Bentley Field Guide To Lens Design PDF

Field Guide to
Lens Design
Julie Bentley
Craig Olson
SPIE Field Guides

Volume FG27
John E. Greivenkamp, Series Editor
Bellingham, Washington USA
FG27 covers and title.indd 3
2/24/12 8:18 AM
Library of Congress Cataloging-in-Publication Data

Bentley, Julie (Julie L.)
Field guide to lens design / Julie Bentley, Craig Olson.
pages cm. (The field guide series)
Includes bibliographical references and index.
ISBN 978-0-8194-9164-0
1. LensesDesign and construction. I. Olson, Craig 1971II. Title.
QC385.B43 2012
6810.423dc23
2012035700
Published by
SPIE
P.O. Box 10
Bellingham, Washington 98227-0010 USA
Phone: +1.360.676.3290
Fax: +1.360.647.1445
Email: [email protected]
Web: http://spie.org
2012 Society of Photo-Optical Instrumentation Engi-
neers (SPIE)
All rights reserved. No part of this publication may be reproduced or distributed in any form or by any means without written permission of the publisher.
The content of this book reflects the work and thought of
the author. Every effort has been made to publish reliable
and accurate information herein, but the publisher is not
responsible for the validity of the information or for any
outcomes resulting from reliance thereon. For the latest
updates about this title, please visit the books page on our
website.
Printed in the United States of America.
First printing
Introduction to the Series

Welcome to the SPIE Field Guidesa series of publications written directly for the practicing engineer or scientist. Many textbooks and professional reference books
cover optical principles and techniques in depth. The aim
of the SPIE Field Guides is to distill this information,
providing readers with a handy desk or briefcase reference that provides basic, essential information about optical principles, techniques, or phenomena, including definitions and descriptions, key equations, illustrations, application examples, design considerations, and additional
resources. A significant effort will be made to provide a
consistent notation and style between volumes in the series.
Each SPIE Field Guide addresses a major field of optical
science and technology. The concept of these Field Guides
is a format-intensive presentation based on figures and
equations supplemented by concise explanations. In most
cases, this modular approach places a single topic on a
page, and provides full coverage of that topic on that page.
Highlights, insights, and rules of thumb are displayed in
sidebars to the main text. The appendices at the end of
each Field Guide provide additional information such as
related material outside the main scope of the volume,
key mathematical relationships, and alternative methods.
While complete in their coverage, the concise presentation
may not be appropriate for those new to the field.
The SPIE Field Guides are intended to be living
documents. The modular page-based presentation format
allows them to be easily updated and expanded. We are
interested in your suggestions for new Field Guide topics
as well as what material should be added to an individual
volume to make these Field Guides more useful to you.
Please contact us at [email protected].
John E. Greivenkamp, Series Editor
Optical Sciences Center
The University of Arizona
Field Guide to Lens Design
The Field Guide Series

Keep information at your fingertips with all of the titles in
the Field Guide Series:
Adaptive Optics, Second Edition, Robert Tyson &
Benjamin Frazier
Atmospheric Optics, Larry Andrews
Binoculars and Scopes, Paul Yoder, Jr. &
Daniel Vukobratovich
Diffractive Optics, Yakov Soskind
Geometrical Optics, John Greivenkamp
Illumination, Angelo Arecchi, Tahar Messadi, &
John Koshel
Image Processing, Khan M. Iftekharuddin &
Abdul Awwal
Infrared Systems, Detectors, and FPAs, Second Edition,
Arnold Daniels
Interferometric Optical Testing, Eric Goodwin &
Jim Wyant
Laser Pulse Generation, Rdiger Paschotta
Lasers, Rdiger Paschotta
Microscopy, Tomasz Tkaczyk
Optical Fabrication, Ray Williamson
Optical Fiber Technology, Rdiger Paschotta
Optical Lithography, Chris Mack
Optical Thin Films, Ronald Willey
Optomechanical Design and Analysis, Katie Schwertz
& James Burge
Polarization, Edward Collett
Probability, Random Processes, and Random Data
Analysis, Larry Andrews
Radiometry, Barbara Grant
Special Functions for Engineers, Larry Andrews
Spectroscopy, David Ball
Terahertz Sources, Detectors, and Optics,
Cridhe OSullivan & J. Anthony Murphy
Visual and Ophthalmic Optics, Jim Schwiegerling

Optical design has a long and storied history, from the
magnifiers of antiquity, to the telescopes of Galileo and
Newton at the onset of modern science, to the ubiquity
of modern advanced optics. The process for designing
lenses is often considered both an art and a science. While
advancements in the field over the past two centuries
have done much to transform it from the former category
to the latter, much of the lens design process remains
encapsulated in the experience and knowledge of industry
veterans. This Field Guide provides a working reference
for practicing physicists, engineers, and scientists for
deciphering the nuances of basic lens design. Because the
optical design process is historically (and quite practically)
closely related to ray optics, this book is intended as
a companion to the Field Guide to Geometrical Optics,
in which first-order optics, thin lenses, and basic optical
systems are treated in more detail. Note that this compact
reference is not a substitute for a comprehensive technical
library or the experience gained by sitting down and
designing lenses.
This material was developed over the course of several
years for undergraduate and graduate lens design classes
taught at the University of Rochester. It begins with
an outline of the general lens design process before
delving into aberrations, basic lens design forms, and
optimization. An entire section is devoted to techniques
for improving lens performance. Sections on tolerancing,
stray light, and optical systems are followed by an
appendix covering related topics such as optical materials,
nonimaging concepts, designing for sampled imaging, and
ray tracing fundamentals, among others.
Thanks to both of our familiesDanielle, Alison, Ben,
Sarah, Julia, and especially our spouses, Jon and Kelly.
The cats will now get fed, and all soccer parents beware!
Julie Bentley
University of Rochester
Craig Olson
L-3 Communications
vii
Table of Contents
Glossary of Symbols and Acronyms
xi
Fundamentals of Optical Design

Sign Conventions
Basic Concepts
Optical Design Process
Aperture and Wavelength Specifications
Resolution and Field of View
Packaging and Environment
Wave Aberration Function
Third-Order Aberration Theory
Spot Diagram and Encircled Energy
Transverse Ray Plot
Wavefront or OPD Plots
Point Spread Function and Strehl Ratio
MTF Basics
Using MTF in Lens Design
Defocus
Wavefront Tilt
Spherical Aberration
Coma
Field Curvature
Petzval Curvature
Astigmatism
Distortion
Primary Color and Secondary Color
Lateral Color and Spherochromatism
Higher-Order Aberrations
Intrinsic and Induced Aberrations
1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
Design Forms
Selecting a Design Form: Refractive
Selecting a Design Form: Reflective
Singlets
Achromatic Doublets
Airspaced Doublets
Cooke Triplet
Double Gauss
Petzval Lens
Telephoto Lenses
Retrofocus and Wide-Angle Lenses
Refractive versus Reflective Systems
Obscurations
Newtonian and Cassegrain
27
27
28
29
30
31
32
33
34
35
36
37
38
39
viii
Table of Contents
Gregorian and Schwarzschild
Catadioptric Telescope Objectives
Unobscured Systems: Aperture Clearance
Unobscured Systems: Field Clearance
Three-Mirror Anastigmat
Reflective Triplet
Wide-Field Reflective Design Forms
Zoom Lens Fundamentals
Zoom Lens Design and Optimization
40
41
42
43
44
45
46
47
48
Improving a Design
Techniques for Improving an Optical Design
Angle of Incidence and Aplanatic Surfaces
Splitting and Compounding
Diffraction-Limited Performance
Thin Lens Layout
Lens Bending
Material Selection
Controlling the Petzval Sum
Stop Shift and Stop Symmetry
Telecentricity
Vignetting
Pupil Aberrations
Aspheres: Design
Aspheres: Fabrication
Gradient Index Materials
Diffractive Optics
49
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
Optimization
Optimization
Damped Least Squares
Global Optimization
Merit Function Construction
Choosing Effective Variables
Solves and Pickups
Defining Field Points
Pupil Sampling
65
65
66
67
68
69
70
71
72
Tolerancing
Tolerancing
Design Margin and Performance Budgets
Optical Prints
Radius of Curvature Tolerances
73
73
74
75
76
ix
Table of Contents
Surface Irregularity
Center Thickness and Wedge Tolerances
Material and Cosmetic Tolerances
Lens Assembly Methods
Assembly Tolerances
Compensators
Probability Distributions
Sensitivity Analysis
Performance Prediction
Monte Carlo Analysis
Environmental Analysis
Athermalization
Stray Light
Stray Light Analysis
Stray Light Reduction
Antireflection (AR) Coatings
Ghost Analysis
Cold Stop and Narcissus
Nonsequential Ray Tracing
Scattering and BSDF
77
78
79
80
81
82
83
84
85
86
87
88
89
89
90
91
92
93
94
95
Optical Systems
Photographic Lenses: Fundamentals
Photographic Lenses: Design Constraints
Visual Instruments and the Eye
Eyepiece Fundamentals
Eyepiece Design Forms
Telescopes
Microscopes
Microscope Objectives
Relays
96
96
97
98
99
100
101
102
103
104
Appendix: Optical Fundamentals

Index of Refraction and Dispersion
Optical Materials: Glasses
Optical Materials: Polymers/Plastics
Optical Materials: Ultraviolet and Infrared
Snells Law and Ray Tracing
Focal Length, Power, and Magnification
Aperture Stop and Field Stop
Entrance and Exit Pupils
Marginal and Chief Rays
105
105
106
107
108
109
110
111
112
113
Table of Contents
Zernike Polynomials
Conic Sections
Diffraction Gratings
Optical Cements and Coatings
Detectors: Sampling
Detectors: Resolution
The Lagrange Invariant and tendue
Illumination Design
Equation Summary
Bibliography
Index
114
115
116
117
118
119
120
121
122
127
129
xi

A
AOI
AR
BBAR
BFL
BFS
BRDF
BSDF
BTDF
c
C
CA
CCD
CDF
CGH
CMOS
CRA
CT
CTE
CTF
d
d
DLS
dn/ dT
DOE
EFL
EPD
ESF
ETD
f
f /#
FEA
Area
Angle of incidence
Antireflection
Broadband antireflection coating
Back focal length
Best fit sphere
Bidirectional reflectance distribution
function
Bidirectional scattering distribution
function
Bidirectional transmittance distribution
function
Surface curvature
Lens conjugate factor
Clear aperture
Charge-coupled device
Cumulative distribution function
Computer-generated hologram
Complementary metal-oxide
semiconductor
Chief ray angle
Center thickness
Coefficient of thermal expansion
Contrast transfer function
Airspace
Thickness
Damped least squares
Thermo-optic coefficient
Diffractive optical element
Effective focal length
Entrance pupil diameter
Edge-spread function
Edge thickness difference
Focal length
f -number or relative aperture
Finite-element analysis
xii

FFL
FFOV
FFT
FOV
GQ
GRIN
h, h0
H
H
HFOV
HO
HOE
HR
i, i0
i, i a
i, i b
ID
IR
L
LOS
LR
LSF
LWIR
m
m
MP
MTF
MWIR
n, n0
n( z), n( r )
NA
NITD
NRT
NUC
Front focal length

Full field of view
Fast Fourier transform
Field of view
Gaussian quadrature
Gradient index
Object/image height
Lagrange invariant
Normalized field coordinate
Half field of view
Higher order
Holographic optical element
High-reflection
Angle of incidence w.r.t. surface normal
Marginal ray angle w.r.t. surface normal
Chief ray angle w.r.t. surface normal
Inner diameter of a lens barrel or mount
Infrared
Radiance
Line of sight
Limiting resolution
Line spread function
Long-wave infrared
Diffraction order
Magnification
Magnifying power (magnifier or
telescope)
Modulation transfer function
Midwave infrared
Index of refraction
Gradient index profile function
Numerical aperture
Narcissus-induced temperature
difference
Nonsequential ray tracing
Nonuniformity correction
xiii

OAP
OAR
OD
OPD
OTF
p
P
PDF
PSF
PSNIT
PST
PV
Q
r
R , ROC
RI
RMS
RSS
RT
s, s0
SA
SLR
t
T
TIR
TIR
TIS
TMA
TML
u, u0
u, u a
u,
ub
UV
V
W
Off-axis parabola
Off-axis rejection
Outer diameter (of a lens)
Optical path difference
Optical transfer function
Pixel pitch in sampled detector arrays
Partial dispersion
Probability distribution function
Point spread function
Point-source normalized irradiance
transmittance
Point-source transmittance
Peak to valley
Sampling ratio
Radial surface coordinate
Radius of curvature
Relative illumination
Root mean square
Root sum square
Reflective triplet
Object/image distance
Spherical aberration
Single-lens reflex
Thickness or airspace
Temperature
Total indicator runout
Total internal reflection
Total integrated scatter
Three-mirror anastigmat
Three mirror long
Paraxial ray angles w.r.t. optical axis
Marginal ray angle w.r.t. optical axis
Chief ray angle w.r.t. optical axis
Ultraviolet
Abbe number
Wave aberration function
xiv

Wi jk
WD
y, ya
y , yb
z
z( r )
Zn
, x , y
, 0
0
, x , y
Wavefront aberration coefficient

Working distance
Marginal ray height at a surface
Chief ray height at a surface
Optical axis
Surface sag/profile function
Zernike polynomial coefficient
Lens shape factor
Wavelength range or bandwidth
Defocus
Obscuration ratio
Transverse ray error
Angle of incidence/refraction
Half field of view
Pupil azimuthal coordinate
Conic constant
Wavelength
Center wavelength
Normalized radial pupil coordinate
System power
Element or surface power
Merit or penalty function
Sign Conventions
Throughout this Field Guide, a set of fully consistent sign
conventions is utilized.
The z axis is the optical axis or the axis of symmetry of
a rotationally symmetric optical system.

The y axis is in the plane of the paper, perpendicular to
the z axis, and the x axis points directly into the paper,
forming a right-handed coordinate system.

All distances are measured relative to a reference point,
line, or plane in a Cartesian sense: directed distances
above or to the right are positive; below or to the left are
negative.
Angles u are measured from the optical axis to the ray
with the smallest of the two angles chosen, while angle
of incidence i are measured from the normal of the
surface to the incident ray.
Counter-clockwise angles are positive, and clockwise
angles are negative.
The radius of curvature of a surface is defined as
the directed distance from its vertex to its center of
curvature and is therefore positive when the center of
curvature is to the right of the surface.
Light travels from left to right (from z to + z) unless
reflected.
A 0 (prime) symbol following a variable denotes a value
after refraction at a surface.
The signs of all indices of refraction following a reflection
are reversed.
To aid in the use of these conventions, all directed
distances and angles are identified by arrows with the
tail of the arrow at the reference point, line, or plane.
Basic Concepts
A lens is often described as a device that creates an image
of an object on a detector. More generally, it can be thought
of as any system that tries to collect and distribute light in
a specified way. Although a lens can consist of a single lens
element, it is often composed of multiple lens elements of
different shapes and sizes, referred to as a lens system or
an optical system.
Optical design or lens design is the process used to
find the best set of lens construction parameters (e.g.,
radii of curvature, thicknesses, airspaces, and materials)
that optimizes the overall performance (including the
manufacturability) of an optical system.
A specification document lists the requirements needed
to design a lens system. It defines basic specifications
such as object and image location, aperture, field of
view, and wavelength range. It contains image quality
measures such as spot size, resolution, and distortion. It
also includes packaging and environmental requirements
such as diameter, length, and temperature.
Optical design software is computer code that aids
in the selection and optimization of lens construction
parameters and performance evaluation. Given a starting
point and a set of variables and constraints, the computer
program drives an optical design to a local optimum, as
defined by a merit function.
A lens designer (or ray bender) finds the variations in
lens parameters that yield the greatest improvements in
optical system performance. Computer optimization and
real ray tracing (use of Snells law to find the position
and direction of a ray after refraction or reflection at
each lens surface) are key components of the process.
A lens designer is also often responsible for developing
a specification document (and resolving any conflicts)
and identifying engineering trades between performance,
packaging constraints, tolerances, cost, and schedule.
Optical Design Process

The basic optical design process is as follows:
A specification document for the optical system is
established, and the appropriate type of lens form (e.g.,

microscope, telescope, camera lens) is selected.
A starting-point design is chosen from previous projects,
historical databases, or patent literature (primarily
based on the aperture, field, wavelength, and packaging
requirements).
If an appropriate starting-point design does not exist,
designers can choose to start from a first-order
solution that identifies focal lengths and object/image/
pupil locations and sizes using thin lenses and paraxial
rays. The first-order solution can be converted to a thirdorder thin lens solution where thin lens parameters
such as radius of curvature and index of refraction
are chosen to minimize third-order aberrations.
Thickness is then introduced into the thin lenses to
arrive at a thick lens starting point.
The starting-point design is assessed by tracing real
rays at multiple wavelengths through the system to
analyze the chromatic and higher-order aberrations.
Compliance to performance requirements (e.g., image
quality, packaging restrictions, fabrication and assembly
tolerances, cost) is evaluated.
If the performance is not acceptable, optimization is
used to improve the design. A set of variables and
constraints is defined, and computer software is used to
find design solutions as defined by a merit function.
If the solution is still unacceptable, designers must
return to a previous step and iterate until the design
meets specifications. This process may require the
selection of a different starting-point design and/or
requesting a significant change in one or more of the
specifications.
The manufacturability of the design is just as important
as the final image quality. Therefore, a complete
design process and performance evaluation includes
tolerancing, stray light analysis, and thermal and
other environmental analyses.
Aperture and Wavelength Specifications

The aperture specification determines the amount of light collected
from an object and establishes a resolution limit to the system. The three
most common ways to specify an aperture for a lens system are entrance pupil diameter
(EPD), f /#, and numerical aperture (NA).
The entrance pupil diameter defines the size of the beam
in object space and is a common way to specify the
aperture of lens systems during computer-aided design.
The ratio of the focal length of a lens to its entrance pupil
diameter is called the relative aperture or f /#.
f /# =
f
EPD
N A = n0 sin u0 .
Numerical aperture is the index of refraction of the image

space media times the sine of the half-angle cone of light.
f /# is typically used to specify systems with distant objects
(e.g., camera lenses or telescopes), whereas NA is used to
specify systems that work at finite conjugates (e.g., relay
systems or microscope objectives).
For infinite conjugate systems, f /# and NA are related
by f /# = 1/(2 N A ). Although this equation only holds for
systems with infinite object distances, the same quantity
is often calculated for finite conjugate systems and is
referred to as the working f /#.
The spectral range of an optical system is identified by
a center wavelength 0 and a bandwidth . Relative
bandwidth is the ratio of the total bandwidth to
the center wavelength. Narrow-bandwidth systems are
called monochromatic (e.g., laser collimators), whereas
larger-bandwidth systems are called polychromatic
(e.g., camera objectives). The optical materials are
ideally chosen for both high transmission at the system
wavelengths and good chromatic correction over the full
bandwidth.
Resolution and Field of View
Diffraction from a finite circular aperture results in an

Airy disc pattern. The Rayleigh criterion determines
the smallest lateral separation X between two Airy
disc patterns for which two distinct points can still be
resolved. Therefore, the fundamental resolution of a
rotationally symmetric optical system is a function of both
its wavelength and its numerical aperture. To increase
resolution, systems move to shorter wavelengths and
higher numerical apertures.
The field of view (FOV) of an optical system is specified
as an object height, object angle, or image height. The
image height can be directly correlated to the size of the
detector (e.g., film or CCD) but
can also represent the size of
the field over which the lens is
well corrected. The half field
of view (HFOV) is the radius
of the object height h or image
height h0 , whereas the full field
of view (FFOV) indicates a
diameter measure.
For lenses with infinite object distances, one can quickly
convert from angular HFOV to image height h0 using
h0 = f tan
where f is the focal length of the system. This equation
is also very useful for converting detector resolution h0
to angular resolution for a focal length f .
Packaging and Environment

The achievable imaging performance of a lens design
is often limited by packaging and interface requirements. For example, the maximum number of elements
that can be used in a design may be dictated by a maximum lens envelope (diameter and length) and/or weight
constraint. Other interface requirements that may directly or indirectly drive the design include working distance, pupil location and size, filter/window/fold mirror
position and size, and centerof-gravity location and shift
(for lenses with movable elements). Special access for
modularity (e.g., an eyepiece) or field service may
also be required.
Most lens systems have a working distance requirement
to keep the lens from crashing into an object or a sensor.
It is important to clarify if a specification references the
vertex of the lens element (back focal length or BFL),
the closest mechanical surface point to the sensor (image
clearance), or a mechanical flange surface (flange-tofocus).
Environmental conditions and outside influences such as
temperature, altitude, and vibration place additional
constraints on a lens design. Moisture or humidity
effects may induce coating delamination, absorption
and scattering from condensed water droplets, and
glass staining. Shock requirements dictate single-event
extreme conditions that the lens must survive (e.g., a
thermal shock, drop, or sudden pressure loss that can
occur during shipping). Extreme environments, such as
highly corrosive salt water, or use in a vacuum chamber
can further restrict design freedom. Lenses operating in
ionizing radiation require special radiation-hardened
glass choices because many optical glasses will fog or
become opaque over time in such an environment.
Wave Aberration Function

Geometrical optics describes the properties of lenses
when the effects of diffraction and interference are
ignored (in the limit where the wavelength approaches
zero). In this case, an object consists of a collection of
independently radiating point sources. Each point source
emits a spherical wavefront (surface of constant optical
path length from a point on the object), and rays are traced
perpendicular to the wavefront.
First-order optics is the study of perfect optical systems
without aberrations. In a perfect system, an incident
spherical wave is mapped to a perfect spherical wave in
the exit pupil, converging on the ideal image location.
Aberrations are imperfections in the image formation

of an optical system. They are measured either by the
deformation of the wavefront in the exit pupil or by
the lateral distance , by which the rays miss the ideal
image point. The distance is called the transverse ray
error, whereas the wave aberration function W is the
optical length, measured along a ray, from the aberrated
wavefront to the reference sphere.
W and are related by a partial derivative with respect to
pupil coordinate , where u0a is the angle of the marginal
ray in image space:

0y =
1 W
n0 u0a y
0x =
1 W
n0 u0a x
Third-Order Aberration Theory

In a rotationally symmetric optical system, one can specify
any ray in the system with only three coordinates: its
1D position in the field plane (typically along the y axis),
denoted by the normalized field coordinate H , and its
2D position in the pupil plane, denoted by the normalized
pupil coordinates x ( sin ) and y ( cos ).
Monochromatic aberrations are described by expanding the wave aberration function W in a power series of
aperture and field coordinates, , , and H :
WI JK H I J cosK
W ( H, , ) = W020 2 + W111 H cos + W040 4 + W131 H 3 cos
+ W222 H 2 2 cos2 + W220 H 2 2 + W311 H 3 cos + O (6)
where the other terms in the power series are forbidden

by rotational symmetry. In third-order aberration
theory (because of the derivative dependence, third-order
transverse aberrations have a fourth-order wavefront
dependence), the higher-order terms O (6) in the wave
expansion are dropped. The coefficients for the remaining
terms are traditionally defined as
W020 :
W111 :
W040 :
W131 :
Defocus
Wavefront tilt
Spherical aberration
Coma
W222 : Astigmatism
W220 : Field curvature
W311 : Distortion
The value of each polynomial coefficient (WI JK ) gives the

magnitude of the aberration and is a function of lens
parameters (e.g., index of refraction, thickness, and radius
of curvature), and marginal/chief ray heights and angles.
Spot Diagram and Encircled Energy

A spot diagram or scatter plot shows the transverse
ray error ( y versus x ) for a given object height H . Rays
are traced into the entrance pupil from a single object
point. A spot is plotted where each ray hits the image
plane with respect to the paraxial image point; several
hundred rays at different wavelengths are usually plotted.
Spot diagrams are good for final system evaluation,

especially for systems with aberrations much larger than
the diffraction limit.
The RMS spot diameter or RMS spot size is a common
metric for the quality of an image point. Many rays are
traced from a single object point, and the centroid or
center of gravity of all of the image plane intersections
is determined. The RMS spot size is then calculated by
taking the square root of the sum of the squares of each
rays distance from the centroid, divided by the total
number of rays.
Encircled energy is an image quality measure for a single
field point, using the intensity
distribution of a spot to calculate the percentage of transmitted energy through an opening
of variable radius. Ensquared
energy assumes that the opening has a square shape.
10
Transverse Ray Plot

Transverse ray aberration plots or ray intercept
plots are plots of transverse ray error versus the relative
position of the ray in the pupil ( y versus y , and x versus
x ) for different field points H . They are similar to spot
diagrams, but instead of tracing a 2D grid of rays in the
pupil, they use a tangential (Y ) and a sagittal ( X ) slice
of the pupil. A perfect lens has a straight line along the
horizontal axis in both plots.
Transverse ray plots and spot diagrams are typically

plotted for three different field points: on-axis, 0.7 field,
and full-field. The shape of the plots at different values
of H yield a lot of information about which aberrations
are present in the lens. An estimate of peak-to-valley spot
diameter () can be easily calculated from the transverse
ray plot by drawing a rectangle around the total error.
11
Wavefront or OPD Plots

The departure of an aberrated wavefront from an ideal
spherical wavefront is called the optical path difference
(OPD) and is typically measured in units of wavelength.
Wavefront or OPD plots are plots of OPD versus y
and OPD versus x for different field points H . Wavefront
plots are proportional to the integral of the transverse
ray error plot and also indicate what aberrations are
present. OPD plots are used instead of transverse ray plots
for well-corrected systems with nearly diffraction-limited
performance.
Wavefront plots show slices of the full wavefront error
of an optical system that would be measured by a test
interferometer at a given field point.
Peak-to-valley (PV)
OPD is the difference
between the wavefront
maximum and the wavefront minimum, and is a
good measure of the image quality of an optical system if the wavefront error is
smooth with small changes in slope. Most real wavefronts
are not smooth, and the RMS OPD is generally a better
measure of image quality. The RMS OPD is calculated by
taking the square root of the mean of the squared OPD
values sampled over the pupil.
The Rayleigh quarter-wave criterion states that if
the total wavefront deviation is less than a quarter of a
wavelength (PV OPD < /4), then the image quality is
good or nearly diffraction limited. Although the exact
relationship between PV OPD and RMS varies with the
shape of the wavefront, a common rule of thumb is RMS
OPD = (PV OPD/5). This is then used to define the RMS
limit, RMS OPD < /20 (0.05), for the image quality of a
good lens.
12
Point Spread Function and Strehl Ratio

The point spread function (PSF) is the impulse response function of
a lens and includes the effects of both aberrations
and diffraction. For a single point object, a PSF
plot shows the full 3D intensity distribution in the
image plane. In practice,
a line spread function (LSF) or edge spread function
(ESF) gives the intensity distribution of a line or edge object and is a more-convenient measure of the lens performance.
The Strehl ratio is defined as the ratio of the peak
intensity of an aberrated lens to the peak intensity of an
aberration-free lens. The
Strehl ratio is a good singlenumber, image-quality design measure for wellcorrected systems but is less
meaningful for highly aberrated systems.
The Marchal criterion for image quality is a Strehl

ratio of 0.82, which is equivalent to an RMS OPD of /14
or 0.071 RMS. A Strehl ratio of 0.8 corresponds closely
to a PV OPD of /4 for a lens with only defocus and/or
spherical aberration. Both criteria are commonly referred
to as diffraction limited.
For a lens with a relatively small OPD, the relationship
between the Strehl ratio and the RMS OPD is
approximated by
2 2
Strehl
= 1 2
= RMSOPD
13
MTF Basics
The modulation transfer function (MTF) is the

magnitude of the complex optical transfer function
(OTF) and is a measure of how well a lens relays contrast
from object to image. MTF is the ratio of the image sine
wave modulation to that of the object as a function of
the spatial frequency, typically specified at the image
in cycles/mm. A cycle is one full period (peak and trough)
of the sine wave. For distant objects, the spatial frequency
can be expressed in cycles/mrad.
The optical cut-off frequency is the spatial frequency
beyond which an optical system cannot transmit information. It equals 2NA/ or 1/( f /#) for incoherent light.
In a perfect image, the MTF is unity at all frequencies.
In a real image, contrast is degraded by both diffraction
and aberrations. The OTF is equal to either the Fourier
transform of the incoherent PSF or the autocorrelation of
the OPD (pupil function). In lenses with large aberrations,
the OTF can be negative, leading to MTF bounce, which
represents a phase reversal of light and dark features
(spurious resolution).
The contrast transfer function (CTF) uses a squarewave modulation to determine constrast and is specified
in line pairs/mm. A line pair consists of one black line
and one white line. MTF specifications often refer to line
pairs/mm when cycles/mm is intended because square
wave targets are easier to generate for measurement
purposes.
14
Using MTF in Lens Design
A complete MTF plot includes performance curves

for several different field points and a diffractionlimited reference line. For off-axis field points, the
MTF depends on the orientation of the lines in
the object and is therefore plotted for both spoke-like
radial modulation (sagittal S) and wheel-like azimuthal
modulation (tangential T), where the modulation
direction is orthogonal to the feature direction. The onaxis MTF is typically a single curve but will be separated
into sagittal and tangential components in lenses with
decentered or tilted elements.
To aid in developing system performance budgets, the
MTF can be indirectly linked to wave aberrations or OPD
using the Hopkins ratio, which describes the ratio of
the aberrated MTF to the diffraction-limited MTF as a
function of RMS OPD.
Imaging performance is quantified by specifying an MTF
value at a particular spatial frequency. For digital sensors,
this frequency is usually associated with the sensors
Nyquist frequency. Although the MTF plots do not
directly tell a designer which aberrations are present in
the lens, MTF is useful for specifying system performance
because it is easily measured and allows the cascading of
MTFs using linear system theory (where the camera,
atmosphere, lens, and other MTFs are multiplied to derive
the overall system MTF). However, optimizing a lens
directly for MTF is computationally intensive, and phase
reversals in the MTF can cause local minima in the merit
function and stagnated optimization.
15
Defocus
In an optical system, focus is the act of changing the
relative position of the sensor with respect to the elements
(usually to improve the sharpness of an image).
The hyperfocal distance is the object distance for a
lens that maintains the depth of focus criteria at the
image plane for all object distances between half the
hyperfocal distance and infinity (without refocusing).
Defocus is the amount of aberration introduced when

moving the image plane axially from the paraxial image
plane and is directly proportional to the amount of
movement z. Defocus is used to balance the blur of other
aberrations and improve the image quality of the system.
It is not typically considered a true aberration.
A wavefront plot for defocus shows a quadratic dependence

on aperture coordinate , whereas a transverse ray plot
shows a linear dependence on that has the same slope in
both x and y and is independent of field coordinate H .
16
Wavefront Tilt
Wavefront tilt is the difference between the paraxial

magnification and the actual magnification of the system.
There is no tilt on axis (H = 0); for objects located along the
y axis, there is no tilt in x. A wavefront plot for tilt shows
a linear dependence on aperture coordinate , whereas a
transverse ray plot shows a constant dependence on that
increases in magnitude with field coordinate H . The spot
diagram shows a perfect point image, but it is not located
at the expected paraxial image point.
The depth of focus of a lens is the longitudinal shift

of the image sensor that produces an image degradation
that is still acceptable for the application. For example,
the Rayleigh quarter-wave criterion for OPD results in a
depth of focus equal to 2( f /#)2 or /(2 N A 2 ). Similarly,
the depth of field of a lens is the allowable shift of the
object position and is related to the depth of focus by the
longitudinal magnification.
17
Spherical Aberration
Spherical aberration is a variation in focus position
between the paraxial rays through the center of the lens
aperture and those through the edge of the aperture.
Spherical aberration is the only one of the five primary

third-order aberrations to appear on axis (H = 0), and it
has the same magnitude for any field coordinate H . A
wavefront plot for spherical aberration shows a fourthorder dependence on aperture coordinate , whereas a
transverse ray plot shows a cubic dependence on that
is the same in both x and y.
Spherical aberration is the first aberration to appear

in most symmetrical lens tolerancing budgets. Because
errors in the airspaces between lens components also
result in spherical aberration, a space adjust procedure
can be devised whereby adjustable spacers are used in the
most sensitive airspaces of a design as compensators to
fix the as-built performance.
18
Coma
For symmetrical optical systems, coma is an off-axis
aberration where annular zones of the aperture have
different magnifications.
There is no coma on axis (H = 0), and for objects located

along the y axis, there is no coma in x when y equals
zero. A wavefront plot for coma shows a cubic dependence
on aperture coordinate , whereas a transverse ray plot
shows a quadratic dependence on that increases in
magnitude with field coordinate H . The spot diagram
shows that the resulting image of a point object looks like
a comet.
Coma is the first aberration to appear in most

asymmetrical lens tolerancing budgets. For example,
errors in the tilt or decenter of lens components result in
coma. Correction schemes can be devised that include an
adjustable push-around element as a compensator to
fix the as-built performance.
19
Field Curvature
A symmetrical optical system with spherical surfaces

naturally forms an image on a curved surface. Field
curvature is an off-axis aberration that describes the
departure of the image surface from a flat surface. Field
curvature looks similar to defocus; however, there is
no field curvature on axis (H = 0), and its magnitude
increases with field coordinate H . A wavefront plot for
field curvature shows a quadratic dependence on aperture
coordinate , whereas a transverse ray plot shows a linear
dependence on that has the same slope in both x and y.
A field flattener is a lens (usually negative) placed close

to the image plane where the marginal ray height is
small. These lenses are used to reduce the field curvature
without significantly changing the image size or adding
spherical aberration.
20
Petzval Curvature
1
The field curvature coefficient
W220 = W222
+ W220 p
W220 can be separated into two
2
components, where W222

is a term whose functional
dependence is equivalent to the astigmatism coefficient
(W222 ), and W220 p is the portion of field curvature due to
Petzval curvature.
The separation of field curvature into two distinct terms is
useful during lens design, as the techniques for correcting
the two aberrations are significantly different. The
amount of astigmatism introduced by a lens element is a
function of the power and shape of the lens and its distance
from the aperture stop, whereas the amount of Petzval
curvature introduced by that same element is purely a
function of lens power and index of refraction. Petzval
curvature is the only aberration completely independent
of ray quantities (angles and heights); this makes Petzval
curvature one of the most difficult aberrations to correct
once the first-order power balance in the design has
been established. The key to fixing the Petzval curvature
(thereby reducing the overall field curvature) in optical
design is to change the first-order power balance between
positive and negative power.
In most positive-power refracting lenses, the Petzval
surface will be curved toward the lens (known as an
inward-curving field). If there is no astigmatism in the
lens, the field curvature reduces to the Petzval curvature.
If Petzval curvature cannot be corrected, overcorrecting
astigmatism can be introduced to artificially flatten the
tangential field, leaving the sagittal field to fall between
the image plane and the Petzval surface, and yielding the
smallest amount of residual field curvature.
Some optical systems use a curved image surface (e.g.,
the retina) to significantly reduce the effects of Petzval.
Modern-day curved digital sensors have the potential to
dramatically simplify lens systems by removing the need
to correct Petzval in the design.
21
Astigmatism
Astigmatism is the aberration that causes a tangential fan of rays to focus at a different location than a sagittal ray
fan. Fundamentally, astigmatism looks like field curvature, but the magnitude of the aberration is different
between x and y. In the wavefront expansion, astigmatism
is defined in terms of a tangential ( y axis) component only.
However, a portion of the field curvature of a lens is also
proportional to the astigmatism coefficient, adding a sagittal ( x axis) component; this term is also included in the
plots.
There is no astigmatism on axis (H = 0), and its magnitude

increases with field coordinate H . A wavefront plot for
astigmatism shows a quadratic dependence on aperture
coordinate , whereas a transverse ray plot shows a linear
dependence on that has a different slope in x and y. A
spot diagram shows the image of a point to be an ellipse.
Design programs plot the longitudinal
astigmatic field curves on a separate
graph. When there is no astigmatism,
the sagittal and tangential image
surfaces coincide with each other and
lie on the Petzval surface. If there
is astigmatism, then the tangential
image surface lies three times as
far from the Petzval surface as the
sagittal image surface.
22
Distortion
Distortion is an aberration in which the magnification

varies over the image height. Object points are imaged
to perfect points with no blur, but straight lines in the
object become curved lines in the image. Design programs
do not include distortion as part of their wavefront or
transverse ray plots. They instead plot distortion as either
a percentage versus field coordinate H or as a deviation
from a grid of points. The percent of distortion is usually
calculated from the difference in height between a real
and a paraxial chief ray for a given field point; however,
a centroid distortion (using either the geometric spot
or PSF centroid location instead) may be a more relevant
evaluation for a particular lens/application.
Pincushion distortion occurs if the magnification
increases toward the edge of the field.

Barrel distortion occurs if the magnification decreases
toward the edge of the field.

Keystone distortion produces a trapezoidal image of a
rectangular object and can result from nonparallel object

and image planes.
23
Primary Color and Secondary Color

Chromatic aberrations arise because the refractive
index of a lens material changes with wavelength
(characterized by the material dispersion). Light from
a single object point at different wavelengths is not
focused to the same
point in the image (e.g.,
for a simple lens, blue
light from an axial,
white-light point source
is focused before red light). For polychromatic lenses, performance evaluations such as transverse ray plots are
computed for at least three wavelengths that span the system bandwidth (one at each end of the spectrum and one in
the middle). For the visible spectrum, these wavelengths
are typically C (red), d (yellow), and F (blue).
The difference between the red and blue focus is
known as primary color (also referred to as axial
color, longitudinal color, or longitudinal chromatic
aberration). Similar to
defocus, the transverse ray
plot shows a linear dependence on that has the
same slope in both x and y
and is independent of field
coordinate H ; but the slope
of the line changes with
wavelength.
If the primary color of the lens has been corrected, the
lens can still have secondary color, which is a difference
between the common red/blue focus position and the
yellow focus position (also indicated by a change in slope
on the transverse ray plot).
When the d line is also imaged
to the same focus as the F
and C lines, the system is said
to be corrected for secondary
spectrum.
24
Lateral Color and Spherochromatism

Lateral color (or transverse chromatic aberration)
is an aberration in which the magnification of the system
is wavelength dependent (the image height varies with
wavelength). For a simple visible lens with the aperture
stop after the lens, the blue image appears to be larger
than the red image.
There is no lateral color on

axis (H = 0), and for objects
located along the y axis, there
is no lateral color in x. A
transverse ray plot shows
a constant dependence on
that increases in magnitude
with field coordinate H (similar to tilt); however, the
intercept value changes with wavelength. Unlike primary
color, the magnitude of lateral color is highly dependent on
the stop location.
Color correction equations developed for the visible
spectrum (d, F, and C wavelengths) can be used outside of
the visible spectrum if a custom Abbe numer V is defined
for a material over the band of interest, with the indices
at the long, short, and middle wavelengths.
Spherochromatism is a variation of spherical aberration with wavelength. In some lenses, there is
very little spherochromatism. In others, the spherical aberration can switch sign (typically blue is
overcorrected and red is undercorrected), and the spherochromatism must be corrected or balanced with axial
color to reduce image blur.
25
Higher-Order Aberrations
Lens systems with large apertures (< f /4) and wide
fields of view (>50-deg FFOV) will show evidence of
higher-order (HO) aberrations. The five third-order
aberrations have fifth-order counterparts (e.g., fifthorder spherical aberration looks very similar to thirdorder spherical aberration, but its dependence on aperture
coordinate increases in order). There are also two new
fifth-order aberrations: elliptical coma and oblique
spherical aberration.
During optimization, the best performance is typically achieved by

balancing low-order aberrations
with high-order aberrations. The
residual errors then will show an
inflection in the transverse ray, distortion, or field curves, indicating a
sign change between the low-order
and high-order aberration contributions.
26
Intrinsic and Induced Aberrations

Intrinsic aberrations are aberrations associated with
a single surface in an optical system and can be
calculated solely from paraxial ray angles and heights
at a given surface. In contrast, induced aberrations
are aberrations generated from the interaction between
aberrated rays coming into a surface and the intrinsic
aberrations of that surface.
A simple example of induced spherochromatism can be

constructed by placing a concave spherical mirror in the
converging cone created by a positive-powered singlet. The
lens sends uncorrected axial color into the mirror surface
so that the marginal ray height at the surface varies with
wavelength; the axial color interacts with the mirrors
intrinsic spherical aberration resulting in a different
amount of spherical aberration for each wavelength. The
mirror itself is not dispersive, so the aberration is not
intrinsic.
Intrinsic and induced aberrations can be used to offset
each other to improve performance. The functional form
of an induced aberration can be identical to an intrinsic
aberration, but it is usually harder to correct the induced
aberrations in a design. Complex, well-corrected designs
are often limited by higher-order induced aberrations
(specifically, higher-order Petzval curvature and sagittal
oblique spherical aberration).
Design Forms
27
Selecting a Design Form: Refractive

Aperture and field size are the two key system
specifications that determine the difficulty/complexity
of a design and can therefore be used to identify an
appropriate starting design form. If the aperture or field
size of a design form is increased outside of its working
region, the performance drops off; if the aperture or
field size is decreased, the design form is likely to meet
the performance requirement but is more complex/costly
than necessary. The classic design-form regions indicated
below are approximate and do not necessarily represent
diffraction-limited performance. Most assume spherical
surfaces and are valid over the visible band. Modern
designs build on these basic forms, often incorporating
aspheres to reduce package size, improve performance,
and/or increase the aperture or field size.
28
Design Forms
Selecting a Design Form: Reflective

Similar to refractive design form selection, aperture
and field size are the two key metrics for choosing a
reflective design form. It is important to first select a
form that can support the FOV requirements. Then the
aperture specification is a strong driver for the number
of mirrors and their asphericity; aspheres are effective
at correcting spherical aberration when increasing the
aperture but are less effective at increasing the field
coverage. The system size requirements will often filter
out most of the candidate design forms. For example, if
the envelope diameter must be less than 23 the EPD,
it is unlikely that a system with a negative primary
mirror that performs well over a wide field can be used.
If an accessible field stop and/or cold stop is required, this
implies that a reimaging system is needed and excludes
non-reimaging forms. Furthermore, if a Lyot stop is
needed with high-quality pupil imaging, more complexity
in the mirrors (e.g., tilts and/or higher-order aspherics)
might be necessary.
Design Forms
29
Singlets
A singlet is a lens used
without any other powered
elements; the stop is typically
located at the lens. Singlets
have relatively poor image quality, limited by large
chromatic aberrations and field curvature. They are often
used as a simple magnifying glass or photon collector.
An aspheric collimator is a singlet with
at least one aspheric surface. The aspheric surface allows for an increased numerical aperture and increased resolution
over a very narrow wavelength range and small field.
These lenses are commonly found in laser collimating applications such as fiber coupling, laser diodes, optical data
storage, and barcode scanners.
Landscape photography requires wide-FOV lenses that
capture as much of the scenery as possible. Shifting the
stop in a singlet away from the lens increases its FOV and
results in a simple landscape lens. As the stop is moved,
the lens shape changes and is generally bent around the
stop to reduce the ray angles of incidence and the off-axis
aberrations. The stop shift also introduces astigmatism
that helps to flatten the tangential field curvature.
A single-element landscape lens still has spherical
aberration and chromatic aberrations and is therefore
restricted to a small NA. For better correction, the
singlet may be aspherized or compounded into a doublet.
The farther the aperture stop
is shifted from the lens, the
larger the lens diameter. Disposable box cameras generally use simple landscape
lenses.
30
Design Forms
Achromatic Doublets
An achromatic doublet
or achromat is composed
of two singlets designed in
combination to correct primary color at two wavelengths (typically red and blue). The individual element
powers of the doublet are determined by the total power
and the individual Abbe numbers of the materials. For
standard refracting materials with positive Abbe numbers, the two elements will have opposite powers, and each
has stronger absolute power than the total doublet power.
The shape of the elements is chosen to minimize spherical
aberration and coma.
The on-axis spot size of an f /5 visible-band achromat is
25 times smaller than a singlet of the same aperture.
Doublet elements can be either cemented together or
separated by an airspace. The cemented doublet is
often preferred for ease of fabrication and alignment. A
Fraunhofer doublet is a positive crown followed by a
negative flint, whereas a Steinheil doublet is a flint-first
doublet. The Fraunhofer form has a shallower curvature
at the cemented interface and is usually the preferred
design form.
Achromats are limited in field to less than a few degrees
by residual Petzval and in f /# to f /4 or larger by higherorder spherical aberration. Fast achromats (< f /4) are rare
because the V ratio that determines element power is
fundamentally limited by existing glass choices. Common
doublet applications include broadband collimating lenses
and telescope objectives. Doublets are also critical building
blocks for complex broadband lenses.
Lenses corrected for axial color at three and four wavelengths are called apochromats and super-achromats,
respectively, and usually require more than two elements/materials.
Design Forms
31
Airspaced Doublets
The speed of a Fraunhofer
doublet can be increased to
f /2.5 by separating the cemented surface to create an
airspaced doublet. The extra curvature and airspace are
used to control higher-order
spherical aberration and spherochromatism. However, the ray angles of incidence in the
airspace between the two elements can become very large,
making the alignment tolerances of the two elements very
tight. A second type of airspaced doublet is a Gauss doublet, which has much steeper curves and significantly
larger higher-order aberrations. As a result, Gauss doublets are not as fast as Fraunhofer doublets and are often
found in stop-symmetric anastigmatic lenses.
Two separated positive and negative elements (or groups of elements) with a relatively large air
gap between them is known as a
dialyte. Although the performance
of a simple two-element dialyte is
usually insufficient (primarily due to its inability to correct lateral color), it is the basis for more-complex telephoto (positive/negative) and reverse telephoto (negative/positive) multi-element designs.
A Schupmann lens is a special airspaced doublet that is
corrected for primary color but uses the same material in
both elements. Because the lens has a virtual image, it is
seldom used alone but rather is often part of more-complex
lens systems.
32
Design Forms
Cooke Triplet
The Cooke triplet
is the simplest design
form
with
enough degrees of
freedom to correct
all first- and thirdorder aberrations and hold first-order imaging constraints
such as EFL and BFL. The outer two elements are positive crowns, and the middle element is a negative flint
with the aperture stop located on either side of the middle
element. The symmetry of the design helps control coma,
distortion, and lateral color. Because it is only corrected
to third order, the Cooke triplet is limited by higher-order
aberrations and is thus capable of relatively slow apertures ( f /6) and moderate fields (1015 deg). Up to 50%
vignetting may be needed in systems with larger fields.
The Cooke triplet is the most basic form of an anastigmat
lens. Strictly speaking, the term anastigmat means free
of astigmatism; however, it is also used to describe
optical systems with reduced astigmatism and relatively
flat fields. In these designs, the field is flattened by
separating positive and negative power along the optical
axis (between surfaces, elements, or even multi-element
groups of components). PLAN microscope objectives
use this principal to flatten the field curvature.
Many photographic objective design forms are derived
from the Cooke triplet by splitting, compounding,
or otherwise modifying elements of the basic triplet.
For example, in the Tessar lens, the last element
of a Cooke triplet is compounded into a doublet.
This design allows for speeds
up to f /4 and/or larger
fields of view than the basic triplet with reduced vignetting.
Design Forms
33
Double Gauss
The double Gauss (or Biotar or
Planar) is a powerful anastigmat
design form used in a wide range
of applications, including fixed-focus
camera objectives, projection lenses,
and high-resolution microscope objectives. The classic
six-element design is nearly symmetric about the stop
(reducing coma, distortion, and lateral color), with outer
positive singlets and inner cemented thick-meniscus
doublets (to minimize Petzval). Astigmatism is controlled
by adjusting the distance of the elements from the stop, as
this separation has little effect on the other aberrations.
Design improvements to the basic form have been studied
extensively in the literature; allowing the system to depart
from symmetry and/or adding elements (by splitting or
compounding) allows one to achieve speeds up to f /1 or
FFOVs larger than 50 deg.
Fast anastigmats with small fields tend to have relatively
short vertex lengths, whereas slow-speed wide-angle
lenses tend to have longer vertex lengths. In Cooke
triplets the vertex length can be controlled by glass
choice (particularly the difference in V between positive
and negative elements).
Many other lens forms use stop symmetry and power
separation to achieve wide fields of view. The simplest
is the Hypergon; two identical meniscus elements
symmetric about the stop achieving an impressive 135deg FFOV. Petzval is reduced by the separation of positive
and negative power in the elements themselves where
each element has a positive and a
negative surface with radii differing by
less than 7%. However, Hypergons are
slow ( f /20) as there is no remaining
degree of freedom to correct the spherical
aberration. More-complex designs (e.g.,
Topogon, Hologon, and Biogon) add
elements to increase their speed.
34
Design Forms
Petzval Lens
The Petzval or Petzval portrait lens is a design
form useful for high-speed applications that require
high-quality imaging over relatively small fields (e.g.,
aerial reconnaissance). Originally designed for indoor
photography, Petzval lenses are an order of magnitude
faster than landscape lenses. The Petzval lens is the basis
of many microscope objectives where aplanatic elements
are added to the short conjugate for increased NA.
Unlike many of the other basic design forms, the Petzval
lens has very little symmetry. The basic layout uses two
separated, positively powered
achromats with the stop at or near the front lens. Negative
astigmatism from the first doublet is balanced with positive astigmatism from the second doublet. Because there
are no lens groups with net negative power, the design
form is fundamentally limited by Petzval, restricting its
field size. Decreasing the airspace between the two doublets can help reduce the Petzval at the expense of increased astigmatism.
Given the focal length f of a Petzval lens, the first-order
thin lens layout can be calculated as follows:
Front element focal length:
2f
Rear element focal length:
Element separation:
Back focal length:
f /2
Improvements to the basic Petzval lens include adding a

strong negative field lens near the image plane to correct
Petzval and increase its field (although this significantly
decreases the working distance).
The Petzval lens is also the base
design form of many extremely
fast ( f /1) lenses, including very
high-speed, 68-element Petzval
projection lenses.
Design Forms
35
Telephoto Lenses
A basic telephoto lens consists of two lens groupsa
positive group followed by a
negative group. The stop is
at or near the front group.
The front principal plane is pushed out in front of the
lens so that the EFL of a telephoto lens is greater than
its physical length (measured from the first lens surface
to the image plane). The ratio of the lens length to EFL is
called the telephoto ratio. Typical telephoto ratios range
from 0.6 to 0.9; the smaller the ratio is, the more compact
the system. The asymmetry of the design form limits its
field (<10-deg HFOV). Moderate-speed ( f /6) lenses can
be built with two cemented doublets, whereas faster lenses
can require the doublets to be air spaced and/or an element
to be added to the first or second lens group.
The focal lengths of any two-element thin lens system
can be calculated given the system focal length f , their
separation d , and the back focal length BFL:
fa =
df
f BFL
fb =
dBFL
f BFL d
For zero-Petzval sum:

f a = f b = f BFL
d=
( f BFL)2
f
Both lens groups are usually separately achromatized as

primary color and secondary spectrum limit the image
quality of telephoto lenses. The telephoto ratio cannot
be arbitrarily reduced without increasing Petzval (which
limits the field coverage of the design). Faster lenses and
smaller telephoto ratios require more-complex designs.
Long-focal-length telephotos are particularly sensitive
to small changes in object distance, and if used for
photography, require an additional floating element that
adjusts the focus as the object position is changed.
36
Design Forms
Retrofocus and Wide-Angle Lenses

A retrofocus lens (or reverse telephoto) is composed of a negative lens
group followed by a positive lens group producing
a BFL much greater than
its EFL (the opposite of a telephoto lens). This form
is extremely useful for short-focal-length lenses that
need a long working distance (e.g., for the placement
of an SLR viewfinder mirror
or a color-separation prism).
This design form is also the
basis for most wide-angle
lenses covering large fields.
Extremely wide-angle or fisheye lenses can cover more
than 180-deg full field.
Retrofocus lenses are large in both diameter and length
(can be 1020 their focal length). Many telephoto design
techniques are also valid for retrofocus lenses (e.g.,
component achromatization and Petzval reduction). A
large airspace between the component groups helps reduce
their individual power and aids in correction. The lack of
symmetry makes correcting secondary lateral color and
chromatic distortion a challenge in these designs. A conic
asphere on the inner surface of the front element can help
the overall correction of a wide-angle lens.
Wide-angle lenses suffer from significant third-order
barrel distortion that is difficult to correct and is often
balanced with a fifth-order pincushion distortion. Fish-eye
lenses have an inherently large barrel distortion that is
not necessarily considered to be an aberration but rather
a necessity for imaging a 180-deg hemispherical scene
onto a flat image plane. The standard h = f tan definition
of focal length no longer holds and may be substituted
with h = f or h = f sin distortion constraint. The barrel
distortion also offsets the illumination fall-off that results
from the cosine-fourth-power rule at large field angles.
Design Forms
37
Refractive versus Reflective Systems

Optical systems can be refractive (lenses), reflective
(mirrors), or catadioptric (both lenses and mirrors).
Because many refractive design forms have analogous
reflective forms (e.g., the telephoto and the Cassegrain,
the retrofocus and the Schwarzchild, or the Cooke triplet
and the reflective triplet), it is useful to understand the
advantages and disadvantages of one type over the other.
Advantages of reflective systems:
No chromatic aberration (a single system can then be
shared among many spectral bands/channels)
Spherical aberration for a spherical mirror is 1/8 of that
for a lens with a glass index 1.5 and an equivalent f /#
Opposite sign of Petzval compared to refractive lenses
Better performance over a wider thermal range
Fewer surfaces needed
Integrated mounting and alignment features
Elements can be independently shaped to reduce weight
Design can be folded to meet packaging constraints.
Disadvantages of reflective systems:

Limited FOV
Aspheres needed for high-quality performance
More sensitive to irregularity and alignment errors
Per-element fabrication cost and schedule
Stray light is harder to manage
Complex off-axis configurations needed to get mirrors in
the light path without obscuring the incoming beam.
In very wide-angle lenses, rays perpendicular to the

optical axis can fail to trace because their direction
cosines are zero. Imaging the stop through the front
group of negative elements also yields an anamorphic
entrance pupil that needs to move forward, shift laterally,
and tilt to get light into the lens as the field angle
increases. In software, ray aiming or a wide-angle mode
(uses real chief rays versus paraxial rays) is needed to
overcome these ray failures.
38
Design Forms
Obscurations
In reflective systems, obscurations are objects that block
a portion of the incident light bundle. For example,
the image from a single on-axis parabolic mirror is
located directly in the incoming beam path. If a detector
is placed at the image plane, it will obscure the
incoming beam. Common on-axis astronomical telescopes
(e.g., Cassegrain or Gregorian) have secondary mirror
obscurations. Although obscurations help reduce package
volume for systems with space constraints, they cause
an overall transmission or throughput loss. The stray
light from obscurations also requires critical attention
(especially in the infrared where the thermal effects of
blackbody radiation from obscuring features must be
considered).
It is important to clarify if the obscuration percentage
is specified as a linear obscuration (% of the aperture
radius) or an area obscuration (% of the aperture area);
linear specifications are more common. A 20% linear
obscuration is 4% by area.
Obscurations modify the diffraction pattern of an object
by moving energy from the central Airy disc to the outer
rings. Central obscurations (blocking light in the center of the pupil) are the most common. Central obscurations primarily reduce the image contrast (MTF) at the
mid-spatial frequencies. The drop in MTF becomes noticeable around 15% obscuration (measured linearly) and increases as the size of the obscuration increases. A second common type of obscuration is a structural support
mount or spider used to hold mirrors in place. Each spider vane produces a diffraction spike in the PSF at a
right angle to the direction of
the vane. Obscurations can also
induce vignetting and, if near the
image plane, can directly alter
the shape of the final image (e.g.,
Hubble Space telescope WFPC).
Design Forms
39
Newtonian and Cassegrain

A single parabolic mirror
used with a fold mirror is
called a Newtonian telescope. It has zero spherical aberration on axis for
an infinitely distant object but is quickly limited off-axis
by coma. A Cassegrain telescope adds a negatively
powered hyperbolic secondary mirror to a parabolic primary. Both mirrors share a common focus point, forming a mathematically perfect, aberration-free or stigmatic image of an infinite on-axis object. The field is
limited by coma and astigmatism to less than a degree,
but the compact package and small telephoto ratio make
the form extremely attractive for
many portable applications. Shifting the mirrors axially from stigmatic
alignment
balances
coma with spherical, improving the
performance in the field.
In a first-order layout, the mirror curvatures of any twomirror system are given by
c1 =
BFL f
2d f
c2 =
BFL + d f
2dBFL
where f is the effective focal length, d is the separation

between the two mirrors, and BFL is the distance from
the second mirror to the image.
The RitcheyChrtien telescope uses a hyperbolic
primary and hyperbolic secondary to correct spherical and
coma exactly to third order. The field coverage is slightly
larger than that of a Cassegrain; however, the primary
mirror can no longer be tested with an on-axis diverging
spherical beam on the finite-conjugate side, requiring the
use of a null lens. Refractive corrector lenses can
also be inserted to improve field performance, but they
add chromatic aberration, limiting the telescopes spectral
region.
40
Design Forms
Gregorian and Schwarzschild

A Gregorian telescope has a
parabolic primary and positively
powered elliptical secondary. By
aligning the focus of each conic section, a perfect on-axis stigmatic image is formed. Many early telescopes were Gregorian; historically, positive concave mirrors were easier to fabricate.
However, the Gregorian with two positive mirrors has
more field curvature than a Cassegrain. Modern mirrorfabrication techniques have made the Cassegrain more
common. The performance of a Gregorian telescope can be
improved by departing from the stigmatic imaging condition and changing the primary mirror to an ellipse.
The 200 00 Hale Telescope at Mt. Palomar is a
Cassegrain; the Hubble Space Telescope (HST) is a
RitcheyChrtien; and the Large Binocular Telescope
(LBT) in Arizona is a Gregorian design.
A Schwarzschild objective has a
small negative spherical primary and
a large positive spherical secondary.
If the mirrors are concentric with the
stop located at their common center of
curvature, the system can be exactly corrected for thirdorder spherical, coma, and astigmatism, and the image
lies on a curved surface whose radius equals the focal
length. The mirror separation d and the mirror curvatures
c 1 and c 2 are given by
d = 2f
c1 =
51 f
c2 =
5+1 f
for a system focal length f . The size of the secondary

mirror and large obscuration ratio (>40%) dramatically
limit the use of a Schwarzschild as a telescope objective,
but its low f /# and long working distance make it an
attractive form for reflective microscope objectives. In
practice, the mirrors may be aspherized and/or lenses
added to the design to increase the NA and field coverage.
Design Forms
41
Catadioptric Telescope Objectives

Catadioptric telescope objectives use weak refractive
elements to correct the aberration of high-power spherical
mirrors without adding significant chromatic aberration.
The simplest example is a Mangin
mirror or back-surface reflective element with two powered surfaces. In
a Schmidt telescope, the aperture
stop is placed at the center of curvature of a spherical mirror, eliminating coma, astigmatism, and distortion. A thin corrector plate with one aspheric surface is then located at
the stop, correcting all orders of spherical while adding
only a small amount of chromatic aberration. The final
image plane is curved with a radius equal to the focal length. A field lens can be added near the image
plane to flatten the image surface, if needed. In practice,
ghost reflections from the corrector plate can be a problem. Spherical airspaced doublet or triplet corrector plates
yield Houghton or Buchroeder systems with improved
achromatization without the use of aspheres.
The Maksutov or Bouwers telescope has a spherical

primary mirror and a weak-meniscus, concentric spherical corrector plate (historically easier to fabricate than
the aspheric Schmidt plate). Nominally, all three surfaces are concentric to the stop; however, the corrector
plate can reduce axial color with a slight departure from
concentricity at the expense of slightly increased spherical. Both the Schmidt and Maksutov can be combined
with other telescope objective forms. For example, the
SchmidtCassegrain and MaksutovCassegrain are
common commercial telescopes for amateur astronomy
that combine the corrector optics with a Cassegrain system.
42
Design Forms
Unobscured Systems: Aperture Clearance

One method to eliminate the obscuration in a reflective system involves
shifting the on-axis beam so that it is no
longer centered on the optical axis (usually at the expense of increased package size). For example, in an off-axis
parabola (OAP), the aperture stop is
simply decentered until the image is
clear of the beam. It is important to note that the object,
image, and mirrors still share a common optical axis in the
unobscured system. The aperture may be made noncircular to keep the package size small, but this will change
the shape of the PSF.
Tips for designing complex multi-mirror designs:
The design process typically starts with definition of
the size and speed of the primary mirror.

On-axis conic sections, when used off-axis, suffer
mainly from coma, and suffer from spherical aberration

at nonideal conjugates.
An analysis of the obscured system with nonapertured
parent mirrors helps determine the aberration
limitations of the unobscured design.
Many unobscured systems use both aperture and field
clearance to eliminate obscurations.
Unobscured two-mirror afocal systems can be constructed using two off-axis
parabolas. These systems
have perfect on-axis imagery
and no coma or astigmatism (Petzval dominates the field).
When set up as two positive confocal parabolas with an
internal image plane, the exit pupil is real and accessible. A Mersenne configuration with a negative mirror has
less Petzval and a smaller package size, but a virtual exit
pupil. Both configurations are typically used as attachments to increase the aperture and focal length of a system or, conversely, shorten its focal length and increase its
FOV.
Design Forms
43
Unobscured Systems: Field Clearance

For unobscured infinite-conjugate systems, field clearance is achieved by tilting the entire system with respect
to the incoming axial beam (this is equivalent to using an
angular field definition that is not centered on the optical
axis). For unobscured finite-conjugate systems, field clearance is achieved by eliminating the on-axis field point and
using the system off-axis in field, where the center of the
field is decentered until the beam is clear of all obscurations (often at the expense of performance and packaging
volume). Although it is often not obvious, the object, image, and mirrors share a common optical axis. If needed,
the image plane and/or mirrors can be slightly tilted or
decentered to improve performance; however, this change
significantly increases the alignment complexity. As most
detectors are rectangular, the smaller dimension of field
is located in the plane of the field decenter to minimize
package size. It is important to make sure that the detector is oriented correctly in the system, as the mirrors are
optimized and sized for one image orientation only.
The Offner relay is a classic
example of an unobscured
finite-conjugate all-reflective
system. It is a 1:1 system
composed of two concentric
spherical mirrors with the object (and image) plane at
their common center of curvature. The stop is located
at the secondary mirror, and the system is doubly
telecentric. The Offner design form eliminates all thirdorder aberrations exactly and is limited by higher-order
astigmatism (but can be improved by deviating from
perfect concentricity). Early high-performance, 1:1 ringfield lithography systems were based on the Offner design
form. This type of design is also the preferred form
for many spectrometers, where the secondary mirror is
replaced by a reflective grating.
44
Design Forms
Three-Mirror Anastigmat
The Cassegrain telescope is a promising design form, as it can fit a long
focal length in a short package while
balancing Petzval with a positive
mirror and a negative mirror. However, similar to many two-mirror design forms, the field coverage of a
Cassegrain is limited to 1 deg by
astigmatism. Adding a third mirror to a Cassegrain provides the degrees of freedom necessary to correct the astigmatism across the field. The result is a three-mirror
anastigmat (TMA) design useful for applications with
HFOVs of 25 deg. In a focal TMA design, the tertiary
mirror is an ellipse; in an afocal TMA design, the tertiary
mirror is a parabola. Both obscured and unobscured (usually using aperture clearance) designs are possible. The
key feature and main advantage of the TMA over other
three-mirror design forms is that it is a reimaging design with an accessible internal image plane and an accessible exit pupil. This allows the placement of field stops,
Lyot stops, and cold stops (for infrared
systems) for improved stray light
suppression. Other TMA variations
include designs with the internal focus
after the primary rather than after the
secondary, or designs with remote exit
pupils for test setups.
Adding a fourth corrector mirror near the internal
focus of a TMA improves the wavefront and reduces the
distortion within a given FOV. The additional mirror also
helps extend the field size in the direction of the mirror
offsets to allow a square or circular FOV. The corrector
mirror can be oriented coaxially with the other three
mirrors or highly tilted to fold the design into a ballshaped enclosure for pod-mounted applications.
Design Forms
45
Reflective Triplet
The non-reimaging TMA or reflective triplet (RT) is the reflective
equivalent of the refractive Cooke
triplet. The stop is at or near a negative secondary mirror with a positively powered mirror on each side of
the stop. All three mirrors are conics
and nominally share a common optical axis (the mirrors may be slightly decentered or tilted
to improve performance). Because there is no intermediate
image, the RT design form can support a larger field than
the reimaging TMA design form; however, an RT design is
not appropriate for thermally cooled systems, as the exit
pupil is virtual and therefore inaccessible for the placement of a cold stop. The RT design form is recommended
for moderate-FOV (up to 10-deg HFOV) applications that
do not require reimaging. RTs perform best with the aperture stop at secondary (distortion is easier to correct due to
the symmetry), but the design form also supports an aperture stop in front of the system, if needed. The beam clearance in an unobscured RT design is provided through both
an off-axis aperture and an off-axis FOV. Systems that are
off-axis in field will tilt and decenter the image plane to
compensate for residual field curvature.
Distortion in an optical system is defined as the difference in image height
between a real ray and a paraxial ray.
For systems with centered FOVs the
specification can be given by a single
number at the edge of the field (e.g., 4% distortion).
However, for unobscured designs with off-axis FOVs, the
paraxial region may not even be in the FOV, and a single
specification is not adequate. In this situation, a series of
new terms such as smile, tow/keystone, expansion,
and anamorphism are needed to describe and quantify
the distortion.
46
Design Forms
Wide-Field Reflective Design Forms

Wide-angle mirror designs with more than
a 10-deg HFOV use a
design principle similar to wide-angle inverted telephoto refractive lenses. A front negative mirror is followed by a positive focusing group. The
aperture stop is located inside the lens (usually near the
positive focusing group). However, wide-angle reflective
designs tend to be large (many times larger than the entrance pupil diameter). The simplest example is an offaxis Schwarzschild, which can also be converted to a
reimaging form for stray light suppression by adding a
concave relay mirror on the other side of focus in the
same way that a Cassegrain is converted to a TMA. The
additional mirror extends the FOV beyond what a standard Schwarzschild can achieve (e.g., up to 20 20 deg or
30 30 deg).
Afocal versions of 3- and 4-mirror reflective designs are
often used as beam reducers for optical systems needing
both large apertures (to collect light from distant objects)
and limited diameter filters, scan mirrors, diffraction
gratings, or beamsplitters that need to be placed in a
collimated beam.
Extremely large (by reflective standards > 20 deg), wideangle reflective designs are based on the design concept
that when the entrance pupil is at one focus of a
negative hyperbolic primary mirror, the reflection is free
of astigmatism. An aspheric secondary mirror is placed at
the aperture stop to control spherical aberration, and a
concave tertiary mirror is used to form the image. This
configuration has been referred to as the three mirror
long (TML), as opposed to the three mirror compact,
another name for the reflective triplet. Variations on
the TML concept include forms with an aerial aperture
stop or WALRUS. The TML family provides the widest
field coverage of the reflective design forms. The main
disadvantage of this form is its very large envelope.
Design Forms
47
Zoom Lens Fundamentals

A zoom lens is a lens whose focal length can be changed
either continuously or in discrete steps. Continuous
zooms use zoom groups with fixed power that move
along the optical axis. Discrete zooms laterally switch
different fixed-power groups into (or out of) the optical path. Some zoom lenses (e.g., liquid lenses) can dynamically alter element powers without motion. True
zoom lenses have a fixed image-plane location. Varifocal
lenses change focal length but allow
the image plane to
move, requiring refocus.
The zoom ratio is
the ratio of the two
extreme
focal
lengths (or, in the
case of a finite conjugate zoom, the two
extreme magnifications). For example, a 25200-mm zoom
lens is an 8 zoom. Zoom lenses are now standard for most
consumer cameras, including point and shoot cameras,
SLR cameras, and video camcorders, where the primary
trade-off is large zoom range versus compact package size.
The f /# of most modern camera zoom lenses varies across
the zoom range. At long focal lengths the aberrations tend
to be larger, and the f /# is slower. A smaller aperture also
reduces the physical size of the outer lens components.
Distortion and lateral color will often change sign through
zoom. The relative illumination and vignetting can change
significantly through zoom.
48
Design Forms
Zoom Lens Design and Optimization

Zoom lenses can be classified according to their distribution of group power (e.g., + + +). A simple mechanically compensated zoom lens has two moving groups:
the variator changes magnification, and the compensator moves to maintain image plane focus. The motion of the compensator group is typically nonlinear;
the zoom groups are linked with a mechanical cam.
The two zoom groups form a zoom
kernel that relays a fixed object
distance to a fixed image plane.
In photographic objectives, a front
focusing group is used to provide
a fixed image position for the zoom kernel and to
accommodate changes to the object distance. Morecomplex zooms add a fourth (fixed) prime group to the
rear of the zoom kernel and/or more moving groups.
Optically compensated zooms move linked groups using
a linear motion but do not have a perfectly fixed image
plane. Many discrete zooms use a form of two-position
optically compensated designs.
Zoom lens optimization tips:
Start with a thin lens layout (e.g., first-order zoom lens
layouts can be designed with Gaussian brackets) or

a predefined starting point; global optimization is not
as useful for finding zoom lens starting points as it is
for fixed focal length lenses.
Independently achromatize each zoom group over
multiple conjugates.
Optimize over at least 35 zoom positions simultaneously and then verify the lens performance between the
optimized zoom positions.
Check that moving groups do not collide with each
other or fixed groups during their zoom motion.
More than two moving groups may be needed to
maintain f /#, aberration control, or telecentricity
across zoom, or for thermal compensation. Adding more
moving groups also helps reduce the package size for
zoom lenses with large zoom ratios.
Improving a Design
49
Techniques for Improving an Optical Design

Many techniques exist either to improve an existing
lens with inadequate performance or to extend an
existing design over new requirements without changing
the basic design form. The most common scenario
arises when a designer must push a legacy design
to increased aperture and/or larger field (for better
resolution/throughput). Such requests often come with a
challenging expectation of no significant increase in the
physical size of the design. Options available to a designer
include:
Split a lens
Raise (or lower) index
Compound a lens
Raise (or lower) V
Aspherize a surface
Use an abnormal partial
Use a diffractive surface
dispersion material
Each of the above techniques seeks to reduce the aberration contribution of a surface/component. A Seidel or
Pegel diagram shows the surface-by-surface aberration
contribution and is a key tool in identifying elements or
surfaces to target for improvement. It can also be very
useful for comparing the sensitivity of different design solutions.
Increasing the aperture or field of an existing design

can lead to ray failures during lens evaluation
and optimization. Examples include total internal
reflection (TIR) or ray intercept failures (where rays
physically miss a surface). The solution is to slowly
walk up to the desired aperture or field in small
increments, optimizing the lens at each step.
50
Improving a Design
Angle of Incidence and Aplanatic Surfaces

Ray angles of incidence (AOIs)
are measured relative to the surface normal at the ray intercept.
A large AOI indicates significant
departure from paraxial refraction, where 6= sin . Ray AOI is therefore an important
indicator of both the aberration contributions (aberration
coefficients depend on the chief and/or marginal ray AOI)
and the tolerance sensitivity of a surface. During the improvement process, elements with large ray angles are
good candidates for modification (e.g., splitting into multiple elements). Examining a lens layout and assessing the
ray angle distribution gives a quick insight into the performance, manufacturability, and improvement potential for
a design. Several design for tolerance optimization metrics have been proposed that minimize ray AOIs throughout a design.
Three conditions exist where a single surface can have
zero third-order spherical aberration:
The object or the image is concentric with the surface so
that rays do not refract (zero AOI);
The marginal ray height is zero at the surface (useful for
field lenses); or
The surface is aplanatic where the marginal ray AOI
( i ) equals the negative of the angle that the ray makes
with the optical axis after refraction (u0 ).
The final aplanatic condition is a
powerful one, as it allows optical
power and large AOIs without
adding spherical aberration (or
coma and astigmatism).
An aplanatic lens exploits two of the three conditions for
zero spherical at a surface: the first surface is aplanatic,
and the second surface is concentric with the image
plane. Microscope objectives and interferometer reference
spheres use aplanatic lenses to increase their speed
without introducing significant aberrations.
Improving a Design
51
Splitting and Compounding

Splitting a lens element into two (or
more) elements of the same material while
maintaining the total power is a highly
successful design technique for improving
an optical system. It
Adds degrees of freedom for optimization.
Reduces the ray AOIs by reducing the individual surface
powers and is therefore very effective at improving

spherical, astigmatism, coma, and distortion.
Reduces spherical aberration when splitting positive
elements in high-NA lenses.
Reduces astigmatism, distortion, and coma when
splitting negative elements in wide-angle lenses.
Does not improve Petzval or primary color since the total
power contribution does not change.
The exact method to split a lens (e.g., literally cutting
the lens in half or using two equi-convex elements) is
design dependent. If the change in element shape is not
significant enough, the lens will revert to its original
local minimum (after splitting, the result will be two
lens elements with a narrow crack-like airspace between
them whose outer shape mimics the shape of the original
element). However, if the change is too large, ray failures
can occur during optimization.
Compounding a lens element into a doublet or triplet (while holding total power)
is another powerful design improvement
technique. It
Uses two (or more) materials to simu-
late a nonexistent but desirable material type (e.g., a

material with a high index, large V , or unusual partial
dispersion).
Primarily reduces axial color, lateral color, and Petzval,
although the extra degrees of freedom can also be used

to correct other aberrations.
52
Improving a Design
Diffraction-Limited Performance
During the design improvement process, performance
targets are useful for guiding the process and ultimately
knowing when to stop optimizing. Many optical systems
use diffraction to set the performance limit, and the lens
requirements may state diffraction-limited by design.
However, different definitions of diffraction-limited exist,
so a designer should clarify the intent of the statement.
Strictly speaking, a diffraction-limited lens is limited solely by diffraction and refers to a zero-wave RMS
OPD; however, these criteria are rarely the intent of
a diffraction-limited specification. Common approximations for diffraction-limited performance include:
a peak-to-valley OPD < /4 (Rayleigh criterion),
an RMS OPD < 0.070 (Marchal criterion), or
a Strehl ratio > 0.8.
Reversing a lens can help a stagnated design during

optimization. Swapping two surfaces of an element in
the lens design program often requires adjusting the
surrounding thicknesses so that the principal planes and
first-order properties remain constant and so that ray
failures do not occur. Changes in ray intercept heights
and AOIs will alter the aberration distribution.
Improving a Design
53
Thin Lens Layout

For a lens whose thickness is much smaller than its focal
length, the marginal ray height is approximately the same
at both surfaces. The ray transfer contribution through
the element can be ignored and the lens treated as a thin
lens with zero thickness. A thin lens layout simplifies
a complex optical system by making each element (or
lens group) a thin lens with optical power. This layout
highlights the balance of positive and negative power
between lens elements and/or lens groups and can be a
vital aid to understanding the performance limitations of
an optical design, especially if the design is limited by
Petzval or color. A thin lens layout is also a useful tool
for starting an optical design (essential for zoom lenses) or
joining independent lens modules together while matching
their pupils in size and location.
c +c
u + u0
a
For a thin lens, the third- = 1 2 C = a
0
c
c
u
u
a
1
2
a
order aberration equations can
be simplified and written as a function of a lens bending
parameter or shape factor that describes what
the lens looks like (defined by the two curvatures), a
conjugate factor C that describes the magnification at
which the lens is used (defined by the input and output
marginal ray angles), and its index of refraction. The
Kingslake G-sums are similar thin lens aberration
expressions, where each aberration coefficient includes
a G coefficient that depends solely on index of refraction.
A y-ybar diagram is a plot of

the marginal ray height y versus
the chief ray height y , where
each point on the plot represents
a lens element (or surface) in
the system. The plot quickly
identifies the locations of objects
and images ( y = 0) and stops and pupils ( y = 0) and their
sizes, and is an alternative to a thin lens layout.
54
Improving a Design
Lens Bending
A key variable when optimizing a design is lens radius.
For a specific index of refraction and thickness, there are
an infinite number of combinations of radii R1 and R2 that
will yield a lens with a given focal length. The result is that
for a fixed focal length, a lens may take on any number of
different shapes or lens bendings, and the aberrations
of the lens (primarily spherical aberration and coma) will
change as the shape is changed.
The thin lens aberration
equations can be used
to show that for a single thin lens with an
object at infinity (conjugate factor C = 1) and
the stop at the lens, a
bending parameter exists either for minimum
(but nonzero) spherical
or for zero coma that depends only on the index of refraction of the
lens, whereas astigmatism, Petzval, and distortion do not depend on the
shape of the lens.
The shape factor for a single thin lens (object at infinity)
for minimum spherical aberration changes from near
convex-plano to meniscus as the index of refraction is
increased from 1.5 to 4.0. This behavior helps explain
why very high-index infrared lenses look different from
visible lenses.
Improving a Design
55
Material Selection
Material selection is one of the most critical choices in
lens design. Changing a lens material during the optimization of a design can significantly alter its performance. In general, raising the index of refraction of
all elements is an effective way of reducing the lens
curvatures for given lens powers, minimizing ray AOIs
and therefore aberrations at each surface. Primary chromatic aberrations can be corrected by selecting materials with the appropriate dispersion (V). In a thin
lens achromatic doublet, the V difference between the two materials determines the individual element powers. Increasing the positive element
V and decreasing the negative element V will reduce the component
powers, the ray AOIs, and the corresponding aberrations.
Old achromats use lead-line glasses with a positive low-index crown and a negative high-index flint (e.g.,
BK7/SF2). New achromats use a high-index positive
crown and a lower-index negative flint (e.g., LaKN17/F2).
A new achromat has 4 less Petzval than an old achromat, but the small V separation due to material limitations restricts it to about half the speed of an old achromat. New achromats are useful in designs driven by field
performance (e.g., wide-angle lenses).
The secondary spectrum of a thin lens achromatic doublet
is proportional to the P /V ratio of the two materials,
where P is the partial dispersion of a material.
Ideally, a pair of glasses with the same P but significant
difference in V is needed to correct secondary spectrum.
Unfortunately, for most glass doublet combinations, the
P /V ratio is nearly a constant. Only a few materials
(e.g., CaF2 ) exist with anomalous partial dispersion
that helps reduce secondary spectrum. A triplet achromat
can correct secondary spectrum by using two of the
materials in the triplet to synthesize a material that can
be matched in partial dispersion with the third material.
56
Improving a Design
Controlling the Petzval Sum

In order to have an optical design with a X j
reasonably flat field, its Petzval sum must be
j nj
small. The Petzval sum is directly proportional
to the ratio of each elements power to its index of
refraction. The reciprocal of the Petzval sum is the
Petzval radius, which equals the radius of the ideal
image surface. The ratio of the Petzval radius to the focal
length of the lens is the Petzval ratio.
Designing a lens to have zero Petzval is difficult, as
positive focal length lenses want to have primarily
positive-powered elements. A key strategy to control the
Petzval sum is to separate positive power from negative
power:
= 1 + 2
t
1 2 .
n
For example, if the radii are equal in a single lens (a

meniscus lens), the surface powers are equal but opposite
sign, resulting in a lens with zero Petzval and a power
that is directly proportional to the thickness of the lens.
In more-complex lenses, where positive lens groups are
separated in space from negative lens groups.
In most positive-power refracting lenses, the Petzval
surface will be curved toward the lens (inward-curving
field). In catadioptric designs, reflective surfaces can
be used to balance the Petzval from positive lenses
because a positive mirror has an index of refraction of
1, yielding an overcorrected (backward-curving) Petzval
contribution equal to its optical power.
Petzval curvature is perhaps the hardest aberration to
correct in an optical system. Once the first-order power
balance has been established, the index of refraction is
the only variable available to control the Petzval sum.
In positive elements, a higher index reduces the inwardcurving Petzval; however, it is not so clearly defined in
negative elements because a lower index increases the
overcorrecting Petzval, but a higher index reduces the
surface curvature, reducing other aberrations.
Improving a Design
57
Stop Shift and Stop Symmetry

Finding the best position for the stop is an important
part of the design process. A stop shift (moving the
stop location and scaling its size to maintain the
f /#) forces the chief ray through different parts of a
lens. Aberrations that depend on the chief ray (coma,
astigmatism, distortion, and lateral color) will change,
whereas spherical, Petzval, and axial color are unchanged
with stop shift. For a thin lens with nonzero spherical
aberration, there is always a stop position that exactly
eliminates coma (it also minimizes the tangential field
curvature). This position is called the natural stop
position.
If the best stop location is not obvious, a floating stop
can be utilized during optimization. The first surface in
the lens is designated as a dummy stop surface with
a variable negative distance. This surface is really the
entrance pupil location and allows the aperture stop
surface to float to its best location. Before finalizing the
design, the aperture stop surface designation should be
moved to the correct location to account for the effects of
any pupil aberrations.
Distortion, coma, and lateral color
can be eliminated in lenses with
stop symmetry. These aberrations all have an odd-order dependence on the chief ray height (or
angle), so lenses on one side of the stop have an aberration contribution equal and opposite to lenses on the other
side. In practice, even lenses with stop symmetry still have
some residual aberrations because the object/image conjugates are not usually symmetric. Stop symmetry is a key
strategy for reducing distortion (similar to Petzval, distortion is a notoriously difficult aberration to correct).
Unlike film, digital sensors cannot accept large ray
AOIs and typically have chief-ray-angle (CRA) matching
requirements. This trait can restrict the positioning of
the aperture stop in the system.
58
Improving a Design
Telecentricity
A telecentric lens is a lens with either the entrance pupil
(object-space telecentric) or exit pupil (image-space
telecentric) at infinity. The degree to which the chief ray
is parallel to the optical axis at either the object or image
plane is called telecentricity. The simplest way to make
a lens telecentric is to place
the aperture stop at a focal
point of the lens.
In a design code, the object space is easily defined to be
telecentric; however, the chief ray must be constrained to
go through the center of the stop during optimization. A
telecentric image space can be attained by targeting the
inverse of the exit pupil distance to zero in the merit
function or with a CRA solve or CRA constraint (usually
at several field points).
The primary benefit of telecentricity is that the system
magnification is insensitive to slight changes in image
distance (focus). The incident cone angle is also constant
over the entire field, improving relative illumination.
This is important for microlithography, microscopes,
machine vision systems, and other types of metrology
lenses. However, the lens elements must be at least as
large as the image, making telecentric lenses larger than
their nontelecentric counterparts.
A 4f optical system is a doubly telecentric finiteconjugate relay (also known as an afocal relay
telescope) with an accessible pupil plane for Fourier
filtering. Each of the two lenses has a focal length of f
so that the total distance from object to image is 4 f .
Improving a Design
59
Vignetting
By definition, the aperture stop limits
the size of the imaging bundle for
an on-axis object point. Vignetting
occurs when another surface (optical or
mechanical) in the system clips imageforming rays for off-axis field points.
The amount of vignetting is measured
as a percentage of the width of the
pupil. Vignetting reduces the amount
of light reaching the image plane for
off-axis field points. Some lenses (e.g.,
lithographic) cannot tolerate any light
loss or drop in relative illumination
across the field, while others (e.g.,
camera lenses) allow as much as 75%
vignetting.
Optical design codes simulate vignetting by approximating the vignetted pupil shape as an ellipse. Vignetting
factors relate the scaling and offset of the elliptical pupil
to the unvignetted pupil. (Note that some programs use
a percentage and an offset, and others define both an
upper and a lower percentage.) Pupil aberrations may
lead to confusing negative vignetting factors (even in the
absence of physical vignetting) because the paraxial entrance pupil must be overfilled to evenly fill the aperture
stop surface.
The minimum clear aperture for zero vignetting at any
surface is the sum of the absolute value of its marginal
and chief ray heights (| ya | + | yb |). If allowed, vignetting
is a powerful technique to improve the image quality of
a design because the pupil rim rays at the edge of the
field typically cause the largest spot flare. Vignetting also
reduces element diameters that are too large to meet
packaging and/or weight constraints. Any clipping surface
should be far enough away from the stop so as to not clip
the axial beam during optimization. Stop shifts can help
increase (or decrease) the amount of vignetting.
60
Improving a Design
Pupil Aberrations
The pupils in an optical design
are not necessarily perfectly
imaged to each other. To derive
third-order equations for pupil
aberrations, the entrance and
exit pupil planes are treated as
conjugate planes (similar to the object/image surfaces),
and the marginal and chief rays in the system are
interchanged in the third-order field aberration equations.
Although pupil aberrations do not directly impact image
quality, they can strongly influence other critical lens
properties. For example, large pupil aberrations make
it hard to pupil match two optical systems (especially
objectives and eyepieces) together.
Pupil spherical is the most commonly encountered pupil

aberration. In wide-FOV eyepieces, pupil spherical causes
the exit beams from different field points to not cross
at a well-defined axial location for an observers eye.
The result is that eyepieces with large pupil spherical
may have a shadowing or kidney-bean effect in the
apparent brightness of off-axis fields. Pupil spherical can
also cause ray tracing ray failures if there is a large
shift between the paraxial entrance pupil location (used
for the initial chief ray aiming) and the full-field pupil
location. Design codes can calculate pupil aberration ray
fans and/or help determine if ray aiming (which forces
explicit x/ y coordinates for the chief ray on a given surface)
is required to continue with the design analysis.
Distortion is less sensitive to changes in object position
if the pupil spherical in a lens is smallan important
trait in photoreducers or magnifiers that require wellcorrected distortion over variable conjugates.
Improving a Design
61
Aspheres: Design
By definition, aspheres are nonspherical surfaces. Standard aspheres are
rotationally symmetric, but off-axis
conic sections and freeform surfaces
are also referred to as aspheres. Many
qualitative terms exist that help describe an aspheric
shape (e.g., mild, weak, steep, or gull wing). In optical
design, aspheres are used to increase performance or
reduce the size and weight of a system. Near a pupil,
aspheres can be used to correct all orders of spherical
aberration. Near an image plane (e.g., on a field lens), they
can be used to correct astigmatism or distortion, or achieve
telecentricity. Early applications for aspheres included
illuminators, infrared systems, and single-wavelength,
high-NA, small-field systems like DVD objectives or laser
diode collimators.
When optimizing with aspheres, a designer should
significantly increase the pupil ray density and, if the
asphere is near an image plane, add more field points.
The sag of a surface is the displacement (along z) of the
surface from its vertex as a function of lateral position
( x, y, or radial coordinate r). The traditional aspheric
sag equation combines a conic expression with an
even-order polynomial, yielding a rotationally symmetric
surface with no vertex discontinuity:
sag = z =
1+
cr 2
1 ( + 1) c2 r 2
+ dr 4 + er 6 + f r 8 + gr 10 +
Both the conic constant and the r 4 coefficient will

affect third-order spherical, whereas the higher-order
coefficients are useful for correcting higher orders of
spherical aberration (in high-NA systems). However,
unneeded high-order terms can create oscillatory surfaces
and optimization convergence problems. Alternatively, a
Q-type (Forbes) polynomial aspheric expansion is an
orthogonal basis set where the coefficients are linked to
aspheric departure and slope, making it more practical for
tolerancing and manufacturing.
62
Improving a Design
Aspheres: Fabrication
Fabrication advances have made high-quality aspheres
practical for almost any design. The aspheric departure
is a measure of the surface deviation from a purely
spherical surface and is one of the most critical fabrication
metrics. A lens drawing typically specifies both a base or
vertex radius and a best-fit sphere (BFS). The BFS
minimizes the amount of material to be removed. Several
different techniques exist for fabricating aspheres (each
with their own pros and cons).
Technique
Conventional
polishing
Ion milling
Material
Glass / Crystals
Deterministic
polishing
Molding
Glass / Crystals
Diamond
turning
Plastic / Crystals
Glass
Glass / Plastic
Comments
High-quality surfaces; long
cycle time
High-precision removal; slow
removal rates and limited
departures
High-quality surfaces;
mid-spatial ripple residuals
Low-cost, high-volume
replication (high NRE for
molds); limited materials and
lower surface quality
Wide range of surface shapes;
serial throughput;
scattering/diffraction from
tool marks
Maximum slope error (the slope is the first derivative

of surface departure) is another metric that drives both
the fabrication and the testing of an asphere. Slightly
oversizing the aspheric clear aperture during the design
will keep the large slope errors that tend to occur at the
edge of the part outside of the critical aperture; however,
the total departure and part diameter will increase.
Aspheric testing can contribute significantly to the
overall part cost. Standard interferometric testing can be
used with small-departure surfaces, but other techniques
are required for large-departure surfaces.
Test Method
Surface profilometry
Null lenses and
computer-generated
holograms (CGHs)
Subaperture stitching
Comments
1D low-resolution scan; probe contacts
part (risk of damage)
High accuracy; costly and specific to a
single surface; expensive and long lead
time; precision alignment required
Useful for a wide range of aspheres;
full 2D surface coverage; limited slope
Improving a Design
63
Gradient Index Materials

A gradient index or GRIN material has a spatially
varying index of refraction typically described by an index polynomial. However, the index polynomial is not directly related to the manufacturing parameters, leading to
lengthy development cycles. Axial gradients have planes
of constant index perpendicular to the optical axis. Their
index equation is convergent; only the linear term N01
is needed to correct all orders of spherical aberration
when combined with a spherical surface. The result is that
axial gradients make fast collimating lenses (similar to aspheres).
Radial gradients have cylinders of constant index centered on the optical axis. Radial gradients have the unique
ability to focus light with flat
surfaces. A thin radial gradient is called a Wood lens, and
its power is proportional to its thickness times its N10 coefficient. Radial gradients are used in fiber couplers and
1:1 relay systems such as endoscopes, boroscopes, and
desktop copiers (replacing complex lens assemblies with
a smaller, less-expensive single component). A quarterpitch rod focuses incoming collimated light at its back
surface, whereas a full-pitch rod relays an upright image
from its front surface to its back.
GRIN aberrations are split into a surface contribution
into the medium and a transfer contribution within
the medium. Axial gradients have large surface and
small transfer contributions, whereas radial gradients
have large transfer and small surface contributions. The
primary manufacturing process for GRIN materials is
ion exchange. Although convenient for modeling and
aberration analysis, the index polynomial is not directly
related to manufacturing parameters, leading to lengthy
material development.
64
Improving a Design
Diffractive Optics
A diffractive optical element (DOE) uses wavelengthscale surface or phase structure to redirect an optical
beam using diffraction rather than refraction or reflection.
Diffraction efficiency is the fraction of light diffracted
into the desired order. Low diffraction efficiencies can
cause significant stray light problems. DOEs work with
both refractive and reflective surfaces and, similar
to conventional elements, they have a focal length,
dispersion, and aberrations. Similar to aspheres, DOEs
can be used for aberration correction or system size/weight
reduction. DOEs can also generate optical power from
flat surfaces or be used for athermalization, but they are
usually more expensive than aspheres.
Diffractive surfaces have a constant negative V (equal to
3.45) that can be used to correct primary color. However,
their abnormal partial dispersion yields a much larger
secondary spectrum than a standard doublet.
Kinoforms are smoothly varying surface-relief DOEs typically made by diamond turning a curved substrate. Kinoforms can be replicated in high volume and are common in infrared systems as hybrid diffractive achromats,
where the DOE replaces the negative
flint element. A binary optic is a stair-step DOE
with a discrete number of steps in either the surface
or phase profile and is formed by multiple lithography steps (the diffraction efficiency is directly proportional to the number of steps). A holographic optical element (HOE) is formed by interfering two wavefronts. A computer-generated hologram (CGH) is a
HOE formed by directly writing a phase profile on a
substrate. Both types of HOEs can
represent arbitrary phase profiles.
CGHs are often used as null lenses
to test large departure and/or offaxis aspheric surfaces.
Optimization
65
Optimization
Given an optical design starting point, modern lens-design
programs use numerical optimization to find design
solutions with improved performance by minimizing a
predefined merit function. Also called a cost or penalty
function, the merit function uses a single number to
represent total lens performance; in general, smaller
numbers denote better design solutions.
Designers first identify the lens variables (e.g., radii,
thicknesses, airspaces, and material parameters) that
are allowed to vary during optimization; all other lens
parameters are kept frozen. Performance defects (or
operands) that depend on the lens variables are then
defined and used to construct the merit function. Example
operands include image-quality measures (e.g., spot size)
and first-order properties (focal length). Constraints
(e.g., minimum lens thickness) can also be applied that
limit the amount of change in a variable.
Because the operands in the merit
function usually outnumber the
variables in the optical design,
optimization algorithms attempt to
find merit function minima, not
exact solutions. The merit function
topology and solution complexity
depend on the number of variables.
A lens with only two variables has a 2D topographical
solution space, whereas systems with more lens variables
will have multiple minima in a multi-dimensional merit
function. Commercial software programs use proprietary
local and global optimization algorithms to find minima in
the merit function.
Local optimizers use gradient methods to find the nearest minimum. The starting variable values determine
which local minimum is found. Global optimizers are
allowed to break out of local minima to find a morecomprehensive solution set.
66
Optimization
Damped Least Squares

The most common method of local optimization is
damped least squares (DLS), which minimizes the
merit function using linear regression approximations.
In DLS, each variable is changed by a (small) step
size that balances good numerical accuracy with fast
convergence time. Calculating the change in each merit
function operand for each variable step change produces a
derivative or gradient matrix and defines a local slope
that allows the optimizer to move downhill and find a
better solution using a numerical gradient search.
An optimizer can stagnate before reaching a good
design solution when the merit function changes very
little with the variable changes. Manually checking the
merit function for overconstrained variables, changing
the operand weighting, or dramatically altering one
or two variable values (e.g., radius of curvature or
airspace) can often overcome stagnation. In some
software programs the variable step sizes (also called
derivative increments) and/or the damping factor
can be modified to prevent stagnation and improve the
convergence of the optimizer.
Standard least-squares algorithms assume linear dependences between targets and variables. However,
in lens design, the relationship between merit function operands and lens variables is highly nonlinear. DLS also approximates the functional dependence
between operands and variables as linear but its
damping factor adds a weighting
factor to the merit function
that heavily penalizes any large
changes in the variables. If
the changes are small, then a
nonlinear function approaches a
linear approximation. Although
the damping factor may slow the
optimization, it is more robust
for highly nonlinear targets.
Optimization
67
Global Optimization
Local optimization uses gradient search to find the
nearest merit function minimum and moves downhill. Global optimization
attempts to find the global
minimum by allowing both
uphill and downhill movement in the merit function.
However, global optimization can require extensive
computation time.
In many cases, the global merit function has a reduced
set of image performance operands and more physical
constraints (e.g., maximum ray AOI) than a local merit
function in order to save time but keep the design
solutions realistic. The result is that global optimizers
do not generally find the absolute best merit function
minima. Global search designs should be followed by a
more-comprehensive DLS local optimization to find the
optimum design solution.
Adaptive simulated annealing algorithms mimic the
physical process of cooling excited atoms and molecules,
where the system has a nonzero probability of reaching a
higher energy state (moving uphill). Genetic algorithms
encode lens variables in a numeric sequence and
selectively choose genes from best-performing parents
to create a new system.
Global optimizers are most often used as global search
algorithms to rapidly scan parameter space for multiple
solutions. Global searches require only a minimal starting
designa series of plane parallel plates with a properly
defined merit function can yield good results. However,
there is no guarantee that global optimization will ever
find the absolute global minimum for a given problem.
68
Optimization
Merit Function Construction

In optimization, a merit function is a single number
that captures all aspects of desired lens performance. The
merit function is constructed by taking the RMS of all
identified operands, where
m
X
i =1
wi (c i t i )
m = number of operands
w i = weighting factor for operand i
c i = current value for operand i
t i = target value for operand i
Squaring each operand serves to magnify the operands

with the worst performance and ensure that positive
and negative operand values do not offset each other in
the sum. Individual operands are relatively weighted
to emphasize their desired contribution to overall
performance. The target value for most operands is zero.
Most design programs have a default merit function
that automatically defines a set of image quality
operands and includes the appropriate pupil sampling and
weighting for fields and wavelengths.
Merit function operands consist of both image quality
measures (e.g., RMS spot diameter) and construction
parameters (e.g., focal length or magnification). Ray
position errors and/or OPD values are used to calculate
the image quality measures. For example, a simple
image quality operand is the distance in the image
plane of a single ray from the centroid (or center
of mass) of a large number of rays, all traced from
the same object point. Construction parameters can
be included in the merit function as heavily weighted
operands (to ensure that they are met), but this approach
can significantly slow down improvements in image
quality. Lagrange multipliers are an alternative to
heavily weighted merit function operands and place the
construction parameter constraints outside of the merit
function, allowing for more stable optimization; however,
each Lagrange multiplier constraint requires one free
variable.
Optimization
69
Choosing Effective Variables

Certain variables are more effective at improving lens
performance than others. Radii of curvature and
airspaces are almost always left free to vary during
optimization. Packaging or manufacturing constraints
(e.g., requirements for a flat surface, equal radii, or
working distance) may restrict their range of values.
In contrast, lens thicknesses are much less effective
variables (a change in a lens thickness is often equivalent
to a change in a nearby airspace). Element thicknesses are
often kept frozen during a majority of the design process.
However, in some cases (e.g., thick-meniscus, Petzvalcorrecting lenses), the thickness of a lens is an essential
variable in improving image quality; thicknesses may also
need to be released to meet edge thickness constraints. If
allowed to vary, optimizers tend to make lenses very thick
(or very thin), slightly improving Petzval with only a small
improvement in the merit function.
A lens element whose thickness is chosen to maintain a
diameter-to-thickness or aspect ratio between 10:1 and
5:1 is often easier and less costly to manufacture.
Glass choice is a critical variable for color correction as
both the index of refraction and dispersion are dependent
on the material choice. Glass differs from most variables
as it is a discrete rather than a continuous variable.
Model or fictitious glasses allow continuous index and
Abbe number for optimization, but a real material must
eventually be substituted from a glass map.
Aspheric coefficients are important variables for
aberration correction in modern designs. If possible,
aspheric coefficient variables should be enabled during
optimization one at a time (starting with the lowest
orders), one surface at a time. It is not necessary in most
designs to vary the conic constant simultaneously with the
fourth-order polynomial coefficient; rather, vary the conic
on strong surfaces and the fourth-order term on flatter
surfaces.
70
Optimization
Solves and Pickups

Solves define a surface curvature or thickness to be
a function of a given system requirement. As the lens
changes, a solve modifies the selected parameter to
maintain the system requirement. Solves are typically
used in the initial setup of a lens, but some lens
requirements such as focal length can be constrained
using solves rather than being defined as merit
function operands. This method decreases the number
of independent variables and corresponding constraints
during optimization, reducing computing time. Examples
include:
A marginal ray angle solve or f /# solve on the
curvature of the last surface can be used to maintain
the system focal length for a fixed entrance pupil size.
A marginal ray height solve on the thickness of the
last surface can be used to locate the image plane at
paraxial focus. If an internal image exists, this solve can
also be used to find and hold its location.
A chief ray angle solve on the curvature of the last
surface can force image-space telecentricity.
Aplanatic solves force the aplanatic condition on a
surface by changing its curvature.
Thickness solves can be used to control lens edge
thicknesses or the total distance between separated
surfaces (useful in zoom or multi-configuration designs).
A magnification solve adjusts the object (or image)
distance to maintain a system magnification.
Many marginal-ray and chief-ray solves require ray
information based on the entrance pupil and cannot be
placed on surfaces before the aperture stop.
A pickup is a special type of solve that picks up or uses
data from another surface. It maintains a set relationship
between two lens parameters as one of them is varied.
Pickups are very useful for optimizing double-pass or
symmetrical systems (e.g., to allow a lens to change its
radius yet maintain symmetry, the first radius is picked
up by the second radius as its negative value).
Optimization
71
Defining Field Points

A lens should be optimized at multiple field points. The
number and location of the field points heavily influence
the optimization process and the final design result.
To ensure that the lens performance does not oscillate
between field points, a final design should be evaluated
at more field points than were used for optimization,
especially if aspheric surfaces are used on field lenses.
For a rotationally symmetric lens, the
field is traditionally divided into circles with equal areas. The normalized
radius of each circle is the normalized field coordinate H . In lenses
with moderate field (<25-deg FFOV)
and no aspheres, three fields (H =
0, 0.7, and 1.0) are generally sufficient
for optimization. For systems with a rectangular sensor,
it is common to circumscribe a circle around the sensor whose diameter equals the diagonal of the sensor,
and this is used to define the field. Asymmetric systems
(e.g., TMAs) require fully defined
rectangular fields in all four quadrants because no rotational symmetry
exists. In some cases, lateral or reflection symmetry reduces the number of
fields required for optimization.
Common field definitions are object height, object
angle, and image height. In complex systems, using
image height fields often requires ray aiming of
reference rays to avoid ray trace failures.
Fields can also be given field weights to balance
or emphasize on- or off-axis performance for a given
application. While projection and lithography lenses
require uniform image quality over the entire field, visual
and photographic systems often allow some performance
loss toward the edge of the field.
72
Optimization
Pupil Sampling
Pupil sampling defines the number and the distribution

of the rays traced through the pupil and is critical
to both optimization and analysis. Tracing fewer rays
leads to faster optimization but also to reduced accuracy.
Obscurations and vignetting will influence the required
pupil sampling and the algorithm used to generate rays.
Pupil sampling methods may weight rays differently
according to their pupil coordinate. The pupil sampling
ray density should be increased for systems with aspheric
surfaces or large aberrations.
Common pupil sampling methods include:
Rectangular grid sampling assumes no symmetry
and often requires a high-density grid 10 10 to obtain
accurate sampling. Rays are generally evenly weighted.
Polar or hexapolar sampling uses periodically spaced
radial and azimuthal rays, and generally requires fewer

rays than rectangular sampling. Rays may be weighted
by ring (concentric circle) or arm (radial line).
Gaussian quadrature (GQ) sampling uses a very
small number of skew rays at very specific pupil

coordinates and weightings. GQ sampling returns a
mathematically exact integral of the pupil with fewer
rays and provides higher sampling near the edge. GQ is
the fastest sampling for the majority of cases.
GQ sampling works best with circular or elliptical pupils

and may not give accurate results with highly obscured or
very noncircular pupils (unless specifically modified in the
design code for this special case).
Tolerancing
73
Tolerancing
Tolerancing is a statistical process that determines
the allowable change or tolerance in each of the
lens construction parameters. All lenses must be
toleranced before being built. The toleranced design needs
to simultaneously satisfy the performance requirements,
minimize manufacturing costs, and minimize alignment
complexity. These goals tend to be contradictory, so a large
part of tolerancing is finding the best compromise among
them.
Tolerances include both optical print values (e.g., radius of
curvature, center thickness, index variation, wedge) and
mechanical print and assembly values (e.g., tilt, decenter,
axial spacing, subassembly alignment). Compensators
are parameters that can be adjusted during the lens
build (e.g., focus, airspace changes, active element
centering) to recover performance losses caused by
other tolerances. Tolerances are often separated into
symmetric (compensated by focus and space adjust)
and asymmetric tolerances (compensated with image
tilt and push-arounds). Tolerances that cannot be easily
modeled (e.g., material inhomogeneity or complex surface
figure errors) form part of an unallocated performance
budget, requiring extra design margin for the unknown
variations. Many different tolerance methods exist, but
they generally follow the same basic process:
1.Choose initial tolerance values for all parameters.
2.Define the performance metrics (e.g., MTF, RMS spot
diameter, boresight error) and the requirements.
3.Run a sensitivity analysis to determine the impact on
performance from each tolerance, and identify sensitive
and cost-driving tolerances.
4.Define compensators and their allowable ranges.
5.Run appropriate statistical analyses (e.g., Monte
Carlo analysis) and evaluate the expected as-built
performance and manufacturing yield.
6.Adjust tolerances and compensators until cost and
performance goals are met or a redesign is needed.
74
Tolerancing
Design Margin and Performance Budgets

Design margin is the
difference between the
required as-built performance and the nominal design performance. The design margin
should be large enough
to allow for all manufacturing errors and
environmental changes. However, the best design for manufacturing is not always the one with the largest design
margin. A lens with slightly worse nominal performance
is usually less sensitive to manufacturing errors, resulting in a much higher yield and lower cost of production.
Designers determine the most manufacturable design by
comparing both tolerances and predicted as-built performance of each design solution.
The design margin for a single top-level requirement
can be allocated to several categories in a performance
budget. The allocations in each category are then
either summed or root-sum-squared (RSS) to a top-level
number. In addition to tolerancing, the budget values
are used for design targets, mechanical structure design
(including thermal management and resonance), and
other environmental requirements. There may be more
than one performance
budget necessary (e.g.,
RMS OPD for image
quality and pointing accuracy for boresight).
Complicated lenses with
many
subassemblies
may also require a performance budget where
each subassembly is
broken out separately.
Tolerancing
75
Optical Prints
An optical drawing or optical print consists of all of
the design and tolerance information necessary for the
fabrication of the part, including:
Material index of refraction, dispersion, and quality
Radii and surface figure (power and irregularity)
Center thickness (CT) and edge thickness
Special surface parameters (e.g., aspheres)
Surface finish (roughness) and quality (scratch-dig)
Outer diameter (OD), surface sag, and edge beveling

Wedge or total indicator runout (TIR)
Clear aperture (CA) per surface
Antireflection (AR) or other coating requirements
Talk with lens fabricators early in the design process

to discuss tolerance requirements, as drawing formats
and achievable tolerances vary from vendor to vendor.
It is also important to make sure that the vendors
measurement methods support your specifications.
Optical prints typically isolate singlets or cemented
doublets or triplets.
Higher-level assembly
tolerances (e.g., group
or barrel axial displacement, and group
tilt and decenter tolerances) are listed on assembly prints or assembly drawings.
ISO 10110 is the international standard
for optical drawings,
although many organizations develop their
own customized drawing formats.
76
Tolerancing
Radius of Curvature Tolerances

Radius of curvature (ROC) tolerances refer to the
change in vertex curvature of a surface and are
specified differently depending on how the lens is to
be measured. Some designs tolerance ROC directly,
assuming a measurement of ROC with either a
spherometer (contact measurement) or a distancemeasuring interferometer (noncontact measurement).
Other designs use test plate fringes to indirectly
tolerance ROC. In this case, it is common to list two
tolerances for ROC, a radius tolerance and a test plate
fringe tolerance, which can appear confusing. The radius
value tells the fabricator to use a test plate within a
certain radius range, and the number of fringes gives the
allowable part deviation from the test plate.
To compare the sensitivity of different radii in a design,
do not apply a uniform radius tolerance in length
units across the lens; instead, use a uniform tolerance
of curvature ( c = 1/R ) or percentage of radius (%R ).
Test plates are precision surfaces used during polishing to
monitor both ROC and surface
irregularity. They are placed in
contact with the lens and viewed
under monochromatic light (typically a 546.1-nm mercury lamp),
forming fringes (or Newtons
rings). The ROC can be calculated from the part diameter and either the number of fringes (flat test plate) or the
curvature of the fringes (spherical test plate). Test plate
fringes represent a double-pass measurement, where one
fringe corresponds to a half-wave of surface deviation. Designs are usually test-plate fit to a specific suppliers catalog of test plates to avoid large test plate tooling costs.
Each radius in the design is individually altered to match
the closest catalog test plate value, and the entire lens is
re-optimized after each change.
Tolerancing
77
Surface Irregularity
Surface irregularity refers to any
measured deviation of an optical surface from its intended shape. During a
standard polishing process, irregularity
is monitored with a test plate by assessing the number of fringes and/or their irregularity. For more-accurate, noncontact measurements,
a surface-measuring interferometer is used to analyze the
surface irregularity.
Thin elements can spring when removed from a
polishing block, resulting in two surfaces with large but
approximately equal and opposite cylinder error. If the
clear aperture on each surface is about the same, the
effect of such cylinder error on lens performance nearly
cancels when used in transmission.
The wide range of possible shapes for surface irregularity makes it difficult to model and tolerance deterministically, and it is often simply specified in fringes or waves of
peak-to-valley (PV) error. Power and cylinder (astigmatism) are common low-order errors produced by standard polishing processes, whereas higher-order errors include an edge roll or edge rip. Irregularity specification
and measurement are sensitive to the exact CA called out.
Allowing extra room between the CA and the outside part
diameter makes it easier to achieve high-quality surfaces.
Simple tolerance models assume that the irregularity
error is pure cylinder. Compensation schemes may include
rotating or clocking an element. More-complex tolerance
models use a combination of Zernike polynomials,
either toleranced individually or as an aggregate RMS
value of all coefficients up to a certain order. Irregularity
from sources such as mounting stress must also be
considered, especially for mirrors. Mechanically induced
surface deformation often has three-way symmetry (3point or trefoil) from kinematic mount designs.
78
Tolerancing
Center Thickness and Wedge Tolerances

A center thickness (CT) tolerance specifies
the allowable error in the vertex thickness
of an element. Fabricators typically leave
CT on the thick side of the tolerance range
to allow for removal of cosmetic defects or
rework during final polishing. The sensitivity
of a design to a CT error is often similar in
magnitude to the sensitivity of the design to
an airspace error on either side of the element.
Thin lenses with diameter-to-CT ratios greater than
about 15:1 typically require special handling during
fabrication, increasing their manufacturing cost.
A wedge tolerance in a spherical element indicates the amount of allowable tilt between the optical axis
of its two surfaces (defined by connecting their centers of curvature)
and a mechanical axis [usually the
outside diameter (OD) of the element]. Wedge is measured on a spinning chuck, either by optical methods such as angular
beam deviation or by mechanical methods such as total
indicator runout (TIR) or edge thickness difference
(ETD). The fabrication process of centering or edging a
lens reduces element wedge and sets the final part diameter. Aspherical elements require special treatment (both
for tolerance modeling and fabrication) because they do
not have a clearly defined optical axis between their two
surfaces.
Meniscus lenses with nearly concentric radii are
extremely sensitive to wedge errors. They can also be
difficult to center because it is hard to remove the wedge
without significantly reducing the part diameter and
encroaching on the CA.
Tolerancing
79
Material and Cosmetic Tolerances

The various parameters required to specify optical
materials must also be toleranced in a lens design. The
two most common tolerances are index of refraction
n and dispersion V . A variation of n 0.001 and
V 0.8% is standard for most commercial glasses. Tighter
tolerances are available at higher cost. High-performance
designs must also consider index inhomogeneity and
birefringence. Inhomogeneity is the PV variation in
bulk index throughout the lens volume and is typically
less than 5 106 for commercial glass. Birefringence is
defined as the maximum OPD between polarization states
over a standard path length (typically less than 20 nm/cm
for commercial glass). The complexity of modeling the
inhomogeneity and birefringence errors often push them
into part of the unallocated performance budget, but such
errors can be toleranced effectively as surface or wavefront
irregularity.
Many lenses are overspecified for cosmetic errors, leading
to reduced yield and additional cost. If the lens surface
is not near an image or visible to the user, demanding
cosmetic specifications may not be required.
Cosmetic defects include scratches, digs, pits, and
staining. Such defects generally have little influence on
imaging performance, with a few exceptions:
Defects in lenses sensitive to stray light will scatter light
onto the focal plane and may add background noise. The
amount of light intercepted by the defect is proportional
to the ratio of the defect area to the beam area and is
generally scattered over 4 sr.
Defects in lenses for high-power laser applications may
act as damage centers or stress intensifiers.
Defects near an image surface or subtending a large
fraction of the beam footprint on a given surface may
appear in the image.
Cosmetic defects and bulk material defects (e.g., bubbles,
inclusions, and striae) are hard to tolerance accurately
and are often part of the unallocated performance budget.
80
Tolerancing
Lens Assembly Methods

To properly tolerance a lens assembly, it is important
to anticipate how the lens elements will be mounted
and aligned. In general, as the required lens centration
and axial spacing tolerances get tighter, the mechanical
complexity, lens assembly time, and cost increase.
Adhesives and flexures can be used to reduce mountinginduced surface deformation. Common assembly methods
include (from low to high precision):
Drop-in assembly: Elements are dropped into a
lens barrel, separated by spacers, and held in place
with a mechanical retainer. The assembly tolerances are
driven by the achievable fit between the element OD and
the barrel inner diameter (ID). These assemblies work
best with low-cost targets in high-volume applications
where ease of assembly is paramount.
Shim-centered assembly: Elements are centered
radially in a lens barrel using shims and then bonded in
place. This method is used for lenses that require both
tight centration tolerances and large radial gaps to allow
for thermal expansion.
Poker-chip assembly: Each element is accurately
aligned to the reference surface of an individual lens cell
and bonded in place. Cell and spacers are inserted as a
stack into a precision-machined lens barrel.
Bolt-together assembly: Each element is accurately
aligned to the reference surface of a precision-machined
lens cell and then held in place on a seat with either
adhesive and/or flexures. The lens cells are precisionaligned to each other either optically or mechanically
and then bolted together.
Tolerancing
81
Assembly Tolerances
Assembly tolerances, such as
decenter, tilt, roll, and axial
spacing, represent positioning
errors of optical surfaces in a
lens system. Assembly tolerances may be applied to lens
surfaces, lens elements, and/or
element subassemblies. Accurate tolerancing requires knowledge of the optomechanical
design and mechanical part tolerances, and a well-defined
assembly plan.
Multiple decenter errors can be added together and
toleranced as a single motion. However, multiple tilt
errors with their own individual pivot points must be
toleranced as separate motions.
Most assembly tolerances are not derived from a single
part error but must be calculated from multiple part errors
using a tolerance stack-up. For example, in a simple
mount, a single airspace tolerance can depend on five
separate part tolerances: two lens surface or bevel sags,
a spacer thickness and diameter, and a lens CT. The
impact of a CT error on an airspace also depends on how
the element is mounted. Elements separated by spacers
shift all subsequent surfaces by their CT errors, thus
maintaining the airspaces between elements, whereas
the CT error in an element mounted on an independent
mechanical shoulder in a lens barrel also changes the
neighboring air space value but may not affect the position
of other elements.
Tilt and decenter tolerances break rotational symmetry
introducing boresight error and field tilt as well as
axial coma, axial astigmatism, and axial lateral color. To
properly apply a tilt tolerance, it is important to include
the pivot point for the tilt motion. For cemented elements,
the tilt and decenter of one element with respect to the
other is specified as roll about the cemented interface.
82
Tolerancing
Compensators
Manufacturing errors that have a similar effect on
performance can often be corrected by modifying a single
lens parameter. Compensators are lens variables that
simulate adjustments made to the lens during assembly
and test to improve as-built performance. The most
sensitive tolerance parameters are usually selected as
compensators. Compensators loosen tolerances and reduce
part cost but add time (and cost) to assembly and
test. Without compensators, most tolerances would be
unreasonably tight. For each compensator, the tolerance
process must define the adjustment range and its
required resolution and accuracy. The optomechanical
design typically constrains the compensator range of
motion.
Compensators should be limited in number, independent
(two compensators should not affect/correct the same
aberration), and practical (e.g., glass dispersion is not a
realistic compensator).
Focus and image plane tilt are almost universally
used as image plane compensators to loosen tolerances.
A custom spacer may be designated as a post-build
airspace compensator whenever focus is insufficient

to control the spherical aberration introduced by
symmetrical tolerance errors.
Asymmetrical tolerance errors typically cause axial
coma. This can be corrected during test by decentering a
single lens element known as a push-around.
In complex lenses, several airspaces can be optimized
prior to assembly using the as-built measured data
(radii, CT, airspaces, and index) in a space adjust.
Tight index-of-refraction or dispersion requirements
may require a melt-recompensation process, in which
the indices of refraction of the actual pieces of material
to be used in the assembly are measured. The design is
then re-optimized (usually only the airspaces are varied)
using the measured glass material data. This method
can help loosen material tolerances.
Tolerancing
83
Probability Distributions
The tolerance extremes for each lens variable limit only
the maximum and minimum values. The probability
distribution function (PDF) defines the likelihood
of obtaining an as-built value between the extremes.
The tolerance PDF depends on the fabrication and/or
alignment process (which may even be specific to a given
supplier) and should be defined for each tolerance in a
Monte Carlo tolerance analysis.
CT tolerances often have skewed PDFs because polishers
will leave lenses on the thicker side of nominal to allow
for possible rework due to cosmetic errors.
Uniform distributions are constant
over the tolerance range with equal

probability between the extremes.
Normal or Gaussian distributions
have a defined mean and standard

deviation, where the mean is the
nominal design value and is the most
likely value to occur.
Because
normal distributions allow

infinite range, a truncated Gaussian
is often used, where the probability is
defined to be zero beyond the tolerance
extremes. This distribution represents
a supplier yield where parts out of
specification are not used.
Skewed Gaussian distributions have
a nonzero higher-order moment, where

the mean or nominal value is not the
most likely value.
Spiked or parabolic distributions
are heavily weighted toward the tolerance extremes, where the nominal design value is the
least likely. They can also simulate binary tolerances.
Most design codes permit the use of custom distribu-
tions to incorporate measured statistical data.
84
Tolerancing
Sensitivity Analysis
A sensitivity analysis determines
the sensitivity of a lens to manufacturing errors by calculating the change
in each performance metric (e.g., RMS
wavefront or MTF) for both a plus and
a minus change (equal to the tolerance value) in each tolerance parameter. Performance changes may not be
symmetric about the nominal design value and often vary
with field point. The analysis is initially performed uncompensated (although focus is regularly included in the
initial run). If needed, a second compensated run (where
the compensator motion is optimized to improve performance) helps gauge the effectiveness of the compensator
and identifies tolerances that drive large compensator motions. During the compensated run, constraints are often
removed from the merit function (e.g., focal length) that
can interfere with performance improvement.
An inverse sensitivity analysis determines the tolerance value that produces a given change in performance,
with the goal of having each tolerance contribute equally
to the performance degradation of the system.
It is often useful to apply the same tolerance value
to all surfaces (or elements) for common tolerance
parameters (e.g., CT 0.025 mm or index 0.001)
and then group the results into tables and/or charts.
A sensitivity table or
chart lets you identify at a
glance the tolerances that
will drive the performance
loss and highlight possible compensators. The tables and charts are also
very useful in comparing
the sensitivities of different
design solutions to possible
manufacturing errors.
Tolerancing
85
Performance Prediction
A basic sensitivity analysis considers the effects on system
performance for each tolerance individually. In order to
predict the image quality of an as-built system, it is
necessary to simulate the effect on performance of all
of the tolerances simultaneously. Most design software
packages compute the as-built performance prediction
using a root-sum-square (RSS) calculation, displaying
sensitivity analysis results in tabular or graphical form.
This approach works well if the performance losses from
all of the tolerances are of similar form.
The RSS of the performance change from each tolerance
gives an estimate of overall predicted performance;
however, it generally overestimates the performance loss
when compared to a Monte Carlo analysis with the
appropriate PDFs and tolerance interactions (e.g., a lens
radius may have a different sensitivity if the radius on the
other side of the lens also changes).
In some cases, part tolerances are divided into symmetric
(surface-centered) or asymmetric (non-surface-centered)
tolerances for separate tolerance runs:
Symmetric
Curvature (radius)
Center thickness
Airspace
Index of refraction
Dispersion
Asymmetric
Surface tilt/decenter
Lens tilt/decenter
Group tilt/decenter
Surface irregularity
Index inhomogeneity
Wavefront differential or derivative tolerancing

generates a multivariable differential tolerance sensitivity matrix containing first-order (linear), secondorder (quadratic), and mixed partial derivatives or cross
terms. It requires a one-time analysis of the nominal
system and can be hundreds of times faster than a full
Monte Carlo analysis.
86
Tolerancing
Monte Carlo Analysis

A Monte Carlo analysis is an alternative tolerancing
approach to predict as-built performance. It simulates
a manufacturing environment by generating a series of
lenses or trials. Each trial run randomly perturbs all
tolerances within its defined extremes according to its
PDFs, and then evaluates the image performance criteria.
By considering all applicable tolerances simultaneously
and exactly (no approximations are made), a highly
accurate simulation of expected performance is possible.
If compensators are needed, each trial can be reoptimized
with only the compensators as variables. If allowed,
relaxing exact constraints (such as focal length or working
distance) in the merit function during compensation can
significantly loosen the symmetric tolerance values.
A rule of thumb for a Monte Carlo analysis is that the
number of trials should exceed the square of the number
of tolerances for good statistical validity.
The result of a Monte Carlo analysis is a probability distribution table and/or cumulative distribution function (CDF) or yield curve showing the percentage of lenses that meet or exceed a given performance criterion. The number of random trials evaluated determines the maximum statistical confidence
level. Because tolerancing is
statistical, a designer needs
to decide in advance what is
an acceptable manufacturing yield (e.g., 98% pass and
2% fail). Individual trials can
be saved for forensic analysis
or to investigate poorly performing outliers.
Monte Carlo analyses generally require significant
computation time and can only give statistical results,
which may not be valid for one-off builds or very small
production runs.
Tolerancing
87
Environmental Analysis
A complete tolerance analysis must include the performance impact on a lens due to changes in its environment.
Environmental requirements such as temperature or
pressure can consist of multiple specification ranges. The
operating range dictates the total extent over which the
lens must meet all performance requirements. A survival
range gives the extent over which the lens need not meet
performance but cannot break or become nonfunctional.
The sensitivity of the lens to environmental perturbations
must be assessed and allocated as part of the performance
budget using an environmental analysis.
In addition to ambient temperature changes, heat
sources such as electronics, motors, light absorption, and
solar radiation can all induce thermal effects in lens
systems. Two key material properties are needed for an
accurate thermal model: The coefficient of thermal
expansion (CTE) gives the percentage change in linear
dimension per unit temperature change, and the glass
thermo-optic coefficient gives the absolute change in
refractive index per unit temperature change. Simple
tolerance models specify the temperature change as
a thermal soak or steady-state effect; however, morecomplex models can include both spatial and temporal
temperature gradients. Pressure or altitude changes
primarily affect the refractive index of the gas material
surrounding the lens elements (typically air) but can also
alter the airspaces between lenses or change the shape of
optical surfaces in sealed assemblies.
Most design codes provide the ability to perturb the lens
temperature and/or pressure and evaluate performance
changes as long as the appropriate material data is
entered. However, complex environmental influences need
finite element analysis (FEA) to help predict changes
in performance [e.g., vibrations can disrupt component
alignment and need FEA to model the vibration effects on
line-of-sight (LOS) stability and image jitter].
88
Tolerancing
Athermalization
As the temperature of an optical system changes, the
lens elements and housing physically expand/contract.
In addition, the glass refractive index will also change.
The primary result of both changes is a focus (or
image plane) shift, although aberrations and pointing
errors (from lateral component shifts) are also possible.
An athermalized optomechanical design minimizes
performance variation with temperature. For example,
focus shift can be actively athermalized by measuring
the temperature and then physically moving either the
lens or the sensor. This approach is not the ideal solution
in many situations, and passive athermalization
(no motor-driven component movements) is employed
with the use of dissimilar materials (e.g., optical
materials with different thermo-optic coefficients or metal
mounting materials with different CTEs). Most infrared
materials have large thermo-optic coefficients, making
athermalization mandatory for large temperature ranges.
A reflective system can be passively athermalized

against isotropic thermal soak conditions by making the
mirrors and the mounting structure out of the same
material. The system then scales in dimension as the
temperature changes but maintains overall focus and
imaging performance.
Stray Light
89
Stray Light Analysis

Stray light is any unwanted
light that reduces performance
by striking the image plane of an
optical system. Primary sources
of stray light include reflections
from uncoated or poorly coated
surfaces, surface or bulk scattering, thermal emission, bright
out-of-field objects, glints from
mechanical structures, diffracted
light from edges, and unwanted diffraction orders from
gratings.
Stray light can be divided into two main categories: ghost
images and veiling glare. Ghost images have structure
(often with sharp edges) and create local image-quality
disruptions. Veiling glare produces a diffuse background
haze on the image plane, resulting primarily in a loss of
overall image contrast.
Tracing rays backwards from the detector (or putting
your eye at the detector) is a highly efficient technique
for isolating potential stray light paths and is much more
effective than tracing from a source to a detector.
A stray light analysis determines both the amount of
stray light at the detector and the different paths that
it took to get there. If the stray light is large enough to
cause a problem, the analysis should also identify any
design changes needed to reduce the stray light. Some
key metrics used to quantify stray light include off-axis
rejection (OAR), point-source transmittance (PST),
point-source normalized irradiance transmittance
(PSNIT), and Narcissus-induced temperature difference (NITD) in infrared systems. The accuracy of a stray
light analysis is limited by the availability of source and
scatter data, the completeness of the computer model, and
the available computer time.
90
Stray Light
Stray Light Reduction

Identifying potential stray light paths during the design
phase is essential to avoiding costly errors in hardware.
A critical object is any surface that can be seen by the
detector, including surfaces that are imaged through lens
elements. An illuminated object receives power directly
from the source and depends on the source location. The
overlap between the list of critical and illuminated objects
highlight key stray light paths. An object on both lists
is called a first-order stray light path, which is a
major contributor to stray light and should be addressed.
Second-order stray light paths exist anywhere that
light can propagate (transmit, reflect, scatter, or diffract)
from an illuminated object to a critical object.
Field stops reduce the number of illuminated objects in a
design because objects behind the field stop cannot be seen
by the sourceonly objects in front of the field stop can be
illuminated. Similarly, placing the aperture stop as close
as possible to the detector reduces the number of critical
objects in a design since objects in front of the aperture
stop cannot be seen by the detector. This approach may
require that the aperture stop be relayed from the front
of the lens system to the rear because the farther the
aperture stop is from a lens, the larger the front elements
become in diameter. Lyot or glare stops are apertures
placed in conjugate pupil planes; such stops might need to
duplicate any aperture stop obscurations from mounting
features for extreme stray light control.
Identified stray light paths can be
eliminated by blocking light with
baffles, vanes, and knife edges,
or by relocating an object so that
it is no longer seen by the detector
or illuminated by a source. Mechanical surfaces and
edges of optical elements can be treated to absorb or
scatter light by blackening or roughening. Rifling or
threading barrels and mechanical surfaces is a common
and economical way of reducing stray light.
Stray Light
91
Antireflection (AR) Coatings

Optical surfaces require antireflection (AR) coatings
to reduce stray light and improve transmission. For an
uncoated surface, the amount of Fresnel reflection loss
depends on the angle of incidence of the light and the
index of the lens material. For light normally incident on a
surface in air, the reflection loss is 4% for a visible glass
with an index of 1.5. In the infrared spectrum, the loss per
uncoated surface can be as much as 36% due to the high
index of infrared materials.
Large changes in ray AOIs across a lens surface can
lead to highly nonuniform pupil transmission and pupil
polarization in high-NA systems. Steep ray angles of
incidence on an AR-coated surface will also blue-shift
the spectral region of a coating, especially coatings with
sharp band edges.
The ideal single-layer ARcoating material for a glassto-air interface has a quarter-wave of optical thickness
and an index of refraction
equal to the square root
of the lens material index.
This condition is difficult to achieve in practice. The closest
material choice for the visible spectrum is a single quarterwave layer of MgF2 (index 1.38), which will reduce the
surface reflection from 4% to 1% at a single wavelength.
More-complex coatings are made of alternating layers of
high and low index, each with a thickness equal to a
quarter-wave optical thickness. Two- and three-layer Vcoats can eliminate the reflection at a single wavelength,
whereas more-complex broadband AR (BBAR) coatings
reduce the reflection loss to <0.5% over an extended
wavelength band. A variety of coating design optimization
software packages exist (very similar to lens design). Most
lens design codes will model (but not optimize) thin film
stacks for accurate transmission calculations.
92
Stray Light
Ghost Analysis
Ghosts are a form of stray light from unwanted
reflections. Traditional ghosts are double-image artifacts
formed by two lens surfaces (two-bounce ghosts) with
imperfect coatings. However, ghosts can also result from
reflections from object, filter, detector, and mechanical
surfaces. In interferometer optics, single-reflection ghosts
create extra reference beams that complicate fringe
analysis. In systems with high-power lasers, ghosts can
produce an internal focus inside a lens material or on
a lens surface that may actually damage the lens. In
thermal infrared lenses, Narcissus reflections from cooled
detector features cause image artifacts. A pupil ghost is a
special ghost image that echoes the shape of the aperture
stop and is typically seen in camera lenses when a bright
source (such as the sun) is inside the object field.
Windows, beamsplitters, concentric mensicus lenses, and
optical surfaces with near-normal incident rays are the
primary sources of in-focus ghost images.
Most of the potential ghost
images in a lens system
are highly aberrated and
out of focus. If the ghost
image is significantly defocused from the image
plane, it produces a general haze on the detector,
reducing image contrast. A
ghost analysis is primarily a graphical technique
whereby sequential ray trace routines generate graphical
ghost paths that analyze every potential two-bounce reflectance in the optical system and then report the image
quality of each ghost in terms of spot size, defocus, magnification, and f /#. The relative intensity of a ghost image
can be estimated from its number of reflections; however,
nonsequential ray tracing is required for a quantitative
radiometric ghost analysis.
Stray Light
93
Cold Stop and Narcissus

Many infrared systems require
a cyrogenically cooled detector
that sits in a thermally insulated
vacuum dewar, reducing background noise and increasing sensitivity. Because the system components also emit the same wavelengths of thermal radiation as
those trying to be imaged by the detector, a cold shield is
needed to minimize the number of warm surfaces (outside of the object scene) that the detector can see. For optimum stray light control, the lens exit pupil should match
the detector cold shield in size and location. Imaging the
lens stop onto the cold shield forms a cold stop, and if
imaged perfectly, the system is said to have 100% cold
shield efficiency.
Surfaces inside the dewar (e.g.,
the focal plane and the cold stop)
are at significantly different temperatures than the scene imagery,
and their reflections from an optical surface will form cold Narcissus ghost images. Such
images are either diffuse background noise or sharp image
artifacts varying in field or time. Retroreflection due to a
small marginal ray angle-of-incidence at a surface or image reflection due to a small marginal ray height are primary sources of Narcissus. The YNI product combines
these into a paraxial quantity that can be used to optimize the lens or locate baffles. Staring detectors allow a
nonuniformity correction (NUC) to calibrate out static
reflections, but any lens motion or environmental changes
after NUC will induce Narcissus. The NITD is the sum of
reflected energy from all optical surfaces.
Techniques for minimizing Narcissus include antiNarcissus coatings, baffles, tilting flat windows and filters,
and designing with Narcissus ray constraints.
94
Stray Light
Nonsequential Ray Tracing

Nonsequential ray tracing (NRT) is an extremely
powerful analysis technique for evaluating stray light.
Unlike standard sequential ray tracing in which rays
intercept surfaces only once and in a given order, NRT
allows a ray to see all of the surfaces in a system
simultaneously and the ray interacts with the first surface
it meets. It is possible for a ray to intersect a single surface
multiple times. NRT is also widely used to design and
optimize illumination and other nonimaging systems.
For accurate results, many orders of magnitude more rays
need to be traced with NRT than sequential ray traces. As
a result, stray light analysis can be time consuming and
computationally expensive.
A stray light analysis with
NRT requires three components:
sources that generate rays, objects that interact with rays,
and receivers that collect the
rays and quantify the resulting
irradiance or intensity distributions. Key features that can be enabled depending on computation limitations are ray scattering and ray splitting, which split energy between a parent ray and multiple child rays. The sum of the transmittance, reflectance,
scattering, and absorption on a surface should equal unity.
Ray databases that include all ray-surface data can be
saved for subsequent filtering and analysis.
NRT models are typically much more complex than
sequential optical surface design models and can include:
multiple sources with rays propagating in any direction;
mechanical features such as lens mounts, barrels,
apertures, baffles, and vanes;
absorption and scattering from bulk materials; and
absorption and scattering from mechanical and optical
surfaces (ground, polished, coated, or painted).
Stray Light
95
Scattering and BSDF

Optical and mechanical
surfaces have a finite surface roughness that can
induce light scattering.
The direction and amount of scattered light depend on the
incoming ray wavelength, polarization, and angle of incidence. Because the scattered light is a potential source of
stray light, a complete stray light analysis requires accurate scatter models for each surface. Ideal scatter models assume that the scatter is either all concentrated in
the specular direction (reflected angle governed entirely
by Snells law) or is perfectly Lambertian (diffuse surface with the same reflected radiance over all viewing angles). Real surface scatter usually lies somewhere between
these two extremes, often with most of the light scattered
around the specular reflection direction.
The bidirectional scattering distribution function
(BSDF) indicates how specular or diffuse a surface is as a
function of both the 2D incident angle ( i , i ) and the 2D
scattering angle (o , o ):
BSDF ( i , i ; o , o ) =
L 1
.
sr
E
The BSDF is defined as exiting radiance divided by

incident irradiance and is centered around the specular
angle defined by i = r . A Lambertian surface has
a constant BSDF for all angles of incidence. The
reflective and transmissive versions of the BSDF are
the BRDF and BTDF, respectively. A scatterometer is
used to measure angle-resolved BSDF, BRDF, and BTDF.
Isotropic surfaces, such as pitch-polished lenses, scatter
light equally in all azimuthal directions, simplifying
the BSDF function. Anisotropic surfaces, such as
diffraction gratings, diamond-turned mirrors, or brushedmetal surfaces, have inherently azimuthal directionality.
Total integrated scatter (TIS) is the integral of the
BSDF and is a common scalar metric for total surface
scattering without any angular resolution information.
96
Optical Systems
Photographic Lenses: Fundamentals

A photographic objective (or camera lens) is a lens
assembly that forms a real image of a real object.
Photographic lenses were historically designed to
image on film, but modern digital cameras use CCDs
or CMOS detectors. Camera objectives are typically
optimized for distant objects, incorporating floating
elements to refocus the image onto a fixed plane as the
user changes the object distance.
The expressions fast, slow, and speed of a lens
originate from photography, where a camera lens with
a smaller f /# results in a wider aperture and more light
collected, allowing a user to set a faster shutter speed for
the same exposure time.
A wide variety of applications for camera lenses exist in
consumer (e.g., point-and-shoot cameras, mobile phones,
camcorders), commercial (professional movie cameras),
industrial (quality control), and scientific markets. Most
modern photographic objectives have zoom capability
(to change magnification), and use aspheres and plastic
materials for both size and cost reduction. Cameras
typically operate in the visible, but they can be designed
for any wavelength band where a detector is available.
Camera lenses are classified by focal length and image
format size (e.g., 100-mm telephoto for 35-mm format).
The focal length for a given image format determines
the angular FOV. A normal lens has a focal length
approximately equal to the image diagonal and a FOV
of about 50 deg. A long-focus lens is a narrow-FOV
lens with a long focal length; the most common type
of long-focus lens is a telephoto lens. A wide-angle
lens is a lens with a FOV wider than 60 deg and a
focal length shorter than normal. Fish-eye lenses have
extremely large FOVs (>180 deg). Macro lenses are
special objectives designed for objects that are smaller
than the image sensor (>1:1 reproduction ratio).
Optical Systems
97
Photographic Lenses: Design Constraints

Most cameras need to be handheld, restricting the size
and weight of a photographic lens. A single-lens reflex
(SLR) camera lens has an additional working distance
requirement. A diagonal mirror is located between the
SLR lens and the image plane to allow the user to view
through the lens itself (the mirror flips out of the way just
before exposure). In contrast, a viewfinder camera has
a separate viewfinder so that lens elements can be much
closer to the image plane. However, the viewfinder FOV
does not perfectly overlap the lens FOV, and what you
see is not necessarily what you get.
MTF (in cycles/mm) is used to specify and measure camera
lens performance. Unlike telescope and microscope
objectives, which are usually diffraction limited at full
aperture, photographic objectives are aberration limited
at full aperture and are rarely used wide open. Stopping
the system down decreases the spot size blur due to
aberrations and increases the depth of focus of system. A
speed of f /8 balances diffraction and aberration effects
in SLR photographic lenses.
Photographic lenses have relaxed requirements for relative illumination (RI), the
ratio of on-axis image plane irradiance to the minimum irradiance over the full image
field. This permits designers
to use vignetting to reduce
aberrations and increase the aperture and/or FOV. For
wide-FOV designs, a 50% RI specification is not uncommon. Distortion requirements for lenses with standard
FOVs range from <1% to 4%. However, in wide-FOV lenses
(especially fish-eye lenses), correcting distortion so that
image height is proportional to f tan() is very difficult;
it becomes easier to control f - distortion, where equal
object-angle increments map to equal image-height increments.
98
Optical Systems
Visual Instruments and the Eye

The human eye is a dynamic optical system that can
accommodate varying object distances by changing the
shape and refractive power of its lens. The refractive
power is typically measured in diopters, where 1 diopter
= 1/meter. The eye forms its image on the retina, a curved
surface containing a series of rods and cones that serve
as detecting units. The resolution of the eye varies over
its FOV, with the best resolution found in the central
foveal region where the cone density is a maximum.
The fovea subtends about a 1-deg FOV with roughly 1arcmin angular resolution. The resolution drops outside
this region (e.g., it is 5 arcmin for a 30-deg off-axis field
point).
There is a practical limit on the maximum useful
magnification for visual instruments. At some point,
the image becomes so large that the details in the
image exceed the resolution of the eye. This is referred
to as empty magnification. In applications where
resolved detail is important (e.g., microscopy), this can
lead to artifacts in the viewed image. However, in
applications where the boresight of the image outline
is more important (e.g., riflescopes), some amount of
empty magnification can be beneficial for relieving
eye strain. As a rule, to avoid empty magnification,
the visual instrument magnification should not exceed
11 the aperture diameter (in inches) in telescopes and
225 the NA in microscopes.
Visual instruments, such as telescopes, microscopes,
riflescopes, endoscopes, periscopes, and binoculars, form
images that can be viewed with the human eye. The
image is typically magnified and placed at a convenient
location for viewing, ranging from 25 cm (10 in) to infinity,
depending on the application. Simple visual instruments
consist of an objective and an eyepiece, while more
complex systems add image inverters and/or relays.
Visual instruments should also have a focus adjustment
to accommodate differences between individual users.
Optical Systems
99
Eyepiece Fundamentals
Eyepieces are essential components of visual systems,
magnifying the image created by an objective and placing
it at a suitable location for viewing with the human eye.
The eyepiece focal length determines its magnification. An
eyepiece is usually designed backwards, starting from
the infinite conjugate (eye side) to the short conjugate
(objective side). To avoid vignetting, eyepieces should be
pupil-matched (in both size and location) to the human
iris; this requires that the aperture stop be completely
outside the lens so that the eye can be positioned a
comfortable distance away from the eyepiece.
High-power microscope objectives often have uncorrectable lateral color that can be offset with a compensating eyepiece.
The distance from the aperture stop to the first lens
element (or first mechanical surface) is called the eye
relief. Eye relief requirements vary from as little as
10 mm to simply clear the eyelashes, to 20 mm for
eyeglass clearance, to as much as 5 in. for protection from
high-power rifle recoil. For maximum light collection, the
aperture stop diameter should be matched to the eyes iris
diameter. Eyepieces are typically designed for a 3 to 4-mm
diameter pupil, but this value may vary between 2 mm (for
bright scenes) and 10 mm (for use in moving vehicles).
Due to the external aperture stop requirement, eyepiece
designs have little symmetry, resulting in significant
astigmatism, lateral color, and distortion (e.g., 810%
distortion is not unusual). The off-axis image quality of
eyepieces can be relatively poorthe eye dynamically
scans the image during its normal operation, employing its
low-quality peripheral vision in static situations. Simple
eyepieces with positive elements have large uncorrected
Petzval, whereas more-complex eyepieces use a negative
field lens to flatten the field. The residual field curvature is
typically evaluated in diopters of required accommodation
(defocus).
100
Optical Systems
Eyepiece Design Forms

Eyepiece design forms can be classified by number of
elements, apparent FOV, and eye relief. Complex designs
with more elements have improved performance, larger
FOVs, and/or longer eye reliefs, but with added cost,
size, and weight (important for handheld or head-mounted
instruments). Modern eyepieces incorporate aspheres
and diffractive elements to reduce size and weight. A key
distinction among eyepiece design forms is the presence of
an accessible image plane, which can be used to overlay a
reticle onto the final image.
Lens distortion changes sign if the optical system
is reversed. This fact is useful when balancing the
distortion between an eyepiece and an objective if the
eyepiece is designed in the opposite direction.
The Huygens eyepiece is a simple two-element design

with a virtual image, while the three-element Kellner
eyepiece has a real image plane. With two identical
doublets, the Plossl design is a low-cost, mediumFOV eyepiece. The orthoscopic eyepiece (known for its
low distortion) is common in astronomical and medical
devices. The Erfle is a standard wide-FOV military
eyepiece with long eye relief but short working distance.
The wide-FOV Nagler design uses a strong negative field
lens group to correct field curvature over a virtual image
plane, but it suffers from strong spherical aberration
of the pupil, leading to a kidney bean effect wherein
portions of the field become alternately light and dark
(vignetted) as an observers eye position changes.
Optical Systems
101
Telescopes
Traditional afocal telescopes magnify distant
objects by enlarging their
apparent angular extent
at the eye. They consist
of a long-focal-length, narrow-FOV objective and a shortfocal-length, wide-FOV eyepiece. The angular magnification of the telescope is defined by the ratio of the objective focal length to the eyepiece focal length. Telescopes
typically have interchangeable eyepieces with different
focal lengths to change magnification. Modern imaging
telescopes remove the eyepiece and use the telescope
objective to image directly onto a sensor. In terrestrial
(spotting) telescopes, an erector is placed between the
objective and eyepiece to form an upright image.
Afocal telescopes and beam expanders are specified by
their magnification (e.g., 4), whereas imaging telescopes
are characterized by their aperture diameter (e.g., 12 in.).
Astronomical telescopes require large apertures to
gather as much light as possible from distant objects.
Astronomical telescopes larger than 12 in. are usually
reflective. In contrast, surveillance spotting scopes
need to be compact and portable and are usually
refractive. A riflescope is a specialized afocal telescope
requiring extended eye relief for recoil. Binoculars are
dual co-boresighted telescopes with upright images.
Beam expanders are afocal telescopes used in reverse
to expand a laser beam diameter and reduce its
divergence. A Keplerian telescope (with a positively
powered eyepiece) has an internal focus, allowing a
possible spatial filter location. A Galilean telescope
(with a negatively powered eyepiece) is shorter and works
well for high-power lasers because there is no internal
focus. A collimator is a reversed telescope objective.
102
Optical Systems
Microscopes
A microscope is an instrument used to magnify and
view objects that are too small for the naked eye.
Applications include biological and medical research,
metallurgy, and industrial inspection. A basic compound
microscope requires an objective and an eyepiece,
where the magnification is the product of the individual
magnifications and is defined relative to the angle the
object subtends at the eye at a distance of 250 mm. The
tube length is the distance from the objectives rear focal
point to the intermediate image.
A modern microscope is designed with an infinitycorrected objective and a tube lens (placed between
the objective and eyepiece). This form allows the insertion
of tilted components (e.g., beamsplitters, filters, and
polarizers) in the collimated space between the objective
and tube lens. The tube lens can also focus the
image directly onto a detector. Microscopes need to be
modular (e.g., interchangeable objectives to adjust system
magnification and resolution). The objective flange-tofocus distance is standardized at either 45 or 65 mm. The
intermediate image diameter ranges from 20 to 28 mm.
Standard tube lengths include 160, 180, 200, and 210 mm.
The design of the illumination system can have a
significant impact on the contrast and resolution of
a microscope. Trans-illumination uses a condenser
to illuminate the sample (light is collected by the
objective on the other side), whereas epi-illumination
uses the objective to illuminate the sample. Dark field
illumination (the opposite of bright field illumination)
is a technique used to enhance contrast where the
condenser has a central obscuration that is larger than the
collecting aperture of the objective lens.
Optical Systems
103
Microscope Objectives
Microscope objectives are small-FOV lenses that
are typically diffraction limited on-axis, telecentric in
object space, and unvignetted. The objective housing is
imprinted with a magnification for a fixed tube lens
focal length and NA (e.g., 20 /0.5). Although there is no
rigid relationship between magnification and NA, if the
magnification is increased, the NA is increased to avoid
empty magnification. Biological microscope objectives
need to correct the spherical and chromatic aberration
introduced by a thin cover slip or microscope slide.
Microscope objectives are typically designed backwards

from the long conjugate to the short conjugate. Low-power
objectives are ordinary achromatic doublets dominated
by field curvature. A medium-power Lister objective
consists of two achromats in a Petzval-like arrangement.
To increase the NA, aplanatic lenses (as in the Amici)
are added to the object end of the lens, decreasing the
working distance. Medium- to high- power objectives are
limited by lateral color, which can be offset by a tube lens
or eyepiece. Inverted telephoto and double-Gauss design
can be turned into microscope objectives with long working
distances. An immersion objective has an extremely
high NA (>1) and is designed to interface with objects in
fluid (index > 1). Objectives are classified by their field
flatness and color correction capability:
PLAN: corrected over a flat field (digital focal planes
require flatter fields than visual instruments).
APO (apochromatic): broad-spectrum color correction.
PLANAPO: corrected for both field curvature and
chromatic aberration for a high level of performance.
104
Optical Systems
Relays
Finite-conjugate relay lenses image a real object to a real
image, typically with a magnification close to one. They
can be grouped into single-stage reproduction lenses
and multi-stage unity-magnification relays, where each
additional stage inverts the image.
Reproduction lenses range from simple document copy/

scan lenses to complex lithography projection lenses.
Their magnification varies between 1:1 and 20:1 (either
reducing or enlarging) and may be adjustable. Since
the applications for these lenses require a high level
of distortion correction, design forms with inherent stop
symmetry (e.g. the double-Gauss) are frequently used.
F-theta lenses are designed to have deliberate barrel
distortion where the image height follows an h0 = f
rather than the standard h0 = f tan relationship. F -
lenses are useful in scanning systems where the f -
distortion creates a scanned spot with a constant velocity
across the field, producing a uniform exposure.
Unity-magnification relays are found in periscopes,
endoscopes, and boroscopes, where light must be
guided down a long tube. Minimizing the lens diameter in
such long-distance relays is critical; this requires multiple
stages, each with a primary imaging lens (usually a
doublet or triplet for color correction). Field lenses are
placed near the internal image planes of the relay to
converge the oblique beams so that they pass through a
smaller clear aperture. Due to their proximity to an image
plane, field lenses must have excellent surface quality and
very few defects. The relay is typically only one part of the
optical system and is coupled to a high-resolution objective
on one end and an eyepiece or camera lens for viewing the
image on the other end.
105
Index of Refraction and Dispersion

The index of refraction (or index) of a material is the
ratio of the speed of light in vacuum to the speed of light
inside the material (typical values range from 1.45 to 4.0).
The change in index as a function of wavelength is called
material dispersion. Normal
dispersion occurs when index decreases with wavelength.
Anomalous dispersion occurs near material absorption
peaks where index increases
with wavelength. The dispersion of a material over a
given spectral band is quantified by an Abbe number or
V-number (V), which is calculated with the index at three
wavelengthsthe two points
near the ends of the spectral
band and a midpointwhich in the visible corresponds to
the F, d, and C spectral lines. A high V corresponds to low
dispersion, and a low V indicates high dispersion.
Custom Abbe numbers can be defined over any band of
interest when designing outside of the visible spectrum
(e.g., the g line is used instead of the F line when
correction at shorter wavelengths is needed). The
dispersion of a material can be vastly different in two
different spectral bands (e.g., visible and MWIR).
Sellmeier
s
Two common dispersion models are the

c1
c2
c3
+ 2
+ 2
n () = 1 + 2
Sellmeier
and
c4 c5 c6
Schott index equaSchottr
tions. Discrete meac 3 c 4 c 5 c 6 sured data can be fit
2
n ( ) = c 1 + c 2 + 2 + 4 + 6 + 8
to one of these equations for use in optical design. However, using dispersion
equations outside of their intended wavelength region can
lead to significant extrapolation errors.
2
106
Optical Materials: Glasses

Common optical materials used in the
visible spectrum include glasses, crystals, and plastics.
Of the three, glass
has the most extensive historical industrial development. The
primary optical properties of a glass are illustrated on a glass map, which
plots the index of refraction of the glass versus its dispersion (typically plotted in decreasing V ). Each dot on the
glass map represents a different material. Low-dispersion
crown glasses have a V > 55, and high-dispersion flint
glasses have a V < 55. Glass makers further divide the
glass map into regions based on proprietary chemistry.
Historically, lead was used to increase the index of glass
during production. The increased lead content also increased the dispersion so that the main sequence of legacy
glasses lie on a lead line. Glasses far from the lead line
tend to be more expensive.
The manufacturer is the ultimate authority on an optical
material. Designs with extreme material sensitivity
should confirm the optical properties with an actual
measurement or melt certification.
Major glass manufacturers now produce environmentally
friendly, lead-free glasses and have preferred glasses
that are in high production and have lower cost than
standard glasses. Other important glass properties not
shown on a glass map include transmission, partial
dispersion, coefficient of thermal expansion, thermo-optic
coefficient (dn/dT), birefringence, and glass transition
temperature.
107
Optical Materials: Polymers/Plastics

Plastics or optical polymers are common in highvolume lens systems. Injection molding is the
preferred method for fabricating low-cost lenses, but
polymers can also be
diamond-turned or conventionally fabricated. Common materials include acrylic
(polymethlymethacrylate/PMMA), polycarbonate (which
can be used as a shatter-resistant material), polystyrene,
and cyclic-olefin copolymer (COC/Zeonex). Plastics have
material properties that are significantly different from
optical glasses or UV/IR crystals.
Advantages
Inexpensive, lightweight material allows extremely
high-volume and repeatable production.
Injection molding allows precision optomechanical
mounting features to be integrated with optical surfaces.
Absorbing dopants or dyes can be included in the optical
material, allowing integrated spectral filtering.
Optical designs can easily incorporate diffractive and
highly aspheric (with inflection points) surfaces.
Many glasses (e.g., Schott B270) can now be injectionmolded and are cost-effective complements to polymers.
Disadvantages
Material selection is limited compared to optical glass.
Injection molding requires large tooling costs.
Very low heat tolerance (<100200 C maximum).
Spectral bandpass is generally only 0.41.6 m.
High thermal sensitivity: 10 larger CTE and 50 larger
dn/dT than optical glass.
Optical coatings can be difficult to apply.
Very large stress birefringence.
Sensitivity to abrasion and chemical exposure.
108
Optical Materials: Ultraviolet and Infrared

The selection of high-quality ultraviolet (UV) materials
for UV applications (e.g., lithography, semiconductor
inspection, broadband spectrometers, and fluorescence
microscopy) is very limited. Most optical materials do not
have good transmission below 400 nm and have increased
bulk scattering. Fused silica (SiO2 ) and calcium fluoride
(CaF2 ) are the two most-common UV materials. Other
materials with high UV transmission (e.g., crystal quartz,
sapphire, lithium fluoride, barium fluoride, and potassium
bromide) have undesirable qualities for lens fabrication
(soft/hard, birefringent, or hygroscopic). Materials with
low visible dispersion are more dispersive in the UV. Many
broadband deep-UV systems are reflective due to the lack
of materials for color correction.
Most visible glasses are opaque (have a long-wave
transmission cutoff) above 2.7 m. Most IR materials are
opaque (have a short-wave cutoff) below 1 m.
Infrared (IR) materials are less limited in selection than
UV materials. Many IR materials are crystalline, and
a large fraction are semiconductors. The two main IR
spectral regions are the mid-wave (MWIR) band from
35 m and the long-wave (LWIR) band from 812 m.
Common IR materials are silicon, germanium, CaF2 ,
zinc selenide, and zinc sulfide. AMTIR, gallium arsenide,
magnesium fluoride, and barium fluoride are also used. IR
materials generally have much higher indices (as large
as 4) than visible glasses. Dispersion can be very low,
with V numbers as high as 1000 in the LWIR. The
longer wavelengths reduce surface scatter and roughness
requirements, making IR materials attractive for costeffective aspheres. However, the material itself can be very
brittle or soft.
IR materials have much higher thermal expansion
and thermo-optic coefficients than visible glasses,
and athermalization is required to maintain the lens
performance over large temperature changes.
109
Snells Law and Ray Tracing

Snells law (of refraction) is the
primary equation used in ray tracing.
It governs the change in direction of a
ray at a dielectric interface between two
materials. Snells law relates two angles
in one plane of incidence; a full 3D ray
trace typically uses direction cosines, the rays unit
vector projection onto each coordinate axis. Because of
the sine functions in Snells law, ray tracing calculations
through multiple surfaces become highly nonlinear for
large angles. Snells law of reflection can also be applied to
a reflecting surface if the sign of the index of the incident
material is reversed after the reflection.
Once the ray is refracted or reflected at a surface,
it must be translated (or propagated) to the next
surface, changing its height (or location in 3D) without
changing its angle. Paraxial ray tracing assumes that
for small angles, sin() is approximately equal to ; this
significantly simplifies ray tracing equations [replacing
the n sin() with n in Snells
law]. The resulting paraxial
refraction and a paraxial
transfer equation allow visual brick (or ynu) diagrams
or matrix math to be set up
quickly to calculate the firstorder properties of an optical
system.
Ray failures generally occur
when either a ray fails to intercept
the next sequential surface or
the ray undergoes total internal
reflection (TIR) at a surface. Optical design codes have significant
difficulty evaluating a lens system
if reference rays (e.g., the chief
ray) fail to trace.
110
Focal Length, Power, and Magnification

Focal length is a fundamental first-order property of a lens. For a thin
lens, given the object distance and focal length,
the Lens makers equation can be used to
quickly find the image
distance. The focal length
also determines the magnification (relative size) between the object and image for
a given object position. The power of a lens is the reciprocal of its focal length. For a thin lens, the power is related
solely to its two curvatures and its index. For a thick lens,
an effective focal length (EFL) is defined that equals
the distance between its focal and principal planes in
either object or image space.
The powered surface of
a plano-convex singlet
made from glass (n 1.5)
has a convex radius approximately equal to half
its focal length. For the
best performance, the surface should be oriented with
the convex side facing the
collimated light or laser
beam.
Objects and images form conjugate
planes. A finite conjugate system has
both a finite object and a finite image
distance. An infinite conjugate system
has either the object or the image located
at infinity. In an afocal system, both the
object and image are located at infinity.
The performance of a lens designed for a
specific conjugate usually degrades when
used at other conjugates.
111
Aperture Stop and Field Stop

The aperture stop (often
shortened to just stop) or
stop surface is the single
aperture in the lens system
that limits the bundle of light
that propagates through the system from the axial object
point. In the absence of vignetting, the ray bundles from
other object points also fill the aperture stop (pivoting
about the center of the stop). The aperture stop is typically
a mechanical surface (for good edge definition) and can
be located anywhere in the optical system except at the
object, the image, or an internal image. The size of the
aperture stop sets the system f /# and NA. For objects at
infinity, the aperture stop limits the input beam diameter.
A ray fan is a slice of
rays that emerge from a
single field point and pass
through a line in the pupil.
The tangential fan (Y-fan
or meridional fan) passes
through the pupils y axis,
and the sagittal fan (or
X-fan) passes through the
pupils x axis.
Field stops also limit
the amount of light that
can propagate through the
system from the object
plane to the image plane by setting the maximum object
height from which a ray bundle can emerge and still pass
through the lens. Field stops are typically the image (or
the object) surface; however, if the system has an internal
image plane, a field stop may be located there (usually for
stray light control), and because this plane is conjugate
with the image, reticles, crosshairs, or linear scales are
placed at the internal field stop to be simultaneously
imaged on the focal plane.
112
Entrance and Exit Pupils

Wavefront performance for imaging lenses is often
measured at the exit pupil. The exit pupil and the
image plane form a Fourier transform pair under
Fresnel propagation. Optical design codes calculate
point spread functions (PSFs) and other diffractionbased metrics using a fast-Fourier transform (FFT) of
the exit pupil wavefront.
The physical stop surface is often buried in
a lens system and difficult to access, leading to the definition
of pupils. The images
of the aperture stop
formed by the lens elements before and after the stop surface are
called the entrance
and exit pupils, respectively. If the stop is
at the first (last) surface, it is also the entrance (exit) pupil. The entrance (exit)
pupil appears to limit the incoming (outgoing) bundle of
rays in object (image) space. Pupil matching is a critical consideration when combining two independently designed optical systems. The entrance pupil of the second
system should have the same position and diameter as the
exit pupil of the first system to avoid losing light and reducing the FOV through vignetting. The entrance and exit
pupils are images of each other forming conjugate planes,
and, consequently, they can have pupil aberrations.
Similar to entrance and exit pupils, internal field stops can
be imaged by the lens elements on either side of the field
stop, forming entrance and exit windows. However, not
all lenses have internal images and, therefore, entrance
and exit windows.
113
Marginal and Chief Rays

Given the surface curvatures, material indices, and
element thicknesses, the marginal and chief rays are
the only two reference rays that need to be traced
to calculate all first-order properties and third-order
aberrations of a lens system.
The marginal ray starts at the base of the object (the
on-axis, H = 0, field point) and passes through the edge
of the aperture stop. The marginal ray locates all image
positions, including any intermediate images; it also
defines the edge of the ray bundle, and, therefore, the
maximum sizes of the
entrance and exit pupils.
The chief ray originates from the edge
of the object (the fullfield, H = 1, field point)
and passes through the
center of the aperture
stop. The chief ray defines the maximum FOV
of the lens system.
Pupils exist wherever the chief ray crosses the optical axis
(the entrance and exit pupil are typically virtual; the chief
ray needs to be extended back until it crosses the optical
axis). The image size is determined by the height of the
chief ray at the image plane.
The chief ray intercept on the focal plane may not define
the center of the energy distribution for a given field
point. In this case, the spot centroid or center of mass
is a better metric for determining true field position at an
image plane. Optical design programs allow a designer
to reference various performance metrics (e.g., RMS spot
diameter, distortion) from either the chief ray location or
the centroid location.
114
Zernike Polynomials
Zernike polynomials are a set of circularly symmetric

orthogonal basis functions defined over a unit circle.
They are 2D functions of both radial and azimuthal
coordinates. In optical design, Zernikes are used to
describe either surface irregularity or system wavefront
(measured in the pupil). Fringe Zernikes are defined
such that each polynomial has a value of unity at the
edge of the unit circle; typically, only a limited 37-term
subset is used, starting at Z1 . Standard Zernikes are
numbered differently (starting with Z0 ) and are used
when many more coefficients are needed (e.g., simulations
of turbulence or time-varying statistical phase surfaces).
However, either type is capable of representing the same
deformation. Standard Zernikes are normalized such that
each polynomial term has an RMS value of 1 over the
unit circle, allowing coefficients to add in quadrature to
determine the overall RMS value. Both types of Zernikes
can be renormalized so that they are orthogonal over an
annular or otherwise noncircular region.
In tolerancing, Zernikes are useful for modeling complex
surface irregularity (beyond cylinder and spherical)
from polishing residuals or mounting effects.
When used to model wavefront error
in the pupil, Zernikes appear identical
to the traditional Seidel aberrations.
However, Zernikes only represent the
pupil wavefront at one field and are
not true expansions of the point characteristic aberration function. Special
Zernike expansions that are functions of both pupil and
field have been developed for lithography applications.
115
Conic Sections
A surface of rotation formed from a conic section is
a special type of asphere. Conic surfaces are defined
using the sag equation and parameterized by the conic
constant . Spheres, paraboloids, hyperboloids, and
ellipsoids are all surfaces formed from conic sections.
Conic surfaces have stigmatic imaging in which one

point is perfectly focused to another point. In a sphere,
the two focal points coincide. In a parabola, one point is at
infinity. Hyperboloids have one real and one virtual focus,
whereas ellipsoids have two real foci. The deterministic
correspondence of foci makes testing conic surfaces easier
than testing a general aspheric surface.
Prolate or pointy ellipsoids (football-shaped) are
ellipses rotated about their major axis and have a negative
. Oblate or squashed ellipsoids (pumpkin-shaped) are
ellipses rotated about their minor axis with a positive .
Oblate ellipsoids are more difficult to test.
Conic surfaces have closedform analytic solutions for all
third-order aberrations. Conic
surfaces used off-axis are limited by coma and, when used
at nonideal conjugates, suffer
mainly from spherical aberration. Conic surfaces are very common in multiple-mirror
reflective telescopes.
116
Diffraction Gratings
A diffraction grating splits incident light into different angles
based on wavelength and grating
spacing d . For a given incident angle, the exiting angle for order m
can be calculated with the grating equation. Because neither incident nor output angles can be greater than 90 deg, the
highest possible grating order is d /. Gratings can be
transmissive or reflective. A Littrow grating is a retroreflective grating with an output angle equal to the incident
angle for one particular diffraction order and wavelength.
Tooling marks on diamond-turned surfaces can be
modeled as gratings for stray light analysis.
Linear gratings can be manufactured by etching or
scribing grooves on a glass/metal surface, photo-reducing
a regular pattern of lines on a glass plate, or by stamping
a master form into a substrate. Curved gratings can also
act as powered optical elements in complex optical systems
but are harder to fabricate and are usually reflective. A
holographic grating is formed by interfering two optical
beams in a photosensitive material.
Light in nondesired diffracted orders can cause significant
stray light. Diffraction efficiency is the ratio of power
in a given order to the total incident power. It is primarily
a function of grating order and grating tooth shape
(e.g., continuous or stepped). Blazed gratings have a
high diffraction efficiency in one order at the design or
blaze wavelength. Grating dispersion is the derivative
of output angle with respect to wavelength and is another
measure of grating efficiency. Increasing the order m will
increase dispersion but reduce the diffraction efficiency.
Free spectral range is the
total range of wavelengths in
a given order that do not
angularly overlap those of
another order.
117
Optical Cements and Coatings

Achromatic doublets can be airspaced or cemented
(where an optical cement is used to bond the achromat
at the common internal surface). Airspaced doublets
have significantly increased sensitivity to manufacturing
tolerances, so cementing is the preferred option if a
transparent cement exists for the wavelength range
(e.g., cements that work well with low absorption in
the UV and IR are difficult to find). Canada balsam
(n = 1.53 to 1.55) and UV-cured synthetic adhesives are
common in the visible and work over large temperature
ranges. The cement layer must accommodate element
CTE mismatches over the full temperature range that
the lens is expected to see. A typical thickness for the
cement layer is 1020 m and is usually not included in
the optical design; however, very high-performing lenses
may need to be optimized with the cement layer in place.
Cemented surfaces do not need AR coatings unless the
index difference between the glasses is large.
High-reflection (HR) coatings are needed on reflective
optics to improve the base metal reflectivity (e.g., an
enhanced dielectric coating can boost the reflectance
of an aluminum mirror from 90% to over 97%).
Dichroics are short- or long-wave-pass filters used
for color balancing or color splitting. Cold and hot
mirrors are special dichroics that transmit or reflect,
respectively, thermal (long-wave) radiation to protect
thermally sensitive optics. Cube beamsplitters require AR
coatings on the entrance and exit faces
as well as an internal coating to divide
the beam. Polarizing beamsplitters
split unpolarized incident light into two
orthogonally polarized beams.
Reflective dielectric coatings generally have very low
absorption, whereas metal coatings have more absorption.
Some reflectors use all-dielectric mirrors to entirely
avoid the absorption from the metal surfaces. Such
mirrors may also provide spectral selectivity.
118
Detectors: Sampling
Modern lenses are used with periodically sampled
digital staring arrays. The Shannon sampling theorem
requires a minimum of two samples per cycle to perfectly
reconstruct a sampled signal. The Nyquist limit defines
the maximum frequency at which features can be sampled
and corresponds to one full whitedark cycle spanning two
pixels. Therefore, the highest spatial frequency that can
be detected equals the reciprocal of twice the pixel pitch
p. Real systems reliably resolve about 70% of the Nyquist
frequency (roughly 2.8 pixels/cycle). Image content above
the sampling frequency (1/ p)
will alias and appear as lowerfrequency content. An approximate sinc-function sampling
MTF is often used for system
modeling and assumes spatially averaged random-phase
frequency content.
Not all detectors have rectangular geometry. Fiber optic
faceplates have hexagonal spacing, leading to a sampling
MTF with non-Cartesian symmetry.
For the best-performing optical
system, the lens and the detector should be matched to each
other. A common design trade
is to optimize the ratio of the
diffraction-limited optical blur
spot to the sampling size. This
ratio is referred to as the Q parameter, which is a comparison of the Nyquist frequency with the lens cutoff
frequency. A Q value of 1 corresponds roughly to 2 pixels per Airy disc diameter. Q values range from 12 for
most sensors; higher values of Q correspond to lens-limited
(oversampled) imaging, and Q values under 1 correspond
to pixel-limited (undersampled) imaging.
119
Detectors: Resolution
Limiting resolution (LR) is the spatial (or angular)
frequency at which an MTF degrades to a desired contrast
level. The lens is only one part of a complex optical
system; besides the diffraction limit and lens aberrations,
the detector is often the most significant contributor
to the system MTF and limiting resolution. In CCDs
for high-performance systems, a charge-diffusion MTF
represents a blurring function caused by photoelectrons
diffusing into neighboring pixels. Linear systems theory
allows multiplication of cascaded component MTFs to
obtain the final system MTF.
Limiting resolution is often measured subjectively using
a square-wave bar target in units of line-pairs/mm
(lp/mm). LR objectively derived from an MTF curve has
units of cy/mm. Although square-waves and sinusoids
are related by an infinite Fourier series, cycles/mm and
lp/mm should not be used interchangeably.
Both periodically sampled detectors (CCD/CMOS arrays)
and film have MTF curves with finite resolution. Classic
film LR (based on chemistry and grain size) can range from
80 cycles/mm for panchromatic film to over 2000 cy/mm
for scientific-grade spectroscopic plates. LR for a digital
array is typically defined by the Nyquist frequency,
equal to the reciprocal of twice the pixel pitch. The Nyquist
frequency for a 3-m pixel is 167 cycles/mm, comparable to
high-performance photographic film.
Limiting
resolution
is
insufficient to determine
overall
system
image
quality because two very
different system MTFs
can have the same LR. A
complete MTF curve is a
much better performance
predictor.
120
The Lagrange Invariant and tendue

In an optical system, the Lagrange invariant (also
called the optical invariant) is a constant that can be
calculated at any surface within the lens. At the image
surface, it equals the product of the image space index,
the image height, and the marginal ray (convergence)
angle. At the object surface, it equals the product of the
object space index, the object height, and the marginal
ray (divergence) angle. At other surfaces, it is given by a
combination of marginal and chief ray angles and heights:
H = n u y nu y
n 2 A = 2 H 2 .
For a fixed image size and f /#, the object size cannot be
reduced (increased) without illuminating the object over
a larger (smaller) NA.
tendue (or throughput) is the 3D optical invariant
and equals the product of the image (object) area times
the solid angle of collection (illumination). tendue is
proportional to the square of the Lagrange invariant; it is
a geometric measure of the total brightness of an optical
system. The conservation of radiance states that the tendue
of an optical bundle can only
remain constant (or increase)
through an optical systemit
cannot decrease and still obey
energy conservation. tendue is purely geometric and has
no inherent relation to imaging performance, representing
only the boundaries of the light bundle and not the actual
distribution. An aberrated blur spot will increase tendue
because the beam area is larger for the same solid angle.
For a flat surface with a uniform solid angle, the tendue
equals A sin2 (), where A is the area of the surface, and
is the half angle of the marginal ray.
121
Illumination Design
Many imaging systems also require optimized
illuminators (e.g., projectors, microscopes, and machine
vision systems). The primary goal of illumination design
is to create a desired spatial and/or angular distribution of
energy at a given surface. Illumination systems are also
found in automobile headlights, spotlights, commercial
lighting,
and
displays.
Collection or concentrator
systems maximize radiation
transfer between a bright
source and a target surface,
such as a simple parabolic
or elliptical concentrator.
Creating a uniform distribution from a complex source
requires homogenization, which can be done by
tailoring a known source distribution or by using
superposition of many source images. Automotive
headlights and streetlights use wedged parabolas as
tailored homogenizers. Light pipes and flys eye arrays
are examples of superposition homogenizers.
Critical (or Abbe) illumination images the source
directly onto the image, increasing brightness but
potentially adding unwanted image structure (e.g., the
lamp filament). In contrast, Khler illumination images
the source into the imaging system pupil for a more
uniform illumination but requires more lenses.
In a microscope, bright field illumination fills the imaging pupil either through the objective (epi-illumination)
or from underneath the sample (trans-illumination).
Dark field illumination puts no light in the imaging
pupilthe imaging system collects scattered light, highlighting edges. Lithography illuminators fill the pupil with
complex ring or dipole illumination. The illumination coherence can play a critical role in high-performance systems by shifting the performance limits (partially coherent
light has a slightly different diffraction-limited spot size
than fully coherent light).
122
Equation Summary
Numerical aperture and f /#:
N A = n0 sin u0
f /# =
f
EPD
Rayleigh criterion:
X =
0.61
NA
Image height as a function of field angle:

h0 = f tan
Transverse ray error:

0y =
1 W
n0 u0a y
0x =
1 W
n0 u0a x
Strehl ratio:
2 2
Strehl
= 1 2
= RMSOPD
Wavefront aberration polynomial:

WI JK H I J cosK
W ( H, , ) = W020 2 + W111 H cos + W040 4 + W131 H 3 cos
+W222 H 2 2 cos2 + W220 H 2 2 + W311 H 3 cos + O (6)
Contrast:
Contrast =
I max I min
I max + I min
123
Equation Summary
Focal lengths of any two thin lens system:
fa =
df
f BFL
d BFL
f BFL d
fb =
Zero-Petzval solution for two thin lenses:
f a = f b = f BFL
d=
( f BFL)2
f
Two-mirror solution:
c1 =
BFL f
2d f
c2 =
BFL + d f
2 d BFL
Schwarzchild solution:
d = 2f
c1 =
51 f
c2 =
5+1 f
Aplanatic condition:
i = u0
Bending and shape factors:
c1 + c2
c1 c2
C=
u a + u0a
u a u0a
Thin lens bending for minimum spherical:

c 2 2 n2 n 4
=
c1
n (2 n + 1)
124
Equation Summary
Thin lens bending for minimum coma:
c 2 ( n2 n 1)
=
c1
n2
Achromatic doublet:
= 1 + 2
1 =
V1
V1 V2
2 =
V2
V1 V2
Petzval sum:
X j
j
nj
Thick lens power:

= 1 + 2
t
(1 2 )
n
Minimum clear aperture for no vignetting:

CAmin = | ya | + | yb |
Aspheric sag equation:
c r2
sag = z ( r ) =
1+
1 ( + 1)( c r)
+ d r 4 + e r 6 + f r 8 + g r 10 + . . .
Gradient index profiles:

n ( r ) = N00 + N10 r 2 + N20 r 4 + . . .
n ( z) = N00 + N01 z + N02 z2 + . . .
125
Equation Summary
Merit function:
m
i= 1
w2i ( c i t i )2
Athermalization condition:
d f Lens
= CTE 1 d 1
dT
CTE 2 d 2
Bireectance scattering distribution function:
L
BSDF ( i , i ; o , o ) = sr 1
E
Sellmeier dispersion:
1+
n () =
c 1 2
c4
c 2 2
c5
c 3 2
2 c 6
Schott dispersion:
n () =
c 1 + c 2 2 +
c3
c4
c5
c6
Snells law:
n sin = n sin
Paraxial ray tracing:
n u = nu y
d
y = y + n u
n
= c (n n )
126
Equation Summary
Lens makers equation and linear magnification:
1
s0
1
f
m=
h0
f
=
h
s+ f
Thin lens power:

=
1
f
= ( c 1 c 2 )( n 1)
Diffraction gratings:
blaze = 2d sin
m = d [sin(m ) sin( i )]
Sampling ratio:
Q=
( f /#)
pixel pitch
Lagrange invariant and tendue:

n2 A = 2 H 2
H = n u y n u y
Etendue =
sur f ace
d A a
127
Bibliography
G. Boreman, Modulation Transfer Function in Optical
and Electro-Optical Systems, SPIE Press, Bellingham, WA
(2001) [doi:10.1117/3.419857].
B. Braunecker, R. Hentschel, and H. J. Tiziani, Eds.,
Advanced Optics Using Aspherical Surfaces, SPIE Press,
Bellingham, WA (2008) [doi:10.1117/3.741689].
R. Fischer, B. Tadic-Galeb, and P. Yoder, Optical System
Design, 2nd ed., McGraw-Hill, New York (2008).
G. W. Forbes, Shape specification for axially symmetric
optical surfaces, Opt. Express 15(8), 52185226 (2007).
J. M. Geary, Introduction to Lens Design, Willmann-Bell,
Richmond, VA (2002).
J. E. Grievenkamp, Field Guide to Geometrical Optics, SPIE Press, Bellingham, WA (2003)
[doi:10.1117/3.547461].
H. Gross, Ed., Handbook of Optical Systems, Vol. 14, John
Wiley and Sons, New York (2005).
H. Karow, Fabrication Methods for Precision Optics, John
Wiley and Sons, New York (1993).
M. J. Kidger, Fundamental Optical Design, SPIE Press,
M. J. Kidger, Intermediate Optical Design, SPIE Press,
R. Kingslake, Optical System Design, Academic Press,
New York (1983).
R. Kingslake and B. Johnson, Lens Design Fundamentals,
2nd ed., Academic Press, New York (2010).
M. Laiken, Lens Design, Marcel Dekker, Inc., New York
(1995).
D. Malacara, Optical Shop Testing, 3rd ed., John Wiley
and Sons, New York (2007).
128
Bibliography
D. OShea, Elements of Modern Optical Design, John Wiley
and Sons, New York (1985).
R. R. Shannon, The Art and Science of Optical Design,
Cambridge University Press, Cambridge, UK (1997).
G. H. Smith, Camera Lenses: From Box Camera to Digital,
SPIE Press, Bellingham, WA (2006).
W. Smith, Modern Lens Design, 2nd ed., McGraw-Hill,
New York (2005).
W. Smith, Modern Optical Engineering, 3rd ed., McGrawHill, New York (1996).
B. H. Walker, Optical Design for Visual Systems, SPIE
Press, Bellingham, WA (2000) [doi:10.1117/3.391324].
W. T. Welford, Aberrations of Optical Systems, Taylor and
Francis, New York (1986).
C. S. Williams and O. Becklund, Introduction to the Optical
Transfer Function, SPIE Press, Bellingham, WA (2002).
W. L. Wolfe, Introduction to Infrared System Design, SPIE
Press, Bellingham, WA (2000) [doi:10.1117/3.226006].
P. Yoder, Opto-Mechanical Systems Design, 3rd ed., Taylor
and Francis, New York (2006).
129
Index
3-point symmetry, 77
4 f optical system, 58
Abbe illumination, 121
Abbe number, 105
aberrations, 7
absorption, 94
achromat, 30
achromatic doublet, 30, 55
actively athermalized, 88
adaptive simulated
annealing, 67
adjustment range, 82
afocal relay telescope, 58
afocal system, 42, 110
afocal telescope, 101
airspace, 69
airspace compensator, 82
airspace tolerance, 81
airspaced doublet, 31
Airy disc, 5
alias, 118
all-dielectric mirror, 117
altitude, 6, 87
Amici lens, 103
anamorphism, 45
anastigmat, 32
angle of incidence (AOI),
1, 50
angular magnification,
101
anisotropic surface, 95
anomalous dispersion,
105
anomalous partial
dispersion, 55
anti-reflection (AR)
coating, 91
aperture, 27
aperture specification, 4
aperture stop, 90, 111
aplanatic lens, 50, 103
aplanatic solve, 70
apochromats, 30
area obscuration, 38
arm, 72
aspect ratio, 69
asphere, 61, 100, 115
aspheric coefficient, 69
aspheric collimator, 29
aspheric departure, 62
aspheric sag equation, 61
aspheric testing, 62
assembly drawing, 75
assembly plan, 81
assembly print, 75
assembly tolerance, 81
astigmatism, 21
astronomical telescope,
101
asymmetrical tolerance,
73, 82, 85
athermalization, 108
athermalized, 88
axial color, 23
axial gradient, 63
back focal length (BFL), 6
baffle, 90
balsam, 117
bandwidth, 4
barrel distortion, 22
base radius, 62
beam expander, 101
best-fit sphere (BFS), 62
bidirectional reflectance
distribution function
(BRDF), 95
130
Index
bidirectional scattering
(BSDF), 95
bidirectional
transmittance
(BTDF), 95
binary optic, 64
binoculars, 101
biogon, 33
Biotar, 33
birefringence, 79
blaze wavelength, 116
blazed grating, 116
blue shift, 91
bolt-together assembly, 80
boresight error, 81
boroscope, 104
Bouwers telescope, 41
brick diagram, 109
bright field illumination,
102, 121
brightness, 120
broadband anti-reflection
(BBAR) coating, 91
bubbles, 79
Buchroeder system, 41
cam, 48
camera lens, 96
camera objective, 96
Cassegrain telescope, 39
catadioptric, 37
catadioptric design, 56
catadioptric telescope
objectives, 41
cemented doublet, 30
center thickness (CT), 78
centering, 78
central obscuration, 38
centroid, 9, 68, 113
centroid distortion, 22
charge-coupled device
(CCD), 96
charge-diffusion
modulation transfer
function (MTF), 119
chief ray, 113
chief ray angle solve, 70
child ray, 94
chromatic aberration, 23
clocking, 77
coefficient of thermal
expansion (CTE), 87
coherence, 121
cold mirror, 117
cold shield, 93
cold shield efficiency, 93
cold stop, 93
collection system, 121
collimator, 101
coma, 18
compensated, 84
compensating eyepiece, 99
compensator, 17, 48, 73,
82
complementary
metal-oxide
semiconductor
(CMOS) detector, 96
compound microscope,
102
compounding, 32, 51
computer-generated
hologram (CGH), 64
concentrator system, 121
condenser, 102
conic, 61
131
Index
conic constant, 61
conic section, 115
conjugate factor, 53
conjugate plane, 110
constraint, 65
construction parameter,
73
continuous variable, 69
continuous zoom, 47
contrast transfer function
(CTF), 13
Cooke triplet, 32
corrector lens, 39
cosmetic defect, 79
cost function, 65
cover slip, 103
critical illumination, 121
critical object, 90
cross term, 85
crown glass, 106
crystal, 106
crystalline, 108
cumulative distribution
function (CDF), 86
curved grating, 116
custom distribution, 83
cutoff frequency, 118
cycles/mm, 13
cylinder, 77
damped least squares
(DLS), 66
damping factor, 66
dark field illumination,
102, 121
decenter, 81
default merit function, 68
defect, 65
defocus, 15
depth of field, 16
depth of focus, 16
derivative increments, 66
derivative matrix, 66
derivative tolerancing, 85
design margin, 73, 74
dewar, 93
dialyte, 31
dichroic, 117
diffraction efficiency, 64,
116
diffraction grating, 116
diffraction limited, 12, 14
diffraction-limited lens,
52
diffractive, 100
diffractive optical element
(DOE), 64
digital camera, 96
digs, 79
diopter, 98
directed distance, 1
direction cosine, 109
discrete variable, 69
discrete zoom, 47
dispersion, 23, 55, 79,
105, 116
distance-measuring
interferometer, 76
distortion, 22
double Gauss, 33
doubly telecentric, 58
drop-in assembly, 80
edge spread function
(ESF), 12
edge thickness difference
(ETD), 78
edging, 78
132
Index
effective focal length
(EFL), 110
ellipsoid, 115
elliptical coma, 25
empty magnification, 98,
103
encircled energy, 9
endoscope, 104
enhanced dielectric
coating, 117
ensquared energy, 9
entrance pupil, 112
entrance pupil diameter
(EPD), 4
entrance window, 112
envelope, 6
environmental analysis,
3, 87
environmental
requirement, 87
epi-illumination, 102, 121
Erfle eyepiece, 100
tendue, 120
exit pupil, 112
exit window, 112
expansion, 45
eye relief, 99
eyepiece, 98, 99
eyepiece design form, 100
fast-Fourier transform
(FFT), 112
fictitious glass, 69
field clearance, 43
field curvature, 19
field curve, 21
field flattener, 19
field lens, 34, 41, 104
field of view (FOV), 5
field point, 71
field size, 27
field stop, 90, 111
field tilt, 81
field weight, 71
fifth-order spherical
aberration, 25
finite conjugate system,
110
first-order optics, 7
first-order solution, 3
first-order stray light
path, 90
fish-eye lens, 36, 96
flange-to-focus, 6, 102
flint glass, 106
floating element, 35, 96
floating stop, 57
flys eye array, 121
focal length, 110
focal plane, 110
focus, 15, 82
focusing, 48
Fourier filtering, 58
fovea, 98
foveal region, 98
Fraunhofer doublet, 30,
31
free spectral range, 116
freeform surface, 61
Fresnel reflection, 91
fringe Zernike
polynomials, 114
full field of view (FFOV), 5
full-pitch, 63
f -theta lens, 104
f - distortion, 97
f /#, 4
f /# solve, 70
133
Index
G-sums, 53
Galilean telescope, 101
Gauss doublet, 31
Gaussian bracket, 48
Gaussian distribution, 83
Gaussian quadrature
(GQ) sampling, 72
genetic algorithm, 67
geometrical optics, 7
ghost, 92
ghost analysis, 92
ghost image, 89, 92
ghost reflection, 41
glare stop, 90
glass, 106
glass map, 69, 106
global optimization, 65, 67
global search, 67
gradient index (GRIN), 63
gradient matrix, 66
grating equation, 116
grating spacing, 116
Gregorian telescope, 40
gull wing, 61
Hale Telescope, 40
half field of view (HFOV),
5
hexapolar sampling, 72
high-reflector (HR)
coating, 117
higher-order (HO)
aberration, 25
hologon, 33
holographic grating, 116
holographic optical
element (HOE), 64
homogenization, 121
Hopkins ratio, 14
hot mirror, 117

Houghton system, 41
Hubble Space Telescope
(HST), 40
human eye, 98
humidity, 6
Huygens eyepiece, 100
hyperboloid, 115
hyperfocal distance, 15
hypergon, 33
illuminated object, 90
illumination, 94
illumination design, 121
illuminator, 121
image clearance, 6
image format, 96
image height, 71
image inverter, 98
image jitter, 87
image plane tilt, 82
image-space telecentric,
58
imaging telescope, 101
immersion objective, 103
inclusions, 79
index of refraction, 55, 79,
105
induced aberration, 26
infinite conjugate system,
110
infinity-corrected
objective, 102
infrared (IR), 108
inhomogeneity, 79
injection molding, 107
interface requirement, 6
intrinsic aberration, 26
134
Index
inverse sensitivity
analysis, 84
inward-curving field, 20
ISO 10110, 75
isotropic surface, 95
Kellner eyepiece, 100
Keplerian telescope, 101
keystone distortion, 22
kidney-bean effect, 60,
100
kinoforms, 64
knife edge, 90
Khler illumination, 121
Lagrange invariant, 120
Lagrange multiplier, 68
Lambertian, 95
landscape lens, 29
Large Binocular Telescope
(LBT), 40
lateral color, 24
lead line, 106
legacy design, 49
lens, 2
lens bending, 54
lens bending parameter,
53
lens design, 2
Lens makers equation,
110
lens system, 2
lens thickness, 69
light pipe, 121
limiting resolution (LR),
119
line spread function
(LSF), 12
line-of-sight (LOS), 87
linear grating, 116

linear obscuration, 38
linear system theory, 14
Lister objective, 103
Littrow grating, 116
local optimization, 65, 66
long-focus lens, 96
long-wave infrared
(LWIR), 108
longitudinal chromatic
aberration, 23
longitudinal color, 23
Lyot stop, 90
macro lens, 96
magnification, 110
magnification solve, 70
Maksutov telescope, 41
MaksutovCassegrain
telescope, 41
Mangin mirror, 41
manufacturing yield, 86
Marchal criterion, 12
marginal ray, 113
marginal ray angle solve,
70
marginal ray height solve,
70
mechanically
compensated, 48
melt certification, 106
melt-recompensation
process, 82
meridional fan, 111
merit function, 3, 65, 68
Mersenne configuration,
42
microscope, 102
microscope objective, 103
135
Index
microscope slide, 103
mid-wave infrared
(MWIR), 108
model glass, 69
modulation transfer
function (MTF), 13
modulation transfer
function (MTF)
bounce, 13
moisture, 6
monochromatic, 4
monochromatic
aberration, 8
Monte Carlo analysis, 73,
86
Monte Carlo tolerance, 83
Narcissus, 92, 93
Narcissus-induced
temperature
difference (NITD), 89,
93
natural stop position, 57
new achromat, 55
Newtons rings, 76
Newtonian telescope, 39
non-reimaging
three-mirror
anastigmat (TMA), 45
nonimaging system, 94
nonsequential ray tracing
(NRT), 92, 94
nonuniformity correction
(NUC), 93
normal dispersion, 105
normal distribution, 83
normalized field
coordinate, 8, 71
normalized pupil
coordinate, 8
null lens, 39, 64
numerical aperture (NA),
4
numerical optimization,
65
Nyquist frequency, 14,
118, 119
Nyquist limit, 118
object angle, 71
object height, 71
object-space telecentric,
58
objective, 98
oblate, 115
oblique spherical
aberration, 25
obscuration, 38
off-axis parabola (OAP),
42
off-axis rejection (OAR),
89
off-axis Schwarzschild
mirror, 46
Offner Relay, 43
old achromat, 55
operand, 65
operating range, 87
optical axis, 1
optical cement, 117
optical cut-off frequency,
13
optical design, 2
optical design process, 3
optical design software, 2
optical drawing, 75
optical invariant, 120
136
Index
optical path difference
(OPD), 11
optical polymer, 107
optical print, 75
optical system, 2
optical transfer function
(OTF), 13
optically compensated, 48
optimization, 3
orthoscopic eyepiece, 100
outside diameter (OD), 78
overcorrected, 24
oversampled, 118
packaging requirements,
6
parabolic distribution, 83
paraboloid, 115
paraxial ray tracing, 109
paraxial refraction, 109
paraxial transfer
equation, 109
parent mirror, 42
parent ray, 94
partial dispersion, 55
passive athermalization,
88
peak-to-valley (PV), 77
peak-to-valley optical
path difference (PV
OPD), 11
Pegel diagram, 49
penalty function, 65
performance budget, 73,
74
performance prediction,
85
periscope, 104
Petzval curvature, 20
Petzval lens, 34
Petzval portrait lens, 34
Petzval projection lenses,
34
Petzval radius, 56
Petzval ratio, 56
Petzval sum, 56
phase reversal, 13
photographic lens, 96
photographic objective,
32, 96
pickup solve, 70
pincushion distortion, 22
pits, 79
pixel pitch, 118
PLAN microscope
objectives, 32
Planar, 33
plano-convex singlet, 110
plastic, 106, 107
Plossl eyepiece, 100
point characteristic, 114
point spread function
(PSF), 12, 112
point-source normalized
irradiance
transmittance
(PSNIT), 89
point-source
transmittance (PST),
89
poker-chip assembly, 80
polar sampling, 72
polarizing beamsplitter,
117
polychromatic, 4
power, 77, 110
preferred glass, 106
pressure, 87
137
Index
primary color, 23
prime group, 48
principal plane, 110
probability distribution
function (PDF), 83
probability distribution
table, 86
prolate, 115
propagation, 109
pupil aberration, 59, 60,
112
pupil ghost, 92
pupil match, 60
pupil matching, 112
pupil sampling, 72
pupil spherical, 60
pupil-matched, 99
push-around, 18, 82
Q parameter, 118
Q-type (Forbes)
polynomial, 61
quarter-pitch, 63
radial gradient, 63
radiation-hardened, 6
radius of curvature
(ROC), 1, 69, 76
range of motion, 82
ray aiming, 60, 71
ray failure, 49, 60, 109
ray fan, 111
ray intercept failure, 49
ray intercept plot, 10
ray scattering, 94
ray splitting, 94
ray tracing, 2
Rayleigh criterion, 5
Rayleigh quarter-wave
criterion, 11
receiver, 94
rectangular grid
sampling, 72
reference ray, 71, 113
reflection, 109
reflective, 37
reflective triplet (RT), 45
refraction, 109
refractive, 37
reimaging, 44
relative aperture, 4
relative bandwidth, 4
relative illumination (RI),
58, 59, 97
relay, 98
relay lens, 104
reproduction lens, 104
reproduction ratio, 96
resolution, 5
retina, 98
retrofocus lens, 36
reverse telephoto, 31, 36
reversing, 52
riflescope, 101
rifling, 90
ring, 72
RitcheyChrtien
telescope, 39
roll, 81
root mean square (RMS)
spot diameter, 9
root mean square (RMS)
spot size, 9
root mean square optical
path difference (RMS
OPD), 11
root-sum-square (RSS), 85
138
Index
sag, 61
sag equation, 115
sagittal, 10, 14
sagittal fan, 111
sampling frequency, 118
sampling modulation
transfer function
(MTF), 118
scatter plot, 9
scattering, 94, 95
scatterometer, 95
Schmidt telescope, 41
SchmidtCassegrain
telescope, 41
Schott index equation,
105
Schupmann lens, 31
Schwarzschild objective,
40
scratches, 79
second-order stray light
path, 90
secondary color, 23
secondary spectrum, 23
Seidel aberration, 114
Seidel diagram, 49
Sellmeier index equation,
105
semiconductor, 108
sensitivity analysis, 84
sensitivity matrix, 85
sensitivity table/chart, 84
sequential ray tracing, 94
Shannon sampling
theorem, 118
shape factor, 53
shim-centered assembly,
80
shock requirement, 6
single-lens reflex (SLR),

97
singlet, 29
skew ray, 72
skewed Gaussian
distributions, 83
slope error, 62
smile, 45
Snells law, 109
solve, 70
source, 94
space adjust, 17, 82
spatial frequency, 13
specification document, 2,
3
specular direction, 95
sphere, 115
spherical aberration, 17
spherical aberration of
the pupil, 100
spherochromatism, 24
spherometer, 76
spider, 38
spiked distribution, 83
splitting, 32, 51
spot diagram, 9
spurious resolution, 13
staining, 79
standard Zernike
polynomials, 114
Steinheil doublet, 30
step size, 66
stigmatic, 39
stigmatic imaging, 115
stop, 111
stop shift, 57
stop symmetry, 57
stray light, 89
stray light analysis, 3, 89
139
Index
stray light suppression,
44
Strehl ratio, 12
striae, 79
super achromats, 30
superposition, 121
surface contribution, 63
surface irregularity, 77,
114
surface roughness, 95
surveillance spotting
scope, 101
survival range, 87
symmetrical tolerance,
73, 82, 85
tailoring, 121
tangential, 10, 14
tangential fan, 111
telecentric lens, 58
telecentricity, 58
telephoto, 31
telephoto lens, 35, 96
telephoto ratio, 35
temperature, 6, 87
terrestrial (spotting)
telescope, 101
Tessar, 32
test plate, 76
test plate fit, 76
test plate fringes, 76
thermal analysis, 3
thermal expansion, 108
thermo-optic coefficient,
87, 108
thick lens, 3
thickness solve, 70
thin lens, 3, 53, 110
thin lens aberration

expression, 53
third-order aberration, 3
third-order aberration
theory, 8
threading, 90
three mirror long (TML),
46
three-mirror anastigmat
(TMA), 44
three-mirror compact, 46
tilt, 81
tolerance stack-up, 81
tolerancing, 3, 73
topogon, 33
total indicator runout
(TIR), 78
total integrated scatter
(TIS), 95
total internal reflection
(TIR), 49
tow/keystone, 45
trans-illumination, 102,
121
transfer contribution, 63
translation, 109
transverse chromatic
aberration, 24
transverse ray aberration
plot, 10
transverse ray error, 7
trefoil symmetry, 77
trial, 86
truncated Gaussian, 83
tube length, 102
tube lens, 102
two-bounce ghost, 92
ultraviolet (UV), 108
140
Index
uncompensated, 84
undercorrected, 24
undersampled, 118
uniform distribution, 83
unity-magnification
relays, 104
unobscured system, 42
V-number, 105
V-coat, 91
vane, 90
variable, 65
variator, 48
varifocal, 47
veiling glare, 89
vertex radius, 62
vibration, 6, 87
viewfinder camera, 97
vignetting, 59
vignetting factor, 59
visual instrument, 98
WALRUS, 46
wave aberration function,
7
wavefront, 7
wavefront differential, 85
wavefront plot, 11
wavefront tilt, 16
wedge, 78
weighted, 68
wide-angle lens, 36, 96
Wood lens, 63
working f /#, 4
working distance, 6
y-bar diagram, 53
yield, 74
yield curve, 86
YNI product, 93
ynu diagram, 109
Zernike polynomial, 77
Zernike polynomials, 114
zoom group, 47
zoom kernel, 48
zoom lens, 47
zoom ratio, 47
Julie Bentley is an Associate Professor at The Institute of Optics,

University of Rochester and has
been teaching courses in geometrical optics, optical design, and product design for more than 15 years.
She received her B.S., M.S., and
Ph.D. in Optics from the University
of Rochester. After graduating, she
spent two years at Hughes Aircraft
Co. in California and then twelve
years at Corning Tropel in New
York. She has experience designing a wide variety of optical systems for both commercial and military markets,
from the UV to the IR, in both refractive and reflective
configurations. She now owns Bentley Optical Design, an
optical design consulting company. Dr. Bentley is a Member of OSA and an SPIE Fellow.
Craig Olson is a Principal Engineer at L-3 Communications. He
received a B.S. in Electrical Engineering from the Georgia Institute of Technology and a Masters
and Ph.D. in Optics from the Institute of Optics at the University of
Rochester. He previously worked in
the telecommunications component
sector and the commercial product
sectors at JDS Uniphase. In addition to developing sensor system
models, he currently designs, builds, and tests active and
passive multispectral sensor payloads for airborne and
ground-based imaging systemsoccasionally finding time
to design a lens. His current research interests involve the
physics of image formation throughout the complete imaging chain. Dr. Olson is a Member of OSA and IEEE, and a
Senior Member of SPIE.

Bentley Field Guide To Lens Design PDF

Uploaded by

Copyright:

Available Formats

Bentley Field Guide To Lens Design PDF

Uploaded by

Document Information

Original Title

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

Bentley Field Guide To Lens Design PDF

Uploaded by

Copyright:

Available Formats

Field Guide to

SPIE Field Guides

Bellingham, Washington USA

FG27 covers and title.indd 3

Library of Congress Cataloging-in-Publication Data

Introduction to the Series

The Field Guide Series

Field Guide to Lens Design

Fundamentals of Optical Design

Field Guide to Lens Design

Field Guide to Lens Design

Appendix: Optical Fundamentals

Field Guide to Lens Design

Field Guide to Lens Design

Glossary of Symbols and Acronyms

Field Guide to Lens Design

Glossary of Symbols and Acronyms

Front focal length

Field Guide to Lens Design

Glossary of Symbols and Acronyms

Field Guide to Lens Design

Glossary of Symbols and Acronyms

Wavefront aberration coefficient

Field Guide to Lens Design

Fundamentals of Optical Design

The z axis is the optical axis or the axis of symmetry of

a rotationally symmetric optical system.

forming a right-handed coordinate system.

Fundamentals of Optical Design

Field Guide to Lens Design

Fundamentals of Optical Design

Optical Design Process

established, and the appropriate type of lens form (e.g.,

Fundamentals of Optical Design

Aperture and Wavelength Specifications

Numerical aperture is the index of refraction of the image

Field Guide to Lens Design

Fundamentals of Optical Design

Resolution and Field of View

Diffraction from a finite circular aperture results in an

Fundamentals of Optical Design

Packaging and Environment

Field Guide to Lens Design

Fundamentals of Optical Design

Wave Aberration Function

Aberrations are imperfections in the image formation

ray in image space:

Field Guide to Lens Design

Fundamentals of Optical Design

Third-Order Aberration Theory

where the other terms in the power series are forbidden

The value of each polynomial coefficient (WI JK ) gives the

Fundamentals of Optical Design

Spot Diagram and Encircled Energy

Spot diagrams are good for final system evaluation,

Field Guide to Lens Design

Fundamentals of Optical Design

Transverse Ray Plot

Transverse ray plots and spot diagrams are typically