The accident occurred on the night of February 6, 1996, at 11:47 p.m. local time during a departure from the International Airport at Puerto Plata with a destination of Frankfurt, Germany, with a technical stop in Gander for fuel.

In accordance with Annex 13 to the convention of ICAO, the DGAC of the Dominican Republic, which was in charge of the investigation, requested technical assistance from the U.S. NTSB. The NTSB sent a team of investigators led by Mr. Jorge Prellezo, who was accompanied by a representative of the FAA, Boeing company, (the airplane manufacturer), and Rolls Royce (the engine manufacturer). Official representation was included from Turkey (the state of registration of the airplane and the state of airline BirgenAir), and the Federal Republic of Germany (as observers), and also representatives of BirgenAir airline who are responsible for the operation (flightcrew and maintenance).

Immediately after the accident, necessary coordination was initiated to locate and recover the flight data and voice recorders. For this the U.S. NTSB and the DGAC of the Dominican Republic coordinated with U.S. Navy specialists in underwater search (SUPSALV) to conduct the operation.

The signals of the recorders were located on the 15th of February 1996 in the territorial waters of the Dominican Republic about 5 miles off the Coast of Puerto Plata at a depth of about 7200 feet.

Immediately the government of the Dominican Republic, the United States, and Germany and the commercial companies of Boeing, Rolls Royce and BirgenAir devised a cost sharing plan to recover the recorders at an estimated cost of $1.4 million.

The recorders were recovered from the Atlantic Ocean on February 28, 1996 and immediately transported with the investigators from the Dominican Republic to the laboratories of the NTSB in Washington, D.C.

The accident airplane was a Boeing B757 registration TC-GEN, owned by the BirgenAir company which operated aircraft from Puerto Plata, Dominican Republic, and which completed the last routine maintenance inspection on January 16, 1996.

Various flights were conducted which terminated on January 23, 1996. Since that day until February 6, the airplane remained on the ground.

During that time the pitot heads were not protected with the appropriate covers. Procedures are not specific.

Examination of the data from the flight recorders revealed:

All the preflight activity on the ground was normal.

During the takeoff roll a check of the airspeed at 80 knots revealed that the captain's airspeed was not functioning.

The takeoff continued and the captain had the copilot call out the airspeeds necessary to rotate the airplane and initiate the flight.

The copilot made the required call outs and the airplane climbed normally, the necessary actions were made to connect the autopilot and continue the flight.

When the airplane reached about 4700 feet, about 2 minutes into the flight, some advisory messages appeared informing of flight control irregularities (rudder ratio & mach/airspeed).

During the following minutes there was an interchange of opinions between the pilots about the nature of the advisory messages.

The radio communications with the airplane were normal and there was no indication of any inflight regularity.

Comments followed between the pilots about confusion that was occurring between the (airspeed) indication systems from the left side airspeed indication system, affecting the indication of the left side airspeed autopilot and activation of the overspeed warning.

The airplane continued flying with the autopilot connected and receiving and erroneous indication in the captain's airspeed (indicating system).

The following flightcrew comments indicated the same confusion from the previous moments.

The recorded sounds and the recorded flight data indicate extreme conditions of flight, one corresponding to overspeed and the other to slow speed (stick shaker), which are contradictory.

The captain initiated an action to correct the overspeed and the copilot advised that his airspeed indicator was decreasing.

The airplane has three airspeed indicating systems and at no time did the flightcrew mention a comparison between the three systems.

The airspeed that the captain mentioned was over 350 knots, the ground radar system registered speeds of less than 220 knots.

The time the flight recoders indicated flight out of control was 1 minute 41 seconds until the end of the recording.

All of the parameters of the engines and different airplane systems were normal and consistent with the different maneuvers indicated by the flight data recorders.

A complete study of the systems by more than 30 experts determined that the anomalies in the flight data recording corresponds to conditions equal to an obstruction in one of the sensors of airspeed (pitot head).

The above conditions were simulated and verified by using a B-757 simulator.

[Signed]
Emmanuel Souffront Tamayo
Mayor, Piloto, FAD
Presidente Junta Investigadora de Accidentes Aereos
DGAC

Back to `Incidents and Accidents'.

Peter B. Ladkin, 1999-02-08
	Last modification on 1999-06-15 by Michael Blume