The investigation of Flight 447

THE INVESTIGATION OF FLIGHT 447

In less than a week after the tragic event, the manufacturer of the airplane that crashed on 1 June 2009 sent an "accident information telex" to its customers having as theme the crash of Flight 447 in the Atlantic Ocean. Obviously, the aircraft manufacturer considered that this information is relevant to prevent future mishaps and so hastened to warn other pilots of similar aircraft.

The manufacturer explains that before doing so, the following data that are being reported have been approved by the Office of Investigation and Analysis for the safety of civil aviation of the country of registration of the crashed plane, taking in account that the normal procedure is that the analysis of aircraft crashes are conducted behind closed doors.

In this type of airplane there is a system that transmits routinely and automatically reports on movement and flight status to the company headquarters. As an airline air network can be very large and the airplane must meet several destinations often one after the other, not to mention passengers connections to other flights, having an automatic tracking of the activity of each aircraft significantly helps to organize the schedules of different flights.

In recent years, with great multitude of computer systems on board aircraft monitoring various vital functions, it was deemed possible to issue a report on the overall health of the aircraft without having to take it to the hangar for maintenance.

Thus each airliner periodically sends data on its internal systems and anomalies, so that the technical staff can plan maintenance for the most opportune date and using the minimum possible time to avoid disruption to the timeliness of routes which serve passengers.

In the case of Flight 447, some fault messages and maintenance orders have been transmitted automatically from the aircraft to the service center of the airline. One of them, at the beginning of the flight, refers to maintenance of one of the lavatories. But the following messages were much more dramatic.

These messages appeared in the form of automatic e-mails broadcast via satellite, minutes before contact with Flight 447 was lost. They indicate that there was an inconsistency between the different measurements in air velocities.

THE SPEEDOMETERS

An airplane has instruments to measure speed, i.e., speedometers. But unlike a car, what matters now is not the speed relative to the ground, but with respect to air, which is the medium in which these machines move.

Aircraft speedometers meet functions quite delicate, far beyond the trivial. Every airplane has a minimum speed at which it can fly. Below that speed, the wind passing through the wings does not generate enough aerodynamic lift and the aircraft falls like a rock. There is also a maximum speed above which the wind force is too high for the plane's structure and may cause damage, some with catastrophic consequences. These speeds and others to consider depend on several factors, including the weight of the aircraft at that point, the altitude at which it is and whether or not certain fins that increase, or others that decrease, the lift are deployed.

In the case of a high-altitude flight the issue is more delicate. Because up there the amount of air molecules is smaller, the aircraft must move much faster to get wind force sufficient enough for the wings to maintain lift. Therefore, in cruise flight at high altitude, the minimum speed required is uncomfortably close to the maximum safe speed.

For all of the above and without prejudging the outcome of the investigation, the data available from the automatic communication system of Flight 447 led to the airplane manufacturer to remind operators of similar models what are the applicable operational recommendations in case of an indication of unreliable airspeed.

The air velocity is measured with hollow tubes called Pitot tubes, positioned so that they point in the direction of flight, whereby the wind force entering into the orifice of their tips is proportional to velocity. This force is calibrated according to altitude, because as we know, the amount of air decreases as we go up and therefore also its force. This calibration is done by other orifices, protected from direct wind, on the sides of the aircraft's fuselage. Typically, there are several Pitot tubes and several of these side holes called of static air pressure, so that if one fails, the other can continue to give information.

PROCEDURES IN CASES OF FAILURES

The recommendations that the manufacturer refers to are in the operating manual of the aircraft, which should always be in the cockpit, at the pilots' hand reach for quick reference at all times.

Every commercial aircraft have at least three airspeed indicators in the cockpit, one for the pilot, one for the co-pilot and a third in the middle of both, as backup. The latter instrument is completely independent of the other, and a failure in one should not affect the other. Obviously, under normal conditions the three should indicate the same speed. If one or all three mark different speeds, there is a problem, certainly delicate, as the crew is confronted to an uncertainty in an information that is vital to keep the plane flying.

For an event of erroneous airspeed information in flight or during take off, or if airspeed indication is lost, the airplane manufacturer asked to perform, among other things, the following:

-Turn off the autopilot and the automatic thrust control of the engines.

-Maintain the flaps and slats that increase wing lift in the position they are.

-Check that the aerodynamic spoiler brake plates are retracted and not deployed.

-If all flaps, slats and spoilers are retracted, move the engine thrust levers to the climb position.

-If the aircraft is at more than 10 000 feet altitude (about 3000 meters), pitch the plane's nose to 5 degrees up.

-According to how the aircraft reacts, respect the alarm of insufficient airspeed (stall warning), and nose dive to gain speed if this is triggered.

-When the flight has stabilized, turn on the Pitot tubes heat to remove any ice residue that may clog them.

-Adjust the nose pitch and engine thrust according to the phase of flight and aircraft configuration to obtain and maintain the design speed.

In the event of a double failure of the Pitot probes heater, do the following:

-If icy weather conditions cannot be avoided: disconnect one of the affected Air Data Reference computers.

There are three of these computers, each independent of the other, which calculate, among other things, speed, altitude, angle of the wings with respect to the wind stream, the rate at which the aircraft gains or loses altitude and the speed of the airplane relative to the speed of sound, whose value this kind of airplanes can not withstand.

SPECULATIONS MUST BE AVOIDED

Again, it is important to note that this wake-up call from the manufacturer to all operators of the model of aircraft involved in this accident does not mean that they have discovered the true and end causes of this particular tragedy.

It is known that at the time of the accident the aircraft route was crossing a tropical area with many updrafts and downdrafts, with several points of localized weather activity. This environmental condition may or may not be involved, in one or more than one way, or in any at all, constituting the cause, an aggravating factor or only remaining as a simple non-consequential anecdote. No one knows.

The said Office of Investigations and Analysis for civil aviation safety warns that "A large quantity of more or less accurate information and attempts at explanations concerning the accident are currently being circulated." The Office "reminds those concerned that in such circumstances, it is advisable to avoid all hasty interpretations and speculation on the basis of partial or non-validated information."

"At this stage of the investigation, the only established facts are:"

"-the presence near the airplane’s planned route over the Atlantic of significant convective cells typical of the equatorial regions;"

"-based on the analysis of the automatic messages broadcast by the airplane, there are inconsistencies between the various speeds measured."

After extraordinary efforts of the country of origin and much tenacity, two years after the disappearance of Flight 447 they were able to locate the wreckage of the aircraft on the seabed. Then, with a special robotic underwater operation, the flight recorders were thankfully recovered. With them the initial telemetry data were confirmed, while they had access to new data that articulate with the first official information and to other subsequent ones:

-When the airplane entered the significant convective cells described, the speed was slightly reduced.

-After the incoherence in the various speed measures began, the nose was pitched 10 degrees up.

-The alarm of insufficient airspeed went off, but the nose was not pointed downwards to regain speed.

-Moments later, the huge airplane fell 12 000 meters in three minutes.

-The plunge occurred in a nose-up attitude, sugesting it was fully stalled.

With these data, many are wondering what the crew did in this circumstance, and why they did what they did. However, to air safety it is much more important to discover what another crew which unfortunately is in a similar situation will do in the future, and how to ensure they will actually do what they have to do, immediately and with minimal margin of error or doubt.

Finally, we should remember that it is very important not to err in determining the true cause of the tragedy, because if we come to a wrong conclusion many similar aircraft will continue to fly with a failure, human or technical, that will remain without being corrected.

Aldo Loup.

Originally published in ABC Color, on 7 June 2009. Map: The place where the airplane of Flight 447 disappeared is practically in the middle of the Atlantic. Retrieving the two flight recorders, one containing all sounds in the cockpit and the other hundreds of additional technical parameters, was a very difficult task. Map credit: Brazilian Air Force.