Flight 587 Crash Hearing Focuses on Rudder Controls

Seconds before American Airlines Flight 587 crashed in a New York neighborhood last November, the frantic pilots realized they had suddenly and without apparent warning lost all control, a transcript released Tuesday shows.

“What the hell are we into?” exclaimed 1st Officer Sten Molin, 34, who was piloting the jetliner. “We’re stuck in it!” The routine takeoff became a nightmare as they encountered forces they could not overpower.

“Get out of it, get out of it!” responded Capt. Edward States, 42. The urgent plea came too late. The tail had ripped off, making the Airbus A300 impossible to fly.

The two pilots in the cockpit were in all likelihood unaware that they were now part of a previously unheard of scenario: the complete separation of a transport aircraft’s vertical tail fin during normal flight.

It took fewer than eight seconds from the time of Flight 587’s second encounter with the wake of a larger plane ahead for the tail to break, according to new data released by the National Transportation Safety Board in the first of four days of hearings.

In the brief interval before the tail broke, the plane’s rudder moved sharply back and forth, stressing the vertical fin beyond its design limits. Normally, pilots make very little use of the rudder in flight.

Tuesday’s hearings focused attention for the first time on the design of the rudder controls and whether their sensitivity played a role in the accident. Moreover, newly released materials showed that four years ago, Airbus had raised concerns with the NTSB that large rudder movements by pilots could over-stress a tail fin. It is unclear whether the information prompted any action.

Investigators say they have reached no conclusions about what caused the crash, which killed all 260 aboard and five people on the ground in the Rockaway neighborhood of Queens. It ranks as the second worst commercial aviation disaster in U.S. history.

In addition to the design of the rudder controls, the probe is focusing on several elements: a possible overreaction by Molin using the rudder, the adequacy of safety standards for the tail fin and rudder and the training given to pilots.

Earlier concerns about the composite material used in place of metal to build the tail fin have been eased, since investigators have been unable to discover any preexisting flaws. But NTSB officials say they are considering recommending more sophisticated inspections of composite components.

“The accident raises questions about the design of the [rudder] system and the training of the pilots who use it,” Greg Feith, an air safety consultant and former senior investigator for the NTSB, said in an interview. “I don’t think that training is as thorough as it should be.”

The rudder, a movable panel on the tail fin, is mainly used to keep an airplane flying straight when landing or taking off in a crosswind, or when an engine fails. To move it, pilots use a system of pedals.

Investigators have said they are specifically interested in a design feature of the A300 rudder controls that requires less pedal force to apply full rudder under certain conditions.

At 250 knots, close to the speed at which Flight 587 was flying, a force of 32 pounds is required to move the rudder to its full limit. At lower speeds, below 165 knots, the pilot must apply a significantly stronger force of 65 pounds.

“This system, as designed, leads a pilot to believe the rudder reaction is proportional to the pedal input, which, in fact, it is not,” Feith said.

Airbus defends its design, noting that about 75% of transport aircraft rely on a similar philosophy. “The rudder system has 15 million flight hours with no concerns expressed prior to this accident about feel and design,” said Clay McConnell, a spokesman for the company.

In a 1998 letter from Airbus to the NTSB, which concerned an incident involving another American Airlines A300, the company warned that pilots should not rely on the rudder to recover from in-flight upsets. “Large or abrupt rudder usage ... can lead to rapid loss of controlled flight,” Airbus said.

Side-to-side rudder movements “can lead to structural loads that exceed the design strength of the fin,” it added. The 1997 incident that prompted the letter did not cause a crash. The NTSB did not warn pilots of the potential danger until February.

“The use of the rudder for boosting control is neither correct nor appropriate,” testified Dominique Chatrenet, an Airbus vice president who oversees flight controls. Capt. Larry Rockliff, an Airbus vice president for training, testified that in the mid-1990s, he was concerned that American pilots had been trained to rely too much on the rudder when recovering from in-flight upsets. American later changed its training to de-emphasize the rudder.

Airbus is at odds with American and with pilots’ groups over the cause of the accident. The airline and pilots say Molin should not be blamed, since pilots until now have been trained that they can safely make full use of all controls at normal flying speeds.

But the investigation has also turned up an incident five years ago in which Molin was criticized by a senior pilot for overusing the rudder to stabilize a Boeing 727. No one was hurt and the captain did not file a report.

An NTSB video reconstruction showed that Molin and States had little time to analyze what they were up against during the two minutes and 24 seconds that Flight 587 was aloft. The plane took off without incident on a clear Monday morning bound for Santo Domingo in the Caribbean.

A few seconds later, Flight 587 crossed the first wake of a larger Boeing 747 about five miles ahead. Airplanes leave two wakes, which curl off the end of each wing like horizontal tornados. “Little wake turbulence, huh?” States said to Molin, who was at the controls.

“Yeah,” Molin responded. The plane was making a left turn, following a standard departure route. But the encounter with the second wake was anything but routine.

The sharp rudder motions began almost immediately. Although investigators said Tuesday they have still not determined that Molin initiated the maneuvers, no mechanical problem has been found that would have caused them.

The cockpit voice recorders picked up the sound of several thumps and mechanical noises. In a strained voice, Molin called for “max power.” He recognized there was a problem and he was trying to regain full control.

“You all right?” asked States.

“Yeah, I’m fine,” Molin responded.

“Hang on to it,” said States. “Hang on to it.”

The cockpit recorder picked up the sound of another thump. Then came a loud bang, which investigators believe was the sound of the tail ripping off. That was followed by grunting sounds from Molin. Then came a roaring noise, rising in volume.

About 14 seconds later, the tape cut off. It was then, investigators believe, that the plane slammed into the ground.