Search for crash clues narrows

When EgyptAir Flight 990 went down, there were suspicions that a thrust reverser, which helps the plane slow during landings, had accidentally deployed in flight. But data from the flight recorder showed that didn't happen.

Then, there were theories that an intruder had entered the cockpit in a hijacking attempt. But Sunday, the taped conversations on the cockpit recorder seemed to rule out that possibility.

Indeed, in the 16 days since the plane went down 60 miles off Massachusetts' Nantucket Island, federal aviation officials are gradually narrowing their search for what caused the fatal crash - although they caution that a determination could take weeks, months, or even years.

While the National Transportation Safety Board (NTSB) almost always pinpoints a cause in the end, the search for mechanical failures - certainly a leading suspect in the EgyptAir crash - can be likened to finding a single white hair on a big brown bear. These days, all the obvious places to look, such as at equipment that caused past accidents, seldom yield answers because those problems have usually been remedied in accord with federal directives.

"The cause of this accident is probably something entirely different than anything that's ever happened before," says one aviation engineer who requested anonymity.

The NTSB bag of tricks

To try to solve this puzzle, the NTSB will employ an extensive array of techniques.

First, it will continue to analyze both "black boxes," which the US Navy recovered using a submersible drone. Yesterday, the agency began the process of matching the flight data recorder with the voice data recorder to give a second-by-second time line of what the pilots were doing and how the plane was behaving. Officials were also listening for the sounds of switches being thrown or other cockpit changes.

Yesterday, officials also finished their investigation of the airplane's maintenance records in Cairo. They were checking whether EgyptAir had performed mandatory checks and repairs.

At the same time, the NTSB is taking the data from the recorders to Seattle to run it on a Boeing computer simulation of a 767 in flight. Investigators will try to replicate the unusual flight pattern of the plane, which started as a deep but controlled dive, then rose again about 8,000 feet, and finally broke up in midair before plummeting into the Atlantic. The computer simulator helped investigators find why a USAir 737 crashed near Pittsburgh in 1994.

"Basically, they feed the data from the flight data recorder into the simulator to at least begin to get an understanding of what was going on in the airplane," says John Dern, a Boeing spokesman.

The simulators may help to solve one of the strangest - and still unexplained - occurrences. The plane's elevators, which are two sets of flaps on either side of the tail, were in opposite directions. The elevators help direct the plane up and down and normally move in concert. But the flight data recorder indicates that one set was deflected up and the other down.

The NTSB's lead investigator, Gregory Phillips, indicated that this might be possible if one of the controls had jammed. But it would require that the pilots push the controls in different directions - something experienced pilots would not be likely to do.

"It's hard to imagine that jammed controls would allow pilots to put it into a dive," says an aviation engineer. "If the controls jammed when they were cruising, that would tend to make the plane roll to one side or the other and that never happened."

Diving for more of the plane?

It's not clear if the NTSB will try to salvage more of the plane. After TWA Flight 800 crashed off Long Island in New York in July 1996, the agency pulled up as much of the plane as it could and reconstructed it in a hangar.

But that plane, a 747, was in shallower water and the crash took place in August, when the water temperature was warmer. In the case of Flight 990, divers would be limited to about 20 minutes on the bottom because of the depth. It would take almost three hours for the divers to come back to the surface.

The debris field in Flight 800 was spread over a large area because the plane exploded at a relatively high level and a high speed. In the case of EgyptAir Flight 990, investigators believe it may have broken up at 10,000 feet. As a result, the debris field is more compact.

It's likely that airline investigators would prefer to raise more of the plane, because they like to have as much information as possible. After a ValuJet DC-9 crashed in Florida in May 1996, for example, many people assumed it was because the airplane was an older model. It wasn't until investigators checked the flight manifest that they discovered the plane was carrying old oxygen containers that had not been stowed properly.

Investigators also are continuing to probe into the mental state of the crew piloting the flight. After a brand new Boeing flight slammed into a hill in Indonesia, investigators initially thought some of the rivets had popped out of the plane, causing it to crash. But further investigation led them to conclude the pilot intended to commit suicide.

To further understand what the EgyptAir pilots were feeling, the NTSB plans to bring over other of the airline's pilots to help them determine the mental states of the two men flying the plane.

All of these investigations will take time. If the NTSB decides to try to raise more of the plane, it may have to wait until next year, when the weather offshore becomes more cooperative.

*Staff writer Faye Bowers contributed to this story.

(c) Copyright 1999. The Christian Science Publishing Society