Cockpit chaos on doomed Air France jet

2011-07-30 18:17

Le Bourget - A confused cockpit crew without proper training to head off high-altitude disaster flew toward it, instead, with incorrect manoeuvres, no task-sharing and perhaps unaware their flight was about to end in the Atlantic Ocean.

Screeching stall alarms and incoherent speed readings from faulty sensors, bad weather in a darkened sky and growing stress make up the chaotic cockpit scenario in the final moments of the Air France flight from Rio de Janeiro to Paris on June 1 2009.

All 228 people aboard the plane were killed.

Friday's third report by France's accident investigation agency, or BEA, lays out almost second-by-second technical data on the flight's deadly trajectory but cannot answer the ultimate question - whether pilot error, equipment failure or other still unknown factors caused the crash.

The BEA's findings raised worrisome questions about the reactions of the cockpit crew - two co-pilots - as the A330 went into an aerodynamic stall and their ability to fly the A330 manually as the autopilot disengaged.

The report expressed broader concern about the state of training of today's pilots flying high-tech planes when confronted with a high-altitude crisis.

BEA officials said they are bringing together a bevy of experts, from psychologists to physiologists, to try to reconstitute the scene from the crews' point of view - the human factor which could include potential disorientation.

Those findings would be included in the final report expected early next year.

Why it happened

Many of the crews' actions "seem contrary to logic and we're seeking rational explanations," chief BEA investigator Alain Bouillard told a news conference, adding that the cockpit crew even seemed unaware the plane had gone into an aerodynamic stall.

"We understood how the accident came about," Bouillard said. "Now we must learn why it came about."

Friday's 117-page report, based on a full reading and analysis of the flight and data recorders dredged from the ocean depths, recommends mandatory training for all pilots to help them fly planes manually and recover from a high-altitude stall.

With the captain of Flight AF447 on a rest break, the report also expressed concern over "non-optimal task sharing" between the two co-pilots. Among the BEA's 10 recommendations, it wants authorities to further define criteria for appointing a relief captain to ensure better synergy among relief crews.

When the captain of the Air France flight returned in the midst of the crisis, "neither of the two co-pilots gave a precise accounting of the problems encountered nor of actions undertaken, except that they had lost control of the plane and that they had tried everything," the report said.

The captain had "implicitly" appointed the younger co-pilot as his relief before taking his regulation nap.

Experts caution against laying blame on the pilots - all experienced and qualified to fly the aircraft.

"The information they're getting from the brain of the airplane, the thing that they've been trained to trust, is sending them all off on tangents," said John Goglia, a former US National Transportation Safety Board member and an expert on airline safety.

Malfunctioning computers

"They've got bells and whistles going off, they've got a face full of lights," Goglia said. And yet, "the pilots had an awful lot of information denied to them to help them deal with the situation" because of malfunctioning computers.

The crew had less than 4.5 minutes to act to correct an aircraft that was sounding alarms and giving sometimes false readings. However, BEA chief Jean-Paul Troadic said that, at the start at least, "the situation was salvageable".

The chilling scenario as described by the voice and data recorders began at 2 hours, 10 minutes and 5 seconds into the overnight flight, when the autopilot and auto-thrust disengaged and a stall warning sounded twice in a row.

The recordings end at 2 hours, 14 minutes and 28 seconds.

The co-pilot designated by the captain quickly took over manual controls of the aircraft, and nosed the plane upward - the opposite of what was needed to give the plane lift.

A basic manoeuvre for stall recovery, which pilots are taught at the outset of their flight training, is to push the yoke forward and apply full throttle to lower the nose of the plane and build up speed. But he nosed up throughout much of the impending disaster and the plane reached a maximum height of 38 000 feet.

The report confirms that external speed sensors obstructed by ice crystals produced irregular speed readings on the plane. Since the accident, Air France has replaced the speed monitors on all its Airbus A330 and A340 aircraft.

The BEA's report noted that Airbus warned pilots in 2008 that incorrect speed readings from the Pitot tubes could cause erroneous stall warnings. But Bouillard, the chief investigator, maintained that the pilots should "always respect a stall warning".

Passengers never knew

Passengers, finishing dinner or napping, were never advised of the plane's plight.

"From what we've been told, nobody realised what was going on. On that level, for my mental and moral comfort I am very pleased to hear this, when you know you had two people on board who were dear to you," said Corinne Soulas, whose 24-year-old daughter Caroline and son-in-law were aboard the flight.

The alarms - computer-generated voices screeching "stall ... stall ... stall" sounded numerous times, and once for a full 54 seconds.

Incomprehension and growing tension ensued.

At an alarm sound, the co-pilot not flying said at one point "What's that about?" Curiously, the crew made no reference in cockpit exchanges to the warnings, the report said.

As described by the BEA report, several calls were made to the captain. The pilot not flying expressed concern several times at the captain's absence, a concern that "probably raised the stress level of [that] pilot as he faces a situation he doesn't understand".

The flying pilot twice said he had lost control of the plane. Then, 27 seconds later the pilot not flying takes control but the designated pilot retakes control "almost immediately without any announcement".

A minute and a half later, the captain arrived, but with no pertinent information from the co-pilots and a lack of information from the control panel he appears not fully aware of the situation - and did not ask questions to better understand.


The report said that "multiple stops and reactivation [of the alarm] probably added to the confusion and disturbed his diagnosis of the situation".

There has been a silent tug-of-war between Air France and Airbus, the plane's constructor, over the crash.

Both were charged last March with involuntary homicide following the accident.

In a statement, Air France said there was no reason to question the crew's technical skills. The airline said the report showed that a series of unlikely failures led to the stall and crash.

The president of an association of Brazilian victims' family members criticised the latest report.

"If the pilots were not well trained, how can France allow them to work on an overnight international flight with passengers on board?" said Nelson Farias Marinho, who lost a son on the flight.

"I completely disagree with this report and I am against it because it is full of contradictions," he said.

  • Dmitri - 2011-07-30 19:08

    What a lot of cr*p. Did this idiotic reporter read the BEA report 1.These folks were properly trained. The Stall warning activated and they did the correct thing (Nose down to decrease the AoA). Stall Deactivated and they then put the nose up again. This occured a few times. The BEA report does not state how many times. 2. Incorrect manoeuvres. Once again false reporting. They deviated 12 degrees to the right. That is a manoeuvres. Joystick and rudder input are not manoeuvres 3. No task sharing. Once again false. The PF was in the right seat and the PNF in the left seat. The PNF read out to the PF what the instruments were showing. Note that the initial situation was caused by faulty/frozen pitot tubes. 4.Unaware that they were falling. Once again an idiotic statement. They were cruising at FL350 and did a "zoom climb" to FL380 where most of the control inputs took place. They then decended at over 10000 fpm. The PNF stated that they were reaching FL 100 Screeching stall alarm. It is loud but not in a "screeching voice"Bad weather...what the hell is bad weather? They were crossing the ITCZ which is always turbulent. There were infra red sat photos of TCU which has been made avail to the public but their wx radar did not "paint" red. I do not work for Air France or the BEA or Airbus in case you folks were wondering. If any other folks who work in the pointy end would like to comment, great stuff. The journalistic ethics are zero.

      Marc - 2011-07-30 21:48

      Some very valid points there Dimitri. Could be that some people are trying to protect their own arses

      Craig Edmondson - 2011-07-30 23:12

      Ok, so what actually happened?

      Jack Johnson - 2011-07-31 00:24

      Dmitri, I agree that News24 does not set the standards on good reporting. That being said:I have never met a driver who thought he (or she) made a mistake. I presume from your use of aviation acronyms and lexicon that you are involved in or at least interested in aviation. I have been following the situation and like most people truly interested in the event, have read the following BEA publications: 1) 2) So let us take a look: 1) These folks were properly trained as far as required by all aviation authorities and the airline standards. This doesn't mean that there wasn't some situations which they were not properly trained to address. As far as the stall warnings go. The report makes it clear that the Stall warning was deactivated during certain conditions, such as when the speed was less than 60kts. So stall warning stopping does not indicate that the aircraft is out of stall. 2) I believe the incorrect manoeuvres referred to were the various pitch-up commands. PS. Joystick? 3) Interaction is not task sharing. The two pilots did not address any of the issues causing the problems. For example the report states: ' Even though they identified and announced the loss of the speed indications, neither of the two copilots called the procedure "Unreliable IAS". For an example of how task-sharing solves problems, see Quantas Flight32

      Jack Johnson - 2011-07-31 00:25

      4) Yes, that is an idiotic statement to make. Especially with the clear statement that they are approaching FL100. Regarding the weather: the interem report ( goes to some length to show the level of hazard that the weather held to AF447. So, in summary- What I believe BEA is saying is that a combination of factors led to the demise of AF447 and the souls she carried. However, while no one could ever say with 100% certainty, proper training of the flight crew in situations of high-alitude stall combined with a loss of the AP's could possibly have brought a different outcome. What people should realise is that saying an accident was "pilot's error" does not equate to *blaming* a pilot for an accident. Stay safe!

      mavuso77 - 2011-07-31 16:14

      And I guess you were there Dmitri?

      buka001 - 2011-08-01 21:11

      Dmitri have you read the BEA report correctly? See this link - It clearly says the pilots nosed up on stall warnings.

  • Tekwini Durban - 2011-07-30 23:20

    for a full and unbiased report - see this link -

      Serias - 2011-07-31 08:44

      Ive read some of your comments on on this and think you understand the subject. All this technical garble confuses me, but im very interested in what happened. My understanding: The probes gave incorrect speed feedback which made the computer believe it was in a stall and thus caused autopilot and auto thrust to disengage. But the plane was actually not in a stall, and the resulting nose up by the pilot actually caused a stall??? Can you maybe explain in "normal" terms, according to you what actually happened.

  • NZdreamer - 2011-07-30 23:50

    Aggggg.....MY biggest nightmare regarding flying!!!

  • Geo - 2011-07-31 04:29

    Just go and look on the history of Air France and you will see how many of their planes have crashed. I have never flown on Air France and will never, even if they give me a 99% discount.

  • pawsaw - 2011-07-31 12:30

    Reading these comments by people who have some knowledge and understanding is somehow comforting, An observation I would make however would be to ask why, if the weather conditions were so abysmal, did the Command Pilot choose to take his rest period so early into the flight. Wouldn't it have made a difference had he waited until they cleared these weather conditions b4 taking a nap? Ultimate responsibility may well be the mechanics and reliance placed by all flying it on the computer brain and shows the extent to which we have all come to be deluded into believing that computers are more intelligent than human beings. The comfort comes from knowing that the passengers probably never suffered the panic and symptoms associated with such a deadly event. Tragic as it was, it was their time to die and it was quick. This gave me great comfort when we were involved in a car accident in which my husband was killed despite his skill as a driver. The time elapsed was possibly 1 minute or two and his suffering was far less than those of us left behind injured and still suffering the results of those injuries on the quality of our lives. Taken against the suffering endured by say the Lindley victims, their deaths were mercifully quick. There is comfort in knowing that your loved ones did not suffer long - they were vital and alive and looking forward and then they ceased to suffer from the vicissitudes of life.

  • RobR - 2011-07-31 12:38

    News24 unfortunately epitomises the modern "low information, sound-bite" trend where reporting focusses on sensationalism to capture the reader's attention, rather than unbiased fact which allows the reader to critically examine the information and draw their own conclusion. The heading alone is an indication of this, leaving no doubt about the implication that is was "chaos in the cockpit" which caused the problem. However, there are a number of facts which have been left out, and I doubt as well that the writer has any aviation experience. I would also like to add that if you have never flown an aeroplane, and have never experienced a difficult situation while Pilot in Command, you will have NO appreciation of how stressful these situations are. I am in no way attempting to absolve blame anywhere. However, commercial aviation is a marriage between those wanting to make money out of it AND those wanting to benefit from its capabilities i.e. travellers. Pilots do the best they can, given the constraints of modern aviation. The biggest fact not evident in any of the reports I have read, is that the Airbus operates in different modes, depending on the reliability of it's instruments. It typically operates in Normal Law where computers essentially prevent the aircraft from operating outside of it's limits. When this information is unreliable, the aircraft reverts to Alternate Law where it operates in a completely different manner to Normal Law [see next comment]...

      RobR - 2011-07-31 12:55

      By it's very nature, Alternate Law is a very difficult area in which to train pilots & most training is performed in simulators. In this mode, the flight controllers are limited i.e. envelope protection is lost. The aircraft feels completely different, there is no low speed protection and the aircraft CAN be stalled. The other thing to consider is the fact that a severely stalled airliner feels like a nightmare in the cockpit where you are subjected to a high level of buffeting. Add in the effect of the severe turbulence and you will get an appreciation of how difficult the operating environment was, for the pilots. No normal passenger/reader will have ANY idea just how stressful & difficult this environment is. There is one comment I would like to oppose [Dimitri's: the pilots did the correct thing i.e. "Nose down to decrease the AoA"]. They did this once, and in fact it is one of the glaring omissions by the pilots i.e. they did not adhere to one of the fundamental tenants of flying - put the nose down when you feel the aircraft is no longer flying properly. However this has to be balanced by the dangers of flying too fast. The airspeed readouts were unreliable but the pilot knew he had the thrust levers at greater than 100%! Given that the aircraft was in Alternate Law, given the difficulties of the cockpit [stall buffeting, weather], given the unreliable instruments & the short duration over which everything occurred, you HAVE to give this headline less credence...

      Dmitri - 2011-07-31 12:59

      Totally agreed RobR. I am hoping they release the full CVR (which I doubt they will). Shades of SA295 in terms of liability coverup???

      Jack Johnson - 2011-07-31 17:19

      You honestly think they are trying to cover up (supposedly)illegal international arms dealing? They could decide not to release the CVR to save further embarrassment to the pilot's families, for all we know. That is even if they should decide not to release it, which hasn't been decided yet!

  • graeme.talbot - 2011-07-31 13:25

    AeroPeru Flight 603 had a similar case, only it was tape was left over the sensor which they are saying failed on the Airbus. The below video can give you an idea of what the pilots might have been faced with except the AirFrance pilots will not have had any radar to assist with air speed readings as in the AeroPeru case which ultimately crashed.

  • Marc - 2011-07-31 13:53

    I was watching a documentary about the airbus, and its amazing how many post-flight complaints there were about the stalling warning of this particular airbus. It's just the pilots managed to handle the situation to protocol. In a flight, you just do not know what goes on in that cockpit. Scary thought

  • Marc - 2011-07-31 13:58

    Im not an expert in this field, but in one of the preliminary reports, the pilots spoke of an odour within the cockpit- which was apparently ozone. Is this normal? My thinking is that they were so high up at one stage that this 'ozone' crept in the plane. I dunno, Im just wondering.

      Serias - 2011-07-31 14:23

      what's scary to me is that on the voice transcripts,it doesn't seem as if they knew whether they were ascending or decending. Like I said previously,is there anyone that can explain what actually happened in "layman's terms". This plane practically fell from the sky without the pilots being able to stop it!

      Zakhele - 2011-07-31 17:07

      Serias- Follow my reading below. In layman's terms what happened is this: If you've in the airplane you would have noticed that whenever an airplane goes through clouds it loses speed and the altitude. The cause of these is that the wings don't get enough lift as a result of slow moving air (practically not much air movement in thick cloud). So in this particular aircraft there was a loss of enough airlift on the wings caused by an airzone which was moving slow as a result of crystals. If this goes on for a prolonged time you supposed to manually pick up the speed of the aircraft to help the wings with lift. But if you don't know what speed you flying and what altitude you are flying the only source of you knowing is by seeing or feeling these yourself. If you are late to react you are doomed as you might be in the unrecoverable deep stall as mentioned below-also considering that flight F-GZCP was T-tailed aircraft. Essentially that's what really happened. We could find many ways if expressing it. The bottom line at what point are pilots told to take over the manual controls of the airplane - considering that by this time you getting conflicting information? One thing we now know is that these planes were not well equipped to survive prolonged crystal air conditions. Let me put it this way Serias: The airline industry always learn of something new after people have died and I am not saying we should blame them-but they need learn to accept their designs shortcomings. Ends.

      Jack Johnson - 2011-07-31 19:33

      Marc- Usually the ozone smell is associated with electrical problems. Zakhele - I disagree, these aircraft ARE well equipped to handle prolonged conditions of ice-formation. Consider that these aircraft are designed to operate at temperatures of less then -40C for most of their operational life. The instruments that measure the air-speed, for example, are equipped with little "heaters" to prevent the build-up of ice and little holes to let the water "drip out" so that it doesn't form a little dam. I agree with you that the conflicting information the pilots received had a big hand in the incident. One effect that has been well documented that occurs in situations like this (I'm not necessarily saying it was the case here) is that pilots focus on one specific issue and ignore everything else, sometimes missing the critical issue. This is what is sometimes referred to as "pilot overload". You are also correct that the industry must and does learn from such unfortunate incidents. PS. The A330 does not have a T-Type tail, but the conventional arrangement, for what it's worth.

  • Zakhele - 2011-07-31 14:48

    "the pilots had an awful lot of information denied to them to help them deal with the situation" because of malfunctioning computers. BEA chief Jean-Paul Troadic said that, at the start at least, "the situation was salvageable". A basic manoeuvre for stall recovery, which pilots are taught at the outset of their flight training, is to push the yoke forward and apply full throttle to lower the nose of the plane and build up speed. But he nosed up throughout much of the impending disaster and the plane reached a maximum height of 38 000 feet. external speed sensors obstructed by ice crystals produced irregular speed readings on the plane. "We understood how the accident came about," Bouillard said. "Now we must learn why it came about." “Why it came about”: Combination of the ice crystals feeding wrong information to the computer and the pilot not taking the manual controls of the plane much, much earlier are the only two reasons why the plane went down. Question to Airbus: At what point are the pilot told to take over the manual controls of the airplane? Write long stories as long as we do not answer this question precisely we will find ourselves re-writing how far we can push these machines-which I don’t think that’s what the relatives of the victims want to hear. Airbus is trying to shift the blame to the pilot. “push the yoke forward and apply full throttle to lower the nose of the plane and build up speed” Did anything tell him that the nose was up?

      Zakhele - 2011-07-31 16:25

      Stalls do not derive from airspeed and can occur at any speed -but only if the wings have too high an angle of attack. Attempting to increase the angle of attack at 1g by moving the control column back normally causes the aircraft to rise. However aircraft often experience higher g, for example when turning steeply or pulling out of a dive. In these cases, the wings are already operating at a higher angle of attack to create the necessary force (derived from lift) to accelerate in the desired direction. Increasing the g loading still further, by pulling back on the controls, can cause the stalling angle to be exceeded -even though the aircraft is flying at a high speed. These "high speed stalls" produce the same buffeting characteristics as 1g stalls and can also initiate a spin if there is also any yawing.

      Jack Johnson - 2011-07-31 17:11

      I have to disagree with your statement that "Airbus is trying to shift the blame to the pilot". The Airbus A330 actually has a brilliant safety record (see ). Two hull-loss accidents out of 796 produced so far. That's 0.25%. Compare: * Boeing 737 (all variants): 148 lost out of 6819 (2.17%) * McDonnell Douglas DC-9 : 101 lost out of 976 (10.35%) * Boeing 747 (all variants): 49 lost out of 1418 (3.46%) The Boeing 777 has an even better safety record, by comparison, of 1 hull loss out of 923 produced. The fact of the matter is that an avoidable combination of events led to the accident. The crew and the aircraft has to act in union to ensure the safe operation of a flight. Sometimes the crew "saves" the aircraft, sometimes, the other way around. In this case, and you said it yourself, the pilot's choice to pull up rather than nose down turned a bad situation into a critical situation. What is important is for the aviation community to learn from this incident, adjust training and statutory requirements, update safety procedures and go forward to have even more safe flights. PS. I have heard so many pilots complain about being denied full authority on certain aircraft. This case proves that denying a pilot full authority in certain situations could save the aircraft.

      Jack Johnson - 2011-07-31 17:16

      Just for further comparison of long-range aircraft safety: * A340 : 5 out of 375 (1.33%) * B767 : 11 out of 1000 (1.1%)

      Zakhele - 2011-07-31 17:53

      @Jack Johnson- Let's say and assume that Airbus is not trying to shift the blame then. But we thing we all agree too is that the pilot acted too late and even after acting seems to have aggravated the problem in front of an experienced captain. The only thing I thing the captain didn't intervene could be that they were receiving conflicting information and the captain could not make up his mind for over 2 minutes. That tells me the machine was in a doomed situation itself. Or perhaps you think the captain might be a terrorist himself?

      Jack Johnson - 2011-07-31 20:13

      @ Zakhele: What you are describing is the classic definition of pilot error. The aircraft was flying at the onset of the event. Even in a high-altitude stall, controlled flight could most possibly have been recovered. The situation that you describe tells me that the crew made the wrong decisions all the way to impact.

  • Zakhele - 2011-07-31 16:27

    Symptoms of an approaching stall: One symptom of an approaching stall is slow and sloppy controls. As the speed of the aeroplane decreases approaching the stall, there is less air moving over the wing and therefore less air will be deflected by the control surfaces (ailerons, elevator and rudder) at this slower speed. Some buffeting may also be felt from the turbulent flow above the wings as the stall is reached. The stall warning will sound, if fitted, in most aircraft 5 to 10 knots above the stall speed.

      Jack Johnson - 2011-07-31 20:31

      Stalling of aircraft are quite well understood in the industry. One of the most important skills a pilot has to show before allowing to fly solo is recovering a stall. Even high-altitude stalls are well understood. I don't know if you ever heard about something called "Coffin Corner" ( ). This is when the aircraft is flying very high in conditions where it is close to stalling. If you look at the document, you will see that U2 pilots fly their aircraft close to this point almost during the whole flight. Pilots CAN be taught to fly in these conditions. One quote from the wikipedia page I refer to: "Some aircraft, such as the Lockheed U-2, routinely operate in the "coffin corner", which demands great skill from their pilots. The FAA is concerned that as jet aircraft become more common, less experienced pilots will be flying those aircraft closer to the altitude of their coffin corners, and that catastrophic accidents will occur as a result". This was identified in a 2003 FAA report. I believe this is a training issue, not an aircraft issue.

  • Zakhele - 2011-07-31 16:31

    Stalling characteristicsL Different aircraft types have different stalling characteristics. A benign stall is one where the nose drops gently and the wings remain level throughout. Slightly more demanding is a stall where one wing stalls slightly before the other, causing that wing to drop sharply, with the possibility of entering a spin. A dangerous stall is one where the nose rises, pushing the wing deeper into the stalled state and potentially leading to an unrecoverable deep stall. This can occur in some T-tailed aircraft where the turbulent airflow from the stalled wing can blanket the control surfaces at the tail. And the pilot still did not ''push the yoke forward and apply full throttle to lower the nose of the plane and build up speed" Must have received conflicting information!

      Zakhele - 2011-07-31 16:34

      - Courtesy of Wikipedia search.

      Caroline - 2011-07-31 16:58

      Zakhele, I'd have to say this sounds like pilot error (I come from an old-fashioned military flying family - my dad was an RAF pilot and pilot examiner in the days when you had to fly the thing yourself). I don't know that much about Airbus avionics but it sounds as though the aircraft computer didn't malfunction. The Pitot is just a pressure sensitive tube which is placed on the hull of the aircraft - if you have one placed where it isn't exposed to the airstream and another one placed where it is, then the pressure differential between the tubes can be converted into airspeed; it's one of the basic instruments on an aircraft. There are obviously more than one pair of Pitot's on something as sophisticated as an Airbus and if they started giving different readings (because of ice) then the computer would detect this as an error, shut down the primary autopilot and ask the pilot to start flying manually - but he couldn't have taken over any earlier. The big mistake he made seems to be ignoring the stall warning and then climbing, and going by the avherald link that someone posted, at one point he cut the throttles back but carried on climbing. He would know he was climbing even in a very basic aircraft from an altimeter and a couple of instruments (like turn-and-bank) which display the aircraft's attitude, and he would also have had the stick back. (see next comment)

      Caroline - 2011-07-31 16:59

      Did it stall when the original stall warnings went off? From the avherald report, maybe not - they were bouncing around and starting to roll, and recovered from that but kept nose up, and this sounds like where it went pear-shaped: "However, none of the pilots called out speeds, vertical speeds, pitch angles or altitude." Stall speed is the threshold below which the wing loses its aerodynamic properties and the aircraft falls out of the sky. It varies with wing angle of attack and other factors, and as you climb, the threshold speed increases, because there is less air pressure (so less lift), so they didn't have much margin of error - interesting technical article here: But the bottom line seems to be that when the aircraft began to stall, they didn't know how to recover, or hadn't practised enough to apply the right procedure automatically. That is not to blame the pilots - as someone else pointed out, it must have been terrifying. But an old hand like my old man would have said that what gets you through those situations alive is practice, practice and more practice, so that your instincts take over, and I think he would have been right.

      Caroline - 2011-07-31 16:59

      The skybrary article is interesting - it says that most stall recovery practice these days revolves around recovering from an *incipient* stall (nose down, speed up, klaxons blaring), not a full stall, which is a different animal. Being the offspring of someone who used to take a Meteor jet bomber (something of a widow-maker) up to 35 000 ft and stall-spin-recover, stall-spin-recover, over and over again, as part of normal on-the-job training, this says to me that the airline (and the regulatory body) is at fault by not ensuring that the pilots regularly receive continuation training in both incipient and full stall recovery. But that's expensive, and as someone else pointed out, difficult in an Airbus, requiring a simulator. (And I daresay you can't spin an Airbus...) It appears the BEA agree: . Better late than never, I suppose.

      Zakhele - 2011-07-31 18:01

      @Caroline- You got the answer there Caroline. I am saying at what point is the pilot allowed to take over and deploy manual controls, especially when the screen doesn't show him the X Y Z he was told during flying lessons and computer simulations? I will get more technical info on the website you shared and will challenge Airbus all the way on this. Thanks.

      Caroline - 2011-07-31 18:17

      @Zakhele, I agree with Jack - that's not Airbus, it's training and procedure, which is down to the airline and the relevant aviation authority. From the little I know about Airbus, it is very much the exception for the pilot to take manual control, but it sounds as though the conflicting airspeed info was reported immediately and the aircraft's computers then required manual flying, which is the point at which the problem *started*. The pilots knew (because the computer told them) that the airspeed indicators were conflicting (and that was an iced-up pitot tube, which is a very basic gadget on the outside of the aircraft - nothing the Airbus computer could have done to fix it, it needed de-icing), and they made the wrong flying decisions. A frozen pitot won't crash an airliner, but the wrong response to a faulty airspeed indicator certainly will and did. It's tragic, it could easily happen again and it sounds as though the regulators are making the right decisions and saying they need to improve pilot training in the area of stall recovery.

  • raspatin - 2011-07-31 17:05

    Stall warning,crystals,computors and $#!t,I aint care get this thing right ,228 people have died !or we will have to go back to shipping like J Van Riebiek or Marco Pollo.Damn I'm sicko this !

  • raspatin - 2011-07-31 17:08

    ...and Zakhele you must be an Air-Force-One pilot seem to have invested in this .

      Zakhele - 2011-07-31 17:42

      Picture this scenario: You set your car to cruise control at 120 km/hr whilst on the straight stretch. All your car does is draw petrol accordingly depending on the force required by the flywheel to maintain the same spin. Now the road changes to go downhill. Once again all you car will do is draw less petrol to keep a speed of 120 km/hr. Now if you leave it there (on cruise control of 120 km/hr) the car will continue driving at 120km/hr until you reach a point where it is practically impossible to drive at 120km/hr due to the road nature. If all of a sudden you now decide to say. oh sh*t I am going to be in trouble I need take over the auto controls and do manual controls it might be that you are late-maybe way too late and we will pick you and your car on the rails. The difference between the car and the airplane is that in your car you make all these decisions yourself but on the airplane. The poor pilot is told don't touch and rush until you see X Y Z. But what if X Y Z doesn't pop up the screen as a result of bad weather with crystals and the computers going out of range? Easy for all of us down here but not easy for the poor pilot at 11.58 km altitude, at night, with snow and computers feeding the wrong information, captain thinking about his girlfriend instead of helping you.......Not easy!

      Serias - 2011-07-31 18:12

      @Zakhele....thanks for previous explanation. Cant we argue then that there was not sufficient back up systems to prevent this. It seems,the more complex and clever they design these systems, the more difficult it gets to actually control manually. Some years ago, there was still a flight engineer also in the cockpit that used to watch the plane systems like a hawk. My personal opinion - you can't replace the human eye, brain and thought process and the airlines, and pilots for that matter, have become so reliant on automated actions and systems, that their "actual" total experience of handling difficulties are just not the same anymore. It seems that all a pilot does these days is take off and land. BUT, flying is still statistically the safest transportation option and its usually a tragedy because so many deaths are involved. I still though can just not get comfortable on a flight!!!!

      Jack Johnson - 2011-07-31 20:04

      @ Serias: It is not a fact of putting more systems into place. It seems that sometimes, the pilots flying these large airliners forget their basic flying skills. @Zakhele - these aircraft are designed to allow the pilot to take control very easily. Moving any of the controls more than a certain bit (in laymen's terms) disconnects the autopilot and hands over control to the pilot. In this case, control was handed over to the pilots when the pitot tubes gave conflicting information. This was quite early in the event. It appears that from here, a sequence of bad decisions caused the demise of AF447. I agree, it's not easy. But that's the profession they choose. Pilots in general are very eager to take the controls of the aircraft. But with that comes the responsibility of the safety of all the souls on board.

      Zakhele - 2011-07-31 20:35

      @Jack Johnson- I guess the French Investigators report is not about to be accepted by the leading experts without a fight. On you statement above aircraft that ''The A330 does not have a T-Type tail'' the pictures showed a F-GZCP with a T-Type tail. I might be interpreting it wrong though. I don't think the NTSB will accept the findings of the French report. There are just far too many contradicting statements all aiming at seeking an easy way out of blaming a man who can no longer defend himself. We need exhaust all avenues before we blame a man who faced a situation nobody wants to be in. Maybe I am thinking with my emotions as a father now. All in all I just hate Airbus pointing a finger at the man before they can conclude on 'why did he fail to bring the plane back to stable condition''. Airbus must first bring all the hard facts of why the man failed to come out of a stall position considering there seems to be an indication that the pilot actually aggravated the problem in ignoring the very basic rule of, ''pushing the yoke forward and apply full throttle to lower the nose of the plane and build up speed". This is probably the third time I hear that all a pilot needed do was to push or pull the yoke. I guess it becomes easy if someone is simulating in an experiment scenario. But to say the pilot had 3 minutes to get the plane back to stability leaves a lot of questions than answers on the Airbus side. 3 minutes is very long time in these situations. Night!

      Jack Johnson - 2011-07-31 22:16

      @ Zakhele: First of all, this is what a T-Type tail looks like: You will note that the horizontal stabiliser is on top of the tail / rudder. The Airbus A330 has the horizontal stabiliser lower on the tail. The NTSB has been consulted during the investigations, as has experts of various different aviation authorities. I doubt that they will contest the findings. Airbus is not pointing any fingers at anybody who can't defend themselves. It is called an investigation. The investigation is so far pointing to pilot error. What I cannot understand is that you identify the correct symptom (i.e. the pilot pulled up, instead of push down). But for some reason you lay the blame at the side of the manufacturer. If a pilot in ANY AND EVERY commercial aircraft flying pulled up in a persistant stall, the outcome would be the same. It is not because of the way Airbus built the plane. Could you explain how you come to the conclusion that Airbus caused the pilot to provide the wrong control input?

  • pawsaw - 2011-07-31 21:02

    Some really well presented and obviously knowledgeable people commenting on this issue here is both reassuring and scary. I still believe that the real problem here is that technology like mankind who invents it is fallible and the more sophisticated the skilled people flying these planes perceive the technology to be the more time they imagine they have to rectify a potentially tragic event such as this. These events are tragedies because they occur so seldom and so there is of necessity less experience for the pilot of a doomed craft to draw from and very little time. I do not believe any pilot would enjoy a situation such as this one and hindsight is always 20/20. The guy in America who landed on a river with virtually no injuries even to his passengers may have responded differently and saved the 220 souls but he may have done exactly what happened here. It was a combination of factors including weather conditions, mechanical and human failure from which we can only learn how future recurrences can be averted thanks to technology and human sacrifice and endeavour. It is the why which is critical so that hopefully there are answers for future potential problems.

      Caroline - 2011-08-01 10:23

      Pawsaw, yes, it sounds like the "why" is inadequate continuation training leading to pilot error. Think about it - under most conditions these days, the aircraft can fly you from start to finish without the pilot touching the controls (ever sat through a Cat III auto-land? I did the other day coming into CPT in fog, on a Mango flight). So the pilot's ability really, really matters when conditions are difficult or things go wrong and the aircraft systems require manual intervention as happened here. If the pilot's training (and practice, re-evaluation etc) focuses mainly on normal flying conditions, then quite honestly, why have them at all? But otoh if they don't spend enough time training and practising for emergencies, then this is the result. But that training takes time during which the pilot is unavailable for commercial flying, requires simulators, perhaps modified aircraft etc, so is very expensive. Before computerised avionics, pilots had to really fly, and were probably much better equipped to handle this kind of thing, but they didn't have the advantage of advanced avionics, computerised air traffic control, etc. So with today's traffic volumes, the accident rate would skyrocket if everyone flew manually all the time, because human error is a fact of life. No easy solutions, but hopefully the training procedures will be improved as a result of this.

  • Gert - 2011-07-31 21:14

    Al very technical but they still have GPS information on speed which is more reliable. don't blame anyone learn about this, basic flying is the most important fact, in any flying training you are suppossed to learn what to do when you lose your instrumentation

  • pages:
  • 1