Techman,

LOL, I'm inclined to agree, I was slightly baffled by that last post as well.

'Smile and nod' might be an appropriate answer.

Quoting Hyperveloce:-



Not your fault, Hyperveloce, because the BEA Note leaves so much out. But the only mention of a 'left noseup input' at any stage before the upset is this:-



The next relevant entry, referring to at least 11 seconds later, says:-

Quote:

---

At 2 h 10 min 16, the PNF said "so, we’ve lost the speeds" then "alternate law […]".

---

Had the PF maintained a 'left noseup input' throughout those eleven seconds the aeroplane would probably have been 'standing on a wing.' But the PNF appears to have been (relatively calmly) going on doing his job and reading and passing on the messages appearing on his screen. That suggests to me that the PF had in fact merely corrected a tendency to roll after the autopilot signed off, and then levelled off and followed the prescribed procedure of 'flying pitch and power.'

The next entry in the BEA note (with no time stated) says:-

Quote:

---

The airplane’s pitch attitude increased progressively beyond 10 degrees and the plane started to climb. The PF made nose-down control inputs and alternately left and right roll inputs. The vertical speed, which had reached 7,000 ft/min, dropped to 700 ft/min and the roll varied between 12 degrees right and 10 degrees left.

---

No mention of any 'nose-up inputs' - indeed, that entry strongly supports the view that the climb was 'uncommanded' and that the PF took appropriate action to counter it?

The PF is reported, much later on, as maintaining 'noseup inputs,' but that was after the stall had occurred. And, as I said earlier, I think that the most likely reason for those is that the pilots thought for quite a while that they were in a dive rather than a stall.

So IMO, as BOAC said, "The climb is still unexplained."

RWA,

Would you not agree that the nose-up request was from the pilot flying (the left part is irrelevant after had correct the roll right and not necessary for the report). Also, you are quite incorrect about 'next mention' of nose up inputs being 'quiet some time later'...do your research mate:

- 2h 10m 51s PF maintained nose up inputs;
- Around 15 second later the PF continued to make nose up inputs;

I can keep quoting, but the most important one is in the 'New Findings' section:

- The inputs made by the pilot flying were MAINLY NOSE UP.

Hyperveloce was quite justified in what he said. In this case, it is 'your fault' mate, do your research.

Control Systems
 

Greetings to you all,
Reading the last stream of posts (were I keep seeing a redundant quoting of BEA´s interim report), it came to me that it would be a nice idea to refresh some basics about reconfigurable Control Systems and a good reading while waiting for BEA´s next report.
The paper covers CS in a general way, but FCS and AFCS are also addressed.

http://users.encs.concordia.ca/~ymzhang/publications/ARC32-2-98Zhang_pp.229-252.pdf

New interim report to be published on 29 July.

Hi,

Indeed (press)
Vol Rio-Paris : rapport vendredi sur "les circonstances exactes" du crash - LeMonde.fr

And .....
Le Figaro - France : AF 447 : le rapport d'étape de l'accident publié vendredi

You can always do what I have done and put him on your Ignore List :)

Tio Pablo, thanks for that link. A bit rough going, but worth the read nonetheless.

Sorry we seem to be disagreeing so much, Poit.

Quoting Poit:-

You’ve flown light stuff, same as me. You’ll know that one almost never uses the stick in one dimension, ‘coordinate the controls’ is the rule. A bank usually entails a loss of lift and also produces a turn. My best guess is that, in order to get back on course and also maintain height, the pilot put on opposite bank and also used a touch of up elevator – and then levelled out. All perfectly normal. And, as I also said the PNF then read out and cancelled a couple of messages, in a pretty calm manner; if the PF was busy standing the aeroplane on its tail the PNF would surely have been saying different things (to say the least :))?

(In this connection, I dislike the tendency of some (particularly that newspaper article) to suggest that the PF was ‘inexperienced.’ If you look him up, you’ll find that he had the best part of 3,000 hours in his logbook, including about 850 on the A330).

Don’t recall saying that? I thought I wrote, ‘much later on’? The BEA’s (very thin) report times the ‘left noseup’ command at ‘2 h 10 min 05.’

The next ‘event’ the BEA describes is the ‘zoom climb.’ No mention of any noseup inputs before or during that; indeed it describes nosedown inputs. All the evidence is that the PF countered the climb and pretty well got the aeroplane back to level flight. The next event is timed as follows:-

I make that 46 seconds between noseup inputs? Surely that’s ‘much later on’ in the context of an accident that only took about four minutes from start to finish?

May help if I explain that I tend mentally to break this (and any other) accidents down into phases. Lacking proper information form the BEA, I currently see these as follows:-

1. ‘Signoff’:- AP/ATH disconnect, pilot takes manual control, corrects an uncommanded bank, and begins flying ‘pitch and power.’

2. ‘The climb’:- No stick input either way at first. Then nosedown to arrest the steep climb.

3. ‘The stall’ – stall warning sounds, pilot applies standard ‘stall avoidance’ procedure at that time (apply TO/GA power and seek to minimise altitude loss). Stall warning stops.

(One thing that can’t be ‘fitted in’ to the various phases, in that we don’t know when it occurred, is that at some time during one of these phases, the THS wound itself up to ‘full up.’ Typically, the BEA note refers to it taking ‘about 1 minute’ but it doesn’t say which minute. I’d appreciate other people’s opinions at to when they think it happened?).

4. ‘The free-fall’ – aeroplane begins losing height at 10,000 feet per minute. I’m on record that my own view, on present evidence, is that the pilot(s) thought that the stall avoidance had worked and that they were in a dive, not a deep stall.

5. ‘The attempted recovery’ – somewhere above 10,000 feet, power was reduced (again, the BEA doesn’t say when) and the PF applies ‘pitchdown inputs.’ Angle of attack improves and the speed indications return. BUT – the stall warning sounds again…….

6. ‘The crash’ – the aeroplane belly-flops into the Atlantic.

To the best of my knowledge, that’s all we know so far. Comments, additional points, corrections, further information from anyone all welcome.

5. 'nose up inputs' ?

Troisième rapport.
 

Hello, just read in the Var Matin that Friday 29th the BEA is coming out with a third report and a analysis. Let's hope we get some more facts. Of course I am thinking about the beloved ones of the deceased but also about the pilots who are flying these aircraft daily.

Ok here is the official announcement.
 

Le troisième rapport d'étape du BEA sera publié le vendredi 29 juillet 2011. Ce rapport présente les circonstances exactes de l'accident avec des premiers points d'analyse et de nouveaux faits établis à partir de l'exploitation des données des enregistreurs de vol.

A cette occasion, un point presse sera organisé au BEA à 14 h 30 le même jour.

Quoting Mr Optimistic_

Many thanks, Mr O, edited! :)

RWA

For the umpteenth time that is :mad:!

Give it a break your theory and misinterpretations are wrong, you are like a broken record and never reply to the posts that contradict you. You wait a few days / posts and post the same rubbish again. If you have nothing NEW to say don't say anything.:ugh::ugh:

You need to read and digest how the airbus flight controls work not how you think they work.

Quoting iceman50:-

On the contrary, I answered your last (also somewhat abusive) post, with a source, in my Post 2086 above.

Revised stall procedures centre on angle-of-attack not power

If Flightglobal (and I) are wrong, please inform us what the recommended procedure actually was back in 2009, and what changes were made recently?

- maybe have a glance at http://www.pprune.org/tech-log/41537...procedure.html ? It may be of interest to you.

BOAC

It is NOT new, it has just been re-emphasised by Airbus Boeing regulators etc as some pilots / checkers were thinking like RWA as in maintain ALTITUDE. The procedure being continually mentioned by RWA was recovery from low speed NOT a stall and even that did NOT mention MAINTAIN altitude. There again perhaps I have been teaching and checking it incorrectly for the last 10 years on the A330 / 340 and been taught stall recovery incorrectly for the 30 years before that.

For a Cessna 172 pilot, this is "experienced." For an air-transport pilot, not.

Making stuff up
 

RWA:

So where is the evidence there was any attempt at pitch and power at this point??? You have simply made this up. We are simply told there was a roll correction and nose up input. Followed by a 3000 foot climb...

There is no mention of any attempt to control power until some point later when the misguided TO/GA selection was made. At least that's the only mention of any power adjustment I can find in any segment of the BEA note until much later when power was reduced.

Hi,
Where did you get that?
There is NO such a "Standard stall avoidance procedure" for A330 due to NORMAL LAW flight envelope protections. If you want to refer to what is really relevant, you'll need to check into the "Abnormal Operations" related to ALTERNATE LAW, where stall avoidance procedures are described.

Here, you'll learn what the correct procedure is 3.04.27 page 6 (Rev 18, 2003). There is two cases: low speed and high altitude.
a) low speed: Apply TOGA and (at the same time) reduce pitch angle (angle-of-attack).
b) high altitude: Relax back pressure on sidestick (reduce angle-of-attack).

Where is then this "standard [imaginatory] procedure" you were talking about for high altitude stall warnings:
=> Apply TOGA and maintain back pressure on sidestick !?!
Both actions recorded are wrong for the situation: TOGA is not needed, neither is this pulling up.

This was the second stall warning, the first was ignored or responded by a pitch up if volontary applied. There is no trace of your:
1. ‘Signoff’:- AP/ATH disconnect, pilot takes manual control, corrects an uncommanded bank, and begins flying ‘pitch and power.’
Pitch was wrong and power was not unlocked (no trace of this in BEA narrative). You are retaining bank correction without pitch correction. Why are you making up this stuff?

Same about this:
How can you say that you don't know when THS moved to 13°?
This is obviously timmed starting at or closely after 0210:51 and then ended also close to 0211:51.

Of course you are completely wrong.
What changed was the emphasis on Pitch rather than Thrust, as large thrust increase may have an adverse effect on pitch reduction at low speed. At high altitude, it doesn't change a glitch as stall warnings would be triggered at the onset of buffet, likely at an already high Mach, without many excess thrust available, hence only a question of AOA at the limits of the safe flight envelope.
Check by yourself the relevant part of A330 FCOM in use before 2009:

http://takata1940.free.fr/stall1.jpg

I suspect that like me, none of the contributors here of late actually know what Air France taught for recovery from the approach to a stall in the A330 in those days, nor how much 'reliance' was instilled in crews in the infallibility of the AB in not 'letting you stall'. It is worth noting that unlike the impeccable Iceman, some trainers were NOT teaching things correctly - in the UK at least.. This is the first para from post#14 in my link (with my emphasis) from April 2010

A very good document from British CAA that has a lot in common with the new procedure:

Applicability: RETRE, TRIE, TRE, SFE, TRI, SFI
Effective: Immediate
STALL RECOVERY TECHNIQUE
1 Recent observations by CAA Training Inspectors have raised concerns that some instructors (both SFIs and TRIs) have been teaching inappropriate stall recovery techniques. It would appear that these instructors have been encouraging their trainees to maintain altitude during recovery from an approach to a stall. The technique that has been advised is to apply maximum power and allow the aircraft to accelerate out of this high alpha stall-warning regime. There is no mention of any requirement to reduce the angle of attack – indeed one trainee was briefed that “he may need to increase back pressure in order to maintain altitude”.


Wait until 29 July?

Hi BOAC,
Maybe a look at AF A330 documentation can give you a hint about that?
In fact, it is in English and they used Rev.24 of the procedure with three cases:
- lift off (low speed)
- flight phases after lift off (low speed)
- high altitude

This "Infallibility" legend of the Airbus "will not let you stall" is certainly not part of this aircraft documentation, check by yourself:

http://takata1940.free.fr/AFstall0.jpg
http://takata1940.free.fr/AFstall1.jpg

- That was a silly comment. Are you seriously suggesting that AB would have put those words in the manual? I would not even expect Ziegler to do that - spout it, yes, but print it? No.

What is your airline experience as pilot, by the way - hours, types? Your ability to produce manuals, graphs and quotes is indeed impressive, but...................

Has the thought crossed your mind that some of the most probably would have had similar words to the AB page in their manuals - and STILL taught otherwise? Your faith is touching, to say the least, and to come from a world where everyone (and, of course, everything) performs faultlessly must indeed be wonderful.

BOAC:
Why ad hominem? :=

Besides, where it's written that those 'some instructors' are/were AB330 instructors? Or from AF?

If uncommanded by PF (or a/c), the climb may have been addressed with PF's 'Nose Down inputs'. I think BEA have not specifically timed those.

Likewise, he would not advance Throttles (he did not).

The second STALL WARN, could actually be predicting the STALL that resulted from loss of energy in this climb (which thus far is not described as to origination). The AoA (not necessarily PITCH) would increase as the Plane slowed, independent of elevators that had lost their effectiveness anyway. As the Plane dropped Nose (Either as a result of controls OR STALL), the AoA would reduce, hence TOGA and 'backstick relief' as the a/c "STALLED". At this point, had PF done nothing, he may have recovered? (No TOGA, no 'back pressure' held). Saying the pilots did not know they were Stalled is presumptuous.

At this point it is not determined if NU inputs are slight (inadvertent) pulls by PF, and not a back (climb) command. It could indeed be a trained APPROACH TO STALL recovery, at a very inopportune time.

Bear, to be clear, that depends on the duration and frequency of nose down inputs as counters to nose up inputs (per how a Sidestick works) which detail is still vague. The BEA seems confident that at least initial climb was initiated by a nose up movement of SS.
Huh?
The climb slowed the plane. Since power was not initially reduced, climb came from pitch up, so ... AoA change was due to ... drumroll ... pitch up. You with me? It has been suggested that when TOGA was selected later in the event chain, nose pitched up more due to how planes like this fly. This tends to increase AoA if not countered by the appropriate pitch adjustment (n-d) to accompany the power increase. This is Flying 101, and would be called for if in alternate versus normal law. There is evidence that pilots were aware of being in alternate law. Not sure of the granularity with which BEA can parse the data from FDR ... we shall see. I suspect it correlates well enough.

Sorry Lonewolf.

I am posting on the other thread, and forgot to mention I assume a substantial UP Draft for the climb, consistent with Harry Mann's proposal that the Right wing drop may be a Tip Dip into Vertical Airmass by Left wing. See, if uncommanded, PF would input Nose Down, to maintain altitude (I reasonably assume he did not want to climb!).

Had a stroke last year, and sometimes (usually?) I leave out key bits. And include some extraneous ones!

Hi,

At the altitude where was the plane .. with the throttle setting he have (before the PA disconnect) I suppose the difference of push by the engines will be minimal when throttles set to TOGA .. and so the nose up effect will be no significant
I understand that at low altitude a TOGA set will produce a significant nose up ...... not at high altitude.

There seem to be a lot of different stall recovery procedures on the market from the time before AF447.

@takata and Iceman

I think, the revised stall recovery procedure, implemented due to some LOC´s without successfull recovery by A+B aircraft made a lot of sense, so it seems to be short sighted to assume the old procedures had nothing to do with the training before AF447 and therefore the applied procedures of AF447. The emphasis of the old procedures was placed on speed increase and maintaining altitude, whereas the revised procedures aim on reducing AOA firsthand and deal with altitude when AOA and speed allows further maneuvering. There are a lot of differences to the new procedures and the procedures as coppied from FCTM A330/A340 (think it´s from Cathay Pacific) from 2006. I admit, i dont know wether those procedures underwent a further change between 2006 and AF447 loss.

Bolding by me

Again bolding by me to highlight some points.
First the old procedure is not named stall approach recovery, but stall recovery. It deals with the situation AF447 was in, alternate law. The actions should be applied simultaneously. Set TOGA thrust. Reduce pitch to 5° above FL200.

That is quite a difference to the new wording

– Apply nose down pitch order on the side-stick
– If needed, reduce thrust in case of lack of pitch down authority
– Ensure wings are level

because a reduction of pitch can be achieved by a reduction in NU force or, if pitch would be less than 5° and PF is focused on the number 5° pitch, he may be motivated to increase pitch to 5°.

The last para is interesting too. It disqualifies the stall warning at high altitude to being just the point of buffet onset (not being stalled already, which might well happened). Again here "relax pressure on SS, no word of ND input. And further on ..... "once stall warning stops, back pressure may be increased again....... not reapplied again.


And the summary of the new procedure says it all (bolding by me)

• Spirit of what is the new procedure
 One single procedure to cover ALL stall conditions
 Get rid of TOGA as first action
 Focus on AoA reduction

If the PF judged the first spurios stall warning at the beginning of the climb as not valid, but the second one at FL375 as valid, then he acted in accordance with the old stall recovery procedure described above by
-applying TOGA and simultaneously reduce backpressure on SS, which for sure was not enough, because a full ND input would have been appropriate. But where do you find a ND input in the procedure of 2009?

By the way, it has to be emphasized that those procedures (the old and the new one) are derived from thinktanks originating from A+B.

I dont think i would have followed this old procedure if trapped in this situation, because my expierience would have hopefully overridden this "nonsense procedure. But did the PF ever expierienced a real stall in a jet?

But if a stall wasn't recognised, why should stall recovery training be relevant ? Ditto THS authority in the absence of sustained ND commands. Mind you, it would be reassuring if they did train it properly in any case, would seem a pity to spend money training and to cock it up. AS SLF I am surprised that with all the instruments and automation to help, it was possible for 3 experienced and trained crew not to be made aware through appropriate indications and prioritised warnings that the a/c had stalled. Of course leaving the stall warner activated when off the ground might have helped, though in this case even that could be doubted.

But only one relevant at the time of AF447 is what I already posted above from Air France A330 2009 FCOM (Rev.24).
Anyway, I don't understand how one may consider that PF actions at this second stall warning (which lasted 50+ seconds) could match any published stall procedure at all without completely twisting the facts. Neither TOGA nor sustained pull-up would be part of it.

Recency of training.

What stall warning response maneuvers were last practicied by either pilot in the cockpit during their last simulator training session?

This is a human factors issue that I've seen raised in more than one aircraft mishap investigation, including one I was personally involved with.

Is there also a FCTM from Air France available with the relevant paragraph like that one posted by myself?


Isn´t the new generation of pilots focused on procedures and SOP´s and a strict obedience of those procedures, whereas my generation was challenging those procedures and their application dependent on the relevant situation and based on their own experience (and improving those by doing so)?

And based on that statement (also it has a ? at the end) and the old stall recovery procedure (not stall procedure, you misnamed probably) the actions of AF447 reflects the application of this procedure to detail, without twisting any facts. We might not understand, why they continued the aplication of this non functioning procedure for so long, but again, see above.

It would be a better approach to question the implementation of the old procedure instead of twisting the application on behalf of the wrong outcome.

retired F4: spot on, and not to mention the training norms and experiences for that particular procedure and situation.

Getting at that may shed some light, but it may also be difficult to do. Fear of blame seems to be infesting corporate culture.

That there is a big problem is recognized by the industry

See flightsafety.org from apr 2011

The more i´m wondering, why we tend to disregard or to deny this problem on the expert level.

Hmm. In which way does maintaining nose-up inputs until pitch attitude ends up at 16° reflect "Reduce pitch attitude to (...) 5° at or above FL200"?

Approach this accident as if the Pilots were not stupid, and Franz starts to make perfect sense. (Probably not just to me!)

It does not at all, and we should ask the question how come?
Maybe that is the reason?

Together with
- lack of training in manual flying,
- lack of training in flying in this altitude ,
- lack of training in manual flying in this altitude in alternate law

continue with

- at night
- in WX /maybe turbulence
- without airspeed indication
- sudden stress

and you might find lots of reasons, why this recovery attempt with a meanwhile outdated procedure failed.

So you think they knew they were stalled ?

If they were unaware of Stall, they would select TOGA exactly, why?