The problem about the Air France beyond the objective troubles of night, tropical thunderstorm, turbulence, etc..
is that in fact the pilot reacted exactly as he was instructed at the simulator. He reacted to a thunderstorm as if they had a windshear.
Windshear are huge changes in directions and speed of wind when flying close to the ground (not at 35000´), but is one of the more trained maneuver as it is a very dangerous situation. If you analize the actions of the pilot flying, you´ll see that he reacts as it´s supposed to react to a windshear advice, full throttle and the sidestick full back, he stayed like that for almost 2 minutes.
When you are close to the ground the engines have the power to make you climb fast and the flight envelope protections keep the plane from entering in to a stall.
That´s why he didn´t understand what was happening, in his point of view he was making the right think to get out of the thunderstorm.
The problem is that he had a totally different situation, a instrument failure. When you have those you only have to keep your thrust and attitude constant and check a specific table that we have for this situation. In that table it´s specified the thurst position and attitude (degrees of nose up seen on the artificial horizont) that you must maintain to keep a cruise speed at a given level.
If he had managed to do just that, staying as they were, they wouldn´t had any problem.
Once they where in a stall, the normal reaction is to pitch down and increase slowly the thrust, once you have enough speed you recover the horizontal attitude. Another problem is that they couldn´t understand that they were in a stall, in normal mode the Airbus can not get in a stall, but as the instrument failed they were able to get below the protections and once they were there the computer didn´t have the possibility to react, it sensed the situation as a data failure and didn´t give alarms some of the times.
There has been too much training involving the automatic safety measures (they are indeed great, but they are not all that it takes to fly) and too little involving normal and abnormal manual flight maneuvers. We have lost the touch of how to flight a plane.
> We have lost the touch of how to flight a plane.
To Greenspun's argument that north-american pilots typically have extensive hours on smaller airplanes (being flight instructor and bush pilots to build up the hours), where they had to fly them manually. In other part of the world, those opportunities are just not there.
Microburst only happens at low altitudes ( http://en.wikipedia.org/wiki/Microburst ), and is the kind of wind shear that´s more dangerous for commercial planes.
There is also fast wind changes at cruise altitude, jet streams bends usually have. But you only notice them as turbulence, when they are very severe you may even loose several thousand feet. It´s not pretty and may result in some nasty injuries to the people that doesn´t have the seat bealt. But is not as big an issue as the low altitude wind shear. Also clear air turbulence is not predictable and is little that you may do beyond recovering the plane control as soon as you suffer it (as opposed to thunderstorms, where you have radar echos to avoid).
Windshear are huge changes in directions and speed of wind when flying close to the ground (not at 35000´), but is one of the more trained maneuver as it is a very dangerous situation. If you analize the actions of the pilot flying, you´ll see that he reacts as it´s supposed to react to a windshear advice, full throttle and the sidestick full back, he stayed like that for almost 2 minutes. When you are close to the ground the engines have the power to make you climb fast and the flight envelope protections keep the plane from entering in to a stall. That´s why he didn´t understand what was happening, in his point of view he was making the right think to get out of the thunderstorm. The problem is that he had a totally different situation, a instrument failure. When you have those you only have to keep your thrust and attitude constant and check a specific table that we have for this situation. In that table it´s specified the thurst position and attitude (degrees of nose up seen on the artificial horizont) that you must maintain to keep a cruise speed at a given level. If he had managed to do just that, staying as they were, they wouldn´t had any problem.
Once they where in a stall, the normal reaction is to pitch down and increase slowly the thrust, once you have enough speed you recover the horizontal attitude. Another problem is that they couldn´t understand that they were in a stall, in normal mode the Airbus can not get in a stall, but as the instrument failed they were able to get below the protections and once they were there the computer didn´t have the possibility to react, it sensed the situation as a data failure and didn´t give alarms some of the times.
There has been too much training involving the automatic safety measures (they are indeed great, but they are not all that it takes to fly) and too little involving normal and abnormal manual flight maneuvers. We have lost the touch of how to flight a plane.