Boeing 747-136 accident at Jamaica, New York, 20 Dec 1995

Accident narrative

On December 20, 1995, at 1136, Tower Air flight 41, a Boeing 747-136, veered off the left side of runway 4L during an attempted takeoff at John F. Kennedy International Airport (JFK) in Jamaica, New York. The flight was a regularly scheduled passenger and cargo flight to Miami, Florida. Of the 468 persons aboard, 24 passengers sustained minor injuries and one flight attendant received serious injuries. The airplane sustained substantial damage and was written off as a constructive total loss.

### The flight

Flight 41 carried 451 passengers, 12 cabin crewmembers, three flightcrew members, and two cockpit jumpseat occupants.

The captain, 53, had accumulated approximately 16,455 total flying hours, including 2,905 hours in the Boeing 747, of which 1,102 hours were as pilot-in-command. The first officer, 56, had 17,734 total flying hours, with 4,804 hours in the 747. The flight engineer, 34, had 4,609 total flying hours, including 2,799 hours as a flight engineer in the 747.

The weather at the time of the accident was partially obscured, with a 700-foot broken cloud ceiling, 1.5 miles visibility, light snow, and fog. The JFK surface weather observation showed that 1.3 inches of snow had fallen between 0645 and 1245.

### Pre-takeoff events

The captain stated that he received a weather briefing before 0830 and was aware of reports of compacted snow on the runways and that some runways were closed. The flight was pushed back from the gate at 1036, and final deicing and anti-icing with Type I and Type II fluids began at 1100. The flight was cleared to taxi to runway 4L at 1116.

The captain stated that the ramp was covered with packed snow and patches of ice. He noted the nosewheel skidded a little in a turn, but braking action was adequate. He stopped the airplane to clear the engines of ice by increasing power to 45 percent N1 for 10 seconds, but the airplane began to slip as power was advanced, and the procedure could not be completed at that time.

At about 1124, the crew heard another flight inquire about the availability of runway 31L. Ground control advised that it was closed and would probably be closed for a couple of hours. The captain stated that based on this information, he did not consider runway 31L a viable option.

Flight 41 was cleared to taxi into position and hold on runway 4L at 1132:06. The captain stated that he centered the airplane, moved the nosewheel steering tiller to neutral, came to a complete stop, set the parking brake, and did the engine anti-ice runup. The airplane did not move. He noted a strip of dark granular material down the center of the runway, with packed snow on either side and snow blowing horizontally across the runway.

### The attempted takeoff

At 1136, the local controller cleared flight 41 for takeoff, reporting winds from 330 degrees at 11 knots. The captain instructed the first officer to hold left aileron for crosswind correction and forward pressure on the control column.

The captain released the brakes and advanced the power, calling "Set time, takeoff thrust" at 1137:04. He stated that the takeoff began normally, with only minor corrections to maintain the runway centerline. Before receiving the 80-knot call, the captain felt the airplane moving to the left. He stated that he applied right rudder pedal without any effect, then added more right rudder and used the nosewheel steering tiller, but both were ineffective. He stated that he had no directional control and the nose continued to turn left.

The captain stated that while the airplane was still on the runway, he applied full right rudder and nosewheel steering tiller, retarded the power levers to idle, and applied maximum braking. He stated that he intentionally did not use reverse thrust because of the slow speed, the long runway, and the possibility that reverse thrust could worsen directional control. The airplane then departed the left side of the runway.

The first officer stated that the airplane veered left at less than 70 knots, and that he commented on the veer to the captain while the captain was attempting to stop. The flight engineer stated that he observed the captain using right rudder and tiller, and recalled that the captain immediately pulled all four thrust levers to idle and applied the brakes just before the airplane left the runway. A deadheading first officer in the jumpseat stated that he felt no swerve and that his first indication of trouble was when the captain retarded the thrust levers. He thought about two seconds elapsed between the power reduction and the airplane leaving the runway.

### Cabin events

When the airplane came to a stop off the runway, the crew performed memory shutdown items and discussed whether to order an evacuation. Determining there was no fire, the airplane was basically intact, and the outside wind chill was low, the captain elected to keep everyone on board.

The purser stated that he tried to call the cockpit on the interphone but received no answer. He ran upstairs to the cockpit, where the captain told him to keep the passengers on board. The purser returned to his station and made a public address (PA) announcement. Flight attendants stated that PA announcements were heard in the front of the airplane, but not in the rear zones. Three flight attendants stated they attempted to use the interphone to communicate with the purser without success. None of the flight attendants attempted to use megaphones.

During the accident sequence, the cabin floor in the forward section was displaced upward approximately two feet. In the aft galley, an ice cart and a beverage cart came loose. The ice cart hit a flight attendant, breaking her shoulder, and the beverage cart blocked an exit. The flight attendant responsible for the aft galley noted after the accident that the secondary latch for the ice cart was bent upward. Another flight attendant recalled noting before departure that the ice cart swing brake was not secured to the retaining tongue, and that she had been unable to lock it.

Based on the recollections of all flight attendants, only those at three positions shouted brace position commands while the airplane was still moving.

### What the investigation found

#### Wreckage and aircraft performance Tire marks on the runway showed the left main wing landing gear departed the left edge of the runway 2,100 feet from the threshold. The right main wing landing gear departed the edge 2,300 feet from the threshold. The airplane came to rest approximately 4,800 feet from the threshold and 600 feet to the left of the runway centerline. The fuselage forward of the No. 2 main entry door was severely crushed upward where the nose landing gear had collapsed. The No. 4 engine pylon was severely damaged and separated.

Because the flight data recorder (FDR) yielded no meaningful data, the manufacturer derived total airplane thrust values from a sound spectrum study of the cockpit voice recorder (CVR). This study showed that engine rpm reached 88 percent, then began to decrease to 75 percent at 1137:13. Within two seconds, thrust began to increase to a maximum of 91 percent N1, about the time the airplane departed the runway.

Using this derived thrust, an aircraft performance study calculated that the airplane departed the runway edge at an airspeed of about 97 knots, at an angle of 10.4 degrees to the left of the runway centerline.

The Board concluded that asymmetric thrust was not a factor in the loss of directional control. Based on the CVR spectrum analysis, the Board determined that the captain abandoned his attempt to reject the takeoff, at least temporarily, by restoring forward thrust before the airplane departed the left side of the runway, which contributed to the severity of the excursion.

#### Flight recorders and maintenance The FDR readout revealed that all parameters except time and synchronization lacked orderliness and reflected random values.

Maintenance records showed the FDR had been removed for an annual check on September 28, 1995. On November 3, Trans World Airlines (which performed the readout) issued a memorandum to Tower Air identifying six suspect data parameters. On December 1, a logbook writeup noted the FDR "OFF" light flickered, and the FDR was replaced. A second writeup noted the system test was inoperative, and the central electronics unit was replaced on December 2. On December 4, the six suspect parameters were entered in the log. On December 7, a digital acquisition unit was replaced.

The mechanic who replaced the unit stated he performed the required functional check after obtaining a tester rented at 0800 for four hours. Tower Air records indicated the airplane departed at 0955 that day. The Board concluded that based on the limited amount of time, Tower Air did not perform the FDR functional test, resulting in the loss of critical FDR data. The Board also concluded that the Tower Air maintenance program deviated in significant ways from its general maintenance manual, and that its continuing airworthiness surveillance and reliability programs were performing inadequately.

#### Simulator study and pilot techniques A flight simulator study was conducted using a Boeing engineering simulator programmed to reflect the accident airplane's parameters. Takeoffs were attempted under dry, wet, snowy, and icy conditions.

The study showed that when pilots attempted to maintain the runway centerline using the tiller under slippery conditions with a 12-knot crosswind, a slight overcontrol at the beginning of the takeoff roll repeatedly led to the nosewheel exceeding its critical angle and losing traction. The airplane then veered left in a weathervaning response to the crosswind. In most of these simulations, the airplane did not completely depart the runway before attaining sufficient airspeed for the aerodynamic rudder to become effective (50-80 knots), allowing pilots to arrest the veer with right rudder.

In contrast, when pilots did not use the tiller and limited control inputs to the rudder and rudder pedal steering, simulated takeoffs were controllable under all slippery conditions, including crosswinds gusting to 40 knots.

The Board concluded that the captain's failure to correct the deviation resulted from his overcontrolling the nosewheel steering through the tiller. The Board also concluded that the captain first relied on right tiller inputs as the airplane veered left, then applied insufficient or untimely right rudder inputs to effect a recovery.

Tower Air's flight manual stated that rudder pedal steering should be used with the tiller guarded until 80 knots. A 1994 company memo stated use of the tiller was not recommended unless rudder steering was not sufficient. Boeing's manual stated pilots should not use the nosewheel tiller during the takeoff roll unless required initially due to crosswind. The Board concluded these procedures provided inadequate guidance regarding the potential for loss of directional control at low speeds with the use of the tiller. The Board also concluded that improvements in the slippery runway handling fidelity of flight simulators used for Boeing 747 pilot training are both needed and feasible.

#### Galley security and flight attendant actions The Board concluded that the material or installation of secondary latches in the galleys was inadequate, as the crash forces were not severe enough that the latch material should have failed. The Board also found that Tower Air flight attendant galley security training was inadequate because flight attendants had not received "hands on" training with all the galley equipment they were required to operate.

Regarding the cabin response, the Board concluded that despite some ambiguity, there were ample indications in most parts of the cabin to have caused a greater number of flight attendants to shout brace commands. The Board also concluded that existing Tower Air procedures provided inadequate g