After a spectacular week-and-a-half in space, late on 7 December the science crew began the process of shutting down the experiments, storing and securing the samples and closing down the Spacelab module and pallet. Meanwhile, Young and Shaw, assisted by Parker, worked their way through the now-routine procedures to prepare Columbia for her descent to Edwards Air Force Base next day. It was at 11:10 am on the 8th - only five hours before their scheduled landing, as they were configuring the Shuttle's computers for re-entry - that gremlins hit the mission: a GPC failed.
When any of the five computers encounters a problem, the prognosis is bad, because they are responsible for monitoring literally thousands of separate functions during the dynamic phases of the mission. Worse was to come, however: six minutes later, a second GPC stopped working! Young and Shaw successfully restarted the second computer, but all their efforts to bring the first one back on-line proved fruitless.
The mood inside Columbia's cockpit was undoubtedly tense. ''About the time that we were reconfiguring the computers,'' remembered Shaw, ''we had a couple of thruster firings, and the big jets in the front fired, and these big cannons - boom! boom! - shook the vehicle. You can really feel it if you're touching the structure. We had one of these firings and we got the big 'X-pole' fail on the CRT [display], meaning the computer had failed. So I got out the emergency procedure checklist and said 'Okay, first GPC fail. Here's what we do.' And in just [a few] minutes we had another one fail the same way: a firing of the jets and the computer failed.''
Post-mission analysis revealed that one of the failures was due to a sliver of solder that had become dislodged during the RCS thruster firings. Young later told reporters at the post-landing press conference: ''My knees started shaking. When the next computer failed, I turned to jelly. Our eyes opened a lot wider than they were before!'' Re-entry was postponed until the computer problem was resolved. Eventually the 'restored' GPC and two others were used to support re-entry.
Then, at 11:42 am, one of Columbia's Inertial Measurement Units (IMUs) - used for navigational purposes - failed. Shaw remembered waking Young up with the bad news: ''It was the end of John's shift and [he] went down to take a nap. As I recall, Bob came up and was up there with me part of the time. During that timeframe, all of a sudden there starts this kind of [banging] noise. The next thing, one of our IMUs fails and we didn't know why. So John comes back upstairs and says, 'I really appreciate you guys making all that banging noise when I'm trying to sleep.' I said, 'Jeez, John, I've got some bad news. We lost an [IMU]. And John's eyes got this big again, because we've had two GPC failures and now an IMU failure.'' Eventually, after nearly eight hours of troubleshooting - during which the crew was kept prepared for re-entry - at 10:52 pm the de-orbit burn was performed and Columbia began her descent back to Earth. Overall, the Shuttle's performance was nominal, until about four-and-a-half minutes before touchdown, when the temperature of one of the APUs rose sharply.
''Then we had another lesson: 'Never let them change the software in the flight-control system without having adequate opportunity to train with it','' said Shaw. ''There were 'gains' in the flight-control system, and [these] changed depending upon what phase of flight you're in. When you're flying a final [approach], there are certain gains that make the vehicle respond a certain way to the inputs the pilot makes on the stick.
''Then, when the main gear touch down, the gains change and the gains are set up so you can de-rotate the vehicle and get the nose on the ground in an appropriate way. We had done all our training in a simulator with a certain set of gains, and then they changed the flight software and these gains so that when it came time for John to land the vehicle in real flight, the gains were different than he'd done all his training on. Certainly, when John started to de-rotate the vehicle, it responded differently than he had trained on.
''So John's flying the vehicle [and] I'm giving him all the altitude and airspeed calls and everything and you feel this nice main gear gently settling onto the lakebed. From downstairs, the rest of the guys [clapped] when the main gear touched the ground very gently. Then John gets this thing de-rotated and we're down to about 150 knots or so when the nose hits the ground and it goes 'smash'. So [the cheers from the rest of the crew] change from 'Yay' to 'Jesus Christ. What was that?'''
Immediately after touching down at 11:47:24 pm, the GPC previously restored by Young and Shaw failed again. Six-and-a-half minutes later, one of the APUs shut itself down, as did a second unit shortly afterwards. The astronauts did not know it at the time, but one APUs was on fire as Columbia sped down the runway.
''We got called the next day,'' recalled Shaw, ''because we had an APU fire. The reason the first one shutdown was it was on fire and the fuel wasn't getting to the catalyst bed, and so it 'undersped' and automatically shut itself down. Then, in response to that, I configured some of the systems and APUs. So the next one didn't shut down until we actually shut it down. But two of them were burning and we had a generic failure of a little tube of metal where the fuel went through and was injected into the catalyst bed and it cracked. When we shut [the APUs] down and shut the ammonia off to them, the fires went out. We had some damage back there, but the fires stopped. But we didn't know anything about that till the next day!''
Was this article helpful?