Nobody is disputing that, we are all in agreement here concerning that fact. But to built this agreement on the wrong statements is not helpful.

As i mentioned before, i see no sense in exchanging arguments of our own by disregarding the meanwhile published final report.

Agreed, there are 220 pages + appendixes plus pictures and graphs, we all should read it and use it. Either we have an oppinion along the report or against the report, but just to sideline it leads us on the wrong path.

@roulishollandais

Thanks, you can be sure that i download the files for my personal library.

Hi Lyman,

Being that you asked,
http://i1166.photobucket.com/albums/...raphic-1SB.jpg
TECHNICAL DATA
RANGE: pitch +/- 90 degrees
roll +/- 180 degrees
ACCURACY: +/- 0.5 degrees
• ATTITUDE according
to TSO C4c:
• ALTITUDE: according to TSO C10b (AS 8009A)

It is called a standby horizon (attitude/altitude) display that is located to the right of the Captain's FD, but to the left of the engine operating display. Below it is two other instruments, one being a clock. Probably, this is how the PNF determined the PF was going up instead of down, level the wings, etc., and later on, the altitude callouts he made as they descended rapidly It may be capable of displaying speed, but not in AF447's UAS situation.

Sure - the point I was making was that we had interesting information *in addition* to the points made by the BEA - in that we had a transient stall warning similar to the DFDR spike without simulated blocked pitot tubes. In mentioning this I was not disregarding the report, simply suggesting that there may be another factor involved.

You feel insulted because you misread the word I used - Theorist. A conspiracist conspires, a theorist theorizes. I think you are a theorist. I think the salient points regarding pilots' actions have been released by the BEA in their final report. But if you want to know how thing were in Rio, what they are going to do upon reaching Paris, etc., well, .....
Theories, most theories tend not to hold water in the end because they have no facts supporting them.
It shouldn't be outrageous at all. The crew actually were unaware of the situation for the entire duration, not just precious seconds. And when the saw "crazy speeds", that didn't trigger the UAS memory list. The ECAM told them of the reversion to alternate law, but somehow the discipline and shared responsibility between PF & PNF wasn't there that night.
SEF is a pretty well known phenomenon when you get into clouds surrounding CBs, what is there to analyze? The heat in the cockpit is warm air suddenly drawn in from the outside, momentarily overpowering the environmental control system. With the warm air came the smell of ozone. It is a pretty common and unique smell. Today, you can actually purchase Ozone perfume by Sergio Tacchini to get a whiff of how ozone smells.
Could be one of two things - Stall Warning or Ice Crystals hitting the windshield.
I think it would be better said this way, "After nine threads, PPRuNe was relying on the BEA's Final Report to confirm their analysis."

It looks like the future is either smarter fail safe computers or more competent pilots. The CEOs will run it through their computer and decide which most helps the bottom line. At least that is how it has developed in the last decade. Good luck.

Thanks RetiredF4 for highlighting this anomaly, and it once more brings to the fore the hazards of "common mode" failures, especially those that then rely on faulty air-data to provide a warning function.

One can only hope that Airbus Industrie will look at using IR data as the point of reference when double or triple ADR disagreements are identified. In this case when ADR Mach values disagreed and the equivalent IR sourced value(s) was higher, then it should be used. This of course leads to the adoption of Vsyn or similar as dealt with earlier in this and other threads.

Hi,

Turbine D
The two go together

A conspirator can not be alone
For have a conspiracy it must be some people sharing the same view (or have a same theory about an event) and they are working together (in a way public or secret) to spreading out this theory to reach their goal :)

TD

By artifact, I meant an actual snap of an instrument reading from the flight. Dozy says these are recorded, I asked to see one.

After reading thousands of posts about this, I guess my conclusions are:
1. Most (bit not all) people are reasonably confident in the independence and thoroughnes of the investigative process, and the conclusions drawn. Many were surprised they managed to recover the recorders, and then the data from the recorders. Perhaps as a result of previous criticism, BEA have made greater efforts to be transparent this time around, and recorded crucial steps (such as the sealing of the recorders). They recovered the two bodies to establish with 100% confidence who was seated at each position.

2. Apart from the pitot icing (a known problem), the aircraft most likely operated as designed. The UAS by itself should not have been sufficient to lead to this tragedy, and numerous previous incidents ended uneventfully.

3. There may be some aircraft design features which need to be reviewed. The stall warning cut out at low speed does seem unnecessarily confusing and may have been unfortunate in this case. There may well be others, such as the sidesticks being out of view of the other pilot.

4. I have read a lot of discussion about various control laws, and am frankly surprised at how much disagreement there is. Have things become too complex? Can pilots be expected to work through this in real time if these issues can be debated for many months here?

5. There should also be some thought given to the human - machine interface when things go wrong. For many reasons, the trend towards increased automation will continue, and pilots of the future will have less real "flying" experience. It is inescapable that these developments create a challenge when the automation fails. This challenge is predicable, and must be addressed. In this case, the pilots were overwhelmed by warnings and messages and conflicting data, and this contributed to the poor situational awareness and loss of control. This is also predictable. In an emergency, information overload will quickly lead to a feeling of helplessness and panic, and also confirmation bias as you search for something (anything) that fits with your (potentially erroneous) understanding of the situation.

6. There are obviously CRM considerations. After the upset obviously. But also prior to the upset. Was a 40 second handover sufficient before the Captain left the deck? At night, crossing ITCZ? Is it reasonable that the Captain suggested that a pilot had a sleep after he (Captain) went for his rest?

7. Video recording would greatly add to the CVR in the event of an incident. I don't believe this should be released to the public - some things do belong behind closed doors. But a visual record would surely strengthen the investigators understanding of exactly what happened, would lead to firmer conclusions, and would enhance lessons learned. In addition, the knowledge that there was a visual record available would help improve confidence in the investigation - in this case, we would know that the displays were functioning. Solid state data storage is cheap, compact and secure. Given that I am recorded when I scan goods through the supermarket checkout or pull out $50 from at ATM or fill my car at the pump, then maybe this is a development whose time has come.

As far as video recording goes, it need not be done in real time. One picture every second would be adequate to determine what was going on at the time. This would allow for a long duration recording without an undue demand on memory space.

The same goes for displays. Instead of trying to shoot the cockpit displays with a camera, why not capture the actual displays and record them? Again, once per second would be adequate to determine what was going on, and yet not require an inexorbitant amount of memory to accomplish.

All of these captures (cockpit cam plus each display - 6, 8, what?) could be recorded on a separate memory module in the FDR, or in a separate Video Data Recorder similar to the FDR. As has been pointed out, the technology certainly now exists. And one capture per second for each camera/display would be sufficient.

Thoughts?

Hiya slats...

You write: ....."1. Most (but not all) people are reasonably confident in the independence and thoroughnes of the investigative process".....

That is not wrong, however let me expand on the hole it leaves for interpretation.

The memorandum BEA released....

Out of sequence, out of scope, and not required by the mission BEA is charged with. The impression left with the public was one of suspicion, and rightly so, it appeared just prior to a massive opportunity for Airbus to pitch its products.

If I was leading the investigative team, I would have been livid to have been approached by any one from Airbus. Do you think this anomaly was spontaneous?

Wanna buy a watch?

The only person who brought it up at the time was you.

Well, you weren't leading the investigation - and in real terms it made sense. I suspect that if the BEA had been asked and they weren't certain they wouldn't have allowed the statement to be released. It's unusual, but logical - and IMO it's a big stretch to call it suspicious.

Dozy...." I SUSPECT that if the BEA had been asked and they weren't certain they wouldn't have allowed the statement to be released. It's unusual, but logical..."

Dozy..."and IMO it's a big stretch to call it SUSPICIOUS.

You may suspect, but I may not?

slats, it is my opinion that your point 7 would be utterly abused by airline management for the 99.99 % of flights that don't end up in extremis.

Bank on it.

UAS was never realized in F-GZCP
 

Hi,

Why Airbus SAS design delegates to the crew UAS diagnosis?

Why Airbus SAS design allow "garbage" to be processed?

Why Airbus SAS design presents to the crew, during crisis, conflicting information?

All this is for:

1)Fault tolerance and graceful degradation?
2)Critical operation and ACCELERATED degradation?

:confused:

Why not to inform the crew IMEDIATELY and PRECISELY UAS BEFORE triggering a complex situation like the cascade of events we had in F-GZCP?

On UAS early warning: Technically speaking this is VERY EASY.

UAS is considered an "insider information" ? :{

slats11;
NOTE: italics are edited/added comments

While reasonable and understandable, this is a very naive view. I, and most crews, don't believe any such recordings should be released to the public either, but they will, in one way or another. Would that it were otherwise.

First, there are diverse interests in "knowing". Legal approaches are legitimate but cannot use the same data sources, (meaning historical crew and aircraft FOQA/QARs data - SSFDRs and CVRs obviously are used) that is used in a safety investigation, otherwise the safety processes which prevent accidents are at risk or destroyed.

The notion of "the public interest" is often legally more powerful than any laws intended to protect proprietary information including safety information. "Trust us" is often what one hears from the regulator, for example. Though it was and is illegal to release the actual CVR, in a lawsuit in Quebec, the CVR of an accident in which the pilots were killed was released to open court for all to hear. Release of proprietary safety informaiton can, and will be done by the courts.

Second, we know from their behaviour that the media has a voracious appetite for ratings and will, not might, pursue the availability of such recordings and they will be provided by someone who has access to the data. The Ostelli book on AF447 is proof of this fact. The "News of the World" illegal phone tapping is another. This is what the media does.

So, if video recording is introduced and an accident occurs, you can expect to see the video on youtube or a "reasonable facsmile", (animation) of same because that is the nature of media and the nature of our present society.

There is no question regarding your point that video recording would be of great use in many, though not most investigations; the Egyptair B763 suicide descent is often quoted. But there are other ways to obtain sufficient data. Though there are significant problems in changing/enhancing any such recording technologies, mandating an increase in the number of parameters and the sample rates of recording would greatly assist investigations. The argument is again from AF447 where the entire right-side instrumentation, (PFD/ND) is not recorded, the justification being a "statistical sample" is good enough. Those doing flight data analysis have argued against the merely-statistical approach for these very reasons but regulators also must listen to the airlines which argue against the huge costs of doing more than the legal minimum when designing data frames. (For a good treatment on data frames, refer to CAP731 Approval, Operational Serviceability and Readout of Flight Data Recorder Systems and Cockpit Voice Recorders)

It is straightforward (meaning that the data is available from the aircraft), though very expensive and technically complex to record all text on CRT screen displays, when they appear/disappear and what causes them to appear. Such a level of recording would be monumentally difficult and expensive, (due to multiple STCs), to achieve and analyze but it is possible.

The argument on the other side, the airlines, is the expense and therefore the ROI. The regulators are sensitive to the tremendous costs of requiring airlines to increase recorded information, especially for older types for which wiring, sensors, hardware/software must be installed to facilitate such recording. But such enhanced levels of recording would render moot most reasons for the installation of video recording without increased risk of the data becoming public.

The popularity of animations of crashes, (youtube etc) proves that recorder traces don't sell soap; videos do.

My last post about the BEA .. so all happy :) (suspicions about BEA)
BEA has a sister branch who is in charge of maritime events ... and although this branch is also subject to suspicion (yet a conspiracy?)
The site:
BEAmer : Bureau Enquêtes Accidents de mer : Accueil
An event made the headlines and continues to surface (no punt intended) from time to time
This the sinking of fishing boat Bugaled Breizh
Here are some links concerning:
Bugaled Breitzh
BEAmer : Bureau Enquêtes Accidents de mer : Rapports d'Enqutes
And search in 2004 for the report about "Naufrage du BUGALED BREIZH"
And
La verité sur le naufrage du Bugaled Breizh
Bugaled Breizh - Wikipédia
Naufrage : Bugaled Breizh : le dossier relancé par un mystérieux témoin - France - TF1 News
Everything is in french .. but it is easy to use an online translator ...
BEA and its maritime branch carry baggages from the past and can not get rid of ...
because every time they respawn when there is an investigation and if this is right or wrong does not matter for the press or the public
Someone released from prison after serving his sentence is still able to re-offend

Michel A. was 9 months in jail..., Marc D., David D., Cedric B., died in AF447...

The Otelli book shows that the guy :

1. has been provided with modified data.
2. or has modified the data himself.
3. or has reported the real data in which case the BEA has modified them.

My vote is for A

Mass media keep quiet when it really matters.
They display obedience and are the tool of propaganda when needed.
Illusion or diversion ... Good Olympics.

Who have "vested" interests ( or who benefit ) with ( to spread ) version "A" ?

@Conf_iture
Otelli is not a proof!
re:Ste Odile : Otelli did just forget to say that
1. on the official IAC (instrument approach card) the mountain BLOSS was enlowerd from 2710 FT to 2500 (the airbus was found af 2620 FT!)
2.on the Air France approach chart (non official, not ICAO conform...) the indication "FAF" was missing ,doing mandatory to maintain 3660 FT QNH to 7 NM to STR VOR! (the aircraft has been found at 8.2 NM)
3.obstacles protection for VOR/DME TACAN approach was 15 per cent ground slope for ICAO,French,civilian or military regulations what the aircraft was doing (220 KT, 3300FT/min...

BEA lied "better" : the Lawyers and Judge could just not understand their slang and BEA said nothing about the 15 per cent rule!

Experts of the Court explained the law to the Judge (d'instruction) with Icao texts in english and not official French text (existing) (The Judge Guichard had learned latin and greek, and "english was not his cup of tea"...

Really Reader's Digest is better than Otelli's books

CONF iture, jcjeant;

I don't wish to divert the thread into yet another rabbit trail but from your comments I don't think you know much about media, especially American media.

Your responses tell me that you are too closely focussed on your particular issues with the BEA and Airbus and quasi-conspiratorial theories to permit a broadly-based understanding of media.

Regardless gentlemen, the debate concerning cockpit video has been done to death even at PPRuNe. My thought was, instead of providing the go-to for the PPRuNe discussion for slats11, I provided some of the well-known and accepted reasons why cockpit video is likely a non-starter, only some of which relate to the issues surrounding media.

Don't be offended, it's just the cut and thrust of discussion :-).

For what it's worth, I think the BEA have been very impartial and
thorough, not only in the analysis of what happened, but also the
recommendations, of which there are many and in a subtle way,
point fingers of responsibility in several directions. The fact that they
are not even more explicit is most likely for legal reasons, but there
is definately an air of almost irritation in parts.

This is the series of threads that refuses to die and now that the
final report has been issued, might it be a good idea to analyse
the recommendations in terms of what they are really saying ?...

Sorry for not been sufficiently clear in my previous post.

I understand the debate and genuine concerns about this. There is a need to balance competing interests, and this means that legitimate concerns must be addressed.

I was not suggestion that video be routinely recorded and made available for internal company review.

For example:
1. It could be triggered by certain parameters, but would not be routinely recorded. This would be different to the routine CVR.
2. If it was too difficult to activate selectively, then an option would be that the video would be routinely erased by the pilots at the end of the flight.
3. Video should not be used for other purposes such as performance management. There are other mechanisms for this.
4. The video should not be easily accessible. It could be stored on a recorder like the CVR and FDR, and only accessed by the appropriate investigative body in event of an incident.

However, I don't believe this suggestion should be automatically rejected out of hand.

Like it or not, there will now often be video recorded of anything abnormal. If there is an engine failure or a fire, or a diversion, or if the oxygen masks deploy, or if the cabin crew are seen to react abnormally, there will be plenty of video captured by the passengers. On previous occasion, this has been emailed to friends and the media as the plane landed.

Seen against this background, the view that there never be video recording of the cockpit seems a little.......quaint.

I've already said my piece on this, .....and I'll repeat it. Everybody in this Godforsaken business eventually hides behind the law that places final responsibility in the pilots hands. It's easier to escape responsibility when you can blame the dead guys.

TTex600
Your comment is paradoxical
Indeed .. if Airbus had presented the information loss of speed in such a way that it is comprehensible in 1 second (and not polluted by other information cascade) the dead pilots of AF447 can be the only responsible for the accident
But as noted by RR NDB. this is not the case and therefore .. one can not blame 100% the pilots ... and it's appears to in filigree (read between the lines :) ) in the recommendations of the BEA

Recognition of the problem from the git would NOT place blame entirely on the crew. One still has deferred, fubared, and AD-d Pitots, an ill understood protocol for mitigation, Altrnate Law that likewise did not arrive and be identified, utter lack of associated hands on high altitude flight, etc. ......

Yet the same tired and tiresome parrotted script. Pilots effed it. Beeyess.....

I do recognize honesty in the report, and I applaud it for the spread of their findings of responsibility.

Progress, but an opportunity to actually reach the goal of unbiased curatives wasted again.

This thread feels a lot less like the Alamo lately, Tex...But Santana still has too many troops.

So, what is this supposed to mean? Is it the BEA and all the other accident investigating bodies that participated in the AF447 investigation muffed it by not issuing a perfect report in your view? I just don't see your logic at all...

US media are not different from any others : They tend to concentrate in the hands of fewer. And monopoly is not the best tool for questioning, is it ?
But let's pretend I don't know for a moment.

AF447 autotrims in the stall.
Is it desirable to autotrim in the stall ?
I can understand someone like DozyWannabe would find any 'good' reason for the BEA to not address the subject, he has no experience, but you have a lot.
Please, would you explain, as you know how to do it, why it is not desirable ?

Or is it a 'quasi-conspiratorial' attitude to question why the BEA is just ignoring the fact.

Those who want to make sure it is all on the dead pilots shoulders in a way we don't question anything else.
More read and accept le figaro than scrutinize a BEA report.

Where would go your vote A-B-C-D ... and why ?

Causal, non causal, fault tolerance and graceful degradation
 

I´ll repeat: It is VERY EASY to diagnose UAS before the garbage coming from the Pitot´s destroy the confidence of the crew in the machine.

And it seems airbus SAS had no competence (an absurd approach imho) to implement a better solution. Delegate was natural. Easy, no investment.

And it would be easy, later to blame dead guys.

But,

If the man-machine interface (a good one, assertively) presented a clear indication the history certainly would* be completely different.

Who would ignore or not believe an alarm (garbage coming from the sensors) AND resulting anomalies like degradation.

I insist, F-GZCP had:

1) A ridiculous design WRT UAS (no redundancy)

2) An absurd man-machine interface behavior aggravating a "minor" issue (a brief probes "cold")

Result:

A critical and dangerous System. And the POB of F-GZCP paid it.


(*) Specially if training was developed to this scenario. BTW a very simple training.

I´m sorry the way i put. It´s sad. IMO the accident was designed by Airbus SAS. The victims were the operator (with some % of responsability, obviously) and all POB.

The crew errors (all, like CRM, etc.) occurred after two very serious design issues.

Technically speaking i repeat: It is very easy to detect the UAS onset. And even to (automatically) disable the subsystems in order to not process garbage and feed it to non properly trained pilots.

Therefore the effective System (plane+crew) could benefit from a non causal approach: It´s output could occur before it´s input. Why? Because after UAS detection you block the garbage to contaminate the System. A System with several other problems as the recommendations suggest.

From the outset .. I will remove the possibility B
Indeed .. these books are a financial income of Otelli
A change on his part (if discovered) cast discredit on his work and his future as a writer would be seriously compromised
Would he take this risk ?
Remaining possibilities A and C
A = a leak that can only come from someone with direct access to the CVR (BEA- Airbus-AF- or other body accredited with the BEA for this survey)
I think the leak coming from A can be a modified version as the person responsible for this leak is not likely to be discovered easily (I suspect that the leak was not directly transmitted to Otelly .. but instead .. via a journalist .. for benefit of french protection law for journalists sources IMHO)
If Otelli is tortured :) for give the name of his source and he give a journalist name .. it's the end of enquiry cause the sources of journalists can't be revealed
The possibility C can be envisaged . .. but unlikely because BEA knowed what was in Otelli book before release of final report (and do not forget the other accredited parties) and because if it's a inquiry about the BEA .. it will be soon discovered if the BEA published a script who does not match the CVR oral recording
This is why I think it's A the most possible option
This option benefit to Airbus principally ....

That is a bit simplistic, ehh?

It only autotrims into a stall if:
1) A serious degradation of systems occurs (Alternate Law 2). (This is by definition a clearly exceptional case) and
2) The pilots continuously (for quite some time) point the Nose of the Airliner where it abso-f$&§ing-lutely (I saw this one in another thread and liked it so much I had to borrow it :O) doesn't belong.

Indeed as a behaviour itself it is not really particularly desirable but this should normally really not be a problem. Why it became a problem in this tragic case is one of the reasons that this thread is doing so well 3 years after the incident. It is simply mysterious and shouldn't have been expected to happen to an experienced Crew. Yet it has.

replaced by post #962 :}

Confiture, if one tries to program the robot to once again interfere with the flying of the aircraft for that exceptional case, the risks that the code in question causes a serial or unrelated error is non trivial, when you are discussing what is being done to over ride a pilot's input to Flight Controls!

In the case of Alt Law 2, it appears that A330 control logic allows for the chance that the robot is wrong, since it already notes some odd signals or sums already, on the way to degradation. That's a conservative approach, and one to my liking.

When degraded already, if the stall warning goes off, but the pilot makes "input x" in certain modes, the robot will apparently concede that most likely, the pilot knows best since it knows something it uses for control is wrong already.

I am more comfortable with that than the alternative, even though every so often the pilot is wandering off the reservation.

That's part of why airline transport aircraft have two pilots.
Redundancy of wetwear has proven to be of value for some decades. :ok: Saved my bacon more than once, in a different environment.

Feel bad about some of the discussion here lately regarding personal views and BEA/AF conspirancy stuff, etc.

I like the discussion of stall entry/prevention training and such. IMHO, it is needed in the "heavy" pilot community, especially those that never routinely few at the "edge" of the envelope or beyond. Don't know how to do it in the sim, but the AF447 tapes show us a decent data point, including what "not to do" when the airspeed is deemed invalid by HAL. The sim trainers should exploit this data.

All of you know by now my feelings about use of AoA for both the flight control laws and the displays. 'nuff said.

The alerts and warnings and chimes and such seem too much to this old pilot. Seems that the most pressing problem should get the most attention. Also seems all the cosmic flight director stuff should simply present something that was last in existance when things went FUBAR with the airspeed sensors and their inputs to HAL. Personal preference is a simple attitude command or nothing at all.

I am also concerned with the aircraft commander briefing to the pilots in the two seats. I did not hear/read a clear delegation of authority. I an also disappointed that the guy with the most experience did not exert his seniority ASAP. Not that all more experienced pilots are better, but this case may have had a different outcome. We shall never know.

Hi,

Gums
I think that from this moment we can say that the experience and authority of the commander are no longer in the cockpit

PNF
2 h 12 min 12,9 We’re pulling
2 h 12 min 14,4
What do you think
about it what do you
think what do we
need to do?

Captain
2 h 12 min 15,5
There I don’t know
there it’s going down

I only flew one 5 day trip to Sao Paulo with a 3 pilot crew and it was only the fact the two FO's were outstanding pilots that made me feel comfortable taking my break.
Once in a while I felt I could not leave the cockpit for a Lav break so made sure it wasn't required because I didn't trust the other guy to handle a problem. I probably wouldn't have left the cockpit with these two guys flying. Neither one seemed qualified to be doing their job.

jcjeant,

Greetings, Sir. You have quoted the Captain, Marc Dubois....

"There I don’t know,
there it’s going down...."

PNF Robert has asked the Captain essentially a specific question about a specific problem, what is that problem?

I believe the problem was getting the Nose down. We all complain about the PF: Why does he pull back so much...? So do the two non flying pilots address the chronic Nose up in this exchange? I think, yes.

Are they merely preoccupied with the Pitch, not knowing Cedric is pulling? Or do they know he pulls for some other reason? Without question they know their problem is Stall, the two unmistakable cues cannot be avoided, Nose Up, and descent.

The message is clear: unto their demise, two pilots cannot figure out why the Nose is Up. If it is because they do not know Bonin pulls, then at least in this aircraft, the method of death is a SideStick that is not visible to other than the one pilot using it.

Solution? For any aircraft that may crash in this manner, the SideStick must be relocated. The problem will be how to identify which aircraft might die this way?
And then reconfigure the sticks. That will be complex, is there an easier way to prevent this stick arrangement from killing more people?

What a dilemma....