Tolerancing in Zemax

Lecture 4

Objectives: Lecture 4
At the end of this lecture you should:
1. Understand the reason for tolerancing
and its relation to typical manufacturing
errors
2. Be able to perform a Sensitivity Analysis
and Inverse Sensitivity Analysis on a new
design
3. Be able to interpret the data from a
Monte Carlo tolerancing analysis of a
new design
March 5, 2009 Optical Systems Design 2
 Motivation
 Having designed a lens, it is important to
know how it will perform once it is built.
 Tolerancing a lens is a very important skill
to have.
 Two approaches:
 Perturbing each element individually and
reoptimizing the system each time. Slow but
accurate. Determines the sensitivities of each
element.
 Find all the sensitivities at once by using
Zemax’s tolerancing function. This method is
very fast, but there is a lot of room for
mistakes.
March 5, 2009 Optical Systems Design 3

Optical System Tolerancing

1. Define quantitative figures of merit for the
requirements
2. Estimate component manufacturing
tolerances
3. Define assembly/alignment procedure and
estimate mechanical alignment tolerances
4. Calculate sensitivities, estimate
performance
5. Adjust tolerances, keeping cost and
schedule in mind
March 5, 2009 Optical Systems Design 4
 System Figure of Merit
 Keep this as simple as possible
 Must propagate all performance specs
through to assembly
 Typical requirements:
 RMSWE (root mean square wavefront error)
 MTF at particular spatial frequencies
 Distortion
 Fractional encircled energy
 Beam divergence
 Geometric RMS image size
 Dimensional limits
March 5, 2009 Optical Systems Design 5

Dimensional Tolerances for

Machined Parts
 Depends on fabrication methods and
equipment
 Rules of thumb for machined parts:
 ± 1 mm for coarse dimensions that are not
important
 ± 0.25 mm for typical machining without
difficulty
 ± 0.025 mm precision machining, readily
accessible
 < ± 0.002 mm high-precision, requires special
tooling
March 5, 2009 Optical Systems Design 6
 Dimensional Tolerances for
Optical Elements
 Diameter
 Clear aperture
 Thickness
 Wedge Angles
 wedge or optical deviation for lenses
 angles for prisms
 Bevels
 Mounting surfaces

Start with nominal tolerances from lens fabricator

March 5, 2009 Optical Systems Design 7

Tolerancing Surface Shape

 Specifications are based on measurement:
 Inspection with test plate:
 Typical spec: 0.5 fringe
 Measurement with phase shift interferometer:
 Typical spec: 0.05 λ rms

 For most diffraction-limited systems, rms surface

gives a good figure of merit
 Special systems require a Power Spectral Density
(PSD) spec
 Aspheric systems really need a slope spec, but
this is uncommon. Typically, assume the surface
irregularities follow low order forms and simulate
them using Zernike polynomials
March 5, 2009 Optical Systems Design 8
 Rules of Thumb for Optical
Assemblies

Base: Typical, no cost impact for reducing tolerances beyond this.

Precision: Requires special attention, but easily achievable in most
shops, may cost 25% more
High precision: Requires special equipment or personnel, may cost
100% more
March 5, 2009 Optical Systems Design 9

Rules of Thumb for Lens

Tolerances

Base: Typical, no cost impact for reducing tolerances beyond this.

Precision: Requires special attention, may cost 25% more
High precision: Requires special equipment may cost 100% more

March 5, 2009 Optical Systems Design 10

 Rules of Thumb for Glass
Tolerances

Base: Typical, no cost impact for reducing tolerances beyond this.

Precision: Requires special attention, may cost 25% more
High precision: Requires special equipment, may cost 100% more

March 5, 2009 Optical Systems Design 11

Zemax Tolerancing
Capabilities
 Can set tolerances in the tolerance data editor
for a wide variety of parameters
 The default tolerance generator can
automatically enter tolerances for: radius of
curvature, surface form, lens thickness,
position, x and y tilt, x and y decentre,
irregularity, wedge, glass index, Abbe
number, and more.
 Can define what compensators to use (e.g.
focus, tilt, position of any optical element) in
sensitivity analysis
 Can select the tolerance criteria (e.g. RMS
wavefront, RMS spot radius)

March 5, 2009 Optical Systems Design 12

 Zemax Tolerancing Tools

 ZEMAX conducts an analysis of the

tolerances using any or all of these
three tools:
 Sensitivity Analysis
 Inverse Sensitivity Analysis
 Monte Carlo Analysis

March 5, 2009 Optical Systems Design 13

I: Sensitivity Analysis
 The sensitivity analysis considers each
defined tolerance independently.
 Parameters are adjusted to the limits of
the tolerance range, and then the
optimum value of each compensator is
determined.
 A table is generated listing the
contribution of each tolerance to the
performance loss.

March 5, 2009 Optical Systems Design 14

 II: Inverse Sensitivity Analysis
 The inverse sensitivity analysis
iteratively computes the tolerance
limits on each parameter when the
maximum or incremental
degradation in performance is
defined.
 Limits may be overall or specific to
each field or configuration.

March 5, 2009 Optical Systems Design 15

III: Monte Carlo

 Monte Carlo analysis is extremely powerful and
useful because all tolerances are considered at
once.
 Random systems are generated using the
defined tolerances.
 Every parameter is randomly perturbed using
appropriate statistical models, all compensators
are adjusted, and then the entire system is
evaluated with all defects considered.
 User defined statistics based upon actual
fabrication data is supported.
 ZEMAX can quickly simulate the fabrication of
large numbers of lenses and reports statistics on
simulated manufacturing yields.
March 5, 2009 Optical Systems Design 16
 Zemax Example
 Open the file DOUBLET-LECT4.ZMX
 Go toEditors drop down menu and
open the Tolerance Data Editor
 Select Default Tolerances from Tools
drop down menu

March 5, 2009 Optical Systems Design 17

Default Tolerances
• Change the default tolerances so that they look like the
screen on the right hand side

March 5, 2009 Optical Systems Design 18

 Tolerance Data Editor
Window

March 5, 2009 Optical Systems Design 19

Tolerance Data Editor

 Here you adjust each of the
tolerances.
 Columns:
 1) Operand number
 2) Four letter mnemonic for the tolerance
 3) Surface number for tolerance
 4) and 5) Skip for now
 6) Nominal value
 7) and 8) Minimum and Maximum
perturbations
 9) Comments

March 5, 2009 Optical Systems Design 20

 Tolerance Mnemonics
 Tolerance
operands
tell ZEMAX
which
parameters
in the
system to
change.
 ZEMAX uses
four letter
mnemonics
for the basic
tolerances

March 5, 2009 Optical Systems Design 21

Zemax Tolerancing
 Choose Tolerancing from
the Zemax Tools drop
down menu
 Select the mode:
Sensitivity (default)
 Select the Criteria: RMS
Wavefront,
 Select the Compensator:
Paraxial focus (default)
 Check Force Ray Aiming
On (slower but more
accurate)
 Check Show
Compensators (to see
how much focus changes
for example).

March 5, 2009 Optical Systems Design 22

 Tolerancing Results

 Here
Here areare the
the results:
results:

 Analysis
Analysis of of Tolerances
Tolerances




File
File :: C:\Documents
Title:
C:\Documents and
Title: Focusing
Focusing doublet
doublet
and Settings\shvisc\Desktop\HW4.zmx
Settings\shvisc\Desktop\HW4.zmx Numbers needed to calculate the sensitivities:

 Date
Date :: TUE
TUE DEC
DEC 55 2006 2006
 Units
Units are
are Millimeters.
Millimeters.
Perturbations Change in merit function


 Paraxial
Paraxial Focus
Focus compensation
compensation only. only.

 WARNING:
WARNING: Boundary
Boundary constraints
constraints onon compensators
compensators will will be
be ignored.
ignored.

 Criteria
Criteria :: RMS
RMS Wavefront
Wavefront Error
Error in
in waves
waves

 Mode
Mode : Sensitivities
: Sensitivities

 Sampling
Sampling :: 20
20

 Nominal
Nominal Criteria
Criteria :: 0.00065152
0.00065152

 Test
Test Wavelength
Wavelength :: 0.6328 0.6328

 Fields:
Fields: YY Symmetric
Symmetric Angle Angle inin degrees
degrees

 ## X-Field
X-Field Y-Field
Y-Field Weight
Weight VDX VDX VDY VDY VCX VCX VCY VCY

 11 0.000E+000
0.000E+000 0.000E+0000.000E+000 1.000E+000
1.000E+000 0.0000.000 0.000
0.000 0.000
0.000 0.000
0.000

 Sensitivity
Sensitivity Analysis:
Analysis:

 |-----------------
|----------------- Minimum
Minimum ----------------|
----------------| |-----------------
|----------------- Maximum
Maximum ----------------|
----------------|

 Type
Type Value
Value Criteria
Criteria Change
Change Value
Value Criteria
Criteria Change
Change

 TRAD
TRAD 22 -0.100000000
-0.100000000 0.001543200
0.001543200 0.000891675
0.000891675 0.100000000
0.100000000 0.001773699
0.001773699 0.001122174
0.001122174

 TRAD
TRAD 33 -0.100000000
-0.100000000 0.000721251
0.000721251 6.9726E-005
6.9726E-005 0.100000000
0.100000000 0.000642781
0.000642781 -8.7431E-006
-8.7431E-006

 TRAD
TRAD 44 -0.100000000
-0.100000000 0.002301125
0.002301125 0.001649600
0.001649600 0.100000000
0.100000000 0.002559580
0.002559580 0.001908056
0.001908056

 TRAD
TRAD 55 -0.100000000
-0.100000000 0.000888400
0.000888400 0.000236876
0.000236876 0.100000000
0.100000000 0.000731590
0.000731590 8.0066E-005
8.0066E-005

 TTHI
TTHI 22 00 -0.100000000
-0.100000000 0.001036787
0.001036787 0.000385263
0.000385263 0.100000000
0.100000000 0.000845486
0.000845486 0.000193961
0.000193961

 TTHI
TTHI 3 03 0 -0.200000000 0.004974029 0.004322505 0.200000000
-0.200000000 0.004974029 0.004322505 0.200000000 0.004698490 0.004046966 0.004698490 0.004046966

 TTHI
TTHI 44 00 -0.100000000
-0.100000000 0.000693073
0.000693073 4.1548E-005
4.1548E-005 0.100000000
0.100000000 0.000828411
0.000828411 0.000176886
0.000176886

 TEDX
TEDX 22 33 -0.100000000
-0.100000000 0.009626142
0.009626142 0.008974618
0.008974618 0.100000000
0.100000000 0.009626142
0.009626142 0.008974618
0.008974618

 TETX
TETX 22 33 -0.100000000
-0.100000000 0.005722684
0.005722684 0.005071160
0.005071160 0.100000000
0.100000000 0.005722684
0.005722684 0.005071160
0.005071160

 TEDX
TEDX 44 55 -0.100000000
-0.100000000 0.009681031
0.009681031 0.009029507
0.009029507 0.100000000
0.100000000 0.009681031
0.009681031 0.009029507
0.009029507

 TETX
TETX 4 5 4 5 -0.100000000 0.011549130 0.010897605 0.100000000
-0.100000000 0.011549130 0.010897605 0.100000000 0.011549130 0.010897605 0.011549130 0.010897605

 TIRX
TIRX 22 -0.100000000
-0.100000000 0.017751740
0.017751740 0.017100215
0.017100215 0.100000000
0.100000000 0.017751740
0.017751740 0.017100215
0.017100215

 TIRX
TIRX 33 -0.100000000
-0.100000000 0.031032656
0.031032656 0.030381132
0.030381132 0.100000000
0.100000000 0.031032656
0.031032656 0.030381132
0.030381132

 TIRX
TIRX 44 -0.100000000
-0.100000000 0.063785240
0.063785240 0.063133716
0.063133716 0.100000000
0.100000000 0.063785240
0.063785240 0.063133716
0.063133716

 TIRX
TIRX 55 -0.100000000
-0.100000000 0.034639912
0.034639912 0.033988387
0.033988387 0.100000000
0.100000000 0.034639912
0.034639912 0.033988387
0.033988387

 TIND
TIND 22 -0.000500000
-0.000500000 0.000890818
0.000890818 0.000239294
0.000239294 0.000500000
0.000500000 0.000735714
0.000735714 8.4190E-005
8.4190E-005

 TIND
TIND 44 -0.000500000
-0.000500000 0.000814628
0.000814628 0.000163103
0.000163103 0.000500000
0.000500000 0.000998033
0.000998033 0.000346509
0.000346509

March 5, 2009 Optical Systems Design 23

Tolerancing Results

 Worst
Worst offenders:
offenders:

 Type
Type Value
Value Criteria
Criteria Change
Change






TIRX
TIRX 44
TIRX
TIRX 4
TIRX
TIRX 55
4
-0.100000000
-0.100000000 0.063785240
0.100000000
-0.100000000
0.063785240 0.063133716
0.100000000 0.063785240
0.063133716
0.063785240 0.063133716
-0.100000000 0.034639912
0.063133716
0.034639912 0.033988387
0.033988387
Worst Offenders

 TIRX
TIRX 55 0.100000000
0.100000000 0.034639912
0.034639912 0.033988387
0.033988387

 TIRX
TIRX 33 -0.100000000
-0.100000000 0.031032656
0.031032656 0.030381132
0.030381132

 TIRX
TIRX 33 0.100000000
0.100000000 0.031032656
0.031032656 0.030381132
0.030381132

 TIRX
TIRX 22 -0.100000000
-0.100000000 0.017751740
0.017751740 0.017100215
0.017100215

 TIRX
TIRX 22 0.100000000
0.100000000 0.017751740
0.017751740 0.017100215
0.017100215

 TETX
TETX 44 55 -0.100000000
-0.100000000 0.011549130
0.011549130 0.010897605
0.010897605

 TETX
TETX 44 55 0.100000000
0.100000000 0.011549130
0.011549130 0.010897605
0.010897605

 Estimated
Estimated Performance
Performance Changes Changes based
based upon
upon Root-Sum-Square
Root-Sum-Square method:
method:

 Nominal
Nominal RMS RMS Wavefront
Wavefront :: 0.000651524
0.000651524

 Estimated
Estimated change
change :: 0.081762126
0.081762126

 Estimated
Estimated RMS RMS Wavefront
Wavefront :: 0.082413650
0.082413650

 Compensator
Compensator Statistics:Statistics:

 Change
Change in in back
back focus: focus:

 Minimum
Minimum :: -0.327629
-0.327629

 Maximum
Maximum :: 0.327965
0.327965

 Mean
Mean :: 0.000030
0.000030

 Standard
Standard Deviation
Deviation :: 0.105018
0.105018
Monte
Monte Carlo
Carlo Analysis:
Monte Carlo

 Analysis:

 Number
Number of of trials:
trials: 20 20

 Initial
Initial Statistics:
Statistics: Normal
Normal Distribution
Distribution

 Trial
Trial Criteria
Criteria Change
Change

 11 0.045548742
0.045548742 0.0448972180.044897218

 22 0.013286277
0.013286277 0.0126347520.012634752

 33 0.036228419
0.036228419 0.0355768940.035576894

 44 0.009442727
0.009442727 0.0087912030.008791203

 55 0.014894832
0.014894832 0.0142433070.014243307

 66 0.020252474
0.020252474 0.0196009490.019600949

 77 0.047652045
0.047652045 0.0470005210.047000521

 88 0.013279680
0.013279680 0.0126281560.012628156

 99 0.009529791
0.009529791 0.0088782660.008878266

 10
10 0.088488208
0.088488208 0.0878366840.087836684

 11
11 0.019946472
0.019946472 0.0192949470.019294947

 12
12 0.014766018 0.014114493
0.014766018 0.014114493

 13
13 0.008394405
0.008394405 0.0077428810.007742881

 14
14 0.069265579
0.069265579 0.0686140550.068614055
15
15 0.005727527
0.005727527 0.005076003
March 5, 2009



 16
16 0.026195678
0.026195678 0.025544154
0.005076003
0.025544154
Optical Systems Design 24

 17 0.009141888
17 0.009141888 0.008490364 0.008490364

 18
18 0.009603029
0.009603029 0.0089515040.008951504

 19
19 0.051217993
0.051217993 0.0505664690.050566469

 20
20 0.037339341
0.037339341 0.0366878160.036687816

 Nominal
Nominal 0.000651524
0.000651524

 Best
Best 0.005727527
0.005727527 Trial Trial 1515

 Worst
Worst 0.088488208
0.088488208 Trial Trial 1010

 Mean
Mean 0.027510056
0.027510056

 Std
Std Dev
Dev 0.022263515
0.022263515

 Compensator
Compensator Statistics:Statistics:

 Change
Change in in back
back focus: focus:

 Minimum
Minimum :: -0.391712
-0.391712

 Maximum
Maximum :: 0.342581
0.342581

 Mean
Mean :: 0.022387
0.022387

 Standard
Standard Deviation
Deviation :: 0.217850
0.217850

 90% <= 0.051217993
90% <= 0.051217993

 50%
50% <= <= 0.014894832
0.014894832

 10%
10% <= <= 0.008394405
0.008394405

 End
End of of Run.
Run.
 Tolerancing Analysis
• From these numbers, can calculate
the sensitivities by dividing the
change in the criteria (RMS
wavefront) by the perturbation.

March 5, 2009 Optical Systems Design 25

Summary: Lecture 4
 Tolerancing is a critical step to ensure that
a lens design can be manufactured and
to predict its expected performance
 Difficult because it involves complex
relationships across diferent disciplines
 Zemax has many very powerful design
tolerancing capabilities
 Important to understand how Zemax
does the sensitivity analysis before you
can blindly use it.
March 5, 2009 Optical Systems Design 26
 Exercises: Lecture 4
 Perform tolerance analysis of the Cooke
triplet lens designed in the exercise for
Lecture 3
 Use precision mechanical dimensional
tolerances and λ/20 RMS surface form error
 What is the mean increase in RMS spot
radius from the Monte Carlo simulation ?
 Which are the three most critical
dimensional tolerances ?

March 5, 2009 Optical Systems Design 27