‘An Awfully Tough Investigation’ Lies Ahead

WASHINGTON —  Moments before Columbia broke apart over Texas, it had begun the most stressing part of its reentry to Earth’s atmosphere, taking a banked turn at 12,500 mph and heating up to 3,000 degrees from atmospheric friction.

It is likely that the cause of the accident is closely related to those stresses. Based on the limited evidence available late Saturday, experts say the possible causes include a failure of the automated flight control system, a breakdown of the heat-protection tiles, a malfunction or fire inside a wing and a failure of the wing structure. But they emphasize that too little is known to be sure of any cause.

Early speculation is focusing on the possibility that the shuttle’s heat-protection system was damaged during its most recent launch on Jan. 16, when a piece of foam broke from the fuel tank and struck the shuttle’s left wing. During the mission, NASA flight officials were concerned but not alarmed.

Ice and other debris occasionally become dislodged from the shuttles during launches, striking the craft and causing minor damage. When Atlantis lifted off in October, foam broke from the fuel tank, as it did when Endeavor was launched in 2000. In 1998, an 11-pound aluminum door tore from Discovery and struck the main engine.

Investigators are taking a new look at Columbia’s liftoff, asking whether the piece of foam might have damaged tiles on the left wing that are intended to protect it from the extreme heat encountered during reentry to Earth’s atmosphere.

Ron Dittemore, manager of the space shuttle program for NASA, said the craft just before the breakup had entered its first so-called “roll reversal,” a turning maneuver that helps reduce speed. Data from the spacecraft indicate that a series of sensors inside the left wing failed within minutes of each other during the turn. Such data are consistent with any number of failure scenarios, experts said.

At an altitude of about 40 miles, the atmosphere becomes thick enough for the shuttle to start behaving like an airplane. It begins using its control surfaces, rather than rocket thrusters. The craft’s auxiliary power unit is turned on to provide power for the control system.

Officials who are gathering debris from the wreckage, which landed in an area covering at least 500 square miles of Texas and Louisiana, said they fear important evidence could have burned up as Columbia fell. They will comb through the constant stream of data that the shuttle sent to flight headquarters in the course of its mission, as well as the craft’s construction and maintenance records.

But officials are already renewing their scrutiny of the foam that dislodged on liftoff, Dittemore said.

Video of the liftoff appears to show the foam hitting the left wing, he said. Officials have asked the NASA team that worked on the external fuel tank, as well as Lockheed Martin, the contractor that built the tank, to isolate certain manufacturing equipment and data that might be relevant to the investigation.

If the foam did damage insulating tiles on the wing, it may have left the craft vulnerable on its return to Earth, when friction with the atmosphere can raise the temperature on the front edge of the wing to 3,000 degrees.

The left wing is also drawing attention because temperature sensors on its rear edge failed about seven minutes before the disaster, providing the first sign to flight officials of a problem with the craft. In the following minutes, other data sensors failed in the left tire well and the left side of the vehicle near the wing. While the evidence is far from clear, those failures may have resulted from the heat damage caused by missing insulation tiles.

Shuttles have previously survived the loss of insulating tiles, and Dittemore cautioned against jumping to conclusions.

Flight officials had been aware of the foam incident and convened technical experts on the day after Columbia’s liftoff to review video of the launch. “We spent a goodly amount of time reviewing that film and then analyzing what that potential impact of debris on the wing might do,” Dittemore said.

“It was judged that that event did not represent a safety concern,” he said. As a result, NASA elected not to photograph the wing from ground or space cameras, as the agency has done on earlier missions.

Dittemore said there will be two parallel investigations, one conducted by the agency, the other by an independent board, whose membership has not yet been announced.

NASA’s first task will be to collect all the debris now scattered across a huge area made up largely of rural areas and small towns. Officials have not yet decided on a central collection area for the debris.

Investigators will work to identify the individual pieces and to determine what failed first and why. Metallurgists from the National Transportation Safety Board, which investigates airline crashes, are assisting in the probe.

“This is going to be an awfully tough investigation,” said Peter Goelz, former managing director of the NTSB, who worked on the crash of TWA Flight 800 in the waters off New York.

“All we have at this point is the result,” he added. “We don’t even know if we saw the initial break-up. By the time people heard the sound, the event was well underway.”

NASA has preserved all information associated with the mission and will turn it over to investigators. Every aspect of the flight and of the shuttle’s design will be examined, including prelaunch activities, mission planning and operations, and any systems or structures that could have failed.

The performance of onboard motors that help guide the shuttle into a proper reentry path will be examined. The motors themselves use hazardous propellants that could create an explosion, although there is no indication that such an incident occurred.

Many questions were raised about the fuel-tank insulation that jarred loose during the launch.

“We have a pretty good idea how big the size of the foam was,” Dittemore said of the piece that came loose during Columbia’s launch. “What we don’t understand [is] what the actual impact did to the tile.”

And even if tiles were knocked off, Dittemore said, there was nothing that could be done about it at that point in the reentry phase.

NASA has long had problems with the insulating foam. In 1995, the agency had to delay launching Discovery after woodpeckers punched dozens of holes in the foam as the shuttle was awaiting liftoff. The delay cost an estimated $1 million.

In 1998, shuttle managers at Kennedy Space Center installed cameras and special sensors on the fuel tank of the shuttle Atlantis to study why foam was falling off during launches.

In 2000, foam insulation peeled away from the external tank when Endeavor was launched. Engineers noted a patch of missing insulation while reviewing video of the shuttle’s liftoff. Later, what was described as “superficial damage” was found on the skirt of one of the booster rockets -- and attributed to the insulation.

The foam insulation is sprayed onto the fuel tank and can be as thick as a few inches. It helps keep the cold, liquid hydrogen propellant from heating up during launch and prevents condensation from collecting on the outside of the tank, which could form ice. Ice would represent both a flight hazard if it broke off and a weight penalty during launch.

The foam is fairly lightweight, similar in density to Styrofoam, though it does have a somewhat harder outer shell, according to Frank Manning, former executive director of NASA’s Aerospace Safety Advisory Panel. He concurred with initial assessments by NASA engineers that a chunk of the foam probably would not do serious damage to a wing in the moments after liftoff.

Albert O. Wheelon, who served on the commission that investigated the Challenger accident, said he was troubled Saturday by disclosures that are reminiscent of the lax safety practices at the root of the 1986 disaster. Although it is far from clear that the foam debris played a central role in the accident, it was a known safety issue that NASA had long intended to address but had put off, he said.

In the Challenger case, NASA had recognized that leaking O-ring seals on the solid rocket boosters were a safety problem, and yet the agency did not take aggressive action before the incident that caused the explosion in 1986.

“The attitude is that ‘we’ll get around to it some time,’ and that’s just what happened with Challenger,” said Wheelon, a former deputy director of the Central Intelligence Agency.

Wheelon said the Bush White House should follow the lessons of the Rogers Commission, which investigated the Challenger disaster, appointing a nonpolitical investigation panel that is free to examine the causes of the accident. “The good news is that we know how to investigate something like this, if we are given the opportunity,” Wheelon said.

*

Times staff writer Rosie Mestel in Los Angeles contributed to this report.