Hello Shadoko,

Thanks for posting.

The article shows a simulation aligned very much with the BEA report we already know well - no new, surprise elements. Which perhaps is an indirect indication of what's coming on Friday....

Turbulence is at a level, where the PNF needs to keep his pointer on the lines of the check-lists he is reading.

PJ2,
The FCTM you quote must be an Airline's customized version ... ?
If so, that Airline did a very good job by stating things clearly, which Airbus has not done yet.
For Airbus, as quoted earlier by A33Zab, the correct procedure is to take initially 5 degrees NU + CLB thrust :

As the messenger for Airbus or BEA as you wish, Le Point a parlé :

• L'action du PF à cabrer sur le manche continue, et aura été majoritaire pendant 95 % des quatre minutes.
• Les données contenues dans les enregistreurs de vol montrent très précisément comment s'est déroulé le crash de l'AF447, et permettent d'identifier les conséquences de l'action à cabrer du PF.

Amen !

Do we have to consider both the PNF could not see what the PF was doing ...

THX for the THS position in your analysis. at which time did they passed SALUP?

the climb to F370 (without trust correcting) has finished 02:10:40
the following correctionTOGA was also not corrected in his pitchup momentum....

and then a stray purring cat was laying on the stick.......

CONF iture;

Thanks for responding.

I don't want to put too fine a point on this as it is just one of a number of notions which have been put forward to come to terms with the pitch-up. The drill and checklist are a problem.

The very fact that there is primary disagreement about how to do this checklist is itself, interesting and revealing.

I do see your point of course, but think about it: What pilot is going to pitch an A330 up at 5 degrees at FL350, expecting that the aircraft is not going to climb? The pitch and power settings for an A332 at FL350 at 205T for 260kts is 3.5deg and 70.9% N1. Five degrees is not going to be a runaway climb but it is going to climb.

The early drills and checklist, (2002, 2003) do indeed state that the "Immediate Pitch Attitude and Thrust Guidance" above FL100 is 5deg of pitch and CLB thrust.

The drill and checklist in force at the time of the accident is confusing because it's flow is not clear nor are the priorities for the pilot. I have made the points before that there are several "bifurcations" in the checklist which are difficult and problemmatic to memorize. The FCTM statements, (2007) which I reference state quite clearly that if the flight is not in immediate danger, then the procedure is to level off and troubleshoot.

The rest of the statement which you quoted from A33Zab's original posting reads as follows, (my bolding in the first sentence and after):

"PART 1: MEMORY ITEMS
If the safe conduct of the flight is affected, the flight crew applies the memory items. They allow "safe flight conditions" to be rapidly established in all flight phases (takeoff, climb, cruise) and aircraft configurations (weight and slats/flaps).The memory items apply more particularly when a failure appears just after takeoff. Once the target pitch attitude and thrust values have been stabilized, as soon as above safe altitude, the flight crew will enter the 2nd part of the QRH procedure, to level off the aircraft and perform trouble shooting. This should not be delayed, since using the memory item parameters for a prolonged period may lead to speed limit exceedance.

PART 2: TROUBLE SHOOTING AND ISOLATION
GENERAL
If the wrong speed or altitude information does not affect the safe conduct of the flight, the crew will not apply the memory items, and will directly enter the part2 of the QRH procedure."

Regardless of what Friday's release has to say, this drill and checklist is a problem.

bear,

sorry, us humans need to sleep from time to time.
To cut it short, no, of course I don't know what happened. Still we can assign different levels of probability to the various suggested scenarios. I'm absolutely with you in firmly believing they were trying to do what from their perspective looked like the best chance to get home.

Audible ice shedding from pitots? Nope...
 

Sadoko:
As one who spends his life creating sound simulations for all classes of aircraft (at the highest fidelity Level D for commercial aircraft and equivalently for military sims), I find it almost impossible to believe that anyone could believe such ice shedding would be audible - the location of the pitot tubes themselves and, above all else, the small quantity of ice involved (due to the physical size of the tube) makes this claim extremely unlikely.

As it stands, for the many simulators I have been involved in, ice shedding from propellers, with the ice being flung from the blades and hitting the cabin sides, is about the only sound cue related to icing that I believe might be audible to any cockpit crew.

Given the altitude, true airspeed and the factors mentioned above, I do not believe that ice leaving the pitot tubes would cause any noticeable audible cue for the crew AT ALL. :=

Hi Neptunus Rex,
post #791
I'm sharing all your comments, only one tid bit should be added.
α Floor is only available in Normal Law, but it is inhibited above Mach 0.53. Below this speed, high AOA protection would reduce pitch while applying α Floor. It is obvious that high AOA protected mode is aimed at low altitude, high pitch and low speed situations which are, by very far, what is frequently needed.

At first sight, stalling at cruise level seems so unlikely, in protected mode or not, that most of the documentation/procedure is not even talking about the case.

At cruise, if one hear the stall warnings, this FCOM sentence "release back pressure on the sidestick" is also inherently implying that the PF is pulling gee's during manoeuver, hence triggering SW, not that he is flying level at the onset for buffet.

If one look at the UAS/Turbulence penetration procedure tables, in clean configuration, safe pitch/thrust settings (=> Mach 0.80):
@ FL370, over 190 t:
- Pitch: 3°
- N1: 94.3%

Hence, at FL375, TOGA would not make that much difference as the aircraft would be very close to its maximum cielling anyway. Also, applying a 5° pitch "above FL100" is in reality aimed for above FL100 and below FL250. It is 3.5° from FL250 to FL370 (90% N1) and 3° above FL370. Also, those settings should be memory items for turbulence penetration speeds.

TOGA + pull up is the Windshear normal procedure. In no way it is related to STALL or UAS procedures at cruise altitude.

Also, while the situation was deteriorating seriously, from 0210:51, the aircraft took another 10 seconds to enter this full stall with all Alarms soundings. Even so, the PF still had its Flight Path Vector (FPV) working. In fact, it was flagged much later when alpha/speeds became again invalid when the stall was much more advanced.

This is another puzzling fact as the FPV (which seems to have been selected) would provide a visual increasing Alpha for the pilots. Moreover, during this early stall sequence, it is also probable that the Flight Directors came back on PFDs and that all speeds/baro altitude were valid and coherent again (but without any characteristic speeds -VLS, VSw, VMax- displayed).

Consequently, my opinion is that the PF should have distrusted all its instruments as everything seems to be working fine in the cockpit displays at this crucial point... but also, during the following 30-40 seconds, in addition to this Stall Alarm which never stopped its warnings.

Le Point article on AF 447 accident simulation
 

For those that are more comfortable with an English version, here are some excerpts from the article. I've selected based on elements discussed on the AF 447 threads:

- the thrust at FL 350 was at %85 - article comment: "at this altitude Thrust level changes other than Idle don't bring a very significant power variation."

- "for changing altitude, the procedure requires a pitch no less than 0 degrees for descent, or 4 degrees for ascent."

- "the altitude change caused a buzzer to sound at 200 ft above FL350."

- "95% of 4 minutes, the NU action was present. However, there were a total of 6 short ND commands."

- "at FL 375 the climb is at 700ft/min, and the plane still flies at 4 degrees attitude and 215knots, certainly slow"

- "when the Captain rejoins, the speeds are invalid, the Stall Warning stops, quite a confusion in the cockpit"

- "while the plane was at 10000ft/min descent, and 40 degree attitude (AoA), the actions of thrust to Idle, and ND don't bring a significant change, AoA remaining above 35 degrees."

GarageYears,

It seems that they've created sounds for the simulation, that replicate the AF 447 cockpit surrounding recorded sounds. Do you doubt that as well?

The article's author interpretation of the "ice cracking" sound which you point to, may be very well wrong.

With your experience of generating sounds from difference sources, for simulation, what's your interpretation, what is that "ice cracking" sound coming from?

Garage Years

Just the man for this. In the manual re: UAS one of the potential causes for loss of airspeeds is a shed radome. Whether 447 lost her fiber nose or not, would the radome contacting the fuselage area near the probes be audible to a crew?

Hi airtren,
This article seems to be a mix between BEA and Presslike-fantasy (simulation). What could be trully learned from it at one day from an official interim report release?
The only interesting information out the whole content is this one:
They are just selling their "Rio-Paris dossier" in today paper issue, as tomorrow, it will be certainly outdated.

bea report one day before !
 

It appears that the french radio "FRANCE-INFO" (MM. FELDZER, POLACKO, aso):ooh: received not only the BEA report, but all informations from the both CVR and DFDR, when it is refused to public use !?
Are OACI procedures no more in use in FRANCE ?

I have seroius doubts about the interpretation that they heard ice cracking sounds on the CVR

Lots of CVR recording over the years and the only ice sounds documented to my knowledge have been impact sounds from hail hitting the frontal faces. The relative velocity of a shed off an aircraft surface pitot would be too low and this "Ice crack sounds" defies imagination. Even the ice sheds off the MD80 incidents could not be heard.

Really? The only thing up on their website is an interview with two of the people involved in the simulator reconstruction sessions - they don't say anything about having a leaked report...

takata

I have not been persuaded by the ICE theory. It is not at all unusual for a radome to come off the nose. After all, it bears mention in the FCOM.

Whether 447 lost her fiber/resin nose or not, the "Sound" in the CVR will be interesting.

I am keeping an open mind.

lomapaseo, your thoughts re: Radome loss?

Hello PJ2,


Indeed, from the start, you've said this consistently, many times....Your confidence and calm which I wish in every pilot of an airplane I am a passenger of, make it sound very simple, very safe....:ok:

Would you consider qualifying a bit the very succinct "do nothing ... ensure the a/c is under control", in regards to the flight law, and turbulence?

Is there anything you would add, is there a need to differentiate the needed actions between the A/C being in "Normal Law, No Turbulence", "Alternate 2, and NO Turbulence", and "Normal Law, with 3 degrees of turbulence: mild, medium, heavy", and "Alternate 2 and mild, medium, heavy turbulence"?.

02:09:45~02:10:25 It's in here.

I don't think that's an accurate statement at all Bear. I think it's extremely unusual. ...and extremely unlikely.

Hello Takata,

The picture is informative.

Would you have a clue, what training center is that?

Often?
 

Hello Bearfoil,

Yeah, we loose radomes by the dozen, they come cheap these days.

Icing of probes, however, does not happen that often, or does it.
The following quote from BEA interim report does not describe a large number of radome loss events.

Quote;
As of 3 November 2009, Airbus had identified thirty-two events that had
occurred between 12 November 2003 and 1st June 2009(18). According to Airbus
these events are attributable to the possible destruction of at least two Pitot
probes by ice. Eleven of these events occurred in 2008 and ten during the first
five months of 2009. (Note by EMIT: destruction should be read as obstruction)
Twenty-six of these incidents occurred on aircraft fitted with Thales C16195AA
probes, two on aircraft with Thales C16195BA probes and one on an airplane
equipped with Goodrich 0851HL probes.
As of 1st June 2009 Air France had identified nine events that might meet
the above-mentioned criteria. After the F-GZCP accident the airline started a
targeted analysis of recorded parameters and identified six additional events
that occurred in 2008.
In addition, a foreign operator began a targeted analysis of recorded flight
parameters recorded after June 2006 on its A330 fleet. As of 18 November
2009 it had identified fourteen events. Only four of them had been detected
and reported by the crews to their airline.
Further, Airbus identified four events that have occurred since 1st June 2009.

Less rare than simultaneous (including rate) obstruction of three Pitot Heads?

Fair question. However, I am basing my question on the FCOM. It's in there.

We are dealing here with most likely an extremely rare confluence of bad things. Yet I witness an unusual presence of "stuckness" and "dismissiveness". I'm not actually taking sides 3hl. What I seek to do is express my thoughts regarding possibilities.

Here, your statement is best directed at Airbus/AirFrance. Again, it's in the FCOM as a cause of duff speeds.

EMIT

Until the data is in, what needs to be acknowledged are all the possibilities, regardless of probability. I am having a hard time trying to convince people that the time to defend a particular opinion is after all of them have been entertained, and the data is available.

If it is too soon to state cause, it is definitely too soon to reject even one, at least conclusively :ugh:

A picture is informative, or will illustrate an article, because it is showing something related (like a flight simulator) not because it was taken specifically for the relevant press article. This one is from news archives or possibly related. It is more likely from news picture archives as it seems taken during an aeronautical convention.
Consequently, I'm pretty sure that this picture was not related to any official (informed) replays of the recorders (Airbus, Air France or BEA).
Now, you may wish it was.

Audible ice shedding from pitots? Nope... BUT...
 

GarageYears
As one who spends his life creating sound simulations for all classes of aircraft (at the highest fidelity Level D for commercial aircraft and equivalently for military sims), I find it almost impossible to believe that anyone could believe such ice shedding would be audible - the location of the pitot tubes themselves and, above all else, the small quantity of ice involved (due to the physical size of the tube) makes this claim extremely unlikely.

As it stands, for the many simulators I have been involved in, ice shedding from propellers, with the ice being flung from the blades and hitting the cabin sides, is about the only sound cue related to icing that I believe might be audible to any cockpit crew.

Totally accept the thrust of your point, but think that there is another source of ice-related sounds.

I dimly remember claims made in the original AF447 thread of hearing ice on the windscreen. Cannot find those posts, but this later one captures the idea.
http://www.pprune.org/6538743-post1871.html
EMIT
I have seen the TAT anomaly (TAT probe icing due to high ice crystal content) when flying through "light green" radar returns in the neighbourhood of CB's, in other words, while avoiding CB's. Turbulence then was only light, occasionally moderate, nothing out of the ordinary. Saint Elmo's and a sound like rain on the windscreen were also present.

To summarise:
a) The claim about hearing ice shedding from the pitots sounds like b/s. [Or a journalistic misunderstanding.]
b) Hearing ice shedding from elsewhere is highly unlikely. [You are the expert.]
c) In de-briefings several pilots have reported hearing something (later identified as ice-crystals) hitting the windscreen prior to a high-altitude "anomaly". So it may well have happened on AF447. [IIRC it is often reported as quite a distinctive sound.]

- this is NOT what the May BEA report said - perhaps you are 'clairvoyant' regarding the 28 July report? The stall warnings sounded BEFORE the climb.

BOAC

airc, THe "STALLWARNING" sounded briefly at PF first input, NURL.


Could that be consistent with a cruise speed input of NOSE UP? If so, would the PF immediately reduce PITCH? If so, it may not be associated with the climb at all?

No idea. The way the (English) report is written (for what it is worth :mad:) there were at least 11 seconds before the a/c began its climb. I was going to sit quietly and await a better report tomorrow but I got woken up by people inventing stuff again:ugh:

france info
They won't tell if they got access to the data, but listening to the interview, it is possible they actually did :

"The pilots didn't understand what was going on, we clearly hear it in the registered conversations in the flight deck, there is no panic."


Also a desire to 'protect' the Airbus technology to the public eyes ... ?

"The airplane did not stall, il s'est enfoncé (it has gone deep ?), thanks to the protections."
"The protections have been effective but not understood by the pilots."

From BEA. At 02:10:16: "So, we've lost the speeds.......ALTERNATE LAW". (PNF). In between the STALL WARNING, and the (PNF) voice recording, there is no official mention of "CLIMB" (or STALLWARNING). Likewise, it would seem a blatant omission by BEA, had the a/c begun the climb, or had experienced notable problems. (Other than UAS).

We may know soon, or not. I expect somewhat less than full disclosure.

Some people are still entertaining the idea that those flight data, CV recorders, are hiding critical informations in order to state the truth, informations that will point to the only one cause of this crash, the only relevant thing they could ever be able to believe as being not a cover up for the manufacturer's wrong doings.

My own feelings is that we won't get more clues tomorrow about this single cause than what is already known fairly exactly today (of course, if one bother to put all the facts together by stopping to make up what he wishes), because nothing more than that was recorded, nothing that could direclty explain PF's actions.

One could replay the scenario many thousands times, it won't help to understand the cause without discarding this way of thinking. Pilot's brain is not recorded. All we are going to learn is more details about their confusion concerning the situation they were facing. It is quite obvious that they never acknowledged the stall situation, neither that they would have tried to recover from it if the stall was not acknowledged at the first place.

That's, in my opinion, the issue the industry would have to deal with in the future. As there will be no single and very easy answer to: what to do then? because it is a very complex one, it's not going to be fixed any time soon.

What is it about "the Pilot's Brain" that is of interest to the a/c? This a/c is protected, flies safely, won't STALL, (at least on its own....)

A thinking Pilot, one 'thinks', is not necessary to the Airbus. One who can memorize, and leave things alone? Now you're talking! Thinking is optional, and more expensive!

Simultaneous Inputs
 

According to the 27 May 2011 BEA update:

"At 2 h 13 min 32, the PF said "we’re going to arrive at level one hundred". About fifteen seconds later, simultaneous inputs by both pilots on the sidesticks were recorded and the PF said "go ahead you have the controls"."

I am interested in the "simultaneous" inputs. Were these consistent, or were they conflicting NU and ND commands?

CVR sounds
 

Firstly - if the assertion was the sound was replicated from the CVR mic recording it must be recognized that the mic is optimized for voice bandwidth. I commented on this previously, but don't have time to search for the post. But the CVR cockpit mic frequency range is limited - I want to say to around 6kHz....?

Secondly transient noises of short duration are very hard to isolate and make any sense of unless very loud. Ice shedding from the pitots would not do this.

Ice crystals hitting the windshield - yes, I can see that being audible, certainly.

Radome damage/departure/etc - something of this nature would certainly lead to a noticeable change in the aerodynamic impact noise footprint.

Both ice crystal noise and radome damage would be heard over the CVR. Ice-shedding not so, in my professional opinion.

Takata, I fully concur with your remarks; tomorrow is not going to "solve" this. Expectations that it will are bound to be disappointed because that isn't the way accident investigation works. Reification of this or that pet theory will not be found in this update.

airtren;

Thanks for your comments and question.

Machinbird made a good point a while back in pointing out to me that "doing nothing" isn't the exact response and of course he was right. By the phrase, I was assuming a great deal on the part of the reader. What I meant was, other than maintaining level flight as best one can using pitch and manual thrust settings, (which would be those which existed just prior to the failure), one "does nothing" in terms of climbing/turning/descending etc. Above all else, one stabilizes the aircraft in level flight. There is absolutely NO reason to change anything during a UAS event.

Now, your question is a good one, made even more important by the extensive and wonderfully-intelligent discussions by many here who do know their stuff, concerning the behaviour of the A330 in Normal, Alternates 1 & 2 and Direct Laws, including simulation of same. I have learned both here and in private communications more about the airplane I flew for many years than I ever knew during my time on the airplane. (There is a notion there that I would like to explore, but time, space and a low desire to do so all call for a break at the moment). The airplane flies and works brilliantly; what is being explored here, as takata notes, are the extreme boundaries of flight in a heavy transport, and at the boundaries, (note: not the limits, but the boundaries), of design and engineering. That is why this accident is extremely complex, involving the above and a healthy dose of human factors, ergonomics, warning systems, and so on.

I assumed the airplane was just about as easy to fly in Alternate 1 and 2 and Direct Laws as it was in Normal. I have done this in the simulator many times and I think it is safe to assume that simulator fidelity is high in these regimes, (vice low fidelity in the stall or upset). It was never a problem flying the airplane thus, nor was it like "balancing oneself on the top of a greased flagpole". I understand the roll direct is brisk, but at altitude, one is always gentle and smooth with a heavy machine in thin air.

In direct response to your question, I think one can manually fly this aircraft at high altitude providing one handles it thus. I have learned through discussions on the reduced damping effects of high altitude flight and the physics of mass and trajectory of a heavy machine as well as the notions underpinning an understanding of PIO, that maintaining level pitch and bank attitudes in the circumstances you describe, would be challenging and perhaps even very demanding depending upon the second-by-second series of sidestick inputs over a short period of time. I think one can lose one's SA if one induces PIO, especially in Roll Direct, a factor which I had not truly been aware of until recently. It will be interesting to see how the BEA deals with this notion and phenomenon, if at all.

These points are why this accident will not be close to resolution and understanding in tomorrow's release by the BEA.

Marshal McLuhan once observed, "In the vortex of process, there are no fixed points of view; understanding is never a point of view." I think that may describe an approach to this report and the final report which will yield the best comprehension of what occurred. If one as a point of view, the report will be disappointing and unfulfilling.

I hope this helps airtren; I am enjoying your contributions and am pleased you entered the conversation.

Thanks for the pointer CONF_iture

From listening to the interview, among the two interviewed pilots comments that I would mention, are references to the shortcomings of the "a/c to pilot information interface", the reviewing of procedures for such situations, and a dialog between Air France and other airline companies, on one side, and Airbus on the other, in that regard which is just at the beginning..

Icon
 

Bearfoil, for once I'll inject an icon: :ugh:

Of course, the FCOM mentions the possibility of a radome getting damaged or lost. Such an event would greatly influence airflow around the nose of the aircraft and would thus adversely influence the pressure measurements as taken by the various pressure sensors mounted around the nose. But FCOM mentions a zillion possible failures.
The BEA Interim Report number 1 mentioned an estimate of aircraft state at impact, based exclusively on damage patterns found on floating debris - almost 2 years later, the recorders proved that the original estimate was remarkably accurate.
Part of the original estimate, was parts of the radome found at sea.

Quote from Interim report 1:
Observations of the tail fin and on the parts from the passenger (galley, toilet door, crew rest module) showed that the airplane had likely struck the surface of the water in a straight line, with a high rate vertical acceleration.

Quote from Interim report 2:
From these observations it can be deduced that:
* The aircraft was probably intact on impact.
* The aircraft struck the surface of the water with a positive attitude, a low
bank and a high rate of descent.
* There was no depressurisation.

I think you can TRUST the BEA that, had there been grounds to suspect that the radome was lost, they would have mentioned it in their reports so far.


To the rest:
Whether it is conceivable that pilots somehow make an error with regards to stall reactions? Yes, it happens, even in good old fashioned Boeings!
Heavy take-off, during flap retraction an upgust underneath a big Cu (note: I do not state Cb), as margin at that moment is very small, it triggers a stick shaker.
PF reaction? Pull up the nose slightly. PNF intervenes, so nose is lowered, a/c continues flight uneventfully.
Query by the PNF (the commander) about the reaction of the PF - well, it pointed to a sort of mix up of the text from GPWS/Windshear procedures, that speak of pulling up, with the stick shaker as upper limit of that action. So, somehow, in the PF's mind, there had formed a connention of "stick shaker, pull up".
More strange reactions in pilots minds: read up in an old Boeing magazine about the 767 that took off from Isla Margarita (1994 or 1995) and turned directly towards the 3.300 ft high mountain on the island. The GPWS calls caused a PF reaction of increasing pitch by 1 degree. Only after the audible impact with a TV antenna on the mountain top, did he progress into a proper GPWS reaction.

Once again: 15 degrees nose up, clearly visible on those beautiful PFD's, is very, very unusual attitude in a transport aircraft around FL350. All recorded pilot actions should have been very much nose down. All sorts of failure flags on the PFD and messages on ECAM should not have distracted the crew from the still perfect attitude indications.
And, as said by many before, the mighty "push" of TOGA at FL350 is nothing more than CLB thrust at that altitude, and as such a disappointingly small amount above normal cruise thrust. Certainly not a big handful of pitch up moment.

Have a look at the fractures evident on the radome after surface retrieval.

Note the linear fracture from the outer perimeter in toward the center of the piece. Can you describe the likely mode of failure of this two phase material with any linearity remaining? I can. Are you at all interested?

Follow up
 

Bear,

Behind the radome is the WX Radar. If the radome gets lost, that radar is not gonna be able to sweep its antenna back and forth any more, exposed to a 480 kts airflow. That is certainly gonna trigger a couple of messages that would have been in the ACARS data stream.

Radome Loss In Flight
 

CVR should go from relatively quiet to noisy as hell.

I doubt that we will hear the CVR for our confirmation, So I will just wait for the BEA to tells us what's significant on the CVR