You are on page 1of 12

See

discussions, stats, and author profiles for this publication at: http://www.researchgate.net/publication/23466686

Method of zoom lens design


ARTICLE in APPLIED OPTICS DECEMBER 2008
Impact Factor: 1.78 DOI: 10.1364/AO.47.006088 Source: PubMed

CITATIONS

READS

34

414

3 AUTHORS:
Antonin Miks

Jiri Novak

Czech Technical University in Prague

Czech Technical University in Prague

109 PUBLICATIONS 250 CITATIONS

92 PUBLICATIONS 191 CITATIONS

SEE PROFILE

SEE PROFILE

Pavel Novak
Czech Technical University in Prague
65 PUBLICATIONS 122 CITATIONS
SEE PROFILE

Available from: Antonin Miks


Retrieved on: 28 October 2015

Method of zoom lens design


Antonn Mik,* Ji Novk, and Pavel Novk
Czech Technical University in Prague, Faculty of Civil Engineering, Department of Physics,
Thakurova 7, 166 29 Prague 6, Czech Republic
*Corresponding author: miks@fsv.cvut.cz
Received 26 June 2008; revised 13 October 2008; accepted 13 October 2008;
posted 15 October 2008 (Doc. ID 97888); published 7 November 2008

Optical systems with variable optical characteristics (zoom lenses) find broader applications in practice
nowadays and methods for their design are constantly developed and improved. We describe a relatively
simple method of the design of zoom lenses using the third-order aberration theory. It presents one of the
possible approaches of obtaining the Seidel aberration coefficients of individual members of a zoom lens.
The advantage of this method is that Seidel aberration coefficients of individual elements of a given
optical system can be obtained simply by solving of a set of linear equations. By using these coefficients,
one can determine residual aberrations of the optical system without detailed knowledge about the
structure of its individual elements. Furthermore, we can determine construction parameters of the
optical system, i.e., radii of curvature and thicknesses of individual elements of a given optical system.
The proposed method makes it possible to determine which elements of the optical system can be
designed as simple lenses and which elements must have a more complicated design, e.g., doublets
or triplets. 2008 Optical Society of America
OCIS codes:
220.0220, 220.3620, 220.1000, 080.0080, 080.3620, 080.1010.

1. Introduction

Optical systems with variable optical characteristics


zoom lens systemscan be divided in two groups.
One group is called transfocators and the other is
vario-objective lenses. A transfocator is a telescopic
optical system with a variable magnification and
it is often placed in front of the objective. A varioobjective lens is an optical system that images in
such conditions that either the object or the image
is in finite distance or the distance between the image and the object is finite. The main functions of
zoom lenses are to provide a continuous change of
the focal length or magnification at almost constant
and sufficient imaging quality, small residual aberrations [111], no change in the position of pupils, etc.,
in the whole range of requested focal lengths or magnification. This change of optical characteristics is
done by the change of position (shift) of some of
the elements of the optical system [1222]. If the
shift is chosen so as not to change the position of
0003-6935/08/326088-11$15.00/0
2008 Optical Society of America
6088

APPLIED OPTICS / Vol. 47, No. 32 / 10 November 2008

the image plane in the whole range of focal lengths


or magnification, we call it a mechanical compensation of the image plane position. In such optical systems, it is necessary that at least one of the system
members is moving nonlinearly. The detailed analysis of zoom lenses with mechanical compensation can
be found in, e.g., [12,13,18,2022]. If we choose the
shift of optical system members to be linear, we call
it an optical compensation and the position of the image plane changes in small increments dependent on
the construction of the optical system. The position of
the image plane is then constant only for a few values
of focal length or magnification. Zoom lenses with
optical compensation were analyzed in detail in,
e.g., [12,15,19].
When the correction is performed on such types
of zoom optical systems, we proceed in the following
way.
First, according to the given optical characteristics, we have to determine the powers and axial
separations between the elements of the thin lens
system and the position of the pupils. Then we shall
do the primary aberration analysis of individual
elements of this optical system on the basis of the

 on the particumagnitude of parameters h and h


lar optical system member, where h is the height of
 is the height of
incidence of the aperture ray and h
incidence of the principal ray on the given element
of the optical system with the power . This analysis
has to be done for each element of a given zoom lens
system and for various values (through the whole
zoom range) of focal length or magnification of the
system (i.e., for various positions of individual
 determine
system members). Values h and h

apertures of the optical system and values h and h
determine the dimension of individual system members. According to the magnitude of quantities h
 we can decide in the primary stage of the
and h,
design process about the construction of individual
optical system members.
Another step is the analysis of the optical system
using the third-order aberration theory. We often use
the third-order theory (Seidel theory) of aberrations
[111] for design of new types of optical systems.
Even though the Seidel theory is only an approximate theory, its great advantage is that it provides
us with a simple analytical expression of aberrations
of the optical system. Moreover, it enables us to have
a deeper insight into the optical system, which is not
possible in commercial optical design software based
on numerical methods. This is important especially
in the primary stage of the design of optical systems,
where it enables us to see which properties each individual element of the optical system should have to
achieve the minimum value of residual aberration of
the whole optical system. Therefore, it is possible to
determine the type of individual elements (single or
compound element) of the optical system and it results in a significant simplification and acceleration
of the primary stage of the design process of zoom
lenses.
The Seidel theory provides us with analytical
formulas for basic types of aberrations, such as spherical aberration, coma, field curvature, and distortion. There are several possibilities for calculation
of aberration coefficients that depend on the parameters of the optical system. The design parameters
are usually the curvature of surfaces, surface axial
separations, and the refractive indices of the glasses
of individual members of the zoom system. We often
start for the first part of optical design from the thin
lens model of the optical system, where we neglect
the thickness of individual lenses. The formulas
for Seidel aberration coefficients are then simpler
and we are able to obtain an analytical solution in
some cases (doublet, triplet, etc.). Many different sets
of variables [1,2,4,6,811] for thin lens theory have
been used. The formulas for Seidel aberration coefficients are very complicated if we need to calculate
the refractive indices of individual lenses, even in
such a simple case as a cemented doublet [2]. In this
case, the solution leads to the fifth-order equation
and it is practically impossible to obtain the solution
for a complex optical system.

The third step consists of the recalculation of thin


lens elements to thick lens elements design, and it
is followed by an optimization of the construction
parameters using some commercial optical design
software.
In this paper we use the modified formulas for
Seidel aberration coefficients that are given in [8].
By using these formulas, it is possible to calculate
the shape and the refractive index of the glass of
the individual lenses of the optical system. Another
advantage is the possibility to determine which
member of the optical system cannot be a simple lens
and must be replaced by a doublet or a triplet. In our
work, the relations for application of Seidel aberration coefficients [8] for calculation of parameters of
zoom lenses (i.e., optical systems with variable optical characteristics such as variable focal length and
transverse magnification) were derived. Basic information about zoom lenses can be found in [1223].
The aim of our work is to present one possible
approach to determine the Seidel aberration coefficients of individual elements of zoom lenses. By
using these coefficients, we can determine the values
of residual aberrations of the investigated optical
system without detailed knowledge of the type of
its individual elements. It also enables us to determine the construction parameters of the optical system, i.e., radii of curvature and thicknesses of
individual elements of the optical system using relations described in [1,2,2325], which is not a subject
of our paper. An optical system designed using the
described method then serves as a starting point
for further optimization using software for optical
system design and analysis, such as ZEMAX and
OSLO [26,27].
2.

Imagery by a Thin Lens

For simplicity, we assume that the lens is placed in


air. The conjugate equation for the thin lens in air is
1 1 1
;
s0 s f 0

where s is the object distance, s0 is the image


distance, f 0 is the focal length of the lens, and is
the power of the lens. For the power , we have



1 1
n 1
;

r r0

where n is the refractive index of the glass from


which the lens is made and r and r0 are the radii
of curvature of the lens. The magnification m can
be expressed as
m

s0
1

:
s 1 s

We now define two parameters [2] (variables), X and


Y. The shape of the lens can be characterized by
so-called shape parameter X. The shape parameter
10 November 2008 / Vol. 47, No. 32 / APPLIED OPTICS

6089

where

X (bending factor) of the thin lens is defined as


X

r0 r
:
r0 r

s0 s
:
s0 s

2n 1
;
X 1

r0

2n 1
:
X 1

s0 s m 1
2
2

1
1 0 :
Y 0
s s m1
s
s

3. Seidel Aberration Coefficients

Assume a thin lens optical system in air that consists


of K lenses. Seidel aberration coefficients [1,2,8]
(Seidel sums) SI , SII , SIII , SIV , and SV are given
by the following formulas:
K
X

h4i M i ;

i1

SII

K
X

 i Mi
h3i h

i1

SIII

K
X
i1

K
X

i1

iNi
hi h

K
X
i1

SV

h2i N i ;

i1

SIV

K
X

K
X

 3 Mi 3
hi h
i

i1

2Mi 2
h2i h
i

K
X
i1

13

K
X

i ;

2N i
h
i

Y i1

i1

2ni 1
;
2ni

Xi

r0i ri
;
r0i ri

hi i
Y 1 1:
hi1 i1 i

14

In previous equations, we denoted hi as the incidence


height of a paraxial aperture ray (auxiliary aper i as the incidence height
ture ray) at the ith lens, h
of a paraxial principal ray (auxiliary principal
ray) at the ith lens, ri and r0i as the radii of the ith
lens, si and s0i as the object and image distances from
the ith lens, ni as the refractive index of the ith lens,
i as the power of the ith lens, SI as the Seidel sum
for spherical aberration, SII as the Seidel sum for
coma, SIII as the Seidel sum for astigmatism, SIV
as the Seidel sum for field curvature (Petzval
sum), and SV as the Seidel sum for distortion.
We can calculate the radii of curvature of the ith
lens in the case where we know its lens power i, refractive index ni , and shape parameter X i. By using
Eq. (6), we obtain
ri

2ni 1
;
i X i 1

r0i

2ni 1
:
i X i 1

yP y2P x2P
y1 3y2P x2P
S

SII
I
1
2s1 s1 3 u31 u0K
2s1 s1 3 u21 u0K u

y0

y21 yP
3SIII I 2 SIV
 21
2s1 s1 3 u1 u0K u

y31
SV ;
 31
2s1 s1 3 u0K u

11

K
X

Fi

s0i si mi 1
2
2
1

1 0 ;
0
si si
mi 1
si i
si i

10

i1

i
;
ni

n2i
;
Ei Bi =2;
4ni 12


1 1
1 1
0 0 ;
i ni 1
ri ri
si si

Di

Transverse ray aberrations y0 and x0 in the image


plane of the optical system that is composed of K thin
lenses in air can be then calculated from the following equations:


i 
1
h
3
;
hi
ni i
12

6090

ni 1
;
ni ni 1

Yi

For the conjugate parameter Y, we have

SI

Bi

Ai

We can see that, for both radii r and r0 , there is only


one shape parameter X. Knowing the lens power ,
its refractive index n, and the shape parameter X,
we can calculate the radii of curvature of the lens.
Using Eqs. (2) and (4), we have
r

ni 2
;
4ni ni 12
3n 2
Ci i
;
4ni

N i 2i Ei X i F i Y i ;

Parameter Y is called the conjugate parameter and is


defined as
Y

M i 3i Ai X 2i Bi X i Y i Ci Y 2i Di ;

APPLIED OPTICS / Vol. 47, No. 32 / 10 November 2008

x0

xP y2P x2P
2y1 yP xP
SI
S
 1 II
2s1 s1 3 u31 u0K
2s1 s1 3 u21 u0K u
y21 xP
SIII I 2 SIV ;
 21
2s1 s1 3 u1 u0K u

where xP and yP are coordinates of the ray in the


plane of entrance pupil of the optical system, y1 is
the distance of the object point from the optical axis,
s1 is the distance of the object plane from the first
surface of the optical system, s1 is the entrance pupil
distance from the first lens of the optical system,
u1 h1 =s1 is the angle of the paraxial aperture
 1 =s1 is the angle of
1 h
ray in the object space, u
the principal ray in the object space, u0K u1 =m is
the angle of the paraxial aperture ray in the image
space, and m is the transverse magnification of the
optical system. Furthermore, we denoted I as the
LagrangeHelmholtz invariant, which is defined
for the optical system in air as
1
I h1 h

1 1

s1 s1

1 u
 1 h1 :
u1 h

Without loss of generality, we can put the


Lagrange_Helmholtz invariant I 1 and h1 1;
then we have
 1 s1s1 :
h
s1 s1



j
X
di1


hj hj h1
;
h h
i2 i1 i
where di denotes the axial separation between the
ith and i 1th lenses.
4. Modification of the Formulas for Third Order
Aberration Coefficients

The aberration coefficients [8] are quadratic functions in X and Y. Their dependence on the refractive
index is very complicated. It is a very difficult problem to obtain a solution for X and n for the complex
optical system because we must solve several nonlinear equations. To obtain the solution for X and
n for a complex optical system, we define new set
of variables. If we put i 1 and Y i 0 (without loss
of generality) in Eq. (13), we can define new
parameters:

where H is the incidence height of the aperture ray.


Other variants of aberration coefficients are given in
Ref. [8]. Upon substituting Eq. (14) into Eq. (15), we
find that


n2i
ni 2
2
i 1
X
M
;
4 ni ni 12 i ni 12

16



1 ni 1

X :
Ni
2 ni ni 1 i

17

These new parameters depend only on the shape


parameter X and the index of refraction n of the lens
and do not depend on the focal length of the lens.
Upon substituting Eq. (17) into Eq. (16), we find that
 i D i Ai N
 2;
M
E2i i
where

The Seidel difference formula (for I 1 and h1 1)


 is given by
between h and h

 i M i i 1; mi 1;
M
 i N i i 1; mi 1;
N
 i Ai X 2 D i ;
 i Ei X i :
M
N

 N=2;

m H 2 M

15

 and N
 describe the spherical
The parameters M
aberration and the coma for the thin lens having unit
focal length and unit magnification (m 1). We
obtain, for the spherical aberration,

Di

n2i
;
4ni 12

Ai ni ni 2

:
E2i
ni 12

As we can see, the value of the expression A=E2 does


not change significantly in relation to n, as is
apparent from the example shown below:
n 1:5 A=E2 0:84;

n 2:0 A=E2 0:89:

We can then consider this expression as a constant


and put approximately A=E2 0:86. Then we obtain
 i Di 0:86N
 2:
M
i

18

As one can see, values of expressions for C and F vary


very slowly with the value n. We can again consider
these expressions as constants and put C 1:06 and
F 1:31 (considering the average value of the index
of refraction approximately n 1:6). Upon substituting Eqs. (16) and (17) into Eq. (13), we find that it
holds approximately
 i 2N
 i Y i 1:06Y 2 ;
M i 3i M
i

19

 i 1:31Y i ;
N i 2i N

20

 i f 03 M i 2f 02 N i Y i 1:56Y 2 ;
M
i
i
i

21

 i f 02 N i 1:31Y i :
N
i

22


s0 2H 2 M;
and for the sine condition

10 November 2008 / Vol. 47, No. 32 / APPLIED OPTICS

6091

These formulas are sufficiently accurate for all practical cases. As we can see, Eqs. (19) and (20) are lin and N.
 The Seidel sums are also
ear in parameters M
linear in these parameters. We have the possibility to
obtain not only shapes of the lenses, but also the refractive indices of the glasses from which the lenses
are made.
The following steps are used in a third-order aberration design of a lens system. First, we choose which
Seidel sum must be corrected and then we solve the
 and N.
 We then
system of the linear equations in M
 and N
 for the individual
obtain the values of M
 and N
 into
lenses. Upon substitution for M
Eq. (18), we have
 0:86N
 2:
DM

optical elements with given values of Seidel aberration coefficients can be found especially in [1,2,23],
where both the necessary theoretical relations and
numerical examples of calculations of such optical
systems are provided. As we can verify by direct
calculation, the approximations used give almost
the same results as exact calculation.
5. Seidel Aberration Coefficients for Zoom Lens
Design

If we substitute Eqs. (19) and (20) into Eqs. (8)(12),


we obtain the following formulas for Seidel
aberration coefficients:
SIj

23

K
X

i2
h4ji 3i M

i1

For the refractive index of the lens we have [Eq. (14)]


p
D
n p
:
D 0:5

2n 1
;
X 1

r0

1:06

SIIj

K
X

 ji 3 M
i
h3ji h
i

1
Y 1:
s0 2

 ji i Y ji 1N
i
h2ji 2i 2hji h

i1

K
X

 ji i Y ji 1:31;
h2ji 2i Y ji 1:06hji h

i1

K
X

 2 3 M
i
h2ji h
ji i

i1

 ji 2 hji h
 ji i Y ji 1N
i
hji h
i

K
X

 ji 2 Y ji 1:06hji h
 ji i Y ji 2:62
hji h
i

i1

25

APPLIED OPTICS / Vol. 47, No. 32 / 10 November 2008

K
X
i1

When designing an optical system in such a way, we


choose the glasses for some lenses and for the remaining lenses we determine the refractive indices
as described above. This situation frequently appears in practice. In the case when the element of
the optical system is composed of several thin
elements in contact (compound element), then our
formulas are also valid. It is only necessary to under and N
 as parameters that charstand parameters M
acterize spherical aberration and the coma of the
whole compound element (e.g., cemented doublet, tri and N,
 we can
plet). By calculation of parameters M
determine the spherical aberration and the coma of
the compound element and, subsequently, we can
simply calculate the radii and refraction indices of
individual lenses. There exist several works, e.g.,
[1,2,2325], where the formulas for such calculation
are given. The detailed description of the calculation
of parameters (radii of curvature and axial distances
between elements) of various types of compound
6092

h4ji 3i Y 2ji ;

K
X

i1

From Eq. (7), we determine the object and image


distances from
1
Y 1;
s 2

K
X
i1

SIIIj

2n 1
:
X 1

i
h4ji 3i Y ji N

i1

24

The real value for the refractive index (1:43 < n < 2)
we obtain 1 < D < 2:76. In the case when the value of
D exceeds the above-mentioned bounds, we must
replace the simple lens by a more complex optical
system, for instance, a cemented doublet or triplet.
The shape parameter X we determine from
Eq. (16) or Eq. (17). We can calculate the radii of
lenses using Eq. (6), repeated here:

K
X

K
X

i ;

i1

SIVj

K
X
i
i1

SVj

K
X

 3 3 M
i
hji h
ji i

i1

K
X

 2 2 2hji h
 ji i Y ji 3N
i
h
ji i

 2 2 Y ji 1:06hji h
 ji i Y ji 3:93
h
ji i

K  
X
hji
i1

i1

i1

K
X

ni

hji


1
;
ni i

where i 1; 2; ; K, K is the number of optical system members, j 1; 2; ; L, and L is the number of

chosen parameters of the zoom lens system (e.g.,


values of focal lengths or magnifications, which are
considered for correction of the optical system).
Moreover, if we denote
 i;
xi M

bj SIj aj;2K1 ;

bjL SIIj ajL;2K1 ;

bj2L SIIIj aj2L;2K1 ;

bj3L SVj aj3L;2K1 ;

we can rewrite the preceding relations in matrix


form. It holds that
Gx b;

 i;
xiK N

where
G apq ;

aji h4ji 3i ;

b b1 ; b2 ; ; b4L T ;

 2 ; ; M
 K; N
 1; N
 2 ; ; N
 K T ;
 1; M
x M

aj;iK 2h4ji 3i Y ji ;

Starget SI1 ; SI2 ; ; SIL ; SII1 ; SII2 ; ; SIIL ; SIII1 ;


SIII2 ; ; SIIIL ; SV1 ; SV2 ; ; SVL T ;

aj;2K1

K
X

1:06h4ji 3i Y 2ji ;

i1

 ji 3 ;
ajL;i h3ji h
i

whereas, for indices p and q, holds: p 1; 2; ; 4L,


q 1; 2; ; 2K. Starget is the vector of required values
of aberration coefficients (usually Starget 0). The solution of this linear equation system can be obtained
in the form
 1 b:
xG

ajL;iK
K
X

ajL;2K1

 ji i Y ji
h2ji 2i 2hji h

1;

 ji i Y ji 1:31
h2ji 2i Y ji 1:06hji h

i1

 2 3 ;
aj2L;i h2ji h
ji i
 ji 2 hji h
 ji i Y ji 1;
aj2L;iK 2hji h
i

aj2L;2K1

K
X

 ji 2 Y ji 1:06hji h
 ji i Y ji 2:62
hji h
i

i1

i ;

aj3L;i

 3 3 ;
hji h
ji i

 2 2 2hji h
 ji i Y ji 3;
aj3L;iK h
ji i

aj3L;2K1

K
X

 2 2 Y ji 1:06hji h
 ji i Y ji 3:93
h
ji i

i1

3:6

 ji
h
;
hji i

26

 1 denotes a generalized inversion of


The symbol G
 1
matrix G (e.g., for 4L > 2K holds G
GT G1 GT , i.e., the solution is obtained by the
least-squares method). By solving Eq. (26), we obtain
 and N
 for individual memthe values of variables M
ber of the optical system. By using Eq. (23), we can
determine which of the optical system elements can
be a simple lens and which must be composed of several lenses (doublet, triplet, etc.). When the calculation is being done, we usually choose a large number
of the states of the zoom lens system, i.e., 4L 2K,
 and N
 for indiand then we calculate the values M
vidual members of the zoom lens using the leastsquares method. In the case we require, e.g., only
correction of aberration coefficients SI , SII , and
SIII , then indices p and q will have values in the
range p 1; 2; ; 3L, q 1; 2; ; 2K. The vector
Sres of residual values of the aberration coefficients
of the optical system can be calculated from
Sres Scalc Starget Gx b;
where x is the solution of the linear system of
Eq. (26), Scalc is the vector of calculated values of
the aberration coefficients, and Starget is the vector
of required values of the aberration coefficients.
The main advantage of our method that uses vari and N
 for the characterization of individual
ables M
optical elements is the fact that the Seidel aberration
coefficients of individual elements of the optical system can be obtained by solving a set of linear equations. In the case that we choose other variables (e.g.,
radii of curvature and their various combinations),
we would have to solve a set of nonlinear equations
10 November 2008 / Vol. 47, No. 32 / APPLIED OPTICS

6093

and that is a very difficult problem. Another advantage of this method is the fact that the values of
 and N
 are always finite, which is not
variables M
necessarily true for other variables.

doublet in contact and comprising two different


glasses or optical materials, the conditions that the
doublet be corrected at three wavelengths F , C ,
and d are

6. Chromatic Aberration

The refractive index is a function of wavelength for


all optical media other than vacuum, an effect sometimes called dispersion; since the properties of an
optical system depend on the refractive indices of
the media, it follows that dispersion gives rise to variation in these properties with wavelength, These
variations are called chromatic aberrations [110].
The formulas for chromatic sums for an optical
system of K thin lenses are given by [1,2,4,9]:
CI

K
X
i1

CII

K
X
i1

h2i

i
P ;
i i

27

i
hi h

i
P ;
i i

nd 1
;
nF nC

nF n
;
nF nC

i1

i ;

2
X
i
i1

0:

30

Suitable glasses are, for example, Schott BK7/SF5


(1 2:0103, 2 1:0103). In the case of a thin
6094

APPLIED OPTICS / Vol. 47, No. 32 / 10 November 2008

0;

2
X
i
i1

Pi d 0: 31

To achieve such color correction at three wavelengths


(apochromatic), the relative partial dispersion of the
two glasses must be equal, i.e., P1d P2d . Suitable
glasses are, for example, Schott FK54/KzFSN2
(1 2:49, 2 1:49). In the case of a thin triplet
in contact that is composed of three different glasses
or optical materials, the conditions that the triplet be
color corrected at four wavelengths (superachromatic) F , C , d , and g are

28

29

2
X
i
i1

i ;

i1

where nd is the refractive index of the glass for wavelength d 589 nm, nF is the refractive index of the
glass for wavelength F 486 nm, nC is the refractive index of the glass for wavelength C 656 nm,
and n is the refractive index of the glass for arbitrary
wavelength .
The sums CI and CII are usually coupled with the
Seidel sums as constituting a group of primary aberrations. The optical system does not suffer from chromatic aberrations if CI 0 and CII 0. We call the
optical system achromatic if chromatic aberrations
are eliminated for two wavelengths, apochromatic
if chromatic aberrations are eliminated for three
wavelengths, and superachromatic if chromatic
aberrations are eliminated for four and more wavelengths.
In the case of a thin doublet in contact and comprising two different glasses or optical materials,
the conditions that the doublet be corrected at two
wavelengths F and C are
2
X

i ;

i1

3
X

where CI is the chromatic sum for a longitudinal


chromatic aberration, CII is the chromatic sum for
a transverse chromatic aberration, i is the Abbe
number of the ith lens, and Pi is the relative partial
dispersion of the ith lens. Abbe number and relative
partial dispersion are defined as

2
X

i1

3
X
i
i1

3
X
i

0;

3
X
i
i1

Pi d 0;
32

Pi g 0:

Suitable glasses are, for example, Schott PK51/


LaK8/FK51 (1 1:7508, 2 2:0464, 3
1:2956).
During the design of individual members of the
zoom lens system, we mostly require that the
individual member of the optical system is free of
chromatic aberration. This can be achieved by
application of Eqs. (30)(32) for individual types of
chromatic corrections.
By the method described in Sections 5 and 6, we
obtain parameters of the optical system that is
consequently used as a starting point for further
optimization using commercial optical design software, such as ZEMAX or OSLO.
7.

Examples

In the following subsections, we will show examples


of the application of previous formulas for the optical
system design. The approximate calculation will also
be compared to calculation with respect to exact
formulas.
A. Calculation of Seidel Coefficients for Thin Cemented
Doublet Using Exact and Approximated Formulas

To show the magnitude of errors due to usage of the


approximate formulas in Eqs. (19) and (20), we present an example of calculation of Seidel aberration
coefficients for a cemented doublet with approximate
formulas. Parameters of the cemented doublet are
shown in Table 1 (linear dimensions are given in
millimeters), where r and r0 are the radii of curvature
of the lens surfaces, d is the thickness of the lenses, n
is the index of refraction, and f 0 is the focal length of
the doublet. Table 2 then presents the results of calculation of aberration coefficients SI and SII for the
mentioned doublet.

Table 1.

Parameters of Cemented Doublet

f0 1
i

ri

r0i

di

ni

glass

1
2

0.4123
0:8482

0:8482
3.5013

0
0

1.57488
1.73212

BaK1
SF1

In Table 2 we denoted M, N, SI , and SII as exact


values calculated from Eqs. (13), (8), and (9) and M a ,
N a , SIa , and SIIa are approximate values calculated
using Eqs. (13), (19), and (20). As we can see from
Table 2, the performed approximation offers very
good results and it can be used in practice. It was
noted that the main advantage of the approximation
is the linearity of formulas for Seidel aberration coef and N.
 Other
ficients with respect to parameters M
examples of applications with detailed numerical
data can be found in [8].
B. Example of Calculation of Two-Element Zoom Lens

Now, we will show the process of calculation of


parameters of the two-element zoom lens using the
third-order aberration theory. Such optical systems
are often used in practice, e.g., as image reverting
systems in riflescopes or as a part of some more complicated zoom lens system. The scheme of the twoelement optical system is shown in Fig. 1. The system
is composed of two elements with powers 1 and 2
and the distance between both elements is d. The
optical system images the object in the object plane
, which is located at the distance s1 from the first
element of the optical system. The image is then created in the image plane 0 , which is located at the distance s02 from the second element of the optical
system. The paraxial aperture ray outgoing from
the point A1 in the plane intersects the first member of the optical system at height h1 from the optical
axis A1 A02. The paraxial principal ray outgoing from
the point B1 in the plane intersects the first mem 1 from the optical axis. The center of
ber at height h
the entrance pupil P1 is located in the distance s1
from the first member of the optical system and in
the distance p1 s1 s1 from the object plane .
The image of the point P1 is the point P02 that is
the center of the exit pupil. The image plane 0 is situated in the distance e from the object plane . The
images of points A1 and B1 are denoted as A02 and B02 .
We obtain, for the power and focal length f 0
1= of the optical system, the following equations:
Table 2.

Seidel Coefficients S I and S I I of the Doublet

 1 0, m 1, s1 2, I 1
h1 1, h
i
1
2

i
M

i
N

Mi

M ia

Ni

N ia

2.0816 0.4917
16.534
16.516 0:8154 0:7765
1.7733 0.6571 10:098 10:196
3.6184
3.6436
SI 6:4364, SIa 6:2303, SII 2:803, SIIa 2:867

Fig. 1. Two-element optical system.

1 2 d1 2 ;

f0

f 01

f 01 f 02
;
f 02 d

where f 01 and f 02 are values of focal length of individual elements of the optical system. We obtain, for an
arbitrary point A1 that is situated in the distance s1
from the first element of the system,

s1 f 0


1
d
1 0 ;
m
f2



d
s02 f 0 1 m 0 ;
f1

where m y02 =y1 is the transverse magnification.


The distance e of the image plane 0 from the object
plane can be calculated from e s1 d s02 .
By substitution into the preceding formula, we obtain, for the distance d between both elements of the
optical system, the following quadratic equation:
d2 ed f 01 f 02 e f 01 f 02 m 12 =m 0:
The solution of the previous equation can be written
as
d


q
1
e  e2 4f 01 f 02 e 4f 01 f 02 m 12 =m :
2

This formula makes it possible to calculate the distance d of the two-element optical system with respect to the transverse magnification m of the
optical system and the distance e between the object
plane and the image plane 0 .
The process of calculation of design parameters of
a two-element zoom optical system can be performed
in the following way. We choose the distance e between the object and the image plane and determine
focal lengths f 01 and f 02 of both elements of the optical
system. The position of the entrance pupil is then given either by the distance s1 from the first lens or by
the distance p1 from the object plane. Moreover, we
set minimum and maximum values mmin ; mmax of
the transverse magnification m of the zoom lens,
the number L of chosen parameters of the zoom lens
system, and the required values of the aberration
coefficients for the design process.
Using given parameters, we can calculate dis1, h
 2 , and elements
tances s1 , s02 , heights h1 , h2 , h
of matrix G and vector b. By solving Eq. (26), we ob 1, M
 2, N
 1 , and
tain the required values of variables M
10 November 2008 / Vol. 47, No. 32 / APPLIED OPTICS

6095

 2 for individual elements of the zoom lens. From


N
these values, which specify aberration properties of
individual elements of the zoom lens, we can determine [using Eq. (23)] which of the optical system
members can be a simple lens and which must be
composed of several lenses (e.g., doublet, triplet).
 determine apertures and values
Values h and h
 determine the dimension of individual system
h, h
members. According to the magnitude of quantities
 we can decide, in the
 1, M
 2, N
 1, N
 2 , h, and h,
M
primary stage of the design process, about the construction of individual optical system members. If
the optical system should be achromatic, then the
power of the lenses of individual elements can be
determined from Eqs. (27) and (28).
The process of calculation of parameters of twoelement zoom lens can be carried out using the
following formulas:
1. Given: f 01 , f 02 , e, s1 (or p1 ), mmin , mmax , K, L.
We require, for example, correction of three types
of aberrations: spherical aberration SI 0, coma
SII 0, and astigmatism SIII 03L > 2K.
2. Calculating: m hmmin ; mmax i,
d


q
1
e  e2 4f 01 f 02 e 4f 01 f 02 m 12 =m ;
2

f 01 f 02
;
s1 f 0
f 01 f 02 d


d
s02 f 0 1 m 0 ;
f1
f0

h1 1;

h2

 2 h2 h
 1 d;
h

Y 1 1

2f 01
;
s1

s02
;
s1 m


1
d
1 0 ;
m
f2

SI SII SIII 0:
 2 0:232, N
 1 0:077,
 1 0:734, M
We obtain M

and N 2 0:102. Residual ray aberrations dx; dy
of the third order that we obtain using the abovementioned equations for the axial point (image
height y0 0 mm) and for the numerical aperture
NAmax 0:1 are shown in Fig. 2 (linear dimensions
are introduced in millimeters). As one can see, the
correction of residual ray aberrations of the third
order is very good in the whole range of required
magnification values.
C.

Example of Calculation of Three-Element Zoom Lens

As another example, we will present the calculation


of the three-element zoom lens that images the object
at infinity. We designate f 0min as the minimum value
of the focal length, f 0max as the maximum value of the
focal length, f 01 as the focal length of the first element, m3 as the transverse magnification of the third
element of the system, s03 as the image distance, and
sc as the position of the aperture stop with respect to
the third element of the optical system. Then we can
determine the remaining parameters of the optical
system using the following formulas:
f 03 s03 =1 m3 ;

f 02 1=1=f 01 1=f 03 ;

d1 f 01 f 02 m3 f 01 f 02 =f 0 ;
d2 f 0 1 d1 =f 01 s03 =f 0 =m3 ;

 1 s1s1 s1 p1 s1 ;
h
s1 s1
p1

Y2 1

Required aberration coefficients:

2f 02
;
s02

where f 02 and f 03 are focal lengths of the second and


third elements of the system, d1 is the distance between the first and second elements, d2 is the distance between the second and third elements, and
f 0 is the focal length of the optical system. Choosing,

1 1=f 01 ;

2 1=f 02 :

Using the previous formulas, we can calculate individual elements of matrix G and vector b and we
can solve Eq. (26) with respect to x. As an example
of the described design process, we will show calculations for two-element zoom system with the following
parameters (Table 3).
Given parameters:
f 01 50 mm;

f 02 50 mm;

mmin 0:333;

mmax 1;

6096

e 0;
K 2;

s1 0;
L 9:

APPLIED OPTICS / Vol. 47, No. 32 / 10 November 2008

Fig. 2. Spot diagrams of two-element zoom lens for y0 0 mm and


SI SII SIII 0.

Table 3.

Parameters of the Optical System (e 0, f 01 50 mm,


f 02 50 mm)

s1

s02

f0

h1

h2

1
h

2
h

0.333 136.607 78.866 57.741 43.297 1 1.732 0 57.741


0.666 111.239 90.824 20.416 122.455 1 1.225 0 20.416
1.000 100.000 100.00
0

1 1.000 0
0

for example, the following values: f 0min 60 mm,


f 0max 120 mm, f 01 100 mm, m3 0:4, s03
70 mm, and sc 40 mm (aperture stop is positioned
between the second and the third elements), we obtain f 02 33:333 mm and f 03 50 mm. Distances d1
and d2 are given in Table 4 for various values of the
focal length f 0 (linear dimensions are introduced in
millimeters).
By using previous formulas, we can calculate individual elements of matrix G and vector b, and we can
solve Eq. (26) for K 3 and L 9 with respect to x.
Required aberration coefficients are SI SII
SIII 0. We obtain the following parameters
 2 0:0154, M
 3 0:554, N
 1 0:473,
 1 0:247, M
M
 3 0:850. Residual ray aberra 2 0:0089, N
N
tions dx; dy of the third order that we obtain using
the above-mentioned equations for the axial point
(image height y0 0 mm) are shown in Fig. 3.
Figure 4 then presents residual ray aberrations for
the off-axis point (image height y0 14 mm). Linear
dimensions in Figs. 3 and 4 are given in millimeters.
The f number of the optical system is 5.6. As one can
see, the correction of aberrations of the third order is
very good and stable in the whole range of required
focal length values.
8. Conclusion

One possible approach to obtaining Seidel aberration


coefficients of individual elements of a zoom lens was

shown. In our method we defined new variables M
 that give simple expressions for the Seidel
and N
sums. The Seidel sums are then linear functions in
 and N
 depend only
these variables. The variables M
on the shape parameter X and refractive index n of
the lens and do not depend on the focal length of the
 and N
 is
lens. The advantage of the parameters M
the possibility to determine the refractive indices
of the lenses, which was not possible before. It is a
powerful tool for design of the optical systems. In
the case when the element of the optical system is
composed of several thin elements in contact (compound element), then our formulas are also valid.

It is only necessary to understand parameters M

and N as parameters that characterize spherical
aberration and coma of the whole compound element
Table 4.

Fig. 3. Spot diagrams of three-element zoom lens for y0 0 mm


and SI SII SIII 0.

(e.g., cemented doublet, triplet). By calculation of


 and N,
 we can determine spherical
parameters M
aberration and coma of the compound element and
subsequently we can simply calculate the radii and
refraction indices of individual lenses. It was shown
on the example of a cemented doublet that the approximations used give almost the same results as
the calculation using exact formulas. The same conclusions are also valid for more complicated optical
systems, e.g. triplets. Another advantage is the possibility to determine which element of the optical system cannot be a simple lens and must be replaced by
a doublet or a triplet. The formulas for calculation of
the parameters of zoom lenses using defined vari and N
 were derived. The main advantage
ables M
 and N
 for the
of our method that uses variables M
characterization of individual optical elements is
the fact that the Seidel aberration coefficients of

Parameters of Three-Element Zoom Lens

f0

d1

d2

60
75
105
120

44.444
48.889
53.968
55.556

91.666
79.166
54.167
41.667

Fig. 4. Spot diagrams of three-element zoom lens for y0 14 mm


and SI SII SIII 0.
10 November 2008 / Vol. 47, No. 32 / APPLIED OPTICS

6097

individual elements of the optical system can be obtained by solving a set of linear equations. In the case
that we choose other variables (e.g., radii of curvature and their various combinations), we would have
to solve a set of nonlinear equations and that is a
more difficult problem. The described process of
optical design was shown on the example of twoelement and three-element zoom lenses. As can be
seen from the presented results, the proposed method enables us to achieve a very good correction of ray
aberrations of the third order. Parameters of an optical system obtained by the described method then
can be used as a starting point for further optimization using optical design software, such as ZEMAX
or OSLO.
This work has been supported by grant
MSM6840770022 from the Ministry of Education
of Czech Republic.
References
1. B. Havelka, Geometrical Optics I, II (Czech Academy of
Science, 1955).
2. D. Argentieri, Ottica Industriale (Hoepli, 1942).
3. H. H. Hopkins, Wave Theory of Aberrations (Oxford, 1950).
4. G. G. Slyusarev, Aberration and Optical Design Theory (Adam
Hilger, 1984).
5. A. Cox, A System of Optical Design (Focal, 1964).
6. P. Mouroulis and J. MacDonald, Geometrical Optics and
Optical Design (Oxford, 1997).
7. A. Mik, Applied Optics (Czech Technical U. Press, 2000).
8. A. Mik, Modification of the formulas for third-order aberration coefficients, J. Opt. Soc. Am. A 19, 18671871 (2002).
9. A. E. Conrady, Applied Optics and Optical Design, Part I
(Oxford, 1929), Part II (Dover, 1960).

6098

APPLIED OPTICS / Vol. 47, No. 32 / 10 November 2008

10. H. Haferkorn, Bewertung Optisher Systeme (VEB Deutscher


Verlag der Wissenschaften, 1986).
11. W. T. Welford, Aberrations of the Symmetrical Optical Systems
(Academic, 1974).
12. A. D. Clark, Zoom Lenses (Adam Hilger, 1973).
13. K. Yamaji, Progress in Optics (North-Holland, 1967), Vol. VI.
14. A. Wather, Angle eikonals for a perfect zoom system, J. Opt.
Soc. Am. A 18, 19681971 (2001).
15. G. Wooters and E. W. Silvertooth, Optically compensated
zoom lens, J. Opt. Soc. Am. 55, 347351 (1965).
16. D. F. Kienholz, The design of a zoom lens with a large
computer, Appl. Opt. 9, 14431452 (1970).
17. A. V. Grinkevich, Version of an objective with variable focal
length, J. Opt. Technol. 73 343345 (2006).
18. K. Tanaka, Paraxial analysis of mechanically compensated
zoom lenses. 1: Four-component type, Appl. Opt. 21, 2174
2183 (1982).
19. G. H. Matter and E. T. Luszcz, A family of optically compensated zoom lenses, Appl. Opt. 9, 844848 (1970).
20. K. Tanaka, Erratum: Paraxial analysis of mechanically
compensated zoom lenses. 1: Four-component type, Appl.
Opt. 21, 3805 (1982).
21. K. Tanaka, Paraxial analysis of mechanically compensated
zoom lenses. 2: Generalization of Yamaji type V, Appl. Opt.
21, 40454053 (1982).
22. K. Tanaka, Paraxial analysis of mechanically compensated
zoom lenses. 3: Five-component type, Appl. Opt. 22, 541
553 (1983).
23. H. Chretien, Calcul des Combinaisons Optiques (Masson,
1980).
24. H. H. Hopkins and V. V. Rao, The systematic design of two
component objectives, Opt. Acta 17, 497514 (1970).
25. M. I. Khan, Cemented triplets: a method for rapid design,
Opt. Acta 31, 873883 (1984).
26. www.zemax.com
27. www.sinopt.com

You might also like