MaxJack

I have followed this thread since it started, and have tried to understand all technical terms, but I have a simple question, and, is there a simple answer?
If in a completed stall, at FL35-38, and then ”falling”. Could you recover, and if yes, how do you recover from the stall during the fall?
According to the Airbus Stall and Recovery Procedure, thrust is not an option, but could rudder settings bring it back again? And how then?
Or is there no way out, as in this case.

infrequentflyer789

G isn't anywhere near 1.65 apart from at the beginning of the climb - look at the traces.
If G limits kicked in, it wasn't in the stall.

AlphaZuluRomeo

Push the stick, thrust idle => dive.
Trade height for speed, while reducing AoA.
Then recover from the dive.

Why the rudder? Best to keep your aircraft symetrical, a spin is worse than a stall

---------

About the THS & difference in normal/alternate law: we already discussed that some months ago.

rudderrudderrat

Correct. That's why the stab trim was allowed to continue trimming.
In Alt Law, with side stick free - the FBW computers will continue to move the elevators whilst attempting to hold the attitude whilst the speed washes off.
In ALT2, the auto stab trim continues despite exceeding Alpha Prot / Max.
The natural nose drop at the stall is masked by the above.
The "Stall Stall" warning is all that remains but is unfortunately turned off when IAS<60kts despite being airborne.

It's going to be one heck of a Human Factors course.

airtren

Of course "knowing your machine" is a very important element, and agree 100% with parts of the post that are about that. The training and policies failed to prepare the pilots to avoid the situation (first), and then to recover from it (second).

But I disagree 100% on the parts of your post about the machine, and responsibilities associated to it, which include the procedures that only a manufacturer can develop.

The machine was brought to an extreme state, by a convergence of elements in which the machine itself, and procedures known at the time, had a complex contributing role. As a recovery from that extreme state was not successful, it was a fatal state. Such a scenario was never tested before, and the machine's behavior was not known in its totality. The machine is known better now, and the important resources spent to recover and analyze the CVR, FDR and parts of the machine were worth for that reason as well.

The procedures have been already changed. Recognizing the machine's contribution is a very important element for the improving and progress of technology, to avoid a repetition with the same type of machine, or for making new generation safer machines.

It's one of the elements that moves the technology forward. Not recognizing that would be a big failure for the technologists, for the manufacturers, for the industry.

CONF iture

What strikes me by reading the Airbus Flight Safety Magazine is :

It is important to know that if such a thrust increase was applied when the aircraft is already stalled, the longitudinal effect would bring the aircraft further into the stall, to a situation possibly unrecoverable.

AlphaZuluRomeo

Indeed. And What strikes me by reading the same sentence is that a situation possibly unrecoverable is not clearly defined. Why would a situation be unrecoverable?
- Because the plane cannot recover ("locked in stall")? That's not demonstrated, and in fact AF447's FDR traces tend to prove the contrary.
- Because the plane may not have the height needed to perform the recovery, this height being larger as the stall is more pronounced/longer held? That's more how I understand "possibly", here.

Lyman

In ALT LAW 2, there is no Bank limit, and it is clear that PF was focused (overly so?) on his rolling. Perhaps to the extent that he neglected PITCH. If he wanted climb (and his stick says so) he may have been eager to address the rolling. Isn't Pitch modulated by the computer? If so, that is my point, not the 2.5 g barrier, but the "comfort" g exhibited in the climb which allowed the a/c to lumber up to 38k low energy, to STALL in mush?

Had the a/c STALLED earlier, before it lost its energy, mightn't there have been a NOSE DROP, and an inescapable cue to get the NOSE DOWN?

I am not saying the PF was tuned to his a/c, but if the a/c wasn't trying to supply comfort instead of reality, could this have turned out differently?

AZR. The language re POWER and then possibility of unrecoverable flight demonstrates that the authors do not know what will happen. No harm in that, but it is crystal clear that they are comfortable in this lack of knowledge. The assumption is that all pilots will acquiesce to the instruction. Some pilots like to know WHY? Given the State of the ART it's also clear that to tell these pilots what will happen serves no purpose. There is a message in that.

CONF iture

AZR,
The way I see it is :
For obvious reasons a flight test cannot be risked.
AF447 was possibly unrecoverable the way the stall was initially engaged ...

airtren

I find too that the article and also the presentation need some refinement. The presentation - link posted also by:
has some ambiguities that may lead to misunderstanding. While the "Loss of speed" for a certain AOA, is mentioned as an important element contributing to a STALL, on page 8, the text "Stall is an AOA problem only. It is NOT directly a speed issue" may be misinterpreted.

While removing the "increase of thrust" from the list of actions for "gaining speed" is understandable because of its effects for the type of engines mount, on increasing the AOA, contributing to the STALL, the "gaining of speed" is still important. It is part of the solution/recovery, and the change of AOA - bring the nose down - is a means to achieve just that. I think more clarity would be helpful, as the training should not have an ambiguity which would allow the interpretation that gaining speed when the nose is down is wrong.

CONF iture

Why not including the complexity of the machine or even the interface in the equation ?

I wish it would be that simple - Unfortunately the complexity involved makes it is not.

What to expect when a Manufacturer puts on the market a product that's supposed to correct the pilot's errors to the point that its designer publicly announces that even his housekeeper could fly it ... ?

Lyman

On a more concrete note, I still have yet to see the timing of the PF's stick for duration that would not invigorate the Trimmable Horizontal Stabiliser. Offered as proof is the mere statement that "The THS didn't trim because the stick did not linger in NU long enough."

I'm calling bs........ Anyone?

Does some neutral party want to index the SS for duration/trim activation? Because if the THS did not move, and the SS commanded its action, what held it back? G? LAW?

Old Carthusian

CONF iture
Quote
'Why not including the complexity of the machine of even the interface in the equation'

Because knowing your machine enables you to deal with its complexity. A level of professionalism is necessary. It is, in fact, a dangerous illusion to regard more modern computerised aircraft as more complex. In fact older aircraft were often the more complex requiring more attention and work. As technology has developed aircraft have in fact become simpler. Certainly aircraft can become even simpler but one cannot attribute what happened to the interface or the complexity of the machine. And if you pause and think how many A330s are flying and how many hours they have flown one cannot call these overly-complex machines. An overly-complex machine could not have survived for such a long time.
We have an aircraft which by all accounts flies well without protections. In fact those who've flown it state that it is one of the better ones to fly manually. We have a culture (that of Air France) in which a certain casualness has developed (an Air France safety audit identifies this). We have certain incidents involving Air France crews which seem to indicate a lack of knowledge of how to handle their aircraft. Putting a machine into an extreme situation then blaming the machine because it is operating outside the parameters it was designed for is futile. The PF put his aircraft into a state where it could not recover. It wasn't the machine that produced this situation but the humans. Understanding why they did so is the important factor here.
As things get safer and we develop more protection against danger paradoxically the danger increases. Humans tend to react to safer situations by dropping their guard. They do not act with caution and thus they pay less attention to safety. This is where training, CRM, culture and psychology all come into play. This is why knowing the machine and professionalism are so important. Remembering that you are responsible for more than one life. But all of this is to do with the people who operate the machine and their responsibility.

AlphaZuluRomeo

CONF iture
Indeed, "possibly" as in "not demonstrated, and will not be either way".

airtren
the "gaining of speed" is still important. It is part of the solution/recovery, and the change of AOA - bring the nose down - is a means to achieve just that.
Uho... Perhaps is the procedure not perfectly written. On the other hand, I disagree when you write "the change of AoA is a means to achieve a gain of speed" (if I understood correctly?)
It's the other way : the gain of speed is a mean to lower AoA (all other parameters equals). And lowering the AoA is what is important to avoid/exit a stall.

DozyWannabe

Here we go again. Once more with feeling, and hopefully for the last time.

• Bernard Ziegler was not the "designer" of the A320, though as Engineering Director at the time he did have final sign off
• The flight characteristics and systems functionality were tested and signed off by Gordon Corps, the archetypal "pilot's pilot" and successor to D.P. Davies at the ARB
• "Correct[ing] the pilot's errors" is only one aspect of Normal Law, another important one being that a pilot can safely fly close to the design limits of the aircraft
• As daft as the statement looks in hindsight, Ziegler didn't say anything about his concierge being able to land the aircraft or handle it in a degraded state
• If the people still holding on to this nonsense as some kind of affront to the piloting fraternity would let it go (after all it did happen 24 years ago), then discourse would be a lot easier

Lyman

Aviation owes its remarkable record of safety in one area.

One: Powerplant. Complexity existed because mechanical solutions could not keep up with how basically simple actual flight is. The Turbine and then the Fan made aviation wicked safe. The question we avoid, the Rhinoceros in the room: Why is it not much safer?

I submit that the actual flight control arena is a Straw man. 447 demonstrates this elegantly.

Carthusian is thus absolutely correct, imo.

Training? Skill? What about the rampant myopia present here on this thread?

AB have crafted for themselves a politically effective position of Carte Blanche from criticism.

With an arrogant ignorance of how simple the machine actually is, and not actually sophisticated as it is presented, they feign "misunderstanding" or "ignorance" on the part of the operator, as always, it is a fact.

It is indeed a machine that is safe and dependable. What is galling is the remarkable stubbornness they display when confronted with obvious blunders in its programming. These blunders include the attitude it takes to reactively defend them.

CONF iture

To the contrary, IMO, the illusion is to regard them as less complex.

Exactly my point.
Cut down in the protections then, they are the ones to bring the over complexity you are not ready to look at.
An even better product will emerge.

If you're not ready to look in that direction as well, I’m afraid you miss a point.
It is a bit like Ziegler who doesn’t want any to touch its baby – It has to be all pilot’s fault :

http://www.crashdehabsheim.net/Films...emoignages.wmv

CONF iture

Bernard Ziegler will be glad to see you did finally promote him from sales department to Engineering Director, not after initially pretending he was not even an engineer at that time.

I hope that in your job or life when you put your mark at the end of a document you’re conscious on your involvement ...

Like it or not, Bernard Ziegler is still the father of the 320, and don’t take it too personal if he said stuff you disapprove.
Who’s talking about feelings …

DozyWannabe

The role he played in the actual specification and design was minimal, as you well know. He signed off on things, maybe did some arbitration - but that's about it.

That's denigrating Captain Corps' role more than a little, and the only reason more don't know about him was because of his untimely death in 1992.

BZ is the "father of the A320" in the same way Bill Gates is the "father of Windows" - it may have been his name on the press releases but the work itself was delegated.

Like certain other posters, you're fighting 24-year-old battles, which frankly makes you look a little ridiculous. This is about AF447, not AF296 - and there is no evidence - none - that complexity, whether perceived or real played any part.

The Airbus FBW design is here to stay - why can't you get over it?

[NB. If the protections were gone, your hero Asseline would not have been alive to lie about the aircraft in the first place. ]

airtren

Your rephrasing of my text is still in line with what I meant, in the context of the presentation, so I will explain from a different angle. I will use material from that presentation, as what I referred to was in its context:

The change of "AoA" affects the lift coefficient CI, according to the

CI curve, as a function of AoA. (page 5 of that presentation).

The CI is a member of the "lift" equation - see bellow - and thus it affects the "lift":

Lift = 1/2 pSv**2Ci, (page 4 of that presentation)

(where: p = air density, S=wing surface, v=CAS, v**2 is square of CAS, Ci=lift coefficient)

Examining just from a pure Mathematics perspective, given that the S is a constant, the other two elements that affect (and will increase) the "lift" are the "p" and "square of v" (if they increase).

Furthermore, if theoretically and abstractly one can assume that AoA does not affect speed - so the speed stays constant - as CI cannot go above a Maximum Value, one may try to change AoA as much as he wants, but that (Max) CI, with an insufficient "v" (speed), and insufficient "p" may still result in insufficient "lift" and thus no return to normal flight.

So my point again is that the change of AoA affecting the CI alone, may not be sufficient to cause the return to normal flight.

It is rather the combined effect of the AOA on the CI, with the AoA effect on increasing the speed and thus the "square of v" that from a possible "speed" insufficiency in a "insufficient lift", to a "sufficient speed", which by resulting in a sufficient "lift" makes the a/c return to normal flight.

I can understand why the written presentation may attempt to reduce the aspects of recovery from Stall that may be look too acrobatic, because of the drastic changes of speeds, and altitude. But I can see how the verbal communication during the presentation, or a training session may be very explicit on the dramatics of the change of AoA to aggressive Nose Down, and letting the a/c in a Nose Down fall to gain sufficient speed, to change the AoA again, to level off, and transform the falling speed into horizontal speed, get sufficient "lift", to get to normal flight.

....

You say, the "gain of speed is a means to lower AOA".

I don't see how the "gain of speed" has necessarily a "lower AoA" consequence. Perhaps your elaborating would help.