AF447 - French prosecutors sends AF to court for negligence

[ATC Watcher]

Today the French State Prosecutor office announced that it will send Air France to court for "negligence" on the AF447 case but absolves Airbus due to lack of evidence .
The "negligence" claim against AF seem to concentrate on not having "informed" ( trained?) its pilots to the emergency procedures during this kind of event .
The full text of the indictment is unfortunately not yet available.

[sonicbum]

Today the French State Prosecutor office announced that it will send Air France to court for "negligence" on the AF447 case but absolves Airbus due to lack of evidence .
The "negligence" claim against AF seem to concentrate on not having "informed" ( trained?) its pilots to the emergency procedures during this kind of event .
The full text of the indictment is unfortunately not yet available.

It will be very interesting to read the full text when available. I guess it comes down to how much exposure the crew received to high altitude handling in alternate law, stall recovery and how much theoretical training was received regarding characteristics and particularities of flights crossing the ITCZ. Let's see.

[KingAir1978]

I sincerely hope that this gets thrown out.

Imagine the precedent if this would result in a conviction? It is simply impossible to train crew for every conceivable scenario... Would be very interesting indeed to see the full text and the eventual outcome.

[MCDU2]

I think it could have positive outcomes for crews if the airline and by default its training department are held to account. Be a welcome wake up call to the airline finance managers that safety costs.

[SeenItAll]

Note that there are many reasons why the court might consider AF culpable for the accident. I believe it was AF that was resisting changing out the Thales pitot tubes that were prone to icing. Perhaps AF training never informed the crew that below 80 knots IAS stall warnings were inhibited. While I am not implying that AF should be criminally liable for these decisions (because perfection is impossible), that is for the court to determine.

[Speed of Sound]

I sincerely hope that this gets thrown out.

It is simply impossible to train crew for every conceivable scenario... .

Nobody is suggesting that it is.

Leaving aside the dodgy pitot for a moment, the case will be that AF failed to ensure that their staff were trained to a certain standard expected of a competent pilot. The court will then look at what that standard is and will decide whether AF attained that standard, not necessarily in the AF447 crew’s case but across its crew training as a whole.

If AF is found guilty, crew training departments across the world will have to look at their own standards and decide if they are up to the mark.

That is a good thing particularly in the current climate where pilot competence and training is being questioned.

[golfbananajam]

the case will be that AF failed to ensure that their staff were trained to a certain standard expected of a competent pilot. The court will then look at what that standard is and will decide whether AF attained that standard, not necessarily in the AF447 crew’s case but across its crew training as a whole.


surely, if this is proven in the court, then the pilots shouldn't be flying/licensed?

[Speed of Sound]

surely, if this is proven in the court, then the pilots shouldn't be flying/licensed?

Not necessarily.

You can receive poor or inadequate training and still be a competent pilot just as you can receive excellent training and still be a poor pilot. The issue here is the overall quality of training, not licensing.

[UltraFan]

It is simply impossible to train crew for every conceivable scenario...

But it wasn't "every conceivable scenario". Nothing they encountered required more then basic flying skills. Pitot tubes freezing and resulting unreliable speed indication requires only one action from the pilot - DO NOTHING. You don't need to change thrust settings, or change the aircraft attitude, or press any buttons, or check any parameters. Just sit there and wait for the heating system to restore the status quo.

I hope this case gets accepted and prosecuted. I would very much like to know the answers to the following:

1. Why did two fully licensed pilots flying for one of the leading airlines, who had over 10,000 FH between them, panic so much that they forgot all procedures, failed to recognize which systems were and which weren't working, failed to even communicate with each other except for "What's going on?" and "I don't know"?
2. Why did a fully licensed Air France pilot with almost 3,000 FH under his belt panic so much that he forgot the most basic maneuvres? Stall = nose down. Yet the guy kept pulling on the sidestick for almost four minutes.
3. And the question that has been bothering me most ever since I read the final BEA report. Why the hell would an experienced 58-year-old captain with over 10,000 FH hours logged refuse the FO's suggestion to fly around the storm, point the aircraft right into icing and turbulence, and then simply leave the cockpit and let his less experienced colleagues deal with the fallout of his decision. I just cannot imagine that. You decide to fly into danger - you see it through and make sure you come out on the other side.

I'm really angry that Airbus has been left out of all this. One of the deciding factors in that accident was inability of the crew to recognize that one of them was doing something horribly wrong. They only realized the FO was pulling on the sidestick when he actually said so out loud. BEA has been very careful to write as little as possible about this, throwing it into footnotes of "contributing factors".

I love Airbus, I love France and Air France, but sadly I see that they are trying as much as they can to downplay this whole accident as human error. I don't want to one day see Airbus in the same position as Boeing is right now. And the worst part - I don't really see any changes being made.

[4runner]

How does a system of aviation oversight, allow pilots to go from private to ATP while never fully stalling an airplane? This is a systematic failure of EASA curriculum and basic muscle memory come home to roost. That, combined with over reliance on technology in an overly complex aircraft.

[Old Dogs]

Today the French State Prosecutor office announced that it will send Air France to court for "negligence" on the AF447 case but absolves Airbus due to lack of evidence .
The "negligence" claim against AF seem to concentrate on not having "informed" ( trained?) its pilots to the emergency procedures during this kind of event .
The full text of the indictment is unfortunately not yet available.

I was a check/training pilot for many years.

Not sure how you train a guy not to hold the stick in the full aft position from 36,000 to sea level. 😏

[Old Dogs]

I sincerely hope that this gets thrown out.

It is simply impossible to train crew for every conceivable scenario.

1) Level the wings - more or less.
2) Put the pitch bar on the horizon - more or less.
3) Put the throttles in the middle - more or less.
4) Stop and think.

[Old Dogs]

Note that there are many reasons why the court might consider AF culpable for the accident. I believe it was AF that was resisting changing out the Thales pitot tubes that were prone to icing. Perhaps AF training never informed the crew that below 80 knots IAS stall warnings were inhibited. While I am not implying that AF should be criminally liable for these decisions (because perfection is impossible), that is for the court to determine.

This has nothing to do with the Thales pitot system or the the fact that below 80 knots the stall warnings are inhibited - the thing won't fly at 80 knots anyway.

It has everything to do with never having learned how to fly an airplane in the first place.

NO competent pilot would hold the controls in the full aft position from 36,000 to sea level.

[eckhard]

1) Level the wings - more or less.
2) Put the pitch bar on the horizon - more or less.
3) Put the throttles in the middle - more or less.
4) Stop and think.

5) Try to figure out what the actual flight path trajectory is, using all available means. Either the aircraft is (a) going roughly where it’s pointing or else (b) it’s going downhill, probably quite quickly.

6) If (a): re-establish desired conditions; If (b): you are stalled and need to recover from that before doing anything else.

[Old Dogs]

5) Try to figure out what the actual flight path trajectory is, using all available means. Either the aircraft is (a) going roughly where it’s pointing or else (b) it’s going downhill, probably quite quickly.

6) If (a): re-establish desired conditions; If (b): you are stalled and need to recover from that before doing anything else.


5) If you get the first three steps right the aircraft trajectory will be more or less straight and level although probably not exactly on desired speed, altitude or heading.

6a) Following the first three steps will pretty much get you to back to the most basic desired condition - stable flight.

6b) I agree, although in 22,000 hours I cannot once recall stalling an aircraft unintentionally.

[Mr Optimistic]

(pax). Not suspecting or recognising a stall when your actions generated it and the system warnings went unheeded should raise some question over training surely?

[Astir 511]

Note that there are many reasons why the court might consider AF culpable for the accident. I believe it was AF that was resisting changing out the Thales pitot tubes that were prone to icing. Perhaps AF training never informed the crew that below 80 knots IAS stall warnings were inhibited. While I am not implying that AF should be criminally liable for these decisions (because perfection is impossible), that is for the court to determine.
Don't believe it's accurate to say they resisted the changeout from Thales to Goodrich units, but they were certainly late adopters of the mod. I don't believe the Mod was driven by an AD (Stand to be corrected if so) so is there culpability in the Risk Assessment of the Regulatory Authorities? As always this lump of Swiss Cheese has many holes, and seems unfair to single out one party. That said you make a valid point around sending a warning to all Airlines around minimising training for costs reasons

[Peter H]

Don't believe it's accurate to say they resisted the changeout from Thales to Goodrich units, but they were certainly late adopters of the mod. I don't believe the Mod was driven by an AD (Stand to be corrected if so) so is there culpability in the Risk Assessment of the Regulatory Authorities? As always this lump of Swiss Cheese has many holes, and seems unfair to single out one party. That said you make a valid point around sending a warning to all Airlines around minimising training for costs reasonsI quoted somebodies timeline on this in https://www.pprune.org/rumours-news/623116-a380-engine-piece-found-groenland-after-9-months-3.html#post10512149

[Uplinker]

No pilot should ever hold full back-stick*, but I for one think training needs a shake up.

Memory drills, in my experience are not taught or practised enough. On rare occasions; you might get two goes each at one of them in the SIM, and if they are flown near enough, the TRE moves on. It might take several years’ worth of biannual SIMs for one to have practised every memory drill !

All the memory drills should be sprung on every pilot on every SIM visit - otherwise how on earth do you practise and stay in practice? (sic). The best we can do is armchair flying or touch drills in the aircraft while in the cruise - it’s not the same thing. You wouldn’t expect a violin player to imagine playing a piece without their instrument and then be expected to play it. Or for someone to read about how to ride a bike then get on and ride it. One needs to physically practise, practise, practise.

We must remember that nowadays, many pilots who fly the big modern jets have not necessarily worked their way up for years on various basic piston and turboprop aircraft in challenging weather, so do not necessarily have the instinctive feel and response to flying that the previous pilot generation did - those who had very basic or no automatics and frequently hand flew their aircraft. Big jets do not have the feel of a small SEP trainer, so recognition of unusual situations and responses need to be taught and practised all over again. Most of us stay out of trouble by keeping within SOPs and limitations but occasionally tragedies occur.

*except for Airbus Abnormal Valpha prot OEB

[Lord Bracken]

As ever, it is worth rewatching the official BEA animation of what the pilots were presented with on their PFDs.

https://youtu.be/n-hbWO0gL6g

It should have been abundantly clear to everyone the aircraft was stalled at a high AOA.

[RobertP]

I too read the BEA report when it was finally issued and have struggled with the same questions as yourself. It has taken far too long to get even to this stage. I hope the prosecution goes ahead.

[Astir 511]

I quoted somebodies timeline on this in https://www.pprune.org/rumours-news/623116-a380-engine-piece-found-groenland-after-9-months-3.html#post10512149
Thanks for the reference, I think the point I was trying to make (albeit not very successfully, my bad!) is that in terms of Timeline to adopt the Mod, if the Regulators had deemed the probe reliability as a flight safety risk they would have issued an AD, and the Timeline for terminating action on the AD would have been driven by the Regulator's Risk Assessment. There wasn't an AD, therefore it's difficult to criticise AFI for a "perceived" slow adoption. At my Airline we embodied the Mod aggressively, as our own Safety Assessment deemed it prudent. But it was a judgement call, as was allowed by the absence of an AD.

[RobertP]

As ever, it is worth rewatching the official BEA animation of what the pilots were presented with on their PFDs.


It should have been abundantly clear to everyone the aircraft was stalled at a high AOA.

A permanent record of terrifying incompetence! It still scares me years later.

[Timmy Tomkins]

I thought there had been a notice ussued by Airbus 3 years before this and AF had elected not to replace the Pitots? Had they done so - competence aside; and it is a shocking video - there may not have been an accident to discuss.

[aixois]

ood evening,

There was also (sorry if it had been already uploaded some months or years ago here) an Airbus notice on "Unreliable Speed" published 2 years before the loss of AF 447 flight.

www.airbus.com/content/dam/corporate-topics/publications/safety-first/Airbus_Safety_first_magazine_05.pdf

See page 14 till 19 of the document.

Sincerely

[ATC Watcher]

The text of the prosecution has been published ; [AF provided] insufficient information to its crews about incidents (of the PITOT probes) that occurred during the previous months, their consequences and the procedure to be applied, in a context of insufficient pilot training at high altitude, of a lack of adaptation of their training, and operational failure ",

The SNPL issued a statement earlier today attacking Airbus . Unfortunately only in French here : https://snpl.com/cp-requisitoire-incomprehensible-airbus/

[blind pew]

That someone was prosecuted for incompetence rather than the cover ups and protecting the establishment that has existed in aviation since the Munich Disaster. Hope it succeeds as it can only enhance safety and politics but....

[tdracer]

Not sure how you train a guy not to hold the stick in the full aft position from 36,000 to sea level. 😏

I don't want to turn this into another Airbus/Boeing bash fest, but I suspect there is some cause/effect due to the pronouncements from Airbus (and others) that 'it's impossible to stall this aircraft'.
In an emergency, the human mind may not remember whatever qualifications may have been put on that statement - e.g. 'in the full up control mode'. He just remembers 'impossible to stall'.
With that in mind, if his brain was telling him it's 'impossible to stall', holding the stick full back while the aircraft drops like a rock (even with the "stall stall" warnings) is somewhat easier to understand.

Tell the pilot that there are anti-stall protections to 'help', but don't tell them they can't stall it because in the right circumstances, they can stall it.

[ph-sbe]

A permanent record of terrifying incompetence! It still scares me years later.

Exactly this. This is bone-chilling to watch. Such a gross incompetence.

That said, I believe it is unfair to prosecute the airline. While the airline has some responsibility to ensure a certain level of training, it is not responsible for an airmen's basic competence. That is the responsibility of the certificating authority, in this case (I assume) the French government themselves. They issued the airman's certificate and appropriate ratings. They are responsible for the airman's checkrides and recurring bi-annuals.

A taxi company is not responsible for a driver's ability to drive; once the driver presents a valid, government issued driver's license in the appropriate class, the company should be able to safely assume that the driver has had the proper training and driver's test. The same is true here.

[UltraFan]

1) Level the wings - more or less.
2) Put the pitch bar on the horizon - more or less.
3) Put the throttles in the middle - more or less.
4) Stop and think.

May I suggest (regardless of anyone present) that this course of thinking was exactly what brought that plane down. I have to do something! Thought the pilot. WHY!? Wasn't on his mind. The wings were level and the plane was flying straight. The attitude was exactly fine. The engines were spinning and served them perfectly well. They reduced speed to react to turbulence. The only thing worth stopping and thinking about in that particular situation was what to order from the galley.

Honestly, I can't think of another major accident, let alone a crash, that resulted from a situation where, to save the plane, pilots had to do exactly nothing. They were busy doing nothing before the incident, they had to remain doing nothing during the incident, and after the incident, they should've continued diligently doing nothing.

[UltraFan]

As ever, it is worth rewatching the official BEA animation of what the pilots were presented with on their PFDs.

It should have been abundantly clear to everyone the aircraft was stalled at a high AOA.

Thank you for this. I couldn't find it. I'm not looking at the PFDs, I'm looking at the timer in the upper center of the screen.

02:10:05 Autopilot disconnects. Speed indication unreliable.
02:10:35 Twenty seconds later Pitot 1 deiced.
02:10:46 Thirty seconds later Pitot 2 deiced.
02:11:07 One minute later Pitot 3 deiced, and the plane was back to normal.

However,
02:10:06 ONE second after AP/AT disconnect, and look at the right sidestick and throttles.

Unbelievable.

[phylosocopter]

May I suggest (regardless of anyone present) that this course of thinking was exactly what brought that plane down. I have to do something! Thought the pilot. WHY!? Wasn't on his mind. The wings were level and the plane was flying straight. The attitude was exactly fine. The engines were spinning and served them perfectly well. They reduced speed to react to turbulence. The only thing worth stopping and thinking about in that particular situation was what to order from the galley.

Honestly, I can't think of another major accident, let alone a crash, that resulted from a situation where, to save the plane, pilots had to do exactly nothing. They were busy doing nothing before the incident, they had to remain doing nothing during the incident, and after the incident, they should've continued diligently doing nothing.

ahh as soon as to AP disconnects, the craft will start to roll! some intervention is required to keep wings level after AP disconnect.

my question is , were they at that point (immediately after AP disconnect) seeing a raw gyro attitude display or were they seeing some flight director telling them to go up?

My view is that BOTH the manufacturer and the regulator should be facing questions here because the state of alarms and displays at time of unreliable airspeed is not defined in regulation (as far as I can see)

edit.... have now found this discussion of the accident report that answers some of my question at
https://www.ainonline.com/aviation-news/2012-07-08/final-af447-report-suggests-pilot-slavishly-followed-flight-director-pitch-commands"A major new finding in the final report concerned the flight director, which normally displays symbology on the pilots’ primary flying displays that give guidance on control inputs to reach a desired steady-state flightpath. After the autopilot and autothrottle disengaged, as the flight control law switched from normal to alternate, the flight director’s crossbars disappeared. But they then reappeared several times. Every time they were visible, they prompted pitch-up inputs by the PF, investigators determined. It took them a long time to “rebuild” what the flight director displayed since this is not part of the data recorded by the flight data recorder.The BEA acknowledged that the PF might have followed flight director indications. This was not the right thing to do in a stall but it seems that the crew never realized that the aircraft was in a stall. Moreover, the successive disappearance and reappearance of the crossbars reinforced this false impression, the investigators suggested. For the crew, this could have suggested their information was valid.

None of the pilots recognized that the flight director was changing from one mode to another because they were just too busy. The PF may have trusted the flight director so much that he was verbally agreeing to the other pilot’s pitch-down instructions, while still actually pitching up.The BEA’s report includes significant recommendations about the flight director. One of them calls for European Aviation Safety Agency to review its “display logic.” The flight director should disappear or present “appropriate orders” in a stall."
from

[Old Dogs]

May I suggest (regardless of anyone present) that this course of thinking was exactly what brought that plane down. ... The wings were level and the plane was flying straight. The attitude was exactly fine. The engines were spinning and served them perfectly well.


Then they have already accomplished Steps 1 through 3, right?

[Uplinker]

I had not seen this video before, but it is immediately obvious that the RHS pilot has not been taught how to correctly use the Airbus FBW side-stick and it is clear that he also did not know or did not remember the correct memory drills for unreliable speed or a stall.

I don’t know what he is looking at but he does not appear to notice the pitch or the IVSI. As soon as ‘Stall Stall’ is first heard, he does not dip the nose (pitch down), as you should in a stall not at lift-off, and a few ‘stall stall’s later he goes to TOGA. I wonder if he is doing Stall at lift-off drill (TOGA and pitch +15°)? He spends most of the time after that holding the side-stick aft of neutral and fussing about keeping wings level, (not a priority for nose high upset), as well as over controlling and almost never pushing forward.

This illustrates lack of sufficient handling training and lack of sufficient memory drill practise. Memory drills need to be instinctive - almost reflex actions. This can only be developed if pilots have sufficient physical practise. You don’t need a six axis FFS SIM to practise memory drills, but you do need some sort of instrument simulator with the same displays, controls and control feel as the real thing.

[infrequentflyer789]

A taxi company is not responsible for a driver's ability to drive; once the driver presents a valid, government issued driver's license in the appropriate class, the company should be able to safely assume that the driver has had the proper training and driver's test. The same is true here.

Not quite, IMO. A taxi company would not usually be responsible, however:
IF the taxi manufacturer notified the operator that they were getting a far higher than expected rate of taxi speedo failures for taxis with speedo model A in bad weather at motorway speeds, speedo failure affects the power steering which makes the vehicle a bit skittish and might cause a crash if the driver isn't up to it, and the manufacturer recommends replacing model A speedo with model B which is designed perform better...
THEN liability might fall on the taxi operator depending on how they responded to that notice

That appears to be what is happening in this case, and is also why this is very different to 737 MAX (despite what SNPL says) - this is not about general pilot competency or training it is about response to the specific risk identified prior to the event and notified by Airbus (in contrast, it appears that if Boeing did identify a risk it chose not to inform operators or pilots).

The taxi operator might, hypothetically, do some or all of the following:

(1) Tells taxi mfr to get stuffed on replacing speedos unless it builds a new adverse weather test rig and proves that model Bs are better than model As (note that model C from another parts mfr is already known to be better)
(2) Does a proper assessment of the increased risk and concludes there is no cause for concern and it's drivers will be able to handle the problem (not the same as assuming they will because they have the correct licence, since the licence training was designed for risk levels which have just been advised to be incorrect)
(3) Issues a warning to their drivers notifying the risk and specifying the correct handling procedures
(4) Checks their drivers' proficiency in handling this situation by testing a random sample in recurrent training/checks, and finds no problems
(5) Ensures all their drivers are tested on this scenario in recurrent training/checks

The question is what would be a negligent response, falling below the standard expected of a reasonable operator. My vaguely educated guess is that any combination involving (4) or (5) would be a solid defence, anything involving (2) or (3) would be less solid defence, actually doing nothing might be ok, but doing (1) and only (1) would have your lawyer shaking their head - because I think (1) may be seen as accepting liability for the notified risk, and "only (1)" means then doing nothing to mitigate it...

Based on many reports I have read, AF actually did (1), question then becomes what else did they do. If the answer is nothing, then I think they have a problem, and I have thought that for several years now - none of this is surprising me, the wheels of French justice turn very slowly, but that isn't surprising either.

[RickNRoll]

ahh as soon as to AP disconnects, the craft will start to roll! some intervention is required to keep wings level after AP disconnect.

my question is , were they at that point (immediately after AP disconnect) seeing a raw gyro attitude display or were they seeing some flight director telling them to go up?

My view is that BOTH the manufacturer and the regulator should be facing questions here because the state of alarms and displays at time of unreliable airspeed is not defined in regulation (as far as I can see)

edit.... have now found this discussion of the accident report that answers some of my question at
https://www.ainonline.com/aviation-news/2012-07-08/final-af447-report-suggests-pilot-slavishly-followed-flight-director-pitch-commands"A major new finding in the final report concerned the flight director, which normally displays symbology on the pilots’ primary flying displays that give guidance on control inputs to reach a desired steady-state flightpath. After the autopilot and autothrottle disengaged, as the flight control law switched from normal to alternate, the flight director’s crossbars disappeared. But they then reappeared several times. Every time they were visible, they prompted pitch-up inputs by the PF, investigators determined. It took them a long time to “rebuild” what the flight director displayed since this is not part of the data recorded by the flight data recorder.The BEA acknowledged that the PF might have followed flight director indications. This was not the right thing to do in a stall but it seems that the crew never realized that the aircraft was in a stall. Moreover, the successive disappearance and reappearance of the crossbars reinforced this false impression, the investigators suggested. For the crew, this could have suggested their information was valid.

None of the pilots recognized that the flight director was changing from one mode to another because they were just too busy. The PF may have trusted the flight director so much that he was verbally agreeing to the other pilot’s pitch-down instructions, while still actually pitching up.The BEA’s report includes significant recommendations about the flight director. One of them calls for European Aviation Safety Agency to review its “display logic.” The flight director should disappear or present “appropriate orders” in a stall."
from

IIRC, one of the UAS procedures is to turn off the FD.

As an aside, the auto-pilot could probably be programmed these days to cope with a UAS event without disconnecting.

[sonicbum]

I don't want to turn this into another Airbus/Boeing bash fest, but I suspect there is some cause/effect due to the pronouncements from Airbus (and others) that 'it's impossible to stall this aircraft'.
In an emergency, the human mind may not remember whatever qualifications may have been put on that statement - e.g. 'in the full up control mode'. He just remembers 'impossible to stall'.
With that in mind, if his brain was telling him it's 'impossible to stall', holding the stick full back while the aircraft drops like a rock (even with the "stall stall" warnings) is somewhat easier to understand.

Tell the pilot that there are anti-stall protections to 'help', but don't tell them they can't stall it because in the right circumstances, they can stall it.

Airbus never stated that any FBW aircraft is impossible to stall. They have always emphasised the fact that flight control protections are there to prevent entering a stall, i.e. exceeding the stalling AOA. If those protections are not available due to a flight control system downgrade, then it is obvious that the stall AOA can be reached and exceeded and this is exactly what thousands of pilots learn during their type rating and it has always been the same for the past 30 years. If a pilot's muscle memory connects "FBW = no stall" and nothing else, then there are issues that need to be tackled somewhere within the SHELL, which is basically what has happened in the past 10 years since the accident, by increasing crew exposure to high altitude hand flying in alternate law, stall recovery and unreliable speed management. Airbus sells You a product, tells You how it works in detail (that is not always the case in the industry apparently) and gives you support to build training programs. They do not have a crystal ball and foresee how each and every airline pilot will react to any system anomaly, as this is part of the operator's duties to ensure proper guidance is given and maintained trough a proper system of recurrent training and checking.

[Sunamer]

I sincerely hope that this gets thrown out.

Imagine the precedent if this would result in a conviction? It is simply impossible to train crew for every conceivable scenario... Would be very interesting indeed to see the full text and the eventual outcome.

yes, it is absolutely impossible to train a flight crew on how to avoid a high altitude stall...
No one has ever done that before and the science still does not know how it occurs. /s

How about, for starters, training pilots that pulling the sidestick all the way aft for prolonged periods of times might not be a good idea, especially when AXXX plane is in a degraded law mode, which lacks protections? How about teaching pilots to do something similar to what AA Advanced aircraft maneuvering course was designed to teach? How about...ah, forget it!

AF447 was a disgrace..and it was a case nowhere near the complexity of some kind of fringe case and a very rare event, which it was not.

[Sunamer]

Airbus shot themselves in the foot with their "get our planes, train your pilots less = big $$$ savings" sales pitch.
They created a false sense of security by adding a ton of automation. Of course, children of the magenta (green, in this case) will abuse that automation and erode their skills until the erosion is sufficient enough to cause a crash.

[FlightDetent]

I suspect there is some cause/effect due to the pronouncements from Airbus (and others) that 'it's impossible to stall this aircraft'.
In an emergency, the human mind may not remember whatever qualifications may have been put on that statement - e.g. 'in the full up control mode'. He just remembers 'impossible to stall'.
With that in mind, if his brain was telling him it's 'impossible to stall', holding the stick full back while the aircraft drops like a rock (even with the "stall stall" warnings) is somewhat easier to understand.

Tell the pilot that there are anti-stall protections to 'help', but don't tell them they can't stall it because, in the right circumstances, they can stall it. The claim that "Airbus FBW are impossible to stall" is not something we say in the assimilated collective, td!

There is a demonstration of a ground escape technique with hard AoA hard protections (pull full back-stick, fear nothing). If that is left to be the strongest impression on the student, many things had gone severely sideways in the training. On the first day of the groundschool, the student gets a card with Golden Rules where number 1 is "This aircraft can be flown like any other aircraft". The reaction to a stall must be instinctive, immediate and decisive - this must be the lymphatic system of a pilot's body, FROM THE PRE-SOLO DAYS!

The fatal actions the AF447 pilot are quite the same as we saw in the Buffalo accident, impossible to make any sense of. Clearly, the Continental Connection Dash8 Q400 crew was not infested with any of what's suggested, and we saw the reaction of the World's best overall aviation system (I mean that honestly in awe) - mandate 1500 hrs.
:ooh: That's a strange compromise. A sad optimum of available solutions, many of our shared community shake heads in disbelief. But it is looking in the right direction - pilot basic skills - even if amalgamating flight hours, experience, and skills into one bucket not too smartly.

In real life, there must be a scope to a pilot type conversion curriculum, any brand or model. Specific features are shown and trained, characteristic behaviour for the class explained. During the exam apart from the type-specific knowledge and skills, the student will as well need to demonstrate the elementary items (stalls, turns, respecting weather minima, deicing the airfoils, DDG compliance) However, those do not come from the training objectives of the aeroplane type rating course itself.

That being said, lives got lost and there shall be no stone left unturned until it is certain that our understanding of what and why it happed is complete. A prosecution may find some more, that is not a bad thing by default. Some ideas of mine, being a fan of technology and Airbus FBW concept (as non-perfect as the aircraft may be, similar to any other) to entertain:

It is my personal hunch is AF447 could have flown into the weather unknowingly. Or some other will (and did since)
- there is a worldwide obsession of pilots to keep the radar OFF. Completely unsubstantiated. Forgetting to switch it on when entering a night segment or IMC happens waaaay too often.
- the Airbus design of the intensity knob for WXR overlay is conducive to turning the ND brightness and radar display to dim after a flight, but only the screen itself back on. Possibly leaving the crew facing a working ND with WXR on, but no radar image on it.

The controlling pilot kept pulling, contributing cues:
- speed so low at extremes of the nose-up pull, the STALL warning ceased due to IAS outside the defined range for the stall warning.
- flight directors re-appearing in a high-nose positon
-> You pull (wrongly) and the horn stops, you release (correctly) and it comes screaming again. You are back of the clock, scared witless, nothing you see resembles anything and all is impossible to interpret at the same time. Yet, now and then, the saviour FD bars flash high up in the ATT indicator. That is seriously bad human factors mojo.
- was there not a glider pilot among the crew, I heard so somewhere.

Ther other pilots present did not intervene enough.
- who had been certainly perplexed too, but perhaps not cognitively lost yet
- disconnected side-sticks, out of sight left side from the right and v.v.
(Similar case for Air Asia 5801. The last two links of the accident chain: x) disoriented F/O pulled into a pronounced unusual attitude z) the ex-F16 captain misunderstood the inputs and later failed to operate the sidestick effectively to recover).

Cockpit video recorders.
- 19 years overdue, seriously.

----------------
Some of the points I raise above have been targeted and are being dealt with. The WXR and side-stick take over not sufficiently in my opinion.

sonicbum: waistline! Otherwise, we cannot claim to be better :O

[flydive1]

How about teaching pilots to do something similar to what AA Advanced aircraft maneuvering course was designed to teach? .

Like pushing hard on the rudder back and forth?

[Icarus2001]

It is my personal hunch is AF447 could have flown into the weather unknowingly. Is this the same weather that other aircraft asked to divert around?

[Maninthebar]

Is this the same weather that other aircraft asked to divert around?

The post made it clear that this would be as a result of THIS crew not using the available wx radar (anecdotal evidence for the prevalence of this practice)

[fatbus]

Just let go ! Don't do a thing . Even if the flight path was displaced in alt law it's not going to fall out of the sky . FL 360 @2.5-3 degrees nose up and I'm guessing A330 @80% without QRH

[Mac the Knife]

I know that I'm wrong and not a pilot and all that, but the image comes to mind of a car with two separate power-steering wheels and no feedback between the two.

Now put up a screen between the drivers so that both can see the road ahead, but not each other.

Set 'em off 'em on a motorway exit and wait for the bang. I doubt whether you'd have to wait more than ten minutes.

Mac (ignorant but not stupid)

[VH-MLE]

If nothing else, the PFD pitch attitude indications (constantly between + 10 to + 17 deg nose up for the first minute or so) should have rang alarm bells for the LHS pilot that this attitude was not sustainable...

[FlightDetent]

I know that I'm wrong and not a pilot and all that, but the image comes to mind of a car with two separate power-steering wheels and no feedback between the two. Mac, there have been saves due to them not interconnected too, the score is neutral.

Staying inside your analogy (which is somewhat offset): the backup driver sees a collision coming but fails to realize that is a result of wrong yet deliberate actions of the leading one, and thus the idea of taking over does not enter his mind at all. That's what I alluded to. And my beef is that the mitigation techniques have not been executed far and wide. AF447 and the AirAsia could have been saved with a a small yet decent probability.

[auv-ee]

[Disclaimer: PP SEL, 200 hr, none in the past 35 years, but intensely followed the original AF447 threads due to interest in the search, and read much of the BEA report when it came out.]

Re: Why did the PF hold the side-stick back? FlightDetent touched on this:


There is a demonstration of a ground escape technique with hard AoA hard protections (pull full back-stick, fear nothing). If that is left to be the strongest impression on the student, many things had gone severely sideways in the training.


What I recall (perhaps incorrectly) from the original threads is that most AF training (and maybe other organizations) in A330 stalls is with respect to approach/departure stalls at low altitude, in Normal Mode, and that the procedure called for advancing thrust (TOGO?) and pulling back on the stick. With these inputs the plane, in Normal Mode, will recover itself with minimal altitude loss. It is entirely possible that the PF retained only this memory, as FlightDetent suggests. I further recall that the pilots verbally acknowledged the transition to Alternate Law 2 (without verbally confirming the implications, whether or not such confirmation should be expected). Evidently (from memory), Alternate Law 2 removes stall protection, retaining over-stress protection. It seems possible, even likely, that the PF didn't understand that AL2 has no stall protection, or didn't internalize the call-out of AL2.

Over and out.

[misd-agin]

Like pushing hard on the rudder back and forth?

Anyone who watches the AAMP tapes knows that pushing the rudder back and forth wasn’t taught in the AAMP class/video.

[vilas]

auv-ee
with your disclaimer there is a severe limitation in a meaningful dialogue on Airbus laws. What you stated about stall training and recovery in Normal Law is completely erroneous. You cannot practice stall recovery in Normal Law because it simply won't stall.
. It seems possible, even likely, that the PF didn't understand that AL2 has no stall protection I am sorry but with that level of ignorance you should be occupying passenger seat and not a pilot's seat. Why did they pull full back stick? Because nobody told them that in cruise in alternate law you never even for your life pull the stick fully back. Even otherwise you shouldn't be pulling on the stick without looking at the PFD.

[Pinkman]

Like pushing hard on the rudder back and forth?

Wry humour which in other circumstances might be amusing but misd-again is correct. It was taught but not on that course. IIRC the PF (the FO) grew up flying taildraggers where that is common (to see ahead apart from anything else).

[VFR Only Please]

(...) The question that has been bothering me most (is) Why the hell would an experienced 58-year-old captain with over 10,000 FH hours logged refuse the FO's suggestion to fly around the storm, point the aircraft right into icing and turbulence, and then simply leave the cockpit and let his less experienced colleagues deal with the fallout of his decision. I just cannot imagine that. You decide to fly into danger - you see it through and make sure you come out on the other side. (...)

There were many reports following the accident from hotel staff and others in Rio saying that the captain appeared absolutely exhausted, or ill, when he left for the flight. (I Think he'd been out on a helicopter sight-seeing trip the day of departure -- anyway, that sort of thing.)

And when the CVR was eventually fished up, there certainly were remarks to the effect that here was a guy awoken and rushed to the cockpit, and then obviously over-passive all the way down. For reasons we'll never know he clearly wasn't in good shape, and didn't take charge.

Perhaps the court proceedings will throw more light on this aspect, yea though an entire decade has elapsed.

[blind pew]

They were on a shag fest more or less as was reported in a French TV report..completely unprofessional and not fit for the flight.
Rather than resting before the flight they decided that sightseeing was the way to go,
From one who flew the route in the early days with two complete crews.
A disgrace who unfortunately took a lot of innocents with them.
Arrogant attitude amongst some in AF at the time.

[auv-ee]

You cannot practice stall recovery in Normal Law because it simply won't stall.

Thanks for the correction. Of course not stall in Normal Law, just low speed, high sink, I suppose. FlightDetent used more accurate wording.

I am sorry but with that level of ignorance you should be occupying passenger seat and not a pilot's seat.

I am somewhat self-aware, thus remaining a passenger for the past 35 years. I will be silent again.

[vilas]

Thanks for the correction. Of course not stall in Normal Law, just low speed, high sink, I suppose. FlightDetent used more accurate wording.
I am somewhat self-aware, thus remaining a passenger for the past 35 years. I will be silent again.
There's certain minimum level of handling skill and procedural knowledge that cannot be breached by a professional pilot. But sometimes there remains a serious defficiency and it doesn't get exposed for decades. When it happens all sophisticated methods of the cause finding are applied (as they should be) but because the experience level of the person involved it is never accepted that, that could be the reason.

[HarryMann]

At the end of the day, if the PIC had left the pitch alone, thought about things for a few seconds; as with any upset stick fixed is a fair starting point, then.... buzz the Captain out of his rest station maybe?
would we even be talking about it ?

Yes, AF do need serious questions to be answered re: training and pilot experience and competence to sit in that seat on that flight.

in court? Yes I can see where they're coming from on this. So disagree with KingAir.

[HarryMann]

Anyone who watches the AAMP tapes knows that pushing the rudder back and forth wasn’t taught in the AAMP class/video.

And none of the class lecturers advocated or suggested such an approach to yaw/ wake upset ?

[Tomaski]

And none of the class lecturers advocated or suggested such an approach to yaw/ wake upset ?

In regards to the AA587 A-300 accident, the yaw doublet that led to the vertical fin failure was the result of a number of interrelated issues of which the AAMP training was certainly a part. However, there were also issues related to rudder sensitivity vs airspeed and Vm limitations that were not well documented in the existing manual which contributed to overcontrol/PIO inputs. There was a lot of good information in the AAMP videos, but the section on the use of rudder in large transport aircraft did not have sufficient warnings regarding the dangers.

[HarryMann]

In regards to the AA587 A-300 accident, the yaw doublet that led to the vertical fin failure was the result of a number of interrelated issues of which the AAMP training was certainly a part. However, there were also issues related to rudder sensitivity vs airspeed and Vm limitations that were not well documented in the existing manual which contributed to overcontrol/PIO inputs. There was a lot of good information in the AAMP videos, but the section on the use of rudder in large transport aircraft did not have sufficient warnings regarding the dangers.

Many thanks Tomaski,
certainly nothing to do with tail draggers though 😏

[Willy Miller]

It's a shame "STALL STALL" had priorty all the way down.

It masked the "DUAL INPUT" calls

Which MAY have woke them up and reset the thought process, starting with "I have control"

I fly A320s and have suggested this to Airbus with no reply.

[Stuka Child]

In regards to the AA587 A-300 accident, the yaw doublet that led to the vertical fin failure was the result of a number of interrelated issues of which the AAMP training was certainly a part. However, there were also issues related to rudder sensitivity vs airspeed and Vm limitations that were not well documented in the existing manual which contributed to overcontrol/PIO inputs. There was a lot of good information in the AAMP videos, but the section on the use of rudder in large transport aircraft did not have sufficient warnings regarding the dangers.

That is simply not true. When is the last time you've watched them?

I've watched them recently, and Warren Vandenburgh repeatedly urges caution when using the rudder because of the sheer size and force of it. Yeah he doesn't say "you might rip the fin off", but that wasn't known at that point - that you might cause the vertical stabilizer to fail even at relatively low speeds.

But there are multiple warnings to be gentle with the rudder in those videos, and certainly no suggestion that it's ok to repeatedly swing it from one side to the other.

[tdracer]

In regards to the AA587 A-300 accident, the yaw doublet that led to the vertical fin failure was the result of a number of interrelated issues of which the AAMP training was certainly a part. However, there were also issues related to rudder sensitivity vs airspeed and Vm limitations that were not well documented in the existing manual which contributed to overcontrol/PIO inputs. There was a lot of good information in the AAMP videos, but the section on the use of rudder in large transport aircraft did not have sufficient warnings regarding the dangers.

A pilot friend once told me there was an anomaly in the American A-300 simulator that - in that sort of wake encounter - you could quickly and easily straighten out the aircraft with a quick 'left-right' or 'right-left' rudder input. This had the unfortunate effect of teaching some of the pilots a bad habit - which ultimately bit them on AA587.
Assuming my friend knew what he was talking about (he usually did), while the maneuver wasn't 'trained' as such, it was inadvertently taught.

[vilas]

AF447 has been beaten to death. More than 24000 posts. Now that the court has said something about 447 so only that aspect should be discussed. There's no need to repeat 24000 posts again. Even in any judicial system when one goes in appeal to higher court against lower court verdict you are not allowed to repeat the whole argument.

[ATC Watcher]

AF447 has been beaten to death. More than 24000 posts. Now that the court has said something about 447 so only that aspect should be discussed. There's no need to repeat 24000 posts again..
Agree. As the one that starter this, I would rather see the discussion moving back to the fact that the airline is prosecuted , and not the manufacturer.
So basically it is the training and not being proactive to previous incidents by said airline ( as opposed to other airlines,( e.g Air Carraibes)
I am still not sure what training other airlines did that AF did not , where they differ , except for not buying the BUSS option.

[Uplinker]

Of course the airline should be prosecuted, not the manufacturer. If you taxied an aircraft and crashed into the hangar or another aircraft, that would not be the manufacturer’s fault.

Those pilots apparently flew through convective clouds, iced up, then failed to correctly follow the Unreliable Speed memory drill. They then mishandled the aircraft and stalled it, then failed to follow the Stall memory drill. Having made these blunders, one of them then incorrectly held rearward stick for most of the subsequent descent into the sea.

We have since learned - if it is true - that the Captain was ill or not properly rested.

So, for sure it is the company that needs to address its training* standards and crew behaviour down-route.

*I think that during every SIM: It should be mandated that every pilot should be allowed to practice and then be checked on all the memory drills, and have to fly complex manoeuvres without the flight directors or auto-thrust. Otherwise, how can we know if we are competent and keep our skills sharp?
.

[ATC Watcher]

Uplinker : We have since learned - if it is true - that the Captain was ill or not properly rested.
Ill? I never saw it mentioned before , not properly rested , yes this is now a fact ,but according the Judiciary report , fatigue is only a secondary contributory factor .

What the judiciary report mentions is the lack of response to 8 previous similar IAS/pitot incidents ( 5 of which in AF) and that 78% of ASR filed by crews never received an answer. ( only 22% were answered by the company , and most of those only on the crew explicit request ) , It also mentions the fact that AF did not qualify the STALL recovery procedure as a memory item, that no training was given to doubtful IAS in high altitude ( only at low alt, where there 12,5 degr pitch is mentioned in the recovery procedure) , and that when converting from A320 to A330, unreliable IAS is not part of the conversion training .
The choice of the Capt not to select the most experienced FO as PF when taking his rest is also mentioned as a lack of directive by AF to its crew , which , according the report. gives complete freedom to the Capt to determine the functions .

What is not clear to me, is if these facts raised by the judiciary are purely specific to AF or if they are common practices in other large airlines, and if it just the judges looking for people to prosecute to satisfy the families ..

[Uplinker]

From reply #52 above:

There were many reports following the accident from hotel staff and others in Rio saying that the captain appeared absolutely exhausted, or ill, when he left for the flight. (I Think he'd been out on a helicopter sight-seeing trip the day of departure -- anyway, that sort of thing.)
.

I did not see the reports myself, hence my disclaimer.

[romiglups]

Judges working on the case since beginning and driving inquiry ('juges d'instruction") have finally turned off prosecutor requests and pronounced "no charge" ("non lieu") for both Airbus and Air France. For the time being there will be no trial, but victims ("parties civiles") and prosecutor can appeal this decision

[PerPurumTonantes]

Air France crash: Manslaughter charges dropped over 2009 disaster (https://www.bbc.com/news/world-europe-49598838)

"Magistrates in France have dropped charges against Air France and Airbus over a mid-Atlantic plane crash in 2009 that killed all 228 people on board.

The Airbus 330 aircraft flying from Rio de Janeiro to Paris stalled in a storm and plunged into the ocean.

On Thursday, the magistrates looking into manslaughter charges brought by victims' relatives decided that there were not enough grounds to prosecute.

They blamed the plane's crew for losing control after speed sensors froze."

[Baron rouge]

people surprised at the way justice is rendered should read again SWIFT.
Nothing has changed!
Why charge AIRBUS and AIR FRANCE when charging dead PILOTS is far less expensive to everybody but the victims.

[T28B]

Following up on the point made by @vilas (https://www.pprune.org/rumours-news/623617-af447-french-prosecutors-sends-af-court-negligence-4.html#post10546737):
The AF 447 library/resource base on PPRuNe (curated by our very own john_tullamarine) is here in Tech Log, (https://www.pprune.org/tech-log/539756-af-447-thread-no-12-a.html?highlight=AF447#post8476687)

I encourage anyone wishing to comment on the accident itself (rather than this legal proceeding) to first review the BEA Final Report (http://www.bea.aero/en/enquetes/flight.af.447/rapport.final.en.php).

[Sunamer]

I sincerely hope that this gets thrown out.

Imagine the precedent if this would result in a conviction? It is simply impossible to train crew for every conceivable scenario... Would be very interesting indeed to see the full text and the eventual outcome.
every conceivable scenario? Please...
For god's sake, the PF stalled the AC in unreliable airspeed situation, and rode it stalled (while pulling SS aft) all the way to the ocean surface. That is hardly a strange or unexpected situation.

[ph-sbe]

For god's sake, the PF stalled the AC in unreliable airspeed situation, and rode it stalled (while pulling SS aft) all the way to the ocean surface. That is hardly a strange or unexpected situation.

Which is something I learned how to recover from when still a pre-solo student.

Stall recovery is something that is something that everyone with a fixed-wing pilot's certificate is trained to recognize and recover from. The fact that this FO in the heat of the moment was unable to apply his training, is not something that can be attributed to the airline or manufacturer.

The other thread about the unfortunate FO suffering from an anxiety attack comes to mind.

[fdr]

That is simply not true. When is the last time you've watched them?

I've watched them recently, and Warren Vandenburgh repeatedly urges caution when using the rudder because of the sheer size and force of it. Yeah he doesn't say "you might rip the fin off", but that wasn't known at that point - that you might cause the vertical stabilizer to fail even at relatively low speeds.

But there are multiple warnings to be gentle with the rudder in those videos, and certainly no suggestion that it's ok to repeatedly swing it from one side to the other.

The A300-605R rudder limiter had (has) a fundamental conceptual flaw in the design. This is not just semantics. On Boeing aircraft, protection of the structure from rudder inputs at high speeds is achieved by either de-powering the system operating pressure, or by altering the ratio of input to output. For the A300, a limiter to the rudder throw was provided mechanically. The alteration of the stops reduces the ultimate force that can be obtained from the rudder, which is the intent of the system.

but...

In a Boeing design, with a rudder ratio changer, the rudder pedals would move the same distance, however the rudder itself would move through a smaller arc. e.g., a 4" rudder pedal movement at low speed may give 100% throw of the rudder, but at high speed, the same rudder pedal deflection would give a partial throw, say 30% rudder deflection. With a rudder limiter design, for the same 4" of pedal movement at low speed, which gives 100% throw, at high speed, the 30% (for arguments sake) deflection is achieved with 1 1/2" of rudder pedal deflection. The force required to move the pedals 1 1/2" in the A300 case is the same for low speed to high speed. The sensitivity of the system has increased by 300%.. In the Boeing case, of ratio, the force to move the pedal 4" is the same at low or high speed, but the deflection is less, so the sensitivity is reduced, by 60% roughly. With the Boeing designs that reduce rudder hydraulic pressure at high speed, the throw is reduced by the high pedal force, but the pedal movement is reduced as well, so the load is protected, and the sensitivity remains similar in both high speed and low speed cases.

The A300 system worked well enough, and was simple, but a pilot doing a doublet would find that the sensitivity was much higher than would have been anticipated, so over control is a likely outcome.

The second issue with the A300 design that was of concern was that the yoke secondary structure was designed in such a manner that if a primary structure failure occurred, the primary failure would act with a lever arm on the yoke that assured secondary structure failure would occur, which it did.

[Global Aviator]

Has anyone got the link to the original prune thread on the accident?

sorry disregard I was looking for the Phuket thread.

[XB70_Valkyrie]

For god's sake, the PF stalled the AC in unreliable airspeed situation, and rode it stalled (while pulling SS aft) all the way to the ocean surface. That is hardly a strange or unexpected situation.

Airbus pilots spend most of their actual flying time in Normal law. Pulling the SS aft causes the aircraft to climb (increases thrust and you get stall protection), that's how it flies. Even with the master caution light on, the ECAM alert for Alternate law is not very obvious. So muscle memory in the Airbus says to climb I pull back on the stick, and the aircraft does everything else - I am sure his brain was thinking, I am pulling back on the stick, why are we still descending? A: aircraft is stalled, in alternate law and there is no stall protection.

My assumption is Airbus pilots get trained in alternate law and the differences in control behavior and envelope protection. But muscle memory is another thing.