HIGH PRECISION STEREO COMPARATOR STUDY MONTHLY NARRATIVE REPORT FOR PERIOD ENDING OCTOBER 1, 1965

• Document Type: CREST
• Collection: NGA Records (Formerly NIMA)
• Document Number (FOIA) /ESDN (CREST): CIA-RDP78B04747A002700040006-3
• Release Decision: RIPPUB
• Original Classification: K
• Document Page Count: 123
• Document Creation Date: December 28, 2016
• Document Release Date: May 6, 2003
• Sequence Number: 6
• Content Type: REPORT

Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
HIGH PRECISION STEREO COMPARATOR
STUDY
MONTHLY NARRATIVE REPORT
FOR PERIOD ENDING OCTOBER 1, 1965
Declass Review by NIMA / DoD
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
The first five sections of this report
summarize the effort and conclusions
reached during the reporting period.
The appendices provide detailed in-
formation.and analyses to justify the
conclusion. It is expected that the
appendices in this and other monthly
summaries will furnish the basis for
the final engineering report.
Approved For Release 2003/05/15 : 6OA-RDP78BO4747AO02700040006-3
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
TABLE OF CONTENTS.
Section
FOREWORD
1.0 Current Status of Work
1.1 Optical System
1.2 Mechanical
1.3 Correlation
1.3.1 Electronic Analog Correlator
1.3.2 Electronic Digital Correlator
.1.3.3 Electro-optical Correlator
1.4 Metering System
1.5 General System Considerations
1.5.1 Overall'Format Observation
2.0 Problem Areas
2.1 Optical
2.2 Mechanical
2:3 Correlation
2.4 Metering Systems
2.5 General Systems
3.0 Projected Work for Next Reporting Period.
3.1 Optical
3.2 Mechanical
3.3 Correlation
3?4 Metering
4.0 Expenditures
5.0 Verbal Commitments and Agreements
Page
4-1
5-1
iii
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
TABLE OF CONTENTS
Section
P
age
Appendix A Condenser Subsystem
A-1
Appendix B Objective and Intermediate Lenses
B-1
Appendix C Reticle Dot Subsystem
C-1
C 1 General Arrangement and.Design
C-1
Considerations
C:2 Reticle Dot Optical Analysis
C-7
Appendix D Optical Correlation
D-1
D 1 Theory
of Operation
D-1
D 2 Labora
tory Models Implemented
D-1
D 3 Perform
ance and Data
D-10
D 4 Limitat
ions of Apparatus and
D-19
Propose
D5 Capabil
d Improvements
ities
D-22
D 5.1
Metrological
D-22
D 5.2
Photometric
D-22
D 5.3
Viewing
D-22
D 5.4
Alignment Function
D-23
D 5.5
Scale Adjustments
D-23
D. 6
D-23
Sp
ot Position Sensor
D 7 Summary
of Features and Capabilities
D-27
Appendix E Analog Correlator
E 1 Theoretical Justification
E 2 Hardware Status
E 2.1 Phase Shift Network
E. 2.2 Correlator
E 2.3 Amplifiers and Phase Split
Amplifiers
E 2.4 Subtractor
E 3 Conclusion
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
iv
E-1
E-1
E -4
E-6
E-6
E-12
E-12
E-12
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
TABLE OF CONTENTS
Section Page
Appendix F High Precision Stereo Comparator Metering F-1
Systems
F 1
F 1.1 Description
F 1.2 Discussion
F 1.3 Conclusions
F 2 L____~etering System
F 2.1 Description
F 2.2 Discussion
F 3
STAT
F-1
F-1
F-2
F-3
F-4 STAT
F -4
F-7
F-8 STAT
F 3.1 Description F-8
F 3.2 Discussion F-10
F 4 Metering System Selection F-10
Appendix.G Material Selection Criteria for a High G-1
Precision Measuring Machine'
G 1 Introduction
G 2 Meehanite Metals
G 3 Granite
G 4 Casting Alloy
G 5 Conclusions
C-1
G-1
G-3
G-5
G-6
STAT
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
 Approved For Release 2003/05/15 : CIA-RDP78B04747A002700040006-3
LIST OF ILLUSTRATIONS
Figure
Page
A-1
Basic Scheme of Koehler Illumination
A-2
A-2
VOS Illumination System
A-4
A-3
Flying Spot Scanner Light Collection System
A-11
A-4
VOS Condenser & FSS Collector, Mechanical
Arrangement
A-13
Table A-1
Optical Components for Condenser Subassembly
A-15
B-1
Visual Observation System, Input Station
B-19
Table B-1
Optical Parameters for Input Side of Visual
Observation System, Design Approach 1
B-2
B-2
Lens Specifications, Design Approach 2
B-3
B-3
Optical Parameters for Input Side of Visual
Observation System, Design Approach 2
B-4
B-4
Optical Parameters for Input Side of Visual
Observation System, Design Approach 3
B-5
B-5
Optical Parameters for Input Side of Visual
Observation System, Design Approach 4
B-7
B-6
Optical Parameters for Input Side of Visual
Observation System, Design Approach 5
B-11
B-7
Lens Specification, Design Approach 5
B-12
B-8
Optical Parameters for Input Side of Visual
Observation System, Design Approach 6
B-14
B-9
Optical Parameters for Input Side of Visual
Observation System, Design Approach 7
B-16
C-1
Reticle Dot Optical System
C-4
C-2
Optical Stereo Micrometer
c-8
vi
Approved For Release 2003/05/15 : CIA-RDP78B04747A002700040006-3
 Approved For Releff0ff 5/11L5LPj04747A002700040006-3
Figure
Page
D-1 Basic Correlator Configuration
D-2
D-2 Optical Correlator Scheme: Transparency
Versus Transparency
D-4
D-3 Two Transparency Equipment
D-5
D-4 Optical Correlator Scheme: CRT to Transparency
D-7
D-5 Equipment Assembly, First CRT to Transparency
Correlator
D-8
D-6 CRT to Transparency Correlator
D-9
D-7 Optical Correlator CRT Vs. Transparency,
with Image Intensifier
D-11
D-8 General Purpose Research Flying Spot
Scanner and Display Apparatus
D-12
D-9 General Purpose Optical Correlator
D-13
D-10 General Purpose Optical Correlator Assembly
D-14
D-11 Dither Output Vs. Displacement of Imagery _
D-15
D-12 Samples of Imagery Studied
D-17
D-13 Dither Output Vs. Displacement
D-18
of Imagery Quotient-Match
D-14
Ster
eo Pair Correlation: CRT to Film
D-20 STAT
D-l5
n
Performance Test Equipment
D-24
E-1
Electronic Analog Correlator System
E-5
E-2
Phase Shift in Degrees
E-7
E-3
Phase Shift and Phase Output with Respect to
Input Signal for L-C Phase Shift Network
E-8
E-4
Ratio of E ut (Phase A and Phase B to E?n
Versus FreQuency for L-C Phase Shift Ne work
E-9
E-5
90 Degree Phase Shift Network
E-10
E-6
Correlation
E-11
E-7
Correlator Output
E-13
E-8
Phase Inverter and Amplifier - Summing Amplifier
E-14
Table F-1
Metering System Comparison
F-12
Approved For Release 2003/05/15 : CIA-RDP78B04747A002700040006-3
vii
 Approved For Release 2003/05/15 : CIA-RDP78B04747A002700040006-3
1.0 CURRENT STATUS OF WORK
1.1 optical System
The definition of the visual observation system is now complete
through the light source, condenser optics, objective turret,
image rotator, and reticle dot, except for establishment of some
of the alignment and positioning tolerances. The method of
integrating a scanner system (for automatic correlation),. has also
been defined and the light-collecting optics for a photomultiplier
have been integrated with the condenser system.
The work on the optical switch which had been scheduled for
completion in September was postponed in order to complete the
above-mentioned subsystems,which have a much greater impact on
the mechanical design.
Detailed descriptions of the optical subsystems are contained in
Appendices A, B, and C.
1.2 Mechanical
a Configuration design of the X-Y measuring stages and optics framework
proceeded during the reporting period. Stress and deflection calcu-
lations are now in progress to determine the basic sections sizes.
Layout of the objective turret, image rotator and reticle dot
assembly was started during the reporting period. It is expected
to be complete early in November.
STAT
Approved For Release 2003/05/1 : CIA-RDP78B04747A002700040006-3
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
A final material selection for the precision stages has not yet
been made. Appendix G contains a discussion of the three most
likely materials: meehanite cast iron, 410 stainless steel, and
granite. In the case of meehanite a problem of corrosion protection
exists. It appears that either a chrome flash or an electroless
nickel deposit may be suitable if build-up can be controlled, and
vendors are being contacted to determine their capability for
handling the anticipated large castings.
1.3 Correlation
1.3.1 Electronic Analog Correlator
The circuitry for the analog correlation is now complete and has
been tested on simulated input signals. In particular,the ability
to show degree of correlation between sine waves of different phase
throughout the design bandwidth (50 KC to 5 MC) has been demonstra-
ted. A description of the theory of operation and circuit design
is contained in Appendix E.
1.3.2 Electronic Digital Correlator
As previously reported, tests of this correlator are scheduled to
start in October.
1.3.3 Electro-Optical Correlator
The equipment modifications described in the previous report have
proceeded. A full description of this correlation system is con-
tained in Appendix D.
1.4 Metering System
STAT
e emons ra on ver e t at the system repeatability was as
Approved For Release 2003/15/13 2CIA-RDP78BO4747AO02700040006-3
 Approved For Release 2003/05/15 : CIA-RDP78B04747A002700040006-3
claimed, and that over a limited range several gage blocks were
measured with a high degree of accuracy. The Isystem STAT
is not to be developed into a standard package and therefore it
would be necessary to design its components into the mechanism
of the Stereo Comparator. This, of course, would also be true
of the laser interferometer discussed in last month's report.
During October drawings will be prepared to show how each of these
systems would be integrated into the X-Y stages.
A description and comparison of the
metering system is contained in Appendix
1.5 General System Considerations
1.5.1 Overall Format Observation
During the reporting period efforts were continued to devise a
technique to permit convenient viewing of the overall format.
The two major criteria were that the technique not require
generation of expendable hard copies of the input photography,
and that the display must maintain very high quality (good resolu-
tion, brightness and grey-scale) for indefinite periods of time.
In addition the technique must not add to the complexity of oper-
ation or significantly reduce the speed of operation.
The first criterion rules out photocopying, diazo copying, etc.,,
and requires the consideration of reversible information storage
systems. Among those considered were: Photochromic dyes, mag-
netic tape and drums, image storage tubes, thermoplastics, rever-
si.ble Xerography, and electroluminescert displays.
As mentioned in the second monthly report, no photochromic or
electroluminescent systems are known which will maintain image
quality over protracted times, particularly at room temperature,
The same is true of image storage tubes, with the possible
STAT
Approved For Release 2003/15/15 3CIA-RDP78B04747A002700040006-3
 Approved For Release 2003/05/15 : CIA-RDP78B04747A002700040006-3
exception of a special tubeI Iwhich
uses an auxiliary light source within the tube to: illuminate an
image produced by electron beam writing on a special face material.
However, the tube requires repeated writing in order to
attain a satisfactory grey scale. Also, erasure requires five to
ten seconds, so that both writing and erasing are incompatible
with system timing requirements.
A trip was made
to discuss the
racticality of a reversible
process, achieved
by omitting the normal fixing operation. It was established that
the technique was probably feasible, and that such a system would
not require extensive development. However, in its basic form
they could guarantee no more than five shades of
grey, which does not sound acceptable.
STAT
STAT
STAT
STAT
STAT
working on a thermoplastic process which STAT
they call FROST , which may have promise for the future, but it
is not expected to be available during the course of the Stereo
Comparator program.
There are three principal companies engaged in the manufacture of
video recording equipmentF I They have been
surveyed to determine the practicality of using a magnetic tape or
drum to store the picture information, with a "stop motion" feature
to permit continuous display of any given frame.
STAT
STAT
Approved For Release 2003/0/15 4CIA-RDP78B04747A002700040006-3
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
a letter from their representative advises. that more than 2 1/2
hours has been achieved. The cause of the loss of picture is
not entirely-clear from conversations with them, but it is believed
to be a combination of abrasion of the tape surface and fouling of
the head gap, accelerated by heat build-up due to re n at
over the same tape section hat
actual head-to-tape contact is necessary in video recording.)
uccessfully used a special lubricant which apparently
permits in efinitely extended re-play. However, although the
lubricant provides successful tape protection they caution strongly
against its use unless absolutely necessary, because of the prob-
ability of harmful accumulations throughout the recorder mechanism.
At the present time a proposal is expecte for a system
incorporating a magnetic drum and multiple read-write heads which
is claimed to have satisfactorily high life and the capability of
continuous scanning or display of any part of the stored video
data. Extensive system integration problems are foreseen, however,
and a search is continuing for more acceptable solutions.
STAT
STAT
STAT
1 - 5
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
2.0 PROBLEM AREAS
2.1 Optical
No new problems are apparent in this area. Tolerance requirements
will be a continuing concern until they are completely defined.
Resolution requirements still appear to be attainable, and it is
planned to obtain confirmation of the soundness of the system
approach by discussions with potential vendors in the fifth month
of the program.
2.2 Mechanical
Work on the film transport mechanism has had to be temporarily
suspended because of more interaction than had been expected be-
tween the optical design and the transport configuration. This
is particularly true of the small working space now predicted for
the highest power objective lens, which complicates transport
removal, and the very shallow depth of field'similarly predicted.
The transport design will be resumed when this data is definitized.
Tests of vacuum hold-down systems have been suspended pending
receipt of representative film samples covering the range of film
thicknesses and widths expected. These samples were requested from
contracting office personn9l at. their last visit to
The X-Y comparator stages and optical frame designs are proceeding
satisfactorily and no technical impossibilities are apparent at
present. The massiveness of the tables is such that careful place-
ment within the allotted room will be necessary. Information is
needed regarding the arrangement of building services (air, power,
and ventilation) within this room, as well as the location of
doorways.
2.3 Correlation
No unforeseen technical problems.
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
2 - 1
 Approved For Release 2003/05/15 CIA-RDP78B04747A002700040006-3
2.4 Metering Systems
At the present time the choice for a metering system is seen as
provide a substantially higher accuracy capability, though it
i
n ca basis the Interferometer, if properly used. is felt to
Interferometer.. On a tech-
s questionable whether this potential can be usefully employed.
provides the greatest protection against miscountin
g
significantly more expensive. The ease of application of either
system will be more apparent after the current month's effort is
completed.
Since some of these trade-offs are economic in nature, and others
involve anticipated usage of the Comparator, a final recommendation
will not be made until after further discussions with the Contracting
Office.
2.5 General Systems
The presentation of the overall format display has always been
regarded as a critically important aspect of the Comparator operation.
The problem has proven much more difficult than originally expected
and no really satisfactory technique has been conceived to date.
It is hoped that new inputs
in the next month
will contribute toward a Solution to the display requirement. If
not, serious consideration will be given to several possible hybrid
combinations of short-term and long-term storage and display systems.
These have been considered casually in the course of the study, but
have not been reported previously because of their objectionable
complexity and probable high cost.
STAT
STAT
STAT
STAT
Approved For Release 2003/05/15 : CIA-RDP78B04747A002700040006-3
2 - 2
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
3.0 PROJECTED WORK FOR NEXT REPORTING-PERIOD
3.1 Optical
Design,will continue on the zoom and relaying system. It is
hoped that all critical elements will be complete by mid-November
so that the design can be presented to potential vendors in time
to take into consideration any recommended changes.
3.2 Mechanical
Casting configuration drawings for the X-Y stages and optical
frames will be completed in October. Structural analysis will
continue. Estimated weights of the stages will'be available
so that drive system design can be started.
3.3 Correlation
Data-taking on correlator performance is expected to begin in
October.
3.4 Metering
Drawings showing integration of the two preferred systems will
be prepared. Final selection will await discussions with the
Contracting Office.
3 - 1
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
 Approved For Release 2003 ?d,195 : CIA-RDP78BO4747AO02700040006-3
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
 Approved For Release 2003/05/15 : CIA=RDP78B04747A002700040006-3
5.0 VERBAL COMMITMENTS AND AGREEMENTS
The Contracting Office has been requested to supply samples of
film and, if possible, stereo photography to be used in forth-
coming tests. Blank or exposed film is needed to perform tests
on the hold-down system, and the samples should cover as wide a
range of anticipated thicknesses and widths as possible. Stereo
photography covering a spectrum of contrasts, scales, imagery
and density would greatly facilitate correlator evaluation.
It is now understood that the requirement for theta alignment.
of the film has been eliminated, since computer corrections for
misalignment are included in the data reduction program.
5 - 1
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
 Approved For Release 2003/05/15 : CIA-RDP78B04747A002700040006-3
APPENDIX A
CONDENSER SUBSYSTEM
The image quality (resolution and "crispness") of an optical
system resembling a microscope depends to a.high degree upon
the proper illumination. Both the object field and aperture
diaphragm of the objective lens (exit pupil EX), have to be
filled with light, the object field, of course, completely and
EX to about three-fourths to four-fifths of its diameter. The
latter is an empirical value which provides crispest image
appearance. It is important to restrict the illumination light
pencil to the necessary dimensions as early as possible along
the optical train in order to avoid stray light, which will
reduce contrast, and therefore severely impair image quality.
This principle is best reduced to practice in the classical
Koehlernillumination scheme (see Figure A-i).
The condenser proper, Cond, is arranged immediately in front
of the object plane, O.P. In its front focal plane, F, is placed
the aperture diaphragm, A-.D., which serves as a secondary light
source. The condenser, Cond, images A.D. into infinity in the
object space. Since microscope objectives are telecentric.lenses,
i.e., their EX coincides with their back focal plane, A.D. is
finally imaged into EX.- Making A.D. an iris provides for its
adjustment so that it fills three-fourths to four-fifths of Ex,
The filament of the light source, L.S., is imaged by a so-called
collector, Coll., into the aperture diaphragm so that its structure
is not noticeable in the object plane, O.P.. The optical components
have to be dimensioned so that an iris behind the collector is
imaged into O.P. by the condenser. This iris is then the field
stop, F. St., which restricts the illumination to the required
field size,in O.P.
Approved For Release 2003/05 A5 CTA-RDP78B04747A002700040006-3
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
L.S.
F. St.
Conti
0P.
\.
MSXC.SCHEME OF KOEHLER ILLUMINATION
FIGURE A-1
Approved For Release 2003/05/154: C1ADP78B04747A002700040006-3
 Approved For Release 2003/05/15 CIA-RDP78B04747A002700040006-3
Implementation of the Koehler-illumination for the Visual
Observation System (VOS) is considered an absolute necessity
for meeting the requirements of highest possible image quality.
However, this implementation is difficult because of the large
range of field size, from 18 to 0.861 mm diameter, and of numer-
ical aperture, from 0.044 to 0.558 (see Appendix B, Table B-7).
It is not. possible to cover this large range without any change in
the illumination system (besides.the change of A.D. and F. St. dia-
meters). An approach is used which is rather common in microscope
design. The whole range is split into two parts, one for low
magnification, large field and small N.A. (3.5x objective), and
one for large magnification, small field and large N.A. (9x and
21x objectives)..
The first approach for both cases (Figure A-2 ) is based on
"thin lenses" and a paraxial ray trace using the formulas of Mil-
Handbook 141 (Chapters 5 and 6):
num (nu)-1 - y F
y+l=y+ut
F = (n-1)(c a-cb)
where: y - height 'of incidence at surface k under consideration.
u - slope angle behind surface k (in radians)
t - distance to next surface (k to k+l)
F - power of lens - 1/f
f - focal length of lens
ca;cb - lens curvature - 1/ra; 1/rb
This set of formulas is also known as "Lange lens formulas",
the only difference between the Mil-Handbook 141 and conventional
Approved For Release 2003/057'157 dA-RDP78B04747A002700040006-3
 Release ~41A00~~00040006-3
L"`f;
F? ?.0046
VOS ILLUMINATION SYSTEM
Approved For Releaagd&(0 /1I : CIA-RDP78BO4747AO02700040006-3
A - 4
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
textbooks being the difference of sign for u. The advantage of
these formulas is that they avoid the reciprocals of the Gauss
lens formula and that they permit direct use of important design
parameters, namely, object and image size and numerical apertures.
The calculation proceeds backwards from the object plane with the
maximum values for the 9x and 21x objectives: y' - 4.0 for the
chief "edge ray", i.e., the ray through the edge of the field
(with a safety margin of 0.5 mm over the actual requirement) and
u - 0.0; N.A. - u = -.5580 for the "position" ray, i.e., the ray
determining the position of the field (y a 0).. The full maximum
value for N.A. has been used though only,three-fourths to four-
fifths of the aperture should be filled with light, in order to
have a safety margin.
The first lens of the condenser proper is,.optically, a hemisphere
following microscope design practice, because a hemisphere is
"aplanatic", i.e., does not introduce spherical aberrations and
coma. (This is important in view of the large aperture angle.) A
small air gap within this lens is without consequence so that the
glass plate for film support can be considered to be split off
this hemisphere. It is important, however, to keep the overall
glass thickness constant, and it must be maintained at the chosen
value of 30 mm in the event that the platen is ever changed.. This
is unavoidable in view of the existing severe requirements. The
angle at the air gap remains below the critical angle of total
reflection, but it is advisable to coat the surface of lens Lll
and the surface of the platen with a highly efficient and hard
anti-reflection coating.
Approved For Release 2003/415: 8IA-RDP78B04747A002700040006-3
 Approved For Release 2003/05/15 : CIA-RDP18BO4747AO02700040006-3
The next two lenses toward the light source, L10 and L9,
bring the position ray back to the optical axis at a distance
of 425 mm from L9, which determines the position of the field
stop, F. St. This position is tentative; it can be increased
without difficulty by decreasing the power of lens L9, if the
mechanical table and condenser subassembly design should require
such an increase. The chief ray intersects, then, the optical
axis at a distance of 14.3667 mm from lens L9, which determines,
the position of the aperture diaphragm, A.D. The height of
incidence of the position ray at this plane,?y - 15.40 mm, deter-
mines the radius of the A.D. iris for the maximum N.A.
Next comes the first beamsplitter, "Prism". Its reduced thick-
ness t/n (assumed n - 1.512) is given in-Figure A-2 , thus
making its consideration for the ray trace unnecessary. But it
can be seen that a size of 40 by 40 mm suffices for this beam-
splitter. A mirror is necessary to bend the optical axis back
into the (approximately) original direction; its size is 50 by 70 mm.
.It can be used for the proper adjustment of thellight direction
with respect to mechanical design as well as for the shop assembly.
Between fieldstop and light source is the collector. It has to
bring the chief ray back to the optical axis. The slope angle
of the chief ray determines the magnification between light source
and. aperture diaphragm, and thus the required size of the filament,
which should, of course, be as small as feasible. A slope angle
u - -.578 is attainable, if a three lens collector is considered to
be acceptable. This results in a magnification of m' - 4x and a
filament size of 8 by 8 mm with a safety margin of 2 by 0.12 mm
for the largest aperture in addition to the safety margin quoted
above (use of the nominal value for the maximum N.A.).
Approved For Release 2003/05/15A CtAfRDP78B04747A002700040006-3
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
A prospective choice for the lamp would be a
lamp DEW or'DGF, 500 w, 4 pin base, 25 hours life, 3200?K color
temperature. Both have the compact type envelope and can be
burnt horizontally as well as vertically. The average luminance
is not quoted and information about it cannot be obtained (as
usual, regrettably), but the filament is very compact (biplane)
leaving only very small gaps between the coils so that the
average luminance should be as high as possible. The difference
between the two types is that DFW has a "proximity reflector"
while DGF has none. Whether the proximity reflector (a reflector
built into the lamp) has an advantage for this application must
still be determined.
The filament size leaves practically no safety margin (8.0 x
7.8 mm). If it is too small a 750 w lamp such as the DEP has to
be used (filament 9.4 x 9.6).
STAT
Of course, other manufacturers have similar lamps. Final choice
will require some experiments, i.e., measurements in connection
with the condenser, at least with the collector. The remarks about
the lamp made here are only indicative of the type of lamp and the
wattage required. Good forced cooling will always be a necessity
in order to preserve lamp life and to avoid heating up of sensitive...
parts of the table by the lamp. Except for the most dense film it
will be possible to reduce the lamp voltage and therefore extend
its life.
It should be possible to handle the case of the 3.5x objective
with the same lamp and general layout since the Lagrange (or
"optical") constant (01 = yu - yu) is 1.460 (with y'(max) - 10.0 mm
and N.A. (max) = u = -.146) as compared to 2.22 for the 9x; 21x
case. Some additional optical elements will be needed. The approach
Approved For Release 200315/157 CIA-RDP78BO4747AO02700040006-3
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
taken I with itsl condenser (removal of one
condenser lens) should be avoided since it deviates from the
Koehler-illumination principle. The filament structure then
becomes noticeable in the image unless it is smeared out by a
ground glass. The latter is not possible because it is accompanied
by a considerable light loss which is intolerable for this appli-
cation.
results in, an increase of the slope angle, u, of the chief ray
in the aperture diaphragm, namely u w + .3649 as compared to
+ .1445 in the previous case. This chief ray has to be brought
back by a lens. The best choice for an additional "flip-in"
lens is'between beamsplitter "Prism" and "Mirror" (see Figure A-2).
The field of 20.0 mm diameter (which includes a safety margin)
Its diameter would still be reasonably small, but y (for the chief
ray) would still be large enough to have enough effect upon the
chief ray with a reasonable lens power.- The power of the lens,
L4, should be chosen so that the (new) chief ray intersects the
former chief ray about 20 mm in front of the field diaphragm. A
negative lens could be introduced in that place to restore. the
former optical train.
The power resulting for L5 .is then F(5) - + .0162 (f - 62 mm).
The power for the negative lens in front of F. St. would only be
F -.0007. This is so small that an attempt was made to omit
it. Both the chief ray and the position ray intersect the plane
of the lamp within the area of the filament, which is all that
is required. The filament is then not exactly imaged into A.D.,
but this very small discrepancy is without any practical conse-
quence within this context. Also, A.D. is not imaged absolutely
STAT
STAT
Approved For Release 2003/05/15k CJAERDP78BO4747AO02700040006-3
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
correctly into the EX of the 3.5x objective since EX does.
not coincide with the back focal plane of this lens, unlike
the case of the 9x and 21x objectives. (This has been considered
in the calculation by having u'- .0555 instead of 0.0 in O.P.).
But again, the discrepancy is so small that it can be neglected
in this case.
However, the omission of L4 does produce an unfeasible design
because the lens diameters for the collector lenses would'have
to be so large that this solution has to be considered unattainable.
If the slope angle of the edge ray is made smaller (since it would
suffice for a proper size of the filament image in the A.D. plane,
namely .3837), the required filament size goes, up, beyond that of
the anticipated value of 8 by 8 mm.
The only way out is to start from the filament with a .chief edge
ray with u + .3837 and to place a negative lens, L4, in front
of the fieldstop, F. St. (a distance of 20 mm has tentatively been
.chosen), which bends the edge ray so that it hits lens L5 at the right
height of incidence (y - -15.1170). The resulting lens powers are
for L4, F - -.0007 and for L5, F - + .0198. Tracing the position
ray through from L5 to the lamp results in a filament size of
4 by 4 mm, which is attainable.
It might be better to build the power of the negative lens 14
into the collector and to use a positive lens instead with the
9x, 21x objectives. The switch from the 9x, 21x case to the 3.5x
case would thus involve removal of one lens and insertion of another.
This might have advantages for the mechanical design. Optically,
the advantage would be that there is a better chance to make lenses
L2 and L3 equal, 'reducing, their price, and use the (new) lens L4
for the correction of some aberrations. This favorable idea has
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
not been implemented at this stage of the development, since
some more changes will have to be made after the mechanical
design requirements have been established. The idea of such a
change will, therefore, be left to the next design cycle; it
will not change the basic design concept.
One disadvantage of this otherwise good approach is that F. St.
will not be exactly correct for the 3.5x case. Its correct
position would be near the front edge of the "Mirror" instead.
Therefore there will be a small amount of excessive light in
O.P. (a small diffuse zone of light around the edge of the field).
This should have a negligible effect upon image quality in the
3.5x case since the 3.5x objective is rather far away and its
numerical aperture is much smaller than the other.
The correlator case differs from the Visual Observation System
(VOS) in that EX and, therefore, A.D. has to be imaged upon the
photocathode, PM, without.the necessity, of having a defined
fieldstop." (The image of the scanning spot in O.P. restricts the.
field sufficiently.) Also, the magnification between A.D. and
PM will have to be larger, since a 2 inch diameter for the photo-
cathode has been contemplated. But two cases still have to be
considered: a) the 9x and 21x objectives and b) the 3..5x objective
(see Figure A-3).
In the correlator case, one lens L7 will be common to a) and
b). For a),'lens L6 augments the power of lens L7 resulting in
a maximum image size of A.D. (that is for the maximum N.A.) of
y 20.2 mm. Lens L6 is taken out and lens L8 is switched in for
b). Lens L8 is again arranged near the beamsplitter prism and bends
the chief ray toward lens.L7. The maximum image size of A.D. on
the photocathode is then y - 10.6.mm.
A - 10
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
PHOTOCATHODE 4 ? /OS9/f O' o) 0
cl~
0
c
2.3344)O'=-32.5/fI f:!0072
L#7
M=3.46Z6 , --20,359+
20.1647 s ? o.O
ly 40033) p} 142920
pf- 9.0.27!2&=-2?5650
40326 ) - =/.5J46 SAD /5.4013; =0.0
= *J690 ; - = 6,7770 /F=+,0097 ^d'=194/0 ~=20760
FLYING SPOT SCANNER LIGHT COLLECTION SYSTEM
Approved For Release 2003~0r1 : CpfARDP78B04747A002700040006-3
A - 11
 Approved For Release 2003/05/15 : CIA-RDP78B04747A002700040006-3
Both switch lenses, L6 and'L8, can be attached to one axis of
rotation since their optical axis coincides. This makes it
possible to turn one lens in while the other lens is turned out.
A tentative mechanical arrangement for both optical trains
.(VOS and correlator, but for the 3.5x objective only) is shown
in Figure:A-4. It is necessary to bend the optical axis for the
correlator train somewhat so that it does not interfere with the
VOS train. Refraction by a small prism attached to the proper
half of the beamsplitter should suffice for this purpose. Dis-
per sion should not be a handicap since there is enough of a safety
margin for y on the photocathode to take care of small image dis-
placements because of dispersion.
One additional element will be required in the VOS train. It
is necessary that the optical axis between lamp and collector be
in an (approximately) horizontal position, even if the lamp.is
used in a horizontal position. Otherwise (that is if the optical.
axis runs vertically as shown in Figure A-4), blackening of the glass
envelope above the filament by tungsten evaporation will soon spoil
the brightness. This will make it necessary to use.a plane mirror
between table housing and collector, which makes it necessary to
increase the distance of fieldstop and collector from the table
housing. Since an optimum solution'will have to result from the
mechanical design, this problem has not been pursued further at
the present.
Addition of the mirror will result only in a proportionately
larger collector and a small power adjustment for the lenses L5 and
L9. It will not change the general design concept, in particular,
not the slope angle (N.A.) between lamp and collector and the
filament (lamp) size. However, this situation makes it unadvisable
A - 12
Approved For Release 2003/05/15 : CIA-RDP78B04747A002700040006-3
 Approved For Release 2003/05/15: CIA-RDP78BO4747AO02700040006-3
A7SS/BLE I AWFERRED
MAX/NUM 1 TABLE
TABLE I OUTLINE
OUTLINE 1
VOS CONDENSER & FSS COLLECTOR,
MECHANICAL ARRANGEMENT
FIGURE A-'4
I-RDP78B04747A002700040006-3
Approved For Release 2003/05/15 : CA
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
at this time to change the "thin" lenses'into thick lenses and
to determine their optimum shape from a (trigonometrical) ray
trace. This task is routine work which should not cause severe
difficulties so that it can be left for later. '
Figure A-4 indicates that-the condenser subsystem can be integrated
in the table design without causing difficulties. Table A-i lists
all components for the condenser subassembly and their tentative
dimensions. "Curvature" ca-cb has been calculated from
ca-cb ? F/(n-1); n - 1.512
and "estimated lens thickness", tL,
D 2 (ca-cb) .
from
Lens diameter D follows from the drawings (Figures A-2 and A-3) by
D ':2 (1 yf + lyl )
A - 14
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
 Approved For Release 2003/05/1 Agjt.;RPR.78B04747A002700040006-3
Sub-
Group
cn
Component Power
F[mm-]
Focal
Length
f[mm]
Free Combined
' Dia. Curvature. Estimated
D[mm] c - c Thickness
a [-1]b tL[mun]*.
Lamp: Biplane
filament
8 x 8 mm,
500 or 750W
Li
+.0048
210
105 .0093 13
L2; L3
+.0040
250
130
.0078 17.
Fieldstop,
F.St., Iris
Diaphragm,
max. dia.
120 mm
L4
-.0007
1450
125 -.0014 7.5
L5
+.0198
50.5
(edge)
38 .0386
Plane Mirror:
50 x 70 mm
Photocathode of
PM: 2 inch
Diameter
L6
+.0032
310
65 .0062
L7 +.0072
140
70 0140
9
L8 +.0125
80
48 .0244
7.5
Prism, Angle
of Deviation
15?
Beamsplitter
Cube 40 x 40 mm;,
T(CORR): T(VOS)
CA P4 85:15
00
o Aperture Diaphragm,
A.D., Iris, Max.
Dia. - 31 mm
L9 +.0097
L)
103
37
.0189
L10 +.0170
.0332
L11 +.0170
59
30
.0333
L
30
-
~..,
A out 3 o the va ,eas~a2&Sj 1g 1'7- 7$ as com-
"
il
pared to o
81
ia ance. 15
 Approved For Release 2003/QWKRIBP78B04747A002700040006-3
OBJECTIVE AND INTERMEDIATE LENSES
Tables B-1 and B-2 represent the first, tentative design approach.,
They are reprinted from Tables Cl and C2 of the second monthly
report. After looking further into the requirements for the
image rotator and for the zoom system.it became,clear that a
distance between primary image and intermediate lens as deter-
mined by Pint - 450 mm is too small. Therefore, Pint ' 540 mm
has been chosen for design approach 2 (Table B-3).
A discussion with Contracting Office representatives revealed
that it is desirable to expand the range of lens 2 so that a con-
ventional work range between 40x and 90x can be covered without
the necessity of switching lenses within this range of magnifi-
cation. This is possible by increasing the zoom range from
m'$ - 1.0l/3.0 to m'z = 1.0-X1/3.5. The overlap between
lenses 2 and 3, and also between lenses 1 and 2 is then consi-
derably increased. In the attempt to realize this objective the
focal length of the intermediate lens is kept at fiat = 540 mm
and the diameter of. the primary image at 2y''equals 18 mm (design
approach 3, Table B-4). This approach requires an increase in'magni-
#ication, m'obj. for lens Nos. 2 and 3, and a. decrease in the
respective F/numbers, or an increase. in the corresponding numer-
ical aperture NA. But this is the inevitable price which has
to be paid for the increased magnification range. On the other
hand, field angles for the first image will slightly decrease from
20' - 6.90? to 6.65? for the full area and from 20' = 2.30? to
1.91? for the center area (area of highest resolution).
The disadvantage. of approach 3 is that the maximum diameter of
the first image, that is before zooming, has to,be 3.5 x 18 = 62.5 mm.
Approved For Release 2003/05/158 CIAjRDP78B04747A002700040006-3
 I t. _ I 6-3
Approved For Release 2003/05/15 : CIA-RDP78B0~4747A0027000 00
1
mlobj.
(Objective)
m,Z
(Zoom)
OPTICAL PARAMETERS FOR INPUT SIDE OF VISUAL OBSERVATION SYSTEM
Design Approach
m C
(Combined)
Object Field
Diameter
2y !mm!
Object sp. Resolution
Numer. Ap. Required
NJ sinol R[lines/mm
0.55
0.121
0.044
Demagnifying Zoom - System 1-->1/3; fiat = 450 mm; diameter of primary image 2y' = 18 mm.
Absolute
Resolution;
Lines Over
Object Field
Prim. Image
Numer. Ap.
NA' - sinal
I o.0275
1260 1 o.0385
0.0403
0.0440
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
Aperture
Angle 2a'
at Prim.
Image
fObj [mo'
F/Number
4.40? 60
4.80? F/1.75
4.60? 150
5.04? 1 Fl4.1
 Approved For Release 203/05/15 :CIA- &P78B84747AOt27000 00
Design Approach
The field angle for the 'full" area is 2r- - 6.88?, for the center' area 2s - 2.30?.
Lens No. 4 refers to "intermediate lens'. Lenses No. 1-3 image into infinity; lens
No. 4 from infinity. The aperture diaphragm of lenses 1 and 2 has to be in their back
focal,planes (telecentric lenses).
3
Assigned
Magnifi-
Focal
Length
Numer. Ap.
for Center
Corres-
Numer. Ap.
Full Fi
ld
R
Lens
No
cation
~
m
fobj.
Area
ponding
NA for
e
Diem.
esolution
Full
.
Obj
F/number
Full Field
2y [mm]
Field
20x
22.5
0.55
o.9
0.265
lines/am
7. 5x
60
0.288
1.75
0.104
3x
150
0.121
4.1
0.044.
0.044
11.5
o.015
aasolution refers to overa11 systasss resolution with refsssws to
object (film plane for leases No. 1 to 3; with selars~ to -
first image plan for lens No. 4. It has basis calculated from the
diffraction limit with a safety factor of 2.0.
8
9
10
Resolution
on
Diameter
Diem, of
in Center
of
Center
Ap. Die-
rea -
Area
Area
phragm
R
lines/mm
a
f?1
1000
0.9
25
525
2.4
34
220
6
37
80
18
50*
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
 ea 4747A 27000~'0006-31~, t t
pprove For Ree'- _ se 2~3105/11:.'CIA-4DP78BS
OPTICAL PARAMETERS FOR INPUT SIDE OF VISUAL OBSERVATION SYSTEM
' 1
m
obj.
2
3
4
5
6
7
8
9
10
(Objective)
Magnification
Range
,
m Z
(Zoom
m C
(Combined)
Object'Field
Diameter
2y final
Object sp.
Numer. Ap.
NA Sind
Resolution
Required
I. lines/mm ]
Absolute
Resolution:
Lines Over
Object Field
Prim. Image
Numer, Ap.
NA' - aind
Aperture
Angle 2a'
at Prim.
Image
fObj.(m,!
F/Number
1)
2Ox
1.0
20a
0.9
0.55
1000
900
o.0275
3
15?
-- --
.
27.0
200 - 67
o.33
6.67x
2.7
o.26
480
1290
0.0392
4.58?
Fl0.9
2)
7.5z
1.0
7.5x
2.4
0.288
525
1260
75 - 25
0.33
2.5x
7.2
0.104
190
1370
0.0385
0.0420
4,40?
4.80?
72.0
Fl1.75
3)
3x
1.0
3.Oa
6
0.121
220
30 - 10
0.33
1.Os
--
o.044'
80
1320 0.0403
1320
0.0440
4.60?
5.04?
180
F14.1
Demagnifying Zoom - System 1-?--I1j3; fiat ` 540 mm; diameter of primary image 2y' = 18 mm.
Design Approach O2
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
 78B84747A0%27000006-3 ~-
Approved For Release 20%3T05/11 : CIA- P
OPTICAL PARAMETERS FOR INPUT SIDE OF VISUAL OBSERVATION SYSTEM
Design Approach n
1
m'ob j
(Obi tlye)
Kagnification
Range
20x
m' Z
(Zoom)
m'C
(C ambined)
Object Field
Diameter
2y fmmj
Object sp.
Numer. Ap,
NA - sind
0.288
1.0 9a
0,288 2.5
o.110
1.0
0.146
Demagnifying zoom system 1-N1 3.5
fi.t-540mm
2y' - 18 am
Resolution
Required
R[lines/mm!
Absolute
Resolution;
Lines Over
Object Field
Prim. Image
Numer. Ap.
NA' - sind
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
This would require an increase of the effective free aperture
diameter of the K-mirror, resulting in an increase of the optical
length through the K-mirror. It is very desirable to make the
maximum free aperture of the K-mirror not larger than 55 mm, which
would then keep the optical path through the K-mirror at about
300 mm. Therefore, an image diameter of 2y' = 15.5 mm has been
chosen in design approach 4 (Table B-5). In order to retain an
image field diameter of 2y'40 18 mm at the ocular, the product
of periscope magnification, m'Per , (until now always assumed to
be 1.0) and of ocular magnification, m' 0C , has to be
ml Per x m'Oc m 11.612x
which should not pose a difficulty. The maximum diameter of the
first image remains then 2y1'(max) - 54 mm.
However, with an optical path of 300 mm for the K-mirror, only 240 mm
remain for the beamsplitter, whose optical path equals t/n, (?Z.;
36 mm), the zoom system (at least 200 mm desired), fine focusing
with the intermediate lens (about 20 mm desired), and a reserve
for mechanical design. Without the latter, 246 mm are required and
only 240 mm are available. A change of first = 540 to f t 'm 600 mm
for the focal length of the intermediate lens would again result
in a chain of difficulties, which can be avoided by the design
approach 5. With the latter approach the input lens and K-mirror
parameter remain the same, but a negative lens is placed between
K-mirror and beamsplitter, bringing the magnification of the first
image again up to 18 mm for the center field and 62.5 for the full
field without the necessity of increasing the dimensions of the
K-mirror. The optical path available for beamsplitter and zoom
system is then increased, a diameter for the primary image of 18 mm
is retained, which makes it possible to examine the primary image
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
 Approved For Release 2003/05/1 CIA-RDP78B04747A002700040006-3~
OPTICAL PARAMETERS FOR INPUT SIDE OF VISUAL OBSERVATION SYSTEM
Uk Obj.
(Objective)
Magnification
Range
1.0
0.2870
Design Approach
Object Field
Diameter
2y 'mm,
18.0 o.861
5.167 3.000
7.714
2.213
Demagnifying Zoos System 1--11/3.483;
fins ' 540 mm
20(min) - 0.0287 - 1.645?; 2y'(min) - 15.5 mm
24(max) - 0.100 - 5.72?; 2y'(==) - 54 mm
mEPer x m'Oc - 11.612 so that 2y' Oc - 18 em
Object sp.
Numer. Ap.
NA - sine(
Resolution
Required
r lines!e!
875 0.0311
1295 0.0459
1370 ~ o.0487
1440 I o.0510
9 1U
Aperture fObj,[~)
Angle 2a'
at Prim. F/Ntmber
Image
3.56
j 30
5.24 F/o.896
5.56 I 180
5.84 F/3.42
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
with the eyepiece used for the observation station. Should,
however, the increase of optical path not suffice, the possi-
bility still exists of increasing the power of the negative
lens. This would further lengthen the optical path, though the
size of the first and the primary image will also increase
(which would not pose a serious difficulty), without the necessity
of changing the design parameter of condenser, objectives, and
K-mirror. Thus, a higher design freedom is gained from this approach
which is an important advantage. Leaving the diameter of the
primary image, for the time being, at 2y'pr - 18 mm, the following
will result:
Making the distance between intermediate lens and negative lens,
t1 It equal 325 mm, the height of incidence, y1 , at the negative
lens for the paraxial chief ray is (with y2 - 27 mm in the plane
of the first image)
yl - 27 x 325/540 - 16.25 mm
and slope angle of this ray in front of the negative lens:
uo - 27/540 - o.0500
The slope angle of the corresponding paraxial position ray is (in front
of the negative lens) (t1 - 540 - 325 - 215)
uo - -y1/t1 - -16.25/215
- -o.075,581
B - 8
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
Resulting in the height, y0, at the intermediate lens of (t - 540)
y0. - -u0 - o.075,581 x 540
- 40.813 mm.
The image diameter in image plane (2)'shall be increased from 15.5
to 18.0. The slope angle behind the negative lens is then, because of
the Helmholtz-Lagrange equation,
ul - u0 x 15.5/18.0
- -0.065.9083.
The required power, F1, of the negative lens follows from
F1 - -(ul - u0) /Yl
f
neg
- -0.010,498/16.25
- -0.000,646
1548 mm
and the distance, tl, from negative lens to image plane increases
from 215 mm to
ti - y1/u
-249mm
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
B - 9
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
The combined focal length of both lenses is
f (int. + neg.) - -Yo/ul
- 627.091 nun
Since the space requirement for fine focusing by meant of the inter-
mediate lens has been considered already (by assuming to - 325nun), this
suffices for the zoom system (t - 200 nun) and the beamsplitter (tops 36 mm),
leeving a reserve of 23 mm for the designer. The negative lens is not
very strong and it may be used for the fine correction of some
aberrations.
The advantages obtained justify the use of this negative lens. The
overall results are shown in design approach 5, (Tables B-6 and B-7). The
question may be brought up again at this time whether the safety fac-
tor of 2 in determining the NA or F/number of the objective lenses as
required by resolution is a realistic one or is too overcautious (see
eq. 4b,.Appendix C2, of the second monthly report). It is planned
to discuss this question in detail with the prospective subcontractor
for the lens and periscope design.
It will, to a large extent, depend upon the proper balance between
complexity of design (and price) on the one hand, and strict adherence
to resolution specifications on the other hand. It should be kept in
mind,'however, that the question of numerical aperture - F/number also
enters into the question of illumination. The exit pupil, EX, for
diffraction limited visual instruments is about 1.0 mm, while EX 1_2.0
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
B - 10
 Approved For Release 2003/05/15: CIA-RDP78B 4747A002700040006-3(
OPTICAL PARAMETERS FOR INPUT SIDE OF VISUAL OBSERVATION SYSTEM
Design Approach @
mlObj.
-_(bi-motive)
Magnification
Range
m,Z
(Zoom)
1.0
0.2870
Object Field
Diameter
2y [mm
Object sp.
Numer. Ap,
NA - sinoi
Resolution
Required
R' lines ?mcn
Absolute
Resolution;
Lines Over
Object Field
432 f 1295
o.0395 4.53 F/o.896
595 1195
0.0365
200 1395
0.0428
265 1370
o.0419
80 1440
0.0440
8
9
10
Prim. Image
Aperture
fObj,(on]
Numer. Ap.
Angle 2a'
NA' - sin ci '
at Prim.
F/Number
%
Image
o.0266
3.05
30
4.18
70
4.91
F/1.529
4.80
180
5.043
F/3.42
Deaoagnifying Zoom System m'z - 1.0---41/3.483; intermediate plus negative leas f(int. + neg.) - 627.091-on keeping light pencil diameter
through K-mirror below 55 on.
For the first image (before action of zoom system) 2y'(min) - 18 on; 2y'(mas) - 62.5 on.
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
 Approved For Release 2003/05/15 : CIA-RDP78B04747A002700040006-3
Table B-7
LENS SPECIFICATIONS
Design Approach V
The field 'angle for the "full" area is 20 - o.100 rod - 5.731?, for the center area 20 - o.0287 sad - 1.645?. Lens No. 4
refers to "intermediate lens", lens No. 5 to "negative lens" assumed to be 325 an behind intermediate lens. Total optical
distance from intermediate lens to first and primary image is then 574 ma or 249 mm optical path from negative lens to image
with a glass path (beamaplitter quartz) of 58 mm; the equivalent optical path of 38.4 mm is included in the 249 mm quoted.
Lenses Nos. 1-3 image into infinity, lens No. 4 and No. 5 from infinity. The aperture diaphragm of lenses 1 and 2 has to be
in their back focal planes (telecentric lenses).
1
2
3
4
5
6
7
8
9
10
Nominal
Focal
flumer. Ap.
.
Magnifi-
Length
for Center
Correa-
Numer. Ap.
Full Field
Resolution
Resolution
Diameter
Dism. of
Lens
cation
f
Area
ponding
NA for
Dian.
for Full
in Center
of Center
Ap. Din-
No
m
ob-
j
F/number
Full Field
2y aal
Field
Area
Area
phragm
.
Obj.
'
NA - sino(
R lines'aa
R lines 'mm'?
1
21x
30
o.558
o.90
o.237
3.0
432
1016
o.86
33.5
2
9x
70
0.327
1.53
0.110
7.0
200
595
2.0
46
3
3.5x
180
o.146
3.4
o.044
18.0
80
265
5.2
53
4
(intern.)
540
o.051
9.8
o.0146
54.0
32.6
93
15.5
50*
5
(negat.)
1548
o.044
11.4
o.0127
62.5
28
80
18.0
52*
Resolution-(columns 7 and 8) refers to overall systems resolution with reference to
object (film) plane for lenses No. 1 to 3, with reference to respective first image planes
for lenses No. 4 and 5. Numerical apertures, NA (columns 3 and 5), refer to object space.
for lenses No. 1 to 3, to image space for lenses No. 4 and 5. They are obtained from the
diffraction limit for the specified resolution with a safety factor of 2(NA - R-A - o.55 x R/1000).
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
was assumed previously; therefore, the safety factor 2x appears
in that context-also. For the time being, further design consi-
derations will be based on the results obtained by design approach
5, Tables B-6 and B-7. An attempt has been made to keep the focal
lengths of the objectives at round numbers so that the possibility
of using commercially available lenses is not blocked. (This
explains why other numbers appearing in Tables B-6 and B-7 are not
round numbers.)
In order to complete the discussion, two further approaches which
were discarded-have to be mentioned.
The first(design approach 6, Table B-8) serves to answer the question
whether a de-magnifying zoom system (m'z - 1.0 ->1/3.5) or a mag-
nifying zoom system (m'z - 3.5---31.0) should be chosen. (Design
approach 6 is comparable to design approach 2, Table B-3). Design approach
6 requires, of course, objectives with a smaller, initial magnification,
mlObb, because an additional magnification will be obtained by the zoom
system. The magnification of 6x, 3x, and lx have been chosen tents-'
tively for m'Obj+, utilizing only m'z - 3.33--4 1.0 for lenses Nos.
1 and 2. Since resolution requirements remain the same, the numerical
aperture, NA, or the corresponding F/number, has to remain the same
for each lens as on approach 2. The resulting lenses (column 10
are then very bulky,and, therefore, expensive. The fact that the
maximum field angle will be reduced from 20(max) - 6.85? (approach 2)
to 20(max) - 1.92? (approach 6) does not weigh very heavily, because
B - 13
Approved For Release 2003/05/15: CIA-RDP78BO4747AO02700040006-3
 .
pproved For Release A'03/05/P :'CIA- DP78E4747A902700A 0V-006-31'-'
l
2
3
4
5
6
7
Obj .
(Objective)
m'Z
m' C
Object Field
Object Sp.
Resolution
Absol
t
Magnification
(Zooo)
(Combined)
Diameter
2y [mm!
Numer. Ap.
NA - sind(
Required
(lines!mm;
u
e
Resolution;
Li
Range
R
nes Over
Object Field
1)
6x
3.33
20x
0.9
0.55
1000
900
200 -.60
1.0
6x
3.0
0.238
431
1295
2)
3x
3.33
lOx
1.8
0.360
655
1180
100 - 30
1.0
3x
6.0
0.1265
230
1380
3)
lx
3.5
3.5x
5.14
0.1463
266
1365
3
1.0
1x
18.0
o.044 80
1440
Magnifying Zoom System m'Z
2y' = 18 m diameter of primary image
fint = 540 mm
OPTICAL PARAMETERS FOR INPUT SIDE OF VISUAL OBSERVATION SYSTEM
f0bj [ m1
F/Number
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
Prim. Image
Numer. Ap.
NA' - sind(
Aperture
Angle 2a'
at Prim.
Image
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
this decrease of field angle does not appreciably help the lens
designer, as compared to the difficulties he encounters in design-
ing an F/0.9 lens with a field of about 7 degrees and a focal length
of about 25 mm (wavelength will be the same in both cases but lens
diameters will change). Another negative aspect of this approach
is the fact that the lx objective with a focal length of f0bj. =
540 mm has to. be integrated into the system, This will obviously
make the system much bulkier. These are reasons enough to drop this
design approach.
Design approach 7, Table 9 will serve to illustrate the possibilities
of a system with only one objective lens, but a zoom system with
which the entire range of m'x - 1.0 1/20 can be covered. In order
to be comparable to design approach 2, this approach is based on the
same optical parameters as approach 2, in particular a focal length
fObj. = 27 mm and a diameter for the primary image 2y' - 18 mm.
Table 9 shows that the objective lens has to cover an extremely large,
range, namely, a change of field from
2y = 0.9 > 18.0 mm
or a corresponding field angle of
20 = 1.84? > 36.8?
while the numerical aperture changes from
NA - 0.55 ---> 0.044
or the corresponding F/number from
F=0.9--> 11.4.
It is very doubtful that the design of such a lens is possible in
view of the high resolution requirements, or even that it would be
practical if the contracting agency is lenient with respect to com-
plete fulfillment of, resolution requirements. Reasons for this
doubt are f6efi pvA F E2ekWe 20QA(Pg/45 fsCl6-ff7JN47470 experienced
B - 15
 Approved For Release 2003/05/11":CIA-ICP78BI4747A#27000L006-31
2
3
4
5
I 6
7
8
9
10
Obi
m'
(Objective
)
Z
m
C
Object Field
Object sp.
Resolution
Absolute
Prim. Image
Aperture
fObj.[mtn
Magnification
(Zoom)
(Combined)
Diameter
2y Imml
Numer. Ap.
Ni - sand
Required
R[lines!mm:
Resolution;
Lines Over
Numer. Ap.
'
'
Angle 2a'
F/Number
Range
NA
- sins
at Prim.
a
Object Field
Image
20x
1.0
2(k
o.9
0.55
1000
900
o.0275 I
3.15?
27.0
200 - 10
1/20
lx
18
0.044
80
1440
0.044
5.04?
F/o.9
Demagnifying Zoom System with m' z - 1.0- -01/20 diameter of primary image 2y' - 18 mm
OPTICAL PARAMETERS FOR INPUT SIDE OF VISUAL, OBSERVATION SYSTEM
Design Approach
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
 Approved For Release 2003/05/15 : CIA-RDP78B04747A002700040006-3
lens designers, that requirements for small field and large numeri-
cal aperture run counter to the requirements for large field and
small numerical aperture. Of course, certainty about this point can
be obtained only by actually working on the design of a lens of this
kind and seeing what results can be obtained from a methodical and
diligent study. However, it is felt that such an approach. deviates
so much from established practice and proven experience that the effort
should not, be undertaken.
There are other points raising doubts in connection with approach 7.
If a one-step zoom system is used, it will again be difficult to
design it (optically) so that-it will in no position impair image
quality. Usually, zoom lenses are farther from the diffraction limit
than conventional, non-zooming photographic lenses. It is very diffi-
cult to minimize aberrations, so that they are negligible, in an opti-
cal system where. the position of the lenses changes. The difficulties
become greater as the position changes, i.e., the zoom ratio, becomes
larger. Also, the large maximum field angle which now appears makes
it advisable to place the zoom lens as close to the objective lens as
possible, otherwise the optical cross section will become impractically
large. But then, other system components cannot be placed as near to
the objective lens as is found desirable, in order to make them insen-
sitive to changes in the optical train, such as the beamsplitter to
bring in the floating dot, beamsplitter for the correlation, and the
image rotator.
A two-step zoom system could be considered, the first step behind the
objective lens, the second step combined with the observation station,
namely, the ocular. But if the zoom-system, or part of it, is inte-
grated with the ocular, the anamorphic eyepiece cannot be placed in
the ocular any more.
Approved For Release 2003/05/15 BCIA-~Q P78B04747A002700040006-3
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
There are other difficulties involved with either a one-step 1:20
zoom system or a two-step zoom system each 1:4.5. Even if these
difficulties do-not make an approach of this kind absolutely
impossible, they will certainly require compromises with other
design details, which are just as important for the overall system
design and its performance.
This discussion leaves design approach 5 (Tables B-6 and B-7) as
the one with the highest promise for a good hardware design. Figure
B-l outlines the arrangement of optical parts for the "input station".
A small compromise was found desirable;'the free diameter of the
intermediate lens was held at 50 mm and that of the negative lens at
52 mm as compared to a full field diameter of 53 mm for the 3.5x
objective. This compromise is necessary to keep the required optical
path through the K-mirror short, but does not seriously impair optical
performance in view of the safety factor of 2.0 in calculating the
required aperture. The beamsplitter cube has been increased to 58 mm
(38.4 mm optical path) to insure good performance.
Approved For Release 2003/05/1 B: CIAAR8DP78B04747A002700040006-3
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-s~MAX,
/d0 FF ~^I I I F+
+~i 444..---~t'CF
VISUAL OBSERVATION SYSTEM INPUT STATION
Approved For Release 2003/05AWU-F3F78B04747A002700040006-3
B - 19
-T-f
L-/B P?9/MARY /MAQE
AND CF
FF ^ FULL F/#LD
CF ^ CENTER F/ELD
FR ? RANGE' FUR F/NE IM&SS/NQ
AD ? APERATURE D/APNRAQM
EN ^ EfVTRANCE P//P/L
N; N ^ FYw t'/PAL PUlN?
GPT - OPT/CAL O/STANCE (H//'/Y/LL #,Q 717 BE CONS/DEREp)
1WTE:.4YL O/MENS/ONS IN M/LL/METERS
--46IFF
i
1 4 /6 MAGE FF _
Z. 70
 Approved For Release 2003IpIPI?N1 f -RJ P78BO4747AO02700040006-3
RETICLE DOT SUBSYSTEM
C 1. General Arrangement and Design Considerations
It is necessary that the floating, or reticle, dot, F.D., be
placed as close to the objective lens of the input station
as possible. Insertion of any additional optical elements
between these components carries the danger of influencing the
position of the reference dot by movements of either the frame
to which the element is mounted (because of temperature effects
or changing loads) or, if the optical element is not rigid, by
undesirable but unavoidable imperfections in its moving mechanism.
The latter applies particularly to the zoom system. It does
require optical elements moving along the optical axis (i.e.,
longitudinally), and a reasonable tolerance for lateral movement
accompanying the necessary longitudinal movement must be estab-
lished because of imperfections in the track and moving mechanism.
If the floating dot is brought into the optical train before it
enters the zoom system the floating dot will participate in the
resulting movements of the image thus serving as a reference
fixed relative to the image.
This is the main reason for insertion of the floating dot imme-
diately behind the objective lens, into the parallel light bundle
between objective lens and the intermediate lens (see Optical
Layout, Appendix B ). The inevitable price which has to be paid
for this approach is that the size of the floating dot has to
change not only by the specification ratio of 1:8 (between 0.5 and
4.0 min of arc) but also by the zoom magnification ratio of 1:3.5.
The total variation in reticle dot size therefore must be 28:1.
In theory it would be possible tb reflect the floating dot through
the objective lens upon the film, utilizing the scattering power
Approved For Release 2003/055/15 :1 IA-RDP78BO4747AO02700040006-3
 Approved For Release 2003/05/15 : CIA-RDP78B04747A002700040006-3
of the film emulsion to bring the floating dot image back.
through the objective lens into the observation: station. How-
ever, this idea has been discarded. A strong argument against
it is the fact that the floating dot image would then depend too
much upon the scattering capability of the film, which varies con-
siderably with film density (size and number of grains in the
emulsion). The floating dot would be'almost invisible in clear
areas of the film where a high dot intensity is required because
of the high background brightness in the film. Also, this approach
does nothing to prevent motion of the floating dot relative to the film
if the frame to which the floating dot mechanism is attached is
not rigid. The relative movement of the floating dot because of
an insufficient rigidity of its frame will be exactly the same if
the floating dot is reflected directly into the optical train
(without the detour over the film) as indicated in the Optical Layout.
The best protection against such dot movements is to mount the
floating dot mechanism into the most rigid part of the instrument
.available for this purpose, that is the yoke carrying the "input
station" and holding it in place over the film table. This approach
does not guarantee absolute freedom from a relative motion between
floating dot and the image of the film area; it only minimizes this
danger. Extreme care will be needed in the mechanical design to
hold the structure within the required tolerance.
There are several approaches possible for changing the apparent
size of the floating dot image. The idea of using an optical zoom
system like that for the VOS has been discarded for two reasons.
First, the required zoom ratio of 1:28 is too large to be practical.
Second, use of zoom lenses would make the instrument sensitive to
possible inaccuracies of the track on which these lenses have to
move. Because the floating dot should be used as a 'reliable pri-
mary reference point, it was considered to be better if this source
Approved For Release 2003/@5/j5 2 CIA-RDP78B04747A002700040006-3
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
of possible error is avoided. This leads to the selection of a
rigid optical system.
The optical system proposed will consist of a high quality
telescope objective T.O. (see Figure C-1) of (at least approxi-
mately) the same focal length as the intermediate plus negative
lens system in the VOS. Its distance from the floating dot beam-
splitter, F.D.BS., is variable within a comparatively large range,
because it images the floating dot, F.D., into infinity. Thus,
the light between T.O. and beamsplitter is'"parallel", in agree-
ment with the general design concept. A 90 degree double mirror,
M1, M2, folds the rays back toward the main tube. The floating
dot, F.D., is formed in the focal plane of T.0'. It is the demag-
nified image of a diaphragm, Dph., the nature of which will be
discussed later. The demagnification is obtained by a commercially
available microscope objective, M.O., a so-called.apochromat
corrected for use without cover glass and a so-called negative
ocular or projector, Pr. The latter is used for highest quality
micro-photography.
STAT
The diaphragm will be illuminated by a lamp, L.S., such as a "Con-
centrated Arc" lamp lor a Zircon-arc lamp, and a con-
denser, Cond. The condenser is divided into two parts with parallel
light between them. A gray wedge, G.W., and if necessary, addi-
tional gray filters are placed in this area for a wide and contin-
uous variation of the "intensity", or better luminance, of the
floating dot image. The gray wedge requires, of course, a small
fixed counter wedge with equal wedge constant in order to equalize
this light distribution over the pupil area. The nature of this
gray wedge will be discussed later.
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
 t
Approved For Release 2003/05/15 :CIA-RDP78B04747A00270004006-3
R. D.
F. D.
Irt. L
M.0,
RETICLE DOT OPTICAL SYSTEM
Figure.C-I
Dph..
_ L.S.
IA
i
ru N
CohaL
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
The exact adjustment of the floating dot in X (parallel to the
eye base line) and Y (perpendicular to the eye base line) is
very important and critical. If the two dots of both optical
branches differ only little in Y, the observer is unable to
fuse these two dots into one dot conceived as floating in space.
'A difference in X determines the parallax between the two dots,
which is equivalent to the apparent distance of the floating dot
in the three dimensional picture,. or its Z coordinate. Because
of the importance of this adjustment, it should have easy access
and should be easy to handle in order to check and readjust the
floating dot position whenever it is deemed necessary. It might
even be advantageous to attach a measuring scale to the adjustment
device in order to be able to compare several adjustments and to
reset a previous adjustment.
It will be shown below that a very large demagnification will
be applied to the diaphragm in order to assure high quality in
the image. This demagnification relaxes the accuracy requirement
for X and Y positioning of the diaphragm and conventional mechanical
design techniques should suffice. However, because there will be a
need. for a mechanism to change the size of the diaphragm, it may be
found desirable in the mechanical analysis to separate these three
adjustments.
If this is found to be the case, it is possible to provide an
optical adjustment which is sensitive and precise and located in
the space between T.O. and the beamsplitter. The best means for
an adjustment of this kind would be a rotating double wedge, DW, one
for the X- and one for the Y-direction, DWX and DWY.
Approved For Release 2003Pb5t155: CIA-RDP78BO4747AO02700040006-3
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
A double wedge consists of two prisms with a small prism angle
mounted so that they can be rotated in opposite directions. In
the zero-position, the two wedges supplement each other to form
a plane parallel plate, which has no effect upon the quality of
the dot image since it is placed where the light rays are paral-
lel. In the position of maximum deviation, the two prisms supple-
ment each other to form a prism with an angle twice the angle of
one prism and a corresponding maximum deviation. This double wedge
is a.sensitive and precise device for measuring, or adjusting, small
angle deviations, which became popular for the first time through its
application in the range-finder of I Icameras. STAT
It would, of course, be necessary to achromatize these prisms, but
this is not a serious problem. It would theoretically suffice to
have only one double wedge and to rotate the two prisms either
counter (to determine the angle of deviation) or together (to move
the point on a circle), but this is discouraged. A separate adjust-
ment in X and Y without cross-talk between them seems a much better
solution for this purpose. Therefore, use of a separate double
wedge for each of the two adjustments is anticipated.
If the X-adjustment is provided with a remote control from the
place of the operator and with a remote readout, it could be used
to measure the difference in Z between two (spatial) image points
in the observation station. It is probable that higher measure-
ment accuracy for the difference of the X. Y, and Z coordinates
of two points could be obtained by the double wedges than by table
motion, provided the two points are visible within the field of
view so that they can be measured with fixed table positions. This
would make it possible to determine the dimensions of certain de-
tails (buildings, vehicles, etc.) with a higher accuracy than points
of larger separation.
C - 6
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
This method could be further improved by splitting the floating
dot into two dots, one with only small adjustments in the center
of the field, the other one with free movements throughout the
field of, say, the 9x objective. This optional device will,be
called "Optical Stereo-Micrometer". (If time permits, a quanti-
tative analysis will be given later.).
To implement this idea, the space between T.O. and F.D.BS. is
again utilized. The light path is split into two branches (see
Figure C-2) by a mirror arrangement similar to that used in
interferometers. (This deviates from the details given in
the proposal; but the approach here is much better.) Each branch
is then equipped with two double wedges, in one branch with only
small deviation angles for the center spot, in the other branch
with larger deviation angles for the dot moving in the field of
view. This device could be used by placing the center dot, say,
on the top of a vehicle, and the other dot on its eight corners.
In this way the dimensions of the vehicle might be obtained inde-
pendently of, and with a higher accuracy, than with the metering
system of two tables moving relative to one fixed floating dot.
IC 2. RETICLE DOT OPTICAL ANALYSIS
If the focal length of T.O. is assumed to be f = 600 mm, a disc
of one minute of arc viewing angle has a diameter of
600 x 0.003 = 0.18 mm
However, this will be magnified by the ocular in the observation
station, which is assumed to have a magnification m' 0C = lox. So
the value above corresponds to 10 minutes of arc. The range has
to cover angular values of 0.5 to 14 minutes of arc because of
the possible demagnification of 3.5x in the zoom, system or for the
F.D. diameter:
d' (F.D.) = 0.009----40.252 mm
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
C - 7
STAT
STAT
 At pproved tFor Re - t a
Release 2003/05/15 : 'CIA-RDP78B4747A00270000006-3
OPTICAL STEREO-MICROMETER
Figure C-2
Approved For Release 2003/05/15 : CIA-RDP78B04747A002700040006-3
 00
STAT
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
which has to be obtained through demagnification by the micro-
scope system mentioned above from a conveniently larger dia-
phragm.
It has been specified that the dot has to be a perfect circle
at all times with a sharp edge gradient. It has to be kept
in mind, however, that the dot is so small that inevitably dif-
fraction effects will come into play. The angular diameter, 6,
of the first minimum of an iI disc follows from
f
oc
? 6 = 2.44 x TINA
where NA = numerical aperture in the image plane of the ocular.
Assuming NA = 0.044, f = 25 mm (m' = lOx), and T = 0.55 x 10-3 mm
oc oc
6 = 6.1 x 10-4[rad] = 2.1 min. of arc.
If NA becomes smaller, 6 will be correspondingly larger. If,
,however, the influence of aberrations is comparable to diffraction
effects, first the light distribution in the disc will change,STAT
then, as the influence of aberrations increases, the Airy disc will
be smeared out more and more.
If the stop which serves to produce the F.D. by demagnification
corresponds to just about the size of an Airy disc, small deviations
from perfect roundness such as those resulting from the lamellas of
an iris will not be visible. The deviations from roundness will not
be resolved by the eye on account of geometrical optical principles
if irregularities are negligibly small as compared to the equivalent
of one minute of arc. Diffraction effects are also negligible
since the diffraction pattern of a many sided polygon approaches
that of a round disc. (See, for instance, Born-Wolf; Principles of
Optics, Section 8.5.2, footnote at the end of the first paragraph.)
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
C - 9
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
This was tested experimentally, found true, and can be demon-
strated. Also note: a starlike diffraction pattern is obtained
not by small deviations of the fieldstop from roundness, but if the
aperture diaphragm in the imaging lens is a distinct polygon (such
as a triangle or square).
.Since the dot is perceived stereoscopically, its position accur-
acy should be compatible with the depth or parallax resolution of
the human eye, which is about ten times as good as the acuity of
the eye (one minute of arc under optimum seeing conditions).. The
position accuracy should, therefore, be better than one-tenth of
the equivalent diameter for one minute of arc.
If we assume that an iris diaphragm is acceptable, a precision
diaphragm of not too small a diameter, manufactured by a well
reputed firm, should be chosen: say a diaphragm between
d (Dph) - 2.5- ..)70 mm,
so that the diameter corresponding to one minute of arc is 5.0 mm.
The centering accuracy would then have to be = 0.5 mm or better.
This seems attainable, though it might require careful laboratory
checks of the products of several manufacturers.
The microscope demagnification required for these dimensions is
m' (micr) = 2.5/.009 - 278x
with an apochromatic objective of N.A. = 0.30 and m' = 12.5x'
or instance), a magnification of 22x remains for the
projector. Since apochromats corrected for use without cover glass
are mostly those designed for metallographic microscopes which, in
most cases, are infinity objectives in combination with an inter-
mediate lens system, it will be wise to anticipate use of this
kind of microscope optics. This makes it possible to restrict
the N.A. to the required value of N.A. = 50/600 - 0.0835 by placing
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
C - 10
STAT
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
an aperture diaphragm between objective and intermediate lens
(as close to the objective as feasible). This aperture dia-
phragm has to be perfectly round and well centered.
C - ll
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
 Approved For Release 200~MND fIABRDP78B04747A002700040006-3
OPTICAL CORRELATION
D 1. Theory of Operation
The operating principle of the optical Correlator is illustrated in
its simplest form in Figure D-1 . In this figure two transparencies,
separated by an arbitrary distance, are placed along a common optical
axis in front of a diffuse light source. An infinite number of bundles
of parallel rays emerge from the light source and penetrate the two
transparencies. If a lens is placed as shown, each of the parallel
bundles of.rays is brought to a focus. in a plane, called the, detection
plane, located at the focal length of the lens.
Because of density variations in the transparencies there will be a
distribution of light and dark areas in the detection plane. If an
extreme example is taken, in which one transparency is a negative of
the other, and if the photography is of such high contrast that trans-
mission is limited to either zero or 100 per cent, then it can be
recognized intuitively that there will be one particular bundle of
parallel rays which will be blocked either at transparency #1 or trans-
parency #2, and therefore there will be one point in the detection plane
which will be completely black. This is the correlation peak, and its
location with respect to the. optical axis gives. the X and Y displace-
ment of the transparencies relative to each other. This location can
be detected by an appropriate sensor. The above example is illustrative
of a "quotient match". If the transparencies were identical copies,
rather than a positive and negative, the correlation peak would con-
sist of a bright spot, and this would be called a "product match".
In cases of practical interest the transmissions are not restricted
to zero or 100 per cent, and for stereo pairs the imagery is not
identical. While the principle of operation remains the same, these
factors combine to reduce the contrast of the correlation peak with
respect to the background. That is, the signal-to-noise ratio is
degraded from that in the idealized case.
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
D-1
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
DETECTION
PLANE
TRANSPARENCY
#1 (T1)
DIFFUSE
LIGHT
SOURCE
LENS
(FOCAL LENGTH - f)
NOTE: Diagram assumes equal scale for T1,.T2. Parallel rays which
penetrate corresponding points of T1, T2 will produce the most
prominent point.in the detection plane. If T2 is smaller scale
than T ., the rays of interest will be convergent, and the detec-
tion plane will therefore translate with respect to the lens.
BASIC CORRELATOR CONFIGURATION
.FIGURE D-1
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
D - 2
 Approved For Release 2003/05/15 : CIA-RDP78B04747A002700040006-3
In practical cases it is found that any slight rotation of one
field relative to the other from the exact registry position
will cause the spot to disappear. A means of improving signal-
to-noise ratio takes advantage of this phenomenon. One trans-
parency can be rotated slightly relative to the other periodically
to cause the spot to pulsate, and this variation can then be picked
out of the relatively constant (DC) background by synchronous
phase detection. This process is called dithering and can be
accomplished either mechanically or electronically.
The photo position sensor chosen
This device was developed as a star tracker
and will indicate the relative position of a spot of light (or
darkness) on the photocathode.
D 2. Laboratory Models Implemented
The first-working optical correlator implemented is a simple
transparency-to-transparency matcher. This device is sketched in
Figure D-2 and presented in the photograph of Figure D-3. Both
fields are presented on film transparencies and illuminated by a
diffuse incandescent light source. The lens focuses the correla-
tion spot and immediately surrounding field onto a photo sensor
cathode. The photo sensor employed F I
The upper transparency is
mounted on a X. Y, and 8 Transit Table. All table drives are
manual. Dither was introduced for data taking purposes by rotating
the top field (in the 6 sense) about 5 degrees. This was enough
to make the spot disperse both to the eye and
The out-
puts F were fed into icrovolt-
meters. Readout was from these meters in microvolts. X and Y
movements required to generate good match curves seldom exceeded
STAT
STAT
STAT
STAT
STAT
STAT
STAT
STAT
Approved For Release 200305/153: CIA-RDP78B04747A002700040006-3
 Approved For Fas
/~W/T1gN CIA-RDP78 INTENSITY
3S 00400
ZFf&,?Pc37P 9 Y POSITION
Ju VOLT-
METER SP09
RELATIVE
POSITION
IND.& REC
X AXIS OUTPUT
1
INCA ES-
CENT
FILAMENT
T,IG
IGH1 5015 CE
)u VOLT- VOLT-
METER METER SPO SPOT REL -
RELATIVE TIVE P0-
INTENS IT S IT ION
RANGE
StRE
IND. REC.
Y AXIS OUTPUT
a------s
TOTAL INTENSITY OUTPUT
SPOT INDICATIVE OF CORRELATION ON PHOTOCATHODE
POSITION-& INTENSITY RTT OR QUAD. PM OR
SENSITIVE PHOTOSENSOR (COMPETITIVE DEVICE
JOINT FOCAL PLANE FOR SPOT PICKUP
TRANSPARENCY #2 NEGATIVE IN QUOTIENT MATCH
POSITIVE IN PRODUCT MATCH
TRANSPARENCY #1 POSITIVE IN BOTH QUOTIENT AND
PRODUCT MATCHING
DIFFUSER
OPTICAL CORRELATOR SCHEME: TRANSPARENCY VERSUS TRANSPARENCY
FIGURE p 2
Approved For Release 2003/05/15 : ClA-RDP78B04747A002700040006-3
D - 4
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
+ 1.5 mm. response to this system is good. This is due to the STAT
fact that theI (natural spectral response is very similar to the STAT
emission spectra of incandescent (2900?K) tungsten.
In order to get a configuration applicable to the Stereo Comparator,
aCRT-to-transparency matcher was implemented next. Figure D-4
presents the general idea, Figure D-5 presents a combination block
and assembly diagram, and Figure D-6 is a photograph of the equip-
ment.
This device functions so as to compare like fields of information
but at different sizes or scales. The first transparency is scanned
by a flying spot scanner and its imagery is presented in the face
of the CRT. The imagery on film transparency to be correlated is of
smaller size (i.e., scaled down) and is placed some experimentally
determined distance away in register on the optical axis. The region
near the apex of the pyramid developed will contain the spot. The
viewer screen I his placed in this region.
STAT
The #2 transparency (on 35 mm film) was held. on a transit table
having X, Y, and 9 movement capability. Dither can be introduced STAT
either mechanically (via 9 mode of table) or electronically (pulsed
X and Y deflection voltages). The outputs are fed into TAT
I I r
microvoltmeters for readouts. Table movements seldom STAT
exceed + 1.5 mm for good match data generation. That is, lock-on
for correlation is about + 1.5 mm from true correlation point.
Rotation required for effective spot dispersion (for dither) is about
5 degrees.
response to the best of CRT phosphors slow) for this
purpose is not good as compared to incandescent tungsten illumina-
tion. Everything else being constant, the illumination level of
the display must be about fifty times that of the incandescent
tungsten 4rfdff@*o$4M et:005M F: P14 904MMOOEM0416MI out
STAT
STAT
STAT
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
X AXIS POSITION SIGN
A i.
TOTAL INTENSITY SIGNAL
Y AXIS POSITION SIGNAL
PHOTO-POSITION-SENSOR
GWjIC
DITHER
X ADJUST
RASTER
S IZE
CONTROL
Y ADJUST
MATCH
(POSITIVE IN PRODUCT
MATCH
NVERTER OR
TRANSPARENCY #1
TRIGGER
INSERTED
IN VIDEO
TAMP
JOINT FOCAL PLANE FOR SPOT PICKUP
TRANSPARENCY (#2) FOR CORRELATION. (NEGATIVE .IN QUOTIENT
MONITOR
CON RAC
GENERATOR I
OPTICAL CORRELATOR SCHEME: CRT TO TRANSPARENCY
FIGURE D-4
Approved For Release 2003/05/1: CIA? DP78BO4747AO02700040006-3
DISPLAY OF IMAGERY ON FACE
(CAN BE NEGATIVE OR POSITIVE)
STAT
STAT
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
MICROVOLTMETER
MICROVOLTMETER
MODIFIED
PATTERN GE
STAT
8
LO V.
A.C.
LINE
INSERT
REGULATED
HV POWER SUPPLY
REGULATED
POWER SUPPLY
REGULATED
POWER SUPPLY
REGULATED
POWER SUPPLY
REGULATED
POWER SUPPLY
SOLA REGULATOR
TRANSFORMER
TI DISPLAY ON CRT
EQUIPMENT ASSEMBLY, FIRST CRT TO TRANSPARENCY CORRELATOR
FIGURE D-5
Approved For Release 20090571?: CIA-RDP78BO4747AO02700040006-3
STAT
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
 roved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
In view of this fact, an image intensifier tube and
associated power supplies would be required for working with small
imagery and high density films. Figure D-7 is an assembly drawing
iW
which suggests how this might be implemented.
STAT
An improved, more versatile version of the CRT-to-transparency
type Optical Correlator is presently being completed. It provides:
(a) variable raster size, (b) automatic variable frequency electronic
dithering, (c) small area vernier scanning (manually, programmed),
(d) a choice of scanning techniques (i.e., standard television, lissa-
jous, triangular, random, etc.), (e) variable frame rate, (f) automa-
tic dither demodulation and readout, (g) contrast and edge enhance-
ment, (h) precision picture stability circuitry. It is illustrated in
Figures D-8, D-9 and D-10.
D 3. Performance and Data
The performance of the optical correlators was largely dependent
0-r AT
T
upon the response Ito the spots generated. This appears
to be a function of the illumination source emission spectrum. It
was found that a blue or green peak sensitivity range ^was not STAT
available. The present near IR peak (8400?) is the best available,
giving excellent response to incandescent tungsten illumination. STAT
The phosphor best matching (in the emission spectra versus STAT
spectral response sense) is the slow F-1 This is also the best in the
the fact that very high brightness is easily obtainable from the
type film. The imagery was illuminated by diffuse incandescent
tungsten (2900?K) light. A positive. transparency was compared with
STAT
The first data was taken using the transparency versus transparency
apparatus of Figures D-2 and D-3 . Sample data from this apparatus
is presented as Figure D-11. This is a presentation of a series
.of match curves about the true match point. The imagery was syn-
thetic (random checker pattern) on (second generation) STAT
Approved For Release 2003/0/1-5 -ICA-RDP78B04747A002700040006-3
 Approved For Release 2003/05/15 : CIA-RDP78B04747AO02700040006-3
Y AXIS SIGNAL
}. A AND TOTAL INTENSITY SIGNAL
X AXIS SIGNAL
STAT
STAT
STAT
STAT
in
FACE (30 FOOT CANDLE AVERAGE STAT
ILLUMINATION LEVEL'OVER 2.2
SPOT AND SPOT AREA DISPLAY
SQUARE INCH)
--r`-IMAGE INTENSIFIER TUBE G 60
(OR EQUIVALENT)
'L
SPOT PICKUP
-- - -I ``- Ti DISPLAYED ON CRT FACE (P4 OR
Am,. ILLUMINATION LEVEL)
VERT.
STAT
OPTICAL CORRELATOR CRT VS. TRANSPARENCY
WITH IMAGE INTENSIFIER
FIGURE' D-7
D - 11
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
OVER 2.2 SQUARE INCHES)
[ ~-~ ~----vT2 ON X, Y, 6 TRANSIT TABLE
HOTOCATHODE (MAXIMUM INPUT = .5 FOOT CANDLE
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
Next 1 Page(s) In Document Exempt
Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
 Approved For Release 2003/05/15 : CIA-RDP78B04747A002700040006-3
STAT
STAT
Ju
VOLT -
METER
(9)
,fix VOLTMETER (X)
X SWEEP
GENERATOR
Y SWEEP
GENERATORI
VIDEO AMP
Ja VOLTMETER (Y)
ELECTRONIC DITHER
CIRCUITRY
X..Y
TRIGGER EDGE ENHANCER
AND DC RESTORER CIR-
CUITRY
PM TUBE P.S. DRIVER
STAT
ON CRT
Z AXIS
INPUT
r ORI
AMP
VER1
MEASURING ENGINE FSS AND ENHANCING
AND READOUT ELECTRONICS ELECTRONICS
GENERAL PURPOSE OPTICAL.CORRELATOR ASSEMBLY
STAT
STAT
STAT
Approved For Release 20031;361 k lR- 78B04747A002700040006-3
D - 14
 80
70
60
50
40
30
20
10
0-
-10
-20
-30
-40
-50
-60
x-70
-80
-90
-100
-110
-120
a
I
C-~ t 1.
rt
!#
I
.
F~L
-
f-
Rid
I
I
A PC
Oil
j
~
1
r
..
4
_
'_ la
r
~
11DR?
._ _...
_ 1 _.
oy
_ I
~~
".
II
_ -
I
,._.
._. _..
I. .,
._I_ .
C.
.._ ..
11
.
...
_.~
.. i..,.
..L ..
i I.
.~ t I~.
t. { ,
..,t._
-Mr-
T
~
f
{{
r
t
}
~
I
tf _ _
+..
_ ..
.. _ ._ . _.
,. ^: .~ _ ~~ ..
? a ...
.
::..
r__I
~~:
~
. X~:.
t
I
-I J
_
II
I
T
_ -I
Fl-
T
f
I
1
r
11
+
~ 1
I A
+
J
fl-
rr
I
HL
t
t i_
11 1
1 1
'
{
i0N
T
t
{}
w.
f
Ln
u
. M. P.. f
E
T-T
'A
I
_
_
I
I
A6,
W-M
 Approved For Release 2003/05/15 : CIA-RDP78B04747A002700040006-3
its positive mate to produce a bright spot (product match). The
top transparency was mechanically and manually dithered in order to
obtain LVy data. The best approximation of the combined X, Y and 9
match (correlation) point was found by eye. The X axis drive was
locked and displacements on Y axis (+ 1.6 mm) and 9 sense ( 5?) were
made. Data was taken at various intervals in Y displacement at
two angular positions (0 = Oo, and 9 = 5?). AVy quantities were
computed by subtracting the Vy reading with dither from Vy with
no dither at each point of axial displacement:
AV -V -V
Y yo y9
In most instances readings were taken every .2 mm increment in
Y displacement. This dither data (iVy vs. Y) is plotted for six
different fixed X positions, five more or less off the X match
position. One curve falls very close to being on the position of
maximum correlation (i.e., true match position). This is indicated
by the - a-a-line in Figure D-11. It is interesting to note that
this curve possesses the greatest excursion, the greatest relative
sharpness, and crosses the Y axis very near the origin. It also
features nearly symmetrical false match point saddle regions.
Figure D-12(a) is a photograph of the imagery.
Results of a study using real imagery with the first field presen-
ted on a CRT and correlated with a second field on a scaled down
transparency is presented graphically in Figure D-13. Both trans-
parencies were on (non-diffusive, very high contrast) STAT
film of a second generation. Figure D-12(c) is a photograph of the
imagery which is constituted by fields, hills and valleys. This STAT
imagery was scanned and presented on. the modified CRT STAT
face. The apparatus used is illustrated by Figure D-4 and asso-
ciated written description. The display was compared with its
negative mate to produce a dark spot in quotient match. The dither
output (i.e., no dither-dither) readings are plotted versus lateral
Approved For Release 2003/05/15 : CIA-RDP78B04747A002700040006-3
D - 16
 Approved For Release 2003/05/15 : CIA-RDP78B04747A002700040006-3
Approved For Release 2003/0p/I11, fr(A R181304 47AO02700040006-3
(a) Random Checker Pattern
(b) River Scene - Stereo Pair
(c) Fields, Valleys, Hills
SAMPLES OF IMAGERY STUD I
 $o
-4.0
N -1-0-4 't'r.TZGE ?'RAPH '+-~'R s('E
pOU ry 7000A0006-3
Approved For Release 2 03/05/15` IA-RDP78B 4747At 2r N C``'~
DITHER OUTPUT VS. DISPLACEMENT OF IMAGERY
QUOTIENT MATCH
0
A roe' For Relea a 003/05/1 : IA-RDP78 0 47A0027pp0 0006-3
-3.0 pp '.~f I. ~ ~.` Z. (~ 3.0 4.0 5.0
FIGURE D-13 MM DISPLACEMENT
 Approved For Release 2003/05/15 : CIA-RDP78BO4747AO02700040006-3
(Y) displacement. Each curve represents a different CRT bright-
ness.
Stereo pair mates with about 60 per cent common imagery were
compared next. The results are presented as Figure D-14. Nega-
tive imagery on the CRT face was correlated with scaled down
positive imagery on a film transparency. The apparatus of Figure
D-4 was used. Since this was quotient matching, a blgc spot
was produced. Some penalty in performance is inherent in this data
due to the facts that the original imagery was (a) at least second STAT
generation, (b) on diffusive emulsion, and (c)' converted to
copy was scanned and displayed on the CRT STAT
STAT
face. It was compared to its stereo mate also on Ifilm in
STAT
quotient match. The correlation dark spot was generated within
lock-on limitations imposed by the nphotocathode face, i.e., STAT