AF447 wreckage found

[aguadalte]

Yes Dozy, thanks for the reply. I meant a procedure training, (already edited my post) i.e., was it common to train the UAS on Flight Sims, at that time.
I just went to my old (1987) version of B737 QRH and looked at the procedure of that time, the basics are the same of course, but this was what was written then, (compared to A330/340 three pages checklist...):

AIRSPEED UNRELIABLE
AIRPLANE ATTITUDE/THRUST.......ADJUST
Adjust attitude and thrust to maintain airplane control. Attitude and thrust information is provided in the performance chapter.
PITOT STATIC HEAT.....................CHECK ON
MACH/AIRSPEED INDICATORS......CROSS CHECK

Pretty simple, hen?

[Feathers McGraw]

Saturn

Thanks for that, hadn't noticed the Tech Log thread, off for a read now.

I shall read the translated report ASAP.

[Graybeard]

Sebaska

Stall AOA is ~7deg when high&fast and ~16deg when low&slow.
Besides, AOA vanes are designed and calibrated for situations when air comes from generally front direction, not side or bottom. AOA was severe enough to cause Pitot readings off, variometer indications got flaky, etc.

Again, I'm quoting DC-10 SW from AOA, because I have the data handy. Its SW from AOA is modified by position of flaps and slats, not "low&slow and high&fast."

According to a prior poster, the mechanical limit of the AOA vane is something like 110 degrees NU. BEA reported FDR trace of >30 degrees AOA, a region where the SW was silenced.

I cannot understand the logic of inhibiting Stall Warning based on low airspeed, unless it's meant as a redundancy to squat switch (WOW) discrete. It would be better to make the WOW more reliable, than add complication to SW logic, but that was an engineer's decision not vetoed by the AB pilots nor certification authorities.

[jcjeant]

Hi,

Since the publication of interim report No. 3 .. all comments are about the training of pilots and also pointed discussions on more technical subjects as well as word by word analysis of the known parts of the CVR
This is all very well and certainly contributes a better understanding of the BEA report.
One thing that seems to be masked by these tests .. and this is the problem of pitot tubes
In its report No. 3 .. BEA speaks very little about the pitot tubes and in its conclusion ... he says that "probably" the Pitot tubes are at the origin of the loss of signal indication and thus the transition from the normal law has alternate
So the BEA does not formally recognize that the pitot tubes are the cause of a loss of airspeed indication.
These Pitot tubes are a very important issue and no training even extreme will solve it ....
Pilots must not have a mission to supplement by their actions a construction defect.
Imagine:
You own a car whose brakes do not work starting from a certain speed.
You can see .. and having escaped the accident you request an explanation from the manufacturer.
It will answer that this is not a problem .. It will give you a free training so you can effectively use the parking brake to help you stop in the abscence of main brakes operation
Are you satisfied with that answer? is what is normal?
Something else ... BEA notes that there is a problem with the crews point of view is .. qualifications .. reliefs on the flight deck... etc..
Why BEA in its recommendations does not require an immediate review by AF (and also other companies) of the composition of crews for this type of flight .. and to be applied immediately.
Possible to immediately implement and do not requires extensive feasibility study.
Afterwards .. regulators may make laws ... but why to wait?

[DozyWannabe]

Originally Posted by Graybeard

I cannot understand the logic of inhibiting Stall Warning based on low airspeed, unless it's meant as a redundancy to squat switch (WOW) discrete. It would be better to make the WOW more reliable, than add complication to SW logic, but that was an engineer's decision not vetoed by the AB pilots nor certification authorities.

I suspect it was not a deliberately designed-in behaviour so much as a point so far outside the envelope that it was not considered. >60kts, in a heavy, at cruise altitude?

[Lonewolf_50]

A previous quote ...

Personally I believe that no pilot would maintain a pitch attitude of 16 deg at height intentionally it would look too weird

In darkness at altitude over an ocean, possibly in cloud, no visual reference to "look" weird.

It also looks weird on your artificial horizon if your intent is to be flying level.
Really, it does.
The dark side (bottom circle) or the AH isn't visible, all you see is the light semi circle ...

PEI 3721, post 2367: http://www.pprune.org/rumours-news/4...ml#post6610842
Nice post, food for thought. (And no mayonnaise, on that, thank you! )

carlos: disagree

This is what should be:
Sound of the stall warning should mean STALL
Not sound of stall warning should mean NOT STALL

Stall warning is a very handy tool in preventing a stall. It lets you act while you still have control of all of your flight surfaces. You get to make a correction while you are still flying, rather than once you have started not flying, but more or less falling ...

Loose Rivets, in re spurious warnings.

I bet your instrument scan was working, since you were flying ...

Denise Moore:
Did the pilots say nothing during the last three minutes? Because I have not seen anything mentioned.

Head over to tech log, a poster (gpc62) did a best effort English translation of what the BEA have released from CVR. It is as he says "unofficial CVR transcript translation - plus VS" (Stall warning)

http://www.pprune.org/tech-log/45687...ml#post6610403

It is some chilling reading.

[Mr Optimistic]

'Extracts' from the CVR start on page 91. It does say 'extracts' so there is more, but not released.

[jcjeant]

Hi,

I suspect it was not a deliberately designed-in behaviour so much as a point so far outside the envelope that it was not considered. >60kts, in a heavy, at cruise altitude?

Well .. when a aircraft stall .. ,he can go anywhere .. and certainly fare outside the envelope.
It's not something discovered after the AF447 crash ... it's known from ages....

[DozyWannabe]

@jcjeant - Fair point, but you're forgetting that engineering is the art of compromise in order to achieve specified goals. As far as I know this particular A330 is pretty much the only aircraft of its size and configuration to end up in that situation, and right at this moment I don't see many pilots queueing up to test a similar envelope excursion in a B767, B777, DC-10, MD-11, A300, L-1011, or Il-8/96 to see how those systems fare in such a predicament.

[britfrog]

pitot failure is known to have occurred but this has no effect on the vsi , the pitot has two sources impact which can get blocked and static the vsi uses the static side
However this could all have been avoided had the crew a handheld gps to crossreference with as this wouild have given them an accurate speed readout as well as height. I have always carried one when i fly along with a handheld icom radio, as long as i have them I will never use them but you can guarantee the day I leave them at home something will go west. However it doesnt get away from the fact that the crew were not up to the job sadly

[MountainBear]

DW

@jcjeant - Fair point, but you're forgetting that engineering is the art of compromise in order to achieve specified goals

.

True enough. But with the authority to make compromises also comes the responsibility to take the heat and bear the costs when those compromises come back to bite you on the ass. Your recent responses in this thread are the perfect illustration of what critics of the airline industry mean when they refer to "engineering by gravestone."

I think the flight crew made compromises; I think Air France made compromises; I think Thales made compromises; I think Airbus made compromises. All those compromises seemed, I am sure, rational and logical and reasonable compromises to the decision makers at the time the decisions were made. The problem is that those rational, reasonable, logical compromises aligned with each other in a way that produced an outcome that was not rational, not logical, and not reasonable. AF447 should not have crashed....but it did.


While the final report is not out yet and so my mind might change, as events stand now I think that in 20 years AF447 is going to be studied as a classic case of system failure. Everyone shares part of the plane from the pilots to the airline to the the manufacturer and its subcontractors. Regrettably, no one really cares about that at this stage; all they want to do is point fingers and shift blame so that when all is said and done it is not they by some other sap who bears the financial cost of this disaster.

[lomapaseo]

No One Cares ??

While the final report is not out yet and so my mind might change, as events stand now I think that in 20 years AF447 is going to be studied as a classic case of system failure. Everyone shares part of the plane from the pilots to the airline to the the manufacturer and its subcontractors. Regrettably, no one really cares about that at this stage; all they want to do is point fingers and shift blame so that when all is said and done it is not they by some other sap who bears the financial cost of this disaster. While the final report is not out yet and so my mind might change, as events stand now I think that in 20 years AF447 is going to be studied as a classic case of system failure. Everyone shares part of the plane from the pilots to the airline to the the manufacturer and its subcontractors. Regrettably, no one really cares about that at this stage; all they want to do is point fingers and shift blame so that when all is said and done it is not they by some other sap who bears the financial cost of this disaster.

This is a surface perception gleened by reading too many discussion posts by people not in the actual chain

You can call it what you will ... even gravestone mentality .. citing a few fatal accidents over the years.

Engineering is not an exact science. Trades are made and lessons learned. But more importantly measuments are applied to avoid worsening our lives and to ensure that we are taking more steps forward than backward.

The confusion about who to blame for the accident du jour is in your mind. It has nothing to do with the learning process and the search for corrective actions that ensure more steps forward than backward. The regulators, the designers, the operators will look to the engineers to effect this process.

So ignore the rhetoric in discussion boards and blogs and look for immediate results in Service buletins, FCOMs, Training, product upgrades and maybe in 10 years a regulatory change.

The fact that you might not recognize the progress has a lot to do whether you are part of the process or not.

[DozyWannabe]

Originally Posted by MountainBear

Your recent responses in this thread are the perfect illustration of what critics of the airline industry mean when they refer to "engineering by gravestone."

My recent responses are of the who what now?

I don't know if I've given the wrong impression hanging around here for - what is it - 5-6 years now? But if you think I'm for automation über alles, airmanship being a thing of the past and an enthusiastic supporter of the race to the bottom then I clearly haven't been expressing myself properly, so please allow me to disabuse you of that notion.

I think that pilots should be pilots. I think that stick and rudder skills are an important thing that should be maintained and checked on a regular basis even if your day job involves flying a FBW airliner that has protections up the wazoo. I think that the current state of affairs in some airlines - and particularly regional US airlines - is worrying bordering on scandalous.

But I am also a software engineer. Not of the calibre that makes the software that runs the systems on aircraft, but I studied with and under the ones that were and are capable of that, and I cannot *abide* the false dichotomy that is frequently presented here blaming the evolution of automation and technology for the decline in piloting skills.

I love flying - have done since I first boarded a BCal BAC 1-11 at the tender age of two on a family holiday and every time since - I still get a buzz when I get on an aircraft even now my line of work makes it happen a few times a year, and nothing - but *nothing* compares to the first time I was taken up as an Air Cadet in a Chuckmonk and the friendly middle-aged guy in front of me said the magic words "you have control". I only wish I had the money to continue that pursuit and hopefully one day I will. Don't think for a second that I don't consider this a privilege.

This here, what we're talking about is an accident that could and should shake up a complacent industry - some good forward steps have been made but we must keep up a basic knowledge of airmanship among pilots. The advances that FBW brought forth are there to assist, but cannot be thought of as taking the place of proper piloting skill. Everything I've seen indicates that the PF did not even begin to understand the problem, let alone work out a way to solve it and if his training was deficient in that regard it needs to be fixed. I'm seeing failures of CRM, failures of communication and major gaps in basic aeronautical knowledge and I've got to say, I'm concerned - because I want the people in front to know what they're doing when I get on the thing and I wonder how many hours of stall recovery training could have been bought with the money used to furnish the airline executives' houses. This isn't just about Air France (although it would appear they have some major work to do), this is about the industry in general.

[BOAC]

Originally Posted by DW

the false dichotomy that is frequently presented here blaming the evolution of automation and technology for the decline in piloting skills.

- your last para was 'spot on' in my book.

Regarding the 'quote' - I think most of us, when we take that 'view', mean the way it has been trained for and implemented by operators rather than the 'luddite' postion to which I think you refer. I, for one, do see a definite 'decline in piloting skills'.

[DozyWannabe]

Originally Posted by BOAC

Regarding the 'quote' - I think most of us, when we take that 'view', mean the way it has been trained for and implemented by operators rather than the 'luddite' postion to which I think you refer. I, for one, do see a definite 'decline in piloting skills'.

Thanks very much BOAC, and I'm honoured. But if what you say above is really the case, why is this ire not more clearly directed? I know it's only a minority that jump on every Airbus incident or accident with tales of conspiracies and concierges - but at the end of the day if the airlines are abusing the tools that the engineers are giving them in order to cut costs then it is the airlines who deserve to be in the firing line - not the engineers!

There is so much that could be done with the co-operation of all parties to make it crystal clear to the industry that while the technological advances can improve safety and can keep costs down - this should not come at the expense of knowing how to deal with things on the rare occasions that things go pear-shaped, but so often on here I see such discussions break down into lamenting the fact that things aren't like they were in the '70s. Lord knows I wish we could undo a lot of what happened in the '80s but we have to work with what we've got.

[ChristiaanJ]

Originally Posted by MountainBear

I think the flight crew made compromises; I think Air France made compromises; I think Thales made compromises; I think Airbus made compromises.

In defense of Thales.... and possibly Goodrich.
(I worked for SFENA, then Sextant, now Thales, but in a totally different sector, so I've no axe to grind).

One thing that seems to have been neglected in recent discussions, is that engineers design to standards, and in particular certification standards.
(When Concorde was designed, an entire new "SST certification standard" had to be written, and I think much the same happened when the all-digital FBW came along.)

Todays certification standards for pitot tubes still date from the Stone Age.
Unless a real job is done to update those standards, and re-define the certification testing with regard to potential present-day (icing etc.) circumstances, "we" (the engineers) just try to do our best.... which, because of the lack of clearly defined design data, may not be good enough.

Both Thales (the 'AB' probes) and Goodrich seem to have tried to do "better than specs", but in the absence of new and more stringent specs and design data, how do you decide what is indeed "better"?

So maybe some of the "blame" should go to the regulators, or rather, there should have been a recommendation in the BEA report for the certification standards for pitots to be "brought up to scratch"?

[bearfoil]

The blame is so diffuse it must be borne by everyone. EVERYONE.

To say this was 1/10.000,000 is the height of ignorance. At every step in the denoument, and mark my words, there will be maths aplenty to show it to be far more likely than that.

Throw in the certain knowledge that the PITOT tubes were a problem, and that AF crew(s) are shown to be lacking in some skills re: problems aloft, AF deferral of R/R for Thales, and their lack of concern about hand flying at altitude, and one gets a better picture.

As for the Engineers, due to their mode d'emploi, I would say the best (and longest) odds are with their work. So Bravo? Mais NON.

It is precisely this assignment of RESPONSIBILITY that is traditionally lacking in an honest approach to safety,

Should I read AD's, SB's, and MEL's to my satisfaction at this point before boarding?

I will if no one else will. The industry is full of pretense and marketing.
As before, and again, mark my words:

The PROBLEM is SYSTEMIC. As such, I look everywhere.

"Bear, don't be ridiculous". Most will be satisfied with BEA FINAL.
It is 'tacit satisfaction' that killed 228.

[jcjeant]

Hi,

Another dumb question (and maybe already asked .. no remember .. and the search engine of this forum kill my nerves )

Why the AP must absolutely disengage when the system detect incoherent speed datas ?
And consequently ..why the procedure for IAS can't be take in charge by the AP for some time ... (pilots alerted) and give to pilots the time to put their gloves before touch anything ?

The fact that you might not recognize the progress has a lot to do whether you are part of the process or not.

Sad to tell .. but methink the 228 victims are part of the process .....

[TCU]

DW as a footnote to your post 2402, this extract from the L1011 website...also in memory of a certain much missed poster:

"Another interesting British Airways certification requirement that came up during TriStar production was FAA Stall Certification - in the US, the FAA requirement for the manufacturer was to only take the aircraft to stall warning (that is, when the stall warning horn sounded in the cockpit) and then recover. For European certification however, the requirement was to take the aircraft to full stall and then recover. So, Lockheed took the 1011 out and full stalled it - then it invited the FAA along and demonstrated to them that the TriStar could enter a deep stall and recover safely - then the FAA pilot tried it and in the deep stall the wing dropped and he could not recover it - film taken in the chase plane showed that the 1011 rolled inverted during the recovery attempt and finally was pulled out at around 10,000 feet - and the test started at about 25,000 feet! During recovery the aircraft pulled an excessive amount of G loads - the No. 1 test ship had to be hangered until it's structure was completely inspected before returning it to flight status.


On later British Airways training flights, I was an Instructor Flight Engineer and sat through some of these deep stalls in the actual aircraft - in the deep stall the aircraft shook so much you could not see clearly - the pilots just had to hold on to the control wheels and push forward to recover - it was quite a wild ride to say the least."

....push forward to recover

[BOAC]

Originally Posted by DW

I know it's only a minority that jump on every Airbus incident or accident with tales of conspiracies and concierges

- as I see it, there is a long-term 'whiff' of 'conspiracy' - that will take time to blow away. The 'Concierge' is something that has created enormous damage and will take even longer, since it has appealed to many - those who wish to learn to fly airliners without lots of effort, those who see the advantages of a 'cannot go wrong' system (namely insurers and accountants - the latter also seeing the savings in training) and management who can insist on no-one touching anything thereby making their jobs easier. There is no doubt that the FBW technology is the way forward. It offers big savings in economy with 'relaxed' stability and enhanced safety in normal operations.

As I have said on t'other thread, the industry needs to claw its way back to a sensible middle ground. Not all pilots agree, of course. As I have also said elsewhere, it is not the fault of the engineers/software writers - they can only work to the best of their ability to produce what the industry asks for.

My last 'elsewhere' quote is that I hope all AB pilots are having quiet contemplative thoughts about the way it works, or doesn't. I also believe this accident will be a big shake up to both AF and to a lesser extent AB

I still like your (oldish) idea of dropping straight to Direct Law in the 447 scenario, by the way - with a concomitant shift in training, of course...............