Scrutiny on NASA Decision Process
A generation after the space shuttle Challenger disaster, NASA’s decision-making process is again under scrutiny amid concerns that a record of survivable damage to the shuttle’s thermal protection system may have lulled the agency into a false sense of security.
“There is a lot of deja vu here,” said Robert B. Hotz, a retired aerospace editor who served on the 1986 presidential commission that investigated the Challenger accident. “They knew they had a problem, but they lived with it. It’s an old issue in flight that if you get away with it once, and you get away with it twice, it can come back and bite you.”
NASA officials say they doubt that the Columbia tragedy was caused by damage from a 2 1/2-pound chunk of insulating foam that struck the orbiter 80 seconds after launch. In more than 100 previous missions, shuttles have survived nicks, cuts and gouges from insulation and other debris falling off the huge external fuel tank.
After the hit on Columbia was detected, engineers and managers spent 11 days trying to assess its consequences. At issue was the extent of damage to U-shaped reinforced carbon panels on the leading edge of the left wing and to black insulating tiles under the orbiter.
Ultimately, it came down to the educated guesswork of experienced people. NASA concluded that damage to the panels on the leading edge of the wing would be inconsequential.
Experts believed that damage to the tiles could range from modest (a missing tile) to fairly serious (an area of missing tiles 7 inches by 30 inches). Some localized structural damage might occur during the searing heat of reentry, but in no event would it result in the loss of Columbia and its astronauts, they concluded.
That conclusion was formally recorded in mission documents on Jan. 28, four days before Columbia was due to return.
NASA’s damage estimate is now a source of many “what ifs.” What if NASA had spotted a catastrophe in the making? Could a rescue have been launched?
Many experts say help could not have gotten there in time. During deliberations over the damage estimate on Columbia, some engineers raised objections, said shuttle program manager Ron Dittemore. However, he described these “reservations” as normal differences of opinion over technical details. No one disagreed with the conclusion that it was safe for Columbia to come back home, he said.
The exact nature of the disagreements is certain to be an area of focus for the board that will investigate the Columbia accident.
The risk that Columbia would suffer an accident was probably higher than the space agency concluded, said an independent expert who reviewed some of the underlying analyses NASA used.
But Elisabeth Pate-Cornell, a risk analysis expert at Stanford University, also stressed that she found NASA’s conclusion acceptable.
“I would have qualified it a little bit more,” said Pate-Cornell, who previously conducted a major study for NASA on the vulnerability of the shuttle thermal protection system. “I would have said that the possibility of an accident had slightly increased.... I would not have said that the flight was doomed.”
NASA has been diligent about trying to make the shuttle’s thermal protection system more resistant, said Pate-Cornell, adding that the agency adopted many of the recommendations for reducing risks in her 1990 report. Once Columbia was in orbit with a damaged left wing, the astronauts had no way of fixing it, according to NASA.
Two shuttle flights earlier, a chunk of insulating foam damaged one of the twin booster rockets that powered Atlantis into orbit Oct. 7. What if NASA had halted further launches at that point to take a closer look at the issue? Could the loss of Columbia have been avoided?
Like the insulation that hit Columbia, the piece that hit Atlantis came from an area where struts connect the external tank to the shuttle orbiter. The booster rockets were jettisoned into the ocean and recovered for reuse. That’s when NASA spotted a problem.
“Superficial damage occurred,” Dittemore said. “When we got the booster back in port and looked at it, evaluated it, reviewed it technically ... we, as an agency, as a shuttle program, decided that it did not represent a technical safety risk to us.”
There were no problems with falling insulation on the the Endeavour mission in November. Then came Columbia’s launch.
The chunk of foam that hit Columbia was probably bigger than any on previous launches, said Michael Kostelnik, a top NASA manager in Washington. But NASA was not overly alarmed, he added, because the incident was not “out of family” with previous problems.
For Dittemore, it raised a flag -- but not a red flag.
“Two occurrences in the last three flights is certainly the signal to our team that something has changed,” Dittemore said. “It did not represent on the first occasion an alarm from a safety point of view. It represented a turnaround processing issue.” That would mean closer inspection of returning shuttles, the possibility that more tiles would have to be replaced, and work to fix the problem with the foam.
NASA said it had reason to be confident that shuttle tiles could sustain damage without catastrophic consequences. An average of 100 tiles per flight experience some kind of impact with debris, Dittemore said. About 25 tiles per flight suffer a gouge of an inch or less. There have been 11 flights with a “significant number of impacts greater than 1 inch,” he added.
Hotz, the Challenger commission member, questioned whether those statistics are solid grounds for confidence.
“Every time they came back OK, everybody said tile damage is not a problem,” he said. “But accidents are never one thing. It’s 11 things lined up in a row. Even if you can survive the individual things, when they get all in sequence,” it will be a problem.
What if the foam from the external tank struck at just the right angle and speed?
“If that debris is traveling at 500 miles an hour it will do some damage,” said Paul Fischbeck, a Carnegie Mellon University risk analysis expert who worked on Pate-Cornell’s study. “If it hit a tile that was less than 100%, then what might happen? The tiles are still a major culprit, how they got damaged or lost is the mystery.”
NASA strongly resists any suggestion that it may have gotten complacent about a familiar risk like damage to the tiles.
“The lessons of Challenger have not been forgotten at NASA,” Kostelnik said. “I am very comfortable that this team did as well as they could in trying to understand what the problem was [on Columbia]. I believe we made the right decision at that time.”
