You are on page 1of 13

Design of Spherical Aberration Free Aspherical Lens

by use of Ray Reverse Tracing Method

Han Seob KIM(hanseob@mail.ulsan.ac.kr), Kyu Yeol PARK(kypark@mail.ulsan.ac.kr),


Won Kyu LEE(wklee@mail.ulsan.ac.kr), Jeong Up JEON (jujeon@mail.ulsan.ac.kr),
Sung Tae PARK(spark@mail.ulsan.ac.kr)

ABSTRACT
In this study, aberration free aspherical lens design method named ray reverse tracing
method is introduced. Differently from the traditional design method, the ray reverse
tracing method traces the shape and location of a real object by use of its virtual image.
From the result, especially spherical aberration free aspherical lens could be designed by
use of the ray reverse tracing method. Furthermore, it could reduce the degree of
dependence of optical characteristics on designer’s ability, because deformation terms and
optimization can be eliminated, which has been performed in conventional lens design
process.
Key Words : Spherical Aberration, Aspherical Lens, Ray Reverse Tracing Method,
Deformation term

1. Introduction

Various studies are going on in the field of material technology and design technology
etc. for lenses which ranges from small-sized one that is used in endoscope for surgical
operation of internal treatment to large-sized one that is used in astronomical telescope, in
order to get more accurate, clear and better image. The most important function required
from optical lense is their optical role that embodies high precision image exactly same as
real object. But there is an undesirable factor in lens characteristics so called aberration
that hinders lens features. When defining an element that transforms a point in an object
space to the corresponding one in an image space as a lens or an optical system, the
idealistic optical system is one that transforms exactly 1:1. And the aberration is the
phenomenon(1) that the real optical system deviates from the idealistic optical system. Thus,
in order to improve the optical system performance, many studies on removal or reduction
of the aberration effect are going on. As a typical example, various methods such as an
analytic method(2) that applies higher degree of order in the mathematical expression of the
aberration and materialistic method(3) that uses new material for one lens to have various
refractive indices are presented. The most widely used method is the one that uses lens for
aberration correction. But, because this method using multiple lenses has some demerit that
the whole system becomes complicated and large-sized, it has its limitation today when
speculating the trend that micronization and light-weightization of the optical system is
rapidly realized.
At present, most optical system changes from spherical lens optical system to aspherical
lens optical system(4)(5), due to the trend of system micronization and necessity of high-
function and high quality including aberration correction. Aspherical lens optical system
has many advantages that it removes spherical aberration by improving focusing
performance and does not need complicated lens combination but still has high
permeability. However, because it has many difficulties in designing and manufacturing,
use of spherical lens that has approximate solution has been preferred(6). In recent years,
instead of the difficulties in designing and manufacturing, study on the application of the
shape of the aspherical lens and its optical merits in many other fields is actively being
proceeded. As a typical example of special use of an aspherical lens, in order to improve
performance and function of high-tech military equipment, DARPA (Defense Advanced
Research Projects Agency) is carrying out a study on the accuracy enhancement in
information by maintaining broader visual field and reducing the aberration from changing
the traditional shape of canopy part of an airplane or head part of a missile which usually
obtains forward information into aspherical shape(7).
Because, differently from spherical lens, aspherical lens is not defined in one curvature,
it has design difficulties. Even though it is an axis-symmetric aspherical lens that is in
general use, it is inevitable to apply design equation for the complicated aspherical surface,
and individual coefficients used in the equation are very important factors that affect the
lens characteristics. And optimization procedure required when designing lens using
commercial software also has great affects on lens characteristics. But, instead this
optimization procedure and selection of individual coefficients play as very important
factors in lens design, these factors are arbitrarily determined by the designer relying on
the designer’s experiences. Thus, eventually, this has the problem that the lens
Spot image
Spot image

Focal region
(a) Spherical aberration free (b) Spherical aberration

Fig. 1 Illustration of spherical aberration

characteristics have to be determined differently depending on designers’ experience and


capability though the initial values for lens design are identical to each other.
Based on this background, in this study, in order to avoid the traditional design method that
mainly depends on the designer’s experience, a new design method named ‘ray reverse
tracing’ method was devised. By using the ray reverse tracing method, it was confirmed
that it was possible to remove the spherical aberration of a spherical lens, and it was also
possible to design a spherical aberration free aspherical lens.

2. Spherical Aberration and Ray Reverse Tracing Method

2.1 Spherical Aberration


Generally, there are various kinds of aberration in spherical lens of optical devices.
Among the various aberrations that could be found in an optical system, spherical
aberration is especially regarded as the most important factor that determines the
performance of the optical device. Fig. 1 shows the effects of spherical aberration. Rays of
light parallel incident along the optical axis of a lens pass the lens and gather on one point,
which is called a focus. But if the spherical aberration of the lens exists, the lens focus can
not form geometrically a single point in a space and has a specific region called focal
region as shown in the figure 1/(b). This phenomenon is caused by the fact that the incident
angle of the ray incident onto the spherical lens is getting larger and larger, as the incident
ray parallel to the optical axis gets far away from the axis, comparing with that of the
incident ray nearer to the axis. In this case, the relation between the incident angle and the
refractive angle can be explained by the Snell’s law(8).
Snell’s law is an equation consisting of incident angle of a ray incident onto a lens
surface, its refractive angle and index of refraction of lens material. The bigger the
diameter of a spherical lens is, the bigger the spherical aberration is. And the size of the
spherical aberration directly influences the size of the focus. Therefore, in a design of a
lens A
Lens α
Ray nSinα=n`Sinβ
?
Ray
? ? ’
Object Ray
Object ? Optical axial
F1 F
? ’
F2 Image
Focal point Optical axis Ray
? ’ nSinα`=n`Sinβ α`
Image

(a) Ray tracing method (b) Ray reverse tracing method

Fig. 2 Principle of ray (reverse) tracing method

high-function and high-quality lens, the spherical aberration that has great influences on
the image forming function of the lens must indispensably be corrected.

2.2 Ray Reverse Tracing Method


In geometrical optics, as a traditional method to draw lens multiplying factor and the
location of an image, ray tracing method has been used(8)(9). Fig. 2 shows the basic
principle of the ray tracing method. This method is a simple one to determine the size and
the location of an image made through the lens using the lens focal length that is known
when the size and the location of an object vary, and has rules as follows.

1.Ray of light ? starting from the object, parallel incident upon a lens, passes F2 .

2.Ray of light ? starting from the object and passing F1 proceeds parallel to optical

axis after passing through the lens.

3.Ray of light ? starting from the object and passing the center of lens proceeds

straight without refraction.

Rays of light starting from the object pass the lens and form the image of the object
depending on a focal length, the distance between the object and the lens, and the size of
the object. In ray tracing method, all the rays of light starting from the object are used to
A1′ A1
B

O p1’ P R F’ F
p1
Fig. 3 Spherical aberration removal method(Ray reverse tracing)

determine the location and the size of an image formed by the lens.
But, in this study, ray reverse tracing method was devised for correction of the
spherical aberration that is inherent to the spherical lens and for generation of aspherical
surface. In this method, an intended virtual image is first made depending on the optical
characteristics of the lens used, and then all the rays of light start not from the object but
from the virtual image and form the object after passing the lens. The principles of the ray
reverse tracing method can be explained as follows. When making an image, at a point on
F, of an object at infinite distance using an arbitrary lens, an arbitrary line is drawn from
the point F to the lens surface. When the end point of the line on the lens surface is denoted
as a point A, line AF is a refractive ray of the lens and the angle between the normal line
at A and the line AF is a refractive angle. By substituting the refractive angle derived
from this and the lens index of refraction into the Snell’s expression, the incident angle
between the lens and the object and the incident ray can be found. But, in this case, if the
lens has aberration, the incident ray with the incident angle that was found from the
refractive angle and the refractive ray that connects the virtual image and a point on the
lens surface will not pass the corresponding point on the object. Therefore, in order for the
incident ray with the incident angle found form Snell’s law to pass exactly the
corresponding point on the object, the curvature of the lens surface must inevitably be
changed. In this study, on purpose of removing the spherical aberration of a lens and
designing the shape of the aspherical lens surface by use of these optical characteristics,
the ray reverse tracing method was devised and utilized.
3. Removal of spherical aberration and generation of an aspherical surface

3.1 Removal of spherical aberration


As described above, the spherical aberration in an optical system, due to which parallel
rays having passed through a lens do not converge into one point called a focus of the lens,
can be the cause of deterioration in an optical characteristic such as resolution. But, in the
case of the lens whose aberration is completely removed, all the parallel rays having
passed through the lens converge into one single point. In ray tracing of an object at a finite
distance from a lens, various rays with various incident angles from the object passing

 
∠OFA = tan −1  h 
 ( OF − p ) 
∠ARF = 180° − ∠ORA = 90° + ∠AOR
 
∠FAR = 90° − ∠OFA − ∠AOR = 90° − tan −1  h  − tan −1  h p  (1)
 ( OF − p )   
∠ORA = 90° − ∠AOR = 90 − tan −1 (h p ) (2)
   
sin 90 − tan −1  h  = n sin 90° − tan −1  h  − tan −1  h  (3)
  p    ( OF − p )   p 
n : Index of refraction

through the lens finally forms an image exactly same as the shape of the object. On the
other hand, if the object is at infinite distance from the lens, all incident rays from the
object onto the lens will be regarded as parallel and form a point image that is same as the
focus of the lens. Therefore, in this study, by using these principles and applying the ray
reverse tracing method, an aberration removal method was suggested.
Fig. 3 shows spherical aberration removal principles using the ray reverse tracing
method. In the case of a spherical lens, its focal length near its radical axis zone can easily
be found from mathematical expression with its radius of curvature and index of refraction
only. The method of spherical aberration removal of an arbitrary lens whose focal length is
already known is as follows. First, a line A1 F is drawn from the lens focus(F) lying on the
optical axis to an arbitrary point(A) on the lens surface where the spherical aberration
removal is required. The line A1 F is a refractive ray between the lens and an image.

Because the point is located on the lens surface, when a line A1 R is drawn from the point
to the center of curvature of the lens, it becomes normal to the lens surface at A1. Thus, the
angle ∠FA1R between the line and the refractive ray becomes a refractive angle of the lens
at A1. An incident angle can be found using a index of refraction of the lens and the
refractive angle shown above. But, because the lens has spherical aberration, it is
impossible to draw an incident ray between the lens and the object using the incident angle
found above. To make it possible, the radius of curvature of the lens must be varied
depending on the point on the lens surface.
In the concrete, in order to correct the spherical aberration using the ray reverse tracing
method, the spherical aberration of the lens must be defined first.
To do this, all incident rays must be parallel to an optical axis, and all refractive rays must
converge into one single point. And for the point F to be the focus, a ray starting from an
object and passing through an arbitrary point A1 must be parallel to the optical axis, and the
angle ∠ORA becomes an incident angle at an arbitrary point A1. Here, the incident angle
and the refractive angle at an arbitrary point A1 on the lens surface must satisfy the Snell’s
law. Accordingly, substituting equation (1) and (2) into the Snell’s law will yield the
equation (3). Because the index of refraction (n), the height (h) of an arbitrary point A1 on
the lens surface and the focal length are constant in the equation (3), the spherical
aberration of a lens becomes a function only of p1 the distance from the origin (O) to an
arbitrary point A1 on the optical axis. Therefore, finding out the correct p1’ makes A1
change to A1’ where eventually the spherical aberration is completely removed.
In the case of real practice, very small value is assigned to h, and A1 is let be the origin
of the next point A2 in the removal process of the spherical aberration. By repeating this
process until the accumulated value of h becomes the same as the radius of a lens, the
spherical aberration all over the lens surface can be removed. When this process is finished,
a new lens surface called an aspherical surface that is different from the original spherical
surface is generated.

3.2 Design of an aspherical surface


As described before, in order to remove the spherical aberration of a lens by use of the
ray reverse tracing method and Snell’s law, focal length and curvature corresponding to
basic data of a lens used are required. But by application of this process, shape of an
aspherical surface that has no spherical aberration can possibly be designed using the focal
length of a lens only, even though the basic shape of the lens is not yet determined. Here,
B A5
h A4
h A3
h A2
h A1
h
O R1 R2 F
p1 p2p3 p4 p5

Fig. 4 Illustration of aspherical lens design using ray reverse tracing method

the focal length is the design data given basically in design of all lenses. The principle of
shape design of an aspherical lens without spherical aberration using the ray reverse
tracing method and a focal length only is as shown in Fig. 4. F is a focus. A focus (F) is
marked at a distance of a focal length given for design from the origin (O). The line
passing the origin and the focus is an optical axis of the lens for design. As described in the
previous section, in order to apply the ray reverse tracing method with the given condition
on the basic shape of a lens, an arbitrary point on the lens surface is picked at the distance
of h from the origin and Snell’s law is applied to this point. But, when the basic shape of a
lens is not given and the ray reverse tracing method is to be applied using the focal length
only, an arbitrary point A1 is picked at the distance of h vertically from the origin. Here, the
location of A1 has a same coordinate as that of the origin in the direction of the optical axis
means that the surface formed from revolution of line A1O around the optical axis is
normal to the optical axis. In this case, because the normal ray incident onto this surface
does not refract and proceeds straight forward and plays a role like the optical axis while
A1 changes its location in the course of design, the problem on selection of an initial
location of the point A1 can be neglected. Thus, the line connecting the focus F and an
arbitrary point A1 becomes refractive ray, and the incident ray is produced by drawing a
line passing through the point A1 and parallel to the optical axis. When drawing a line
connecting an arbitrary point A1 and the origin, and drawing another line A1 R1 passing A1

and normal to the line A1O , the angle between the line A1 R1 and the refractive ray becomes

a refractive angle, and the angle between the line A1 R1 and the incident ray becomes an
incident angle. After this process, in the same manner as in the spherical aberration
removal method described in the previous section, substituting the refractive angle, the
incident angle, and the index of refraction of
a lens material into the Snell’ law and
expanding it result in an equation (3). It is a
function of the distance between the origin
and the coordinate of A1 in the direction of
optical axis. Therefore, the value of p1 that
satisfies the equation (3) becomes the
coordinate of A1 in the direction of the
optical axis. In the Fig. 4, A1 is a temporary
origin to find out the location of A2, and in
turn A2 is for A3 and so on. By repeating
these process until the accumulated value of Fig. 5 Aspherical lens design Module
h becomes the same as the design radius of a
lens, the shape of an aspherical lens is designed. It can be decided that applying this
method has advantage in that design process can be standardized in comparison with the
traditional design method which uses complicated and high order equation for aspherical
lens design.

3.3 Design module for aspherical lens


Fig. 5 shows a design module for an aspherical lens to implement on a PC (personal
Computer) the design process of an aspherical lens using the ray reverse tracing method
suggested in this study. In this module, by entering the index of refraction of a lens, lens
diameter and the curvature of a spherical lens, it is possible to design an idealistic shape of
aspherical lens surface free of spherical aberration through the aberration removal process
of a spherical lens.
And, this module is so structured that even in the case of entering the focal length only
without entering the shape data of spherical lens, it is possible to design aspherical lens that
has required optical performance.

3.4 Generation of an aspherical curvature and coefficients


Generally, in the design of an aspherical lens, following design equation for the
aspherical lens has been used. In this equation, k is defined from the basic shape of a lens,
cr 2
z (r ) = + Ar 4 + Br 6 + Cr 8 + Dr10 .... (4)
1 + 1 − (1 + k )c 2 r 2

r 2 = x2 + y2
c = Vertex curvature
k = Conic constant
A, B, C, D : Deformation term

and design variables are vertex curvature and aspherical coefficients. But, these variables
are floating during the course of designing, and finally decided by optimization. In this
paper, design method for an aspherical lens using the ray reverse tracing method was
suggested, and in consideration of the its correlation with a general design method using
the equation mentioned above. An example of the process for derivation of vertex
curvature and deformation term from the aspherical lens shape data was shown in Fig. 6.
Here, an arbitrary aspherical surface of 15mm radius was used.
Fig. 6/(a) shows the process to find out the curvature of the spherical shape closest to a
given aspherical shape. When the curvature of a spherical shape close to an aspherical
lens shape is plotted, the spherical shape that is the closest to the aspherical shape
corresponds to the one that makes the least area by crossing the aspherical shape, and its
curvature is selected as that of the aspherical shape. Here, the selected curve has a radius of
17.82mm, hence the vertex curvature is 0.0561167.
Fig. 6/(b) shows the process of evaluating the error occurred when the aspherical
surface does not perfectly coincide with the spherical surface. The error shows the
difference in y coordinates at the same x coordinate. And Fig. 6/(c) shows the curve type of
the error when drawn in different y scale. It is possible to generate deformation term by use
of the error data obtained in this procedure. That is, if the error data obtained above are
expressed in the aspherical surface design equation aforementioned, the coefficient for
each term in the equation will become the corresponding coefficient in the aspherical
design equation. But, because the aspherical design equation consists only of even order
terms, using only the coefficients of even order terms as they were in the polynomial
obtained from the error curve is inadequate. Therefore, when a polynomial is made using
the error curve, curve fitting is performed with the polynomial consisting only of even
order terms except odd order terms and constants. Then, the coefficient of each term
Sphere(15.82

Sphere

Aspheric

Sphere

(a) Calculation of basic curvature for (b) Calculation of error curve


aspherical shape

(c) Curve fitting (d) Comparison with aspherical curves


Fig. 6 Generation of aspherical curvature and deformation terms

obtained can be used as deformation term. By applying this process, it was possible to
evaluate the deformation terms as follows.
A= 0.0000399112 B= -0.0000003442 C= 0.0000000006
Fig.6/(d)is the result of comparison between aspherical shape generated from this
process using the deformation terms and the initial basic aspherical shape. In Fig. 6/(d) as
compared with Fig.6/(a) which shows the shape error of spherical shape that has a radius
closest to that of the initial basic aspherical shape :17.82mm, vertex curvature : 0.0561167,
it can be confirmed that an aspherical shape in which almost no shape error exist is
obtained.
It can be concluded that this result shows that it is possible to calculate the deformation
terms for lens design from the shape data of an aspherical lens, and by use of these terms it
is possible to trace design information on an aspherical lens in an arbitrary shape.

4. Conclusion

In this study, as a removal method of spherical aberration that hinders optical lens
performance, ray reverse tracing method that is new and different from the traditional lens
design method was suggested. And it was confirmed that by use of the ray reverse tracing
method, not only removal of spherical aberration that could be found in a spherical lens but
also design of aspherical lens is easy
The conclusion of this study is arranged as follows.
1. It is possible to design an aspherical shape free of spherical aberration by use of the
ray reverse tracing method.
2. In the case of designing an aspherical lens by use of the ray reverse tracing method,
vertex curvature and the deformation terms which have been used as the initial
design data in the traditional design method for an aspherical lens is unnecessary,
and it is possible to design an aspherical shape with a given focal length only.
3. In the design of an aspherical lens using the ray reverse tracing method,
optimization process that is required in the traditional design method can be
omitted, hence the ray reverse tracing method has advantage that the design process
can be standardized.
4. It is possible to calculate the deformation terms used in the lens design from the
shape data of a aspherical lens, and by use of these, it is possible to trace back the
design information about an arbitrary aspherical shape.

Acknowledgement

This study was carried out as a project on commission “development of an intelligent


grinding machine“ managed by KITECH and operated by Ministry of Commerce, Industry
& Energy of the government of Korea.

References

1. T.-S. Song, M.-D. Lee, H.-G. Jun, N.-C. Park, and Y.-P. Park, “Technical Trend of
Optical Information Storage Device” J. of KSPE, Vol. 18, No. 4, pp. 25 - 36, 2001.
2. J. I. Beamonte, "Stability of the spherical aberration up to the fifth order in cemented
doublets", J. Opt. A: Pure Appl. Opt. 2, pp. 161 - 168, 2000.
3. Roman Ilinsky, "Gradient-index meniscus lens free of spherical aberration", J. Opt. A:
Pure Appl. Opt. 2, pp. 449 - 451, 2000.
4. Suzuki, H., Hara, S., and Matsunaga, H., “Study on Aspherical Surface Polishing
Using a Small Rotating Tool,” Journal of the Japan society for precision engineering,
Vol. 59, No. 10, pp. 1713 - 1718, 1993.
5. Itoh, S., “Study on Measurement of Axi Symmetrical Form Generated by Ultra-
Precision Machining(3rd Report),” Journal of the Japan society for precision
engineering, Vol. 61, No. 3, pp. 391 - 395, 1995.
6. J.-S. Lee, M. Saeki, T. Kuriyagawa, K. Syoji, “A Study on the Mirror Grinding for
Mold of a Small Aspherical Lens,” J. of KSPE, Vol. 18, No. 12, pp. 82 - 87, 2001.
7. Erick Kreifeldt, “DARPA Turns Researchers Loose On New Class Of Optics,” Optics
and photonics News, Vol. 8, No. 1, pp. 6 - 10, 1997.
8. Hecht, Eugene., "Optics", Addison-Wesley Pub. Co., 2nd ed., pp. 211 - 239, 1987.
9. Jesus Atencia and Manuel Quintanilla, "Ray tracing for holographic optical element
recording with non-spherical waves", J. Opt. A: Pure Appl. Opt. 3 pp. 387 - 397, 2001.