Professional Documents
Culture Documents
GPO PRICE $
CFSTI PRICE(S) $
Microfiche (MF)
ff 653 July 65
NASA SP-57
ORBITING
SOLAR
OBSERVATORY
SATELLITE
OSO I
The Project Summary
Prepared by
Goddard Space Flight Center,
Greenbelt, Maryland
Ill
CONTENTS
Page
ooo
FOREWORD ............................................................... HI
Chapter 1
Chapter 2
SPACECRAFT DYNAMICS
Analysis of Spacecraft Rigid-Body System .................................. 13
Nutation Damper Development ........................................... 22
Effects of External Torques on Spacecraft .................................. 47
Control of Torque Effects ................................................. 67
Chapter 3
Chapter 4
CONTROL SYSTEMS
Introduction ............................................................ 89
Erectrical Control Systems ............................................... 91
Gas Control Systems ..................................................... 141
Performance ............................................................ 151
.d
Chapter 5
Chapter 6
POWER SUPPLY Page
Introduction ............................................................ 189
Power and Temperature Estimates ........................................ 189
Solar Cell Array ........................................................ 189
Buttery ................................................................ 190
Power Margin .......................................................... 190
Power Supply Performance ................................................ 191
Chapter 7
THERMAL CONTROL
Introduction ...................................................... _..... 193
Analysis of Simplified Mathematical Model ................................ 193
Analysis of Shielded Model of Spacecraft Wheel ............................ 206
Evaluation of the Shape Factor Integrals .................................... 211
f Chapter 8
Chapter 9
TEST PROGRAM
Introduction .......................................................... 243
Test Equipment .......................................................... 243
Developmental Testing .................................................. 247
Prototype Testing ...................................................... 252
Flight Model Testing .................................................... 267
Prelaunch Testing ........................................................ 273
Miscellaneous Tests ...................................................... 275
vi
INTRODUCTION AND PROGRAM DESCRIPTION
This chapter presents a description of the GENERAL DESCRIPTION OF THE ORBITING
Orbiting Solar Observatory program, a brief SOLAR OBSERVATORY
description of the scientific experiments aboard,
orbit criteria, spacecraft design history, and The Orbiting Solar Observatory (Figure 1-1)
the development sequence of the spacecraft. was primarily a stabilized platform for solar-
oriented scientific instruments. I n addition,
INTRODUCTION experiments which did not require solar orien-
tation were contained in the observatory.
It has become evident that solar radiation
and particle emissions, which have a great ef-
fect upon our atmosphere, must be measured
before their character is changed by the atmos-
phere. Until recently, many solar emissions
could be studied only through the effects they
have on the atmosphere. However, to study
both cause and effect, it is essential that these
phenomena be measured both from above and
below the atmosphere.
The studies decided upon for the f i s t attempt
a t continuous measurement of solar phenomena
above the atmosphere mere spectrometric
studies in the ultraviolet and in the sh(rt wa~\.e-
length x-ray regions of the solar electromagnetic
radiation spectrum. The value of the data
gathered from above the atmosphere is two-
fold. In the &st place, measurements de- FIGURE
1-1.-Orbiting Solar Observatory (OS0 1).
rived directly from the sun with no atmos-
pheric attenuation may be used to arrive at Electrical power required for Observatory oper-
explanations for purely solar phenomena and ation was supplied by an array of solar cells
g v e greater insight into the mechanics of the mounted on the upper section. A complete
production of the measured emissions in the telemetry and command system was provided
sun. Secondly, the direct measurements from to transmit information back to earth, and to
the sun when correlated with observed effect.: receive commands from a ground station.
produced in the earth’s atmosphere may be
used to explain the mechanisms by which Basic Spacecraft
these atmospheric phenomena occur.
There can be no doubt that the ability to The observatory consisted of two main
measure unattenuated solar emissions is a sections: the wheel structure and the upper
giant stride in the advancement of knowledge structure. The lower wheel structure was
of the sun and its influence on the earth. The composed of nine wedge-shaped compartments.
Orbiting Solar Observatory gave us this abilitv. Five of these compartments contained scientific
1
ORBITING SOLAR OBSERVATORY
UNIV. OF MINNESOTA
IMTEWPV P A W W R mm
Of H M M R4YS fOJ-5 M€V)
WZ 3/.5/&. PlDWER 0.6 W
W~83Ih.powERa2 w
UNIV. OF ROCHESTER
ikrER0Y mUM mys
U€WROUmUIT &BOW /00 MW)
M€MUR€M€UT WZ /6.8/& POW€# h/ W
r;;r 0
WHEEL EXPERIMENTS
FIQURE1-2.-Wheel experiments.
experiments (Figure 1-2); the other four con- held in position by two bearings, one a t the
tained functional hardward such as the wheel top of the wheel and one a t the bottom.
electronic controls, batteries, telemetry system, Mounted on the shaft between the bearings was
radio command system, and the in-flight data a high-pressure nitrogen gas tank which carried
storage system (Figure 1-3). Three fiberglass the gas supply for precession jets. These jets
spheres on extended arms contained pressurized were mounted atop the solar array structure.
nitrogen gas for the spin control system. A torque motor mounted on the base of the shaft
The upper structure of the observatory con- drove the shaft and upper structure relative to
tained the solar oriented experiments (Figure the wheel. This motor actively controlled the
1 4 ) , and was mounted on the wheel. The azimuth orientation of the stabilized upper
fan-shaped array to which silicon solar cells structure by driving it a t an equal rate b u t
were attached made up the larger part of the opposite rotational direction with respect to the
upper structure. wheel. Also mounted on the base of the shaft
The two main structures were connected by was a slip-ring assembly that transferred power,
an aluminum shaft (Figure 1-5) extmdirig from telemetry signals, and control signals from the
the base of the casting containing the pointed upper section into the wheel.
instruments, through the center of the wheel, The observatory was designed for maximum
tmd terminating in the support ring structure on utilizntion o f the payload section of the Thor-
the underside of the wheel. This shaft was Delta lnunch vehicle. The maximum wheel
L
+ INTRODUCTION AND PROGRAM DEscRIpnON
WHEEL
ELECTRONIC SYSTEMS
FIGURE1-3.-Wheel electronics systems.
diameter allowed by the Thor-Delta shroud was of the wheel was reduced to approximately
44 inches. During launch, the three nitrogen 30 rpm by jet action. This spin rate was main-
gas containers on th,e extended arms mere folded tained within *5 percent of nominal value by
down along the sides of the third-btage mohr. gas jet,? attached to each of the spherical gas
When the nitrogen tanks were extended out- containers. These jets were actuated by a
vi-ard after third-stage burnout, the diameter signal from photodetectors and an electronic
of the payload was increased to 92 inches. control system which computed the instan-
Since these spheres were carried a t the ends of taneous period of rotation of the wheel with
the extended arms, the axial moment of inertia reference to the sun. The three gas-filled
mas larger than either transverse moment of spheres were interconnected to assure that
inertia. The overall height of the observatory unbalance of the rotating body would not occur
was 37 inches, and total weight waq about due to an unequal gas flow rate through the jets.
460 pounds.
The Orbiting Solar Observatory utilized the Control System
,qVroscopic properties of a spinning bod>- for
stability. Prior to third-stage firing, the third The biaxial point.ing control system of the
stage and the spacecraft were spun up to ap- obserratorp utilized the entire vehicle as the
proximately 100 rpm bj- a system of small rocket controlled platform. Coarse elevation position-
motors (Figure 1-6). After third-stage burn- ing of the stabilized sect,ion was accomplished
out, the arms supporting the three gas con- b1 determining pitch error with pitch control
tainers were extended. The spacecraft was photodetectors and controlling pitch attitude
separated from the rocket, and the spin rate with on-off jets (Figure 1-7). This was pos-
3
ORBITING SOLAR OBSERVATORY
GOOOARO !SPACE
/--- FLIWT .
Ik.
aw
tD SPACE FLIGUT
FIGURE
14-Solar oriented experiments.
sible because of the gyroscopic properties of mounted on the stabilized section, torques of
the spinning body. By exhausting nitrogen either sense, normal to the plane of the solar
gas (after stabilization) through nozzles array, could be produced to precess the space-
craft in elevation. The spin axis was thus
positioned perpendicular to the solar vector
within about three degrees. There was no
control about the sun vector (roll), but rates
mNa6
around this axis were very slow due to the
- awnxunm rigidity of the gyro.
Fine elevation and azimuth positioning of
-.ucDzN
the instruments was accomplished by electrical
- b-afms servo motor control. As described earlier, the
i MMmIUDIHI
elevation servo motor was mounted on the
casting which supported the pointed instru-
A U M W IYMCIDTW ments. This motor drove the instruments
W?UrnS
relative to the spin axis for fine positioning
in elevation. The azimuth servo motor,
AZIMf77f SHAFT ASSEMBLY DETAIL mounted to the shaft connecting the stabilized
upper section to the spinning wheel, is men-
P'IGURE 1-5.--A~iniuth shaft assemt)ly. tioned i n tm earlier paragraph.
4
b INTRODUCTION AND PROGRAM DESCRIPTION
T h e aziniuth and elevation servo motors were coarse detectors provided a signal to the azi-
actuated by signals from two tj-pes of photo- muth servo drive system vchich would despin
detectors mounted on the stabilized upper the upper section. As the upper section spin
section of the spacecraft. These were: (1) rate was reduced to zero, the coarse detectors
coarse detectors mounted on the sail structure pointed the solar instruments to within two
(Figure 1-7) and (2) fine detectors mounted to three degrees of the solar vector. At that
directly to the pointed instruments. time. a disabling detector mounted on the
There \%-erefour coarse detectors, each having pointed instruments actuated a rela>- which
a 90-degree field of view. which provided tiirned the coarse detector control off, leaving
a frill 360 degrees of position control. There the azimutb positioning servo with fine detector
were two fine detectors which provided a control only. K h e n the azimuth servo had
differential signal oi-er about 10 azimuth degrees positioned the iipper section normal to the
in each direction from the solar vector. The d a r vector, the elevation servo then positioned
elevation servo position error sign:& originated the pointed in.;truments in elevation. AS ex-
from the 10-degree fine detector.: only. plained earlier, the fine detectors for both
When the spacecraft was in siinlight and elerution and nzimuth position control were
the iipper section was spinning, the azimuth mounted directly t o the pointed instruments
5
ORBITING SOLAR OBSERVATORY *
FIGURE1-7.-Control systems.
Two of these detectors were used for each con- the spacecraft attitude had been returned to
trol loop. Operating as pairs, the detectors within one degree of the desired position.
provided a differential signal which was the Other photodetectors used in the pointing
error signal for the particular servo. Short- control system were: (1) turn-on detectors and
term pointing accuracy of the fine control (2) spin control detectors. T h e turn-on de-
system was better than one minute of arc in tectors, as the name implies, were used to
elevation and azimuth. Long-term accurttcy actuate electrical equipment (turned off in the
was better than two minutes of arc. dark) each time the satellite emerged from the
T h e pitch control detectors were detection earth's shadow. A set of six detectors was
units for the pitch jet control system. This distributed around the outer surface of the
block of four detectors was mounted facing the spinning wheel t,o tjiim on equipment. bot,h in
sun on the front of the stabilized structure. the wheel and on the pointed section. T h e
Whenever the spin axis of the spacecraft detectors observed the sun once per revolution
changed pitch more than three degrees from of the wheel, providing the spin controller with
t,he normal to the solar vector, one of these the necessary information for the determination
detectors turned on the appropriate jet to of wheel spin rate.
precess the spacecraft back toward the desired
Power Supply
attitude. Two of the detectors were required
for this function, one of either sense. The Solar energy was the sole source of power for
other two detectors turned off the jets whenever operating electrical equipment and experiments
6
• INTRODUCTION
ANDPROGRAM
DESCRIPTION
on,hespacecraft.Thesolar-cell
array mounted frequency was used in the tape speed compen-
on the stabilized section converted solar energy sation networks of the receiving station to com-
into electrical energy whenever the spacecraft pensate for permanent changes in tape recorder
was in sunlight. This energy was stored in speed and for transient changes, such as wow
nickel-cadmium batteries contained in the and flutter, up to 300 cps.
spinning wheel. During daylight, when the The subcarrier oscillator output signals were
spacecraft was pointing, the solar-cell srray recorded and played back at equal amplitudes.
was normal to the sun vector within three A pre-emphasis filter preceded the transmitter
degrees. input to provide an amplitude taper commen-
The total surface area of the solar-cell array surate with transmitter noise level and a
was 3.72 square feet. This array was composed bandwidth of 100 kc.
of thirty-one, 18-volt, 60-cell modules which The transmitter consisted of two parts: the
produced a power output of approximately 27 modulated driver, and the RF power amplifier.
watts. The average day-night power available Both units used solid-state active elements.
for use from storage batteries was approximately During the 90-minute record phase of the orbit,
16 watts. Since the telemetry, data system, the driving signal to the RF amplifier was re-
and control system required approximately duced to permit an output of 300 row. By
seven watts, nine watts were available for the ground control initiation of playback, the driv-
various scientific experiments. ing signal to the power amplifier was increased
to permit an output of 1.75 watts during the
Tkermal Control five-minute tape recorder playback interval.
Since only one transmitter operated at a time,
Spacecraft temperature was controlled by both were able to operate at the same frequency.
making use of carefully selected external sur- The antenna system utilized two of the space-
faces. Special paints were developed for the craft arm supporting structures as radiating
upper structure back surfaces and the wheel rim elements. Antenna polarization required diver-
surfaces. Polished aluminum covers were used sity combining of vertically and horizontally
for the fiat top and bottom wheel compartment polarized signals at the receiving station. Due
covers. to a two-db loss in the antenna and RF multi-
plexing system, the radiated power was nomi-
Data Acquisition and Command System nally 1.0 watt.
The command system for the observatory
The observatory telemetry was accomplished was a 7-tone AM system. It was capable of
with an FM-FM system. There were two inde- actuating 10 distinct command functions.
pendent parallel multiplexing systems. Each
system had a tape recorder, a set of subcarrier DESCRIPTION OF THE OBSERVATORY
oscillators, and a transmitter. EXPERIMENTS
The tape recorder, running at approximately
0.65 ips, recorded the complex signal (eight non- Solar X-Ray Spectrometer
standard low frequency sub-carriers) for 90
minutes of the 95-minute orbit. During a five- A solar x-ray spectrometer was employed for
minute interval, when the spacecraft was within the primary study of short wavelength x-rays
receiving range, the tape recorder would play (See Figure 1-4). It was designed with a sensi-
back the recorded signal at 18.35 times the tivity adequate to yield a spectral distribution
record speed. At this playback speed, the sub- over the 60A-400A wavelength band. With
carrier oscillator frequencies became standard the experiment continuously pointed at the
IRIG frequencies and modulated the spacecraft sun, sunlight entered the instrument through a
transmitter. slit. The sunlight was dispersed at the 2-degree
One subcarrier oscillator was a fixed fre- grazing angle grating into the x-ray spectrum,
quency, highly stable oscillator. This oscillator and was focused spectrally on the "Rowland"
ORBITING
SOLAR
OBSERVATORY
circle. The detectorwasmountedon a motor- storage register which was read into a shift
driven carriage which moved along a track co- register and presented as digital words for
incident with the Rowland circle and thus could output data. This gamma ray monitor experi-
continuously scan and measure the spectrum ment was developed at the University of
presented. The spectrometer was developed at Minnesota by J. Winkler and L. Peterson.
Goddard Space Flight Center by W. Behring
and W. Neupert. 100-500 Mev Gamma Ray Monitor
An experiment for monitoring the 1.8A This experiment was designed to monitor the
x-ray region was developed at Goddard Space 0.510 Mev electron-positron annihilation line
Flight Center by K. Frost, W. White, and R. using a NaI (T1) scintillator. It was developed
Young. It utilized two Be(Xe) ion chambers
by K. Frost and W. White at Goddard Space
connected in parallel and collimated to reduce
Flight Center.
low-energy electron effects.
3800A-4800A Monitor
0.2-1.5 Mev Gamma Ray Monitor
An experiment for monitoring the 3800A-
This gamma ray monitor used two NaI (T1) 4800A band was built using a special photo-
scintillators: one was shielded for a penumbral diode with a filter. This experiment was
detection angle of 20 °, and the other was un- capable of detecting variations of solar flux in
shielded and nearly isotropic in detection ability a bandwidth as small as 0.1 percent. This
so that it could be used as a background control band of wavelengths is important in the total
detector for the shielded detector. This experi- energy balance studies of the sun. The experi-
ment was developed at Goddard Space Flight ment was developed at Goddard Space Flight
Center by K. Frost, W. White, and K. Hallam. Center by K. Hallam, W. White, and H.
Murphy.
50 Key-3 Mev Gamma Ray Monitor 1100A-1250A Monitor
Neutron Monitor
9
ORBITING
SOLAR
OBSERVATORY
precisely,momentums,were such that con- bearings to operate reliably in the near absoblte
siderablymore scientific instrument weight vacuum conditions of space. Early investiga-
couldbehandledthanwasoriginallyexpected. tions indicated that bearings had been tested in
The designweight was to be approximately laboratory vacuums before, hut with unsatis-
400lb, 100lb lessthanthemaximumcapability factory results. This indicated that a new kind
of the vehiclesystem. of lubricant and lubricating technique was re-
quired. Practically all of the well known
DESIGN OF THE OBSERVATORY lubricants, lubrication concepts, and lubrication
philosophy were of no help in tile solution of the
Tile Orbiting Solar Observatory was origi- problem. A new approach was used, and a dry
nally conceived as a gyroscope. The spinning lubricant technique was developed and success-
portion was to be an open mechanical structure fully tested in the laboratory. The same prob-
carrying gas bottles and batteries at the ex- lem existed for slip rings and brushes. Again,
tremes for spin-up impulses and energy storage, a new technique was developed and tested with
and a stationary structure (with respect to the success.
sun) holding the experiments to be pointed at
the sun. This stationary structure was to be DEVELOPMENT OF THE OBSERVATORY
shaft-mounted through the open spinning wheel
structure, with a bearing supported shaft The observatory development was begun by
through the center of a large gas bottle con- conducting three simultaneous programs. These
taining the pitch control gas supply. As the programs were the development of the electrical
design evolved, it became evident that a higher control system, the basic spacecraft, and the
wheel spin momentum was required than could data acquisition and command system.
be made available with the structure in mind,
and that a life longer than a few days would be Electrical Control System
desirable. At this point, it was realized that
the additional wheel weight required to increase The basic approach in the development of the
the spin momentum could be used for payload, electrical control system was to design a mech-
thus scientific experiments were added to the anized system which could be rapidly assembled
wheel structure. The batteries which were and would perform in tim same manner as the
required to provide the power had to be charged, proposed orbit system. The initial effort cul-
so a solar-cell array was added. minated in a mock-up of the entire system on an
These developments led to the design of a air-bearing-mounted test fixture. The air-
two-section structure: a wheel structure that bearing fixture was built to simulate the space-
consisted of compartments to provide space for craft dynamically. The moments of inertia of
the placement of the wheel controls, batteries, the fixture were the same as the final payload.
telemetry equipment, and experiments; and a This fixture consisted of a large-diameter fiat
large fl_t "sail" structure to carry solar cells. plate mounted horizontally on an air bearing by
From this point, the design was not changed means of a conical center section that allowed
basically, but instead of putting the spin-up gas the center of gravity of the fiat plate wheel to be
bottles in the wheel structure, the bottles were lower than the mounting sphere of the air
put on extendable arms, such that the arms bearing. The wheel then represented the spin-
could be stowed for launch and extended after ning portion of the spacecraft and provided a
orbit was achieved. The reason for this was mounting surface for the control equipment that
to increase the spin momentum. The most would actually be mounted in the observatory
obvious problem that this configuration pre- wtleel structure. An upper structure was then
sented was the limit of the life of the bearings mounted on the air-bearing-supported wheel.
required to support the solar oriented portion This upper structure was shaft-mounted on the
upon the spinning portion. The success of the wheel structure to enable it to rotate wittl
entire venture depended upon the ability of respect to the wheel about an axis (spin axis)
10
" INTRODUCTION AND PROGRAM DESG'RII_ION
no_nal to the plane of the wheel. A bearing- concepts were made as a result of the air-bear-
mounted shaft, perpendicular to the spin axis ing fixture work.
of the wheel, was attached to the upper structure After the operation of the control system
for mounting simulated solar oriented instru- had been proven successful, the next step in the
ments. This method provided a fixture which electronics development was the prototype
had a two-axis _mbal mounting for instruments effort. The breadboard design, in progress as
similar to the actual spacecraft. The center of the air-bearing work was being done, was
gravity of the fixture was adjusted to coincide finalized using experience gained in the air-
with the center of the spherical air bearing bearing work. Prototype electronics were pre-
suspension ball. When balanced, this fixture pared in flight configuration, and flight elec-
essentially provided a free body that would tronic components were requisitioned.
react to external torques as would the actual Prototype electronic subassemblies were in-
observatory in orbit. stalled in the prototype spacecraft after they
The most efficient method for providing the had successfully passed qualifying vibration and
simulated spacecraft with control system elec- thermal-vacuum tests. After the prototype
tronics was to adapt an existing pointing control system was installed and integrated, the entire
to the simulated spacecraft. The system prototype spacecraft was subjected to a series
electronics were constructed from the drawings of environmental tests. Problems which were
of a balloon-borne pointing control system encountered during prototype testing were
previously flown, enabling the use of readily solved and the solutions applied to the flight
available components. Gains and limits were model control system. The environmental
adjusted to suit the new application. A test testing of flight subassemblies was completed
fixture was built to test the electronics before and the subassemblies were integrated into the
the controls were applied to the air-bearing flight spacecraft which was then environ-
fixture. This test fixture was essentially iden- mentally tested.
tical to the air-bearing fixture, except that it
Basic Spacecraft
was fixed to the floor. The test fixture could
rotate and had biaxial gimballing and could As decisions were made with respect to
wobble about the rotational axis by means of an providing a vibration-resistant structure, the
eccentric adjustment at the drive axis. This basic spacecraft design evolved from the antic-
wobble table fixture was used to check out the ipated shape and size concepts into actual
electronic and gas controls. The table fixture hardware. The design to survive the ex-
could be rotated, given specified wobbles, and pected launch thrust was straightforward;
could otherwise recreate the conditions of the however, a design to survive the expected
air-bearing fixture with the exception of the launch vibration was not as simple. A struc-
pitch and roll capabilities of the air-bearing ture which was designed as a flight structure
fixture. was assembled and designated the vibration
At this point, a complete simulated spacecraft structural model. Simulated instruments and
was available for control system development. control equipment were mounted on this struc-
The control system was then installed on the ture to simulate the eventual flight model
air-bearing fixture. The fixture was suspended configuration. Extensive vibration testing was
on its bearing, and the wheel structure was spun accomplished on the structural model. Only a
up to the expected spin rate of the actual ob- minimum of useable numerical data were de-
servatory. An indoor lamp was directed at the rived from this testing but the structure proved
device from a distance to simulate the sun for to be adequate for the vibration loading im-
control system acquisition and positioning. posed. From this point of the development,
The test on the air-bearing fixture provided the only minor changes were made as the work
necessary confidence in the method of control, progressed to the prototype structure. The
the feasibility of the gas systems, and the nutation prototype structure was assembled, and pro-
damper. Several modifications in the original totype instruments and controls were installed.
ii
7_2-552 0--66-----2
ORBITING SOLAR OBSERVATORY
Vibration testing revealed that the only mod- redundant tape recorder. The tape recorders
ifications required were minor changes in the were both running during the launch accelera-
mounting bracketry hardware for some of the tion and vibration to prevent brinnelling of the
equipment in the wheel structure. Testing of bearings. With orbit power turned on, data
the arms in the erection sequence indicated recording began on the main tape recorder.
that the arm structure required strengthening. Also, at 200 seconds after timer start the
Following these modifications, the prototype despin gas jets were turned on. The spin rate
structure survived vibration testing far in ex- for third-stage thrust stability was a nominal
cess of that required to prove structural 120 rpm, but the spacecraft operational spin
integrity. The flight model structure, with rate was required to be 30 rpm. The extension
modifications made as the prototype testing of the arms reduced the spin rate somewhat, but
exposed problems, successfully completed the it was necessary to despin from this rate to 30
flight environmental test with no problems. rpm. The despin gas jets slowed the spin rate
until the automatic spin-up controller sensed
Data Acquisition and Command System
that the rate was at 30 rpm. At this time, the
The data acquisition system for the observa- automatic spin-up controller turned on the spin-
tory was designed to perform the following up gas jets to produce a known incremental
functions: spin-up impulse and permanently disconnect
1. Accept the analog outputs of seven data the despin circuitry so that no accidental despin
channels* and one reference channel and com- was possible. This sequence left the spacecraft
bine them into a frequency multiplex of eight spinning at the nominal control rate.
subcarrier oscillators. After 400 seconds, the upper structure and
2. Demodulate and select the particular tone nutation damper bob, which had been latched
command channel via the receiver and decoder during the powered flight phase of the launch
respectively. to prevent mechanical damage, were released
3. Actuate ten spacecraft command functions. by explosive squib-actuated pin pullers. At
the same time, a latching relay applied the error
Launch Sequence
voltage to the azimuth torque motor drive
The launch and orbital acquisition phase electronics so that the azimuth positioning con-
most severely tested the spacecraft system. trol system could begin to orient the upper
The orbital acquisition sequence was pro- structure to an attitude fixed relative to the sun.
grammed by digital timers which were started The upper structure, which was spinning with
by centrifugal switches. These switches sensed the wheel when latched, slowed and then locked
the centrifugal force produced when the third on the sun. The unlatched nutation damper
stage and payload were spun up by the solid- quickly absorbed the energy of nutation. No
propellant rockets mounted on the spin table detectable nutation was present after the azi-
between the second and third stages. muth pointing controller had positioned the
The 'switches closed properly at spin-up and upper structure.
the timers started properly. One hundred At 800 seconds after spin-up (the last timed
seconds later the timed release of the stowed function of the timers), tile elevation gimbal
arms by means of explosive squib-actuated pin was unlatched by means of an explosive squib-
pullers was properly accomplished, and the actuated pin puller. Tile elevation positioning
arms extended and locked as anticipated. serve mechanism pointed the upper structure
Shortly after arm release, third-stage engine experiments to the center of the sun at this
separation occurred. The launch vehicle timed time.
this separation. At 800 seconds, the timers also repeated all of
Two hundred seconds after spin-up, the the previous functions to give a higher prob-
timers turned on obrit power and turned off the ability that all functions had successfully oc-
*In some cases, several experimental measurements curred. At the end of this sequence, the timers
were combined into a single chatmel. turned themselves off.
12
Chapter 2
SPACECRAFT DYNAMICS
Ys =
z3
cos. .,,,
--cos e sin 4,"
sin _ sin ¢"
cos e cos 4'"
--sin e cos 4'"
sin
cos
(2.2)
13
ORBITING
SOLAR
OBSERVATORY
Z
x2 M
z3 Y2
X
2 SOLAR VECTOR
x2
S
SPIN AXIS
14
° SPACF_RAFF DYNAMIC_
-_-= mgt_a sin _ 0+w,J,3(--sin 0 cos _")+_,3Iv3 (sin 0 sin ¢" cos t
+cos 0 sin t)+_zpr,3(--sin 0 sin #' sin t+cos 0 cos _)
d/\_)T3 2 ,, ,, ,,
_,_--)= m_r_,(/_sin 0+at)sin 20)+I,_t_,3(-sin ocos _ )+ _(-0 cos acos _ +6 sin 0 sin V")]
+I_3[_,3 (sin esin#' cos t+eos 0 sin t) +t)_,_ (cos _ sin ¢" cos t--sin 0 sin t)
+_"_ sin 0 cos _" cos _+i_,.,(--sin 0 sin #' sin t+cos 0 cos t)]
+l,_[b,_(--sin _ sin _" sin t+cos 0 cos t)--t|,% (cos _ sin _" sin t+sin _ cos t)
--z"_
_' z3 sin e cos #' sin t-i_,_ (sin 8 sin ¢" cos t+cos 0sin _)1 (2.6)
_T_.=0
_a
_T_
= m_t)A-_,j,_ sin _"A-w,ff,_ cos _" cos t--_,31._ cos #' sin t (2.7)
d (_-_)--=m_-4-I,_(&,3
dt sin cb" +d/'t_=a cos ¢")-4-I,a(&,_ cos ¢" cost
--d_"'" _3sin d" cos t--tw_" cos #' sin t)
_Ts=l_ m_/_ sin 2_--w,sI,3 cos 0 cos #'+,_w4,s (cos _ sin _" cos t
2
--sin 0 sin t)--awj_s (cos _sin _" sin t +sin 0 cos t) (2.9)
d / \i_T3
-- (2.10)
dt |_-2-r7]--I_a(_,a
\ t_v /
sin tA-i_ cos t)+Iz_(b_ cos t--i_,3 sin _)
6T_ .
b-_=I,_(a_,,a sin 0 sin #'+0_,_cos#')+(_,sin 0cos #'--0sin #')(_J,_cos t--_,_L_sin t) (2.11)
i5
ORBITING SOLAR OBSERVATORY
d /bT3\ .
bT3
--:-(_ sin 0 sin 4_"+0 cos _ tt )( Ca)_I_3 sin e+O_Jz3 cos e)
+ (4." +& cos 0) (,%1_ 3 cos e-- _,al.3 sin _) (2.13)
Equations (2.3) through (2.13) may be simplified by assuming a, _b", &, _ small and 0=2--0' where
0' is small. Then
_3 _ --h+i w_3_O _3-----_"
"" (2.14)
and
dt
since Iq = Iv, = Iv. The moment of inertia of the wheel about its spin axis is 18.
16
SPACECRAFT DYNAMICS
1
(2.19)
1
v(=_ k(_+_+_) (2.20)
1
F,=_ c(g+_+_) (2.21)
In order to write the Lagrangian for the damper, it is necessary t,o express Equation (2.19)
in terms of the x2, Y2, z2 coordinate system. The equation of transformation between these two
systems is given by
or
[x\ /cos. coso --sin.
_y_=/sin
\z/ \
a cos 0
--sin 0
cos a
0
cos sm:)(cos00
inO
sin a sin
cos 0
i)(x) sin ¢" cos ¢"
0
y_
z2+d4
c°s
sin
I.x1
--COS a cos 0 sin _r_
Therefore,
z=x_ (cos _ cos 0 cos ¢"--sin a sin ¢")--y_ (cos a cos 0 sin ¢"+sin a cos ¢")+ (z2+dO cos a sin 0
y = x2 (sin a cos 0 cos O' ' + cos a sin ¢" ) + y2 (--sin a cos 0 sin O" + cos a cos O" ) + (z 2+ d4) sin a sin 0
or
x=cos a cos O(x2 cos ¢"--Y2 sin q,")--sin a(z2 sin ¢"+Y2 cos 4,")+ (z2+d4) cos a sin 0
y=sin a cos O(x2 cos ¢"--Y2 sin ¢")+cos a(z2 sin ¢"+Y2 cos ¢")+ (z2+d4) sin a sin 0
or
17
ORBITING SOLAR OBSERVATORY
(2.24)
/
=-do,+ _-0p j
where
(2.25)
Therefore,
+y_")](x_:'+y_y)+2_[_i'+$_'_'+_"(x_y--y_i')][ec(x_y--y_")
+ (z_+d4)e,]+ 2(_i[(x:Cy--y:Cy) cos 20+ 2( z_+d_)O,e_]-- 2&ie,(x_Y + y_CY)
• 2 t! t! t! tt X _t2 ?22 (2.27)
By assuming ^ _
that _=_-- 0' , where 0' is small, and that a and _" are small; and by dropping third
_n_(_+_+_)___
T_=_- { __+y_+
._ _2_-_b,,_(x2+_)+2¢
_ .,, (_:x_--22Y2)_-&_[_-(z2+d_) _]
+_[_+(z2_d_)_]_20(22-_y2)(z2_d4)._2&(_'x_)(z2+d_)-2&i2y_-20i_x_2&_x2y_} (2.28)
In order to write the Lagrangian equations, the following quantities are derived:
d /5T4\ .. _ ,, ,,
d-t (-_j)=m4{ a[_+(z_+d,) l+a[29_y_+2_(z_+d4)l+@_+¢, x_+_ 2_)(z_+d_)
bT,
5_-- m_{ _[x_+ (z_+d4) 2]+ (2:--ih' ' y_) (z_+d,) -- i2x2+ &x_y2}
:It
(z_+d.)
18
° SPACECRAFT
DYNAMICS
or °
d bT. ,,. (2.30)
_r4 ,, x
_-_7= m,[ff (x_-q-_) q- (_2x2-- "2Y_)--_(z2-q-d,) -q-&x_(z_+d,)]
d bT4 "" z ....
-_(_-77)=m'[(_q-_)q-2¢ ( 2Xz_-y2YZ)'_-(Y22:'2 = 2Yz)
d c)r4 ,, ,,
_(_):m,[_2--_ y_--¢, _-I-_(z_--l-d,)q-#_]
or
d (bT,'_,,,m,t_c__q_(z2q_d,)O l (2.32)
dt \b2#=
_T_ ,,
_ff2= m,[_2q-ff x2"-I-a(y2-q-d,)]
d bT4 ....
(2.34)
_T4
-m,i,b " If2 l!
y_-¢, .¢_+a_yr(_ l!
(z_+d,)--_,z_+aOz_] (2.37)
by_
bT'=m_[(&2q-_)(z_q-d_)q-_(" 2,_--4_"'y_)q-a(y_-q-cb"
" ""x3)] (2.38)
bV bV bV_ o (2.39)
ba -- ba _-7_ --
(2.40)
(2.41)
(2.42)
.i.,,t
ORBITING SOLAR OBSERVATORY
5F 5F 5F= 0 (2".43)
b_ -- b0 -- 56"
bE (2.44)
bE . (2.45)
_22= cy2
bF (2.46)
d bT_ _ (2.47)
+ ¢,'"_(--/_2+I_+I,2--/_3+ha+Iz3)
m_i)2+c_)2+ky2=--ma(z2+d_)5_ (2.54)
T_=F_la (2.58)
and
20
SPACECRAFT
DYNAMICS
" If the damper is considered to have only two degrees of freedom, that x_, Y2 and z2 are small
compared to d4, and that the upper structure azimuth angular velocity and pointed instrument
angular velocity are small, Equations (2.48) through (2.58) simplify to:
(I,_+I,,)Yh"----A(s)
I# '=--A(s) (2.59)
(_--a)I,s=E(s)
_+ 2ny2-}-p'_y2 =-d4&
where
C }
2n-----
m4
(2.60)
m4
Equations (2.59) were solved on the Beech The outputs recorded were the two servo
Aircraft CRC analog computer using the follow- pointing errors and the angular velocities of the
ing initial conditions: spacecraft and nutation damper. (Velocities
Time history traces of the computer study tion at precession torque removal is given on
were obtained on a s[x-channel Offner recorder. page 36.
21
ORBITING SOLAR OBSERVATORY
22
SPACECRAFT DYNAMICS
LINE OF
NODES
(ROLL
AXIS)
SPIN !
% SOLAR
AXIS
VECTOR
JET THRUST
PRODUCING
TORQUE
LINE OF NODES
23
ORBITING SOLAR OBSERVATORY
to eliminate such random motion in order to of the behavior of the damper selected for .use
assure spacecraft stability. was imperative in order to evaluate accurately
the effect of gravity on its performance. Sim-
Design Requirements plicity of and confidence in the ground testing
operation thus became strong design consider-
The general design requirements that were
ations.
established for the nutation damper are as
follows:
General Approach to the Solution
1. Damping rate.--Compatible with a time
constant of ten (10) seconds for the exponential Various Types of Dampers Considered
decay of the nutation amplitude.
In the process of finding a damper that met
2. Weight of damper.--Minimum weight for
all of the specified requirements, many different
established damping rate.
types of damping devices were considered.
3. Size of damper.--Compact as possible and
Most of those given serious consideration were
compatible with overall spacecraft geometry.
passive dampers; tbat is, dampers driven by the
4. Nutation ]requency.--Dependent upon
nutational motion they eventually damp out.
principal moments of inertia and spin rate of
Since the Orbiting Solar Observatory spun
spacecraft. Calculated frequency of 4.71 rad/
about its axis of greatest moment of inertia, a
sec was based on the following values:
passive damper could be used to absorb the
¢0----0.5 rps nutation energy without adversely affecting the
I----30 slug-f_ 2 spin. A passive damper was preferred over an
]I----20 slug-ft 2 active one for this application because of its
reliability and simplicity of operation, and
5. Maximum anticipated external torque.-- because it did not require an outside power
1 lb-ft (reaction of elevation servo motor on source.
instrument platform during initial acquisition The three general types of dampers investi-
of instrument). gated were pendulum dampers, inertial dashpot
6. Ang_dar amplitude of nutation due to moxi- dampers, and a torus partially filled with
mum torT_e.--Calculated value of 7.4 minutes mercury. Simple, spherical, and torsional pen-
of arc, based on spacecraft parameter values dulums, and _ ball rolling in a curved tube were
above.
considered. Practicality of mounting the var-
7. Minimum aUowable residual n_ltation ampli- ious type dampers on either the rotating or
tude.--Less than l arc-minute (compatible with nonrotating sections of the spacecraft was also
accuracy of fine elevation pointing control). taken into consideration. Some of the damping
8. TemperatT_re range.--O ° to 50 ° C. schemes included among the various combina-
9. Test requirements.--The ground opera- tions of damper type and position were easily
tional test requirement for the nutation damper seen to be impractical. The investigation of
was that its in-flight performance be conclusively each of the other combinations continued until
demonstrated in the laboratory. Appropriate the damper could be ruled out or until a theoret-
test apparatus had to be devised to meet this ical evaluation of its performance could be made.
requirement. In the following discussion, the results of
The presence of gravitational acceleration the analysis of each type of damper concluded
during ground testing in contrast to the absence to be feasible for spacecraft use are summarized.
of acceleration in orbital flight contributed For those types not chosen for this particular
significantly to the complexity of the problem. application the reason is given for not selecting
The behavior of many type_ of dampers is the damper.
altered by the effect of gravity, often to the Mercury Ring Damper--A hollow toroidal
extent that a device that performs well in one ring partially filled with mercury or some other
environment will not function in the other. more viscous liquid, has been proven to be an
Furthermore, a complete mathematical analysis effective nutation damper. A theoreticalanaly-
24
° SPACECR_ DYIq_CS
25
ORBITING SOLAR OBSERVATORY
26
• SPACECRAFT DYNAMICS
pound, a typical value for the wheel radius is friction of the ball, or shock-absorbing
three to four inches. Thus, the simple pen- "bumpers" may be placed at either end of the
dulum is sex to eight times more efficient, in tube to absorb energy.
terms of damper weight for a given damping This type of damper exhibits one distinct
rate, than the torsional pendulum. advantage over any of the types mounted on
Rolling Ball Damper--A ball free to roll in a the nonrotating body. The frequency of vi-
closed curved tube can be used to damp out bration of the damper, like the nutation fre-
nutation if mounted on the rotating body as quency, is directly proportional to the spin
shown in Figure 2-6. The tube may be filled rate of the spacecraft. Therefore, the damper
with a viscous fluid to combine viscous damp- remains tuned or in resonance with the
ing with the damping caused by the rolling disturbing frequency for any given spin rate.
27
782.-552 0--66--3
ORBITING
SOLAR
OBSERVATORY
The principal disadvantageof this or any motion as the spherical pendulum damper.
othertype of dampermountedon the rotating Spherical Pendultem Damper--Selection of the
body was that the frequencyof the distur- spherical pendulum damper was based on the
banceit experienced wasequalto thedifference following considerations:
between spinfrequencyandnutationfrequency. 1. Damping efficiency of this damper was
This disturbingfrequencywasalwayslessthan concluded to be equal to or better than any of
nutationfrequencyfor a disk-shaped rotating the other feasible types.
bodyspinningaboutits axisofgreatestmoment 2. It was possible to make a theoretical
of inertia. For this spacecraft,it was about analysis of its behavior that could be conclu-
one-thirdof the frequencyof nutation. sively substantiated in simple experimental
The undesirableeffectof a lowerfrequency tests.
disturbancecan be seenby examiningthe 3. The effect of gravity on its behavior could
energyabsorptionequationderived for the be evaluated accurately.
variousdampers.In eachcase,the absorption 4. A modified version of the damper could be
rate at resonance is proportionalto the fourth easily incorporated into the spacecraft ground
powerof the disturbingfrequency. To main- simulation tests on the air-bearing fixture.
tain a given dampingrate as the disturbing The possibilty of mounting this type damper
frequencyis lowered,eitherthe inertialmassof on either the rotating or nonrotating sections of
the dampermustbe increased or the damping the spacecraft was investigated. Valid argu-
constantdecreased.Decreasingthe damping ments for choosing either location could have
constantresultsin an increasein the required been made. The decision to mount it on the
amplitudeof vibration of the inertial mass. nonrotating body was based on the following
Within the spacelimitations of the rotating conclusions:
sectionof theOSOI, this typeof damperwould 1. On the rotating body, a pendulum damper
havebeendifficult to designto satisfyboth the would have to be mounted in an inverted posi-
dampingraterequirement andminimumweight tion in the centrifugal force field in order to tune
requirement. the damper to the relatively low frequency dis-
Other problemsassociatedwith this type turbance it would experience. The load on the
damperwere: wire supporting the bob of an inverted pendu-
1. Becauseof static friction, the damper lum becomes critical as the natural frequency
wouldnot havecompletelydampedout nuta- of the pendulum approaches zero. The tuned
tion andcouldhavecauseddynamicunbalance frequency required in this application was suffi-
of therotatingbody. ciently low that a slight increase in spin rate
2. The effectof gravity on the behaviorof above nominal value resulted in buckling of the
the damperduringgroundtestingwassignifi- support wire. The resulting restriction on spin
cant for the relativelylow spinrate specified rate deviation was considered too severe.
for the OSOI. The groundcheckoutof the 2. Since the frequency of the disturbance ex-
nutationdamperin spacecraft simulationtests perienced by the damper mounted on the rotat-
wouldhavebeendifficult. ing body was much lower than nutation fre-
Simple Pendulum Damper--The simple pen- quency, the damping efficiency was also lower
dulum and spherical pendulum dampers are than at the alternate location, assuming fixed
analogous to the two types of inertial dashpot values for mass and damping constant. Re-
dampers discussed earlier: the simple pendulum duction of the damping constant to compensate
is restricted to one degree of freedom, whereas for the lower frequency value was not feasible
the spherical pendulum has two degrees of in this case. The resulting amplitude of vibra-
freedom. Otherwise, the principle of operation tion of the pendulum bob was large enough that
and the analysis of the behavior of the two types the support wire would have failed in bending
is essentially the same. The simple pendulum at maximum displacement.
damper was not chosen because it was only 3. Ground testing was greatly simplified by
one-half as efficient in damping out nutational selecting the location on the nonrotating body.
28
• SPACECRAFT DYNAMICS
An=lysis of the Spherical Pendulum Damper metric study of the system was considered to be
impractical until some knowledge could be
There were certain disadvantages to mount-
gained concerning the general behavior of the
ing the nutation damper on the nonrotating
nutation damper and the servo mechanisms
body, e.g., the damper did not function when
connecting the three rigid bodies of the spac_
the spacecraft was in the shadow of the earth. craft.
During that time, the servo motors were inop- The more practical approach adopted was
erative and the nonrotating body gradually ac- first to analyze the behavior of the damper
quired angular momentum from the rotating independent of system operation. Tiffs step
section until both bodies were spinning at the was particularly important in simplifying
same rate. When the pendulum damper was the selection of the most efficient damper of
rotated, the normal static equilibrium postion many types considered. The analysis of each
of the bob was no longer a stable point, and of the dampers under steady forced vibration
the bob was moved by centrifugal force to a was used for this ptupose. The approximate
position of lower potential against the damper values for damper parameters obtained from
case. It maintained this fixed position relative this analysis were then used to design models
to the case for as long as rotation continued. that could be checked out experimentally on
In assuming the new position, the bob created ground test fixtures which simulate the oper-
a small dynamic unbalance in the spacecraft. ation of the damper-spacecraft system during
Both the loss of the damper and the small flight. Using the Optimum values for damper
amount of wobble caused by the dynamic un- parameters gained from the simplified analyses
balance could be tolerated since both condi- and accompanying experimental work, numer-
tions were remedied shortly after the spacecraft ical solutions of the equations of motion of the
emerged into sunlight again. Another unde- complete rigid-body system were obtained.
sirable feature was that the natural undamped Steady Forced Vibration Ana/ys/s--From this
frequency of vibration was fixed. Since the simplified analysis, the equations of motion of
nutation frequency varied directly as spin rate, the damper under sinusoidal excitation of con-
the damper would not have remained sharply stant amplitude and frequency were obtained.
tuned to nutation frequency as the spin rate An expression was then derived which gave, in
varied between allowable limits. However, the terms of damper parameters, the rate at which
damper was designed to be rather efficient in the damper was capable of dissipating energy.
damping out nutations with frequencies 10 per- When the required energy dissipation rate was
cent above or below the nominal value. Dis- substituted into this expression, approximate
turbances of these maximum or mimimum al- values for the damper parameters could be deter-
lowable frequencies were damped out with a mined. Of particular interest was the weight
time constant of about 20 seconds, or twice the of the vibrating mass of the damper since
optimum value. damping efficiency is directly proportional to
Because of the complexity in operation of this weight.
the spacecraft-damper rigid-body system, a The steady forced vibration analysis could not
complete mathematical solution of the rigid- be used to determine whether the torque applied
body problem was not attempted initially. to the spacecraft by the damper was in the
Even when small angle approximations are proper direction to damp out nutation. This
used, the differential equations which describe determination was made in each case by direct
the motion ol the system are nonlinear and form analysis of the vector diagram of the rigid-
a set of seven simultaneous second order body system.
equations. Analytical solution of these Another deficiency of this analysis was that
equations is not possible. Numerical solutions it did not take into account the effect of the
are possible using an analog computer, but damper in decreasing nutation amplitude as
such solutions require numerical values for kinetic energy was taken out of the system.
all of the parameters of the system. A para- Since (in this approach) constant amplitude
29
ORBITING SOLAR OBSERVATORY "
of the forcing function was assumed, a constant complete analysis was also of importance in
energy dissipation rate was obtained. Actually, evaluating dynamic stability of the serve-
both of these quantities constantly decrease as connected spacecraft rigid-body system and in
nutation is damped out. It was reasoned that gaining a more complete understanding of the
a more accurate description of the motion of the gyroscopic motion of such a complex system.
damper-spacecraft system during nutation could As previously mentioned, the equations of
be obtained from the analysis of tbe behavior motion of the system formed a set of seven
of a two-body system during free vibration. simultaneous, second order, nonlinear differen-
Two-Body Free Vibration Analysis--The two- tial equations. These resulted from the seven
body system used in this analysis was the degrees of freedom of the complete system.
spherical pendulum damper mounted on the The nutation damper had two degrees of
pendulum platform test fixture. The problem freedom and the spacecraft five (three for the
was simplified by considering only single- rotating body, an additional freedom in yaw
plane motion of the system. Use of the damp- for the nonrotating section, and a further one
er-test fixture system in single-plane vibration in pitch for the pointed solar experiments).
as a mathematical model of the damper- Since the three rigid bodies making up the
spacecraft system during nutational motion spacecraft were connected by the azimuth and
greatly simplified the analysis of the problem. elevation serve motors, the dynamic response
Despite the simplifying assumptions, this of these serves had to be taken into account
approach was still considerably more complex when analyzing the dynamic behavior of the
than the steady forced vibration analysis.
spacecraft. For this reason, the equations of
The differential equations of motion of the
motion of the spacecraft without the damper
system were easily written but an analytical
were derived first and solved on an analog com-
solution in terms of system parameters was not
puter. This analysis was essentially a para-
possible. Because of the lengthy and tedious
process involved in obtaining numerical solu- metric study of serve characteristics and was
tions by hand, an electronic analog computer used to augment breadboard tests of the atti-
was used. The parametric solution of the tude control system. The nutation damper was
equations, obtained with the aid of the com- incorporated into the system only after satis-
puter, yielded optimum values for damper factory response of the serve system had been
parameters. Results of this study were sub- established.
stantiated by experimental tests of the damper
on the pendulum platform test fixture. Design Procedure
Analysis o[ Complete Rigid-Body System--
This analysis was performed to verify the In the following discussion the pertinent
results of the simplified analyses and prelimi- results of the mathematical analysis of the
damper would damp out nutational motion A' sin pt, and
of the spacecraft but also that the results of y=A p' cos (pt--a) (2.63)
the simplified analyses and the tests performed
on the pendulum platform were correct. The in the plane in which the forcing function is
3O
SPACECRAFT DYNAMICS
A"cos pt. In Equations (2.62) and (2.63), the In Equations (2.68) through (2.70) the nutation
following notation is used: is as follows:
31
ORBITING
SOLAR
OBSERVATORY
32
• SPACECRAFT
DYNAMICS
the,desirednaturalfrequencyfor thependulum Damper Model Design
couldnot be attainedin the upright position A preliminary model of the spherical pendu-
withoutunreasonable pendulumlengths.There- lum damper was designed and built, using the
fore,the invertedpositionwasassumed for all values for damper parameters obtained from
ground tests conductedto evaluatedamper the steady forced vibration analysis. The
performance.Equation (2.76) was used to dimensioned drawing of this model is presented
determinethe diameterof a steelmusicwire in Figure 2-7.
requiredto yieldthe desireddamperfrequency. The damper was designed for easy modifica-
In this calculationthependulumbobweightof tion. The case could be easily disassembled,
onepoundpreviouslyselected wasused,andthe and the spring wire pendulum arm was screw-
value, [ =3 in., was arbitrarily chosen. attached at either end. Thus the damping
2 2
For a spherical bob, pl=_'r 2, and for a one- constant could be varied by changing fluid, and
pound brass sphere, r=0.92 in. ; then p_----0.339. the frequency could be varied by adjusting
The modulus of elasticity for steel, E, is 3X pendulum arm length or by replacement with a
107 lb/in 2. Therefore, to obtain the desired wire of different diameter. The cylindrical
frequency, p=4.70 rad/sec; I=1.40 X10-7-- - wall of the case was made of clear plastic so that
_d4/64 and d=0.041 in. To obtain the desired visual observation of pendulum bob tests could
frequency for the damper in orbital flight, be made.
Equation (2.77) is used to determine the correct As discussed previously, the two fixed param-
wire diameter; d=0.025 in. eters for the model were: weight for bob of one
5 IN. DIAMETER •
---_'_
1/4 - 28UNF 3B
!
/ i
V
q
7/16
STEEL COVER
\
4.000
7/32 IN. DIAMETER
4. 001
HOLES ON 4-1/2 IN.
SPHERICAL BC
RADIUS
\- PLEXIGLASS TUBE
(POLISH)
4IN. OD, 3IN. ID
/
/
//PLEXIGLASS BASE
13/32 !
i
I
i
I
8 - 32NC - 3A SCREWS
i 3-1/4 IN. DIAMETER -----_ i
ON 3-5/8 IN. BC
4 IN. DIAMETER
33
ORBITING
SOLAR
OBSERVATORY °
34
• SPACECRAFt
DYNAMIC8
Far the air-b_aring fixture: minus 10 percent to nominal value plus 10
m_=0.234 lb-see2/in. percent to determine the response of the damper
m2=2.61 X 10 -3 lb-sec2/in. to this variation in forcing frequency.
p1=3.92 rad/sec Results of the study are presented in Figures
3.53/sec _P2 < 4.32/sec 2-8 through 2-11.
c,=0 or 5X 10 -3 lb-sec/in. Figure 2-8 is the plot of time constant versus
5X 10-4 <e2< 10 -2 lb-sec/in. the damping constant, c2, obtained from Step 1
above. The optimum value of c2=0.0012
where m2, t'2, e2 refer to mass of pendulum bob, determined from this study was later found to
frequency, and damping constant of damper, be slightly in error due to the assumption that
respectively; and ml, Pl, and c_ refer to similar p_=p2. ._ter completion of Step 2, Step 1 was
parameters of the platform. The nonzero value repeated, and the value c2=0.00175 was found
of c, was assumed to correlate computer results
to be the minimum value for which beating did
with test results from the platform fixture.
not occur when the damper was sharply tuned.
This damping constant for the decay of free Figure 2-9 shows the results obtained from
vibrations of that fixture was determined
Step 2. This plot of time constant versus
experimentally.
damper frequency shows that the damper had
The procedure that was followed in optimiz- to be tuned to a frequency slightly less than
ing damper parameters was:
nutation frequency for maximum damping rate.
1. To optimize c2 for fixed value of p2 equal Figure 2-10 is the plot of time constant versus
to p_ (it was determined from the test work
pendulum bob weight obtained from Step 3.
that the optimum value of P2 was very nearly A bob weight of about 0.75 lb yielded a 10-
this value).
second time constant for the assumed nutation
2. Using the value of c2 obtained in Step 1, frequency of 4.13 rad/sec.
to find the optimum value of p2.
3. To determine the value of m2 that yielded r, SEC
the desired time constant of 10 seconds for
20
optimum values of c2 and p2.
4. Using the obtained values of m2, P2, and
c2, to vary p, over the range from nominal value
18_
r, SEC
Pl
16-
14.
12
9 5 BEATING
• --o_-c_.--i_--_-
lO
35
ORBITING SOLAR OBSERVATORY
dH,, , , "_ •
24
22
r i : i
! clH.
t-_,,H,--o_,H,,=
, ,
M,_ I
I
[
20- ..... l-
i
dH,. • _o'
,,,H ,_-- _o'
,,H,,,=M,I
.)
18-----
_ 10-
The body angular velocities are then those of
the trihedron plus body motion with respect
8-
to the trihedron, namely,
r
251
rate with nutation frequency. Within the
frequency range of nominal value ± 10 percent,
the damper was very efficient.
r,SEC 20 o - -
Pl' RAD/SEC
Equations of Motion--Euler's equations re-
lating to the trihedron axes OMNS (Figure FIGURE 2-11.--Time constant versus nutation fre-
2-12) are written as follows: quency (p_=4.1 rad/sec; c_=0.0012 lb/in./sec).
36
SPACECRAFT
DYNAMICS
6
Z If 0=_/2 initially and M.=M,=0, this angle
will not differ greatly from its initial value
during torquing, and Equations (2.85) are
further simplified. From the last of these
n
equations, it is apparent that
LINE OF
N NODES
or that _=_=constant.
The first two equations then are written:
Y
(2.86)
M
For the asmmaed initial conditions: a(0)=
O-x-y-z : INERTIAL FRAMF
&(0)=0=0; O(O)=,r/2, the Laplace transforms
O-_---S
O-M-N-S
: FIXED
= TRIHEDRON
IN BODY
WITH O-N
of Equations (2.86) are:
LINE OF NODES
O-S IS SPIN AXIS OFSPACECRAFT
I,O--I,& _ sin 0 cos 0-4-Id_sin 0 (2.84) The inverse transforms of Equations (2.88)
(_b+a cos o)=M,, are:
d M. t-M, sin pt ]
I_ (_b+& cos O)----M, a= _ I_p
(2.89)
Equations (2.84) are simplified by assuming
O=2+/_p (1--cos Pt)
that the velocities of the trihedron, _ and &, are
small compared to $. Eliminating all terms
From Equations (2.92) it is determined that
with products or higher powers of the small
the curve of the intersection of the figure axis
velocities in Equations (2.84), we have:
and a spherical surface described about the fixed
point of the gyroscopic body is a cycloid. This
--Ila sin O+IOgh=M., ]
curve is shown in Figure 2-13. The cycloidal
(2.85) motion of the figure axis has the period,
I_O+IiM, sin o=M_
r=2_/p, and the amplitude, a=2MJI_,p.
I d (¢+acosO)=M.J Since the body is spinning about this axis of
37
ORBITING SOLAR OBSERVATORY
z i
x Y
"_-PATH OF ANGULAR
MOMENTUM VECTOR
In terms of the unit vectors, i, j, and k, along -_( I_ \M, (1-cos pt) =0
the S, N, and Z axes, respectively, we have
Therefore, H remains in the x-y plane during
constant torquing.
H=I_0cos 0,--_ i--Io_ sin 0,-- k Next, the rate at which H moves in this
plane is determined. In Figure 2-14 let
+Ii0j +I,&k (2.91)
38
• SPACI_RAFT DYNAMIC8
!
or
(2.98)
--I_°=0
aH=a,+sin-' It@ (2.94)
Ls_g(s)--I_sOo--Ii_o+Ios/(s)--I,,,_=0 J
from which
If & and $ are sufficiently small in compar-
ison to _, Equation (2.94) can be simplified to:
(2.96)
_ ,0_o_0o_ospt+_ sinpt 1
a--ao-l-p -_ (2.100)
Thus, except for small pulsations brought
about by the (_+_) term in Equation (2.94)
H moves at the constant rate, MN/Io, in the
xy-plane.
The magnitude of these variations from the Thus, the figure axis performs harmonic motion
average rate can be determined by differen- about the position,
tiating Equation (2.94) with respect to time
and then making the appropriate substitutions
for a and 0 and their time derivatives from
Equations (2.89). The resulting expression described by the equations:
reveals that at
Since
39
ORBITING SOLAR OBSERVATORY
1. Upper asse_nbly:
(2.102)
O----00--_=2-t-/_pp (i--cos pt 0)
Hm_------Imp& sin 0 (2.103)
H_=In_ }
Mn (1--cos pto)-----_ H,2= ,I _a cos
l_op
2. Instrument:
From Equation (2.95), the a coordinate of
the angular momentum vector is found to be:
to Mn
Hm3------Im3a sin $ (2.104)
sinpt0
H,3= 1,/x cos
which is the same as the a coordinate of the
center of the circle given in Equation (2.102), where the subscripts n, m, and s refer to the
above. Since it was shown in the preceding moment of inertia or angular momentum of
section that the angular momentum vector, H, the body about the N, M, and S axes, respec-
remains in the xy-plane (8----,r/2) at all times tively, and the subscripts 2 and 3 refer to the
during torquing, it is conchlded that when upper assembly and instruments, respectively.
torquing of the gyroscopic body is stopped, the Equations (2.103) and (2.104) are derived
residual motion is a circular motion of the spin assuming that the azimuth servo motor pre-
vector about the angular momentum vector on vents rotation of the upper assembly about
4O
° SPACECRAFT DYNAMICS
the spin axis and that the elevation servo pre- Laplace transforms of Equations (2.108) for
vents rotation of the instrument about the the initial conditions, a(O)=a(O)=O(O)=O, 0(0)
M-axis. The inertial term, Ira3, then applies =7/2, are:
41
ORBITING SOLAR OBSERVATORY
OT
The kinetic energy is --=m (_ + A'p cos pt)
bi
1 1 OT
T=_ mv2----_ m(_'2A-?_'2A- _ '_) -_= m (y-- A p sin pt ) ;
or 5T
1
T=_ m[(_-4-A'p cos Tt)2-4-O)--A'p sin pt)2-4 - _2] _) ..
(2.113)
i_) ..
The potential energy due to the strain energy
dt
of bending the pendulum arm wire is, as pre-
viously derived:
--
d(s_)
dt =my--mA'p
" 2 cos pt
:rE1 SX/d'ry
V'=Jo 2-Pp
_dz_= 2 \dz 2] dz (2.114)
and from Equation (2.119)
42
• ,SPACECRAFT DYNAMICS
Letting 2n-----c/m and p'_=k/m, Equations the spherical pendulum damper, it is first
(2.121) become: necessary to show that for small vibrations
the natural frequency of the damper is the
same when operating as a simple pendulum
#T2n_+p'2y=A' fcospt (2.122) as when performing conical pendulum motion.
_ + 2n_c +p'2x=A' p2sin pt }
+2n_--0 This is done by considering the free undamped
oscillations of the damper during each type of
Solution of the third of Equations (2.122) motion.
is easily determined to be:
m
x-----A" cos
y=A" sin (pt--a)
(pt--a) } (2.124)
where
From Equation (2.125), the energy dissipation where k is determined from the spring force
rate is determined to be: of the wire and the gravity force, rag. The
Lagrangian equation of motion is:
m#+ky=O (2.129)
cp _
___A_'2 (2.126)
(1--p_/p'2) 2+ 4n2p2/p ' ' Equation (2.129) is the familiar expression
describing free undamped sinusoidal oscilla-
At resonance, this expression becomes:
tions of frequency, p=_/m.
43
782-552_
ORBITING SOLAR OBSERVATORY
(2.131)
44
t
$PACECRAPT DYNAMI(_
Y(x)=a (1--cos 2 )
where the positive sign is used for the upright
pendulum and the negative sign for the in-
the above expression may be rewritten to: verted position.
(c) The gravitational potential of the dis-
tributed weight of the wire is:
Furthermore, we have
p,= ± _t [w fo" l (dy_'\dx]dx] dx (2.139)
d#=,, apT. Tx
d_ -- _- sln with the sign convention the same as in Equa-
tion (2.138).
and at Y_-Yt:
Substitution of Equations (2.133) through
dymaz ap_- (2.139) into Equation (2.132) yields the fre-
&---2t quency equation:
w _ dY_dx]
p2____ (2.140)
45
ORBITING SOLAR OBSERVATORY
y=a (l--cos _)
El fo ! _-_x2)/d2y\2"ax=
a21r4EI w fo I fdy'_ _ , a_Ir2wp_
EI + O.405W [ _q-O.12Ow[ a
p = 34.4 4W [3 q_ 2.467W [ p_-f- 0.227w [4 q_ 1.234w[ _p] (2.141)
When the weight of the wire is small compared Kinetic energy of the system is:
to bob weight, Equation (2.141) can be sim-
plified to the expression: T=½ m,.C_-t-½ m_(.C,+ Jcz)2 (2.144)
... f EId:O.405W[ 2
(2.142)
--I x2 I'--
kI _2
p= 1 •745 _/m[3+2.467mg,
E1 p_ (2.143)
46
• SPACECRAFT DYNAMICS
T=½[rnl+m2(l+_1)*]x2+[m_-m2(l+fll)(l+_,)]._+½[ml+rn,(1-4-_,)2]?) 2 (2.148)
The cross product terms in Equations (2.148) and (2.149) are eliminated if:
The Lagrangian equations of motion may be written from Equations (2.151) as follows:
a11x,-4-bux-_-CllZ-_ b12y=O _t _
a=_t-t- bz2_ A-c22y+ b12_=0 J (2.152)
47
ORBITING
SOLAR
OBSERVATORY
SPIN-ROLL pLANE
SPIN AXIS
ELEVATION POSITIONING
_ SOI_R DIRECTION
SPIN-PITCH pLANE
SPIN AXIS
ANGLE i / _
V -- CENTER OF MASS
PITCH AXIS
PITCH-ROLL PLANE
FIGURE 2-17.--Aerodynamic forces during an orbit.
48
a
SPACECRAFT DYNAMICS
y2 z2
R=radius of the earth
r_ cos2/_ _-_-=1 z__0 (2.154)
and
where f_is the angle from the spacecraft velocity h=height of spacecraft above the earth
vector to the flatplate normal. The integrands
of Equation (2.153) become During the sunlight portion of the orbit, the
upper structure area presented to the aero-
dA' =cos [3_/_--z_dz (2.155) dynamic force is
and
zdA' :cos _(z_-'-_dz) (2.156) Az=A cos 0 (--0' <0<_r[2)
and
The integrations over the area give
As=--A cos 0 (r/2<0_r+0')
D_=D sin 0
The torque upon the spacecraft during the =oV_A('Y+d) [--4 cos O'+ (1--sin2 o')]
orbit night is
-- /'2_r-O, (2.161)
T_-=CD-_ P V_A('_+d) J.+o, sin ado
In order to evaluate the net torque, an
=--pV,A_+d) (4cosO ") (2.158) altitude of 300 nautical miles=346 statute
miles--556 kilometers was assumed. The air
where
density, from damping calculations on Sputnik
d=distance from the lower edge of the It WaS
49
ORBITING SOLAR OBSERVATORY
The density at h=556 kilometers is from the earth was less than the force on tt_ose
parts nearer the earth. This difference pro-
p_58----(4 X 10-13) e -0.44 duced a torque. The magnitude of this torque
largely determined the amount of control gas
= 6.4X10_10 gm 1.24X10_,5 slug
cc ft a required for correction of externally caused
motions.
Now Figure 2-18 shows the coordinate system used
in this analysis. Axes x, y, and z form a geo-
p55BV2A= (1.24) (6.125) (5.275 X 10 -v) lb s
centric ecliptic system with the sun always along
=4.01 X 10 -6 lbs the x-axis. Unit vectors in this system are i,
j, k. The u, v, w sysrem is a geocentric equa-
and for
torial system with the u-axis along the equa-
torial-orbital line of nodes. The x', y', z' sys-
tem is a rotating geocentric system with the
spacecraft always on the x'-axis. The unit
the torque is vectors are i', j', k'.
I
a=/_ t (2.163)
Z' _ X
/ sec
(1.3X10 -6 lbrft) (3600X24
x. day/(57.3 deg)
r-I-
(26 lbrft-sec 2) (21r rad_ rad
x, see / _y
deg EQUATORIAL
=0.04 day NE
5O
• ,SPACECRAFT DYNAMICS
X,fi,+X,ecfic
(2.164)
Wl \ es_. --e._c e¢I \Zl
where
_,_=cos, e.=sin e, etc.
,I?._ U
(2.165)
z' l \ o -i.
and
x"\
y"]=/0
z"/
[1
\0 -a.
0
-c o)(:)
,,.
Ol
=(c) (2.166)
-i_(x,n,-x_,a,)] }j Let
R=R,+r' (2.170)
+ [,,(X,e,) + n, (i,e,+ i,Xee,) ]k
Where Re is a vector from the center of earth
=(Ax cos n+A2 sin n)i 1 to the center of mass of the spacecraft.
The wheel of the spacecraft is continuously
+ (Bx cos+(6'1 cos sin
v/-_-a2 '7)j sin _)k _
_+C_ (2.167) rotating, giving the spacecraft gyroscopic sta-
where bility. The sail with its solar cells and experi-
ments is directed at the sun during the orbit
Al=cos X cos fi--sin X cos e sin fi
day and, because of bearing friction, rotates
A2=sin isin esin fi--cos i (sin Xcos f_ with the wheel during the orbit night. The
+cos X cos e sin fi) spin axis is maintained nearly perpendicular to
the solar vector by means of a pitch gas system.
Bl=cos X sin _+sin X cos e cos
(2.168) Let
B2=--sin i sin e cos _--cos i
51
ORBITING
SOLAR
OBSERVATORY °
Then R-S=(R. R)-a/_=[(Ro+c,+r,). (R_+c,+r,)] -a¢_
L (Rc+c,+r,)]pdv =[R,. Rc+2R_ • (c,+r,) + (c,-_-r,) 2] -s/2
------gR_ f, R-3[(c,+r,) X
_=l
The quantity R -s in the integrand may be and
approximated as follows: iI. i' = 1
3i'.(c,+r,)'] (2.174)
R-3_-_R:3 [ 1"_ 2i"(c'-_-r')_-3/2 .J R:a [ 1-_c J
and
-----R-_I,gR_
f R_×(c, _ My+ f. r,Mv)-3 f_ [i'× (c,+r,)][i'. (c, +r,)]pdv } (2.175)
f pdv=m_
(/(
•"
Y"
\z"/
=--/_ _
\ o
a_
B_ 0 |Y'"/
o
0) \z"'/
(2.177)
where
where m_ is the mass of the i th part. Further-
more, we have
B_can have the value
t_'=sail position angle
_(R_Xc,) f ev=R,X (m,c,-l-m,c2)----O or
52
SPACECRAFt DYNAMICS
A_idingtheselasttwo equationsyields
The spacecraft positional vector it, may be expressed in the i t', jtt, ktt system as
=ai+bj+ck
where
b=(it.j)
c= (i t. k)
a= (i t. i)
J=ba,+ca,
K=--b_,+_
=i" K {af3_._,(ytt "-- z' " _)+ J[ (z' " + c,) _-- (a_.2zt' t 2+ f3!_ytt ' _)] + ...}
+ktt[(aBi+dB_)(a&*--dB,*) (x"t2--y'"2)+...l
L=_3oJ2(itt{K}{a_B_ (I:--I,)
t t + d[#. t' (I.--1,)
, t +_o,, (I,--I,)
t l +_mt] }+j" { K} {J_,13,
, t (I,--I.)
t i
--a[#, t t (I.--I,)+f3.
f t l t f t
(I,--1,)+_m,]}+k ?t t i i i f
[(aBc+JBD(aB,)(aB_--J{3,)(I_--I_)])
= -- 3o 2(i t' { (aca,--aba.)_I_ (I _--- I;) + [ (c_- b2)a,a,+ bc (2a) ,ltOf(I',--P.) + O_,
' (I_, -I _-)+din i] }
55
ORBITING SOLAR OBSERVATORY
where the superscripts on the moments of LsD -- torque due to the sail (orbit d_y
inertia indicate that the moments of inertia can only)
be those of the sail or of the wheel. Now let LsN----torque due to the sail (orbit night
L_----torque due to the wheel (orbit day only)
and orbit night)
• W W
Due to the symmetry of the wheel about the spin axis, Ix -----Iv and the equation for L reduces to
where where
l;:sail moment of inertia about the spin
I_=moment of inertia of the wheel about axis
the spin axis I_-----sail moment of inertia about the solar
vector directed through the sail
Ir--I:--I_:transverse moment of inertia
center of mass
of the wheel.
l_:sail moment of inertia about the per-
pendicular to the solar vector and
During the orbit day, when the sail is directed the spin axis and through the sail
at the sun and 8_----0. The torque due to the center of mass
L,8N-_-_k--oo_
[ o _i,,_ac ab _.... 'P z-- Px-_[(c --b )a,a_-_bc(2a)¢][Ss (I_--Is)Tf3_ (Iys Ii)
_ _ olc-- oQJPsPe_ y) 2 2 s2 s s s2
s2 s s p p
+8s (Iy-- Is) +_m.] } _-k" { [a _- (bacWcas)2]8_Si - (aba_-_aca.) (28')o } ( _-- _)]
Ls_--_ 3_2 { [Ii-- ½(P-+ I_) -- _m,]i[ (c2- b 2)a,a_+ bc(2 at)]+j[a ba_--aca_] ) (2.182)
54
SPACECRAFT DYNAMIC8
1 1 o
----(1) (_)+0=_ for a-----45
The torque about the j"-axis results in a roll about the solar vector and
1 1 _ o
=_ (_)+_ (_)=-_-for a=45
-----1 )÷1(01,
2 (1 = era=90 °
The torque about the spin axis was applied torque due to gravity gradient by driving
to the sail only during the sunlight portion of against the wheel. The action resulted in a
the orbit, which averaged about _ of the total spin-up or spin-down of the wheel (change in
orbital period. The serve controlling move- angular rate). Now
ment about the spacecraft spin axis resisted the
1 1 _ o
=_ (-_)+_ (_=-_-for a=45
5_
ORBITING
SOLAR
OBSERVATORY
The torqueon thewheelis
L 3 / 2_ \3
_=2 _96----_) (6.42) (i" + _/_j") ----1.14 X 10 -5 (i" +_f_j) lb-ft (2.183)
The torque on the sail during the sunlight portion of the orbit is
The torque on the sail during the night portion Then the momentum per volume, p, is given by
of the orbit is
E
1 3033 . • C
cp=E
L=lO-5(1.265V'_-l.64j" _Ol147k '') (2.186)
or since a change in momentum per unit area
The k component of the torque changed the
is a force per unit area, the radiation pressure
(rad_/day for absorption, P,, is
spin rate of the wheel by 5.91X10 -a \s-_//
or a 5 percent change in nominal wheel P,=E
rate in 26 days. Thus the spin rate time
If the incident beam is reflected from the sur-
constant was 312 days.
face, the change in momentum per unit area
From the formula
in one second is
2cp = 2 E
___T_T (2.187)
--./_03
and the radiation pressure for reflection, Pp, is
Suppose a stream of radiaLion in a vacuum Then the pressure due to the absorbed radia-
falls normally on the surface of a body. Accord- tion, p,, is
ing to Einstein, we know that
3.
E
m=-_ (2.188) and the pressure due to the reflected radiation,
m
p. is
where
_p= 2pE= 2(1--,_) E
re=massvolume
E=mean energy/volume and the pressure per unit area becomes
and
p=_+p_-_-(2--a)E (2.189)
c-velocity of light
56
SPACECRAFT DYNAMICS
"For a radiation of mead energy per volume The pressure is the normal component of the
impinging upon the surface with a velocity, v, force per unit area and
the energy per unit area in one second, i.e., the
solar constant, S, is
Pa(O)= -S--cos' 0 (2.192)
S-_-cE
and S=0.14 watt/cm'. By using the above If an incident beam in which the energy
expression for E, the radiation pressure per unit density, E, is specularly reflected from a sur-
area for normal incidence may be written face at the angle of incidence, 0, the force,
A A A
F (e)
=E cos
A
Z
X X __
57
ORBITING SOLAR OBSERVATORY
=lrBldS1
F;.,,=aoF,(O) =0.3 (9 cos _ E lb/ft 2
The component of the total flux in the direction The total force, F_. _, on the wheel due to solar
of the normal is absorptivity becomes
73-=fBldS'c°s2#'( dS'c°s#2
)r'
=2rBldS1 cos _ 81 sin 01d#1 where A_ is the cross-sectional area of the wheel.
fo r 12
The cross-sectional area of the wheel is approxi-
2 /,,/2 mately 4.2 square feet, and F_., becomes
=--_ _BldS_ cos 3 #1Jo
F_. ,= (0.3) (9.7X 10 -s) (4.2)
=_ lrBldS1 =1.22)<10 -7 lb
58
SPACECRAFT DYNAMICS
For the wheel and the case of specular reflection For the OSO I spacecraft made with an alu-
minum shell: a:22 in.=0.559 m; a=3.8><.10'
2
F, ,----(0.7) (2 _ (_ A..) mho/m; g,= 1.25 >(10 -e henry/m
d,,==0.090 in.----2.28X10 -_ m
0.7X2X23 (S) A,
q= _(1.25 X 10 -6) (0.559) (3.8 X 107)
(2.28)< 10 -3) = 0.030 see
0.7X2X20.3 SAc,
3X0.3 c
o.,o=0.5 cp=3.14/sec q_o=0.094
or Z I
T= s
1+ ":a__'
432
or since s=l/o'd
T= T_a%dB2 (2.198)
1+4 bt_a2aad_a)2
59
782-552 0---66------5
ORBITING SOLAR OBSERVATORY
For the coordinate system shown in the figure above, the transformation equations Rre:
f co ,co o
Y"/=/--sin/3
z"J _. sin
cos a sin--,i cos l
cos B
0
sin a sin t_
COS ot
(2.203
and the relationbetween the primed and un- x=r cos a cos _b 1
primed axes may be written y=r sin ¢' (2.208)
[ 0cos
z')
10Sol[:]
(.sin a 0 cos_ .j z
(2.206)
The potential
cos O'/r _= + Voz/r 3 or
z =--r
of the magnetic
sin a cos _'
field, V= + V0
T[o 1
z --sin a 0 cos z'
field,
gradient
it can
of the
be obtained
potential.
by taking the negative
By= bV_by V°z [--3 (x2 +Y_ + Z2) -_/2 "2y]=3 (V--_) yz
B,=----=Oz
i_V Vo { r-a+z [ --3(x_+y2q-z2)-s/z. 2z ]} =Vo {13z2_
\fi----_-]
6o
SPACECRAFT DYNAMICS
(SX 13)- (SX B) = (S. B) (B- B)-- (S. B) (S. B) and because S- S-= 1; S- B----B_', Equation (2.211)
becomes:
-4-2 sin _ cos fl cos otBzBy42 sin a cos a sin _ _B,B,--2 sin a sin _ cos ¢B_B, (2.213)
B,--(Y_)3xz=(Y_)(-3r_sinacosacos"¢h')=--3(-_)sinacosacos'4_"
B,=(-_) 3yz=(-V/) (-3r2 sin a sin _' cos _')=-3 (Y_) sin a sin _" cos ¢ '
(2.214)
B, =tv°'ikr
5] (2z2--za--Y_)=(-_) (2r _sin_a cos _¢'--r _ cos2a cos __'--r _ sin 2 _')
B- B--B'y'_--- 1.4,3 sin 2 a cos __'--9 sin _a sin 2_' cos _¢'4sin _ ¢_sin 2 a(6 cos 2_'
--1.4.9 sin _ _b' cos _ _b'--9 cos 4 _b')46 sin/_ cos/3 sin _a(3 sin _' cos _ _' sin ¢' cos _')]
(2.215)
Equation (2.215) is now used to evaluate Equation (2.212).
1 (_o)2[_
=_---x -- -4-3 _sin _a- (I _sin _t_sin _ )]'- o
or
_{roy
'_--\ r_][_ rl+_8 sin_ (1-- i sin. a) (1_2 sin , ' )]
=_g(_, 8) (2.216)
61
ORBITING SOLAR OBSERVATORY
p =,ra2_rd[ Bl
For (9) 16-inch panels, we calculate the values
a=O. 559 m/a-3.8Xl07 mho/m; d-_2.38
[=16 in.=0.4064 m; [2=--0.165 ms
X10 -a m; [----0.254 m
h----lOin.----0.254
m; h2----0.0645
m2
B1=B_g(a, f3)/h(x)
62
o
SPACECRAFT DYNAMICS
The cylinder of the wheel was the only source ---_XIO -7 watts
of damping torque not of negligible magnitude
when considering a six-month orbital life. The where
only effective way to reduce this damping torque
was to insulate the compartment cover plates V_=volume of ferromagnetic materials
and the rim panels to break up eddy current in the wheel, in cm a
paths. It was not practical to insulate the rim
Since the spacecraft was in the dark portion
panels since they were structural members.
of the orbit a maximum of 35 percent of the
The cover plates were not difficult to insulate
orbital time, the maximum effective value for
from the rest of the structure, so this was done.
A slow-down time constant of between 41 the average power loss in the upper structure,
p_, due to hysteresis was
and 187 days was anticipated considering all
torque effects present. This required a spin-up
gas expenditure about every four days in the p,= (0.35)-_X10 -7 watts
extreme case. (See the discussion of Chapter where
4 on the gas system.) Spin-up cycling at four-
day intervals would have used up the spin gas V_=volume of ferromagnetic materials in
in a little over a year; well in excess of the six- the upper structure, in cm 3
month life supply requirement.
This average power loss value is now com-
Magnetic Damping Due to Hysteresis Losses pared with the power losses due to eddy cur-
Ferromagnetic materials were used in the **R. M. Bozorth, Ferromagnetism, Princeton, N.J.;
OSO I spacecraft. Hysteresis losses will occur D. Van Nostrand Co., Inc., 1951.
63
ORBITING
SOLAR
OBSERVATORY
rents in the rotating wheelof the spacecraft. 688-day time constant is, however, significant
In Equation (2.200), the torque exerted on the compared to the actual eddy current losses
spacecraft due to eddy currents in the cylin- remaining due to the rib plates and inner cylinder
drical shell is given as conducting paths.
T=--k60
Magnetic Coupling Torques
where
k=_raS[ _dB_
Torques Due to Permanent Magnets in Pointed
Instruments.
The instantaneous power loss is related to the
torque as follows: The open-end photomultiplier in the Goddard
pointed experiinent used 8 Alnico permanent
magnets for magnetic focusing. These mag-
P= dt = --d-i
nets were arranged parallel to each other and
or
were mounted with their poles aligned with the
pitch axis of the spacecraft. The torques pro-
Therefore duced by the interaction of their magnetic
moment with the earth's field had either spin
or roll components. The combined pole strength
of these 8 magnets was not known, but theywere
or
866 1732X43 T! z
Tx _-
r_=v_ V_ rB= V_+O.35V_ days
2 F0.35 -_ T,=--T;
Tv=T'y=0
For the OSO I, V_ and V_ are estimated to be
100 cm 3 and 25 cm 3, respectively, and therefore
regardless of the direction of H.
we determine that
Spin decay due to hysteresis and eddy current
1732 damping in these magnets at night when the
rH-_ 1--_- (43)=688 days instrument is spinning should be negligible.
Since a second unit of magnets probably
Therefore, the effect of hysteresis losses in would not exactly match the magnetic moment
ferromagnetic materials would not be significant of those in the photomultiplier, calculations
when compared to the eddy current losses were were made to determine how much residual
the eddy current paths not broken up. The torque could be tolerated.
64
• SPACECRAFT DYNAMICS
tj=-_=5,720 sec
67
ORBITING
SOLAR
OBSERVATORY
z This is shown in Figure 2-20. After 180 a of
rotation the torque due to the earth's field
acting either along or normal to the spin axis
is reversed in direction. In the figure. Ty is
the reaction of the magnet to the earth's field
when the magnetic north pole is in the _y
direction, etc.
From Figure 2-21 note that the torque pro-
duced by a magnet mounted parallel to the
T _ 12_ - / spin axis does not change direction due to spin
and is therefore cumulative. If in Figure 2-21
the sun is always in the _-y direction, the earth's
field vector oscillates back and forth through
the y-z plane in the manner previously de-
scribed, and the major cumulative component of
X_ this torque is a roll torque producing precession
in pitch. As the spacecraft rolls, this torque is
FIGURE2-20.--Torques produced by a magnet mounted 90 ° out of phase with that produced by the
normal to the spin axis.
magnets mounted on the pointed instrument, but
it still requires pitch jet gas to counter its effect.
Z
Therefore, this torque must be added to the
residual torque of the other magnets in satisfying
the value for Tm given above.
55
o
SPACECRAFT DYNAMICS
reasonable number, consider the spin axis to be Equations (2.225) and (2.226) show that the
oriented at an angle of 45 ° with respect to the same axis initial position error is unchanged;
ecliptic plane. In this case, the necessary the same axis initial rate error has an average
pitch correction would be 0.707 degrees per day. value of zero. The cross axis initial rate error
also appears and has an average value equal to
Vehicle Separation F..ffecls the initial rates divided by p.
67
ORBITING SOLAR OBSERVATORY
where _ has been substituted for _ and from Integration of Equation (2.232) yields
the last equation we note that _ is a constant.
Differentiation of the first of Equations (2.227) _.. M_ sinpt M'_
(t) =-_-. --p-- =Td sin pt (2.233)
results in
and
_--p0"=0 (2.228)
where
(1-cospt) (2.234)
Therefore, the application of a torque about
the N-axis causes an average pitch velocity of
Substituting the value for 0 from the second of &=M,/Io_, while the average roll velocity (t_)
is zero.
Equations (2.227) into Equation (2.228) gives
For a torque about the M-axis (pitch axis)
Equations (2.227) become
i_--pO=--Mm/Ir
Taking the Laplace transform, with initial
values zero, gives b+p,_=o
p_ 1 p=constant (2.235)
&(s)---Td M, 8(82__p2 )
(2.229)
These equations may be combined to form the
or
expression
a(t)=7-M_d (1--cos pt) (2.230) ... p2
# q-p_=_ Mm (2.236)
and
68
Chapter 3
The configuration of the Orbiting Solar Figure 3-2 illustrates the second stage of
Observatory as presented in the initial proposal evolution. This layout was completed in early
is shown in Figure 3-1. At that time it con- 1960, prior to the first experimenters' meeting.
mined all of the basic elements of the final design At the time, dimensional layouts of the Delta
except the external arms which carried the spin shroud had been received as had the preliminary
gas supply bottles. vibration input specifications and the steady-
As the various design criteria, i.e., power state acceleration specification. At this point,
requirements, functions of the spacecraft, size the over-all weight was to be 350 pounds and the
and shape of the payload envelope, and load pointed experiments were to be not more than
inputs from the launch vehicle became known, 80 pounds.
the present OSO I configuration gradually Figure 3-2 shows all of the basic elements
evolved. contained in Figure 3-1 with some change.
The maximum diameter of the wheel was known
Function to be 44 inches and the pointed experiments
were to be 4 inches wide by 8 inches high and
The function of the spacecraft structure was
twofold: to carry all experiments, controls, and no longer than 39 inches. The 4-by-8 inch
components safely through the launch phase, dimensions were compatible with the Aerobee
and to serve as a part of the control system pointing control experiment space in the antici-
when placed in orbit. pation that OSO I pointed experiments might
i SOLAR CELLS
! !IIi[!l_
/ Ill \ ..... .
' 1; t !lIllllll
_J [ [ llli' _llll!ii1!ll
..,..I I JM/ _ I :l j / ; |
L_
ORBITING SOLAR OBSERVATORY
(Spin Axls)
be tested in Aerobee shots. The length was
Yaw Axis
determined by laying out the necessary compo-
nents. Starting at the third stage mounting
interface the components were stacked until the
position of the pointed experiments was deter-
mined in relation to the shroud curvature.
At that time it was known that the power
requirements were much greater than originally
thought, and the solar cell array was shown Instruments
f- (Line of Sight
The major changes were the division of the to Sun)
7o
STRUCTURAL DESIGN AND FABRICATION
Burnout Condition:
hid
Steadystate- - _ - __ _ _ - - _ _ _ _ - _ - =+0
-12.1 g
Vibration
Sinusoidal (50-500 cps)---= 1 7 . 1 g
Random _________________=O
Transverse
Vibration
Sinusoidal (50-500 c p s ) - - _ = f l . 4 g
Ptandon _ _ _ _ _ _ _ _ _ _ _=-O_ _ _ _ _
It was necessary to estimate magnification
and attenuation factors to arrive at. f k a l load
factors on the spacecraft components. By
using the conservative assumption that the
lowest primary structural resonant frequency
FIGURE
3-4.-Basic wheel assembly.
is at 100 cps and that the magnification factor
with the third-stage interface mounting surface. is 2.5, final loads on the spacecraft. components
The third-stage rocket and the spacecraft were and structure became :
clamped with a Tee-band. The wheel height Maximum qa Condition:
was established by the longitudinal dimension An'al
required of slip rings, bearings, and precession Steady-state-_- _ _ - - - - - - - - - - - - _ = 0 +
gas bottle. The cover panels were used pri- -2.8 g
marily for thermal shielding. It was necessary Vibration
that each corer be electrically insulated to _
Sinusoidal- - - _ _ _ _ _ _ _ _ _ _ _ = f17.8 g
Random _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ = f 5.1 g
reduce the eddy current losses.
Loads and Siresses
The critical flight conditions for structural
design occurred a t the time dynamic pressure
mdtiplied hy hooster angle of attack was a
maximum, and at the time of Thor zmk-engine
burnout. These conditions are referred to as Total_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ = f 8 . 6 g
maximum pa. and burnout, respectively. iMaxi-
mum transverse loadings occurred a t the Burnout Condition:
AXifZl
masimum pa condition and maximum axial
loads at the burnout condition. Limit loading _
Steady-stat,e- - - _ _ _ _ _ _ _ - - _ _ _ _ =+0
-12.1 g
inputs a t the Delta vehicle-spacecraft interface Vibration
for these conditions are as follows: _
Sinusoidal- _ - _ _ _ _ - _ _ - _ _ _ = -+ 17.8 g
Maximum ga Condition: Random-- _ - - - - - _ _ - _ - - - - - = 0
Azial __ +
Total _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ = 17.8 g
Steadv-state- _-
-__ -____ _ _ _ _ _ _ =+ 0 -29.9 g
-2.8 g Transverse
Vibration Vi bration
Sinusoidal (50-500 cps) *- .- = i7.1 g
Random_- ___
- _ _ _ _ - _ _ _ _ _ _ = 0.047 g*/cpS Sinusoidsl- - - -..__ _
- - - - - - = f3.5 g
Transverse Random_______--__-_----=O
~-
Vibration
Sinuscidal (5&500 cps) *_-- = & 1.4 g Total_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ = +3.5 g
Random _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ =0.047 g P / c p s
For design purposes, ultimate loadin,,UQ were
*It is assumed that the eparecraft is vibration critical!
in t h e 50-500 cps range. assumed to be 1.5 times the above values.
71
ORBITING
SOLAR
OBSERVATORY
43.50 IN. pages, the following equations were derived
for the spacecraft-launch vehicle interface
loads. Results of the equations appear in the
table following.
Y-_,F,=O
M,,W,+n,,_W,,--PP=O (3.1)
5-I F,=O
nz,Ws+n,,,W,o-V--=O (3.3)
V=n,,W,+n,,_W,, (3.4)
__,M----0
20n,,Ws-F7.9n,,,W,o+M=O (3.5)
P
43.00 IN.
M= --20n,,W,--7.9n_,_W,_ (3.6)
FmusE 8-&--Geometry and loading on O80 I.
Item Units Condition
Spacecraft Geometry
Figure 3-5 is a sketch of the spacecraft Max. qa
I Burnout
showing the general configuration and loading
315 315
while in the boost phase. The symbols used Ww ib .........
72
STRUCTURAL DESIGN AND FABRICATION
P
: 15.81
0.064
"---- 10.56
U I t
SECTION A- A
HOBIEAcS
TALR__ZON 1.8 IN. 2117-T3 RIVETS
(SINGLE SHEAR)
_-_.M=0
Bes__ng Allow./Rivet=810X 1.05=850 ]b
-- 10.56P+g.5R,=0 _It.
R 10.56 850
M.8.----_--1=_.47
i
Wheel Center Casting Assembly
Figure 3-9 is a free-body diagram of the
center casting assembly showing the expected
loadings. This casting had to react all loadings
from the oriented portion of the spacecraft and
from the wheel ribs and deck assembly. In
arriving at final loads, the weights and load
factors shown on page 3-12 were used in con-
junction with the ultimate load factor of 1.5.
Ps, Ro,, and R h are the loads produced on the
- T5.25 D
FIGURE 3-6.--Typical deck.
UPPER CASTING--\ IPSRus
_K/ 29/32
_r..
--, .ooo
DECK
ATTACH
35_:: 09 0°5T.,:
10
BASECAST,.O-- Vlt
_£W L ;:.3_ __ _.5_,.
R,----2520--1870=650 lb ult.
P,: M,W,= 25.7 (145) (1.5) = 5590 lb ult.
0.43n_ W_---- (0.43) (8.6) (315) (1.5)---- 1750 lb ult.
FIGURE 3-8._Loading on rim panel during arm
74
STRUCTURAL DESIGN AND FABRICATION
75
782-552 0--66--------6
ORBITING SOLAR OBSERVATORY
tests, however minor changes in the locations experiment field of view would not be occluded
of wiring holes were required. by the extended arms nor would the experi-
ment physically strike the wheel structure.
UPPER STRUCTURE A necessary design requirement for the upper
structure was that all of the moving parts be
General
locked during the launch phase. Both the
The upper structure of the spacecraft con- elevation and azimuth motions of the upper
sisted of: (1) the azimuth casting which was structure were locked with squib-actuated pin
mounted on the azimut'h shaft; ( 3 ) the elevation pullers.
casting, which carried the pointed experiments An open framework was used for the solar celI
and was mounted on bearing supported shafts support structure. This was made of 0.025-
within the azimuth casting; (3) the sail structure inch thick aluminum sheet stampings riveted
which supported the solar cell array; (4) the together, and formed a light rigid unit readily
azimuth shaft; ( 5 ) the elevation motor and adaptable to mounting components as well as
bearings; (6) the nutation damper. solar cells. The sail structure form was deter-
The upper structure contained the motorized mined by the cross-sectional area available in
servo elevation drive system and the precession the vehicle shroud.
gas control system, including the precession gas Since it was necessary to locate the precession
jets. It also supported the upper structure jet nozzles as far from the spacecraft center of
electronic packages and controls. Figure 3-1 1 gravity as possible to develop the greatest
is a photo of the sail structure. torque, t,hey were located at the periphery of
Because the moment of inertia of the upper the sail within the open framework.
structure added to the spacecraft transverse The pointed experiment mounting method
moment of inertia and contributed nothing to and the mounting of the solar detector eye
the spin moment of inertia, it was necessary to blocks on the pointed experiments were prob-
keep this moment of inertia as small as possible lems which had to be solved early in the design
relative to the spacecraft center of gravity. phase in order to provide design criteria and
This meant that all large masses were to be specifications for the experimenter groups.
mounted as near to the bottom of the upper The sail structure design had to await correla-
structure RS possible. However, the servo tion of the total spacecraft power requirements
elevation control of the pointed experiments and the solrtr cell output capability before it
had to permit a t least * 5 degrees of movement could be completed.
of the experiments about the elevation axis. The azimuth casting supported several elec-
The elevation axis, or pivot point, had to be tronic components and control system sub-
chosen so that in the depressed position the assemblies. Since the design of the casting was
predicated upon final configuration of these
components and subassemblies, the final con-
figuration of the casting was the last piece of
structural hardware design completed.
The azimuth casting bolted to the azimuth
shaft. The shaft was supported by bearings in
the wheel assembly as illustrated in Figure 3-12.
Axial motion of the shaft was restrained a t the
upper bearing, and the lower bearing controlled
radial play only. The shaft was surrounded by
a precession gas storage b o t h . The bottle was
threaded on the shaft at the upper end and was
free to slide on O-ring seals for normal strain
deformation a t the lower end. A hole was
FIGURE
3-ll.-Basic sail structure. provided in the shaft to allow flow of the gas U P
76
STRUCTURAL
DESIGN
ANDFABRICATION
5590 LB
1870 LB
_--AZ CASTING ,
BEARING _
5590 LB
R
!1 TOR OTT
OE E
/ v
JI WHEEL
650 LB
"O" RING SEAL-- _ S ]
77
ORBITING
SOLAR
OBSERVATORY
have been to initiate a major redesign _nd
locate the motor at the top of the shaft. This
was done on the following OSO program. It
was too late in the S-16 program to allow time
for this redesign.
Azimuth Casting
-i - The azimuth casting (see Figure 3-14) is
critical for axial and transverse loadings result-
ing from the maximum qa condition. The
S casting is loaded by the sail assembly and
equipment trunnion-mounted to the casting.
I Assumed loadings are as shown in Figure 3-15.
5M
i
SEC C-C
_U =0 =f M_ d_ (3.8)
5X1 El,
SEC A-A
(3.9)
FIGURE 3-14.--Azimuth casting sections. bX2 J EI_
q2 9.5 _1 q2
ql
/ B 5Xs 3 EA
(3.10)
I1
Center of Gravity of Sail Structure Without
P2
Azimuth Shaft:
I1
I2 L V
I1 -1
z = 16 IN.
c.g. of azimuth
craft separation
shaft
plane is:
Z,h_t = 11 in. approx.
c.g. of complete
Z=20 in.
from
upper structure
Delta-space-
is:
78
STRUCTURAL
DESIGN
ANDFABRICATION
" Weight of elevation casting and pointed Final ultimate bending moments (in.4b) are
experiments---- 75 lb shown in the diagram below.
-297 -- -- -762
(50) (25.7) t (1.5) = 101 lb-in.
q2-- 2(9.5) A A
X_M0=0
L J
wq-qi=37.2 (3.11)
-2323
8377
/C.\ 7953 16330
\ 4175
79
ORBITING SOLAR OBSERVATORY
+- -+-I-
80
STRUCTURAL
DESIGN
ANDFABRICATION
81
ORBITING SOLAR OBSERVATORY
COMPONENT MOUNTING
FIGURE 3-20.--Nutation damper assembly.
82
STRUCTURAL DESIGN AND FABRICATION
P
FIGCRE3-23.-Tape
I
53
ORBITING SOLAR OBSERVATORY
84
"" STRUCTURAL DESIGN AND FABRICATION
The maximum loads on the arms that oc- In the differential equation above the terms
curred during arm extension will be discussed Mc and Md are both nonlinear in 0 and _,
first. Later, the loads due to vibration inputs which are functions of time. This equation
and thrust will be reviewed. does not lend itself to closed-form analytical
The typical arm and arm damper assembly solution.
in Figure 3-25 shows the geometry of an arm To control the time duration of extension, an
in the stowed position. The forces acting on orifice had to be provided between each damper
the arm at the instant of release are: cylinder, and the size of the orifice determined.
For a piston of specified size and stroke, the
Centrifugal force Fc=Mrw 2 (3.13) volume of fluid to be moved is known. The
8_
o
ORBITING SOLAR OBSERVATORY
Since the dampers form a closed hydraulic heavier at the bolts, and this was where the
system, thermal cycling could cause internal low flexure stress occurred. The remainder of
pressure changes which could limit the design the arm was very nearly a constant stress
values. The differential expansion between the section.
fluid and the container system was calculated The safety margin for the arm flexure stress
and converted to pressure by using the com- was raised for several reasons. First, the cal-
pressibility factor for alcohol. Assuming that culated loads were based on nominal torques at
the system was filled at low temperature and the nominal spin rates. The torque increase with
maximum rise was 35°C, the pressure was wheel spin velocity was not linear. A 10 per-
found to be 5000 psi. cent velocity increase raised torque by 21 per-
In establishing the design load for the hy- cent. This reduced the margin of safety to
draulic system, there was one other factor 1.47. Second, the two arm sections were
investigated. As mentioned, the system was to machined from plate stock, and a fairly loose
provide a mechanical tie between the arms in dimensional tolerance was desirable.
case one arm hung up. If one arm did hang up, The lug that was bolted to the arm web and
there would be no fluid flow through the orifices, tied to the piston had a margin of safety of 2.1
and under static conditions the pressure would in shear and 1.6 in tension. The piston clevis
become twice the potential maximum seen by pin which mated with the lug was 4130 cold-
any one cylinder, because the arm which was drawn steel having a margin of safety of 2.5 in
hung up would exert no pressure on the system. shear and 4.7 in tension. Other detail parts
Since the maximum pressure per cylinder was were stressed similarly.
calculated to be 1750 psi, this design criterion is
Vibration Stress
less than that for thermal expansion. Thus,
5000 psi was established as the design value for For vibration, the arm was again treated as a
the hydraulic damper system. An arm damper simple beam and the transverse input of 11.8 g
cylinder may be seen in Figure 3-24. random vibration used for load criteria (Figure
3-27). Checking various points along the arm,
Arm Stress During Extension
open, and
Maximum
the arm was extended
moment
was normally
moduli were computed
occurred
approximately
143 ft-lb.
for various
when
30 degrees
points
Section
along
<> ..........
X
the arm and stresses were calculated to range FIGURE 3-27.--Equivalent beam load.
from 12,000 psi to 5660 psi. Based on a yield
strength for 2024-T4 aluminum, this design the maximum flexure stress was found to be
provided a margin of safety of 2.0 to 7.1. The 10,000 psi, resulting in a margin of safety of 3.6.
arm was fabricated in two sections and bolted Using the endurance limit, the margin was re-
.together (see Figure 3-26). The section was duced to 1.8, based on 5X 10 s cycles. However,
this level of input is reached only during ran-
dora transverse testing for four minutes in each
axis, and not enough cycle reversals occur to
justify this rating.
The vibrational loads are directly transferred
H to tile pin connections on tile linkage. The
stresses were calculated and the minimum mar-
FmURE 3-26.--Typical arm construction. gin of safety involved was found to be 4.3.
86
"" STRUCTURAL DESIGN AND FABRICATION
O
age occurred primarily in the heat treating
operation.
At this point, the design was changed to
an arm fully machined from plate stock. In
redesigning, the same section modulus and
weight were maintained for a different geome-
try. Fabrication proceeded without further
FIGURE 3-28.--Spin acceleration loading.
trouble, and the arms produced were uniform
Spin Acceleration Stresses and within tolerance specification. Further
tests were made to fully qualify the latter
During spin acceleration of the third stage
design.
booster, the arm hinges were loaded in shear
No difficulty arose in the fabrication of other
as shown in Figlrre 3-28. The spin accelera-
parts of the system except for the insulated
tion was specified by Douglas Aircraft Company
tie member between the linkage and the wheel.
to be radians per second per second. The
First, an insulated bushing system was tried,
margin of safety for the hinge pins in shear
but this allowed too much electrical capacitance
was calculated to be 5.0.
between the charged linkage and the grounded
Hydraulic System Stresses wheel. Then a fitting in the form of a tee
was made from laminated fiberglass board.
The tie bolts were 303 stainless, the cylinder
This, however, was not strong enough in shear
was 304 stainless, the end caps were 2025-T4
across the laminations.
aluminum, and the piston was 416 stainless
steel. Finally, a design was incorporated using
fiberglass epoxy laminations where the lamina-
Based on a maximum load of 5000 psi
tions were not in shear (Figure 3-29). A
internal pressure, and the yield strength of
heated, closed metal mold was used to produce
the materials involved, the detail parts had
the section desired. The electrical and me-
the following margins of safety:
chanical performance was satisfactory.
Tie bolts ......... 1.28 (tension)
Cylinder ......... 1.07 (burst) WHEEL
Aluminum ends___ 12.2 (shear about O-rings)
Fabrication
S7
,o
Chapter 4
CONTROL SYSTEMS
INTRODUCTION axis about the pitch axis. The pitch gas
control system maintained the spin axis perpen-
The basic approach to the problem, that of dicular to the solar vector within about three
providing a stabilized solar-oriented space degrees. No control about the roll axis (sun
platform from which scientific instruments vector) was provided, but rates about this
can continuously monitor the sun, was to axis were slow due to the rigidity of the gyro.
utilize the inherent gyroscopic stability of a Fine elevation positioning of the pointed
rotating body and to orient a separate structure instruments was accomplished by an electrical
which could be shaft-mounted to the spinning servo called the elevation servo. The elevation
body. The separate structure could then be servo motor was mounted on the casting which
oriented to the sun by means of a gimballed supported the pointed instruments.
axis arrangement. The control problems then Prior to third-stage engine firing, the entire
were: third stage and spacecraft with arms in the
1. Control the rate of spin of the rotating stowed condition were spun up to approxi-
body and the orientation of the spin axis of mately 120 rpm by a system of small rocket
the rotating body with respect to the sun. motors. This spin-up started the operation
2. Control the gimballed axis orientation of the spacecraft sequencing timer at zero
of the oriented structure with respect to the time.
sun. At approximately 100 seconds (sequencer
Problem (1) was solved with gas control time), and after third-stage engine burnout,
systems for spin-rate maintenance and for the arms were released and, due to centripetal
spin-axis orientation. Problem (2) was solved force, raised and locked in the extended posi-
with electrical servo-mechanisms, one for each tion. Shortly after this time, the spacecraft
of the two gimballed axes, deriving orientation was separated from the third stage.
information directly from the sun. The fol- At 200 seconds the spin rate of the spacecraft,
lowing paragraphs describe in a generM way with the wheel and upper section mechanically
how the control systems performed the orienta- latched together, was reduced to approximately
tion task. The remainder of this section 30 rpm by the action of the despin jets, located
describes in detail the control system parame- at the ends of the arms. After completion of
ters, design, and mechanization. this despin, the despin system was permanently
The OSO I spacecraft utilized the entire d eenergized.
vehicle as a controlled platform. Coarse Turn-on eyes, a set of six eyes mounted on
elevation or pitch control of the platform was six of the nine sides of the wheel, were used to
accomplished by the ON/OFF pitch gas control actuate electrical equipment each time the
system. By exhausting nitrogen gas through spacecraft emerged from the earth's shadow.
nozzles mounted on the stabilized upper section, The same eyes were used to turn the electrical
torques of either sense were produced about servos and the pointed experiments off during
the roll axis. These torques, which were the satellite night.
applied when the spacecraft upper structure This spin rate was maintained within ±Sa/v
was oriented, precessed the spacecraft spin of nominal value by spin-up jets located at the
89
ORBITING SOLAR OBSERVATORY "°
ends of the extended arms. The jets were The pitch control eyes were detection usits
actuated by a signal from the electronic spin for the pitch control system. This block of four
control system which computed the instantaneous eyes was mounted on the side of the stabilized
period of rotation of the wheel relative to the upper structure facing the sun. An 8-second
sun. The primary sensors for this system were time delay was incorporated into the pitch
two spin eyes connected in parallel and mounted control circuitry to insure that the upper section
on the rotating wheel. The spin eye output was pointed at the sun before using pitch gas.
was a narrow pulse generated as the eye swept For these conditions, and with the spacecraft
past the sun once per wheel rotation. spin axis exceeding 3 degrees from the normal to
At 400 seconds (sequencer time), the azimuth the solar vector, one of these eyes turned on the
release squibs and the mutation damper release appropriate jet to precess the spacecraft back
squibs were fired and power was applied to the toward the correct position. Two of the eyes
azimuth servo. were needed for this function, one of either
The azimuth servo actuating the azimuth sense. The other two eyes turned off the iets
DC torque motor drove against the wheel to whenever the spacecraft attitude had been
despin the upper section and point it at the overcorrected one degree in the opposite direc-
sun. The azimuth servo position error signals tion.
originated from two types of detectors. These An unstable condition could have resulted
were coarse eye detectors and fine eye detectors. without a time delay in the pitch control system.
There were four coarse eye detectors, each This could have occurred for a situation in which
having a 90-degree field of view, giving a full the upper structure was pointed at the sun and
360 degrees of position control about the spin the spacecraft had a large nutation or wobble
axis. They were mounted to the solar array angle, such that the eyes would call for pitch
support structure and had a stable null perpen- correction during only part of the wobble period.
dicular to this structure and an unstable null The gas was called for when the spacecraft spin
in the antisolar direction. Their field of view axis was at its maximum error relative to the
was limited to 15 degrees above and below a solar vector. If no gas was used and if the
plane perpendicular to the spin axis. This nutation damper should have failed to work, the
limiting of the coarse eye field of view eliminated maximum error would have remained the same.
the periodic signals caused by sunlight reflecting With the nutation damper working properly,
into the coarse eyes from the three rotating spin the wobble or nutation amplitude was reduced
gas bottles and reduced any reaction that could to zero and no problem existed. If pitch gas
be caused by the earth's albedo. This also was expended only near the portion of maximum
means that the spacecraft could not acquire the error, the amplitude of any wobble or nutation
sun if its spin axis exceeded 15 degrees from a was increased for each impulse of pitch gas and
plane perpendicular to the solar vector. the spacecraft spin axis nutated in an ever-
The composite coarse eye amplitude curve widening cone. The time delay, as noted
was such that its first derivative provided a above, prevented pitch gas being expended un-
favorable despin signal for approximately 80 less the upper section was oriented, and further-
more prevented an unstable condition if the
percent of an upper structure rotation. As the
nutation damper should have failed.
upper section spin rate was reduced to zero,
At the time that the coarse azimuth servo
the coarse eyes point the upper section to within
had pointed the upper section to within 2 to 3
2 to 3 degrees of the solar vector. This acquisi-
degrees of the solar vector, a disabling eye
tion of the sun had to be accomplished every mounted on the pointed instruments actuated a
spacecraft dawn. relay that turned the coarse control off, leaving
Coarse pointing of the upper section per- the azimuth positioning servo with fine eye
mitted the operation of two additional systems: control only.
the azimuth fine servo control system and the The disabling eye had a field of view of
pitch control system. 4-6 to 4- 8 degrees in both the azimuth and ele-
9O
•* CX3NTROL SYSTEMS
vati_n directions. A 1- to 2-second time delay moved through large amplitudes for small nu-
was used in conjunction with the disabling cir- tations. The damper was very effective for
cuit to prevent transients in the coarse servo. nutations small enough that the bob did not
These transients would have resulted if the hit the walls of the container (less than
servo were disabled for a short time each time degree). For these small amplitudes the damp-
the upper structure faced the sun during ing time constant was about 10 seconds. At
acquisition. greater amplitudes the damping rate appears
The coarse servo was disabled during the to have been a constant of approximately one
fine pointing mode in order to eliminate the degree per minute.
errors caused by the coarse eyes seeing the earth Another cause of nutation during the regular
and those errors caused by differences in null orbiting condition was the unbalance of the
position between the coarse and fine eyes. pointed instruments due to the movement of
There were two azimuth fine eye detectors internal photodetector scanning mechanisms.
with a field of view of 4-10 degrees. The eye The detectors were placed at some specific point
amplitude curve away from the null position had for ground checkout balance of the spacecraft.
been shaped to minimize hunting during acqui- However, during orbit operation, the detectors
sition when large initial position errors and moved from the balance position. No nutation
velocities were encountered. occurred during the time the instrument was
The azimuth fine servo positioned the pointed pointed at the sun, but as soon as the pointing
instruments to within one arc minute of the control turned off at night and the upper struc-
center of the sun. Long-term accuracy was ture began to spin, a dynamic unbalance
better than 2 minutes of arc. occurred that caused a wobble. When the
After the nutation damper was unlocked at servo controls pointed the experiments at the
400 seconds (spacecraft sequencer time) and sun the next spacecraft day, the wobble was
the upper structure was oriented at the sun, then converted to a nutation which had to be
the nutation damper damped out nutation due damped out.
to the wobble caused by the burned-out third At 800 seconds (spacecraft sequencer time)
stage, the unsymmetrical thrust of the separa- the elevation gimbal was released and the ele-
tion spring, and the unsymmetrical motion of vation servo pointed the instrument in elevation.
the arms as they were released and swung out. The fine eyes for the elevation servo were
A nutation of several degrees, which could mounted to the pointed instruments and were
have been caused by these effects, would have the same type as the azimuth fine eyes. There
caused high power usage by the servo system were no elevation coarse eyes since the pre-
and therefore had to be quickly removed. cession control system provided the coarse
The nutation damper was provided to remove elevation control.
these undesirable motions. The damperworked The Orbiting Solar Observatory is shown in
on the principle that if energy is removed from Figures 4-1 and 4-2. Figures 4-3 through
a freely rotating body, the body tends toward 4-8 show the launch sequence events and sche-
a rotation about the axis of maximum moment matic representations of the various subsys-
of inertia. This axis was made to coincide tems as they were integrated into the spacecraft
with the azimuth shaft axis by careful balance
before launch. ELECTRICAL CONTROL SYSTEMS
91
782-552 0--66--7
ORBITING SOLAR OBSERVATORY -.
1~
Yaw Axis
3 the azimuth shaft since the upper section spun
=le; Pitch Control up a t night. This meant that the azimuth
\ Solar Cell Array servo had to be able to acquire the sun from a
spinning start. The coarse mode servo was
required to give proper error signals from
all possible sun angles about the azimuth axis
and to bring the upper section into the oper-
Pitch ,\ ,, I ointed Instruments ating region of the fine detectors.
The azimuth coarse servo was designed to
acquire the sun from a spinning start with the
spacecraft spin axis *15 degrees from the
J’ normal to the solar vector. This value was
?
Spin-up ‘
L &-spin
Nozzle Nozzle
selected because of the 10-degree tolerance
i C a s Contalner lut Spin Rate Control
iissigned to spin-axis orientation a t orbit
FIGURE4-2.--0rbiting Solar Observatory systems. injection.
92
CONTROL SYSTEMS
os0 1
FLIGHT
The coarse servo eyes, therefore, had to during the first acquisition imriiediat,ely follow-
have a 360-degree field of view about the spin ing third-stage separation, it was still serious.
axis and a & 15-degree field of view perpen- If the earth were acquired, the pitch control
dicular to the spin axis. T h e field of view of system would precess the spacecraft spin nsis to
the unshielded coarse eye was &90 degrees in a position perpendiculw t o the earth's local
the latter direction. Shields were installed vertical. The spacecreft spin nxis would be
to permit an unobstructed view of the stin precessed so far out of perpendicularity to the
at an angle of -+15 degrees and to occult the solar vector that acquidion of the sun a t the
sun at angles greater than *25 degrees. followinz dawn would be highly unlikely.
This arrangement permits the coarse eyes to The itxiniuth fine eyes httd n field of view
see the siin throughout the design range and liniited to i10 degrees. Beyond this range,
limits their view of the earth to the maximum the inputs to the azimuth systerii consisted
possible extent. entirely o f the signal5 from the coarse ej-e\.
T h e object of this, of course, mas to niake the At the ?rt;irt of :icquisition both the fine iind
sun the brightest object to the coarse eyes so coitrbe wrvos were in operiition. The coarse
that the sun would be ttcquired tind not, the servo despun the upper structure iind oriented
eitrth. Even thougli this problem arose only it :it the sun to within * 3 degrees (within the
93
0 .
ORBITING SOLAR OBSERVATORY
CONTROL I
SYSTEMS
FIGURE
4-4.-OSO I control systems.
fine eye field of view). If the upper structure The mechanical constants of primary im-
maintained this orientation for a period of 3 portance in the servo design were friction,
seconds, the input signal from the coarse moment of inertia, and the motor damping.
eyes was disconnected. This action was nccom- The moment of inertia of the upper structure
plished b y a time delay circuit activated by a about the spin axis ol' the spticecrnft included
target eye having a 5 6 - to °ree fieId of the moment of inertia of the experiment pack-
view in azimuth. The fine azimuth servo then age which could rotnte about a pitch axis
pointed the experiment pnckage to within f 10 perpendiculnr to the spin axis. For a rotation
degrees of the sun. o f y nhout this pitch axis, the contribution of the
Disconnection of the coarse eyes eliininttted experiment package to the moment ol inertin
pointing errors which might have resulted from about the spin axis was reduced by the factor
coarse eye unbalnnce due to stray light from the cos2 y. Since the pitch rotation was limited to
earth or reflections from the spncecrnft. i - 5 degrees nnd the square of the cosine of 5
The elevfition servo was required to work over degrees is 0.993, the moment of inertia about the
IL mnximuni range of k 5 degrees becnuse of the spin nxis wns constnnt to within 1 percent.
active control of the spin axis orientation. The The friction present W(LS the coulomb friction
fine elevation detectors hnve t~ field of view of due t o bearings, slip ring, and motor brushes.
f 10 degrees which eliminnted the requirement This frict ion was relatively consttint in nitLgni-
for tin elevation coarse control mode. t ude, mid its sign opposite t o that of the angular
94
*a
CONTROL SYSTEMS
FIGURE
4-5.-Gas control systems.
velocity. T h e value of the coulomb friction for operating with a full output despin torque and a
t h e azimuth system was approsimat el? 0.05 d u t y factor of 0.6 was
lb-ft .
A torque motor was selected as the actuator
because it could be connected directly to the
azimuth shaft, thus eliminating the need for
3 Ib-ft-sec2 rad
gearing with the attended problems of backlash
and increased friction. The torque motor also
[; (0.54XO.6) lb-ft] [. G]
had a high efficiency. ~ 3 sec
0
A 1.8-lb-ft (364-oz-in.) motor. as selected.
T h e mot or, wit 11 the spacecraft volt age available, In writing the overall transfer function, it is
produced a maximum of about 0.54 Ib-ft of assuined tliat the eyes and the amplifiers were
torque. It provided a dynamic range (i.e.. the not frequently sensitive in the frequency range
ratio of full output torque t o friction torque) of of interest. The frequency range of interest
10 and wis 21 convenient size meclianicall;\-. TWS below 100 cps, and the frequency response
T h e time t o acquire the sun ivit h this motor of the solar-cell eyes and amplifiers WHS flat
ORBITING SOLAR OBSERVATORY
JUUCI4OU BOX
WHEEL
ELECTRONIC SYSTEMS
FIQURE
4-6.-Wheel electronics systems.
where
I,v=rnonient of inertia of the body about
the axis of revolution
40=t~ngulttrposition of t h e hodg
K d = motor dttniping const itnt
In itri effort to control the rnotion of the body,
tt control torque, T,(t), and gain, K , are
introdurd, where c(s) is the trttiisfer fuiiction of the w ~ i i -
where pensat ing networks needed t o stabilize the
T,(t)=
L
I Pf
0-t
Ei(t--r)M’(T)dT (4.2)
system and 4in(s)is the Ltiplttce t,ritnsforiii of
the input t o t h e cornpenstit irig network.
I *
CONTROL SYSTEMS
#NyIFR L 3 w
0 SPACE F L I W
FIGUREPi.-Solar-oriented esperiments.
97
8 .
ORBITING SOLAR OBSERVATORY
UNIV. OF MINNESOTA
/W€rn/W P A U # # R ~ / i W m w
OF mMfla RAYS fo.1-s MEyj
Wt: 3/S/&s. mW€U 0.6 W
UNIV, OF
CALI FOR NIA I
WHEEL EXPERIMENTS
FIGURE
4-8.-Wheel esperiments.
(4.10)
•° CONTROL SYSTEMS
+(,,,+1)-] T,,+I
-;- -I s(r2s--I-1)
LR _(o)=(1+_) -_K ¢(0) (4.14)
F 1
TF I
I
I
I ¢1
IL
KA'r' 2 s+ ] _ (_
1
1
t
s_e
In the block diagram the following notation The open-loop transfer function for the
is used: block diagram can be written as
KR=eye gain (ga/rad)
K(rls+I)
KA=DC amplifier gain (volts/ga)
(4.15)
Kv----power amplifier and torque motor G(S)--Kals('r2s+ 1)(£+ 1)
gain (lb-ft/volt)
K= KFKAKv= 250 lb.ft-rad The lead-lag equalization network readily
TF=friction torque=0.05 lb-ft stabilizes this system; however, it was found
Ke = motor damping = 0.038 lb-ft-sec in development of the control system that
/----moment of inertia of upper section structural resonances significantly altered the
about the spin axis control open-loop transfer characteristics. The
Io_= 24 lb-ft-sec _ dominant resonance at 29.5 cps had ampli-
r_=lead time constant (sec) fication of approximately 40 db, which was
r2=lag time constant (sec) sufficiently high to cause instability when the
99
ORBITING
SOLAR
OBSERVATORY '°
K(0.225s-_ 1)
(4.16)
Gc(s)=(O.O115s+1)[1.35(10-7)s3 + 6(10-5)s2-_ 2.18(10-2)s-t-1]
The Bode plot of this transfer function and the troller filter succeeded in attenuating suffi-
measured frequency response of the flight ciently the resonant peaks.
hardware are shown in Figure 4-9. The azimuth coarse mode used the same
The Bode plots of the total loop transfer roll-off network that was used for fine. The
function, including the controller character- lead-lag equalization and the system gain
istics and the structural resonances, are shown were different, however. The lead-lag net-
in Figure 4-10. To obtain this plot, a linear work can be represented as
power amplifier was used instead of the pulsed
G 3.2s+1
power amplifier.
The gain margin of the linear system is
shown to be 7 db. The gain margin of the
The Bode plot of the actual flight hardware
actual system correlated closely with the
equalization, including the roll-off network, is
linearized model. The fast roll-off of the con-
shown in Figure 4-11.
The torque gain for the coarse mode was
(0 225s * l)
3O
40
20
30
_ ,o
_o
-20
-lO ........ i
01 1
Jl
FREQUENCY (CPS}
-20
0.1 1 I0 I00
FIGURE 4-9.--Azimuth fine mode equalization. FREQUENCY (CPS)
_AM PLITUDE
Elevation Servo
_,.^s_:
A_GL_: \ _'44: -200 ip_a=K rlp_l
(e--a)+Kdp(e--a)W T_ (4.17)
-320
10o
"* CONTROL SYSTEMS
[I_y(r_s+ 1) + Kas (r2s + 1) +g(rls+ 1)][e (s) -- a(s) ]=-- _-_ (r_s+ 1) + [I_3s(r2s+ 1)
At each satellite dawn, d(0)=&(0)=$(0) axes, the azimuth axis and the elevation axis.
=&(0)=0, and the servo error is given by The two servo systems were independent of one
another, but were nearly identical. The azi-
T_ muth servo system was slightly more compli-
cated because of the large acquisition angles
required. It operated about the azimuth of
The servo error for a Tp of 0.05 lb-ft and a gain,
spin axis of the spacecraft and was used to
K, of 150 lb-ft/radian is approximately 1
minute of arc and, unlike the azimuth servo, hold the oriented section fixed in space by
cannot be biased out. driving against the rotating wheel. The eleva-
tion servo rotated the instruments a few
The elevation servo block diagram is identi-
degrees in a plane containing the solar vector
cal to that of the azimuth servo except that
and the spin axis.
the motor damping input from the wheel does
not exist. A block diagram of the aximuth servo system
The parameters used on the elevation servo is shown in Figure 4-13. Error signals pro-
are as follows:
K= 150 lb-ft/radian
COARSE I
Ka=0.015 lb-ft.sec
I, 3 = 1.3 lb-ft-sec _
r_=lead time constant-_0.22 sec
TO _ I
T2=lag time constant=0.016 sec
-- I t
101
ORBITING SOLAR OBSERVATORY
102
.. CONTROL SYSTEMS
10”
1‘09
I
i i
1
,
L E F T FRO?*T
COARSEEYE
TO SUN
A RIGHT FRONT
COARSE E Y E
\ I
\\ 1 50
25 EYE
‘ci
ANGLE
t
i
575 1 J
75i
100
t
- - - - - - -INDWIDL
INDWIDLA.AL
L EYE
\\
COMPOSITE CL‘RVE
FIGVRE
4-19-AqwIar characteristics of the coarse
rye.
103
_Q
ORBITING
SOLAR
OBSERVATORY
NO BACK APERTURE
WITH I BACK I
APERTURE I ! _,__
0o 2° 4' 6° 8° 10 °
(A) SINGLE EYE DEGREES
t. - q
9° 6° 3o 3° 6° 9_
ANGLE FROM SOLAR VECTOR
DEGREES
was provided with the switch so that switching
would not continually take place during the (B) TOED-IN EYE PAIR
initial acquisition.
FIGURE 4-21.--Fine eye response.
Fine Eyes--The fine eyes for both elevation
and azimuth were mounted directly on the
pointed instruments. Fine-eye detectors pro- 2 5
104
° ° CONTROL SYSTEMS
105
ORBITING SOLAR OBSERVATORY
R
$
PROPORTIONAL TO R L
106
,o
CONTROL SYSTEMS
1
Z = Z OF DIODE
T_- Ie
= Z OF ZENER
/
Zd rb
\
I
I
f
:E I /
I r
I
I e
-
I \
I I e V
1 t
I
I
¢ <
< I
I
Zz 1 4
I
I J-- -- --
I
\
The input impedance seen by the eye is Adding Equations (4.22) and (4.23) gives:
Z= E
I (4.20) O=--I(R_+ Z+)+ I_ _(R,+r,+ I_) (rb+Z_) ]
(4.24)
To find I, the loop equations are solved. First,
we have the relation The matrix of Equat, ion (4.21) plus Equation
(4.24)
E=I(R_+Zz+Za)--I.(1-_flZz+R. ) (4.21)
where
\ +# /] I
E I R_+Z'+Za
R+ is the emitter resistor n , , rb-4-Z2l
Zz is the zener diode impedance n+tr+t 1-_j I_
Z_ is the impedance of CR1
is current gain of the transistor (4.25)
and
If the voltage dropped between the transistor . E a (4.26)
junction and ground is V, then: Z-=I R +r, ±r_+Z_
e T_--_
V=IR++le(Re+r,) (4.22) or
107
782-552 o--_i(i--s.
"l
ORBITING SOLAR OBSERVATORY
(v.-G-v_)
Re=emitter resistor
since Vr is completely independent of current.
Therefore, we have Then
.aV,
(4.30) Aie=aAi_=
108
• CONTROL SYSTEMS
+15V
CRI
CR2 ["
-I _L
i,!
"°I DIFFERENTIAL
o
OUTPUT
1
O
Rll
1
r_ C2 R8
t ! ,
FIGURE 4-29.--Servo preamplifier.
109
I i
ORBITING SOLAR OBSERVATORY
03 ÷
' RI3
I
2OK °1 G BL
" _ MDA "r -L..IFf MDA 2N328A
T _._RI
< IOK
BR
<
DOT 25
TI
-AN
I Mo. <_3v l R3
IOK
FOR + OUTPUT
I r Q4 / ml RI4
Q2
2N328A
20K
MDA
RQ6 _.
MDA _ 4.15V _
ZEYE I/.d
F o-
po _ RT3
MO
oR4;
_J cR3t
WO
_L_
t R23 _ CR6
:Y
CR5
PS595
T -
EL
o u
R.O.
CR8
PS 595 j - AZ R.0.
• _ Q7 I_1
VO ZC
• _ - 2NII32 : _ - 2NII32
TO Xo
R22
RI8
IOK
IOK
MDA
R20 MDA R24
3.9K 3.9K
RI9
R21
,>2K r 2K
OF+IS
provide the necessary, stability for the servo were tied together in order to minimize the
system. The differential output of the lead effect of differential temperature coefficients.
networks is then sent to the chopper and ac Thus, the temperature drift of this stage was
amplifier. the difference in the drifts in the base-emitter
The azimuth coarse and fine preamplifiers voltage, the collector and emitter resistors, and,
and the elevation fine preamplifier were identi- to a small degree, differential drifts in a. The
('al except for different value components needed collectors of the preamp were biased to 10v to
by differing coarse and fine eye currents and allow about a -t-5v peak swing.
differing lead network parameters.
Modulators
The preamplifier was mechanized as a dif-
ferential amplifier because it was self-tempera- In order to provide the necessary gain for the
ture stabilized to a first approximation, it had servo h)op, a fairly high-gain voltage amplifier
inherent rejection to supply voltage changes, was required. The standard method of syn-
and good common mode rejection. The bases chronous chopping (modulating) of the input
11o
eo
CONTROL SYSTEMS
CZ
DOT_
TI
FOR+ G
OUTPUT
X EYE CI
-(-15 V
FOR + OUTPUT
I' 2NTIGA
119
lOOK _ IN4S3
• 5 COARSE
._*CR5
ING45
C4
and output of an ac amplifier to provide tile Thus, the modulator drive voltage had to be
necessary low drift characteristics was used. greater than 4.2v peak, and the base-emitter
The modulator alternately sampled between diode had to have a breakdown _eater than
the differential outputs of the preamp. The the peak-to-peak signal, or 8.4 v. The rea-
output of the modulator was thus a square wave soning is as follows.
with peak-to-peak amplitude equal to the If the maximum signal were applied to the
differential outputs of the preamp. modulator, the voltage at Q1 emitter (Figure
At a frequency, such that the reactance of 4-32) would be 14.2v and at Q2 emitter, 5.8v.
the lead capacitor was small compared to 20K, When Q1 is conducting, the collector voltage
the signal output voltage to the modulator was of Q2 would be about 14.2v. To keep Q2 from
nearly equal to the collector voltage swing, drawing current through the collector-to-base
or 4.2v peak. Therefore, the modulator tran- diode, its base voltage must be greater than
sistors had to be able to hold off this voltage. 14.2v. Since the emitter of Q2 is at 5.8v, the
111
ORBITING SOLAR OBSERVATORY
R¢ ON FOR STATE 1
20K R1
: 1 R2
20K
', i
t Ra ---_ - I/2 R2
ON FOli STATE 2 L
v +(v or V )
T a b ALL VOLTAGES
RE. GND
112
•" CONTROL SYSTEMS
"Rc) was about 0.14 vsec. Since the rise times The measured values of C1 and C2 in one trans-
of the drive signals were considerably longer, former were 19 _f and 26 _f. The voltage
the voltage across R1 was the true derivative unbalance caused by these values of interwind-
of the drive voltage, which only had a value ing capacitance is:
during the rise or fall of the drive signal and
was much smaller than the peak-to-peak 2080X 106X7 X 10-12---- 14 mw
voltage.
The first two effects gave positive and nega- which is close to the measured unbalance of
tive spikes on alternate half-cycles because the about 17 my due only to chopper drive.
spike was the derivative of a square wave. The effects on the amplifier of the first two
These spikes were of greatest concern because problems were not serious since the spikes were
of the pulse widths which were 10 to 20 msecs. well filtered at the output. However, the
This pulse width was determined by the time spikes caused a reduction in gain proportional
constant of the 1000 mtf shunt capacitor of the to the spike area and the area of the signal out-
ac amplifier discharging through the preamp put square wave. This resulted in minor cross-
collector resistor. Without this capacitor the over distortion since the ratio became less with
height wotfld have been much greater but the increasing signal level. The spikes also had
width would have been much less. The spikes various minor effects because the amplified
were nearly eliminated by putting 0.1 uf spike could cause the output stages of the ac
capacitors from the collector to ground of each amplifier to limit.
transistor in the preamp. Since these capacitors The major problem was the change in C1 and
were across t.he collector resistor, they had C2 with temperature causing a change in un-
little effect on the frequency response. balance voltage. This was the major cause of
The third problem, caused by the trans- drift.
former interwinding capacitance, was an off-
Modulator Driver
balance at the modulator input. This was
caused by the rectifier action of the choppers. The modulator driver (see Figure 3-34) pro-
When chopper Q1 is conducting, C1 dumps vided the switching drive to the modulator and
C(V) coIdombs on one lead capacitor, and C2 demodulator. Its driving requirements were
removes C2(V) from the same lead capacitor. not severe except for rise and fall time. The
When the drive switches polarity, C1 re- circuit had to supply approximately 1 ma of
moves CI(V) cmdombs from the opposite lead drive current.
capacitor, and C2(V) is injected into it. The rise and fall time was important because
With both preamp inputs at lOv, the voltage of its effect on output noise. In a carefully
across C1 for state 1 in Figure 4-33 is 4v. designed system, the output noise from spikes is
When it switches to the opposite state, it _ill primarily at tbe switching rate, since the
charge to --9 volts if the drive is symmetrical. majority of energy is at this frequency and the
Thus, the total voltage change across each output filter has the least attenuation at this
capacitor regardless of state is 13v. The cur- frequency. To keep the noise energy at the
rent into one lead capacitor is fV(C1-C2) and switching frequency to a minimum, the switch-
into the other lead capacitor is fV(C2-Cl). ing time nmst be kept at a minimum, thus
Since most of this current flows through the requiring a fast rise and fall time.
c.wo 20K resistors, the unbMance voltage is the To accomplish fast rise and fall times with
difference between these currents multiplied nornlal audio trausformers, the transformer
by 40K or drivin 7 impedance must be mucb less than
rated. Tiffs also reduced droop. Double emit-
40K X 2K(_X 13V [(C1-C2)-- (C2-C1)] ter followers were used to provide this low
impedance.
or
The driver circuit used a unijunction trigger
20S0 )< 106 (C1-C2) circuit to provide trigger pulses to a flip-flop.
113
ORBITING ,SOLAR OBSERVATORY "o
- I / .....
I ZN338
_M
Ii rc::Z .... 5 P
IOOK CRI4
IC"_SZ
1 T..... .y
-._%;;, ,
_N_38
CR5
PSS9
It was found that on occasion tile fall of tile high currents needed to accomplish this with an
trigger pulse turned on the transistor that the emitter resistor at drives approaching the
trigger rise had just turned off. To eliminate supply voltage.
this, a diode shunted by a resistor to discharge The transformers were connected from side to
the coupling capacitor was used in the trigger side instead of one side to ground to eliminate
line. It was later found that the high-speed dc in the primary, thus reducing droop.
diodes used for CR12 and CR13 also cured this
problem. AC AmpliFier
The flip-flop was common except for the steer- The ac amplifier amplified tile square wave
ing circuit. The diode of the off-side of the from the modulator (see Figure 4-34). It had
circuit was reverse-biased so tile turn-off pulse to have sufficient bandwith to preserve the rise
would not reach the base. However, tile diode and fall times, minimize droop, and maintain
in the on-side was not reverse-biased and the stable ac gain. The output impedance had to
turn-off pulse could readily pass to tile base. be low and not put dc in the primary of trans-
This scheme provided positive steering of the fl)rmer, T2, so that _ good square wave could be
pulses. retained at the transformer secondary.
The flip-flop had good rise and fall times. To Lowest noise level from the modulator was
keep the flip-flop from being loaded and to obtained when the impedance that the modu-
provide a low-impedance drive for the trans- lator worked into was high. Also, since this
formers, c()mplementary emitter followers were impedance was high in comparison to the shunt
used on each output. This circuit provided tile impedance in the rate network (20K), the varia-
low impedance of an enfitter follower on both tion in the amplifier impedance had little
half-cycles without tile requirement for the very effect on tile lead network. Therefore, an
114
**
CONTROL SYSTEMS
"efnitter-follower input stage was decided upon. nected to a voltage divider circuit which used
It was desired that the input stage clip at a capacitors to keep this end of the transformer
level somewhat greater than that required for at signal ground.
output saturation. To accomplish this, a Feedback was taken from the transformer
resistor was put in series with the base and primary to reduce the transformer driving
collector lead. When the signal went positive, impedance and thus help to compensate for
the collector voltage decreased to the emitter transformer deficiencies.
voltage and clamped the emitter voltage at a
Demodulator
value determined by the resistance divider
made up of emitter and collector resistors, or The demodulator consisted of a 7.5v refer-
about 5v. The resistor in the base lead ence, synchronous demodulator, 7r-section filter,
limited the current when the clamping voltage and output complementary pair emitter fol-
was reached. The value of collector and base lower for low-impedance drive (see Figure 4-34).
resistor had to be kept fairly low or the high- Since it operated synchronously with the
frequency response would have been jeopardized modulator, its output was a dc voltage propor-
due to capacitance. Negative clipping was tional to the amplified solar detector eye signal.
accomplished by transistor cutoff. Because it The voltage reference was used to provide
was originally planned to clip at +0.5v, the the quiescent output voltage for no ac signal
emitter quiescent voltage was set at 0.5v. To voltage. The actual reference came from
stabilize this voltage and thus the clipping matched zener diodes. These zeners were put
voltage and quiescent current against tempera- in series between +15v and ground with a
ture drift, a diode was put in the biasing circuit. resistor between them. The resistor was center-
To adjust gain, the emitter resistance was tapped, thus giving 7.5v. The 7.5v was fed
made up of 2 resistors as a voltage divider. to a complementary pair of emitter followers
The two amplifier stages were conventional like those used for the ac amplifier output.
grounded emitter circuits, with feedback to the This provided a low-impedance 7.5v so that the
emitter of the first stage. The amplifier had a reference was not loaded at full output.
lag network in the first collector to control The demodulator consisted of the driver
amplitude and phase at high frequencies to transformer, current-limiting resistors, and the
improve amplifier stability. demodulating transistors. The transformer
Decoupling was provided by a 1K resistor provided a peak voltage in excess of the peak
and a zener to ground. A zener diode was used voltage expected from end to end from the ac
instead of the conventional capacitor so that output transformer to assure that the demodu-
the gain would not be modulated by low- lator transistor block was cut off for one-half
frequency ripple of supply voltage. cycle.
The output stage used complementary emit- The resistor limited the drive current to a
ter followers for a low-impedance output. value several times that drawn from the demod-
This stage differed from that used in the modu- ulator during the conduction half-cycle.
lator driver outputs by the addition of the The transistors were alternately forward and
diodes in the base circuit. The purpose of the reverse-biased. When forward-biased, the col-
diodes was to provide proper bias for the emitter lector was clamped to the emitter and the drive
followers. The diode voltage drops tracked current was limited only by the transformer
the base emitter voltage needed to keep a small resistance, current-limiting resistor, and forward
quiescent current flowing in the stage and thus diode drop of the base-emitter diode.
eliminate crossover distortion. They also pro- When reverse-biased, no current could flow
vided low-impedance coupling to the bases of in the base-emitter or base-collector diode if
the two transistors. the voltage across the transformer was equal
To eliminate dc from the transformer pri- to instantaneous end-to-end voltage across the
mary, the end of the transformer which was not amplifier transformer. More explicitly, at max-
connected to the emitter followers was con- imam output, the voltage at the input to the
115
*Q
ORBITING SOLAR OBSERVATORY
3-PSSS5
, 4"I_V OpI.K
:N 38 -
o_J _,_
47K PSSS5
' ,
30
RIR R;3
IOK
h 2N ZON_
R$3 PI-A TO 47_Z RW RES ON P_
47K Oil PI-80mV£ S_O_AL TO PA
& - _,. RK "=e
t2_ 5K
2N338 I IN44_A I
9_ J ",_. I
5K
R30 _74GA
RSI R7
64K 47K t_O36 "*OOMFD
R -tl_, BZA -It¢
O_ H
' R;_2. ctl iov
_s,o__o_A
I=
NO _3
3- PS595 COMpARATO_ (- )
filter was 13v. The emitter voltage of the OFF Thus the output current divided by the tran-
transistor at this output was 11 volts less, or sistor _ at the coldest operating temperature
2 volts. Since the ba_e-collector diode was to had to be smaller than 2v divided by base-
be reverse-biased, the transformer had to put biasing resistance.
the base at a voltage in excess of 13v. Thus
the transformer had to supply more than 13 Pulse-Width Modulator
minus 2, or llv. Obviously, it was necessary The pulse-width modulator supplied pulses
that the base-emitter diode be able to withstand
with a duration proportional to the error signal
this reverse voltage. The only transistor found
from the demodtdator. These pulses drove the
with this high rating was the Raytheon 2N328A,
power amplifier.
which had a VEto greater than 70 volts, ten
The schematic of this circuit is shown in
times the maximum for the next highest tran-
sistor found. Figure 4-35. The block diagram below illus-
The filter was a simple _-section filter used to trates the operation.
filter the transient from switching, as well as to
generate a rapid roll-off in both azimuth and
elevation channels. The roll-off was necessary COMPARATOR
FORWARD J._
SECTION
SIGNAL
system could not oscillate, the resonance was PHASE INVERTING
NONINVERTING
AMPUFIER I REVERSE
116
.o
CONTROL SYSTEMS
• _RROR
SIGNAL
10
AND
COMPARATOR RAMP TO
VOLTAGE FORWARD
COMPARATOR 8
=OF
DEAD ZONE
ee
ERROR RAMP TO "_ 6
SIGNAL REVERSE
COMPARATOR 0
4
I i I I E
m- TIME m
I 'l I l
I
j t I I
• I z 2
i I I I
I J I
t I
I
DRIVE I I I I DEAD ZONE
I _u
: I I [-,
VOLTAGE I I I I
1 *
I 0
I t I I
I OUTPUT PULSES
FROM REVERSE O 2
E.
LIMITING O
AM PLIFIER 2_
:- TIME _ 4
H7
"o
15 MH 15 MH
fv-v
1200UF 18V
7 II,
I
120 Q2 Q4 120
C35G
\ / 0.33
47
I0()
__ )2N1016C 5
FORWARD
_ TO REVERSE
SIGNAL
POWER SIGNAL
MONITOR
A positive pulse of about 10 volts amplitude the line through Q4, Q6, and CR1 with brush A
from the pulse-width modulator was applied positive and brush B negative.
to the base of the emitter follower Q1. This Some typical instantaneous power distribu-
pulse fired the silicon-controlled rectifier Q2 tion figures for this circuit with the motor
and drove the transistor Q3 into saturation. stalled are as follows:
The torque motor was then connected across Motor armature resistance of 10 ohms
the line through Q2, Q3, and CR1 with brush B
Supply voltage of 18 volts
positive and brush A negative. Tile average Voltage drop due ;o Q2 and Q3, and
torque developed by the motor was controlled CR1 =3 volts
by the duration of pulse applied to Q1. Wqaen
Modulator power=0.05 watt
the signal was removed from the base of Q1, Driving power from Q1=1.8 watts
Q3 ceased to conduct and reset the silicon- Power dissipated in Q2, Q3 and CR1=4.5
controlled rectifier. The voltage transient, due wat ts
to the motor inductance was limited to a safe Power delivered to tim motor-----22.5 watts
peak value by the 3M capacitor across the Total power delivered by the supply
motor. Reversal of the motor was accom- =28.8 watts
plished by applying a positive pulse to the base Power efficiency of the power amplifier
of Q6, which drove Q5 into saturation and
_28.8--6.3X100%=78 %
fired Q4. This connected the motor across 28.8
118
•" CONTROL SYSTEMS
180
because with it in the circuit, most of the current
i, f was supplied by the solar-cell array. Without
Z
60 / //' the filter the peak currents
be supplied
resulting
partially
decrease in overall
would have bad to
by the battery
efficiency.
with a
p, 4O
jY 15. Voh Regulator
/
30
/ Many circuits in the OSO I control system
electronics required a stable supply voltage.
2O The main spacecraft power was obtained from
/
batteries, and the voltage on the power line
10
/
f could vary due to the state of charge of the
.7' , ,
batteries and due to varying loads. An elec-
II 40 50 61 10(
PERCENT FULL LOAD tronic regulator was developed to operate from
the battery voltage and supply a -}-15-volt out-
FIGURE 4-39.--Control electronics power section
put with a regulation of better than ± 1 percent
efficiency.
for the expected line and load variations.
Figure 4-41 shows the schematic diagram for
This efficiency was independent of the power this regulator. The input voltage from the
delivered to the load except at small load power batteries was connected to pins M and P. The
when the power required by the modulator regulated 15 volts appeared on pins K and P.
and voltage regulator became appreciable com- Transistors Q6, Q5, and Q3 were the high gain
pared to the power delivered by the power dc amplifier. Changes in the output voltage
amplifier. were coupled to the input of the dc amplifier by
Figure 4-39 shows how the efficiency of the D1, D2, and R12. These changes were ampli-
power section of the control electronics varied fied and the signal appearing at the output was
with load. This figure also shows how the coupled to the base of Q4, the series regulating
efficiency of a conventional linear power section transistor. If the output voltage of the regula-
varied with load. It can be seen from Figure tor tended to increase, then Q4 operated to
4-39 that a pulse-width modulation system minimize this increase. Figure 4-42 shows
requires much less power than a conventional typical transfer characteristics for the regulator.
linear amplifier system except at near maximum
loads. The normal load on the power amplifier Spin Jet Control
was between 5 and 10 percent of maximum.
The purpose of the spin jet control was to
This was the load after acquisition when the
maintain the spacecraft wheel section rotating
serve was maintained at null. During acquisi-
at a nearly constant rate so that the nutation
tion the load on the power amplifier increased
damper would work effectively. The desired
to maximum value, but this lasted only for
period of rotation for the OSO I was 0.495
about one minute at the start of the 60-min-
seconds %0.025 seconds. The nutation fre-
ute sunlit portion of the orbit. The circuit
quency was directly proportional to the wheel
schematic (Figure 4-40) shows the power am-
spin rate. The nutation damper'was a sharply
plifiers, drivers, and monitor circuits.
tuned energy absorber designed to operate in a
A double L-section energT storage network or
narrow range of wheel spin rates. A change in
filter was used in series with the 18-volt power sup-
nutation frequency of about 5 percent reduced
ply to provide peak currents for the power am-
the effectiveness of the damper by a factor of 2
plifier. The purpose of the filter was to reduce
which was about all that could be tolerated.
the pulse current load on the power supply,
thereby preventing modulation of the power Torques to maintain the spin rate were provided
supply voltage at the power amplifier pulse rate. by gas jets mounted on the arms. Gas was
The filter also increased the efficiency somewhat supplied to these jets by a solenoid valve that
119
ORBITING SOLAR OBSERVATORY
ON AZ C_STING I RIO
UPPER STRUCTURE
2K
I DRIVE SIG W( R8
[ FRO_ PWM
IOK ISK
3
S2
Q2 IK
2N338
2N1056
C3
N( +lOV
AZ POWER MONITOR
C( +lSv
I
O3
C35G'
30UF
RIO7
330fl
CW DRIV_ SIGNAL
FROM PWM
FIGURE 4-40.--Servo power amplifier and driver circuits and power monitor circuits.
actuated when the spin decreased below the put was a ramp which had a peak value propor-
preset value. tional to the period of rotation. The voltage
Figure 4-43 is a block diagram of the spin jet ramp was compared with a reference voltage by
control for OSO I. A photocell located on the a circuit designed to furnish an output only if
rotating wheel supplied a current pulse every the ramp exceeded the reference voltage.
time its field of view swept across the sun. Therefore, as the wheel slowed down, the peak
'['his current pulse was shaped to obtain a con- of the ramp increased. As soon as the ramps
stant amplitude pulse with small rise and fall had increased above a critical value, the com-
times. The differentiated output of the pulse parator supplied a trigger to the 4-second mono-
shaper was used as a trigger input to a bistable stable. The 4-second pulse from the mono-
mnltivibrator. The bistable multivibrator stable multivibrator passed through one stage
switched from one stable state to the other of power amplification and opened a solenoid
once every revolution of the wheel; therefore, valve. This valve supplied gas to the jets from
the duration of the output pulses of this circuit gas storage bottles for 4 seconds, which was
was eqllal to the period of the wheel rotation. sufficient to speed up the wheel by the desired
A bootstrap-type integnttor was used to ob- amount.
tain the time integral of the pulses from the Several problems associated with this tech-
bistable multivibrator. The integrator out- nique are not obvious. As the spacecraft comes
120
+
• CONTROL SYSTEMS
I
I-I
ul_l.t
l+_v)
llOV l'lSV --
into the sunlight, it passes through the penum- to the spin-up system before the spin eyes are
bral shadow where only a part of the sun is producing pulses of sufficient amplitude, gas
visible. If the turn-on circuitry applies power would be expended when not required. Thus,
very high-spin eye sensitivity was required.
Furthermore, to take another precautionary
....... 30°C
measure, the regenerative trigger discussed
-- 25-C
121
ORBITING
SOLAR
OBSERVATORY ""
This pulse is differentiated by C3 and th6
HBISTABLE /
1
MU LTIVIBRATOR;
AMPLIFIER
POWER [
SOLENOID
SPIN-UP
VALVE H GAS JETS
When the flip-flop is switched
stable state, the base drive to Q5 is removed
to the other
and
the current through R16 charges C8. Capaci-
__I ....
,,$----]
"K"
9 + 15 VOLTS ] ="C" TO SPIN RATE MON
+,,voL_ 02 T T
,
"$ "¢_--'--
L R,..
1
i
I q
LI
DOT R
c.
22K IOK 20V _R56 15/d
®:5,K
R4 2N1056 R37 2.7 MEG '
IOK Q2 2201'
-c
ZS, ,w_ 22K
XJMPER I
2N358 _338
R_22 K 47K
¢I C5
ST400
O19
SM72
122
•" CONTROL SYSTEMS
-7
The voltage comparator consists of the zener
I
diode D14 and diodes D10 and D12. D14 m 4 4_
z
supplies the reference voltage which appears 8 i i
t_
The power amplifier used to drive the sole- (°C) resistance resistance
123
782-552 0--66----9
o.
PHOTOCELLS
REAR APERTURES The pitch control system kept the spin axis of
FIGURE 4-47.--Pitch control eyes schematic. the spacecraft approximately normal to the
solar vector so that the elevation servo system
craft sequence timer. At 200 seconds after was within its operating range. The controller
third-stage spinup, the sequence timer energized operated whenever the spin axis was more than
the despin relay. (The third-stage engine had 3 degrees from the normal to the solar vector.
been separated from the spacecraft and the The controller operated gas jets which precessed
spacecraft's arms had been extended prior to the spacecraft until the spin axis was approxi-
200 seconds.) The despin relay applied power mately one degree out of normal to the solar
to open a solenoid valve which allowed gas to be vector in the opposite direction. A schematic
supplied to the despin jets. These jets slowed of the pitch control eye system is shown in
the wheel down until the lower limit of the spin- Figure 4-47, and a block diagram illustrating the
control circuit was passed, at which time the total pitch control system is given in Figure
;normal spin-control circuit actuated and allowed 4-48.
SUPPLY VOLTAGE
AMP
_ _ 1 _
NO" BISTABLE MULTI- [ I
l I J ,JVIBRATORNOI
F'I AMP
NO. 1
VALVE
NO.
UP
JET
T ] -] AmP _ [ I
NO2 l |
T [ HIGH
| PRESSURE
N_,GULATOR
30 PSI
SUPPLY VOLTAGE [ N 2 GAS
f iAMP I-'1
. H n"MP
o. vA,.vE 2 /
r
DOWN
PITCH
JET I
::R4 INO. 4 I
A, B, A', AND B' ARE PHOTOCELLS
124
"" CX)NTROL SYSTEMS
brator No. 1 to switch to its original stable The six photocells located around the wheel
are connected in parallel and supply the input
state. This removes the driving signal from
signal to the day power turn-on circuit. The
Power Amplifier No. 1 which closes the solenoid
photocells are connected between the base of
valve and removes the torque from the space-
Q1 and the 18-volt supply bus. In sunlight
craft.
the combined resistance of the eyes drops to
For _<89 °, the operation is the same except less than 15,000 ohms so that sufficient cur-
Photocells A' and B' and their associated rent is supplied to the base of Q1 to cause it to
control electronics are employed. Figure 4-49 saturate. The drop in potential at the col-
shows the electrical schematic of the automatic lector of Q1 is coupled to a magnetic latching
pitch control system. relay through transistors Q5, Q6, and Q7.
125
ORBITING SOLAR OBSERVATORY
"A"
:B2
_oK
R,'
IOK', :
!!
I "B"
04
O+IBV
D3 2N338 RII
I000.
CI C3 IN483A 2W
D5
[-IN253
"D"
o UP OUT
OI 12N338_
2N33e _7
;T 400
DI
__ IN751A
PC'A'UP
"C" IN483A IN485A _'_ +1C4 R,O <
i ; ,D6
R9 35V ¢ "SM72
33K IK .01
-- 7 MFO lOOK •
GN
PC-B-UP
01 "H"
PC-A-DOWN
"R" "S" P.A. TEMR PROBE
telSV
_RI5 RI9 ,
IOK IOK '
P O=P" + 18V
011
C5 C7 R24 <
IN253
_O,P.<
120 2W
b7K
RI6 RIB _M"
QI DOWN OUT
IN4B3A __1
113
._N3_
_400
INT_IA
07
_47K 47K I
!N550
1
RESISTORS ARE 5% EXCEPT AS NOTED
The relay is latched over and turns on the day driver transistor to remove power from the
power. In order to conserve power, a set of relay coil.
contacts on this same relay is used to remove
Undervoltage Switch
the voltage from the last energized relay coil
once the relay is latched. The undervoltage switch was located be-
When the satellite enters the dark portion tween the spacecraft batteries and the space-
of the orbit, the photocell resistance increases craft battery load.
and Q1 is cut off. The positive potential at The spacecraft batteries were nickel-cad-
the collector of Q1 is coupled through tran- mium cells which had a rather flat discharge
sistors Q2, Q3, and Q4, thus energizing the curve until they became ahnost fully discharged.
OFF relay coil. The relay latches over to Near the fully discharged state, the voltage
remove power from the day power bus and at would fall abruptly and the impedance would
the same time grounds the base of the relay rise so that very little power could be drawn
126
•" CONTROL SYSTEMS
_2
11
I.)lcr ECT()I{
o, ORIENTATION OF' 'rHF I _._ NO.I
Z 6 TURN-ON
EYE , +,
_ 10 DETECTORS AROUND [ _ _ ^ /_
THE SPACECRAFT ( N / o I
w. EL \ I\ / I
I_ RELATIVE
_ 9
I _ WHEEL
_z
°i
r_
_ 6
,,
20 40 ,
60 ,
80 ,
I00 120 140,,, 160 180 200 2 0 240 260 280 , , , b
3 0 3 0 3 0 '
360
RELATIVE WHEEL ROTATION ANGLE (DEGREES)
from them. If the batteries had become al- spacecraft and permit the batteries to charge
most fully discharged they would not have been from the rotating solar-cell array. When the
able to deliver the current necessary to allow batteries were about 10 percent charged, the
the servo to catch and orient the upper section. undervoltage switch turned on the spacecraft
The azimuth servo would have tried to ac- again. A block diagram of the undervoltage
quire the sun throughout the daylight portion switch is shown in Figure 4-52.
of the orbit and thus would never have per- All the power to the spacecraft arrived
mitted the batteries to recover. This would through a 6-pole double-throw latching relay.
have led to a disabling of the spacecraft. By Five of the poles carried spacecraft power and
ground command, sections of the spacecraft one was used for circuit switching by the under-
could be turned off to conserve power so that voltage switch.
the undervoltage condition could be corrected. Figure 4-53 shows a typical charge-discharge
However, if the command system was not cm-ve for an OSO I battery pack. Reference to
operative and the batteries discharged to the this figure shows that the battery voltage
point where the voltage dropped abruptly, a starts dropping rapidly once it reaches 17
last measure was available to prevent the loss volts, and at 15.5 volts the drop in voltage is
of the vehicle. This was the undervoltage so great that the battery pack can be con-
switch which could shut down the entire sidered discharged at this point.
127
ORBITING SOLAR OBSERVATORY
RI ISK
1
OI 2N338 CRI PS595 O8
2NI711
R7 I
39OK 2N'_
_ISK
R8 _
CR2jPS595
_2N338 0_ 47K :'r-io0 _.f
2,0:', I
82K 33K IOK
Q3 Q6 RIh
2N338 2N338 _2K,
04 Q7
2NI482
V j H
X Y Z W
IK
-- SPARE A- SPIN
_RED DOT (DELAYED ISV)
]P?lCOEpq.
[
-,, ,,, I
I 18V DAY TO AZ
I CASTING
SLIP RING
#9
I8V NIGHT I
®
(ORBIT)
K9
BATT # I THRU
()n charge the battery voltage rapidly in- upper swit('hin_ point of the undervoltage
creased until it was 19.0 volts, or about l0 switch was dictated by the temperature charac-
per('ent charged. Above 19 volts, the rate of teristics of the batteries (see Figures 4-54
increase of voltage was approximately constant and 4-55). The turn-on point of the under-
until a peak was reached at about 90 percent of v(,ltage switch had to be chosen h)w
full charge. enough s(, that if t he batteries were warm, that
The un(lerv(dtage switch was designed t- is, 30 ° C, the v(,ltage would increase above the
remove tile l()ad from the batteries if their set point. If the batteries were cold, that is
v()ltage dropped below 15.5 volts, and it would -10 ° C, the set point had t(, be high enough
not reconnect the load until the battery voltage to insure that the batteries receive a minimum
increased to 19.9 volts. The choice of the charge before the load was reconnected.
128
•" CONTROL SYSTEMS
A
SATE LLITE POWER
i A-
{
I 47 UF
{ 47 UF
I
i
±
TO BATTERIES
_ b
I i, I DRIVER
"ON"
T A [ :
i TURN-OF_ ±
,
[ i TURN-OFF
VOLTAGE
REFERENCE
SIGNAL
I -
LATCHING REGENERATIVE
RELAY I SWITCH
(SHOWN IN I [ TURN-ON
"POWER ON't--_ VOLTAGE
CONDITION) " REFERENCE
FIGURE 4-52.--Undervoltage
switchblockdiagram.
Figure 4-56 is the schematic diagram of the circuit having a characteristic similar to a
undervoltage switch together with some tem- Schmitt circuit. The collectors of both these
perature monitoring circuits and a voltage transistors are coupled through emitter fol-
monitor circuit. lowers Q3 and Q6 to power transistors Q4 and
Transistors Q1 and Q2 are connected so Q5. When the voltage level at Q1 exceeds a
that they operate as a regenerative switching preset value determined by the zener diode
D5, both Q1 and Q2 switch to full conduction
and the collector of Q2 goes positive. This
positive voltage causes the power transistor
CHARGE AT 200 MA Q6 to conduct and operate the magnetic
lat ching relay whose contacts connect the load
to the battery packs. This "has a set of con-
tacts which switch between two different
voltage divider circuits at the input to Q1.
When the load is connected to the batteries,
the voltage divider keeps Q1 in saturation
until the battery voltage drops below a lower
15
limit determined by the setting of R7. When
l
2 4 6 e 1o f2 ii i',i_ i0 _2 this lower limit is reached, Q1 and Q2 switch
TD_E(HOURS) off and the collector of Q1 goes positive causing
FIGURE _--53.--Typical room temperature battery' Q4 to conduct and latch the relay over to
charge discharge curves. remove the battery load.
129
ORBITING
SOLAR
OBSERVATORY "°
1BY
,-
/.................. FAMII,Y OF CIIAItGE CI]IIVF,S FOR
NI-CI) I_ArI'TERY 'I'EST _ACK
17V .................. _ DECEMBER 6-21, 1960
I POWER SUPPLY 200 MA
I U.V.S. 15MA F
........ 20 °C
, _ _ __IOOC
...... Ooc
-- +10oC
........... +20 °C
........ +30"C
I I I I I
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
HOURS
The load is not reconnected until the battery the temperature monitors, was 27 ma at 18
voltage increases to a value sufficient to enable volts, or 49 milliwatts.
the second voltage divider to overcome the
Monitor Circuits
bias provided by D5 and cause Q1 to conduct.
The upper battery voltage suffMent to ('ause The following section describes the operation
(,ondu(,ti(m of Q1 is set by adjustment of R3.
of the spacecraft monitoring circuits.
From the explanation of the circuit operation,
it can be seen that this circuit disconnected the Temperature Monitor
battery load when the voltage dropped below a The temperatures of various points through-
level determined by R7 and reconnected the out the spacecraft were measured by the tem-
load when the voltage increased to a level deter- perature monitor circuits. The temperature
mined by the setting of R3. A set of contacts monitors furnished an output of 0 to 5 volts,
on the relay was used to disconnect the relay covering the temperature range of --40°C to
coil last energized to conserve power. +80°C, with the exception of the solar-cell sail
Figure 4-57 shows the effect of temperature temperature monitors. The solar-cell sail tem-
on the upper and lower switching points of the perature was monitored by two circuits ; one cir-
undervoltage switch. These measurements cuit furnished an output of 0 to 5 volts cover-
were taken with the upper switching point set ing the range of --70°C to +20°C, and the
for 18.5 volts at room temperature. Power other circuit output varied from 0 to 5 volts
consumption of the circuit board, which includes covering the range of +10°C to q-120°C.
130
•" CONTROL SYSTEMS
i i
FAMILY OF DISCHARGE CURVES FOR
NI-CD BATTERY TEST PACK
DECEMBER 8-21, 1960
22V
R LOAD 49 OHMS
I POWER SUPPLY 200 MA
I U.V.S. 15MA
21V. __
....... 30oc
.......... 20 °C i
......... 10°C
20V .....
OoC
..... 10°C
........ 20oC
19V'_ _
18V
.'_
"i
16V
15V
I
J
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
HOURS
Bead thermistors were used as the tempera- circuit had to have a resistance voltage char-
ture sensitive elements. These devices have a acteristic which was the same as the resistance
resistance-temperature characteristic, which is temperature characteristic of the thermistor.
closely approximated by the following equation: This requirement (shown in Figure 4-59) of
the monitor circuit was nearly achieved by
n(To) e_/r (4.33) the circuit, shown in Figure 4-60.
R(T)-- e_/_°
Figure 4-61 is a calibration curve for a
where
typical temperature-monitor circuit. These
monitors have proved to be stable in calibration
R(T) is the resistance at absolute tempera-
and will repeat the calibration curves within
ture, T
± 2 percenb. Figure 4-62 is a spacecraft
R(To) is the resistance at absolute tempera-
schematic showing some temperature monitors.
ture, 7'0
8 is a constant depending on the type of Battery Voltage Monitor
thermistor
The battery voltage monitor circuit was
located on the undervoltage switchboard. The
Figure 4-5,_ shows a plot of the resistance
purpose of Hfis circuit was to supply an output
characteristics of a Fenwall Type GA51L2 bead
thermistor. This refit has a resistance of which varied from 0 to 5 volts as the battery
100 kilohms at room temperature. voltage varied from about 14 to 20 volts. The
In order to obtain a linear temperature region of 14 to 20 volts on the battery charge-
voltage characteristic, the temperature monitor discharge curves represented the region which
131
ORBITING
SOLAR
OBSERVATORY
I I 12
0=2
• INTSIA
i PI -T I
D¢3
IN751A
R24
p-_/_ EOUT TO COMM
2ZK PI-X
., .... ,...... o_
PI-D LIMIT 02 _23,
SlOn,
SPARE CIRCUIT /
---o Pt- a
+lSv
lOOK THERMISTOR PROBE
LOCATED *N WHEEL
T_ I TM 2 T_ 3
_000T CC_L
CONTACTS CLOCKWISE
FILTO_, IN¢
LLI2EIBA
12 VOLTS
21 _" 8O
{-
_o
\
>
T -40
-_o b _ 4b s_ sb _bo
2 _ _ _oo 2b 3b 5_o _doo
TEMPERATURE (°C) RESISTANCE (KILORMS)
FIGURE 4-57.--Temperature dependence of the switch- FI(;URE 4-58.--Fenwall type GA51LZ bead thermistor
lug poi_ts of the undervoltage switch. rcsistor temperature characteristics.
132
"" CONTROL SYSTEMS
r.a
O O
4
D
O 2
O
_ 1
0 0
RESISTANCE (KILOHMS) z
O
X
FIGURE 4-59.--Temperature monitor circuit RI versus
-_o do
E0 characteristics. 6 z0 40 _o tim
TEMPERATURE ('C)
p +i 5 VDC
FIGURE 4-61.--Calibration curve for a typical temper-
ature monitor circuit.
% control
The
telemetering
spin
circuit.
rate
the
of the
The
timing
peak
wheel
ramp
value
was monitored
from
of
the
this
spin
ramp
by
could _ve some information on battery state telemetry system. Figure 4-65 shows the
of charge. Calibration curves for a typical calibration curve taken on the flight model
monitor.
circuit are shown in FiglLre 4-63.
normal to the spacecraft spin axis, was moni- two sets of fine eye detectors (readout eyes)
tored by means of a solar detector mounted on were aligned to point parallel to the optical
the sail structure. The monitor had a half- axis of the pointed experiments and hence
opaque circular aperture through which sun- boresighted to the control fine eye detectors.
light was _dlowed to fall on a silicon cell. As The control eyes were moun_ed on one pointed
the pitch angle changed, the edge of the opaque experiment and the readout eyes on the o_her.
half of the entrance aperture shadowed more or In this manner, information could be recorded
less of the cell. This gives a current outpu_ to show any relative shift between the pointed
133
ORBITING
SOLAR
OBSERVATORY I, o
OI e2 03 Q4
,_oL R5
_OOK
RH
tOOK
,,. I% t%
,? r2_ RT3
CI) I%
R$ R7
lOOK lOOK
°.
---t
07 2N1036 RI7 D
Ri6 196K IN755-A "R" REGULATOR OUT +ISV
I% 08
750_
RIB - _NM" COMM."i_M.+ISV.
5K - 35V
05 2N338
- _ "N" COMM. TM. GNO.
RI4
RI3 _Q6 2N697
1" REGULATOR OUT +ISV
D2
2NB38 51K IR_I
_
I IN -754-A
,_T_
---_ _ ii 12 13 14 15 16 17 18 19 20
Launch Sequence Timer
BATTERY VOLTAGE (VOLTS)
134
9*
CONTROL SYSTEMS
WHT
0 +I 5VDC
CI':IA,
4.64K
R2
BLK
NI132
P5"_95 ¢2tl
CR3
oOUT
i+C1
CRI R1
CR2
IN754A 5K _I IN7 53A
, O GND
after the third-stage spin-up. Figure 4-66 relays which turned orbit power on and the
shows the block schematic diagram of this unit. redundant tape recorder off; at 400 seconds, a
The timer was started by an acceleration switch ground closure operated the relays which fired
which was closed by the radial acceleration the azimuth release squibs and the nutation
produced by the third-stage spin-up. The damper release squib; at 800 seconds, a pulse of
following outputs were programmed: 100 sec-
onds after timer start, a ground closure was
supplied to a relay which fired the arm release 200-SEC ORBIT POWER
AMPLIFIERS
ated the despin relay which started the satellite 400-SEC AZ AND NUTATION RELAy
5-
O CONSOLE
ARMING
RESET
AND RE-SET
a_ ARMING
CIRCUITS AND ]
4
$d
b_
.J COUNTER
[ BINARy
t- 3
ACCELE RATION
[.-, SWITCHES
CIRCUITS
_ 2
STARTING
CONSOLE
0
REMOTE
_0. C _- 30_C
START
.-i
0
, , , , , , ,
.5 1.0 1.5 2.0 2.5 3.0 3.5
FIGURE 4-65.--Spin rate period (sec/rcv). FIGURE 4-66.--Launch sequence timer block diagram.
135
ORBITING SOLAR OBSERVATORY J_
+18 volts was provided to operate the relay The timer outputs were derived from multi-"
which fired the elevation release squib. In vibrator stages. A buffer amplifier was used
addition, all the previous outputs of the timer to isolate the outputs from the binary divider.
were repeated at 800 seconds. Immediately An RC coupling network having a time constant
after supplying the 800-second output, the of about 2.5 seconds was used at the input to
timer turned itself off. Also, to assure reliable the buffers so that the outputs were negative
sequence, a completely redundant timer was pulses lasting from 2 to 3 seconds. The outputs
used in parallel. The timer consisted of an of these buffers were either used directly or
8-stage binary divider, a unijunction oscillator, were used to operate relays which in turn
buffer amplifiers, a 15-volt regulator, and start- supplied outputs.
ing and stopping logic circuits. Figure 4-67 At the last timer output (800 seconds after
shows the schematic diagrams of the timer, timer start) an 800-second output was supplied
Figure 4-68 shows the printed board wiring of and all of the previous outputs were repeated.
the timer, and Figure 4-69 shows the assembled This was done to increase reliability, i.e., in
timer package. case one of these outputs had not performed its
The timer was started by an acceleration function, a second redundant output pulse
switch closure caused by the radial acceleration might then have produced the desired result.
due to the third-stage spin-up. Either of two After the completion of the 800-second outputs,
redundant acceleration switches could start the timers shut themselves off and could not be
both of the timers. There were two require- started again unless a console reset and arming
ments on timer starting: (1) the timers must signal was applied followed by the proper
not start the timing sequence by intermittent acceleration switch closure.
ch)sures of the acceleration switches due to For ground testing, a remote starting input
shock and vibration during launch, and (2) the was used. This input latched the starting
timers must have a reset signal present when circuits into the running mode.
armed and ready to be started by the accelera-
Commutator
tion switch. The requirements on starting the
timers were met as follows.
The OSO I utilized an 8-channel FM telem-
When either acceleration switch closed, it
etry system. One channel of the data system
operated a double-pole, double-ttlrow relay.
was used for spacecraft systems data monitor-
One set of contacts was used to apply +15 volt
ing. In order to monitor the many poin[s of
power to the two timer unijunction oscillators.
interest (52), it was necessary to commutate
The other set of contacts was used to energize
the information into the data system. This was
a second relay whose contacts were used to
accomplished by two similar commutators
ground the divider reset lines to maintain a
whose outputs were sequenced into the data
positive reset on the counters. The unijunction channel.
oscillators supplied pulses with periods of 6.25
The requirements imposed on commutator
seconds. If one acceleration switch remained
design were: (1) the power drain had to be low;
closed for 12.5 seconds, an output was obtained
(2) offset voltages had to be low (1 mill|volt or
from the second stage of the counters. Tiffs
less); and (3) the hardware procurement had to
output was used to operate a DPDT latching
be compatible with the OSO I development
relay: One set of contacts were used to hold
schedule. The relay commutator was selected
+15 volt power on the unijunction oscillators, over alternate methods such as solid-state
and the other set of contacts were used to hold
switching because it best fit the overall re-
the reset relays energized. If the acceleration quirements. The final design had a sampling
switch did not remain closed for 12.5 seconds, rate of 0.5 to 1 see which was compatible with
power was removed from the oscillators and relay operation, and it had a total power drain
reset relays when the acceleration switch opened. of 137 mill|watts. The contact switching also
The counters were also reset to zero. guaranteed near zero dc offset voltages.
136
•" CONTROL
SYSTEMS
_37
ORBITING SOLAR OBSERVATORY
FIQTJRE
4-68.-Launch sequence timer circuitry.
T h e OS0 I coniiiiut ator was niechitnized tis tt output (see Figure 4-70). T h e channels to be
ring counter in which tnttgnetic latching relays conirnutated were connected t o t h e contacts of
were the active elenients. One pole of tt relay this set. T h e other set of contitcts were used
wits in the ring counter circuit ttnd tlie other wtts to control commutation. When chitrine1 1 was
used for the comrnutttted signal. on, 18 volts was ttpplied t o the O F F coil of
The cotimiutator, which was driven by R c.hnnnel 1, itnd the ON coil of channel 2. When
unijuriction clock, had a feature such that if tlie clock circuit grounded t h e return line of the
vertitin of the components had fttiled, the coin- (.oils (all chitnnels connected in parallel) ,
mutator would have reset itself t o channel 1. clittnnel 1 would throw t o OFF tind channel 2
The conitnut at or would thus continue t n operate t o O N . This action was assured by t h e charge
lifter n failure between channel 1 ttnd the chtm- stored in C9 as follows: C9 charged in less
riel wit 11 the bad coinponent. time than t h e activation time of the relay.
The cotn~nut2tt or piickitge also contained IL Therefore, if cliannel 1 relay rtctivtit ed befnre
voltnge divider from the regulated 15-volt clinnnel 2 reltty had turned on, the clittrge in the
supply t Iirit wis used t n calibrate t lie t elernet ry capwit o r ttcross the channel 2 ON coil would
chitnnel 2ind ti circuit t o niensure the 15v and httve thrown the r e h y 0 1 1 .
I8v supplies. By this action, channel 2 cttine on. Also,
The c w t i i n i u t 111 o r Iititl one DI’DT ltttching power was supplied to the high side of the OFF
reltty for cvic~licwrnniut tit ctl clirtnnel. One pole coil of channel 2 ttnd the ON coil of chitnnel 3.
o f Cti(’I1 relily wtis connec~ted t 0 the coltllllut ator These coils activated when the next ground
138
.*
CONTROL SYSTEMS
FIGCRE
4-69.--4ss~mhled timer package.
closure occurred on the coil return line. This and )et short enough so that each channel can
process repeated until the last channel \vas on, be sampled often enough.
at which time voltaFe was applied to the OFF The unijunction drives a filter circuit used to
coil of the last channel and through a diode to keep stray, high-frequency pulses from the Fate
the O S coil of the first channel. When the of the controlled switch, Q3. TThen the uni-
nest grounding of the coil return line occurred junction fired, the pulse from its base fired the
the last channel turned off and the first channel controlled switch. T h e snitch clamped the
turned on again. This completed the cycle. relay coil return line to ground as long as the
The clock timing \$-as acconipliihed with :t anode current remained above about 0.1 pa.
unij unc tion t ransiq tor in conjunction with TThen the previously O S relay was turning off,
timing elements R1, R2 and C'2. C2 had R power \%-asinterrupted to the rela- coil return
pwitive temperature coefficient n-hich waq linP for sin interral determined by the relay
partially c o m p e n ~ t e dfor by the reduction with trmsfer time, about 2 ms. The only current
temperature of the tripger voltage of the uni- through the controlled switch during this time
junction. The uncompensated drift I \ * N ~about \\-as from the discharge of stray capacitance.
6 percent over the teniperitture range of - 2 O O C Thus, the controlled switch mould then open
t o +30°C'. Thi, drift \\-ti> not berious becttuse until the nest trigger pulse. The resistor, R7,
the only requirement on timing W'HS that each \\-as a current limiter.
readin: hztve sufficient duration to establish ti The reset circuit protected against nialfunc-
good avertye in the preqence of telemetry noke tion, (that resulted in remora1 of +18r from
*e
___/9_ _:_;
d_'llN '_2 IN '=_5 IN #4 _N _t* .5 IN #6 iN #7 IN ;" _321N
-_ -i I ±
..... Ty_2cog
_::ii_ !_]]'_*°'_'I ...... l i i l ± I i.... i ± I r ±_ ;---i:
_ ±_ T I-- "_,o__1 __
COUMUTATOR 2N4S_
_g
MUTATOR rN
S IG OUT
15V _ 50v
OK R24 CHANN 25
'1 R 29
I.....
178K_ TO CHANN
. -ol
TO CHANN
the reset line) by recycling the commutator to coefficient. Since a 2 percent, or 50 my,
channel 1. The circuit protected against the change in the normal telemetry output voltage
following problems. represents 50 my, or 0.33 percent, of the 15v
1. An ON channel goes off, but the succeeding regulated line, or 0.28 percent of the +18v
channel does not go on. line, the supply voltages could be read quite
2. A relay that gets hung up between contacts. accurately.
The above conditions were sensed by the Calibration signals were provided by tapping
reset line. When voltage to the reset line was down the 15v regulated line. (The 15-volt
removed, the voltage across C6 decayed. supply was essenLially insensitive to temper-
When the voltage dropped below the zener ature.) Generation of calibrate signals with
voltage of CR14, Q7 was no longer held clamped, zeners with a suitable voltage coefficient would
and base drive was provided to Qs, which have required a larger power dissipation.
energized the ON coil of channel 1. At the Also, since _he 15v bus was read to a higher
time of the next clock pulse, power would flow accuracy than the calibration voltages, a
through the ON coil of channel 1 on. This correction factor for both the calibrate signal
broke the circuit to the + 18v side of the reset and the 15v bus can be derived as follows:
circuit, and thus the controlled switch dropped
out and the commutator proceeded normally. Vm----15v monitor output as read by telem-
CR13 prevented a base-emitter breakdown in etry
Q5 and Q6 when the commutator was working V,=actual output voltage of 15v monitor
normally; i.e., there was ÷18v on the cathode V=actual voltage of the 15v line
of CR13 and ground at the base of Q6. V_=actual calibrate voltage
The voltage monitors were mechanized with V_=V/'3
zener stacks which were chosen to provide V 5
?n = -- Va
140
"" CONTROL SYSTEMS
i4i
° •
, AZIMUTH ......
PSi- RELEASE
, NUTATION DAMPER
P86-LATCH AS$Y C-6852
K351
B-7407
__}.-LtJ-LL-L--L_
K 24 WHEEL
ARM 12
RELEASE
ARieS
RELEASE
ARMel
RELEASE
ARId 12
RELEASE
ARM II 3
RELEASE
The electronic circuit for despin is described the despin valve off permanently by a latching
on page 143. This circuit measured thespin rate relay.
of the spacecraft, and whenever tile spin rate
Pitch Control System
dropped to a preset level a little below nominal,
the circuit opened the spin-up nozzle for four The purpose of this system was to nmintain
seconds. Since the daytime spin moment of pitch attitude so that the spacecraft spin axis
inertia was about 23.5 slug-ft 2, one burst of remained within 3 degrees of perpendicular
the spin-up system increased the daytime spin to the solar vector.
rate by 0.032 rps. Thus, the spin rate was Figure 4-74 is a schematic diagram of the
held ch)se t() the nominal value. pitch control system. The locations of the
The (lespin valve was used only once, just components in the spacecraft are shown in
after separatism of the spacecraft from the spin- Figure 4-73. The nitrogen gas, stored at an
stabilize(I laun('h vehicle. The valve was initial pressure of 3000 psi, was piped through a
opened hy the hmnch sequence timer. In about filter to a pressure regulator which dropped the
140 sec()n(ts the spin rate (Ir()pl)ed below its pressure to 30 psia, and then to two solenoid
re)mired value. Then, when the electroni(' valves. When one wdve was open, gas dis-
circuit switched tim spin-up valve on, it switched cln_rged through a nozzle which aimed up from
142
e
CONTROL SYSTEMS
DE -SPIN
SOLENOID
LAUNCH WHEEL
VALVE
SEQUENCE TIMER ELECTRONICS BOX
DE -SPIN
REGULATED_
JET
N 2 SUPPLY) _
I
18Vc
200-SECOND i
MOMENTARY I
!
GROUND CLOSURE
TO INITIATE
DE -SPIN
SPIN-UP
SOLENOID
SPIN-UP
VALVE
JET
I _GULATED_._....___.._
N 2 SUPPLYJ _
CONSOLE DE-SPIN,
RE-SET GROUND
CLOSURE SPIN-UP 18V
4-SECOND
GROUND
CLOSURE
143
ORBITING SOLAR OBSERVATORY
SUPPLY VOLTAGE
> B
AMPLIFIER
NO. 1
MU LTIVIBRATOR AMPLIFIER
[ AMPLIFIER BISTABLE
NO. 1 H POWER
NO. 1 [
I
R2
NO. 2
¢
H PITCH [
_i_ _
VALVE
SOLENOID
NO. 1
30 PSI
HIGH-I)RFSSIJRE
Ng, GAS N2
SU Pl)LY VOLTAGE
SOLENOID
VALVE DOWN
REGUI ,ATOR _ I NO. 2
JET ]
NO. 3
AMPLIFIER
6[ AMPLIFIER
NO. 4
MULTIVIBRATOR
NO. 2
AMPLIFIER
NO. 2 'I
A, B, A', AND B' ARE PHOTOCELLS
R3 , R4
v
the right side of the sail, producing a torque 0.5 rps was 73.8 lb-ft-sec. This resulted in a
directed toward the sun. This caused the 0.0965
spinning wheel to precess in the pitch down pitch rate of _ rad sec-', or 0.075 deg sec-',
direction. When the other valve was open, thus each pitch attitude correction took about
gas discharged through a nozzle aimed up from 53 seconds.
the left side of the sail producing a torque
directed away from the sun. This caused the Basic Design Decisions
spacecraft to pitch up.
The electronic circuit whicil controlled the This section presents a discussion of decision
valves is described on page 126. Its effect was from which most of the details of the system
to open the appropriate valve when the pitch design were derived.
attitude deviated from the solar vector by
more than three degrees, and to keep the valve Mass Ejection
open until the pitch attitude deviated in the Tim requirement here was to change the
opposite direction by about one degree. (This angular momentum of the spacecraft. This
oversho,,t was provided for one of the could have been accomplished in only two ways:
experiments.) By reacting against some external object (such
The nozzle thrust was 0.05 lb, and the lever as the earth's magnetic field), or by shoring part
arm was 1.93 ft, so the torque was 0.0965 lb-ft. of the spacecraft angular momentum in some
The angular momentum of the wheel with a internal object (such _ts a spinning disk or a
moment of inertia of 23.5 slug-ft 2 spinning at high-velocity gas stream), and then throwing
144
•* CONTROL SYSTEMS
tlfis object away. Very early in the program For the nine gases listed above, we may
the decision was made to use this mass ejection calculate the following values for It:
method. A gas ejection system is most practi-
Ir
cal because of the ease of ge_iing highl velocity Pound (force)
sec/pound
at a large radius, and thus a high angular mo- Gas L (weight)
mentum per pound of ejected mass. H: ............. 274 8. 1
He ............ 167 9. 6
Propellant Selection Ne ............ 74 17. 5
A ............. 52 19. 8
The gas which was ejected from the nozzles Kr ............ 36 20. 2
had to have as little effect as possible on the CO ............ 74 22. 3
sensitive optical surfaces in experimental instru- N2 ............. 74 22. 3
ments. This consideration, and the desire for NO ............ 72 22. 6
system simplicity, dictated the selection of a O2............. 70 23. 0
passive propellant.
Of the four gases with the highest IT, nitrogen
If the propellant had been stored as a liquid
was chosen because of inertness.
rather than as a gas, less tank weight and
volume would have been required. However, Quantity of Propellant
this would have introduced serious balance and
phase separation problems. Therefore, it was Despin aud Spin-Up System--When the
decided to store the propellant as a gas. To spacecraft was placed in orbit, it had a moment
simplify development, a gas known not to con- of inertia of 19.7 slug-ft 2 and was spinning at
dense when passed through pressure regulators, about 3 rps. Thus, the initial angular momen-
valves, and nozzles was chosen. The propellant tum was 247 slug-ft s sec -_. Aiter despin, the
therefore had to have a critical temperature moment of inertia was 24.4 slug-ft 2 and the
well below the expected storage temperature spin rate was about 0.5 rps, so the angular
or a critical pressure well above the storage momentum was 77 slug-ft _ sec-_. This change
of 170 slug-ft 2 sec -_ in angular momentum was
pressure. These considerations limited the
choices to nitric oxide, oxygen, carbon mon- caused by the despin gas jets, which acted at
oxide, nitrogen, hydrogen, helium, argon, a radius of 3.97 ft. This requires a jet im-
krypton, and neon. pulse of 43 lb sec. With nozzle efficiency
An important criterion for the selection of a of 90 percent giving an effective specific im-
pulse of 67 lb sec/lb weight, despin consumed
specific propellant is tbe impulse which can be
0.64 lb of gas.
obtained from a given weight of gas-filled
To estimate the amount of gas required for
storage bottle. The storage capacity of a
spin-up it would be necessary to calculate the
spherical fiberglass bottle of the type used is
angular impulse delivered to the spacecraft
about 1.4 (10 s) psi (in 3) pound weight of bottle.
over its six-month life by eddy current damping
Now, at 460 ° R, 1 psi (in 3) of an ideal gas weighs and by aerodynamic drag. Neither of these
1.095 (10 -7) MW pounds, where MW is the impulses can be calculated, so it was arbi-
molecular weight of the gas; so to store 1 pound trarily decided to provide enough gas to give
of gas we need 65.3/MW pounds of bottle. an angular impulse of 1000 shig-ft _ sec -1,
Specific impulse, /_, is defined as the impulse or 3.76 lb of gas. This would have been
obtained from 1 lb weight of propellant. Then enough gas even if the drag were great enough
to reduce the spin rate by 7 percent per day.
(The actual spin rate decay, measured on the
lr-- 1, 18MW
65.3 MW+65.3 (4.36) orbiting spacecraft, was 0.2 percent per day.)
1-_ MW The despin and spin-up system carried 4.4
lb of nitrogen.
where Ir is the impulse per pound of (gas+ Pitch System--The pitch gas had to do two
bottle). jobs. First, it had to pitch the spacecraft
145
ORBITING
SOLAR
OBSERVATORY o.
enough to keep up with the change in direction reliable electronically, and was the metiro_l
of the solar vector due to the earth's orbital used.
motion. Second, it had to buck out any The valve-on time interval had to be long
external torques tending to pitch the space- compared to the response times of various
craft. In estimating the gas requirements, we parts of the pneumatic system, yet short enough
assumed that these two jobs were independent. to be regulated by a simple RC network.
The solar vector rotates 1 degree per day. Any interval between 2 and 6 seconds was
If the spacecraft spin axis makes an angle 0 considered usable.
with the ecliptic plane, the spacecraft must Considering the various tolerances involved,
pitch 1° cos _ per day to keep up with the if the spin rate had to be held within a 15-
solar vector. Assuming the worst case, the percent range, the spin rate increment had to be
spacecraft would have had to pitch 180 ° no more _han 8 percent and no less than 4 per-
during its six-month life. Since the angular cent of nominal rate. Tha_ is, the angular
momentum of the spacecraft was 77 slug-ft 2 impulses had to be between 3 and 6 ft-lb-sec.
sec -1, 180 ° pitch required an angular impulse of These considerations required _hat each spin-
77_r=242 lb-ft-sec. The pitch jets acted at up nozzle provide a thrust between 0.04 and
a radius of 1.93 ft, and provided 67 lb-sec 0.25 lb. Using the same thrust on despin,
impulse per pound of gas, so 180 ° pitch re- between 60 and 350 seconds would have been
quired 1.87 pounds of gas. required to be completed.
A precise analysis of the external torques It was decided to use 0.10-1b thrust on both
applied to the spacecraft would have been spin-up and despin nozzles.
extremely costly; for this reason only a crude Pitch--The pitch jet _hrust level was set by
worst-ease analysis was made. This analysis, the allowable nuta_ion. It was necessary to
which included the effects of gravity gradient, reduce the nutation amplitude to near 1 minute
aerodynamic drag, radiation pressure, and of arc, or 2.91 (10 -_) radians, to be certain that
current loops, showed that the angular impulse the nuta_ion would not interfere with pointing
about the roll axis adds up to less than 270 accuracy or cause excessive control power
lb-ft-sec during the six-month life of the useage.
spacecraft. To buck out this angular impulse, As shown in chapter 2, the amplitude A of
2.09 pounds of gas were required. nutation caused by control torque L is
The total pitch gas requirement then was
4 pounds. Since there was a little extra A-- LL
__ (4.37)
space available, tire pitch system was designed
where
to carry 4.4 pounds oi nitrogen.
The actual gas usage was much less than
I is tire spin moment of inertia
anticipated since the actual pitch roll was
I t is the transverse moment of inertia, and
about 4 degrees per week (about }_ of that
o_ is the spin rate of the spacecraft..
expected).
Despin az_d Spiy_-l'p--There were two ob- A_--3.01 (10 -_) L, or L=0.0967 ft-lb
vious schemes, either of which could have been
used to control spin rate. One was for a spin for 1-minute nutation amplitude. Since the
rate sensor to tm'u the jets on when the spin pitch jets act at a radius of 1.93 ft, the jet
rate reached some low limit, and to turn the thrust of 1-minute nutation amplitude is 0.05
jets off when the spin rate reached some high lb. This was tile t,hrust level used.
limit. Tile other was for the spin rate sensor
Storaye Pressure
to turn the jets on for some fixed time interval
whenever the spin rate reached the low limit. There was a variety of pneumatic components
This second method was simpler and more rat, ed for 3000 psi service, but components rated
146
"" CONTROL SYSTEMS
147
ORBITING SOLAR OBSERVATORY "°
then 0.05-lb. thrust requires pressure was 3000 psi, Pc should not be mu, cl_
less than 30 psi. The value Po=30 psi was
rh=1.96(10 -5) slug sec -1 selected.
Expansion Area Ratio--Figure 4-75 shows a
From Equation (4.38), we have plot of relative nozzle efficiency as a function
of e, derived from Equations (4.39) and (4.40).
AtPc=2.76 (10 -5) lb
This plot shows that although a large e is
desirable, there is not a great deal to be gained
This means that
by increasing e beyond 30 or 40.
if Po--100 psi, the throat diameter=0.019 If Pe is less than Pa, the behavior of the
in., nozzle is unpredictable. In orbit, P, is negligi-
if Pc=30 psi, the throat diameter=0.034 ble, but during laboratory testing it is not.
in., The testing was done in a vacuum chamber of
if Pc=10 psi, the throat diameter----0.059 2(106) cm 3 volume. The nozzle discharged
in. about 2(10 -5) slug sec -1 of nitrogen, so the
pressure in the vacuum chamber, P,, increased
A throat diameter less than 0.030 in. is difficult
at about 2(10 -3) psi sec-'. Figure 4-76 shows
to control, so Pc should not be much more than
the time required for P, to rise to P+, as a func-
30 psi.
tion of e. ]n order to allow 20 seconds for a
A simple single-stage pressure regulator can
be designed for a pressure ratio of 100, but not measurement, it was necessary to make _ less
for a much higher ratio. Since the gas storage than 31. It was decided to make , about 31.
98
96
94
92
90
86
Z
_. 84
82
80 ..-
> 78
576 - :
74
72
7O
FIGURE 4-75.--Relative nozzle efficiency as a function of expansion area ratio, for a nozzle operating in a vacuum.
148
•" CONTROL
SYSTEMS
Pc = 30 PSIA
F = 0.05 LB
NOZZLE IS DISCHARGING INTO A CHAMBER OF 2(106) CC VOLUME
AT INITIAL PRESSURE OF 0 PSIA.
v I i
--_-_ i
_ ..... : - '1........ I .....
i
z
0
o
r_
, !
I
=,,,,,,,=,=,,=,_,,,,,.,,,=,_.,,,=,_
i
Entrance Cone--Nozzle performance was only of 0.036 inch would produce 0.05 lb. thrust at
very slightly affected by the shape of the con- Pc=30 psia. However, it was felt that a throat
vergent section of the nozzle. A right circular with a sharp entry would behave as though its
cone with 45-degree half-angle was quite satis- diameter were less than its actual value.
factory on many large and small nozzles. Therefore, to get 0.036 in. effective diameter, it
Therefore, for the sake of simplicity, a 45- was decided to try 0.039 in actual diameter.
degree entrance cone was chosen. Ezit Cone Shape--If the expanding section
Throat Shape--In large nozzles, efficiency is ot the nozzle is a right circular cone, the exhaust
bad unless the throat is carefully contoured to gas stream is divergent. The thrust is not
provide a smooth path for gas flow. It was quite as great as if each gas particle were moving
felt that this contouring would probably be parallel to the axis. To correct for this stream
unnecessary in a very small nozzle. It was
divergence, the thrust indicated by Equation
decided to use a straight cylindrical throat with
(4.40) must be multiplied by _/ (l+0), where 0
no rounding at the ends.
It was shown earlier in this section that it is the hall-angle of the exit cone. This function
is plotted in Figure 4-77.
Pc----30 psia and F=0.05 lb, and if the nozzle is
100 percent efficient, the throat diameter should In applications where a small increase in
be 0.034 in. If the nozzle efficiency were 90 efficiency is important, it is customary to
percent, then a throat with an effective diameter contour the expanding section of the nozzle to
149
°Q
ORBITING SOLAR OBSERVATORY
99
,oo
94_
.... 90°
91
92_
EXIT CONE HALF-ANGLE, DEGREES
"\ -- (CONE DIA) //_
FIGURE 4--77.--Correction for stream divergence• \\' ,_ . 288 REF i/" .032
150
CONTROL
SYSTEMS
otherfactors,but it is clearthat therewasnot was anticipated that the bearing friction might
roomfor significantimprovement. increase slightly after a number of orbits and
Tberewereno highly critical dimensions in thus reduce the anticipated error toward zero.
the nozzle. Efficiencyis insensitiveto changes As can be seen from the curve of Figure 4-80,
in throat diameter,throat length, entrance the initial error was not in the direction antici-
angle,exitangle,andexitdiameter. Thismade pated, but the long-term error was as expected.
the nozzleeasyto build andinspect. The initial error was about 2 minutes of arc.
Thrustmeasurements confirmedthe estimate This was reduced to about 1 arc-minute after
asto theeffectivethroatdiameter. In vacuum, a few orbits and was from that point on about
the nozzlegave0.05lb. thrustat Pc----30.0 psia 1.5±0.5 arc-minutes. The constancy of the
and 0.10 ]b. thrust at Pc----60.0 psia. positioning demonstrates the long-term relia-
This nozzle was completely satisfactory for bility of the positioning control system.
its purpose. One possible explanation of the initial error
being in the direction opposite to that
PERFORMANCE
anticipated is that the dry lubricated bearings
The performance discussed in this section is may have been run-in at the 1 g load ground
in some cases necessarily that of the complete test position and may have had to be run-in
control system loop since this is all that the again at the orbit unloaded position. The fact
that the error was reduced over several orbits
information transmitted from the spacecraft
from the initial error supports this reasoning.
can report.
An interesting hump appears in the curve of
Control Systems
Figure 4-80 near orbit 600. The azimuth
Azimuth Positioning control system power usage, shown in Figure
4-81, shows the increase in power required to
The accuracy of azimuth positioning was
maintain point during this time.
designed to be i I arc minute with the actual
design requirement being ±5 arc minutes.
18
The system was adjusted for zero pointing error I IOSO[ I I
AZ]MUTH POWER
16
on the gound with the wheel section spinnin_ CHANNEL 40
DATA POINT TAKEN
14
at 0.5 rps. APPROXIMATELY
SATELLITE MID-DAY
'° I]11 I I I I t
OSO I 0 100 300 500 700 900 1100 1300
-- OPTICAL AXIS LEFT OF
AZIMUTH ANGLE
THE SOLAR VECTOR ORBIT NUMBER
CHANNEL
DATA POINT TAKEN
APPROXIMATELY
SATELLITE MID-DAY
FIGURE 4-81.--Azimuth power, 1000 orbits.
1S1
ORBITING SOLAR OBSERVATORY
six eyes in the block. The data from this eye 1 I [ I IIII j
-- OPTICAL AXIS ABOVE O_ I
pair corroborated the positioning accuracy as THE SOLAR VECTOR
EL_ATION ANGLE
I C_NNEL _
signed such that there is a drive dead zone for ORBIT NUMBERS
It is this variable friction problem that must than the edges of the brinnelled fiats on the
be solved when more accurate positioning is bearings. As can be seen from the figure, this
required. Liquid lubricants can produce a is not the case.
much lower and a much more constant bearing This leaves only the flexible leads to produce
friction level, and for this reason are being con- the error which is obviously a function of the
sidered for use on future spacecraft of the same pitch angle. The flexible leads were painted
design. with the same white paint that was used on the
back of the solar array to dump heat. The
Elevation Positioning flexible leads then must have been very cold
Elevation positioning was designed to be and thus were probably considerably stiffer than
_11/2 arc-minutes with the design requirement when the pointed experiments were dynamically
being ±5 arc-minutes. The system was ad- balanced on the ground at room temperature.
justed on the ground to zero error and it was It. is assumed then that the flexible leads acted
anticipated that the positioning error would as springs attempting to keep the pointed exper-
drift between the dead zone set points of ± 1}{ iments in a position normal to the spin axis
arc-minutes with perhaps one direction of error and that the elevation control system was
predominating due to magnetic coupling effects forced to use considerable power, as seen in
or to flexible lead unt)alanced spring action. Figure 4-84 to maintain point against tile spring
As can be seen from Figure 4-82 this was action of the flexible leads.
detinitely not the case. This problem has been solved for future space-
Figure 4-83 which shows the pitch angle and ('raft by using much hmger and more flexible .
the elevati(m error, quickly exphfins why the leads and by making the surfaces of the fiat
anticipated ('()nditions were not achieved. The leads shiny to keep their temperature near
(rely mechanical connections between the space- room temperature.
152
CONTROL
SYSTEMS
.6
4-
2- AZIMUTH POWER
(WATTS)
0
AZIMUTH POINTING
+2-
(MIN OF ARC)
0
-2-
6-
ELEVATION POWER
4- (WATTS)
2.
ELEVATION POINTING
+2.5 (MIN OF ARC)
+I. 5
0
-1.55:
+3"
PITCH ANGLE
+2"
(DEGREES)
+I"
_1 ° .
J j
_2 °
4 - _, , ) I i J I I
O 90 180 270 360 4.50 .540 (;30 720 8|0 900 990
ORBIT NUMBER
153
ORBITING SOLAR OBSERVATORY
SATELLrTE MID-DAY
of a much more serious nature.
f
As noted in Chapter 1, the spin-down and
I." // I
_,r j / / / / spin-up sequencing during initial acquisition
,., ¢ II / I
.J .......... fl
was perfect. Furthermore, spin-up control
was proper for many orbits thereafter. How-
I
OPTICAL AXIS BELOW THE
ever, beginning with orbit 895 the spin control
-6 -- SOLAR VECTOR
system produced several bursts of gas when
none were required to maintain spin rate. The
-10
50100 300 500 700 900 1100 1300 )500
last burst occurred near orbit 1150. There
ORBIT NUMBER
have been none since. A spin rate curve is
FIGURE 4-85.--Pitch readout, 1000 orbits. shown in Figure 4-86.
25°C
as a direct function of their own response _ 3O o C
154
•" C_£)NTROL
SYSTEMS
"16ad. This sequence of events was continu- with time. This cause of malfunction is
ously repeated until the spin rate gradually improbable.
was reduced by atmospheric drag to about 0.74 A missing trigger pulse for the integrator
rps at which time, during orbit 1622, the circuit is a probable cause of malfunction.
azimuth servo regained control. There are three likely possibilities that are
During the period of time from orbit 1139 to all functions of the deteciors. These are as
orbit 1622, the only data recorded were during follows:
those periods when the spacecraft was not in 1. Gradual Reduction of Detector Illuminated
the undervoltage condition and was within Sensitivity: There are two reasons why this
range of a receiving station. is thought to be an unlikely explanation of the
The undervoltage condition described above malfunction. First, the detectors of similar
was repeated after orbit 1822 until about orbit type ("on target" detectors and "day-night
3320 at which time the azimuth servo again turn-on" detectors) have not shown a similar
acquired control at about 0.66 rps spin rate. problem. Second, if this had been the problem,
The reason for loss of control after orbit several spin-up cycles over a short period of
1822 was that magnetic coupling effects be- time would have beeD the result. This was
tween on-board magnets and the earth's field not the case.
caused a sinusoidal variation in spin rate. 2. Detector Continuously Illuminated: The
This variation added to the spin rate after detectors were used in such manner that only
the azimuth servo regained control to again minimmn illumination was required to produce
spin the wheel too fast for servo t)ointing control a resistance low enough to trigger the pulse
to be maintained. This coupling effect integrator. This was done to insure that the
amounted to an increase of only 0.05 rps, but turn-on eyes would give a dark signal to turn
was enough, when added to the already high off spin-circuit power at an illumination level
spin rate, to cause the nonoriented condition considerably higher than that required for
to occur again. The spacecraft was mag- reliable spin-circuit operation. The reason
netically compensated before flight. Had this for this was to prevent the spin circuit from
not been done, the coupling effects would have missing a pulse just before power turn off.
been much greater. A missing pulse would result in a spin-up burst
The most probable reasons for the second at every spacecraft dusk. If the pulse shaper
were saturated continuously for one revolu-
acquisition being at 0.66 rps rather than 0.74
tion of the wheel, a spin-up burst would be
rps after the initial spin-up problem are
generated. A continuously illuminated de-
build-up of brush material on the torque motor
tector would have produced this result. No
commutator causing higher apparent com-
mutator resistance and thus reducing torque, possible earth-sun-spacecraft geometry could
and/or an increase in bearing, slip-ring, and have produced the condition of an illuminated
motor-brush frictional drag. earth such that the spin detectors could see
the earth continuously for one revolution.
In the following paragraphs some probable
Thus, a continuously illuminated detector was
and improbable causes of the spin-up mal-
function are discussed. Refer to the circuit an improbable cause of the malfunction.
3. Detector Dark Resistance Reduced: If the
of Figure 4-44.
the dark resistance of the detectors were reduced
It is difficult to find a place for a malfunction
from a nominal 500K ohms or greater to the
beyond the burst generator. It is thus as-
range 200K-280K ohms, the pulse shaper
sumed that pulses triggered the burst generator. would have been saturated. If the dark resist-
Stray pulses could have triggered the burst ance were decreased gradually (a possibility
generator. No such triggering occurred in if in a radiation environment), the pulse shaper
prelaunch checkouts or for the first 895 orbits. would gradually have become insensitive as the
If this were the malfunction cause, the pulses circuit neared saturation. Just before per-
either grew more severe and then less sensitive manent saturation, the pulse shaper would
155
782-552 O--_16--11
ORBITING
SOLAR
OBSERVATORY
_15
_o o e • such that the power was cycled on and off about
five times before either staying off at dusk or
staying on at dawn. This switching was no
consequence to any spacecraft functions, but
l;_o 2'00 3bo _o'o 5_o 60'0 _o so'o Wo lOOO was unfortunately very confusing to the Uni-
ORBIT NUMBER
versity of Minnesota experiment. The experi-
FIOURE 4-87.--15-volt power amplifier monitor chan- ment used circuitry that detected and counted
nel 45, readout data. each power on-off cycle as one orbit. The
experiment then had been counting orbits in
very rapid groups of 5 or 6 which made dat_L
reduction more difficult than anticipated.
Silicon detectors of much more reliable
characteristics are being used on future space-
oo _ e
Undervoltage Switch
1_o 2b 3_o 4bo 5(;o 6oh 7_o 860 9bo 1ooo
ORBIT NUMBER
Unfortunately, as described fully in the spin-
FIGURE 4-88.--15-volt linear amplifier monitor chan- circuit performance description, the under-
nel 44, readout data. voltage circuit was given ample opportunity to
156
44
CONTROL SYSTEMS
"d_monstrate excellent performance. The circuit The detector output varied with good agree-
worked properly several hundred times. The ment to the theoretical variation in the solar
circuit was designed to give a last measure of constant. This information indicated that the
life to the spacecraft. As it happened, the life degradation measured for the fine pointing
of the spacecraft without the protection would detectors was due entirely to effects on the lens
have been only two and one-half months. glass (probably ultruviolet-radiation-induced
color center formation causing some attenu-
Monitor Circuits
ation), and that the degradation of the cells was
As can be seen from the curves of this section, negligible when protected by the filter glass.
all of the monitor circuits performed well. Of considerable interest would be data con-
Data was permanently lost from most of the cerning the effects of high energy electron flux
monitors after the commutators failed. on the detectors. Unfortunately, the detector
outputs were reduced to the lowest level that
Sensor Monitoring Experiment
was readable by about orbit 2000. This was
The space environment effects upon the Ball about the time the 9 July 1962 high-altitude
Brothers engineering experiment detectors are nuclear device, which caused the electron flux,
listed below. was exploded. Since then the coarse detector
De_tiou eye did not continue to give on-scale readings
(percent per
Detector Type month) after that time, as would have been anticipated
Open cell ..................... 3. 6 since the solax constant was on the increasing
Fine pointing .................. 1.8 half-cycle, it was assumed that some radiation
Fine pointing .................. 1.2
change had occurred. This change would have
Coarse pointing ................ 0
had to be only about 3 percent to produce the
These degradation figures were derived over effects measured, thus no real limits can be
the first 1000 orbits and were, for this period of estimated. It is thought that the effect of
time, linear functions of time. It was assumed electron flux on the detectors would be evidenced
that the data could be extrapolated to 6 months only in the lens glass from the x-rays caused by
with some confidence if the electron flux problem the electrons impinging on the lens light shields.
was ignored. It is thought that these x-ray effects may be of
The open-cell detector exposed the silicon cell greater magnitude than the ultraviolet effects.
directly to solar irradiation. The degradation The effects of the x-rays on the silicon cells is
was due to direct ultraviolet irradiation damage. thought to be negligible since the energy levels
The fine pointing detectors, with lenses and of the x-rays produced are less than the destruc-
narrow transmission band falters, were the type tive energy level threshold for silicon semi-
used for the servo-positioning-control error conductors.
detectors. The two detectors were not spec- The azimuth position readout pair of detec-
trally matched. This is the reason for the tors performed well. After the upper commu-
slightly different degradation rates. The servo tator failure about orbit 1900, this detector pair
gain decrease in 6 months (9 percent average) became the only readout available to allow
due to eye output reduction caused a positioning determination of azimuth pointing accuracy.
error of the same percentage applied to the A slight shift of about _ arc-minute in
initial error. That is, with a 1.0 arc-minute relative azimuth pointing direction of the
initial control accuracy, the error after 6 months pointed experiments occurred during launch.
was increased by 0.1 arc-minute. The variation After this shift had occurred, agreement was
in positional accuracy caused by the different found unchanged between the actual azimuth
degradation rates was negligible. pointing readout detectors and the pair used
The coarse pointing detector was identical to on this experiment. After 5000 orbits the
the coarse detectors used for the positioning experimental pair was reading azimuth posi-
control system. No degradation was measured. tioning errors of less than one arc-minute.
157
"o
ORBITINGSOLAR OBSERVATORY
DO
Two curves of interest are presented concern- orbits) of about 10 percent in output produced
ing the experiment. The curves of Figure 4-90 low-end-oJ-scalereadings. Thus, this curve could
show the combined effects of degradation and not be obtained. Of some interest, however,
solar constant variation over the first 1000 is the fact that the detectors did decrease this
orbits. The solar constant change was a 10 percent over the period of time from about
decrease of 3.2 percent over this period. The 21 seconds prior to power turn-off to about 13
coarse detector, as noted previously, followed seconds prior to power turn-off. This indicates
this change very closely. that there is some atmospheric attenuation in
The curve of Figure 4-91 shows the coarse the narrow infrared bandwidth to which the
detector output variation over one orbit. The detectors are sensitive prior to that attenua-
peaks after dawn and before dusk were of tion due to the spacecraft entering the earth's
variable amplitude and varied somewhat in penumbral shadow.
shape and in time as the spacecraft-earth-sun
Commutators
orientation changed. The peaks were due to
earth light inputs to the coarse detector. The Four commutators were incorporated in the
curve plotted is typical. OSO I. These were the two spacecraft house-
Of value to the circuit designers would have keeping channel commutators, providing 30
been a time-attenuation curve for solar intensity points of information in the wheel section and
as the spacecraft emerged from and went into 22 points in the upper section, the alternate
the earth's shadow. Due to the sensitivity of pointed experiment 20-point commutator, and
the fine detector readout, a decrease (first few the Rochester 20-point commutator.
158
.s
C,ONTROL SYSTEMS
_06
DAY POWER DAY POWER
TURN ON TURN OFF
104
_'IGHT I' DAY r
\ !
/
102
100 i
Z / i
8 ................
I
i
z
_ 98 t
o
z i
_ 96
! ' I
i I
94
I
1
92
0 400 800 1200 1600 2000 2400 2800 3200 3600
TIME (SEC)
Of these four devices, two have failed. The was not accomplished in this time. The lower
spacecraft lower commutator failed during commutator has been operating in this sequence
orbit 341, causing the loss of 22 of the 32 lower since the failure of the relay that switched the
commutator data points. The spacecraft upper eleventh data point.
commutator failed after orbit 1900 with all of The reason for the upper commutator failure
the upper commutator points being lost. The was not understood. The commutator was
experiment commutators were working properly not cycling. The method of operation of the
after 3500 orbits. two commutators was as follows. The lower
The housekeeping data commutators were commutator ran its sequence in 32 seconds
designed such that the failure of a switching with 1 second dwell time data sampling and
relay to make contact would cause the device then reset to point one. At the reset time the
to reset to channel 1 and would thus allow output line from the upper commutator, which
cycling from channel 1 to the channel switched was cycling continuously, was switched into
by the bad relay. This was accomplished by a the output line through the lower commutator.
circuit that allowed the relays 0.5 seconds to The lower commutator completed its sequence
throw and would cause a reset if relay closure while the upper commutator was read out.
159
oo
Chapter 5
161
ORBITING
SOLAR
OBSERVATORY
162
•° DATAACQUISITION AND COMMAND SYSTEM
• Signal-to-noise ratio achieved under ideal Worst Conditions: By increasing the sky tem-
conditions: perature to 5170°K (the remaining parameters
being unchanged), the margin of safety becomes:
Pt" G: X_
(5.3)
S/N=kT . Tt (worst)=5170+290+290
=5750°K
where:
163
ORBITING SOLAR OBSERVATORY
i
]
1.
"7
164
a*
I)firing the 90-minute phase of the orbit, the nominal modulation index of 5. The choice of
transmitter output was reduced to 300 milli- subcarrier frequencies was predicated upon
watts and the low-frequency, real-time sub- limited response of the tape reccrder in con-
carriers were transmitted. When the satellite junction with tape storage capacity and the
was within receiver range, the transmitter out- necessity to operate at a reasonable trans-
put was increased to 1.75 watts and the high mission bandwidth to conserve transmitter
frequency subcarriers from the tape recorder power.
were transmitted.
Channel 1: Center frequency 164 cps Uni-
The power amplifier was designed to operate versity of Minnesota--PCM, 1 bit per sec.
without spurious oscillations into a tuned load, Channel 2: Center frequency 212 cps OSO I
such as the diplexer, with a VSWR as great as spacecraft status--52 PAM channels, sub-
2:1.
commutated at 1 sample per second.
Channel 3: Center frequency 294 cps God-
Antenna System
dard alternate pointed experiment--20
Considering the vector sum of the vertical PAM channels, subcommutated at 1
and horizontal polarization components, the sample per second.
antenna patterns exhibited nulls no greater than Channel 4: Center frequency 400 cps Uni-
--6 db for 95 percent of all aspect angles. The versity of Rochester--20 PDM channels,
vertical and horizontal polarization components pulse per second.
were received at the Minitrack ground station
Channel 5: Center frequency 572 cps Uni-
via two parallel receiving channels. As the
versity of California--pulse rate informa-
spacecraft rotated at _ revolution per second,
tion at 4 pulses per second.
antenna nulls were realized in one polarization Channel 6: Center frequency 790 cps God-
component or the other. A post-detection dard wheel experiment--pulse rate in-
diversity combiner performed the summing of formation at 4 pulses per second.
the two signals and provided a smoothed video Channel 7: Tape speed compensation refer-
output continuously as great as the stronger of ence oscillator, fixed at 955 cps.
the two signals. Channel 8: Center frequency 1200 cps God-
dard pointed experiment--pulse rate in-
FM Muitiplexin9 System formation at 6 pulses per second.
The OSO I telemetry system was an FM- When the tape recorder played back at 18.35
FM system (Figure 5-1). The outputs from times record speed, channels 1 through 8
eight nonstandard low-frequency subcarrier became 3 kc, 3.9 kc, 5.4 kc, 7.35 kc, 10.5 kc,
oscillators were multiplexed into a magnetic 14.5 kc, 17.5 kc, and 22 kc, respectively. The
tap recorder for orbital storage. The tape playback ratio was critical since the sub-
recorder was interrogated and played back the carrier frequencies had to remain within the
recorded signal when in range of one of the -4-7.5 percent bandwidth of the ground station
Minitrack ground stations. filters.
At the same time that the subcarrier multi- During playback, an additional channel at
plex was being stored, it was also being fed to 2.3 kc was mixed with the above frequencies
the transmitter for continuous real-time trans- at the primary transmitter input. The signal
mission. When the tape recorder was played originated at a subcarrier in the Ames emissivity
back, the real-time subcarrier signals were experiment.
removed from the transmitter and the stored To increase reliability, two sets of sub-
frequencies were transmitted. At the end of carrier oscillators were provided. Only one
playback, the real-time frequencies were again set was turned on. No switching was re-
connected. quired at the input of the oscillators since each
The subcarrier input data rates from the of the data channels was connected to each of
sources indicated below were dictated by a the redundant oscillator pairs.
165
ORBITING SOLAR OBSERVATORY
Receiver
T h e receiver WBS crystal-controlled AM with
- 100 dbni sensitivity a t 75 percent modula-
tion. It was limited to receiving one tone a t
Airborne Decoders
T h e dccoder, consisting of R series of seven
bandpnss filters, wtts designed so that tfhetrans-
iuissiori of two toaes in succession was required
t o actunte IL given coiniumd chtmnel. Tone 1
permitted the selection of the OFF condition
in prepnratiori to actutite tones :< through 7.
Tone 2 permitted the selection of the ON
condition in preparation to actuate tones 3
FIGURE5-3.--Subcarrier oscillator package. through 7. Once a given condition was estab-
•" DATA ACQUISITION AND COMMAND SYSTEM
167
oo
ORBITING SOLAR OBSERVATORY
ing array were brought out separately; these In order to permit quick-look capability at
were passed through notch filters which rejected Fort Myers, the video output of each of the two
the command frequency and thereby protected Mod I receivers was combined in a Defense
the receiver preamplifiers which were also Electronics Model TDC-1 diversity combiner.
mounted on the tower. From the preampli- The composite signal was sent to the EMR dis-
tiers the two outputs were brought into a criminators where it was reduced to the Sanborn
polarization differentiation unit which allowed display.
selection of all possible modes of polarization, The two video signals were also recorded on
either simultaneously or one at a time. To channels 1 and 2 of the station tape recorder.
assure correct polarization differentiation, the They were subsequently combined in a similar
electrical length and relative phase of both combiner at Greenbelt prior to formal data
transmission lines, from the antenna output reduction.
terminals to the polarization selector box, were
Command Transmitting System
made precisely equal by means of a line
stretcher. All stations referred to previously had com-
Receiving and commanding antenna re- mand capability for the OSO I. The Collins
quirements were as follows: 242G-2 VHF transmitter was used for interro-
Telemetering: gation purposes at these ground stations. The
Center frequency: 136.5 Mc transmitter was crystal-controlled and ampli-
Frequency range: 135-138 Mc tude modulated. It was capable of operation
Gain: 19.2 db above isotropic over a frequency range from 108 to 152 Mc.
Polarization: horizontal, vertical, right and Special coders were used to supply the modu-
left circular lating signal necessary to actuate the space-
"E" plane pattern: 19.5 ° at half-power craft command functions. An output power
points of 200 watts was available. A command timer
"H" plane pattern: 19.5 ° at half-power was used which turned the transmitter on for a
points predetermined length of time (2 seconds) after
VSWR: under 1.1 at 136.5 Mc under 2.0 the switch was thrown and permitted trans-
from 135-139 Mc mission of the selected command for the timed
Nominal impedance: 50 ohms interval.
IINASA Satellite Instrumentation Network Facilities The transmitter operated into a load con-
Report, pages 15 and 16 nected through a 52-ohm transmission line
168
DATA ACQUISITION AND COMMAND SYSTEM
GUARD
_OSCILLATOR
_- .........
DRIFT GUARD BAND +
2_
i%
MODULATION
TOTAL
DEVIATION
ALLOWABLE
15%
sE_rrlvITY+6%
I
SUBASSEMBLY DESIGN AND
QUALIFICATION
B* _ 18-22
R_
VI_ ,
TO
XMTRS
7. Signal overvoltage protection: the input
sensitivity characteristic limited the devia-
tion to 4. 7_ percent.
955 CPS 27O MY SIGNAL _C_ANNEL 2
percent.
FIGURE 5-4.--Subcarrier oscillator subassembly block 9. Linearity: 3 percent of design bandwidth
diagram. from the best straight line.
10. Long term drift: see Table 5-2.
Design Specifications
Reference oscillator--less than 4-0.25 per-
The following design specifications provided cent of the reference frequency.
oscillators which operated properly within Subcarrier oscillator--less than 4- 0.75 per-
channel bandwidths for a period of six months cent of their center frequency.
during qualification testing and nine months in 11. Output signal: 270 mv rms.
orbit. 12. Temperature: --20°C to -+-55°C.
1. Power requirements per oscillator: 24 13. Weight and size: 2_ oz.; 1.88 cubic
milliwatts at 6 volts. inches.
169
ORBITING SOLAR OBSERVATORY
Table 5-1
Long term
Playback IRIG
Channel No. -0% _o (cps) Bandwidth drift limits Intell. required of each frequency channels
channel
+6% (cps) +cps (fc) X18.35 (playback) re
(cps)
Table 5-2
164 cps 212 cps 294 cps S.N. 400 cps 572 cps 790 cps 955 cps 1200 cps
S.N. 1580 S.N. 1547 1327 and 1324 S.N. 1310 S.N. 1287 S.N. 1280 S.N. 1284 S.N, 1299
Test time
Date
!(days)
*Oscillators readjusted. [] Adjusted for 0-n input signal, later changed to 0-5 vdc.
170
DATA ACQUISITION AND COMMAND SYSTEM
vortage per unit bandwidth at the output of the Due to the nonlinearity of the transmitter
FM carrier demodulator is directly proportional deviation sensitivity curve (Figure 5-6) the
to the modulating frequency, N _f, and in- percentages given did not indicate the relative
creases with frequency at a constant rate of amplitudes. The amplitude providing a pre-
6 db per octave; and (2), the subcarrier band- scribed deviation is a function of the transmitter
width in an FM-FM system is a fixed percent- sensitivity at the deviating frequency.
age of the subcarrier frequency and will, On the OSO I, a filter was used to obtain the
therefore, increase as the subcarrier frequency relative amplitudes. A master gain control was
increases (BW=f). Since noise voltage in- then used to vary the amplitude of the composite
creases as the square root of the bandwidth, signal to control the deviation of the transmitter
N _ (BW) _/_, and bandwidth is proportional to within the restraints of bandwidth.
frequency, then there is a second increase in It was determined chiefly by empirical
the noise voltage, N =fl/L methods that the multiplex described could be
The total noise introduced into the input of contained in a bandwidth of 100 kc with
the subcarrier discriminator is then the product negligible loss of sidebands during peak devia-
of these two sources. N=f.P/_ of N=P/2. tion. In adjusting deviation to conform to
Therefore in order to maintain a constant bandwidth restrictions, doppler shift as well as
signal-to-noise ratio over the multiplex spec- transmitter and receiver drift were considered.
trum, the deviation of the transmitter per There was no indication of drift in the sub-
subcarrier signal had to follow the f_/_ taper. carrier oscillators during the first nine months of
The first steps in pre-emphasis design were orbital life. The nonlinearity of the alternate
accomplished using theoretical data, but the 212-cps oscillator was increased beyond the
final adjustment was a matter of trial and error. design specifications during this period. No
Table 5-3 shows the ideal pre-emphasis. other changes were noted during this time. A
Figure 5-2 shows how the ideal had to be copy of the data obtained from the spacecraft
modified to keep the low frequencies above the in orbit is shown in Figure 5-7.
transmitter background.
Tape Recorder Subassembly Design
Table 5-3
Design Specifications
Ideal Pre-Emphasis Schedule.for the OSO I
Since the spacecraft was beyond the range of
Transmission System
the receiving station for 90 minutes of its 95-
minute orbit, it was necessary to store data.
f f3/, Percentage of total
deviation The multiplexed subcarrier oscillator tones plus
a fixed reference tone were recorded on tape for
2,300 cps ......... 110, 000 1.11 90 minutes. The tape recorder was a single-
3,000 cps ......... 165, 000 1.66 reel, continuous-loop device that recorded con-
3,900 cps ......... 290, 000 2. 92
tinuously until commanded to play back.
5,400 eps ......... 400, 000 4. O3
7,350 cps ......... 630, 000 6. 35 During the playback interval of 5 minutes, the
10,500 cps ........ 1,080, 000 10.9 entire tape loop passed over the playback head.
14,500 cps ........ 1,750, 000 17. 6 The recorder then reverted to the record state
17,000 cps ........ 2, 200, 000 22. 2 automatically. The record time to playback
22,000 cps ........ 2, 300, 000 33. 2
time ratio of the tape recorder caused the
recorded tones to be transmitted at the fre-
9925X 10_ 99.97
quencies indicated in Table 5-1. The fre-
quencies were standard IRIG FM-FM telem-
The next step in pre-emphasis design was to etry frequencies. The reference tone, 17.5
adjust the amplitudes of the subcarriers to kc, was used for airborne tape-speed compensa-
permit the percentage deviations indicated. tion.
171
782-552 O--66-----12
ORBITING SOLAR OBSERVATORY
I00
EM = 3.0V RMS
:E M =8.0VRMS
:E M = I. 5V RMS
:MI =8 E M=0.98VRMS
MI =6 E M=0.74VRMS
MI =4 EM =0.50VRMS
MI = 2 E M = 0.249V RMS
MI=I E M=0.12VRMS
.I 10 100
MODULATION FREQUENCY (KC)
The following design specifications provided 7. Wow and flutter: less than 1_ percent
tape recorder storage which met all systems' peak-to-peak and less than 300 cps.
requirements adequately. 8. Input impedance: 20K.
1. Power requirements at 19 volts dc: 9. Input signals:
Record operation: 1.03 watt (90 minutes) Channel 1: Input consisted of 4-frequency-
Playback operation: 1.4 watts (5 minutes) multiplexed subcarrier oscillators. Center
Average orbit power: ].,05 watts (95 min- frequencies are 164 cps, 212 cps, 294 cps,
utes) and 400 cps.
2. Record-to-playback ratio: 18.35±2 per- Channel 2: Input consisted of 4 frequency-
cent. multiplexed subcarrier oscillators. Center
3. Tape Speed: frequencies are 572 cps, 790 cps, 955 cps,
Record speed: 0.65 ips. and 1200 cps.
Playback speed: 11.9 ips.
The signal level for both channels was 540
4. Tape transport : continuous loop, single reel.
millivolts rms.
5. Motor: the motor used in the tape recorder
10. Erase head: a permanent magnet.
started as a dc motor. Centrifugal force
11. Record method: signal superimposed on
removed the starting brushes and the motor
ran as a synchronous machine from an a bias oscillator.
inverter power supply. 12. Output signal both channels: 1.07 volts
6. Tape capacity: 300 ft. rms. across a 5K load.
172
.* DATA ACQUISITION AND COMMAND SYSTEM
.t-
FIGURE5-7.-Data format.
173
ORBITING SOLAR OBSERVATORY *.
-
FIGURE
5-9.-Tape recorder motor, electronics, transport drive and reel.
tiid liibricnted by Alinnesotit Mining and During preliminary testing it was determined
A I t i n I i f tictiiring ( h r n pnny wits used. t h a t the Vector power amplifier was incom-
patible with the ESCO diplexer. T o correct
Transmitter Subassembly this problem, a lumped-cons t ant diplexer was
purchased from Vector. However, the power
The transmitter siibassenibly consisted o f amplifier continued to radiate spurious signals
two parts : the driver ttnd the power amplifier. unless the VSWR was precisely adjusted. T h e
Both units were initally obtained from the adjustments were considered t o be too critical
Vector Manufacturing Company. The driver,
Figure 5-10, was capable of delivering 250 milli-
wntts to the power amplifier. T h e power
iimplifier was capable of 1.75 watts output.
There were two modes of operation for the
transmitter subassembly, low power during 90
minutes of the orbit and high power during
commanded plttyback of the tape recorder. T h e
change to low power oiitpiit was ttccotnplished
by switching ti resist,or in series with t,he B +
line to the exciter. 'I'lie resulting reduced drive
to the power ttniplifier provided tin output o f
0.3 wittts. During high power trttnsmission--
Iast,ing five minutes of cilch orbit8---tlle B + t,o
t'he driver w i t s n o r i n i d t i n d t tie power Itrnplifiet.
ollt~pllt#
WIlS 1.75 WiLtt#S. FIGURE5-lO.-Transmitter subassembly driver.
171
.*
DATA ACQUISITION AND COMMAND SYSTEM
C6
LI c4
R F IN
>
-
FIOLTRE
.?-12.-OSO I power amplifier schematic.
175
t •
ORBITING SOLAR OBSERVATORY
e_
#¢ I.
,.ql, ?
£
;;_ x• _
_, P _ _
_ g
w
F-
- I I T I '
NO l l ' I I
L_f rb,'
7 ._
176
.* DATA ACQUISITION AND COMMAND SYSTEM
Modulation: AM
Input impedance: 50 ohms.
6 d b bandwidth: 35 kc.
Sensitivity: -dbm (75 percent mod.).
Dynamic range: 2 t o 2000 microvolts mini-
mum.
Image rejection: 80 db.
Spurious response rejection: 60 db.
Local oscillator radiation: 200 ppw max.
Local oscillator stability: f 2 kc.
I.F. stability: f 2 kc.
Audio amplifier response: f l d b 3 kc to 7kc.
Audio output: 50 milliwatts (5v rms across
500 ohms).
Standby power: 150 milliwatts (19 vdc
FIGURE5-14.-OS0 I receiver. source).
Power during interrogation: 250 milli-
confidential. Seven tones actuated seven watts.
channels. The receiver output was filtered by Input connector: SA4 male.
highly selective bandpass filters at the input
Decoders
to the decoders. When the presence of the
particular channel tone was detected, the relay The decoders were procured from AVCO
driving circuit of the decoders was biased into Corporation. They were operated in parallel
conduction and connected one end of the to the point of at,tachment to the relays (see
driven relay field to ground. This was a Figure 5-15). One decoder contained the
momentary connection during the presence of
the command tone.
Figure 5-13 illustrates the decoder. Its
operation is as follows. Each of the tone dis-
sector channels is connected to either one end of
a single relsy field (channels 1 and Zj, or t u a
pair of relay fields (channels 4 and 5), or in the
case of channels 6 and 7 to three relay fields.
Channels 1 and 2 are used to select the ON or
I DECODER
5183116
Receiver
The receiver was procured from the AVCO
Corporation. It was Model No. AD153114
(Figure 5-14) and had the following chnrscteris-
t,ics:
Frequency: 1 10-1 50 M c .
The timing is fised and is set t o a specified
frequency at the manufacturer's plant FIGURE
5-15.-Commartd system diagram
177
ORBITING SOLAR OBSERVATORY 0 .
v
stacked nrrcingement its shown in Figure 5-16.
HYBRID
* multiplexing system, isolation from transmitter A special fiberglass test fisture was used on
to receiver of 40 db, isolation from receiver the BBRC antenna tower. This fkt.ure was
t o receiver of 30 db, and an energy pattern essentially a modified gimbal which allowed the
within -6 d b of a unity gain antenna. Small recording of antenna patterns a t angular posi-
angle nulls were allowed. The entire antenna tion increments of IO degrees. The fixture and
and R F multiplexing system had t o weigh less tower used for antenna pattern measurements
than 3 lb. are shown in Figures 5-18 and 5-19.
Figure 5-20 is a pictorial view of OS0 I,
Antenna
inchiding the antenna elements. The spacing
Preliminary antenna design was accomplished between stubs \\-as approximately one-half
with a one-third scale model of the Orbiting wavelength. Each antenna element consisted
Solar Observatory. Antenna patterns were of a support bracket and a stub. The support
siibmitted for a two-dricen-element array and for brackets also supported the extendable arms of
a three-dricen-element array. The radiation the spacecraft. Each stub was electrically
characteristics of the two arrays were similar. connected to its respectire support bracket to
For simplicity and reliability, the tvio-driz)en- form a three-element array. The three ele-
element array was chosen. Impedance match- ments mere completely insulated from the space-
inp was then accomplished using a full-scale craft by the use of nylon and fiberglass epoxy
model of OSO. Principal-plane antenna pat- material. Two of the elements were excited
terns were recorded and submitted for approval. and the third was parasitic. The element exci-
The antenna patterns recorded on the BBRC tation was applied at the point where the sup-
antenna range are included in Appendix A . port bracket connected to the spacecraft arm.
These include 18 transmitter patterns and 3 The electrical length of each element was
principal-plane receiver patterns. approsimately one-half wavelength.
I t was decided that the spacecraft transmitter
should radiate during the launch for tracking
purposes. Since the arms were folded down
during this phase, the antenna configuration
was much different from that of the orbital
position and the resulting radiation pattern
w-as severely ciisiorted. This W T R S Rcrompanied
by a high 105s of radiated power. In spite of
this distortion and loss of power, adequate gain
was avnilahle for tracking and the mismatch of
impedances caused no harni to the R F power
amplifiers.
The antenna system was utilized for reception
:is n-ell as transmission with the two frequencies
differing by 12 percent. Impedance matching
was accomplished by single-stub matching
techniques. Single-stub matching is a narrow-
band method. but it was possible to obtain n
\-SWR of less than 1.5:l a t the tranwiitter
frequency and less than 2:l a t the receiver
frequency.
Diplexer
-4 helical reionat or bandpass diplexer wab
deqgnecl hJ- ES('0. During s\-\tem integra-
FICCRE5-1s.-.iiiteniia tou er. tion. it i w s found that the T'ector solid-statp
179
ORBITING SOLAR OBSERVATORY
180
.. DATA ACQUISITION AND COMMAND SYSTEM
EXCITATION
OPERATIONS
General
The worldwide Minitrack network was re-
sponsible for tracking and for receiving data
from the Orbiting Solar Observatory.** The
SATELLITE OS0 subsatellite plot is shown in the world map,
. Q
$2
FIGURE5-21.-OSO I dipleser.
181
ORBITING SOLAR OBSERVATORY
FIGURE
5-22.-OS0 I hybrid. FIGURE5-23.-OS0 I coaxial relay.
182
_o
DATA ACQUISITION AND COMMAND SYSTEM
o Telemetry recordings were required from the centimeter deflections. After conversion to
stations listed below: functions, the 12 channels were TWX'd to
Pr/mary Boulder daily.
Antofagasta, Chile The following functions indicated the status
Blossom Point, Maryland of the spacecraft on a quick-look basis:
Fort Myers, Florida Channel Parameter
Lima, Peru
2 Battery temperature.
Quito, Ecuador 5 Calibrate 0 volts dc.
Santiago, Chile 6 Calibrate 2.5 volts dc.
Secondary 7 Calibrate 5.0 volts dc.
Mojave, California 8 Spacecraft voltage monitor.
Woomera, Australia 34 Elevation positioning accuracy read-
Johannesburg, South Africa (early orbit out
only) 35 Azimuth positioning accuracy readout.
36 _-5.0 volts dc calibration.
NOTE: Secondary telemetry acquisition sta- 37 2.5 volts dc calibration.
tions were used only during early orbits or during 38 0 volts dc calibration.
those occasions where conflicts developed and 40 Azimuth positioning power monitor.
a primary station was unavailable to command 41 Elevation positioning power monitor.
and record the telemetered data.
There were two types of data transmitted by
Launch and Early Operation
the Orbiting Solar Observatory. These were
real-time and stored data. Stored data, re- The first station to command the Orbiting
corded in the spacecraft during one transit Solar Observatory to play back its recorded
around the earth, were played back and trans- data was the Fort Myers Minitrack station.
mitted upon command once per orbit by one of This station was provided with special tape-
the above stations. Recovery of real-time speed compensation and diversity-combining
telemetry data was not required at any of the equipment to provide a detailed quick-look
stations until both of the spacecraft recorders analysis of the spacecraft and experiment status
failed to function. immediately after launch. BBRC personnel
In order that the status of the spacecraft and the GSFC spacecraft controller were
could be evaluated on a current basis, a special present at this station during the first two
quick-look procedure was established for the weeks of orbit to monitor the spacecraft oper-
analysis of the "housekeeping" channel. ation and to advise a course of action in case of
During the first two weeks after launch, a spacecraft emergency.
Fort Myers was the primary station, and all The stations listed in Table 5-4 were able to
housekeeping data received there were also receive telemetry signals in the order listed and
reduced there. The outlying stations could in the manner indicated during the launch and
read only chart deflection of the selected func- early orbit period. They were equipped to
tions listed below and TWX this information make a Sanborn recording in real-time from
to Fort Myers. one antenna polarization and TWX PAM
After two weeks, GSFC at Greenbelt received deflections (in centimeters) to the Fort Myers
all quick-look housekeeping data in the form of station.
183
ORBITING SOLAR OBSERVATORY " •
TABLE 5-4.--0S0 I Readout and Command Schedule for the First 19 Hours After Launch
T+4hr. 59' ...... AGASTA ..... 12 54 667 414 Playback_ __ Yes .... REC
TTShr. 01' ...... SNTAGO .... 12 16 1468 912 None ...... NO ..........
T+5hr. 23' ...... JOBURG .... 12 61 614 382 None ...... No ...........
TT5hr. 39' ...... LIMAPU .... 09 11 1768 1098 None ...... NO ...........
T+ 5hr. 40'_ ..... AGASTA ..... 12 37 851 529 None ...... No ...........
T+ 6hr. 41' ...... SNTAGO .... 12 50 697 433 Playback_ __ Yes .... REC
T+7hr. 05' ...... JOBURG .... 10 13 1623 1008 None ...... No ...........
T+ 8hr. 22' ...... SNTAGO_ 12 86 551 342 Playback_ __ Yes .... REC
T+8hr. 23' ...... AGASTA ..... 11 22 1212 753 None ...... No ...........
T+ 10hr. 03' ..... SNTAGO .... 12 65 597 371 Playback_ __ Yes .... REC
I
T+ 10hr. 04' ..... AGASTA ..... 12 29 1013 629 None ...... No .... I .......
i
T+ llhr. 45' ..... AGASTA ..... 12 74 563 35O Playback_ __ Yes .... REC
T+llhr. 45' ..... SNTAGO .... 11 25 1114 692 None ...... No .... I .......
T+llhr. 46'_ .... LIMAPU .... 11 12 1653 1027 None ...... NO ...........
TT13hr. 27' ..... AGASTA ..... I0 17 1394 866 None ...... NO .... ' .......
T+13hr. 27' ..... LI MAPU .... 11 56 641 398 Playback_ _ Yes .... REC
T+13hr. 29' ..... QUITOE ..... I0 14 1548 962 None ...... No ...........
T+14hr. 33' ..... OOMERA .... 09 13 1608 999 None ...... No ...........
T+15hr. 08' ..... LIMAPU .... 11 17 1402 871 None_ NO ...........
TT15hr. 09' ..... QUITOE ..... 12 75 556 345 Playback_ __ Yes .... REC
T+16hr. 12' ..... OOMERA .... 12 49 709 441 None ...... No ...........
T+16hr. 54' ..... FTMYRS _ __ II 2O 1279 795 Playback_ __ Yes .... REC
TT17hr. 53' ..... OOMERA .... 13 69 587 365 None ..... No ...........
T+18hr. 34' ..... FTM YRS .... 12 73 574 357 Playback_ _ Yes .... REC
TT18hr. 37' ..... BPOINT ..... II 21 1257 781 None ..... Yes .... REC
The normal operations started approximately Data reduction consisted of editing and
two weeks after launch. All stations assigned analyzing 52 PAM housekeeping channels.
command and telemetry acquisition responsi-
bilities during this period were expected to Key Channels--Ouick-L.ook Reduction
record a malimunl amount of stored data during Certain of the 52 channels contained the key
the playback period for which they were as- performance information of the spacecraft.
signed responsibility. Each station, in ad- Each receiving station was required to forward
dition, wits req_fired to submit quick-h)ok centimeters deflection of these channels to the
information (Sanborn deflections in centi- Operations Control Center at GSFC. After
meters) to GSFC at Greenbelt where the TWX'ed the function status to Boulder on a
housekeeping channels were red,iced and conversion to the function, the Control Center
TWX'ed to Boulder for analysis. daily basis.
184
DATA ACQUISITION AND COMMAND SYSTEM
• The key channels are: Flexowriter typed out data sample numbers 1
Battery No. 2 temperature. through 90 down the page. The result of this
Spacecraft voltage monitor. tabulation then was "Function Magnitude ver-
Spin gas pressure. sus Orbit Position," since sample number and
Elevation readout. orbit position could be correlated.
Azimuth readout. By editing the oscillographic records, thirty-
Pitch readout. three of the fifty-two channels were selected
Ammuth power monitor. for computer processing in each orbit. The
Elevation power monitor. process was repeated three times with three
sebs of data and calibration tapes. The com-
Semi-Automatic Comprehensive Reduction of puter required 15 minutes to process one orbit.
Housekeeping Channels
OVERALL PERFORMANCE
Magnetic tapes supplied by GSFC were
General
used to generate paper strip charts. Sanborn
pen and Heiland Visicorder records were The FM-AM telemetry system of the space-
generated simultaneously. The Visicorder rec- craft performed well for over 1000 orbits. The
ords were used for housekeeping data reduction signals received from the spacecraft at the
and the Sanborn records were used to qualita- receiving station during tape recorder playback
tively determine the condition of experiments. were at least 20 db above threshold and free of
noise resulting in data of excellent quality.
Oscar K Readout A record of a data transmission received at
Oscillographic records were read by the Fort Myers, Florida, receiving station compared
Bensen Lehrner, Oscar K, oscillogram semi- favorably with one made in the laboratory
automatic reader. The output of the Oscar during spacecraft checkout. It was possible
to receive real-time data transmission from the
K reader was automatically recorded on an
electrotyper. Recorded data included time, spacecraft in Boulder, Colorado. The signal
channel number, frame number, and data. was of sufficient quality that information con-
The data were recorded as a decimal digit cerning the operation of the spacecraft and the
from 0.00 to 5.00 volts. These data were not in experiments was routinely being gathered at
a form for graphical presentation in that scalar Boulder.
mathematics had to be performed to convert The communications system continued to
voltages to quantities of temperature, pressure, give excellent performance through the tenth
angular degrees, etc. Week during which time 100 percent of the
An additional output of the Oscar K provided data recorded was usable. An average of 10
an octad format which was recorded on punched to 15 percen_ of the data could not be recorded
paper tape. due to spurious commands received by the
spacecraft as well as minor ground station
Computer ReducTion problems.
The CDC 160 computer with a Fortran pro- RF System
gram was used to process the octad format tape The RF systems (both primary and alternate)
from the Oscar K. Special computer program continued to operate satisfactorily. At the
tapes were prepared which contained co- time of this writing there had been no significant
efficients of calibration curves. The data tape change in transmitter frequencies or receiver
and the calibration tape were fed to the com_ signal-to-noise ratio since launch.
puter. The computer output tape was fed The slant rangett versus the available signal
to a Frieden Flexowriter. The Flexowriter level at the receiver input terminals is shown
typed out the magnitude of eleven channels in Figure 5-25. As can be seen, the combined
in eleven rows across the page. Ninety data ttTaken from NASA-GSFC Operations Plan 1-62,
samples per channel were taken per orbit. The Orbiting Solar Observatory (S-16).
185
"l
ORBITING SOLAR OBSERVATORY
O¢2 I ........ f
<
120 i i
--
7_ / _
./ --_1 _
-%4c2
.,
9O _.. L_ .I
<
>
8O i 1 I I I _ Ill II
<
E
> I i ,
200 300 400 ' 560 700 1000 2000 ao0o ;,ooo'5o'oo'ooo
SLANT RANGE- KILOMETERS
polarization curve lay within the data threshold to nearly zero during orbit 372. The low-
at 300°K and 5000°K up to a slant, range of frequency channel was not affected. The de-
2000 miles. crease in amplitude could have been caused by
l'rior to the faihire of the primary tape re- cations were that the basic recorder, tinier, and
corder, the high frequency channel amplitude motor continued to operate normally. Data
decreased fronl it normal level during orbit 364 obtained during orbits 3472 and 3489 indicated
185
#B
DATA ACQUISITION AND COMMAND SYSTEM
that the motor and driver were still operating have been caused by a modulated carrier or
properly. One of the following items appar- by two carriers heterodyning and generating the
ently failed: exact command tones.
1. Bias oscillator. GSFC transported a transmitter to the
2. Latching relay. particular region of North Africa over which the
3. 15-volt regulator. signals were received. The transmitter was
4. Mylar belt (tape drive). modulated with the ON command tone. This
modulation desensitized the receiver and de-
Command System coder to any OFF commands sent by the source
Soon after the launch of the spacecraft, it of interference. In this way, the sequence of
became apparent that spurious commands were ON-OFF commanding which was disastrous to
being received when the spacecraft was between the primary tape recorder was eliminated.
20 ° and 30 ° east longitude and between 15 ° and No problems other than spurious command-
35 ° latitude (see Figure 5-24). The specific ing were apparent with-the command system.
source of the spurious command signals was not All commands were received properly by the
determined. The spurious commanding could spacecraft.
187
782-552 0--(16-----13
Chapter 6
POWER SUPPLY
INTRODUCTION flective coating on one side and a red antire-
flective coating on the other. The purpose of
Electrical power for the spacecraft was con- these coatings was to transmit the light to which
verted from sunlight by a sun-oriented array of the cell was sensitive (to increase power output)
silicon solar cells. Part of this power was used and reject the light to which it was insensitive
to charge a storage battery which stabilized the (to decrease temperature).
voltage, supplied peak demands, and provided The cells were soldered together into modules
power during the dark part of the orbit. The of four and five cells each, and the modules were
solar-cell array and battery constituted the interwired to form strings of 52 cells in series.
power supply. There were 36 such strings.
Solar-cell modules occasionally fail by open-
POWER AND TEMPERATURE ESTIMATES circuiting, but rarely by shortcircuiting. If
modules are connected in simple series strings, a
It was estimated that at the nominal voltage failure of one module takes an entire string out
of 19 volts, the power requirements would be: of the circuit. However, if parallel crosswiring
Day -,_gh_
is provided between series strings, the failure of
(watts) (watts) a single module is much less serious. The
Control system .................... 6. 6 2. 8 current that should flow through that module is
Communications system ............ 4. 5 4. 5
carried by those in parallel with it, in addition
Experimental instruments ........... 6. 5 4. 3
to their own normal current. In this particular
Total ....................... 17. 6 11.6 array, if 6 or more series strings were cross-
connected the failure of a module would have
The thermal analysis (Chapter 7) predict a cost only the amount of power normally con-
solar-cell array operating temperature of about tributed by that module.
70 ° C and a battery temperature of about 10 ° C. If all the series strings had been crosscon-
nected, then a short circuit across part of one
SOLAR CELL ARRAY string (caused, say, by meteorite damage) would
have wiped out the entire array. Therefore,
The solar-cell array had to provide 17.6 watts this array was divided into five sections, three
day power, and also had to provide enough containing 6 series strings and two containing 9
battery charge to support the l l.6-watt night series strings. The strings within each section
load. Assuming that the battery efficiency was were crosswired, and each section had its own
75 percent on an ampere-hour basis in a 76 per- blocking diodes.
cent sunlight orbit, the solar-cell array had to Current-voltage curves for the completed
deliver 22.5 watts. In a 63 percent sunlight array are shown in Figure 6-1. Curve A is an
orbit the solar-cell array had to deliver 26.7 experimental curve taken in sunlight in Boulder,
watts. Colorado. Curve C is this experimental curve
The array comprised 1872 silicon solar cells, corrected to the expected orbital conditions.
Hoffman Type 120CG. Each cell had a 0.006- Curve C shows that the array will deliver 29.4
inch-thick cover glass with an ultraviolet re- watts at 19 volts at 70 ° C in space sunlight.
189
ORBITING SOLAR OBSERVATORY ""
CURVE A: EXPERIMENTAL CURVE TAKEN IN BOULDER SUNLIGHT amperes with a maximum voltage of 20.7 voltg.
OF 103 MW CM -2, WITH PANELS AT 5_°C.
CURVE B: CURVE A ADJUSTED FOR 140 MW CM- OF JOHNSON In normal operation, the battery was cycled
LIGHT, PANELS AT 52°C.
CURVE C: CURVE A ADJUSTED FOR 140 MW CM -2 OF JOHNSON to a discharge depth of about 2 percent. It
LIGHT, PANELS AT 70°C.
1.6
completely, the battery alone could have
powered the spacecraft for about 10 orbits.
_1.4
190
"" POWER SUPPLY
• The day power switch, as described in detail produce 29.4 watts, and for a maximum sun-
in Chapter 4, was provided to remove power light, 26.4 watts.
from those circuits and experiments not re- The array temperature in orbit was 60°C,
quired at night. 10°C lower than predicted. This reduction in
The voltage regulators, also described in operating temperature allowed the array to
Chapter 4, were provided for those spacecraft produce about four percent more current than
functions critical to supply voltage variations predicted so that the estimated initial orbital
such as the servo amplifiers and most of the output of the array was 30.6 watts. Because
spacecraft instrumentation circuitry. of this, the minimum sunlight orbit power
Fuses were installed in series with the power margin was about 4 percent and the maximum
lines to each of the experiment and spacecraft sunlight orbit margin 16 percent.
systems which could fail without destroying the The reason for the narrow margins was that
more control system drive power was zequired
usefulness of the remaining systems. The
than was originally estimated. The anticipated
fusing was set at a level such that a drastic drive was 2.5 watts and the actual drive was
overcurrent demand was required to remove
5.5 watts. Had the anticipated drive values
the circuit from the supply. The fuses were and array temperature been realized, the mini-
derated for vacuum use based upon actual mum sunlight orbit power margin would have
vacuum test data. been 11 percent and the maximum sunlight
The pointed and wheel experiments could orbit power margin 27 percent.
have been removed from the power lines by
command should the power supply capability Long Term Orbital Results
have been reduced, or if a malfunction had
Voltage" measurements made during the first
occurred that was insufficient to blow the fuse. four months of orbital life indicated that the
power system was performing as expected and
POWER SUPPLY PERFORMANCE that the array output was sufficient to maintain
the entire observatory load.
Preflight and Initial Orbital Resuhs
Measurements of battery voltage were made
The actual power usage for the spacecraft, during orbits 3500 to 3700 about four months
after the 9 July 1962 high-altitude nuclear
using the last preflight checkout data and the
device explosion which caused considerable elec-
data measured in orbit, was as follows:
tron flux intensification at the 300-nautical-
mile altitude of the spacecraft. It was deter-
Day Night
(watts) (watts) mined at that time that the spacecraft array
could not supply the required 23 watts day
Control electronics ........... 6.2 1.2 power. However, the array was able to charge
Control drive ................ *5. 5 0
the batteries with the 3A-watt wheel experiment
Communications ............ 4.1 4.1
Experiments ................ 7.2 3.4 load commanded off. If the following assump-
tions are made it can be concluded that the
Total .................. 23.0 8.7 array was producing about 21.3 watts, and thus
that the array had degraded about 30 percent
*3.5 watts azimuth, 2 watts elevation drive.
in the four-month period that the radiation had
Test results from battery charge cycling indi- existed.
cated an 80 percent charge efficiency. Thus, Array temperature was still 60°C.
the array was actually required to provide 23 The loads were the same as when the space-
watts day power and enough additional charge craft was launched.
to maintain the 8.7-watt night load. For a There was no micrometeorite damage to the
minimum sunlight orbit, the array had to panel.
191
• 6
The original 30.6-watt array output estimate Solution of the Degradation Problem
was accurate.
In future spacecraft of the same type and
Half of the wheel experiment load could have mission a more satisfactory answer to the
been handled. problem of degradation would be one of the
There were other considerations relating to following:
the degradation determination outlined above. 1. N on P silicon cells can be used to derive
One of these was that there was some prob- about the same space sun output as the OSO I
ability that, due to brush material buildup on array a.nd should have a degradation figure low
the commutator of the azimuth torque motor. enough (about 11 percent) to be useful for a
the torque gain of the motor may have been six-month lifetime. This solution will un-
reduced. This could have had an appreciable doubtedly be the one used.
effect on degradation, but it was estimated that 2. The other approach would be to use higher
it caused no greater than a 20 percent increase efficiency P on N cells, to produce as much
(0.7 watts) in required azimuth pointing power. power as is practical for the given area, antici-
A further consideration was that the pointing pate the degradation, and plan to selectively
power varied over wide limits due to the spring command experiments off when the array can
effect of the elevation gimbal flexible leads no longer handle the full load.
against which the elevation torque motor had With no increase in panel area being con-
to drive. This effect caused a variation in the sidered, the first solution has an experiment
elevation drive power from 0 to 6 watts with power limit of about 9 watts (averaged over the
the average being the 2 watts assumed in the orbit) assuming the present basic power require-
previous discussion. The 30 percent degrada- ments exist. If this experiment power is ex-
tion figure arrived at above then could have ceeded, there will be no choice other than to
been as low as 15 percent or as high as 36 increase the size of the array or to reduce the
percent; but 30 percent is the most probable six-month life expectation for full-time experi-
figure. ment utilization.
192
Chapter 7
THERMAL CONTROL
INTRODUCTION maintain the temperature of equipment located
in this section within the specified temperature
In the Orbiting Solar Observatory it was range for all orbital conditions.
important that temperatures be maintained 3. Temperature profiles around the orbit for
within specific limits in those sections of the the wheel interior and exterior cover panels
spacecraft containing temperature sensitive when the surface radiative properties previously
equipment. determined, are assumed.
The efficiency of the silicon solar cells pro- An exact mathematical analysis of the actual
viding electrical power for the spacecraft de- physical problem was not possible because of
creases as temperature increases. For this the complex geometry of the two spacecraft
reason the nonrotating structure was designed sections under consideration. Simplified models
to operate at the lowest possible temperature. of these geometries had to be assumed to
The nickel-cadmium storage batteries located provide a feasible solution method, and other
in the wheel section operated at good efficiency simplifiying assumptions were made in the
over the range 35°C to --10°C. Other com- analysis that was performed.
ponents and equipment located in this section This chapter presents the methods and
were temperature limited, but all were capable results of the mathematical analysis and the
of operating satisfactorily in the range from accompanying computer program conducted
--10°C to 35°C. Therefore, this temperature in support of the thermal design program for
range was specified for all wheel experiments OSO I. Derivations of the heat transfer
and equipment. equations for determining spacecraft tempera-
Many of the thermal design specifications tures are given in detail, including evaluation
established for the OSO were based on funda- of the shape factor integrals associated with
mental heat transfer considerations and did the heat exchange rates between the various
not require mathematical analysis. For ex- spacecraft outer surfaces and the earth.
ample, by painting all interior wheel surfaces The simplified analysis of the wheel was
with "thermally black" paint and providing run with an Intercom 500 computer program
good conduction between adjoining structural written for a Bendix G-15 computer. The
members, temperature gradients in the wheel analyses of the solar-cell array and the shielded
were minimized. This procedure tended to model of the wheel were run from Fortran
prevent local "hot spots" from developing in programs written for the CDC.1604 computer.
components such as the storage batteries and
electrical equipment where heat was generated. ANALYSIS OF SIMPLIFIED MATHEMATICAL
Mathematical analysis was performed in MODEL
support of the overall thermal design problem
to provide the following information: In the first solution of the problem, the
1. Expected operating temperature of the rotating and nonrotating sections of the space-
solar-cell array for various orbital conditions. craft were assumed to be isolated, isothermal
2. Required values of absorptivity and emis- masses, but with the same outside geometry and
sivity for the outer surfaces of the wheel to thermal capacity as that of the actual space-
ORBITING
SOLAR
OBSERVATORY t
craft. The presence of the armsandspin-up The radiative power received by the body
gas bottles on the wheelstructure and the from the earth consists of reflected sunlight and
pointedinstrumentsin the upper structure thermal radiation. These heat input rates are
was neglected.The heat transfer rate be- given by
tween the two sectionswas consideredto
benegligiblysmall.
This simplified model of the spacecraft
qe_ __ =F_Ae_,_oreEo (above sunlit earth) }
=0 (above dark earth)
provideda fairly accuraterepresentation
of the
physicalproblemsincegoodconductionpaths (7.3)
between adjoining structural membersare
and
providedby filling all rivetedor boltedjoints
with high-conductionepoxy resin, and all qe_---- F_Ae_ae T$ (7.4)
interior surfacesin the wheel are painted where
"thermallyblack." Thesestepsprovidemaxi-
mumheatdistributionwithin thetwo structures F,=geometric shape factor for reflected
and prevent the occurrenceof significantly sunlight
largetemperat_re gradients. Fb=geometric shape factor for thermal
radiation from the earth
Derivation of Heat Transfer Equations qe =heat input rate due to reflected sun-
light
The rate of change in telnperature of a
thermally isolated section qeb=heat input rate due to thermal radi-
of the spacecraft ation from the earth
is:
Ae_ =surface area of the body receiving
reflected sunlight
dTe,_q_--qo
dt mc (7.1) Ae%=surface area of the body receiving
thermal radiation from the earth
where re _ coefficient for diffuse reflection from
the earth (albedo)
Te-_=temperature of tile spacecraft ae=coefficient of absorptivity of the
t=time
body for receipt of thermal radi-
q_=radiative power absorbed by the ation at 250°K
spacecraft Te=the effective black-body tempera-
qo=power radiated from the spacecraft ture of earth_250°K
mc=the total heat capacity of the space- z-_the Stefan-Boltzmann constant,
craft
0.1714 X 10- s BTU/hr/ft_/o R4
go=AreeaT 4 (7.5)
f=A_agEo (in sunlight) where
qo_ (7.2)
L =0 (in the earth's shadow) At= total radiating surface area of body
where
_e_=coefficient of emission for thermal radi-
ation from the surface areas
A,=the projected area of the body in the
direction from the sun
The internal heat generated in the wheel
aG=the coefficient of absorptivity of the section of the spacecraft is assumed to be:
body for solar radiation
Eo=emissive power of the sun, 442 qe_=8.5 watts (in sun); 8.5 watts (in dark)
BTU/hr-ft? (7.6)
194
•" THERMAL CONTROL
where
Substituting Equation (7.9) into Equation (7.8), Oo=sin-X (h_o)=sin-X (0.9199) =66°55'
we have
ro----htan 00= (345) (2.3464) =809.5 statute miles
A¢-,, cos _, cos ¢_dS
dq¢a=aereEe -- x cos _ p2 The angle 8 in Equation (7.10) is constant
over the entire surface of the disk at any given
or
time and is equal to its value for the element
qe = F_Ac-, aor e Ee (7.3) dS directly below the spacecraft at that time.
With the above assumptions, we have the
where
following relationships:
To simplify integration of Equation (7.10), the ["Fb cos _ for cos _ > 0
following assumptions are made:
/_'= t 0 for cos 8<0
The earth is a fiat disk of radius, r0, with the
spacecraft a distance, h, above the center of the Thus, in determining the factor, F_A¢-,, in
disk, where Equation (7.3) for any particular surface of the
195
ORBITING SOLAR OBSERVATORY " °
n /_. TO dS . SUN
F__I [" cos ¢1 cos ¢_dS (7.12) n='j sin _--ff cos
_" ,)8 p2
196
•° THERMAL CONTROL
Likewise
PLATE
p2=h2 sec 2 0
Case 1 (0 <13"]<_/2--So):
Case2 13"1
<_/2):
1 2 1
FIGURE 7-2.--Plate orientation. F--_ sin 00 cos 3"d-_ sin 2 3"
1
1. The plate surface "sees" the entire disk COSO0_/sin 20o--COS 2 7
_F
representing the earth (0 _ 13"]<T/2--0o).
2. The plate surface "sees" more than half,
but not all, of the disk (_/2--0o_<13'1<_/2). _1T sin2 3"tan-' \_/sin 2COS0o
00_cos 2 3")
3. The plate surface "sees" a small part (less
than half) of the disk 0r/2 _ 13"1<_/2+0o).
+1 cos2 _' tan-'(Icos 3"l_/tan 2 3"tan 2 00--1)
4. The plate surface "sees" none of the disk T
(_/2 + Oo_<I",'I-<_)-
+1 cos _, sin 2 00 sin -_ (cot 3"cot 0o) (7.16)
The boundary between the regions of the disk
is given by
Case 3 (,r/2_< 1,1<¢/2+0o):
u-n=sin O sin _ sin 3,-4-cos O cos 3,-=-0
or Same expression as Case 2.
_=--Sin -1 (cot 0 cot 3")
Case 4 (lr/2+0o< 13"1
<¢):
The direction cosine terms in Equation (7.12) F=0 (7.17)
are
These functions can be considered for
COS _bl_ cos O
0 _<_ < _r only, since 3" will be determined by a
cos ¢_ = n- U----sin 0 sin _ sin 3"+ cos 0 cos 3" cosine expression and the choice of quadrant
for 3" is thus partly arbitrary. A plot of F
O and _ are used as the variables of integration versus 3" for a flat surface is shown in Figure
over the disk. Since
7-3.
r:h tan O Shown in Figure 7-3 is a graph of the
and function:
dr=h sec 2 ado
0.846 cos" (0.63453") cos (0.63453")>_ 0
dS becomes
197
ORBITING SOLAR OBSERVATORY °°
1.0_
so that the areas under the two curves are equal:
expression for F and is sufficiently accurate. The development of this expression is given on
The constants in the approximation were chosen page 211.
1.0
0.8
V + 0.514
0,6'
F, TOP
F, BOTTfi
F, COMPOSITE
0,4-
0.2
0.0
0o 40 ° ' 8 l0o ' ' °
120 160 °
y, DEGREES
198
" " THEKMAL CONTROL
dT dT dn dT . 1
--_=-_ "-_= d-_ "'=--_c j ('Y' 8, T)
or
dT__ 1
dn _/(% 8, T) (7.23)
199
ORBITING SOLAR OBSERVATORY
dTe-__ 1
• { qo+cos flrcEo[Ae-%aot (0.514+0.332 cos 2 _) +A ¢-%a_,(0.254+0.077 sin 2 _) ]
dT1 mc_
-4-qe,--(2A$_,_e_t-+-Ae_.:¢__),)aT$_ } (7.24)
dTe-__ 1
d_ mc¢7 [(A_KFB:_e_K+AF_FF_eF_)(TT_+(ABKFs_:aOsK
0.846 cos 2 [0.6345 cos -_ (sin 77cos e)], In Equations (7.26), n_, and roUt are given in
if cos [ ] > 0 degrees from an initial position n=0 °, and
FBK _ f
0 if cos [ ]<0
R=radius of the earth
h----orbital altitude of the spacecraft
cos B= f sin n cos _ if sin n cos _>__0
Finite Difference Form
I. 0 if sin n cos _<0
Euler's Modified Method was used for the
and numerical integration of the differential equa-
tion. This method is ordinarily not satisfactory
fA_eEe for n<_n_, or n>__no., for extending a solution very far from a starting
qe= point because of the error build-up, but in this
]
kO for no.t>n>n,. case we wished only to carry the solution around
a fixed orbit repeatedly until the temperature
The expressions for determining the position cycle became stable. The speed of this method,
angles of the satellite when it goes into and which was a prime consideration, can hardly be
emerges from the earth's shadow are, respec- exceeded.
tively, The method consists of finding a "zeroth"
approximation for T_+_, namely, T_+_,
0 by the
use of the relation
n_"=57'296 [_r+tan-_ _ / (RI--(R/R+h)_/R
-4-h) _-- sin _ eJ
T (°) --T-I- hv
'*'-- ''mc;_ J(')'" B,, T,) (7.28)
_o,_=57.296 I2_r--tan -_ (-R/R+h)2--sin_ i]
This is simply a linear extrapolation of the
(7.26) solution from the point (w, TO, using the slope
for
of the solution curve at that point. Succes-
[(R/R +h)'_--sin 2 e]>O, sively better approximations to (_+_, T_+_) are
and then found using the iteration formula
"qi.= %,t:270 ° (7.27)
for "r(,+,, _ T_t_ ___v. j(3,,, B,, T,)+.[('_,+,, 0,+_, _,+,:T(")
_+_ -- _mc_l
[(R/R+h)_--sin _ _] < 0 (7.29)
2OO
•" THERMAL CONTROL
_I'his is actually an extrapolation process from determine their effect on spacecraft tempera-
(T_, T0 using the average slope between the tures:
points
(]tO t, Oq_ b, OtO s
(T,, Td and (n,+l, n,(,)
_i+l]-
t_ t, t_O b, O/a s
;7= 3.7895°/min 210.00O 458. 550 .000 4.142 13. 336 -13. 084
225. 000 458. 560 .00O 3. 905 13. 099 -13.085
R= 3960 statute miles 240. 000 458. 270 .00O 3. 667 4.14'/ -I 3. 051
255.000 457.690 .000 3.493 3.973 -12,986
The following parameters were changed in FIGURE 7-6.--Example of G-15 printout format-first
value at various times during the program to computer program.
201
ORBITING SOLAR OBSERVATORY
202
THERMAL CONTROL
Table 7-I
I
Maximum Maximum Minimum T $___ Maximum T __#
Run No. (deg) _"(deO in _) _out(deg) qalb qbb
(BTU/min (BTU/min)
(°R) (°K) (°R) (°K)
203
782-552 O--66------14
ORBITING ,SOLAR OBSERVATORY ° "
Table 7-2.--Summary oJ Results oJ Computer Runs Top and Bottom Surfaces Partially Unpainted"
the three previous assumptions are revised to through the sides, particularly from direct sun-
approximate more closely the actual situation, light, should more than offset the additional loss
this configuration can be modified to stay well of heat through these surfaces. As can be seen
within the specified temperature range of from Run 1 of Table 7-3, this reasoning proved
-- 10°C to 35°C regardless of attitude or orbital to be incorrect. The computer solution shows
inclination. that the converse is true, and that the minimum
Alternate Approach to Raising Mean Temper- temperature drops below --30°C.
ature.--The proposed solution of leaving part of Effect of Various Assumptions on Mean
the top and bottom surfaces unpainted is not Temperature--Runs 2 ttu'ough 6 of Table 7-3
very desirable from the ground handling stand- were made to determine the individual effect
point. Extreme care must be exercised during on temperature of various assumptions made in
ground operations to assure that these surfaces previous runs. Albedo values of 0.3, 0.4, and
are not scratched or exposed to grease and other 0.5 were assumed for different runs keeping
substances that might leave a film on the sur- other parameters constant. Results showed
face. A slight Inodification in snrface charac- that a change of 0.1 in the assumed albedo value
teristics of highly polished metals can substan- produced only about a 5°C change in spacecraft
tially alter their radiative properties. temperatttre.
As a possible alternate method of raising The assumption cos O=l (Run 5) produced
spacecraft tenlperature ahove 00( 4, the scheme a substantial temperature change of 14°C. This
of painting top and bt)ttonl surfaces with result points out the importance of further
ahuninunl paint and the sides with thermally investigation into the nature of reflected sun-
black paint was analyzed tm the computer. It light from the earth. This heat source is
was reasoned that the increased heat input significant for low-orbit spacecrafts, and until
2o4
° THERMAL
CONTROL
Table7-3.--Summaryof Results oj Computer Runs Sides Painted Black and Various Other
Assumptions
Max.
a_ t aO. Maximum [ Maximum [ Max_num I Min.
nN_. re a(Dt • _. q_b (BTU/_ q_b (BTU/[ qi_ (BTU/ ! TO Remarks
e_ t rain) rain) min) (°K) (K)
data on its nature and intensity are made were fixed, and the heat transfer computer study
available from present spacecraft flights, ac- was concerned only with determining the ex-
curate temperature predictions for future space- pected operating temperature of the cells for
crafts will be difficult to make. various orbital conditions.
Runs 7 and 8 of Table 7-3 show the tempera- Computer runs were made for the seven dis-
ture fluctuations in a homogenous shell during crete values of the orbital inclination angle,
one orbit. Due to its low thermal capacity the _, listed previously in this report. Variations
shell temperature varied as much as 32 °C from in spacecraft attitude angle, _-, were unnecessary
minimum to maximum. However, the mean because, if shadowing by the wheel is neglected,
temperature of the shell was about the same as the upper structure is exposed to the same heat
that of the entire wheel for similar surface input rates from the sun and earth regardless of
characteristics. Thus, any particular isolated attitude.
mass inside the wheel will probably fuctuate The following fixed values of parameters were
more significantly than the :k2°C indicated for assumed for this study.
the total mass, but its mean temperature can
AFR=A_r:=4.52 ft _
be expected to be about the same as that of the
O£(_FR _ f(_--_FR : 0.70
total mass.
aCBK _ _--*SK-----0.85
inertia, the cells would heat up rapidly when to the front and back sides of the structure,
sunlight. This objective was acheived by surfaces, at each of the 72 points around the
painting the back surface of the solar cell panels orbit.
and all support structure with a special metallic The temperature data for the minimum sun-
oxide paint. Therefore, the surface radiative light orbit (}=0 °) and maximum sunlight orbit
parameters for tlds section of the spacecraft (}=56.78 °) are plotted in Figure 7-8. Tern-
2o5
o
ORBITING ,SOLAR OBSERVATORY
INTERIOR MASS
8O
6O
?
4O
__ 20
_-V 2
N 0 -Q
M
-20 "
perature extremes for the five other orbits were FIGURE 7-9.--Shielded model of the spacecraft wheel
section.
bracketed by the two cases plotted. These
curves indicated that solar-cell temperature
is, the shape factor between the panels and the
remained in the range from --20°C to 75°C for
interior mass and the emissivity of all interior
all orbits, and averaged about 55°C in sunlight.
surfaces were unity.
2O6
_o THERMAL CONTROL
-the net heat exchange between this mass and the adjacent surface of the interior
three cover panels. These net heat exchange mass in this model)
rates are given by the general expression: F12=shape factor between the two
surfaces
dt (me)4
dT_ A1
-d-_ =i_( mc_ [F2_ (cos _ae_rcEo +act_ T_) -- (l + _ ¢-_ )a T_ q-_ T_,]
(7.36)
dT_ 1 4 4 4__ _ T _ 4
d,_ i_(mc)4 [q¢-)+A,a(T,--T,)+A,a(T_ Tw)+A_a( _--T_)]
2o7
ORBITING
SOLAR
OBSERVATORY °•
where The values listed for the following parametexs-
were assumed constant for this program:
A,=projected area of the side panel in
the direction of the solar vector (mc)l= (mc)2= 1.21 BTU/°R
(mc)3----2.82 BTU/°R
(mc)4= 55.96 BTU/°R
ae_=0.32
X----fractional part of top and bottom ae_=ee_:O.343
surface areas of condition 1 (shiny) ae. :0.25
I-),----fractional part of the same areas of
aot_=0.32
condition 2 (painted) and the other
symbols the same as previously
defined. a_t2=ee-%----0.343
A_/=5 °
The external shape factors, FIE, /_E, and
_hg, in Equations (7.36) are derived in the In addition, the constant values for A, A_, A_,
same manner as was used for the simplified Ee, _, q_, _, T_, R, and h listed on page 201
model. The analytical expression for F3E, for were used.
the sides, is the same as Equation (7.19); The variable parameters in this study were
however, since in this case the top and bottom r_, _,, _, and i'. The orbit and spacecraft
panels are not generally at the same temper- orientation angles, _ and i', were varied in the
ature, their shape factors cannot be combined same manner as before. The earth's albedo, r$,
into a single expression of the form of Equation was assumed to have the values 0.3, 0.4, and
(7.18). The individual relations, Equations 0.5 in various computer runs, and the fraction, _,
(7.15), (7.16), and (7.17), must be used. These was varied from 0.00 to 1.00. The results of
complex expressions are closely approximated this study are summarized in Figure 7-10 and
by the following simplified equations: Table 7-4.
Figure 7-10 shows the temperature variation
_E=0.846 cos 2 [0.6345 cos -x (--sin _ sin of the wheel interior and the three cover panels
around the minimum sunlight orbit (_----0 °) for
sin _--cos v cos _)], if cos [ ]>0 _,----1.00 and an assumed value of 0.4 for the
earth's albedo. These curves indicate that for
or
this configuration all temperatures will remain
208
THERMAL CONTROL
Table 7-4
Average temperstm'es
(:K)
(deg) r (deg)
Top Bottom Sides Interior
2O9
ORBITING SOLAR OBSERVATORY
Table 7-4
210
THERMAL CONTROL
SIDES
(9
o
20 INTERIOR
lO BOTTOM
0 ! I !
computations. As shown in Figure 7-12, this plate is always in the yz-plane. The polar
difference would theoretically be expected to coordinate angle is ¢ while 0 is the angle
reduce temperatures of the interior of the space- between the z-axis and the line between the
craft to a mean of 9°C, 8°C lower than the plate and dS. The angel between the z-axis
value computed for the 100 percent polished and the plate normal is 7, while r is the polar
condition. The temperature curves of Appen-
coordinate distance from the origin to dS, the
dix B indicate the results accomplished by the area element on the earth disk. The distance
passive temperature control work. The azi-
from dS to the plate is p. Thus Equation (7.12)
muth casting temperature was about 2°C and becomes
the electronics package on the back of the sail
structure maintained a temperature between 0°
0 sin q_ sin 7
and 10°C. The solar panels were about 15 ° C
cooler than expected with the high day tem-
+sin e cos e cos 7)dq, dO (7.37)
peratures being about 60°C and the low night
temperatures being about --35°C. Where the integration is to be performed over
the region of the disk "seen" by the plate.
EVALUATION OF THE SHAPE FACTOR
INTEGRALS T
Figure 7-2 shows the geometry of the plate Figure 7-13 shows the limits of plate positions
and the earth disk. Note that the normal to the for Case 1. Since the entire earth disk is
211
• b
ORBITING SOLAR OBSERVATORY
20
I0
_ 0
_ -lo
-20;
>
"_ -30
20 4'0 60 80 1(}0
PERCENTAGE OF TOP AND BOTTOM SURFACES POLISHED
(9 \
\
/i/
\ //
a.,
20 /, PLATE
17.4°C
o
iii;iiiiiiiiii .." _t / \,
•_0o_
/
\\_x__POSITIONS
\
LIMITS OF PL_,TE
FOR _, > 0
z \\
-10
0.50 0. '6O 0.70 0. i_0 0.'90 1 .'00 t. I 0 FIGURE 7-13.--Limits of plate position (case 1).
x, PERCENTAGE OF TOP AND BOTTOM WHEEL SURFACES SHINy
FIGURE 7-12.--Average interior wheel temperature as advantage of symmetry, however, and write
function of top and bottom survace conditions.
212
THERMAL CONTROL
o-,.-(;-Oo_<,,K;):
/--PLATE
¢/---,' =_
PLATE ........... : -_
LIMITS OF PLATE
•v = -.,
Z -e o i
I
/_-- POSITIONS FOR _ > 0
(
EARTH DISK
/ _=!
/ 2
/
/ F I1
Figure 7-14 shows the limits of plate positions when the plate "sees" more than half, but
not all, of the earth disk. Also shown is a diagram of the two regions of the earth disk to
be covered in the evaluation of the integral. Again symmetry properties may be used to write
F= 2lr fo,/2-, f..2/2 (si n2 0 sin c sin -/+sin 0 cos ¢ cos ,/)dcd0
=II+L
I2-=---- f:,o
/ 2- 7
sin g 0 sin 7 ]cos 61-a
--sin-_ (cot 0 cot 7)
dO+ _J r.o
.I2- 7
sin 0 cos 0 COS .,,
,/[(_]--Sin-](cot0COt7)
do
213
ORBITING
SOLAR
OBSERVATORY °
In/2.1 let x=sin 2 O. Then dx-=2 sin 0 cos odo, and . "
dx dx
do=2 sin 0 cos O--2_'x _/l--_--xx
Thus
{',,,' o0x41_
I2.,= 2 sin _,j¢o, (1_1) cot 2 3` 2_/x_/i--x
dx
-- sin_r 7f":'
ocos _" (x ese2 1--x
"--e°t2 Y) 1/2dx
I 1 0o 1 1 (7.41)
2.2= 3 cos 3` [sin 2 0] f/2-7 =2 sin2 0o cos 3`-- 3 cos a _,
I2.a may be integrated by parts, letting I_=sin -1 (cot 0 cot 3`) and dr=sin 0 cos odo. Then
I 2.a--;reosTsin20osin-'
_1 (7.42)
(cot 3, cot 00)--3 1 eosa 3`÷ 1 cos-l, [_cot, 3, f, °o
/2_, _/tan2 tanOdO
3, tan2 0_ 1
Let this remaining integral be called I...a.,. In this integral, let x=tan 3,tan 0, so that dx----tan 3,
see 20dO, or
de:=
1_ -x2 e Ot2 "7
Then
I2.3a ---1 cos 2 y sgn (sin 3`) tan-' ([cos 3`[ _/tan 2 3` tan 2 001) (7.43)
7_
If 0_<y_<Tr it is not necessary to include the factor sgn (sin 3`) in this expression. The shape fac-
tor for Case 2 is, therefore,
1 1 1
÷_ sin2 3,÷_ sin 2 0o cos 3,--_ cos a 3"+1_r COS 7 sin2 00 sin -1 (cot 3' cot 0o)
_12 c°sa y÷_17r cos_ "r sgn (sin y) tan -_ (Icos 3,[4tan 2 3` tan 2 00--1)
214
* THERMAL CONTROL
t_L
l 1
=_ sin s 0o cos *A-_ sin 2 _,--_ cos 0o_/sin 2 0o--cos 2 ,-- sin2 , tan -1 \_/sin _ 0o--cos _ _' _r cos X
sin z 80 sin (cot -¢ cot 80)+ 1 cos _ -¢sgn (sin -¢) tan -1 ([cos -¢[_/tan _ _¢tan 2 0o--1) (7.44)
T F
Note that as -¢-_--80, approaches
=sin s 8o
This must agree with the value of F in Case 3 for the same angle.
Case 3: (- -)
__<_'<_-t-0o (see Figure 7-15):
_oR_
.......... _ .._ "-_'_-__
' ='i -I /
\
Y / \
\
.80 _'\
\
/
LIMITS OF PLATE /
215
ORBITING SOLAR OBSERVATORY °
=-- sin 2 O sin 3"[cos _] dO+ cos 3' sin O cos O [_] dO
"/r -:t/2 -sin-Z (cot a cot -t) -_/2 --sin-' (cot a cot 3")
Since the integrands are the same as those in Case 2, we have for Js. 1:
1 0o 1 i
(7.46)
J2.s= 3 cos 3" [sin 2 013"_,n=2 sin s eo cos 3,-- 3 cos a 3"
J2.3= 2 cos 3"([1 sin 2 O sin-l(cot -10o 1 cot3"3"! f3±o-,/_ a/tan 2tan0d0
0 cot 3")J3"_,/2-+-_-r[cot 3"tan s 0--1 )
1
=- cos 3" sin s 0o sin-l(cot 0o cot 3")--1 cos3 3" sin-_(--1)
71" 7r
(cos3") Icot3"!cot23"
cot33" 1 / cotS^/ Wtan_ 1 /cot23"(x2--1) _l-,
I tan3"tan°°
or
J2.z= 1 cos 7 sin s O0sin-'(cot eo cot 3")-F_ cos 3 3"+1 c°s s 3"sgn (sin 3") t,m-l([cos 3"[_/tan 2 eo tan 2 3"--1)
(7.47)
The total results for F are:
1 1 1 a 1
+2 sins 3"+3 sin_ O0cos 3"+3 cos 3"-t-;-r cos 3" sin s O0sin-l(cot Oocot 3")
1 1
q-_ cosa 3"q-_r cos _ 3" sgn (sin 3") tan-l(Icos 3"14 tan2 3" tan_ Oo--1)
216
* THERMAL CONTROL
_)la
F----_1 sin 2 00 cos _+_ 121 sin 7--_ cos 00 _/sin 2 00--cos _ _/_ sin2 ,_ ta__l \_/sin_
( 00-cos2
COS00 7 ).__1 _ cos
sin 2 00 sin -1 (cot 00 cot 7)-t -1 cos 2 _ sgn (sin y) tan -1 (Icos 7l _tan _ 7 tan-" 0o--1) (7.48)
T
2"
F=--_sin1 a 0o+_Cos
1 2 0o--lcos0osin 2 0e--sin 2 0o--_ 1 COSS0o(_)--lsin 3 0osin -l (-- l) +l T sin2 0o tan -1 (0)
T
1 "3 1 2 1
=--_ sm 0o-{-_ cos 0o--_ sin 3 00=0
/ _ I _ ",/ = T
o- / .... :e,,, dS 2
LIMITS
POSITION
OF PLAT
FOR /
/
0_,_ 7 / _
217
ORBITING SOLAR OBSERVATORY
F=I fcos4), 1
_ j_ (; sin as) dS 4 , fo°O sin 0 Jo
=--71"2 t"/2 _/1--siId 0 sm _ 0d¢dB
or
-4
--71.2
foO°sinOE(2, sinO) dO (7.52)
F_l ff cose
--_ ./j_ sin a_h 2 see 2 e tan ed_fO
where E is an elliptic integral. Numerical inte-
(7.49)
gration of this result gave
where the integration is over the entire earth
disk (except for a small hole formed by the F _4
--_ (0.816)----0.331
projection of tbe cylinder along its axis to
intersect with the disk. This unseen region
can be neglected because of the small size of the The extreme values of F(_) were thus
cylinder in comparison with the earth disk).
The counterpart of ¢._ in the case of the flat F(0) =1 (0o--sin 0o cos 0o)=0.254
_r
plate is a,.
Hence,
_/1--(sin 0 sin 4) sin 3,--cos 0 cos 7)_dOd4) (7.50) F(-y)=0.254+0.077 sin 2 "Y (7.53)
218
Chapter 8
219
782-552 O--6G--15
ORBITING SOLAR OBSERVATORY "
'1
SUN
ORIENTED EXPERIMENTS
FIGURE 8-1.--Solar-oriented experiment.
22o
• ORBITING SOLAR OBSERVATORY SCIENTIFIC EXPERIMENTS
per mm on a one-meter radius of curvature The spectrometer used about 1.3 watts sup-
blank. No reflecting coatings of any kind plied from the spacecraft power supply at a
were used on this grating. The diffracted rays nominal 18 volts dc. About one watt was used
then continue to the exit slit in front of the by an oscillator which drove the three-phase
detector. The detector was mounted on a synchronous detector scandrive motor. The
carriage moving on a rail so that the detector motor required about 300 mw at 137 cps to
input slit would follow a Rowland circle where yield 100 mw of mechanical output power. The
the spectrum was in focus. The plane of the remainder of the power was used in the multi-
detector slit remained approximately perpendic- plier amd pulse-handling circuitry. Power to
ular to the diffracted ray at any position along the experiment was switched off when the
the rail, thereby keeping the spectral passband spacecraft entered the earth's shadow and
nearly constant for all angles of diffraction. switched on at the beginning of a new space-
The 50-micron entrance and exit slits provided craft day.
a spectral passband of 1.7 A and permitted Since the diffraction grating was quite sus-
resolution of lines 0.85 A apart. ceptible to contamination, all of the exposed
The detector used was an open-window materials inside the spectrometer were tested
photomultiplier developed by the Bendix Cor- at a pressure of about t0 -B mm Hg to eliminate
poration specifically for use in this spectrometer any material having a high vapor pressure.
and designated the M-306. Its operation was Because the electronic circuits were potted in
as follows: Photoelectrons from the exposed a compound having a high vapor pressure, they
tungsten cathode move along cycloidal paths were enclosed in sealed boxes which were vented
in crossed electric and magnetic fields between to the outside through holes in the spectrometer
two glass strips coated with a semiconducting base plate. During assembly, all the parts
secondary-emitting oxide layer. One of the exposed to the interior were carefully cleaned
glass strips serves as a continuous dynode. to be free of oil, grease, and other contaminants.
Each photoelectron is multiplied into a pulse of The temperature of the instrument was con-
approximately 106 electrons at the anode. trolled by controlling the radiation balance.
The data format of the x-ray spectrometer Most of the outer case was polished. The
experiment used a pulse-counting technique central section was black in order to thermally
whereby narrow, randomly occurring pulses "tie" the experiment to the spacecraft. A
from the detection system were converted to thermistor was mounted in the experiment to
bi-value waveforms by means of binary, scalers. monitor temperature during orbital flight. A
To extend the dynamic range over a wider plot of this temperature for the first 1,000
range of input pulse rates, three different orbits is shown if Figure 8-4. The output of
scaling factors were used. The three bi-valued
waveforms thus generated were then coded
and combined in such a way as to permit later +5O
unambiguous identification of the particular
contributions of the different scaling factors to +40
the composite waveform. A typical data out-
+30
put format is shown in Figure 8-3. This format
was the input of the 1200-cps subcarrier oscil- +20
lator providing the frequency modulating for
+10
1the data transmitted.
0
-- DAY _'_ NIGHT_
t | i J 1 I I t I
-10 0 lO0 200 3004005006007008009001000
ORBIT NUMBER
FmvoRZ 8-3.--Goddard x-ray spectrometer output
format. FZGURE 8-4.--X-ray spectrometer temperature.
221
ORBITING SOLAR OBSERVATORY
222
ORBITING SOLAR OBSERVATORY SCIENTIFIC EXPERIMENTS
224
ORBITING
SOLAR
OBSERVATORY
SCIENTIFIC
EXPERIMENTS
) , _
®
RI: >
>
INPUT A
®
!
/ [ \
,%-_-
INPUT
T
j
B
CRl
I_[ _
R61 >
R10
I OUTPUT
[ <
l
i I R2A
?
CR 2 _" R2 ,' R2A _ <.I
2 CR 3 R 5
!
! .<R2 R7 ! R9 !
: :b ®
GROUND
15V
tion of this device (see Figure 8-9) is the same
as the readout amplifiers previously described 36K 5K
in Chapter 4.
The sensor monitoring experiment required
SILICO; -- MEA
a voltage shifter designed to accept an input CELL P5595 7.5K
signal ranging from 0 to 5 volts and to supply DETECTOR
o
an output ranging from about 4 to 9 volts.
The requirement for this unit arose from the
fact that
subcarrier
it was desirable
oscillator which
to use an existing
required a 4- to
)
9-volt input. -- OUT
I
Figure 8-10 shows the schematic diagram of
21.82K
this circuit. The requirements on the voltage MEA
shifter were that it present the proper input and IN 754A
225
ORBITING
SOLAR
OBSERVATORY .
+IBV
_2
network was added to the output of the level
_R 2
,4.64K shifter to limit the maximum ouptut to 10 volts.
>I_ I 8W Q4
D2 Performance
- IN483A
R4 10K I OUTPUT
Information on the performance of the four
R3
4, 64K
_ D3
IN'/58A
GSFC solar-oriented experiments is not included
141, Ow _ (10v)
.... J in this chapter, but may be obtained by referring
GROUND
8.
The degradation figures were derived over
0 the first 1,000 orbits and were, for this period
of time, linear functions of time. It was
ee assumed that the data could be extrapolated
[., 7-
to six months with some confidence if the
electron flux problem were ignored.
The open-cell detector exposed the silicon
O 6.
[-. (,.ell directly to solar irradiation. The degrada-
tion was due to direct ultraviolet irradiation
0, damage.
The fine pointing detectors with lenses and
O 5-
narrow transmission band filters were the type
used for the servo positioning control error
detectors. The two detectors were not spec-
4.3 I
4 traUy matched. This is the reason for the
0
slightly different degradation rates. The servo
INPUT FROM EYE AMPLIFIER (VOLTS)
gain decrease in six months (nine percent aver-
Fm_mE _-11.--Voltage shifter transfer curve. age) due to eye output reduction caused a
pt)sitionin_ error of the same percentage applied
to the input signal. Transistors Q1 and Q2 to the initial error; that is, with a 1.0 arc-minute
are used to present the proper impedance to the initial control accuracy, the error after six
signal source. Transistors Q3 and Q4 present months was increase(t by 0.1 minute of arc.
the proper impedance t<_ the subcarrier oscil- The variation in positional accuracy caused by
lator. Tim outpu[ of the voltage shifter appears the different degradation rates was negligible.
acrt)ss R3. Figure 8-11 shows the t ransfel The coarse pointing detector was identical
characteristi(,s of a typical unit. Typically, the to the c,oarse detectors used for the positioning
level shifter has a linearity ,)f about plus or conlrolsystem. No degradati(m was measured.
minus i_
/ percent and a temperature stability The detector output varied, in agreement with
,)f about l)lus ,)r minus 1 percent over the range the theoretical, as the variation in the solar
-f temperature, --20 ° to +50°C. A limiting constant. This information indicated that the
226
o ORBITING SOLAR OBSERVATORY SCIENTIFIC EXPERIMENTS
2.1
2.0
1.9
_1.8
Z
LU - 10°C
1.7
,-,I
0.6
MA COARSE ERROR DETECTOR
I , I i I i I , I i I * I i I I I
1.4 .
0 1O0 200 300 400 500 600 700 800 900 1000
ORBIT NO.
degradation measured for the fine pointing The azimuth position readout pair of detectors
detectors was due entirely to effects on the lens performed well. After the upper commutator
glass (probably ultraviolet radiation-induced failure about orbit 1900, this detector pair
color center formation causing some attenua- became the only readout available to allow
tion) and that the degradation of the cells them- determination of azimuth pointing accuracy.
selves was negligible when protected by the A relative shift in azimuth pointing direction
filter glass. between the pointed experiments (about one
Data concerning the effects of high-energy arc-minute) occurred during launch. After
electron flux on the detectors would be of 3500 orbits, the experimental pair was reading
considerable interest, but unfortunately, the azimuth positioning error changes from the
detector outputs were reduced to the lowest initial position of less than one minute of arc.
level readable by about orbit 2000. This was Two curves of interest are presented con-
about the time the 9 duly 1962 high-altitude cerning the experiment. The curves of Figure
nuclear device was exploded. Since the coarse 8-12 show the combined effects of degradation
detector eye did not continue to give on-scale and solar constant variation over the first 1000
readings after that time, as was anticipated orbits. The solar constant change was a
since tile solar constant was on tile increasing decrease of 3.2 percent over this period. The
half-cycle, it was assumed that some radiation coarse detector, as noted previously, followed
change occurred. Tiffs change needed to be this change very closely.
only about two percent to produce the effects The curve of Figure 8-13 shows the coarse
measured, thus no real limits could be estimated. detector output variation over one orbit. The
ORBITING SOLAR OBSERVATORY
14-1/8" MAX_
DAY POWER DAY POWER
TURN ON TURN OFF
ioo IGHTJ_
98
96
I;il/x2"
/ II SECTOR
OF
92 0 2800 , r SATELLITE
TiME (SEC)
scription of each are shown in Figure 8-15. The within the solar cycle.
total weight of the wheel experiments in OSO Solar Ultraviolet Experiment
I was ll3 pounds.
Monitoring of the solar flux from 1100 to
Goddard Wheel Experiments 1250 A was accomplished with a LiF-CS2 ion
chamber suitably stopped down to retard
The Goddard Space Flight Center wheel deterioration. Considered of primary impor-
experiment package for the OSO I is shown in tance, this wavelength region included the
Figure S-16 and comprised the following Lyman-alpha chromospheric emission line which
experiments: is strongly associated with solar activity.
Intensity variations were measured to allow
Solar Radiation Experiment
correlation to be made with active features
This experiment utilized a special photo- on the solar surface.
diode with a filter which restricted its spectral
response t() the region between 3800 and Solar Gamma-Ray Experiment
4800 A at the half-intensity point, very similar This gamma-ray experiment was designed to
to a band widely used in stellar photometry. detect gamma rays from the sun and analyze
228
ORBITING SOLAR OBSERVATORY SCIENTIETC EXPERIMENTS
WHEEL
FIGURE8-15.-Mrheel experiments.
their energy spectra from 0.2 to 1.5 MeV. 0.5 Mev eiectroo-pcsitron annihilation line and
The experiment utilized for detection two to study any possible temporal variations.
S a 1 (TI)scintillators, backed by photomulti- The detectors looking outward from the wheel
pliers. One detector was shielded to achieve a spent a good fraction of the time scanning
penumbral acceptance angle of 20 degrees.
fLKNMK
The second detector, which was unshielded,
a as used in anticipation of background events
induced by the presence of spacecraft mass
and, in particular, by the shielding of the other
detector. The unshielded detector had nearly
iqotropic response.
Pulse signals from the photomultipliers were
fed into w 16-channel differential pulse-height
analyzer which \\-a$ time-shared by these
two detectors and by one in the composite
pointed experiment. From the analyzer, the
4gnnls were coded and fed to the spacecraft
telemetry *>-stem. Of w particular intere-t FIGVRE +16.-Goddard Space Flight Cpnter wheel
in this experiment was the ability to detect the experimeri t package.
229
ORBITING SOLAR OBSERVATORY °
FIGURE 8-17.--(loddard wheel experiment data output This experiment was designed to provide
format. preliminary measurements of the intensity and
directional properties of low-energy gamma rays
Performance
in space. These photons have energies of about
The earth sensor received signals froll_ scat- 1 Mev and are similar to light rays or x-rays
tered light reflected from the cover during each in that they travel with the same speed and are
rotation past the sun. This scattered light not deflected by electric and magnetic fields.
produced an ex*ra pulse in the output signal Tiros, unlike true cosmic rays, but like starlight,
but did not interfere with the normal function the gamma rays arrive from the same direction
of the earth sensor. The solar radiation diode in space from which they originated. They
230
ORBITING SOLAR OBSERVATORY SCIENTIFIC EXPERIMENTS
23 1
ORBITING SOLAR OBSERVATORY .
t.n4 cu _ p I
Ijl --I._i.JEaU
was generated
pointed
programmed
every time the telescopes
at the sun and during
gates allowed
22.5-degree
counts
one other
sector interval.
registered
were
pre-
These
in the telescope
to pass into the storing and accumulating binary
scalers. The sector gate was programmed so
that after every night-day sequence, the sector
FIGURE 8-20.--University of Minnesota gamma ray was advanced one 22.5-degree interval. Hence,
experiment block diagram. after every 16 orbital passes of the satellite, the
telescopes would have been gated at each 22.5-
tion properties of gamma rays between 50 kev degree interval around the wheel for one orbit.
and 3 Mev energy and observed time variations The sun interval was always gated ON during
of the solar flux in this region. Advantage was every orbital pass and every daytime wheel
taken of the fact that the wheel spin axis was revolution.
maintained normal, within a few degrees, to a The six 9-binary storage registers accepted
plane containing the solar vector. The tele- events from the detectors in the following
scope, pointing radially outward from the manner:
wheel, then swept by the sun, across the sky, 1. From the 50 to 150 kev lead pipe telescope
and probably across the earth with every two- during the sun gate interval.
second rotation period of the wheel. 2. From the 50 to 150 kev lead pipe telescope
The 50 to 150 kev telescope, shown in the during the sector gate interval.
block diagram of Figure 8-20, consisted of a 3. Scaled-down counts from the 0.3 to 1.0
NaI (T1) crystal scintillation counter in a lead Mev isotropic counter continuously.
pipe shield which looked outward through the 4. From the 0.3 to 1.0 Mev Compton tele-
wall of the spacecraft. A directional detector, scope during the sector gate interval.
operating in the 0.3 to 1.0 Mev energy regions, 5. Scaled-down counts from the 1.0 to 3.0
was obtained by using tw o scintillation counters Mev isotropic counter continuously.
connected as a Compton coincidence telescope. 6. From the 1.0 to 3.0 Mev Compton tele-
Photons, after being scattered in the first crys- scope during the sector gate interval.
tal, were absorbed in the second scintillation Events from the detectors were continuously
detector. The directional properties were ob- accumulated in the storage register during the
tained by selecting only photons scattered daytime portion of the orbit. The storage
forward in the first crystal. The absorbing registers were read nondestructively in sequence
NaI (T1) crystal was surrounded by a plastic into a shifting register which advanced at a
phosphor fence which rejected events having two-bit-per-wheel-rotation rate (1 bit-per-sec
proper energy but caused by edge effects of at 30 rpm). Six words of 13 nonreturn-to-zero
particles in the crystal. Photon counters in bits were supplied to the telemetry subcarrier
the 0.3 to 1.0 and 1.0 to 3.0 Mev energy region oscillator. The first four, or "ID," bits of each
232
ORBITING SOLAR OBSERVATORY SCIENTIFIC EXPERIMEN_
_'ord were of five-volt amplitude and identified gamma rays presumably were produced by
the beginning of the six-word sequence, the cosmic rays. That most of the counting rate
word number, the position of the sector gate, was not caused by local radioactivity in the
and the position of the earth horizon. The last satellite itself was inferred from a preflight
nine bits of each word were of three-volt ampli- calibration. The calibration, performed a few
tude and corresponded to the reading of the days before the launch with the entire payload
storage register at the time of the start. Num- and vehicle configuration in the gantry, gave
bers read from the storage register continually lower background rates than in any previous
changed due to the continued accumulation of situation. These were about a factor of ten
counts from the detectors. below those observed in orbit.
During the night, the readout of t ae storage Of the first 1000 passes, about 50 passes had
and operation of the shifting register were in- been completely reduced and another 50 were in
hibited, and the cosmic-ray single counter and process at the time of this writing. Since the
telescope data were recorded. The outputs of Compton telescopes had the possibility of indi-
detectors were suitably scaled and the resultant cating the presence of an extraterrestrial
square waves were added together and fed into gamma-ray flux in the most unambiguous
the telemetD" subcarrier oscillator. manner, the data from 48 passes, numbered
generally m the 400's, are presented in Table
Performance 8-3.
Inspechon of the early telemetered passes at Events occurring when the spacecraft was in
Cape Kennedy indicated that the instrument the trapped radiation areas, as well as noisy
was performing properly. Later passes re- records, were eliminated. Events were accu-
vealed that occasionally incorrect digital words mulated when the earth detector was illumi-
were received. This effect, apparently due to nated, when it was off, and when the telescopes
an electrical transient resetting the shift- looked at the sun. Since the acceptance angle
register logic, was particularly troublesome on of the earth telescope was about the same as
passes 50 to 150 and had vanished by pass 400. the Compton telescope at 1.0 Mev, this formed
No actual loss of mformation resulted, but auto- a simple means for distinguishing earth albedo
matic data reduction became more difficult. gamma rays from those that were incoming.
No detrimental effect was observed due to the The rates of the isotropic counters, averaged
lower than anticipated operating temperature of over the same time intervals as the telescopes,
the satellite, and no evidence of a gain change in are given. The near identity of these rates indi-
the scintillahon counters was noted. The cated the absence of any large systematic effect
signal-to-noise ratio on most passes was quite due to a nonrandom variation of the isotropic
adequate for automatic data reduction. flux during the intervals in which counts were
The background gamma-ray rates in orbit tallied: for example, from radiation belt back-
were about a factor of three higher than those ground. The results would indicate an extra-
predicted by a reasonable extrapolation of bal- terrestrial flux greater than the earth's albedo.
loon results at Minneapolis. Most of these In order to actually determine the flux and t_)
Table 8-3
233
ORBITING
SOLAR
OBSERVATORY "
I,OOO _--
I I
respectively, at the equator. A variation of"
RATES VS. _ME about 2 to 1 in the rates from the equator to
OSO-I
PASS 15
geomagnetic 40 ° is observed.
The first daytime portion, in addition to the
general latitude variation, shows effects due to
the trapped radiation which were observed on
COUNTS most passes over the South Atlantic and Africa.
SEC
234
ORBITING SOLAR OBSERVATORY SCIENTIFIC EXPERIMENTS
craft wheel sweeps the circle. tively low orbit, as OSO did, those neutrons
counted were predominantly from the earth.
The output data is presented to the tele-
These neutrons are thought to leak out of the
metering system in a pulse duration-modulation
atmosphere of the earth. They are thought
form (PDM). The output pulse sequence
to be produced in the atmos|)here hy cosmic
contains the folh)wing kinds of data:
rays striking oxygen and nitrogen nuclei. Cal-
Solar gamma-ray counts--Channel 1 culations have been made as to how many of
Nonsolar gamma-ray counts--Channels 2 these leakage neutrons should t)e detected and
through 16 what their energy spectrum should be.
236
ORBITING SOLAR OBSERVATORY SCIENTIFIC EXPERIMENTS
The previous experiments by the University were 20 milliamps a t 18 oolt,s, or about 360
of California on two Atlas rockets have given milliwatts.
reasonable measurements of the neutron fluxes
in space for short periods of time. The values Proton-Electron Flux Experiment
in space for two short periods of time. The T h e objective of this experiment was to
values obtained were different, but one flight determine the time and position variations of
was during a solar storm, so this difference waq the fluxes of protons of 2.0 Mev energy and
not surprising. T h e OS0 spacecraft flight electrons of 60 k e r energy in the region below
provided measurements of the neutron flux in the Van Allen belts. I n particular, the experi-
hpace for an extended period of time. The ment distinguished between electron ionization
experiment had to be able to measure the time erents and proton (and other heavy particles)
variations of the flus, and hopefully, be able to ionization erents. This experiment was chosen
study the neutron enhancement effects during a by GSFC because it was expected that the
solar storm. It was intended to try to de- largest fraction of high-energy radiation in the
termine the latitude dependence of this en- plasma sphere and in outer space would be of
enhancement. these t-ypes. T h e proton-electron (P-E) de-
Besides measuring the quiet dav neutron tector, shown in Figure 8-25, was the first space
flus and time variations of flux, the experiment application of ft phenomenon that had been
looked for flus variations a t sunset to see if an well known in nnclear physics for several years.
effect could be observed related to solar neutrons It was based on the fact that in certain scintil-
or x-rays. lators. protons and electrons produce fluorescent
Understanding the neutron flux wns im- pulses of distinctly different decay times.
portant in itself, but i t was also important This made i t possible through use of time-
because the deea>*of these neutrons forms one comparing and discriminating circnits to em-
of the important sources of the Tan Allen
ploy a single scintillator on a single photo-
radiation belt. Understanding the neutrons
qiiantittttively will help understand the radia-
tion belts quantitatively.
T h e experiment is shown in Figure 8-24.
The unit weighed approximately 26 pounds.
The power requirements of this experiment
-
F I G E K 8-?4.-l.-niversity
E of California neutron flu\ FIGUHE
S-25.--l-niversity of California proton-electron
experiment. detector.
237
ORBITING
SOLAR
OBSERVATORY
multiplier to detect and separately count both 12 (the 10.5 kc channel). A sample format o_
protons and electrons. the putput signal is shown in Figure 8-26.
The basic detector consisted of a 15 cm- The output signal is made up of the following
diameter by 1.0 cm-high stilbene crystal on an items:
RCA C7151 ruggedized photomultiplier tube. G(W)--Geiger tube with magnet (electrons)
Side and front shielding provided collimation of G"--G"Fa neutron detector
the crystal and excluded protons of Ep=-100 G(o)--Geiger tube without magnet
Mev. A quartz light pipe provided similar BI°_B_°F3 neutron detector
shielding from the rear. The electronic com- P÷E--Proton plus electron scintillator
ponents were mounted on foam-covered circuit detector
boards which compactly surrounded the photo- Pl--Proton identifier
multiplier housing. The unit weighed about 9 (P+E)/64
pounds. Power requirements were 14.5 ma at B_°_B_°F3 neutron detector
18 volts, or about 265 mw.
The output signal commutator required 51.2
To measure the neutron flux, two BF3 pro-
seconds to complete a frame. Every sixteenth
portional counters, one BI°F3 and one Bur3 frame was an automatic calibration.
were used. The neutrons were counted by
the reaction, BI°÷N_Li÷a. The BnF3 count- Performance
er was used to measure the background counting
The neutron experiment worked well except
rate. The background due to (p,n) events in
for improper cmmt presentation from the GM
the vehicle and detector was estimated to be
counters. These counters were to tell if some
low, but it also was measured by exposing the
counts in the BF3 counters were caused by
detector to a high-energy proton beam. The
neutrons produced in the spacecraft by a large
BF3 counters were covered with about lh /
charged-particle flux, rather than by incident
inches of epon moderator to make the neutron
neutrons. This function was also performed
count rate larger. This detector had a nearly
by the proton-electron package, making the
constant efficiency from one kev to several Mev.
GM counters less necessary. The calibration
Two GM counters to study charged particles
system in the neutron package indicated that
were also used. This was not a primary
the threshold sensitivity of all channels follow-
measurement. It was intended to use the
ing the detectors had changed by less than 20
GM counter count rates to determine whether
percent (which was insignificant with good
increases in the neutron count rate were in time
plateaus).
coincidence with large charged particle count
The proton-electron experiment operated
rates. This deternfined whether protons pro-
successfully. The calibration system on board
ducing neutrons in the detector were responsible
indicated that the threshold sensitivity of the
for the neutron count rate, or if neutrons were
proton-plus-electron channel following the photo-
then_selves incident on the detector. One
multiplier had changed by less than 4 percent.
GM counter had a sweeping magnet so that it
The generally high electron and proton flux
would count mostly protons. The second GM
region over the South Atlantic was mapped for
com_ter had no magnet so it would count pro-
several orbits, but a comprehensive data
tons and electrons.
accumulation has yet to be achieved. Some
The output signal from the two experiment
indication of time dependence of the proton
signals was combined into one and was the
fluxes observed over North America was also
input to the 572-cps telemetry subcarrier oscil-
found.
lator, which during playback became Channel
238
ORBITING SOLAR OBSERVATORY SCIENTIFIC EXPERIMENTS
*thermal radiation to gain or lose heat to their changes in the temperature of the surfaces with
surroundings. Therefore, an important con- time, it was possible to determine the amount
sideration in the design of space vehicles is the and rate of change in the thermal radiation
type of surface used for temperature control. characteristics of the surfaces.
For example, although most polished metal Four basic types of surfaces were studied:
surfaces reflect a large part of the sun’s rays, 1. A flat absorber which absorbed radiation
they give up the heat they do absorb a t a rery at a uniform rate throughout the wavelength
low rate, and consequently reach high tempera- spectrum.
ture equilibrium in the sun. White paints and 2. A flat reflector which reflected radiation at
enamels, on the other hand, while reflecting a uniform rate throughout the spectrum.
most of the incident sunlight, emit heat, at a 3. A solar absorber which absorbed radiation
high rate, causing them to reach low-tempera- in the solar wavelength region and was useable
ture equilibrium. Stability of the thermal to reject heat at its own temperature in the
characteristics of temperature control surfaces infrared region.
is very important in the design of spacecraft. 4. -4 solar reflector which reflected solar
The space enrironment can cause changes in radiation and rejected heat in the infrared
the radiation properties of surfaces, leading to region.
inadequate temperature control. The magni-
Six test surfaces were placed in thermally
tude and rate of such changes is a t present not
isolated mountings and the equilibrium tem-
known for most materials.
perature of each surface was measured in space.
Surface temperature provided a measure of the
ratio of heat absorbed to heat emitted from the
surface. The surface was selected to provide
information on the emissivity stability of
various coatings applicable to spacecraft tem-
perature control. The surfaces were of three
types :
1. Surfaces which were being used or ap-
peared t o be useful for temperature control.
2. Surfaces which were expected t o change
on!y nnder certain conditions of space
eniironment .
3. A surface serving as a reference. Com-
parison of the temperatures of the test surfaces
with that of the reference surface provided a
basis for evaluating changes in the thermal
FIGCiRP; 8-27.--Ames Research Ceritw ernissix it y characteristics of the test surfaces.
stability esperimmt.
The temperature of the surfaces was detected
by thermistors. The varying resistance of the
A project in progress a t the time of this thermistors changed the frequency of an
writing a t h i e s Research Center involved oscillator. This, in turn, modulated the fre-
measuring changes in the radiation character- quency of transmitter So. 1 together with the
istics of a number of surfaces intended for use recorded output of either tape recorder 1 or 2
in thermal control systems during *paceflight. dnriny playback. This was the only experi-
The experimental package shown in Figure 8-27 ment qignd normally transinit ted in real time.
waa developed to perform thebe tests on the Figire 8-25 sliov,-s a typical output signal
spacecraft. The experiment consisted of meas- format.
uring the temperature of several test siirfaces Te5t .iirface- and their relittive character-
during exposure to conditions in space. From i,tic- are as shown in Table 8-4.
ORBITING SOLAR OBSERVATORY *
Performance
3 4 5 6 7 8 9
l 39 --3 12 7 76 69 85 177
Fort Myers, day ........ 190 I
6O 3/7/62 ............. --2 13 9 77 70 86 178 191 I
124 0 16 ll 81 72 8X 180 194 I
180 1 16 14 83 72 89 1_4 196
......................
240
, ORBITING
SOLAR
OBSERVATORY
SCIENTIFIC
EXPERIMENTS
•TABLE 8-5.--_/leasured Temperature o[ Sensor on Ames Emissivity Experiment on OSO/--Continued
3 4 5 6 7 8 9 10
I Location and date of acquisition
: (seconds)
13 I 21 I --2 --26 17 38 22 92 91
--22 20 39 27 98 96
1710 --1
--1 --21 21 40 34 99 97
3/12/62, night.
86 -- 49 --45 6 36 1 62 59
3/13/62, night ....... [ --
89 1 --9 35 39 45 118 128
3/13/62, day .......... ]
103 3/14/62, day ......... I -3I --20 --26 14 33 23 88 88
118 3/15/62, --15 --22 18 36 26 94 94
day .......... I --28
177 --18 --20 14 38 24 94 89
3/19/62, day ......... --26
241
ORBITING SOLAR OBSERVATORY
given for orbit 89. The reference-surface tem- optical surfaces of experiments and to avoid.
perature was 35°F in both cases. For this weight loss which could disturb the spacecraft
position in the orbit, the temperature of the balance. Outgassing problems were controlled
epoxy-base paint rose from --21°F in orbit 29 by the proper choice of materials on board.
to --14°F in orbit 89, an interval of 4 days. If an experiment contained a magnet, addi-
Similar comparisons can be made between other tional magnets had to be added and oriented to
temperature listing, and for the other coatings. effect a magnetic balance for the overall space-
A study of the table will reveal a certain amount craft system.
of normal scatter in the measurements. The All experiments were electrically checked on
ch_mges in temperature noted above, however, the bench for proper operation prior to in-
were sufficiently marked to indicate definite stallation into the spacecraft. The final
degradation in the surface characteristics. calibrations of the University of California and
the University of Minnesota experiments had
EXPERIMENT INTEGRATION to be performed in the spacecraft. The cali-
brations required that the spacecraft be sus-
Assurance of the compatibility of the space-
craft electrical and mechanical systems with pended in the air a minimum of 10 feet away
each experiment was a necessary undertaking from any surrounding object or the ground.
from very early ill the program until _ check- Due to the hazards to the spacecraft involved
out of the entire prototype spacecraft was in suspending it high in the air, the prototype
completed. Problem areas in which much coordi- was used. Calibration was accomplished using
pleted. Problem areas in which much coordi- various radiation sources to obtain calibration
nation was required were the outgassing prop- curves with the actual spacecraft mass properly
erties of materials used in experiments, the moderating the sources. Various background
magnetic balance requirements due to experi- levels associated with these sources were meas-
ment detector magnets, the command require- ured. System and environmental testing re-
ments of certain experiments, the balance of quired the use of various radiation sources to
the pointed experiments, the electrical inter- check out the instruments contained in the
ference between experiments and spacecraft spacecraft. Due to the nature of radiation
system_ or between the experiments themselves, which some of the experiments were measuring,
and the experiment checkout procedures re- it was not practical to provide all the sources
quired during system and environmental required to completely check out the experi-
testing. ment. Instruments of this type were checked
The amount of outgassing had to be con- electrically and by using known background
trolled in order to avoid contamination of the count levels as sources.
242
Chapter 9
TEST PROGRAM
INTRODUCTION Air-Bearing Test Device
Extensive testing was required throughout The air-bearing test device was designed to
the development of the spacecraft to demon- duplicate the free gyroscopic motion of the
strate feasibility of design and to evaluate the orbiting spacecraft.
performance of the various systems. This test The air-bearing test device comprised three
program proceeded through the developmental, main sections: the spherical air bearing, the
prototype and flight model stages of design. gyro platform, and the azimuth assembly (see
In many instances, successful testing required Figure 9-2).
the use of special test equipment and fixtures. The spherical air bearing provided minimum
A description of each of these test dex4ces is friction and eliminated as nearly as possible the
included herein as a preface to the actual test influence of all external torques. It was capable
program. of supporting up to 600 pounds with an air
supply of 30-70 psi, and provided for a spin axis
TEST EQUIPMENT tilt of +40 ° .
on which the pointed instrument experiments elevation positioning system with eyes and
could be mounted for testing. The entire device controls. It also provided a yoke with a trans-
was mounted on a fixed base. verse spindle for mounting the elevation point-
This base also supported a motor and a drive ing instrumentation.
mechanism for rotating the wheel with respect The gyro platform spin rate was maintained
to the fixed base. Also mounted on the fixed by three gas jets mounted tangentially to the
base was a drive mechanism which rotated an outer surface of the rotating section. In order
excentric through which the lower wheel shaft to maintain a fixed position, the upper structure
bearing was mounted. The excentric drive was driven by a torque-motor in the opposite
speed was adjustable to give the wheel wobbling direction to the gyro platform at the required
motions at any frequency from zero to two speed.
cycles per second. The driven excentric could The principal moments of inertia of the fix-
also be adjusted to provide wobble amplitudes ture were of the same order of magnitude and in
from zero to several degrees. Figure 9-1 is an approximately the same ratio to each other as
illustration of the wobble table. those of the rotating section of the spacecraft.
243
ORBITING SOLAR OBSERVATORY
I 1
A-A
L A
FIGURE
+l.-Wobble table fixture.
944
TEST PROGRAM
Periscope
full-opening, allowing convenient installation
A periscope fixture was designed t o itllow
and removal of the spacecraft (see Figure 9-3).
known sun angle changes to be obtained for
checking out the pitch readout and the pitch Electrical lead-throughs were provided t o
control eyes. The fixture consisted of two operate the spacecraft, monitor the spacecraft
rnirrors mounted pnrsllel to etich other. One operation, control the pitch and spin gas control
iiiirror WLS fixed and one WHS rotated by mems systeiiis and monitor temperatures and pitch
of a micrometer. The micromet,er mirror readout.
motion was calibrated such that the apparent Spin Table Test Device
w n angle shift caused by the micrometer motion
Two spin table test fistires were built. One
could be itccuratelp reztd.
was a table fixture, tlie other was floor mounted.
Shadow Alignment Tool Both tilbles had a. rotatable shaft supported by
21 rigid base. The shaft had an attach clamp
The shitdow iiligirnmt t o ~ Wl A Y it very simple fitting i o :itate with the attach clamp fitting
device built t o determine when the upper of the spacecmft. A :c;rque rnotor installed
htructure of the spacecraft w-tih normal to the on the shaft rotated the spacecraft to simulate
solar rector. I t consisted of ill1 occulting disk expected orbital motion. The spacecraft con-
on i~ long itrni with calibrated ringh tit the base trol input connections were provided by slip
of the arm mounting. Thih de1Tice W H ~also rings connected throngh the spin table t o an
weful for determining the zero pitch angle oiitpiit connector. This permitted the space-
pohition of the pitch mi+. craft to be operated froin a console while
rot at in p.
Thermal-Vacuum Chamber
Balancing Machine
The t lien rial-vacuii n i chamber provided :I
siniulated space environment for testing the balancing iiiacliine was required t o plnce
spiicecrnft and its 1-arioiis sJ-steriis. It was the spacecraft in static and dynamic balance.
capable of ii wide range of temperatiire and The hii1;incing mnrliine had a platform on
pressure. The chuinber was 10 ft. in diaiiieter, nliicli the spiicemift coidd be inoiinted and
large enougli t o acconiiiiodnte the sp;icecrnft rotated. The pliitforin wis p w t of it pendidulil
and test table. I t hiid t h e e windows to periiiit \ - o h . Dyiiiiiiic iiiih:il:ince in tlie payload pro-
observation ilild iillow the use o f the art ifici;d duced motion in the pcndiiliiiir yoke, both in
siinlight source. The door of the chaniber was plirise and :iiiiplitiide. lTnbalunce niotiients
ORBITING SOLAR OBSERVATORY
FIGURE
9-4.-Balancing machine.
were determined and proper balancing weights of the support wires, The weight of the steel
positioned on the spacecraft. plate could be niade such t h a t when the plat-
The bdancing machine is shown in Figure form was oscillating a t a given frequency and
94. amplitude, its total energy (kinetic plus poten-
Pendulum Platform Test Fixture tial) wits equal to the kinetic energy of nutation
of the spacecraft. Thus, in damping out oscil-
The pendulum platform wtts used ns a test,
hitions of the test fixture, the damper dissipated
apparatus because it could be niade t o perform
the same amount of energy as in the actual
the circular translational motion experieiicetl
ttpplictttion. B y adjusting the frequency of
by the dnmper diiring nutiLtion.
vibration ttnd the weight of the steel plate, this
The pendiilutn platform test fixture (Figure
s p p m t t u s could also be used to evnlntite datriper
+ 5 ) consisted of a circiilnr steel plute siispended
perfornisnce on the air-betiring fixture.
from a rigidly nioiinted plttte of similar dinien-
sions by three ttdjustnble lengths of steel wire. Magnetotropometer
Multiple wire supports were utilized to elinii-
nate tip rtntl tilt of the pltitforin diiring lnterd
oscilhtion. Thqiiency of oscilltition of the
plntforin WIIS controlled by rLdjiisting the length
246
TEST
PROGRAM
TRIPOD CHAIN
I
J HULL
POST
/
FIGURE
TUB
9-6.--Magnetotropometer.
DEVELOPMENTAL TESTING
i
Control Sy_em
if,
In the early stages of development it was
necessary to prove that the initial concepts of
the control system components were sound.
FIC.URE 9-5.--Pendulum platform test setup. Testing was accomplished on the wobble table
and the air-bearing table.
field of the earth, thus determining the magnetic
£lectronic Subsystem
dipole moment of that object (see Figure 9-6).
The hull was 45 inches in diameter. The A mockup electronic system was built from
vertical clearance in the hull was about 22 easily obtainable components with no concern
inches. Both tub and hull were deep enough given to acquiring or using qualified space
to allow 150 pounds weight in the floating hull. flight components. This mockup system was
magnetotropometer could accommodate the used to demonstrate soundness of the design
entire sail structure of the OSO I spacecraft concept.
with pointed instruments installed. The next step in testing the electronic sub-
Figure 9-6 shows the cylindrical hull floating system involved the use of engineering model
in a tub of oil. The hull is held in the center of printed circuitry designed for the spacecraft,
the tub by three slack chains which drape from but not packaged as anticipated for the final
three stationary posts to a tripod attached design. This circuitry was installed on the all-
to the hull. The chains are attached to the bearing test device to demonstrate the per-
tripod at a small distance from its axis. There- formance capability of a two-axis control
fore, if the hull and tripod are rotated from system.
rest position, the chains apply a restoring
Pointing Control Subsystem
torque. If the dimensions of the apparatus,
the weight of tile chains and the displacement Tim various components of the closed loop
angle are known, the restoring torque can be pointing control were tested individually before
calculated. Then, knowing the orientation of the complete loop was checked out. These tests
the object and the direction and intensity of the involved measuring the open-h)op transfer
magnetic field of the earth, one component of characteristics of the servo-amplifiers, the
the unknown dipole moment can be calculated. power handling capability of the servo motor
247
ORBITING
SOLAR
OBSERVATORY "
adequate. W_ sin 2 0
(9.2)
2 cos0
Gas Control Subsystem
where
The gas handling components of the gas
control system (spin control and pitch control)
re=mass of platform
were tested before being integrated into the
v=linear velocity of platform
control systems on the test apparatus. This
r=radius of the circle described by the
involved the leak testing of valve, regulators,
center of gravity on the platform during
and fittings, the flow testing of regulators and
this motion
filters, calibration of pressure transducers and
co=angular velocity of center of gravity
nozzle design testing to determine nozzle effi-
about the center of the circle
ciency and impulse properties.
248
t TEST PROGRAM
/
_age 248 that the spherical pendulum damper platform motion had not been performed and
dissipates energy at twice the rate of the simple only the approximate results of the steady
pendulum damper. The damping rate of the forced vibration analysis were available. Fur-
damper-platform system is, therefore, the same thermore, the spacecraft design program was
during either type motion. still young and values for spacecraft parameters
It can also be determined from Equations being quoted at that time were changed con-
(9.1) through (9.3) that the total energy of the siderably during subsequent design modifica-
platform during either type motion is directly tions. The numerical results of these early
proportional to the weight of the platform. tests were therefore not directly applicable to
Since the energy dissipation rate of the damper final damper design. They were important
is independent of this parameter, it was reasoned primarily in bracketing damper parameters
that the time constant of the damping rate for more closely than had been possible from the
a given damper-platform configuration should results of the steady forced vibration analysis
be directly proportional to platform weight. included in Chapter 2. The importance of
Therefore, rather than change the platform performing a mathematical analysis of damper-
weight to duplicate the total energy of either platform motion to optimize dampel parameters
the spacecraft or air-bearing fixture, a single rather than using experimental results for this
series of tests with fixed platform weight was purpose was also made evident by the results
conducted. Nutation damping rate of the of these tests:
damper for a particular spacecraft or air- 1. Maximum damping efficiency did not
bearing fL<ture configuration was then calcu- occur when the natural frequency of vibration
lated from the results of these tests. This of the damper was equal to the disturbing
approach seemed advisable in view of the frequency as had been indicated by the steady
constantly changing spacecraft and air-bearing forced vibration analysis. The slight differ-
fixture designs during the development pro- ence in value of these two frequencies at the
gram. The calculated values for damping tuned condition was observed to be influenced
rates using this simplified approach were by the damping constant of the damper.
verified by the results obtained from the analog 2. The damping constant, in turn, was ob-
computer in the mathematical analysis of the served to have an optimum value for fixed
problem. values of the two frequencies. When using
Instrumentation was provided to measure fluid with a kinematic viscosity of 100 cen-
the amplitude and frequency of oscillation of tistokes or greater, the damping rate of the
the platform. A thin metal strip was positioned system decreased with increasing viscosity.
against the edge of the platform in such a way On the other hand, fluid of less than 10-centi-
that any lateral oscillation of the platform stoke viscosity was determined to be too thin
resulted in bending of the metal strip. A since beating, or the alternate transfer of energy
strain gauge was attached to the strip to between the damper and platform, occurred.
measure this deflection. Output of the strain These results indicated the necessity for a
gauge, which was an approximately linear parametric study in order to arrive at an
function of platform displacement, was ampli- optimum damper configuration. This study
fied and fed into a strip-chart recorder. The was obviously more readily conducted on an
strip-chart plot of platform displacement as a analog computer than with the test setup.
function of time was then used to calculate the The series of tests performed using the
rate at which free oscillations of the platform spherical pendulum bob served two purposes;
were damped out. namely, to verify the results of the mathe-
Damping Ey_ciency--In the first series of matical analysis of system motion and to
tests conducted on the pendulum platform test determine experimentally the association be-
fixture, the cylindrical pendulum bob was used tween damping constant of the damper and the
in the nutation damper. At the time of these kinematic viscosity of the fluid used to fill the
tests the theoretical analysis of the damper- damper. This latter determination was neces-
249
ORBITING SOLAR OBSERVATORY o
sary because of the lack of theoretical informa- tion of the platform did not decay exponentially.-
tion available that was pertinent to this From the curves in Figure 9-7, the following
particular geometry. Stokes' equation, relating conclusions can be made:
damping force and kinematic viscosity for a 1. The time constant for the decay rate of the
sphere immersed in a fluid, yielded damping amplitude decreases with decreasing fluid vis-
constants considerably smaller than those cosity, but a moderate variation in viscosity will
calculated from experimental results. This not seriously affect damper efficiency.
discrepancy was due to the small volume of 2. As fluid viscosity (and therefore damping
fluid used. The flow of fluid over the surfaces constant) decreases, the damper must be more
of the cylindrical container probably contrib- sharply tuned to disturbing frequency to obtain
uted significantly to the damping force. Three good damping efficiency.
different series of tests were made with this 3. Maximum damper efficiency occurs when
configuration using 10-, 50-, and 100-centistoke the natural frequency of the damper is slightly
silicone fluid in the damper. With the natural less than the disturbing frequency.
frequency of the damper held fixed, the fie- 4. For the range of fluid viscosities tested, a
quency of vibration of the pendulum platform time constant of twice the optimum value is
test fixture was varied by changing the length obtained when the disturbing frequency is ap-
of the support wires of the platform. From proximately 10 percent above or below nominal
the recorder plot of platform amplitude versus value.
time, the time constant for the decay rate of These qualitative results of the experimental
the amplitude was determined in each case. tests were later substantiated by the results of
The graph of time constant versus platform the mathematical analysis of the problem.
frequency for the tests using 50-and 100- In order to determine the minimum viscosity
centistoke fluid is presented in Figure 9-7. fluid that could be used without encountering
A similar p]ot for the tests using 10-centistoke beating, additional tests were conducted with
fluid could not be made since beating occurred 20 and 30-centistoke fluid. Beating occurred
during these tests, and the amplitude of vibra- using the 20-centistoke fluid and did not occur
when the 30-centistoke fluid was used. Since
only a small difference in time constant could be
detected with a change of 10 centistokes in fluid
r, SEC _CULATED P2 _ CENTISTOKE
FLUID50
viscosity, further optimization of damping con-
stant was not attempted. The 30-centistoke
15 / _,4 -- 50 CENTISTOKE fluid was selected for use in the spacecraft
damper.
To obtain an accurate evaluation of damper
1o 5 NTI_TOKE
performance
instrumentation
nonrotating
a simple setup was devised.
upper section
of a photodetector
cell) mounted
and
This
on the
partially hidden
from the constant-direction light source by a
knife-edge whenever the spin axis of the test
fixture was nearly vertical. Slight variation in
3.5 4.0 4,5
position of the spin axis resulted in variation of
Pl, RAD/SEC the amount of the solar cell surface exposed to
light. Since output current of the solar cell is
NOTE proportional to the amount of light it detects, it
Pl = undamped natural frequency of platform
was a simple matter to use tiffs output signal to
P2 undamped natural frequency of damper determine test fixture attitude during nutation.
With the attitude control system operating
FIGURE 9-7.--Time constant versus platform frequency. and tim wheel spinning at nominal rate, torques
250
TEST PROGRAM
-were applied to the test fixture to induce quency was not feasible because of the difficulty
nutation. Because of its relatively large ampli- of measuring damped frequency of the pen-
tude and low frequency, the resulting motion of dulum in the sealed, fluid-filled assembly.
the fixture was easily followed visually. When Attempts were made to measure damped fre-
the nutation damper was caged, this motion quency with one end of the case removed, but it
continued indefinitely_ With the damper un- was found that the damping constant of the
caged, nutation was quickly damped out, and fluid with the end removed was different from
the fixture remained "at sleep" until another its value in the sealed unit.
torque impulse was applied. Verification of the frequency equations for
The recorder plot of nutation amplitude the pendulum in the upright and inverted
versus time was compared with similar plots positions was accomplished with a spring wire
obtained for the same damper during tests on of sufficient diameter to support the bob weight
the pendulum platform. The time constant of in the inverted position. Frequency measure-
about 8 seconds measured for the air-bearing ments were made by visual observation and the
tests checked satisfactorily with the value use of a stop watch. Within the accuracy of
calculated from the results of the pendulum this measurement technique, the two measured
platform tests, taking into account the greater frequencies were found to agree with the calcu-
amount of energy the damper had to take out lated values. During these tests, the end of the
on the air-bearing fixture for _4brations of equal wire attached to the bob was adjusted to various
amplitude. Thus, these tests proved conclu- positions within the adjustment range of the
sively that the damper would damp out nutation setscrew. This determined the extent of the
of a gyroscopic body at the rate predicted from double pendulum effect created whenever the
the results of the pendulum platform tests and wire end was not positioned at the center of
two-body free vibration analysis.
gravity of the bob. It was found that by com-
Frequency Tests.--Since the frequency of
puting an effective wire length from the meas-
vibration of the damper during orbital flight
ured upright frequency in each case, the
could not be checked by direct measurement in
frequency of the inverted damper could be pre-
the laboratory, it was important that the
frequency equations derived for the damper in dicted with one percent accuracy. The double
the gravity field be fully substantiated by tests. pendulum effect was therefore considered to be
It was reasoned that if the undamped frequency negligible.
of the damper in the inverted position could be
Integrated Control Systems
predicted accurately from the knowledge of the
necessary damper parameters and the undamped The laboratory tests of the gas control system
frequency in the upright position, the effect of and observation of the concurrent behavior of
gravity on frequency would be firmly estab- the sun sensing eyes, torque motors, electronic
lished. control circuits and the nutation damper re-
The procedure for calculating frequency of the quired a freely rotating platform. For this
inverted damper from a known frequency of the reason the entire engineering model of space-
upright damper was also followed in tuning the craft control systems was tested on the air-
damper to nutation frequency of the air- bearing test fixture.
bearing fixture. This was necessary because of The control system was tested for pointing
the low frequency of nutation. The wire pen- accuracy, solar acquisition, spin-up, pitch cor-
dulum arm required to obtain the tuned fre- rection, day-night cycling, data acquisition and
quency of the fluid-filled damper in the inverted data transmission. The test fixture was given
position was unable to support the weight of the wobble during simulated night condition and
pendulum bob in air. The undamped frequency nutation during day conditions. All possible
in the inverted position could, therefore, not be motions of the actual spacecraft were simulated
measured directly. Calculation of undamped and the resulting systems reactions were
frequency from the measured damped fre- measured.
Z)l
782-5520---66-----17
ORBITING SOLAR OBSERVATORY
Many minor changes and adjustments were Other parts subjected to strain gage measure-
made to the engineering model. One unforeseen ments were the arm mounting rim panels,
problem that would have caused the actual decks, and ribs. Other than in the upper wheel
spacecraft to fail was detected as a result of the casting, the highest stress measured was 1100
air-bearing table testing. The problem, an p.s.i, on a deck panel. All other stresses were so
unstabilizing interaction between the pitch low as to be negligible.
control and a large nutational distrurbance was
quite easily solved by the addition of a time Data Acquisition and Command System
delay in the turn-on of the pitch control gas jets.
The subassemblies of this system were sub-
Mechanical Structure contracted by Ball Brothers Research Corpora-
tion. All developmental testing of these sub-
During the developmental testing program assemblies was performed by the subcontractor.
for the control system, a concurrent effort was Ball Brothers Research Corporation subjected
underway to develop the mechanical structure each subassembly to a rigorous qualification
for the spacecraft. After this structure was procedure upon receipt from the subcontractor.
built and assembled, static structural load A detailed analysis of the antenna design was
testing was performed where necessary. Tests made by running patterns of the actual array
were also run to determine that the erection of on an antenna tower utilizing the full scale
the extendable arms at launch spin rates did not structural model spacecraft as a test model.
damage the spacecraft or produce any appre-
ciable wobble input to the spacecraft. PROTOTYPE TESTING
252
TEST PROGRAM
The component vibration test specifications Martin Company in Denver. These tests con-
were as follows : sisted of 1220 minutes of low-level sinusoidal
input sweeps in three axes, 15 minutes of high-
Accelndion Displacemcd
Frepueney rnnge (cpa) (peak g) level sinusoidal input sweeps in three axes, 12
5-17-9 - ------ minutes of random ribration input in three axes,
17.9-50 6. 4 - and 10 minutes of 15 g sustained acceleration
50-1000 15.6 -
100-2000 21.6 -
input in the thrust axis.*
Figure 9-9 shows the structural model in place
T o t d duration-15 minutes.
on the shaker, and Figure 9-10 shows the
W h e e l Structure structural model in place on the centrifuge.
After the fourth complete vibration
prototype mechanical structure, two decks
carrying large, heavy experiments were found
to have cracks in the side flange bend radii.
These failures were considered to be insignificant
because test results were far in excess of flight
load expectations a t this point. However, a
program of careful investigation of the bend
area on other decks was instigated.
There was one other consideration of the
loading under actual flight conditions, the
nature of the loads on the deck was changed by
the thrust acceleration, as shown in Figure 9-8.
C
3 FIGURES9.-Structural model vibration test.
2
-
During these tests no failures or problems
occurred in either casting. The derices used to
lock out both elevnt,ion and azimuth freedom
FIGURE
S8.-Flight loading vibration plus thrust. failed and required redesign, hut, these did not
directly affect the castings.
Superimposing the ribration on the thrust During t,he acceleration tests, seven strain
curre shows that the load is predominantly gage readings were obtained on various points
downward on the deck. Thus, full cycle of the azimuth casting and four readings on the
reversals during launch would not have occurred elevation casting. S o undul3- high stress levels
as frequently as during vibration testing. were indicated.
Furthermore, the time duration of vibration The structural model azimuth casting was
input during flight was only 5 minutes, compared later used for .z x-ibration test of the honeycomb
to 45 minutes of testing time. This point is solar cell pmels. The setup consisted of the
made not to suggest that the design criteria casting hard-mounted on a fixture. wit'h the
should have been changed, but, to indicate th:tt sail structure and solw cell pmels at,tached.
in evaluating the cracks which appeared there The awemblg was not balanced. The test was
was justification in accepting the design :is run for 1 7 minutes with sinusoidal swept inputs
qualified. up to 1 7 g. At this t h e , during a transverse
Elevation a n d Azimuth Castings ~
253
ORBITING SOLAR OBSERVATORY
W E P -E 1
test, the control accelerometer mounted on the ances were changed. There were no further
fixture fell off, and the vibrator exciter at- problems during subsequent testing.
tempted to put infinite forces into the fixture.
The exciter was turned off by overload cutouts, Sail Structure
but the input loads reiiched were not known. No problems of a structural nature were en-
After this occurrence, the azimuth casting w m countered in the sail structure during the
found to be btidly cracked in a lower corner stri~cturslmodel tests. For tliese tests, the
riidius. The crcxk wtis considered t o be clue to original modular stamped solar cell panels were
n tesl failure rather than II component fttilure. used, and two of the modules had dummy solar
cells bonded in place.
Bearings
After the striictiiral model tests, the sail
After the vibrntion and ncceler2ition tests a t striictiire wns removed arid independent vibra-
hftwtin Company, the shnfts and henrings were tion tests were conducted on it. It wns found
disnssembled for inspection. The benrings still that diiring thrust itxis vibration, the horizontal
functioned properly, h i t the tiximiitli sh11ft wns tie members were flexing with lnrge tmiplitudes,
bndly gttlled ~~irtic.iila~.ly
lit t h e lower benring. cniising excessive crosstnlk oiitpilt in the trans-
As tt result of this condition the slirift, fit toler- verse plnne. Stiffeners were ndded to the hori-
254
• TEST PROGRAM
zontal members and the 'test was re-run with Static and Dynamic Balance
satisfactory results. This same test was repeated
The wheel structure had to be aligned to spin
when the honeycomb panels replaced the about the figure axis and the structure had to
modular stamping. Even after the test failure,
be aligned about the spin axis to prevent a
there was no damage to the sail structure. dynamic unbalance which would have caused a
Vibration tests were conducted on the nuta-
wobble in the orbiting payload. The upper
tion damper to evaluate the caging mechanism.
structure had to be dynamically in balance to
As a result of these tests the original design was prevent the night spin-up of the upper structure
modified. At first an 0 ring was utilized from introducing wobble. Dynamic balance
for a bob seat during caging but it was found was also necessary to keep the spacecraft within
that the bob rotated, twisting the pendulum the alignment tolerances of third stage thrust
wire and deforming it. Also, the previous for the stowed launch condition.
design utilized a solenoid for releasing the bob,
Balance Tolerances
but this would not successfully withstand vibra-
tion. The solenoid was replaced with a squib Static Balance--Launch Condition.--Accord-
actuated pin puller unit, and the caging seat ing to Douglas Report No. 7843144, the center
was changed to metal and later serrated to of mass of the spacecraft had to lie within
0.0033 inches of the axis of the master diameter
prevent the bob from rotating during shake.
of the attach fitting. This meant that the
Summary spacecraft static balance had to be within 25
oz-in.
After all prototype subassemblies had re-
Dynamic Balance.--I.aunch Condition.--Ac-
ceived rigorous qualification and testing, they
cording to Douglas Report No. 7843144, a
were installed in the prototype spacecraft and
principal inertial axis had to lie within 0 radians
integration tests were begun. of the axis of the master diameter of the attach
fittings, where
Comprehensive System Tests
L+4.5
Mter all components and subsystems had 0=0.001 ---
been individually tested and successfully inte-
grated, the first comprehensive, covers-off, I, is the moment of inertia about the spin axis
operational system test was performed. This in slug-ft 2, and Ir is the transverse moment of
was an extensive test, in accordance with a inertia in slug-ft _.
For the OSO I it was determined that
formal test procedure, designed to provide
assurance that all subsystems in the prototype /_= 19.7 slug-ft 2
model were working properly, and that no and It---- 15.3 slug-ft',
adverse interactions between subsystems and so 0----0.0055 radian, or 18.9 minutes.
instruments had occurred.
This meant that the dynamic balance had to be
In these tests, every measurable parameter of within 1800 oz-in3
significance to spacecraft performance was
Orbital Condition
measured. All work was accomplished with
the spacecraft mounted on the spin table. In orbit, static balance is not important.
Comprehensive system tests were performed However, all methods producing good dynamic
on the prototype spacecraft before during and balance also produced good static balance.
after environmental tests. Dynamic unbalance of the wheel causes the
Prior to the actual environmental testing, it spacecraft to wobble. If the wobble becomes
was necessary to balance the prototype space- too large, errors in experiment pointing causing
craft, measure the moments of inertia, and de- higher servo power result. Therefore, an
termine the location of the center of mass. arbitrary limit of one minute of wobble was
255
ORBITING
SOLAR
OBSERVATORY
chosen.This meantthat the wheelhadto be one horizontal axis. The apparatus can ba
dynamicallybalancedwithin 144 oz-in. 2 changed to use the lower pair of trunnions,
Dynamic unbalance of the sail would have allowing rocking about a different horizontal
caused a wobble at night which would have axis. If the object is balanced, it has no
become nutation when the upper structure tendency to rock about either axis.
acquired the sun the following day. An To determine the effect of an unbalanced
arbitrary limit of 10 minutes was chosen for condition, a small mass, m, is placed on the
night wobble. This meant that the wheel and upper edge of the cylinder at an angle, a, from
sail together had to be dynamically balanced reference line drawn on the cylinder. This mass
within 2150 oz-in? will cause rocking torques in two ways. First,
the weight of m causes a torque about the rock-
Combined Tolerance ing axis of toga sin (_t_a). Second, the mass
Control over stowed balance and orbital applies a centrifuge force of ma_ _ at a distance
balance is not independent. All the balance _4--_2 above the rocking axis, producing a
requirements were met as follows: torque of ma_2([4--[2) sin (_tTa). The total
1. With the arms extended, the wheel was rocking torque, then, is
balanced within 10 oz-in, static and 140 oz-in. 2
dynamic. ma[gq-_([4 - t2)]sin (_t+x). (9.4)
2. With the arms extended and the sail locked
in the launch position, weights were added to If the mass were placed at the bottom edge of
the cylinder, _4 would be replaced by [3. If the
the sail to balance the spacecraft within 10
apparatus were changed so that it pivoted on
oz-in, static and 300 oz-in? dynamic.
the lower trunnions, [2 would be replaced by Ix.
3. With the arms stowed in launch position
Note that even if a weight is so placed that its
and the sail locked in launch position, the
static and dynamic effects just balance and it
spacecraft was checked to assure a balance
produces no rocking torque about one pair of
within 20 oz-in, static and 100 oz-in. _ dynamic.
trunnions, it does produce a rocking torque
Theory of the Balancing Method. about the other. Thus, if the cylinder is out of
balance, it tends to rock about at least one pair
Consider a right-circular cylindrical object
of trunnions.
that is to be balanced statically and dynamically
Further note that even if the cylinder is
about its figure axis. It is possible to bring this
object into complete balance by attaching two initially out of balance it is possible to place
weights, one at each end. The problem is to some weight on one end of the cylinder in such
find out experimentally what weights to put on a position that the rocking torque about one
and where to put them. pair of trunnions due to the initial unbalance is
Figure 9-11 shows a cylindrical object of just cancelled by the rocking torque due to the
radius a, spinning about its vertical figure axis weight.
at _ radians per second and free to rock about To simplify the notation, let
a]g+_2([4 - [2)]----A,
a[g+_2(ts--t2)]-_B,
a[g_-_2(t3 - ix)l= C, (9.5)
256
TEST PROGRAM
Now shift to the lower trunnions and attaeh Now we remove mn and m2 and place ran
some weight, m2, to the bottom rim at # so that (whose value is now known) exactly where m_
the apparatus does not rock. The two weights was. Then, still pivoting on the lower trun-
together apply a torque of nions, find a weight for the top rim that stops
the rocking. This is mr, and the cylinder is
maD sin (_tWa)+rnaCsin (_t+#). balanced.
Two weights must now be found, one for the top The steps in balancing the cylinder, then, are
these:
rim and one for the bottom rim, such that the
two together will apply the same torque as ml 1. Using the upper trunnions, find a trail
about the upper trunnions and the same torque weight which, when mounted at the proper
as ml and m2 together about the lower trunnious. point on the top rim, stops the rocking. Leave
If ml and m2 are removed, and these new weights it on.
installed, the apparatus will not rock about 2. Using the lower trunnions, find a trial
either pair of trunnions. The cylinder will be weight which, when mounted at the proper
balanced. point on the bottom rim, stops the rocking.
Designate the top weight, Mr, and its posi- 3. Remove and weigh the bottom trial weight.
tion, 0. Designate the bottom weight, rnb, and Replace it with a weight greater by a known
its position, ¢. They apply a torque about factor. Remove the top trial weights.
the upper trunnions of 4. Still using the lower trunnions, find a
weight which, when mounted at the proper
tara sin (o_t-i-O)--i-mn B sin (_otq-4J), point on the top rim, stops the rocking. Mount
it. The cylinder is balanced.
and a torque about the lower trunnions of
Balancing Method
mr D sin (o_t+O)+mn Csin (o_t+_).
Aligning the Tab/e.--Before the spacecraft
Then
was mounted on the balancing machine, align-
ment of the table to the true bearing axis of the
mlA sin (o_t+a)=mrA sin (o_t+O)
spindle was checked with a dial indicator. The
+mnB sin (wt+_), (9.6)
indicator was set up to measure vertical motion
and
4.5 inches from the center of the table. The
m_D sin (cot+a)WmaC sin (_otq-#) table was adjusted for a reading less than
0.0003 inch TIR. The table was then aligned
=mrD sin (cot+O)+mnCsin (_t+_). (9.7) within 0.12 minutes of arc.
257
ORBITING SOLAR OBSERVATORY
This meant that the nubbin was well aligned wheel compartment covers.) The permanent
to the spacecraft bearing axis. Then, with the weights were fabricated and installed.
sail locked, the balancing machine spindle was 6. Steps (2) through (5) were repeated until
rotated and the spacecraft was moved hori- the wheel was in balance within the effect of u
zontally on the table until the nubbin ran within five-gram weight .with only permanent weights
0.0007 inch TIR. This meant that if the nub- installed. This usually required two reitera-
bin was to lie on both axes, with both axes tions, and the wheel was then balanced.
perpendicular to the attach fittings, then the Balancing the Sail.--With the sail locked,
two axes were coincident. The spacecraft was the wheel balancing steps were repeated with
now aligned to the balancing machine. weights being placed on the sail rather than on
Selecting the Pivot Point_.--The best sensi- the wheel. More reiterations were required
tivity and accuracy were attained when _1= [3 because of the difficulty in putting weights
and _2=[,. Since the most convenient places where they were i_equired. This was continued
to mount balance weights on the wheel were until the wheel and sail together were balanced
the top and bottom rims, the two pivot points within the effect of a ten-gram weight. The
(the flexures that take the place of trunnions) spacecraft was then in balance.
were set at the heights of the top and bottom
rims. Checking Balance in Stowed Condition
Tuning the Balancing 2tlachine.--With the With the sail locked and the arms down, a
spacecraft installed and the pivot points set, check was made to be sure that the spacecraft
the tuning springs and weights were adjusted was in balance in the launch condition within
so that the natural frequency of the rocking effect of a twenty-gram weight.
motion about each pivot was within one percent
Moments of Inertia
of the spindle rotation frequency. Since the
operation was moderately damped, closer Spacecraft moments of inertia were measured
tuning was not necessary. to provide numerical data for use in tuning
Balancing the Wheel.-- the nutation damper, the setting of the spin-up
1. The sail was tied to the balancing machine circuit parameters and the spin and pitch
yoke so that it could not rotate. gas iet nozzle ON times.
2. Using the upper pivot, the magnitude and The measurements were made by suspending
phase of the rocking torque was measured. A the entire spacecraft from a point and measur-
weight of the proper size was taped on the ing pendulum motion periods.
wheel at the proper position on the top rim of Spin-Axis Moment.--Figure 9-12 shows the
the wheel to reduce the rocking torque. This experimental setup. The spacecraft was sus-
step was repeated with trimming weights until pended from three vertical wires of length, s,
a five-gram change caused overshoot. attached at a tmrizontal distance, r, from the
3. Using the lower pivot, Step (2) was center of gravity. The spacecraft was caused
repeated on the bottom rim. The bottom to oscillate as a torsional pendulum, and the
weight was then removed and weighed. A frequency of oscillation, w, was measured.
weight 12 times as great was then put in the For smM1 amplitudes of oscillation, the mo-
place of the original weight. ment of inertia is
4. Steps (2) and (3) were repeated until the
wheel was in balance within the effect of a I =mgr2 (9.10)
8¢0 2
five-gram weight for both pivots. This usually
required about three reiterations. where m is the mass of the spacecr, ft and g is
5. The position and size of the two weights the local value of gravitational acceleration.
was noted and a location where permanent A small correction was made for the mass and
weights could be installed to give the same moment of inertia of the wires.
effect as the two temporary weights was found. With the sail locked to the wheel, the mo-
(Usually the best place was the inside of the ment was measured with arms out and with
258
• TEST PROGRAM
?
J
FIGURE 9-12.--Experimental
determination.
setup for spin moment
J
The mass and c.g. location of the bracket and
the mass of the spacecraft were known, so the
c.g. location of the spacecraft could be cal-
FIGURE 9-13.--Experimental setup for CG location.
culated.
This measurement was made with arms
extended and with arms stowed. The effect of transverse moment of the spacecraft was
gas in the bottles was calculated. calculated.
Transverse Moments.--The single wire used This measurement was made with arms
in the c.g. determination was replaced by two extended and with arms stowed, with the
wires (see Figure 9-14) spaced equally from the sail in a vertical plane and the sail in a horizontal
combined c.g. of the spacecraft and bracket.
plane, and with each of the three arms toward
The spacecraft was swung as a torsional pendu-
the floor. The position of the sail made very
lum, and Equation (9.10) was used to determine
little difference. The measurements with dif-
the moment of inertia of the assembly about
ferent arms toward the floor were almost identi-
a vertical axis through the combined c.g.
From this moment and the known moment of cal. The effect of gas in the bottles was
259
ORBITING
SOLAR
OBSERVATORY
Results--
Spin moment, wheel, arms extended:
Bottles full: 24.39 slug-ft 2
Bottles empty: 22.29 slug-ft _
Spin moment, wheel and sail, arms extended:
Bottles full: 27.66 slug-ft 2
Bottles empty: 25.5C slug-ft 2
Spin moment, arms stowed:
Bottles full: 19.71 slug-ft 2
Transverse moment, arms stowed:
Bottles full: 15.26 slug-ft _
Transverse moment, arms extended:
Bottles full: 17.72 slug-ft 2
Bottles empty: 16.56 slug-ft 2
Center of gravity, arms extended:
Bottles full: 12.68 in. above interface
Center of gravity, arms stowed:
Bottles full: 11.85 in. above interface
Environmental Tests
Acceleration Tests
260
TEST PROGRAM
Sinusoidal Test.-
Thrust Axis Specification
Acceleration Vdocity
FrcPwrrcY (CPS) (9) (inisec)
5-50 2.3 ________
50-500 ________ 2.6
500-2000 21. 0 _-____-_
2000-3000 54. 0 ________
Transverse Axis Specification
L
Acceleration
f i e q w n c y (CPd (9)
5-50 0. 9
f.
50-500 2. 1
500-2000 4. 2
2000-3000 17.0
FIGURE+15.--Centrifuge test.
Random Jf otion Thrust and Transverse Spec-
Table 9-1 ijication.-
Test Appr.
Freqlrrnrw Duroiion PSD L e d Accel.
Test Equipment Used for Prototype Testing Vtbratmn Arts Range (cps) ( m m) (qz'cps) (q-rm)
Thrust 20-510 4 0. 07 *5. 85
(Z axis) 510-2000 4 0. 07 *lo. 40
Exciter- - - - - - - -. Ling Model 275 rated a t -
7500 Ih peak force. Total- -8
Power supply - - - Ling Model PP 40/60, 40 Transverse
( S axis) and 20-510 1 0. 07 5. 85
KYA amplifier. 510-2000 4 0. 07 10. 40
Control- - - - - - - -. Ling Model R1007 sine and (Y axis) -
Total- -8 min. (each axis)
random control console.
Tape facility- - - . Ampex Model FR-1000, 7- *Within amplitude limit of vibration generator.
channel F h l record with
2-channel Fhl playback. Test Arrangemenfs
Transducers- Endevco Model 2221C
acceierometex The test arrangements for the various con-
C E C Model 41G2.%re!ority figwxtions are described below.
pickups. Thrusi Ark Tests.--The p r o t o t n e space-
Enderco Model 2106 force craft mas clamped with the actual payload
guages.
Associated Endevco Model 2216
third-stage interface retaining clamp to a test
transducer amplifiers. fixture having the proper matching interface.
I equipment. Endevco Model 2622 power The clamp bolts were torqued to the proper
supplies. level for the actual explosire bolt interface
Mosely Model 5s and clamps as specified by the Douglas design speci-
Model 2D x-p plotters.
Henlett-Packard Model
fication. The test fixture was bolted to the
302-4 wave analyzer. exciter table such that the input forces were
Mo4ey Model GOB loga- applied through the test fixture to the space-
rithmic converters. craft. The test fixture consisted of a 3-inch
Nosley Model SO strip-chart plate which bolted to the exciter, an aluminum
recorder.
Exciter B and Ti Model 1016 servo
casting resembling the frustrum of n cone
Controller controller. tapering from 20 inches in diameter a t the
bottom to 10 inches in diameter at the top, and
a mtchined adapter flange which provided the
Testinfj S13~cific(ltions.-The specified tests to interface fitting. For the effective weight testa
be accomplished and the specifications are as described below, force gauges were inserted
follo\vs. between the adapter flange and the casting.
ORBITING SOLAR OBSERVATORY
262
TEST PROGRAM
263
ORBITING SOLAR OBSERVATORY .
duration and is recorded. A tape loop is cut the effect upon the spacecraft of the actual
and run through the analyzer and further force input expected.
filtering adjustments are made. At this point, Effective weight measurements were made
the specified level random test may begin. by measuring the phase and amplitude of the
Usually minor adjustments are made during forces and accelerations involved. This
the first few seconds of the actual test run due measurement was made only at 550 cps because
to the fact that the tape loop may not contain one of the force gauges became inoperative
enough time for proper statistical averaging of after the measulement had been made at that
the high-level response. These adjustments frequency. The effective weight measured
must be only minor due to the complex over- was 27 lb.
lapping nature of the filtering networks. Transverse X-Axis Effective Weight.--The
The transverse Y-axis random test was effective weight was somewhat variable over
accomplished after the equalization discussed the 550 to 650 cps bandwidth (about 20 percent
above. The test was run with full spin and variation). The average was about 21 lb.
pitch gas bottle pressure. Checkout revealed Transverse Y-Axis Effective Weight.--Since
no failures. the highest effective weight measurement in the
X-axis diJection was measured at 550 cps, the
Thrust Axis Sine Input (Second Run).--The
Y-axis was measured only at that frequency.
test was run with full bottle pressure at specifi-
cationlevel. The Goddard pointed experiment The weight was about 27 lb.
Resonant Burning Dwell Tests.--After effec-
carriage pin was broken. The test had to be
tive weights had been determined for each
repeated.
axis, a slow sweep was run from 550 to 650 cps
Thrust Axis Random Input.--Equalization
in 30 seconds at an input level of 600 lb,
was accomplished and the test was run at
divided by the resultant effective weight for
specified input levels with the gas bottles fully
the thrust axis, and 100 lb divided by the
pressurized. No failures occurred.
resultant effective weight for the transverse
Thrust Axis Sine Input (Third Run).--This axis. The 600 lb and 100 lb figures represent
full-level test was repeated since some equip- the estimated thrust axis and transverse axes
ment had not yet survived the test. This time
components of the third-stage engine thrust
no failures were noted. Leak tests indicated
force, respectively.
no bottle leakage. The problems that were encountered with
Thrust Axis Effective Weight.--Because some both the spacecraft equipment and the scientific
of the third-stage engines used in the Thor-Delta instruments during the prototype vibration
vehicle configuration had been shown in test testing serve to point out the importance of this
firings to resonate during the burning time at a testing. The input levels at which vibration
frequency between 550 and 650 cps, it was testing should be accomplished are argumenta-
considered necessary to test the spacecraft pay- tive, but the necessity for the testing is cel tainly
load extensively at this bandwidth. In order obvious in the light of the test results during
to do this, the effective weight of the spacecraft this program.
had to be known. A complete comprehensive system test was
Effective weight is that quantity derived by
performed after the complete vibration test.
dividing the measured force input to the space-
craft by the resulting measured acceleration Thermal Tests
produced. To do this, it is necessary to
separate into components the reactive and the Thermal tests of the prototype were made in
direct forces, and accelerations as measured the thermal-vacuum chamber, and rough vacuum
for both the fixture and the composite space- was used to expedite thermal changes. The
craft and fixture, and to measure the phase chamber is 10 ft in diameter. Figures 9-20,
angles involved. As a simplified view of the 9-21, 9-22, and 9-23 show the 10-ft.chamber
problem, what is required is the knowledge of and the operating equipment for the chamber.
264
TEST PROGRAM
FIGURE
9-20.-Large thermal vacuum chamher with FIGURE9-22.-Roughing pump system.
32-inch dual-diffusion high vacuum pumping system.
265
ORBITING SOLAR OBSERVATORY
FIGURE
9-23.-Spacecraft on spin table in vacuum chamber.
266
TEST
PROGRAM
at 35° C and a minimum test duration of five portion of the thermal-vacuum test, the cold
days at -- 10°C. cycle was started. The vacuum chamber
Throughout the test, chamber temperature pressure was raised to 30 mm Hg to accelerate
and pressure were recorded hourly and during the temperature change from 35 ° to --10°C.
any temperature and pressure changes that oc- This temperature change required eight hours.
curred between hourly recordings. Temperature of --10°C and pressure of
The temperature sensor mounted on the 1X10 -5 mm Hg was achieved on August 12.
spacecraft wheel hub was used to maintain the The test was successfully run at --10°C and
specified spacecraft temperature throughout pressures lower than 10 -5 mm Hg for a total
the thermal-vacuum test. The spacecraft spin duration of 119 hours. During this time, the
table temperature and the chamber thermal prototype spacecraft was successfully operated
liner temperatures were also monitored. in the same manner previously described.
The first attempt to run the thermal-vacuum Upon completion of thermal-vacuum testing,
test started with the high-temperature cycle a complete and final comprehensive systems
on July 26, 1961, and was stopped on July 28. test was made, and the prototype spacecraft
This test ran for 47 hours at 35°C, and was was completely qualified.
halted to repair spacecraft instruments. This
attempt was not included as part of the required FLIGHT MODEL TESTING
seven-day test.
Subsystems Tests
The second attempt was started July 31.
This attempt was stopped to correct problems All of the flight model spacecraft subsystems
occurring in the spin jet system. The total and major components were tested to the com-
elapsed time of this attempt was 22 hours. ponent environmental test specification. This
None of this time was applied to the seven-day specification required that the components
test requirement. or subsystems be successfully operated before,
The final and successful attempt to run this during (in most cases), and after three-axis
test was started with a pressure of 1 X 10 -5 mm vibration tests and high and low thermal-
Hg and a temperature of 35°C on August 2, vacuum tests.
and ran successfully until complete for a dura-
tion of nine days and three hours. Basic Electronics Paclcages
Throughout this test run, the prototyPe The basic electronics for the control systems
spacecraft was operated continuously in simu- were packaged in three separate boxes. (See
lated orbiting day-night cycles. Real-time Figures 4-14 and 4-15 for two of these.)
data was transmitted to the telemetry ground These are as follows:
station allowing records to be made of all real- Servo Amplifier Box.--This package, mounted
time spacecraft output channels. Complete on the sail structure, consisted of the servo
playback records were made at intervals ap- preamplifiers, lead networks, ac amplifiers, mod-
proximating the orbit intervals expected. The ulators and demodulators, voltage regulators
spacecraft day-night operation was simulated required for the circuits, and positioning readout
by using a timer to operate the day-night amplifiers.
switching circuitry within the spacecraft. Power Amplifier Box.--This package, also
By controlling small lamps inside the cham- mounted on the sail structure, contained the
ber, the pitch control system, the spin-up pulse-width modulators, power amplifiers, pitch
system, and the pitch readout system were control electronics, and necessary voltage regu-
tested periodically. The spacecraft was oper- ]ator circuits.
ated throughout the thermal-vacuum test Wheel Electronics Box.--This package,
according to the OSO I spacecraft thermal- mounted in the wheel, contained the spin con-
vacuum test procedure and the thermal- trol electronics, undervoltage circuitry, day
vacuum test specifications. turn-on circuitry, spin-rate monitoring circuit,
Upon completion of the high-temperature and necessary voltage regulators.
257
782-552 0--66--18
ORBITING SOLAR OBSERVATORY °
Each of these electronics packages was tested versa. These subassemblies were allowed t_
to the flight model environmental test specifi- reach an equilibrium temperature at --20 ° C
cations which are defined as follows: in a cold chamber and then moved directly to
a hot chamber at -_80 ° C and allowed to come
Sinusoidal Sweep Vibration Tests.--The tests
were conducted by sweeping the applied fre- to equilibrium_ The procedure was repeated
quency from the lowest to the highest frequency three times. Angular response curves were
once for each of three mutually perpendicular run before and after the temperature cycling
axes. The time rate of change of frequency to determine that no mechanical or optical
was four octaves per minute making each sweep changes or damage had occurred. Response
about 2.5 minutes duration for a total time of curves were also run before and after the
7.5 minutes on each subassembly. vibration and thermal-vacuum tests.
moves from orbit night to orbit day and vice Calculated tuning frequency, 2.04 cps
268
• TEST PROGRAM
The flight spare nutation damper specifications this adjustment was made, the top pendulum
wer6: screw was locked with a jam screw and sealed
Radius of bob, 0.926 inches with C-7 epoxy.
Bob weight, 0.825 lb The bottom housing was oriented and as-
Diameter of wire, 0.024 inches sembled complete with caging piston and spring,
Pendulum wire length, 3.12 inches but without the squib block. The bottom
Calculated tuning frequency, 2.03 cps housing was then bolted to the upper housing.
Locktite was used to secure the bolts.
Calibration of the pendulum was accomplished The squib block was assembled and two
by suspending the pendulum from a special capstan nuts were adjusted for proper caging
optics bench fixture and adjusting the pendu- position. The squib block was mounted on the
lum length until the natural frequency of the damper and the damper was caged.
pendulum agreed with the calculated frequency. Prior to filling, the damper was carefully
This frequency adjustment was made by count- weighed. After filling, the damper was weighed
ing the frequency of interruption of a light beam again to assure that the damper was properly
by the pendulum with a photomultiplier tube. filled with oil, and the dampers were leak-tested
The output of the pbotomultiplier was compared once again with a helium leak detector.
with a standard frequency on a two-channel The nutation dampers were then connected
strip-chart recorder. Comparisons and adjust- to the filling system, illustrated by the schematic
ments were made until the bob frequency agreed of Figure 9-24 and vacuum-filled with Dow
with the calculated frequency. When the bob Corning 200, 30-centistoke silicone fluid. This
was calibrated, the mechanical adjustment was filling was done with hot, air-free oil in a vacuum
sealed with Locktite to prevent any frequency to avoid air being trapped inside the damper.
change under vibration or temperature extremes. The damper oilwas then cooled to --10°C to
Adjustment of the pendulum bob caging was complete the filling.After the dampers were
accomplished by screwing the bob wire into the filled, the fillingsystem was valved off. The
upper housing until the bob was nearlyrestricted dampers were then heated to 85°C to cause a
from swinging by the brass caging ring. When pressurerisewhich was absorbed by the bellows.
THERMOMETER, °C
VACUUM AND
PRESSURE
GAUGE
MECHANICAL PUMP
259
ORBITING
SOLAR
OBSERVATORY •
The pressure
waslessthan 17 psi. If the pres- using the same method as was used on the proto-
sure in either damper had risen to over 20 psi, type spacecraft.
the bellows would have been judged defective.
Vibration Tests
Without the bellows the damper would have
burst with a slight temperature rise over the The flight model spacecraft was subjected to
--10°C filling tempelature. After the pressure input levels 50 percent less than those used for
test the dampers were again brought to a tem- the prototype spacecraft, and the test duration
perature of --10°C and sealed. Each tempera- was one-half of that used for the prototype
ture cycle was one hour in duration. model. The effective weights measured on the
Vibration test inputs for the nutation dam- prototype were used for the flight model
pers were programmed to correspond to data resonant-burning dwell tests. The flight model
taken from the azimuth casting accelerometer vibration tests went well. The first test was
during tile transverse axes prototype vibration repeated when a lead wire shorted. The lead
test. The duration of the tests were 5 minutes wire had been too tightly routed around a sharp
in each axis. The thrust axis vibration test was corner of a grounded component retainer clip.
of 8 minutes duration. Tile frequencies and The insulation had cold flowed, probably ac-
"g" levels are listed below: celerated by vibration, and shorted. The lead
wire was re-routed and the complete test series
Acceleration
Frequency (cps) (peak g)
was run with no failures.
5-40 2.5 The general concept of environmental testing
40-80 17. 5 is that the anticipated environmen'_ should be
80-100 8. 5
realistically simulated and that the level of
100-300 7. 05
300-3000 11. 25 simulation be increased by some safety factor
margin in order to gain confidence that any
After the vibration tests were completed, the equipment surviving the simulated conditions
nutation dampers were tested to determine that will survive in the actual environment.
the caging mechanism had held tile penulum The general impression of the series of vibra-
bob securely. The caging release squib actu- tion tests performed at Ball Brothers is that the
ators were then fired to cheek the proper opera- test specifications were unduly rigorous. Since
tion of the caging release mechanism. The as much time and money can be wasted in
squib block was then removed, cleaned, and testing to unreasonably rigorous specifications
replaced. The nutation dampers were painted as are wasted in testing to specifications that
with the spacecraft white paint used on most of are too mild, it is recommended that data be
the upper structure which faced away from the gathered from future flights with a view toward
sun in orbit. providing more realistic test specifications.
Thermal-vacuum tests were run with the
dampers at pressures lower than 10 -_ mm Hg Thermal-Vacuum Test
and temperatures of 0°C and S0°C. Each This test was performed to assure that the
temperature was maintained for 24 hours. No spacecraft would operate in the orbita,1 environ-
silicone fluid leakage occurred during the ment. Flight model test specifications required
thermal-vacuum tests. a minimum test duration of three days at 25°C
and a minimum test duration of three days at
Comprehensive System Test
0°C. Test specifications also required a vacuum
The flight model spacecraft was tested in of 10 -_ mm Hg, or less.
accordance with tim same format docuInented Throughout the test the spacecraft hub
test procedure used for the prototype system temperature, the spin table temperature, and
tests. Prior to performing the various tests the chamber thermal liner temperature were
necessary to qualify the ttight model, it was monitored to provide control data for main-
necessary to bahmce the spacecraft. The flight taining the specified temperature of the space-
model was 1)ahmced statically and dynamically craft. The control temperature sensor was the
270
" TEST
PROGRAM
To=wc=horizontal force at P in
Any object which possesses a magnetic dipole
moment tends to act like a compass. That is, the direction of the chain (9.13)
the earth's magnetic field exerts a torque on it
which causes it to tend to line up with the field. tBBRC TN 61-68.
271
ORBITING
SOLAR
OBSERVATORY
STATIONARY
POST
_/ /CHAIN
OUTER CHAIN _ ._
A TTpACII-I#¢NT- / \ / _
INNER CHAIN_ / /
ATTACHMENT _ / /
POINT _
272
TEST
PROGRAM
assumingall errors to be random and inde- Torque .............. (16.0) (50.3) = 805
pendent,theoverallaccuracywasestimatedto dyne-cm
Torque .............. 0.1 NA i H
beabout6 percent. H = 0.229 gauss
Componentsofmagneticdipolemoment
Instrument
Forward Right up
Goddard X-Ray Spectrograph (with compensating magnet)_ --83 358 454 gauss-era s
California Neutron Experiment ......................... Spinning Spinning --236
Upper Commutator ................................... 646 0 0
Alternate Pointed Experiment Commutator .............. o --646 0
Components were chosen on the basis of From the detailed analysis of Equation (9.18)
highest quality, commercial availability, and as applied to the OSO spacecraft, the need was
reliability proven by past history under rocket seen for surface preparations that would have
and spaceflight environmental conditions. An a/e ratios of about 1, 5, and as low as possible
incoming quality control inspection program to cool the solar-cell panel.
was instituted to measure the important electri- Ball Brothers Research Corporation con-
cal and mechanical parameters to assure that structed surface preparations with the follow-
only those components meeting the purchase ing characteristics:
specifications would be accepted.
Paint No. 80U:
Many components used in the control and
communications systems had to be of higher c_--_0.32 to 0.34
quality than was guaranteed by the routine
e=0.32 to 0.34
incoming inspection tests. Therefore, a selec-
tion criterion was established and only com- Paint No. 63W:
ponents that had measured parameters near
the average for that component were selected. a=0.2 to 0.25
to 0.3
Furthermore, those components handling large e=0.83 to 0.85
amounts of power were required to meet param-
eter specifications after being power-aged. Aluminum:
Each component test specification was derived
c_--_0.25 to 0.30|
from a knowledge of the importance of each
parameter in the actual circuit application. _=0.04 to 0.05J _----_5
All component selection was accomplished on
the basis of the requirements of the individual The 63W (white) and the 80U (flat) coating
system application in addition to the basic are enamel-like coatings developed for space-
component quality requirement. craft use. They had high thermal stability,
excellent resistance to ultraviolet damage,
Thermal Stability Tests high film flexibility, low evaporative loss in
The temperature of the spacecraft is depend- high vacuum, and known optical properties.
ent on the radiation heat balance between the When properly applied and cured, these coats
spacecraft and its surroundings and on the heat have excellent durability, water repeHency,
dissipated in the craft by such things as elec- and weatherability.
tronic gear. The rate of change of spacecraft The aluminum is mechanically polished
temperature is described by Equation (9.18): aluminum. These aluminum surfaces are the
plates on top and bottom of the spacecraft
dT¢---2-_ q'--q-° (9.18) body proper.
dr mc
where Optical Properties
275
ORBITING
SOLAR
OBSERVATORY
TO WATER SATURATOR AND FLOW METER
SUN
t_ TO WATER BUBBLER
TO VACUUM PUMPS ._--__ _..
Z
QUARTZ
GUARD VESSEL
m
STEEL TUBING
E
STAINLESS
, ./_ STEEL OR
COPPER OR =_
_'_ ALUMINUM
VACUUM
LUCITE ROD
SURFACE
THE TEMPERATURE
MEASURED
OF TFrIS WARMER o
WALL CAN BE VARIED
FROM 76:K TO 340°K
SCALE
_2_0A_T
rT'-- T 4_ (9.20) when the sample has reached a steady-state
q--,o+_,(1--_o)A,/Ao _ _ 'J
temperature, where
276
- TEST PROGRAM
r TSA.PLE2 1_c I I
SAMPLE 12 100_C
.00_1
,ffi
I .....
SAMPLE 8 100_C
SAMPLE 4 100"C
f I
__._...---- SAMPLE 31 _C
FIGURE 9-29.--Weight loss per unit area in vacuum of FIGURE 9-31.--Weight loss per unit area in vacuum of
BBRC point 63W. BBRC point 80U.
SAMPLE 1 : AIR DRIED, TESTED AT 50_C SAMPLE_ BAKEDAT375°FFOH I HOUR, TESTED AT I00°C
SAMPLE 2: AIR DRIED, TESTED AT 100"C SAMPLE 11: AIR DRIED, TESTED AT 50°C
SAMPLET: BAKED AT 375°F FOR I HOUR, T_STEDATS0_C SAMPLE 12: AIR DRIED, TESTED AT 10(PC
SAMPLES: BAKEDAT375°FFOR I HOUR, TESTED AT100_C
• 08
,07
!
I: SAMPLE 12 100"C
! i
_.oe
I
_05
. I p !
i i i .
"_. 03
: [ SA.PLE. _0_ I _
i i . i ..
.01
FIGURE 9--30.--Weight loss in vacuum of BBRC FIGURE 9-32.--Weight loss in vacuum of BBRC
point 63W. point 80U.
intensity for about 500 hours. (The quartz prepared as if to paint, up to the point where an
envelope of the lamp cuts out wavelengths organic type primer was used. Aluminum
less than about 0.2 p. There was no visible alloys were prepared per MIL-C-5541 (for
evidence of discoloration even on the white example, the Okite process), and were then
(63W), but there was some slight mud-cracking either chromacoated or primed per MIL-C-
of the surface, too slight to seriously influence 15328A before final coatings were applied.
the temperature of the spacecraft. Excellent film flow in spraying was obtained by
thinning slightly with Butyl Cello-solve.
Application of Coatings Air dry produced a Sward hardness of about
The coatings were applied over most clean 10 and after a minhnmn bake of 30 minutes at
metallic surfaces by spraying, flow coating, 400°F gave a maximum hardness of about 25.
brushing, or dipping. A clean (degreased) For any future applications for use above 350°F
surface with all loose particles removed was it is recon_nended that an additional bake-out
necessary for optimum film adhesions. How- of 15 minutes at 480°F be used to reduce the
ever, for hard service the metal surfaces were thermoplasticity or tack. If applications are
277
ORBITING SOLAR OBSERVATORY
not for long-time use in ultraviolet, the addition support combustion. Other samples were im-
of Octasol cobalt drier will lower the thermo- mersed in liquid nitrogen with no noticeable
plasticity. change. Two samples were exposed to outdoor
Samples of the enamel formulation were put environmental conditions for nearly seven
on/_/_8-inch-thick 2024 aluminum. An acetylene months. The samples were in direct sunlight
torch was applied to the unpainted side. The and facing south. No noticeable change in
aluminum substrate melted away leaving a properties was observed. The surfaces re-
paint film free-standing. The paint did not mained remarkably clean of contamination.
278
APPENDIX A
ANTENNA PATTERNS
The following antenna patterns were recorded on the Ball Brothers Research Corporation
antenna range. There are 18 transmitter patterns and three principal-plane receiver patterns.
60 °
300 °
llO °
250 °
240 °
230 °
220 °
2O0
170 °
_' = 0"
Polarization: EO x E_ x
Remarks: Arms Up
279
ORBITING
SOLAR
OBSERVATORY *
0 = +90 °
0° ]0 °
350 °
340
30 °
3300
40 °
320 °
80 °
300 °
70 °
290 °
80 °
280 °
270 °
[10 °
250 °
110
0o
220 °
2]0 °
160 °
200 °
200 °
100 _
170 ° 180_ 190 °
0 = -90° _=0 °
Polarization E0 x_ E_ x
Remarks: Arms Up
28O
" APPENDIX A--ANTENNA PATTERNS
0 = +90 °
350 _ 0° lO °
10 ° 350 _
340 °
30 °
30 °
50 o
310 °
270 °
I10 °
240 c
120 240 °
23O
230 °
220 _ ,0 °
220 Q
210 °
f
I 0 °
160° 200 °
150"
zTo
° 180' zgo"
9 = -90" ¢, = 10"
Polarization E0 x E_b x
Remarks: Arms Up
281
ORBITING
SOLAR
OBSERVATORY '
i0°
350 °
340 °
40°
320 °
60°
300 °
70°
290
290 °
70'
280 ° 80 °
2800
-10--20 30--4(
270 _
100 °
110 °
250 °
110
240
° ,0°
240 °
120
30 °
130 230 °
0o
220 °
21
150 210 °
200 °
160 ° 200 °
0 = -90 ° = 20
Polarization E9 x E_ x
282
. APPENDIX A--ANTENNA PATTERNS
(] = +90"
0° lO °
350 °
270 °
100°
240 °
210 °
200 c
160 °
0 = -90" ,_ = 30
Polarization EO x E_ x
Remarks: Arms Up
283
782-552 O--66_19
ORBITING SOLAR OBSERVATORY •
340 °
40 °
320°
50 °
310°
80 °
280 °
O=0_0'
ZTO_
90 c
llO °
250 °
0o
240 o
L30°
230 o
0o
220¢ 220 a
210 °
oo
200
200 °
190°
¢ = 40 °
Polarization: E8 x E¢ x
Remarks: Arms Up
284
• APPENDIX A--ANTENNA PATTERNS
9 = +90"
330 30 °
310 °
290
270 °
100 °
LIO°
110 250 _
120"
120 240 °
23O
230 °
210 °
200 C _Do
160 200 °
0 =- 90" _ = 50"
Polarization: E6 x_ E_ x_
Remarks: Arms Up
285
ORBITING
SOLAR
OBSERVATORY .
9 = +90 °
350 ° 0 ° 10°
10 ° 350 °
340 °
)ic
330 30 °
30 °
40 °
320 °
310 °
60 °
300 °
70 °
290 °
80 °
2800
,0_=ff
270 °
110 °
110 250 °
0o
120 240 °
230 , .30 °
230 °
140
)o
210 °
zTo
o 180° 190 °
9 = -90 ° ¢ =60 °
Polarization: E0 x E¢ x
Remarks: Arms Up
286
, APPENDIX A--ANTENNA PATTERNS
0 = ÷90 °
_: 0° 1o
°
350 Q
340 °
)ic
330
320
310 °
2<30e
70
260
100 °
.10 °
110 250 °
240 °
30 °
230 °
0o
220 °
17o
° 180° zgo
o
0 = -90 ° = 70 °
Polarization: E0 x_ E¢ x_
Remarks: Arms Up
287
ORBITING
SOLAR
OBSERVATORY
0 = +90 °
350°10
o 0 ° 100°o Isotropic
30 °
30 °
40 °
320 °
60 °
300 °
70 °
290 °
80 °
280 °
90°0=0 °
270 °
O0 °
!60 o
I10 °
250 °
110
!0 °
240 o
120
230 , 230 o
,0 °
220 °
140
210 °
200: 2o
160° 200 °
190 _
17o
° 180° 190 °
0 = -90 ° ¢ = 80 °
Polarization: EO x E¢ x
Remarks: Arms Up
288
• APPENDIX A--ANTENNA PATTERNS
30 °
30 °
40 °
320"
270 •
100
10 °
11o 2s0 °
12o
1.30 230 °
220 °
210 °
200 _ 0 o
160' 200 °
17o
° 180° 190 °
9 = -90" ¢ = 28o.
Polarization: EO _. F__ .X_
289
ORBITING
SOLAR
OBSERVATORY °
G = +90 °
0° )ic
340 °
30 _
330 °
40 °
320°
60 °
300 °
o 70 °
290 290 °
70 _
80 °
280 °
270 _
O0 °
260'
!60°
1o0
llO°
250 °
i20 °
240 o
120
30 °
230 o
130
)_
21o °
160 °
200 °
190
17o
° 180° 10o
o
0 = -90° (h = 290"
Polarization: E9 x E¢0 X
Remarks: Arms Up
290
* APPENDIX A--ANTENNA PATTERNS
340 °
330 30 °
330 °
310
270 °
100e
110
240 o
120
23O
230 °
220 °
210 °
)o
160 ° 200 °
190"
17o" 180° 29o°
O = -90" = 300 °
Polarization: E0 x E¢ X
Remarks: Arms Up
291
ORBITING
SOLAR
OBSERVATORY
0 = +90"
350 0° 10°
3500
40 °
50 °
310 °
60 °
300 °
290° 70 °
290 °
80 °
280 °
270 °
llO °
11o 250 o
:0°
120 240 •
30 °
230 o
0o
220 °
Clo
21 210o
200 ° 0 °
160 ° 20Q °
170
° 180° 1soo
0 = -90 ° _ = 310"
Polarization: EO x E_ x
Curve Plotted in: Voltage
Remarks: Arms Up
292
• APPENDIX A--ANTENNA PATTERNS
0 = +90"
0°
0=0:
$0 °
120
220 °
9 = -90" _ = 320"
Remarks: Arms Up
293
ORBITING
SOLAR
OBSERVATORY
340 °
20 °
330
500
320 °
50 °
310 ° 310 °
70 °
290 °
130°
280 °
LIO°
250 °
110
120°
240 °
230 °
130
220 °
po
210 °
,o o
20O
200 °
17o
° 180° 19o
o
0 = -90" _ = 330"
Polarization: EO X E_ X
Remarks: Arms Up
294
APPENDIX A--ANTENNA PATTERNS
_Isotropic
30 °
50 o
310 =
270 o
LIO°
110 250 °
240 °
230 o
220 o
210 °
0 o
200 °
0 = -90 ° _ = 340"
Polarization: E8 x_ E_ x_
Remarks: Arms Up
295
ORBITING
SOLAR
OBSERVATORY '
9 = +90"
0 ° 10 °
350 °
340 ¢
110 °
250 °
12C 240 o
220 °
160 °
200 °
0 = -90" = 350"
Polarization: E9 x_ E¢ x_
Remarks: Arms Up
296
• APPENDIX A--ANTEb,rNA PATTERNS
: 270"
o"
Isotropic
11o
= 90" e =0"
Polarization EO x E_ x
Remarks: Arms Up
297
ORBITING
SOLAR
OBSERVATORY
340 °
270 °
110 °
250c 250 e
110
240 °
120
SO°
230 °
_0o
220 °
210 °
)o
e = -90 °
_----0 e
Polarization: EO x E_ x_
Curves Plotted in.:-Volta__
Remarks: Arms up
298
• APPENDIX
A--ANTENNA
PATTERNS
90 °
270 o
110 110e
_0 °
240 °
30 °
230 °
210 °
180°
0 = -90" _, = 90"
Pola_ization: E0 x E_ x
Remarks: Arms Up
299
782-552 0--66------20
APPENDIX B
The following curves indicate the results accomplished by the passive temperature control
work.
+40
+30
L)
o
_20
+10
-10
100 200 300 400 500 600 700 8OO 900 1000
ORBIT NUMBER
+40
+3O
L)
o
+20
+10
J
_
0
-10 00
1(
ORBIT NUMBER
3OO
APPENDIX B--PASSIVE TEMPERATURE CONTROL
30-
o
20-
10
0
i
-10 i
0 0 5C 0 70O 80O 9_ _0 1000
ORBIT NUMBER
FIGURE B-3.--Battery No. 3 temperature monitor, channel 4.
+40
+30.
+30-
r,.)
o
+20- +20-
+I0" _+i0-
J _f
_ 0 0-
I
i
i
-10 -10
0 100 200 300 400 0 I00 200 300 400
ORBIT NUMBER ORBIT .NUMBER
FIGURE B-4.--Battery No. 5 temperature monitor, FIGURE B-5.--Battery No. 6 temperature monitor,
charnel 27. channel 28.
301
ORBITING SOLAR OBSERVATORY °
+40
+5O
+40
+30
+30
(9
o +20
(9
C_
o +20
C_
f_ +i0
+10
ol
-I0
0 100 200 300 500
-I0
ORBIT NUMBER
100 200 300 400
ORBIT NUMBER
FIGURE B-8.--Bottom skin temperature monitor,
channel 24.
FIGURE B-6.--Battery No. 7 temperature monitor
channel 29.
+40-
+30-
+40
(9
o
- +20 +30
?
+20
+i0-
[-.
+10
g_
[., 0
-I0
-i0
100 400 0 100 200 300 400 500
ORBIT NUMBER ORBIT NUMBER
FIGURE B-7.--Top skin temperature monitor, chan- FIGURE B-9.--Rim temperature monitor, channel 15.
nel 14.
302
. APPENDIX B--PASSIVE TEMPERATURE CONTROL
+30
_20
C.) P
O +20
-lO
+I0
-10
-I0
100 2(0 300 0 I00 200 300 400
ORBIT NUMBER ORBIT NUMBER
FIGURE B-10.--Hub temperature monitor, channel 26. FIGURE B-11.--Tape recorder No. 1 temperature
monitor, channel 10.
+30
+20
_+10
0
[-,
-I0
0 i00 200 300
ORBIT NUMBER
303
ORBITING SOLAR OBSERVATORY
+40
.+30
o
+20
_+I0
J
0
-I0
0 100 200 300 400 500 600 700 800 9( _0 1000
ORBIT NUMBER
+40
_9 +30
o
+20
+I0
A
0
-I0
}0 300 400 500 600 7 _0 900 1000
ORBIT NUMBER
304
APPENDIX B--PASSIVE TEMPERATURE CONTROL
+40
+30
I
L) i-
o
+20
F_
J
gh
+I0
F
gfl
F_ E
-10
0 I00 }0 300 400 500 600 700 800 9C0 I000
ORBIT NUMBER
+50
! I l
i
t i
+40 i ' !
i I
, !
L) +30- i
o
i
I
j
i
+20- i '
[-,
! ! , i I
_f
+10-
' 1
k]
[.., I t
i
1
-10
0 I_)0 200 300 400 500 600 700 800 900 1000
ORBIT NUMBER
305
ORBITING SOLAR OBSERVATORY
6O
5O
30
2O
L)
o
10
-i0
-20
-3O
-4O
0 160 260 300 450 500 66o 760 800 900 1000
ORBIT NUMBER
3o6
U.S. GOVERNMENT PRINTING OFFICE : t966 0--782-552
"The aeronautical and space activities of the United States shall be
conducted so as to contribute . . . to the expansion of human knowl-
edge of phenomena in the atmosphere and space. The Administration
shall provide for the widest practicable and appropriate dissemination
of information concerning its activities and the results thereof."
_NATIONAL AERONAUTICS AND SPACE ACT OF 1958