Near Proof of Shuttle’s Fatal Flaw

Investigators laid out a dramatic and detailed outline of the causes of the Columbia tragedy for the first time Tuesday, pointing to problems that began with foam debris from the space shuttle’s external tank slamming into the left wing during launch and that ended with superheated gases melting away the internal structure of the wing.

Barring radical new evidence in the weeks ahead, the scenario issued by the Columbia Accident Investigation Board is likely to represent something close to what the board will conclude when it writes its final report this summer and issues recommendations for ways to improve space shuttle safety.

The scenario begins with a large piece of foam that struck the leading edge of the wing 82 seconds after launch and probably dislodged a piece of structure that was seen floating next to the shuttle during the mission.

By the time the shuttle reentered the atmosphere on Feb. 1, a breach in the leading edge of the wing allowed superheated gases to melt the wing, which led to the breakup over East Texas and killed all seven crew mem-bers.

Although that sequence of events has been discussed widely in public, it now represents a theory that is accepted by the accident board and by senior officials at the National Aeronautics and Space Administration, according to Harold Gehman Jr., chairman of the accident board.

Gehman acknowledged that the panel may “never have irrefutable proof” of the direct causes of the space shuttle accident, but said it would nonetheless be able to construct a circumstantial cause and make key safety recommendations that will allow a return to spaceflight.

“We are careful not to say the foam knocked a hole in the leading edge of the orbiter, because we can’t prove it,” Gehman said. “Now that is not to say that we don’t believe that is what happened.”

Indeed, when the board issues its report, it may qualify its finding only by saying the foam “possibly or maybe” knocked a hole in the wing, Gehman said.

The board hopes that additional evidence on that point will come from tests over the next month when investigators will shoot foam chunks from a large gas gun at pieces of leading edge material at Southwest Research Institute in Texas.

If the tests show that foam could punch a hole in the panels, the board is likely to be more confident that it has identified the cause of the accident. The leading edge panels are constructed of a thin material known as reinforced carbon carbon, the only known substance that can withstand the 3,000-degree temperature created by air friction when the shuttle reenters Earth’s atmosphere from space.

Even without sure knowledge that the foam caused the disaster, Gehman said the board has uncovered strong evidence of problems that need to be addressed by NASA and said he would have a range of issues that the space agency needs to address before resuming flight.

“Fixing the foam and leading edge is not enough,” Gehman said.

At a news conference in Houston, Gehman laid out the investigative scenario and the evidence in three segments: the launch, the 16-day orbital flight and the reentry into Earth’s atmosphere.

Detailed analysis of photographs taken during the launch shows that a large piece of foam detached from the orbiter at 81.7 seconds after launch and struck the wing where, evidence would later show, the breach occurred. That point was most likely on leading edge panels seven or eight or at the T-seal that covers the gap between the panels. But the photographs are not detailed enough to show a hole or other damage.

A review of space shuttle history shows that pieces of foam at the attachment point known as the bipod ramp have detached and struck the shuttle on six previous missions and investigators believe three other incidents may have occurred at night or when the shuttle could not be photographed. Although such incidents violated flight safety rules, NASA grew to accept them, board member Kenneth W. Hess said.

On the second day of the mission, radar images showed a mysterious object floating near the shuttle for the next two days before it fell back to Earth. Air Force investigators at Wright-Patterson Air Force Base in Ohio have conducted radar tests on 24 shuttle parts that might show up the same way on radar. Only two items cannot be eliminated: portions of a T-seal and portions of a leading edge panel, lending further credence to the theory that a hole was punched into the leading edge. How such a piece survived the ascent into space and then floated away cannot be explained.

The destruction of the orbiter began almost immediately after it hit the atmosphere, still far out over the Pacific Ocean. Temperature sensors began registering abnormal readings just five minutes after the orbiter reached the upper atmosphere and the analysis of those readings again points to a breach at leading edge panels seven or eight. When investigators later recovered those panels, they found edges that were burned away as superheated gases flowed into the wing.

Once the superheated gases penetrated the leading edge, it took only three minutes to melt a hole into the aluminum wing spar, the heavy structure that bears the tremendous aerodynamics loads supported by the wing. After the spar was breached, wire bundles inside the wing began to melt and more than 600 temperature sensors and other gauges began to shut down.

Gehman said the board is still trying to compute how big a hole the leading edge had to melt the spar in three minutes. The hole had to be big enough to account for the damage but small enough for the shuttle to make it all the way to Texas, Gehman said.

Molten metals -- aluminum, stainless steel and Iconel -- began spraying inside the wing, leaving heavy telltale marks at panel seven and eight. And as the wing began to heat up, its shape deformed, Gehman said, and heat resistant tiles began to pop off.

Observers on the ground, from campers to space buffs, began to see unusually bright pieces of debris leaving the shuttle as it crossed over California and Nevada.

At the same time, the aerodynamics of the craft began to change and the computers attempted to adjust the control surfaces to counteract the melting away of the wing. By now, heat had penetrated all the way into the wheel wells.

Moments later, the orbiter’s nose suddenly pitched down and the computers in one last desperate attempt to save the craft fired the reaction control rockets to stabilize the craft. But it was too late and within half a minute all signals were lost from the ship, Gehman said.

Although the pilots flying the shuttle discussed the higher-than-normal temperature readings with ground controllers, they probably did not know the magnitude of the problems.

The safety board hopes to conclude its evidence gathering and testing in June and then move its operation from Houston to Washington, D.C. It would issue its report in July.

Gehman said the board would address the disaster on many levels, including NASA’s broader budgetary and cultural issues that contributed to the accident. The board is likely to make observations and recommendations that may have no direct link to the disaster, but would be aimed at improving overall safety, he added.

“The safety margins have changed over the years and we want to restore those,” Gehman said. “This is a good chance to make the shuttle safe to fly.”