Gums,

Both Dozy and PJ2 (a former A-330 pilot) did recovery exercises in the SIM. Going back to the original AAF447 Thread #9, this is what PJ2 had to say about one of his recovery attempts:
One other thing that I think is important from a AF447 historical perspective that PJ2 pointed out discusses "Startled" in a collection of posts:

I like PJ2's thoughts here, in effect, everyone is startled when something unexpected happens, it's what you do or don't do after being startled that determines the outcome.

A last point: There is a difference between stall prevention, given the warnings and what is done to recover from an actual stall. This Airbus presentation discusses these differences on pages 11 & 12:

http://fucampagne2008.u.f.f.unblog.f...lprocedure.pdf

Cheers!

Startled or Man with hammer syndrome
 

Bonin was convinced he needed to pull the stick back all the time.
So convinced he ignored the nose up attitude, decaying airspeed, stall warnings, alarms, incursion into other FL and ignored Robert repeatedly telling him to descend. Bonin clearly verbalizes that he was stick back all along which tells us that it was a conscious decision. Not only was Bonin behind his airplane he was ignoring every fact being given to him. Startled is an over simplification of Bonin's mental state and warrants further consideration along with all the other failings. I leave it to the pysch experts to explain his fixation with stick back e.g. man with hammer syndrome, confirmation bias.

02:13:40 (Robert) Remonte... remonte... remonte... remonte...
Climb... climb... climb... climb...

02:13:40 (Bonin) Mais je suis à fond à cabrer depuis tout à l'heure!
But I've had the stick back the whole time!

At last, Bonin tells the others the crucial fact whose import he has so grievously failed to understand himself.

02:13:42 (Captain) Non, non, non... Ne remonte pas... non, non.
No, no, no... Don't climb... no, no.

02:13:43 (Robert) Alors descends... Alors, donne-moi les commandes... À moi les commandes!
Descend, then... Give me the controls... Give me the controls!

Chaps,

Tech Log seeks to discuss and debate technical issues - which, obviously, includes the human aspects of technical subjects.

Sometimes, some of us have very strong views on this and that.

However, we don't need to venture into personal commentary of an adverse nature where that is not materially pertinent to the underlying discussion - where this might be permitted, it will need to be justified objectively as being pertinent.

As always, the desired aim is to play the ball, not the players.

A slight possibility exists, but I very much doubt it - for one thing, the fact that the aircraft can stall when in a degraded flight law is down in black and white in both training and ops materials that I know of. Also, the crews in the other similar incidents I mentioned earlier weren't flying FBW Airbus types, and the PF in those cases disregarded not only the Stall Warning and stick-shaker, but also their crewmates telling them they were in a stall!

@Turbine D:

I have a lot of respect for PJ2, and agree with the vast majority of what he writes. However what bothers me about writing off the consequences of what has recently been termed "startle response" as a
is that while the industry has - as he said - been dealing with scenarios such as this for a very long time, the fact remains that there is always room for improvement.

I was enjoying one of my periodic re-reads of "Fate Is The Hunter" the other week, and there was an embryonic version of the subject even then. Gann refers to it using the concepts of "fright" and "fear":

(Emphasis mine)

If no less a legend of aviation than Gann had a handle on it many decades ago, then I'd venture to suggest not just that there may well be something in it, but that there may be some mileage in studying the effect properly.

What bugs me about how a lot of the hearsay around this case has gone down is that there's been a worrying groundswell of dismissive opinions from some quarters along the lines of: "Well, he was low-hours/automation-dependent/a 'magenta child', so he didn't really know how to fly". Which raises the worrying assumption on the part of people making such comments that they're dead certain that it would never happen to them. Part of the reason I've been banging on about the other incidents where captains with in excess of ten thousand hours and/or with a successful military career behind them have ended up doing more-or-less exactly the same thing is because people making those assumptions, in the face of these contrary facts, frankly scare the bejeezus out of me!

@Machinbird,
Pardon. " As we know, the transcription of the acronym "PIO" has been modified and is still unfixed so long the solutions to avoid and recover from a variety of "oscillations" are still to fix and understand better. My propositions are leading too far of common definitions of theory of systems. I don't want to contest them (instability, gain, human operator transfer function). I have to write that clearer.:{

AF447 discussion deuce
 

Well, JT, I think we need to let this thead go on. Maybe re-label it and start over with existing posts, and that's your call.

There are a lot of techniques and personal war stories and lessons-learned and such for folks to learn from, huh?

We have had a diverse group of pilots and engineers here for almost 5 years. As a "light" pilot with an engineering degree, I was accepted by this current group to offer my lessons from an early FBW jet and to comment about other aspects of the aviation mistress we all served.

Seems all of us want/desire to help prevent accidents if all crews knew the data from the previous episodes to include crew techniques/actions, past problems of mechanical or electronic nature, and the beat goes on.

I vote to keep the thread going, but maybe re-label it and move to another forum if we have to.

On the road now to high country, owe some decent responses to some posts and will try to keep up with my morse code telegraph line from cabin to 'net connection. Carrier pigeons won't stay here, and chipmunks are unreliable.

Xcitation makes a telling observation about the progression of events in the AF447 cockpit.
Have you ever had a wasp in your vehicle as you were driving? Darned distracting, wasn't it? In fact, this type hazard is so distracting that drivers experience a very high frequency of traffic accidents during these events because they cannot properly prioritize hazards and give their full attention to driving.

I have a personal theory (that is only a theory) that all 3 pilots were rapidly sent past their maximum sensory processing thresholds and thus they were filtering out all background noises. The development of this process in all 3 crewmembers seems to have followed a triggering sequence.

Bonin started the sequence by encountering roll PIO which dominated his full attention in a manner similar to the wasp phenomena that I mentioned at the beginning. His control of the aircraft was clearly deficient.

Bonin's flying greatly concerned Robert but since he felt constrained by the Captain's briefing, he did not give relief to his concerns by assuming control, but instead monitored the gent (big wasp) in the other seat who was effectively trying to kill him at the moment and concentrated on calling the Captain.

By the time Captain Dubois arrived on the flight deck, the situation had progressed to the point that the aircraft was not properly following the controls. Both the copilots loudly made exclamations that the aircraft was not under control. In effect, he had walked into a cockpit full of wasps. Apparently nothing in his background had prepared him for unraveling such a mess.

I know that things are never supposed to get this far out of hand in an airline cockpit, but it happened in real life. A triggering event followed by a triggered sequence is one way to explain how something like AF447 happened.

They never mentally heard the stall warning because there was no relief from the mental pressure of the event. The hazard kept increasing as the altitude unwound. There was no clear AHA moment.

gums,

The thread is far too interesting not to let it run its course ....

Did the Cpt Dubois experienced the SW or "inverted SW" only ?
 

And there is no certainty that even the SW was still on at the moment of his instrument scan.
And of course nobody want even to try to establish if at that time a determined (or yes, desperately) maneuver would allow recovery (no, not only full fwd stick , but rather a forced direct law + manual trim full ND at structural limit, even if that would made things to fly in the cabin, followed by a recovery at the g limit again).
After all many "impossible" situations were saved by pilots performing unstandard and untrained maneuvers. But yes - in a good situational awareness, which was not the case of AF447 - this plane set trap after trap to his crew, and of course they are only to be blamed for falling in ...

Quote:

---

Did the Cpt Dubois experienced the SW or "inverted SW" only ?

---

The answer of course is that he only heard the inverted Stall Warning.The stall warningstopped as he entered the cockpit because the airspeed was too low for it to make sense (NCD).

Even if he heard it, it didn't make good sense to him either. It went into the too hard to understand right now category.

Dozy, re:

I was well aware of fear/fright factors from the day one of my private licence and have felt such fright as all pilots do, a number of memorable times hence, from the early years right up to the year of my retirement. Such things are visceral and not subject to "improvement", (thank goodness). But physiological and psychological human responses which sublimate rational responses may be mitigated, as I went to some trouble to define, by thorough training, an abiding mentality of cockpit discipline when the going gets interesting, and recent developments borne of the results of such cockpit chaos and rapid loss of SA as we see in the AF447 cockpit; these developments are known as CRM and SOPs.

To me, the cockpit is where the buck stops; it is those hands and minds that govern everyone's lives on board. But that does not excuse or turn focus away from organizational circumstances which "planted the garden" for this failure.

But truthfully, really, there is some hindsight bias at work in this. In terms of this organization and the larger industry, was UAS on anyone's radar as a top training priority? No, it was not, even bearing in mind that much had changed since the two initiating accidents which brought the loss of airspeed information into the industry's 'consciousness' from which it created the UAS cockpit responses. AF had addressed the issue as early as 2006, providing training as to when and how to apply the UAS drill. As with other professions with similar responsibilities and authorities, the standard is constant learning and engagement over and well above the formal recurrent training schedule and requirements. I don't think that's too tough a standard at all but, again similar to other professions, some do "just come to work"...

We all know that thirty-one other crews experienced this event and for all of them it was a log-book entry, just as it should have been. I'm sure each one would verify an initial fright - in fact one would wonder about a pilot who did not exhibit that initial reaction. To be sure, there was "startle", (if that's the way some want to put it), in every one of these events - I can tell you that there was such during a subtle failure of airspeed info (blocked pitot - airspeed acting as altimeter) on a B767 over the Rockies one dirty night!). But that's what training, SOPs and CRM are for - to replace fear quickly and provide the road map forward which sublimates initial fright to engender disciplined action. It works - thousands of minor incidents like this occur in airliners every day and they are non-events.

Gann would also have known and intuitively appreciated that 230T of mass does not lose energy in a few seconds to the point of falling out of the sky; - there is a "pilot's (sixth) sense" regarding the airplane involved here which seems to have been missing in the crew combination of AF447. A number of things come to mind: - the lack of clear command-and-control structure left by the captain and engendered by the airline, (ie, the F/O in the left seat should have taken control but out of deference to structure, did not), the possibility that sublimation of the airplane's situation, (ie, "not-serious-but-requiring-attention"), a desire to "do something" which led to the instantaneous, (2-second) reaction by the PF to remembered-drills for UAS done right after takeoff in the sim instead of doing 'nothing' for the moment, (as I had suggested in June, 2009).

Stall warning or no, there is nothing unclear about buffet, an unwinding altimeter that is going through FL's once every three seconds and the inability to arrest a 10,000fpm rate of descent.

I think OG has nailed the PIO matter. I've tried it in the sim, (not just the UAS sim session). The reducing oscillations are to me a successful response to the very sensitive roll that the airplane has in roll-direct; the PF got it under control. The sustained and increasing pitch is significant enough to be separately-intentional and not a result of inadvertent input while getting the roll under control.

- this is the crux of the matter. We can talk about 'startle'/'fright'/PIOs/FBW/UAS training/QRHs etc etc until the cows come home, but as PJ characteristically has pointed out, there was a big 'hole' in comprehension throughout the cockpit, and I recall beating my gums (sorry Sir...:) ) over and over again about 'symptoms of the stall' in days of yore, and a low (or in the AB case, NCD) IAS, high nose attitude and ginormous rate of descent SHOULD have switched on the light. Surely it is not unreasonable to expect that at least one of the three would have noticed?

And that is *exactly* what I meant by "improvement", so please accept my apologies if that wasn't clear.

Absolutely - and all I was saying was that there may be some possible solutions to plug the gaps where the situation still goes pear-shaped in spite of those things.

I certainly agree that the pitch input was separate from the roll inputs, and always have. What I do question is whether the intention was conscious or sub-conscious.

@DW
How do you reconcile the PF saying that he had been "stick back all the while" with a subconscious action. Surely his words explicitly state that PF intentionally stick backed as soon as he took control initiating the zoom climb.
IMHO one can question why he chose to stick back as a response to UAS however it is without doubt that it was intentional. I think an earlier poster suggested that his arm must have been aching with that continual pull pressure on the side stick. Maybe thats why he eased off momentarily a couple of times.

Fair point, but there's an important distinction between "all the while", as you put it, and the actual English translation which is:

(Final report Annexe 1 - p31)

"All the while" would imply that he's been pulling full stick back since the onset of AP disconnect, which does not tally with the DFDR trace. "For a while" implies that he has been consciously doing it for a period of time, but gives no clue as to when that period of time began.

Towards the end of the sequence, as the aircraft was descending in a mushy stall, it's fairly straightforward to reason why he may have decided to pull up. But at the beginning of the sequence, when the aircraft was still flying and the two F/Os had just discussed why they could not safely climb higher, it becomes a much more problematic issue to fathom out.

Mental perception: clever sophisticated machines can fail in sophisticated ways . Crew convinced themselves that they were faced by something complicated and nothing should be taken at face value, including rushing noise, speed readings, altitude. Inverted stall warning treated accordingly.

From the SSFDR traces, regarding the PF's sidestick position from initiation of the climb to impact:

http://www.smugmug.com/photos/i-GRhj...GRhjsb6-XL.jpg

Right, but to give it a bit more context:

http://i1088.photobucket.com/albums/.../fdr-munge.png

The descent begins around 02:11:17, shortly followed by the PF's emphatic back-stick to the stop after 02:11:30 (which has the unfortunate consequence of completing the THS's drive to the full nose-up position, fouling the pitot tubes and consequently rendering the Stall Warning intermittent). Though it was completely the wrong thing to do, I can understand the instinct to pull the nose up when the altimeter begins to wind down.

What I have more trouble understanding is the motivation behind pulling the nose up past half-stick deflection around 02:10:09 (i.e. at AP disconnect), when the PF has just been explaining to the PNF why they are unable to safely climb to a higher flight level.

Dozy;

The elevator had sufficient authority to get the nose down - it was never stalled, (discussed at length in The Threads). Had the stick been held forward the aircraft would have recovered, while the THS was returning to it's cruise position, (that's the behaviour we saw in the sim).
Yes, it is instinctual and must be, to the extent possible, "trained out". Difficult though - better to avoid such a requirement altogether as such instinct has killed at least two other crews, (DC8 - test flight, B727 - placement flight) that we know of. It may have been a factor in the Trans-Canada-Airlines DC8 crash at St. Therese in the mid-60's.
It isn't productive to wonder because, although we may imagine all kinds, I doubt if there is a specific reason. By this time the PF was lost/confused but unwilling to yield control to the more experienced man, who never pressed the take-over button for 15" or whatever it takes to lock the other sidestick out.

Right, and it's what we saw in the sim as well (my summary on the megathread is linked here):
http://www.pprune.org/tech-log/46062...ml#post6793521

The caveat was that it took between 5 to 8 seconds of constant full nose-down on the sidestick for the THS to return to a "normal" position. Release to pitch-neutral too early and you run the risk of the THS causing the aircraft to pitch up again. Of course, if you're certain of what you're doing that's no problem - but if the environment is as disjointed as the AF447 flight deck seems to have been, I'd argue it'd take quite a bit of nerve to follow through.

Right again - I guess the angle I'm coming from is asking whether there may be more consistently effective methods of training it out than those we have now. And while it may not be productive to merely wonder, as you say, the idea that 80% of a random sampling of pilots instinctively pulled up when startled by a Stall Warning - i.e. the effect was such that it completely negated their training - worries me greatly. It also can't help but make me think we're missing something.

The warning "Stall" seems to invite some pilots to put the stick on their stomach ... So replace this "Stall" warning by another like "Noze down"
Maybe more pilots will better understand ...... if any
Simple problem .. simple solution :)

PIO
 

Some basic definitions and references about control theory, stability, oscillation, gain and PID,etc.

Oscillation - Wikipedia, the free encyclopedia

Control theory - Wikipedia, the free encyclopedia

PID controller - Wikipedia, the free encyclopedia

You seem so sure that the PF would have had no reason to climb intentionally. I'm not so sure.

At 01:52:23, when he pointed out to the Captain: "You see the REC MAX it’s changed to three seventy five", the Captain's only reaction was to start making preparations for his rest break. Later, when he said to the PNF in the briefing at 02:00:33 "So what we have is some REC MAX a little too low to get to three seven", I can imagine him thinking "...and this doesn't seem to bother our stubborn old man at all."

From the report (2.1.1.2):

Hi Dozy;
Re,Sorry, I may be gradually losing it but where do I say, "...the idea that 80% of a random sampling...", etc? I don't believe I'd say such a thing because I don't believe the statement to be true in the first place! ;-)

Sorry - ambiguous wording. "as you say" refers to your assertion that "It isn't productive to wonder".

You may not believe the 80% figure, but the stat and its source are very real and discussed earlier in the thread (Tarom A300 over Paris Orly).

I agree with you
PJ2 and don't believe to the existence of such an experiment.

Well from an informal sampling of USN carrier pilots, 99.999% will reduce angle of attack when receiving a stall warning and the ones that don't are generally facing an imminent collision with something solid and are trying to at least get the cockpit above the solid object before impact.:}

Ah, understand thanks. I am indeed losing it or over-interpreting...!

I read the posts earlier in the thread earlier, (and again, now) - I was looking at the "80%" figure more broadly and not in the rare specific cases cited. It's not a matter of belief, the stats are indeed there, (I was thinking training scenarios, etc).

While the sample rate is to small (Tarom, Colgan, Thomson, Turkish...are there others?) to take this statistically-seriously, I could understand the notion that it, (pulling up and in some cases not adding power), at first warning of a stall, should never occur in such a high-risk event does take us beyond statistics into a mandate to do something.

In this then, I don't think it is possible to over-train.

I believe in Gladwell's notions, (Outliers). I play an instrument which is both a cognitive and muscle-memory activity and the only way to play accurately and fast is to practise, practise, practise.

Both cognitive response time and muscle memory improve dramatically and, (if one is performing in public, which I certainly do not!), reduces performance anxiety, and startle tremendously regardless of whether it is an airplane, typing on a keyboard, playing the piano or reciting/speaking text from memory. The mind-and-muscles "go to the right place", to loosely describe it.

I know you and many here know this already - it isn't new, though Gladwell has put a refreshing interpretation on it. Here, not only was there not a correct response, (or a response to incorrectly-perceived cues) but no concept of what was about to rapidly happen to the aircraft's energy in a sustained climb. There seemed no concept or awareness of an airplane's physical state or environment. It should not be possible to get into the cockpit of a transport aircraft absent these basics.

In the light of AF447 especially, but also in view of the "80%" perspective, I wonder what our regulators consider "stall practice" these days?

I know that in Canada, once one has been initially certified for the airplane, there is no requirement whatsoever to demonstrate approach-to-the-stall or stall recovery in fly-by-wire aircraft. I have no idea what the justification is for this because it seems to imbue FBW technology with some form of magic when it's just another way of moving the flight controls. I suppose they're thinking of the Airbus protections but we all know now that a more sophisticated approach to such training needs to be done. What do the JARS have to say?

Also, PDF version in English: http://www.bea.aero/docspa/1994/yr-a...r-a940924a.pdf (Page 30, Para. 5)

and French : http://www.bea.aero/docspa/1994/yr-a...yr-a940924.pdf (Page 29, Para. 7)

The document exists, and if you believe there was no experiment to glean those statistics, then you're effectively accusing the BEA of making the figure up - a serious allegation!

[EDIT : NB. My above comment is directed only at roulis. Cheers for the clarification, PJ2!]

@Noske

Thank you for setting the context prior to AP disconnect of PF anxiety and a desire to climb above the weather. So apparently the PF was already biased to climb before AP disconnect. In stress such bias can be amplified.
That said why would he not verbalize his action to increase flight level to his colleague when he initiated the zoom climb. My own view is that it was a conscious action and a failure to verbalize/communicate to PNF rather than a subconscious.

xcitation;

Re,
Yes, I think so. This is consistent with all cockpit behaviour noted by the BEA - absence of SOPs and CRM at critical junctures where it counted and likely would have made a difference in the outcome.

Anyway, there is little to re-argue/re-hash that isn't in The Threads, so returning to lurking while awaiting something new and interesting.

I'm not denying it's a possibility - but on the other hand, if the control input was an entirely conscious move to get to a higher FL, why did he not level out at a sensible point and instead continue to drag the aircraft up towards "coffin corner"?

Paradoxically .... unconscious can do things consciously !

Bias
 

@DW
Excellent point. I think this is where bias comes in. When all you have in your hand is a hammer everything looks like a nail. To him he had pre-occupied his mind with getting higher to fix the problem. Then after AP disconnect he got confirmation bias coupled with an endorphin rush. I think this is the gremlin that we referred to as startled. As a pilot it is critical to understand that the brains perception can become biased during stress. This is effect amplified by modern highly automated cockpits.

While I certainly wouldn't disregard the possibility noske brought up, I had a quick look back at the report because I was sure I was forgetting something - which turned out to be this:

Therefore there exists the possibility that the "problem" as you describe it - namely the weather/turbulence situation and how to avoid it, may have been considered as solved in the mind of the PF with the PNF's lateral course change.

Solved? I don't think I'd give that interpretation a very high probability.

Crews usually consult on the best route through weather. The process has enough variability that you would never consider a weather problem solved until it is behind you. Best you can hope for is to achieve the least hazardous route.

Semantics seems to strike again...
 

For Gods'sake Sir, I think it's possible that the "overpedantic " gentlemen had a slip of mind with "solved" and would better used "dealed by..' or "treated by..."

NOT semantics. This relates to the mind state of the person flying. While picking your way through weather, if the person flying consideres that an instantaneous choosing of the apparent best heading has completely taken care of his weather problem and he can relax and think about other things, then he probably doesn't even belong at the controls.:mad:

We can therefore consider in addition to other possible causes for the persistence of pulling (and the uncommon pitch of 12-15° followed), the initial intention to avoid the storm by climbing, pursued simply omitting to move the throttles and to let the other guy know what he intended...

@M'bird:

Where do I say that "instantaneous choosing of the apparent best heading has completely taken care of his weather problem and he can relax and think about other things", or words to that effect?

Maybe it's because I've been reading Dr. Al Diehl's book on human factors recently, but I'm talking more from a psychological perspective than a practical/piloting one. Of course a minor course correction isn't going to completely solve the weather problem - and I doubt the PF thought so, but generally I find that people tend to get their stress levels elevated more when they perceive that there's a problem and nothing is being done about it, and while it doesn't exactly go away, it does tend to subside somewhat when they recognise that attention is at least being paid to solving the problem.

Incidentally, I completely recognise what you're saying about USN carrier pilots, but I think sending every trainee airline pilot to go shoot carrier landings before giving them their ATPL is probably overkill... :}