Radar records show that TWA Flight 800 apparently continued to fly for at least 24 seconds after the cataclysmic event that doomed it, raising the possibility that some of the passengers and crew may have been alive in the last horrifying seconds before the crippled jet broke apart in a fireball, investigators said Friday.
The radar information does not give a clue as to whether the cockpit crew had any functioning controls to work with after the plane was disabled at 13,700 feet by either a mechanical problem, a bomb or a missile _ investigators still do not know which.
But the investigators said the records indicated that the jetliner's engines probably continued to run as it moved over the water at more than 400 mph. The plane was also descending rapidly, and by the time it reached about 8,500 feet, a fireball erupted, fed by the jet fuel that investigators presume was gushing from the plane's tanks.
The scenario, based on information from a radar dish that takes 12 seconds to make a 360-degree sweep of several thousand miles, did not make investigators any more certain of the cause of the problems. But it did help establish the framework for a sequence of failures.
According to the investigators, the radar made two sweeps after the initial event in which the plane appeared to be basically intact. But by the time the radar made its third 12-second sweep, two pieces of the plane became visible on radar.
Federal officials said that there were no signs of problems on either the flight data recorder or cockpit voice recorder except for a loud noise at the end of the voice recorder and except for the obvious fact that they quit working a fraction of a second after the noise was heard.
Investigators also said they had found two of the plane's four engines. But they said they had decided to delay raising them and assessing their condition _ which could provide major clues _ because doing so would delay the recovery of bodies. Searchers have now brought up the remains of nearly two-thirds of the 230 people on board, and nearly half have been turned over to next-of-kin.
According to the National Transportation Safety Board, the last moment of normal functioning was at about 8:31:12 p.m., when the plane was climbing toward 15,000 feet with its wings level. Within 12 seconds of then _ the time that a radar at Islip, N.Y., takes to make a full-circle snapshot of planes in its sector _ the beacon that broadcasts the plane's identity and altitude failed.
About the same time, possibly at the same instant, power was cut to the flight data recorder and the cockpit voice recorder. And sometime around then the pilots' radios probably ceased to function.
But the plane was in the air for another 2.5 nautical miles, another 41 seconds. For at least 24 of those seconds, the jetliner continued at a more or less constant speed over the water, about 380 knots, but its speed through the air was apparently rising rapidly, because it was descending to the point where it became a fireball. A normal airliner descends at about 1,500 feet per minute; this would have been about five times faster.
In that rapid descent it apparently was shedding some large, though relatively light, pieces, which were found in what investigators now call Debris Field No. 2, an area southeast of the plane's track over the water. These may be sides of the fuselage, investigators say.
As technicians in Washington studied the recorders, investigators here laid out this chronology:
The plane, having arrived from Athens earlier in the day with no mechanical problems, took off lightly loaded from Kennedy International Airport and climbed at the standard speed of 250 knots, or 287 mph, to 10,000 feet. Above that height it began climbing with its airspeed indicator at 300 knots or 345 mph.
But as the air gets thinner with altitude, an airspeed indication at a constant speed means increasing speed over the ground; that plus a partial tailwind of 20 knots put the airplane's speed over the water at 279 knots at the time of the first event.
That event probably happened at 13,700 feet, because that is the last altitude reported by the on-board transponder, which gives the plane's identity and altitude. By the time the radar looked again, there was only the "primary return," or ordinary radar echo. From then on, the radar picture was only two-dimensional.
Robert T. Francis, vice chairman of the safety board, said Friday afternoon that the cockpit voice recorder ended with "a loud, unknown noise." The flight data recorder also ended abruptly, with no indication of trouble beyond the obvious: that power to it was cut off.
And experts surmise that the radios must have failed too, or the pilots would have sent a distress signal, or at least said that they could not maintain their climb; failure to report that would be to invite a collision.
But they do not know the sequence of these events, and precisely how each of those pieces of equipment _ the radios, the recorders and the transponder _ are supplied with electric power. They believe, though, that they can establish those facts from evidence on the ground, and then, perhaps, figure out where on the airplane the problem originated.
Investigators have no evidence of a major breakup of the plane before the fireball. They are convinced that the fireball happened 5,000 feet below the initiating event by careful correlation of reports from pilots on other planes with an unusual data source, a private company on Long Island that provides a kind of shadow air traffic tracking service, more sophisticated than what is done by the Federal Aviation Administration.
The company, Megadata, of Bohemia, N.Y., uses a radar at Islip to listen for beacon reports by the aircraft and was tracking about a dozen of them in the area at the time of the fireball.
That is an unofficial source of data, but investigators have used it to assure themselves that the pilots of other planes were in position to report what they say they saw, a fireball at about 8,500 feet.
One of them was Eastwinds Flight 507, a Boeing 737 en route to Trenton from Boston, flying at 16,000 feet, northwest of the TWA 747. A pilot on board said he saw the nose and left side of the TWA flight and saw the fireball below him.
A pilot on a USAir commuter flight to New Haven, Conn., at around 11,000 feet, also saw the fireball below him, according to investigators.
Witnesses have also told investigators that something came off the airplane before the fireball.
Investigators still do not know what could have happened to crack open the plane sufficiently that it would begin spilling its 165,000 pounds of fuel into the air, where it was ignited, but they believe that it was doomed before the fireball began.
Both Francis and James K. Kallstrom, the FBI's chief official on the case, said, though, that they were certain that the cause could be derived from wreckage.
At a news briefing, Kallstrom said his agents have interviewed a number of witnesses who recall seeing something in the air "ascending" toward the aircraft, possibly the track of a ground-to-air missile.
Kallstrom said it is more likely that the witnesses saw something streaking away from the doomed aircraft, possibly burning debris from the initial explosion.
Kallstrom said he was not yet ready to declare absolutely that the plane was felled by an on-board bomb.
"I think we've moved the ball a few more yards down the field," he said. "We're not here to declare what everybody is discussing in the newspapers (a bomb) but I think we are closer to that."
There will be many other indications, according to experts. In the bombing of a Pan Am 747 over Lockerbie, Scotland, for example, one additional way that investigators knew the cause was a bomb was evidence of damage on the tail that was not inflicted by the impact, but rather by debris that had been blasted out of the front of the airplane and struck the forward edges of the tail.
"There is evidence down there that's going to tell us what happened to this aircraft," said Francis in a briefing Friday afternoon.
_ Information from the Los Angeles Times was used in this report.