IMPROVED SCREEN FOR REAR PROJECTION VIEWERS

• Document Type: CREST
• Collection: NGA Records (Formerly NIMA)
• Document Number (FOIA) /ESDN (CREST): CIA-RDP79B00873A002000010084-0
• Release Decision: RIPPUB
• Original Classification: C
• Document Page Count: 13
• Document Creation Date: December 28, 2016
• Document Release Date: November 8, 2012
• Sequence Number: 84
• Publication Date: May 27, 1966
• Content Type: REPORT

Declassified in Part - Sanitized Copy Approved for Release 2012/11/08: CIA-RDP79B00873AO02000010084-0 '5X1
CORNING GLASS WORKS
ELECTRO-OPTICS LABORATORY
RALEIGH, NORTH CAROLINA
Technical Report No. - 10a
Date - May 27, 1966
Period Covered - April 4, 1966
to
May 27, 1966
EXCLUDED FROM A UT, Q ATIC
RERRAMIG, MIT) IR 5200.10
OCTANT CONTAINS INF04:MATU3ri XFFVCT1% 6.
%At rkT-E": E F T:`E U +IiED Tr,i 18 4:"s4z"
4
5.
A4 iVLrz OF THE LR LW Mot L4
?& 7433 ANO T AVMR Tt~ A t!NA01 MV 0
S W tIC 4 U tM~'6 a+
1% VRwjV':rw VV LAt.
EOHF~DuTW
Declassified in Part - Sanitized Copy Approved for Release 2012/11/08: CIA-RDP79B00873AO02000010084-0
 Declassified in Part - Sanitized Copy Approved for Release 2012/11/08: CIA-RDP79B00873A002000010084-0
Technical Report No. 10
I. Materials Investigations
A program to obtain samples of a different class of
materials, consisting of crystalline metallic oxides
in a matrix of translucent CoramR alumina, was
initiated. In these materials the particle size,
number density, and relative refractive index can
be carefully controlled. Samples of CoramR alumina,
which contain magnesium oxide particles, have already
been obtained.
Other samples, consisting of both glass-ceramics and
Fotoform R glasses, are due from our manufacturing
facilities in Corning, New York. These will be
evaluated as soon as possible and the results repor-
ted. From an analysis of these materials a new set
of specifications will be formulated and forwarded
for fabrication of improved samples.
Some time in the next period will be spent in evaluating
competitive rear projection screen materials. It is
important that we measure these other materials using
our instrumentation as each different piece of equip-
ment has different instrument factors which influence
the data taken. This often makes a comparison of the
data taken on two different instruments invalid; and,
furthermore, it is sometimes difficult or impossible to
correct the two sets of data for these instrument
factors.
II. Theoretical Investigations
To make our theoretical work and subsequent experimental
data as useful as possible, it is important that the angular
Declassified in Part - Sanitized Copy Approved for Release 2012/11/08: CIA-RDP79B00873A002000010084-0
 Declassified in Part - Sanitized Copy Approved for Release 2012/11/08: CIA-RDP79B00873AO02000010084-0
K 41' L
scattering functions be normalized to yield the angular
gain distribution function in its most meaningful form.
There are two possible ways in which to normalize the
angular scattering functions, and both use the same perfect
isotropic scatterer as the reference to which they are
normalized. This problem has been discussed in the open
literature1-8 .
First, Gain1 (d), can be defined as the ratio of power con-
tained within a given solid angle, from an experimental
screen to that from one composed of perfect isotropic
scatterers) provided there is the same amount of power
radiated into the forward hemisphere. Mathematically this
can be stated as
I(6,o)A6,Lo/jj. I(e,0)de,do
r~;n (Q nA _n(,) = hemisphere (i)
I0(6,0')Le,Lo/-(j ~(9de,do
fihemisphere
where I(6,0) and I0(6,0) are the angular scattering function
of the experimental screen and the isotropic screen respectively.
The denominator of both the top and bottom of equation (1)
represents the power scattered into the forward hemisphere.
The second, and more widely used approach, normalizes to the
total incident power,
Gain2 (e,O4e,Lo) =
I(e,~)Le I(e,c5)de,do
sphere
10(e,0)4e,AO/S Io(e,0)de,do
sp ere
Generally the scattering function is measured in the 36 = o
plane. For our application there is rotational symmetry,; hence,
I(e,0) = I(0,0) ., (3)
Now let 0 = o, choose a value for pO, and consider the gain
function only dependent on e. Clearly the integral in (2),
over the sphere,is constant, and is independent of the
Declassified in Part - Sanitized Copy Approved for Release 2012/11/08: CIA-RDP79B00873AO02000010084-0
 Declassified in Part - Sanitized Copy Approved for Release 2012/11/08: CIA-RDP79B00873AO02000010084-0
3
scattering function, provided there is no absorption; thus,
$ Gain2 (e) de = K2 (4)
If absorption exists;. the effective transmittance is
just
I Gain2 (6) do
(5)
2
When comparing two different samples using Gain 2(e),,not only
can the relative shapes of the scattering functions be seen
but there is also a direct indication of their relative
efficiencies, Figure 1. The fraction of light scattered
6
Figure 1. Comparison of two gain curves
using Gain 2(e)
into either hemisphere can be found by taking the ratios of
the integral of Gain2 (e ), over the proper hemisphere,.to the
integral over the full sphere.
If one chooses to use Gain1 (e , 0) instead of Gain2 (0)/,only
a comparison of the relative shape of the scattering function
is possible. This can be seen by comparing two different
rear projection screens. Let them both have scattering
Declassified in Part - Sanitized Copy Approved for Release 2012/11/08: CIA-RDP79B00873AO02000010084-0
 Declassified in Part - Sanitized Copy Approved for Release 2012/11/08: CIA-RDP79B00873AO02000010084-0
,~ pp
functions of the same shape in the forward hemisphere, and
also let only screen #1 have some backscattering. The
function Gain 1(0) would show the two screens as having
identical scattering properties, as it only has meaning to
0 = i-r/2. Gain 2(0) would show the shape in the forward
hemisphere to be identical; in addition, it would show
that more incident power was required to get the same amount
of power into the forward hemisphere of each screen. The
ratio of the amount of input power, needed to scatter the
same amount of power into the forward hemisphere, could
also be computed from Gain 2(e).
Assume now, the shapes of both scattering functions are
identical over all angles, and let screen #1 have some neutral
density. Again, the Gain1(6) curve would show both screens
as being identical, while Gain 2(0), besides showing the shapes
to be identical, would also indicate that screen #1 is less
efficient than the other; and an analysis of the function
Gain 2(0) would indicate by how much.
By now it should be clear that Gain 2(0) is by far a more
complete measure of the operational performance of a rear
projection screen than is Gain 1(g). Gain 2(6) was used in
computing the various data given in the Phase II Summary
Report.
One additional word should be said about the isotropic, or
as it is sometimes referred to, the Lambertian scatterer.
By definition,it radiates uniformly in all directions. If
we limit the size of a Lambertian screen, composed of
Lambertian scatterers, by means of an opaque plate with a
hole in it; we find the subtended area changes as cosine Ty,
Figure 2. This is equivalent to illuminating an area on a
rear projection screen which is small compared to the angular
field of view of the detector. The cosine 'Y term must be used
to correct for the obliquity of the detector. On the other
hand, if the field of view is held constant and is always
filled by the screen, no angular correction is necessary.
'5 1
IT pr4V-F~
Declassified in Part - Sanitized Copy Approved for Release 2012/11/08: CIA-RDP79B00873AO02000010084-0
 Declassified in Part - Sanitized Copy Approved for Release 2012/11/08: CIA-RDP79B00873AO02000010084-0
-5-
Figure 2. Geometry of the cosine correction
for oblique viewing
III. Instrumentation
A. Goniophotometer
All mechanical and electrical work, including minor
modifications, is complete. All of the optics in
the collimator, and the detector have been aligned
and the two sample holders have been adjusted to hold
the samples normal to the incident beam. The angular
resolution is better than ?10, which is set by play in
the various gear trains, by the electronics of the
servo system, by the collimation of the incident beam
and by the size of the apertures in the detector unit.
To ensure as much signal as possible, the apertures in
both the collimator and detector should be as large
as can be tolerated. This is determined by measuring
the scattering function of a typical screen using larger
and larger apertures in the detector unit. At first no
observable difference of the scattering functions will
Declassified in Part - Sanitized Copy Approved for Release 2012/11/08: CIA-RDP79B00873AO02000010084-0
 Declassified in Part - Sanitized Copy Approved for Release 2012/11/08: CIA-RDP79B00873AO02000010084-0
-6-
seen. Finally an aperture size will be reached where
be
the measured scattering function is significantly
different from the others previously measured. This
then is the first aperture which is too largewhence,
the next smaller aperture is optimum in that it gives the
best angular resolution and at the same time the most
signal power. This aperture now sets the angular
resolution of the instrument?and all other apertures
should be optimized to give the same resolution.
This optimization of the goniophotometer is being con-
cluded. Calibration and alignment procedures are being
established. This is to ensure all of the data taken with
it will have some common scale factor, thus permitting
data, taken weeks apart, to be validly compared.
B. Modulation Transfer Function Analyzer
Construction of the MTF control console, including the
control electronics,is finished; all that remains is the
final adjusting of the motor speed controls. The rotating
polarizer mount and film transport have also been com-
pleted and checked out.
Some preliminary MTF data on a rear projection screen
have already been taken. A Fresnel zone plate made by
interferometric techniques was used as the variable sine-
wave pattern, Figure 3, and used in the optical system
shown in Figure 4. Information on contrasts in the image
with and without the screen were displayed on an
oscilloscope and photographed. These data were then
reduced to contrasts as a function of spatial frequency
and subsequently to values for the MTF of the screen,
Figure 5. This work was done primarily to check again.:
our concept of the MTF analyzer. We found an oscilloscope
to be quite useful for preliminary alignment, as only a
few seconds are required for each data scan. However,
because of the reduction in size of the display, con-
I in Part - Sanitized Copy Approved for Release 2012/11/08: CIA-RDP79B00873AO02000010084-0
 Declassified in Part - Sanitized Copy Approved for Release 2012/11/08: CIA-RDP79B00873A002000010084-0
U
CONFIDENTI&
Figure 3. Fresnel zone plate type of
sine-wave resolution target
Light Source
Fresnel Mask in Rear Projection
Transport
Projection
Lens
To Oscilloscope
Figure 4. Optical system used to measure the MTF
of a rear projection screen
[ONFIDENTIp(
Declassified in Part - Sanitized Copy Approved for Release 2012/11/08: CIA-RDP79B00873A002000010084-0
 Declassified in Part - Sanitized Copy Approved for Release 2012/11/08: CIA-RDP79B00873AO02000010084-0
,
I ~:
3r 1 IENi u / I I--I_ ! ~Ll~ii~l^
I
_.
1 :
I a
i + : Y
, I!
[+
MC
AN
i .
. oE,.
.
oil
C-1 Nis
l
m El
y ~ G ?~ ?i
mop
?
?'i~
rr
:
,-
/
may"
X
#4
,
ME
T
M
-T
.
77
N
:I:
* :; Y: ii :I?
t
I::
-T*
t
r
T
144,
46
44-1
i_:
C)
I ....
['I.
,
j
1.
r N-
:ii
+-:
.i .:j\
.{.
~
?T:
} F
I
i+
~
FI
T
.
;
}*=.:
:}.1
}.. .I?_
.
.
Y
t?
Ii
rl
tr?
r
I
...
'?t?
: i :
t
:rl:
I~
.+
r
xr
I
*'
t-r
tj! 7-
:
Lr
4L
:.I
r
44.
lnt~
TT 7
T'
1.4
Declassified in Part - Sanitized Copy Approved for Release 2012/11/08: CIA-RDP79B00873A002000010084-0
 Declassified in Part - Sanitized Copy Approved for Release 2012/11/08: CIA-RDP79B00873AO02000010084-0
siderable information is lost "through-.- uncertainties
in measuring the photograph. Therefore, the x-y
plotter in the main console is definitely required for
the final recording of MTF data. An oscilloscope
display should be considered for initial alignment.
The feasibility of using an interferometer to make sine-
wave masks has been established. This approach is possible
if some curvature of the fringes can be tolerated.
Straight fringes can be obtained; however, the change
of space frequency with position is too small to make
these patterns of any value for the MTF analyzer. They
are very good as constant spatial frequency patterns. Thus-,
the sine-wave masks must be formed in an unequal arm inter-
ferometer using a non-parallel beam of light which produces
the conventional Fresnel zone plate type of mask,
already shown.
Two different approaches will be followed in attempting
to make the special sine-wave mask. In the first, a slit
almost in contact with the film, will be illuminated by
a cylinderical lens, Figure 6.
Parallel Light
From Modulator
Cylindrical Lens
Moving Film
Figure 6. Making the sine-wave by placing a mechanical
slit in contact with a moving emulsion
Declassified in Part - Sanitized Copy Approved for Release 2012/11/08: CIA-RDP79B00873AO02000010084-0
 Declassified in Part - Sanitized Copy Approved for Release 2012/11/08 : CIA-RDP79B00873A002000010084-0
This slit is semi-adjustable down to 10 microns in
diameter. In the second approach, a s.l.;t. image will
be directly projected onto the film by a high quality
lens, Figure 7.
Opal Glass Diffusor
Figure 7. Making the sine-wave mask by projecting
a slit image onto a moving emulsion
The slits for this approach are being made by shadowing
very fine threads of wire on glass, with aluminum, and
also by photographic reduction techniques. These slits
will be projected onto the film at a reduction of about
3:1.
The first part of our effort to make the sine-wave masks
will be concerned with determining the film character-
istics and getting the individual components together and
working as a system. The time requirement to complete
this first phase is expected to be 1.5 weeks. A total
of 4 weeks is scheduled in which to get the system
Declassified in Part - Sanitized Copy Approved for Release 2012/11/08: CIA-RDP79B00873A002000010084-0
 Declassified in Part - Sanitized Copy Approved for Release 2012/11/08: CIA-RDP79B00873AO02000010084-0
II Li'~ d~'?' ytl: _, ~." ~ ~"