(source: Electronics World, Mar. 1966)
By DONALD A. IMGRAM [OAO Project Engineer, Grumman Aircraft Engineering Corp.]
Probably the most complex satellite built to date, the Orbiting Astronomical
Observatory uses the latest electronic techniques to enable scientists to
see the universe without interference.
THE Orbiting Astronomical Observatory (OAO) Is the largest, heaviest, and
most electronically complex satellite being built under NASA's unmanned
observatory program. Astronomy is the OAO's mission, and why NASA should
be willing to invest over one hundred million dollars, literally in pursuit
of the stars, is not too difficult to understand. All heavenly bodies emit
energy at frequencies spanning the electromagnetic spectrum. But the atmosphere
which surrounds the earth is a hazy, shimmering veil which absorbs much
of the radiation which reaches us from outer space, and only a small portion
of the incident radiation is unaffected as it attempts to penetrate the
atmospheric barrier. Hence, all current astronomical theories are based
almost exclusively on information received by ground observers through a
narrow optical "window." Just how much present concepts will be
shaken when man can observe the entire universe from an unobstructed vantage
point high above the atmosphere is a matter of speculation, but who amongst
us could have imagined the contrast between the two photo graphs making
up Fig. 1? Similar revelations undoubtedly await the scientists preparing
experiments for use on board the OAO, once detection equipment is placed
outside the earth's atmosphere where the view will be unobstructed com pared
with terrestrial observation as shown in Fig. 2.
The OAO
Each OAO, and it is expected that as many as ten launches will be made
over the course of the next decade, is composed of two main elements. These
are the spacecraft itself and the associated experimental package.
The spacecraft system consists of four electronic subsystems. These are
the stabilization and control, data-processing, communications, and power
supply subsystems. The functional interplay of each can best be understood
by detailing the operation of the satellite after it is inserted into ES
in tended orbit. Immediately following booster separator, the stabilization
and control system halts the tumbling motion induced by separation forces
and begins the initial stabilization and orientation process by seeking
and acquiring the sun. Both of these functions are accomplished simultaneously
by two different kinds of sensors. The first is a system of three rate gyros
that senses tumbling rates about the three spacecraft control axes pitch,
yaw, and roll). In the second system, error signals a-e generated by a system
of eight "coarse" solar sensors. These are silicon solar cells
which when illuminated produce an output that is a cosine function of the
angle of incident sunlight on the cell. The eight cells have hemispherical
fields of view and are installed such that four provide pitch axis error
information while the other four sense yaw motion. No roll control data
is provided by the solar sensors. Two of the four eyes for each axis are
paired on the front (anti-sun) and rear (sun-pointing) faces of the satellite.
The polarity of one cell in each pair is reversed with respect to d at of
LE mate, and the outputs of each pair are summed algebraically. The resultant
composite signals are roughly sinusoids with the zero (null) crossing points
corresponding to zero displacement of the spacecraft from the sun line.
Driving signals in the form of error rate and error displacement are hence
available and can be used to position the observatory so that at the conclusion
of the coarse solar sensing and the rate stabilization phase, the spacecraft
is aligned with the sun line. All spacecraft torquing during this phase
is accomplished using the high-thrust gas-jet system whose operation is
similar to those used in the Mercury and Gemini spacecraft.
Ultimately, the error angle will be reduced to zero and the spacecraft
will eventually approach a limit cycle motion of ±2 degrees with a residual
rate less than 0.03 degree per second. At this time the spacecraft will
have its rear face pointing towards the sun.
Mounted on the rear lice of the observatory is a solar-cell "disable" eye
having a restricted field of view (±-10 degrees).
Also affixed to a common fine-pointing assembly are eight additional solar
cells. Their function is to provide finer solar-sensing control signals
which are used until the spacecraft alignment to the sun line is maintained
at ±0.25 degree. Upon achieving the fine solar pointing, the spacecraft
begins one of the more challenging stabilization operations that the OAO
will be called upon to perform. Termed the "roll search" maneuver,
its success depends upon proper functioning of the spacecraft's star trackers,
the most-critical components on board the OAO.
---- Fig. 1. A hint of things to come is shown in this pair of photographs
of the Sagittarius Constellation. The two pictures are identical except
the one made using near ultraviolet light (right) discloses considerable
detail not re corded by the visible light emanating from the same source.
Many similar spectacular pictures will be made by OAO.
Six gimballed star trackers are employed on the OAO, and each is an electro-mechanical/optical
device used for the detection, acquisition, and tracking of selected "guide" stars.
Guide stars chosen for the OAO total 31 in number, all having an apparent
brightness (S-4 response) of second magnitude or brighter. Operation of
a tracker is shown in Fig. 3.
The heart of the device is a 3.5-inch-diameter aluminized beryllium mirror
with a focal length of 5 inches. Light entering the telescope-like barrel
is reflected off the main mirror onto a prism beam splitter having two secondary
plane mirrors offset at an angle of 45 degrees with respect to each other.
The offset arrangement permits splitting the incident light into two beams,
each of which is used to provide error data about a specific spacecraft
control axis (i.e., pitch and roll). The two beams subsequently pass through
slits in two orthogonal vibrating reeds and illuminate the cathode surface
of a ten-stage photomultiplier. Since one reed vibrates at 350 cps and the
other at 450 cps, appropriate circuitry can be used to obtain two-axis control
information which facilitates automatic tracking of the guide stars. Filtered
outputs from the photomultiplier are synchronously demodulated to yield
d.c. voltages whose polarity is a function of phase (direction of displacement
of the star image) and whose amplitude is a function of the displacement
of the star from the center of the tracker's one-square-degree field of
view.
Star-tracker usage begins immediately upon the completion of the fine sun-pointing
mode. After the observatory settles into the +- 0.25 degree limit cycle,
the roll gyro output is biased to induce a roll search motion about the
sun line.
While the OAO is rolling, each tracker sweeps out a one-degree-wide circle
on the celestial sphere. Careful pre-selection of gimbal angles-each tracker
can be gimballed through 45 degrees on each axis-is performed before launch
on ground computers such that at a specific and unique roll angle, all non-occulted
star trackers simultaneously detect guide stars. When this occurs, the trackers
generate star-presence signals which cause the gyro bias to be removed and
allow the trackers to lock onto and track the stars. Sub sequent observatory
control is accomplished using stellar control, the solar sensors and gyros
serving no further purpose.
All experimentation will obviously not be done with the OAO aligned with
the sun line; hence, the observatories are provided with a capability for "slewing" to
any arbitrary position in space. Imagine that the spacecraft is aligned
with the sun line and that it is desired to reorient the OAO to a position
30 degrees away, say to point the main experiment telescope at a star in
the region of Orion. The first required operation is that ground operators
at the OAO Central Control Station at Goddard Space Flight Center in Greenbelt,
Maryland develop a series of spacecraft commands on a digital computer.
These commands will be sent by teletypewriter to one of the three OAO remote-control
stations located at Rosman, North Carolina; Quito, Ecuador; and Santiago,
Chile. These sites were chosen and equipped with special OAO ground-support
equipment because at least one of these stations has a 10-minute r.f. contact
with the satellite once every 100-minute orbit. In turn, the remote stations
relay the appropriate commands to the observatory. When received on board
the spacecraft, the commands are routed to the data processing subsystem.
The means by which the OAO slews from one region of the sky to another
are three "coarse" inertia wheels. Inertia wheels, positioned
one on each control, are simple a.c. motors which are accelerated and braked
as a result of commands sent from the ground. When, for example, the pitch
wheel is accelerated, the spacecraft will be caused to rotate about its
pitch axis but in an opposite direction in accordance with the principle
of conservation of angular momentum. Braking the wheel will cause the spacecraft
to cease rotation. By careful calibration of the wheel's inertia properties,
acceleration, and braking characteristics, it is possible to accurately
predict the slewing actions of the spacecraft in response to wheel motion
and hence to orient the OAO to any desired position in space.
The observatory is also equipped with a set of three "fine" inertia
wheels which are used to maintain the satellite in a stabilized attitude
once the experiment is pointed at the region of interest in the celestial
sphere. When supplied with signals directly from the star trackers, the
wheels can stabilize the spacecraft to any arbitrary position in inertial
space with an accuracy of +-1 minute of arc and restrict drift off this
pointing to rates less than 15 arc seconds in 50 minutes of time. Some later
spacecraft will be equipped with a "fine error sensor" as part
of the main experiment optics. Use of this device will enable future OAO's
to be accurately attitude stabilized to less than ±0.1 second of arc.
Spacecraft Commands
Commands may be sent to the OAO as either "real-time" or "delayed-mode" (stored)
commands. Real-time commands are used when the observatory is within line-of-sight
of a ground station and are executed immediately. Delayed-mode commands
may also be executed during line-of-sight contacts (at some time after their
initial decoding on board the spacecraft), but generally the delayed commands
are placed into memory storage for subsequent execution after the satellite
terminates its ground contact. The recognition by the spacecraft as to what
type of command has been received is accomplished in the command decoder
and distributor portion of the primary processor. A unique feature of the
OAO is its ability to verify commands transmitted from the ground. To understand
how this operation takes place, it is necessary to review the OAO's command
format. The entire OAO data-processing system is digital in nature. Commands
are sent to the OAO as a series of four 32-bit command messages. The four
words consist of the first and second command words, and the first and second
verification words. Verification words are the "ones" complement
of the command words, meaning that, for example, if the first command word
began its message by the binary bits "11010," the first verification
word would be registered as "00101." This format is employed so
that on-board verification of received commands can be accomplished by comparing
the command word and the verification word on a bit-by-bit basis. Only if
an exact comparison of the bits can be made is the command acted upon by
the satellite.
Commands are received by the OAO's command receiver at a 1-kc. rate. The
spacecraft's system clock, which is a crystal-controlled sine-wave oscillator
supplying a fundamental clock frequency of 1.6 mc., provides the basic observatory
timing reference. A shaper-driver network forms the clock signal into a
square wave and transfers the resultant waveform to a ten-stage frequency
divider. Outputs from the frequency divider are decoded to provide clock
pulses and bit gate for use by all OAO equipments. Received commands are
synchronized to the primary processor's 50-kc. bit time rate and are verified
on board as well as retransmitted to the ground for a further verification
(or "echo check"). When it is desired to store a command for later
use, the message is entered into the OAO's command-storage unit.
Command-storage capability is included to facilitate quad redundant storage
of 128 two-word commands (verification words are not stored since they are
of no further use after a command has been validated). The storage medium
consists of the MARS (Multi-Aperture Reluctance Switch) de vices. These
are tiny ferrite cores which allow random-access, non-destructive readout
capability. The observatory also incorporates a 200,000-bit data-storage
unit which is used to record both spacecraft as well as experiment data.
Again, the MARS devices are utilized. The entire data memory can be used
to store information in a non-redundant fashion, or the memories can be "halved" to
permit redundant storage of 100,000 bits of data. The Gemini computer was
derived from the OAO data-processing unit.
Fig. 2. The earth's atmosphere absorbs certain portions of the electromagnetic
spectrum.
Once above this medium with the OAO, total spectrum observation is very
easily attained.
Fig. 3. The star tracker splits incident starlight into two beams, each
modulated at a different frequency by the vibrating reeds to provide error
data about a specific axis.
Fig. 4. Block diagram of the spacecraft communications system.
Fig. 5. This special telescope permits detailed observation of the ultraviolet
spectrum in two-angstrom-wide portions.
Spacecraft Electronics
As in all spacecraft, data transmission is vital. In the OAO, the spacecraft's
command receiver (CRE) is part of the communications subsystem, a block
diagram of which is shown in Fig. 4. R.f. signals are received in two channels,
corresponding to the two v.h.f. slot antennas. The command signal sent to
the CRE is a 148.260-mc. v.h.f. carrier, amplitude-modulated by the command
modulation. Command modulation consists of the superposition of two subcarriers.
One subcarrier, known as the command-message modulation, is frequency-shift
keyed (FSK) in accordance with the sequence of "1's" and "0's" in
the message. The other, known as the message rate clock modulation, is a
sinusoid at 1042 cps. The FSK subcarrier may be considered to consist of
the alternate existence of two audio tones of equal amplitude, referred
to as the "1" tone and the "0" tone. The command modulation
is formally known as pulse-code modulation, non-return-to-zero, frequency-shift
keyed, amplitude modulation or [PCM (NRZ) /FSK/AM]. Also part of the communications
system are the radio-tracking beacon and the wide- and narrow-band telemetry
transmitters. Operating at 136.440 mc., the beacon permits accurate orbital,
tracking of the space craft. The narrow-band transmitter (NBT) is used for
space craft telemetry data and for the echoing of commands for ground verification.
In addition, although it is not a primary mode, the narrow-band transmitter
can be used for experiment-data transmission. The output of the NBT is PCM
(split-phase) phase-shift keyed (PSK ) and is sent at a rate of 1042 bits
per second. The NBT, like the wide-band transmitter (WBT), is a completely
solid-state unit. Each NBT consists of a crystal-controlled oscillator,
a phase-shift keying modulator, an FM isolation and amplification stage,
three stages of class-C amplification, a driver stage, a power output stage,
a varactor doubler stage, RFI filter, and telemetry output circuitry. The
WBT is capable of transmitting either analog or digital data. A video bandwidth
of 62.5 kc. is used for analog information, with a total transmission bandwidth
of 250 kc. Digital data can be transmitted as PCM /NRZ at a rate of 1042
bits per second in real time if it is received from experimenter's data-handling
equipment, or at 50,000 bits per second from the data-storage unit or directly
from the experiment. Power output of the WBT is eight watts over temperatures
ranging between 0 and 130°F. The transmitting frequency is 400.550 mc.
Completing the spacecraft's electronic subsystems is the power supply subsystem.
All power used on board the space craft is generated by the satellite's
solar array, which is capable of generating approximately 1000 watts under
the most favorable conditions with respect to incident sunlight.
The array employs p-n cell types of between 13% and 14% efficiency, and
the total area devoted to solar cells is approximately 114 square feet.
Each of the three OAO batteries consists of 22 series-connected, sealed
nickel-cadmium cells of the 20-amp-hour size. Each battery is capable of
supplying approximately 535 watt-hours of energy. Control of the battery
during recharging is a critical operation and is performed by the battery-charge
and sequence controller. A regulator-converter, which provides regulated
d.c. outputs of several voltages, accepts unregulated 23 to 34 volts d.c.
from either the array, if the spacecraft is in the sunlight, or the batteries
if the spacecraft is occulted from the sun. Two multivibrator type power
oscillators (master and slave) convert the un regulated d.c. inputs to square-wave
pulses at 2 kc. The combined outputs of the master and slave oscillators
are added in an autotransformer connection in the slave oscillator and a
secondary connection in the master unit. The resulting output is full-wave
rectified and filtered through diode rectifiers and LC section respectively.
Voltage regulation is achieved by varying the phase angle between the master
and slave oscillators. The a.c. demands of the inertia wheels as well as
the gyros are supplied by an inverter.
All spacecraft electronics are mounted peripherally around the central
structural cylinder which houses the main experiment package. By employing
this structural arrangement, it is possible for the OAO to accommodate a
variety of experiment packages without substantial changes to the basic
spacecraft design. Hence, we find that the spacecraft is often referred
to as one of the family of "streetcar" satellites, meaning that
it accepts a variety of "passengers."
Experiment Packages
All current experiments are intended to explore regions of the electromagnetic
spectrum that are essentially invisible to earthbound observers. A total
of nine experiments will be included on board the first four OAO launches.
A few of the key experiments will be discussed here. The initial OAO will
have an experiment whose function will be to gather spectral energy distribution
information on selected stars and nebulae in the ultraviolet range (100
to 4000 A). As a secondary function, the experiment will measure time-varying
spectral-intensity data on particular stars. A total of seven observing
instruments will be used to take these measurements. Four stellar photometers,
each of which covers a bandwidth of approximately 1000 A, are used. In turn,
each photometer is equipped with a programmable filter to further subdivide
the coverage into 250-A bands. Also provided in the first OAO will be two
scanning spectrometers. One covers the range of 1000 A to 2000 A while the
other scans between 2000 A and 4000 A. A unique cycling device permits scanning
the range in 100 steps, yielding intensity bandwidth of 10-A to 20-A wide.
The final instrument is a nebular photometer capable of measuring the spectral
intensity of star clouds as observed through five programmable filters,
each covering 600-A bandwidths between 1500A and 3800A.
Another experiment is intended to measure the brightness of 50,000 main-sequence
stars in the ultraviolet region between 1200 A and 2900 A. Called Project
Celescope, the experiment contains four independent Schwarschild telescopes,
each employing an imaging "uvicon" ( ultraviolet vidicon) system.
The bandwidths to be surveyed are 1200 A to 1600 A, 1300 A to 1600 A, 1600
A to 2900 A, and 2300 A to 2900 A. Be cause of its high rate of data output,
this experiment will be done only during real-time contacts. A typical experiment
routine would call for the warm-up of calibrator lamps in the five minutes
just preceding ground contact. Three minutes of calibration will follow
after con tact has been made. To use the Celescope, the uvicons are scanned
in the digital-direct mode, and the first picture, including calibration
information, is transmitted to the ground via the WBT.
Then, one or more standard Celescope data sequences are commanded. Each
sequence includes a 60-second exposure and digital-direct scan for each
camera.
A two-degree by two-degree region is mapped, with some overlap for ease
in data interpretation.
Still another experiment will examine approximately 14,000 stars a year,
at first producing a resolution of 2 A, which later will be upgraded to
resolutions be tween .04 A and .05 A. Operation of this unique telescope
is shown in Fig. 5. Ultraviolet light from a stellar source enters the telescope
and is reflected off the primary 38-inch-diameter beryllium mirror onto
a smaller quartz secondary mirror which converges the light onto the en
trance slot leading to the spectrometer mirror. The spectrometer mirror
directs the light to the spectrometer grating.
The grating is a ruled square which disperses the light and sends it back
to the spectrometer mirror and thence to a series of exit slits in a grating.
The light then passes through the 2-A slits in front of six photon-scintillation
detectors. At appropriate times, the grating is synchronously rotated to
facilitate illumination of the detectors by different portions of the spectrum,
thereby permitting scanning of all the spectrum between 1000 and 4000 A.
Additional experiments will continue to survey the ultraviolet region while
others will experiment in the x-ray and gamma-ray ranges.
The OAO program may prove to be one of the most interesting scientific
projects yet conceived.
|