KENNEDY SPACE CENTER—When NASA scrubbed the shuttle Endeavour's final launch here on Friday, engineers said there was a best-case and a worst-case scenario. Well, guess what: it was the worst case. The trouble began when an electric heater for the hydraulics system failed to turn on. When engineers opened the hatch into the left aft engine compartment (arrowed in photo), they found the problem lay not in a faulty thermostat (best case: it would have been fixed in a matter of minutes) but in an electrical subpanel called the load control assembly (worst: it takes days to replace and retest). "The power's not getting out to the heaters," said Mike Moses, the launch integration manager.
NASA hasn't set a new launch date yet, except to say it won't be until May 8; even if the repair goes well, the Air Force has May 6 reserved for a rocket launch of its own. The Endeavour astronauts have flown back to Houston, although they remain in medical quarantine (which basically just means no contact with young children or anyone sniffly) at NASA's Johnson Space Center. Support crews from other NASA centers have gone back home, too, and the press corps is dispersing. I got a parking spot in the first row this morning, and the press-conference room—which was standing-room-only on Friday—was only half full this afternoon.
Engineers are in the process of draining fluids from the ship and powering it down. They plan to pull out the glitchy load control assembly on Monday and install a new one Tuesday. It is a box about the size of a microwave oven, weighing 40 to 50 pounds, packed with switches and fuses. It feeds power into nine separate systems, not just the hydraulics but also life support and the main engines. All those systems will have to be retested with the new box in place—a process that takes a full two days. Although a failure of the box would not have shut down those systems, since there are backup power lines, NASA policy is to fly the shuttle only when everything, both primary and backup, is good to go.
In the scheme of things, the loss of a load control assembly is not a major problem. Electronics do fail. Much the same thing happened to the shuttle Discovery during preparations for takeoff in July 1995. To be sure, engineers need to take apart the box to make sure it was indeed to blame. If not, they'll clearly need to do more detective work. "How the failure occurred, we just don't know yet," said shuttle launch manager Mike Leinbach. This is one reason NASA has not set a new launch date—the better to let engineers stay focused on the problem at hand, rather than add the extra stress of a countdown.
If the delay goes beyond May 8, the shuttle may run into a scheduling conflict with the undocking of a Russian Soyuz spacecraft from the space station, and the final shuttle launch—of Atlantis in late June—may have to be pushed back.
Image: Courtesy of Ken Kremer