Verified by 
Until the “black boxes” are found and studied, we will not be sure of the cause of the AirAsia crash. However, two pieces of information point to a replay of the Air France Fight 447 crash five years ago when the plane's speed system malfunctioned.
One piece of information is radar that shows the plane, just before disappearing from the screen, flying much too slow. A second clue is the distance from where the plane was last seen on radar and the area in which debris has been found. A plane gliding down from a cruise altitude of 34,000 feet would cover a distance of over a hundred miles. But, the debris area is much closer than that to the plane’s last position on radar. The short distance suggests that the plane, rather than gliding down at a normal rate of descent, descended quite rapidly. The most likely explanation of such a descent is speed too slow for the wings to support the weight of the plane.
On an Airbus the computers that control the plane are dependent upon accurate information being fed into the computers. When computers first became common, people used the term GIGO, which stood for Garbage In, Garbage Out. If bad information goes into a computer, bad information comes out. It's not acceptable to allow "Garbage Out" to fly an airplane, so when Airbus designed the computer-controlled Airbus A320, its computers were programmed to recognize Garbage In. How?
To control the plane's speed, all three of the speed sensors send information to the computers. If all three send identical speed information, the computers accept the information. But, if one sensor says 220 mph, another says 280, and yet another sensor says 50, the computers respond as if to say, "Hey, I can't work with garbage so I'm turning the plane over to you, Mr. Pilot. You figure out what's going on and tell me how to fly the airplane. Give me manual inputs using the joy stick."
Yes, a joy stick, just like on a computer game. Other input devices substitute for the throttles and rudders on a conventional airliner. When the computers turn to the pilot for manual imputs, the pilot is expected to do the following:
- Understand that the computer regards the information provided by sensors as unreliable. It could be that one speed sensor is right. But which one? Or, none could be right.
- Fly the plane by hand.
- Follow the procedure for this situation: don’t change the altitude or power setting (or use settings prescribed in the flight manual) to maintain a speed that causes the right amount of air to flow over and under the wing.
- Continue until the sensors again produce reliable information.
The Cause Of Unreliable Speed Information
Some thunderstorms produce tiny ice crystals that, if plentiful enough, collect in the pitot tube, a vital part of the speed sensing system. A pitot tube is an open pipe that is aimed straight ahead of the plane so that the faster the plane goes, the more pressue builds up inside. The amount of pressure is used to determine the speed of the plane through the air. Pitot tubes are heated so that ice crystals, instead of clogging the tube, melt.
The standard pitot tube for this airliner, built in the U.S. by Goodrich, initially had some blockage problems, but after modification, has been reliable. Some Airbus planes were equipped with a pitot tube built in France by Thales. Some of the original Thales tubes have, in a few cases, become temporarily clogged by ice crystals. Redesigned Thales tubes are also considered reliable. Safety expert John Goglia’s comments on pitot tube icing at this link http://www.forbes.com/sites/johngoglia/2014/12/28/asiaair-flight-8501-what-are-pitot-tubes-and-how-could-they-affect-flight/
Pilot Response To Unreliable Speed Information
If the pilot recognizes the speed information is unreliable, he or she is expected to follow the memorized unreliable speed indication procedure and follow that with the troubleshooting procedure. See images.
Unreliable Speed Indication Procedure
If the pilot doesn't recognize the speed indications are bad and changes the plane’s speed based on bad information, the plane may be flown too fast or too slow. In the case of AF 447, the pilots – believing the bad information to be accurate - caused the plane to fly too slow, When the speed indications became reliable, the pilots believed them to be unreliable and continued flying too slow, causing the plane to continue descending rapidly until it crashed into the ocean.
Flying The Plane Via Manual Inputs To The Computers
Pilots flying a computerized airliner may fly too little by hand to maintain their skills. If hand flying skills have deteriorated, if the plane unexpectedly turns the flying over to the pilots, they may not be up to the job. Flying by hand may have been more difficult in this case. The procedure for unreliable speed indication is to maintain the plane's altitude and the power settings used prior to the unreliable speed indications. But, it appears the pilot was climbing when the speed readings went bad. According to the procedure, he should have leveled the plane and set the power at the power needed to maintain level flight at that altitude. After transitioning from climbing to level flight, he may have needed to look up the power setting information in the flight manual, complicating the task of flying the plane manually.
Troubleshooting Checklist
It has been said that in a crisis, a person does not “rise to the occasion” but, instead, descends to the level of his or her training. To adequately handle this situation, a pilot would need to have been thoroughly trained by performing the procedure in a simulator. In addition, the pilot woulld need to fly the plane by hand frequently enough to maintain basic flying skills.
Action Taken In Response To The Air France 447 Crash
At Air France, the problematic pitot tubes were replaced with reliable ones. At Airbus, a system to produce airspeed information synthetically was developed and is standard on the A380 and available as an option on other Airbus models. At Boeing, synthetic airspeed is standard on the 787, and - since earlier models are not so computerized - synthetic airspeed is not needed.
For in-depth information see this article by an Airbus engineer at http://aviationtroubleshooting.blogspot.com/2009/06/af447-unreliable-speed-by-joelle-barthe.html
