New Questions on Shuttle Tile Safety Raised

When Boeing advised NASA last month that damage to the space shuttle Columbia’s tiles wouldn’t prevent a safe return, it specifically rejected a time-honored computer program that had predicted major problems for the orbiter.

Instead, Boeing’s sanguine recommendations were based, at least in part, on 1999 tests by a Texas research laboratory.

But when those test results were released Wednesday, they did not offer any evidence that Columbia could have safely returned to Earth.

In fact, the tests conducted by Southwest Research Institute found the tiles were susceptible to damage from foam debris that could fall off an external tank during launch. Although the potential for such damage was generally known, the Southwest report provided detailed evidence that very small pieces could cause notable damage.

On Wednesday, NASA also released internal e-mails revealing that the space agency’s engineers, with increasing urgency, questioned -- fewer than 24 hours before Columbia was scheduled to land -- whether its crew would survive. Some NASA engineers seemed dumbfounded that the debate had not been resolved that late in the mission.

“Why are we talking about this on the day before landing?” NASA engineer William C. Anderson wrote to a colleague on Jan. 31. A day later, Columbia disintegrated while descending through the atmosphere, killing its seven-member crew.

The disclosures bring into sharper focus the intense debate within NASA and its contractors over whether the space shuttle was facing a risky flight back to Kennedy Space Center in Florida.

As the shuttle lifted off on Jan. 16, foam debris fell from an external fuel tank and struck the left side of the craft. NASA asked Boeing to assess whether the foam had damaged the tiles.

Boeing had repeatedly assured NASA that its crew would enjoy a “safe return.” The cause of the incident remains unknown, but the foam remains a primary line of inquiry.

In Houston on Wednesday, Boeing NASA Systems spokesman Ed Memi said the company’s analysis was correct, given what engineers knew at the time -- “unless the investigators prove otherwise. We stand by what we provided to NASA,” he said. A NASA spokeswoman agreed that the analysis was sound.

Boeing’s assurances were based in part on the study by the Southwest Research Institute, an independent, nonprofit engineering and physical science research organization in San Antonio. NASA released that study Wednesday.

Institute officials did not return phone calls seeking comment.

The study was designed to supplement a software program, Crater, that NASA has long relied on to evaluate tile damage.

Crater predicted that Columbia’s tile system was damaged, but Boeing engineers told NASA to disregard that prediction because the software system tends to “overpredict” damage.

In a memo Jan. 23, Boeing said it had reviewed the Southwest report when it told NASA the shuttle would return safely. It is not clear from the Boeing report what else it might have based its recommendation on.

It appears that Boeing used the report to cast doubt on Crater, according to Paul Fischbeck, a Carnegie Mellon University professor who co-wrote a study looking at the damage that foam debris causes on tiles.

However, Fischbeck said that nothing in the Southwest report appears to support Boeing’s advice, either to ignore Crater or to conclude that the shuttle was undamaged.

Members of the independent Columbia Accident Investigation Board said in interviews last week that they are interested in determining whether Boeing’s analysis was complete and adequate.

Among other issues, panel members questioned whether Boeing engineers had any scientific basis for overruling the Crater software program.

The release of the Southwest report has only compounded those sorts of concerns within the aerospace community.

For instance, using a high-powered helium gas gun, Southwest fired a number of pieces of debris at samples of space shuttle tiles. However, the tests were confined to pieces that were a fraction of the size of the 2.7-pound piece of foam that struck Columbia on Jan. 16.

The institute had planned to conduct tests on larger pieces of foam, but NASA told the researchers not to bother, largely because the smaller pieces had caused so much damage.

NASA officials conceded Wednesday that they have never tested the potential impact of larger pieces of foam insulation, because the pieces of foam that broke off external tanks and struck the space shuttle in previous flights were far smaller.

“Historically, the impacts and damage we had seen from foam insulation had been small,” said NASA spokeswoman Kylie Moritz. “What was used in this test, we thought, was adequate for what we had observed in flights.”

Finally, Southwest’s tests only measured the size of dents that foam insulation caused to individual, free-standing tiles, not the impact of foam on a bank of tiles that more accurately represents the shell of the space shuttle, said Paul Czysz, professor emeritus of aerospace and mechanical engineering at St. Louis University and a veteran NASA consultant.

Czysz, among other engineers, suspects that the foam insulation severely damaged the tiles covering Columbia’s landing gear door, perhaps enough that the door broke loose early in the shuttle’s descent, allowing superheated gas inside and destroying the craft.

Boeing engineers, Czysz said, “should have assumed the worst, not the best.” That didn’t happen, he added.

Transcripts of the internal e-mails also released by NASA on Wednesday show that debate over the potential impact of the foam insulation raged inside the space agency even as Columbia’s astronauts were preparing to land.

The e-mails show that debate was restricted to a small technical community within NASA, mainly landing-gear experts, but that it failed to gain wide recognition at higher levels, said John Pike, a space expert at GlobalSecurity.Org. “Why is this whole debate going on among a bunch of wheel guys?” Pike asked in an interview Wednesday. “I don’t see that it ever reached the top level administrators.”

In one internal e-mail, an engineer compared the discussion among more than a dozen NASA engineers to what is known within space agency circles as the “burning rock” event.

That was a reference to an engineer’s suggestion that Apollo 11’s rockets might set fire to moon rocks, causing a fatal explosion. In that case too, the engineer waited until the moon mission was well underway before raising the issue.

Others were clearly frustrated about the engineers’ inability to convince top NASA administrators that their theories should be taken seriously.

“Any more activity today on the tile damage [discussion] or are people just relegated to crossing their fingers and hoping for the best?” Robert H. Daugherty, an engineer at NASA’s Langley Research Center in Hampton, Va., wrote on Jan. 28.

Anderson, Daugherty and the other engineers, according to NASA documents, debated a series of catastrophic scenarios that, in hindsight, turned out to be what happened to Columbia.

Among other things, the engineers suggested that Columbia might have suffered a breach that would allow entry of superheated gas known as plasma, that its left wing might have a “big hole” in it and that the shuttle’s landing gear door might have blown off -- all theories that investigators are now seriously exploring. But word of the e-mail exchange reached NASA administrators only after the shuttle had broken up.

“It was considered to be a ‘what-if’ exercise,” one NASA official, asked to summarize the e-mail exchange, wrote to Ron Dittemore, NASA’s space shuttle program manager, on Feb. 11.

Despite NASA’s repeated assurances that it was unified in the belief that Columbia would return to Earth safely, the e-mails show that several engineers had enough lingering concerns about the damaged shuttle that they discussed drastic measures late in the mission.

Two engineers, for example, discussed the pros and cons of asking the crew to bail out versus performing a “belly” landing without landing gear.

On Jan. 27, Daugherty also raised the prospect of asking one of the astronauts to perform a spacewalk that would enable the crew to check the underside of the shuttle for damage.

Dittemore has said such a maneuver would have been too risky because astronauts wearing bulky spacesuits could have brushed against the delicate shuttle’s tiles and damaged them.

“Can’t imagine that an astronaut [even on a crappy tether arrangement] would cause MORE damage than he is going out to look for!” Daugherty wrote.