Four years ago Air France flight 447 fell out of the sky and crashed into the Altantic Ocean killing everyone on board. When the black box was located, two years after the event, it was determined from the data recorded, that the trigger for the event was the pitot tubes, which measure flight speed and altitude, freezing up in severe weather. This stopped the flow of flight speed data, which caused the autopilot to trip out, returning control of the plane to the pilots.
Who is to blame for what happened next remains a contentious issue, but the combination of confusion in the cockpit about what was happening, confusing and contradictory alarms and a lack of any visual information due to it being night and stormy conditions appears to have led to the plane being put into a stall from which it was never recovered.
Prior to the crash it had already been recommended that models of Thales pitots were upgraded to a different model or replaced with a model made by Goodrich Sensors and Integrated Systems which are regarded as being more resilient to adverse conditions.
A report issued by the AAIB in September revealed that two British Airways aircraft made emergency landings last year after their systems suffered similar weather induced malfunctions in the same way as flight 447 four years ago.
This time, the training of pilots to deal with just this sort of malfunction appears to have made the difference.
So what are pitot tubes and what is the problem?
Pitot tubes measure the flow of a fluid, in this case air, and based on the data returned, a calculation can be made of an airplane’s height and speed. The probes are fitted with heaters to stop them icing up. But during severe weather conditions the heating to the tubes can be overwhelmed and the tubes freeze up. When the data the aircraft is receiving degrades, the autopilot, rather than try to fly the plane with poor data, trips out and returns the control of the aircraft to the pilots. This happens abruptly and in difficult conditions.
When control is returned to the pilots under such conditions, it can be returned with some of the usual safety controls in place, or even none. The usual safety controls keep the plane in a protective envelope, preventing, for example, the plane being put into a stall. So when the auto pilot cuts out, some, or all, of this protection is lost. Two of the landings were made entirely manually with none of the safety systems operational.
The AAIB report reveals three separate incidents affecting two aircraft.
In the first incident on April 20 2012, a BA Airbus 321 with 183 passengers from Stockholm to Heathrow flew into storm clouds as it approached London. The tubes froze forcing the autopilot to switch off and sending erratic and contradictory data to the pilots. The flight diverted to Stansted, where it touched down safely without the instruments telling the pilot his air speed.
A second incident, affecting the same aircraft, happened on June 16. The plane was flying from Edinburgh to Heathrow with 183 passengers and crew, when its autopilot switched off while it was climbing through clouds. Once again an emergency landing was made at Stansted under manual control.
Despite the same aircraft being affected twice AAIB investigators blame the weather rather than the plane.
And on August 20 a British Airways Airbus encountered a -23C air pocket at 26,800ft, causing the instruments to give conflicting and incorrect information. On this occasion they were reset and the aircraft continued on its journey.
So why, given that there are known problems with the pitots haven’t they all been replaced?
Some people point to the regulation. The pitot tubes performance exceeds that which is required by the regulators, a point which Airbus has made repeatedly. Also, there is a low incidence of the tubes malfunctioning. But when they do malfunction, the results are potentially very dangerous.