Wide-Angle Microwave Lens Design
Wide-Angle Microwave Lens Design
273 611
AD
m/e
UNCLASSIFIED
NOTICE: When government or other drawings, speci-
fications or other data are used for any purpose
other than in connection with a definitely related
government procurement operation, the U. S.
Government thereby incurs no responsibility, nor any
obligation whatsoever; and the fact that the Govern-
ment may have formalated, furnished, or in any way
supplied the said drawings, specifications, or other
data is not to be regarded by implication or other-
wise as in any manner licensing the holder or any
other person or corporation, or conveying any rights
or permission to manufacture, use or sell any
patented invention that may in any way be related
thereto.
AFCRL 62-18
273 611
7 Wide -angie Microwave Lens for
Line Source Applications
W. ROTMAN
R.F. TURNER
JANUARY 1962
PROJECT 4600
TASK 46008
JANUARY 1962
ii
CONTENTS
1. Introduction 1
2. Theory 2
2. 1 Derivation of Design Equations 2
2. 2 Selection of Optimum Parameters 8
2. 3 Lens Contour and Phase Error Calculations 10
3. Experimental Studies 16
3. 1 Wide-Angle Lens Design 16
3. 2 Radiation Characteristics of Lens Model 23
4. Conclusions 27
Appendix A: Lens Contour Calculations 29
TABLES
Table 1. Lens Contour
a= 300, and rParameters
= 0.597 for g = 1.137, 29
Table 2. Lens Contour Parameters for g = 0.90
<0.05>1.200 and a= 300 33
A)g= 0.900(r= 3.70); (b)g= 0.950
r= 1.53); () g= 1.000 (r= 1.00);
j) g= 1.050(r= 0.771); (e) g= 1.100
(r = 0.651); (f) g = 1.150 (r = 0.582);
(g) g= 1.200 (r= 0.541)
Table 3. Normalized Path Length Errors, A I, for
Microwave Lens a= 300 41
(a) g= 0.900; (b) g= 0.950; (c) g= 1.000;
Sg= 1.050; () g = 1.100; () g = 1.150;
(g) g= 1.200
Table 4. Normalized Path Length Errors A for
Microwave Lens a = 300 and g = 1. 137 55
References 56
V
ILLUSTRATIONS
Figure Page
1. Parallel-Plate Microwave Lens 3
2. Ray Trace Diagram for Two-Dimensional Electromag-
netic Lens 4
3. Microwave Lens Parameters 9
cc)G = Lens
4, Microwave 1. 137F; (d) G = (a)
Contours, G = 1. 10F# (.? G = F;
1. 20F 11
5. Path Length Errors in Microwave Lens, (g) G = F;
(.b) G = 1. 10F; (c) = 1. 13T 13
6. Parallel-Plate Lens as Line Source Feed for Reflector 18
7. Experimental Microwave Lens Antenna (Front View) 21
8. Experimental Microwave Lens Antenna (Rear View) 22
9. H-Plane Radiation Patterns of Microwave Lens Antenna,
(a) 0 = 00 (On-Axis); (b_)0 = 150; (g) 0 = 300 24
10. Microwave Lens Antenna with Phase Shifters in Lens
Elements 25
11. Radiation Patterns of Microwave Lens Antenna with
Phase Shifters in Lens Elements 26
vii
WIDE-ANGLE MICROWAVE LENS FOR
LINE SOURCE APPLICATIONS
1. INTRODUCTION
The wide-angle scanning characteristics of two-dimensional micro-
wave lenses have been extensively investigated 1 and applied to the design
of radar antennas. Ruze 2 has shown, for example, that constrained
lenses of the no-second-order type are capable of generating 10 beam-
widths and of scanning these beams over angles as great as one hundred
beamwldths. The no- second- order designation refers to a lens with two
perfect off-axis symmetrical focal points and an on-axis focal point for
which the second- order phase deviation is zero. For no-second-order
lenses both front and back lens faces are curved.
Ruze also discusses the design of a straight-front-face lens re-
quired for line source applications as the primary illuminator for a para-
bolic cylindrical reflector or as the feed for a rectangular planar array.
This straight-face lens has excellent scanning characteristics since both
second and third order coma may be almost eliminated by proper de-
focussing. It has two perfect, symmetrical, off-axis focal points and a
highly corrected on-axis focal point. For very narrow beam antennas,
however, its higher order coma aberrations may still be objectiunable.
A further improvement in coma aberrations may be achieved by
applying general lens design principles, developed by Gent et al. 3, 4
to the special case of the straight-front-face lens. This results in a
lens design with three perfect focal points, two symmetrically located
off-axis and one on-axis. It is the purpose of this.paper to obtain the
design equations for the improved straight-front-face lens from Gent's
generalized equations, to evaluate its phase aberrations and scanning
capabilities, and to demonstrate fabrication techniques applicable to this
type of design.
2. THEORY
2. 1 Derivation of Design Equations
A two-dimensional schematic representation of the straight-front-
face lens is shown in Fig. 1. The basic external difference
between this lens and Ruze's design is that the lens elements are
lengths of coaxial transmission line, rather than waveguide. This
permits the connection of arbitrary points on the front and rear surface
of the lens so that corresponding front and rear surface distances,
N and Y, for a single lens element, are not necessarily equal (as they
are in Ruze's design). This additional degree of freedom permits
specifying four independent conditions to determine the lens uniquely,
rather than the three conditions that were available to Ruze. In the
present lens design, these conditions were selected as the straight-
front face, the two symmetrical off-axis focal points, and an on-axis
focus. The ends of the coaxial cables that form the transmission line
elements of the microwave lens are connected directly to a straight
line of radiators to form a line source.
The formulation of the lens design equations and notation follows
3
that of Gent. In Fig. 2, the lens surfaces are shown two-dimensionally
by the cross sections Z 1 and Z 2 . The first contour, Z I, determines
the position of the probe transitions between the parallel plates and the
coaxial cables. The second contour, 2 2' is straight and defined by the
location of the radiating elements that comprise the line source. Corres-
ponding elements on contours Z, and Z2 are connected by a transmisson
line TL.
The contour Z1 is defined by the two coordinates (X, Y) that are
measured relative to a point 01 on the central axis of the lens. Points
on the straight contour Z2 are similarly determined by the single co-
ordinate N, measured relative to the point 02 on the axis. The points
01 and 02 lie on contours Z and Z2 respectively and are connected by
a transmission line TL0 of electrical length Wo . The point P, defined
by the coordinates X and Y, is a typical probe element in Z2i and is con-
nected to point Q, which lies on Z 2 and is defined by the coordinate N,
by the transmission line TL of electrical length W. The three quantities
X, Y, and W can be chosen at will; thus this straight-front-face lens has
3
FCACLE
TOP VIEW FC
INPUT PARALLEL
HORN PLATES
SECTION A -A
FIG. 1. Parallel-plate microwave lens.
4
P(XY)~~ I 21C
c~(-Fcosa,Fsina) electrical K
/Section of
wovef rant
FIGtrce
2.Ra iaramfortw-diensoLelcrmgtilns
5
three degrees of freedom. Other types of lenses, (including Ruze's
design in which Y= N) have at least one less degree of freedom.
Actual values for the three degrees of freedom will now be
selected to obtain wide-angle scanning characteristics. These design
parameters include (Fig. 2) two symmetrical off-axis focal points,
F 1 and F2, and one on-axis focal point, G, having coordinates (-F cos a.,
F sin a), (-F cos a, -F sin a), and (G,O) respectively relative to the point 0.
A ray through the lens at the origin is represented by F 1 0 1 0 2 M and
F 1 PQK represents any other typical ray.
The lens is now designed so that the three focal points F 1 , F 2 ,
and G give perfectly collimated beams of radiation at angles to the axis
of - a, + a, and 00 respectively.
In our special case Gent's 3 equations for the optical path-length
conditions for path-length equality between a general ray and the ray
through the origin are:
(F 2 P)+W-Nsinc=F+W o , (2)
and
(G 1 P)+W=G+ W o , (3)
where
(F 1 P) 2 =F 2 +X 2 +y 2 +2FXcos a-2FYsina 3 (1)
(G 1 P) 2 (G+ X) 2 +y
2 (3a)
and
W-W
W= F g=G/F.
6
Also
ao =cos a, b,= sin a.
Equations (1a) to (3a) may then be written
(F1 X) 2 2
F 2
1+x 2+y 2+ 2ax - 2 boy,
F2 - (M)
(F 22P) 2 2
F +x +y2 + 2aox + 2boy,
(2D)
and
FGP2 =(g +x) 2 +yy2 (3Db)
F2 2
(F 1 P) 22 2
F2 F =(1 -w-bo1 ) 2
- &-v-.)
Since the off-axis focal points are symmetrically located, the lens
surfaces must also be symmetrical about the center axis. This means
that, if q is replaced by - 11 and y by -y3 Eq. (lc) remains unchanged.
Equation (1c) can therefore be separated into two independent equations;
one contains only odd powers of y and il while the other contains the
remaining terms. Thus,
- 2b o 11 + 2bow= - 2boy
or (4)
y = (1-w).
-1
7
Also
x 2 + 2 + 2aox = w 2 + bo2 I22w. (5)
Equations (3) and (3b), relating to the on-axis focus, may likewise
be written
Equations (5) and (6) can be combined to give the following re-
lation between w and 11
aw2+ bw + c 00 (7)
where
a= - _ io
c= 22
1bo -b1 2 2]
ao -4(g - ad T)
This completes the solution for the lens design. For fixed values
_G 1+c2
1+ 2" (8)
-6
.£
1
ic a
L-
/ 0)-
cr 04
Li /
4
/0,j~ 04
xL T
10
the difference in path length between the central ray and any other ray)
may be shown to be:
where
AL =path length error,
h H/ F = normalized distance from point on focal arc to origin
01 of surface Z1 . H is determined from the triangle with
sides R, H, and G-R and with included angle (Fig. 3).
0 = angle between central axis and point on focal arc.
R = radius of focal arc (determined by the three points G, F, and
F 2 on the arc).
2.3 Lens Contour and Phase Error Calculations
The preceding section indicated that an optimum set of lens para-
meters exists, in the sense that they provide minimum phase aberrations
over a prescribed range of angles. This selection of parameters assumes
that the correction for second and third order coma terms also results
in the minimization of the higher order aberrations. Since this is not
obvious, an investigation was conducted to determine the phase errors
in lenses of this type for parameters that may differ from the optimum
value.
The total scan angle, 2a, for the lens was fixed at 600 (a= 300),
as representative of a wide-angle lens system. From Eq. (8), the
optimum on-axis focal point is given by g= 1. 137. Accordingly, calcu-
lations were made on a digital computer of the normalized lens contour
parameters, y, x, and w and also of the normalized phase error, A4
from Eqs. (4), (6), (7), and (9) for the following range of values:
g= 0.90 <0. 05 >1. 20, and 1.137.
0= ± 50 ±150 ±25o ±35%
r = 0 <0.25 > 0. 80.
For = 00 and±300, At, is zero.
Selected lens contour curves are shown in Figs. 4a to d for
g = 1.00, 1. 10, 1. 137, and 1. 20. Their tabular values are given in
170Y 17,y
.5 .5
4*.4 .4
.3 -3
FF
On-ais pirft . 0.
.,ffa point0.4
x G~ 0 -.--
-12 -1.1 -1.0 -.9 -8 -. 6 n5 -4 -3I 2n -. -1-2 -11 -1.0 -.8 -.8 -.7 6- -4 -'
surface .6r .
Contur for .
7 y lens
Ft. .6
F,(Ruzis design)
5
-4 .
F .3 F
0.l OF0.
x - -tl -1.0 -.8 .8 -.7 -. 1I n -3
-62 5i - ~ 0O -.
1.65.0 -. 4F
It 1.0 -5~ e 7 .6 -.5 -.4 .3 -2 -.1
0
-1
-.2
F -3F
-4.
Fe Ire5
(c)
Appendix A for g = 0.90 <0.05> 1.20 and 1.137. The light lines
between the rear and front lens contours indicate corresponding
values of y and i which are the junction points for the coaxial lens
elements.
A comparison is made in Fig. 4c between the rear lens con-
tour for the Gent (y ? i) and the Ruze (y = 11) designs for the opti-
mized value of g = 1. 137. In the latter case the straight-front-face
lens equations are
222
x 2 +ao y 2 + 2aox =0, (10)
'
j . __ , __ ,0
I'
"i - ______ - II
j ,. Q....
/ . ,4I
__ __ / __
! I I " , I +"
_____ ____ -; A
14
metry with respect to y] while those for g = 1. 10 and g = 1. 137 have odd
symmetry. (The path length error curves show only positive values
for e while I takes on positive and negative values. Alternatively,
both positive and negative values of e could be used with only positive
values of Tb)
The minimum beamwidth obtainable from a microwave lens is
determined by the size of the aperture, the operating wavelength, and
the maximum permissible phase or path-length aberrations. The
beamwidth for a rectangular aperture antenna with a cosine illumina-
tion taper and 23-db sidelobes (typical of the usual antenna practice) is
given by
HPBW=690x (11)
D
where
H P B W is the half-power beamwidth,
Ji 'max 0. 55,
G/ -
G/2max = . 035,
-30°<0 <+ 30 ° ,
and
(A)max = 0.00013 (from Fig. 5c and Appendix B.
Then
(HP BW)min = 0.0650 (For 0 =00)
=0.0750 (For 0 =300).
Thus, we can scan a beam of less than 0. 10 over 600 for a total scan
angle of greater than 600 beamwidths.
As a second example, we will try to minimize the ratio G/D
(focal length to aperture ratio) at the expense of a somewhat greater
minimum beamwidth.
The following parameters are therefore chosen:
g=1. 100,
()max = 0.0005,
-300 <0<+300,
max w=0.80
G/D =0.687(For 0=00),
and
<HPBW)mtn=0.l 8 0(0 =00)
=0.210 (0 = 300).
For both these examples, the scan angle 0 can be extended to
+35 0 with very little deterioration of radiation pattern. Note, also,
that defocussing from the assumed circular focal arc to some other
noncircular contour that also passes through F 1 , F 2 , and G does not
accomplish much in reducing coma aberrations (except for the case
16
of g = 1.00) since defocussing primarily affects the even- order aber-
rations whereas the residual phase aberrations are primarily of odd
order.
The g = 1. 137 lens contour is thus optimum in the sense that it
minimizes the obtainable beamwidth for a reasonable F/D ratio. The
value of g = 1. 10 seems more suitable, however, if the F/D ratio must
be minimized and if the permissible beamwidth is greater than 1/40.
Either of these two designs should be equally applicable, however, to
the majority of lens design problems.
3. EXPERIMENTAL STUDIES
3. 1 Wide-Angle Lens Design
The analysis of the previous sections has shown that the design of
wide-angle microwave lenses, with beamwidths on the order of fractions
of a degree, is theoretically feasible. A two-dimensional model of such
a microwave lens was constructed to determine the problems inherent
in this design. Design specifications include the following parameters:
fo = 3.0 Gc (design frequency),
g = 1.137,
lmax = 0. 60,
D/% =18,
emax = 300,
HPBW = 3 0 (Nominal Half-Power Beamwidth),
Primary Illuminator- Open- Ended Waveguide Horn,
Lens Elements-RG-9/U Coaxial Cables.
Refer to Fig. 4c for the lens contour and to Fig. 5c for the normalized
path length error At.
The beamwidth for this lens was chosen to keep its physical dimen-
sions within reasonable limits. On the other hand, the theoretical phase
errors inherent in this model are so small that they cannot be detected
by their effect on the radiation pattern or other electrical characteris-
tics of the lens. For example, either the Ruze (y = 11) or Gent design,
for g = 1.00, 1.10 or 1. 137 (Figs. 4 and 5) results in lenses with equiva-
lent electrical performance when the beamwidth is approximately 30 .
17
Our objective is not to obtain an experimental comparison between the
competitive Ruze and Gent lens designs,. but rather to demonstrate
techniques unique to the construction of microwave lenses with variably
spaced coaxial lens elements.
The microwave lens model (Fig. 6) uses the parallel plate and
coaxial TEM modes to obtain maximum bandwidth. Microwave radia-
tion from the primary horn illuminators, located along the focal arc,
propagates between the parallel plates to the reflector-backed probes
that form the rear lens contour. These probes extract the energy from
the parallel-plate region and feed it into the coaxial cable lens elements
which, in turn, excite the probes on the straight-front-lens contour.
These front probes form a linear array between a second set of parallel
plates that radiate into space through a short TEM horn transition. The
lens in Fig. 6 is shown feeding a cylindrical parabolic reflector that
collimates the beam in the elevation plane.
The primary horn illuminator is designed for a prescribed ampli-
tude distribution along the front lens face by selection of its aperture
dimensions and orientation. Its required radiation pattern between the
parallel plates is first determined graphically by ray-tracing, equating
the power radiated per unit length along the front lens face to the angu-
lar sector subtended by this power flow at the horn position on the focal
arc. The aperture and orientation of the horn illuminator are then se-
lected7 to give the closest approximation to the desired primary pattern.
The realizable lens illumination differs somewhat from the theoretical
value because the required primary pattern is not symmetrical (except
f or the on-axis position e = 00) and does not conform exactly to the
physically realizable patterns obtained from a uniformly-phased horn
illuminator. The farfield radiation pattern is, however, not very sensi-
tive to small changes in illumination taper. Since the input horn's
directivity is a function of its position along the focal arc, its dimensions
and orientation depend upon its focal position. Also, the peak of radia-
tion is not, in general, in the direction of the vertex of the lens. For
example, the required horn parameters in the present model vary from
an aperture of 1. 2% (primary HPBW = 41.0) for the on-axis focal point
18
(I))
ww z
WZZ
I-8
-S
o<
z 2
zoLaa. j
J0~~ ir CA:'.
wo
2)
U)
0 4
W
W)
0
m-
z w
19
to 1.45% (primary HPBW = 340) for the 0 = 300 focal point. Since
several stationary horn illuminators are used in the experimental
model, this change in horn dimensions causes no difficulty. If the
beam scanning were accomplished by moving a single input horn along
the focal arc, a compromise aperture dimension (for example, 1. 3%)
would have to be selected.
The dimensions and spacing of the probes must be chosen to
ensure adequate coupling of the lens elements to the parallel plate
structures over a wide range of incident angles. The performance of
each probe is affected by mutual coupling to its neighbors. The prob-
lem of analyzing the behavior of a set of probes located along a curved
contour and excited by a curved wavefront is quite difficult. An approxi-
mate evaluation can be obtained, however, by considering the perform-
ance of an infinitely long linear array of uniformly spaced probes
phased to radiate a plane wave front at an angle, y, normal to the array.
The relation between this angle of radiation and the electrical phasing of
the probes, Yp, is given by
M
0
C)l
z 0
0 w
-J -
IL
00
ClD
I-
col
wr
01 Cou
o
CL
a I1
c-a
w
z 0
Cl)
22 c
0
0 z 0
0 (D
z Q w
o 0)
z w
U))
zw
CL
LLJ -i
a)
I>
LL C'
I-r
ma
ca)
U))
Li-
23
of the line source, which forms the straight front face of the lens,
from the rest of the lens structure.
3.2 Radiation Characteristics of Lens Model
Many radiation patterns of this microwave lens were measured
for different angular positions and sizes of the horn illuminators and
for focussing adjustments over a range of frequencies. Some repre-
sentative patterns are shown in Figs. 9 a to c for a frequency of
3 Gc and for e = 0*, 150, and 300. The horn was located on the focal
arc and had an aperture of 1. 45%. In all cases the position of best
focus was found to be within one inch of the focal arc and was not par-
ticularly critical. The radiation patterns were also measured over
the frequency range of 2. 8 to 3. 2 Gc with no significant deterioration
of performance. The measured sidelobe levels, as high as -18 db,
are somewhat inferior to the -22 db design value. This deviation is
probably caused by a nonoptimum selection of horn aperture size
(which would also account for the measured beamwidth being some-
what smaller than the theoretical value), as well as by constructional
tolerances.
An alternate technique of scanning the antenna beam of a wide-
angle microwave lens is the introduction of a linearly progressive,
time-variable phase delay in the coaxial lens elements. This in turn
causes a progressive phase shift between radiating probes of the
straight front face of the lens and a corresponding shift in the position
of the radiation pattern's peak. Thus, by putting phase shifters in the
coaxial cables of the lens, a beam from a stationary input horn illumi-
nator may be scanned over a limited region of space.
The original concept for the antenna studies described in this
paper was that of a microwave lens that would combine a series of
angularly spaced, stationary input horn illuminators with coaxial phase
shifters in the lens elements. Although the beam generated through the
lens by each horn points in a different direction, all the beams can be
moved simultaneously by varying the phase shifters with progressively
increasing phase. Thus, by spacing the horns at the proper angular
24
Hill i . ,1 1,1
l';.. -- T
-5 -2N
4 !il 1.24 3
Ti i JA
~~IT~U 1.'L V Ij
L.'
4-2 b
-3 -4
. - ' 1- . 6 12-' 1 824 0
j+1~~ 4.FT1
1-l;:, r -H"
1:1 17' r I-
441 7'T 11' III
it It1111 i-
.... . . ~ - X: - ;P 7
. ........ ... ?.
.. -
I:4 Is:10 25
U14-~
FIG. ralatio
9 H plne pattrns o len ane,
Nicrwav
(.a~~~~;
b .10 ~0=3g 0(nAi) 7
25
w
cr.,
w LL
(.
z c,,
IL _ _
-Jn
CO)
L
26
r= (
-F-4 Md
-A*-3["
31jj*
LN
*Q- ~O
N N Y IJO N
i~a)
pip,
r--
OPd
j - -: I,
27
intervals, the entire field of view of the lens may be covered by
several independent beams, each of which scans only a small sector.
This antenna concept was designated as the Multiple-Beam Interval
Scanner (MUBIS) system. Although its development has been delayed
(replaced by a competitive antenna system 8 with a wide-angle micro-
wave lens and coaxial organ-pipe scanners), a study was made of the
action of the coaxial phase shifters in beam scanning. Figure 10
shows the wide-angle microwave lens model, modified by adding
trombone-type coaxial phase shifters (line stretchers) to each of the
coaxial cable lens elements. The phase shifters are adjusted to in-
crease the cable lengths linearly along the face of the lens. A set cf
radiation patterns is shown in Fig. 11 for the case in which the input
horn is located at 300. The line stretchers are adjusted to scan the
beam ± 120 about this central value in 30 steps. Only the region in
the vicinity of the peak of each radiation pattern is shown. It can be
seen that the gain of the antenna remains fairly constant and that the
pattern shape does not deteriorate severely even for these large angles
(420 maximum) from the normal. Similar patterns have been obtained
with the input horn set at 00 and 150.
4. CONCLUSIONS
The design principles presented in this report for two-dimensional
lens structures may also be applied to three-dimensional lenses. The
additional degree of freedom obtained by allowing nonuniform lens-
element spacing (y 9 1), instead of the more conventional uniform
spacing (y = ij), permits the specification of two symmetrical off-axis
and one on-axis focal points for a two-dimensional straight-front-face
lens. In the Ruze (y = 11) design the inner lens contour is uniquely de-
termined by specifying the scan angle a. It is also known 1 that no so-
lution is possible for the design of a three-dimensional, axially sym-
metrical lens if the Ruze constraints are applied. The inner contour
of the Gent (y 3 q) two-dimensional lens design is however determined,
not only by the required scan angle a, but also by the normalized on-
axis focal length, g. Figure 4 shows that the shape of the inner lens
contour depends strongly upon the parameter g.
28
A three-dimensional straight-front-face lens design may be derived
as a figure of revolution of an appropriate two-dimensional lens contour.
The parameter g that controls the inner lens contour should be selected
to minimize the lens' astigmatism, as measured in the principal plane
orthogonal to that of the selected focal points. A three-dimensional
phase-error analysis would be required to assure that aberrations
are within permissible tolerances over the entire lens surface.
It is also possible that the basic Gent lens equations may contain
a solution for the design of a three-dimensional lens with two perfect
and symmetrical off-axis focal points for which both lens surfaces can
be nonplanar. Since these three-dimensional lens designs have not
been extensively investigated under the present study, no numerical
solutions are available.
The analysis presented in this paper has shown that microwave
lenses that are capable of scanning wide-angle sectors with beamwidths
on the order of a fraction of a degree are theoretically feasible and
also practical to construct. These line sources may be used as primary
feeds for parabolic cylindrical reflectors or planar arrays in large
radar antenna systems.
29
Appendix A
Table 1.
Lens Contour Parameters for g = 1.137, .a= 300, and r = 0.597
1 w -x y
0.00 0.00000 0.00000 0.00000
0.01 0.00000 0.00005 0.01000
0.02 0.00002 0.00019 0.02000
30
Table 1. (Contd
w -x y
0.03 0.00004 0.00043 0.03000
0.04 0.00007 0.00077 0.04000
0.05 0.00011 0.00121 0.04999
0.06 0.00016 0.00174 0.05999
0.07 0.00021 0.00237 0.06999
0.08 0.00027 0.00309 0.07998
0.09 0.00034 0.00391 0.08997
0.10 0.00042 0.00483 0.09996
0.11 0.00051 0.00584 0.10994
0.12 0.00060 0.00695 0.11993
0.13 0.00070 0.00815 0.12991
0.14 0.00080 0.00945 0.13989
0.15 0.00091 0.01084 0.14986
0.16 0.00103 0.01233 0.15984
0.17 0.00115 0.01391 0.16980
0.18 0.00127 0.01559 0.17977
0.19 0.00140 0.01736 0.18973
0.20 0.00153 0.01922 0.19969
0.21 0.00166 0.02118 0.20965
0.22 0.00179 0.02323 0.21961
0.23 0.00192 0.02537 0.22956
0.24 0.00205 0.02761 0.23951
0.25 0.00217 0.02993 0.24946
31
Table 1. (ontd)
w -x y
0.26 0.00230 0.03234 0.25940
0.27 0.00241 0.03485 0.26935
0.28 0.00253 0.03744 0.27929
0.29 0.00263 0.04013 0.28924
0.30 0.00273 0.04290 0.29918
0.31 0.00281 0.04575 0.30913
0.32 0.00289 0.04870 0.3190.8
0.33 0.00294 0.05172 0.32903
0. 34 0.00298 0.05484 0. 33899
0.35 0.00301 0.05803 0.34895
0.36 0.00301 0.06130 0.35892
0.37 0.00298 0.06466 0. 36890
0.38 0.00293 0.06809 0.37889
0.39 0.00284 0.07160 0.38890
0.40 0.00273 0.07519 0.39891
0.41 0.00257 0.07884 0.40895
0.42 0.00237 0.08257 0.41900
Table 1. (Contd)
w -x y
0.50 -0.00142 0.11461 0.50071
0.51 -0.00227 0.11883 0.51116
0,52 -0.00325 0.12309 0,52169
0.53 -0.00435 0.12738 0.53231
0,54 -0.00560 0.13168 0.54302
0.55 -0.00701 0.13600 0.55385
0.56 -0.00859 0. 14032 0.56481
0.57 -0.01038 0.14463 0.57592
0.58 -0. 01238 0. 14892 0.58718
0.59 -0.01464 0.15318 0.59864
0.60 -0.01717 0.15739 0.61030
0.61 -0.02001 0.16153 0,62221
0.62 -0.02321 0.16559 0.63439
0.63 -0.02681 0. 16954 0.64689
0.64 -0.03086 0. 17335 0.65975
0.65 -0.03543 0. 17699 0067303
0.66 -0. 04060 0.18042 0.68679
0.67 -0.04645 0. 18359 0.70112
* 0.68 -0.05910 0.18646 0.71611
0.69 -0.06068 0.18895 0. 73187
0.70. -0.06935 0,19097 0.74855
0.71 -0.07932 0.19243 0.76632
0.72 -0.09085. 0,19320 0.78541
0,73 -0. 10427 0.19311 0.80611
33
Table 1. (Contd)
I w -x y
0.74 -0.12000 0.19194 0.82880
0.75 -0.13861 0.18940 0.85395
0.80 -0.3172 0.1349 1.054
Table 2.
Lens Contour Parameters for g = 0.90<0.05>1. 200 and a= 300
w -x y
0.00 0.0000 0.0000 0.0000
0.05 0.0015 0.0028 0.0499
0.10 0.0060 0.0113 0.0994
0.15 0.0136 0.0254 0.1480
0.20 0.0243 0.0451 0.1951
0.25 0.0381 0.0704 0.2405
0.30 0.0551 0.1012 0.2835
0.35 0.0754 0.1375 0.3236
0.40 0.0991 0.1792 0.3604
0.45 0.1263 0.2262 0.3932
0.50 0.1572 0.2785 0.4214
0.55 0.1920 0.3360 0.4444
0.60 0.2308 0.3984 0.4615
0.65 0.2739 0.4659 0.4720
0.70 0.2312 0.5381 0.4752
0.75 0.3727 0.6154 0.4705
0.80 0.4272 0.6983 0.4583
35
Appendix B
Table 3.
Normalized Path Length Errors, Al, for Microwave Lens a= 300
Table 3a g= 0.900
0 = 50 0 = 100 0 = 150 0= 200 0 = 25' 0 = 350 0 = 400
0.00 -0.000000 -0.000000 -0.000000 -0.000000 -0.000000 -0.000000 0.000000
0.05 -0.000003 -0.000008 -0.000014 -0.000017 -0.000014 -0.000030 0.000081
0.10 -0.000015 -0.000039 -0.000065 -0.000080 -0.000065 -0.000135 0.000360
0.15 -0.000042 -0.000106 -0.000172 -0.000204 -0.000164 -0.000337 0.000896
0.20 -0.000092 -0.000223 -0.000352 -0.000415 -0.000329 -0.000668 0.001767
0.25 -0.000173 -0.000408 -0.000632 -0.000737 -0.000580 -0.001168 0.003076
0.30 -0.000296 -0.000682 -0.001044 -0.001207 -0.000945 -0.001892 0.004962
0.35 -0.000473 -0.001074 -0.001630 -0.001876 -0.001464 -0.002914 0.007615
0.40 -0.000721 -0.001622 -0.002450 -0.002811 -0.002191 -0.004345 0.011310
0.45 -0.001064 -0.002380 -0.003587 -0.004116 -0.003210 -0.006351 0.016460
0.50 -0.001534 -0.003425 -0.005167 -0.005946 -0.004650 -0.009192 0.023700
0.55 -0.002173 -0.004872 -0.007389 -0.008557 -0.006728 -0.013300 0.034040
0.60 -0.003066 -0.006907 -0.010580 -0.012390 -0.009839 -0.019450 0.049160
0.65 -0.004307 -0.009836 -0.015340 -0.018330 -0.014790 -0.029120 0.071970
0.70 -0.006077 -0.014210 -0.022870 -0.028300 -0.023510 -0.045350 0.107300
0.75 -0.008639 -0.021020 -0.035870 -0.047920 -0.042440 -0.074340 0.162000
0.80 -0.011520 -0.029640 -0.055880 -0.091820 -0.118800 -0.116600 0.228300
42
Table 3c (Contd)
0 = -35 ° 0 =-400
6 = -50 0 =-100 8 =-150 0 = -200 e =-250
&
47
Table 4.
Normalized Path Length Errors Al, for Microwave Lens - a= 300 and g = 1.137.
0 = -50 6 = -i0o = -15' 0 = -200 = -250 9 -35' 0 = -400
0.00 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000
0.05 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 -0.000001
0.10 0.000000 -0.000001 -0.000001 0.000000 0.000000 0.000000 0.000000
0.15 -0.000001 -0.000002 -0.000002 -0.000002 -0.000001 0.000001 0.000002
0.20 -0.000003 -0.000005 -0.000005 -0.000005 -0.000003 0.000004 0.000009
0.25 -0.000005 -0.000009 -0.000010 -0.000009 -0.000006 0.000008 0.000020
0.30 -0.000008 -0.000013 -0.000016 -0.000015 -0.000010 0.000014 0.000034
0.35 -0.000010 -0.000018 -0.000021 -0.000020 -0.000013 0.000020 0.000048
0.40 -0.000011 -0.000019 -0.000023 -0.000022 -0.000014 0.000023 0.000057
0.45 -0.000007 -0.000012 -0.000016 -0.000015 -0.000010 0.000019 0.000049
0.50 0.000006 0.000009 0.000009 0.000007 0.000003 0.000000 0.000008
0.55 0.000036 0.000059 0.000067 0.000060 0.000037 -0.000047 -0.000098
0.60 0.000095 0.000160 0.000186 0.00018. 0.000106 -0.000145 -0.000322
0.65 0.000210 0.000354 0.000414 0.000378 0.000240 -0.000337 -0.000761
0.70 0.000431 0.000729 0.000855 0.000783 0.000500 -0.000710 -0.001616
0.75 0.000875 0.001483 0.001743 0.001601 0.001024 -0.001437 -0.003355
0.80 0.001883 0.003205 0.003781 0.003485 0.002237 -0.003226 -0.007403
RE FERENCES
1. Final Engineering Report on Investigation of Variable Index of
Refraction Lenses, Sperry Engineering Report No. 5224-1233,
Signal Corps Contract No. DA 33-039-sc-15323, Sperry
Gyroscope Co. Great Neck, New York, September 1952.
2. J. RUZE, Wide-Angle Metal-Plate Optics, Proc. IRE 38:53-58,
January 1950; Also CFS Report No. E5043, Cambridge Field
Station, Cambridge, Mass., March 1949.
3. S.S.D. JONES, H. GENT, and A. A.L. BROWNE, Improvements
In or Relating to Electromagnetic-Wave Lens and Mirror Systems,
British Provisional Patent Specification No. 25923/53, August 1953.
4)~ ~~
~ ~)4U
4
L d ~ .. ) 4)
4
.. 4
C* ) U z~I
zJ2
z)
11 o ~ ~ 1 4
cd~ .8 0 ,4 0
'93
0 ! o 2
V
Umu
i ii Iv 0 0 d2A
00 ., 4b
O)vc A H.M,9
>4)
k$
o~.U: m~o 2t
0~4)D ~
0~0 ~ $4~.- 0
0. 4k 4 AC 00 ' 4
uz 4 " w : o g .. g 0'-
104 U
p
A 0 4)0.4 w 0 W-2
tO M uU
0$4-5
0 ~ ~
Cd', RA . 2-0 I - 10
rQ
z: z z
Eccd
,* 02
- 4 Of 4)
0 *;
~G)Go
* 02C
0n
00 0
0) 4 0.
Uw W~o
d0
.- W 1
0 00 C
4, .0-,
0~k 00~
ED '0
0 Od W,
4N a~' v 2
C.)
(5z
'-4 0) 0-
Cd' 00O 4 )>C
4) 00 - .440U V0~
" 500 0 0d
0 W wiio W o - j I
4). 4)
00 *W 80 0
0u Q -0,o 2 ,I iH ilI ..
c rc-WE,.. E-0000.M
'a -.
l Uo
00
S )U
....... °...............................................................................
.8
'I
•
i i