Lyman, TTex600;
Well, we know what documents are available that deal with the UAS and ADR abnormals. We don't yet know what's trained. I referenced some documents in a post to Flyinheavy, on June 1: http://www.pprune.org/tech-log/48235...ml#post7222304 .

So we know at least, that the information to deal with the UAS event and/or failed ADRs was available and provided very specific guidance on how to do the memory items, the checklist items and what the thinking was behind the 2006 update. The thinking is clear, but again we don't know what the training actually was with reference to these documents. Also, these documents may not be available to all operators. They are not "required" for action in the same sense as, say, an AD.

O.C.

I thought about your response and agreed with it because to pilots it makes sense..."If not the UAS drill/checklist specifically, why not the standard SOPs at least?"

To be clinically accurate with reference to alf5071h's point concerning hindsight bias, we still have hints of the statement, "Why didn't they stick to SOPs?" One way to examine the question is, "How far back do we go before making up our minds as to what happened?" Can we ever make up our minds and say?


The other side of this same question is here, in PPRuNe: We have spent nine threads and three years and are still unable to say why, and we remain unable to say why the aircraft was pitched up and, more importantly, why it was held there when all of us who fly transports know that the airplane is going to run out of energy with the pitch attitudes recorded in the data. We know this to be true and expect that others who do this work would know too. How do we sort that out so that realistic, preventative action may take place? Or is hindsight bias increasingly serving the courts?

Acknowledging the phenomenon of hindsight bias still permits learning and change from "mistakes" but not because of the assumptions we may about such (assumed) "mistakes" or pilot behaviour which we can read in the tiny little bits of data we have from the recorders. How we learn is perhaps captured in the statement we're all familiar with: "The crew did not wake up that morning intending to have an accident." We might even extend this to organizational thinking.

From Dekker's, Drift Into Failure:

"The idea of the amoral calculator, of course, works only if we can prove that people knew, or could reasonably have known, that things were going to go wrong as a result of their decisions. Since the 1970's, we have 'proven' the time and again in accident inquiries (for which the public costs have risen sharply since the 1970's) and courts of law. Our conclusions are most often that bad or miscreant people made amoral trade-offs, that they didn't invest enough effort, or that they were negligent in their understanding of how their own system worked. Such findings not only instantiate, but keep reproducing the Newtonian-Cartesian logic that is so common-sense to us. We hardly see it anymore, it has become almost transparent. Our activities in the wake of failure are steeped in the language of this worldview: Accident inquiries are supposed to return probable 'causes.' The people who participate in them are expected by media and industry to explain themselves and their work in terms of broken parts (we have found what was wrong: here it is). Even so-called 'systemic' accident models serve as a vehicle to find broken parts, though higher upstream, away from the sharp end (deficient supervision, insufficient leadership). In courts, we argue that people could reasonably have foreseen harm, and that harm was indeed 'caused' by their action or omission. We couple assessments of the extent of negligence, or the depth of the moral depravity of people's decisions, to the size of the outcome. If the outcome was worse (more oil leakage, more dead bodies), then the actions that led up to it must have been really, really bad. The fine gets higher, the prison sentence longer.

t is not, of course, that applying this family of explanations leads to results that are simply false. That would be an unsustainable and useless position to take. If the worldview behind these explanations remains invisible to us, however, we will never be able to discover just how it influecnes our own rationalities. We will not be able to question it, nor our own assumptions. We might simply assume that this is the only to look at the world. And that is a severe restriction, a restriction that matters. Applying this worldview, after all, leads to particular results."

- Dekker, Sidney. Drift Into Failure, Surrey, Ashgate, 2011, p.5-6

PJ2 done wrote:
This puts me right back to the simplistic thinking (and sorry to snip the rest of your totally germane post), that there HAD to have been something wrong with the displays. It's the only answer that makes sense to me. Data has been withheld in the interim reports, as has been stated here many times.
Go ahead, shoot me down; I don't care; I can't fall very far from my armchair. :8

Organfreak;

Regardingf what was on the PF's PFD/ND displays, it has been argued before, but inconclusively, like the argument that the PF was following the FDs. We just don't know and there is no data which tells us that this is what occurred.

What's more, once open, those kinds of doors swing both ways - as much as one may be totally convinced that one's notion is correct regarding what occurred and why, someone else may posit an entirely reasonable counterexample and there you are...mere theory, no fact, no conclusion or worse, dangerous conclusion(s).

We are not paralyzed; we are in a new "evidence space" which directs responses differently.

Well PJ2, I fully realize there are no facts or evidence to support-- I'm indulging myself in a completely groundless speculation, just for once, because mindless certainty feels so much better than the unknown. :)

Yes, it does, until it doesn't! ;-)

The notion of "amoral calculation" was used by Vaughn in her book on Challenger. She slowly shifted in her view that NASA engineers and managers intentionally placed the shuttle and the crew at risk due to production pressures and long-term cost-controls. She eventually concluded that the very opposite was occurring...that everyone thought that they were doing exactly the right thing.

I think we can take it that the PF thought the same thing. I am hoping that the BEA HF Group can do for the crew of AF447 and all those affected, the same thing that Vaughn did for the crew of Challenger.

The hard data is in the ACARS...The crew were not privy. Neither were they privy to data supplied for the post mortem via the recorders. Unwinding a little bit of the conclusionary tone here, Organfreak has repositioned the discussion.

Unless and until the complete record is available, even the BEA report will need a salt shaker.

PJ2:

"....say why the aircraft was pitched up and, more importantly, why it was held there when all of us who fly transports know that the airplane is going to run out of energy with the pitch attitudes recorded in the data."

Without knowing the state of the displays, the rhetoric has the crew in the corner, perhaps forever. It remains to be seen that the conditions were benign re: recovery. By benign I mean framed so by those who would condemn......

Re: Vaughn/Challenger. I never thought the Challenger case due malevolence. The incident happened post launch, thus inviting hindsight bias from the outset.

What I do think, is that Staff made a decision that involved risk management, out of their job description, and without proper oversight. Doubt can save the poodle, or kill it....In the Challenger case, the structural process was abdicated/co-opted, by those who ignored the safeguards. Negligent? Of course....

It is chilling to think that these situations can crop up just as they did when we did not know "better".....

Trim, a technical question.
 

When considering the differences between events, some of the initial climbs were with autopilot engaged, AF447 was not.
Is there any difference in the pitch trim rate between autopilot engaged and manual flight, between normal and alternate law, or between any of the previous options and that which might be seen with over-speed protection?

Lyman,
Strictly by accident, I assure you. I doubt I have that power. Sorry if I was rude to you last week, even if you did deserve it. :O

Safetypee, great handle!
It is my understanding, as discussed in thread #s Gazillion, that this particular flavor of ALT Law included full auto-pitch trim. Apparently, to catch you up, it did exactly as PF commanded, which was to wind-up to almost full nose-up. Doubtful if this helped. It has been opined by somebody here that there's no good reason for auto-trim in ALT2B, and probably added to the confusion.

Organfreak, thanks. My line of thought is if there were to be a difference in trim rates, then the manual-flight trim-follow-up could result in a more extreme attitude (or lower trimmed airspeed) than for autopilot/autotrim for the same duration; where duration might be the satisfaction of a flight guidance (FD) command.

An alternative view of this might be any difference in the flight control pitch rate limit between manual and autoflight (or over speed pull-up); expecting that manual flight would enable higher pitch rates than for the autopilot.

Hi

Answering here to A-FLOOR, from the thread "Humbling sim experience", in an effort not to hijack it for too long ;)

Re: Overspeed and understanding of stall situation in AF447
Yes I did, thank you. En français dans le texte, qui plus est. :)
1/ At one time, the PF though the aircraft was (too?) speedy ("J’ai l’impression qu’on a une vitesse de fou non qu’est-ce que vous en pensez ?") but the aircraft was not. As the overspeed protection listen to its sensor inputs, and not to the PF mind, I maintain: no overspeed detected, hence no activation of overspeed protection.
2/ I fail to see where the transcript shows the crew was aware of the stall. OTOH, there are parts which show they weren't ("comment ça se fait qu’on continue à descendre à fond là ?")

Releasing the back pressure on the stick may be not enough, as the FBW will then try to maintain path/g load. If the stick is placed and held nose down, the nose will drop. It takes more time as long as the THS full nose-up, but it works, and if maintained the THS will come back from its "full nose-up" position.
NB: I do however agree that it's not a good idea for the auto-trim to go full nose-up when the SW is on (point discussed at lengh in one of AF447's dedicated thread).

OK, wasn't aware of that. Thx. :)

CVR transcript
 

Yes, of course.

Three years after the horrific event, and despite the fantastic PPRuNe members knowledge, F-GZCP wreckage remains incomprehensible.


IMO, here is a very interesting link where you can see both CVR and FDR transcripts.

I have printed it (3 pages needed to see some data), and also IR#3 (En) pages 29-30-31 (and take a pencil to add some "zero lines" or other line ...)


Regards

Thanks, now we found out it is Airbus chief test pilot, mr Rosay, could now someone please translate what he actually said, for the benefit of us not understanding French?

I'll refrain from definitive judgment, as I don't feel qualified enough to draw the line of normality. However, on normality scale, finding UAS so ordinary to not even report it is far, far more normal than trying to climb the aeroplane above its ceiling just because airspeed indication got messed up.

How would anyone know what the rules were? The data is de-identified, you don't know whether there was safety management system with compulsory/confidential reporting scheme in the place at the time of the occurrences, let alone what airline did the alleged offenders fly for.

Rest of your post is completely irrelevant to AF447; it refers to earlier replacement of Rosemount probes with either Goodrich or Thales. Yup, Thales probes were found to be superior to Rosemount's.

This statement of yours is radically at odds with the truth, as what I mentioned is clearly stated on page 65 of interim2. Interim doesn't mean it is not official. Just not final. Yet.

The airline is not identified in the report. Your notion it's AF might eventually turn out to be correct but for the time being it's pure conjecture.

Anyone suffering from this condition can be quickly and efficiently cured by gazing upon interim1, page 69 or interim3, page 59.

In the article about spatial disorientation, published in internal air force magazine, some 23 years ago (if you wonder: my father was reserve captain in the army that started to crumble at the time so he could get away with sneaking away a few issues of GRViPVO to his aeroplane-crazy son). There was nice illustration, allegedly based on radar plot, of MiG-15 UTI entering the cloud, reversing its heading and completing two full spirals before hitting the ground. Point of the article, with which I fully concur, was that spatial disorientation can happen even to the best and most experienced pilots. That former fighter pilot can be offended by the suggestion is quite surprising to me. Frankly, I find it even a bit appalling.

So how does it fit in with your "We are talking about trained airline pilots here" umbrage? He was astronaut, far better trained than any airline jockey, yet he died making what uninformed call "beginner's mistake".

Sure it is. It's my opinion based on DFDR readouts, precisely rendered on pages 29-31 of interim3. Where do you find fault in it?

Again, I find such a notion coming form an carrier qualified ex-phantom driver a bit disturbing. Aeroplane doesn't change a bit, it is the unfortunate pilot that gets into phase with its oscillation.

Anyway, maximum roll before the aeroplane stalled was 11°, which can be called significant only by severe stretch of imagination. However, from DFDR data it is pretty clear that CM2 actively and successfully dampened the roll, therefore he was very aware of the aeroplane's bank so we can lay the theory "he did not see instruments" to rest.

Please everyone: read the following paragraph carefully. Reread it if it's not clear after the first reading. Come back and ask questions if still unsure you understood it. Refer to interim2 for more information.

There were more than thirty cases of UAS similar to AF447 on A330 and A340 aeroplanes. Almost all of them included autopilot disconnect and reversion to alternate law. All of them ended without damage to aeroplane or injury to passengers. Airlines involved are not explicitly stated, yet from the date of the occurrence and MSN of the aeroplane it can be easily determined. There were some incidents involving the Air France.

Therefore:
Notion that unreliable airspeed must be fatal on Airbus is false.
Notion that most of the contemporary Airbus pilots are unable to handfly the beast in alternate law at cruise level is false.
Notion that every Air France pilot is so ill prepared for the UAS on 330/340 they are bound to crash if and when it occurs is false.

If you were to guess only based on CVR, it would be pretty reasonable thing to assume but there is no need to do that as DFDR data is available. No electronic or mechanical failure was detected. Yes, you have read it correctly: no failure! Sensors kept on measuring the pressure which was no longer actual total, as pitot orifices were blocked by ice and that's what cascaded into AP loss and reversion to ALTN2. Per design and a good one, too.

That's something both CM2 and CM1 forgot just before the accident. I'm speculating but perhaps they were too knackered to think straight?

Not just that. We can be pretty certain that CM2 strongly believed he was saving himself and everyone on board from a threat that he unfortunately did not name for the sake of CVR. That his actions were just the opposite of needed is beyond tragic.

You must be privy to a CVR version somewhat different to that published by the BEA.:confused:

The Captain did say at 2 h 11 min 52, "Well look take take that".

So why persist that the crew knew the aircraft was stalled, when there is no supporting evidence.:ugh:

Could anybody explain the specific difference(s) between ALT2"A" (I assume there is such a thing) and ALT2B?

Standing by...:)

mm43

Not sure re: CVR, but it was reported, among other comments, that "I understand nothing", "We have crazy speed", AND (from Captain, on return) "What are you doing, this is a Stall".

The latter did not end up in the CVR releases, but a lot of other stuff may not have as well. "Er.. What are you doing" did make the report. For now, I reject the comment re; STALL, however, unless the entirety of the audio is released, we will not know, ever. Likewise, I am unclear as to why CONFiture's comment is unaddressed: "Vmode trace" not in the record.

Not at all .. this show how much earlier (far before the AF447 event) was the preoccupation about the Pitot tube problem and it's possible consequences ....
I want to remind you what you posted
clandestino
And what I posted ...
jcjeant
Yes .. indeed .. despite all the earlier preoccupations .. it was unluckely no rush ......
clandestino
In the video .. the interviewer ask Mr Rosay .. what the AF447 pilots had to take as action when the event occured
His aswer was .... do nothing !
Clandestino
Gagarin was not really an experienced pilot ... (check his flight hours ...)
And when he was incorporated in the space program .. he fly very little time .. (or not at all) just for keep up to date his pilot papers (license)

Reference
Starman by Jamie Doran and Piers Bizony

@OK465,

Heh heh, thought I had read of it here, so I parroted it. Looking at the FCOM, I see nothing of the sort. I sit, corrected. :\

OF:

No need for you to sit corrected. The IR refers to ALT2B.

And you are correct about the other documentation.

That's the point. :)

Lyman,

Where did "this is a stall" come from? I don't think I have seen this anywhere. IMO, what seemed obvious, the Captain never figured out what was happening upon return to the cockpit.

ALT 2B, or not to be?
 

OK465,
WOW! That is damned interesting.

Here are the three pertinent pages of A330 FCOM, and sorry for the squashing of the images- (ah...the Mysteries of Photshop).

http://www.organfreak.com/images/A330controllaws1.jpg

http://www.organfreak.com/images/A330controllaws2.jpg

http://www.organfreak.com/images/A330controllaws3.jpg

@ LymanThe BEA have made it quite clear that comments made that were not relevant to the conduct of the flight were eliminated from the CVR transcript. If the stall warning had been acknowledged verbally in any form, or some speech construct inferring the like, then the BEA would be less than honest not to include it. I have no reason to suspect that either was the case.:=

@ OK465

It would seem that the major differences in ALT Laws are:-

ALT1:
Pitch attitude (Θ) protection lost.
Hi Speed and Stall speed are alternate.

ALT1A:
As ALT1 but Stall protection is lost.

ALT2:
As ALT1 but lateral normal law is lost and replaced by
lateral alternate (Roll = DIRECT; Yaw = Alternate)

ALT2A:
As ALT2 (Stall protection is lost)

ALT2B:
Pitch attitude (Θ) protection lost.
Hi Speed and Stall protection lost.
Bank angle protection lost

Source - A33Zab in AF447 Thread No.6, and in

ALT2B no Hi and Lo speed stability (= VMO2/ Vc prot),

as explained earlier in this thread.

There is enough data to affirm FD were displayed at times. So there are more than obvious reasons to investigate what those FD were displaying for those periods. Everybody is questioning why the PF did what he did, the answer could be just there ...

Are there pilots that follow the flight director blindly? Didn't happen when I was flying.

1. pulling can lead to high-mach-low-threshold transient stall warning if aeroplane is jerked into climb, reasonably smooth pull from about 1.5 to 5° will avoid it
Easy stuff in turbulence

2. when settled at 5° pitch, AoA will be near cruise pitch and will gently increase towards 5° as speed is bled off. By the time it gets there, you won't be at high mach anymore.
Easy stuff in turbulence

3. If no valid mach, stall warning reverts to low threshold
I like that ''If''

4. stall warning is not stall itself - a fact cheerfully ignored by those unable to tell the difference between "approach to stall recovery" and "stall recovery".
It does not matter, both should trigger the very same procedure (the new one) NOSE PITCH DOWN + THRUST REDUCE

5. when arguing about Habsheim showoff, you repeatedly claimed that alpha prot prevented aeroplane from achieving higher lift at even higher AoA (backside of the power curve, anyone?) and now all of a sudden, stall warning is considered to be the edge? Are you having it both ways or are you about to make a breakthrough in aerodynamics - discovery of the area of fantastic aerodynamic performance between alpha max and stall warning?
Because Alpha Max is simply not Alpha Stall Warning and Airbus did that for a reason.
Alpha Max = protection = do whatever you like = still flying
Alpha Stall Warning = Do NOT stay there = still flying but for how much longer ?

Airbus chief pilot has got all of it now : No more 5 degrees pitch up - Finished : "When you lose the speed indications in cruise, that is the most simple procedure to apply : You have to do NOTHING"

Clandestino, it could be understandable you recommend to apply 5 deg of pitch up as a first measure the time to get some figures from the QRH, but it is absolutely non sense you pretend to be comfortable to apply it vitam eternam. It tells a lot how you see yourself as a pilot ... as it shows in your writing style too.

Thanks for reposting this list.

When BEA and others like Clandestino state, that other UAS events had a positive outcome, can the difference to AF447 partially be found in this list? BEA does not state in which kind of Alternate Law the other events ended, i couldn´t find it at least. So all those events could be just in simple ALT 1, quite a difference to the ALT2B AF447 dropped to.

Hi Lyman

Re: your #1215

• I don't remember releasing "the entirety of the audio" is a standart practice (source: other accidents). More the contrary, in fact. In that light, presenting this release as "waited" as you do will only lead to disapointment/rise in conspiration theories when this release does not occur.
• I'm aware of -at present time- two versions of the transcript:
  - the official version (by the BEA, in IR3)
  - the version "leaked" by Mr Otelli in his book, build from an earlier transcript of the CVR than that used by the BEA in IR3, as far as I can tell.
  Even if both versions have been quoted numerous on times by various media, I'm not aware of any version/sub-version that contains the apparent recognition of the stall by any of the crew member. Would you be so kind as to provide a link to that version/sub-version/comment, in order for its credibility to be assessed?

Thanks :)

Right! So where have we got to:

Either do nothing or -
- 5° pitch up will not stall the airplane
- pull up gently to avoid stall warning
- if stall warning occurs: reduce pull up
- use small, brief stick inputs, watch the result before making further inputs
- get the QRH for correct pitch and thrust setting

This is a Stall
 

AZR:

Doomed Flight AF 447: An Airbus Programming Error? - SPIEGEL ONLINE


"Exactly what orders he issued are not part of last Friday's report. But sources close to the investigation are saying that he said: "This is a stall. Reduce power and nose down!""

No comment on credibility of article, reporter, source, etc.

MM43:

Thanks for re-posting the list.

I respect A33Zab's obvious knowledge, but that list leaves a lot to be desired and has some flaws of logic and terminology in it if you were to use it for training, i.e.

...as if stall 'protection' (wrong term) is available in ALT2, but not lost until reconfig to ALT2A?? In any case both stall 'protection' & theta protection take a hike in all laws other than normal, so 1A, 2A are what exactly?? Just vanilla 1 & 2??

My question regards the difference in failure modes that would create one or the other level of reconfig. I would have thought that 2A & 2B are a function of how many ADR's get flaky, i.e.

1 flaky ADR = continued Normal Law
2 flaky ADR = ALT2A
3 flaky ADR = ALT2B

I guess I'd like to know how do I then get into each of these conditions and then how do I recognize which one I have, which ones 'latch', which can be recovered to a different level of reconfig, or do I even need to know any of this?

Simulators generally only provide a limited range of constrained failures to achieve one or the other reconfiguration level for training, so does the available documentation need to be enhanced?

A33Zab is an excellent source of info, but there's probably a lot of A330 pilots who don't even know he exists. :)

B44:

I trained a whole lot of pilots in the 727, and yes there were some who could get caught by this, even back in 'your day'.

COnf:
Conf, I will presume that your point is for stall at high altitude, or for an aircraft with underslung engines, since
At high alt/high mach, overspeed can be a risk
Reducing power on underslung engines removes some of the "pitch up" input from thrust vector being below, as opposed to coincident with, the longitudinal axis.

As I ponder the point you make, it seems a simpler response to lower the nose, as doing so you will decrease AoA without touching power.
However, if you reduce power => you reduce thrust => decrease airspeed / Mach ... so ... might your power reduction risk ofsetting a nose reduction and thus keep you near stall AoA if you are trying to avoid stall?

IS part of the nose down/power reduce response intended to cause an altitude reduction and thus change the performance environment of the aircraft?

In chewing over your point, I conclude that your recommendation is confined to approach to stall/stall in the high altitude / high mach environment.

Do I understand you correctly?

thermalsniffer

Thanks for the link. :ok:
I forgot about that... errr... is that a novel? ;)

Well, a simple thing:

"At no point" on the cockpit voice recorder "is the word stall ever mentioned," Chief Investigator Alain Bouillard said in an interview.

So, as far as I'm concerned, source is dismissed. Thanks again. :)

I'm somewhat surprised at the comments that the crew never mentioned being in a stall at all, as this is exactly how I remember it from reading an early report about the CVR transcript (capt telling his crew they are in a stall shortly after he returned to the FD). This is exactly one of the things which really stuck with me from reading that, so I'm dumbfounded how it ended up not being mentioned at all in the interim report.

Surely there must be other people who remember reading it as well?

Nope :=

In fact that "stall" was never mentioned was what stuck in my mind.

@A-FLOOR:

Well....no, sorry. That's the very first time I've heard that one, after following these threads closely since 6/09. This Der Spiegel article is full of other mistakes, misrepresentations, and inaccuracies as well. Forget it.

We've spent three years here trying to figure out how they didn't know they were stalled. (Not to mention why they pulled up, so I won't mention it.)

A330 C*
 

Has anybody some information about
1. A330 feedback equations including speed parameter V ? and information about A330 cross-over velocity Vco?
C* =? Knz.nz + Kq.q and is it standard C* ?
2. relative position of the accelerometer ? in normal flight and stalling deeply,
3. where does the "speed" input in the feedback come from in the UAS case?

Thank you :ooh:

The answer was in the revised procedure that never showed up, following the known and discussed cases of Air Caraibe. Airbus acknowledged it was necessary to think about something to help a crew in such circumstances … We are still in the wait … Is it part of the recommendation chapter in the coming report … ?

Airbus initially wrote the reply to UAS in cruise is to adopt 5 deg of pitch up. The BEA confirmed such procedure in their press conference. Now, Airbus, through his chief pilot, says differently …

Big clarification up is necessary regarding UAS procedures, more specifically at high altitude. Pilots need clear procedures and relevant TRAINING !

The last time I did practice the exercise, it was just after take off and the first warning we got was : WINDSHEAR WINDSHEAR ... consistant with my airspeed going in the red.

Trial
 

The decision of the Court of Cassation of France, that was expected June 15, finally arrives September 24, 2012.
It responds to the request of the Lawyer Me Soulez-Lariviere about the jurisdiction of the French criminal courts for a disaster in the non-territorial waters.
This was the case TOTAL, but also concern the flight AF447, since Mr. Soulez is one of the lawyers defending the leaders of major French civil aviation and has already expressed his opinion hostile to that jurisdiction to the crash of Mont Ste-Odile.