I'm pretty sure that's not the case - to the best of my knowledge the annunciator system is separate from the flight control logic. The Control Law logic itself is triggered by failure detection scenarios between the sensors and flight control computers. It should not trigger when one of those sensors records something outside of the envelope, only if there is a confirmed failure of one of the systems.

In short, in the unlikely event there's a failure of the system and a concurrent entry into stall warning territory, then the control law will degrade, but a value that triggers stall warning by itself will not trigger law degradation as long as the system checks out as operating normally.

In the QF72 accident a faulty ADIRU emitted erroneous data for various parameters. The crew received stall and overspeed warnings, the protections responded to an erroneous AoA value by pitching the airplane nose-down, but Normal law was maintained.

@retiredF4, dozy, hn39

I checked, I'm wrong you're right.

Should re-read and revise before posting rather than rely on memory.

Perpignan also stayed in normal for ten secs or so during SW, before other parameters kicked it out of normal.

Remember that the root cause of the anomalies at Perpignan were damaged static ports caused by improper maintenance procedures during cleaning. The level of systems issues caused by loss of valid static information is considerably more severe than those caused by loss of pitot tube data (e.g. comparison of Birgenair 301 [pitot] to Aeroperu 603 [static]).

Reading the recent comments, I am struck by how quickly the nose up Airbus lost altitude, with less than 90 seconds available to recover from an initial altitude of - was it 38,000 feet? It calls to mind that early display when an Airbus planked into a grove of trees when the engines didn't spool up quickly enough....in that case did the protection prevent the stall?

Yes, and it's likely that the death toll was significantly reduced as a result.

In a more serious setting than an internet forum, a statement like that would be borderline slanderous. Do you have any evidence for this supposition? Or is it more likely that they *did* analyze the technical aspects and found that no technical failure matched the evidence?

I've asked before and I'll ask again - I can't see any benefit to the BEA by smoothing things over for Airbus. It would also open Airbus to greater risk of problems if the issue repeated itself. What do you believe either would stand to gain?

No technical failure (exept pitots of course) was early announced by Airbus after a BEA note (if I remember well ) released around 27 May 2011

@jcj:

Well yes - that does not mean that potential technical failures were not considered prior to that press statement however. We know they brought up the avionics compartment - all such a statement would mean is that they didn't find anything wrong besides the pitot tubes. It doesn't therefore follow that they didn't check.

What we do have is:

• The PNF calling out "Alternate Law" - which means ECAM was working
• The Captain indicating all three ADIs (horizon) - which means they were working
• Multiple altitude call-outs during the descent - which means the altimeters were working

This would seem to indicate that the idea they were in a dead flight deck with no displays is a non-starter.

I don't care what airplane you fly but if all your airspeed went away in the clouds with a little turbulence and everything else worked like altimiters, VSI, attitude indicators and engine power indications would it be a major problem? I didn't think so. Why them? Hands on flying maybe?

Doze...

This.... In short, in the unlikely event there's a failure of the system and a concurrent entry into stall warning territory, then the control law will degrade, but a value that triggers stall warning by itself will not trigger law degradation as long as the system checks out as operating normally.

At what point then do you consider the a/c STALLED? It changes LAW to accommodate the STALL? How tight does Mr.Pilot have to play it? Should he wait, so that Mr airbus can still claim, "Won't STALL...." In Normal Law

Here....

The PNF calling out "Alternate Law" - which means ECAM was working
The Captain indicating all three ADIs (horizon) - which means they were working
Multiple altitude call-outs during the descent - which means the altimeters were working


No, he may have called Alternate Law because he knew without speeds, Normal goes away

Probably, but you still have to admit, they may have been working, and wrong....

Multiple altitude callouts. from Altimeter? Again, working and congruent, but....

And In a more serious setting than an internet forum, a statement like that would be borderline slanderous. Do you have any evidence for this supposition? Or is it more likely that they *did* analyze the technical aspects and found that no technical failure matched the evidence?

The St Elmo's fire is not analyzed, the PNF statement is given as "Proof"

The additional heat in the cockpit is not enlarged,

The Smell is allowed to be proven by PNF again, a known authority on Nasal powers...

"What's that". Again the proof is a good guess...

An analysis of prior events re: no record of RHS....


BEA is relying on PPRUNE to vett their analysis. I think their ploy was a good one...

One gigantic leap of faith in an audience that will accept wool as an eye covering...

Lack of high-altitude manual handling training is almost certainly a factor, but the sad truth is that no matter how good you are, sometimes it just isn't your day.

The kind of Stall Warning I was referring to was a transient warning - on the order of a second or two at most. In such an instance the aircraft is not stalled and is not in any danger of stalling, but because of the pre-emptive nature of the warning, it sounds regardless. In many respects it's like the "G"-induced warning that sounded after the PF pulled the nose up to 15 degrees - the aircraft was actually a long way from reaching approach to stall, but because of the abrupt shift in data the warning sounded. Surely in this case it's better to design the systems to be over-cautious rather than the opposite?

Highly unlikely - he was monitoring pilot and knew to go to ECAM at the first sign of a problem. Given that he did not immediately take control away from the PF in light of the inappropriate attitude, it's probably reasonable to assume that he was not especially well-versed in the consequences of a drop to Alternate.

The chances of that are exceptionally slim given the evidence at hand.

Ah, the STALL warning is explained...


"No Harm, No foul..".


Really? Transient... K.......

The first Stall was real... And transient. "Oh oh where's the Stall bug?" Your readiness to dismiss the Warning is presumptuous, no offense. If I hear the Stall Warn, and I no longer have a Vsw prompt, well, I may be concerned. Perhaps especially if I fear Overspeed at the same time, as has happened before?

The explanations are long, and legend, with precious little concurrence even between and among the pros?... The pilots had seconds... Are you not a little embarrassed?

The first warning was not. This is not coming from me, this is coming from the TRE on the sim day, who was well-versed in the event.

Hey, I'm just repeating what I was told by a veteran pilot. And the evidence indicates that the actual AoA of the aircraft was not consistent with stall at that point.

Actually, most of the "pros" that I trust tend to agree on the issue.

Remember to read one more time the report before spreading more erroneous information.

Is it again something of your own making or you will produce a Reference this time ?

Who's talking about technical failure ?
As designed, the autotrim helped to put deeper the aircraft into the stall ... Which manufacturer would like to see such brilliant design discussed widely on the public place ... ?

The first warning was real and should have been longer, dixit the BEA.
I thought your faith in the Institution had no limit ... but now a 'veteran pilot' would make you doubt ... ?

Lyman blurted out:
I'm sorry, but..................

ROFL!!!


:}

I've always thought Airbus speed indications to be very sensitive.(almost like the hole in the front of the pitot is too large) so the sudden onset of turbulence would give very rapid reading changes. Interestingly we got into some rough stuff (a320)at 37000 over the alps the other night. Air speeds were fluctuating rapidly+15 to -25 kts, with the speed trend arrows going off the scale. Whilst the auto (guidance) pilot stayed engaged and auto thrust did not chase the speed.
But whilst the turbulence was quite sharp edged I would say only moderate but the speed indications would have had you believe the aircraft was out of control. Obviously it was not but the speed indications looked horrendous / stupid. In my mind this would have added the confusion to the crew after the uas event.
(note I have more hrs on Boeings product than AB ( quite experienced on both )), & find the Boeings airspeed indications more damped than the AB)
Whilst 16 pitch at height would look totally un-real I suspect this crew were totally confused by ALL the indications & warnings. Imagine the Capt coming back in and seeing hearing all that, would he as well be confused thinking 16 up going down noise warnings coming & going that can't be right "what has happened to the flight instruments they are reading crazy they must be at fault."
Humans are not good at multiple shock analysis coupled with fear & sceptasizm.

My mistake - it was water from the cleaning freezing inside the AoA vanes. I'm sure I remembered reading "static port" in Flight or something though...

I believe the old Approach To Stall procedure was in the FCOM Supplementary Procedures - not an ideal place for a memory item, certainly! I have it on good authority that the new procedure has been added to the QRH.

Are we talking about the same warning? I'm talking about the single "blip" on the traces at about 02:10:08 - that was a single "G"-induced warning, and the aircraft was not about to stall at that point. The first non-transient warning (where the aircraft was actually approaching stall) began at approx. 02:10:48 according to the traces.

You clearly don't know me well - it's not about having "faith in [any] Institution", but a desire for factual accuracy and an intense distaste for politically-motivated conspiracy theories.

The BEA's only job is to establish the relevant facts - you can bet that one of the ambulance-chasers will try to make hay out of the autotrim feature. After all, the biggest source of misinformation as it applies to aviation accidents tends to come from lawyers releasing scuttlebutt to the press.

And yes, I'll take a veteran pilot's word, but not if they're irredeemably arrogant (Asseline) or clearly delusional (Jacquet/Marnet-Cornus).

DozyWannabe, #901

I brought my two lads up to question everything, especially the
things they trust the most :-). You seem to be basing absolute
trust in the BEA, when of course, everything they say has political
/ legal implications and will be phrased in a particular way. I'm
not saying that anything they say in the report is innaccurate, or
that they haven't done a thorough job, but they above all should be
subject to the the most rigorous scrutiny. Not only for what they
say but in particular, what they don't.

With respect, to be truly impartial, you need to question a lot more,
rather than appearing as an apologist at every turn :-)...

Originally Posted by DozyWannabe
ConfitureDozyWannabe
DozyWannabe, why do you still reference your statements to outside sources like your TRE from the sim, when it is all written in the BEA final report?

BEA Final report 1.16.3.2 Analysis of the operation of the stall warning

Then you misunderstand me. I am very much a believer in the "question everything" ethos. However, where a lot of people seem to stop is at questioning apparent authority and leaving it there. I'm also inclined to "question the questioners" as it were. It's by adhering to the first while neglecting the second that the really dumb conspiracy theories propagate - from claims that the moon landings were faked to all the 9/11 bobbins, and yes - the idea that the BEA are somehow politically motivated to protect Airbus.

Based on everything I've read over the years it would seem that of all the players, the BEA are the least likely to be politically motivated - their reports tend to be very dry and factual on the whole, not unlike the UK AAIB (both being effectively departments within the relevant civil service, albeit independent from the regulatory office). It is the lawyers in the civil and criminal proceedings who seem to be the real political animals, aided and abetted by a press who have an insatiable appetite for a juicy scandal.

It is through this alliance that most of the rumour surrounding these cases seems to have been established, and on thorough investigation, few if any of these rumours turn out to be credible - however they've done their work, as the rumours tend to be propagated years or even decades later

No accident investigation report can print things that have not been proven, nor do they tend to print aspects of the investigation that turned out to be dead ends at the end of the investigative/experimental work - that's true of all the major agencies in the world and always has been.

I'm a little offended at that insinuation to be honest - I've explained my take on questioning and impartiality above.

The fact is that the motivation behind a lot of the "questions" on here is based on incorrect information relating to the BEA, and a quixotic desire to hurt Airbus because of a misunderstanding of the motivation behind the introduction of FBW - as well as an incident more than two decades ago that didn't go their way.

[EDIT : In addition, those clamouring for the BEA to release information on aspects of the investigation that led to potential causes being ruled out seem to forget that this information is not released within the reports themselves, but tends to come in the years following, from interviews conducted with the investigators for books like MacArthur Job's - or for TV/film purposes. I noted a while back that the Mayday/ACI series on NatGeo has blocked it's final show of the 2012 season to be on AF447 - could be interesting, even if it's a high-level overview. ]

Stall warning not real?
 

Figure 62 of the final report shows what happened: Between 2:10:10 and 2:10:19 the AoA exceeded the stall warning threshold three times. Alpha-max is approximately 1 degree higher than the SW threshold and was exceeded at least once. I know that alpha-max in not alpha-stall, it is where pre-stall buffet starts. I don't know the stall AoA at M=0.8, do you? Does your TRE know?

HN39 - that's sort-of what I was saying. The threshold was exceeded in a transient sense (due to "G" loading and possibly turbulence), but the aircraft was not in danger of actually approaching stall until the apex of the climb.

The point I was trying to make was that while the warning was caused by exceeding the threshold, the aircraft was not in danger of stalling at that point - I was trying to head off another flight of fancy.

The airflow over the wing separates when the AoA exceeds a critical value, which may be transient or not.

Which is why (quite correctly) the warning does not differentiate between transient and non-transient. However because we have the full DFDR traces we can say with a measure of certainty at which points during the incident timeline the aircraft was approaching stall and stalled and when it was not.

I think this can be explained in more shades
The BEA is certainly not politically motivated to protect anyone
By cons .. number of politicians (or government) are often highly motivated to protect the interests of the country (economy .. defense .. prestige .. etc)
It should probably not be recalled here that the BEA is a body under the french Ministry of Transport (sous tutelle)
In fact .. Mr Troadec (chief of staff BEA) and all are civil servants
If one day he (Troadec) would disagree with the Ministry of Transport .. the only way he could know would be to resign

@jcj:

But civil servants are not politicians - by and large they just do the job that they have been asked to do. Accident investigation has always been something of a unique case in that regard, because the purpose must always be to find what went wrong and prevent it from happening again.

The one time in living memory when it was proven that an aviation authority colluded with politicians and a manufacturer to soft-pedal design problems (overriding the accident investigators in the process), it ended up blowing up in their faces, and condemning the airliner involved (the DC-10) to a chequered reputation in perpetuity. To do so is clearly not worth the risk.

Hi,

DW
This is ignoring the fact that politicians can take risks at any price with no fear
Courtesy .. with the help of the press ... see under ..
For the frenchs .. just remind the Rainbow Warrior case :rolleyes:
They made it before .. why no make it after ... ( this is actually some cases pending ... )

Like I said - because in the long run it did more harm to their reputation than good.

(And you're talking about a military operation, which is a far cry from accident investigation)

I'm reminded of an anecdote I once read that took place during the Comet investigation - no less a politician than then-PM Winston Churchill (who I have no special regard for, but respect his tenacity) asked for the ambassador to Italy, only to be told he was off 'showing the flag'. Churchill's response was along the lines of "Tell him to get back at once and sort out this Comet problem, or there won't be a flag to wave!".

HazelNuts39

Was there any evidence of preStall buffet with the first SWs? I am wondering if it is at all possible that PNF had felt or heard buffet consonant with the warning? I am willing to agree his comment was related to the SW only, but is it eliminated in the traces it could be something else? I think the BEA have left this open?

Thanks HN

And we return to the question ... (However because we have the full DFDR)
Do the pilots had the full DFDR unfolding under their eyes ... ?
And if he had been .. do they have had time to analyze it?
How pilots could have known the difference ?
This shows the gap that exists between an analysis of the actions (pilots and plane) ... and actions in real time
Reconciling the two is difficult
PS
BTW it's unfortunate that the full DFDR (listing) is not annexed to the BEA report

@jcj:

It doesn't matter in this case, because even if the transient warnings were acknowledged as a sign of imminent stall, to pull the nose up to 15 degrees and hold it there is not and never has been the correct response.

There is no evidence of that in the traces available. It is also highly improbable that the pilots would have perceived it, given the small magnitude and duration of buffet onset exceedence and masking by turbulence.

But that is not what BEA is saying. See my bolding.

BEA Final report 1.16.3.2 Analysis of the operation of the stall warning
BEA said, it´s neither spurious and it should have lasted longer. The warning threshold increased due to the false low mach (frozen pitots), which made the warning stop. What information do you both have, which contradicts the BEA final report? Do i understand the report wrong?

@franzl

We're talking at cross-purposes here. The BEA say they should have sounded for longer based on the data they have - I'm not disagreeing with that. The nature of that warning seems to be that it would have stopped during the brief period the nose starts to come down after the initial 15 degree pitch-up. Had 15 degrees been held it would have definitely approached actual stall earlier. However. after bringing it back down for a few seconds the PF pulled the nose up again - to just shy of 20 degrees this time, and the stall warning sounded continuously until the indicated speed bled off to the AoA vane cutoff point (i.e. <60kts).

Note that the BEA are referring to transient bursts of the stall warning which do not correlate with the DFDR graphs. The transient warning I was referring to does indeed appear on the graph.

@DW
I do not agree, please recheck.
BEA is talking about the identical two warnings in the same timeframe. It adresses the logic of the stall warning, and that it should have sounded longer than it did and three times instead of two times. And it didn´t sound that long, because the threshold, at which the stall warning is activated, increased to about 10° AOA (which is not depicted in the FDR- trace with the green line) due to the abnormal (not actual) Mach number. The threshold is lower at high mach, higher at lower mach. As mach was wrong due to the iced pitots, the warning threshold was set too high and therefore the warning sounded too short.

If you compare the mentioned times from BEA with the figure 62 of Final report, you will see that the warning sounded shorter than depicted looking like spurious and thus not understood by the crew. And that is exactly what BEA states under 1.16.3.2 Analysis of the operation of the stall warning.
Why the wrong movement of the threshold like mentioned by BEA is not depicted in Figure 62 FDR traces with the green line, i dont know. But the faulty mach is depicted above. Maybe this line has a different origin and is not mach dependent.

The warning was valid, however it was not presented to the crew in the apropriate long time frame due to the iced pitots.

I'm not disputing that, I was simply simply stating that in my admittedly limited experience we had a similar transient warning without iced-over pitot tubes (which are not part of the sim logic). The behaviour uncovered by the BEA certainly warrants looking at from a technical viewpoint

Nevertheless it does not alter the fact that it appears the crew did not respond in the correct manner to either the transient or sustained stall warnings, instead keeping the nose up.

for a little more science :rolleyes::
MIT OpenCourseWare | Aeronautics and Astronautics | 16.333 Aircraft Stability and Control, Fall 2004 | Lecture Notes