The Dallas Morning News August 5, 2005
NASA admits landing risks
But engineers say shuttle problems from loose blanket very unlikely
By Bruce Nichols
HOUSTON – As with space shuttle Columbia in 2003, Discovery will land early Monday with a problem that NASA engineers don't fully understand. In this case, it's a loose piece of heat-shield blanket that they decided not to fix.
The decision to land came despite acknowledgment from engineers that they can't be absolutely certain there won't be problems.
"I'm not here to tell you that we're 100 percent confident that there is no risk," Wayne Hale, chairman of the Mission Management Team, told reporters. "We've assessed this risk to the very best of our engineering knowledge, and we believe it is remote."
During a visit to Johnson Space Center on Friday, NASA Administrator Michael Griffin expressed full confidence. "We strongly feel we will have a clean, very nominal re-entry," he said.
And this time, the preferred landing track takes Discovery to Florida over Central America, the Caribbean Sea and Cuba. NASA decided after the 2003 Columbia disaster to avoid populated areas if possible.
Mr. Hale acknowledged "eerie" similarities between the decision to land Discovery and the decision to land Columbia. But he contends NASA has new tools and new attitudes that result in a much more complete analysis of potential problems.
"I've got to tell you, the difference between where we were before and where we are now is profound," he said.
Columbia headed for a Florida landing with a problem that engineers thought they had fully analyzed. A piece of insulation was spotted flying off the fuel tank during launch and striking the left wing. But mission managers decided the risk was low.
They were wrong.
The insulation had knocked a hole in the wing's leading edge, and the hot gases that surround the shuttle during re-entry into the atmosphere at 17,500 mph ate into the wing and destroyed Columbia. Seven astronauts died.
They could be wrong again, said NASA critic John Pike, formerly of the Federation of American Scientists and now with Glo balsecurity.org. "If you look at where we've been over the last 20 years, you have to wonder," Mr. Pike said.
But there's a greater likelihood that they're right, said a space policy analyst at American University, because NASA officials have collected more data, and they've more seriously considered it.
"Some of this [debate] is a consequence of a much more rigorous checkout of the shuttle," Dr. Howard McCurdy said. "The old way, for landing in particular, you didn't really check it out so, if these things were occurring, you ignored them."
Despite talk of eliminating risks after Columbia, flying the shuttle is always going to involve balancing risks, he said.
"There are a whole series of known risks, and if you total it all out, there's a 1 percent failure rate," he said. "This little blanket is part of that 1 percent."
Among the parallels between the Columbia decision and the Discovery decision:
•The model NASA used to analyze where the blanket would go if it came off had not been used for that purpose before. Investigators found that the model used in Columbia wasn't suited to analyze the launch debris impact.
•There was dissent in meetings leading up to the decision to land Discovery, just as there was in the decision to land Columbia.
•An engineer who figured prominently in the Columbia Accident Investigation Board report questioning that Mission Management Team's analysis, Rodney Rocha, also cited unknowns in the analysis that led to the Discovery decision.
But there appear to be big differences this time:
•The model used to study the heat-shield blanket on Discovery is "several generations more sophisticated" than the model used during Columbia's ill-fated flight, Mr. Hale said, and it was supported by wind-tunnel tests of a mockup of a damaged blanket.
•The dissent in ranks was brought to the attention of the Mission Management Team and fully discussed, Mr. Hale said, whereas during Columbia's mission dissent was squelched, the accident board said.
•Mr. Rocha wasn't listened to before the Columbia disaster. This time, he "gave us what I considered to be an outstanding assessment of the unknowns, the variations, the risks that were inherent in all of this analysis," Mr. Hale said. The concern was a "poofed out" piece of blanket under the left cockpit window. NASA's not sure what caused the damage, but the chief suspect is impact, perhaps by a cover that's popped off a guidance-control jet shortly after launch.
The blanket is made of silica-fiber cloth filled with silica-fiber batting.
The loose piece is 20 inches long and 4 inches wide and weighs less than an ounce, but analysis indicates that if it came off at supersonic speeds, there's a 1.5 percent chance it could knock a hole in a rudder speed brake, a hole that could grow to six feet in length.
That kind of damage sounds big, but engineers calculate that Discovery's control systems could overcome the problem and land safely, officials said.
Mission managers decided not to send an astronaut on a spacewalk to try to tie the blanket down or remove it because a spacewalk is dangerous, and they hadn't figured out how an astronaut could fix it.
"When in flight, many of your decisions are not absolute decisions. They are risk vs. risk decisions," Mr. Hale said. "You say which is riskier, going out and fooling around with this or coming on in given the knowledge we've got."
The first opportunity for landing is at 3:46 a.m. Dallas time Monday, but landing could be postponed a day or two or diverted to Edwards Air Force Base in California if Florida weather is bad.
Because Florida weather is changeable, officials said they won't know until shortly before re-entry. The preference is to land in Florida because it means NASA doesn't have to piggyback the shuttle from California on a 747 to prepare for the next flight.
If landing is postponed, the track moves westward with every delay, but re-entry over Dallas is unlikely largely because Discovery flew to the International Space Station, which orbits at a 51.6-degree inclination in relation to the equator. Columbia flew at 28.5 degrees.
Meanwhile, in-orbit operations went smoothly Friday, with the shuttle and station crews returning a big logistics module to the shuttle payload bay and restowing the 50-foot sensor boom used to look for damage. The shuttle is scheduled to undock from the space station today.
At briefings Friday, officials said they had just begun analyzing the biggest problem facing the shuttle program, the big piece of foam insulation that flew off the fuel tank during launch July 26. Shuttles won't fly again until that's fixed, officials have said.
A special team appointed by Mr. Griffin to look into the problem arrived at NASA's Michoud Assembly Facility in New Orleans on Tuesday. "I'm pretty much letting them fact-find," said space station manager Bill Gerstenmaier, who's heading the team.
He said he expected a preliminary report early next week.
© Copyright 2005, The Dallas Morning News Co.
