Inquiry of Shuttle Will Cite Foam

Investigators said Tuesday that foam insulation that fell off the space shuttle Columbia’s external tank and slammed into its left wing represents the “most probable cause” of the accident that killed seven astronauts and destroyed the $2-billion spacecraft.

The statements are the strongest public finding yet that the foam insulation, suspected from the very earliest days of the investigation, is the direct cause of the shuttle’s breakup over Texas while reentering Earth’s atmosphere on Feb. 1.

Roger Tetrault, a member of the investigating board who has focused on the technical causes of the accident, said the 84,900 pounds of debris collected in Texas have yielded “compelling” evidence that the foam damaged the Columbia’s delicate heat shield on its leading edge during launch.

The statements all but ruled out alternative theories of the accident, which have included micrometeoroids, space junk, explosive bolts that were ejected during launch and other remote possibilities.

But because those possibilities can never be disproved, the Columbia Accident Investigation Board will qualify its findings by saying the foam is the most probable or highly likely cause when it issues its final report next month, said board Chairman Harold Gehman Jr.

The crippling damage to the wing occurred on the eighth leading edge panel on the left wing, the apex where the wing’s angle changes. The eighth panel, constructed of a heat-resistant material known as reinforced carbon carbon, is the largest and most complex part of the leading edge.

Portions of the recovered panel show that 3,000-degree superheated gases melted away its bottom portion, leaving knifelike edges. No other recovered leading edges have similar damage.

What’s more, molten iron and nickel were sprayed behind the panel, presumably from the fittings that held the panel in place, Tetrault said.

By the time the shuttle got to Texas, the superheated gases had entered the interior of the wing and were melting away the bonding that held silicon heat tiles onto the wing surface.

Tetrault showed a map indicating that debris from panel 8 and outboard to panel 22 clutter the area farthest west and therefore fell off earliest. Other debris from the left wing separated farther east. And finally parts of the right wing were recovered even farther east toward the Louisiana state line.

Near the end, a large piece of the left wing broke away near leading edge panel 8, though investigators are unsure how much of the wing fell off.

One theory is that almost half of the shuttle’s wing broke off, including the entire section behind panel 8, Tetrault said. But other experts have disagreed and said the fracture removed only a portion of the wing behind panel 8.

In either case, the loss of that much wing surface crippled the shuttle, even though its computers continued to operate control surfaces and rocket thrusters in a doomed effort to maintain stability.

At about the same time, a large portion of the shuttle’s vertical stabilizer tore off and landed in the middle of the left-wing debris field in Texas.

“You can conclude from the debris alone that the breach occurred at panel 8,” Tetrault said.

The board has assembled substantial evidence that corroborates its debris assessment. Wire bundles in the wing melted sequentially and caused signals to go silent, giving investigators a good understanding of where the heat first entered the wing. Again, the evidence points to panel 8.

And high-speed photographs taken during launch showed that the foam debris, weighing 1.67 pounds, struck the leading edge between panels 6 and 9.

“When you put all that together, the foam is the most probable cause of the accident,” Tetrault said.

Gehman concurred with Tetrault’s assessment, though he said the board has not decided how strongly it will word its finding. That will partly depend on the results of additional foam tests to be conducted July 7 in San Antonio. Investigators will use a gas-powered gun to shoot a foam block at a replica of Columbia’s left wing, aiming at panel 8. Unlike a previous test that used only a single panel, the upcoming shot will use three leading edge panels of reinforced carbon carbon, according to board member G. Scott Hubbard.

Although increasingly confident that foam damaged the leading edge, the board has failed to solve the mystery of why the foam fell off the tank. It has determined that on seven previous flights, foam from the so-called bipod attachment fell off.

Five of those flights involved the Columbia and one the Challenger, which was destroyed by another cause in 1986, Gehman said. Both the Columbia and Challenger were heavier than the other shuttles, though whether that is a factor is unknown, he added.

Since NASA is redesigning that portion of the foam insulation, it will not be necessary to determine the precise causes of the failure, Gehman said.

Gehman said he did not see any reason why NASA could not resume flights within six to nine months, though he spelled out three recommendations that could require extensive shuttle modifications in the years ahead.

NASA must fix the foam debris problem, though it may never be completely sure it can prevent all foam losses, he said.

Then, NASA must find a way to toughen the shuttle’s thermal protection system and improve its ability to absorb debris hits. It has already developed tougher tiles, but has not used them because they weigh more and would reduce the payload that NASA could launch into orbit, he said. As for the leading edge panels, he noted they are only a quarter-inch thick and suggested, “Why not five-eighths of an inch?”

Finally, he said, NASA must develop ways to inspect and repair damage in orbit if the first two measures still fail to protect the thermal shield system.

A crew escape or ejection system will not be recommended, Gehman said. He said such a system might be helpful, but the board has decided to let NASA assess whether it’s needed.

Gehman said the first flights of the shuttle will be among the safest because NASA is very good at engineering fixes. But longer term, the agency will need to make substantial changes to its management to maintain safety. About half of the report will deal with NASA management and policies, he said.