NASA TECHNICAL NOTE NASA TN D-6849

C-./
--1
m

COPY: RE'
�
,...l
AFWL (001� i
KIRTLAND AFE! � !
a-=
lr1 .>
;��;
1:-' � "11
CD
z
==

APOLLO EXPERIENCE REPORT -

LUNAR MODULE LANDING RADAR
AND RENDEZVOUS RADAR ,_. , r\::�:�� ·�.�- -. lr • .· ._/ ,--
/�··.' . ' ,-

· -. ·
I\

by Patrick Rozas and Allen R. Cunningham / ,'

_
_:
J • . . '·

Manned Spacecraft Center ·� . . .

Houston, Texas 77058

NATIONAL AERONAUTICS AND SPACE ADMINISTRATION • WASHINGTON, D. C. • JUNE 1972

, ;
1
..��
I
I
r
 TECH I,.IBRARY KAFB, NM

llllll llllllllllllllllll l llllllllllllll �II

[]133651

7. Author(s) 8. Performing Organization Report No.

Patrick Rozas and Allen R. Cunningham, MSC MSC S-311

·-
10. -·--·-
Work ----- -··---�-----;
Unit No.

9. Performing Organization Name and Address 914-13-00-00-72

Manned Spacecraft Center
Houston, Texas 77058
�Co--;;- ;;�-;:-
t c
-
G ;;nt
·
No.

-------1� -
Type of Re or P" t andP;riod Covered

12. Sponsoring Agency Name and Address

Technical Note
National Aeronautics and Space Administration ----··- - --------;
Sponsoring Agency Code
Washington, D. C. 20546

15. Supplementary Notes

The MSC Director waived the use of the International System of Units (SI) for
this Apollo Experience Report, because, in his judgment, use of SI Units would impair the usefulness
, of the report or result in excessive cost.
-----------�
16. Abstract

A developmental history of the Apollo lunar module landing radar and rendezvous radar subsystems
is presented. The Apollo radar subsystems are discussed from initial concept planning to flight
configuration testing. The major radar subsystem accomplishments and problems are discussed.

1----- ---------. - ·---- -

17. Key Words (Suggested by Author(s)) 18. Distribution Statement

· Compatibility ·Reflectivity
·Doppler · Scatterometer
· Quadrature · Gyromotor

·1·
· Boresight . Cycle Slip

\--- --- -· ---

19. Security Classif. (of this-- --
.. ;;�: .. -. --- - --
.. - - f2i. �
-
:/2.

J
-
:w: s-;;� lassif -
(of this page ) No. of Page Price*
-
report)

None
--- . _ !
... __ _________ :_�------- . 3 · o_o
_$__ _ ___,

• For sale by the National Technical Information Service, Springfield, Virginia 22151
 APOLLO EXPERIENCE REPORT

LUNAR MODULE LANDING RADAR AND RENDEZVOUS RADAR

By Patrick Rozas and Allen R. Cunningham

Manned Spacecraft Center

SUMMARY

A technical history of the Apollo lunar module landing radar and rendezvous
radar subsystems is presented. Radar subsystem accomplishments and problems are
presented with discussions of the program plan; subsystem design, development, and
testing; subsystem performance, reliability, and quality control; and subsystem prob­
lems and changes. Conclusions and recommendations applicable to future space pro­
grams are also presented.

INT RODUCTION

In the development of the Apollo lunar module (LM) landing radar and rendez­
vous radar subsystems, the program was managed chiefly through the prime contrac­
tor, who coordinated closely with the various subcontractors to ensure maximum
communication. The first prototype units of the radar subsystems evaluated subsys­
tem performance through special tests such as environmental exposure and aircraft
flight tests, which simulated actual mission conditions. The deficiencies detected
during this series of tests were corrected, and the final-configuration flight units
were built. The first flight units were subjected to a full qualification test program
and to additional aircraft flight tests to ensure the integrity of the subsystems and the
fulfillment of all design goals. The vehicle-interface and subsystem performance
tests on the Apollo spacecraft were next in a series of tests to ensure subsystem com­
patibility. The final subsystem tests were performed during the early Apollo flights.
The successful operation of the rendezvous radar and landing radar subsystems during
the Apollo missions demonstrates that accurate and highly reliable subsystems have
been developed for lunar missions.

PROGRAM PLAN

The program plan called for NASA to monitor and direct the contractor's work,
which required extensive analyses, design studies, testing, quality control, et cetera.
Monthly technical reviews of the subsystems and periodic design reviews were con­
ducted. The NASA Manned Spacecraft Center (MSC) provided the technical guidance
to ensure the technical advance of each subsystem. The resident Apollo spacecraft
 program office at the prime contractor facilities provided the level of support that was
required to resolve some of the technical problems as they occurred.

The contractors also established offices for program management, material re­
view, cost control, and quality analysis and for control of engineering and manufactur­
ing procedures that were used in the design and fabrication of the radar subsystems.
Periodic design reviews and technical review meetings were held to provide maximum
communication between MSC and the contractors.

The delivered equipment included several subsystems that were flight prototypes.
These subsystems provided electrical and electronic parameters from which the final
radar configuration was determined. Tests were performed at the contractor's plant
and at MSC. The contractor performed the subsystem qualification through a series
of tests. During the spring of 1966, radar antenna boresighting was performed at
MSC. The flight test program was conducted by MSC with contractor support at the
White Sands Missile Range (WSMR) , New
Mexico. Figure 1 presents a schedule of
significant program events.
Event 1 1963 i � l br
1964 l965 l96 196 lrl968 1969

The radar subsystem for the MSC

LM rendezvous radar and
CSM )
radar transponder '"'
!
!
boresight tests was delivered in April 1966; contract awarded

equipment for the flight test program was LM 1

...
landmg radar contract

delivered in August 1966. The first space

awarded

flight use of the Apollo radar subsystems -

was the Apollo 9 ( spacecraft LM-3) mis­
sion in March 1969. The radar subsystems
were first used for lunar landing during the
Apollo 11 (spacecraft LM-5) mission in
July 1969. The radar subsystems were
required during four space flights and two / Radar flig hts
I""'
lunar landings to provide data for rendez­ - l

vous or for both lunar landing and rendez­

vous. The performance of the radar Figure 1. - Schedule of significant
subsystems was excellent on each occasion. program events.

SUBSYSTEM DESCRIPTION

Landing Radar

The landing radar senses the velocity and slant range of the LM relative to the
lunar surface by means of a three-beam Doppler velocity sensor and a radar altimeter.
The velocity and range information is processed and made available to the LM guidance
computer ( LGC) in serial binary form and to the LM displays in the form of pulse
trains and de analog voltages. Table I presents significant landing radar parameters.
A block functional diagram of the Apollo landing radar is shown in figure 2; the beam
configuration is shown in figure 3.

2
I

TABLE I.- SIGNIFICANT LANDING RADAR PARAMETERS

Type of system:
Velocity sensor cw, 3-beam
Radar altimeter cw, FM

Weight (nominal), lb 42.0

Size:
Antenna assembly
Length, in. .. 20. 0
Width, in. 24.6
Height, in. .. 6.5
Electronics assembly
Length, in. 15. 75
Width, in. 6.75
Height, in. .. . . 7.38

Power consumption:
Maximum de consumption, W ... 1 32
Antenna pedestal tilt actuator, W . 15
Antenna heater (maximum), W .. 63

Altimeter antenna:
Type ...... Planar array, space duplexed
Gain (two-way), dB .. 50. 4
Beam width (two-way)
E plane, deg .. . 3.9
H plane, deg ... 7. 5

Velocity sensor antenna:

Type .......... Planar array, space duplexed
Gain (two-way), dB .. 49. 2
Beam width (two-way)
E plane, deg 3.7
H plane, deg 7. 3

Transmitters:
Type ........... Solid state
Frequency
Velocity sensor, GHz 10.51
Radar altimeter, GHz 9.58

Output power:
Velocity sensor (minimum per beam), mW 50
Altimeter (minimum per beam), mW ... . 87.5

Altimeter modulation:
Type . . . • • . . .
Sawtooth F M
Modulation frequency, Hz . 130
Deviations
Low (altitude > 2500 feet), MHz ±4
High (altitude< 2500 feet), MHz •
±20

3
 Antenna assembly Electronics assembly

} To
lGC

Rl
Velocity-
sensor
rece i ver Rz ---v
antenna y
array
Crystal
R3 balanced �o---v
--
z
mixers

sensor 1

'"'";"� {
transmitter T2
antenna
array
T
3

Altimeter
transmitter Range sense
antenna {TR
array Key:
R ·Receiving array RR ·Range receiver
Altimeter tR T Transmitter
•
D Doppler
receiver Quadrature siqnal V Velocity
•

L
•

antenna T R ·Range transmitter

•

array

Figure 2.- Apollo landing radar block diagram.

The landing radar, which is located

v
X Note: The beam 2 and beam3 in the LM descent stage, is packaged in
configuration angles are
nomrnally rdentical to the two replaceable assemblies. The antenna
beam 1 configuration angre. assembly forms, directs, transmits, and
receives four narrow microwave beams.
To perform these functions, the antenna
assembly is composed of two interlaced

�
phase arrays for transmission and four
space-duplexed planar arrays for recep­

�
tion. The transmitting arrays form a
Beam 2 platform; four quadrature-pair balanced
Altimeterbeam
eam3 microwave mixers, four dual audio­

:_� :;
�
\ frequency preamplifiers, two solid-state
\
� -- 't --- microwave transmitters, a frequency mod­
Beam1 - - - --- - --- ulation ( FM) modulator, and an antenna
..,""' -- ---\.....:
. /
./
pedestal tilt mechanism are mounted on
Beam group
/ -- /
-
-

center l1ne
__
-
..._ --<"
-
,.,. ----_..,.

/
,.. the platform. The electronics assembly
-__,..
/
/
/
contains the circuitry that is required to
track, process, convert, and scale the
Doppler and FM/continuous wave (cw) re­
Figure 3. - Landing radar beam con­
turns, which provide the velocity and slant
figuration. Velocity coordinates
range information to the LGC and to the
are shown with respect to the
display panels.
vehicle and the antenna.

if .�:
�-.oil
 The transmitting antenna radiates the cw microwave energy from the solid-state
velocity-sensor transmitter to the moon. Three separate receiving antennas accept
the reflected energy. The received Doppler-shifted energy, which is split into quad­
rature pairs, is mixed with a portion of the transmitted energy by microwave diodes
that function as balanced mixers. The output of the crystal balanced mixers gives the
frequency difference between the received signals and the transmitted signals. This
frequency difference is the Doppler shift, which is directly proportional to the LM ve­
locity with respect to the lunar surface along the detected microwave beam.

The output of the altimeter transmitter (a sawtooth waveform) is frequency mod­

ulated at 130 hertz and is transmitted by a second antenna. The reflected energy re­
ceived by the receiving antenna is split to form a quadrature pair and, with a sample
of the transmitted signal, is coupled to balanced microwave mixers. The frequency
difference at the output of the balanced mixers is proportional to the time difference
between the transmission and the reception of the modulated energy, plus a Doppler­
shift factor. The undesired Doppler-shift factor is compensated for in the range
computer.

The quadrature outputs of the three velocity sensors and the altimeter balanced
mixers are routed to the four audio-frequency amplifiers. The wideband signals at the
audio-frequency amplifier outputs are used as inputs for frequency trackers, which are
located in the electronics assembly. The frequency trackers search for the signal over
the expected frequency range with a narrowband tracking filter; once the signal is ac­
quired, the frequency trackers follow the signal with a high degree of accuracy. The
tracker output is an average frequency, equal to the frequency that corresponds to the
center of power of the received Doppler signal spectrum. The Doppler sense is re­
tained. The frequency trackers also provide a de step voltage to indicate tracker lock.

The tracker outputs are routed to velocity and range data converters, where beam
velocity information is resolved into velocity components. The coordinate system is
referenced to the body coordinates of the antenna and a line drawn at right angles to the
face of the transmitting arrays, which in turn is parallel to the beam group center line.

The velocity data, which are computed with respect to the beam group center
line, are given in a pulse train form that is superimposed on a 153. 6-kilohertz refer­
ence frequency to facilitate a determination of the sign of the velocity. These velocity
pulse trains and the range pulse train are routed to the signal data converter. The
signal data converter forms an interface with the LGC by accepting strobe signals from
the computer and using these signals to assemble and read out the range and velocity
data in serial binary form. The serial binary radar output information is given to the
LGC.

Rendezvous Radar

The rendezvous radar is a space-stabilized cw tracking radar for the Apollo

lunar missions. This lightweight, highly reliable, and accurate radar subsystem func­
tions in operational environments that are encountered on the earth, in space, and on
the moon. The rendezvous radar is a solid-state coherent tracking radar that is used
in the LM for performing rendezvous with the command and service module (CSM) in

5
 lunar orbit. A block diagram of the rendezvous radar is shown in figure 4, and sig­
nificant rendezvous radar parameters are presented in table II.

Antenna assembly Electronics assembly

t ,----,
X 96
IModulation tones 1
1
r---- -

Phase - frequency �� ;-1 _

1 _ Frequency
c a_r.;,.
rier
!
modulator multiplier 1-_.;;. _
__
_
synthesizer
Transmitter j 1 Trunnion--. Automatic gain control monitor
power monitor '---------.Ref vo tage controlled oscillator Strobes
J Trunnion
l f f fer en t '-
--' 1
-- If output 1 - From LGC
.- -. -- differenc e ,.---'- - --
----, 1 � l � �di� � � : l-- ,---J-- ----, Range Signal data Range
Circulator.....__.... Comparator � Su m p. I
f sum If sum F rate t
ifuf.-- To displays
,--
....,. -J Mixers and Receiver ,......r-- requencyF-- processor �
diplexer Shaft preamplifiers If different tracker data
I '----' difference � ToLGC
I _]'� '----.-' Lockon --- - \'ri'dl��fi�n
\1 Shaft _j Modulating indicator Data good �
tones � logic ___ Telemetry
I + (indicator _ _ Data good to
1 �c
I 1 Self-test Range t ange Range
I L---L.-1 oscillator , -=--- To displays
� tracker
I
I
I
I
I
J
f Mo ulated
tones
Trunnion and shaft angle
errors mon1tor
1---- Resolvers
I Trunnion and shaft
I
I angles to displays
I
I 800 28V

�·---------�---l
Hz
I
I "----- Angles to LGC
Antenna designate
\__j Seco•o<oe< from LGC
Manual slew
J . [ == : =-l--+---� Servo control
r--:Stabilization and amplifier Automatic track enable
28v Igyromotors 1
To subassemblies
28V
---- r-::::1 : r;:--'--:- - __ll!2_ V
__L 5 ..:.._
--�t�· � t1..,.
Power
_

�-------------------r:l --�
Trunnion and shaft
angle rates
I To displays LS:U:PP::._IY___t-""'28'-'V-
Antenna temperature Temperature I
sensor I
l
Telemetry
I

Figure 4.- Apollo rendezvous radar block diagram.

In conjunction with a transponder that is located in the CSM, the rendezvous

radar measures line of sight (LOS) range, LOS range rate, LOS angle, and LOS angle
rate with respect to the CSM. Both the rendezvous radar and the transponder use
solid-state frequency multipliers as transmitters. Transmission and reception are
both performed in the cw mode. Gyromotors (located on the rendezvous radar antenna
assembly) stabilize the aperture LOS against variations in LM body motion, which per­
mits accurate measurements of angle rate.

Angle tracking is achieved by using an amplitude-comparison monopulse technique

to obtain maximum angle sensitivity and boresight accuracy. Range rate is determined

6
 TABLE II.- SIGNI FICANT RENDEZVOUS RADAR PARAMETERS

Radiation frequency, MHz . . . 9832.8

Received frequency, MHz ... . 9792.0 +Doppler
Radiated power (nominal) , mW 300
Antenna design . . . . Cassegrain
Angle tracking method Amplitude monopulse
Antenna reflector:
Primary (parabolic) 24-inch diameter
Secondary (hyperbolic) . . . . 4. 65-inch diameter
Antenna gain (at beam center) , dB 32
Antenna beam width, deg . . . . 3.3 to 4
Antenna sidelobe level (adjacent
to main lobe) , (minimum) , dB - 13
Angular coverage, deg ±70 by 225
Number of gyroscopes 4 (2 pair; 1 pair for redundancy)
Modulation . . . . . . . Phase modulation by 3 tones
(200 Hz, 6.4 kHz,
204. 8 kHz)
Receiver channels . . . . . . . . 3 (reference, shaft, and trunnion)
Receiver noise figure, dB . . . . 10
Receiver i.f. frequencies, MHz 40.8, 6. 8, 1. 7
Maximum range (unambiguous), n. mi. 405
Minimum range, ft . . . . . . . . . 80
Minimum discernible signal (for full
specification operation), dBm . . - 122

by measuring the two-way Doppler frequency shift on the signal that is received from
the transponder. Range is determined by measuring the time delay between the
transmitted-signal modulated waveform and the received signal waveform. A three­
tone phase-modulation system is used to obtain high-accuracy range measurements.

The rendezvous radar includes an antenna assembly and an electronics assembly.

The antenna assembly converts VHF signals to microwave-modulated signals and trans­
mits them. The return signal is converted into an intermediate frequency (i. f. ) signal
and is sent to the electronics assembly. The antenna assembly locks on to, and con­
tinually tracks the source of, the return signal.

The electronics assembly furnishes crystal-controlled signals that drive the an­
tenna assembly transmitter and provide a reference for receiving and processing the
return signal. This assembly also supplies servodrive signals for antenna positioning.
The electronics assembly consists of a receiver, a frequency synthesizer, a frequency
tracker, a range tracker, servoelectronics, a signal data converter, self-test cir­
cuitry, and a power supply.

In addition to the microwave-radiating and the gimbaling elements, the antenna

assembly includes internally mounted gyromotors and resolvers, a multiplier chain,
a modulator, and mixer-preamplifier components that avoid the necessity of micro­
wave rotary joints and permit the use of flexible coaxial cables between the outboard

7
111111111

antenna assembly and the inboard electronics assembly. A flexible cable wrap system
is used at each rotary bearing point. The antenna assembly has two axes: the trunnion
(azimuth) axis lies parallel to the LM X-Z plane, and the shaft (elevation) axis lies
parallel to the LM Y-axis. When the trunnion and shaft angles are 0 o, the antenna
boresight is parallel to the LM positive Z-axis.

DESIGN AND DEVELOPMENT

The design of the LM radar subsystems was unique because the subsystems
were the first solid-state radar subsystems to be operated in the space environment.
After the first landing radar and rendezvous radar engineering models had been com­
pleted, a thorough design review was held in May 1966. Changes and improvements
were made to the radar subsystems after this review, and the design of the production
model was finalized. To incorporate all the new features into the production model in
an organized manner, another design review was held in April 1968.

During the subsystem design phase, many factors were considered that would
affect the operation of the LM radar subsystems (landing radar, rendezvous radar, and
transponder). One important factor was the wide range of the thermal environment
conditions to which the radar subsystems would be subjected. The electronics assem­
bly for the radar subsystems was required to operate from 0 o to 160 o F; how-
ever, because the antennas are located outside the spacecraft, they had to withstand
temperatures of -240 o to +240° F . The radar subsystems were also designed to oper­
ate under widely varying shock and vibration conditions during the launch and boost
phases of the Apollo missions.

One of the most interesting tradeoff studies was of the antenna selection for the
landing radar. The tradeoff study considered size, weight, gain, beam width, and
ease of fabrication. Two antenna types with a given aperture are compared in table III.

TABLE III.- SUMMARY OF DISH AND ARRAY ANTENNAS

Parameter
,, - Dish system
-J Array system
_]
Gain, dB . 23. 8 24. 9

Beam width E, deg 4. 28 3. 00

Beam width H, deg 6. 4 1 5. 49

Depth, in. . . . . . 14. 0 3. 0

8
 If magnesium is used for the waveguide on both the dish and the array systems and if
aluminum honeycomb is used for sandwich structures, the weight difference is approx­
imately 0. 9 1 kilogram, with the dish system being heavier. The boresight technique
for the array system is less involved than that for the dish system.

After the array-type antenna was chosen, there was one problem in that the pre­
dicted transmitter array beam pointing angles did not agree with the measured results.
A series of tests was performed; these tests indicated that the tilted interlaced altim­
eter array caused a shift in the beam placement. This effect was sensitive to the
tuning elements between the velocity sensor and the altimeter radiators. This study
'-
resulted in a redesign of the array elements.

Special manufacturing techniques were used because the LM radar subsystems

are required to be highly reliable, lightweight, and compact. For example, multi­
layer boards and cordwood construction were chosen, because they fulfill the reliabil­
ity and packaging requirements better than other techniques which were considered.
Also, to reduce the number of unacceptable solder joints, failure records were kept
on each employee in the production line.

SUBSYSTEM TESTS

Landing Radar Boresight Test

The objective of the landing radar boresight test was to acquire sufficient data
to provide a basis for analysis of the static effects on landing radar antenna beam
geometry and to provide the value of the velocity bias errors to be used in the LGC.
This test was conducted at MSC. A detailed description of this test can be found in
reference 1.

Rendezvous Radar Boresight Test

The objective of the rendezvous radar boresight test was to obtain sufficient data
for the following tasks: (1) to aline the rendezvous radar with the LM vehicle naviga­
tion base, (2) to verify the functional operation of the rendezvous radar, (3) to deter­
mine the pointing accuracy of the rendezvous radar, and (4) to acquire sufficient data
to analyze rendezvous radar target acquisition and angular tracking performance.
This test was conducted at MSC. A detailed description of this test can be found in
reference 2.

Rendezvous Radar Performance Eva I uation Flight Test

The objective of the 1967 rendezvous radar performance evaluation flight test
was to verify the capability of the rendezvous radar to perform as required during the
Apollo missions. The tests were conducted under flight conditions, which simulated

9
several CSM-to-LM orientations along each of the probable LM rendezvous and lunar­
orbit trajectories to demonstrate that the rendezvous radar performed within the re­
quired accuracy range at distances representative of the design range. The objective
of the simulated rendezvous test was to verify that the tracking, ranging, and velocity
loops of the rendezvous radar operated properly during a simulated lunar stay. A
T-33 jet aircraft and a helicopter were used for the tests at WSMR. A detailed de­
scription of the flight test plan can be found in reference 3.

Landing Radar Performance Evaluation Flight Test

The objective of the 1967 landing radar performance evaluation flight test was to
demonstrate the capability of the landing radar to meet performance requirements
under dynamic flight conditions and to secure data that were used in evaluating the
LGC performance. The tests were conducted, within the capabilities of the test air­
craft, under flight conditions that simulated numerous points along each of the prob­
able LM lunar-descent trajectories.

The objectives of this series of tests were ( 1) to evaluate the performance of the
landing radar under dynamic flight conditions, (2) to verify the landing radar mathe­
matical model, (3) to evaluate the performance of the landing radar and the LGC,
(4) to verify the adequacy of the landing radar to meet mission requirements, and
(5) to define the constraints or necessary design changes. A more detailed description
of this flight test can be found in reference 4.

Perform ance Evaluation of the Apollo Rendezvuus

and Landing Radar Flight Test

The 1968 performance evaluation of the Apollo rendezvous and landing radar
(PEARL) flight test was an extension of the 1967 flight test and was necessary to cor­
rect some of the questionable data that resulted from timing errors in the 1967 flight
test. In addition to correcting the data, the PEARL program provided data for new
profiles, which aided in the evaluation of the landing radar for expected lunar-descent
trajectories.

Landing Radar Reflectivity Test

The objective of the 1968 landing radar reflectivity test at WSMR was to improve
the estimate of reflectivity as a function of the near-vertical incidence angle, obtained
from the 1967 RF scatterometer test. Modifications to the PEARL test aircraft and
the landing radar were made to conduct this test. The modifications consisted of
changing the antenna mount and the location of radar monitoring points. The electri­
cal properties of the terrain were measured to permit an extrapolation of the reduced
data to the lunar environment. Results of this test are incorporated in the present
lunar reflectivity model. A more detailed description of this test can be found in
reference 5.

10
r

Apollo 7 Rendezvous Radar Overflight Test

The objective of the rendezvous radar earth-orbital flight test during the Apollo 7
(spacecraft CSM- 10 1) mission was to determine the performance of the rendezvous ra­
dar transponder link under a simulated overpass condition at maximum range. The
test conditions were to simulate the lunar-stay phase of a lunar mission by requiring
the rendezvous radar to track an orbiting CSM that was within operative range to verify
that the tracking, ranging, and velocity loops of the rendezvous radar and the tracking
loops of the transponder could function at the extreme limits of their capabilities. The
tests were made in the mode II operation configuration (long-range acquisition). A de­
tailed description of this flight test can be found in reference 6.

Radio-Frequency Scatterometer Test

The primary purpose of the 1 967 RF scatterometer test was to provide meas­
urements of the backscattering coefficient per unit surface area a for various
0
types of earth terrain. The angular dependence of the backscattering cross section
per unit surface area a (e) and the absolute magnitude are measured by relating the
0
power density of the reflected energy for each Doppler frequency to the respective in­
cidence angle.

Both the accuracy and the altitude capability of the radar subsystems that are
used in surface track systems depend upon surface reflectivity characteristics. For a
rough surface, a knowledge of the value of a as a function of the variable is usually
0
sufficient to describe surface reflectivity. Therefore, another objective of the reflec­
tivity program was to learn as much as possible about the reflectivity characteristics
of various earth surfaces. This information would aid in the design and evaluation of
radar for earth, lunar, and planetary missions. The reflectivity program included
the following:

1. Reflectivity of various types of surfaces, including sand, desert, and vol­

canic formations

2. Reflectivity as a function of time for a given surface

3. Reflectivity as a function of altitude

A more detailed description of this test program can be found in reference 7.

Apollo 9 Landing Radar Test

Because the LM landing radar had never been tested in a space environment be­
fore the Apollo 9 (spacecraft LM-3) flight, special instrumentation was installed to
measure the signals in the velocity and altimeter preamplifier outputs, during the
landing radar test. If only crystal noise were present in the channels during the test,
the radar was operating properly. However, during the Apollo 9 mission, spurious
signals appeared, which were attributed to flaking of the Mylar thermal blanket during

11
 the lunar-descent engine burn. This flaking necessitated changing the Mylar thermal
blanket to an ablative paint on the lunar-descent stage.

Radio-Frequency View Factor Test

The purpose of the RF view factor test was to determine any false lockon ef­
fects caused by Doppler returns from LM structural vibrations during lunar-descent
engine firings. Three areas of special interest were the LM legs, the LM engine
skirt, and the LM bottom structure.

Results of the test indicated that some degradation of radar performance had oc­
curred. For this reason, three changes were made to correct the problem.

1. The preamplifier rolloff was changed to decrease the landing radar sensi­
tivity to the low-frequency vibrations exhibited by the LM structure.

2. The antenna was rotated 6° to prevent the landing radar beam from impinging
on the LM leg structure.

3. A baffle was installed to shield the radar beams from the lunar-descent en­
gine bell reflections.

I NTEGRATION AND CHECKOUT TESTS

Test Philosophy

The objective of subsystem testing was to demonstrate the integrity of the equip­
ment after installation on the spacecraft. Subsystem tests were conducted at the LM
contractor 1 s plant and at the Kennedy Space Center (KSC). These tests provided a
functional verification of the replaceable electronics assemblies to validate the inte­
grated subsystem.

The objective of integrated testing was to determine the physical, functional and
operational compatibility of all subsystems. The functional compatibility of all LM
subsystems was demonstrated during simulated flight modes. Integrated tests were
performed at the LM contractor 1 s plant and at KSC.

Test Flow

The test flow, which includes testing at the factory and KSC, is shown in table IV.

12
 TABLE IV.- TEST FLOW

(a) Factory .testing

I
CSM-101, CSM-104, CSM-106,
LM-3 and LM-4 LM-5 and subsequent
CSM-107, and subsequent

1. Preinstallation test 1. Preinstallation test 1. Rendezvous radar transponder

functional verification test
2. Rendezvous radar 2. Radar subsystem functional
functional verification verification test 2. CSM integrated checkout test
test
3. FEAT
3. Landing radar functional Plugs in
verification test Plugs out

a
4. FEAT
Plugs in
Plugs out
-

(b) KSC testing

CSM-101, CSM-104,
LM-3 and LM-4 LM-5 and subsequent CSM-107 and subsequent
and CSM-106

1. Rendezvous radar 1. Rendezvous radar 1. Combined systems 1. Combined systems

boresight test boresight test test (O&C building) test (O&C
building)
2. Combined systems 2. Combined systems 2. LM-to-CSM interface
b test (O&C test 2. Combined systems
test (O&C
building and test (VAB)
building)
VAB) 3. Flight readiness test
3. Flight readiness test
3. LGC interface test
3. Flight readiness
test
4. Combined systems
c
test (VAB )

5. LM-to-CSM inter-
face test

6. Flight readiness
test

a
Formal evaluation acceptance test.
b
Operations and control.
c
�
Vehi le assembly building.

13
 Rendezvous Radar Test Problems

Gyromotor leakage. - As a result of one gyromotor failure at the manufacturer's

plant, a special gyromotor leakage test was incorporated into the LM rendezvous radar
test program. The gyromotor failure was caused by a leakage of suspension fluid into
the gyromotor float; this leakage resulted in a gravity-sensitive drift. Drift tests were
performed on the LM-3 and LM-4 rendezvous radar subsystems at KSC and on the
LM-5 rendezvous radar at the manufacturer's plant to determine whether a gravity­
sensitive drift term with time dependence was present. The only anomalies that were
encountered during this special test were a no-spin-up situation and excessive drift.
The special test was deleted after the manufacturer had shown by endurance tests that
gyromotors with leakage problems did not have either decreased life or degraded
performance.

Gyromotor spin-up failure. - During the special test for gyromotor leakage that
was performed on rendezvous radar 18 {spacecraft LM-4) at the boresight range, one
gyromotor failed to spin up. The failure was attributed to a nonconcentric rotor bear­
ing, and the faulty gyromotor was replaced.

Gyromotor drift. - The special test for gyromotor leakage performed on the
LM-4 rendezvous radar indicated a possible excessive drift in one gyromotor. Al­
though the gyromotor was replaced, later tests showed that the drift of the suspect
gyromotor was within acceptable tolerance.
Electromagnetic interference. - During the combined systems test on spacecraft
LM-3 at the operations and control {O&C) building, electromagnetic interference
(EMI) was encountered and traced to a harmonic of the 1. 024-megahertz clock in the
pulse code modulation and timing electronics assembly. During the rendezvous radar
boresight test on the same electronics assembly on spacecraft LM-3, EMI problems
had been traced to a harmonic of the high-frequency tone. To ensure adequate screen­
ing of EMI problems, the pilot test was instituted at KSC for spacecraft LM-3 and
L:JYl-4, and at the contractors' facilities for spacecraft LM-5 and subsequent spacecraft.
The pilot test indicated the susceptibility of the rendezvous radar frequency tracker to
spurious signals from the 40. 8-megahertz preamplifier. The problems were generally
correctable by slight tuning of the 40. 8-megahertz preamplifier in the rendezvous
radar.

Minimal discernible signal leakage. - During the flight readiness test of space­
craft LM-3 and during the combined systems test on spacecraft LM-4 and LM-5 at the
vehicle assembly building (VAB), range tracker lock could not be acquired. The prob­
iem was caused by RF leakage in sections of the waveguide that connected the rendez­
vous radar with the transponder. A flexible waveguide was incorporated in the ground
support equipment to correct the leakage for the flight readiness test of spacecraft
LM-5 and for subsequent tests.

Voting logic. - Because of a design deficiency, the voting (gyromotor select) cir­
cuit did not automatically select the preferred gyromotors. Consequently, a manual
switch was installed for gyromotor selection. This problem was first noted during the
combined systems test of spacecraft LM-4 gyromotor torquing at the O&C building,
during the rendezvous radar boresight tests on spacecraft LM-5, and during the pre­
installation test on spacecraft LM-6. The gyromotor manual selection switch was in­
stalled on spacecraft LM-5 and subsequent spacecraft.

14
 Cycle slip and moisture absorption. - A cycle slip in transponder 20 occurred
during the combined systems test on spacecraft LM-5 and spacecraft CSM- 107 at VAB.
Initially, the problem was thought to be caused by excessive input signal strength,
which would overdrive the microwave phase modulator in the transponder and result in
ambiguous ranging because of improperly weighted midtone and hightone inputs to the
rendezvous radar up-down counter. However, the cycle slip was later attributed to
moisture absorption in the rendezvous radar and transponder ranging tone filters.
Sin.ce that time, both the rendezvous radar and the transponder were found to have
phase-shift problems in the filters for the ranging tones. Extensive testing and bakeout
procedures were integrated into the contractor and KSC test cycle to ensure identifica­
tion and correction of these problems before launch, because design modifications of
the rendezvous radar or the transponder to correct this deficiency were too costly.
The procedure involved obtaining extensive heater operation prior to launch or any
range-tone phase measurement. Adjustment of the phase calibrator circuits of the
rendezvous radar or the transponder ensured normal accuracy ranging under nominal
mission conditions.

Landing Radar Test Problems

Long-line capacitance. - During the combined systems test on spacecraft LM-5

at VAB, the blanking pulse in the landing radar altimeter low-frequency sweep gener­
ator was inhibited by long-line capacitance. The corrective action was to shorten the
cable between the deviation inhibit point in the low-frequency sweep generator and the
deviation inhibit switch at the bench test console (BTC).

Velocity bias error. - During the landing radar subsystem functional verification
test for spacecraft LM-5, a logic race at the input to the landing radar electronics
assembly shift register caused a one-count bias error in velocity. Appropriate logic
circuit alterations were made to eliminate the logic race condition. In addition, the
Gaussian distribution, which had been assumed for the test limits when the Doppler
spectrum simulator was used, was corrected to account for the presence of more en­
ergy in the tails of the distribution because of the poor approximation of a Gaussian
spectrum in a simple three-stage resistance-capacitance low-pass section.

Transponder Test Problems

Self-test. - The original wiring of the test selection switches on CSM panel 10 1
could impair operation of the rendezvous radar transponder during rendezvous (when
other normal functions of this switch, such as reaction control system quad tempera­
ture measurements, were performed). This situation resulted from the fact that po­
sitions A, B, and C of the right-hand test switch activated the self-test oscillator of
the transponder. This problem occurred during the rendezvous radar transponder
functional verification test for spacecraft CSM- 10 1. To correct the situation, the test
switch was rewired so that the transponder self-test operated from a separate self­
test enable switch.

Low supply voltage. - Excessive noise in the phase-lock loop and loss o.f phase
lock were encountered during the CSM integrated checkout test for spacecraft CSM- 10 1

15
 (transponder 13) because of poor transponder inverter power supply regulation. The
problem was traced to a low de input voltage that was caused by excessive line length
as well as circuit breaker and isolation diode voltage drops. The de input voltage was
raised to an adequate level by modifying the previous routing of the rendezvous radar
transponder de supply wiring with a direct connection through remotely operated relay
controls to the service module power distribution terminals.

Minimum discernible signal. - During the rendezvous radar transponder func­

tional verification tests for sp-acecraft CSM- 103, unreasonably sensitive values of the
minimum discernible signal (- 142 dBm) were obtained. The unreasonable values were
determined to be the result of poor attenuator calibration and poor procedure. The
attenuator calibration was degraded because of signal leakage from high-power to
low-power paths in the BTC microwave plumbing. The original procedure involved
observation of the phase-lock discrete to indicate the lockon point. However, because
the spectrum analyzer gave a more accurate indication of the lockon point, the orig­
inal procedure was deleted in favor of the spectrum analyzer method.

Ground loop of 6. 4 kilohertz. - Excessive midfrequency phase error was en­

countered during the rendezvous radar transponder functional verification tests for
spacecraft CSM-104. The problem was caused by a long signal ground path that had
not been properly reconnected to the de ground when the new de voltage supply con­
nections were made to correct the low supply voltage problem. To correct the prob­
lem, the signal ground path was properly connected to the de ground at the transponder.

Bench test console leakage. - During the CSM integrated checkout test for space­
craft CSM- 101, the transponder phase locked on X-band energy that leaked from the
BTC. For spacecraft CSM- 104 and subsequent spacecraft, the X-band source in the
BTC was turned off during periods in which operation of the transponder beacon mode
was desired to ensure electromagnetic compatibility, in the most severe condition, be­
tween the rendezvous radar transponder and the CSM- systems.

SUBSYSTEM PERFO RMANCE

The radar subsystems performed very well in flight as shown by the successes
of the Apollo 7 (spacecraft CSM- 10 1) mission, the Apollo 9 (spacecraft LM-3) mission,
the Apollo 10 (spacecraft LM-4) mission, the Apollo 1 1 (spacecraft LM-5) mission, and
the Apollo 12 (spacecraft LM-6) mission. The results of these flights are summarized
in the following paragraphs.

Apollo 7 Mission

The Apollo 7 (spacecraft CSM- 10 1) overflight test of the rendezvous radar at

WSMR fulfilled the test objective. All parameters of the rendezvous radar that were
tested (shaft, trunnion, range, and range rate) showed the performance to be consist­
ent with observed errors from the rendezvous radar PEARL flight test series at WSMR.
Exceptional performance was noted for shaft, trunnion, and range measurements from
the rendezvous radar. The rendezvous radar range rate bias appeared to be greater
than the master end-item specification.

16
 Apollo 9 Mission

During the Apollo 9 (spacecraft LM-3) earth-orbital flight, the landing radar de­
tected spurious signals that were attributed to flaking of the aluminized Mylar coating
during the lunar-descent engine burns. This problem was corrected; therefore, the
confidence for reliable operation during an actual lunar landing was increased.

Apollo 10 Mission

Because only limited radar data were available from the Apollo 10 (spacecraft
LM-4) flight, an estimate of the lunar reflectivity was made in the vicinity of acquisi­
tion only. However, the reflectivity calculation that was based on the Apollo 10 mission
data added confidence to the reflectivity model for the LM landing radar performance
simulation.

Apollo 11 Mission

The landing radar performed well during the Apollo 1 1 (spacecraft LM-5) lunar­
descent and lunar-landing maneuvers. The data appeared to be well within specification
limits, except a few points at low velocities near zero Doppler shift where the landing '
radar was not expected to track. The two questionable data points were probably
caused by poor data processing during the LGC overload alarm. The lunar-surface
reflectivity was determined to be in close agreement with the present smooth-surface
model at the velocity beam 1 acquisition point.

Apollo 12 Mission

On the Apollo 12 (spacecraft LM-6) flight, the landing radar operated as expected;
lockon was obtained early in lunar descent. Calculations based on flight data indicated
a higher value of lunar reflectivity than had been expected, which might have been the
result of local lunar terrain slopes that gave high angles of beam incidence.

Overall Performance

On all missions up to Apollo 12, the rendezvous radar has performed well, as
indicated by the successful rendezvous. On the Apollo 1 1 (spacecraft LM-5) and
Apollo 12 (spacecraft LM-6) flights, the rendezvous radar range data were compared
to the VHF ranging system data. In both cases, the range data were in very close
agreement. As an example, on the Apollo 12 flight, the mean bias between the
rendezvous radar range data and the VHF ranging system data was less than 0. 04 per­
cent of the median range at which the comparison was made.

17
 PROBLEMS AND CHANGES

Failure of Multilayer Printed-Circuit Boards

The multilayer printed-circuit boards failed during the rendezvous radar quali­
fication test program. The interlayer columns exhibited open circuits at hot- and cold­
temperature extremes. The vendor believed that the boards which had passed thermal
cycling were flightworthy, but further tests indicated that the boards would fail after
thermal cycling. The corrective action was to identify the manufacturing problem and
then to change the process. Therefore, the multilayer boards were replaced with an
improved type of board. The improvements resulted from changes in manufacturing
techniques.

Landing Radar Detection of Vibrating Structural Members

During vibration testing of the landing radar subsystem that was mounted on an
LM mockup, the radar locked on to false targets. Vibrating structural members were
generating Doppler interference signals in the reflected signals, and the radar locked
on to these Doppler signals. To correct the problem, a metal shield was installed be­
tween the radar antenna and the vibrating members to block the view of the members
by the radar. In addition, the low-frequency response of preamplifiers was reduced
to attenuate the low-frequency false Doppler signals further, and the antenna was ro­
tated to move the beam from the LM structure.

Range Errors in the Rendezvous Radar

During flight tests of the rendezvous radar at WSMR, errors were found in the
range readings, the magnitudes of which were in multiples of 2400 feet. The
errors were found to be caused by cycle slips in the range tone tracking phase-lock
loop. Each cycle slip, or phase shift of 360°, caused a change of 2400 feet in the
range reading. The cause of the cycle slips was a low signal-to-noise ratio and a tone
phase-shift bias inherent in the design. Therefore, a limiter was added in the trans­
ponder tone amplifiers to restrict the peak noise to an amplitude at which the noise
would not cause cycle slips.

Arcing in Frequency Multiplier

During thermal vacuum testing, arcing occurred in the transmitter frequency

multiplier chains. The problem existed in both the rendezvous radar and the landing
radar and was caused by high voltages and inadequate separation of high voltage points.
The solution was to rearrange the components to obtain greater separation.

18
 Cracked Solder Joints

On the landing radar, solder joints cracked; this cracking was caused by a buildup
of conformal coating in critical locations. The conformal coating had a thermal coef­
ficient of expansion which was different from that of the component leads. As the
temperature changed, stress was exerted on the solder joints by the expansion and
contraction of the conformal coating. As a result, the solder joints cracked. The
problem was solved by changing the manufacturing technique to prevent buildup of large
amounts of the conformal coating in spaces where it could exert stress on the solder
joints.

Rendezvous Radar False Carrier Lockon

During testing of the Apollo 9 (spacecraft LM-3 } radar subsystems, the rendez­
vous radar locked on to false signals. The rendezvous radar was found to be locking
on to a harmonic of the 204. 8-kilohertz range tone. The solution was to improve the
shielding on the cables between the antenna assembly and the electronics assembly.

Landing Radar Lockon

Data which were obtained from the Apollo 9 (spacecraft LM-3 ) flight indicated
that false Doppler signals were received, which could cause radar lockon. The false
Doppler signals were found to be caused by reflections from flaking aluminized Mylar
thermal coating, located on the bottom of the lunar-descent stage. When the lunar­
descent engine fired, some of the Mylar burned and flaked off. The flakes then caused
radar energy reflections that contained Doppler frequencies which were related to the
velocity of the flakes. The solution was to replace the aluminized Mylar with a non­
flaking thermal paint.

Rendezvous Radar Range Tone Phase-Shift Drift

The phase shift of the rendezvous radar range tone filters was found to vary with
time. The problem occurred on the Apollo 9 (spacecraft LM-3 ) spacecraft and subse­
quent spacecraft. The phase-shift drift was most serious on the midtone (6. 4 kilohertz)
filter. The effect of excessive phase shift was to cause range errors that were in
multiples of 2400 feet. Turning on the tone filter heaters tended to stabilize the
phase-shift drift. The solution was to adjust the filter phase shifts to a small negative
value initially, and then to operate the heaters long enough to obtain the phase shift
near the desired value of 0 o.

CONCLUSIONS AND RECOMMENDATIONS

The success of the Apollo flights and the excellent operation of the radar subsys­
tems have shown that the radar subsystem design, construction, and testing are satis­
factory. Nevertheless, a few recommendations may be helpful in planning future space
programs.

19
 Careful planning should be provided for all flight tests. In particular, flight tests
should not be strictly mission oriented (where the subsystem is tested only under antic­
ipated mission profiles). Instead, tests should also be conducted to evaluate the sub­
system capabilities and performance limits. Such data become very important when
predictions must be made to indicate subsystem performance under new conditions.

For filtering of range tones, digital filters should be considered to avoid the
phase-shift drift problem that was encountered in the rendezvous radar. Digital filters
were not practical when the rendezvous radar design was finalized. However, recent
advances in the state of the art indicate that digital filters should be seriously con­
sidered in future space programs to avoid problems of phase-shift drift.

Consideration should be given in future space programs to compensation in the

guidance computer for the Doppler effect in the range channel of the landing radar.
Presently, the Doppler effect is removed in the landing radar by subtracting a scaled
Doppler shift that is obtained from two of the velocity beams. When the spacecraft
velocity is zero, the Doppler shift is zero, and the velocity trackers lose lock. A
resulting transient appears in the range channel, which causes a range data transient.
If the Doppler effect in the range channel were removed by the guidance computer, a
zero Doppler-shift transient would not affect the range data.

Interface control documents should be updated to reflect the flight hardware in­
terface requirements. Several testing problems could have been avoided with updated
interface control documents. Provisions for mandatory modification of ground support
equipment to meet test requirements should be included in the ground support equip­
ment contract to ensure that the ground support equipment is current. A statement of
permissible field adjustments and the required ground support equipment capability to
support field adjustments should be included in the interface control documents. All
testing groups must maintain close communication. Firm requirements for justifica­
tion of any deviation in test procedure, equipment configuration, or test stimuli should
be negotiated by all testing groups before the test program is begun. In particular,
testing of the first two or three vehicles, and the subsystems, should have nearly one­
to-one correspondence in test procedure from vendor to launch.

Manned Spacecraft Center

National Aeronautics and Space Administration
Houston, Texas, November 30, 1 97 1
9 14--13-00-00-72

20
 REFERENCES

1. Byne, G. E. ; and Miller, J. C. : Landing Radar Antenna Boresight Test Plan

(rev. 1). Tech. Rept. 05952-H034-R0-0 1, TRW Systems/Houston Operations,
Mar. 3 1, 1967.

2. Rood, E. J. : Rendezvous Radar Boresight Test Plan. Tech. Rept. 2240-H002-

RO-OOO, TRW Systems/Houston Operations, July 12, 1966.

3. Roberts, L. W. : Rendezvous Radar Flight Test Plan. Tech. Rept. 05952-H035-

RO-OO, TRW Systems/Houston Operations, Sept. 15, 1966.

4. Miller, J. C. : LM Landing Radar Aircraft Flight Test Plan. Tech. Rept. 05952-
H023-RO-OO, TRW Systems/Houston Operations, Sept. 7, 1966.

5. Anon. : White Sands Low Altitude (SH3A) LM Landing Radar Flight Test Plan.
LEC Document 64 1D. 3 1. 628, Lockheed Electronics Co. , Sept. 20, 1968.
(Also available as MSC/IESD Document 3 1- 19. )

6. Dobby, S. D. ; and Sulava, J. F. : Apollo Spacecraft Systems Analysis Program.

Analysis of CSM- 10 1 Rendezvous Radar Overflight Test Data (rev. 1). Tech.
Rept. 1 1 176- H 153-R0-0 1, TRW Systems/Houston Operations, Apr. 7, 1969.

7. Floyd, W. L. : Scatterometer Data Analysis Program. Rept. 57667-2, Ryan

Aeronautical Co. (San Diego, Calif. ), Sept. 2 1, 1967. (Also available as
NASA CR-62072. )

NASA-Langley, 1972 - 31 S-31 1 21

 I
N AT I O N A L A E R O N A U T I CS A N D SPACE A D M I S T R AT I O N

WA S H I N G T O N , D . C . 20546
POSTAGE A N D FEES PAI D
NATIONAL AERONAUTICS A N D
O F F I C I A L B U S I N ESS
SPACE A D M I N ISTRATION
FI RST CLASS MAIL
PENALTY FOR P R I VATE U S E $300 U.S.MAIL

NASA 451

D ::· r, 1 r) :- � . I T� \ T :� �, :; � ··�
-
{

\ :;- \·! :� 1\ :' l l :J -; T_. ; �' ( I ? ) r: )

�· :, C l ' �� T ""2 r,> r.\ . � '{ / ···�· I J .-1 L I
!\ ' ;' ,... ': :
. :;: L . i.J r 1 ,' ) •.: v _; '\J 1 := 1 1 T f=' r.•

-( I C/ : :__ '\ \: I \ .-. i� . ' ·.: '� 7 1 1 7

If Undeliverable ( Section 1 5 8
POSTMASTER: Postal Manual ) D o Not Return

'The aerona11tical and space activities of the United States shall be

conducted so as to contribute . . . to the expansi01z of human knowl­
edge of phmomena in the atm osphere and space. The Administration
shall provide for the widest practicable and appropriate dissemination
of information concerning its actit•ities and the 1"eStllts thereof."

- NATIONAL AERONAUTICS AND SPACE ACT OF 1 95 8

NASA SCIENTIFIC AND TECHNICAL PUBLICATIONS

TECHNICAL REPORTS : Scientific and TECHNICAL TRANSLATIONS : Information

technical information considered important, published in a foreign language considered
complete, and a lasting contribution to existing to merit NASA d istribution in English.
knowledge.
SPECIAL PUBLICATIONS : I nformation
TECHNICAL NOTES : Information -less broad derived from or of value to NASA activities.
in scope but nevertheless of importance as a Publications include conference proceedings,
contribution to existing knowledge. monographs, data compi lations, handbooks,
sourcebooks, and special bibliographies.
TECHNICAL MEMORANDUMS :
Information receiving limited distribution TECHNOLOGY UTILIZATION
because of preliminary data, security classifica­ PUBLICATIONS : Information on technology
tion, or other reasons. used by NASA that may be of particular
interest in commercial and other non-aerospace
CONTRACTOR REPORTS : Scientific and
applications. Publications include Tech Briefs,
technical information generated under a NASA
Technology Utilization Reports and
contract or grant and considered an important
Technology Surveys.
contribution to existing knowledge.

Details on the availability of these publications may be obtained from:

SC I ENT I F I C AND T EC H N I CA L I N FO R MAT ION O F F I C E

N AT I O N A L A E R O N A UT I C S A N D S PA C E A D M I N I S T R AT I O N
Washington, D.C. 20546

---
---·-
- ·-· .... . . . . . . . ·-·-- ·--.. --
. . . .. .. .