On the morning of November 12, 2001, American Airlines Flight 587 followed a Japan Air Lines Boeing 747 out to the runway at New York’s Kennedy airport. It was an Airbus A300 bound for Santo Domingo, with 260 people aboard. It lifted off about a minute and 40 seconds after the bigger jet.
A large airliner leaves a wake of turbulent air behind it, and Flight 587 flew into that wake twice. The first time, the captain said, “little wake turbulence, huh?” The first officer, who was flying, said, “yeah.”
The second encounter came at about 9:15 and 51 seconds. It was not severe. The National Transportation Safety Board later found that it “did not place flight 587 in an upset condition,” and that the airplane’s response “did not indicate that an upset was imminent.”
The first officer reacted hard. He turned the control wheel right, then all the way left, then right, then all the way left again. On the floor, his feet worked the rudder pedals: right, left, right, then further right, then left, then right. Each push swung the airplane’s nose sideways. The NTSB later examined whether the airplane’s response to his first push set off a loop, with each new input answering the last. The captain asked, “you all right?”“Yeah, I’m fine.” The captain said, “Hang onto it. Hang onto it.”
At 9:15 and 58 seconds, under seven seconds after the pedal inputs began, a fitting that held the tail fin to the airplane broke. The fin and rudder came off and fell into Jamaica Bay. Then the engines separated. Flight 587 came down on the Belle Harbor neighborhood of New York. All 260 people aboard and 5 people on the ground were killed.
The NTSB’s finding was that the loads which broke the fin came from “the first officer’s unnecessary and excessive rudder pedal inputs.” It named two contributing factors. The A300-600’s rudder system was unusually sensitive at high speed, with the lightest pedal forces of any transport airplane the board looked at. American’s upset-recovery training may have reinforced his tendency to respond aggressively to wake turbulence. On an earlier flight, another captain had told the first officer his rudder use was “quite aggressive.”
One sentence in the report stands out. The swing that built up to the fatal load “was solely the result of these rudder pedal inputs and was not associated with the effects of wake turbulence.” If he had stopped at any time before the fin broke away, the board wrote, “the natural stability of the airplane would have returned the sideslip angle to near 0º, and the accident would have been avoided.”
The airplane had been trying to settle itself. Each correction kept it from doing so.