Computers in Spaceflight: The NASA
Experience
- - Chapter Three -
-
- The Skylab Computer
System
- [66] Skylab, America's
first orbital workshop, carried a highly successful computer
system. For much of the operating life of the space station, the
computer was not just the fourth crew member but the only crew
member. It made a large contribution to saving the mission during
the 2 weeks after the troubled launch and later helped control
Skylab during the last year before re-entry. The entire system
functioned without error or failure for over 600 days of
operation, even after a 4-year and 30-day interruption. It is
significant as the first spaceborne computer system to have
redundancy management software. The software development for the
system followed strict engineering principles, producing a fully
verified and reliable real-time program.
-
- The record of the computer system stands
in contrast to that of the workshop itself. NASA launched Skylab
on May 14, 1973 on a Saturn V booster. The first two stages put
the modified S-IVB third stage into orbit. The S-IVB contained the
workshop, which included a solar telescope mount and living and
working quarters. The plan was to launch the first crew the next
day aboard a Saturn IB carrying an Apollo command and service
module. However, shortly after achieving orbit, telemetry from the
unmanned Skylab indicated that one of the two wings of solar
panels was missing and the other had not deployed. The panels on
the Apollo Telescope Mount (ATM) had opened properly but they were
too small to supply power for the whole workshop. In addition, the
gyros were drifting and the thermal shield was damaged. These
failures caused concern that the interior of the space station
would overheat and destroy the equipment. The damage was so
serious that for the first 3 or 4 hours the ground controllers
felt that NASA would be fortunate if the systems were to function
for 1 day1. However, by using the computer system that
controlled the workshop's attitude, the ground controllers were
able to keep the Skylab at angles to the sun such that the
equipment would be exposed to tolerable temperatures in the
laboratory in concert with generating adequate power from the
remaining solar panels. At times these were conflicting
requirements. This had to be done for 2 weeks while engineers
prepared repair materials for the crew to fix the workshop.
Controller Steven Bales remembered that time as "the hardest 2
weeks I have ever spent," since a 24-hour watch had to be
maintained on the attitude and temperature2.
-
- The computer system again served as
"captain" during the entire Skylab reactivation. The workshop
systems were shut down on February 9, 1974, after the last crew
left. NASA expected that the Skylab would stay in orbit until the
mid-1980s. By that time the Space Shuttle would be operational
and, it was thought, could be used to bring up rockets to boost
the laboratory into a higher orbit. However, unexpected solar
activity in the mid-1970s resulted in an increase in the density
of the atmosphere, so the Skylab's orbit decayed at a much faster
rate than projected3.
-
-
-
[67]
-
-
- Figure 3-1. Skylab in orbit. Note
the foil sun shield above the center section and the missing large
solar panel. The Apollo Telescope Mount is the section with the
"windmill" solar panels. (NASA photo 74-H-98)
-
-
-
- By 1978, the predicted re-entry time was
to be late that year or in early 1979. NASA decided to attempt to
change the attitude of the workshop so that minimal drag would
ensue. In this way, the orbit might be maintained until the
Shuttle could rescue the space station. Engineers reactivated and
reprogrammed the computer to maintain the proper attitude and,
later, to control the re-entry when NASA abandoned the attempt to
maintain orbit. They accomplished this over 4 years after the
computer was shutdown.
-
- The need for the computer system that
served Skylab so well was not apparent until the original "wet
workshop" concept (the laboratory to be assembled in space inside
of the empty propellant tanks of the last stage of the launch
vehicle) had progressed through more sophisticated designs to the
eventual "dry workshop"4. In December 1968, NASA decided to acquire a dual
computer system to help control attitude while in
orbit5. Attitude control was crucial to the success of the
solar experiments. In fact, the name of the computer reflects
this: Apollo Telescope Mount Digital Computer (ATMDC). Two of
these computers were a part of the Skylab Attitude and Pointing
Control System (APCS), which consisted of a number of other
components, [68] such as an interface unit, magnetic tape memory,
control moment gyros, the thrustor attitude control system, sun
sensors, a star tracker, and nine rate gyros6.
-
- Marshall Space Flight Center devised this
complex system-a pioneering effort because it represents the first
fully digital control system on a manned
spacecraft7. Its mission-critical status led to the use of
extensive redundancy in its design, in both hardware and software.
The computer system not only managed its own redundancy, but all
redundant hardware on the spacecraft8. The uniqueness and complexity of the control laws
associated with the control moment gyro attitude system led one
NASA engineer to refer to it as "a crazy
animal"9. It was up to the Skylab computer system to tame
it.
