“Right at liftoff, we got several alerts”
How close the crew of STS-93 came to disaster still remains unknown, but astronaut Steven A. Hawley explains the moment engine controllers were disabled by an electrical short
What happened?
About five seconds after liftoff an electrical short disabled the centre engine's primary digital control units, along with the right engine's secondary digital control units. Meanwhile, a shortage of fuel during the ascent of STS-93 was traced to a huge hydrogen leak. This could have caused the engine to dangerously overheat.
What mission were they on?
Hawley was on STS-93 on a primary mission to deploy the Advanced X-ray Astrophysics Facility (now known as the Chandra X-ray Observatory). “We had scrubbed two previous attempts at launching STS-93: the first was on 20 July 1999 due to a data spike in hydrogen pressure just before main engine ignition, and the second was on 22
July when there was a lightning storm. But a day later, we were ready to go.
“By this point, Columbia had flown 25 missions, but right at liftoff we got several alerts that, taken together, suggested an AC electrical short. Mission control had notified us right away that it was a short and that one of the consequences was loss of redundancy in the engine controllers on two Space Shuttle main engines (SSME). The loss of another controller on one of those engines would have shut down the SSME. No single-engine failure would result in a bailout.
“I was looking at our procedures for a contingency abort should we lose two engines, although I wasn't particularly worried about that really happening. In flight, I didn't think that was likely because the Shuttle systems are robust and I had a lot of confidence in their reliability. Of course I didn't know the cause of the AC short so, in retrospect, another AC short was probably more likely than I had assumed at the time.
“At liftoff, however, another failure had happened which we didn't know about at the time. A gold pin had been dislodged from a liquid oxygen tank (LOX) post during the main engine ignition sequence and, as it was violently ejected, it struck the inner surface of the engine nozzle. This damaged several cooling tubes containing hydrogen, and the damage allowed hydrogen to leak during ascent (it was leaking 1.5 kilograms [3.3 pounds] of hydrogen every second). Consequently we had a low-level cutoff of the three SSMEs for only the second time in Space Shuttle program history.
“I could tell immediately from the onboard displays that we were lower than we were supposed to be at Main Engine cutoff (MECO). However, I knew they were high enough to be able to do the mission. I wondered why we were short, but I didn't dwell on it since it didn't appear to have any mission impact. The digital control units that were not shorted operated the centre and right engines.
“But in a day or so, the ground was able to tell us about the fuel leak. One interesting consequence of the AC short was that the secondary controller (and I don't recall now which engine it was) had a bias in the pressure transducer which controlled the mixture ratio in a way that actually used less fuel than planned.
“Without the AC short the fuel leak would probably have had a greater effect. As it was, we achieved our intended orbit and completed the mission as planned. After the flight I found out that the ground controllers were aware of the engine performance issues during ascent, but we weren't on board.
Poorly routed wiring rubbing on an exposed screw head was found to be the cause of the electrical short. They did a good job of figuring it out.”