating is presumed to grow
into the space between the surfaces. Correctly computing the phase of the ray requires back propagating the
electric field a distance equal to the physical coating thickness, and adjusting the coating phase to be measured
along the ray vector rather than the normal vector. ZEMAX calls the s and p values "Ray" coefficients when
using this convention, and always uses these values for ray tracing. A similar computation yields the factors for
reflective surfaces, or for surfaces where the coating stack has automatically been reversed, and ZEMAX
automatically chooses the correct factors in all cases.
The default in ZEMAX is to convert the "field" coefficients to "ray" coefficients. However, this conversion may
be disabled, see Convert thin film phase to ray equivalent on page 91.
The polarization ellipse
A convenient way of describing the polarization state of the electric field for rays parallel to the local Z axis only
is to ignore the Ez component and consider only Ex and Ey. By plotting on a cartesian graph the endpoint of the
vector (Ex, Ey) as time moves through one period, a figure is traced out which may be a straight line, a circle, or
most generally an ellipse. This curve is defined by the major and minor axis lengths, and the angle the major axis
makes with respect to the X axis on the cartesian graph. Note the minor axis may have a length of zero, in the
case of linear polarization. For circular polarization the major and minor axis lengths are identical.
Whenever the polarization ellipse is used, it is important to remember that the Ez component is ignored; which
limits the utility of the polarization ellipse in beams that are not nearly collimated along the local Z axis.
Definition of terms
There are a number of ways to define parameters of interest for describing polarized light beams. All of the
expressions generally depend upon angle of incidence, wavelength, and polarization orientation. Here are the
terms used frequently by ZEMAX and their definitions:
Intensity, I
The intensity is the sum of the squares of the individual components of the electric field, or equivalently, the
dot product of the electric field and its complex conjugate:

I = E E,

where the symbol E denotes the complex conjugate of the vector E . Subscripts are applied to the symbol I
to indicate which intensity is listed, for example, Ix would be the intensity in the x direction only.
Phase, P
The phase of any one component of the electric field can be computed with the arg function, which is the
arctan(imaginary / real). The phase is given by
P = arg ( E ) .
Chapter 20: POLARIZATION ANALYSIS 498

Subscripts are applied to the symbol P to indicate which intensity is listed, for example, Ps would be the phase
for the S component of the electric field.
Path length through medium,
For homogeneous medium, the path length is the distance in lens units along the ray from the previous surface
to the current surface. Note the index is accounted for separately; this is not the "optical" path length. For gradient
index medium, the effective path length (for purposes of computing internal absorption) is the actual integral of
the optical path length divided by the base index of the medium (the index at x=0, y=0, z=0). The path length is
only used for computing the internal absorption. The path length will be negative for virtual propagations.
Internal absorption per lens unit,
This factor accounts for the internal or bulk absorption propagating through glass. The absorption is computed
from the data provided in the glass catalog. See Defining Transmission Data on page 484.
Internal Transmittance, IT
The internal transmittance is computed using the path length and internal absorption :

IT = e .
Propagation Phase Factors, pc, ps

As the ray propagates from surface to surface, the electric field rotates according to
i
e , = ( 2i ) ,
where is the path length and is measured in the medium. The propagation factors are written as the cosine
and sine of the angle . The "E Field After" from the previous surface will be multiplied by these factors to yield
the input electric field to the current surface. ZEMAX adopts the negative sign convention for the propagation
phase factors to be consistent with the convention of negative imaginary index to indicate absorption in coating
materials as discussed in Defining coatings in ZEMAX on page 501.
Amplitude Reflectance,
The reflectance amplitude is the complex valued coefficient of reflectance for the electric field. These are
computed for both S and P polarizations and using both field and ray phase conventions, as described above.
Amplitude Transmittance,
The transmittance amplitude is the complex valued coefficient of transmittance for the electric field. These are
computed for both S and P polarizations and using both field and ray phase conventions, as described above.
Intensity Reflectance, Rs, Rp
Reflectance intensity is measured normal to the surface. It is given the symbol R, and is always real valued
between 0 and 1. Reflectance can be computed from the reflectance amplitude as follows:
R = * .
Intensity Transmittance, Ts, Tp

Transmittance intensity is measured normal to the surface. It is given the symbol T, and is always real valued
between 0 and 1. Transmittance can be computed from the transmittance amplitude as follows:
T = * .

Intensity Absorption, As, Ap
Absorption intensity is the intensity of the electric field which is neither transmitted nor reflected:
A = 1.0 T R .
Diattenuation, D
Diattenuation means "two attenuations", and is used to compare the loss of intensity of the S polarized light
compared to the P polarized light. Diattenuation is defined as
Ts Tp
D = ----------------
- .
Ts + Tp
Coating Phase, Ps, Pp

The phase of the transmitted beam is generally different for the S and P polarizations. The phase is given by
P s = arg ( E s ), P p = arg ( Ep ) .
Retardance, S
For transmission, the retardance is the phase of the P polarization minus the phase of the S polarization:
S = arg ( E p ) arg ( E s ) .
For reflection, the retardance is the phase of the P polarization minus the phase of the S polarization plus :
S = arg ( E p ) arg ( E s ) + .
The different conventions are used to agree with common practice in the thin films industry.
Phase Difference Between X and Y, Pxy
This quantity is the phase difference between the X and Y components of the electric field, or
Pxy = Px Py .
Note this value ignores the Z component of the electric field.

Major and Minor semi axis of the Polarization Ellipse, EM, Em
An ellipse has a major axis and a minor axis; the major axis being the longer of the two. These magnitudes,
along with the angle of the polarization ellipse (Ap) uniquely define the polarization state in the XY plane.
Angle of XY Polarization Ellipse, Ap
This parameter is the angle relative to the local +x axis that the major axis of the polarization ellipse makes. If
the polarization ellipse is a circle, this angle is meaningless.
Properties of uncoated surfaces
At normal incidence only, uncoated surfaces between dielectric media have transmittance and reflectance
amplitudes which are described by the Fresnel expressions:
n1 n2 2 4n 1 n 2
R = ----------------- , and T = -----------------------2- .
n1 + n 2 [ n 1 + n2 ]

At angles other than normal incidence, the relations are more complex. For example, a surface between air
and BK7 at a wavelength of 0.55 micrometers has reflectance and transmittance as a function of incident angle
as shown in the following figures.
Note that the reflectance increases substantially with incident angle, until nearly all of the light is reflected
rather than transmitted at grazing incidence angles. Note also that the S and P polarization states have different
transmittance. This leads to a polarization dependent apodization of the aperture.
Defining coatings in ZEMAX
For dielectric materials, the index of refraction is purely real, and therefore the imaginary part of the index is
zero. For metals, the index of refraction is complex. There are two sign conventions in common use for the
imaginary part of the index of refraction. ZEMAX uses the following convention:
= n + ik ,
where n is the usual index of refraction and k is the extinction coefficient, a negative value for absorbing
materials. The alternate sign convention is = n ik where k is then a positive value for absorbing materials,
but ZEMAX does not use this convention. For example, using the ZEMAX convention, the index of aluminum is
approximately given by
= 0.7 7.0i .
Note the extinction coefficient is negative using this convention for typical absorbing materials. This choice of
index sign convention is related to the sign convention used by the propagation phase factors, see Propagation
Phase Factors, pc, ps on page 499.
ZEMAX uses an ASCII file format to define all coating data. The file may be of any valid file name. A sample
file called COATING.DAT is supplied with ZEMAX, and COATING.DAT is the default coating file name for any new
lens. Multiple coating files may be defined, and the name of the coating file use by any particular lens file is defined
on the System, General, Files tab dialog box. It is highly recommended that any modifications to COATING.DAT
be saved in a file with a different name, so that subsequent updates to ZEMAX will not overwrite the changes
made to COATING.DAT.
In the coating file, The keywords MATE (for material), TAPR (for taper profile), COAT (for coating), TABLE (for
coatings defined by a table of points) and IDEAL or IDEAL2 (for ideal coatings) are used to define different types
of coating data. All of the material and taper definitions come first, then all of the coating definitions.
Coating file data syntax
The coating file consists of sections of data defining materials, tapers, and coatings. The data uses the
following format and syntax. Each data section is described in detail in the text which follows.

! Any line starting with the ! symbol is a comment line
! For information on the MATE format see "The MATE data section" below.
MATE <material name>
wavelength real imaginary
....
MATE <next material name>

....
! For information on the TAPR format see "The TAPR data section" below.
TAPR <taper name>
DX decenterx
DY decentery
AN rotation_angle (degrees)
RT termnumber radialterm
CT termnumber cosineterm
PT termnumber polyterm
....
TAPR <next taper name>

DX decenterx
DY decentery
....
! For information on the COAT format see "The COAT data section" below.
COAT <coating name>
material thickness is_absolute loop_index tapername
...
COAT <next coating name>

...
COAT I.transmission
! For information on the IDEAL format see "The IDEAL data section" below.
IDEAL <name> <Transmitted intensity> <Reflected intensity>
! For information on the IDEAL2 format see "The IDEAL2 data section" below.
IDEAL2 <name> s_rr s_ri s_tr s_ti p_rr p_ri p_tr p_ti no_pi_flag
! For information on the TABLE format see "The TABLE data section" below.
TABLE <coating name>
ANGL <angle in degrees>
WAVE <wavelength 1 in micrometers> Rs Rp Ts Tp Ars Arp Ats Atp
...
ANGL <next angle in degrees>
...
For details on each of these coating definition keywords, see the discussions which follow.

The MATE data section
Syntax:
MATE <material name>
....
The material names may be any user defined name up to 20 characters in length, with no spaces or special
characters permitted.
ZEMAX uses the material data in the following way:
The wavelength argument is always in micrometers. The wavelengths must be specified in ascending order.
The real value is the real index of refraction of the material at that wavelength.
The imaginary value is the extinction coefficient.
If a single wavelength value is provided for a material, then the real and imaginary parts of the index of
refraction are used no matter what wavelength is being traced. Dispersion of the coating material is therefore
ignored.
If two or more wavelengths are defined for a material, then for wavelengths shorter than the shortest defined
wavelength, the shortest wavelength data is used. For wavelengths longer than the longest defined wave-
length, the longest wavelength data is used. For wavelengths in between, linear interpolation is used.
The TAPR data section
Syntax:
TAPR <taper name>
DX decenterx
DY decentery
....
Normally, ZEMAX assumes the coating thickness is uniform across the entire optical surface. Because of the
way some optical coatings are applied, a thin film coating may have a thickness that varies over the optical
surface. A tapered coating is defined by applying a taper function which computes a dimensionless coefficient to
multiply the nominal coating thickness. Any layer of any coating may have a taper applied, and each layer may
have a different taper. The form of the taper may be a radial polynomial, a cosine function, or a general polynomial
in the x and y directions. The various models may all be used simultaneously, and the taper function may
optionally be decentered and rotated.
The effective coating thickness is:
d eff = d ( f r + f c + f p ) ,
where d is the nominal coating thickness and fr, fc, and fp are the radial, cosine, and polynomial taper factors,
respectively. The taper factors are functions of the decentered and rotated coordinates (x, y), which are
computed from the local surface coordinates using these expressions:
x' = ( x x d ) cos ( y y d ) sin ,
y' = ( x x d ) sin + ( y y d ) cos ,

where is the taper function rotation angle specified in degrees. The radial coordinate is then defined as:
2 2
r = ( x' ) + ( y' ) .
The radial taper factor is defined as:
20
i r ,
i
fr =
i=0
The cosine taper factor is defined as:
i ( cos ) ,
i
fc =
i=0
where is defined as:
sin = --r- ,
R
where r is the radial coordinate as defined above and R is the radius of curvature of the substrate.
The polynomial taper factor is defined as:
20
fp = i Ei ,
i=0
where E i are the polynomial terms in x and y. The E i terms are a power series:
2 2 3
E 0 = 1 , E 1 = x' , E 2 = y' , E 3 = ( x' ) , E 4 = x'y' , E 5 = ( y' ) , E 6 = ( x' ) , etc.
Note the cosine taper factor will yield different coatings on substrates that have a different radius of curvature.
To define a coating whose thickness varies exactly as cos , the 0 cosine taper factor should be zero and the
1 cosine taper factor should be 1.0. When applying cosine taper factors to coatings on non-sequential
component objects; the substrate radius is ignored and treated as a plane. The radial and polynomial terms are
still used.
If the sum of the taper factors evaluates to a negative number, ZEMAX assumes it has a value of zero. The
parameters x d , y d , , i , i , and i are all defined in the taper data section of the coating file. Note that both
even and odd powers of the radial coordinate are used, and that the zero order term is used. The coefficients
generally have units and magnitudes that depend upon the lens units.
!Care should be taken to check that the proper lens units are being used when defining or
using tapered coatings.

The taper names may be any user defined name up to 20 characters in length, with no spaces or special
ZEMAX uses the taper data in the following way:
If the first keyword on any line following the TAPR keyword is DX or DY, then the data value after the DX or
DY keyword on that same line is assumed to be the x d or y d in lens units, respectively.
If the first keyword on any line following the TAPR keyword is AN, then the data value after the AN keyword
on that same line is assumed to be the taper rotation angle .
If the first keyword is RT, the second data value must be an integer between 0 and 20, inclusively. This integer
determines the radial term number. The third data value is the coefficient on that term, i .
If the first keyword is CT, the second data value must be an integer between 0 and 5, inclusively. This integer
determines the cosine term number. The third data value is the coefficient on that term, i .
If the first keyword is PT, the second data value must be an integer between 0 and 20, inclusively. This integer
determines the polynomial term number. The third data value is the coefficient on that term, i .
Any terms ZEMAX uses are assumed to be zero unless otherwise defined.
If the entire layer of a coating is to be scaled by a different amount on each surface, or if the coating thickness
needs to be optimized, see the discussion about coating multipliers in Surface coating tab on page 74 and
Optimizing coatings with ZEMAX on page 508.
The COAT data section
!The order of the coating layers is important! See the discussion in a following section.
Syntax:
COAT <coating name>
...
or
COAT I.transmission
The coating names may be any user defined name up to 20 characters in length, with no spaces or special
ZEMAX uses the coating data in the following way:
When the coating file is first read, ZEMAX verifies that each coating consists of materials that were defined in
the material section. If referenced materials are not defined, an error is issued.
The coating thickness is measured either in units of primary wavelength thickness in that medium (relative
definition), or in units of micrometers independent of any wavelength (absolute definition). If relative definition
is used, the same coating used by two different lens files, with two different primary wavelengths, will have
different actual coating thicknesses. The actual thickness of the coating is determined by:

d = ----0- T ,
n0
where 0 is the primary wavelength in micrometers, n 0 is the real part of the index of refraction of the coating
at the primary wavelength, and T is the "optical thickness" of the coating specified in the coating file. For
example, in a coating material whose real part of the index of refraction is 1.4, a quarter wave coating (T =
0.25) at a primary wavelength of 0.550 micrometers would be 0.0982 micrometers thick. Note that only the
real part of the index of refraction is used for determining the thickness of the layer.
If the is_absolute integer is zero, the thickness is interpreted to be relative, otherwise, the coating thickness is
absolute in micrometers. If the is_absolute, loop_index, and tapername are omitted, then is_absolute and
loop_index are both assumed to be zero, and no taper is assumed.

The loop_index parameter determines how ZEMAX replicates groups of coating layers. See the next section
for details.
Defining replicated groups of coating layers
Some coatings contain repeated groups of coating layers. For example, a coating definition may be defined
as follows:
COAT 3GROUPS
MAT0 0.25 0 0
MAT1 0.25 0 0
MAT2 0.50 0 0
MAT3 0.25 0 0
MAT1 0.25 0 0
MAT2 0.50 0 0
MAT3 0.25 0 0
MAT1 0.25 0 0
MAT2 0.50 0 0
MAT3 0.25 0 0
MAT4 0.25 0 0
Note the sequence of layers with materials MAT1, MAT2, and MAT3 is repeated 3 times. This is a perfectly
acceptable syntax, however, 11 text lines and 11 layers are required to define the coating. A shorthand syntax
that replicates the layers is available using the loop_index parameter. The loop_index integer parameter is set to
the number of times a group of layers is to be repeated. The loop_index must be set at BOTH the first AND the
last layer (this is to indicate the range of layers, and ZEMAX also reads the coating data in both directions, as
described in a following section). The coating above could be written as:
COAT 1GROUP
MAT0 0.25 0 0
MAT1 0.25 0 3
MAT2 0.50 0 0
MAT3 0.25 0 3
MAT4 0.25 0 0
Note the "3" parameter appears on both the first and third layers listed. This syntax not only saves typing, it
reduces the chance of careless typing errors, eases editing, and conserves the total number of layers; which is
limited for each coating definition (see the following section on coating data limits). The coating above only
requires 5 "layers" to define, although there are physically 11 layers modeled.
There are some important points regarding the use of the loop_index:
There are no limits on the number of different groups that may be defined, or on how many layers each group
contains, as long as the total number of lines under the COAT heading does not exceed the maximum number
of allowed layers, defined below.
A loop_index of zero is the same as a loop_index of 1; that is, zero does not mean do not include a group at all!
Loops may not be nested, nor may they overlap.
Defining simple ideal coatings
The COAT keyword also may be used to define simple ideal coatings. The syntax is
I.transmission
where transmission is numerical data. For example, the command
COAT I.75
will create a coating named I.75 which transmits 75% of the ray energy. Ideal coatings defined in this way
transmit the specified fraction of light, and reflect the rest, independent of the ray wavelength, angle of
incidence, incident material, or substrate material. Note this type of ideal coating has no absorption, and the
transmission must be less than 1.0 and greater than 0.0. For a more general ideal coating, see the IDEAL and
IDEAL2 coating definitions which follow.
The IDEAL data section
Syntax:
IDEAL <name> <Transmitted intensity> <Reflected intensity>

Ideal coatings are defined only by the intensity transmission and reflection coefficients. Ideal coating names
may be any user defined name up to 20 characters in length, with no spaces or special characters permitted.
After the ideal coating name, the T and R values are listed. If T+R exceeds 1.0, the values will be scaled so that
T+R = 1.0. The absorption coefficient is computed automatically via A = 1.0 - R - T, to conserve energy. To make
a coating that absorbs 100% of the energy, set both T and R to zero. ZEMAX assumes the phase of transmission
is zero. The phase of reflection is zero if the incident index is higher than the substrate index, otherwise the phase
of reflection is .
The IDEAL2 data section
Syntax:
IDEAL2 <name> s_rr s_ri s_tr s_ti p_rr p_ri p_tr p_ti no_pi_flag
The Ideal2 keyword provides an alternate means of defining ideal coatings. Ideal2 coatings are defined by the
real and imaginary amplitude transmission and reflection coefficients for S and P polarized light separately. Ideal2
coating names may be any user defined name up to 20 characters in length, with no spaces or special characters
permitted. After the coating name, the reflection real, reflection imaginary, transmission real, and transmission
imaginary values are listed for S polarized light, followed by these same four values for P polarized light. To make
a coating that absorbs 100% of the energy, set both transmission and reflection values to zero. The phase of
reflection and transmission is defined by the complex coefficients. If the incident index is lower than the substrate
index, ZEMAX changes the sign of the reflected real and imaginary coefficients to yield an additional phase
change upon reflection of , unless the no_pi_flag is set to any value other than zero; in which case no changes
are made to the phase of the coating.
The TABLE data section
Syntax:
TABLE <coating name>
ANGL <angle in degrees>
...
ANGL <next angle in degrees>
...
Table coatings are similar to IDEAL coatings, except the transmission and reflection may be a function of
incident angle and wavelength and may be specified separately for S and P polarizations. For multiple discrete
angles of incidence and wavelengths, four intensity values are defined: Rs, Rp, Ts, and Tp; for intensity reflection
and transmission and the s and p subscripts indicate the polarization state. There are also four angles (in degrees)
defined: Ars, Arp, Ats, and Atp; these values define the phase angle of the R and T values so ZEMAX can compute
the real and imaginary parts of the transmission and reflection. The absorption is computed from A = 1 - R - T for
each polarization state. If the incident index is higher than the substrate index, the phase of reflection is changed
by a factor of .
The primary purpose of the table model is to allow accurate ray tracing where the layer by layer prescription
of the coating is not known or is unavailable because the coating vendor will not provide the design data.
The syntax consists of the keyword TABLE followed by the name of the coating. After the TABLE keyword,
blocks of data are defined for each incident angle. After each angle definition, R and T values are defined for each
wavelength in the table. An example syntax would be:
TABLE MYTABLECOAT
ANGL 0
WAVE 0.35 0.014 0.014 0.986 0.986 0.0 0.0 0.0 0.0
WAVE 0.70 0.013 0.013 0.987 0.987 0.0 0.0 0.0 0.0
ANGL 5
WAVE 0.35 0.015 0.015 0.985 0.985 0.0 0.0 0.0 0.0
WAVE 0.70 0.012 0.012 0.988 0.988 0.0 0.0 0.0 0.0
ANGL 45

WAVE 0.35 0.065 0.070 0.935 0.930 0.0 0.0 0.0 0.0
WAVE 0.70 0.062 0.065 0.938 0.935 0.0 0.0 0.0 0.0
ANGL 80
WAVE 0.35 0.600 0.600 0.400 0.400 0.0 0.0 0.0 0.0
WAVE 0.70 0.600 0.600 0.400 0.400 0.0 0.0 0.0 0.0
There are several very important points to note:
Angles of incidence less than the smallest angle defined will use the smallest angle data. Angles of incidence
greater than the largest angle defined will use the largest angle data. Other values are linearly interpolated in
cosine of the angle space. Note ZEMAX interpolates but does not extrapolate. Angles must be listed in
ascending order. If data for an angle of incidence of 0.0 degrees is defined, the S and P data should be
identical, since S and P are indistinguishable at normal incidence.
Angles of incidence are assumed to be in the incident media. For coatings placed on surfaces going from glass
to air, or on a non-sequential object, ZEMAX will automatically reverse the coating if required by computing
the correct angle of incidence to use.
Wavelengths in between wavelength data points will linearly interpolate in wavelength space.
All wavelengths listed must be identical for each angle. The same ascending order values for the wavelengths
is required at each angle for accurate interpolation. An error message will result if the wavelengths are not
properly defined.
ZEMAX assumes the phase of transmission is zero if no phase angles are defined. The phase data is angles
measured in degrees if the data is provided. If the incident index is higher than the substrate index, the phase
of reflection is changed by a factor of .
Generally much more than the brief set of data listed here is required for adequate accuracy. ZEMAX currently
allows up to 20 table coatings in each coating file, each with up to 20 angle points and 100 wavelength points per
table coating. If more data points are needed to model the coating, contact technical support for assistance.
Adding comments to the coating file
Any text following the "!" symbol is considered a comment.
Limits on the amount of coating data
There are limits to the amount of data that may be stored in any one catalog as follows:
50 different coating names.
200 layers per coating (see the section Defining replicated groups of coating layers on page 506 to work
around this limit if required).
50 different materials.
120 dispersion data points per material.
20 different tapers.
22 taper terms per taper.
One way to work around these limits is to use a different coating file for each lens, if required. These limits only
apply to a single coating file; multiple coating files may be used for different lenses.
Editing the coating file
There are a few ways of accessing the data in the coating file. First, the file can be edited outside of ZEMAX
using any ASCII text editor. Second, there is a "Edit Coating File" option under the "Tools" menu on the main
menu bar. This invokes the Windows NOTEPAD editor. Also, there is a "Coating Listing" option under the "Tools"
menu, which lists the coating and material definition data. If the coating file is edited, the file must be reloaded in
order to update the new coating data; there is a menu option on the Tools menu for reloading the catalog.
Optimizing coatings with ZEMAX
ZEMAX is not intended to be an optical coating design program. Optimization of coating designs generally
requires some automation of the coating material selection and number of layers.
However, ZEMAX does have the ability to optimize and tolerance coating layer thicknesses, once the number
of layers and material selection is made. ZEMAX uses a dimensionless "coating multiplier" which linearly scales

the thickness of any layer of a coating. Optimization may then be made on the usual coating performance data
using the CODA operand. To optimize the layer thicknesses, first define a coating with the correct number of
layers and materials for each layer. The initial coating thicknesses should be set to either a known good starting
value, or 1 wave (in this latter case the coating multipliers may be interpreted to be in units of waves). Do not use
zero for a coating layer thickness as this thickness cannot be linearly scaled. Coating multipliers may not exceed
the value 10.0.
Coating multipliers may then be defined on the surface coating tab described in Surface coating tab on
page 74. This tab supports setting the multipliers to a variable status, which allows for optimization of the relative
coating layer thickness.
Default materials and coatings supplied with ZEMAX
The default COATING.DAT file supplied with ZEMAX contains several materials and a few common optical
coatings. The following table describes some of the included materials and coatings.
DEFAULT MATERIALS AND COATINGS
Materials Description
AIR Unity index, used for including air gaps in coatings
AL2O3 Aluminum Oxide, index 1.59
ALUM Aluminum, index 0.7-7.0i

ALUM2 An alternate definition for aluminum
CEF3 Cerous Flouride, index 1.63
LA2O3 Lanthanum Oxide, index 1.95

MGF2 Magnesium Fluoride, index 1.38
N15 Imaginary material of index 1.50
THF4 Thorium Fluoride, index 1.52

ZNS Zinc Sulphide, index 2.35
ZRO2 Zirconium Oxide, index 2.1
Coatings Description
AR General anti reflection, defined as a quarter wave of MGF2.
GAP A small gap of air used to show evanescent propagation
HEAR1 High performance anti reflection coating

HEAR2 High performance anti reflection coating
METAL A thin layer of aluminum used to make beamsplitters
NULL A zero thickness coating used primarily for debugging
WAR Anti reflection "W" coat, defined as a half wave of LA2O3

followed by a quarter wave of MGF2.
These materials and coatings are meant to be used as examples, and may not be applicable in any particular
situation. Always check with coating manufacturer or designer for details on coating material specifics. For more
detailed information on how these materials and coatings are defined, use the Coating Listing report described
in the chapter "Reports menu".

Specifying coatings on surfaces
Once a coating has been defined, it can be applied to a surface by specifying the coating name in the coating
column; located at the extreme right side of the lens data editor. ZEMAX interprets the coating definition using
one of four rules:
If the surface specified is a boundary going from air to glass, the coating layer order is interpreted exactly as
specified in the coating file. The incident media is air, then the outermost layer is listed first (at the top) of the
coating definition, then the next layer, etc., with the substrate being the glass type on the surface. The definition
of the coating should not include the substrate index or material definition. The term glass here means the
glass type is not "MIRROR", and not blank (which is treated as unity index for air). Gradient index lenses are
considered glass.
If the surface specified is a boundary between air and air, or glass and glass, the coating is also interpreted
exactly as for air to glass, with the appropriate calculations done for the incident and substrate media.
If the surface specified is a boundary going from glass to air, then the order of the layers is automatically
reversed, so that the coating is the same as if it had been applied going from air to glass. Therefore, a coating
defined by ALHHS, would be interpreted as SHHLA if the boundary goes from glass to air.
If the surface type is a mirror, then the coating definition must include the substrate index. The last layer in the
coating definition is then assumed to be a semi-infinite thickness of substrate material.
ZEMAX may also model certain limiting cases, such as frustrated TIR; see Modeling frustrated total internal
reflection on page 512.
What ZEMAX does if no coating is specified
If the coating column is left blank, then the following is assumed:
If the surface is a mirror, then the surface is assumed to be coated with a thick layer of aluminum, with an index
of refraction 0.7 - 7.0i. The aluminum layer is assumed to be thick enough that no light propagates past the
layer.
If the surface is a dielectric interface, bare, uncoated surfaces are assumed.
Defining the incident polarization
The individual analysis settings, such as those for the polarization ray trace (see the Analysis Menu chapter)
allow for specification of the input polarization. The polarization is defined using a Jones vector:
Jx
J = ,
Jy
where Jx and Jy have both a magnitude and a phase, and the symbol J is used instead of E to distinguish the
2D Jones vector from the 3D electric field vector E. ZEMAX normalizes the specified Jx and Jy values to have
unity magnitude, and then scales the intensity as appropriate if any pupil apodization has been specified. The Jx
and Jy values are therefore measured in terms of relative electric field amplitude.
Let the ray vector be K, which has X, Y, and Z direction cosines (l, m, n). For rays traveling parallel to the Z
axis, or K = (0, 0, 1), the electric field in the Z direction is identically zero, and the conversion from the Jones
vector to the Electric field is trivial:
E x = J x, E y = J y, E z = 0 .
For a more general ray, the conversion of the Jones vector (Jx, Jy) to the 3D electric field values (Ex, Ey, Ez)
is ambiguous. It is not possible to interpret the Jx and Jy values as meaning that for any ray, the Jx value should
be applied in such a way to leave the Ey zero, and for the Jy to be applied so that the Ex is zero. The reason is
that the E resulting for the Jones vectors (Jx = 1, Jy = 0) and (Jx = 0, Jy = 1) must be orthogonal to both K and
to each other.

ZEMAX allows user selection of three different methods to perform the conversion from J to E. In each method,
the vector K refers to the ray vector, the Jx value is the field along a vector S, and the Jy value is the field along
a vector P. Note that K, S, and P must all be unit vectors and orthogonal to each other. The three methods are:
X Axis Reference: The P vector is determined from K cross X, and S = P cross K. This method is the default.
Y Axis Reference: The S vector is determined from Y cross K, and P = K cross S.
Z Axis Reference: The S vector is determined from K cross Z, and P = K cross S.
When the object is at infinity, the method selected will change the polarization orientation of S and P for different
fields, but all rays from the same field will have the same polarization since all rays are parallel to each other. For
finite conjugates, especially when the object space numerical aperture is large, the S and P vector orientations
will vary for different rays in the pupil. No matter which method is selected, the transmission results for unpolarized
light will be unaffected because any two orthogonal rays may be traced to compute transmission. For systems
where a particular polarization is desired, care should be taken to verify that the conversion from J to E yields the
intended polarized ray. To review the detailed conversion results, see Polarization Ray Trace on page 182.
Defining polarizing components
Any boundary between two media can polarize a beam. However, ZEMAX supports an idealized model for a
general polarizing device. The model is implemented as a special "Jones Matrix" surface type for sequential ray
tracing, and a "Jones Matrix" object type for non-sequential ray tracing. The Jones matrix modifies a Jones vector
(which describes the electric field) according to
E' x Ex
= A B ,
E' y C D Ey
where A, B, C, D, Ex, and Ey are all complex numbers. In the lens data and in the non-sequential components
editor, ZEMAX provides cells for defining A real, A imag, etc.
Note that what happens to the z component of the electric field is not defined by the Jones matrix surface,
which implicitly assumes the ray is at normal incidence. To compute the E component of the electric field, the
boundary condition is
k E= 0 ,
where E is the electric field after the Jones matrix and k is the ray vector. Since the X and Y components of
E are known, this expression can be used to compute the Z component.

This general Jones matrix can be used to define a wide variety of polarizing components. For example, see
the sample Jones matrices in the following table. The format of the numbers is (a, b) where the a value is the real
part and b is the imaginary part.
SAMPLE JONES MATRICES
Component Matrix
Null matrix
( 1, 0 ) ( 0, 0 )
( 0, 0 ) ( 1, 0 )
X analyzer
( 1, 0 ) ( 0, 0 )
( 0, 0 ) ( 0, 0 )

Component Matrix
Y analyzer
( 0, 0 ) ( 0, 0 )
( 0, 0 ) ( 1, 0 )
Quarter-Wave plate in X direction
( 0, 1 ) ( 0, 0 )
( 0, 0 ) ( 1, 0 )
Quarter-Wave plate in Y direction

( 1, 0 ) ( 0, 0 )
( 0, 0 ) ( 0, 1 )
Half-Wave plate in X direction

( 1, 0 ) ( 0, 0 )
( 0, 0 ) ( 1, 0 )
Half-Wave plate in Y direction
( 1, 0 ) ( 0, 0 )
( 0, 0 ) ( 1, 0 )
Intensity filter, 25% transmission

( 0.5, 0 ) ( 0, 0 )
( 0, 0 ) ( 0.5, 0 )
What ZEMAX can compute using polarization analysis

ZEMAX can generate plots of R, T, A, D, P, and S as a function of either wavelength for a given incidence
angle or as a function of angle given the wavelength. See the "Analysis" chapter for details.
Using the "Polarization Ray Trace" feature, ZEMAX can compute and tabulate the detailed calculations used
to trace the polarization state through the system for any given ray. Other features directly compute polarization
dependent results; see the "Analysis Menu" chapter for details.
Bulk absorption and transmission
ZEMAX also can accurately compute transmission of individual rays, or compute an average over the pupil.
Transmission calculations are based upon surface effects, such as reflection losses, as well as bulk transmission
through glass according to Beer's law:

t = e ,
where is the absorption coefficient and is the path length through the glass. The parameter generally
depends upon wavelength and has units of inverse length. See the Chapter "Using Glass Catalogs" for details
on defining transmission data. Polarization effects are not considered in all ZEMAX computations.
Modeling birefringent materials
Birefringent materials are modeled using the birefringent in/out surface types, described in Birefringent In and
Birefringent Out on page 235.
Modeling frustrated total internal reflection
Frustrated TIR (FTIR) occurs when a ray of light traveling through glass strikes an interface at an angle
exceeding the critical angle. If another dielectric medium, such as another piece of glass, is placed close, but not
touching the interface, some light will transmit through the thin gap and propagate, even though the refraction at

the boundary cannot satisfy Snell's law. Both the reflected and the transmitted beams will be affected, depending
upon the thickness of the gap. In the limit of the gap having zero thickness, the light will continue through as if
there were no boundary. In the limit of a large gap, more than a fraction of a wavelength, then virtually all of the
light is perfectly reflected.
Although this appears to be a special case, the very general model incorporated into the polarization routines
of ZEMAX handle FTIR as any other coating problem. The trick is to define a coating that reflects the situation.
One of the default coatings is GAP, defined to be 0.1 waves of AIR, with index unity. Using the tilted surface with
a Y-Tangent of 1.0, and placing the coating GAP on the boundary yields the following system, with correctly
modeled polarization properties. Note the air gap may be changed by editing the coating file as described in
Defining coatings in ZEMAX on page 501.
With the GAP coating in place, ZEMAX can now compute either transmitted or reflected polarization properties.
Transmission vs. angle for a FTIR beamsplitter. A FTIR beamsplitter in transmission. The tilted surface be-
tween the two glass blocks has a GAP coating.
However, the ray tracing for other purposes (such as computing system transmission, ray fans, optimization) will
only proceed along the transmitted path. To model the reflected path, give the internal tilted surface a glass type
of MIRROR, add a 90 degree coordinate break after the tilted surface, and change the sign of the remaining
thicknesses as usual.
The coating in this case now needs to be modified to indicate the presence of the second dielectric block, even
though no ray tracing will be conducted in the transmitted path; just the reflected path. A new coating needs to
be defined in the coating file (called FTIR in this example) that looks like:
COAT FTIR
AIR .1
N15 1
The "coating", when applied to a mirror, treats the mirror as a gap of air followed by piece of glass. Note that
detailed modeling of the index and dispersion of the substrate glass is certainly possible by definition of a glass
in the coating file; here N15 was used for simplicity. The thickness of the last material, N15, which defines the
substrate, is irrelevant because ZEMAX always assumes the substrate is semi-infinite when applied to a
MIRROR. The resulting lens figure and reflection plot is shown. Both the reflection and transmission plots are of
course identical for either system model.
Finally, note that conventional TIR can also be modeled by increasing the gap thickness. Distances much
greater than a fraction of a wavelength are essentially equivalent to an infinite gap. However, very large
thicknesses for evanescent wave propagation should not be defined in the coating file because of "underflow"

numerical errors which arise because the transmission is exceedingly low. One wave, or an optical thickness of
1.0, is usually sufficient.
Reflection vs. angle for a FTIR beamsplitter. A FTIR beamsplitter in reflection. The tilted "mirror" surface
has a FTIR coating.
Limitations of polarization analysis

There are a few limitations to the polarization analysis capabilities of ZEMAX. The data in the coating file should
always be carefully checked for accuracy whenever fabrication decisions are being made. Also, the reflectivity
and transmission curves generated by ZEMAX should always be presented to the coating manufacturer as a
check to ensure the coatings will behave as intended.
The polarization ray tracing algorithm generally does not return correct results for certain types of surfaces,
such as paraxial surfaces, diffraction gratings, binary optics, or Fresnel surfaces. Generally, only surfaces which
have refraction or reflection described by Snell's law have valid polarization data computed. In particular, the
polarization algorithm assumes that rays remain in the plane of incidence after refraction or reflection. This is true
for conventional refractive optics (it is a condition of Snell's law) but is not true in general for diffractive optics and
"imaginary" surfaces such as paraxial lenses.
The polarization effects of propagation through gradient index media are not considered, although the surface
effects (reflection, transmission coefficients) are accounted for. Because the rays take curved paths through
gradient media, the polarization vector is rotated in a manner ZEMAX does not account for.

Chapter 21 PHYSICAL OPTICS PROPAGATION
Introduction
!This feature is only available in the EE edition of ZEMAX. Physical optics propagation is
complex, and the user is urged to read and understand this chapter before using the feature.
Geometrical optics is the modeling of optical systems by tracing rays. Rays are imaginary lines which represent
normals to the surfaces of constant phase, called the wavefront. Either rays or wavefronts can be used to
represent a beam. However, rays and wavefronts are propagated differently. Rays propagate along straight lines
without interfering with one another, wavefronts propagate while coherently interfering with themselves. For this
reason, the ray model and the wavefront model yield different representations of the beam as it propagates
through free space or through optical components. The ray method is fast, flexible, and extremely useful for
modeling almost any optical system. However, rays are not well suited to modeling certain important effects,
primarily diffraction.
ZEMAX does have some ray based diffraction computations, such as the diffraction MTF or PSF. These
diffraction computations make a simplifying approximation: that all the important diffraction effects occur going
from the exit pupil to the image. This is sometimes called the "single step" approximation. Rays are used to
propagate the beam from the object, through all the optics and intervening spaces, all the way to the exit pupil in
image space. The ray distribution in the exit pupil, with transmitted amplitude and accumulated OPD used to
compute the phase, is used to form a complex amplitude wavefront. Then, in a single step, a diffraction
computation is used to propagate this complex amplitude wavefront to the region near focus.
Geometrical optics and the single step approximation work quite well for the majority of traditional optical
designs, where the beam is not near focus anywhere except the final image. However, the model breaks down
for several important cases:
When the beam comes to an intermediate focus, especially near optics that truncate the beam (rays by
themselves do not predict the correct distribution near focus).
When the diffraction effects far from focus are of interest (rays remain uniform in amplitude and phase,
wavefronts develop amplitude and phase structure).
When the propagation length is long and the beam is nearly collimated (collimated rays will remain
collimated over any distance, real beams diffract and spread).
Physical optics is the modeling of optical systems by propagating wavefronts. The beam is represented by an
array of discretely sampled points, analogous to the discrete sampling using rays for a geometric optics analysis.
The entire array is then propagated through the free space between optical surfaces. At each optical surface, a
transfer function is computed which transfers the beam from one side of the optical surface to the other. The
physical optics model allows very detailed study of arbitrary coherent optical beams, including:
Gaussian or higher order multi-mode laser beams of any form (beams are user definable).
Beams may be propagated along any arbitrary field position (skew beams).
Amplitude, phase, and intensity may be computed at any surface in the optical system.
Effects of finite lens apertures, including spatial filtering, may be modeled.
Accurate computation of propagation through any optical component ZEMAX can model via ray tracing.
The physical optics model is generally more accurate at predicting the detailed amplitude and phase structure
of the beam away from focus than conventional ray tracing. However there are some disadvantages to the
physical optics propagation analysis:
Physical optics is generally slower than geometrical optics.
515 Chapter 21: PHYSICAL OPTICS PROPAGATION

Because the entire beam array must be stored in computer memory at once, the required RAM may be
quite large for large sampling arrays.
The sampling limits the amount of aberration in the beam that can be accurately modeled. For highly
aberrated systems, geometrical optics should be used.
The following sections will summarize the physical optics propagation algorithms and the information needed
to properly use this advanced ZEMAX feature.
Support for multiple processors
The physical optics propagation routines in ZEMAX are designed to work on multiple CPU computers.
Computers with two or more CPUs will execute the physical optics propagation feature significantly faster than
single CPU computers.
Diffraction propagation
The theory and methodology of diffraction propagation is well understood and widely detailed in available
literature. The methods used in ZEMAX are based upon these references:
Goodman, Joseph W., Introduction to Fourier Optics, McGraw-Hill, New York (1968).
Lawrence, George N. "Optical Modeling", in Applied Optics and Optical Engineering, Volume 11, R. R.
Shannon and J. C. Wyant, eds., Academic, New York (1992).
Only the material relevant to using the physical optics propagation feature in ZEMAX will be summarized here.
Representation of the electric field
The electric field may be represented in three dimensions as
E ( x, y, z ) = E x x + Ey y + E z z ,
where the E values are all complex and x , y , and z are the cartesian unit vectors. The coordinate system
used by ZEMAX is that the beam propagates primarily down a local Z axis. The Z axis used to represent the beam
is aligned with a reference chief ray in each optical space, and therefore this Z axis is not generally the same as
the Z axis defined by the Lens Data Editor which is used to position optics. Because the beam is propagating
along the local Z direction, the first approximation made is to neglect the E z component. Since the electric field
must always be normal to the ray propagation direction, E z can be reconstructed from other data when required,
as will be described later.
By keeping track of the electric field components along both the X and Y axes, effects due to polarization may
be studied, such as transmission and reflection losses, polarization aberrations, and of course the polarization
state of the beam. If polarization effects are not required, the Y component of the field may be ignored, speeding
the computations.
The Fresnel number
A very useful concept in physical optics modeling is the Fresnel number. Strictly speaking, the definition of the
Fresnel number only applies to unaberrated rotationally symmetric beams with a finite extent. However, the
concept is still useful in cases that do not meet these criteria. The Fresnel number depends upon the diameter
of the beam, the radius of curvature of the wavefront phase, and the distance to an observation point where the
complex amplitude of the field is desired. Conceptually the Fresnel number is the number of annular "Fresnel
zones" from the center of the beam to the edge. Fresnel zones are the radial zones where the phase as seen
from the observation point changes by .
Chapter 21: PHYSICAL OPTICS PROPAGATION 516

A perfectly collimated beam will have a Fresnel number given by
2 2 2
Fn = --- [ Z + A Z ] ,

which for Z greater than A reduces to approximately
2
A
F n = ------ ,
Z
where A is the radial size of the beam and Z is the distance from the beam to the observation point. The
Fresnel number becomes small as Z grows large.
For beams that are not collimated, the concept is the same. A converging beam will have a very small Fresnel
number if the observation point is near focus. A perfectly spherical beam converging to focus will have a Fresnel
number of zero, since there are no zones where the observed phase reaches . As the observation point moves
from the focal region, the Fresnel number increases.
Near and far field
If the Fresnel number is small, less than roughly 1, then the beam at the observation point is said to be in the
"far field" relative to the current beam. For Fresnel numbers larger than 1, the beam at the observation point is
said to be in the "near field" relative to the current beam.
It is important to consider the terms near and far as being relative to the propagation from the present location
of the beam to the observation point at which the Fresnel number is computed, rather than having any rigid
relationship to the beam position alone. For example, a beam in the exit pupil of an optical system is typically
called the near field because the far field is at focus. However, a short propagation from focus to a slightly out of
focus observation point is likely a near field propagation if the defocus is small.
The decision as to whether a propagation is in the near or far field will determine the choice of diffraction
propagation methods.
Angular spectrum propagation
It is well known that plane waves remain plane waves while propagating in homogeneous medium. A plane
wave is represented by
ik z
e ,
where k is the wave vector, with magnitude ( 2 ) , and z is the local z direction. The vector k points along
the normal to the wavefront in the direction of propagation. This normal vector has direction cosines , , and
, where
2 2 2
+ + = 1.
The plane wave can then be written as
i2
-------- ( x + y + z )

e

Now recall the definition of the Fourier transform and the inverse Fourier transform:

i2 ( x + y )
FF [ A ] [ A ]e dx dy ,

i2 ( x + y )
[ A ]e
1
FF [ A ] d d .

Let G = FF [ E ] , that is, G is the Fourier spectrum representation of the electric field E . By definition then

i2 ( x + y )
E ( x, y ) = G ( , )e d d .

The electric field can therefore be interpreted as being the integral of a collection of plane waves propagating
with direction cosines
2 2
= , = , and = 1 ( ) ( ) .
Eliminating , and making the approximation that the plane wave propagates at a small angle with respect
to the Z axis (for more information on these approximations see Algorithm assumptions on page 533), the plane
wave equation can be rewritten
i2z iz ( + )
2 2
i2z
----------- 1 2 2 ----------- ----------------------------------

e e e .
i2z
-----------

The e term is just a phase propagation term that is normally neglected. The term which depends upon
2 2 2
and is the transfer function for a plane wave in free space. Defining = + the plane wave transfer
function can then be rewritten as
2
iz
G ( , , z ) = G ( , , 0 )e .
To propagate an electric field from one plane to another, the field needs to be Fourier transformed, the plane
wave propagator applied, and then the resulting distribution inverse Fourier transformed. These operations may
be summarized by defining the plane to plane (PTP) operator:
E ( x, y, z 2 ) = PTP [ E ( x, y, z 1 ), ( z 2 z 1 ) ] , where
1
PTP ( E, z ) FF [ T ( z )FF [ E ] ] , and
2
iz
T ( z ) = e .

Note that the transfer function T ( z ) has unity amplitude but a complex phase. This phase varies slowly from
2 2
point to point in the frequency domain representation G if z is small. But if z grows large, the phase
variations become increasingly rapid. If the phase changes by more than about 2 between adjacent points in
the finite array, the phase becomes ambiguous, and a phenomenon known as aliasing occurs. For this reason,
the angular spectrum method works very well if the propagation distances are fairly short or if the beam is nearly
collimated. Although the diffraction theory is accurate for any propagation distance, when the beam is represented
by a finite sized array of discrete points, the phase of the beam cannot be accurately represented if the phase of
the angular spectrum propagator changes too rapidly between points.
When using the angular spectrum propagator, the phase of the electric field is measured relative to a plane.
Positive phase indicates the wavefront is advanced along the local +z axis relative to the plane, regardless of the
direction of propagation.
The angular spectrum propagator is useful when the Fresnel number is large. This includes the important case
of propagating a beam a short distance. However, the angular spectrum propagator also works well for
propagating a large distance when the divergence of the beam (and thus ) is small. A good rule of thumb to use
is that if the beam does not change size significantly, the angular spectrum propagator may be used. To propagate
beams with small Fresnel numbers, where the beam will change size significantly, requires a separate theoretical
and numerical method.
Fresnel diffraction
For small Fresnel numbers, the appropriate theory is Fresnel diffraction. For a complete discussion and
derivation, see the Goodman reference given in the introduction. The key assumptions in Fresnel theory require
that the field being computed is not too close to the initial field, namely, if z 2 z 1 = z , then z is large compared
to the region over which the field at z 2 is to be determined. Another way of saying this is that the beam cannot
diverge too quickly; very fast F/# beams cannot be accurately modeled with Fresnel diffraction theory (for more
information on these approximations see Algorithm assumptions on page 533).
In the Fresnel region the electric field distribution is given by
i2
ikz ---------- ( x 1 x 2 + y 1 y 2 )
e z
E ( x 2, y 2, z 2 ) = ------------ q ( r 2, z )
iz E ( x1, y1, z1 )q ( r1, z )e dx dy , where

( ir ) ( z )
2
q ( r, z ) = e .
Each of the terms in the above expression has a useful physical interpretation. The leading term indicates that
as the beam propagates, the phase changes along the z axis, just like the plane wave described earlier. The
amplitude also decreases linearly with distance, or the intensity (E*E) falls quadratically. The expression for
q ( r, z ) , called the quadratic phase factor, indicates that the phase is referenced to a sphere of radius z (strictly
speaking it is a parabola, but we have already assumed in the Fresnel development that r 2 z ). This is a very
useful property; all that is required in the representation of the electric field is the phase difference relative to the
reference sphere. This significantly reduces the number of sample points needed to accurately define the phase
of the beam.
When using the Fresnel propagator, the phase of the electric field is measured relative to a reference sphere
with a radius equal to that of the distance from the beam waist. This is not the same radius as the phase radius
of curvature of the Gaussian beam. Positive phase indicates the wavefront is advanced along the local +z axis
relative to the reference sphere, regardless of the direction of propagation.

Another important property of q ( r, z ) is that as z gets larger, q ( r, z ) , varies more slowly in phase. This
is the opposite of the T ( z ) operator, which varies rapidly in phase as z gets larger. Accordingly, Fresnel
diffraction is useful when the Fresnel number is small.
Selecting the correct propagator
ZEMAX will automatically choose the angular spectrum propagator when the Fresnel number is large, and the
Fresnel propagator when the Fresnel number is small. However, there are times when the angular spectrum
propagator is a better choice than the Fresnel, and ZEMAX supports a surface specific option to choose the
angular spectrum propagator rather than the default choice.
Fraunhofer diffraction
Consider the Fresnel diffraction expression. If z is very large, then q ( r, z ) may be neglected. This yields
the Fraunhofer diffraction expression, which is
i2
ikz ---------- ( x 1 x 2 + y 1 y 2 )
e z
E ( x 2, y 2, z 2 ) = ------------
iz E ( x1, y1, z1 )e dx dy , or

E ( x 2, y 2, z 2 ) = FF ( E ( x 1, y 1, z 1 ) ) ,
where phase and amplitude factors are omitted. The far field distribution is just a scaled version of the Fourier
transform of the near field distribution. Fraunhofer diffraction is only valid if the Fresnel number is nearly zero.
The ray based diffraction features in ZEMAX, such as the diffraction MTF and PSF assume Fraunhofer
diffraction. This is why ZEMAX cannot compute the ray based diffraction MTF or PSF if the beam is too much out
of focus. The Fraunhofer assumption is never used by the physical optics propagation algorithm in ZEMAX, it is
presented here for completeness.
The pilot beam
Consider a Gaussian beam with waist 0 . The Rayleigh range is given by
2
0
z r = ---------- .

The phase radius of curvature of the beam is a function of the distance from the beam waist, z :
2
zr
R ( z ) = z + ---- .
z
Note that radius is infinite at z = 0 , reaches a minimum of 2z r at z = z r , and asymptotically approaches

infinity as z . The phase of the Gaussian beam along the axis is defined by the Gouy shift, given by
1 z
( z ) = tan ---- .
z r

For example, the axial phase is 4 at a distance of plus one Rayleigh range. The beam size is also a function
of the distance from the waist:
12
( z ) = 0 1 + ----
z 2
.
z r
Note for large distances the beam size expands linearly. The divergence angle of the beam is given by
1
0
= tan ------ .
zr
Now consider the problem of numerically representing this beam by a discrete sampling of points. If a constant
spacing between points is used, the beam will expand beyond the edges of the array if the propagation proceeds
too far from the waist. Therefore, far from the waist, a linearly expanding coordinate system where the point
spacing is proportional to z is best. However, near the waist, the beam size does not decrease to zero, but remains
reasonably constant. In this domain, a constant sampling is most convenient.
The compromise sampling system is to use a constant spacing near the waist, and a linearly scaled spacing
far from the waist. The transition point is somewhat arbitrary, however for optical design applications a value of
twice the Rayleigh range has proved useful, and ZEMAX uses this threshold. For this reason, in the following
discussion about propagation in and out of the Rayleigh range, the actual threshold point may be different than
the Rayleigh distance. The concept however remains the same.
The diffraction theory developed earlier did not assume any particular shape or form to the electric field being
propagated. The algorithms are for the most part independent of the field distribution. However, the problem of
sampling remains. It is also impractical to compute (or even define) Fresnel numbers for arbitrary, aberrated
beams with irregular or non-existent apertures.
For this reason, a pilot beam is used to assist the physical optics propagation algorithm in determining which
propagation algorithm to select. The pilot beam is an ideal Gaussian beam, with a waist, beam size, phase radius,
and relative z position. The initial parameters may be generated by fitting the Gaussian beam equations to the
initial distribution. The pilot beam is then propagated from surface to surface. At each surface, new beam
parameters, such as the new waist, phase radius, or position are computed. The properties of the pilot beam are
then used to determine if the actual distribution is inside or outside the Rayleigh range, and what propagation
algorithms are appropriate.
After passing through an aperture that significantly truncates the beam, such as a pinhole aperture, it may be
required to recompute the pilot beam parameters, as described later in this chapter.
Sign conventions for phase data
It is very important to understand the phase sign conventions when constructing beams using DLLs, files, or
when performing fiber coupling calculations where phase matching is critical.

As shown in the previous sections, the angular
spectrum propagator works best when the beam is
Inside Rayleigh Range Wavefront
nearly collimated, while the Fresnel theory works
best when the beam is diverging. When using the
angular spectrum propagator, the phase of the elec-
tric field is referenced to a plane.
Once the beam propagates past the Rayleigh
range, the Fresnel propagator is used, and the
phase of the electric field is referenced to a sphere
whose radius is the distance from the beam waist to
the current position of the pilot beam. Reference plane
The sign of the phase is positive if the wavefront Outside Rayleigh Range
Wavefront
is to the "right" of the reference surface, with "right"
being towards the positive local Z axis direction. For
a beam propagated to the +z side just inside the
Rayleigh range, the phase slope is negative be-
cause the wavefront is left of the reference plane.
Just outside of the Rayleigh range, the phase slope
is positive because the wavefront is now to the right
of the reference sphere. Reference sphere
Therefore, the slope of the phase of the electric

field will "flip" from negative to positive when cross-
ing from inside to outside the Rayleigh range on the
+z side.
For example, consider a Gaussian beam. At the waist, the phase is zero everywhere on the reference plane.
If the beam is propagated to just inside the Rayleigh range on the positive side of the waist, the phase at the
1
center of the beam will change to 4 radians, this is the Gouy shift given by = tan ( z z r ) . The phase radius
of curvature of the beam will become twice the Rayleigh range distance. The wavefront phase will be increasingly
negative relative to the center of the wavefront as the radial aperture is increased, because the curved wavefront
phase is measured relative to a plane.
If the beam now propagates a small distance so the pilot beam is now just on the far side of the Rayleigh range,
the phase will remain 4 radians at the center but will now be referenced to a sphere whose radius is the
distance to the waist, which is the Rayleigh range for this case. Because the radius of the reference sphere is
smaller than the curvature of the wavefront, the slope of the phase of the beam will now "flip" to be increasingly
positive with radial aperture.
Finally, as the beam propagates further past the Rayleigh range, the phase relative to the reference sphere
will tend toward a constant value of 2 radians (the limiting value of the Gouy shift).
The two representations of the beam are equivalent, however care must be taken to account for the proper
phase reference surface when constructing or comparing different beams.
Propagating in and out of the Rayleigh range
The naming conventions used in this section are from the Lawrence reference cited earlier in this chapter.
It has already been shown that short propagations are well modeled using the PTP operator. This operator
has the property that the sample spacing remains constant.

It is most convenient to use the Fresnel propagator when the beam is already at the waist of the pilot beam,
and the field is desired far from the waist relative to a reference sphere. For this reason, the Fresnel propagator
is redefined as the waist to sphere (WTS) operator:
E ( x 2, y 2, z 2 ) = WTS [ E ( x 1, y 1, 0 ), z 2 ] , where
1 s z
WTS ( E, z ) ---------- FF [ q ( r 1, z 2 )E ] , s = ------2- ,
iz 2 z2
z2 z2
x 2 = -------------- , and y 2 = -------------- ,
n x x 1 n y y 1
where nx and ny are the number of points in the x and y directions of the array. The last two expressions yield
the new linearly scaled sample spacings after application of the operator. Reversing the order of operations results
in the sphere to waist (STW) operator:
s
STW ( E, z ) q ( r 1, z 2 )FF [ iz 2 E ] ,
with a similar change in the sample spacing. There are four possible general propagation cases to consider:
II: propagation from inside to inside the Rayleigh range.
IO: propagation from inside to outside the Rayleigh range.
OI: propagation from outside to inside the Rayleigh range.
OO: propagation from outside to outside the Rayleigh range.
All these cases can be handled with the appropriate combination of the PTP, WTS, and STW operators:
II(z1, z2) = PTP(z2 - z1)
IO(z1, z2) = WTS(z2 - z0)PTP(z0 - z1)
OI(z1, z2) = PTP(z2 - z0)STW(z0 - z1)
OO(z1, z2) = WTS(z2 - z0)STW(z0 - z1)
where z0 is the pilot beam waist position, z1 is the starting beam position, and z2 is the end beam position.
Separation of X and Y propagation
For rotationally symmetric optics, a single rotationally symmetric pilot beam and propagation method is
appropriate. When the beam is astigmatic or anamorphic, and/or the optical components are cylindrical or
toroidal, then a more efficient model for the beam is to separate the X and Y propagation. Separation of X and Y
direction propagation is accomplished by choosing the appropriate option, see Beam Definition Tab: on
page 190.
If the propagation in X and Y is separated, then there will be two pilot beams; one each for the X and Y direction.
The phase references will also be distinct in the X and Y directions. In place of a spherical reference surface, a
"bispherical" surface is used instead. The sag of a bispherical surface is given by:
2 2
cx x + cy y
z = ------------------------------------------------- ,
2 2 2 2
1 + 1 cx x cy y

where
1 1
c x = ----- , c y = ----- .
Rx Ry
The pilot beam, including beam phase radius, waist, position, Rayleigh Range, divergence, are all distinct in
the X and Y directions. There is a small amount of additional computational overhead in propagating the X and
Y portions of the beam independently, and so this option should only be used when required.
Comments about point spacing and sampling
Although the total number of array points n x and n y , remain constant, the array size and point spacings x
and y will change as the beam propagates. If the array width is very large at the beam waist relative to the waist
size, then there are relatively few points across the beam waist. This will result in a smaller array size far from
the waist, with a relatively large number of points across the beam size. Conversely, if the array size is small at
the waist, the array size will grow large compared to the beam far from the waist, leaving few sample points to
represent the beam. This inverse relationship is a necessary but frequently inconvenient product of the Fourier
transform theory used to model the diffraction. The exact equations describing the change in point spacing are
given in the previous section.
There is clearly a tradeoff between good sampling of the beam near the waist and good sampling far from the
waist. It can be shown that to achieve approximately uniform sampling relative to the beam size at both the waist
and far from the beam waist the array size at the beam waist in the X and Y directions should be
X = 0x n x , Y = 0y n y .
The physical optics analysis feature settings include an "Auto" button which will use this formula to set a
suggested initial array width. See Physical Optics Propagation on page 190.
Propagation through arbitrary optical surfaces
The methods described thus far are good for propagating through homogeneous space. However, the prime
interest is in propagating the beam through optical components such as lenses and apertures.
It is not practical to directly perform a diffraction propagation through an arbitrary surface shape. The difficulty
lies in the representation of the complex amplitude at discrete points in a plane or spherical phase referenced
array. When the beam is incident on a curved surface, different parts of the beam intercept the surface at different
points along the local z axis.
To avoid this problem, the properties of the pilot beam are used to generate a set of rays that represent the
wavefront incident upon the surface. This set of rays, called the probing rays, is then traced through the optical
surface using conventional geometric optics. The path length of each ray, and the positional and angular
aberrations generated, are then used to reconstruct the pilot beam and complex amplitude after the surface. The
probing ray set is used to generate the transfer through the surface.
The probing ray set can be used to determine:
The effective power of the surface with respect to the pilot beam.
The direction of the beam after leaving the surface.
The polarization phase and amplitude transmission of a surface.
The vignetting of the surface with any supported surface apertures (or ray errors). Note that surfaces without
hard apertures defined will pass the entire beam without vignetting, independent of how the system aperture
is set.
The phase aberrations.
The anamorphic and non-linear stretching and compression (distortion) of the beam.

The probing ray method can be applied to propagation through a single surface, or multiple surfaces at once.
This is a very desirable property, because geometrical optics may be used to propagate through whole optical
components that would be difficult to model with physical optics propagation. These include highly tilted surfaces
and gradient index lenses, to name a few. Propagating through multiple surfaces at once using rays also speeds
up the analysis. Some special surfaces, such as the ABCD matrix surface, and some types of Fresnel surfaces,
do not allow POP analysis at all because there is no way to compute the effective phase of these surface types.
A warning is issued by ZEMAX if the POP analysis cannot proceed due to the presence of these special surface
types.
Propagating through non-sequential surfaces
Non-sequential component (NSC) surfaces pose a difficult challenge for physical optics propagation. A NSC
surface may represent a large number of discrete components. The incoming beam may not hit all or any of these
components, or perhaps a portion of the beam will hit one component and split off from the rest of the beam, or
any number of other possibilities. The Fresnel propagation method used by POP is not suitable for computing the
effects of propagation through arbitrary NSC surfaces.
There are a limited number of cases where POP can be used to propagate a beam through a NSC surface
with reasonable confidence in the results. These include:
-NSC surfaces with a single lens or mirror.
-Prisms with an effectively sequential path and apertures large compared to the beam.
Even in these cases, POP can only use ray tracing to model the effective phase the NSC surface imparts to
the beam. Diffraction propagation is not possible. For this reason, only relatively slow beams propagating through
thin NSC surfaces should be used. Always check the POP results carefully before having confidence in results
of propagation through NSC surfaces.
Accounting for polarization
If polarization is used, ZEMAX will use polarization ray tracing to determine the properties of the probing ray
set and corresponding transfer function. Polarization ray tracing permits the modeling of the effects of optical
coatings on the phase and amplitude of the transmitted or reflected beam.
ZEMAX assumes the Ex and Ey portions of the field account for all the energy in the beam at any given array
point. However, the Ez component is required for polarization ray tracing. ZEMAX recreates the Ez as needed by
applying the condition that E must be perpendicular to the propagation vector k. The vector k is computed from
the pilot beam properties and the phase errors present on the beam. The Ex and Ey fields are renormalized to
account for the correct beam intensity. After polarization propagation, the Ex and Ey components are renormal-
ized to again hold the energy present in the Ez component, which is then discarded.
Memory requirements
Geometric optics ray tracing generally requires very little RAM, because each ray may be traced from object
to image, independently of all other rays. Physical optics is far more demanding, because the entire beam must
be stored at once. ZEMAX uses 8 byte "double precision" floating point numbers to store the beam data. There
are actually 16 bytes per array position because the electric field amplitude is a complex number. Here are some
guidelines for how much RAM is required for various parts of the propagation algorithm:
Task RAM Requirement

Storing an unpolarized beam 16 bytes per grid point
Storing a polarized beam 32 bytes per grid point
Computing the surface transfer probing ray set 48 bytes per grid point
Some surface types require additional data and may
require up to 64 bytes per grid point

For example, propagating a 2048 x 2048 unpolarized beam will require about 256 Mb. ZEMAX currently
supports a minimum array size of 32 x 32 and a maximum array size of 8192 x 8192. ZEMAX requires the
sampling to be an integral power of 2; for example, 32, 64, 128, 256, etc.
If the computer does not have sufficient RAM, it may "page" to the hard disk. This will allow the computation
to proceed, but at a rate hundreds or even thousands of times slower than if sufficient RAM were available. The
Windows OS does not allow ZEMAX to allocate more than 2.0 Gb of RAM, and in many cases, the actual limit is
much lower. This limit is imposed by Windows, and not ZEMAX.
Defining the initial beam
The menu option Analysis, Physical Optics, Physical Optics Propagation invokes the Physical Optics Propa-
gation feature. The settings box for this feature allows user selection of many options, including those needed to
define the initial beam, sampling, surface ranges, and field positions. See Physical Optics Propagation on
page 190.
The X- and Y-Sampling determine the number of points used to sample the beam. Larger values are generally
more accurate at the expense of longer computation times and larger RAM requirements.
The X- and Y-Width are measured in lens units. The larger the width, the more empty space around the portion
of the beam with non-zero intensity there will be. This space is called the guard band, and it is vital that sufficient
empty space exist around the beam. The empty space allows room for the beam to expand as aberrations are
introduced to the beam. If portions of the beam get too close to the edge of the array, they will "alias" and reflect
back into the beam, decreasing the accuracy of the computed results.
The initial beam may be defined using any of these options:
Gaussian Waist
Gaussian Angle
Gaussian Size+Angle
Top Hat
User defined table in a file
User defined DLL program
The beam may then be aligned along the chief ray for any defined field position, in the optical space preceding
any surface. Propagation then proceeds from the starting surface to the ending surface. Definitions of beam waist,
Rayleigh range, and divergence angle may be found in Propagating the embedded beam on page 188.
Gaussian Waist
The Gaussian Waist beam is an optionally truncated and decentered Gaussian defined by:
x dx 2 y dy 2
-------------- + --------------
0x 0y
E ( x, y ) = E 0 e , if
x-------------
dx- 2 y-------------
dy- 2
Ay 1
+ , and
Ax
E ( x, y ) = 0 otherwise.
The dx and dy values are the decenter values, Ax and Ay are the truncating aperture values, and E 0 is chosen
to yield the peak irradiance in power per unit area defined in the settings box. If Ax or Ay are zero, no truncating
aperture is used. A smoothing function is used near the edge of the truncating aperture to avoid pixel related
errors. The smoothing function weights the pixel amplitude by the area of the pixel inside the truncating aperture.

The truncating aperture is useful for modeling receiver fiber modes, where the truncating aperture is typically 15%
greater than the core size. The beam is defined at the waist, so the initial phase is zero across the full width of
the beam. The beam will diverge to a larger beam size as it propagates away from the waist. See Gaussian Angle
and Gaussian Size+Angle below.
Gaussian Angle
The Gaussian Angle beam is similar to the Gaussian Waist. The distinction is that the Gaussian Angle beam
is defined by the far-field divergence half-angle in degrees, measured in air. ZEMAX uses the specified
divergence angle to compute the beam waist. The beam waist is given by:

0 = --------------- .
tan
Note that the angle is the half-angle of divergence in degrees, and the wavelength is not scaled by the index
of the starting medium. If the x and y angle values are different, then an elliptical beam will be generated. The
rest of the beam definition is then identical to the Gaussian Waist.
Gaussian Size+Angle
The Gaussian Size+Angle beam is similar to the Gaussian Waist and Angle beams. The distinction is that the
Gaussian Size+Angle beam is defined by the beam size (not waist) at the starting surface and the far-field
divergence half-angle in degrees, measured in air. ZEMAX uses the specified beam size and divergence angle
to compute the beam waist, position, and phase. The beam waist is given by:

0 = --------------- .
tan
Note that the angle is the half-angle of divergence in degrees, and the wavelength is not scaled by the index
of the starting medium. If the resulting beam waist is larger than the specified beam size (a physical impossibility),
the waist value will be used for the beam size. The z position of the beam relative to the waist is then computed
using:
2
z = z r ------ 1 ,
0
where z r is the Rayleigh range. If the x and y values are different, then an elliptical beam with a possibly
toroidal phase will be generated.
Top Hat
The top hat beam is defined as
E ( x, y ) = E 0 if -------------- + -------------- 1 , and

x dx 2 y dy 2
wx wy
E ( x, y ) = 0 if -------------- + -------------- 1 ,
x dx 2 y dy 2
wx wy
where E 0 is chosen to yield the peak irradiance in power per unit area defined in the settings box.
The width of the array should be at least 6-10 times the waist size w, in each direction.

User defined table
The table of values must be placed in either a binary or ASCII format file and read from disk. The file must end
in the extension ZBF (for ZEMAX Beam File). The binary format is identical to the format written out by ZEMAX
if the "Save Output Beam To:" option is selected. The Ex and Ey values are defined such that the Ex*Ex + Ey*Ey
is in units of watts. If the units flag indicates the beam units are different from the current lens units, the beam is
automatically scaled to the current lens units when read into ZEMAX.
!All ZBF Beam Files must be placed in the \POP\BEAMFILES subdirectory off the main ZEMAX
directory.
Beams in ZEMAX are always centered on the chief ray for the selected field and wavelength. Therefore, the
data in the beam file should be positioned relative to the chief ray that will be used to align the beam. The center
point in the beam file is at the coordinate (nx/2+1, ny/2+1). ZEMAX requires the values of nx and ny to be an
integral power of 2; for example, 32, 64, 128, 256, etc. The minimum sampling is 32 and the current maximum
sampling is 8192. Fiber coupling data is ignored when reading beams, and will be zero if fiber coupling is not
computed on output. Note the total fiber coupling is the product of the receiver and system efficiency. The first
data point is at the -x, -y corner, and the data proceeds across the x rows first. The Rayleigh distance is ignored
on input and is automatically recomputed by ZEMAX. The wavelength value stored in the ZBF file is scaled by
the index of the media the beam is currently in.
ZEMAX Beam File (ZBF) binary format
The ZBF binary file format is defined as follows. All integers are 4 bytes, all doubles are 8 bytes.
1 integer: The format version number, currently 0.
1 integer: The number of x samples (nx).
1 integer: The number of y samples (ny).
1 integer: The "is polarized" flag; 0 for unpolarized, 1 for polarized.
1 integer: Units, 0 for mm, 1 for cm, 2 for in, 3 for meters.
4 integers: Currently unused, may be any value.
1 double: The x direction spacing between points.
1 double: The y direction spacing between points.
1 double: The z position relative to the pilot beam waist, x direction.
1 double: The Rayleigh distance for the pilot beam, x direction.
1 double: The wavelength in lens units of the beam in the current medium.
1 double: The waist in lens units of the pilot beam, x direction.
1 double: The receiver efficiency. Zero if fiber coupling is not computed.
1 double: The system efficiency. Zero if fiber coupling is not computed.
1 double: The z position relative to the pilot beam waist, y direction.
1 double: The Rayleigh distance for the pilot beam, y direction.
1 double: The waist in lens units of the pilot beam, y direction.
2*nx*ny double: Ex values.
If polarized, 2*nx*ny Ey values follow the Ex values.
ZEMAX Beam File (ZBF) ASCII format
The ZBF ASCII file format is defined as follows. The first line must be the single character "A", followed by the
other data values specified.
A: indicates an ASCII file.
version: The format version number, currently 0.
nx: The number of x samples.
ny: The number of y samples.
ispol: The "is polarized" flag; 0 for unpolarized, 1 for polarized.
units: 0 for mm, 1 for cm, 2 for in, 3 for meters.
unused 1: Currently unused, may be any value.
dx: The x direction spacing between points.
dy: The y direction spacing between points.
zx: The z position relative to the pilot beam waist, x direction.
Rx: The Rayleigh distance for the pilot beam, x direction.

lambda: The wavelength in lens units of the beam in the current medium.
wx: The waist in lens units of the pilot beam, x direction.
re: The receiver efficiency. Zero if fiber coupling is not computed.
se: The system efficiency. Zero if fiber coupling is not computed.
zy: The z position relative to the pilot beam waist, y direction.
Ry: The Rayleigh distance for the pilot beam, y direction.
wy: The waist in lens units of the pilot beam, y direction.
Ex real value for point 1
Ex imaginary value for point 1
Ex real value for point 2
Ex imaginary value for point 2
etc... for 2*nx*ny Ex values.
If polarized, followed by 2*nx*ny Ey values.
User defined DLL program
To define a beam using a program, the algorithm which computes the initial complex electric field from user
defined parameter data must be written and compiled into a Windows Dynamic Link Library, or DLL. Sample DLLs
are provided with ZEMAX with source code. New DLLs may be easily created with a suitable compiler.
Beam DLL parameters
beam properties. These values are defined by the DLL and are only used by the DLL. The values may be entered
or edited directly on the settings box when the DLL is selected as the beam type.
Creating a new Beam DLL
UserBeamDefinition
UserParamNames
When initializing a beam using a DLL, ZEMAX passes to the UserBeamDefinition function the source
parameters, the wavelength, and other data. The UserBeamDefinition function then is required to compute the
electric field values. These values are returned to ZEMAX and are used to populate the beam array. The function
UserParamNames is used to define the names of all used parameters. These names appear on the settings
dialog box.
The best way to learn the use of Beam DLLs is to copy and study an existing DLL. The sample DLLs provided
with ZEMAX include extensive documentation and comments on the data format; see any of the sample source
code files for examples.
!All Beam DLLs must be placed in the \POP\DLL subdirectory off the main ZEMAX directory.
Surface specific settings
Each ZEMAX surface supports these settings relevant to Physical Optics Propagation:
Use Rays To Propagate To Next Surface: If checked, then the diffraction propagation algorithms will not be
used to propagate the beam to the next surface. Instead, rays will be traced to the next surface and the
resulting ray transfer function will be used to propagate the beam and update the pilot beam parameters.
Multiple sequential surfaces may all have this option checked, in which case the beam will advance through
all the surfaces using only ray propagation. Use of this feature is required for non-sequential, birefringent,
and gradient index surfaces. This feature may also be used to speed up the propagation algorithm by
propagating through several surfaces at once, if the distance between the surfaces is small enough so that
the diffraction propagation is unimportant. The default is unchecked.
Re-Compute Pilot Beam Parameters: Some surfaces significantly alter the beam characteristics. A good
example is a pinhole aperture. After passing the aperture, the beam waist size, divergence, and position
may be significantly altered, even though technically the pinhole has no optical power (this is in fact the
core difference between geometric and physical optics). After passing the pinhole, the pilot beam parame-
ters need to be recomputed for subsequent accurate propagation. Checking this option will invoke an

algorithm which finds the pilot beam parameters that best fit the actual beam.
Do Not Rescale Beam Size Using Ray Data: By default, ZEMAX will use the ray grid to determine any
distortion, stretching, scaling or other change in the beam shape. A good example of this is the passing of
the beam through a grating. The beam will become compressed along the direction of the diffraction.
However, there are times when this computation fails. One such case is when the ray grid enters a caustic.
The rays no longer accurately represent the beam, and should not be used to determine the beam shape.
ZEMAX automatically skips this step if it detects the ray grid is in a caustic, however, there may be other
cases where the algorithm should not be applied, and this option allows user selective disabling of the
algorithm.
Use Angular Spectrum Propagator: If selected, the angular spectrum propagator will be used for propaga-
tion through the surface instead of the propagator ZEMAX would automatically select. This option should
only be used if the beam size does not change dramatically over the propagation distance. The array width
will remain constant using the angular spectrum propagator. The default is unchecked.
Draw "beamfile name" on shaded model: If selected, the ZBF of the displayed name will be drawn on
shaded model layout plots at the location of the surface. The file name is the same as the name of the lens
file, with a 4 digit suffix equal to the surface number. This is the same name convention that automatically
saved beam files use. See Display Tab: on page 191 for information on saving beam files at every surface.
If this option is selected, and the correct beam file name is found, then the surface itself will not be drawn,
just the beam file at that surface. For this reason, this option works best on dummy surfaces. Drawing large
beam files uses considerable amounts of memory and will slow down the drawing of shaded model displays.
Resample after refraction: This control allows the beam sampling and width to be changed after propagating
through any surface. If selected, the new x and y sampling and beam width may be specified. Note that this
control is ignored if rays are used to propagate through the surface. See also "Auto Resample" below.
Auto Resample: If checked, the beam will be automatically resampled. The algorithm used first recomputes
the pilot beam parameters as described in "Re-Compute Pilot Beam Parameters" above. The X and Y- width
is then set according to the equations described in Comments about point spacing and sampling on
page 524. The number of points used in the X- and Y- sampling will not change.
Reference Radius: A rotationally symmetric beam has an identical phase radius of curvature in the X and
Y directions. However, in the general case, there will be a difference between the X and Y phase radii. For
cylinder or toroidal lenses, this difference may be extreme. The choice of reference radius determines how
the pilot beam is propagated, which in turn affects the choice of propagation algorithm, array sampling,
phase sampling, and other important data for determining the further propagation of the beam. Usually, the
default "Best Fit" is the best choice; this selects the optimum trade-off between the X and Y radii to minimize
the phase excursions relative to the reference surface. "Shorter" selects the shorter of the X or Y radius,
"Longer" selects the longer of the two, and "X" and "Y" makes the choice explicitly. The "Plane" option will
reference the phase to a plane, and this works well in cases where the deviation of the rays near the chief
ray is a poor representation of the surface power, as is the case with axicons. The "User" option allows
direct specification of the X and Y radius to use for the reference sphere. Use the value of zero to define a
plane. Use "Best Fit" unless you are certain that another choice will yield higher accuracy for a given
application. If "Separate X, Y" is selected on the beam definition tab, then the best fit radius in the X and Y
directions is automatically selected, unless "Plane" or "User" is chosen for the reference radius. If the
reference radius is set to "User" and "Separate X, Y" is not selected; the X radius value is used for both X
and Y directions.
Computing Fiber Coupling

The physical optics propagation algorithm may be used to compute fiber coupling efficiency. A ray based
method is also supported, see Fiber Coupling Efficiency on page 174.

The fiber coupling receiver efficiency is defined as a normalized overlap integral between the fiber and beam
complex amplitude:
T =
Fr ( x, y )W ( x, y ) dx dy
-------------------------------------------------------------------------------------------------------------------- ,
Fr ( x, y )Fr ( x, y ) dx dy W ( x, y )W ( x, y ) dx dy
where F r ( x, y ) is the function describing the receiving fiber complex amplitude, W ( x, y ) is the function
describing the complex amplitude of the beam coupling into the fiber, and the symbol represents complex
conjugate. Note that these functions are all complex valued, so this is a coherent overlap integral. Maximum
receiver efficiency (T = 1.0) is achieved when the mode of the beam perfectly matches the mode of the fiber in
both amplitude and phase at all points. Any deviation in mode shape, or phase, will reduce the value of T to less
than 1.0. Optical aberrations typically introduce phase deviations which reduce receiver efficiency.
Additional system losses may be caused by apertures which vignette the beam, losses due to reflection from
air-glass boundaries, or bulk absorption. These loss factors reduce the system efficiency. The total coupling
efficiency into the fiber is the system efficiency multiplied by the receiver efficiency.
Where the integral is computed
The fiber coupling receiver efficiency integral defined above is valid and may be computed anywhere in the
optical system; including at the location of the fiber, or at some location near or far from the fiber. The critical
requirement is that the fiber mode function describes the mode of the fiber at the point the integral is computed,
which is always the "end" surface (see Physical Optics on page 187 for a list of analysis options, including
specification of the end surface and the fiber mode). If the end surface represents the beam at the location of the
receiving fiber, then the fiber mode parameters must correspond to the mode at the receiving fiber. If the end
surface is 10 mm away from the fiber, then the fiber mode parameters must be selected to define the mode 10
mm away from the fiber.
Defining the fiber mode
The fiber mode may be a Gaussian or Top Hat function, or may be defined by a DLL or a data file. This allows
very general and arbitrary fiber modes to be described, including multi-mode, aberrated, or arbitrary amplitude
and phase fibers. The fiber mode may be defined using all the same options allowed for defining initial beams,
see Defining the initial beam on page 526.
Decenters and tilts
It is usually valuable to determine fiber coupling for decentered and tilted fibers.
Decenter X/Y values are in lens units and may be defined on the fiber data tab of the Physical Optics
Propagation settings box (Physical Optics Propagation on page 190). These values decenter the fiber mode.
The decenter values shift the center of the mode within the array defining the fiber mode at the end surface. For
this reason, the decenter values should be small enough to keep the majority of the mode energy inside the array
width/height at the end surface.
The tilt values are in degrees. Tilt of the fiber is modeled as adding a linear phase shift to the fiber mode. The
phase shift is proportional to the fiber tilts and the mode width/height.
Note that if the end surface is at the receiving fiber location (typically near focus) then a tilt of the fiber
introduces phase tilt. If the end surface is far from the fiber, then tilt of the fiber introduces a decenter of the mode.
Therefore, great care must be used in selecting the tilt and decenter values; proper use of these values depends
upon knowledge of where the overlap integral is being computed (at the end surface!) relative to the fiber location.
Choosing the location for the receiving fiber
The receiving fiber mode may be either centered on the incoming beam chief ray or at the end surface vertex.
In this latter case, the fiber mode is decentered by the amount equal to the incoming chief rays x and y coordinates
on the end surface, and the fiber is tilted to be aligned with the local Z axis rather than along the chief ray. The
surface vertex option is not permitted when using the File option for defining the fiber mode.

Suggestions for use
It frequently takes some experimentation with the sampling, width, and surface specific settings (see above)
to get good, accurate results from this feature. Keep in mind the following suggestions:
Use an appropriate width for the initial beam. The best initial width to use is almost always that computed
by the "Auto" button next to the width control, and the use of this control is highly recommended. Note that
if the width is too large, then there will not be enough points across the beam to yield adequate sampling,
if the width is too small, aliasing will occur. If the beam waist is smaller than the wavelength at any surface,
including the initial beam, the POP results are probably not accurate, see Algorithm assumptions on
page 533.
!Always check the initial beam by setting the end surface to the start surface in the analysis
window, and observe that the beam is properly sized for the selected array width.
Use adequate sampling. As the guard beam increases in size relative to the non-zero amplitude portion of
the beam, the number of points that have non-zero data will decrease. It may be required to increase the
sampling to be sure enough points sample the beam.
If there are any cylindrical or toroidal optics in the system, or if the beam itself is astigmatic or anamorphic,
try using "Separate X, Y" on the beam definition tab. Separating the X and Y propagation greatly reduces
the sampling required for accurate results because a reference bispherical surface is used instead of a
reference sphere, and the bisphere is capable of more closely fitting anamorphic beams.
If the results seem unbelievable, debug the propagation by propagating one surface at a time through the
optical system, and study the results for plausibility. Typically this process will highlight at which surface the
algorithm fails. It is frequently the case that certain surfaces, such as gradient index optics or highly tilted
surfaces, are best handled by ray tracing. For these surfaces, choose the "Use Rays To Propagate To Next
Surface" option.
Imagine a large diameter beam with a wavelength of 1.0 micrometers (0.001 mm), an array width of 64 mm,
and a sampling of 64 x 64 points. The delta between points is 1.0 mm. Now imagine the beam going through
a collimating lens with a focal length of 100.0 mm. The sampling spacing near focus is given by:
z2
x 2 = -------------- ,
n x x 1
which yields a new sampling spacing of 0.0015625 and a total width of 0.1 mm. To decrease the sample
spacing in the focal plane while keeping the array width constant, the product n x x 1 must increase. Since
the sample spacing is the array width divided by the number of points, both the number of points and the
initial array width would have to be doubled. If the aberrations are so large that the beam size is on the
order of the array width in the focal plane, then the array width must increase. Note the array width in the
focal plane is given by:
z2
n x x 2 = ----------
-.
x 1
Increasing the array width can be accomplished by increasing the number of points while leaving the initial
array width fixed.
!Note that the beam sampling and width may be manually adjusted at any surface.
See Surface specific settings on page 529 for information on resizing the beam at any surface.

Algorithm assumptions
There are numerous approximations made in the model and algorithms used for physical optics modeling.
Some of these assumptions will limit the accuracy of the results in certain cases. The major assumptions are
described below.
Both the angular spectrum and the Fresnel diffraction propagation algorithms are developed assuming the
beam is not too fast. Neither theory accurately predicts the correct diffraction results if the F/# of the beam
is too low. The errors in the propagation theory will reveal themselves as discontinuities in the data as the
propagation algorithm crosses over the Rayleigh range boundary. The loss of accuracy cannot be precisely
quantified (if ZEMAX could precisely quantify the error, we could apply that as a correction to have a more
accurate theory). ZEMAX will report a warning in the propagation report if the Gaussian pilot beam waist
becomes smaller than the wavelength. This is a good rule to use as a limit of the accuracy of POP in general.
Gaussian beams whose waist is small compared to the wavelength, such as fibers, act essentially as point
sources radiating into a sphere. In this domain, the ray model is perfectly acceptable; in other words, dont
use POP for very fast beams! The big differences between ray based and POP based results show up in
very slow diffracting beams, not fast beams, so use ray based computations in ZEMAX to model very fast
beams with confidence. Specifically, use the ray-based fiber coupling algorithm rather than POP for fast,
accurate results in systems where the beam is fast and the diffraction effects from the edges of the lenses
are not significant; this applies to virtually all fiber coupling systems.
Scalar diffraction theory applies. The vector nature of the electric field is ignored. For very fast beams, scalar
diffraction theory is not valid and the results should be used with caution. Note that propagation of fast
beams is generally well handled by geometric ray tracing, except near focus. Although there is no exact
cut-off to the accuracy of scalar diffraction theory, beams faster than F/1 are probably not accurately
modeled with scalar theory. This limitation is another reason not to use POP for very fast beams, as the
previous paragraph explains.
The probing ray set is an adequate model for the propagation near lenses. When propagating through
surfaces, the probing ray set is used to determine the effective phase aberrations and change in overall
reference sphere and pilot beam shape. The diffraction along the distance the probing rays trace is ignored.
The probing ray set is not an accurate method of determining beam propagation near the focus of a ray
bundle.
Diffraction effects of surface structures are ignored. Gratings, binary optics, and diamond turned surfaces
have a complex micro surface structure that can significantly effect the local electric field. This may or may
not affect the propagation of the beam. Note that ray tracing, which also ignores the surface structure, has
proved to be very useful and accurate for modeling these devices. Any surface adequately modeled by ray
tracing will also work fine with physical optics propagation, since ray tracing is used to compute the surface
transfer function.
Samples
Many sample files are included with ZEMAX to demonstrate some of the applications and proper use of the
physical optics modeling features. The files are located in the \Samples\Physical Optics directory. Some of these
files are described below. Because of the large amount of computation involved in physical optics propagation,
some of these files take some time to compute and display the results.
Free space propagation
The sample file "Basic Propagation" illustrates the case of a beam propagating through free space. The beam
is a Gaussian defined to have a waist of 0.1 at surface 1. The wavelength is chosen to be micrometers so that
the Rayleigh range is 10.0 mm. At a distance of 1 Rayleigh range, the beam expands in size by 2 , and the peak
irradiance drops to 0.5. At a distance of 2 Rayleigh ranges, the peak irradiance drops to 0.2, and at 3 Rayleigh
ranges the irradiance decreases to 0.1.
Note that beam may be virtually propagated backward using the normal ZEMAX sign convention of a negative
thickness. Choosing the "End Surface" to be any of the surfaces in the LDE will show the beam intensity at that
surface. Note the phase of the beam along the axis has the correct Gouy shift.

A pinhole aperture
The sample file "Pinhole Aperture" illustrates spatial filtering. The first lens forms an aberrated image, which
can be seen by selecting the end surface to be 5. Surface 6 has a small circular aperture, which only allows the
central portion of the beam to pass. The second lens recollimates the beam. The spatially filtered beam can be
seen at the image of the stop on surface 10. There are several interesting things to note:
The transmitted irradiance drops to 0.12. The smaller the pinhole, the lower the transmitted irradiance.
Because the pinhole aperture changes the properties of the beam, the pilot beam parameters must be
recomputed. Note the option to do so is selected on surface 6.
A lens array
The sample file "Lenslet Array" uses the user defined surface DLL "US_ARRAY.DLL" to create a 7 x 7 array
of rectangular lenses. The focal distance is 100.0 mm, and the pilot beam will focus down to a waist size of about
10 micrometers. However, the lenslet array forms multiple focal spots. To see all of the spots requires high
sampling so that all the spots can be seen without aliasing. Notice the decrease in intensity and the aberrations
apparent in the spots towards the outside. It is also possible to see the rectangular structure in the diffracted image
spots due to the rectangular apertures of the lenslets.
Talbot imaging
Talbot imaging refers to the property of a beam forming an image of itself as it propagates. The image reveals
itself in both phase and amplitude modulations as the beam propagates. The sample file "Talbot Imaging" shows
a user defined aperture on surface 1 that consists of 20 rectangular slits, like a grating. A propagation of 20 mm
reveals the phase reversed Talbot image of the grating. An additional 20.0 mm shows the restored image of the
grating. The image is not perfect because of the finite extent of the grating. Note the region near the center of the
grating has a well formed image.
Fresnel lens
The sample file "Fresnel Zone Plate Lens" illustrates the focusing power of an aperture consisting of concentric
rings. The spacing of the rings is selected to block light from every other Fresnel zone, so that at a distance of
200.0 mm the light diffracted from all the zones interferes constructively. The constructive interference creates a
bright focused spot on axis; an effect not predicted by ray tracing. This file also illustrates the intensity of the beam
at several planes between the Fresnel zone plate and focus; note the concentric ripples caused by interference
from the edges of the zone plate.
Since ZEMAX does not have a multiple-zoned circular aperture, the zone plate was created by placing one
annular obscuration on each of several surfaces; all located at the same plane. The option to "Use Rays To
Propagate To Next Surface" is used to speed the computation through all the aperture surfaces.

Chapter 22 ZEMAX PROGRAMMING LANGUAGE
Introduction
The ZEMAX Programming Language (ZPL) is a macro language specifically designed for use with ZEMAX.
ZPL offers the power of user-extensibility. This means that if you need a particular calculation or graphical display
which is not "built in", you can write your own ZPL macro to do the job. ZPL macros can be stored on disk and
called from within ZEMAX.
ZPL is similar to the BASIC programming language, except not all BASIC constructs and keywords are
supported, and some new capabilities and functions unique to ray tracing have been added. If you are familiar
with BASIC, you will learn ZPL quickly. If you are not, or if you have never programmed before, don't panic! ZPL
is easy to use, and this chapter will give you instructions and examples to get you started.
!For a more powerful, compiled programming capability see ZEMAX EXTENSIONS on

page 607.
Creating ZPL macros

To create a ZPL macro, it is probably easiest to start with an existing macro that performs a task similar to the
one you want to achieve. If you are attempting to write your first ZPL macro, you may want to read the example
sections at the end of this chapter. Some example ZPL macros can also be found in the \MACROS directory, they
are described in the chapter "ZEMAX example files".
Use any ASCII text editor to create the ZPL file (such as the NOTEPAD editor). Name the file any descriptive
title you wish, but the file must end in the .ZPL extension.
There is a limit to the allowed complexity of any single line in a ZPL macro. If the "line too long" error occurs,
try breaking the line into several smaller lines.
Running ZPL macros
To run your ZPL macro, from the main menu, select Macros, Edit/Run ZPL Macros. The ZPL control dialog
box will appear with the following options:
Active File: A drop-down list of macros available. All the listed macros are ASCII files ending in the extension
.ZPL. The files must be in the directory for ZPL Macros; see the Chapter "File Menu" in the section on the
Directories tab of the Preferences dialog box. Macros may also be placed in any subdirectory within the macros
directory.
Close After Execution: If checked, the ZPL control dialog box will automatically close after the macro execution.
Quiet Mode: If checked, the default output text window will not be shown. This is useful for graphics macros
that do not generate useful text.
Status: During execution of the macro, ZEMAX will use this area to print a status message stating the line
number of the macro being executed. The status message is updated every quarter second.
Terminate: The terminate button will stop execution of the macro currently running.
Cancel: The cancel button terminates the current macro if one is running. If no macro is running, cancel closes
the ZPL control dialog box.
Edit: The edit button invokes the Windows NOTEPAD editor. The editor can be used to modify or rename the
macro.
View: The view button will display the contents of the macro file in a text window which can be scrolled or
printed. No editing is allowed in the view window.
Select the macro to run from the "Active File" list, and then click on Execute.
ZEMAX will begin running the macro. The entire macro will run, and any text output from PRINT statements
or error messages will be placed in a text file. When the macro has terminated, the text file will be displayed in a
window. The display of this output window can be suppressed using the CLOSEWINDOW keyword.
535 Chapter 22: ZEMAX PROGRAMMING LANGUAGE

Assigning ZPL macros to buttons
Frequently used macros may be assigned to buttons for convenient access. See Buttons 1-16, Buttons 17-
32, Buttons 33-48 on page 63 for details.
An overview of ZPL
A ZPL macro consists of a series of commands (called statements) which are stored in an ASCII text file. You
can create a ZPL macro with any text editor outside of ZEMAX (you can also edit the macro from within ZEMAX;
this will be discussed in a subsequent section). The ZPL language uses a syntax similar to BASIC for most (but
not all) commands. For example, the statement
x = 5
is one type of valid ZPL statement. There are several important things to note in this statement. First, no
declaration of variables is required. This means "x", which is called a variable, did not need to exist before the
value of 5 was assigned to it. If "x" had already been assigned a value, it would now be reassigned. There is no
need to declare the variable before using it. Second, no special symbol is required to terminate a statement, such
as ";" in C. Because of this, each ZPL statement must be on a line by itself.
ZPL supports "inline" function calls. This allows statements of the following forms to be used:
x = SQRT(5)
y = SINE(x)
z = SQRT(x+5*(7-x))
The functions SQRT (square root) and SINE (sine) are built in to ZPL. There are many of these single-operand
functions, all of which are defined in a subsequent section. Note that ZPL is not case-sensitive; SQRT() and sqrt()
are the same function. This documentation will use the convention of capital letters for functions and keywords,
and lower case for everything else.
All of the previously shown statements have one thing in common: they are all assignments. This means the
expression on the right side of the equals sign is evaluated, and the result is assigned to the variable designated
on the left. There is one other kind of statement in ZPL, and it is called a keyword.
One example of a keyword is PRINT. The PRINT keyword is followed by a list of items, separated by commas,
to be printed. For example, the ZPL statements
x = 3
y = 4
z = SQRT( x * x + y * y )
PRINT "The hypotenuse is ",z
will print the following to the screen:
The hypotenuse is 5.0000
Note that ZPL enforces operator precedence. Within any pair of parentheses, operations are carried out in a
particular order. ZPL uses the following precedence from highest to lowest: functions (such as SQRT), logical
operators (such as ==), multiplication and division, and then addition and subtraction. Parentheses always
override precedence, in this sense they have the highest priority.
There are many other keywords which will be described in a subsequent section.
ZPL variables
Variables provide temporary storage for numerical quantities whose exact value may not be known when the
macro is written, but is defined when the macro is run. ZEMAX performs most of the work for you when you need
a new variable. For example, the statement
x=5
presented earlier will cause ZEMAX to allocate memory for the new variable "x" and keep track of the value
associated with it. Once the variable is defined, it may be used in any subsequent expression. There are however
a few rules regarding ZPL variables:
Variable names must not contain any "special" characters that ZPL uses for logical operations or delimiting
such as (, ), =, +, -, *, /, !, >, <, ^, &, |, #, ", and the space character.
A variable cannot take on the same name as a keyword or function, such as THIC or RAYX. Since ZPL is not
Chapter 22: ZEMAX PROGRAMMING LANGUAGE 536

case-sensitive, you cannot use rayX or Thic to avoid this rule.
Each variable name is limited to 28 characters.
Violating these rules may generate a syntax error, however the macro may appear to run without errors and
will not compute what you really want. All ZPL variables are stored internally as 64-bit double precision numbers.
ZPL functions
ZPL functions can be used on the right hand side of an assignment, and in expressions which are arguments
to keywords. These functions may require no arguments, one argument, or multiple arguments. All functions
return a single value. Certain functions, such as PWAV() (primary wavelength) , return a value which does not
depend upon the argument, and therefore it is not required to provide one. The parentheses however, are still
required.
Many functions have corresponding keywords. For example, RADI() is a function which returns the radius of
the surface corresponding to the argument (RADI(3) returns the radius of surface 3). There is also a keyword
called RADI which is used to set the radius of that surface. See the keyword descriptions for more information.
In the following table, all ZPL functions are listed. If the syntax is given as FUNC(), then no arguments are
required. FUNC(x) indicates one argument is required, FUNC(x,y) indicates two arguments, etc.
ZPL FUNCTION LISTING
Function Argument Return value
ABSO(x) Numeric expression The absolute value of the expression.

ACOS(x) Numeric expression Arc cosine in radians.
APMN(x) Valid surface number The minimum radius value. For spider
apertures, this is the width of the arms. For
rectangular and elliptical apertures, it is the x-
half width of the aperture.
APMX(x) Valid surface number The maximum radius value. For spider
apertures, this is the number of arms. For
rectangular and elliptical apertures, it is the y-
half width of the aperture.
APXD(x) Valid surface number The aperture x-decenter value.
APYD(x) Valid surface number The aperture y-decenter value.
APTP(x) Valid surface number An integer code describing the aperture type on
ASIN(x) Numeric expression Arc sine in radians.
ASPR() (null) Aspect ratio of the current graphics device.
ATAN(x) Numeric expression Arc tangent in radians.
ATYP() (null) System aperture type code: 0 for EPD, 1 for F/
#, 2 for NA, and 3 for float by stop size.
AVAL() (null) System aperture value.
CONF() (null) Returns the current configuration number,
which is between 1 and the number of
configurations, inclusive.
CONI(x) Valid surface number Conic constant of the surface.
COSI(x) Numeric expression in radians Cosine of the expression.
CURV(x) Valid surface number Curvature of the surface.

EDGE(x) Valid surface number Edge thickness at the semi-diameter of that

surface.
EOFF() (null) End of file flag. Returns 1 if the end of file has
been reached, otherwise returns 0. Only valid
after execution of READ keyword.
ETIM() (null) The elapsed time in seconds since last TIMER.
EXPE(x) Numeric expression e to the power of the expression.
EXPT(x) Numeric expression 10 to the power of the expression.
FICL(vec#) Ve ctor # , be twe en 1 a nd 4 The fiber coupling efficiency. See Using the
inclusive. FICL() function on page 546
FLDX(x) Valid field number X-angle or height of specified field.
FLDY(x) Valid field number Y-angle or height of specified field.
FTYP() (null) Returns 0 if the current field type is angles in
degrees, 1 if object height, 2 if paraxial image
height, and 3 if real image height. Heights are
always in lens units (see UNIT).
FVAN(x) Valid field number Vignetting angle of specified field.
FVCX(x), FVCY(x) Valid field number Vignetting compression x or y of specified field.
FVDX(x), FVDY(x) Valid field number Vignetting decenter x or y of specified field.
FWGT(x) Valid field number Field weighting of specified field.
GABB(x) Valid surface number The glass catalog Abbe number of the glass for
GETT(window, line, col- The window number, line number, Extracts a numerical value from any open text
umn) and column number of the window. This feature allows computations
numerical value to extract. Each using any value ZEMAX can display in a
Columns are delimited by spaces. window.
GIND(x) Valid surface number The glass catalog d-light index of the glass for
GLCA(x) Valid surface number Global vertex x vector of the specified surface.
GLCB(x) Valid surface number Global vertex y vector of the specified surface.
GLCC(x) Valid surface number Global vertex z vector of the specified surface.
GLCM(surf, item) Surf must be a valid surface For item equal to 1-9, the return value is R11,
number, item is an integer between R12, R13, R21, R22, R23, R31, R32, or R33.
1 and 12. For item equal to 10-12, the return value is the
x, y, or z component of the global offset vector.
See Global Coordinate Reference Surface on
page 93.
GLCX(x) Valid surface number Global vertex x-coordinate of the specified
surface.
GLCY(x) Valid surface number Global vertex y-coordinate of the specified
surface.

GLCZ(x) Valid surface number Global vertex z-coordinate of the specified

surface.
GNUM(A$) Any string variable name If the string A$ is the name of a valid glass, such
as BK7, then GNUM returns the number of the
glass in the glass catalog. The glass number
can subsequently be used by GLAS to set the
glass type on a surface. If A$ does not
correspond to any glass in the catalog, GNUM
returns 0. GNUM returns -1 if the string is
"MIRROR".
GPAR(x) Valid surface number The glass catalog partial dispersion coefficient
of the glass for the specified surface.
GRIN(s, w, x, y, z) Surface #, wavelength #, x, y, and Returns the index of refraction at the specified
z coordinates x, y, z position on surface s at wavelength
number w. Works for gradient and non-gradient
media.
IMAE() Image analysis efficiency. Returns the geometric image analysis
efficiency. See the discussion Optimizing with
the IMAE operand on page 428.
INDX(surf) Valid surface number Index of refraction at the primary wavelength.
See ISMS.
INTE(x) Numeric expression Returns the largest integer not greater than the
argument.
ISMS(surf) Valid surface number If the surface is an odd mirror, or follows an odd
mirror but is not a mirror, the return value will be
one, otherwise the return value is 0.
LOGE(x) Positive numeric expression Log base e of the expression.
LOGT(x) Positive numeric expression Log base ten of the expression.
MAGN(x,y) x and y are any real numbers Computes the square root of x squared plus y
squared.
MAXF() (null) The maximum radial field angle in degrees, or
radial object or image height in lens units.
Which value is returned depends upon if the
fields are defined by angle, object height, or
image height.
MAXG() (null) The number of glasses currently loaded.

MCON(row, config, da- Row (operand number), Extracts data from any row and configuration of
ta) Configuration number, and data the Multi-Configuration Editor. This function is
value to extract from the Multi- similar to MCOP with extended capabilities for
Configuration Editor. extracting data.
If the row and config number are both zero,
MCON returns either the number of operands,
the number of configurations, or the active
configuration number for data = 0, 1, and 2,
respectively.
If the row number is between 1 and the number
of multi-config operands, and the config number
is zero, MCON returns the operand type,
integer 1, integer 2, integer 3, and string flag for
that specified row, for data = 0 through 4
respectively. The 3 integer values are used for
various purposes for different operands, such
as surface and wavelength numbers. The string
flag is 1 if the operand data is a string value,
such as a glass name, or 0 for numerical data.
If the row number is between 1 and the number
of multi-config operands, and the config number
is valid, MCON returns either the numerical
value or the string data for that operand.
Note that all string data returned by MCON
must be extracted with the $buffer command
after the call to MCON. For example, the
following code sample will place the name of
the operand on row 1 in a$:
dummy = MCON(1, 0, 0)
a$ = $buffer()
See also keyword SETMCOPERAND.
MCOP(row, config) R ow (o pe ra nd n um b er) an d Extracts data from any row and configuration of
configuration of the Multi- the Multi-Configuration Editor. A configuration
Configuration Editor. nu m be r o f z ero in dic ate s the c urre nt
configuration. See also keyword
SETMCOPERAND.
MFCN() (null) MFCN updates the lens, validates the merit
function, updates the merit function, then
returns the current merit function value. See
OPER.
NCON() (null) Returns the number of configurations.
NFLD() (null) The number of defined fields.
NOBJ(surf) Valid non-sequential surface The number of defined objects in the non-
number sequential surface.
NPAR(surf, object, surf is the surface number, object Returns a value from the parameter columns of
param) is the object number, and param is the non-sequential components editor. See
the parameter number of the value SETNSCPARAMETER on page 590.
to return.
NPOS(surf, object, surf is the surface number, object Returns a value from the position columns of
code) is the object number, and code is the non-sequential components editor. See
0-6 for reference object, x, y, z, tilt SETNSCPOSITION on page 591.
x, tilt y, and tilt z position to return,
respectively.

NPRO(surf, object, surf is the surface number, object Returns a numeric or string value from the
code, csg) is the object number, code values property pages of objects defined in the non-
are as defined for the sequential components editor. For information
SETNSCPROPERTY keyword, on the code values, see SETNSCPROPERTY
and csg is the coating and on page 591. String values may be extracted
scattering group number. using the $buffer() function after calling this
function with a code that returns string data. For
example, to extract the comment column of
object 5 from NSC surface 2 use these two
commands:
dummy = NPRO(2, 5, 1, 0)
a$ = $buffer()
The value dummy may be ignored. The string
variable a$ will contain the object comment.
NSDC(surf, object, pixel, The surf value indicates the If the object number corresponds to a detector
data) surface number of non-sequential object, then the coherent intensity data from the
group. specified pixel is returned. Coherent intensity
The object number indicates the consists of 4 numbers which are each summed
desired detector. over all incident rays: the real part, the
Pixel indicates which pixel value is imaginary part, the amplitude, and the
returned. incoherent intensity.
Data is 0 for real, 1 for imaginary, 2
for the amplitude sum, and 3 for the
coherent intensity.
See also NSTR on page 575.
NSDD(surf, object, pixel, The surf value indicates the If the object number is zero, then all detectors
data) surface number of non-sequential are cleared and the function returns zero. If the
group. object number corresponds to a detector object,
The object number indicates the then the incoherent intensity data from the
desired detector. specified pixel is returned.
Pixel indicates which pixel value is If the pixel number is zero, then the sum of the
returned, see the discussion to the flux or average flux/area for all pixels for that
right for details. detector object is returned. If the pixel number
Data is 0 for flux, 1 for flux/area, is -1, then the maximum flux or flux/area is
and 2 for flux/solid angle pixel. returned. If the pixel number is -2, then the
minimum flux or flux/area is returned. If the pixel
See also NSTR on page 575. number is -3, the number of rays striking the
detector is returned. If the pixel number is -4,
the RMS deviation from the mean is returned.
If data is 2 then the pixel number must be
greater than 0, otherwise, a value of 0 is
returned. Only data values 0 and 1 (for flux and
flux/area) are supported for faceted detectors.
NSUR() (null) The number of defined surfaces.
NWAV() (null) The number of defined wavelengths.
OCOD(A$) Any string variable or literal that is The optimization operand code number used by
the name of a ZEMAX optimization the OPEV function.
operand.

ONUM(A$) Any string variable name If the string A$ is the name of a valid
optimization operand, such as EFFL, then
ONUM returns the id number of the operand.
The operand id number can subsequently be
used by SETOPERAND to set the operand type
in the merit function editor. If A$ does not
correspond to any operand name, ONUM
returns 0.
OPDC() (null) The optical path difference. Valid only after a
RAYTRACE or RAYTRACEX call. OPDC will
not return valid data if the chief ray cannot be
traced.
OPER(row, col) R ow (o pe ra nd n um b er) an d Extracts data from any row and column of the
column (data type) of the Merit Merit Function Editor. The row is the same as
Function Editor. the operand number; the column is 1 for type, 2
for int1, 3 for int2, 4-7 for Hx-Py, 8 for target, 9
for weight, 10 for value, and 11 for percent
contribution. OPER does not update the lens or
the merit function, it returns the current data.
S e e a l s o O P E V, M F C N a n d k e y w o r d
SETOPERAND.
OPEV(code, int1, int2, Code is the optimization operand Computes the same value as any optimization
hx, hy, px, py) code (see function OCOD), and operand would, without the need to add the
int1, int2, hx, hy, px, and py values operand to the merit function. This is useful for
for the operand. See Optimization computing numbers already available from the
operands on page 391 optimization operands. For example, to
compute the EFL from the EFFL operand use
this code:
C = OCOD("EFFL")
E = OPEV(C, 0, 1, 0, 0, 0, 0)
See also OPER, OCOD, MFCN and keyword
SETOPERAND.
OPTH(x) Valid surface number The optical path length along the ray to the
specified surface. Unlike RAYT and RAYO,
OPTH considers the phase added by diffractive
surfaces such as gratings, holograms, and
binary optics. Valid only after a RAYTRACE or
RAYTRACEX call. OPTH will not return valid
data if the chief ray cannot be traced.
PARM(n,s) Valid parameter number and Parameter "n" of surface "s".
surface number
PARn(x) Valid surface number Obsolete, use PARM instead.
PMOD() (null) 0 if paraxial mode is off, else 1.
POWR(x,y) x is positive, y is any number Computes x to the power of y. X must be
positive, or the absolute value is used.
PVHX() (null) "Hx" parameter from the ZPLM optimization
operand.
PVHY() (null) "Hy" parameter from the ZPLM optimization
operand.
PVPX() (null) "Px" parameter from the ZPLM optimization
operand.

PVPY() (null) "Py" parameter from the ZPLM optimization

operand.
PWAV() (null) The primary wavelength number.
RADI(x) Valid surface number Radius of curvature of surface. If the surface
has an infinite radius, RADI returns 0.0. This
possibility must be considered to avoid potential
divide by zero errors.
RAGX(x) Valid surface number The global x coordinate of the ray intercept.
Valid only after a RAYTRACE or RAYTRACEX
call.
RAGY(x) Valid surface number The global y coordinate of the ray intercept.
call.
RAGZ(x) Valid surface number The global z coordinate of the ray intercept.
call.
RAND(x) Positive numeric expression Random floating point number uniformly
distributed between 0 and the expression.
RANX(x) Valid surface number The X-direction cosine of the surface normal.
call.
RANY(x) Valid surface number The Y-direction cosine of the surface normal.
call.
RANZ(x) Valid surface number The Z-direction cosine of the surface normal.
call.
RAYE() (null) The ray-trace error flag, 0 if no error. Valid only
after a RAYTRACE or RAYTRACEX call. See
the RAYTRACE keyword for details.
RAYL(x) Valid surface number The X-direction cosine of the ray following the
surface. Valid only after a RAYTRACE or
RAYTRACEX call.
RAYM(x) Valid surface number The Y-direction cosine of the ray following the
RAYTRACEX call.
RAYN(x) Valid surface number The Z-direction cosine of the ray following the
RAYTRACEX call.
RAYO(x) Valid surface number The ray optical path length from the previous
surface to the specified surface. The optical
path length is the path length times the index of
refraction, either or both of which may be
negative. For rays inside a non-sequential
surface, RAYO returns the sum of the path
length times the index of refraction of all objects
the ray passed through.Valid only after a
RAYTRACE or RAYTRACEX call. See also
OPTH and RAYT.

RAYT(x) Valid surface number The ray path length from the previous surface
to the specified surface. The path length may be
negative. Valid only after a RAYTRACE or
RAYTRACEX call. See also OPTH and RAYO.
For rays inside a non-sequential surface, RAYT
returns the sum of the path lengths along all
segments. See also RAYO.
RAYV() (null) 0 if ray was not vignetted, else vignetted surface
number. Valid only after a RAYTRACE or
RAYTRACEX call.
RAYX(x) Valid surface number The X-coordinate of the ray intercept. Valid only
after a RAYTRACE or RAYTRACEX call.
RAYY(x) Valid surface number The Y-coordinate of the ray intercept. Valid only
RAYZ(x) Valid surface number The Z-coordinate of the ray intercept. Valid only
RELI(f) Relative illumination RI for the specified field position.
SAGG(x,y,z) x,y are the coordinates in lens units Computes the sag in lens units at the specified
at the surface number z point on the surface.
SCOD(A$) Any string variable or literal that is The surface type code of the surface. See
the name of a ZEMAX surface function STYP and SURFTYPE on page 598.
type. The valid names are 8
character strings displayed on the
prescription data report for each
surface type in use. These are not
the same names that are displayed
in the Lens Data Editor.
SCOM(A$, B$) Any two string variable names If the strings are equal, SCOM returns 0. If A$
is less than B$, then SCOM returns a value less
than 0; otherwise, a value greater than 0.
SDIA(x) Valid surface number Semi-diameter of surface.
SIGN(x) Numeric expression Returns -1 if the argument is less than zero, 0 if
the argument is zero, and +1 if the argument is
positive.
SINE(x) Numeric expression in radians Sine of the expression.
SLEN(A$) Any string variable name The number of characters in the string variable
A$
SOLV(surf, code, Surf is the surface number. Code is The return value is data about the solve type for
param) 0 for curvature, 1 for thickness, 2 the specified surface and data. If param is zero,
for glass, 3 for conic, 4 for semi- then an integer corresponding to the solve type
diameter, and 5 for TCE. Param is is returned. For param between 1 and 3, the
an integer between 0 and 3. data is the solve parameters. For more
information on the solve data and the meaning
of the solve parameters, see SOLVES on
page 379.

SPRO(surf, code) Surf is the surface number and Returns a numeric or string value from the
code values are as defined for the property pages or editors of surfaces defined in
S E T S U R FA C E P R O P E R T Y the lens data editor. For information on the code
keyword. values, see SETSURFACEPROPERTY on
page 593. String values may be extracted using
the $buffer() function after calling this function
with a code that returns string data. For
example, to extract the comment column of
surface 4 use these two commands:
dummy = SPRO(4, 1)
a$ = $buffer()
The value dummy may be ignored. The string
variable a$ will contain the surface comment.
SQRT(x) Positive numeric expression Square root of the expression.
STDD(surf, parm) Surface number, tilt and decenter The value of the specified surface tilt and
data parameter decenter parameter. For a description of the
parameter numbers, see SETSTDD on
page 593.
STYP(n) Surface number n The surface type code of the surface. See
function SCOD and SURFTYPE on page 598
SVAL(A$) Any string variable name String value. Returns a floating point value of
the string A$.
TANG(x) Numeric expression in radians Tangent of the expression.
TMAS() Total mass in grams The total mass in grams of the lens from surface
1 to the image surface.
THIC(x) Valid surface number Thickness of the surface.
UNIT() (null) Returns 0, 1, 2, or 3, if the current unit type is
millimeters, centimeters, inches, or meters,
respectively.
VEC1(x) Positive subscript value Returns the value of the array variable at the
specified subscript.
VERS() (null) Returns the four digit version code. A typical
version number is 10000 for version 10.0.
Subsequent versions of ZEMAX will return
larger version codes.
WAVL(x) Valid wavelength number Wavelength in micrometers.
WWGT(x) Valid wavelength number Wavelength weighting.
XMIN() (null) The minimum x-coordinate in the graphics
window.

XMAX() (null) The maximum x-coordinate in the graphics

window.
YMIN() (null) The minimum y-coordinate in the graphics
window.
YMAX() (null) The maximum y-coordinate in the graphics
window.
Using the FICL() function

The FICL function computes fiber coupling. Because the function takes so many arguments, the arguments
are placed in a vector array and then the function is called. The vector array must be one of the four VEC arrays
already defined. The values placed in the vector array are defined as follows:
0 = Sampling (1 for 32 x 32, 2 for 64 x 64, 3 for 128 x 128, etc..)
1 = Wavelength #
2 = Field #
3 = Ignore source flag (0 for false, 1 for true)
4 = Source NA x-direction
5 = Receiver NA x-direction
6 = Source x angle (deg)
7 = Source y angle (deg)
8 = Receiver tilt about x (deg
9 = Receiver tilt about y (deg)
10 = Receiver decenter x
11 = Receiver decenter y
12 = Receiver decenter z
13 = Align source to chief ray flag (0 for false, 1 for true)
14 = Align receiver to chief ray flag (0 for false, 1 for true)
15 = Use polarization flag (0 for false, 1 for true)
16 = Source NA y-direction (if zero then NAy = NAx)
17 = Receiver NA y-direction (if zero then NAy = NAx)
The fiber coupling is computed by calling FICL(n) where n is the vector number containing the argument list.
ZPL math operations
ZPL macros can include basic math operations such as add, subtract, multiply, and divide. The syntax for each
is shown below.
x = y + z
x = y - z
x = y * z
x = y / z
All other operations are supported only through the use of ZPL functions or ZPL logical operators, both
described in subsequent sections.
ZPL logical operators
Logical operators are used to construct complex statements which ultimately evaluate to one or zero. Most
logical operations take the form (left_expression) (operator) (right_expression), similar to mathematical state-
ments such as 1 + 2. The exception is the not operator "!" which takes only a single argument, of the form !
(right_expression). The logical operators use the convention that zero is "false" and any non-zero value is "true".
The not operator returns 1 (true) if the (right_expression) is 0 (false) and returns 0 (false) if (right_expression) is
non-zero (true). One common use of the not operator is in IF statements such as:
IF !x THEN PRINT "x is zero."
The other logical operators can be also be used as part of the argument in IF statements. For example, an IF
statement may contain two conditions which must both be true for the THEN statement to be executed:
IF ( x > 1 ) & ( y < 2 ) THEN PRINT "Both conditions are true."

These two conditions are related by an "and" expression denoted by &. Note the parentheses are used to force
precedence. ZPL supports the logical operators described in the following table.
ZPL LOGICAL OPERATORS
Logical Description
& And, returns 1 only if both expressions are non-zero.

| Or, returns 1 if at least one expression is non-zero.
^ Xor, returns 1 if only one expression is non-zero.
! Not, returns 0 if (right_expression) is non-zero, else returns 1.
== Equality, returns 1 if expressions are equal.
> Greater than, returns 1 if left_expression is greater than right_expression.
< Less than, returns 1 if left_expression is less than right_expression.
>= Greater than or equal to, returns 1 if left_expression is greater than or equal to right_expression.
<= Less than or equal to, returns 1 if left_expression is less than or equal to right_expression.
!= Inequality, returns 1 if expressions are unequal.
ZPL string variables and operations

ZPL supports string variables and basic string operations. String variables can hold a maximum of 260
characters. String variables do not need to be declared, but can be created at any time using a defining statement
such as:
newstring$ = "Here is the new string"
Note that string variables are distinguished from numeric variables by the presence of the $ character at the
end of the string. String variables can be concatenated using the + operator. The syntax is:
C$ = A$ + B$
Concatenation can also include constant strings:
total$ = "A$ is " + A$ + " and B$ is " + B$
There are also string functions which can be used to extract text data, such as
title$ = $LENSNAME()
Note the function $LENSTITLE() starts with the $ character. This identifies the function as returning a string
result. The string functions can be used in a defining statement such as
this$ = "Here is the lens title: " + $LENSNAME()
String variables are printed just like other strings:
PRINT "Here is A$: ", A$
Note that the PRINT function can only print single string variables; there is no support for the concatenation
operand or string functions inside print statements. The correct procedure is to concatenate the strings into a new
string and then print the new string:
A$ = B$ + C$
PRINT A$
Alternatively, the comma acts as a concatenation operand:
PRINT A$, B$, C$
String functions cannot be printed directly like this:
PRINT $LENSNAME() ! NOT CORRECT !!!
Instead, the correct procedure is to assign the function result to a variable and then print the variable:

Z$ = $LENSNAME()
PRINT Z$
One very important function is $STR(). This function accepts as an argument any expression which evaluates
to a number. $STR is useful for formatting numeric data into strings:
A$ = "The expression evaluates to " + $STR( SQRT(3*3 + 4*4 + z*y) )
The inverse function is SVAL(A$), which converts a string into a floating point value.
The following table lists the available string functions.
ZPL STRING FUNCTIONS
Function Description
$BUFFER() Returns the current string in the lens buffer. This function is used to extract string
data from various ZPL keywords and functions.
$COAT(i) Returns the coating name for the ith surface.
$COMMENT(i) Returns the comment string for the ith surface.
$DATE() Returns the current date and time string. The formatting is specified by the Date/
Time control settings described under Miscellaneous on page 60.
$EXTENSIONPATH() Returns the path name for ZEMAX extensions.
$FILENAME() Returns the current lens file name, without the path.
$FILEPATH() Returns the current lens file name, with the complete path.
$GETSTRING(A$, n) Returns the nth sub-string for the string A$. For example, if A$ = "one two three",
then $GETSTRING(A$, 2) returns "two". Spaces and commas are used as
delimiters, although commas count as separate sub-strings. See $LEFTSTRING
and $RIGHTSTRING.
$GLASS(i) Returns the glass name of surface number i.
$LEFTSTRING(A$, n) Returns the left most n characters in the string A$. If A$ has fewer than n
characters, the remaining spaces will be padded with blanks. This allows
formatting of strings with a fixed length. See $GETSTRING and $RIGHTSTRING.
$LENSNAME() Returns the lens title defined in the General System dialog box.
$NOTE(line#) Returns the notes information defined in the General System dialog box. Because
the notes may be very long, $NOTE returns the characters from the notes in
groups called lines. A line ends when a newline (carriage return) character is
found, or when the total number of consecutive characters on the line reaches 100,
whichever comes first. The line# indicates which line in the notes is to be returned.
For example, $NOTE(1) returns the first line in the notes, up to the first newline
character. If there are more than 100 consecutive characters before the first
newline is found, then line 1 will be the first 100 characters, and line 2 will be the
remainder of the line up to the first newline, or the next 100 characters, whichever
comes first.
The maximum number of characters in the notes is currently 4000 (because of
space limitations, it may not be possible to edit this many characters in the General
dialog however). $NOTE will return a null (empty) string if there are no defined
characters in the notes for the specified line.
The function SLEN can be used to determine the actual number of characters
returned by $NOTE using this syntax:
A$ = $NOTE(1)
N = SLEN(A$)
The value N will be the integer number of characters in the string A$.
$OBJECTPATH() Returns the path name for NSC object files.

Function Description
$PATHNAME() Returns the path name only for the current lens file. This is useful for determining
the directory where the lens file is stored.
$RIGHTSTRING(A$, n) Returns the right most n characters in the string A$. If A$ has fewer than n
characters, the remaining spaces will be padded with blanks. This allows
formatting of strings with a fixed length. See $GETSTRING and $LEFTSTRING.
$STR(expression) Returns a string formatted using the format defined by the FORMAT keyword. The
numeric expression may be any equation, including combinations of constants,
variables, and functions. See function SVAL(A$) to convert strings to numbers.
$TEMPFILENAME Returns the name of a temporary file, with complete path, suitable for temporary
storage of text or binary data. See keyword GETTEXTFILE.
$UNITS() Returns either MM, CM, IN, or M, depending upon the current lens units.
ZPL string logical operators

String logical operators are very similar to the (numeric) logical operators defined in a previous section. The
key difference is that the expressions being compared are strings rather than numbers. The supported string
logical operators are defined in the following table.
ZPL STRING LOGICAL OPERATORS
Logical Description
$== Equality, returns 1 if left_string and right_string are identical.

$> Greater than, returns 1 if left_string is greater than right_string.
$< Less than, returns 1 if left_string is less than right_string.
$>= Greater than or equal to, returns 1 if left_string is greater than or identical to right_string.
$<= Less than or equal to, returns 1 if left_string is less than or identical to right_string.
$!= Inequality, returns 1 if left_string and right_string are not identical.
For example, an IF statement may compare strings as follows:

A$ = "TEST"
BS = "TEST"
IF (A$ $== B$) THEN PRINT "Strings are identical."
ZPL keywords
ZPL keywords provide the capability to direct program flow, to generate output, and to perform certain crucial
tasks, such as ray tracing and modifying the lens prescription. Each keyword is described in detail in the following
sections. Certain keywords with similar or related functions are grouped together.
APMN, APMX
Purpose:
Used to set the aperture minimum/maximum radius of a surface.
Syntax:
APMN(surface) = (new_value)
APMX(surface) = (new_value)
Discussion:
These keywords requires two expressions, one to specify the surface number and the other to define the new
value. The expression (surface) is evaluated and then rounded down to an integer to yield the surface number.
If the surface number is less than zero or greater than the number of surfaces, the command is ignored.

Otherwise, the expression (new_value) is evaluated, and assigned to the aperture minimum (APMN) or aperture
maximum (APMX) radius.
The aperture minimum/maximum radius data is used by all of the aperture types, although the meaning of the
data depends upon which aperture is being used, as described in the following table.
APERTURE TYPES AND APMN USAGE
Aperture type APMN is used for: APMX is used for:
None. Nothing Nothing

Circular aperture Setting the minimum aperture Setting the maximum aperture
radius radius
Circular obscuration Setting the minimum aperture Setting the maximum aperture
radius radius
Spider Setting the width of each spider Setting the number of spider arm
arm
Rectangular aperture Setting the x-half width Setting the y-half width
Rectangular obscuration Setting the x-half width Setting the y-half width
Elliptical aperture Setting the x-half width Setting the y-half width
Elliptical obscuration Setting the x-half width Setting the y-half width
User aperture Nothing Nothing
User obscuration Nothing Nothing
Floating aperture Nothing Nothing
Example:
APMN(3) = 1.75
APMX(3) = 3.50
APXD, APYD
Purpose:
Used to set the aperture x/y decenter of a surface.
Syntax:
APXD(surface) = (new_value)
APYD(surface) = (new_value)
Discussion:
These keywords requires two expressions, one to specify the surface number and the other to define the new
Otherwise, the expression (new_value) is evaluated, and assigned to the aperture x-decenter (APXD) or aperture
y-decenter (APYD).
Example:
APXD(3) = -3.6
APYD(5) = -1 * APYD(3)
APTP
Purpose:
Used to set the aperture type of a surface.
Syntax:

APTP(surface) = integer_code
Discussion:
This keyword requires two expressions, one to specify the surface number and the other to define the new
Otherwise, the expression (new_value) is evaluated, rounded to an integer and assigned to the aperture type.
The aperture type is stored as an integer code described in the following table.
APERTURE TYPE CODES
Integer code Aperture type

0 None.
1 Circular aperture
2 Circular obscuration
3 Spider
4 Rectangular aperture
5 Rectangular obscuration
6 Elliptical aperture
7 Elliptical obscuration
8 User aperture
9 User obscuration
10 Floating aperture
Example:
APTP(3) = 1
ATYP, AVAL
Purpose:
Used to set the system aperture type and value.
Syntax:
ATYP = (code)
AVAL = (new_value)
Discussion:
These keywords are used to define the system aperture type and value. The UPDATE command must be
issued before the new data takes effect. The aperture type is defined using one of the following codes:
SYSTEM APERTURE TYPE CODES
Aperture type Code
Entrance Pupil Diameter 0

Image Space F/# 1
Object Space Numerical Aperture 2
Float By Stop Size 3

Example:
! Set the EPD to be 35
ATYP = 0
AVAL = 35.0
BEEP
Purpose:
Makes an audible beep sound.
Syntax:
BEEP
Discussion:
This command can be used to alert the user when a calculation is finished or input is required.
COAT
Purpose:
COAT is used to set the name of the coating on a surface.
Syntax:
COAT (surface), coatingname
Discussion:
This keyword requires the surface number, and the name of the new coating to apply. Leave the name blank
if the surface is to be uncoated.
Example:
COAT 2, WAR
A$ = "AR"
COAT 3, A$
Related Keywords:
UPDATE
CLOSE
Purpose:
Closes an ASCII file previously opened by the OPEN command.
Syntax:
CLOSE
Discussion:
See the description for OPEN.
CLOSEWINDOW
Purpose:
Suppresses the display of the default output window. This keyword can also be used to close any open analysis
windows.
Syntax:
CLOSEWINDOW
CLOSEWINDOW n
Discussion:
CLOSEWINDOW (with no argument "n" provided) is used to run the ZPL macro in "quiet" mode. The text
window normally displayed at the end of the macro execution will not be displayed if the CLOSEWINDOW
keyword is included at any line in the macro. CLOSEWINDOW has no other effect on macro execution.
CLOSEWINDOW (with an integer argument "n" provided) will close analysis window number n.

COLOR
Purpose:
Sets the current pen color to use for graphics line and text functions.
Syntax:
COLOR (n)
Discussion:
The value n must evaluate to an integer. If the argument n is zero, the color will be set to black. Otherwise,
the color defined by the integer value will be used for all subsequent line and text commands in graphics mode.
There are 12 valid colors plus black; numbered 0-12. The colors are normally used by ZEMAX for fields and
wavelengths; color 1 is for either field 1 or wavelength 1, etc.
COMMENT
Purpose:
COMMENT is used to set the comment of a surface.
Syntax:
COMMENT (surface) = "literal string"
COMMENT (surface) = stringvariable$
Discussion:
This keyword requires two expressions, one to specify the surface number and the other to define the comment
string. COMMENT also accepts a string variable which contains the comment.
Example:
COMMENT 4 = "Comment for surface 4"
COMMENT 6 = A$
Related Keywords:
UPDATE
CONI
Purpose:
CONI is used to set the conic constant of a surface.
Syntax:
CONI (surface) = (new_value)
Discussion:
Otherwise, the expression (new_value) is evaluated and assigned.
Example:
CONI 1 = -1
Related Keywords:
UPDATE
COPYFILE
Purpose:
COPYFILE is used to make a copy of a file.
Syntax:
COPYFILE sourcefilename newfilename
Discussion:

This keyword requires two file names, defined as literal string expressions in quotes or as string variables. The
file sourcefilename is copied into the new file newfilename. If newfilename already exists it is overwritten without
warning.
Example:
COPYFILE "C:\source.dat" "C:\copy.dat"
Related Keywords:
DELETEFILE
RENAMEFILE
CURV
Purpose:
CURV is used to set the curvature of a surface.
Syntax:
CURV (surface) = (new_value)
Discussion:
Example:
CURV(3) = - CURV(4) + .001
Related Keywords:
UPDATE
DELETE
Purpose:
Deletes a surface from the spreadsheet.
Syntax:
DELETE (n)
Discussion:
The value n must evaluate to an integer expression. See also INSERT.
Example:
DELETE 5
DELETE i+2*j
DELETECONFIG
Purpose:
DELETECONFIG deletes an existing configuration in the multi-configuration editor.
Syntax:
DELETECONFIG (config)
Discussion:
The value config must evaluate to an integer expression greater than 0 and less than or equal to the current
number of configurations. See also INSERTCONFIG and DELETEMCO.
DELETEFILE
Purpose:
DELETEFILE is used to delete a file.
Syntax:

DELETEFILE filename
Discussion:
This keyword requires a file names, defined as literal string expression in quotes or as a string variable.
Example:
DELETEFILE XFILE$
Related Keywords:
COPYFILE
RENAMEFILE
DELETEMCO
Purpose:
DELETEMCO deletes an existing operand in the multi-configuration editor.
Syntax:
DELETEMCO (row)
Discussion:
The value row must evaluate to an integer expression greater than 0 and less than or equal to the current
number of operands. See also INSERTMCO and DELETECONFIG.
DELETEMFO
Purpose:
DELETEMFO deletes an existing operand in the merit function editor.
Syntax:
DELETEMFO (row)
Discussion:
number of operands. See also INSERTMFO.
DELETEOBJECT
Purpose:
Deletes an existing NSC object.
Syntax:
DELETEOBJECT (surf), (object)
Discussion:
The value surf must evaluate to an integer expression, and the specified surface must be a non-sequential
component surface. Use a surf value of 1 if the program mode is non-sequential. The value object must evaluate
to an integer that is between 1 and the current number of objects plus 1, inclusive. The specified object will be
deleted, renumbering other objects as required. See also INSERTOBJECT and SETNSCPROPERTY.
Example:
DELETEOBJECT 1, 1
EDVA
Purpose:
Sets the extra data values for a surface.
Syntax:
EDVA surf, ed_value, new_value
Discussion:

The surf expression indicates which surface is to be modified. The ed_value expression must evaluate to an
integer corresponding to one of the extra data value positions. See the chapter "Surface types" for information
on the extra data values. The new_value expression is evaluated and is used for the new value of the extra data
value on that surface.
Example:
EDVA 5, 6, x+y
END
See GOSUB.
EXPORTBMP
Purpose:
Exports any graphic window as a BMP file.
Syntax:
EXPORTBMP winnum, filename, delay
Discussion:
The winnum value may be either an integer or an expression that evaluates to an integer. The integer winnum
corresponds to the graphic window number that should be saved to a file. ZEMAX numbers windows sequentially
as they are opened, starting with 1. Any closed windows are deleted from the window list, without renumbering
the windows which remain. Any windows opened after another window has been closed will use the lowest
window number available. The filename should be the full file name including the path, but with no extension.
ZEMAX will automatically add the BMP extension. The optional delay parameter specifies a time delay in
milliseconds. For some complex graphics, a delay is required to allowed the graphic to be completely redrawn
and the screen capture to complete. If the JPG files appear incomplete, try a delay value of 500 - 2500
milliseconds. See also EXPORTJPG and EXPORTWMF.
Example:
EXPORTBMP 1, "C:\TEMP\MYBMPFILE"
EXPORTBMP k, A$
EXPORTCAD
Purpose:
Exports lens data as an IGES, STEP, SAT, or STL file for import into CAD programs.
Syntax:
EXPORTCAD filename
Discussion:
For a full description of CAD file export, see Export IGES/SAT/STEP Solid on page 208. The filename can
be a literal string, such as "C:\DATA\FILE.IGS" or a string variable name, such as MYFILENAME$. There are
many other parameters required to define the export. These other data are placed in the vector 1 variable as
follows:
VEC1(1): The file type. Use 0 for IGES, 1 for STEP, 2 for SAT, 3 for STL.
VEC1(2): The number of spline points to use (if required on certain entity types).
VEC1(3): The First surface to export.
VEC1(4): The last surface to export.
VEC1(5): The layer to place ray data on.
VEC1(6): The layer to place lens data on.
VEC1(7): Use 1 to export dummy surfaces, otherwise use 0.
VEC1(8): Use 1 to export surfaces as solids, otherwise use 0.
VEC1(9): The ray pattern. Use 0 for XY, 1 for X, 2 for Y, 3 for ring, 4 for list, 5 for none.
VEC1(10): The number of rays.

VEC1(11): The wave number. Use 0 for all.
VEC1(12): The field number. Use 0 for all.
VEC1(13): Use 1 to delete vignetted rays, otherwise use 0.
VEC1(14): The dummy surface thickness in lens units.
VEC1(15): Use 1 to split rays from NSC sources, otherwise use 0.
VEC1(16): Use 1 to scatter rays from NSC sources, otherwise use 0.
VEC1(17): Use 1 to use polarization when tracing NSC rays, otherwise use 0. Polarization is automatically
selected if splitting is specified.
VEC1(18) Use 0 for the current configuration, 1 - n for a specific configuration where n is the total number of
configurations, n+1 to export "All By File", n+2 to export "All By Layer", and n+3 for "All At Once". For a more
detailed explanation of the configuration setting, see Export IGES/SAT/STEP Solid on page 208.
Example:
VEC1(1) = 0
VEC1(2) = 32
VEC1(3) = 1
etc...
VEC1(18) = 0
EXPORTCAD "C:\TEMP\MYCADFILE.IGS"
EXPORTJPG
Purpose:
Exports any graphic window as a JPG file.
Syntax:
EXPORTJPG winnum, filename, delay
Discussion:
See EXPORTBMP.
Example:
EXPORTJPG 3, "C:\TEMP\MYJPGFILE"
EXPORTJPG k, FILENAME$, 500
EXPORTWMF
Purpose:
Exports any graphic window as a Windows Metafile. Unlike EXPORTBMP and EXPORTJPG, EXPORTWMF
requires the filename to have the desired extension included in the filename argument. The metafile format will
be whatever the current selected metafile type is, see Graphics on page 59.
Syntax:
EXPORTWMF winnum, filename
Discussion:
See EXPORTBMP.
Example:
EXPORTWMF 3, "C:\TEMP\MYWMFFILE.EMF"
EXPORTWMF k, FILENAME$
FINDFILE
Purpose:
Used to find the names of files stored on disk.
Syntax:
FINDFILE TEMPNAME$, FILTER$
Discussion:

This keyword requires two expressions, one to specify the string variable name to store the file name in, and
another string variable which contains a "filter" string. The filter string usually specifies a path name and wildcards
appropriate to the desired file type. See the example below.
FINDFILE is useful for listing all files of a certain type in a directory, or for analyzing large numbers of similar
lens files. To reset FINDFILE back to the first file of any type, just call FINDFILE with a different filter, then call
FINDFILE again with the original filter name. Each time FINDFILE is called with a new filter, it resets back to the
first file that meets the filter specifications.
Example:
FILTER$ = "C:\ZEMAX\*.ZMX"
PRINT "Listing of all ZEMAX files in ", FILTER$
FINDFILE TEMPFILE$, FILTER$
LABEL 1
if (SLEN(TEMPFILE$))
PRINT TEMPFILE$
FINDFILE TEMPFILE$, FILTER$
GOTO 1
ENDIF
PRINT "No more files."
FLDX, FLDY, FWGT, FVDX, FVDY, FVCX, FVCY, FVAN
Purpose:
FLDX and FLDY are used to change the field point definitions. FWGT is used to change the field weight. FVDX
and FVDY change the x and y vignetting decenter factors. FVCX and FVCY change the x and y vignetting
compression factors. FVAN changes the vignetting angle.
Syntax:
FLDX (field number) = (new_value)
FLDY (field number) = (new_value)
FWGT (field number) = (new_value)
FVDX (field number) = (new_value)
FVDY (field number) = (new_value)
FVCX (field number) = (new_value)
FVCY (field number) = (new_value)
FVAN (field number) = (new_value)
Discussion:
These keywords require two expressions, one to specify the field number and the other to define the new value.
The expression (field number) is evaluated and then rounded down to an integer to yield the field number. If the
field number is less than 1 or greater than the number of fields, the command is ignored. Otherwise, the
expression (new_value) is evaluated and assigned. The UPDATE command must be issued before the new data
takes effect.
Example:
FLDX 1 = 0.0
FLDY 1 = 12.5
FWGT 1 = 1.0
Related Functions:
FLDX, FLDY, FWGT, MAXF, FTYP, FVDX, FVDY, FVCX, FVCY, FVAN
Related Keywords:
UPDATE, FTYP, NUMFIELD
FOR, NEXT
Purpose:
The FOR and NEXT keywords define a program block which is executed a specific number of times in a loop.
Syntax:
FOR variable = start_value, stop_value, increment
statements...

NEXT
Discussion:
The keyword FOR marks the beginning of a group of statements to be executed a multiple number of times.
FOR requires a variable to be specified which acts as a counter (it need not be an integer), a starting value for
the counter, a stop value, and an increment. The NEXT keyword marks the end of the group of statements. FOR-
NEXT loops may be nested. The number of FOR and NEXT statements must be the same.
Upon reaching a FOR statement, the expressions for the start, stop, and increment values are evaluated and
saved. The stop and increment values are not evaluated again, even if the expressions defining the values consist
of variables whose values change within the program block. Only the values valid at the beginning of the FOR
loop are used.
If the start value and stop value are the same, the loop executes exactly once. If the start value is less than
the stop value, then the loop continues until the counter variable is greater than the stop value. If the start value
is greater than stop value, then the loop continues until the counter variable is less than the stop value.
Example:
FOR i = 1, 25, 1
PRINT i
NEXT
j = 5
k = 0
FOR i = j, j + 5, 2
k = i + j + k
NEXT
FORMAT
Purpose:
Specifies the numerical precision format for subsequent PRINT statements.
Syntax:
FORMAT m.n [EXP]
Discussion:
The integers m and n are separated by a decimal point. The value m refers to the total number of characters
to be printed, even though some of them may be blank. The value n refers to the number of places to display
after the decimal point. Therefore, FORMAT 8.4 will cause subsequent PRINT statements to print 8 characters,
with 4 numbers after the decimal point. FORMAT .5 will cause PRINT to show 5 decimal places, and as many
total places as needed. FORMAT only affects numeric output from PRINT. If a number is too large to fit within
the m decimal places, then the m portion of the FORMAT statement will be ignored. The optional keyword EXP
after the m.n expression indicates exponential notation should be used.
Example:
x = 1.123
FORMAT 12.0
PRINT "An integer portion=", x
FORMAT 12.8
PRINT "A decimal portion = ", x
FORMAT 12.8 EXP
PRINT "Exponential notation = ", x
FTYP
Purpose:
FTYP is used to change the type of field definition between angle in degrees, object height, and image height.
Syntax:
FTYP = (new_value)
Discussion:

The (new value) is an expression that must evaluate to 0, 1, 2, or 3. Use 0 to set the field type to angle in
degrees, 1 for object height in lens units, 2 for paraxial image height in lens units, 3 for real image height in lens
units. Setting the field type does not change the field values, nor is the system updated. Usually, the field values
must be adjusted with the FLDX and FLDY keywords whenever the field type is changed. The UPDATE command
must be issued before the new data takes effect.
Related Functions:
FLDX, FLDY, FWGT, MAXF
Related Keywords:
UPDATE, FLDX, FLDY, FWGT
GCRS
Purpose:
GCRS is used to change the Global Coordinate Reference Surface. See Global Coordinate Reference
Surface on page 93.
Syntax:
GCRS = (new_value)
Discussion:
The (new value) is an expression that must evaluate to a valid surface number.
GDATE
Purpose:
GDATE will place the current date under the lens name in the text box on user defined graphics screens.
Syntax:
GDATE
Discussion:
GDATE is primarily used for making your graphics look like other ZEMAX graphics.
Example:
See the section GRAPHICS.
GETEXTRADATA
Purpose:
Retrieves the extra data values from the extra data editor. The data is placed in one of the vector array
variables (either VEC1, VEC2, VEC3, or VEC4).
Syntax:
GETEXTRADATA vector_expression, surface_expression
Discussion:
The data is stored in the specified VECn array variable. For example, if the command GETEXTRADATA 1, 5
is issued, the extra data for surface 5 will be placed in VEC1. The data is stored in the following format, where
the first number in each line refers to the array position:
0: The number of extra data values in the vector
1: The first extra data value
n: The nth extra data value
See the chapter "Surface Types" for descriptions of the extra data values.
GETGLASSDATA
Purpose:
Retrieves the data for any glass in the current catalogs. The data is placed in one of the vector array variables
(either VEC1, VEC2, VEC3, or VEC4).
Syntax:

GETGLASSDATA vector_expression, glass_number
Discussion:
The data is stored in the specified VECn array variable. For example, if the command GETGLASSDATA 1, 32
is issued, the data for glass number 32 will be placed in VEC1. The glass number is returned by the GNUM
function. The data is stored in the following format, where the first number in each line refers to the array position:
0: The number of data values in the vector
1: Formula number: 1 for Schott, 2 for Sellmeier 1,
3 for Herzberger, 4 for Sellmeier 2, 5 for Conrady.
2: MILNUM
3: Nd
4: Vd
5: Thermal coefficient of expansion -30 to +70 c
6: Thermal coefficient of expansion +20 to 300 c
7: Density in g/cm^3
8: Deviation from normal line P gf
9: Lambda min
10: Lambda max
11-16:Constants of dispersion (meaning depends upon formula)
17-22:Thermal constants of dispersion
23+: Internal transmission coefficient (per mm) alpha, T = exp(alpha * path).
The alpha for wavelength 1 is stored in 23, wavelength 2 is in 24, etc.
Related Functions:
GNUM
GETLSF
Purpose:
Calculates the geometric edge and line response functions, similar to the Geometric Line/Edge Spread on
page 137.
Syntax:
GETLSF wave, field, sampling, vector, maxradius, use_polarization
Discussion:
Wave is an integer corresponding to the wavelength number to use for the calculation. A value of zero indicates
a polychromatic calculation. Field must be an integer between 1 and the maximum number of fields. The value
indicates which field position to use. Sampling may be 1 (32 x 32), 2 (64 x 64), 3 (128 x 128), etc... up to 2048 x
2048. The vector argument must be an integer value between 1 and 4, and specifies which vector array the data
should be placed in. The maxradius argument is the maximum radial coordinate of the edge and line spread
functions; this is the half-width of the data range. Use 0 for a default width. If any of the arguments fall outside the
valid ranges, then the nearest acceptable value is used instead.
The data is returned as an array of values in the specified vector. Vector position 0-3 will hold the number of
points "N", the starting x coordinate (this is the negative of the half width of the data range), the delta coordinate,
and the offset (defined below), respectively. The offset is the first position in the vector that holds the edge or line
spread data. Starting at the offset, the first N value are the tangential LSF response. The next N values are the
sagittal LSF response. The tangential and sagittal ERF values are in the next two groups of N data values.
If the current vector size is not large enough, ZEMAX will automatically increase the size of the vectors to hold
the LSF data in the manner described in SETVECSIZE.
Example:
! Macro computes and prints the LSF and ERF for polychromatic light at field 1.
!
! Syntax is GETLSF wave, field, samp, vector, maxradius, usepol
!
GETLSF 0, 1, 3, 1, 0, 0
N_BINS = vec1(0)
STARTX = vec1(1)
DELTAX = vec1(2)
OFFSET = vec1(3)

FORMAT 15.0
PRINT "Number of Bins = ", N_BINS
FORMAT 15.3 EXP
PRINT "Starting Coordinate = ", STARTX
PRINT "Delta Coordinate = ", DELTAX
FORMAT 15.0
PRINT "Offset = ", OFFSET
OFF1 = OFFSET
OFF2 = OFF1 + N_BINS
MAXI = N_BINS-1
FORMAT 16.3 EXP
PRINT
PRINT " X TLSF SLSF TERF SERF"
PRINT
FOR i = 0, MAXI, 1
PRINT STARTX + DELTAX*i,
PRINT vec1(OFF1 + i),
PRINT vec1(OFF4 + i)
NEXT i
GETMTF
Purpose:
Calculates tangential and sagittal MTF, real part, imaginary part, phase, or square wave response data for the
currently loaded lens file, and places the data in one of the vector arrays (either VEC1, VEC2, VEC3, or VEC4).
Syntax:
GETMTF freq, wave, field, sampling, vector, type
Discussion:
The freq argument is the desired spatial frequency in line pairs per millimeter. If the frequency is less than
zero, or greater than the cutoff frequency, GETMTF returns zero. Wave is an integer corresponding to the
wavelength number to use for the calculation. A value of zero indicates a polychromatic calculation. Field must
be an integer between 1 and the maximum number of fields. The value indicates which field position to use.
Sampling may be 1 (32 x 32), 2 (64 x 64), 3 (128 x 128), etc... up to 2048 x 2048. The vector argument must be
an integer value between 1 and 4, and specifies which vector array the data should be placed in. The type
argument refers to the data type: 1 for MTF, 2 for real part, 3 for imaginary part, 4 for phase in radians, 5 for square
wave MTF. If any of the arguments fall outside the valid ranges, then the nearest acceptable value is used instead.
This calculation uses the FFT MTF method (see FFT MTF on page 114).
The data is returned in one of the vector arrays with the following format: Vector position 0: tangential
response; Vector position 1: sagittal response.
Example:
! This macro computes the T & S response at 30 lp/mm
! for the currently loaded lens, polychromatic,
! at the maximum defined field,
! and a 32x32 grid density (sampling = 1).
! Data will be placed in vector 1.
! This is all it takes to get the data:
GETMTF 30, 0, NFLD(), 1, 1, 1
PRINT "Tangential response:", vec1(0)
PRINT "Sagittal response :", vec1(1)
GETPSF
Purpose:
Calculates the diffraction point spread function (PSF) data for the currently loaded lens file, and places the
data in one of the vector arrays (either VEC1, VEC2, VEC3, or VEC4).

Syntax:
GETPSF wave, field, sampling, vector, unnormalized, phaseflag, imagedelta
Discussion:
Wave is an integer corresponding to the wavelength number to use for the calculation. A value of zero indicates
a polychromatic calculation. Field must be an integer between 1 and the maximum number of fields. The value
indicates which field position to use. Sampling may be 1 (32 x 32), 2 (64 x 64), 3 (128 x 128), etc... up to 2048 x
2048. The vector argument must be an integer value between 1 and 4, and specifies which vector array the data
should be placed in. The unnormalized flag is zero if the data should be normalized to a peak of 1.0, if the
unnormalized value is 1, then the data is returned unnormalized. If phase flag is zero, the data returned is intensity,
if 1, then the phase in degrees is returned. The imagedelta value is the spacing between PSF points in
micrometers; use zero for the default spacing. The wavelength must be monochromatic to compute phase data.
If any of the arguments fall outside the valid ranges, then the nearest acceptable value is used instead.
The data is returned in one of the vector arrays with the following format:
Vector position 0: the total number of PSF data points in the vector array. Usually, this number will be 4*n*n
where n is the sampling size (32, 64, etc.). For example, if the sampling density is 2, the pupil sampling will be
64 x 64, and there will be 128 x 128 or 16,384 values in the array. This will require 8 bytes per number, or a total
of 131 kb. A sampling density of 1024 will require at least 8 Mb just for the array; another 64 Mb or more to
compute the PSF. Position 0 also returns other values as error codes. If position 0 is zero, then the computation
was aborted. If -1, then the vector array is not large enough to hold all the data. Use SETVECSIZE to make the
array bigger. If -2, then there is not enough system RAM to compute the PSF data.
Vector position 1 through 4*n*n holds the PSF data intensity, normalized to unity. The first 2n values are the
first row, going left to right from -x to +x, then each subsequent block of 2n values is another row, going from +y
to -y. Vector position 4*n*n+1 holds the spacing between data values in micrometers.
Example:
! This macro computes the PSF
! for the currently loaded lens, polychromatic,
! at the first field,
! and a 32x32 grid density (sampling = 1),
! data will be placed in vector 1,
! normalized to 1,
! no phase data,
! default image delta.
SETVECSIZE 4500
GETPSF 0, 1, 1, 1, 0, 0, 0
np = vec1(0)
IF (np == 0)
PRINT "PSF Computation aborted."
GOTO 1
ENDIF
IF (np == -1)
PRINT "SETVECSIZE too small for PSF data."
GOTO 1
ENDIF
IF (np == -2)
PRINT "Not enough system RAM for PSF data."
GOTO 1
ENDIF
PRINT "There are ", np, " data points, spaced ", vec1(np+1), " micrometers apart".
LABEL 1
GETSYSTEMDATA
Purpose:
Retrieves most system specific data, such as effective focal length, working F/#, apodization factors, and other
data not associated with any particular surface. The data is placed in one of the vector array variables (either
VEC1, VEC2, VEC3, or VEC4).

Syntax:
GETSYSTEMDATA vector_expression
Discussion:
The data is stored in the specified VECn array variable. For example, if the command GETSYSTEMDATA 1
is issued, the system data will be placed in VEC1. The data is stored in the following format, where the first number
in each line refers to the array position:
0: The number of system data values in the vector
1: Aperture Value
2: Apodization Factor
3: Apodization Type (0:none, 1:gaussian, 2:tangent)
4: Use Env Data (1 if true, 0 if false)
5: Temperature in degrees c (valid only if Use Env Data true)
6: Pressure in ATM (valid only if Use Env Data true)
7: Effective Focal Length
8: Image Space F/#
9: Object Space Numerical Aperture
10: Working F/#
11: Entrance Pupil Diameter
12: Entrance Pupil Position
13: Exit Pupil Diameter
14: Exit Pupil Position
15: Paraxial Image Height
16: Paraxial Magnification
17: Angular Magnification
18: Total Track
19: Use Ray Aiming (1 if true, 0 if false)
20: X Pupil Shift
21: Y Pupil Shift
22: Z Pupil Shift
23: Stop Surface Number
24: Global Coordinate Reference Surface Number
25: Telecentric object space flag (0 for false, 1 for true)
26: The number of configurations
27: The number of multi-configuration operands
28: The number of merit function operands
29: The number of tolerance operands
GETTEXTFILE
Purpose:
Creates an ASCII text file from any ZEMAX analysis window that supports text.
Syntax:
GETTEXTFILE "textfilename", "type", "settingsfilename", flag
Discussion:
The textfilename must be in quotes, or be a string variable name, and include the full path, name, and
extension for the file to be created. The string function $TEMPFILENAME can be used to define a suitable
temporary file name.
The type argument is a 3 character case-sensitive label that indicates the type of analysis to be performed.
The 3 letter labels are identical to those used for the button bar in ZEMAX. A list of codes may be found on the
"Buttons" tab of the File, Preferences dialog box. The labels are case sensitive. If no label is provided or
recognized, a standard ray trace will be generated.
If a valid file name in quotes, or a string variable name, is used for the "settingsfilename" argument, ZEMAX
will use or save the settings used to compute the text file, depending upon the value of the flag parameter.
If the flag value is 0, then the default settings will be used. If the lens file has its own default settings, then
those will be used; these are the settings stored in the "lensfilename.cfg" file. If no lens specific default settings
exist, then the default settings for all ZEMAX files, stored in the file "ZEMAX.CFG" will be used, if any. If no
previous settings have been saved for this or any other lens, then the default settings used are the "factory"
defaults used by ZEMAX.

If the flag value is 1, then the settings provided in the settings file, if valid, will be used to generate the file. If
the data in the settings file is in anyway invalid, then the default settings will be used to generate the file. The only
valid settings files are those generated by ZEMAX, then renamed and saved to a new user defined file name. For
example, on the Spot Diagram settings dialog, pressing "Save" will generate a "SPT.CFG" file with the saved
settings. If this file is renamed to "MySPT.CFG", then the file, with a full path, may be used as the "settingsfilena-
me" argument to GetTextFile.
If the flag value is 2, then the settings provided in the settings file, if valid, will be used and the settings box for
the requested feature will be displayed. After the user makes any changes to the settings the file will then be
generated using the new settings. The dialog boxes used to change the analysis settings use the data from the
lens currently in the Lens Data Editor.
No matter what the flag value is, if a valid file name is provided for the settingsfilename, the settings used will
be written to the settings file, overwriting any data in the file.
Only text, and not graphic files, are supported by GETTEXTFILE.
See also OPEN, CLOSE, READ, READNEXT, and READSTRING.
Example:
! Macro sample to generate cardinal points in ASCII file.
! Get a temporary file name
A$ = $TEMPFILENAME()
! Compute the data and place in the temp file
GETTEXTFILE A$, Car
! Open the new file and print it out
OPEN A$
ABEL 1
READSTRING B$
IF (EOFF()) THEN GOTO 2
PRINT B$
GOTO 1
LABEL 2
CLOSE
GETVARDATA
Purpose:
Retrieves the current number, type, and value of all optimization variables. The data is placed in one of the
vector array variables (either VEC1, VEC2, VEC3, or VEC4).
Syntax:
GETVARDATA vector
Discussion:
The data is stored in the specified VECn array variable. For example, if the command GETVARDATA 1 is
issued, the data will be stored in VEC1. The data is stored in the following format, where the first number in each
line refers to the array position:
0: n, the number of variables
1: The type code for the first variable
2: Surface number for the first variable
3: Parameter number for the first variable
4: Object number for the first variable
5: The value of the first variable
5*q-4: The type code for the qth variable
5*q-3: Surface number for the qth variable
5*q-2: Parameter number for the qth variable
5*q-1: Object number for the qth variable
5*q: The value of the qth variable
etc...
The integer q goes from 1 to n, where n is the number of variables. If n is zero, then no valid data is returned.
The value of the number in array position zero, n, is always valid. The type codes for variables is as described in
the following table. The surface number, parameter number, and object number may or may not have meaning
depending upon the type of variable.

GETVARDATA TYPE AND ID CODES
Variable type Type Code Surface Parameter Object
Curvature 1 surface # - -
Thickness 2 surface # - -
Conic 3 surface # - -
Index Nd 4 surface # - -
Abbe Vd 5 surface # - -
Partial Dispersion P g, F 6 surface # - -
TCE 7 surface # - -
Parameter Values 8 surface # parameter # -
Extra Data Values 9 surface # extra data # -

Multi-configuration Operand Values 10 oper # config # -
Non-Sequential Object Position X 11 surface # - object #
Non-Sequential Object Position Y 12 surface # - object #

Non-Sequential Object Position Z 13 surface # - object #
Non-Sequential Object Tilt X 14 surface # - object #
Non-Sequential Object Tilt Y 15 surface # - object #
Non-Sequential Object Tilt Z 16 surface # - object #
Non-Sequential Object Parameters 17 surface # parameter # object #
GETZERNIKE
Purpose:
Calculates Zernike Fringe, Standard, or Annular coefficients for the currently loaded lens file, and places them
in one of the vector arrays (either VEC1, VEC2, VEC3, or VEC4).
Syntax:
GETZERNIKE maxorder, wave, field, sampling, vector, zerntype, epsilon, reference
Discussion:
The maxorder argument is any number between 1 and 37 for Fringe or between 1 and 231 for Standard or
Annular coefficients (see the discussion of zerntype below), and corresponds to the highest Zernike term desired.
Wave and field are the integer values for the wavelength and field number respectively. The value for sampling
determines the size of the grid used to fit the coefficients. Sampling may be 1 (32 x 32), 2 (64 x 64), etc.... up to
2048 x 2048. The vector argument must be an integer value between 1 and 4, and specifies which vector array
the data should be placed in. The zerntype is 0 for "fringe" Zernike terms, 1 for "Standard" Zernike terms, and 2
for "Annular" Zernike terms. See Zernike Fringe Coefficients on page 165, Zernike Standard Coefficients on
page 168, and Zernike Annular Coefficients on page 171 for descriptions of these different types of Zernike
terms. For Annular Zernike Coefficients epsilon is the annular ratio; this value is ignored for other Zernike types.
To reference the OPD to the chief ray, the reference value should be zero or omitted; use 1 to reference to the
surface vertex. If any of the arguments fall outside the valid ranges, then the nearest acceptable value is used
instead.
The data is returned in one of the vector arrays with the following format: Vector position 1: Peak to valley in
waves; Vector position 2: RMS to the zero OPD line in waves (this value is not physically meaningful but is
provided for reference); Vector position 3: RMS to the chief ray in waves; Vector position 4: RMS to the image

centroid in waves (this is the most physically meaningful number related to image quality); Vector position 5:
Variance in waves; Vector position 6: Strehl ratio; Vector position 7: RMS fit error in waves; Vector position 8:
Maximum fit error (at any one point) in waves. The remaining vector positions contain the actual Zernike
coefficient data. For example, Zernike term number 1 is in vector position 9, Zernike term 2 is in position 10, and
so on.
Example:
! This macro computes the first 37 Zernike Fringe coefficients
! for the currently loaded lens, at wave 1, field 1
! and a 32x32 grid density (sampling = 1). The coefficients
! will be placed in vector 1. First get the data:
GETZERNIKE 37,1,1,1,1,0
! Now print it out:
FORMAT 16.6
PRINT "Peak to Valley : ", vec1(1)
PRINT "RMS to chief : ", vec1(3)
PRINT "RMS to centroid : ", vec1(4)
PRINT "Variance : ", vec1(5)
PRINT "Strehl ratio : ", vec1(6)
PRINT "RMS Fit Error : ", vec1(7)
PRINT "Maximum Fit Error : ", vec1(8)
i = 1
label 1
FORMAT 2.0
PRINT "Zernike #", i, " = ",
FORMAT 16.6
PRINT vec1(8+i)
i = i + 1
if (i < 38) THEN GOTO 1
PRINT "All Done!"
GLAS
Purpose:
GLAS is used to set the glass type of a surface.
Syntax:
GLAS (surface) = (glass_number)
GLAS (surface) = glass$
Discussion:
This keyword requires two expressions, one to specify the surface number and the other to define the number
of the glass. If the glass number is -1, the glass type is set to "mirror". If the glass number is 0, then the glass type
is blank, or unity index. Otherwise, the glass number must be between 1 and the number returned by MAXG.
GLAS also accepts a string variable which contains the name of the desired glass, such as "BK7".
Example:
GLAS 3 = MAXG()
A$ = "BK7"
GLAS 4 = GNUM(A$)
GLAS 4 = A$
Related Keywords:
UPDATE
GLENSNAME
Purpose:
GLENSNAME will place the current lens name at the top left corner of the text box on user defined graphics
screens.
Syntax:
GLENSNAME
Discussion:

GLENSNAME is primarily used for making your graphics look like other ZEMAX graphics.
Example:
GOSUB, SUB, RETURN, and END
Purpose:
These four keywords are used together to define and call subprograms within the ZPL macro file. Each
keyword has a special purpose. GOSUB is used to direct the program flow to a defined subroutine. SUB is used
to define the subroutine name, as well as indicate the beginning of the subroutine body. RETURN indicates the
macro execution should continue at the point where the most recent GOSUB call was placed. END indicates that
the macro should terminate immediately.
Syntax:
See the Example section for sample syntax.
Discussion:
There can be no more than 100 defined subroutines per ZPL macro file. Each subroutine must be terminated
by a RETURN statement. More than one return statement may be placed in the body of a subroutine. If
subroutines are defined, at least one END statement must be used to indicate the end of the main macro body.
The main macro body must be at the top of the file.
There can be no more than 100 "nesting levels" used in the ZPL macro. For example, if subroutine ABC calls
subroutine XYZ, then the nesting level is 2. If subroutine XYZ then calls subroutine DEF, the nesting level is 3.
All variables in ZPL are global. Any variables used or defined within a subroutine exist within the main macro
as well.
Example:
x = 1
y = 2
GOSUB add
print "the sum of ", x, " and ", y, " is ", z
END
SUB add
z = x + y
RETURN
GOTO
Purpose:
Normally, each macro line is executed in turn. GOTO allows execution to resume at an arbitrary point in the
macro. GOTO is always used in conjunction with the LABEL command.
Syntax:
GOTO label_number
GOTO text_label
Discussion:
There must be a LABEL command with the corresponding label_number or text_label somewhere in the
macro, or an error will result.
Example:
LABEL 1
x=RAND(10)
if x <= 5 THEN GOTO 1
PRINT " X is greater than 5 "
GRAPHICS
Purpose:
Creates a standard ZEMAX graphics frame with ruling lines for the plot title.

Syntax:
GRAPHICS
GRAPHICS NOFRAME
GRAPHICS OFF
Discussion:
If GRAPHICS is specified alone, then a standard ZEMAX graphics window will be created. If the optional
argument NOFRAME is supplied, then the standard frame for the graph title will be suppressed. All subsequent
graphics commands will be sent to this newly created window. GRAPHICS OFF will close any existing open
graphics windows, and then display the closed window.
Example:
graphics
xmx = xmax()
xmn = xmin()
ymx = ymax()
ymn = ymin()
xwidth = xmx-xmn
ywidth = ymx-ymn
xleft = xmn + ( .1 * xwidth )
xrigh = xmn + ( .9 * xwidth )
ytopp = ymn + ( .1 * ywidth )
ybott = ymn + ( .7 * ywidth )
line xleft,ytopp,xrigh,ytopp
line xrigh,ytopp,xrigh,ybott
line xrigh,ybott,xleft,ybott
line xleft,ybott,xleft,ytopp
gtitle " the rain in spain falls mainly on the plain"
glensname
gdate
gtext xmx/2,ymx/2,0, " start this text in the center."
gtextcent ymx*.05, " center this text near the top."
gtext xmx*.05,ymx*.75,90, " place me vertically near left edge."
gtext xmx*.15,ymx*.68,30, " orient me at 30 degrees."
graphics off
Related Keywords:
LINE, GITITLE, GLENSNAME, GDATE, GTEXTCENT, GTEXT, PIXEL
GTEXT
Purpose:
GTEXT is used for labeling graphics plots with user defined text.
Syntax:
GTEXT x, y, angle, user_text
GTEXT x, y, angle, A$
Discussion:
The coordinates x and y refer to the left edge of where the text string user_text will appear. "user_text" may
be either a constant string in quotes or a string variable name. Angle specifies how the text is rotated with respect
to the graphics frame, and defaults to 0 degrees (horizontal). See also SETTEXTSIZE.
Example:
GTEXTCENT
Purpose:
GTEXT is used for centering labels on graphics plots with user defined text.
Syntax:
GTEXTCENT y, user_text
Discussion:

The coordinate y refers to the vertical position of the text string user_text. See also SETTEXTSIZE.
Example:
GTITLE
Purpose:
GTITLE is similar to GTEXT except only the text needs to be specified, and the text will appear centered in
the title bar on the graphics display. GTITLE is useful for making your ZPL graphics functions look like standard
ZEMAX graphics.
Syntax:
GTITLE user_title
Discussion:
GTITLE is primarily used for making your graphics look like other ZEMAX graphics.
Example:
HAMMER
Purpose:
Invokes the Hammer optimization algorithm to optimize the current lens with the current merit function.
Syntax:
HAMMER
HAMMER number_of_cycles
Discussion:
If no argument is provided, then the hammer optimization runs 1 cycle. If an argument is provided, it must
evaluate to an integer between 1 and 99, and the Hammer optimization algorithm will run the specified number
of cycles. Currently, the ZPL implementation of this function does not take advantage of multiple processors.
Related Functions:
MFCN
Example:
PRINT "Starting merit function:", MFCN()
HAMMER 3
PRINT "Ending merit function :", MFCN()
IF-THEN-ELSE-ENDIF
Purpose:
IF provides conditional macro execution and branching.
Syntax:
IF (expression)
(statements)
ELSE
(statements)
ENDIF
or
IF (expression) THEN (statement)
Discussion:
The IF-ELSE-ENDIF construction is used for conditional execution of either the group of statements following
the IF statement or the statements following the ELSE statement, but not both. The value of expression is
considered false if it is zero, otherwise it is considered true. The expression can be any valid ZPL expression,
composed of functions, variables, operands, and constants. The IF statement must be paired with an ENDIF,
although the ELSE is optional. IF-ENDIF pairs may be nested to any level.

The IF-THEN construction is handy for conditional execution of a single instruction. If the THEN keyword is
specified, the IF statement is terminated, and there is no need for an ENDIF. The ELSE keyword is not supported
in the IF-THEN construction.
Example:
x = 1
y = 2
if (x < y)
PRINT "x is less than y"
ELSE
if (x==y) THEN PRINT "x equals y"
if (x > y) THEN PRINT "x is greater than y"
ENDIF
IMA
Purpose:
Computes the Geometric Image Analysis feature and saves the result to a BIM file. For a description of the
BIM (Binary Image) format see The BIM format on page 144. For a description of the Geometric Image Analysis
feature see Geometric Image Analysis on page 142.
Syntax:
IMA outfilename infilename
Discussion:
This keyword requires the name of the output BIM file, and optionally, the name of the input IMA or BIM file.
If the extension to the outfilename is not provided, the extension BIM will be appended. The extension must be
provided on the infilename. The filenames must be enclosed in quotes if any blank or other special characters
are used. The outfilename will be placed in the \ImaFiles subdirectory. The infilename must also be placed in the
\ImaFiles subdirectory. No paths should be provided with the file names.
The settings for the Geometric Image Analysis feature will be those settings previously saved for the current
lens. To make adjustments to the settings, open a Geometric Image Analysis window, choose the appropriate
settings, then press "Save". Be certain to choose "Show As" as anything other than "spot diagram". All
subsequent calls to IMA will use the saved settings. The exceptions are the output file name, which is specified
as the first argument after the IMA keyword, and the input source file, which is optionally specified as the second
argument after the IMA keyword.
Example:
IMA "output.BIM" "LETTERF.IMA"
Related Keywords:
IMASHOW, IMASUM
IMASHOW
Purpose:
Displays an IMA or BIM file in a viewer window. For a description of the IMA and BIM file formats, see The
IMA format on page 144 and The BIM format on page 144.
Syntax:
IMASHOW filename.ima
Discussion:
This keyword requires the name of the IMA or BIM file. The extension must be included. The filename may be
enclosed in quotes if any blank or other special characters are used. The file must be located in the \ImaFiles
subdirectory. This command will open a new window to display the file.
Example:
IMASHOW "LETTERF.IMA"
Related Keywords:
IMA, IMASUM

IMASUM
Purpose:
Sums the intensity in a BIM file. Despite the name of this function, currently only BIM files, and not IMA files,
may be summed. This limitation is due to the low number of grey scales that are supported in the IMA file format.
For a description of the IMA and BIM file formats, see The IMA format on page 144 and The BIM format on
page 144.
Syntax:
IMASUM filename1 filename2 outfilename
Discussion:
This keyword requires the names of three BIM files (which need not be distinct). The BIM extension must be
provided on all three file names. The filenames may be enclosed in quotes if any blank or other special characters
are used. The files must be located in the \ImaFiles subdirectory, which is where the output file will be. The two
source files must have the same number of pixels and color channels, or an error message will result.
Example:
IMASUM "A.BIM" "B.BIM" "sum of A and B.BIM"
Related Keywords:
IMA, IMASHOW
IMPORT_ED
Purpose:
Imports data into the extra data editor from a file.
Syntax:
IMPORT_ED surface filename
Discussion:
The surface may be any valid surface number that uses extra data values. The data to be imported should be
a single column of numeric data in ASCII format. See "Tools" under Extra Data on page 79, and Importing grid
data on page 259.
INPUT
Purpose:
INPUT provides a means for prompting the user for numeric or text data when the macro is run.
Syntax:
INPUT "Prompt String" , variable
INPUT variable
INPUT "Prompt String" , string_variable$
INPUT string_variable$
Discussion:
The variable may be any valid variable name. If the variable name is a string variable, then the input will be
interpreted as a literal string; otherwise, as a numeric. The INPUT command will use a "? " prompt if no prompt
string is supplied. The prompt is always displayed on the screen, and the input always is accepted from the
keyboard only.
Example:
INPUT "Enter value for x:", x
PRINT "X = ", x
INPUT "Enter a value for A$:", A$
PRINT A$
INSERT
Purpose:
INSERT inserts a new surface in the lens data editor.

Syntax:
INSERT (surf)
Discussion:
The value surf must evaluate to an integer expression. See also DELETE and SURFTYPE.
Example:
INSERT 5
INSERT i+2*j
INSERTCONFIG
Purpose:
INSERTCONFIG inserts a new configuration in the multi-configuration editor.
Syntax:
INSERTCONFIG (config)
Discussion:
The value config must evaluate to an integer expression greater than 0 and less than or equal to the current
number of configurations plus 1. See also DELETECONFIG and INSERTMCO.
Example:
INSERTCONFIG 4
INSERTMCO
Purpose:
INSERTMCO inserts a new multi-configuration operand in the multi-configuration editor.
Syntax:
INSERTMCO (row)
Discussion:
number of operands plus 1. See also DELETEMCO and INSERTCONFIG.
Example:
INSERTMCO 7
INSERTMFO
Purpose:
INSERTMFO inserts a new merit function operand in the merit function editor.
Syntax:
INSERTMFO (row)
Discussion:
number of operands plus 1. See also DELETEMFO and SETOPERAND.
Example:
INSERTMFO 23
INSERTOBJECT
Purpose:
Inserts a new NSC object.
Syntax:
INSERTOBJECT (surf), (object)
Discussion:

The value surf must evaluate to an integer expression, and the specified surface must be a non-sequential
component surface. Use a surf value of 1 if the program mode is non-sequential. The value object must evaluate
to an integer that is between 1 and the current number of objects plus 1, inclusive. The new object will be inserted
at the specified object number, renumbering other objects as required. See also DELETEOBJECT and SET-
NSCPROPERTY.
Example:
INSERTOBJECT 1, 1
LABEL
Purpose:
LABEL provides a destination for the GOTO command, see the section "GOTO" for details.
Syntax:
LABEL label_number
LABEL text_label
Discussion:
The label_number must be an integer value greater than zero, such as 1 or 7. If a text label is used, the text
label must not contain spaces or other special characters that are used as delimiters. LABEL has no effect on
program flow by itself. There is a limit of 300 label statements in a macro.
Example:
LABEL 7
LABEL startover
LINE
Purpose:
LINE is the primitive line drawing function for graphical displays.
Syntax:
LINE oldx, oldy, newx, newy
Discussion:
LINE will evaluate the four expressions and draw a line connecting the points defined. The coordinates refer
to the current graphics frame, and must be contained within the bounds defined by XMIN, YMIN, XMAX, and
YMAX. Although only integer pixel values can actually be plotted, LINE will accept real values as arguments and
round the coordinates to the nearest integer equivalents. LINE is only valid in graphics mode. The color of the
line drawn is controlled by the current pen color, which is specified using the COLOR keyword.
Example:
LOADCATALOG
Purpose:
Reloads glass and coating catalogs for the currently loaded lens.
Syntax:
LOADCATALOG
Discussion:
When lenses are loaded, any associated glass catalogs and data files, including the COATING.DAT file, are
automatically loaded if they are not already loaded. However, if these catalogs have been modified, perhaps by
the ZPL macro itself; then the LOADCATALOG keyword may be used to force a reload of the catalogs. Use of
this keyword is not required unless the COATING.DAT or glass AGF catalog files have been modified since the
start of the current ZEMAX session.
LOADLENS
Purpose:

Loads a new lens file from disk, replacing any lens file currently in memory.
Syntax:
LOADLENS "filename" [appendflag, session]
LOADLENS file$ [appendflag, session]
Discussion:
LOADLENS will load a new lens file from disk. If the filename contains the complete path, such as
C:\MYDIR\MYLENS.ZMX, then the specified file will be loaded. If the path is left off, then the default directory for
lenses will be used (see the File menu chapter under Environment).
If the appendflag is zero or absent, then LOADLENS loads the file. If the appendflag is greater than zero, then
the file is appended to the current lens starting at the surface specified by the value of the appendflag. If the
session flag is non-zero, any associated session file will be loaded with the lens and all windows will be updated
otherwise the lens session file is ignored.
Example:
LOADLENS "COOKE.ZMX"
Related Keywords:
SAVELENS
LOADMERIT
Purpose:
Loads a merit function file from disk, replacing the merit function in the current lens.
Syntax:
LOADMERIT "filename"
LOADMERIT file$
Discussion:
LOADMERIT will load a new merit function from disk. If the filename contains the complete path, such as
C:\MYDIR\MYLENS.MF, then the specified file will be loaded. If the path is left off, then the default directory for
lenses will be used (see Directories on page 59). See also SAVEMERIT.
LOCKWINDOW
Purpose:
Locks any open window.
Syntax:
LOCKWINDOW winnum
Discussion:
See Graphic windows operations on page 36.
Example:
LOCKWINDOW 7
Related Keywords:
UNLOCKWINDOW
NEXT
See FOR.
NSTR
Purpose:
Initiates a Non-sequential trace.
Syntax:
NSTR surf, source, split, scatter, usepolar, ignore_errors, no_random_seed, save,
savefilename, filter

Discussion:
Surf is an integer value that indicates the number of the Non-Sequential surface. If the program mode is set
to Non-Sequential, use 1. Source refers to the object number of the desired source. If source is zero, all sources
will be traced. If Split is non-zero, then splitting is on, otherwise, ray splitting is off. If Scatter is non-zero, then
scattering is on, otherwise scattering is off. If Usepolar is non-zero then polarization will be used, otherwise
polarization is off. If splitting is on polarization is automatically selected. If ignore_errors is non-zero, then errors
will be ignored, otherwise ray errors will terminate the non-sequential trace and macro execution and an error will
be reported. If no_random_seed is zero, then every call to NSTR will use different random rays, otherwise, the
rays will be the same every time, which is required when using NSTR for optimization.
If save is omitted or is zero, the arguments savefilename and filter need not be supplied. If save is not zero,
the rays will be saved in a ZRD file. The ZRD file will have the name specified by the savefilename, and will be
placed in the same directory as the lens file. The extension of savefilename should be ZRD, and no path should
be specified. If save is not zero, then the optional filter name is either a string variable with the filter, or the literal
filter in double quotes. For information on filter strings see The filter string on page 371.
Related Functions:
NSDD
Example:
NSTR 1, 0, 0, 0, 0, 1, 0, 1, "saverays.ZRD", "h2"
NUMFIELD
Purpose:
Sets the total number of defined fields. The UPDATE command must be issued before the new data takes
effect.
Syntax:
NUMFIELD <expression>
NUMWAVE
Purpose:
Sets the total number of defined wavelengths. The UPDATE command must be issued before the new data
takes effect.
Syntax:
NUMWAVE <expression>
OPEN
Purpose:
Opens an existing ASCII numeric text file for reading by the READ command.
Syntax:
OPEN "filename"
OPEN A$
Discussion:
The filename provided must be a valid file within quotes, or a string variable name containing the file name.
See the keywords READ and CLOSE. Always CLOSE a file after all the data has been read.
Related Functions:
EOFF
Related Keywords:
READ, READNEXT, CLOSE
Example:
PRINT "Reading the double-column file TEST.DAT!"
OPEN "TEST.DAT"
READ x1, y1

READ x2, y2
READ x3, y3
CLOSE
OPTIMIZE
Purpose:
Invokes the optimization algorithm to optimize the current lens with the current merit function.
Syntax:
OPTIMIZE
OPTIMIZE number_of_cycles
Discussion:
If an argument is provided, it must evaluate to an integer value between 1 and 99, and the optimization
algorithm will run the specified number of cycles. If no argument is provided, or number_of_cycles evaluates to
zero, then the optimization runs in "Automatic" mode, stopping when the algorithm detects the process has
converged. To update the merit function without optimizing, use the MFCN function.
Related Functions:
MFCN
Example:
PRINT "Starting merit function:", MFCN()
OPTIMIZE
PRINT "Ending merit function :", MFCN()
OPTRETURN
Purpose:
Used to return numerical values back to the optimization algorithm through the use of the ZPLM optimization
operand.
Syntax:
OPTRETURN (datafield) = (result)
Discussion:
OPTRETURN takes two arguments, separated by an equal sign. The (datafield) expression must evaluate to
an integer between 0 and 50. The datafield refers to a position in an array where the value of the expression
(result) may be stored. The sole purpose of OPTRETURN is to be able to optimize values computed within a ZPL
macro.
The optimization operand ZPLM must be used in the merit function to call the ZPL macro and retrieve the value
returned by OPTRETURN. See the chapter "Optimization" for details.
Example:
x = sqrt(thic(3) + radi(5))
OPTRETURN j = x+5
OUTPUT
Purpose:
Specifies destination for text output. Output is either to the screen, or to a file.
Syntax:
OUTPUT SCREEN
OUTPUT "filename"
OUTPUT "filename" APPEND
OUTPUT A$
OUTPUT A$ APPEND
Discussion:
If OUTPUT SCREEN is specified alone, then all subsequently executed PRINT statements will be directed to
the screen. If a valid filename is provided, then subsequent PRINT statements will output to the filename
specified. To close the file created earlier, use OUTPUT SCREEN which will direct subsequent PRINT outputs

to the screen. SHOWFILE will close the file and send it to the text viewer program for screen display. PRINTFILE
will close the file and print it on the currently defined printer.
If the keyword APPEND follows the file name, then subsequent output will be appended to the file. Otherwise,
the contents of the file will be overwritten.
Example:
OUTPUT "x.txt"
PRINT "This will not appear on the screen, but in the file x.txt."
OUTPUT SCREEN
PRINT "This will appear on the screen."
OUTPUT "x.txt" APPEND
PRINT "This will appear after the first line in the file x.txt."
Related Keywords:
CLOSE, OPEN, SHOWFILE, PRINTFILE
PARM
Purpose:
PARM is used to set parameter values on a surface. The parameter values are defined in the chapter "Surface
Types".
Syntax:
PARM (number), (surface), (new_value)
Discussion:
This keyword requires three expressions: the parameter number, surface number, and the value.
Example:
PARM 2, 4, 250
Related Keywords:
UPDATE
PARn
PARn is obsolete, use PARM instead.
PARAXIAL
Purpose:
Used to control whether ray tracing is done with paraxial or real rays.
Syntax:
PARAXIAL ON
PARAXIAL OFF
Discussion:
The current paraxial mode can be established by a call to the function PMOD, which returns 0 if paraxial mode
is off, 1 if it is on. This feature is used for switching between real and paraxial ray traces. Certain calculations,
such as measuring distortion and computing first-order properties such as effective focal length, require the
tracing of paraxial rays.
Example:
mode = PMODE()
IF mode THEN PRINT "Paraxial mode is on!"
IF !mode THEN PRINT "Paraxial mode is not on!"
PARAXIAL ON
PRINT "Now paraxial mode is on!"
PRINT "Restoring original mode..."
if !mode THEN PARAXIAL OFF
PAUSE
Purpose:

Pauses macro execution while displaying a status message. The status message can be a string or numerical
value. The macro continues once the user presses the "OK" button on the status dialog.
Syntax:
PAUSE
PAUSE x
PAUSE "Ready to continue..."
PAUSE x+sqrt(5)
Discussion:
This feature may be used for debugging, presenting results, or pausing the execution of the macro.
PIXEL
Purpose:
Turns on a single pixel on the current graphics screen.
Syntax:
PIXEL xcoord, ycoord
Discussion:
This feature is useful for making spot diagrams.
POLDEFINE
Purpose:
Defines the input polarization state for subsequent POLTRACE calls.
Syntax:
POLDEFINE Ex, Ey, PhaX, PhaY
Discussion:
The POLDEFINE keyword is used to define the input polarization state for subsequent polarization ray tracing.
POLDEFINE requires the normalized Ex and Ey electric field magnitudes, as well as the X and Y phase angles
in degrees. The default values are 0, 1, 0, and 0, respectively. Once the polarization state is defined, it remains
the same until changed.
Example:
POLDEFINE 2.0 2.0 45.0 -66.0
Related Keywords:
POLTRACE
POLTRACE
Purpose:
Calls the ZEMAX polarization ray tracing routines to trace a particular ray through the current system.
Syntax:
POLTRACE (Hx), (Hy), (Px), (Py), (wavelength), (vec), (surf)
Discussion:
The expressions Hx and Hy must evaluate to values between -1 and 1, and represent the normalized object
coordinates. The pupil coordinates are specified by the expressions Px and Py, which also must be between -1
and 1. For more information about normalized coordinates, see the chapter "Conventions and Definitions" under
"Normalized field and pupil coordinates". The wavelength expression must evaluate to an integer between 1 and
the maximum number of defined wavelengths. The vec expression must evaluate to a number between 1 and 4,
inclusive. The surf expression must evaluate to an integer between 1 and the number of surfaces, inclusive.
The input polarization state of the ray is defined by the POLDEFINE keyword.
Once the ray is traced, the polarization data for the ray is placed in the vector variable specified by the vec
expression. For example, if the command "POLTRACE Hx, Hy, Px, Py, w, 2, n" is issued, the data will be stored
in VEC2. The data is stored in the following format, where the first number in each line refers to the array position:

0: n, the number of data entries in the vector
1: The ray intensity after the surface
2: E-Field X component, real
3: E-Field Y component, real
4: E-Field Z component, real
5: E-Field X component, imaginary
6: E-Field Y component, imaginary
7: E-Field Z component, imaginary
8: S-Polarization amplitude reflection, real
9: S-Polarization amplitude reflection, imaginary
10: S-Polarization amplitude transmission, real
11: S-Polarization amplitude transmission, imaginary
12: P-Polarization amplitude reflection, real
13: P-Polarization amplitude reflection, imaginary
14: P-Polarization amplitude transmission, real
15: P-Polarization amplitude transmission, imaginary
16: E-Field X direction phase Px
17: E-Field Y direction phase Py
18: E-Field Z direction phase Pz
19: Major axis length of polarization ellipse
20: Minor axis length of polarization ellipse
21: Angle of polarization ellipse in radians
If the value in array position 0 is 0, then an error occurred and the polarization data is invalid. This may occur
if the specified ray cannot be traced. See the RAYTRACE command to extract extended error information.
Example:
POLDEFINE 0, 1, 0, 0
POLTRACE 0, 1, 0, 0, pwav(), 1, nsur()
PRINT "Transmission of chief ray at primary wavelength is ", vec1(1)
Related Keywords:
POLDEFINE, RAYTRACE
POP
Purpose:
Computes the Physical Optics Propagation (POP) of a beam through the optical system and saves the surface
by surface results to ZBF files. For a description of the POP feature see Physical Optics Propagation on
page 190. For information on the ZBF file format, see ZEMAX Beam File (ZBF) binary format on page 528.
Syntax:
POP outfilename, lastsurface
Discussion:
This keyword requires the name of the output ZBF file, and an expression that evaluates to the last surface to
propagate to. The filename must be enclosed in quotes if any blank or other special characters are used. The
created ZBF files will be placed in the \POP\Beamfiles subdirectory. No paths should be provided with the file
names.
The settings for the POP feature will be those settings previously saved for the current lens. To make
adjustments to the settings, open a POP window, choose the appropriate settings, then press "Save". All
subsequent calls to POP within ZPL will use the saved settings. The exceptions are the output file name, which
is specified as the first argument after the POP keyword, and the last surface number, which is optionally specified
as the second argument after the POP keyword.
Example:
POP "pop_output.ZBF" 12
PRINT
Purpose:
Print is used to output constant text and variable data to either the screen or a file, depending upon the current
status of the keyword OUTPUT.

Syntax:
PRINT
PRINT X
PRINT "The value of x is ", x
PRINT " x = ", x, " x + y = ", x + y
Discussion:
PRINT alone will print a blank line. PRINT with a list of text arguments and expressions will print each text
string (enclosed in double quotes) and the numeric value of each expression. PRINT uses the numerical output
format specified by FORMAT. If the last item in the list is followed by a comma, PRINT will not end the line with
a carriage return.
Example:
X=3
PRINT "X equals ",x
Related Keywords:
REWIND
PRINTFILE
Purpose:
Prints a text file.
Syntax:
PRINTFILE "filename"
PRINTFILE NAME$
Discussion:
The filename must be a valid filename within quotes or the name of a string variable which contains a valid file
name. The file must be an ASCII file (as would be created by OUTPUT and PRINT statements in ZPL) and must
be in the current directory. PRINTFILE also closes the file if no CLOSE statement has been executed.
Example:
OUTPUT "test.txt"
PRINT "Print this to the printer."
PRINTFILE "test.txt"
Related Keywords:
OPEN, OUTPUT, CLOSE, PRINT, PRINTFILE
PRINTWINDOW
Purpose:
Prints any open graphic or text window.
Syntax:
PRINTWINDOW winnum
Discussion:
corresponds to the window number that should be printed. ZEMAX numbers windows sequentially as they are
opened, starting with 1. Any closed windows are deleted from the window list, without renumbering the windows
which remain. Any windows opened after another window has been closed will use the lowest window number
available.
Example:
PRINTWINDOW 5
PWAV
Purpose:
Sets the primary wavelength.

Syntax:
PWAV (wavelength number)
Discussion:
The expression (wavelength number) is evaluated and the primary wavelength is set to the specified number.
The UPDATE command must be issued before the new data takes effect.
Example:
PWAV 1
Related Functions:
WAVL, WWGT, PWAV
QUICKFOCUS
Purpose:
Adjusts the thickness of the surface prior to the image surface to minimize the specified criteria.
Syntax:
QUICKFOCUS mode, centroid
Discussion:
The expression for mode should evaluate to 0, 1, 2, or 3 for RMS spot radius, spot x, spot y, or wavefront OPD.
The expression for centroid should evaluate to 0 or 1 to indicate the RMS should be referenced to the chief ray
or image centroid, respectively. The "best" focus is chosen as a wavelength weighted average over all fields.
Example:
! Focus at best RMS wavefront to centroid
QUICKFOCUS 3, 1
RADI
Purpose:
RADI is used to set the radius of curvature of a surface.
Syntax:
RADI (surface) = (new_value)
Discussion:
Example:
RADI i+m = -1200
Related Keywords:
UPDATE
RANDOMIZE
Purpose:
RANDOMIZE seeds the random number generator.
Syntax:
RANDOMIZE (seed)
Discussion:
If seed evaluates to zero or a negative number, ZEMAX seeds the random number generator with a value
based upon the CPU clock. Otherwise, the value provided is used to seed the random number generator. Using
the same seed will reproduce the identical series of random numbers created by the RAND function.
Example:

RANDOMIZE
RANDOMIZE 250
Related Functions:
RAND
RAYTRACE
Purpose:
Calls the ZEMAX ray tracing routines to trace a particular ray through the current system.
Syntax:
RAYTRACE (hx), (hy), (px), (py), (wavelength)
Discussion:
The expressions hx and hy must evaluate to values between -1 and 1, and represent the normalized object
coordinates. The pupil coordinates are specified by the expressions px and py, which also must be between -1
and 1. For more information about normalized coordinates, see the chapter "Conventions and Definitions" under
"Normalized field and pupil coordinates". The wavelength expression is optional, defaulting to the primary
wavelength, but if supplied must evaluate to an integer between 1 and the maximum number of defined
wavelengths.
Once the ray is traced, the ray intercept coordinates and direction cosines may be determined using the ZPL
functions RAYX, RAYY, RAYZ, RAYL, RAYM, and RAYN. If an error occurred during ray tracing, the function
RAYE (for RAY Error) will return a value other than zero. If RAYE is negative, it indicates that total internal
reflection occurred at the surface whose number is the absolute value of the value returned. If RAYE is greater
than zero, then the ray missed the surface number returned. Checking RAYE is optional, however, the RAYX,
RAYY, ... functions may return invalid data if RAYE is not zero. The functions RANX, RANY, and RANZ return
the intercept surface normal direction cosines, and OPDC returns the optical path difference for the ray. The
function RAYV returns the surface number at which the ray was vignetted, or it returns zero if the ray was not
vignetted. Values returned for surfaces past the surface of vignetting may not be accurate.
Example:
PRINT "Tracing the marginal ray at primary wavelength!"
n=NSUR()
RAYTRACE 0,0,0,1
y = RAYY(n)
PRINT "The ray intercept is ", y
PRINT "Tracing the chief ray at maximum wavelength!"
RAYTRACE 0,1,0,0,NWAV()
y = RAYY(n)
Related Keywords:
RAYTRACEX
RAYTRACEX
Purpose:
Calls the ZEMAX ray tracing routines to trace a particular ray from any starting surface through the current
system.
Syntax:
RAYTRACEX (x), (y), (z), (l), (m), (n), (surf), (wavelength)
Discussion:
The expressions x, y, z, l, m, and n define the input ray position and direction cosines in the local coordinates
of the starting surface. The surface expression must evaluate to an integer between 0 and the number of surfaces
minus one, inclusive. The wavelength expression is optional, defaulting to the primary wavelength, but if supplied
must evaluate to an integer between 1 and the maximum number of defined wavelengths.

!If the object has a thickness of infinity, and the surf parameter is zero, then the input
coordinates are assumed to be relative to the first surface rather than the object surface;
although the ray will still be defined in object space media. Otherwise, ZEMAX uses the
specified coordinates without alteration.
Once the ray is traced, the ray intercept coordinates and direction cosines may be determined using the ZPL
functions RAYX, RAYY, RAYZ, RAYL, RAYM, and RAYN. Note only data from surface AFTER the "surf" surface
will be valid.
If an error occurred during ray tracing, the function RAYE (for RAY Error) will return a value other than zero.
If RAYE is negative, it indicates that total internal reflection occurred at the surface whose number is the absolute
value of the value returned. If RAYE is greater than zero, then the ray missed the surface number returned.
Checking RAYE is optional, however, the RAYX, RAYY, ... functions may return invalid data if RAYE is not zero.
The functions RANX, RANY, and RANZ return the intercept surface normal direction cosines, and RAYT returns
the optical path length up to the surface for the ray. The function RAYV returns the surface number at which the
ray was vignetted, or it returns zero if the ray was not vignetted. Values returned for surfaces past the surface of
vignetting may not be accurate.
Example:
n=NSUR()
RAYTRACEX 0,1,0,0,0,1,0,NWAV()
y = RAYY(n)
Related Keywords:
RAYTRACE
READ
Purpose:
Reads data from an existing ASCII numeric text file opened for reading by the OPEN command.
Syntax:
READ x
READ x, y
READ x,y,z,a,b,c,q
Discussion:
The ASCII file must be already open, see the keyword OPEN for details. Each READ command reads a single
line from the file. The first valid data field from this line is placed in the variable first listed. The data from the
second field is placed in the second variable listed, if any. Therefore, the number of variables listed in the read
statement should match the number of columns in the text file. Numeric data in the file should be delimited by
spaces. The data may be in free-form, and is internally promoted to double precision. A maximum of 2000
characters can be read in on any single line. The maximum number of variable arguments is 199; for reading
longer lines with more arguments use READNEXT instead. The variables listed must be valid ZPL variable
names.
Always CLOSE a file after all the data has been read. See the function EOFF.
Example:
PRINT "Reading the double-column file TEST.DAT!"
OPEN "C:\DATA\TEST.DAT"
READ x1, y1
READ x2, y2
READ x3, y3
CLOSE
Related Functions:
EOFF
Related Keywords:
OPEN, CLOSE, READNEXT, READSTRING

READNEXT
Purpose:
Reads data from an existing ASCII numeric text file opened for reading by the OPEN command.
Syntax:
READNEXT x
READNEXT x, y
READNEXT x,y,z,a,b,c,q
Discussion:
READNEXT is almost identical to READ. The key difference is READ will read the entire data line from the
opened file, up to the newline character, while READNEXT reads only enough characters to fill the number of
arguments.
For example, if a data file contains a line with this data:
3.0 4.0 5.0
The following two READNEXT commands will read the values 3.0, 4.0, and 5.0 for x, y, and z:
READNEXT x, y
READNEXT z
READNEXT is more useful than READ if the line is very long, or the number of arguments is large.
Example:
OPEN "C:\DATA\TEST.DAT"
READNEXT x1, x2
READNEXT x3
CLOSE
Related Keywords:
OPEN, CLOSE, READ, READSTRING
READSTRING
Purpose:
Reads data from an existing ASCII text file opened for reading by the OPEN command.
Syntax:
READSTRING A$
Discussion:
The ASCII file must be already open, see the keyword OPEN for details. Each READ command reads a single
line from the file. The entire line read is placed in the variable listed. The variable listed must be a valid ZPL string
variable name, although it does not need to be previously referenced. Always CLOSE a file after all the data has
been read. See the function EOFF.
Example:
PRINT "Reading the contents of file TEST.DAT!"
OPEN TEST.DAT
READSTRING A$
PRINT A$
CLOSE
Related Keywords:
OPEN, CLOSE READ, READNEXT
RELOADOBJECTS
Purpose:
Reloads NSC objects into the NSC Editor.
Syntax:
RELOADOBJECTS surface, object
Discussion:

The expression for surface must evaluate to an integer surface number corresponding to the non-sequential
component surface. For NSC mode, use 1. The expression for object must evaluate to an integer object number,
or zero to reload all objects. This keyword may take a significant amount of time to execute, depending upon the
number and type of objects defined.
Example:
RELOADOBJECTS 1, 0
REM, !
Purpose:
REM is used to indicate the rest of the line is a remark.
Syntax:
REM text
! text
Discussion:
The exclamation symbol may also be used to indicate a remark. Both the REM command and the "!" symbol
are only recognized as remark indicators if they appear at the very beginning of the line, prior to any spaces, tabs,
or other characters. Any other use will cause a syntax error at run time.
Example:
REM any text can be placed after the REM command.
! any text can also be placed
! after the exclamation symbol.
REMOVEVARIABLES
Purpose:
Sets all currently defined variables to fixed status.
Syntax:
REMOVEVARIABLES
RENAMEFILE
Purpose:
RENAMEFILE is used to rename a file.
Syntax:
RENAMEFILE oldfilename newfilename
Discussion:
This keyword requires two file names, defined as literal string expressions in quotes or as string variables. The
file oldfilename is renamed newfilename.
Example:
RENAMEFILE AFILE$ BFILE$
Related Keywords:
COPYFILE
DELETEFILE
RETURN
See GOSUB.
REWIND
Purpose:
REWIND erases the last line printed by the PRINT statement, up to the previous end of line. This allows printing
a counter or other data over an existing line in the text output file.
Syntax:

REWIND
Example:
PRINT "First line"
REWIND
PRINT "New First line"
Related Keywords:
PRINT
SAVELENS
Purpose:
Saves the current lens file.
Syntax:
SAVELENS ["filename"]
SAVELENS NEW$
Discussion:
SAVELENS will save the current lens file to disk. The name of the current lens in memory will also be changed.
If the file name is absent, then the lens data is stored in the current file name.
Example:
SAVELENS
SAVELENS "NEWCOPY.ZMX"
SAVELENS NEW$
Related Keywords:
LOADLENS
SAVEMERIT
Purpose:
Saves the current merit function to a file.
Syntax:
SAVEMERIT "filename"
SAVEMERIT file$
Discussion:
SAVEMERIT will save the current merit function to a file. If the filename contains the complete path, such as
C:\MYDIR\MYLENS.MF, then the specified path will be used. If the path is left off, then the default directory for
lenses will be used (see Directories on page 59). See also LOADMERIT.
SAVEWINDOW
Purpose:
Saves the text from any text window to a file.
Syntax:
SAVEWINDOW winnum, filename
Discussion:
corresponds to the text window number that should be saved to a file. ZEMAX numbers windows sequentially as
they are opened, starting with 1. Any closed windows are deleted from the window list, without renumbering the
windows which remain. Any windows opened after another window has been closed will use the lowest window
number available.
Example:
SAVEWINDOW 1, "C:\TEMP\TEXTFILE.TXT"
SAVEWINDOW 3, A$

SCATTER
Purpose:
Used to control whether sequential surface scattering is done while tracing rays.
Syntax:
SCATTER ON
SCATTER OFF
Discussion:
The default condition at the start of a macro is SCATTER OFF; and all rays will be traced deterministically. If
SCATTER ON is executed, then sequential surface scattering will be enabled for all subsequent RAYTRACE
commands.
SDIA
Purpose:
SDIA is used to set the semi-diameter of a surface.
Syntax:
SDIA (surface) = (new_value)
Discussion:
If new_value evaluates to a positive number, then the semi-diameter status will be changed to "Fixed", no
matter what the initial status was. If new_value is negative, then the status is set to "Automatic" which will be
updated the next time "UPDATE" is called.
Example:
SDIA 5 = SDIA(4)
Related Keywords:
UPDATE
SETAIM
Purpose:
Sets the state of the ray aiming function.
Syntax:
SETAIM state
Discussion:
This keyword requires one numeric expression that must evaluate to either 0 or 1. The state is a code which
is 0 for ray aiming off and 1 for ray aiming on.
Example:
SETAIM 1
Related Keywords:
SETAIMDATA
SETAIMDATA
Purpose:
Sets various data for the ray aiming function. Syntax:
SETAIMDATA code, value
Discussion:

The code values are used as follows:
Code Property
1 Sets "Use Ray Aiming Cache" to true if value is 1, or false if value is 0.
2 Sets "Robust Ray Aiming" to true if value is 1, or false if value is 0.
3 Sets "Scale Pupil Shift Factors by Field" to true if value is 1, or false if value is 0.
4, 5, 6 Sets the value of the x, y, and z pupil shift, respectively.
Example:
SETAIMDATA 5, 0.34
Related Keywords:
SETAIM
SETAPODIZATION
Purpose:
Sets the system apodization type and factor.
Syntax:
SETAPODIZATION type, factor
Discussion:
This keyword requires the type to evaluate to an integer expression. The type is 0 for none, 1 for Gaussian
and 2 for tangential. The factor sets the apodization factor. See Apodization Type on page 86.
Example:
SETAPODIZATION 1, 0.7
SETCONFIG
Purpose:
Sets the current configuration for multi-configuration (zoom) systems.
Syntax:
SETCONFIG (confignumber)
Discussion:
This keyword requires one numeric expression that must evaluate to an integer between 1 and the maximum
number of configurations. The expression is evaluated and then rounded down to an integer to yield the
configuration number.
Example:
SETCONFIG 4
Related Functions:
CONF, NCON
SETDETECTOR
Purpose:
Sets the coherent or incoherent detector data for any pixel on a detector rectangle or detector surface object.
Syntax:
SETDETECTOR surf, object, pixel, datatype, value
Discussion:
This keyword requires numeric expressions that specify the surface, object, pixel, and data type (all integers)
and the new detector value. Surf is the surface number of the non-sequential group; use 1 when using non-
sequential mode. Object is the object number of the detector. The object must be either a detector surface or a

detector rectangle. Pixel is a number between 1 and the number of pixels the object supports. Datatype should
be 0 for incoherent intensity, 1 for incoherent intensity in angle space, 2 for coherent real part, 3 for coherent
imaginary part, and 4 for coherent amplitude. For more information on the use and meaning of these values see
Detector Rectangle object on page 332. The units for incoherent intensity are source units (see Source Units
on page 87). The units for coherent amplitude are the square root of source units.
Example:
SETDETECTOR 1, 5, 12, 0, 1.0056
Related Functions:
NSDD
SETMCOPERAND
Purpose:
Sets any row or configuration of the Multi-Configuration Editor to any numeric value.
Syntax:
SETMCOPERAND row, config, value, datatype
Discussion:
This keyword requires numeric expressions that evaluate to integers specifying the row and configuration of
the Multi-Configuration Editor.
If the config number is 0, then the value is interpreted as follows:
datatype = 0, value is a string literal or variable that specifies the name of the operand.
datatype = 1, 2, or 3, value is the number 1, 2, or 3 value used as part of the multi-configuration operand
definition. See the description of the multi-configuration operand numbers defined in SUMMARY OF MULTI-
CONFIGURATION OPERANDS on page 471.
If the config number corresponds to a defined configuration then the value is interpreted as follows:
datatype = 0, value is the value of the operand.
datatype = 1, value is the pickup offset of the operand.
datatype = 2, value is the pickup scale of the operand.
datatype = 3, value is the status of the operand, 0 for fixed, 1 for variable, 2 for pickup, 3 for thermal pickup.
datatype = 4, value is the pickup configuration number.
datatype = 5, value is the pickup row number.
Example:
SETMCOPERAND 3, 4, somevalue, 0
SETMCOPERAND 1, 0, "THIC", 0
Related Functions:
MCOP
SETNSCPARAMETER
Purpose:
Sets the parameter values of any object in the NSC editor.
Syntax:
SETNSCPARAMETER surface, object, parameter, value
Discussion:
This keyword requires 3 numeric expressions that evaluate to integers specifying the non-sequential compo-
nents surface number, the object number, and the parameter number. The fourth argument is the new value for
the specified parameter.
Example:
SETNSCPARAMETER 4, 2, 15, newp15value

Related Functions:
NPOS, NPAR
Related Keywords:
INSERTOBJECT, SETNSCPOSITION, SETNSCPROPERTY
SETNSCPOSITION
Purpose:
Sets the x, y, z or tilt x, tilt y, tilt z position of any object in the NSC editor.
Syntax:
SETNSCPOSITION surface, object, code, value
Discussion:
nents surface number, the object number, and a code. The code is 1 through 6 for x, y, z, tilt x, tilt y, tilt z,
respectively. The fourth argument is the new value for the specified position.
Example:
SETNSCPOSITION 4, 2, 2, newyvalue
Related Functions:
NPOS, NPAR
Related Keywords:
INSERTOBJECT, SETNSCPARAMETER, SETNSCPROPERTY
SETNSCPROPERTY
Purpose:
Sets properties of NSC objects.
Syntax:
SETNSCPROPERTY surface, object, code, csg, value
Discussion:
nents surface number (use 1 if the program mode is non-sequential), the object number, a code which specifies
what property of the object is being modified, and the coating and scatter group number (use 0 if not applicable
to the property being set). The fifth argument is the new value for the specified property, and it may be either text
in quotes, a string variable, or a numeric expression. The code is as follows:
Code Property
0 Sets the object type. The value should be the ASCII name of the object, such as "NSC_SLEN"
for the standard lens. The ASCII names for each object type are listed in the Prescription Report
for each object type in the NSC editor. All NSC object names start with "NSC_".
1 Sets the object comment.
2 Sets the reference object number.
3 Sets the "inside of" object number.
4 Sets the object material.
5 Sets the coating name for the specified csg.
6 Sets the profile name for the specified csg.
7 Sets the scatter mode for the specified csg, 0 = none, 1 = Lambertian, 2 = Gaussian, 3 = ABg.
8 Sets the scatter fraction for the specified csg.

Code Property
9 Sets the number of scatter rays for the specified csg.
10 Sets the Gaussian sigma for the specified csg.
11 Sets the reflect ABg data name for the specified csg.
12 Sets the transmit ABg data name for the specified csg.
13 Sets the "User Defined Aperture" flag. Use 1 for checked, 0 for unchecked.
14 Sets the "User Defined Aperture" filename.
The following codes set values on the Bulk Scattering tab of the Object Properties dialog.
81 Sets the "Model" value on the bulk scattering tab. Use 0 for "No Bulk Scattering", 1 for "Angle
Scattering", and 2 for "DLL Defined Scattering".
82 Sets the mean free path to use for bulk scattering.
83 Sets the angle to use for bulk scattering.
84 Sets the name of the DLL to use for diffraction splitting.
85 Sets the parameter value to pass to the DLL, where the csg value is used to specify which
parameter is being defined. The first parameter is 1, the second is 2, etc.
The following codes set values on the Diffraction tab of the Object Properties dialog.
91 Sets the "Split" value on the diffraction tab. Use 0 for "Dont Split By Order", 1 for "Split By Table
Below", and 2 for "Split By DLL Function".
92 Sets the name of the DLL to use for diffraction splitting.
93 Sets the Start Order value.
94 Sets the Stop Order value.
95 Sets the parameter values on the diffraction tab. These are the parameters passed to the
diffraction splitting DLL as well as the order efficiency values used by the "split by table below"
option. The csg value is used to specify which parameter is being defined. The first parameter
is 1, the second is 2, etc.
The following codes set values on the Sources tab of the Object Properties dialog.
101 Sets the source object random polarization. Use 1 for checked, 0 for unchecked.
102 Sets the source object reverse rays option. Use 1 for checked, 0 for unchecked.
103 Sets the source object Jones X value.
104 Sets the source object Jones Y value.
105 Sets the source object Phase X value.
106 Sets the source object Phase Y value.
107 Sets the source object initial phase in degrees value.
108 Sets the source object coherence length value.
109 Sets the source object pre-propagation value.
Other code values will be added in the future to set other properties of NSC objects as requested by users.
Example:

SETNSCPROPERTY 1, 2, 0, 0, "NSC_SLEN"
Related Functions:
NPOS, NPAR, NPRO
Related Keywords:
INSERTOBJECT, SETNSCPARAMETER, SETNSCPOSITION
SETOPERAND
Purpose:
Sets any row or column of the Merit Function Editor to any numeric value.
Syntax:
SETOPERAND row, col, value
Discussion:
This keyword requires numeric expressions that evaluate to integers specifying the row and column of the
Merit Function Editor. The col integer is 1 for operand type, 2 for int1, 3 for int2, 4-7 for Hx-Py, 8 for target, and 9
for weight. To set the comment string associated with an operand use col 10. To set the operand type, use a col
value of 1 and the integer operand returned by the function ONUM. An alternate method for setting the operand
type is to use a col value of 11 with the string name of the operand type, such as EFFL or DIST. If col is 10 or 11,
value should be a string constant or variable. Note the value and percent contribution columns cannot be set but
must be computed.
Example:
SETOPERAND 1, 8, tarvalue
SETOPERAND 3, 11, "EFFL"
SETOPERAND 5, 10, "Operand Number 5"
Related Functions:
MFCN, OPER, ONUM
SETSTDD
Purpose:
Sets surface tilt and decenter data.
Syntax:
SETSTDD surf, parameter, value
Discussion:
This keyword requires 2 numeric expressions that evaluate to integers specifying the surface and tilt/decenter
parameter, and a third parameter for the new value. The parameter is used as follows:
0: Before order. 0 for dec/tilt, 1 for tilt/dec.
1-5: Before decenter x, decenter y, tilt x, tilt y, tilt z, respectively.
6: After status. 0 for explicit, 1/2 for pick up/reverse current surface, 3/4 for pick up/reverse current-1 surface,
5/6 for pick up/reverse current -2 surface, etc...
7: After order. 0 for dec/tilt, 1 for tilt/dec.
8-12: After decenter x, decenter y, tilt x, tilt y, tilt z, respectively.
Example:
SETSTDD 3, 1, before_dx_value
Related Functions:
STDD
SETSURFACEPROPERTY
Purpose:
Sets properties of surfaces.

Syntax:
SETSURFACEPROPERTY surface, code, value
Discussion:
This keyword requires 2 numeric expressions that evaluate to integers specifying the surface number, and a
code which specifies what property of the surface is being modified. The third argument is the new value for the
specified property, and it may be either text in quotes, a string variable, or a numeric expression. The code is as
follows:
Code Property
0 Sets the surface type. The value should be the ASCII name of the object, such as "STANDARD"
for the standard surface. The ASCII names for each surface type are listed in the Prescription
Report in the Surface Data Summary for each surface type currently in the Lens Data Editor.
1 Sets the comment.
2 Sets the curvature (not radius) in inverse lens units. Use zero for an infinite radius.
3 Sets the thickness in lens units.
4 Sets the glass.
5 Sets the conic constant.
20 Sets the surface aperture type. See APTP on page page 550 for the aperture type codes.
21, 22 Sets the surface aperture min, max value respectively.
23, 24 Sets the surface aperture decenter x, y value respectively.
25 Sets the User Defined Aperture (UDA) file name.
30 Physical Optics setting: "Use Rays To Propagate To Next Surface". Use 1 for true, 0 for false.
31 Physical Optics setting: "Do Not Rescale Beam Size Using Ray Data". Use 1 for true, 0 for false.
32 Physical Optics setting: "Use Angular Spectrum Propagator". Use 1 for true, 0 for false.
33 Physical Optics setting: "Draw ZBF On Shaded Model". Use 1 for true, 0 for false.
34 Physical Optics setting: "Recompute Pilot Beam Parameters". Use 1 for true, 0 for false.
35 Physical Optics setting: "Resample After Refraction". Use 1 for true, 0 for false.
36 Physical Optics setting: "Auto Resample". Use 1 for true, 0 for false.
37 Physical Optics setting: "New X Sampling". Use 1 for 32, 2 for 64, etc..
38 Physical Optics setting: "New Y Sampling". Use 1 for 32, 2 for 64, etc.
39 Physical Optics setting: "New X-Width". New total x direction width of array.
40 Physical Optics setting: "New Y-Width". New total y direction width of array.
41 Physical Optics setting: "Reference Radius". Use 0 for best fit, 1 for shorter, 2 for longer, 3 for
x, 4 for y, 5 for plane.
80 Sets the scatter code; 0 for none, 1 for Lambertian, 2 for Gaussian, 3 for ABg.
81 Sets the scatter fraction, should be between 0.0 and 1.0.
82 Sets the Gaussian scatter sigma.
83 Sets the ABg name.

Other code values will be added in the future to set other properties of surfaces as requested by users. Some
of the code values used are redundant with other ZPL keywords, for example, the command
CURV 5 = 23.4
is equivalent to
SETSURFACEPROPERTY 5, 2, 23.4
The function SPRO uses a very similar syntax and identical code values to "get" rather than "set" these same
values. Usually, changes to surface properties do not become effective until after the UPDATE keyword is
executed.
Example:
SETSURFACEPROPERTY 1, 4, "BK7"
Related Functions:
SPRO
Related Keywords:
UPDATE
SETTEXTSIZE
Purpose:
Changes the size of the characters drawn by the GTEXT command.
Syntax:
SETTEXTSIZE xsize ysize
Discussion:
The arguments refer to the fraction of the graphic screen width that each character represents. For example,
the default text size is 70 40. This means each character is 1/70 of the graphic screen width, and 1/40 of the
screen height. An argument of zero restores the text size to the default.
Example:
! Make text twice default size
SETTEXTSIZE 35 20
! Restore text size to default
SETTEXTSIZE
SETTITLE
Purpose:
Sets the title of the lens that normally appears on all plots.
Syntax:
SETTITLE A$
SETTITLE "Here is the lens title"
SETUNITS
Purpose:
Sets the current lens units.
Syntax:
SETUNITS code
Discussion:
This keyword requires one numeric expression that must evaluate to an integer between 0 and 3. The code
is 0 for mm, 1 for cm, 2 for inches, and 3 for meters. This function does not scale or convert the lens data in any
way, it only changes how the lens prescription data is interpreted.
Example:
SETUNITS 0

SETVAR
Purpose:
Changes the state of variables for optimization.
Syntax:
SETVAR (surface_expression), VARCODE, (status_expression), (object_number)
or
SETVAR (config), M, (status_expression), (operand_number)
Discussion:
The surface_expression must evaluate to an integer argument between 0 and the maximum number of
surfaces, or an error will result. The surface_expression determines which surface will be made variable. The
VARCODE must be one of the following ASCII mnemonics (or a string variable containing the mnemonic):
R for radius of curvature
T for thickness
C for conic
G for glass
I for glass index
J for glass Abbe
K for glass dpgf
Pn for parameter n
D for thermal coefficient of expansion
En for extra data value n
M for multi-configuration data, see discussion below
Nn for non-sequential component position data, 1-6 for x, y, z, tx, ty, tz
On for non-sequential component parameter data, where n is the parameter number
If the status_expression evaluates to 0, then the variable status is removed. Otherwise, the value is made
variable. If the VARCODE is Nn or On; the object number must be provided; otherwise, it should be omitted. If
the VARCODE is M, then the syntax for this command is as shown under "syntax" above.
Examples:
SETVAR j+3, R, 1
SETVAR 5, P6, 0
SETVAR surfk+2, E06, status
SETVAR config, M, status, operand
SETVAR 1, O32, 1, 5
SETVECSIZE
Purpose:
Changes the maximum size of the VEC1, VEC2, VEC3, and VEC4 arrays.
Syntax:
SETVECSIZE (expression)
Discussion:
The expression must evaluate to an integer argument between 1 and 2,000,000. All four vector variables are
always the same size. All data in the vectors will be lost during the resize. The initial size of the vectors is 1000.
SETVIG
Purpose:
Sets the vignetting factors for the lens.
Syntax:
SETVIG
Discussion:
See the Chapter "Conventions and Definitions" for a description of vignetting factors.
SHOWFILE
Purpose:

Displays a text file to the screen using the ZEMAX file viewer.
Syntax:
SHOWFILE "filename" saveflag
SHOWFILE NAME$ saveflag
Discussion:
The filename must be a valid filename within quotes or a string variable containing a valid file name. The file
must be an ASCII file (as would be created by OUTPUT and PRINT statements in ZPL) and must be in the current
directory. Once the file is displayed, it may be scrolled up and down and printed like any other text file. The ability
to scroll and print the data is the primary advantage of using OUTPUT and SHOWFILE instead of PRINT
statements. SHOWFILE also closes the file if no CLOSE statement has been executed. If the saveflag is zero or
omitted, then the file is erased when the window is closed. if saveflag is any value other than zero, then the file
remains even after the window is closed.
Example:
OUTPUT "test.txt"
PRINT "Print this to a file."
SHOWFILE "test.txt"
Related Keywords:
OPEN, OUTPUT, CLOSE, PRINT, PRINTFILE
SOLVETYPE
Purpose:
Changes the solve status on a given surface and value. Only some solve types are supported; contact ZEMAX
technical support for information on setting other types of solves.
Syntax:
SOLVETYPE (surf_expression), CODE, (arg1), (arg2), (arg3)
Discussion:
The surf_expression must evaluate to an integer argument between 0 and the maximum number of surfaces,
or an error will result. The CODE must be an ASCII mnemonic as listed in the following table, or the name of a
string variable holding the mnemonic. The arg1, arg2, and arg3 expressions evaluate to the first, second, and
third solve parameters as specified in SOLVES on page 379. For non-sequential pickup solves, the arguments
are the first, second, and third lines that follow the "Solve Type" selection on the NSC solve dialog box. Some
codes do not use any or all of the arguments, but dummy values for the unused arguments (such as 0) must still
be provided and be separated by commas or a syntax error will result.
CODES FOR SOLVETYPE KEYWORD
Solve Type CODE
Curvature: Fixed (turn solve off) CF
Curvature: Pickup CP
Thickness: Fixed (turn solve off) TF

Thickness: Edge Thickness TE
Thickness: Marginal Ray Height TM
Thickness: Pickup TP
Glass: Fixed (turn solve off) GF
Glass: Model GM
Glass: Offset GO
Glass: Pickup GP

Solve Type CODE
Conic: Pickup. KP
Parameter: Parameter 0-16 Pickup. P0, P1, P2...P16

Non-Sequential Component Pickup X, Y, Z, Tilt-X, Tilt-Y, NSC_PX, NSC_PY, NSC_PZ, NSC_PTX, NSC_PTY,
Tilt-Z, Material. Append an underscore then the object num- NSC_PTZ, NSC_PMAT
ber to the code, for example, NSC_PX_14 will set a pick up (see left)
solve on object 14.
Non-Sequential Component Parameter Pickup. Append an NSC_PP

underscore then the object number, then another under- (see left)
score and the parameter number to the code, for example,
NSC_PP_11_7 will set a pick up solve on object 11, param-
eter 7.
Example:
! The following line will add a glass pickup solve
! on surface 7, picking up from surface 5:
SOLVETYPE 7, GP, 5
! Add a thickness pickup with a scale factor of -1:
SOLVETYPE 7, TP, 5, -1
Related Functions:
SOLV
STOPSURF
Purpose:
STOPSURF sets the current stop surface location by number.
Syntax:
STOPSURF (surface)
Discussion:
This keyword requires one expression which evaluates to an integer between 1 and the number of surfaces
minus one. The UPDATE command must be issued before the new data takes effect.
Example:
STOPSURF n+2
Related Keywords:
UPDATE
SUB
See GOSUB.
SURFTYPE
Purpose:
Changes the surface type.
Syntax:
SURFTYPE (surf_expression), CODE
Discussion:
The surf_expression must evaluate to an integer between 0 and the number of surfaces. The CODE is an
integer value which indicates the surface type. The integer codes may be determined at run time by the SCOD
function. The SURFTYPE command cannot be used to set a grid sag, grid phase, or user defined surface.
Example:
SURFTYPE j+1, SCOD("STANDARD")
B$ = "EVENASPH"
SURFTYPE 5, SCOD(B$)

TELECENTRIC
Purpose:
Sets the telecentric object space mode. See Telecentric Object Space on page 87.
Syntax:
TELECENTRIC = (new_value)
Discussion:
If (new value) evaluates to 0, telecentric mode is turned off; otherwise, telecentric mode is turned on.
TESTPLATEFIT
Purpose:
TESTPLATEFIT calls the test plate fitting routine. See Test Plate Fitting on page 197.
Syntax:
TESTPLATEFIT tpd_file log_file method number_cycles
Discussion:
This keyword requires string expressions for the test plate data file, and the name of a file for the output log
file. The method is an integer between 0 and 4, inclusive, for try all methods, best to worst, worst to best, long to
short, and short to long, respectively. The integer number_cycles is 0 for automatic or the maximum number of
optimization cycles to execute. Note the tpd_file name should NOT include the path, since all test plate files are
in a fixed directory; while the path should be included for the log file.
This keyword may take a long time to execute. It is advisable to display the log file after completion of the
fitting, or use some other means to indicate when the fitting is complete.
Example:
TESTPLATEFIT "optico.tpd" "c:\temp\logfile.dat" 0 0
SHOWFILE "c:\temp\logfile.dat"
THIC
Purpose:
THIC is used to set the thickness of a surface.
Syntax:
THIC (surface) = (new_value)
Discussion:
Example:
THIC i+m = THIC(i+n)-1
Related Keywords:
UPDATE
TIMER
Purpose:
Resets the internal clock. This feature is used in conjunction with the ZPL function ETIM() for measuring the
elapsed time since the last TIMER command.
Syntax:
TIMER
Discussion:

TIMER and ETIM() are used primarily for testing execution efficiency of the ZPL interpreter and various
program architectures.
Example:
i=0
TIMER
LABEL 1
x=RAND(1000)
i=i+1
if i < 10000 THEN GOTO 1
FORMAT .1
PRINT "Elapsed time:", ETIM(), "Seconds"
UNLOCKWINDOW
Purpose:
Unlocks any locked window.
Syntax:
UNLOCKWINDOW winnum
Discussion:
See Graphic windows operations on page 36.
Example:
UNLOCKWINDOW 2
Related Keywords:
LOCKWINDOW
UPDATE
Purpose:
Updates pupil positions, index data, paraxial constants, semi-diameters, maximum field normalization values,
and solves. The UPDATE keyword MUST be used before tracing or evaluating a system if the prescription data
(such as radii, thicknesses, system apertures, wavelengths) have been altered since the last UPDATE. If the
UPDATE command is followed by the keyword "ALL", then all open windows will be updated as well. If the
UPDATE command is followed by an expression which evaluates to an integer corresponding to an open window,
then the specified window is updated.
Syntax:
UPDATE
UPDATE ALL
UPDATE n
Example:
THIC 5 = THIC (5)-1
UPDATE
VEC1, VEC2, VEC3, VEC4
Purpose:
These keywords are used to set the vector variables VEC1, VEC2, VEC3, and VEC4. Each vector can store
an array of double-precision floating point numbers.
Syntax:
VEC1 (array_subscript) = (new_value)
VEC2 (array_subscript) = (new_value)
Discussion:
VEC1..4 are used to store any user data in an array. The array_subscript value can be any expression which
is rounded down to an integer. The resulting integer expression must be between 0 and the current maximum
vector size, which is initially 1000, or an error will be reported. The ZPL functions VEC1..4 can be used to extract
the data.

To change the default size to a value different from 1000, use the SETVECSIZE keyword.
Example:
i = 0
LABEL 1
i = i + 1
VEC1(i)=i
IF i < 10 THEN GOTO 1
j = 0
LABEL 2
j = j + 1
VEC2(j) = VEC1(j) * VEC1(j)
IF j < 10 THEN GOTO 2
i = 0
LABEL 3
i = i + 1
PRINT "x = ", VEC1(i), " x*x = ", VEC2(i)
IF i < 10 THEN GOTO 3
PRINT
PRINT "All done!"
WAVL, WWGT
Purpose:
WAVL and WWGT are used to change the wavelength value and weight.
Syntax:
WAVL (wavelength number) = (new_value)
WWGT (wavelength) = (new_value)
Discussion:
This keyword requires two expressions, one to specify the wavelength number and the other to define the new
value. The expression (wavelength number) is evaluated and then rounded down to an integer to yield the
wavelength number. If the wavelength number is less than 1 or greater than the number of wavelengths, the
command is ignored. Otherwise, the expression (new_value) is evaluated and assigned. The UPDATE command
must be issued before the new data takes effect.
Example:
WAVL 1 = 0.486
WAVL 2 = 0.587
WWGT 1 = 1.0
WWGT 2 = 0.65
Related Functions:
WAVL, WWGT
Related keyword:
NUMWAVE
XDIFFIA
Purpose:
Computes the Extended Diffraction Image Analysis feature and saves the result to a ZBF file. For a description
of the ZBF (ZEMAX Beam File) format see ZEMAX Beam File (ZBF) binary format on page 528. For a
description of the Extended Diffraction Image Analysis feature see Extended Diffraction Image Analysis on
page 151.
Syntax:
XDIFFIA outfilename infilename
XDIFFIA out$ in$
Discussion:
This keyword requires the name of the output ZBF file, and optionally, the name of the input IMA or BIM file.
If the extension to the outfilename is not provided, the extension ZBF will be appended. The extension must be

provided on the infilename. The filenames must be enclosed in quotes if any blank or other special characters
are used. The outfilename will be placed in the \POP\Beamfiles subdirectory. The infilename must be placed in
the \IMAFiles subdirectory. No paths should be provided with the file names.
The settings for the Extended Diffraction Image Analysis feature will be those settings previously saved for the
current lens. To make adjustments to the settings, open an Extended Diffraction Image Analysis window, choose
the appropriate settings, then press "Save". All subsequent calls to XDIFFIA will use the saved settings. The
exceptions are the output file name, which is specified as the first argument after the XDIFFIA keyword, and the
input source file, which is optionally specified as the second argument after the XDIFFIA keyword.
Example:
XDIFFIA "output will be in this file.ZBF" "SOMEIMAFILE.IMA"
Related Functions:
ZBFCLR, ZBFSHOW, ZBFSUM, ZBFMULT
ZBFCLR
Purpose:
Clears the complex amplitude data in a ZBF file.
Syntax:
ZBFCLR filename
Discussion:
This keyword requires only the name of the ZBF file.
General comments about all ZBF related keywords
For a description of the ZBF (ZEMAX Beam File) format see ZEMAX Beam File (ZBF) binary format on
page 528. ZEMAX requires the filename extension to be ZBF, and will append or replace the extension as
required. The filename may be enclosed in quotes if any blank or other special characters are used. The file must
be located in the \POP\Beamfiles subdirectory. All ZBF output files will be placed in this same directory.
Example:
ZBFCLR "some beam file name.ZBF"
ZBFCLR N$
ZBFMULT
Purpose:
Multiplies the complex amplitude data in a ZBF file by a complex factor.
Syntax:
ZBFMULT filename, Ax, Bx, Ay, By
Discussion:
This keyword requires the name of the ZBF file and the real (A) and imaginary (B) parts of the complex number
to multiply every point in the ZBF file by. There are separate A and B for the x and y polarized light directions.
The resulting data is written back to the same file name.
See ZBFCLR on page 602 for comments that apply to all ZBF related keywords.
Example:
ZBFMULT "some beam file name.ZBF", 0.0, 1.0, 0.0, -1.0
ZBFPROPERTIES
Purpose:
Opens the specified ZBF file and places various data about the beam in a vector variable.
Syntax:
ZBFPROPERTIES filename, vector
Discussion:

This keyword requires the name of the ZBF file and a value for the vector number to place the data in. The
value of vector must be between 1 and 4, inclusive. After the ZBFPROPERTIES function executes, the following
beam data will be placed in the specified vector: nx, ny, dx, dy, waist_x, waist_y, position_x, position_y, rayleigh_x,
rayleigh_y, and wavelength (in lens units); the values are placed in vector positions 1 through 11.
Example:
ZBFPROPERTIES "TEST.ZBF", 1
rayleighx = VEC1(9)
ZBFRESAMPLE
Purpose:
Resamples a ZBF file to a new width and point spacing.
Syntax:
ZBFRESAMPLE filename, nx, ny, wx, wy, decenterx, decentery
Discussion:
This keyword requires the name of the ZBF file and six numbers. The beam will be resampled and interpolated
as required to create a new beam file with nx and ny points, of total width wx and wy, in the x and y directions
respectively. The nx and ny values must be powers of 2; such as 32, 64, 128, etc. The decenterx and decentery
values may be provided to optionally decenter the new beam relative to the old beam. If either nx or ny is zero,
no change is made to the existing beam sampling. If either wx or wy is zero, no change is made to the existing
beam width. The resulting data is written back to the same file name.
Example:
ZBFRESAMPLE "TEST.ZBF", 128, 128, 25.4, 25.4, 0, 0.4
ZBFSHOW
Purpose:
Displays a ZBF file in a viewer window.
Syntax:
ZBFSHOW filename
Discussion:
This keyword requires only the name of the ZBF file. This command will open a new window to display the
beam file.
Example:
ZBFSHOW "new beam data.ZBF"
ZBFSHOW N$
ZBFSUM
Purpose:
Sums either coherently or incoherently the data in two ZBF files and places the resulting data in a third ZBF file.
Syntax:
ZBFSUM coherent filename1 filename2 outfilename
Discussion:
This keyword requires an integer to indicate if the sum is coherent (any value other than zero) or incoherent
(zero), and the names of three ZBF files. If an incoherent sum is performed, then the output data will be real
valued only. If the two source files do not have the same number of data points and point spacing in both x and
y directions, the second source file listed is first scaled and interpolated as required to match the sampling and
point spacing of the first file, then the sum is performed. The outfilename may be the same as one of the source
file names; in which case the original file is overwritten.

Example:
ZBFSUM 1 "a.ZBF" "b.zbf" "coherent a plus b.zbf"
ZBFSUM 0 "a.ZBF" "b.zbf" "incoherent a plus b.zbf"
ZBFSUM 0 A$ B$ C$
ZBFTILT
Purpose:
Multiplies the data in a ZBF file by a complex phase factor to introduce phase tilt to the beam.
Syntax:
ZBFTILT filename, cx, cy, tx, ty
Discussion:
This keyword requires the name of the ZBF file and four numbers. The phase of the beam is modified by a
phase angle given by = ( x cx )tx + ( y cy )ty . The cx and cy values are the center of the phase tilt, and tx
and ty are the slopes of the tilt in units of radians per beam unit length. The coordinates x and y refer to positions
within the beam file, with the center coordinate (x = 0, y = 0) being at the point (nx/2 + 1, ny/2 + 1) where nx and
ny are the number of points in the x and y directions. The resulting data is written back to the same file name.
Example:
ZBFTILT "TEST.ZBF", 0.0, 0.0, 0.0, 0.01237
Example 1
Suppose you needed a macro to print out the image plane intercept coordinates of the chief ray at every
defined field angle. This is more commonly required than you may think, for example, ZEMAX spot diagrams are
all plotted relative to the chief ray coordinates, and so the intercept coordinates are needed to properly offset the
data. The same is true for transverse ray aberrations; they are referenced to the chief ray.
The best way to learn ZPL is to study previously written macros. Look closely at the following macro and try
to follow the macro logic.
nfield = NFLD()
maxfield = MAXF()
n = NSUR()
FOR i=1,nfield,1
hx = FLDX(i)/maxfield
hy = FLDY(i)/maxfield
PRINT "Field number ", i
RAYTRACE hx,hy,0,0,PWAV()
PRINT "X-field angle : ",FLDX(i)," Y-field angle : ", FLDY(i)
PRINT "X-chief ray : ",RAYX(n), " Y-chief ray : ", RAYY(n)
PRINT
NEXT
PRINT "All Done!"
The first line of the macro calls the NFLD() function which returns the number of defined field angles, and
assigns it to the variable "numfield". The second line calls MAXF(), which returns the maximum radial field angle
in degrees, and stores the value in the variable "maxfield". The number of surfaces is then stored in the variable
"n" by calling the function NSUR().
The FOR loop is then defined with i being the counter for the field position, starting at 1, with a maximum value
of nfield, and an increment of 1 on each loop. The two macro lines which define "hx" and "hy" use the functions
FLDX() and FLDY(), which return the x- and y-angles for the current field position number "i". The chief ray is then
traced by the keyword RAYTRACE. Note the chief ray intersects the center of the pupil, which is why the two pupil
coordinates are both zero. The PWAV() function returns the primary wavelength number, which is usually used
for the chief ray.
The various PRINT statements will output the chief ray coordinates on the image plane to the screen. If you
like, try typing the macro into a file ending in ZPL (for example, CHIEF.ZPL) and then run it from within ZEMAX

to test it. The coordinates reported should be the same as those listed on the spot diagram when the chief ray is
the reference point.
Example 2
Here is an example ZPL macro to estimate the RMS spot radius (on axis) of the current optical system. The
macro traces many random rays through the system, and records the radial offset from the primary wavelength
chief ray. The current wavelength weighting is applied to estimate the RMS spot radius.
PRINT "Primary wavelength is number ",
FORMAT .0
PRINT PWAV(),
FORMAT .4
PRINT " which is ", WAVL(PWAV()), " micrometers."
PRINT "Estimating RMS spot radius for each wavelength."
! How many random rays to trace to make estimate?

n = 100
! Initialize the timer

TIMER
! Store the number of surfaces for later use

ns = NSUR()
! Start at wavelength 1
weightsum=0
wwrms = 0
FOR w = 1, NWAV(), 1
rms = 0
FOR i = 1, n, 1
hx=0
hy=0
angle = 6.283185 * RAND(1)
! SQRT yields uniform distribution in pupil
radius = SQRT(RAND(1))
px = radius * COSI(angle)
py = radius * SINE(angle)
RAYTRACE hx,hy,px,py,w
x=RAYX(ns)
y=RAYY(ns)
rms = rms + (x*x) + (y*y)
NEXT
rms = SQRT(rms/n)
wwrms = wwrms + ( WWGT (w) * rms )
weightsum = weightsum + WWGT(w)
FORMAT .4
PRINT "RMS spot radius for ", WAVL(w),
FORMAT .6
PRINT " is ", rms
NEXT
wwrms = wwrms / weightsum
PRINT "Wavelength weighted rms is ", wwrms
FORMAT .2
t = ETIM()
PRINT "Elapsed time was ",t," seconds."
Note the use of the trailing commas in the first two PRINT statements. These allow the data printed from the
first three PRINT statements to appear on the same line. The intervening FORMAT statements changes the way
the numerical output is printed, even on the same line. The "!" character is used to indicate a comment, which is
ignored when the macro is run.

Chapter 23 ZEMAX EXTENSIONS
Introduction
ZEMAX has a very powerful feature which allows another Windows program to establish a communication link
with ZEMAX, and for this other program to get data from ZEMAX on the lens. The idea is that a program can use
ZEMAX to trace rays through the lens, and then the data is sent to the other program for further analysis or
computation.
!Extensions are more complicated than macros; for simple programming jobs see ZEMAX
PROGRAMMING LANGUAGE on page 535.
There are three closely related ways this feature can be used to extend the capabilities of ZEMAX:
First, any stand-alone Windows program may be designed to establish a link with ZEMAX which is used to
extract data about a lens that ZEMAX can provide- typically ray trace or other optical data. The program may
use this data any way it sees fit.
Second, a stand alone program can do its analysis "hidden" from the user, and then the data generated may
be displayed in the usual ZEMAX graphic or text window. When using this mode, the program is called an
"Extension" to ZEMAX. Extensions appear as menu options under the "Extensions" menu in the ZEMAX
program. The Extensions must be placed in the directory \Extend for ZEMAX to run them.
Third, the stand alone program may be used to compute data that ZEMAX can optimize. In this mode, the
program is called a "User Defined Operand" or UDO. These UDO's must be placed in the directory \UDO for
ZEMAX to run them.
Applications that require complex user interfaces, with many inputs, menus, and buttons are better suited to
be stand alone applications that do not use ZEMAX to display the computed data. Applications that may be
complex computationally, but require only a single dialog box type user interface, are better suited to be integrated
with ZEMAX as Extensions. Extensions have the advantage of looking and acting just like regular ZEMAX
analysis windows. Features implemented as Extensions may be viewed within ZEMAX, updated, have settings
changed, and be printed, copied to the clipboard, or be panned and zoomed just like any ZEMAX analysis window.
UDO's are for applications that compute data that must be optimized by the ZEMAX optimizer. UDO's are
described in detail in the chapter "Optimization" under "User defined operands".
Well written applications may even be designed to operate either as a stand alone program or as an Extension,
as shown in the examples section.
The communication between application and ZEMAX is accomplished using Dynamic Data Exchange (DDE).
DDE is a protocol defined within the Windows operating system for sharing data between programs. Two
programs can establish a DDE link, with one program acting as the "server" and the other the "client". The client
generally requests specific data from the server, and the server sends the data back to the client.
ZEMAX is intended to be the server, and any Windows program with the right modifications can be made a
client. This chapter defines the DDE interface to ZEMAX, so that the user may write software programs which
take advantage of the ZEMAX ray trace engine.
Requirements for writing Extensions
The purpose of this chapter and the DDE features is to assist experienced programmers in writing code that
can interface to ZEMAX. This feature allows programmers to write third party software programs that work in
conjunction with ZEMAX. The third party program may use ZEMAX as a ray trace engine, but provide an interface,
graphics, input and output, and extended calculations that are customized for a particular type of analysis.
Extensions can be easily implemented using any program or programming language that supports DDE.
Examples of such programs include Visual Basic, Excel, and MatLab, but there are many others and DDE is a
607 Chapter 23: ZEMAX EXTENSIONS

fairly common interface between scientific programs. If you are writing an application in C, then using this feature
requires a knowledge of Windows and C programming, with at least some mastery over message passing,
message loops, memory management, and global atoms, handles, and pointers. An excellent book that contains
all the information needed to learn DDE programming is "Programming Windows" by Charles Petzold, Microsoft
Press. The examples and code sections in the text of this chapter are not complete programs, but are code
fragments intended only for illustration.
Although programming a Windows application is somewhat complex, the good news is that all of the difficult
code has already been written, and is available for study and re-use. The details on establishing the communi-
cation link with ZEMAX have already been worked out, and the sample code may be "cut and paste" with only
the code relating to the specific computation required needing change.
There is also a C source code file, called ZCLIENT, that vastly simplifies the DDE communication with ZEMAX
by encapsulating all the communication in wrapper functions. ZCLIENT provides a number of functions that can
be called to extract data from ZEMAX without requiring knowledge of DDE programming. ZCLIENT is described
later in this chapter.
There are complete examples of client programs and Extensions, including source code, provided with ZEMAX
and discussed at the end of this chapter.
The power of the Extension comes at a price, although it is a reasonable one. Use of the DDE features does
require that the user have a suitable compiler or development tool that can generate 32 bit Windows executables.
It is also assumed that the user can write the required code, and most importantly, ensure that the code is reliable
and bug-free. To maximize speed, ZEMAX performs very little error checking on data returned by the client, and
so buggy Extensions are capable of bringing ZEMAX to a crash, or placing ZEMAX in an infinite loop.
For this reason, technical support on the implementation of DDE or Extensions is strictly limited to illustrating
that the provided sample files work correctly. If you need a ZEMAX Extension, and do not possess the desire or
ability to write them yourself, please feel free to contact ZEMAX Technical Support for a quote on developing a
custom program to meet your requirements. We have considerable experience in developing these types of
programs, and can generally write Extensions at very competitive rates on short notice.
Establishing the link
To establish a DDE link with ZEMAX, the client program must broadcast a message to all top level windows
that includes a reference to the application name, and the topic name. For ZEMAX, the application name is
"ZEMAX", and the topic name can be any non-NULL string. ZEMAX does not use the topic name, just the
application name and the "item". The item indicates to ZEMAX what data is being requested. The various items
ZEMAX supports are described later.
See the many sample DDE client codes for specific code examples.
Once the DDE link is established, any analysis that can be done based upon ray tracing may be performed
by the server, and the data returned to the client.
Terminating the link
When all computations are complete, and the DDE link is to be terminated by the client, the client sends a
termination message to the ZEMAX server. This message will terminate only the server portion of ZEMAX, not
ZEMAX itself.
Extracting data from ZEMAX
ZEMAX supports a number of capabilities under DDE. Each function is given a name, called the "item", and
this item is passed to the ZEMAX server using the WM_DDE_REQUEST message. Any data ZEMAX requires
(for example to trace a ray) is passed to ZEMAX by encoding within the item name. Most items in ZEMAX have
names that require no further encoding, but some items have coded values appended to the name. The values
are appended to the item name and are delimited by commas.
ZEMAX responds to each item request with a WM_DDE_DATA message, and the message contains a pointer
to a block of memory holding a string. Most data is passed from ZEMAX back to the client application in a string,
this corresponds to CF_TEXT format. The client application must extract the string and free the memory for the
string.
See the many sample DDE client codes for specific code examples.
Chapter 23: ZEMAX EXTENSIONS 608

The data items
Here are the supported data items:
CloseUDOData
This item is used to close the user defined operand buffer, which allows the ZEMAX optimizer to proceed. The
only time this item name should be used is when implementing a User Defined Operand, or UDO. UDO's are
described in detail in the chapter "Optimization". See that chapter for details. The syntax for CloseUDOData is:
CloseUDOData, buffercode
See also GetUDOSystem and SetUDOItem.
DeleteConfig
This item inserts deletes an existing configuration in the multi-configuration editor. The syntax is:
DeleteConfig, config
See also InsertConfig.
DeleteObject
This item deletes an existing object. The syntax is:
DeleteObject,surf,object
See also InsertObject.
DeleteSurface
This item deletes an existing surface. The syntax is:
DeleteSurface,surf
See also InsertSurface.
ExportCAD
This item export lens data in IGES/STEP/SAT format. The syntax is:
ExportCAD,filename,filetype,numspline,first,last,rayslayer,lenslayer,exportdummy,
usesolids,raypattern,numrays,wave,field,deletevignetted,dummythick,split,scatter,
usepol,config
For details on this feature and the meaning of all parameters, see EXPORTCAD on page 556, and Export
IGES/SAT/STEP Solid on page 208.
There is a complexity in using this feature via DDE. The export of lens data may take a long time relative to
the timeout interval of the DDE communication. Therefore, calling this data item will cause ZEMAX to launch an
independent thread to process the request. Once the thread is launched, the return value is the string "Exporting
filename". However, the actual file may take many seconds or minutes to be ready to use. To verify that the export
is complete and the file is ready, use the ExportCheck data item. ExportCheck will return 1 if the export is still
running, or 0 if it has completed. Generally, the ExportCheck data item call will need to be placed in a loop which
executes until ExportCheck returns 0. A typical loop test in C code might look like this:
/* check if the export is done */
still_working = 1;
while(still_working)
{
// Delay for 200 milliseconds
Sleep(200);
PostRequestMessage("ExportCheck", szBuffer);
if (szBuffer[0] == '1')
{
// Still working!
}
else
{
// All done!
still_working = 0;

}
}
If ExportCad returns "BUSY!" then the export thread is already running from a previous request, and the export
will not be performed.
ExportCheck
This item indicates the status of the last executed ExportCAD data item. The return value is 1 if still running,
or 0 if it has completed. See ExportCad above for a discussion of the use of this data item.
FindLabel
This item finds the surface that has the integer label associated with the specified surface. The syntax is
FindLabel,label
The returned value is the surface number of the first surface with the identical integer label, or -1 if no surface
has the specified label. See also SetLabel and GetLabel.
GetAddress
This item extracts the address line 1, 2, 3 or 4 from the environment settings. The syntax is
GetAddress,1
with similar commands for lines 2, 3, and 4. If the user has selected to hide the address box, then all strings
return as "\r\n".
GetAperture
This item extracts surface aperture data. The syntax is
GetAperture,surf
The returned string is formatted as follows:
"type, min, max, xdecenter, ydecenter, aperturefilename"
This item returns the type as an integer code; 0 for no aperture, 1 for circular aperture, 2 for circular
obscuration, 3 for spider, 4 for rectangular aperture, 5 for rectangular obscuration, 6 for elliptical aperture, 7 for
elliptical obscuration, 8 for user defined aperture, 9 for user defined obscuration, and 10 for floating aperture. The
min and max values have different meanings for the elliptical, rectangular, and spider apertures than for circular
apertures; see Aperture type and other aperture controls on page 68 for details. See also SetAperture.
GetAspect
This item extracts the graphic display aspect ratio and the width or height of the printed page in current lens
units. For example, If the current aspect ratio is 3 x 4, the aspect ratio returned will be 0.75. Knowing the correct
aspect ratio is required when drawing isometric plots. The format of the returned data is aspect, width. If the aspect
ratio is greater than 1, then the plot is taller than it is wide, and the format of returned data is aspect, height.
The syntax is:
GetAspect,filename
where filename is the name of the temporary file associated with the window being created or updated. If the
temporary file name is left off; then the default aspect ratio and width (or height) is returned.
GetBuffer
The GetBuffer item is used to retrieve client specific data from a window being updated. The syntax is:
GetBuffer,n,tempfile
where n is the buffer number, which must be between 0 and 15 inclusive; and tempfile is the name of the
temporary file for the window being updated. The tempfile name is passed to the client when ZEMAX calls the
client; see the discussion "How ZEMAX calls the client" for details. Note each window may have it's own buffer
data, and ZEMAX uses the filename to identify the window for which the buffer contents are required. See also
SetBuffer.

GetComment
This item extracts the surface comment, if any. The surface for which the comment data is returned is
appended to the item name, for example, to get the comment for surface 5, the item name should be
"GetComment,5".
GetConfig
This item extracts the current configuration number, the number of configurations, and the number of multiple
configuration operands.
"currentconfig, numberconfig, numbermcoper"
See also SetConfig.
GetDate
This item extracts the current date and time, in the format selected by the user in the ZEMAX environment
dialog box.
GetExtra
This item extracts the extra surface data. The syntax is
GetExtra, surf, column
"value"
See also SetExtra.
GetField
This item extracts field data. The syntax is
GetField, n
where n is zero or the field number. If n is zero, the returned string is formatted as follows:
"type, number"
The parameter type is an integer; either 0, 1, or 2, for angles in degrees, object height, or paraxial image height,
respectively. The parameter number is the number of fields currently defined. If n is not zero, but corresponds to
a valid field number, the returned string is formatted as follows:
"xfield, yfield, weight, vdx, vdy, vcx, vcy, van"
These 8 values are exponential format floating point numbers for the various field data. See also SetField.
GetFile
This item extracts the full name of the lens, including the drive and path. Extreme caution should be used if
the file is to be tampered with; since at any time ZEMAX may read or write from/to this file.
GetFirst
This item extracts first order data about the lens. The returned string is formatted as follows:
"focal, pwfn, rwfn, pima, pmag"
Values in the string are the EFL, paraxial working F/#, real working F/#, paraxial image height, and paraxial
magnification.
GetGlass
This item extracts some data about the glass on any surface. The syntax is
GetGlass,surf
If the specified surface is not valid, is not made of glass, or is gradient index, the returned string is empty. This
data may be meaningless for glasses defined only outside of the FdC band. Otherwise, the returned string is
formatted as follows:

"name, nd, vd, dpgf"
GetGlobalMatrix
This item returns the matrix required to convert any local coordinates (such as from a ray trace) into global
coordinates. For details on the global coordinate matrix, see Global Coordinate Reference Surface on page 93.
The syntax for this item is:
GetGlobalMatrix,surf
The returned data string is in the format
R11,R12,R13,R21,R22,R23,R31,R32,R33,Xo,Yo,Zo
GetIndex
This item extracts index of refraction data for any surface. The syntax is
GetIndex,surf
If the specified surface is not valid, or is gradient index, the returned string is empty. Otherwise, the returned
string is formatted as follows:
"n1, n2, n3..."
where the index values correspond to the index at each defined wavelength, in order.
GetLabel
This item retrieves the integer label associated with the specified surface. The syntax is
GetLabel,surf
The returned value is the label. See also SetLabel and FindLabel.
GetMetaFile
This item creates a Windows Metafile of any ZEMAX graphical analysis plot. The syntax is
GetMetaFile, "metafilename", type, "settingsfilename", flag
The metafilename must be in quotes and include the full path, name, and extension for the metafile.
recognized, a 3D Layout plot will be generated.
If a valid file name is used for the "settingsfilename" argument, ZEMAX will use or save the settings used to
compute the metafile graphic, depending upon the value of the flag parameter.
the requested feature will be displayed. After the user makes any changes to the settings the graphic will then be
lens currently in the Lens Data Editor, so the item PushLens must be used prior to using flag = 2.

Only graphic, and not text files, are supported by GetMetaFile.
See also GetTextFile and OpenWindow.
GetMulticon
This item extracts data from the multi-configuration editor. The syntax is
GetMulticon, config, row
If the config number is greater than zero and less than or equal to the number of configurations, the returned
string is formatted as follows:
"value, num_config, num_row, status, pickuprow, pickupconfig, scale, offset"
If the config number is zero, the multicon operand type data is returned as follows:
"operand_type, number1, number2, number3"
The numbers are used as described in Summary of multi-configuration operands on page 471. The status
integer is 0 for fixed, 1 for variable, 2 for pickup, and 3 for thermal pickup. If status is 2 or 3, the pickuprow and
pickupconfig values indicate the source data for the pickup solve.
See also SetMulticon.
GetName
This item extracts the name of the lens. The returned string is the current lens name as entered on the General
data dialog box.
GetNSCData
This item gets the various data for NSC groups. The syntax is:
GetNSCData, surf, code
Surf refers to the surface number of the NSC group, use 1 if the program mode is Non-Sequential. Currently
only code = 0 is supported, in which case the returned data is the number of objects in the NSC group.
GetNSCMatrix
This item returns the rotation and position matrices for NSC objects relative to the NSC surface origin.
The syntax for this item is:
GetNSCMatrix,surf, object
The returned data string is in the format
R11,R12,R13,R21,R22,R23,R31,R32,R33,Xo,Yo,Zo
GetNSCObjectData
This item gets the various data for NSC objects. The syntax is:
GetNSCData, surf, object, code
Surf and object refer to the surface number and object number. The code is one of the integer values in the
table below; the returned value is the data item indicated. See also SetNSCObjectData.
Code Date returned by GetNSCObjectData

0 Object type name. (string)
1 Comment, which also defines the file name if the object is defined by a file. (string)
2 Color. (integer)
3 1 if object uses a user defined aperture file, 0 otherwise. (integer)
4 User defined aperture file name, if any. (string)
5 Reference object number. (integer)

Code Date returned by GetNSCObjectData
6 Inside of object number. (integer)
7 and up Reserved for future expansion of this feature.
GetNSCParameter
This item gets the parameter data for NSC objects. The syntax is:
GetNSCParameter, surf, object, parameter
Surf and object refer to the surface number and object number, and parameter is the integer parameter
number. The returned string is the parameter value. See also SetNSCParameter.
GetNSCPosition
This item gets the position data for NSC objects. The syntax is:
GetNSCPosition, surf, object
Surf and object refer to the surface number and object number. The returned string is
x, y, z, tilt-x, tilt-y, tilt-z, material
See also SetNSCPosition.
GetNSCSettings
This item extracts the maximum number of intersections, segments, nesting level, minimum absolute intensity,
minimum relative intensity, glue distance, miss ray distance, and ignore errors flag used for NSC ray tracing. The
data is returned in a string as follows:
Max_Intersections, Max_Segments, Max_Nesting, Min_Absolute_Intensity, Min_Relative_Intensity,
Glue_Distance, Miss_Ray_Length, Ignore_Errors
Ignore_Errors is 1 if true, 0 if false. See also SetNSCSettings.
GetOperand
This item gets the operand data from the Merit Function Editor. The syntax is:
GetOperand, row, column
Row is the operand row number in the Merit Function Editor. Column is 1 for type, 2 for int1, 3 for int2, 4-7 for
hx-py, 8 for target, 9 for weight, 10 for value, and 11 for percent contribution. The returned string is the numeric
value of the requested parameter. To update the merit function prior to calling GetOperand, use the Optimize item
with the number of cycles set to -1. See also SetOperand and Optimize.
GetPath
This item extracts the full path name to the directory where ZEMAX is installed, and the path name to the
default directory for lenses. The returned strings are separated by a comma.
GetPolState
This item extracts the default polarization state set by the user. The data is formatted as follows:
"nIsPolarized, Ex, Ey, Phax, Phay"
If nIsPolarized is anything other than zero, then the default polarization state is unpolarized. Otherwise, the
Ex, Ey, Phax, and Phay values are used to define the polarization state. Ex and Ey should each be normalized
to a magnitude of unity, although this is not required. Phax and Phay are in degrees. See also SetPolState.
GetPupil
This item extracts pupil data. The returned string is formatted as follows:
"type, value, ENPD, ENPP, EXPD, EXPP, apodization_type, apodization_factor"
The parameter type is an integer indicating the system aperture type, a number between 0 and 5, for entrance
pupil diameter, image space F/#, object space NA, float by stop size, paraxial working F/#, or object cone angle,

respectively. The value is the system aperture value, unless float by stop size is being used, in which case the
value is the stop surface semi-diameter. The next 4 values are the entrance pupil diameter, entrance pupil
position, exit pupil diameter, and exit pupil position, all in lens units. The apodization_type is an integer which is
set to 0 for none, 1 for Gaussian, 2 for Tangential. The apodization_factor is the number shown on the general
data dialog box.
GetPolTrace
This item is very similar to GetTrace, with the added capability to trace polarized rays through the system. See
the GetTrace documentation for details. The syntax for the GetPolTrace item is as follows:
GetPolTrace, wave, mode, surf, hx, hy, px, py, Ex, Ey, Phax, Phay
The arguments are identical to GetTrace, except for the additional Ex, Ey, Phax, and Phay arguments. Ex and
Ey are the normalized electric field magnitudes in the x and y directions. The quantity Ex*Ex + Ey*Ey should have
a value of 1.0 (with an important exception described below) although any values are accepted. Phax and Phay
are the relative phase, in degrees.
If Ex, Ey, Phax, and Phay are all zero, and only in this case, then ZEMAX assumes an "unpolarized" ray trace
is required. An unpolarized ray trace actually requires ZEMAX to trace two orthogonal rays and the resulting
transmitted intensity be averaged. If any of the four values are not zero, then a single polarized ray will be traced.
For example, to trace the real unpolarized marginal ray to the image surface at wavelength 2, the item string
would be
GetPolTrace,2,0,-1,0.0,0.0,0.0,1.0,0,0,0,0
For polarized rays, the data message comes back in the following format:
"error, intensity, Exr, Eyr, Ezr, Exi, Eyi, Ezi"
The integer error will be zero if the ray traced successfully, otherwise it will be a positive or negative number.
If positive, then the ray missed the surface number indicated by error. If negative, then the ray total internal
reflected (TIR) at the surface given by the absolute value of the error number. Always check to verify the ray data
is valid before using the rest of the string!
The intensity will be the transmitted intensity. It is always normalized to an input electric field intensity of unity.
The transmitted intensity accounts for surface, thin film, and bulk absorption effects, but does not consider
whether or not the ray was vignetted. The Ex, Ey, and Ez values are the electric field components, with the r and
i characters denoting the real and imaginary portions.
For unpolarized rays, the data message is:
"error, intensity"
Although GetPolTrace is easy to program and use, there is a significant disadvantage to using GetPolTrace:
only one ray per DDE call is traced. The overhead of passing data via DDE can be large compared to tracing the
ray, and so if a large number of rays are required to be traced, the execution may be relatively slow. For
information on tracing large numbers of rays, see the discussion "Tracing large numbers of rays" elsewhere in
this chapter. See also GetPolTraceDirect.
GetPolTraceDirect
GetPolTraceDirect provides the same direct access to the ZEMAX ray tracing engine that GetTraceDirect
does, and allows for polarized ray tracing as well. See GetTraceDirect for important details on the direct access
version of GetTrace. The syntax for the GetPolTraceDirect item is:
GetPolTraceDirect, wave, mode, startsurf, stopsurf, x, y, z, l, m, n, Ex, Ey, Phax,
Phay
The Ex, Ey, Phax, and Phay parameters are exactly the same as defined for GetPolTrace.
The data message comes back in the same format as described in GetPolTrace.
GetRefresh
This data item causes ZEMAX to copy the lens data from the LDE into the stored copy of the server. The lens
is then updated, which means ZEMAX recomputes all pupil positions, solves, and index data. If the lens can be

updated, ZEMAX returns the string "0", otherwise, it returns "-1". If the GetUpdate returns "-1", no ray tracing can
be performed.
All subsequent commands will now affect or be executed upon the newly copied lens data. The old lens data,
if any, cannot be recovered. See also GetUpdate and PushLens.
GetSag
This item computes the sag of any surface. The item name is formatted as "GetSag,surf,x,y", where surf is the
surface number, and x and y are the coordinates on the surface for which the sag is computed. The returned
string is formatted as "sag, alternatesag". The values for x, y, and the sag are all in lens units.
GetSequence
This item returns the sequence number of the lens in the Server's memory, then the sequence number of the
lens in the LDE, separated by a comma.
GetSerial
This item returns the serial number of the ZEMAX key.
GetSettingsData
This item retrieves the settings data used by a window. The data must have been previously stored by a call
to SetSettingsData. The data may have been stored by a previous execution of the client program. The syntax is:
GetSettingsData, tempfile, number
The tempfile data is the name of the output file passed by ZEMAX to the client. ZEMAX uses this name to
identify for which window the GetSettingsData request is for.
The number is the data number used by the previous SetSettingsData call. Currently, only number = 0 is
supported. This number may be used to expand the feature in the future.
The return value is the string that was saved by a previous SetSettingsData command for the window and
number.
See also How ZEMAX calls the client on page 634 and SetSettingsData.
GetSolve
This item returns data about any solve on any surface. The syntax is GetSolve,surface,code where code is
an integer code indicating which surface parameter the solve data is for. The solve data is returned in the following
formats, depending upon the code value.
GetSolve Code Returned data format

0, curvature solvetype, parameter1, parameter2
1, thickness solvetype, parameter1, parameter2, parameter3
2, glass solvetype (for solvetype = 0)
solvetype, Index, Abbe, Dpgf (for solvetype = 1, model glass)
solvetype, pickupsurf (for solvetype = 2, pickup)
solvetype (for solvetype = all other values)
3, semi-diameter solvetype, pickupsurf
4, conic solvetype, pickupsurf
5-12, parameters 1-8 solvetype, pickupsurf, offset, scalefactor
1001+, extra data values solvetype, pickupsurf, scalefactor
1+

The solvetype is an integer code, and the parameters have meanings that depend upon the solve type; see
the chapter SOLVES on page 379 for details. See also SetSolve.
GetSurfaceData
This item gets surface data. The syntax is:
GetSurfaceData, surf, code
Surf refers to the surface number. The code is one of the integer values in the table below; the returned value
is the data item indicated. See also SetSurfaceData.
Code Date returned by GetSurfaceData

0 Surface type name. (string)
1 Comment. (string)
2 Curvature (not radius).
3 Thickness.
4 Glass. (string)
5 Semi-Diameter.
6 Conic.
7 Coating. (string)
51 Before tilt and decenter order; 0 for Decenter then Tilt, 1 for Tilt then Decenter.
52-56 Before decenter x, decenter y, tilt x, tilt y, and tilt z values, respectively.
60 After status; 0 for explicit, 1 for pick up this surface, 2 for reverse this surface, 3 for pick up this
surface - 1, 4 for reverse this surface - 1, etc.
61 After tilt and decenter order; 0 for Decenter then Tilt, 1 for Tilt then Decenter.
62-66 After decenter x, decenter y, tilt x, tilt y, and tilt z values, respectively.
For example, to get the conic data for surface 8, the item name should be "GetSurfaceData,8,6".
See also SetSurfaceData and GetSurfaceParameter.
GetSurfaceDLL
This item extracts the name of the DLL if the surface is a user defined type. The surface for which the surface
DLL name is returned is appended to the item name, for example, to get the DLL name for a user defined surface
5, the item name should be "GetSurfaceDLL,5".
"dllname, surfacename"
The dllname is the name of the defining DLL, the surfacename is the string displayed by the DLL in the surface
type column of the Lens Data Editor.
GetSurfaceParameter
This item extracts surface parameter data. The surface and parameter number to be returned is appended to
the item name, for example, to get the parameter 5 data for surface 8, the item name should be "GetSurfacePa-
rameter,8,5".
See also GetSurfaceData and SetSurfaceParameter.
GetSystem
This item extracts system data.

"numsurfs, unitcode, stopsurf, nonaxialflag, rayaimingtype, useenvdata, temp, pressure, globalrefsurf"
This item returns the number of surfaces, the unit code (0, 1, 2, or 3 for mm, cm, in, or M), the stop surf number,
a flag to indicate if the system is non-axial symmetric (0 for false, that is it is axial, or 1 if the system is not axial),
the ray aiming (0 for off and 1 for on), the use environment data flag (0 for no, 1 for yes), the current temperature
and pressure, and the global coordinate reference surface number.
See also SetSystem and GetSystemAper.
GetSystemAper
This item extracts system aperture data.
"type, stopsurf, aperture_value"
This item returns the system aperture type (0, 1, 2, 3, 4, or 5 for Entrance Pupil Diameter, Image Space F/#,
Object Space NA, Float by Stop Size, Paraxial Working F/#, or Object Cone Angle, respectively), the stop surface
number, and the system aperture value. If the aperture type is Float by Stop Size, the aperture value is the stop
surface semi-diameter.
See also GetSystem and SetSystemAper.
GetTextFile
This item creates an ASCII text of any ZEMAX analysis that supports text. The syntax is
GetTextFile, "textfilename", type, "settingsfilename", flag
The textfilename must be in quotes and include the full path, name, and extension for the file to be created.
recognized, a standard ray trace will be generated.
If a valid file name is used for the "settingsfilename" argument, ZEMAX will use or save the settings used to
compute the text file, depending upon the value of the flag parameter.
the requested feature will be displayed. After the user makes any changes to the settings the file will then be
lens currently in the Lens Data Editor, so the item PushLens must be used prior to using flag = 2.
Only text, and not graphic files, are supported by GetTextFile.
See also GetMetaFile and OpenWindow.
GetTol
This item extracts tolerance data. The syntax is

GetTol, n
where n is zero or the tolerance operand number. If n is zero, the returned string is formatted as follows:
"number"
where number is the number of tolerance operands defined. If n is not zero, but corresponds to a valid
tolerance operand number, the returned string is formatted as follows:
"toltype, int1, int2, min, max"
GetTrace
This item requires that the client provide additional data. In order to trace a ray, ZEMAX needs to know the
relative field and pupil coordinates, the wavelength, the mode (either real, mode = 0 or paraxial, mode = 1) as
well as the surface to trace the ray to. All of this data is encoded and appended to the item name. This is
accomplished by delimiting the various parameters with commas as follows:
GetTrace, wave, mode, surf, hx, hy, px, py
For example, to trace the real chief ray to surface 5 at wavelength 3, the item string would be
GetTrace,3,0,5,0.0,1.0,0.0,0.0
Although this looks cumbersome, it is easy to program. ZEMAX receives the item, recognizes the "GetTrace"
at the beginning of the name, strips off the rest of the string, then parses it, and traces the ray. Usually, the ray
data is only needed at the image surface; setting the surface number to -1 will yield data at the image surface.
Note 0 is reserved for the object surface.
The data message comes back in the following format:
"error, vigcode, x, y, z, l, m, n, l2, m2, n2, intensity"
The integer error will be zero if the ray traced successfully, otherwise it will be a positive or negative number.
If positive, then the ray missed the surface number indicated by error. If negative, then the ray total internal
reflected (TIR) at the surface given by the absolute value of the error number. Always check to verify the ray data
is valid before using the rest of the string!
The parameter vigcode is the first surface where the ray was vignetted. Unless an error occurs at that surface
or subsequent to that surface, the ray will continue to trace to the requested surface.
The x, y, and z values are the coordinates on the requested surface.
The l, m, and n values are the direction cosines after refraction into the media following the requested surface.
The l2, m2, and n2 values are the surface intercept direction normals at the requested surface.
The intensity is the relative transmitted intensity of the ray, including any pupil or surface apodization defined.
Although GetTrace is easy to program and use, there is a significant disadvantage to using GetTrace: only
one ray per DDE call is traced. The overhead of passing data via DDE can be quite large compared to tracing the
ray, and so if a large number of rays are required to be traced, the execution may be relatively slow. For
information on tracing large numbers of rays, see the discussion "Tracing large numbers of rays" elsewhere in
this chapter. See also GetTraceDirect.
GetTraceDirect
GetTraceDirect provides a more direct access to the ZEMAX ray tracing engine. Normally, rays are defined
by the normalized field and pupil coordinates hx, hy, px, and py. ZEMAX takes these normalized coordinates and
computes the object coordinates (x, y, and z) and the direction cosines to the entrance pupil aim point (l, m, and
n; for the x-, y-, and z-direction cosines, respectively).
However, there are times when it is more appropriate to trace rays by direct specification of x, y, z, l, m, and
n. The direct specification has the added flexibility of defining the starting surface for the ray anywhere in the
optical system. This flexibility comes at the price of requiring the client program to carefully ensure that the starting
ray coordinates are meaningful.
Like GetTrace, this item requires that the client provide additional data. In order to trace a ray, ZEMAX needs
to know x, y, z, l, m, n, the wavelength, the mode (either real, mode = 0 or paraxial, mode = 1) as well as the
starting and stopping surfaces to trace the ray from and to. All of this data is encoded and appended to the item
name. This is accomplished by delimiting the various parameters with commas as follows:

GetTraceDirect, wave, mode, startsurf, stopsurf, x, y, z, l, m, n
The data message comes back in the following format:
"error, vigcode, x, y, z, l, m, n, l2, m2, n2, intensity"
where the parameters are exactly the same as for GetTrace, except for intensity.The intensity is the relative
transmitted intensity of the ray, excluding any pupil apodization defined. Note GetTrace includes pupil apodiza-
tion, GetTraceDirect does not. Both include surface apodizations.
Although GetTraceDirect is easy to program and use, there is a significant disadvantage to using GetTrace-
Direct: only one ray per DDE call is traced. The overhead of passing data via DDE can be quite large compared
to tracing the ray, and so if a large number of rays are required to be traced, the execution may be relatively slow.
For information on tracing large numbers of rays, see the discussion "Tracing large numbers of rays" elsewhere
in this chapter. See also GetTrace.
GetUDOSystem
This item is used to load a particular lens from the optimization function memory into the ZEMAX server's
memory. The only time this item name should be used is when implementing a User Defined Operand, or UDO.
UDO's are described in detail in the chapter "Optimization". See that chapter for details. The syntax for
GetUDOSystem is
GetUDOSystem, buffercode
where buffercode is an integer passed to the UDO on the command line. See also SetUDOData.
GetUpdate
This data item causes ZEMAX to update the lens, which means ZEMAX recomputes all pupil positions, solves,
and index data. If the lens can be updated, ZEMAX returns the string "0", otherwise, it returns "-1". If the
GetUpdate returns "-1", no ray tracing can be performed. To update the merit function, use the Optimize item with
the number of cycles set to -1. See also GetRefresh and Optimize.
GetVersion
This item returns the version number of ZEMAX. This number is generally a 5 digit integer with no decimal
place, such as 10000 for version 10.0. Later versions will have larger numbers.
GetWave
This item extracts wavelength data. The syntax is
GetWave, n
where n is zero or the wavelength number. If n is zero, the returned string is formatted as follows:
"primary, number"
The parameter primary is an integer which indicates which of the wavelengths is the primary one. The
parameter number is the number of wavelengths currently defined. If n is not zero, but corresponds to a valid
wavelength number, the returned string is formatted as follows:
"wavelength, weight"
These 2 values are exponential format floating point numbers which correspond to the value and weight for
the specified wavelength. See also SetWave.
ImportExtraData
This item imports extra data and grid surface data values into an existing surface. The syntax is:
ImportExtraData,surf,filename
For details on how ZEMAX imports extra data, see Extra Data on page 79. For information on importing grid
surface data, see Importing grid data on page 259.
InsertConfig
This item inserts a new configuration in the multi-configuration editor. The syntax is:
InsertConfig, config

The new configuration will be placed at the location indicated by the parameter config. See also DeleteConfig.
InsertObject
This item inserts a new NSC object. The syntax is:
InsertObject,surf,object
The new null object will be placed at the location indicated by the parameters surf and object. See also
SetNSCObjectData to define data for the new surface and the DeleteObject item.
InsertSurface
This item inserts a new surface. The syntax is:
InsertSurface,surf
The new surface will be placed at the location indicated by the parameter surf. See also SetSurface to define
data for the new surface and the DeleteSurface item.
LoadFile
Loads a ZEMAX file into the server. Note that loading a file does not change the data displayed in the LDE;
the server process has a separate copy of the lens data. The file name to be loaded is appended to the LoadFile
item name, and must include the full path. For example: "LoadFile,C:\ZEMAX\SAMPLES\COOKE.ZMX". The
returned string is the same as for the GetUpdate item; after updating the newly loaded lens file. If a value other
than 0 is returned, the Update failed, if -999 is returned, the file could not be loaded. See also GetPath, SaveFile,
and PushLens.
LoadMerit
Loads a ZEMAX .MF or .ZMX file and extracts the merit function and places it in the lens loaded in the server.
Note that loading a merit function file does not change the data displayed in the LDE; the server process has a
separate copy of the lens data. The file name to be loaded is appended to the LoadMerit item name, and must
include the full path. For example: "LoadMerit,C:\ZEMAX\SAMPLES\MyMerit.MF". The returned string is format-
ted as follows:
"number, merit"
where number is the number of operands in the merit function and merit is the value of the merit function. If
the merit function value is 9.00e+009, the merit function cannot be evaluated.
See also Optimize.
MakeGraphicWindow
This item notifies ZEMAX that graphic data has been written to a file and may now be displayed as a ZEMAX
child window. The primary purpose of this item is to implement user defined features in a client application, that
look and act like native ZEMAX features. The item string must be formatted as follows:
MakeGraphicWindow, filename, modulename, wintitle, textflag, settingsdata
or
SetSettingsData, 0, settingsdata
MakeGraphicWindow, filename, modulename, wintitle, textflag
The filename is the full path and file name to the temporary file that holds the graphic data. This must be the
same name as passed to the client executable in the command line arguments, if any. The modulename is the
full path and executable name of the client program that created the graphic data. The wintitle is the string which
defines the title ZEMAX should place in the top bar of the window.
The textflag should be 1 if the client can also generate a text version of the data. Since the current data is a
graphic display (it must be if the item is MakeGraphicWindow) ZEMAX wants to know if the "Text" menu option
should be available to the user, or if it should be grayed out. If the text flag is 0, ZEMAX will gray out the "Text"
menu option and will not attempt to ask the client to generate a text version of the data.
The settings data is a string of values delimited by spaces (not commas) which are used by the client to define
how the data was generated. These values are only used by the client, not by ZEMAX. The settings data string

holds the options and data that would normally appear in a ZEMAX "settings" style dialog box. The settings data
should be used to recreate the data if required. Because the total length of a data item cannot exceed 255
characters, the data item SetSettingsData may be used prior to the call to MakeGraphicWindow to specify the
settings data string rather than including the data as part of MakeGraphicWindow. See How ZEMAX calls the
client on page 634 for more details on the settings data.
A sample item string might look like this:
MakeGraphicWindow,C:\TEMP\ZGF001.TMP,C:\ZEMAX\FEATURES\CLIENT.EXE,ClientWindow,1,
0 1 2 12.55
This item indicates that ZEMAX should open a graphic window, display the data stored in the file
C:\TEMP\ZGF001.TMP, and that any updates or setting changes can be made by calling the client module
C:\ZEMAX\FEATURES\CLIENT.EXE. This client can generate a text version of the graphic, and the settings data
string (used only by the client) is "0 1 2 12.55".
MakeTextWindow
This item notifies ZEMAX that text data has been written to a file and may now be displayed as a ZEMAX child
window. The primary purpose of this item is to implement user defined features in a client application, that look
and act like native ZEMAX features. The item string must be formatted as follows:
MakeTextWindow, filename, modulename, wintitle, settingsdata
or
SetSettingsData, 0, settingsdata
MakeTextWindow, filename, modulename, wintitle
The filename is the full path and file name to the temporary file that holds the text data. This must be the same
name as passed to the client executable in the command line arguments, if any. The modulename is the full path
and executable name of the client program that created the text data. The wintitle is the string which defines the
title ZEMAX should place in the top bar of the window.
The settingsdata is a string of values delimited by spaces (not commas) which are used by the client to define
how the data was generated. These values are only used by the client, not by ZEMAX. The settingsdata string
holds the options and data that would normally appear in a ZEMAX "settings" style dialog box. The settingsdata
should be used to recreate the data if required. Because the total length of a data item cannot exceed 255
characters, the data item SetSettingsData may be used prior to the call to MakeTextWindow to specify the settings
data string rather than including the data as part of MakeTextWindow. See How ZEMAX calls the client on
page 634 for more details on the settings data.
A sample item string might look like this:
MakeTextWindow,C:\TEMP\ZGF002.TMP,C:\ZEMAX\FEATURES\CLIENT.EXE,ClientWindow,6 5 4
12.55
This item indicates that ZEMAX should open a text window, display the data stored in the file
C:\TEMP\ZGF002.TMP, and that any updates or setting changes can be made by calling the client module
C:\ZEMAX\FEATURES\CLIENT.EXE. The settingsdata string (used only by the client) is "6 5 4 12.55".
NewLens
This item erases the current lens. The minimum lens that remains is identical to the lens in the Lens Data
Editor when File, New is selected. No prompt to save the existing lens is given.
OpenWindow
OpenWindow will open a new analysis window on the main ZEMAX screen. The syntax is
OpenWindow,Type
"Buttons" tab of the File, Preferences dialog box. The labels are case sensitive. See also GetMetaFile.

Optimize
Optimize calls the ZEMAX Damped Least Squares optimizer. The syntax is
Optimize,n
where n is the number of cycles to run. The return value is the final merit function. If the merit function value
returned is 9.0E+009, the optimization failed, usually because the lens or merit function could not be evaluated.
If n is zero, the optimization runs in automatic mode. If n is less than zero (for example, n = -1), Optimize updates
all operands in the merit function and returns the current merit function, and no optimization is performed.
PushLens
The syntax is:
PushLens, flag
PushLens will take the lens currently loaded in the server's memory and push it into the Lens Data Editor. This
operation requires the permission of the user running the ZEMAX program. To check or set this permission status,
see the PushLensPermission data item described below.
The returned string is the same as for the GetUpdate item; after updating the newly pushed lens file. If a value
other than 0 is returned, the Update failed, if -999 is returned, the lens could not be pushed into the LDE.
If the flag value is zero or is omitted, the open windows are not updated. If the flag value is 1, then all open
analysis windows are updated.
See also PushLensPermission, GetPath, GetRefresh, LoadFile, and SaveFile.
PushLensPermission
The syntax is:
PushLensPermission
The return value is 1 if ZEMAX is set to accept PushLens commands, or 0 if extensions are not allowed to use
PushLens. The PushLens data item can only succeed if the user has set the appropriate permissions in ZEMAX.
To allow a DDE extension to use PushLens, the option "Allow Extensions To Push Lenses" must be checked.
This option is found under File, Preferences on the Editors tab, and is described in Editors on page 61. The
proper use of PushLens is to first call PushLensPermission. If the return value is 0, then the client should display
a dialog box instructing the user to turn on the "Allow Extensions To Push Lenses" option before proceeding.
ReleaseWindow
When ZEMAX calls the client to update or change the settings used by the client function, the menu bar is
grayed out on the window to prevent multiple updates or setting changes from being requested simultaneously.
Normally, when the client code sends the MakeTextWindow or MakeGraphicWindow, the menu bar is once again
activated. However, if during an update or setting change, the new data cannot be computed, then the window
must be released. The ReleaseWindow serves just this one purpose. If the user selects "Cancel" when changing
the settings, the client code should send a ReleaseWindow item to release the lock out of the menu bar. If this
command is not sent, the window cannot be closed, which will prevent ZEMAX from terminating normally. The
ReleaseWindow item takes just one argument: the name of the temporary file. The syntax is:
ReleaseWindow, filename
The return value is zero if no window is using the filename, or a positive integer number if the file is being used.
SaveFile
Saves the lens currently loaded in the server to a ZEMAX file. The file name to be used for the save is
appended to the SaveFile item name, and must include the full path. For example: "SaveFile,C:\ZEMAX\SAM-
PLES\COOKE.ZMX". The returned string is the same as for the GetUpdate item; after updating the newly saved
lens file. If a value other than 0 is returned, the Update failed, if -999 is returned, the file could not be saved. See
also GetPath, GetRefresh, LoadFile, and PushLens.

SetAperture
This item sets surface aperture data. The syntax is
SetAperture, surf, type, min, max, xdecenter, ydecenter, aperturefile
"type, min, max, xdecenter, ydecenter, aperturefile"
This item uses an integer code for the surface aperture type; 0 for no aperture, 1 for circular aperture, 2 for
circular obscuration, 3 for spider, 4 for rectangular aperture, 5 for rectangular obscuration, 6 for elliptical aperture,
7 for elliptical obscuration, 8 for user defined aperture, 9 for user defined obscuration, and 10 for floating aperture.
The min and max values have different meanings for the elliptical, rectangular, and spider apertures than for
circular apertures; see Aperture type and other aperture controls on page 68 for details.
If SetAperture is used to set user defined apertures or obscurations, the aperturefile must be the name of a
file which lists the x, y, coordinates of the user defined aperture file in a two column format. For more information
on user defined apertures, see User defined apertures and obscurations on page 69.
See also GetAperture.
SetBuffer
The SetBuffer item is used to store client specific data with the window being created or updated. The buffer
data can be used to store user selected options, instead of using the settings data on the command line of the
MakeTextWindow or MakeGraphicWindow items. The data must be in a string format. The syntax is:
SetBuffer,1,any text you want......
There are 16 buffers provided, numbered 0 through 15, and each can be set using SetBuffer,0,.....; Set-
Buffer,1,.....etc. The text following the "SetBuffer,n," is the only text that is stored; and it may be a maximum of
240 characters.
Note the buffer data is not associated with any particular window until either the MakeTextWindow or
MakeGraphicWindow items are issued. Once ZEMAX receives the MakeTextWindow or MakeGraphicWindow
item, the buffer data is then copied to the appropriate window memory, and then may later be retrieved from that
window's buffer using GetBuffer. See also GetBuffer.
SetConfig
This item switches the current configuration number and updates the system. The desired configuration is
appended to the item name. For example, to switch to configuration 3, the item name is "SetConfig,3".
"currentconfig, numberconfig, error"
The currentconfig is the new configuration number, which will be between 1 and the value of numberconfig.
Normally, this will be the desired configuration requested in the item name, as long as it was a valid configuration
number. The error code is the same as returned by the GetUpdate item, and will be zero if the new current
configuration is traceable. See also GetConfig.
SetExtra
This item sets the extra surface data. The syntax is
SetExtra, surf, column, value
"value"
See also GetExtra.
SetField
This item sets field data. The syntax is
SetField, 0, type, number
or

SetField, n, xf, yf, wgt, vdx, vdy, vcx, vcy, van
If the value for n is zero, then the field type and the total number of fields is set to the new integer values. If n
is a valid field number (between 1 and the number of fields, inclusive) then the field x and y values, field weight,
and vignetting factors are all set. See also GetField. The returned data is the same as for GetField, n.
SetFloat
This item sets all surfaces without surface apertures to have floating apertures. Floating apertures will vignette
rays which trace beyond the semi-diameter. The syntax is
SetFloat
The returned value is "OK".
SetLabel
This item associates an integer label with the specified surface. The label will be retained by ZEMAX as
surfaces are inserted or deleted around the target surface. The syntax is
SetLabel, surf, label
The returned value is the label. See also GetLabel and FindLabel.
SetMulticon
This item sets data in the multi-configuration editor. To set multi-configuration data, the syntax is
SetMulticon, config, row, value, status, pickuprow, pickupconfig, scale, offset
If the config number is zero, SetMulticon may be used to set the operand type and number data using this
syntax:
SetMulticon, 0, row, operand_type, number1, number2, number3
The returned string is formatted as defined for "GetMulticon". The numbers are used as described in Summary
of multi-configuration operands on page 471.
See also GetMulticon.
SetNSCObjectData
This item sets the various data for NSC objects. The syntax is:
SetNSCData, surf, object, code, data
Surf and object refer to the surface number and object number. The code is an one of the integer values defined
in the table in the discussion for GetNSCObjectData on page 613. The returned value is the data item indicated.
SetNSCPosition
This item sets the position data for NSC objects. The syntax is:
SetNSCPosition, surf, object, code, data
Surf and object refer to the surface number and object number. Code is 1-7 for x, y, z, tilt-x, tilt-y, tilt-z, and
material, respectively. Data is either a number (for codes 1-6) or a string with the material name (for code 7). The
returned value is that of GetNSCPosition after the new data has been applied. See also GetNSCPosition.
SetNSCParameter
This item sets the parameter data for NSC objects. The syntax is:
SetNSCParameter, surf, object, parameter, data
Surf and object refer to the surface number and object number. Parameter is the integer parameter number,
and data is the new numeric value for that parameter. The returned value is that of GetNSCParameter after the
new data has been applied. See also GetNSCParameter.
SetNSCSettings
This item sets the maximum number of intersections, segments, nesting level, minimum absolute intensity,
minimum relative intensity, glue distance, miss ray distance, and ignore errors flag used for NSC ray tracing. The
syntax is:

SetNSCSettings, MaxInt, MaxSeg, MaxNest, MinAbsI, MinRelI, GlueDist, MissRayDist,
IgnoreErrors
The returned value is identical to that of GetNSCSettings. Since the MaxSeg value may require large amounts
of RAM, verify that the new value was accepted by checking the retuned string value.
See also GetNSCSettings.
SetOperand
This item sets the operand data in the Merit Function Editor. The syntax is:
SetOperand, row, column, value
Row is the operand row number in the Merit Function Editor. Column is 1 for type, 2 for int1, 3 for int2, 4-7 for
hx-py, 8 for target, and 9 for weight. The returned string is the same as for GetOperand. To set the operand type,
the value can be the name of the operand, such as EFFL.
To update the merit function after calling SetOperand, use the Optimize item with the number of cycles set to
-1. See also GetOperand and Optimize.
SetPolState
This item sets the default polarization state set by the user. The data is formatted as follows:
SetPolState, nIsPolarized, Ex, Ey, Phax, Phay
See GetPolState for a complete description.
SetSettingsData
This item sets the settings data used by a window in temporary storage before calling MakeGraphicWindow
or MakeTextWindow. The data may be retrieved using GetSettingsData. The syntax is:
SetSettingsData, number, data
Currently, only number = 0 is supported. This number may be used to expand the feature in the future.
See also How ZEMAX calls the client on page 634 and GetSettingsData.
SetSolve
This item sets solve data and returns data about the new solve on any surface. The syntax is:
SetSolve,surface,code,data...
where the code is as defined in the discussion of GetSolve and the data... are the exact parameters as
defined for GetSolve. The parameter list depends upon the code listed. The return string is the same as GetSolve
returns for the newly defined solve data.
SetSurfaceData
This item sets surface data. The item name is formatted as follows:
SetSurfaceData, surf, datatype, data
Where the values for datatype are the same as documented for the GetSurfaceData item. The return string is
the newly set surface data formatted as documented for the GetSurfaceData item. See also GetSurfaceData and
SetSurfaceParameter.
SetSurfaceParameter
This item sets surface parameter data. The item name is formatted as follows:
SetSurfaceParameter, surf, parm, newvalue
where surf is the surface number and parm is the parameter number. The return value is identical to
GetSurfaceParameter. See also SetSurfaceData and GetSurfaceParameter.
SetSystem
This item sets system data. The syntax is
SetSystem, unitcode, stopsurf, rayaimingtype, useenvdata, temp, pressure,
globalrefsurf

The returned string is formatted as defined for GetSystem. See also GetSystem for details on the arguments.
SetSystemAper
This item sets system aperture data. The syntax is
SetSystemAper, type, stopsurf, aperture_value
The returned string is formatted as defined for GetSystemAper. See GetSystemAper for details on the
arguments.
SetUDOData
This item is used to pass data computed by the client program to the ZEMAX optimizer. The only time this item
name should be used is when implementing a User Defined Operand, or UDO. UDO's are described in detail in
the chapter "Optimization". See that chapter for details. The syntax for SetUDOSystem is:
SetUDOData, buffercode, data0, data1, data2,...data50
where buffercode is an integer passed to the UDO on the command line. When the ZEMAX server receives
the SetUDOData item, it places the data values in a buffer, then closes the buffer for further input. This item allows
multiple items to be passed at once to the server, however, there is a limit of 255 characters in the input line. To
circumvent this limitation (of Windows) use the SetUDOItem and Close UDOData items. See also GetUDOSys-
tem.
SetUDOItem
This item is used to pass just one datum computed by the client program to the ZEMAX optimizer. The only
time this item name should be used is when implementing a User Defined Operand, or UDO. UDO's are described
in detail in the chapter "Optimization". See that chapter for details. The syntax for SetUDOItem is
SetUDOItem, buffercode, datanumber, data
where buffercode is an integer passed to the UDO on the command line and data is the data item number
being passed. Using this item, the servers buffer may be filled one item at a time, which largely circumvents the
255 character limit on item names. However, the ZEMAX server does not know if any more data is coming, and
so the buffer remains open. After the last data item has been sent, the buffer must be closed using the
CloseUDOData item before the optimization may proceed. A typical implementation may consist of the following
series of item messages:
SetUDOItem, buffercode, 0, value0
CloseUDOData, buffercode
See also GetUDOSystem.
SetVig
This item sets the vignetting factors. For more information, see Vignetting factors on page 96. No arguments
are required. SetVig always returns "OK" as the return string.
SetWave
This item sets wavelength data. The syntax is
SetWave, 0, primary, number
or
SetWave, n, wavelength, weight
If the value for n is zero, then the primary wavelength number and the total number of wavelengths is set to
the new integer values. If n is a valid wavelength number (between 1 and the number of wavelengths, inclusive)
then the wavelength in micrometers and the wavelength weight are both set. See also GetWave. The returned
data is the same as for GetWave, n.
WindowMaximize
This item will maximize the main ZEMAX window or any analysis window ZEMAX currently displays. The
syntax is

WindowMaximize, winnum
Use 0 for winnum to maximize the main ZEMAX window, otherwise, the integer window number.
WindowMinimize
This item will minimize the main ZEMAX window to an icon. See WindowMaximize for details.
WindowRestore
This item will restore the main ZEMAX window to its previous size and position. See WindowMaximize for
details.
Tracing large numbers of rays
If only a few rays need to be traced, then it is easy and fast enough to use DDE items such as GetTrace or
GetPolTrace. However, if more than perhaps 100-500 rays need to be traced, it is far faster to use the array ray
tracing technique.
Instead of tracing the rays one at a time, an array containing a list of all the rays to be traced is created, and
then the entire array passed to ZEMAX at one time. ZEMAX then traces all the rays, and passes the entire array
back to the client.
This method mimics the behavior of GetTrace, GetTraceDirect, GetPolTrace, and GetPolTraceDirect, with the
added power that any number of rays to be traced at one time, although it is slightly more complex to program.
There are 2 steps required:
1) Fill an array of structures with the required data
2) Pass the array to ZEMAX
Step 1: Placing the ray data in the array
The data defining the list of rays to be traced is placed in an array of structures of the following form:
typedef struct
{
double x, y, z, l, m, n, opd, intensity;
double Exr, Exi, Eyr, Eyi, Ezr, Ezi;
int wave, error, vigcode, want_opd;
}DDERAYDATA;
Inside the client program, an array of type DDERAYDATA is created with a declaration such as:
DDERAYDATA MyData[101];
There are four trace modes supported, numbered from 0 to 3, and they correspond to the actions of GetTrace
(mode 0), GetTraceDirect (mode 1), GetPolTrace (mode 2), and GetPolTraceDirect (mode 3). Each mode
requires a slightly different set of data to be placed in the DDERAYDATA structure. The mode is set by the opd
parameter in array position zero.
Each ray requires 1 array element. If there are "n" rays to be traced, then the array structures 1 through n will
be used to hold the data for these n rays. Array position zero is reserved for some header data that needs to be
passed with the array.
Mode 0: Similar to GetTrace
Like GetTrace, the rays are defined by the relative field and pupil coordinates, the wavelength, the mode (either
real, mode = 0 or paraxial, mode = 1) as well as the surface to trace the ray to. Before the rays can be traced,
the array should be modified in array positions 1 through n to contain the definitions for each ray. The field, pupil,
and wavelength data are all stored "in place" in the array as follows:
MyData[i].x = hx;
MyData[i].y = hy;
MyData[i].z = px;
MyData[i].l = py;
MyData[i].m = 0.0; /* ignored on input */
MyData[i].n = 0.0; /* ignored on input */
MyData[i].opd = 0.0; /* ignored on input */
MyData[i].intensity = 1.0; /* initial intensity */
MyData[i].Exr = 0.0; /* ignored */

MyData[i].Exi = 0.0; /* ignored */
MyData[i].Eyr = 0.0; /* ignored */
MyData[i].Eyi = 0.0; /* ignored */
MyData[i].Ezr = 0.0; /* ignored */
MyData[i].Ezi = 0.0; /* ignored */
MyData[i].wave = wavenumber;
MyData[i].error = 0; /* must initially be zero */
MyData[i].vigcode = 0; /* must initially be zero */
MyData[i].want_opd = 0; /* 0 if OPD data is not needed, 1 if it is */
Note the argument i should be between 1 and n, the number of rays. The array may be of any size, however,
it is bad programming practice to trace too many rays at once, because it makes it difficult to accept a user
interrupt if the ray tracing is to be terminated. The initial intensity will be scaled by any pupil apodization or surface
apodization that occurs. Note mode 0 considers pupil apodization, while mode 1 does not.
The array position 0 is reserved for other data that applies to all the rays. The data in array position 0 is used
to tell ZEMAX the mode, the number of rays, and the surface to trace the rays to. The data is placed in the array
as follows:
MyData[0].x = 0.0; /* ignored */
MyData[0].y = 0.0; /* ignored */
MyData[0].z = 0.0; /* ignored */
MyData[0].l = 0.0; /* ignored */
MyData[0].m = 0.0; /* ignored */
MyData[0].n = 0.0; /* ignored */
MyData[0].opd = 0; /* sets mode 0 */
MyData[0].intensity = 0.0; /* ignored */
MyData[0].Exr = 0.0; /* ignored */
MyData[0].Exi = 0.0; /* ignored */
MyData[0].Eyr = 0.0; /* ignored */
MyData[0].Eyi = 0.0; /* ignored */
MyData[0].Ezr = 0.0; /* ignored */
MyData[0].Ezi = 0.0; /* ignored */
MyData[0].wave = mode; /* 0 for real rays, 1 for paraxial rays */
MyData[0].error = numrays; /* the number of rays in the array */
MyData[0].vigcode = 0; /* must initially be zero */
MyData[0].want_opd = lastsurf; /* -1 for image, or any valid surface number */
The number of rays must be placed in the error variable. Note that any declaration for an array of DDERAY-
DATA types must be 1 bigger than the number of rays, since array position 0 is reserved for the data described
above. All rays in the array must be traced to the same surface using the same mode, but they all may have
different wavelengths and field and pupil coordinates.
If the variable want_opd is 0, then the usual x, y, z, l, m, and n data will be computed for the ray. If want_opd
is anything other than zero, then the OPD for the ray will be returned along with the usual data. Computing the
OPD takes additional time beyond that for regular ray tracing, and ZEMAX only performs this additional
computation if requested to do so.
Computing the OPD is also more complicated for the client program, because OPD means optical path
difference; which means two rays must be traced rather than just one. To compute OPD for an arbitrary ray,
ZEMAX must trace the chief ray, then the arbitrary ray, then subtract the phase of the two to get the OPD. Rather
than trace the same chief ray over and over, which is slow, usually the chief ray is traced once, and then the phase
of the chief ray is subtracted from each subsequent ray. ZEMAX uses the sign of the want_opd parameter to
specify which calculation should be done. If want_opd is less than zero (such as -1) then the both the chief ray
and specified ray are requested, and the OPD is the phase difference between the two in waves of the current
wavelength. If want_opd is greater than zero, then the most recently traced chief ray data is used. Therefore, the
want_opd flag should be -1 whenever the chief ray changes; and +1 for all subsequent rays which do not require
the chief ray be traced again. Generally the chief ray changes only when the field coordinates or wavelength
changes. This method is much faster if there are many rays being traced from the same field point, as is the case
for many optical analysis calculations.
The opd can only be computed if the last surface is the image surface, otherwise, the opd value will be zero.

Mode 1: Similar to GetTraceDirect
Like GetTraceDirect, the rays are defined by x, y, z, l, m, and n coordinates on any starting surface, as well as
the wavelength, the mode (either real, mode = 0 or paraxial, mode = 1) and the surface to trace the ray to. Before
the rays can be traced, the array should be modified in array positions 1 through n to contain the definitions for
each ray. The field, pupil, and wavelength data are all stored "in place" in the array as follows:
MyData[i].x = x;
MyData[i].y = y;
MyData[i].z = z;
MyData[i].l = l;
MyData[i].m = m;
MyData[i].n = n;
MyData[i].opd = 0.0; /* ignored on input */
MyData[i].error = 0.0; /* must initially be zero */
MyData[i].vigcode = 0.0; /* must initially be zero */
MyData[i].want_opd = 0; /* ignored */
interrupt if the ray tracing is to be terminated. The initial intensity will be scaled by any surface apodization that
occurs. Note mode 0 considers pupil apodization, while mode 1 does not.
as follows:
MyData[0].x = 0.0; /* ignored */
MyData[0].y = 0.0; /* ignored */
MyData[0].z = 0.0; /* ignored */
MyData[0].l = 0.0; /* ignored */
MyData[0].vigcode = startsurf; /* the surface on which the coordinates start */
The behavior of this mode is otherwise the same as in mode 0. Computation of the OPD is not permitted in
this mode.
Mode 2: Similar to GetPolTrace
This mode is very similar to GetPolTrace. Before the rays can be traced, the array should be modified in array
positions 1 through n to contain the definitions for each ray. The field, pupil, and wavelength data are all stored
"in place" in the array as follows:
MyData[i].x = hx;
MyData[i].y = hy;
MyData[i].z = px;
MyData[i].l = py;

MyData[i].m = 0.0; /* ignored */
MyData[i].n = 0.0; /* ignored */
MyData[i].opd = 0.0; /* ignored */
MyData[i].Exr = Exr /* Electric field X real */
MyData[i].Exi = Exi /* Electric field X imaginary */
MyData[i].Eyr = Eyr /* Electric field Y real */
MyData[i].Eyi = Eyi /* Electric field Y imaginary */
MyData[i].Ezr = Ezr /* Electric field Z real */
MyData[i].Ezi = Ezi /* Electric field Z imaginary */
interrupt if the ray tracing is to be terminated.
The intensity parameter is used to return the relative transmitted intensity of the ray. If all six of the electric field
values are zero; then ZEMAX will use the Ex and Ey values provided in array position 0 to determine the electric
field. Otherwise, the electric field is defined by these six values. The defined electric field vector must be
orthogonal to the ray vector or incorrect ray tracing will result. Even if these six values are defined for each ray,
values for Ex and Ey in the array position 0 must still be defined, otherwise an unpolarized ray trace will result.
as follows:
MyData[0].x = Ex; /* Electric field amplitude in x */
MyData[0].y = Ey; /* Electric field amplitude in y */
MyData[0].z = Phax; /* Phase in degrees for Ex */
MyData[0].l = Phay; /* Phase in degrees for Ey */
MyData[0].vigcode = 0; /* must initially be zero */
If all size of the electric field values are zero; then ZEMAX will use the
The behavior of this mode is otherwise the same as in mode 0.
Mode 3: Similar to GetPolTraceDirect
Like GetPolTraceDirect, the rays are defined by x, y, z, l, m, and n coordinates on any starting surface, as well
as the wavelength, the mode (either real, mode = 0 or paraxial, mode = 1) and the surface to trace the ray to.
Before the rays can be traced, the array should be modified in array positions 1 through n to contain the definitions
for each ray. The field, pupil, and wavelength data are all stored "in place" in the array as follows:
MyData[i].x = x;
MyData[i].y = y;
MyData[i].z = z;
MyData[i].l = l;
MyData[i].m = m;
MyData[i].n = n;
MyData[i].opd = 0.0; /* ignored */

MyData[i].Exr = Exr /* Electric field X real */
MyData[i].Exi = Exi /* Electric field X imaginary */
MyData[i].Eyr = Eyr /* Electric field Y real */
MyData[i].Eyi = Eyi /* Electric field Y imaginary */
MyData[i].Ezr = Ezr /* Electric field Z real */
MyData[i].Ezi = Ezi /* Electric field Z imaginary */
interrupt if the ray tracing is to be terminated.
The intensity parameter is used to return the relative transmitted intensity of the ray. If all six of the electric field
values are zero; then ZEMAX will use the Ex and Ey values provided in array position 0 to determine the electric
field. Otherwise, the electric field is defined by these six values. The defined electric field vector must be
orthogonal to the ray vector or incorrect ray tracing will result. Even if these six values are defined for each ray,
values for Ex and Ey in the array position 0 must still be defined, otherwise an unpolarized ray trace will result.
as follows:
MyData[0].x = Ex; /* Electric field amplitude in x */
MyData[0].y = Ey; /* Electric field amplitude in y */
MyData[0].z = Phax; /* Phase in degrees for Ex */
MyData[0].l = Phay; /* Phase in degrees for Ey */
MyData[0].vigcode = startsurf; /* the surface on which the coordinates start */
The behavior of this mode is otherwise the same as in mode 1.
Mode 5: For tracing non-sequential rays
Mode 5 is used to define and trace a ray inside of a non-sequential group. Unlike the other modes; there is no
text based command for tracing NSC rays, and only 1 ray at a time may be traced. However, rays may split or
scatter into multiple paths, and mode 5 will trace them all and return the entire tree of ray data.
To define the starting ray, use array position zero. The data is placed in the array as follows:
MyData[0].x = 0.0; /* starting x coordinate */
MyData[0].y = 0.0; /* starting y coordinate */
MyData[0].z = 0.0; /* starting z coordinate */
MyData[0].l = 0.0; /* starting x direction cosine */
MyData[0].m = 0.0; /* starting y direction cosine */
MyData[0].n = 1.0; /* starting z direction cosine */
MyData[0].opd = 5+nMaxSegments; /* sets mode 5, see comments below */
MyData[0].intensity = 1.0; /* initial intensity */
MyData[0].Exr = 0.0; /* initial E field if doing pol ray tracing */
MyData[0].Exi = 0.0;
MyData[0].Eyr = 0.0;
MyData[0].Eyi = 0.0;
MyData[0].Ezr = 0.0;
MyData[0].Ezi = 0.0;

MyData[0].wave = 0; /* wavelength number, use 0 for randomly selected by weight */
MyData[0].error = 1; /* NSC group surface, 1 if the program mode is NSC */
MyData[0].vigcode = 0; /* controls polarization, split, and scatter */
MyData[0].want_opd = 0; /* the inside of flag, use 0 if ray is not inside anything */
Only array position zero is used to define the starting ray. ZEMAX will trace the ray through the NSC group
and fill the remaining array positions with ray data. Since rays may split or scatter into multiple child rays, a very
large number of segments may be returned. The nMaxSegments number is the maximum allowed size of the
array for ZEMAX to return. The array should also be large enough to hold at least nMaxSegments elements. The
nMaxSegments value ZEMAX is using for the current optical system may be determined by the GetNSCSettings
command.
If polarization ray tracing is used, the initial electric field values must be provided. The user application must
ensure the defined vector is orthogonal to the ray propagation vector, and that the resulting intensity matches the
starting intensity value, otherwise incorrect ray tracing results will be produced.
The integer vigcode value determines if polarization ray tracing, splitting, and scattering are to be used. To
determine the vigcode value, use 0 for no polarization, 1 for polarization, plus 2 if splitting is used, plus 4 if
scattering is used. The resulting integer will be between 0 and 7 inclusive. Note if ray splitting is to be used,
polarization must be used as well.
The want_opd flag indicates where the ray starts. If zero, then the ray is assumed to not start in any object.
Otherwise, the ray starts inside the media defined by the specified object number. This is identical to the "inside
of" flag set on the NSC Editor in ZEMAX.
The array is passed to ZEMAX just like the other modes, see Step 2: Pass the array to ZEMAX on page 633.
When the ray tracing is complete, the array will contain the entire ray tree. Array element zero also contains
the total number of segments stored in the array element want_opd:
nNumRaySegments = MyData[0].want_opd;
The other values are interpreted as follows:
i = the segment number
MyData[i].wave = segment level
MyData[i].want_opd = segment parent
MyData[i].vigcode = inside of object number
MyData[i].error = hit object number
MyData[i].x,.y,.z,.l,.m,.n = the ray coordinates and cosines
MyData[i].intensity = the ray intensity
MyData[i].opd = the path length to the hit object
See also GetNSCSettings and SetNSCSettings.
Step 2: Pass the array to ZEMAX
Once all the rays are defined, the array must be sent to the ZEMAX server. The easy way to do this is using
the ZCLIENT PostArrayTraceMessage function:
PostArrayTraceMessage(szBuffer, RD);
The RD variable is the pointer to the DDERAYDATA array holding the list of rays. The ZEMAX server will trace
all the rays, then pass the data back in the same array. For more details on the code sample, see the section "a
sample client program" at the end of the chapter.
Note that an extension should never request that more rays be traced at one time than can be traced within
the timeout period specified in the zclient.c code. Within zclient.c there is a constant called DDE_TIMEOUT that
has a value of 5000 (the units are milliseconds, so this is 5 seconds). If more rays are traced in one call to
PostArrayTraceMessage than can be traced in the time allowed by DDE_TIMEOUT, an error will occur, and
PostArrayTraceMessage will return -1. If the ray trace is successful, PostArrayTraceMessage will return 0. It is
good programming practice to trace rays in small groups of 1000-5000 rays at a time. Once this many rays are
traced, the overhead of multiple calls using PostArrayTraceMessage becomes negligible. Alternatively,
DDE_TIMEOUT can be set at a higher number, but this is not preferred.

How ZEMAX calls the client
ZEMAX looks in the \Extend directory for Extension executables. Each program in this directory ending in the
usual .EXE extension is assumed to be a valid ZEMAX Extension program. The name of the executable is placed
in the pull down menu under "Extensions" on the main menu bar.
When any of the listed Extension names is selected from the menu, ZEMAX executes the Extension program
with the following syntax:
programname textflag optionsflag tempfile {settings data}
The programname is the full path to the executable, for exam ple, "C:\ZE MAX\EXTEN-
SIONS\DDE_DEMO.EXE".
The textflag is initially always 0. The convention ZEMAX uses for Extensions is the same as for internal ZEMAX
features: if the feature generates both a graphic and a text format for output, show the graphic version first. If the
feature does not have a graphic display, then show a text window. The Extension must generate a graphic file if
it can and is requested to do so. The user can click on "text" to look at the text version in a new window; if this
happens ZEMAX will call the Extension again with the text flag set to 1. If the Extension only generates text, then
it can generate a text file and open it by sending the MakeTextWindow item. ZEMAX will never send the textflag
set to 1 unless the Extension specifically states it can support a text mode; this is why the MakeGraphicWindow
returns a flag to indicate if the feature supports text, and why ZEMAX always ask for a graphic mode first.
The optionsflag is set to 0 if the user does not wish to see the "settings" box; or 1 if the user does. There are
two ways this flag may be set; either the user clicked on "Settings" on a window already open, or, the user has
selected "Show Options First" from the Environment options within ZEMAX. In this latter case the settings box
may need to be shown before the calculation is done the first time. If this flag is set the Extension must prompt
the user for setting options using a dialog box or other interface mechanism. If there are no user defined options,
then a message box stating "This window has no options" should be displayed.
The tempfile is the fully qualified path and file name of the temporary file to hold the data generated by the
Extension. Initially, this file will not exist, and the Extension needs to create it, write either text or graphic data to
the file, and then call MakeTextWindow or MakeGraphicWindow as appropriate. If the user is updating an existing
window, then the file may already exist, and it should be overwritten. The file should never be opened and read
as the data may not be what the Extension originally put there. ZEMAX does some translation of the Extension
created graphic file in particular. No other file should be used for output, because ZEMAX expects windows which
are being updated to retain their original temporary file names. Only one window may be created by a single call
to the Extension.
When ZEMAX receives a MakeTextWindow or MakeGraphicsWindow request from the Extension, one of the
arguments encoded in the atom name is a space delimited string of Extension defined values. This string should
be used to hold the "settings data" for the settings the Extension requires, such as the surface number, field,
wavelength, or other data for which the data in the window was computed. When ZEMAX updates a window, it
will call the Extension and provide the settings data string back to the Extension.
It is important that the Extension does not attempt to store the settings data itself, either statically or in a file.
The reason is that ZEMAX may have multiple text and graphic windows open at once, all using the same
Extension code to generate the data displayed. For example, a Extension that computes and displays the power
of a single surface may be used by several windows simultaneously, each for a different surface. Only ZEMAX
"knows" which window is using which setting data; and so ZEMAX passes this data on the command line. On the
initial call to the Extension, the settings data string is not supplied, and the Extension must determine it's own
default settings.
Because DDE data items cannot be longer than 255 characters, it is a good idea to use the SetSettingsData
rather than return the settings data appended to the end of the MakeGraphicsWindow or MakeTextWindow
commands. If the extension uses SetSettingsData to store the settings data with the window, then GetSettings-
Data should be used to retrieve the data whenever needed.
Generating a text window
Generating a text window for display by ZEMAX is accomplished with the MakeTextWindow item. First, open
the filename passed to the Extension using the typical C language fopen function, with "wt" as the mode for "write

text". Then, output data to the file, close the file, then send the MakeTextWindow item to ZEMAX. ZEMAX will
read the ASCII data and display it without modification or interpretation. A test code might look like:
FILE *out;
out = fopen(szFile, "wt");
fputs("Write this line\n", out);
fclose(out);
Do not forget to close the file!
Generating a graphic window
The Extension generates a graphic window by writing out graphic format data to the temporary file name
provided on the command line. The graphic data is created using a script language similar to that used by the
ZPL macro language or the annotation feature.
All ZEMAX graphics are written on a virtual screen whose coordinates are 1.00 wide by 1.00 high, with the 0,
0 origin at the lower left corner. The screen aspect ratio is normally 4 wide by 3 high, or 5 wide by 3 high, depending
upon user preference (See the File Menu chapter under Environments for details). Therefore, the virtual pixels
being drawn are wider than they are tall.
Each line in the script file should be followed by a newline "\n" character. After writing all of the graphic
commands to the file, the file should be closed for writing.
The following graphic script commands are supported.
ADDRESS
Syntax: ADDRESS
The address command will cause the "address box" in the lower right edge of the plot to appear, unless the
user has selected this box not to appear.
BOX
Syntax: BOX x1 y1 x2 y2
The BOX command will draw four lines, connecting the four points of a rectangle. The coordinates are the
upper left and lower right coordinates of the rectangle.
DATA
Syntax: DATA n "string"
The DATA command accepts an integer argument between 0 and 5, and a text string. The integer argument
indicates on which row of the data box the string should be written. Normally, the lens name goes on row 0; and
the date on row 1. The LENSNAME and DATE commands do this automatically. The DATA command allows any
string to be placed on any of the lines.
DATE
Syntax: DATE
The current data and time is normally written on the second line of the data box below the plot title. This
command has no arguments, and will print the date and time (depending upon the User's Environment
preferences) on this second line.
FRAME
Syntax: FRAME
Either a FRAME or a NOFRAME command must be the first line in the script file. FRAME will cause a standard
ZEMAX graphics frame to be drawn, with a rectangle around the entire perimeter, and two ruling lines at about
0.2 in the y direction. The plot title can be written centered inside these ruling lines using the TITLE command.
See also NOFRAME.
LENSNAME
Syntax: LENSNAME

The current lens name is normally written on the first line of the data box below the plot title. This command
has no arguments, and will print the lens name on this line.
LINE
Syntax: LINE x1 y1 x2 y2
This is the most basic graphic command; it draws a line connecting the two points. Note that the coordinates
should be between 0 and 1.0.
NOFRAME
Syntax: NOFRAME
Either a FRAME or a NOFRAME command must be the first line in the script file. NOFRAME will cause a
standard ZEMAX graphics frame to be drawn, but without the two ruling lines. See also FRAME.
PENSTYLE
Syntax: PENSTYLE ncolor nstyle nwidth
The PENSTYLE command changes the current pen color, style, and width. The integer ncolor must be 0 for
black, or between 1 and 12, for the corresponding pen color. The colors may be defined by the user in the File,
Environments dialog. The nstyle integer can be 0 for a solid line, or an integer between 1 and 4 for various styles
of dashed lines. The integer nwidth is the relative width of the pen. The default value is 1. A value of 2 will make
lines twice as wide as the default width. If the pen style is not a solid line (nstyle = 0) then the pen width must be
1 pixel; Windows does not support thick lines which are not solid. All subsequent text and lines drawn will be in
the new color, style, and width.
TEXT
Syntax: TEXT "string" x y angle width height
The text command is used to place text data on the graphic. The x and y coordinates are the starting
coordinates for the string. The angle is in degrees. The width and height values affect the font size. The default
value of 10 for each yields the same font size ZEMAX normally uses; which is 1/70 of the screen width and 1/40
of the screen height. Numbers larger or smaller than 10 will yield proportionally scaled font sizes.
TITLE
Syntax: TITLE "string"
If FRAME was used to define the graphic rather than NOFRAME, then the plot title would normally be centered
within the ruling lines along the bottom 20% of the plot. The "string" provided will be displayed in this position by
the TITLE command.
A sample Extension program
Within the \Extend directory is a sample application called DDE_DEMO which illustrates the use of DDE with
ZEMAX. The source code is included. DDE_DEMO can operate in either of two modes: stand alone, or as an
Extension.
If run directly from a command prompt, Windows Explorer, or by double-clicking on the DDE_DEMO icon, it
will run as a stand alone application, conducting the DDE in the background. The data it generates will be
displayed in the DDE_DEMO window. To run DDE_DEMO in stand alone mode, first run ZEMAX, load a lens file,
then launch DDE_DEMO by double clicking on the DDE_DEMO icon. Try loading different lens files, or editing
the one in the LDE, and watch the DDE_DEMO window to monitor the changes. DDE_DEMO will list out some
of the system, surface, field, and wavelength data, and will trace a few rays, and list the x, y, and z coordinates
of the rays on the screen. The program automatically updates itself every few seconds.
To run DDE_DEMO as an Extension, run ZEMAX, load a lens file, then select Extensions, then DDE_DEMO.
Note that any .EXE application placed in the \Extend directory will be listed in the ZEMAX Extensions menu;
whether or not it is a valid stand alone or Extension capable of communicating with ZEMAX. DDE_DEMO will
execute in a "hidden" window, establish the DDE link with ZEMAX, extract the system and ray data, write the data
to a file, notify ZEMAX that the file is ready (using the MakeTextWindow item) and then close. The client's job is
finished once it has generated the data; and it should terminate itself when done.

When ZEMAX receives the MakeTextWindow item, it reads the contents of the file and displays the data in a
standard ZEMAX text window. If the window is now "updated", or the "settings" option is selected, ZEMAX will
call the client program again and ask for the data to be computed again.
Note that the exact same program can act as both a stand alone client and as an Extension. The secret to the
bimodal operation is that if ZEMAX calls the client, it passes command line arguments to the client telling it what
type of data to create and where to write the data to. If no command line arguments are provided, the client can
assume it was launched from the operating system, and therefore it should run in stand alone mode. The DDE
exchange and subsequent processing is the same either way; only a few key lines of code are required to
determine whether the window should be visible or hidden, and whether the data is written to a file or the client's
own screen.
The DDE_DEMO code
The source code for DDE_DEMO is provided. It contains the basic algorithms, variables, declarations, and
Windows message handling required of all client programs. There are portions of the code which illustrate the
use of WM_DDE_POKE for tracing arrays of rays as well.
DDE_DEMO is a rather difficult program to use as a starting point for an Extension, although it does show
some high level techniques. A simpler way to implement ZEMAX Extensions is described below.
A simplified technique for implementing Extensions
Obviously, DDE programming for Windows is not for the faint of heart; it involves numerous programming skills
and significant familiarity with message loops, DDE, pointers, atoms, and global handles.
However, most of the code used to communicate with ZEMAX is standard "boiler plate" code which is common
to any DDE client program. This code has all been written already, and has been placed into a single source code
module program called ZCLIENT. ZCLIENT is provided in source code form with ZEMAX, and may be freely
copied (as long as the copyright is retained), and used in new Extensions.
The ZCLIENT program
ZCLIENT handles all of the communication with ZEMAX transparently. Embedded within ZCLIENT is a call to
a single function given the name "UserFunction". This function is placed in a separate C program provided by the
user, and is compiled along with ZCLIENT to create the Extension executable.
When ZEMAX calls the Extension, execution begins within ZCLIENT. ZCLIENT establishes the DDE commu-
nication, then calls UserFunction. Within UserFunction, two functions provided by ZCLIENT, PostRequestMes-
sage and GetString, are all that are usually required to get the data needed from ZEMAX. The data is then
formatted as either a text or a graphic, and sent back to ZEMAX for display.
The syntax for PostRequestMessage is as follows:
PostRequestMessage(szItemName, szBuffer);
szItemName contains the name of the item desired; the available item names are described earlier in this
chapter. The desired data is then sent back in the string szBuffer. This function returns 0 if the data transfer was
successful, otherwise, it returns -1. ZEMAX normally provides comma delimited data, and the function GetString
extracts the individual items. The syntax for GetString is:
GetString(szBuffer, nItem, szSubString);
The szBuffer string is the buffer returned by PostRequestMessage. The integer nItem is the number of the
item desired; 0 for the first item, 1 for the second, etc. The character szSubString holds the string from the nItem
position.
For example, to get the name of the lens, the code would be:
PostRequestMessage("GetName", szBuffer);
GetString(szBuffer, 0, szLensName);
The string szLensName now contains the name of the lens. Because "GetName" only returns 1 item, this code
could be shortened to just one line:
PostRequestMessage("GetName", szLensName);
There is also a modified version of PostRequestMessage for doing fast tracing of large numbers of rays. The
function is PostArrayTraceMessage. This function accepts two arguments: the buffer string, and the address of

the array of ray data. A sample array ray trace code using PostArrayTraceMessage is provided, it is called
ARR_DEMO.C.
To trace rays using PostArrayTraceMessage, the ray list must first be defined, then a single call to PostArray-
TraceMessage must be made. Here is a sample of the technique:
/* Fill RD array position 0 with the header data. */
RD[0].x = 0.0;
RD[0].y = 0.0;
RD[0].z = 0.0;
RD[0].l = 0.0;
RD[0].m = 0.0;
RD[0].n = 0.0;
RD[0].opd = 0.0; /* this is where we set the mode, mode 0 is like GetTrace */
RD[0].intensity = 0.0;
RD[0].wave = 0;
RD[0].error = 25; /* trace 25 rays */
RD[0].vigcode = 0;
RD[0].want_opd = -1;
/* Define the 25 rays. Obviously, you can define any rays you want... */
k = 0;
for (i = -2; i <= 2; i++)
{
for (j = -2; j <= 2; j++)
{
k++;
RD[k].x = 0.0;
RD[k].y = 0.0;
RD[k].z = (double) i / 4.0;
RD[k].l = (double) j / 4.0;
RD[k].m = 0.0;
RD[k].n = 0.0;
RD[k].opd = 0.0;
RD[k].intensity = 0.0;
RD[k].wave = 1;
RD[k].error = 0;
RD[k].vigcode = 0;
RD[k].want_opd = 0;
}
}
/* Now go get the data */
PostArrayTraceMessage(szBuffer, RD);
/* The output data is now stored in RD, ready to use! */

/* Note just one line of code to trace all rays. */
Using ZCLIENT is vastly easier than writing all the DDE communication code yourself. This sample array trace
is taken from ARR_DEMO, one of the sample DDE codes that use ZCLIENT.
A sample Extension program using ZCLIENT
The source code for a complete Extension which uses ZCLIENT is provided. The Extension is called
PHASPLOT, and it uses ZCLIENT.C as well as PHASPLOT.C. PHASPLOT generates either a graph or a text
listing of the phase of a binary 2 surface. The code illustrates how to make a graphic and a text window, how to
get data out of ZEMAX, and how to format output for graphics and text display.
PHASPLOT also has a ZEMAX style "settings" box, the source code for which is included in PHASPLOT.C in
the "SurDlgProc" function; which uses the PHASPLOT.RC file to define the dialog box appearance.
Writing Extensions from the PhasePlot template
The best way to learn how to write Extensions is to start with PhasePlot, and start editing! Make a copy of
PhasePlot.C and PhasePlot.RC, and rename the copies to something else (MY_CODE.C, MY_CODE.RC). Then
create a new project using any of the good commercially available C compilers (Borland and MicroSoft both make

excellent tools) and load these 3 files into the project. Recompile and build the project into a test EXE to make
sure everything is in place.
Then, edit the code within UserFunction and the settings dialog function to suit your requirements.
Getting analysis data using GetTextFile and GetMetaFile
One often overlooked capability of Extensions is the ability to create text and graphic data files from analysis
already supported by ZEMAX by using a single call to zclient. For example, to generate a text listing of the spot
diagram, use the command
PostRequestMessage("GetTextFile, \"C:\\OUTPUT.TXT\", Spt, , 0", szBuffer);
The spot diagram data will be placed in the text file "OUTPUT.TXT" in the specified directory. For more
information, see GetTextFile on page 618 and GetMetaFile on page 612.

Annotate
INDEX Graphics 37
Annotations
Symbols On graphic windows 38
Annular aspheric surface 278
m 87 Annulus object 298
A Anti-nodal planes 45
Anti-principal planes 45
Aperture
Abbe number
System aperture value 85
Tolerance on 448
Aperture checking
Abbe V number
With footprint diagram 160
Definition for 485
Aperture type 68
For MILNUM glasses 485
Apertures 69
Operand 405, 407
Floating 69
ABCD surfaces 228
Removing 206
Aberration
Soft edged 279, 280
Longitudinal 160
Surface aperture 53
Pupil 110
System aperture 53, 85
Ray fans 108
User defined 69
Wavefront 128
Aplanatic condition
Aberration coefficients 164, 165, 416
Solving for 381
Annular Zernike 171
Apodization 471
Extended Zernike 169
Cosine cubed 86
ABg model 354
Defined 45
Catalog of data 203
Gaussian 87, 390
Wavelength scaling of data 204
Apodization factor 471
Aborting
Defined 45
Editing of cell values 36
Specifying 87
Long computations 40
Apodizations
Absorbers 347
User defined 45, 277
Absorption 484, 500
Appending
Active configuration
Lens files 58
Defined 45
AR
Active Cursor 37
Codes 479
Address box
Arms
Preferences 58
For pupil integration 388
Afocal systems 266
Array
Air pressure 491
Of diodes 337
Airy disk
Array variables
Defined 111
ZPL macros 596, 600
Alternate Even surface 228
Arrays
Alternate Odd surface 229
Of lenses 278
Alternate surface intersection 228, 229
Arrays of objects 215
Analysis units 87
Arrow
Anamorphic asphere surface 278
Drawing on plots 38
Angle of incidence 414
Arrow source
Angular distributions 332
Size of 96
Angular magnification
Aspect ratio 37, 39
Defined 45
Printing 60
Operand 393
Aspheric object 308, 309, 318
Angular Spectrum 530
Aspheric surface object 298, 299, 331
Angular spectrum 517
Hologram 313
Animation 218
641
Aspheric surfaces 243 Boresight error 394
Fast tracing 92 Tolerancing 441, 455
High order terms 244, 246 Boundary conditions
Superconic 271 On (center) thickness 396
Asterisk On curvature 396
Hyperhemispheric surfaces 52 Boundary operands
In data cells 61 Understanding 421
Indicates surface apertures 65 BSDF 354
Astigmatism Bubbles 323
Calculating 164 Bugs
Operand 393 Reporting 33
Athermal lenses Bulk absorption 352, 512
Analyzing and optimizing 496 Bulk scattering 349
Atmospheric refraction 229 Bulk transmission 512
Atmospheric turbulence 258 Button bar
Axial color Defining 63, 64
Operand 394
Axial symmetry 389
C
Axicon 270
CAD files
Axicon surface object 299
Importing 314
B Candela 88
Capping
Back focal length GRADIUM profiles 256
Defined 45 Capturing
Beam footprint 160 Screen images 42
Beam splitter Cardinal planes 45
Defining coatings for 352 Cardinal points 45, 178, 220
Beamsplitter Cell
Polarization modeling 512 Editing contents 36
Beep command 552 Cell width
Beers Law 512 Changing 60, 61
Bending Celsius
Using fold mirrors 207 Defining temperature 491
Best fit glass 219 Center thickness
Best fit sphere 177 Operand 396, 405, 407
Biconic 226 Centroid 111
Surface type 229 Determining coordinates of 113
Biconic lens object 300 Encircled energy 135
Biconic object 300, 301 Optimization 410, 417
Bicubic spline 259 Check for updates 61
Binary 1 object 302 Checking for 61
Binary 2 object 303 Chief ray
Binary optics 230, 232, 282, 284 Definition of 45
Sign conventions for 232 Normalized coordinates 49
Biocular analysis 154, 155 Solve 380, 381, 382
Birefringent materials 235 Chief ray solve
Bitmap images 146 On coordinate breaks 384
Bitmaps Chord Tolerance 314
As sources 322 Chromatic aberration 162
Exporting 42, 556, 557 Circular 69
BMP 146
Creating 37
642
Clipboard Conjugate
Copying graphics to 37 Surface type 238, 239
Copying text to 39 Conrady
Cut, Copy, and Paste 75, 76, 77, 78, 81 Dispersion formula 481
Selecting metafile format 60 Consider objects list 351
Using 42 Constant
Cloning windows 37, 39 Operand 396
Coating file 91, 202, 501 Construction optics
Coating groups 352 Constraining 395
Coating multiplier 74, 505, 508 For holograms 265
Coatings Contours
Adding to surfaces 203 Setting 131
Defined by tables 507 Convergence 154, 155
Editing 202 Convert semi-diameters
Loading coating file 203 to apertures 203, 206
Multi-configuration control of 472 Converting to NSC groups 214
Optical 184 Coordinate axes
Properties of 184 Definition of 46
Tapered 503, 505 Coordinate breaks
Coherence length 350, 359 And paraxial rays 92
Coherent imaging 149 Reversing with "order" 239
COLOR Solves for 384
ZPL macros 553 Surface tilts and decenters 73
Color Surface type 239
True color PSF 127 Coordinate reference
Color of rows 67, 348 Drawing 349
Color slides 322 Coordinate system
Colors Conventions for 46
Changing default 59 Copy
Defining 63 Operand data 76, 77, 79
Of objects 348 Surface data 75, 81
Of surfaces 67 Corner object 319, 329
Coma Cosine lamp 339
Calculating 164 Cosine source 339
Operand 396 Cosine space 112, 332
Comments 66 Cosine surface 267
Hide/Show menu 75 Coupling efficiency 142, 145, 175, 400, 538, 546
In the merit function using BLNK 394 CPC 305
Company name CPUs
On plots 58 Number of 364, 423, 437
Compensators CR
For tolerancing 450 Codes 479
Compound parabolic concentrator (CPC) 294 Creating POB files 217
Cone angle 85 Cross sections
Cone object 304, 305 Point Spread Functions 125
Configuration CSGs defined 352
Changing 97, 475 Cubic spline
Selecting on windows 37 Extended surface type 245
Zoom position 97 Cubic spline surfaces 240
Conic Curvature
Description of 270 Boundary conditions on 396
Solve for 384 Operand 396
Tolerance on 442 Cut
Surface data 75
643
Cylinder arrays 278 Diffraction gratings 241, 268
Cylinder Fresnel surface 240 On cylinders 273
Cylinder lens object 315, 317, 327 On toroids 273
Cylinder pipe object 305, 306 Diffraction Imaginary Transfer Function 115, 116, 117,
Cylinder volume 396 120, 121, 122, 125, 126, 128
Cylinder volume object 306, 307 Diffraction limited
Cylindrical lens Meaning of 46
Paraxial model for 267 Diffraction Modulation Transfer Function 115, 116,
Cylindrical surfaces 229, 272, 273 117, 120, 121, 122, 125, 126, 127, 128
Gratings placed on 248, 273 Diffraction Phase Transfer Function 115, 116, 117,
D 120, 121, 122, 125, 126, 128
Diffraction Real Transfer Function 115, 116, 117, 120,
Date
121, 122, 125, 126, 128
Diffractive optics 258
Printing on plots and text 61
Digital Mirror Device (DMD) 317
DDE 607
Diode 337
Debugging
Dipvergence 154, 155
Tolerancing 468
Directories
Decenters
Default path for output 59
Tolerancing 445
Directory names
Decenters and tilts
Preferences 59
On surfaces 73
Dispersion 162, 163, 178, 179, 198
With coordinate break surfaces 239
Dispersion coefficients 480
Decimals
Distortion 156, 158
Changing default 60, 61
Calculating 164
Default merit function 385
Definition of 156
Start at row 398
F-Theta 156
Default path
Grid 158
For output 59
Operand 398
Defaults, changing 59
Representing 146
defined 289
Divergence 154, 155
Delano diagrams 161
DLL 275
Deleting
dn/dt 495
Operands 76, 77, 78, 81
dPgF 479
Surfaces 65, 74
Drawing
Surfaces in ZPL macros 554, 558
Element 104, 105
Density
Dual zone 234
Discussion of 220
Dynamic Data Exchange (DDE) 607
Of lenses 220
Dynamic Link Library 275
Detector object 332, 334, 335, 348
Detector viewer 369 E
Detector volume 335
Dialogs 35 Edge margin 93
Diameter Edge Spread Function 125, 137
Operand 398 Edge thickness
Diameter to thickness ratio Controlling 382
Constraining 406, 407 Defined 46
Diattenuation Determining 219
Of coatings 185 Operand 399, 406, 407, 418
Diffraction 515 X- direction 419
Diffraction grating 307 Editing
Multiple orders 350 Spreadsheet values 36
Diffraction grating object 307
644
Editor windows 35 Exporting
Menu 65 Graphics and text 42
Preferences 61 IGES and STEP files 209, 556, 609
Editors Extended cubic spline
Preferences 60 Surface type 245
Effective F/# 139, 140, 399 Extended polynomial
Effective focal length Surface type 247
Defined 47 Extended Polynomial Lens object 308
Displaying 219 Extended Polynomial Surface object 309
Operand 398 Extended sources 142, 146
Single element 219 Encircled energy computations 137
Element drawings 104, 105 Extensions 223
ISO 10110 105 Extra data
Element power 219 Defined 47, 225
Solve 381 Operand 418
Ellipse Optimization of 426
Defining 270 ZPL Macros 560
Elliptical 69 ZPL macros 555
Elliptical object 307 Extraordinary rays 235
Elliptical surface
Defining 238, 239
F
Encircled energy 135
F/#
END
Capped at 10,000 94
ZPL macros 568
Curvature solve 379
Ensquared energy 135, 136, 138
Displaying 219
Entrance pupil diameter
Effective 140
Defined 47
For fields, wavelengths 220
Operand 399
Method of computing 94
Setting 85
Paraxial working 51
Entrance pupil position
Working 55
Defined 47
Fabrication support 177
Operand 399
Faceted object 325, 326
Entry ports 291
Faceted surfaces 249
Environment
Far field 517
Defining 491
Far field pattern 339
Temperature and pressure 491
Far-field cosine space 332
EQUA
Fast Asphere Trace 92
Operand 399
Fiber coupling efficiency 142, 145, 175, 400, 538, 546
Escape
Fiber optic object 328
Aborting long computations 40
Field angles
Even Asphere 226
Conventions for 47
Even Asphere lens object 308
Field curvature 140, 142, 153, 154, 156
Even Aspheric 308
Calculating 164
Examples 32
Operand 400
Exclude Substitution 478
Field curvature plot
Exit ports 291
Meaning of 154, 155, 156
Exit pupil
Offsets on axis 156, 161
Significance of 91
Field positions
Exit pupil diameter
Defining 96
Defined 47
ZPL macros 576
Exit pupil position
Field type
Defined 47
Multi-configuration control of 472
Operand 399
File menu 35, 57
Exponential notation 61
645
File names
On plots 58
G
Files
Gases
Finding names in ZPL macros 558
Thermal modeling 484
Filter string 364
Gaussian Beam
Filters
Analysis 187
User defined neutral density 277
Operands 400
Fitting
Paraxial analysis 187
Test plates 197
Physical optics analysis 515
Fletch
Skew analysis 189
Size of 60
Gaussian factor
Fletch Rays 101
Optimization 389
Flip
Gaussian Quadrature 387
Lens elements 212
Gaussian scattering 354
Float by stop size 85
Gaussian source 339
Defined 47
General data 85
Floating apertures 69
Geometric MTF 119
Focal length
Vs. Field 121
Scaling by 213
Geometry errors 364
Focus
Ghost reflections 48, 215, 362, 371
Quick focus tool 213
Glass
Fold mirrors
Best fit 219
Inserting 207, 215
Codes 479
Font size
Constraining 425
Changing 60, 61
Cost factors 479
In text windows 61
Dispersion formulas for 480
Setting in ZPL macros 595
Entering type 48
When printing 62
Excluding from optimization 478
Footprint
Gradient index 250, 349
Beam 146
Model 219
Footprint diagram 160
Optimizing 424, 437
FOR loop
Optimizing with RGLA 415
ZPL macros 558
Pick up solves 383
Foucault analysis 129
Removing in multi-configurations 475
As optimization operand 400
Substitution 383, 426, 437, 478
FR
TCE 479
Codes 479
Templates 195
Fraunhofer diffraction 520
Transmittance of 163, 179
Fresnel
ZPL macros 548
General polynomial form 249
Glass catalog 162, 163, 178, 179, 198, 203, 477
Zone plates 286
Adding new glasses to 480, 482
Fresnel diffraction 519
Copying or moving 479
Fresnel lens 309, 310, 534
Creating new 479
Fresnel lens object 309
Managing multiple 477
Fresnel lenses 326
Melt data 483
Fresnel number 516
Obsolete glasses 490
Fresnel object 327
Selecting 477
Fresnel surfaces 240, 245, 249
Transmission data 484
Fringes
Global coordinates 93, 174, 612, 613
Test wavelength 452
From ZPL macros 543
F-theta
Multi-configuration control of 472
DISC operand 397
Operand 401, 414
F-Theta distortion 155, 156
Reference surface 67, 93, 560
Full width half max 138
646
Global optimization 195, 433, 435 Holograms 261, 262
Glue distance 95, 345 Aspheric non-sequential 313
GOSUB 568 General polynomial form 230, 232
Gouy shift 189, 522 Multiple orders 350
Gradient index 250, 251, 252, 253, 257, 258 Non-sequential solid 312
Boundary operands for 429 Optically fabricated 264
Dispersion model 252 Optimizing construction optics 265
Operand 403 Toroidal 273
Vignetting 93 Used in reflection 261
Gradient Lens Corporation 254 Using Zernike coefficients 282, 284
Graphic frames 60 Huygens
Graphics Point Spread Function 126, 127
Colors 63 Hx
Copying to clipboard 37 Defined 49
Exporting using Clipboard 42 Hy
Preferences 59 Defined 49
Grating mode 268 Hyperbola
Gratings 241, 268 Defining 270
Elliptical 242, 243 Hyperhemispheres
Multiple orders 350 Preventing with HHCN 403
Toroidal 248, 273 Hyperhemispheric 67
Grid phase surface 258 Described 52
Grid sag surface 259 Testing for 403
Guard band 526 Hyperhemispheric lens object 323
H Hyperhemispheric object 301
Hyperhemispheric surface object 324
Hammer optimization 195 I
Hardware lock
Installing 32 Ideal coatings 507
Policy regarding 33 IGES files
Harvey-Shack scattering 354 Exporting 208, 209, 210, 211
Heads-Up Displays (HUD) 154, 155 Importing 314
Help Ignoring objects 348
Button 40 Illumination 146
Menu option 35 Image analysis 142, 146, 148, 151
Herzberger Aborting long computations 40
Dispersion formula 481 As optimization operand 403
Hexapolar rings Editing IMA files 143, 149, 152, 153
Defined 48 IMA file format 144
Hidden line drawings 100, 102, 103, 107, 108 Image height
Hiding Field definition 47
Edges of surfaces 68 Image plane
Rays on layout plots 68 Coordinates 112
Highlight color 60, 63 Image space F/#
Highlighting layouts 60 Operand 404
Setting 85
Images
Appearance of 142, 146
Importance sampling 351
Imported object 314
Incoherent imaging 149
Increment
tolerancing 455
647
Index data Knife edge test 129
Operand 404
Optimizing 424
L
Index of refraction 482, 485
Lagrange invariant 165
Determining 219
Operand 404
Dispersion plot 162, 178
Lagrange multiplier 390
Table of values 220
Lambertian 354
Temperature and pressure effects 491
Landscape orientation 62
Inequality
Last zoom 37
Operand 410
Lateral color 161
Initial Phase 350
Operand 404
Input
Layout
ZPL macros 572
Element drawing 104, 105
Input focus 36
Layouts
Inserting
Scale bar width 63
Operands 76, 77, 78, 80
Three-dimensional 100, 102, 103, 107, 108
Surfaces 65, 74
Two-dimensional 99
Surfaces in ZPL macros 554, 555, 572, 573
Learning ZEMAX 32
Inside of
Lens catalog 199
Placing sources 336
Lens Data Editor
Installation
Printing contents of 220
Of the hardware lock 32
Lens design
Integer
References on 31
Rounding to using INTE 539
Lens element
Intensity 88
Reversing 212
Interferogram
Tilting and decentering 212
Data for surface deformation 282, 283, 284, 285
Lens files
Drawing 128
Deleting 58
Interferometers
Inserting 58
Using data from 282, 284
Loading 57, 58, 574, 575
Internet 61
Opening 57
Invariant 165
Recently used list 64
Irradiance 88
Saving 57, 587
Irregular
Lens mounts
Surface type 262
Thermal expansion of 491
Irregularity 445, 446, 447
Lens object 323
ISO drawings 105
Lens search 199
J Lens units
Defined 48
Jones matrix 511 Specifying 87
Jones matrix object 295 Lenses
Jones matrix surface 263 Plotting to scale 100
Jones vector 497 Lenslet object 315, 317
JPG Lightpipes 294, 295, 304, 305, 320
Creating 37 Line Spread Function 125, 137
Jx, Jy 350 Linear interpolation 259
K Liquids
Thermal modeling 484
Load and Save buttons 40
Keyboard
Loading
Shortcuts 40
Lens files 58
Use of 40
Lock window 37, 39, 575
Keyboard shortcuts 36, 40
Kinoforms 230, 232
648
Long computations Mirrors
Aborting from 40 Conventions for 53
Longitudinal aberration 139, 160, 161 Deleting fold 207, 215
Operand 405 Inserting fold 207
Longitudinal aberrations Specifying 48, 66
Calculating 164 Model glass 219, 383, 424, 485
Longitudinal color Modulation Transfer Function 115, 116, 117, 120, 121,
Calculating 164 122, 125, 126, 128
Lost energy 364, 365 Monte Carlo ray tracing 363
LPTD Mouse
Operand 429 Shortcuts 37, 39
Ltk 479 MTF
Luneberg lens 280 Aborting from long computation 40
M Calculating in macros 562
Geometric approximation 119
Operand 406
M Squared 187
Optimization of 421
Macros
Visualizing 142, 146
ZPL macros 535
Vs. Field 116, 117, 121
Magnification
Vs. Focus 115
Displaying 219
Multi-configuration 471
Operand 393, 412
Operand values 405
Main window 35
Optimization and 475
Marginal ray
Tolerancing 467
Definition of 48
Multi-mode fiber coupling 142, 145
Normalized coordinates 49
Multiple CPUs
Solve 380, 381
Parallel processing 364, 423, 437
Margins 93
Multiple orders 350
Setting for printing 62
Multi-threading 364, 423, 437
Mass
Of lenses 220, 417 N
Maximum aspheric deviation 177
Melt data Narcissus
fitting 483 YNI contribution 419
Melt Frequency 479 Near field 517
Merit function Nesting object limits 345
Default 385 Nesting sources 336
Defining 385 Neural networks 435
Formula for 391 Neutral density filters
Gaussian apodization of 390 With user defined surfaces 277
Ignoring operands in 399 Nodal points 45, 220
Loading and saving 76 Non-paraxial system
Loading from ZPL 575, 587 Defined 49
Modifying 390 Non-sequential 293, 336
Percent contribution of components 420 Converting surfaces to 214
Printing 195, 196, 197 Defined 49
ZPL macros 570, 577 Multi-configuration operands for 473
Metafile Object definitions 293, 336
Creating 37 Optimization of systems 427
Windows clipboard format 60 Optimization operands for 408, 410
MicroElectroMechanical System (MEMS) 295 Ray tracing 289
Micrometers 87 Ray tracing limits 94
Microns 87 Restricting 94
MIL number glasses 485 Non-sequential program mode 57
649
Normal distribution 460 Operand data
Normalized coordinates 390, 579, 583 Cut, Copy, and Paste 76, 77
Defined 49 Operands
Note files Defining complex 423
Editing 104 User defined 429
Notes 88 Optical design
NSC 289 References on 31
With ports 290 Optical invariant
without ports 290 Lagrange invariant 404
NSG America, Inc. 257 Optical path differences
Null object 318 Operand 410
NURBS surfaces 269, 274 Referencing 91
O Solves 382
Optically fabricated hologram 264
Optimization 195, 385
Object
Global 195, 435
User defined 329
Global minimum 433
Object plane
Global search 435
Tilting 272
Hammer 195, 435
Object space N.A.
Merit functions for 385
Setting 85
Of glass 424
Object space numerical aperture
Operands 390, 423
Defined 50
Requirements 385
Operand 410
Selecting variables for 385
Objects 293, 336
Suggestions for use 433
Arrays of 215
With external programs 431
Ignoring 348
With multi-configurations 475
Inserting and deleting 81
ZPL macros 429
Overlapping 345
Optimization operands
Replicating 215
Plot of 180
Objects of revolution 325
Orders
Obscuration type 68
Splitting rays by 350
Obscurations 68, 73
Orientation
Circular 69
Graphics 62
Elliptical 69
Orientation indicator 101
Rectangular 69
Size of 60
User defined 69
Overlapping objects 345
Vignetting from 140
Overlaying graphics 37, 457
Obsolete
glasses 490 P
Odd Asphere 227
Odd Asphere lens object 318 Pan and zoom 39
Offense against the sine condition 411 Parabasal
Ogive surface 281 Discussion of 50
Opacity Parabola
Of objects 348 Defining 270
Of surfaces 67 Parallel processing
OPD Multiple CPUs 364, 423, 437
Optical path differences 410 Parameter data
OpenGL Defined 50, 225
Shaded model graphics 103, 108 Tolerancing 448
Opening Parameter solve
Lens files 57 On coordinate breaks 384
650
Paraxial 227 POB files
Definition of 50 Creating 217
Working F/# 51 POB format 319
Paraxial focus 381 Point Spread Function
Paraxial image height Cross sections 125, 127
Defined 51 FFT 122, 126
Paraxial lenses 266, 267 For NSC rays 333
Paraxial magnification Huygens 122, 126
Defined 51 Point spread function 122, 126, 127
Paraxial ray data Polarization 525
Operands 411 Analysis 497
Paraxial rays Defining global state 90
And coordinate breaks 92 In non-sequential ray tracing 352
Tracing 174 NSC sources 350
Paraxial surfaces 266, 267 Partial 350
Paraxial XY 227 Phase of detectors 333
Paste Ray trace 182, 190, 193
Operand data 76, 77, 79 Reflectivity, Transmission, Absorption 184
Surface data 75, 76, 77, 79, 80, 81 ZPL macros 579
Pasting Polygon object 319
Graphics and text 42 Polygon objects
Paths Defining 373
For data files 59 Polynomial
PAUSE Extended surface type 247
ZPL macros 579 Polynomial surfaces 268
Peak to valley Pop-up screens
Wavefront error 165 Configuring 59
Pen Portrait orientation 62
Changing width of 62 Ports in non-sequential tracing 290
Pen color Power
ZPL macros 553 Function 542
Pen colors 63 Maintaining with glass change 66
Pen width 60, 62 Operand 413
Periodic surface 267 Solve 381
Petzval Watts 289
Operand 411 PR
Phase Codes 479
Point Spread Function 122 Preferences 58, 59
Phase of a surface 131 Address box 58, 59, 60
Photometric units 88 Directory names 59
Physical optics 515 Editors 61
Pick up solves 379 Printing 62
Pilot Beam 529 Pre-Propagation 350
Pilot beam 521 Prescription Data 220
Pinhole 529 Pressure
Pinhole aperture 534 Of each surface 495
Placing sources 336 Setting 90
Plot width 62 Pressure effects 491
Plots Primary wavelength 96
Changing pen width 62 Defined 51
Plotting Multi-configuration 473
All zoom positions 102 Principal planes 45, 220
Optimization operands 180
651
Printing 36 Radius
ASCII from ZPL macros 581 Curvature 396
Changing pen widths 62 Radius of curvature
Coating definitions 203 Conventions for 51
Contents of a window 219 Random number generator
Font size 62 Seeding 452
Graphic windows 37 Random Polarization 350
Landscape mode 63 Ray aiming 67, 88
Merit function 195 While tolerancing 457
Orientation of 62 ZPL macros 588, 589
Portrait mode 63 Ray database files 367
Preferences 62 Ray fans 108, 110
Prescription data 220 Ray splitting 289, 362
Spreadsheet data 220 Ray tracing 174
Surface data 219 Rayleigh range 188, 520, 521, 522, 527
System data 219 Raylist format 102
Text margins 63 Raylists 102
Text windows 39, 219 Real propagation 51
Tolerance data 196, 197 Real ray
Using ZPL macros 580 Tracing 174
Windows in ZPL macros 581 Recovering lost files 82
Prisms 319 Rectangular 69
Probing rays 524 Rectangular object 320
Profiles Rectangular pipe object 320
Scattering 91, 349 Rectangular volume grating object 322
Program mode 57 Rectangular volume object 322
Propagation report Redo 82
For POP analysis 191, 533 Reference Radius 530
ProSource 341 Reference sphere 91
Pupil aberration References 31
Controlling 88 Referencing OPD
Pupil positions Described 91
Displaying 219 Reflections
Pupil shift 89 Ghost 215
Px Reflectivity
Defined 49 Of coatings 184
Py Reflectors
Defined 49 Specifying 48, 66
Q Refractive index
Index of refraction 219
Relative Cost 479
QSUM
Relative cost of glasses 479
Operand 414
Relative illumination 139, 415
Quadratic sum
Remote Desktop 61
Operand 414
Replicating objects 215
Quality factor 187
Report graphics 220
Quick focus 213
Resolution
Quitting ZEMAX 64
Estimating 142, 146
R Resolving
Two field points 113
Radiant Imaging 341 Retardance
Radii Of coatings 186
Entering 66 RETURN
ZPL macros 568
652
Reverse Rays 350 Scattering 352
Reversing ABg 354
Lens elements 212 Catalog of data 203
RGLA Gaussian 354
Operand 425 Harvey-Shack 354
Rings 387 Lambertian 354
RMS On sequential surfaces 72
Spot size 112, 132, 133, 134, 385 On spot diagrams 112, 122
Wavefront error 132, 133, 134, 165, 385 Towards an object 351
X and y spot size 112 User defined 356
Robust Viewer 204
Ray aiming mode 89 ZPL macros 588
Roofs 321 Scattering profile file 91
Rotating Scattering profiles 349
3D layout plots 101 Schott
Rotation matrix 93, 220, 344 Constants of dispersion formula 480, 482
Row color 67, 348 Screen capture 42
S Scripted tolerancing 463
Searching
For stock lenses 199
Sag
Seidel coefficients 164, 165, 169, 171
Constraining 416
SELFOC lenses 257, 258
Operand 416
Sellmeier
ZPL macros 544
Dispersion formula 480
Sag of a surface 130
Semi diameter margin 93
Sag table 177
Semi-diameter
Sagittal
Accuracy of 92
Definition of 51
Semi-diameters
Sampling
Accuracy of 52
Hexapolar rings 48
Converting to apertures 206
SAT files
Defined 52
Exporting 209
Entering 66
Importing 314
Sequential program mode 57
Save and Load buttons 40
Sequential ray tracing 289
Saving
Defined 52
Lens files 57
Session files 57
Scale bar
Set to default 385
On layout plots 99
Set vignetting
Setting accurately 63
On field dialog box 96
Scale factor 100
ZPL macros 596
Scale lens 213
Variables 74
Scaling
Settings
On printers 63
Saving with the lens 40
Scatter groups 352
SETVECSIZE
Scatter Towards list 349, 351
ZPL macros 595, 596
Shaded model 103, 108
Shear distance 258
Shortcuts
Keyboard 35, 36
Summary of 40
Sign conventions
For binary optics 232
For radius of curvature 51
Simulated annealing 435
653
Single mode fiber coupling 175 Spline
Slide object 322 Bicubic using grid surface 259
Slider control 218 Linear interpolation using grid surface 259
Soft apertures 279 Spline surfaces 240, 269, 274
Solid modeling 100, 102, 103, 107, 108 Spot diagrams 111, 113
Solves 379 Configuration matrix 113
Aplanatic 381, 383 Full field 113
Center of curvature 383 Matrix 113
Chief ray angle 380 Spot size 132, 133, 134
Chief ray height 382 Spreadsheet
Chief ray normal 381 Printing Lens Data Editor contents 220
Compensator 382 Spreadsheet editor
Concentric 381, 383 Cursor controls for 65
Edge thickness 382 Shortcuts 65
Marginal ray angle 380 Square root
Marginal ray height 381 Operand 404, 405, 416
Marginal ray normal 381 Square wave MTF
Optical path difference 382 Operand 406
Pick up 380, 382, 383, 384, 476 SR
Position 382 Codes 479
Setting and removing 74 Standard deviation
Suggestions for use 384 Wavefront error 166
Thermal Pick Up 493 Standard lens object 323
Turning off 379 Standard surface object 324
Sorting Status bar
Tolerance data 79 Modifying 64
Source arrow STEP files
Size of 96 Importing 314
Source bitmap 322 STL format 324
Source diode 337 STL Object 324
Source object STL objects
Ellipse 339, 343 Defining 376
Filament 340 Stock lens catalog 199
File 340 Stock lenses
Gaussian 342 Using search to find 199
Point 342 Stop surface
Rectangle 343 Defining 67
Tube 343 Zooming in multi-configurations 474
Volume cylinder 343 ZPL macros 598
Volume ellipse 344 STP files
Volume rectangle 344 Exporting 209
Source units 87 Importing 314
Sources 336 Straddling objects 345
Extended 142, 146 Strehl ratio 46, 52, 126, 127, 132, 133, 134, 167
User defined 340 ZPL macros 567
Specular ray 352 String length
Spherical aberration ZPL macros 544
Calculating 164 String operators 549
Operand 416 String variables
Spherical object 323 ZPL macros 547
Spherochromatism Strings
Operand 416 Comparing 544, 545
Spider Subdirectories 59
Spider aperture 69
654
Subroutines TCE 479
ZPL macros 568 Adding data 495
Substitute Data defined 493
Glass 383 In glass catalogs 479
Substitution Technical support 33
Templates for 195 Telecentric mode 87
Superconic Temperature
Surface type 271 Of each surface 495
Supergaussian 337 Setting 90
Surface apertures Temperature effects 491
Defined 52 Templates for glass substitution 195
Multi-configuration values 471 Test plate fit files 198
Specifying 68, 73 Test plate fitting 197
Surface data Test wavelength 452
Cut, Copy, and Paste 75, 81 Tolerancing 450
Surface numbering 65 Text
Surface Phase 131 Copying to clipboard 39
Surface power 219 Text data
Surface Sag 130 Exporting 564
Surface tilts and decenters 73 Text windows
Surface types 225 Extracting data from 538
ABCD 228 Thermal effects 491
Alternate intersection 228, 229 Auto setup 78
Conic 67, 225 Thermal modeling
Coordinate breaks 239 Of gases and liquids 484
Cubic spline 240 Thickness
Diffraction gratings 241, 268 Boundary conditions on 396
Fresnel 249 Conventions for 53
Holograms 261 Operand 417, 418
NURBS 269, 274 Solves on 381
Plane 67, 225 Thicknesses
Selecting 67, 225 Entering 66
Surfaces Thin films 184
Deleting 65 Coatings 184
Inserting 65 Thin lens
Miscellaneous data 219 Cylindrical 267
Reversing order of 212 Surface type 266
System aperture Through focus
Defined 53 Geometric MTF 120
Multi-configuration value 471 Tilted surface 262, 272
Specifying 85 Tilts
System data Tolerancing 445
ZPL Macros 560, 563 Tilts and decenters
System menu 85 On surfaces 73
T Time
With coordinate break surfaces 239
Printing on plots and text 61

Table coatings 507
TIR 292, 293
Tabulated Fresnel object 327
Frustrated 512
Tabulated object 325, 326
Total Internal Reflection 53
Talbot imaging 534
Tolerance data
Tangential
Cut, Copy, and Paste 78
Definition of 51
Loading and saving 79
Tapered coatings 503, 505
655
Tolerance sensitivity
Optimizing for 426
U
Tolerancing 196, 441
UDA Scale 69, 348
Debugging 468
Undo 82
Defining compensators for 453
Disabling 61
Increment 455
Uniform distribution 460
Irregularity 445
Units 87
Using scripts 463
Determining in ZPL macros 545
Toolbar
Setting lens units 87
Defining 63, 64
Source, irradiance, and flux 87
Tools menu 195, 223, 289
ZPL macros 595
Top Hat 527
Universal plot 180
Toroidal 228
Unlock window 37, 39, 600
Toroidal faceted object 326
Unpolarized 90
Toroidal Grating 228
Unzoom 37
Toroidal lens
Update
Paraxial model for 267
Described 85
Toroidal lens object 301, 327
Update all
Toroidal surfaces 229, 272, 273
Described 85
Gratings placed on 248, 273
Updates 61
Thermal effects 494
Use polarization
Torus surface 321, 328
Polarization state 90
Torus volume 321, 328
User defined
Total Indicator Runout
Apertures and obscurations 69, 352
Tolerancing 445
Optimization operands 431
Total integrated scatter 204
Surface 275
Total Internal Reflection
Surface apodization 277
ZPL Macros 583, 584
User defined object 329
Total track
User defined sources 340
Defined 53
User defined surface
Operand 417
Diffractive optics 277
Transfer Functions 115, 116, 117, 120, 121, 122, 125,
DLL 67
126, 127, 128
Gradient index 277
Transmission 183, 484
Lenslet arrays 277
Internal 163, 179
Refractors and reflectors 276
Of coatings 185
Using linked code 275
Transmission data
Using look up table 258, 259
Glass catalogs 484
Transparent objects V
Opacity of 348
Transparent surfaces Vacuum
Opacity of 67 Index of refraction 491
Transverse aberrations Modeling 492
Calculating 164 Variable line space gratings 282
Transverse color Variables
Calculating 164 Defining within ZPL macros 596
Triangle object 329 Removing 195
Triangular corner object 329 Setting and removing 74
True color ZPL macros 586
PSF 127 Variance
Tutorial 32 Wavefront error 165
Vector variables
Setting the size of 596
656
Vignetting 68, 73, 109, 140
ZPL macros 583, 584, 596
Y
Vignetting factors
YNU tracing 174
Adjusting during optimization 416
Y-Ybar diagrams 161
Defined 53
Field curvature and distortion 158 Z
Multi-configuration values 472
Removing 206 ZBF beam file
Setting 96 Decentering 603
Virtual propagation 55 Properties of 603
Volume Resampling 603
Of an element 219 ZBF beam files
Operand 418 Macro commands related to 602
Volume detector 335 ZBF file format 528
Volume sources 336 ZBF files
Volumes Macro commands for 572, 603
Discussion of 220 ZEMAX Beam File 528
Volumetric scattering 349 ZEMAX Programming Language
Voxels 314, 335 ZPL macros 535
Zernike
W Deformed surfaces 282, 283, 284, 285
Zernike coefficients
Watts 289
Calculation of 169, 171
Wavefront aberrations 164
Fringe 167
Wavefront error 132, 133, 134
Optimization of 420
Peak to valley 165
Standard 169
RMS 165
Zernike phase surface
Wavefront propagation 515
Optimizing coefficients 283
Wavelengths
Zernike surface 331
Conventions for 55, 491
Optimizing coefficients 244, 283
Weights for 96
Zone plates 286
ZPL macros 576
Zoom and pan 39
Weight
Zoom in, out 37
Configuration 472
Zoom lenses
Of lenses 220
Maintaining length of 382
Wide angle apertures
Maximum semi-diameter solve 384
In object space 85
Multi-configuration feature 471
Window size
Operand 420
Changing default 59, 60
Phantom elements 475
Windows
Zoom positions
Editor properties 61
Changing 97
Graphic 35
On plots 58
Resizing and moving 35
Plotting all 102
Shortcuts 40
Showing number on plots 58, 59
Text 35
ZPL macros 223, 224, 535
Types of 35
Editing 535
Wire frame model 103
Examples 604
Working F/# 85
Optimization of 429
Defined 55
Printing 535
Method of computing 94
Status message 535
Operand 418
String variables 547
Paraxial 51
ZRD file format 367
657

Ze Max Manual

Uploaded by

Copyright:

Available Formats

Ze Max Manual

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Ze Max Manual

Uploaded by

Copyright:

Available Formats

What is ZEMAX used for?

What is ZEMAX used for?

What does the manual cover?

What does the manual cover?

ZEMAX

Optical Design Program

ZEMAX Development Corporation

ZEMAX is a registered trademark of ZEMAX Development Corporation Copyright ZEMAX Devel-

About this document

!This feature is only available in the EE edition of ZEMAX.

Author Title Publisher

Hecht Optics Addison Wesley

35 Chapter 2: USER INTERFACE

Chapter 2: USER INTERFACE 36

37 Chapter 2: USER INTERFACE

Chapter 2: USER INTERFACE 38

39 Chapter 2: USER INTERFACE

Ctrl + Tab Moves the input focus from window to window.

Chapter 2: USER INTERFACE 40

41 Chapter 2: USER INTERFACE

Enter Equivalent to pressing the highlighted or default button in a dialog box.

Using the Windows clipboard

Chapter 2: USER INTERFACE 42

43 Chapter 2: USER INTERFACE

45 Chapter 3: CONVENTIONS AND DEFINITIONS

Chapter 3: CONVENTIONS AND DEFINITIONS 46

where l, m, and n are the x, y, and z direction cosines.

47 Chapter 3: CONVENTIONS AND DEFINITIONS

Chapter 3: CONVENTIONS AND DEFINITIONS 48

49 Chapter 3: CONVENTIONS AND DEFINITIONS

n sin = n' sin ' .

For small angles this becomes

Chapter 3: CONVENTIONS AND DEFINITIONS 50

51 Chapter 3: CONVENTIONS AND DEFINITIONS

Chapter 3: CONVENTIONS AND DEFINITIONS 52

53 Chapter 3: CONVENTIONS AND DEFINITIONS

P x = VDX + Px ( 1 VCX ) , and

P x = P x cos P y sin , and

Chapter 3: CONVENTIONS AND DEFINITIONS 54

55 Chapter 3: CONVENTIONS AND DEFINITIONS

57 Chapter 4: FILE MENU

Chapter 4: FILE MENU 58

59 Chapter 4: FILE MENU

Chapter 4: FILE MENU 60

61 Chapter 4: FILE MENU

Chapter 4: FILE MENU 62

63 Chapter 4: FILE MENU

Chapter 4: FILE MENU 64

65 Chapter 5: EDITORS MENU

Chapter 5: EDITORS MENU 66

67 Chapter 5: EDITORS MENU

Chapter 5: EDITORS MENU 68

69 Chapter 5: EDITORS MENU

Entity Name/Syntax Description

Chapter 5: EDITORS MENU 70

71 Chapter 5: EDITORS MENU

Surface properties scattering tab

Chapter 5: EDITORS MENU 72

!The order of the decenters and tilts matters!

73 Chapter 5: EDITORS MENU