Clues in shuttle's final minutes
By Paul Recer
February 03, 2003

NASA engineers settled into their long, joyless task of figuring out how space shuttle Columbia broke apart, saying conditions in the shuttle's final minutes point to a possible problem with its critical heat-protection tiles.

NASA says new evidence shows that the temperature on Columbia's left side shot up and the ship was buffeted by greater wind resistance before it disintegrated over Texas, killing all seven astronauts aboard. Those conditions forced its automatic pilot to quickly change course.

The combination of these events suggests that thermal tiles may have been damaged during launch. The shuttle's exterior is covered with thousands of tiles designed to protect it from the extreme heat of re-entry.

Despite the possible clues, shuttle program manager Ron Dittemore stressed yesterday that the information was only preliminary.

"We've got some more detective work," Dittemore said. "But we're making progress inch by inch."

While engineers at the Johnson Space Centre in Houston analysed billions of bits of electronic data radioed to Earth by Columbia on Saturday morning, state and federal officials collected bits and pieces of the shattered spacecraft over a broad swath of east Texas and Louisiana.

The debris was being catalogued and trucked to an Air Force base in Louisiana. Some human remains also have been recovered from the astronaut crew.

President George W Bush had arranged a meeting today with NASA administrator Sean O'Keefe to get an update on the disaster.

Computer data indicates that moments before Columbia broke apart on Saturday on its way toward a landing in Florida, temperatures rose in the wheel well and on the fuselage on the left side of the shuttle.

The abnormal readings were on the same side of the craft that was hit by peeling fuel-tank insulation during the craft's January 16 launch, NASA engineers said.

Dittemore said engineers also were planning to examine 32 seconds of computer data that earlier had been ignored because it was considered flawed. The data came just before all communications with Columbia were lost.

NASA engineers spotted the peeling fuel tank insulation on high speed cameras that recorded the launch of Columbia. Dittemore said the possible effects on the tiles from the insulation were studied aggressively while the shuttle was still aloft, but engineers concluded "it did not represent a safety concern".

"As we gather more evidence, certainly the evidence may take us in another direction," he said.

Dittemore said engineering data shows a temperature rise of 11 to 17 degrees Celsius in the left wheel well about seven minutes before the spacecraft's last radio transmission. There followed a rise of about 32 degrees over five minutes in the left hand side of the fuselage above the wing, he said.

The shuttle temperature rose the normal eight degrees on the right side over the same period, he said. All the readings came from sensors underneath the thermal tiles, on the aluminium hull of the craft.

The temperature spikes were accompanied by an increased drag, or wind resistance, that forced Columbia's automated flight control system to make rapid adjustments to maintain stability. Dittemore said the corrections were the largest ever for a shuttle re-entry, but still within the craft's capability.

Lockheed, the maker of the fuel tank under scrutiny, said yesterday that NASA used an older version of the tank, which the space agency began phasing out in 2000. NASA's preflight press information stated the shuttle was using one of the newer super-lightweight fuel tanks.

Harry Wadsworth, a spokesman for Lockheed, said most shuttle launches use the "super-lightweight" tank and the older version is no longer made. Wadsworth said he did not know if there was a difference in how insulation was installed on the two types of tanks.

Wadsworth said the tank used aboard the Columbia mission was manufactured in November 2000 and delivered to NASA the next month. Only one more of the older tanks is left, he said.

Dittemore said the tank, though no longer manufactured, had been used for many years and was between 2700kg and 3150kg heavier than the newer version. Still, "we had no reason to doubt its capability".

Earlier yesterday, O'Keefe named a former navy admiral to oversee an independent review of the accident, and said investigators initially would focus on whether the piece of insulation caused the damage that brought down the shuttle.

While O'Keefe stressed that the space agency was not locking into a single scenario of what caused the crash, the insulation was "one of the areas we're looking at first, early, to make sure that the investigative team is concentrating on that theory".

For a second day, searchers scoured forests and rural areas of east Texas and western Louisiana for bits of metal, ceramic tile, computer chips and insulation from the shattered spacecraft.

State and federal officials, treating the investigation like a multi-county crime scene, were protecting the debris until it can be catalogued, carefully collected and then brought to Barksdale Air Force Base in Louisiana.

The effort to reconstruct what is left of Columbia into a rough outline of the shuttle will be tedious and painstaking.