Air France Airbus 330 – Unreliable Airspeed Problems Since 2006

Unreliable speed is one of the difficult situations that a pilot has to face. Once the failure has been identified, a procedure, based on pitch angles and thrust settings, will assist the pilot in safely flying the aircraft...

...the main difficulty is to rapidly detect an unreliable speed situation. Reaction time is crucial, since the aircraft may stall and overspeed conditions could cause aircraft damage.

...‘We’ were already aware of what the problem was the morning of the accident (the famous erroneous airspeed). There are passed records that display the exact ACRAS messages sent by the AF477.

The same causes had already produced the same effects. It is likely that the AF447 experienced the same problems as all the other aircraft that encountered a Pitot defect, the only difference is that those other aircraft had managed to free themselves from that situation.

...The BEA’s communications through the media have raises questions. We believe that the families of the victims have been left in confusion, dragging in despair all that has been said in the press. In the end, we feel that an explanation of the origin of the accident is probably dramatically simpler.

... At the date of this document (file made in June 2008, note dated August 2008), it is written that there had already been 6 cases that had produced the same alarms. The Pitot probes were designated undoubtedly as the cause of these problems...

...It is written that the probe type BA have solved the problems: “The new pitot probe corrects the problems with enhanced water trap and relocated drain holes.”

The critical problem with pitot issues on a Bus330 is that the airplane is essentially computer controlled. Even when the pilot is “hand flying” the aircraft, the computer will restrict the effect of direct pilot control input if it computes any “flight limitation” exceedance.

Therefore, if the computer is receiving bad information from the various data sources and restricts the input of the pilot who recognizes the erroneous inputs, the computer becomes like HAL 1000 and a battle ensues. Midpoint over the Atlantic in a thunderstorm would be very bad timing for this to happen, but it's conceivable.

The pitot history of this particular airframe is quite telling and will no doubt continue to be the main focus of the investigation into this crash.

FYI: The pitot system takes in ram air and compares this dynamic pressure to static pressure. The difference is, essentially, indicated airspeed (IAS). It's the speed displayed to the pilot on his primary instrument panel. (Ergo, if the aircraft is not moving, dynamic pressure equals static pressure and the airspeed indicates zero.)

Speaking as a private pilot with NO experience in the big airplanes, if the weather is clear and it's daylight and you have visual contact with the surface and the horizon, and that's a lot of “if’s”, it is not impossible to detect and correct for an impending stall or excess speed. But, and it's a HUGE BUT, at night in the clouds, in a thunderstorm, experiencing severe turbulence, flying a machine big as football field, it's very difficult, impossible, to keep the aircraft under control once you lose the ability to reference your instruments. Any one of these conditions would make it a real challenge to keep flying, and every time you add one to the pile the difficulty goes up by the square of the number on the pile.

In smaller aircraft you are getting all sorts of clues from your surroundings as to what's going on, things like the sounds, the view and the feeling from your butt. I would think fly-by-wire would deprive you of most if not all of this input.

My guess is when they get it all figured out it will be readily apparent that the crew performed in a heroic manner in attempts to fly the plane, but in the end was simply overpowered by erroneous data and/or the loss of a valuable part of the aircraft, like the tail section or a wing, as a result of turbulence.

My heart and prayers certainly go out to both the crew and the passengers, not to mention the families.

Those methods are fine for situations where time and exactness is not critical, but in moment to moment flying they are neither accurate enough or fast enough to be relied upon. It’s interesting that even with all the improvements that have been made to the science of flying in the last 100 plus years we STILL depend on a device that was with us when the first aircraft left the ground.

At 400 mph plus just the time it takes to communicate with a satellite, even passively, then cipher the information and do the calculations to make the info useful you’ve gone a long ways. Also, it’s air speed that’s important to you, and GPS and radar report ground speed. Their can be a big difference, enough to bring you down.

You’re right on both counts. Airline pilots utilize autopilot to a much greater extent today than 15-20 years ago. Part of that is the complexity of aircraft has increased exponentially, the airframe designs are much more critical in terms of limitations (for speed and effeciency reasons), and there are now only two pilots in the cockpit. Back in the day of flight engineers (FE), there was a third guy to operate aircraft systems while the other two navigated and aviated. Automation made the job of the FE obsolete.

Flight automation (autopilot) was employed more and more to alleviate the flying pilots from distraction in the event aircraft systems required pilot attention.

Scheduling of pilots for long days, multiple legs and minimum rest also adds to the pressure to use autopliot more. Fatigue is a major issue in the industry, and autopilot does take a load off during cruise flight. In fact, during cruise flight with the new 1000ft minimum vertical separation (RVSM) between aircraft on the high altitude airways, autopilot use is required by law (FAR).

A lot of pilots make a point to turn off the autopilot frequently to keep their “handflying” skills current. Then again, many don’t.

The fly by wire in the Bus330 will most certainly be a factor in the loss of this aircraft. Remember when the chief test pilot for airbus crashed the jet (A320) at an air show because the fly by wire computer interafce wouldn’t let him do what he wanted to do. He wanted to add power but he had put the computer into a mode that made it impossible?

Then there was the 3rd world country flight crew that was electrically trimming the flight controls of their A320 with the autopilot on. The computer dutifully countered their trim inputs until it could no longer do so and the autopilot tripped off. The aircraft became uncontrollable, control was not regained, the aircraft and passengers were lost.

Computers are a frustrating thing while sitting at a desk as they refuse your coaxing and cajouling. In an airplane going 550 knots...

Hope that helps a bit.

Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.