What Went Wrong: The Crash Of Flight 587

NASCAR Team's Final Flight Photo by George Tiedeman/Newsport On October 24, while Hendrick Motorsports star Jimmie Johnson was winning his sixth race of the 2004 season, NASCAR officials shielded him from tragic news: Earlier that day one of the team's planes had crashed in heavy fog 10 miles from the track in Martinsville, Va., killing both pilots and eight relatives and associates of team owner Rick Hendrick. According to the preliminary NTSB report, the Beechcraft King Air 200 turboprop overshot the Martinsville airport and flew for 3 miles, descending to 1800 ft. It had started to climb when it crashed into 3211-ft. Bull Mountain. The NTSB's initial findings suggest pilot error. It noted, however, that the plane's GPS device was not certified for instrument-only flight, and the aircraft wasn't equipped with a ground-proximity warning system, which the FAA plans to require on planes this size, starting in March.--Tracy Saelinger

It was thought that the Airbus pilot, hitting rough air, overcorrected with the rudder, creating stresses that ultimately ripped the vertical stabilizer from the plane. To most nonpilots, that scenario seemed dubious: Could a patch of turbulence, combined with a pilot's heavy foot on the rudder pedal, really rip the tail from a plane?

According to the final report of the National Transportation Safety Board (NTSB), this improbable scenario was indeed what brought down Flight 587. But contributing factors made the tragedy more complex--and controversial.

When a plane flies, each wingtip trails a swirling vortex that is like a horizontal tornado. If another plane enters this invisible turbulence it can suddenly rise, sink or roll violently. Such turbulence is rarely fatal, however, because air-traffic controllers maintain adequate separation between jets. The plane preceding Flight 587, Japan Airlines Flight 47, a Boeing 747, was 5 miles downrange by the time the Airbus took off, normally sufficient for the turbulence to dissipate.

When Flight 587 hit the 747's vortex 96 seconds later, the bump was mild enough that First Officer Sten Molin, who was at the controls, corrected only with the ailerons. The second bump 15 seconds later was a little worse. If Molin had done nothing, the plane would have flown through it without incident. "His actions were unnecessary and excessive, because the effect of the wake turbulence on the aircraft was minor," John O'Callaghan, an NTSB investigator, announced at the October NTSB board meeting to review the final report. Why, then, was Molin's response so aggressive?

Part of the problem lay in Molin's flying style. During the NTSB investigation, two fellow pilots said that Molin tended to overcorrect for wake turbulence. Making matters worse, American Airlines' flight training encouraged pilots to use the rudder to recover from upsets in flight (incorrectly, Airbus maintains). Even worse, the NTSB found that the training program used a flight simulator that was unrealistic in portraying how an Airbus A300 would respond during wake turbulence. After the accident American changed its training program.

Some critics contended that the advanced composite materials used to reduce the Airbus's weight have not stood the test of time. But the NTSB found that as Flight 587 fishtailed through the sky, its vertical stabilizer withstood the fast-building stresses as it had been designed to. The rudder was certified to handle only a full deflection--that is, to travel from neutral all the way right or left, but not to swing alternately from far right to far left. It wasn't until the rudder had done this three times, with load forces almost twice the fin's design capacity, that the vertical stabilizer finally gave way.

Not to say that Airbus was faultless. Compared to other commercial jets, the NTSB found that the A300 had "the lightest pedal forces of all the transport-category aircraft evaluated." In other words, it had a hair trigger, a tendency of which the pilot was fatally unaware.

That a fully laden commercial jet could meet such a catastrophic fate merely as a result of pressing the rudder pedals was one that, tragically, probably surprised the two pilots as much as it did horrified observers on the ground.

96 SECONDS AFTER TAKEOFF
An American Airlines Airbus A300-605R hits the wake turbulence created by a Japan Airlines Boeing 747-400. For experienced pilots like First Officer Sten Molin and Captain Ed States, the bump is relatively minor.

111 SECONDS
The Airbus hits the wake again. This time Molin responds aggressively. In the next 7 seconds, his vigorous use of the rudder pedals causes the plane to fishtail through the air as the rudder swings back and forth.

118 SECONDS
With the plane yawing to the left and the vertical stabilizer exposed to the full force of the 300-mph slipstream, Molin throws the rudder back to the right. The stress rips off the fin. The pilots lose control of the plane.

134 SECONDS
As the plane plunges into a steep dive, tremendous forces rip the engines from their mounts. Flight 587 crashes in Belle Harbor, Queens. The engines land several blocks away; the tail fin tumbles into Jamaica Bay.

DISASTER DEBRIS: The vertical stabilizer, recovered from Jamaica Bay. PHOTO BY P. DERELLA/AP