AF447 wreckage found

[goldfish85]

Basil:

At last a reasoned comment. I have reviewed most of the posts and have concluded that we should all wait for BEA to review the data and, at least, provide a complete time-line and FDR traces.

[SeenItAll]

To re-enter the translation discussion ...

Language is often quite flexible, and requires context and inflection to discern its particularly-intended meaning. The BEA transcripts give us neither. We don't know the precise context in which these phrases were uttered -- because the preceeding and trailing conversations are not provided; and because we only have the BEA text and not the actual CVR recording, we don't have the tone and inflection.

Once (or if) the full CVR is made available, then we may better know exactly what these pilots really meant or were doing when they said these things. Of course it is also possible that the BEA selected its particular English translation because it thought that its selected words conveyed better the complete nuance of the conversation (for which they know the context and inflection) than the would be conveyed by a literal translation of the French.

But I will note that it would be hard, on the basis of the BEA listening to a CVR, for it to conclude that the captain "attended" the briefing if he said absolutely nothing.

[Good memories]

two important views !

Please compare post 1219 and post 1491. We are dealing here with two very experienced men . The Shadow's possible analysis complies a lot with the experience of a many hours 330 skipper. In 1491 once more is brought forward the design of stationary throttles. From the beginning of the 320 design this was a mistake according to line pilots from various companies, Airbus interviewed.

To my best knowledge in the very beginning of the 320 ,Indian Airlines had a accident at Madras during push back and engine start . There the tow pin broke and the tow bar sheared the ground/air wiring on the nose strut. The acft. thought it was in the air with 0 IAS and the computer gave GA thrust without the pilots noticing it. It ended up with it's nose in the terminal.

Imhu if we continue flying these planes and we will, training and pilot's system knowledge should get more attention. Yes it will cost a little more money.

[RWA]

I have reviewed most of the posts and have concluded that we should all wait for BEA to review the data and, at least, provide a complete time-line and FDR traces.

Just so long as we don't have to wait as long as last time, goldfish85.

If you've noted my post 1486 above, the BEA analysed a remarkably-similar accident that befell an Air New Zealand A320 in November 2008. But, at first, they issued only a pretty 'non-committal' report that very closely paralleled the recent 'note' on AF447. It wasn't until September 2010 - many months after the AF447 accident - that they issued their 'final report' on the 2008 ANZ accident.

OK - I'll stick my neck out. I think that the evidence provided by the Air New Zealand accident, plus AF447, already provides enough evidence to point to an urgent need to review both aircraft systems (particularly the operation of stall warnings, and such things as the THS 'default mode') and pilot training (particularly the difference between 'stall avoidance' and 'stall recovery').

And that such measures should be embarked upon next week at the latest. NOT in a year's time or more, when the BEA eventually issues its final report.

On the face of it, as things stand, the same sort of accident could happen again tomorrow?

[bearfoil]

Yes on the face of it. Likewise another airburst of a Rolls, a rollback of a TRENT, a runaway THS, etc. Not that that will happen. What's missing is an open disclosure of the mods, the research, and transparency to the flying public.

Oh.....and the ODDS. Informed decisions are the only kind that mean a thing.

[Lonewolf_50]

[quote+EMIT]
In general.
On the subject of pitot icing: yes, it can happen, on Airbus as well as Boeing.
TAT probe icing happens as well, probably more often than pitot icing. It can be observed from the TAT anomaly phenomenon (again, read the Boeing Bulletin). For crystal icing to occur, you do not have to fly through any solid wall of red, just the light green outskirts will suffice. [/quote]
1. If TAT probe ices up, how much impact would that have had on the pilot's observation during the crew brief (all three pilots) in re temperature not changing as forecast? How much impact would that have on the T value used by the ADC for various computations, and thus both displays and inputs to the Flight Control Computers?

2. If AoA vane/probe ices up, is there a warning associated with that which crops up on ECAM? I'd not expect an iced AoA probe to be noted in ACARS for maintenance, since AoA would most likely un ice as the aircraft descends. (Put another way, flying in and around ice isn't a maintenance fixable issue in re AoA performance in flight. )

If AoA is suffering from ice degradation, is that of any use in discussing this mishap? It appears that the AoA probe kept feeding FDR data all the way down. Is there any reason not to assume the AoA dat reliable?

[RWA]

On the subject of pitot icing: yes, it can happen, on Airbus as well as Boeing.

Just as a matter of interest, Lonewolf 50 mate, whoever said that it can't happen to both or either company?

[ChristiaanJ]

Originally Posted by Lonewolf_50

2. If AoA vane/probe ices up, is there a warning associated with that which crops up on ECAM? .....
If AoA is suffering from ice degradation, is that of any use in discussing this mishap? It appears that the AoA probe kept feeding FDR data all the way down. Is there any reason not to assume the AoA data reliable?

It would seem the AoA sensor on the A330 is the classic "vane" (not a "probe"), so icing would at the most cause some "asymmetry" (not fully calibrated value), and with both AoA sensors affected, probably not even enough to get a "vane disagree" (i.e. a comparator trip).

Please, let's not confuse AF447 with Perpignan, where it seems, during washing on the ground, water got into the AoA sensor part, which then froze at altitude, stopping the vane from rotating freely.

Is there any reason not to assume the AoA data reliable?

Depends on what you call "reliable". The "AoA data" finally are nothing more than the angle of the vane with respect to the fuselage.
They can be corrected for AoA + IAS + Mach, etc. only within the flight envelope explored during test flights....

I would suggest the "AoA data on the FDR" are likely to be good enough to be accepted as a first ballpark figure, until we get the full records.

[goldfish85]

We should review the incident to an A-340 (with the same control logic) which experienced a zoom during cruise over the North Atlantic. While this was reported as an near-miss incident, the circumstances have unmistakable similarity.

The A-340 was in turbulence and entered alpha-protection which, with no stick input commands alpha-prot (about 4.2 deg at M=0.6). Achieving this resulted in a zoom of about 2300 ft (as I remember). The angle of attack reported in the latest BEA summary is about the same.

I'm traveling, so I can't refer to the report, but the incident was to A-340, TC-JDN on 2 October 2000. The report was AAIB Bulletin 6/2001.

[Lonewolf_50]

RWA, I went back a few pages to try and figure out the context of that which you isolated, and could not find the post. I think what was in my mind was that we have had a number of different mishaps and incidents discussed here, and pitot anomalies have been featured in either family of aircraft. (As contributing causes to incidents or mishaps/crashes).

I am not sure if I sensed an AB bash and was responding to that, or why I phrased it just so.

So I can't answer your question, as I am not sure what post (in total) you were referring to.

[TioPablo]

@ Goldfish85

There you are:

http://www.aaib.gov.uk/cms_resources...%2006-2001.pdf

[captdaddy]

1491-Excellent analysis regardless of accident findings

As much as the author of 1491 has spent in the Airbus , I have spent in Boeing and Douglas aircraft and although I finished my career( 26,000 hrs) in glass cockpits I consider myself a "steam gauge" pilot. In the days when we have the glass all over the place the problem in recreating this scenario will be determining "what the pilots saw"....what information was given to them and what screens were just blank?
Not too long after the Delta L-1011 crashed at DFW the airlines came up with an AAMP (Advanced Aircraft Maneuvering Program) that was meant to show the pilots of transports how to get max performance out of their aircraft...one thing we at our airline found out that when the sim was put out of the normal flight envelope and then handed over to blindfolded flight crew they were able to recover the aircraft by reference to the attitude indicators and airspeed alone....EXCEPT for our Airbus sims...all the screens went black with a big RED X across them...apparently since they were outside their flight envelope ...I say apparently because we could never get a response from Airbus on why this occurred except the same caveat I read in response to the professor from University of Berlin ...face it , if any fault is laid at the Airbus door for bad design/ cause of accident, that will be the end of Airbus....and the EU will not stand by for that....blame the pilots , they can't talk back.
There's been enough doubt cast over the whole Airbus laws of control to warrant a serious review....One thing I do know....if I wanted to split -s a 747 from FL 410 , it would roll over and do what I asked it to . Nobody here has shown that the Airbus would or could. I am not trying to say that they don't have some great features but for 3 and one half minutes three very qualified pilots tried to avoid dying and were unable to do so because either they didn't know what was REALLY happening or because they couldn't do anything about it. Either way AIRBUS owes them and us a better answer than WE need more training...maybe they do in how to design Pilot's airplanes...

[Mr Optimistic]

Was this significant in terms of priorities and delays,

the commander took manual control of the aircraft because neither autopilot would engage.

[WilyB]

As much as the author of 1491 has spent in the Airbus , I have spent in Boeing and Douglas aircraft and although I finished my career( 26,000 hrs)...

Daddy,

I have zero hours in a simulator, but I am pretty positive that simulators can only accurately reproduce the real life data they have been fed.

Meaning that unless Airbus or Boeing deliberately fly an aircraft outside its envelope, and record the data, the simulator will not be able to reproduce accurately the aircraft reactions.

[Man Flex]

Guys,

The mid-atlantic A340 zoom climb has little bearing on the AF447 accident. Try reading the reports.

The high speed protection was triggered by turbulence and the aircraft did what it was designed to do.

The AF climb was pilot induced, it's there in the report quite explicitly. Why the pilot made nose up inputs is the real mystery here and it may possibly have something to do with his airspeed reading.

The high speed protection was disabled when the aircraft went into alternate 2 law and was never available.

[bearfoil]

"...The AF climb was pilot induced, it's there in the report quite explicitly..."

No it in't. BEA made NO conclusions, they only reported data. YOU say it was PF induced, conclusively. There is a mild difference.

[IcePack]

Good post 1491 (Basil) interesting in that the upper air handling is a challenge. BEA say trim was 29% usually that would be 38% I.e more aft.. That may be because of the AF configuration.. (note the cofg moves aft in flight due fuel transfer to tail) which makes the a/c even more sensitive in pitch.
I agree IMHO a very very difficult situation when one takes surprise/shock into account.

[captdaddy]

Sim programming

WilyB,

The sims have a sets of equations describing everything from Engine Performance to turbulent weather to day/night to ACARS ...the data points to create that would take decades if they were collected and then programmed. The whole idea of the sim is it will do exactly what the plane would in the real world....if they had to program captured data points that means that the below glide slope approach some guy flew in the sim wouldn't be possible unless they'd captured the data points beforehand....the whole reason they can train pilots to fly the 787 or 380 before the plane ever flies is because they have detailed equations to describe all phases of flight....what happened to our Airbus sims is that the equations only described what the programmers in Toulouse expected pilots to see in normal flight regimes.

[Turbine D]

Man Flex,
I agree with you.

From the AF447 report: From 2h 10min 05sec, the autopilot then auto-thrust disengaged and the PF said, "I have the controls". The airplane began to roll to the right and the PF made a left nose up input. The stall warning sounded twice in a row. At 2h 10min 16sec, the PNF said, "So, we've lost the speeds" then "alternate law". The airplane's pitch attitude increased progressively beyond 10 degrees and the plane started to climb."

It certainly sounds "quite explicitly" to me. It is reported data, no conclusion necessary.

Relative to the mid-Atlantic A-340 zoom incident, I agree that was different in many respects. To begin with, erroneous speed was not the reason for the AP/AT disconnect, as was the AF447 AP/AT disconnect. For AF447, one of the first fault messages with cockpit effect was "PROBE-PITOT 1X2 /2X3 / 1X3 (9DA)", and IMO, this is what caused the AP/AT disconnect and lead to the cascade of subsequent events.

[bearfoil]

"From the AF447 report: From 2h 10min 05sec, the autopilot then auto-thrust disengaged and the PF said, "I have the controls". The airplane began to roll to the right and the PF made a left nose up input. The stall warning sounded twice in a row. At 2h 10min 16sec, the PNF said, "So, we've lost the speeds" then "alternate law". The airplane's pitch attitude increased progressively beyond 10 degrees and the plane started to climb." "

Show me where, exactly, this concludes (or even says) that the Climb was initiated by the pilot. Careful.