AF447 [Archive] - Page 7

lomapaseo

14th Jun 2009, 21:24

There is way too much blame being assigned here to a single entity be it a manufacturer or a single system.

The investigation is on-going and damn few facts are available to establish a causal chain. This thread is begining to look amateurish in its over simplification of what went wrong and who is to blame.

The idea that anybody involved with the investigation would not like to find the critical evidence (FDR/CVR etc.) is outlandish and demeans a supposedly professional internet site.

This is not about blame, lawyers or insurrance carriers, it's about finding enough information to prevent a similar accident.

Root cause is hogwash. Nobody can effectively eliminate any system malfunction whether it be pitot tube interactions with computers or a weather radar outage.

Our efforts need to be focused on minimization (of an occurrence) and mitigation (of the resulting hazard) To do this you need data on all the interactions that were at play in this accident including the pilots.

In today's machine you can not take the pilot out of the accident causal chain, yet nobody is even considering the what-ifs in this discussion. Remember for every system failure including the pilots there is a cause effect and that is ultimately how the corrective action will be accomodated.

ChrisVJ

14th Jun 2009, 21:30

JD-EE

Back in the days of the first Jets control runs became more complex and at higher speeds their weight became too much for comfortable flying so servos were introduced. The problem with servos was that they did not always feed the control surface pressure back to the pilot so an artificial "feel" was added to control columns. IIRC it was called "Q Feel" and was considered rather revolutionary at the time.

Q Feel was necessary so pilots could judge, ( based on the old direct systems) how much their control surfaces were deflecting or pressing on the air and secondly, as the control pressure became significantly greater, would not deflect a control surface beyond what was safe.

Then true FBW came along. The control inputs are fed into a computer which judges the amount of control surface deflection required for the indicated manoeuver and applies it. The computer also ensures that the control surface is not deflected too much for the conditions so 'Q' is not necessarily required.

Now personally this worries me a bit, but, as someone pointed out on another thread where we were discussing this, the difference is not the risk of electric failure but the difference in risk between electric failure and physical component failure eg. cable or rod bearing etc., in systems which are both complex. There is little evidence so far in statistical terms that FBW is any less safe than the old physical connection controls.

Pontius Navigator

14th Jun 2009, 21:31

wheelie my boeing (http://www.pprune.org/members/244426-wheelie-my-boeing)

In otherwords turbulence is incapable of moving the joystick despite it's significant moment arm and weight out on that moment arm?

I find that hard to believe.

Why?

Even your Microsoft Flight Sim pilot knows that early joysticks had no physical feedback. Only later games joysticks have feedback.

I don't know Airbus but there is no need for physical feedback. You place the joystick (computer mouse or whatever) in the direction you want it to go then you observe the effect. This may be an indication of control surface position, the movement an instrument etc. The feedback is visual and not physical.

JD-EE

14th Jun 2009, 21:35

Assuming, that no VHF coverage is avail. ACARS is communicating via SATCOM. To contact the satellite, the A/C has to know his position in regard to the satellite to adjsut the dish.
ACARS is linked with most of the computers , so it should be easy to track it by sending LAT/LON datas.

I believe it's been established there is no dish. They use Inmarsat as their carrier. And the antenna is omnidirectional for this mode.

JD-EE

14th Jun 2009, 21:41

Wheelie

Who is talking about feedback?

In severe turbulence, sufficient to damage a plane, is the pilot's seating secure enough that the transient accelerations of the plane are not going to move the pilots arm out of the pilot's control in response to the accelerations?

Will a stick with no hand on it stay perfectly still poking up there in the air above the mounting point as the plane accelerates with rapid jerks in random directions?

This is independent of whether the stick's output is controlling the plane or not. It's a question about whether under very heavy turbulence the input to the stick from the effects of the turbulence is stable enough to fly the plane. If not and the pilot HAS control isn't it possible that the bouncing around could lead to improper or damaging imputs to the flight control system?

Shore Guy

14th Jun 2009, 21:43

Could someone explain how stall warning is triggered in the A-330? There is reference to it being tied into air data information....is there not an AOA probe?

Thanks in advance....

BOAC

14th Jun 2009, 21:48

ISIS - essentially for the last week or so I have been trying to establish whether the poor souls would have ANY attitude reference left with all the 'failures' - with little success! WITHOUT structural breakup, they would have a fighting (but difficult) chance using standby attitude and power (let's assume the standby speed is knackered too). Without an independent attitude.................. I'm very much with aguadalte #1507 on this.

Backtracking a bit

Picking up on the ISIS:

A33Zab (#1518 ish) - Thanks - full freedom in pitch and roll.

captainflame IT ALSO GET'S INFO from ADIRUs 1 and 3 ! (#1523 ish) - 'how much and what' could be significant with those 2 down! Not quite a 'standby'?

Mad FS So if you were to have some kind of catastrophic design flaw that took out every ADM the ISIS should still be running. I don't think the air data system designers are as clueless as you seem to think. (#1527 ish) and The backup - ISIS - is dissimilar. (#1535 ish) - as barrymung says - you have to wonder? Dissimilar but with common inputs?

NotPilotAtALL

14th Jun 2009, 21:51

Hello,

There is way too much blame being assigned here to a single entity be it a manufacturer or a single system.

The investigation is on-going and damn few facts are available to establish a causal chain. This thread is begining to look amateurish in its over simplification of what went wrong and who is to blame.

The idea that anybody involved with the investigation would not like to find the critical evidence (FDR/CVR etc.) is outlandish and demeans a supposedly professional internet site.

This is not about blame, lawyers or insurrance carriers, it's about finding enough information to prevent a similar accident.

Root cause is hogwash. Nobody can effectively eliminate any system malfunction whether it be pitot tube interactions with computers or a weather radar outage.

Our efforts need to be focused on minimization (of an occurrence) and mitigation (of the resulting hazard) To do this you need data on all the interactions that were at play in this accident including the pilots.

In today's machine you can not take the pilot out of the accident causal chain, yet nobody is even considering the what-ifs in this discussion. Remember for every system failure including the pilots there is a cause effect and that is ultimately how the corrective action will be accomodated.Very interesting and informative post about different matters ..

Now .. if we give a raw sight on the aviation accidents statistics .. the figures show .. in almost 100 accidents .. pilot error is responsible for 80 .. so a average of 80% (nothing so far proof this new accident will inflate this 80%)
This average was different in the past .. but cause the progress of technologies .. technical failures are reduced at best engineering can.
So it's seem's that the corrections for avoid accidents work mainly when applied to technical parts and not so best on humans.
So we can conclude .. the technology progress (sometime with big jumps) but unfortunately the human stay the same .....
It's possible the strong response for avoid at max the possibilities of accident will be to reduce the most possible the human actions in a plane .. and ultimately (after some more big technology jumps) remove humans pilots from plane ??

Regards. http://smilies.sofrayt.com/%5E/a/bye.gif

JD-EE

14th Jun 2009, 21:52

Pontius, I think you misunderstand me.

I am envisioning a joystick that is simply sitting up in the air. Pick one for your favorite game. Hold the base. Now move it rapidly left and right. What is the effect as read on your computer screen? Does it stay put? Does it move as if somebody had their hand on the stick and was exerting control?

Experience suggests you will find that your computer gets input that it does not distinguish from that you might have given yourself. Heck, you can see the stick flopping around.

Now imagine it runs your automobile. Forward means go forward, the further forward the faster. And so forth. Now suppose you hit an obstacle, a speed bump. Your body jerks forward. Your car lurches forward into a crash. This is not a problem if you're sitting comfortably with no sudden accelerations over a small fraction of a G. But add in jerks from turbulence and you might find your arm involuntarily inputting a control surface change that was out of the range the plane could tolerate.

Meanwhile, what are your feet doing as you bounce six inches up and down? Might such a bounce have led to the pilot overcontrolling in several of the accidents attributed to "stupid pilot over controlled the plane?"

(Maybe a high-G accelerometer set needs to be mounted in the cockpit so that the data recorders can note that it's not the pilot's fault he could not control the plane, nothing could.)

ClippedCub

14th Jun 2009, 21:59

and ultimately (after some more big technology jumps) remove humans pilots from plane ??

You don't want that in passenger planes. The software specification writers can't think of all combination and sequence of failures. Besides, computers gives up when they reach, 'that does not compute', the pilot never gives up.

Some of those pilot error causes are due to poor training, and should be labeled training requirement error - like Colgan 3407 for instance.

h3dxb

14th Jun 2009, 22:02

@JD-EE

I believe it's been established there is no dish. They use Inmarsat as their carrier. And the antenna is omnidirectional for this mode.

Believe what U want :}

For antenna control steering and computed doppler correction, the following ARINC 429 octal labels are transmitted from the ADIRU to the SDU:
310 Present position - Latitude
311 Present position - Longitude
312 Ground speed
313 Track angle
314 True heading
324 Pitch angle
325 Roll angle
361 Inertial altitude

This is required for:

The high gain antenna is an electronically steerable phased array. Simultaneous transmission and reception of satellite signals (full duplex operation) provides two bands of operation: receive (Rx) band, and transmit (Tx) band.
Beam steering of the antenna is performed via serial transmission of phase shifter data from the BSU. RF signals are transmitted to and received from the D/LNA unit.
The high gain antenna provides +12dBic nominal gain with near hemispherical coverage and enables transmission of high rate data and voice communications.

But U can ask me every time :ok:

Graybeard

14th Jun 2009, 22:16

Thanks for the translation of the TCAS Fault message, h3dxb. Now, let's see if your manuals can confirm or refute the following:

Ref: "There are five official, damning reports on A330 and A340 aircraft operated by major airlines suffering complete failure of all flight instruments in flight. Pitot probes iced up in meteorological conditions at night, by day, in cloud and in clear air."

The early failures reported by AF447 ACARS all seem to be explainable by iced pitot tubes, save for the TCAS Fail. TCAS does not use airspeed, just altitude via the transponder.

Air data Altitude failure will cause the transponder to revert to Mode A. I'm not sure if it would report failure of its altitude input to the CMS/ACARS, but it didn't report a Fault.

Transponder reversion to Mode A will not cause a TCAS Fail condition, but a TCAS OFF condition.

Therefore, we have a TCAS Fault that occurred nearly simultaneously with the airspeed sourced faults, but unrelated.
-------

GB

Smilin_Ed

14th Jun 2009, 22:27

Therefore, we have a TCAS Fault that occurred nearly simultaneously with the airspeed sourced faults, but unrelated.

I don't think we can conclude that they are absolutely unrelated. They may well be unrelated from a systems viewpoint, but what if the failures were both the result of catastrophic airframe failure or lightning strike?

Mike-Bracknell

14th Jun 2009, 22:27

I am envisioning a joystick that is simply sitting up in the air. Pick one for your favorite game. Hold the base. Now move it rapidly left and right. What is the effect as read on your computer screen? Does it stay put? Does it move as if somebody had their hand on the stick and was exerting control?
One would hope that the sprung self-centring of the joystick versus the movable part's weight with no hand on it would far exceed the transient g-forces available in flight to move it on it's own?

I have no experience of the A330 stick, but I assume it's got to have a certain amount of sprung(?) self-centring otherwise it's one hell of a choppy ride when the pilot wants to scratch his chin.

Gringobr

14th Jun 2009, 22:41

Some of this info is new.. Passengers unconscious due to decompression.
Aircraft hit the sea probably in a near horizontal attitude.
I am taking this information from a local magzine called Veja dated 17 June (but available for sale today), page 69.. It is a top quality magazine and has the best analyses etc I have seen on this accident.
Summarizing because I ( and others) have already posted some details here
1 There are no burns etc (no bomb/explosion/fire) and the bone structures are basicaly preserved.. means the plane did not fully disintegrate in the air and/or at least the part where they were sititng arrived at the ocean relatively intact
2 What the doctors here call "4 fractures" upper parts of arms and legs ( 4 parts) broken.. This happens with accidents in which aircraft hit the sea with passengers sitting in their seats
3 Rosy (coloured) teeth. this symptom means the passengers were subjected to decompression , sudden exposure to rarified air and -56 degrees, they lost consciousness quickly before dying (all bodies rescued to date)
4 Clothes and documents, many of the bodies (but not all) had their clothes relatively intact, with documents and boarding passes. This means a goo part of the part of the cabin where they were seated hit the sea relatively intact

bravoxray

14th Jun 2009, 22:43

I have been reading this very informative thread with great interest and as a one time private single engine VFR pilot learned an awful lot . More a passenger now but not a terrified one. Air travel is still statistically very safe IMO compared to other mediums.
Cannot understand the obsession of this low tech ram air pitot tubes as an explanation for this accident. They are basically all the same in principle regardless who manufactured it. How the various computers process the input data is a different ball game.
The accident was surely a sum of various problems, intrinsic and extrinsic.
I do take offense though for even suggesting/insinuating pilot errors in this accident without any shred of evidence. They would not have flown into the center of a CB as very qualified experienced pilots and am sure were aware of past (temporary) icing problems on subject pitot tubes but was not considered a nogo.
I do have a question though for prof. com. pilots. Is there any reason why there are not multiple FDR's and CVR's located in the aeroplane, excluding the cost factor? The weight factor appears to be minimal.
The FDR info, if certain flight/monitoring devices failed, would that not diminish the outputs recorded on the FDR at critical moments?
I apologize if I do sound ignorant.
On the other hand, some of the comments I have seen are so ludricous that would not be surprised to see a recommendation for parachutes for all passengers. Kind of funny at 35 fl. , probably better shape upon impact but still dead.

HarryMann

14th Jun 2009, 22:50

Will a stick with no hand on it stay perfectly still poking up there in the air above the mounting point as the plane accelerates with rapid jerks in random directions?Mike, you beat me to it! It's breakout force would hopefully exceed any likely in-flight accelerations.

This is independent of whether the stick's output is controlling the plane or not. It's a question about whether under very heavy turbulence the input to the stick from the effects of the turbulence is stable enough to fly the plane. If not and the pilot HAS control isn't it possible that the bouncing around could lead to improper or damaging imputs to the flight control system?It is possible though, that under extreme turbulence, consistently accurate pilot flight control inputs in Alternate or Direct Law may be less easy than with a larger, less sensitive device having larger amplitude stop-to-stop movements e.g. a conventional control column - judged without having any experience with an AB style joystick.

Jetdoc

14th Jun 2009, 23:26

I have a drawing of the A330 side stick controller showing the guts of it. Its a pretty beefy unit. I don't think it will flop around too much. I don't have anyway of posting it but if anyone wants to do so, I'll send it to them.

Here's something about ACARS messages. I was sitting on an A320 the other day. It was on overhaul and some testing was being done on it. The system under test failed and within seconds we got a message that the failure had been transmitted.

I would bet that alot of people here probably just learned that fault messages are transmitted on ACARS. Remember that the primary reason for doing so is to give advance warning of system faults so that maintenance personnel can prepare for troubleshooting and repair. This is especially useful during short turnarounds. There is no need for Lat/Long info and there is no information that would be hidden by Air France or Airbus.

Jetdoc

14th Jun 2009, 23:29

(Maybe a high-G accelerometer set needs to be mounted in the cockpit so that the data recorders can note that it's not the pilot's fault he could not control the plane, nothing could.)

There is one installed on the aircraft specifically reporting to the Flight Data Recorder.

Jetdoc

14th Jun 2009, 23:36

http://i658.photobucket.com/albums/uu308/Jetdoc1/SideStick.jpg

WNcommuter

15th Jun 2009, 00:02

ChrisVJ wrote:
There is little evidence so far in statistical terms that FBW is any less safe than the old physical connection controls.

Suppose hypothetically that FBW reduces errors and improves safety overall but suppose that physical connection reduces pilot error in conditions exceeding the abilities of the FBW software. Wouldn't tort lawyers have a field day blaming FBW design for its shortcomings, or blaming non-FBW for its shortcomings, depending on what type of accident happened? This could be a no-win situation for aircraft designers.

WhyIsThereAir

15th Jun 2009, 00:03

Jetdoc -

Thanks for the image, that looks like a nice beefy stick with redundant linkages. I will be willing to believe that the Airbus engineers considered turbulance and saw to it that the stick is dynamically balanced against side thrusts, *when sitting there untouched*.

Now: Is there a procedure for dynamically balancing the sidestick *taking into account the mass of the PF's hand and arm*? Arms and hands of PFs (and PNFs) are also subject to lateral accelerations in turbulance. If the hand is coupled to the stick, is it not possible that the moment arm of the hand and arm might couple unintended movement into the stick?

If there is a procedure to balance the stick *with hand and arm attached* against spurious accelerations, then this is not a problem. Otherwise perhaps it could be in some cases.

JD-EE

15th Jun 2009, 00:04

h3dxb (http://www.pprune.org/members/294815-h3dxb)

For the Satellite Telephone system and other high speed connections I well believe there is an aimed antenna. ACARS could be using something like, but more modern than, Inmarsat-C which had a low data rate, small messages, and an omnidirectional antenna. It "sort of" makes sense to send ACARS through such a connection. That is why I believed it when someone else mentioned that ACARS used an omnidirectional antenna. Telephones (and Internet) would need a high gain steered patch antenna to be effective. For ACARS that might not be an issue.

Mad (Flt) Scientist

15th Jun 2009, 00:14

The early failures reported by AF447 ACARS all seem to be explainable by iced pitot tubes, save for the TCAS Fail. TCAS does not use airspeed, just altitude via the transponder.

Air data Altitude failure will cause the transponder to revert to Mode A. I'm not sure if it would report failure of its altitude input to the CMS/ACARS, but it didn't report a Fault.

Transponder reversion to Mode A will not cause a TCAS Fail condition, but a TCAS OFF condition.

Therefore, we have a TCAS Fault that occurred nearly simultaneously with the airspeed sourced faults, but unrelated.

Question: does the entire ADR section of an ADIRU "fail" as one unit? That is, having detected that it is producing "unreliable" airspeed data (through the ADR DISAGREE method) is the entire ADR output regarded as suspect by other systems?

If so then you don't need any fault on the altitude to trigger the TCAS message - the airspeed problems cause the other aircraft systems - including TCAS - to regard all ADR info from the identified "unreliable" ADIRU(s) as suspect.

From a quick scan of such AB docs as are available easily online, I can only find references that say things like:
Each ADIRU is divided in two parts, either of witch [sic] can work separately in case of failure of the other.
Which would tend to suggest that the whole of the ADR part is considered to fail as a unit.

I can see logic for either approach - once a source of data is suspect, you may wish to disregard it entirely, but there's also an argument for trying to use what may still be valid data streams from a partial;ly invalid source. Hopefully someone know which applies.

Jetdoc

15th Jun 2009, 00:14

There are about 73 pages of maintenance manual procedures regarding the various parts of the side stick and 8 pages of testing. There is some 'rigging' involved so to speak regarding the transducers but no balancing.
I do know that the pilots arm rest adjustment is important when using the side stick but thats best explained by an airbus pilot.

md80fanatic

15th Jun 2009, 00:17

I would like to advise caution when chasing a single goose. It's still possible the pitots were not frozen over and the plane never entered extreme turbulence. We simply do not know anything yet. We have perhaps one, maybe two holes in the cheese, with certainly many to follow. We should keep this discussion academic IMO ..... trying hard not to cast, or giving the impression of casting, blame in any particular direction. I'm finding this thread very educational, and I would like to thank everyone for their contributions.

h3dxb

15th Jun 2009, 00:24

@JD-EE

Unfortunanly it's called SATCOM Antenna not ACARS Antenna, and this one is steerable.

@Graybeard

Anyhow, going through the downlinked ACARS messages like published in "Der Spiegel" 24/2009 page 129, all warnings happened from 0210Z.

There are 2 failures downlinked with the same time tag:
341115 and 279334 .
Checking the TSM for the operational MSG F/CTL ALTN LAW with failure in 341115 it points to PROBE PITOT 1+2/2+3/1+3 detected by EFCS1 or 2.

279334 does not refer to ALTN LAW, its going direction RUD TRVL LIM FAULT

Referring TSM
TASK 27-91-00-810-822
Disagree of the Pitot Probe Data in the FCPCs

The rest on you guys , I'm out now, let my name out of the game

JD-EE

15th Jun 2009, 00:47

h3dxb (http://www.pprune.org/members/294815-h3dxb)

Thank you, that resurrects some of my prior articles, mostly deleted, that suggest the plane was relatively stable through 0214Z. After that it may have all happened very suddenly. Even with a phased array antenna steering is "slow" compared to the motions of a plane that has gone out of control. (I have written such loops in the past, for reference.)

thesandbox

15th Jun 2009, 00:56

JD-EE wrote.....
Now imagine it runs your automobile. Forward means go forward, the further forward the faster. And so forth. Now suppose you hit an obstacle, a speed bump. Your body jerks forward. Your car lurches forward into a crash. This is not a problem if you're sitting comfortably with no sudden accelerations over a small fraction of a G. But add in jerks from turbulence and you might find your arm involuntarily inputting a control surface change that was out of the range the plane could tolerate."

This is exactly my point....not the joystick receives feedback but that in this situation if factors are as assumed who would have a steady hand in trying to recover this a/c in direct law. Any little bump or quick movement would translate rather quickly to the joystick if the pilots hand is on it....not as easy to inadvertently shove the control column full forward in turbulence when "hand flying" it.

ClippedCub

15th Jun 2009, 01:12

Found this article on the A320 side stick controller. Claims Captain and FO can cancel each other out.

A320 sidestick description + references (Re: Airbus safety) (http://www.kls2.com/cgi-bin/arcfetch?db=sci.aeronautics.airliners&id=%[email protected]%3E)

Remember the F-16 side stick didn't move at all on the prototype, but it moves slightly now with artificial feel. and the forearm is supported. Forward travel is extremely limited though because they found under negative g, the pilot would over control.

EMIT

15th Jun 2009, 01:13

JD-EE, pot #1544 and on.

Your concern about turbulence bumping the stick, with or without hand attached, around so much that the inputs would be enough to exceed aircraft structural limits ? NO WAY.

Side stick steering is accurate enough to keep an F-16 aimed within a few mils of a target, while flying at 500 kts at low level in very bumpy conditions (1 degree is 17.5 mils).
And in the A-320 it is comfortable enough to enable greaser landings in 30 kt crosswind conditions that cause a 1.25 g bump at 32 ft radio altitude and 2 seconds later a 0.8 g bump at 25 ft RALT.

BOAC #1546.
I agree that flying without any attitude refernce would make the task at hand nigh impossible, but what about Inertial Ref systems 1 and 3?
Somewhere in the endless list of postings I have seen the interpretation that IR2 was reported as failed by, amongst others, IR1 and IR3.
Apparently some (many?) people had interpreted the involved data format of ACARS as a report of failure of IR2 and IR1 and IR3.

Mad Flt Scientist #1563.
I agree with you that the ADIRU parts are the Inertial Reference part and the Air Data part. I do not think that in failure modes the static and dynamic pressure sections of the Air Data part are treated as separate entities. I am not sure about the Angle Of Attack Signal - whether that would also be rejected - have to go back to the books on that one.

By the way, why not throw in a new subject. Isn't it a great coincidence that together with the pitot static failures, another totally unrelated system failed, IR2?
Could it be that, troubleshooting ADR2, by accident ADIRU2 was switched OFF? I always found that a very easy to make error in the A-320.
In everyday life you always manipulate the rotary selector, never the ADR push button, the system is called ADIRS and failure of the Air Data (Computer part) of the ADIRU is annunciated on ECAM as
NAV ADR2 FAULT.
Now that looks very much like ADIRS, which you always manipulate via the rotary selector (with shuts down the IR part).

Edit: corrected the mil specification)

The Chaser

15th Jun 2009, 01:23

Not implying causality in any way shape or form by providing these links.

Whilst looking for something else, I read through the systems linkages/interactions (discussed in the first link) and am wondering if it might provide an answer to Greybeard, BOAC and Capt Crunch’s questions regarding sys links between some of the input data to ISIS.

http://www.casa.gov.au/airworth/airwd/ADfiles/over/a330/a330-095.pdf

Airbus Model A330-200 and -300 Series Airplanes, and (http://rgl.faa.gov/Regulatory_and_Guidance_Library%5CrgAD.nsf/0/3CF0C40419204A2F8625756100508BEF?OpenDocument)

I am not certain so no comment.

I am sure this has been asked, but I cannot find a solid reference. Was AF447 equipped with Northrop-Grumman (formerly Litton) Air Data Inertial Reference Units (ADIRUs), part number (P/N) 465020-0303-03ZZ (with ZZ from 09 up to 16 inclusive), or the Honeywell units?

EDIT: TIME (http://timeinc8-sd11.websys.aol.com/time/world/article/0,8599,1902907,00.html?iid=tsmodule)

According to Airbus spokesman Justin Dubon :-

"manufacturer Northrop Grumman makes the ADIRUs for Qantas, and Honeywell for Air France."

TWT

15th Jun 2009, 01:38

I recall reading earlier in this thread that AF447 had Honeywell ADIRU's fitted

Mad (Flt) Scientist

15th Jun 2009, 01:45

I recall reading earlier in this thread that AF447 had Honeywell ADIRU's fitted

this (http://www.pprune.org/rumours-news/375937-air-france-a330-200-missing-42.html#post4973169) is the only post I could find which appears to contain that information. It's hardly conclusive.

The Chaser

15th Jun 2009, 01:45

TWT Thank you.

EDIT: See above post#1569 re: Airbus confirmation of Honeywell being the manufacturer of Air France's A330 ADIRU's

Would both ADIRU models output data in the same way/configuration to linked systems?

Captain-Crunch

15th Jun 2009, 01:51

Graybeard said:
Without the FDR to bring in a surprise, the majority of a civil jury (US law) would conclude that Thales, Airbus and Air France share in the negligence that allowed this fleet to contine after five serious incidents caused by icing of obsolescent pitot tubes. I don't see why anyone would try to thwart recovery of the FDR. The evidence from it could hardly be more damning.

GB

Statistically, A330's are good safe airplanes. It's just that we who desire that trend to continue, wish for meaningful successful mods to the factors responsible for this tragedy. And we are just a rumor board, NOT an investigative body. We have only skimpy info available to us; therefore any speculation by us here is strictly personal opinion only.

Graybeard, I'm not sure that negligence is the right word. Industry wide "Overconfidence in automation" is the term I would use, and is the non-criminal culprit that I suspect. Whether that be an operational training attitude, or a design issue (I can't know that, I've never flown the A330 or a FBW aircraft) or a combination of a whole lot of things (factors), the fact is that a problem likely exists, and all parties appear to me to be doing something about it. Did the past fixes work? Doesn't look like it.

I'm sure its most likely the crew was current and qualified. But what about being able to fly partial panel, at night, in weather, with nothing but the whisky compass and a flashlight in your teeth? Could you set say 8x% N1 and make a gradual 180 on the Whiskey compass with no attitude, or just using the IRU true heading in rough weather? I don't know if it can be done....

We never really train for that one do we? Because, before now, it was believed it was unlikely to ever happen. But by the same token, for a long time, we never believed volcanos could take out all four engines, and before that we never believed in a thing called microbursts after a pilot made a so-called "bad" landing.

Right?

My humble opinion is that simulators are marvelous training devices, but they are no substitute for hand flying the climb and descent and hand flying a manual landing on a daily basis in the actual aircraft far away from the "Magic Kingdom" (the sim.) Without practice in this art of manual flying (occasionally all the way up to altitude and then S&L for a few minutes.) crews don't stand a chance when "Otto" gives up at the worst possible time.

There's only one man we can turn to in an uncertain situation like this. A good Union Man. President Obama was wise indeed, to appoint the former head of the Air Line Pilots Association to head the FAA.

I would call on the very-capable, just appointed, FAA Administrator: Captain Randy Babbit, to review the over-reliance/distraction with automation that crews now face, and make sweeping changes to the industry in this regard. :8

Fraternally,

Captain Crunch

.

Graybeard

15th Jun 2009, 01:52

misd-agin

15th Jun 2009, 02:06

swish 266 - post 1510 -
"The more important thing is WHY IT HAPPENED!
WHY AF477 ENDED UP BEING WHERE THEY SHOULDN'T HAVE BEEN IN THE FIRST PLACE?
This is the million $ question!
Why didn't they deviate like IB or LH?"
*******************

The IR satellite picture shows cold temperatures for hundreds of miles. IB supposedly deviated 30 miles from the track which isn't much and would put them in the same airspace on the IR picture.

Has anyone stated what exact route AF 447 took? Perhaps they were deviating just like the Iberia flight.

JD-EE

15th Jun 2009, 02:11

misd-agin (http://www.pprune.org/members/128797-misd-agin)

Nobody knows. And the supposed positions at last ACARS report seem to be conjecture. Two quite separated locations have been published for it.

{o.o}

ACLS65

15th Jun 2009, 02:11

The Chaser:

Though not a definitive answer this would make me think Honeywell.

"The NTSB said, they have accepted an invitation by the BEA to assist in the investigation and have dispatched Bill English as their representative joined by engineers from the FAA, General Electrics (manufacturer of the CF6 engines) and Honeywell."

Mad (Flt) Scientist

15th Jun 2009, 02:16

Mad(flt)Scientist, "Question: does the entire ADR section of an ADIRU "fail" as one unit? That is, having detected that it is producing "unreliable" airspeed data (through the ADR DISAGREE method) is the entire ADR output regarded as suspect by other systems?"

The communication from the Air Data (Module, Computer) to the transponder, for example, is on an Arinc 429 serial digital data bus. There are several words transmitted in sequence, each with its own identifier and validity state. CAS, IAS, TAS, SAT, Pressure Altitude, Baro Altitude, Vertical Speed, etc. will be on the bus.

The transponder will pick off Pressure Altitude, and ignore the rest. The ADC transmits all words repeatedly. If the airspeed goes invalid, for example, Fail Warn will be indicated in the Sign Status Matrix of that word.

Therefore, an iced pitot will have no effect on the transponder or any other device that uses Altitude, nor on the TCAS, which is even farther removed.

GB

OK, but in this case is it not the FMGEC that's calling the ADR DISAGREE, not the ADIRUs themselves? After all, an iced pitot may present no useful symptoms to the individual ADM or ADIRU that can be used to detect the possible problem. It's only at a higher, integrated, level where the comparisons can be carried out between the different ADIRUs and a problem possibly identified. In this case the ADIRUs may all have been claiming their data was valid, since no system failure had occurred. (And as I understand it there's no ADIRU failure on the ACARS list - which implies the boxes themselves thought they were ok; it was the customer systems that identified the problem.

If the FMGEC (or whatever its called?) is spitting out an ADR DISAGREE, is that going to then cause either the ADR parts to consider themselves potentially invalid, or other systems to do so? (The FMGEC in effect is "policing" the ADR data in that model)

misd-agin

15th Jun 2009, 02:18

FE Hoppy (post 1246), Lost in Saigon(post 1248)(thanks for the picture), and ttcse(post 1251) -

A poster stated that the AF vertical stabilizer had no fluids around it unlike the A310 vertical stabilizer. The difference might be that the A310 vertical stabilizer picture could have been taken as soon as 9 hours after the accident. Based on the picture the sea state seems fairly calm.

The AF 447 vertical stabilizer was not found for several days. Reports stated that the search was difficult due to weather and sea conditions.

Those differences may, or may not, explain the different levels of POL's near the vertical stabilizer.

Right now it's too premature to state the vertical stabilizer must have seperated inflight.

CONF iture

15th Jun 2009, 02:28

BOAC #1546.
I agree that flying without any attitude refernce would make the task at hand nigh impossible, but what about Inertial Ref systems 1 and 3?
Somewhere in the endless list of postings I have seen the interpretation that IR2 was reported as failed by, amongst others, IR1 and IR3.
Apparently some (many?) people had interpreted the involved data format of ACARS as a report of failure of IR2 and IR1 and IR3.

Surprised the following post is not avail anymore ?

http://www.imagefrog.net/out.php/i24465_AF447011.gif

Regarding the sidestick, whatever the law, the less you touch it, the better.
Actually very easy to overcontrol.

Rollingthunder

15th Jun 2009, 03:49

loss of vertical surfaces in Canadian fleet prior to
AA A300,

What the hell? I worked for them and the only loss I'm aware of is an A320 vertical stabilzer clipped by another aircraft. Replacement flown in by Airbus Beluga. Interesting day.

barstow

15th Jun 2009, 04:51

A "maintenance professional" is one helluva long way separated in so many ways from an engineer, or a materials scientist that I'm surprised anyone would give any weight to their opinion.

A lot of laypersons (including personnel employed to bolt and unbolt things on things that take to the air) seem that they just can't hurdle the concept of a non-metallic aircraft/parts. I even read someone equating keel breakages on super-maxi yachts to Airbus design flaws....

Not even close! :ugh:

RWA

15th Jun 2009, 05:38

Rollingthunder, I think the poster meant the Air Transat case, where the entire rudder broke off while cruising on autopilot. That was in 2005, long after AA587.

AirDisaster.Com Accident Photo: Air Transat A310-308 C-GPAT (http://www.airdisaster.com/photos/c-gpat/photo.shtml)

This was one of two incidents (the other was a Fedex aircraft) that finally solved the A300/310 rudder problem; which was found to have been not any basic weakness but delamination caused by leaked or spilled hydraulic fluid.

WhyIsThereAir

15th Jun 2009, 05:58

> Must admit though, it seems increasingly likely to have been the aeroplane breaking up.

No, not really. The message (as I understand it) indicates that the plane is probably descending faster than normal. With the result that the pressure outside is increasing faster than the pressure inside. That is not what you would expect with a broken cabin, since you then should have about the same pressure on both sides, since they would be connected.

We have no way of knowing if that is a controlled descent or uncontrolled. However, the fact that it was received by the satellite indicates that the antenna was still tracking, and that argues that the plane was reasonably stable at that moment.

Graybeard

15th Jun 2009, 06:01

Mad(flt)Scientist: "If the FMGEC (or whatever its called?) is spitting out an ADR DISAGREE, is that going to then cause either the ADR parts to consider themselves potentially invalid, or other systems to do so? (The FMGEC in effect is "policing" the ADR data in that model)"

The sensors police themselves, but do not police each other. It's up to the data user to compare inputs and apply logic. If there is an airspeed miscompare among the 3 Air Data units for example, the user that receives and compares them will report the outlier to the Centralized Management Sytem, which relays ADR 2 Fault to the ACARS. The user will continue to use and compare other ADR outputs that are valid.

This is why the TCAS Fault stands alone in the reports at 02:10. It does not directly relate to the other faults that could be traced to iced pitot tubes. It may indicate intermittent cabling shaken by turbulence, a lightning strike to an antenna, or a myriad other possibilities, so it may point to the initial aircraft failure - or not.

GB

captainflame

15th Jun 2009, 07:23

Greaybeard:

As mentionned meany times before on this thread, "TCAS inop" is a resulting listed inop system in the ECAM "ADR disagree" STATUS page on the Airbus.

It is so decribed in FCOM Vol3, under the list of ECAM messages which present ECAM checklists/actions AND STATUS page.

IT SO HAPPENS THAT ALL THE ACARS MESSAGES from AF447 ARE CONSISTENT WITH AN ADR DISAGREE MESSAGE.

For the last time !

Finn47

15th Jun 2009, 07:27

French investigators (BEA) have announced a press conference coming up on Wednesday:

The BEA will be holding a press conference concerning the ongoing operations at sea in its premises at Le Bourget on Wednesday, 17th June, from 10 to 11 am.

captainflame

15th Jun 2009, 07:53

Greaybeard:

When the system compares ADR data and sense one is going astray, it disregards and eliminates that source and only uses the 2 remaining.

ADR DISAGREE comes on after, when the 2 remaining ADR don't agree either. It's the PRIMs (F/CTL computer) that reject this data. Resetting them will not restore Normal Law.

Thresholds to reject an ADR : CAS and TAS, 16Kts for 10s, MACH 0,05 for 10s, ALT 3000ft for 1s, AOA 3,6 for 10s, Static pressure 5HPA for 1s. This means a sudden and significant variation of the ADR data will eliminate that Air data source !

Since the air data is corrupted, erroneous, wrong or along those lines, the FMGEC eliminates the CAS from the PFDs (SPD LIM Red flag, loss of "Vls").

Also, the A/C being in "Alternate Law without portections", the "flight envelope" parts of both FMGECs are considered INOP !! thus triggering a fault message on the FMGECs.

Note that the LOAD FACTOR protection is still working !!

Similar incident on another A330, another company, another day, but with same pitots, WAS attributed to icing of the probes! It lasted 2 minutes 20s, from 22h22,20s to 22h24,40s (reengagement of FDs and APs !!) and triggered ECAM messages and PFD flags similar to AF447 ACARS messages.
For 1m26s, the PF is in manual flight, without APs, without FDs, without FPV (Bird) and without A/THR.

Cheers

Dave Gittins

15th Jun 2009, 08:31

Without good news reports in this part of the world and having tried to wade through 80 pages of the thread with intermittent success, is it fair to say the position is ????? :

The AF447, an A 330, departed from the cruise at FL350 without any radio calls.

It appears to have suffered an inflight break up or at least a partial break up which made it uncontrollable.

Potential reasons are overspeed or stall due to issues with speed sensing or violent turbulence in CBs. The aeroplane's composite structure may have some or no bearing.

AF are replacing some parts of the pitot system because of issues of icing on one type of pitot assembly found on some airbusses. Some other owners in the world fleet of A330 and other models of Airbus are following suit or may even be further advanced with retofitting.

A number of ACARS messages over a 4 minute period indicate various malfunctions or changes in system configuaration but without specialist access to the data it is not clear what that is telling us.

When / If the FDR and CVR are recovered that may / probably will, tell us more.

Without shooting me out of the sky for not having read all the posts (I have an airport to build) is that a fair summary ? or can somebody please help me understand the full story to date.

Dave Gittins

paros

15th Jun 2009, 10:03

As I am not an expert in accident investigation I just wonder if the outcome would have been the same if......

1) All large transport aircraft operating under ETOPS rules were required to have "RealTime" weather information superimposed on the Map/Weather display. (Available on many exec jets)

2)The bottom line in event of multiple failures that the pilot is able to fly the aircraft albeit with cables and rods. (as in Boeing)

3) The Maximum Normal Operating Altitude be reduced from the manufacturers optimum altidude by 2000 feet.

The above would of course be at a cost.......BUT what are you worth?

onetrack

15th Jun 2009, 10:05

JD-EE - I am not a composites expert, and neither am I making any claims - I merely stated a personal opinion, that composites appear to rely on the resins for the composite material behaviour and abilities.

One has to examine all facets of a disaster, to improve knowledge. There are obviously, real, current reservations, about the effects of aging, long-term UV exposure, and contamination by fluids and other chemicals on composites durability and strength.
One thing I find rather amazing, and to me, quite simplistic - is the "tap" test on composites, to determine if internal de-lamination has occurred. One would think, that with aircraft, a much more sophisticated examination process would already be in place.

The following Airbus composites testing conference paper looks very impressive - but the testing appears to lack at least two (vital, I would have thought) additional features - the effects of UV aging, and chemical and fluid contamination on the composites, at the same time as destructive testing was taking place.

As we all well know, laboratory and factory testing is excellent in many applications - but rarely, does it take the place of actual in-field service, with all its additional factors that cannot always be considered, in a lab and engineering planning environment.
I notice that Eurostar (by Evektor), along with other manufacturers, insist that their metal frame is still superior to composites.

New Scientist article - Composite aircraft may hide dangerous flaws - tech - 22 November 2007 - New Scientist (http://www.newscientist.com/article/dn12951-composite-aircraft-may-hide-dangerous-flaws.html)

Airbus Composite testing conference (2004) - http://www.aer.bris.ac.uk/comptest2004/proceedings/presentations/comptest_109.pdf

HarryMann

15th Jun 2009, 10:29

In photos 1 and 3 one sees in more detail how the rudder failed. The bent mounting pins that are visible are bowed in the center, no doubt levered from their mounting lugs in the empennage, showing that the tail was pulled off by aerodynamic forces and not bent randomly or forward by impact forces. This again is consistent with the visible damage to the lower part of the rudder. We also see how the composite material has failed along the opposed cross hatch of fibers.

Really! That's terrific stuff

If I'm looking at the same (1 & 3) pictures of the VS you are then I'd say...

You could earn a fortune as a 'remote accident investigator' just using small low resolution pictures taken around the other side of the world, saving BEA, FAA, AAIB an absolute fortune.

== UV damage to composites ==

Bearing in mind the comments in several posts above:
Is this really a serious concern, when there are multiple layers of (UV resisting) paint and UV light however strong would doubtless penetrate one or two laminate layups?

OVERTALK

15th Jun 2009, 11:06

Graybeards said (at

post 1549 (http://www.pprune.org/rumours-news/376433-af447-78.html#post4997293) that:
"Ref: "There are five official, damning reports on A330 and A340 aircraft operated by

major airlines suffering complete failure of all flight instruments in flight. Pitot probes

iced up in meteorological conditions at night, by day, in cloud and in clear air."

The early failures reported by AF447 ACARS all seem to be explainable by iced pitot tubes, save

for the TCAS Fail. TCAS does not use airspeed, just altitude via the transponder.

Air data Altitude failure will cause the transponder to revert to Mode A. I'm not sure if it would report failure of its altitude input to the CMS/ACARS, but it didn't report a Fault.

Transponder reversion to Mode A will not cause a TCAS Fail condition, but a TCAS OFF condition.

Therefore, we have a TCAS Fault that occurred nearly simultaneously with the airspeed sourced faults, but unrelated."

It would appear that some are flummoxed by the TCAS failure msg on ACARS (TCAS needing and acquiring static pressure only).
Correct me if I'm wrong but, depending upon exactly how the ADIRU derives and passes the static pressure, the pitot icing failure may be related.
i.e.
Pitot pressure = dynamic pressure (i.e. IAS) plus static pressure

The ASI derives a normally correct airspeed by balancing the pitot derived static against the static pressure sensed from the static ports (and fed into the ADIRU as a digital signal by a transducer).

IAS (or dynamic pressure) = pitot pressure minus static port sensed pressure

So if your static lines ice up due to trapped water freezing as you climb, then those two differently derived static pressures will become increasingly unequal and the ASI will wind back to zero (in a climb/opposite in a descent). In the cruise, the IAS will simply "freeze".

So it depends which static pressure is being fed to the ADIRU and at what stage? Is it simplistically static port sensed pressure (with its built-in pressure error) or is it the resolved average of the two differently sourced static pressures? Once the pitot ices over, that average would be duff.

DC-ATE

15th Jun 2009, 11:16

Captain Crunch -

Nice post.....[#1573] UNTIL those last two paragraphs.

1st - We'll see if the appointment was "wise indeed".

2nd - Simply isn't gonna happen.

Erik.G

15th Jun 2009, 11:28

Cabin crew seat found floating in the Atlantic

http://gfx.aftonbladet-cdn.se/multimedia/dynamic/00953/belt480_953629l.jpg

rmn_76

15th Jun 2009, 11:42

Just as a general reply to all the non-composite experts and non-stress engineers.

Yes, when we test composites we artificially age and damage test pieces. We also typically test in a hot wet condition (humidity in particular has a negative effect on the structural properties of CFRP).

Can we all stop making out that the engineers (i.e. me) have no idea how to design composite structures. After 20 years of design and testing we do.

The rudder failures that so many of you are so excited about were design failures not composite problems. My point being that ingress of hydraulic fluid requires something to leak the fluid on to the composite.

And finally, whoever it was that brought up that some manufacturers say composites are no good. This will probably be because they are not capable of manufacturing composite primary structure. Airbus was saying that composite fuselages were a bad idea, right up until they canned the old A350 program and restarted with a composite fuselage.

Can I please ask that anyone writing a post that begins with the phrase "I'm not an expert" dosn't. There are a lot of experts on this forum but it's annoying to have to trudge through the wild speculation to find the interesting/insightful comments.

Graybeard

15th Jun 2009, 11:52

In the late 1980s, McDouglas bought back the high time DC-9, fuselage #2, and subjected it to pressurization cycles, bringing the total to 200,000, IIRC. Has similar testing of high time composite components been performed?

Slats One

15th Jun 2009, 11:54

All very true, but as I have studied composite construction and use for 30 years, I can no longer remain silent as I read this thread.

Composites are safe.

Composites age and we have few methods of assessing them as they do so.

But what a lot of you are missing is that these composite tails have a history. Have you forgotten the litres of water drained from AA Airbus tails/ have you forgotten the issue of internal collapse of the structure due not just to oil based contaminants, but also the effect of repeated freezing and expansion in flight cycles of ingress affected composites.

And what about the manufacturing flaws - flaws over and above a critical size. Who is going to look for those. What about the repaired new fins that are alleged to have existed.

The cause of this accident remains unknown. until we know ALL the facts, we will never know what happened as a casue or a consequence.

Best wait until then before we unpack past accidents and past debates.

BOAC

15th Jun 2009, 11:58

It appears to have suffered an inflight break up or at least a partial break up which made it uncontrollable.
Potential reasons are overspeed or stall due to issues with speed sensing or violent turbulence in CBs. The aeroplane's composite structure may have some or no bearing. - in answer to your request, DG, I would say the jury are still undecided on this. It is even possible (as far as I can ascertain) that it might even have attempted to ditch.

Eastwest Loco

15th Jun 2009, 12:07

I am totally technically uninformed apart from osmosis over 30 years in the industry plus being the son of a LAME and will sign and post that.

In this tradgedy, 228 souls have been lost.

My 2 bob's worth is that I feel an aeroplane has been bitten by "Coffin Corner" being too heavy and too high for a warm pool of air sucked up by CB's and found itself in a stall or overspeed conundrum.

The pilots may well have chosen a "box canyon" way out that overcomplicated the situation that led them directly into an unavoidable CB and the coup de grace. The flight deck crew took that knowledge with them.

The bottom line is we have lost some wonderful aircrew and passengers.

They seem to have been a little lost in the analysis.

Hopefully no blame will be appointed or apportioned until some solid facts come to light, and Airbus should not be hung over the Coffin Corner side of things until all facts are in.

The last major CC events were from memory transcontinental Boeing 727-200s.

Rest peacfully good souls on board AF447. May your loss not be in vain.

EWL

cpdlcads

15th Jun 2009, 12:22

Graybeard

15th Jun 2009, 12:26

Overtalk: "So if your static lines ice up due to trapped water freezing as you climb, then those two differently derived static pressures will become increasingly unequal and the ASI will wind back to zero (in a climb/opposite in a descent). In the cruise, the IAS will simply "freeze"."

Why are you introducing static line freezing, when the only known suspect is the pitot tubes? Airspeed needs static pressure, as by definition it is static pressure subtracted from ram air pressure. Altitude, (static pressure) does not need Airspeed, except for trimming correction at high angles of attack. Altitude errors provide a whole different list of faults.

There is nothing in the ACARS reports suggesting Altitude Fault or disagree. Remember, the CMS report to Maintenance is a guide to the mechanic on which box to change on a short turnaround, so it may say ADR Fault, or ADR disagree. It doesn't say Airspeed Fault, or Altitude Fault. Those are left for the mechanic to find in the BITE, Built-In Test Equipment, testing.

Beyond that, a Pressure Altitude Fault will not trigger a TCAS Fault, as the Transponder is in between, relaying own ship altitude to the TCAS. Again, a Pressure Altitude Fault will cause the transponder to revert to Mode A, putting the TCAS in OFF mode. Remember the Legacy collision in Brazil? The transponder in Standby had put the TCAS in OFF, and didn't trigger a TCAS Fail, that the crew would have more likely seen.

GB

Graybeard

15th Jun 2009, 12:48

Capitan Flame: "When the system compares ADR data and sense one is going astray, it disregards and eliminates that source and only uses the 2 remaining."

So why wasn't there an ADR Fault transmitted at 0210, if that were the source of the TCAS Fault as you say?

There has been no rational symptom of Altitude Fault shown in all these pages. All the ACARS reported faults downstream of the ADR are traceable to Airspeed Fault, save for the TCAS. Indeed, in the prior incidents, airspeed indication was lost or faulted, with no mention of altitude indication fail or disagree.

GB

RWA

15th Jun 2009, 12:50

Desertia

15th Jun 2009, 13:13

As nothing more than a curious member of the SLF, I've obviously not posted my groundless theories on what could have happened, preferring to read the myriad of alternatives offered on here, and keeping an open mind given that it's obvious that there is still insufficient evidence to identify the cause of this crash.

But one thing I would like to say, having been in the IT field in a variety of positions over the years, is that you cannot underestimate the risk of human error, or at least limitations in putting together the millions of line of code required to automate something as complex as flight.

I remain of the opinion that, irrespective of the causes of this accident, I would still rather an experienced pilot have a piece of wire or hydraulics controlling the various surfaces under unusual conditions than having one or more computers trying to work it out.

Artificial intelligence still does not - and probably will not for decades to come, unless there is a magnificent leap in technology - mirror the ability of the human brain to eschew logic and do what thousands of hours of flying, and experience of abnormal circumstances allow, which is, for want of a better word, THINK.

Best wishes to all, and the efforts of you chaps up the front are always greatly appreciated.

Sober Lark

15th Jun 2009, 13:23

Paris based AXA led the coverage brokered by London based JLT Aerospace inc. Other insurers American International Group (12.5% of the coverage but loss capped at €21 million). AIG Munich..

The hull was valued at $100 million. It is too early to speculate on the likely size of the liability loss because those payments vary widely depending of the personal details of the victims such as earning potential and dependants, and which country takes the case. That could be complicated – the plane was owned cby Air France, built by European consortium Airbus with parts supplied from around the world, and crashed in international waters with passengers from residents of 32 countries.

If the Air France loss proves to be very significant then it could put upward pressure on premiums.

Airlines must pay $25000 to family victims to cover immediate economic needs within 15 days of identifying appropriate recipients.

The average liability for for aviation deaths in the US is 1.5 to 2 million and about $150k outside the US.

Liability figures are based in part on passenger numbers carried. (projected passenger demand falling 8% this year)

Before AF there had been $706 million in airline losses this year.

Source: Business Insurance Z Philips

clipstone1

15th Jun 2009, 13:35

with a world annual airline premium of about $1.6bn, less reinsurance costs etc, meaning potentially with this loss the insurance market has just about hit "claims = premium" with another 6 months of the year to go...not good news in general for airlines. The market has been trying to increase premiums for a couple of years and this loss is quite possibly the one to turn the rates upwards just when airlines have no money.

RWA

15th Jun 2009, 13:36

"I remain of the opinion that, irrespective of the causes of this accident, I would still rather an experienced pilot have a piece of wire or hydraulics controlling the various surfaces under unusual conditions than having one or more computers trying to work it out.

"Artificial intelligence still does not - and probably will not for decades to come, unless there is a magnificent leap in technology - mirror the ability of the human brain to eschew logic and do what thousands of hours of flying, and experience of abnormal circumstances allow, which is, for want of a better word, THINK."

Truly great comment, IMO, Desertia.

Can't help it, but whenever the subject of computerised flying controls comes up it puts me in mind of a programmer who has never learned to ride a bike trying to program one that would work automatically.

One can imagine the programmer saying to his cyclist adviser, "About turns, I'm not clear about them yet. If someone wants to turn, do they turn the handlebars first, or lean over first........?"

Flying has always been a mixture of art and science. Ever since 1903......

No way you can boil the whole thing down to a mere computer program. Not in ALL possible situations, anyway.

Lost in Saigon

15th Jun 2009, 13:54

I would say at least two of your assumptions are incorrect.

Without good news reports in this part of the world and having tried to wade through 80 pages of the thread with intermittent success, is it fair to say the position is :

The AF447, an A 330, departed from the cruise at FL350 without any radio calls.

There is no way to confirm this without the CVR. They may have made calls that no one heard.

It appears to have suffered an inflight break up or at least a partial break up which made it uncontrollable.

Up to this point in time, there is no real evidence of an inflight breakup. They could have impacted the water intact. The aircraft could have broken up during a planned ditching. There have been lots of possible explanations for the spread of debris and bodies

barit1

15th Jun 2009, 14:19

Lost in Saigon

15th Jun 2009, 14:19

Dave Gittins, I replied to you recommending that you read this story, which will do a lot to bring you up to date:-

Air France crash: What is known so far | csmonitor.com (http://www.csmonitor.com/2009/0614/p02s09-usgn.html)

For some reason my post was deleted. Maybe because I went on to make some further points. So I'll leave it at that.

Just 'click,' mate........ :)

Your post was probably deleted because the article you link to is almost pure speculation. Not "Known Facts".

Example:

The plane itself also offers potentially telltale clues: The last of the airplane computer's messages reported that cabin pressure, usually maintained at an equivalent to an elevation of 10,000 feet, was dropping at a rate of 1,800 feet per minute.

Cabin pressure had been lost and one obvious cause could be that the plane was falling apart.

There is no proof cabin pressure had been lost.

An aircraft doing an unplanned emergency descent, or even a planned ditching, could quite conceivably experience a cabin rate of descent of 1800 feet per minute at some point in the flight.

I would expect that an aircraft cabin in a free-fall descent (after "falling apart") would descend at something closer to 18,000 feet per minute. Not 1,800 feet per minute as stated.

Gringobr

15th Jun 2009, 14:22

Up to this point in time, there is no real evidence of an inflight breakup. They could have impacted the water intact. The aircraft could have broken up during a planned ditching. There have been lots of possible explanations for the spread of debris and bodies
.
There is... see below.. Rosy coloured teeth
.
What the bodies say
Some of this info is new.. Passengers unconscious due to decompression.
Aircraft hit the sea probably in a near horizontal attitude.
I am taking this information from a local magzine called Veja dated 17 June (but available for sale today), page 69.. It is a top quality magazine and has the best analyses etc I have seen on this accident.
Summarizing because I ( and others) have already posted some details here
1 There are no burns etc (no bomb/explosion/fire) and the bone structures are basicaly preserved.. means the plane did not fully disintegrate in the air and/or at least the part where they were sititng arrived at the ocean relatively intact
2 What the doctors here call "4 fractures" upper parts of arms and legs ( 4 parts) broken.. This happens with accidents in which aircraft hit the sea with passengers sitting in their seats
3 Rosy (coloured) teeth. this symptom means the passengers were subjected to decompression , sudden exposure to rarified air and -56 degrees, they lost consciousness quickly before dying (all bodies rescued to date)
4 Clothes and documents, many of the bodies (but not all) had their clothes relatively intact, with documents and boarding passes. This means a good part of the part of the cabin where they were seated hit the sea relatively intact

Lost in Saigon

15th Jun 2009, 14:26

Sudden decompression does not have to mean inflight breakup.

Still, it is possible that there was an inflight breakup. But it can not be called a "Known Fact"

DC-ATE

15th Jun 2009, 14:27

Desertia -
.....one thing I would like to say, having been in the IT field in a variety of positions over the years, is that you cannot underestimate the risk of human error, or at least limitations in putting together the millions of line of code required to automate something as complex as flight.
I remain of the opinion that, irrespective of the causes of this accident, I would still rather an experienced pilot have a piece of wire or hydraulics controlling the various surfaces under unusual conditions than having one or more computers trying to work it out.
Artificial intelligence still does not - and probably will not for decades to come, unless there is a magnificent leap in technology - mirror the ability of the human brain to eschew logic and do what thousands of hours of flying, and experience of abnormal circumstances allow, which is, for want of a better word, THINK.

Very well put and, I'm sure there are countless pilots that feel the same way. I for one. Once again, I fear computers have played a part in a tragedy.

Dave Gittins

15th Jun 2009, 14:32

I am grateful for being put straight about my assumptions.

My question was whether those assumptions about known facts (or at least things with a high probability as alternatives) were correct.

I suppose it is fair to say that only time will tell what assumptions are actually correct or not. Perhaps all we do know with certainty is that the holes in the swiss cheese lined up, but we do not as yet know which holes maybe not even which sort of cheese.

No investigation takes place without starting with facts and then making informed assuptions (either positive or negative) and then setting out to try and deduce whether any of the assumptions are valid.

I might for example make the assumption that a ditching was attempted (suggested by BOAC).

What I then need to establish is whether any facts fit or disprove that.

For example; was the wreckage in a localised area or spread over 10s of miles ? If it was in one place, was the recovered debris so disrupted as to indicate a very high speed impact ? or a slow speed ditching ? Were there indications of the angle at which the aeroplane hit the water ?

My original intention was not to presume, but having been out of the loop to determine what theories were current and seemed to have legs and which had already been discounted.

It seems that at present practically no assumption is safe simply because I have no idea which of the statements I have read is supported by verifiable facts yet.

Dave Gittins

ekw

15th Jun 2009, 14:54

On the contrary - computers are much better at mathematics, trigonometry and other complex calculations at high speed, and they relieve the pilot of a lot of workload. Airbus Systems have all been well tested with multiple redundancies built in. Something else happened that made both automatic and manual control impossible.

Much speculation concerns the ACARS messages - but if there was a catastrophic failure some of the more informative fault reports may not have gotten through to the COM link or they may have been queued for transmission but never made it. We shouldn't assume that the faults transmitted were the principal faults that occurred. For instance if another unreported fault disrupted power to the pitot heating, then the latter problem would be a symptom and not a cause.

RWA

15th Jun 2009, 15:16

"For example; was the wreckage in a localised area or spread over 10s of miles ? If it was in one place, was the recovered debris so disrupted as to indicate a very high speed impact ? or a slow speed ditching ? Were there indications of the angle at which the aeroplane hit the water."
Dave, the evidence so far is that all bodies recovered so far have been in two distinct 'zones', at least 50nms. apart.

Of course, currents and winds can do extraordinary things - especially in mid-Atlantic. But the evidence so far indicates that the aeroplane broke in half at high altitude. Say a 75% probability.

Additionally, according to the pathologists, none of the dead recovered so far drowned. Nor had any of them inhaled smoke. Which tends largely to rule out the aeroplane crashing into the sea intact, and also any sort of bomb........

I tend to think that the pathologists are being a bit reticent. Not because of a cover-up, but out of consideration for the relatives. Ever since the Comet crashes in the 1950s, it's been known that there are two signs that 'shout' explosive decompression - skull fractures and 'all four' fractures - meaning both legs and both arms. No news on skulls yet, but the path. guys say plenty of 'all fours.'

To make my own position clear - I've always done 'bits' of all kinds of things. Like flying. But one thing I had to do for years (to my cost, in emotional terms) was to investigate fatal accidents in industry. Take my word for it, it's utterly harrowing.........

So all I hope and pray for (literally) is that 'they' find out what happened to AF447 in quick time.

Because I guess that all but the least imaginative among us realise that, as we stand, the exact same thing could happen again tomorrow.......

Sneiled

15th Jun 2009, 15:30

I have been reading mostly all posts on Airliners and pprune about AF447 crash since the breaking news on 1st of June.
From day one, some media made their headlines with the title AF447 Mystery
Big Mystery; as pilot did not even give any sign of what went wrong or send a May Day (we all know that under some situation, crew has other priority before they send any sign of distress)
Two weeks after the crash it remains a mystery...
We get more clue, more information with the debris, bodies, ACARS messages, testimony of professionals including pilots who were following similar flight path on that night and of course TONS of assumptions.
We have been hearing weird and erroneous details from media (the best was the SMS and phone calls made by the passengers to their loved ones minutes before the crash...)
Investigation teams will obviously have more clue with the review of the debris and the bodies examination.
But as long as we do not find the FDR and VCR, we won't be able to know what actually happened over the Atlantic ocean that night
.
I am hoping that rescue team find it VERY SOON.
Every morning,the first thing I do is to turn on CNN to watch the headlines, go on airliners to know the latest, log on the Brazilian army web site to read the reports.
And we are a lot in this situation I guess...
Stop guessing, giving assumptions, criticize each others and medias and so on.
For sure, we will know the truth, it is just a matter of time. (Hopefully short...)

BOAC

15th Jun 2009, 15:44

I might for example make the assumption that a ditching was attempted (suggested by BOAC). Dave, so as NOT to get RWA too excited about his cs article, which has serious flaws as I posted (deleted) and PM'd (not deleted:)) and confirmed by Lost in Saigon, I should point out that I merely said, in #1599,It is even possible (as far as I can ascertain) that it might even have attempted to ditch.Not saying they did - or didn't, just pointing out that we do not know. Beware the somewhat goulish train of 'injury analysis' too - the jury is out on that one too.

Your last paragraph is a good comment.

MY personal opinion (which I detect is shared by many here) is that I have never known an aeroplane break up and then carry on for '50 or so' miles. Space Shuttle - yes. A330 I do not believe it is remotely possible.

MacBoero

15th Jun 2009, 15:45

There is... see below.. Rosy coloured teeth
Pink coloured teeth is a common feature found in corpses, and is particularly associated with corpses that have been exposed to water. It is often found in the bodies of people who have been strangled too.

NotPilotAtALL

15th Jun 2009, 16:16

Hello,

I fear computers have played a part in a tragedy.

Sure they played their part ...
They just performed nicely the job for what they are designed. :ok:
When the job required (or informations given) go over their design or limits .. (due to fail safe limitations coded) they go out the game and give hand to the pilot(s)

Regards.

PappaAlphaX-ray

15th Jun 2009, 16:16

@ RWA

A few pages back - it was invented to determine speed in water.

Lemurian

15th Jun 2009, 16:32

RWA :
it's actually called a 'pitot tube.' Invented by a French engineer called Henri Pitot about 1718. It's basically a 'Venturi tube' -
HMMmmm !That's a gem, if knew one !

Triskel

15th Jun 2009, 16:36

Re-comment by Dave Gittens:
MY personal opinion (which I detect is shared by many here) is that I have never known an aeroplane break up and then carry on for '50 or so' miles. Space Shuttle - yes. A330 I do not believe it is remotely possible.

There could be a couple of reasons for this:
1 - if the a/c broke up in the air with one part in a violent CB updraught and the other perhaps in downdraught, or falling into a different updraught, significant separation could easily occur.
2 - perhaps more likely - debris 'floating' at different depths could be sepatared by currents in the area e.g. the 'surface currents' are only a few hundred metres deep but move several miles per day relative to deeper water.

BOAC

15th Jun 2009, 16:42

Chaps and chappesses - to avoid further confusion/embarrassment/laughing at xxx/distraction from the thread, can we all refer to them FROM NOW as Prandtl tubes please? 1732 is a better date too. Don't forget good old Henry Darcy as well. Now, where were we...........?

EDIT: Thank you Triskel...........

dicksorchard

15th Jun 2009, 16:44

on yahoo news .
search for air france photos .

a few new pics of wreckage from different angles .
you guys may be interested in them .

particurly images 10 & 11 .

They show A technician of the French Bureau of Investigation, B.E.A, taking pictures of debris belonging to crashed Air France flight AF447. from the quayside .

sorry could not get them to load up .

JD-EE

15th Jun 2009, 16:46

onetrack

Not testing for extended UV performance is a critical failure on the part of AirBus. Of course, requiring that the composites always remain well covered with UV opaque paint should solve that problem.

ClippedCub

15th Jun 2009, 16:52

Can't help it, but whenever the subject of computerised flying controls comes up it puts me in mind of a programmer who has never learned to ride a bike trying to program one that would work automatically.

One can imagine the programmer saying to his cyclist adviser, "About turns, I'm not clear about them yet. If someone wants to turn, do they turn the handlebars first, or lean over first........?"

Flying has always been a mixture of art and science. Ever since 1903......

The programmer doesn't decide how the program works. Programmers write software defined by the engineers/pilots and communicated in the form of a Software Specification. Each element of the software is tested to verify compliance with the specification.

ClippedCub

15th Jun 2009, 17:00

I have never known an aeroplane break up and then carry on for '50 or so' miles.

It doesn't. The 50 miles of debris can be explained by trajectory analysis if it broke up in flight.

http://www.asc.gov.tw/author_files/Ballistic.pdf

steamchicken

15th Jun 2009, 17:28

We've had a LOT of people basically assuming that the A300 and A330 are identical, cos they're French like and they've got a computer thingy. We've also had two detailed reports of similar events with an A330 (Air Caraibes and Air France).

But the A340 is meant to need minimal conversion from the A330. And, well, designed 20 years later than the A300. Have there been any comparable events with the A340?

DC-ATE

15th Jun 2009, 17:58

NotPilotAtALL -
Sure they played their part ... [referring to the remark I made about computers possibly playing a part in this accident]
They just performed nicely the job for what they are designed.
When the job required (or informations given) go over their design or limits .. (due to fail safe limitations coded) they go out the game and give hand to the pilot(s)

You'll have to refer to the post that contained the info I was writing about. And if your Screen Name reflects your connection to aviation, I suggest you let those associated with the flight deck decide.

wizele

15th Jun 2009, 17:58

Green-dot

15th Jun 2009, 18:00

Airbus Systems have all been well tested with multiple redundancies built in. Something else happened that made both automatic and manual control impossible.

Much speculation concerns the ACARS messages - but if there was a catastrophic failure some of the more informative fault reports may not have gotten through to the COM link or they may have been queued for transmission but never made it. We shouldn't assume that the faults transmitted were the principal faults that occurred. For instance if another unreported fault disrupted power to the pitot heating, then the latter problem would be a symptom and not a cause.

Could solar weather conditions have been a contibuting factor, possibly triggering single event efects (SEE) and/or a solar proton event (SPE)?
Source:

SolarSoft Latest Events (http://www.lmsal.com/solarsoft/last_events_20090602_0723/index.html)

There was an increase in X-ray activity from the Sun (ref. GEOS 10 in upper graph), starting late May 31st which continued through June 1st and in the same time frame the coronal hole graph show a pronounced dip at the approximate time the aircraft went missing (Kyoto Det/PFSS in lower graph). The accident site is also slightly north of the South Atlantic Anomaly, an area of radiation in the form of trapped protons where the lower Van Allen belt reaches down into the atmosphere. The X-rays preceded a solar flare which occurred approx. 12 hrs after the aircraft went missing.

Pontius Navigator

15th Jun 2009, 18:02

I have a question:
It has been stated that the aircraft was facing rough weather (a strong turbulence).
But... how comes the belts of the crew seats were not fastened?? Do we know anything about the passengers: i.e. if the belts on the seats they found were fastened, if the bodies showed ligature marks due to the belts...
If the belts were not fastened the turbulence would not be a key factor anymore

I presume you mean the cabin crew. I suggest you read the thread as there are many possibilites.

As for the passengers, short answer is No. Nothing, as far as has been posted here, has commented on the passenger seat belts.

Gringobr

15th Jun 2009, 18:04

.............But... how comes the belts of the crew seats were not fastened?? Do we know anything about the passengers: i.e. if the belts on the seats they found were fastened, if the bodies showed ligature marks due to the belts...
If the belts were not fastened the turbulence would not be a key factor anymore
We do not know if the crew seat belts were fastened.. I presume you are referring to the photos of the 2 jump seats.. These are seats only used occasionally by cabin crew...
The fact that most of the bodies found so far have the " 4 fractures" ( legs and arms broken), suggests they had their belts fastened, and this at the time of relatively a horizontal impact with the sea,,

PJ2

15th Jun 2009, 18:51

RWA, DC-ATE;
Once again, a non-flying, non-experienced, non-trained, non-pilot offers an "IT" opinion on something he knows nothing about and says so, and you think the remark is "brilliant", a "truly great comment...", or "very well put..." ? Why?

Desertia states right at the start of the post:
As nothing more than a curious member of the SLF, I've obviously not posted my groundless theories on what could have happened, preferring to read the myriad of alternatives offered on here, and keeping an open mind given that it's obvious that there is still insufficient evidence to identify the cause of this crash.

But one thing I would like to say, having been in the IT field in a variety of positions over the years, is that you cannot underestimate the risk of human error, or at least limitations in putting together the millions of line of code required to automate something as complex as flight.
I have flown 18,000+hrs since 1967 and have flown as captain and as an A320 instructor, 319s, 320s, 330s and 340s since 1992, (I'm now retired) and even with that background I do not consider myself beyond learning, (fearing to go where angels tread), am not an IT or structural engineer, but I know a thing or two about flight safety work and about the Airbus (and other types, B767, B727, L1011, DC8, DC9). There are many people here with similar and greater qualiications who, because they are experts, know what they dont' know and are prepared to ask questions instead of making sweeping, grandiose pronouncements on "Airbus dangers", "fly-by-wire" or comparing the Airbus with the difficulties of designing "computerized bikes". There is simply no comparison with the metaphor in this last.

Gentlemen, no one at Airbus is underestimating the risk of human error, nor are they at Boeing. The record of millions of takeoffs, landings and hours flown by both manufacturer's transports without fatal accident proves this to be true. To support statements that claim otherwise is to offer thoughts in basic ignorance of the aircraft involved.

There is no "smoking gun" latent in these designs which good training and experience cannot address. Neither the B777 nor the 340 designs have such faults; nor are they invincible when circumstances overwhelm the man-machine interface.

Neither one of you fly the Airbus or even know the first thing about its systems or how it's built yet you're willing to condemn the airplane based on the flimsiest notions of someone who isn't even in aviation.

This is not how flight safety or investigative work is done, and if I may make the observation yet again, THAT is the problem with this thread, while the contributions of those who are the experts and who do the work go ignored, unrecognized or unread.

Specifically - RWA, you recommended the CSM (Christian Science Monitor) article as something to read to understand this accident. Frankly, it is composed of the same imaginative analysis and reporting that exists here and in any other media; it's noise and it's print filling the vacuum. For example, the article states that, "the cabin had lost pressure and was climbing at a rate of 1800fpm".

Except perhaps from the thousands of non-expert, non-pilot, non-engineer, non-maintenance comments, where did they get that from? Where does anyone get this information from? Even if it's "18,000fpm"...where does that come from? There are other problems with this article as many know.

I have the ACARS messages just like the rest of the world now does and I have access to an AMM. Interpreting the thin, tiny bit of information available through the ACARS messages is not a straightforward exercise in reading the AMM. From the start of the last thread and this one, I and many experts have advocated reserved and cautious approach when interpreting these messages. For many reasons, this list could be, (and almost certainly is, given the intent and design of the ACARS-CMS-FIDS systems and the SITA protocols handling the messages) much larger but we do not have that information.

Here is an example of just a few of the ATA messages that can be generated by the BITE system just within the CPC, (Cabin Pressure Controller). This is by no means a complete ATA fault diagnosis list for the cabin pressue controller. As far as I know, there is not ATA code for "cabin depressurization, Cabin Rate of Climb = nnnnn feet-per-minute":

http://i277.photobucket.com/albums/kk76/batcave777/CabPressCtrlATA_2009-06-15_085352.png

Here are examples of ATA fault/failure codes for pitot-static probes/sensors. The same issues obtain - the list is by no means complete.

Meaningful interpretations of ATA codes are not straightforward even if the correct documentation, tools and resident expertise are all available. This will not be a straightforward process for the BEA, for Airbus or for AF. It is folly to endeavour to think or do otherwise. Regarding the pitot "faults/failures", notice that the 3 sensors have the same six-digit ATA code. Someone here wisely observed that "underneath" this bread-and-butter ACARS message dump is very likely a highly-detailed message. We do not know if that message was transmitted or received or available.

http://i277.photobucket.com/albums/kk76/batcave777/ProbesSensorsATA_2009-06-15_110850.png

The way to find things out here in this example is to answer the question, Which is the right ATA code?, and Where is the evidence that tells us we are right?

There is no ATA code for "pitot icing". The theory is the result of hind-sight bias and not the result of any knowledge or evidence. There is only the cryptic ATA message indicating that the receiving computers, (ADIRS, TCAS, CPC, DMUs) were not receiving data that was within specified parameters and therefore caused a BITE and a latched fault/failure message. Nobody anywhere, knows why, yet.

frontrow

15th Jun 2009, 19:02

Are any details of autopsies on web?

BOAC

15th Jun 2009, 19:06

PJ - thank you for the claw back to sanity. Do you have any take on the jungle of confusion over the ISIS?

Pontius Navigator

15th Jun 2009, 19:11

Are any details of autopsies on web?

Yes. Read this thread.

mickyman

15th Jun 2009, 19:16

Any similarities with the Lauda air 767 and/or the Birginair 757?

Perhaps the 'weather' is a red herring?

MM

ukwomble

15th Jun 2009, 19:31

Informative, PJ2.

The following paper discusses the design of the 320/330/340 fault tolerant computer systems, for anyone interested in learning about the software side.

http://personales.upv.es/juaruiga/teaching/TFC/Material/Trabajos/AIRBUS.PDF

grizzled

15th Jun 2009, 19:52

(Mods: Please bear with me . . .)

If you are reading this and are new to PPRuNe -- welcome! If and when you are tempted to post for the first time:

1 Read the whole thread.
2 If you feel you don't have time for (1) above, click on the "search this thread" function and type in a word related to what you were about to comment on or ask about. Read ALL the resultant posts.
3 If you still have a reason to post, type your post into a word document (rather than directly onto the PPRuNe page).
4. Go make a sandwich.
5. Re-read your post, asking yourself: Does this post advance the thread, contribute something worthwhile to the discussion/debate, or ask a question that HAS NOT ARLEADY BEEN ASKED?
6. If the answer is "yes", click on "Submit Reply"
7. If the answer is "no", enjoy the sandwich whilst you contine reading PPRuNe.

DC-ATE

15th Jun 2009, 19:55

PJ2 -

Thank you for your comments.

My answer to Desertia concerning his post related to his comments:

....."I remain of the opinion that, irrespective of the causes of this accident, I would still rather an experienced pilot have a piece of wire or hydraulics controlling the various surfaces under unusual conditions than having one or more computers trying to work it out."

and....."Artificial intelligence still does not - and probably will not for decades to come, unless there is a magnificent leap in technology - mirror the ability of the human brain to eschew logic and do what thousands of hours of flying, and experience of abnormal circumstances allow, which is, for want of a better word, THINK."

Neither one of you fly the Airbus or even know the first thing about its systems or how it's built yet you're willing to condemn the airplane based on the flimsiest notions of someone who isn't even in aviation.

Quite true. But, I have read enough from those that HAVE flown these types to know that I'm glad I never had to. It is utterly amazing to read the differences expressed by those that fly these things as to how various systems operate. There are disagreements everywhere. The manuals are simply too full of information that not even a computer scientist could possibly remember/recall in the time of need. Who ever heard of an aircraft having THREE different flight control laws?!

Whatever happend to AF447, apparently happend so fast, the crew had no chance to sort out all the warnings being displayed on their panel. They probably had no chance to simply FLY their aircraft.

Like it or not, his (Desertia) reference to having a "...piece of wire or hydraulics controlling..." says alot, along with those of us, yourself included, who learned how to "fly" airplanes and not manipulate a computer to do it for us.

But, we digress.....

DingerX

15th Jun 2009, 19:56

Right on, PJ2.

One detail though (and this also addresses PBL's quandary): there is out there a reasonably-sized data set for ACARS messages sent during A330 pitot-tube icing events. Air Caraibes had a couple of cases (although from the report pasted, the principal diagnostic detail was the early rudder travel limiter fault). Air France had something like 6, and they obviously have the ACARS messaging practices that interest us. Presumably, other carriers should have something analogous.

So, while everyone's running around crowing about a cause, I'd suggest that, if that is the cause, someone has a body of information with a good shot at confirming or falsifying the hypothesis pretty quick. While it helps to do the piece-by-piece analysis, it helps even more to associate the whole list of messages with similar cases where the cause is known.

...not that we have access to that sort of information, mind you, but someone who did might find it illuminating.

WhyIsThereAir

15th Jun 2009, 20:09

This is a question from the unknowledgable trying to obtain knowledge by asking those who know. An explanation or a pointer to some available document would be greatly appreciated.

In the kinds of planes I know about the rudder pedals are pretty much at the ends of your legs, and you adjust them so you comfortably have full rudder travel even with the sholder harness and belt tightened.

But I've read here that transport pilots only use the rudder on takeoff and landing, and at any other point using the rudder is stupid and dangerous; you only use ailerons. (If this is incorrect I'd appreciate enlightenment, but I've read this in basically these words at least 3 times in this thread.)

If you don't use the rudder in cruise, are you still using the pedals as footrests? Do you have other footrests you can use instead? Is there a pedal disable button to turn them into footrests?

I'm sure from the kind of responses I see here to questions that many people will read this question as pushing some agenda and get angry about it. I'm not pushing an agenda. I'm asking what you do with your feet in cruise to make sure you don't input rudder movements. Is there anytihng in the design of the cabin and controls to help you, or do you just have to be consious to never move your feet? This isn't a problem with toy planes, because you most often do use the rudder in them.

Thanks, and have a nice day!

einhverfr

15th Jun 2009, 20:12

A couple points about FBW systems (I do some embedded/control sytem work, not FBW, but close enough I feel able to dispel some misinformation here). Personally I don't think FBW-issues are at all likely factors in this accident though autopilot/autothrust might be (However, in non-FBW systems these would be factors too).

The idea that it takes something like a million lines of code to automate flight is based on a number of logical mistakes. I don't doubt that once one starts adding more complex elements of avoinic systems, it might approach a million lines of code, but, this is somewhat meaningless. In short folks assume that airplane avionics control software and Microsoft Windows face similar challenges because they are both "software." Actually, this is not the case. (I have done testing on some versions of Microsoft Windows, and some embedded system design, so...)

The first thing is that software engineers who are going to work on a specific problem are going to have to understand the nature of the problem well enough to have intelligent governsations with specialists as to finer points. Specifications need to be understood in context, and software engineers need to be involved in the specification process. The idea that avionics systems are designed by folks who have no general aviation knowledge is silly (and there would have been a lot more accidents if this were the case).

The second thing is that most of the avionics units are actually going to be reasonably simple matters of sensor input and data output. ADIRU's, the TCAS, VSC, etc. all fit into this category. These systems can be validated both symbolically and via automated testing in ways that general-purpose computer programs can't. A simple rule to bear in mind here is that the effort required to validate software is proportional to 2^n where n represents the compexity of input. Simple input means robust testibility. ADIRU systems for example, would have reasonably simple inputs. Also simple inputs have the advantage in that it is practical to symbolically analise source code here to determine that it WILL handle all sensor input appropriately. Some systems seem less subject to full validation (TCAS systems for example) than others (ADIRU's, for example).

However, since one system's output is usualy an input elsewhere, more restricted output improves testability of the entire system.

Controllers, such as those found in avoinics systems, are remarkably simple, testable, and even verifiable via non-testing methods in a way that general purpose computer software is not. Embedded controllers are everywhere-- in our offices, cars, homes, etc. and have been shown to be reliable. Some elements may still be beyond testability (in particular, you can't protect against sensors inputing consistently incorrect data in this way, nor can you rule out every possible timing issue, but you CAN experientially and even mathematically prove that correct inputs will be acted on correctly).

Now, I do see one POSSIBLE issue with computerized systems in the AF447 incident involving the issues of self-healing systems. Pitot icing detection works on the assumption that there are at least mild differences in severity in the icing in different tubes (hence different stagnation pressures detected). The computers typicaly vote out the outlying system and in this case, this can be dangerous because it is possible that the outlying system is more correct than the two (more consistently iced up) tubes report. In this case, autothrust might have been increased prior to autopilot disconnect and the plane might have been going substantially faster than anyone understood at the time it was handed back to the pilots.

All in all, I think it is possible that self-healing element of FBW systems played a role in this incident. However, this isn't a matter of a bug or a glitch but rather the (human) difficulty in maintaining situation awareness with a computer with which one interacts in limited ways. Such degraded situation awareness may have been a contributing factor. I don't think this means that systems shouldn't be self-healing, but rather that pilots might do well to have more transparent diagnostic information available to them (including outlying readings from instruments).

Will Fraser

15th Jun 2009, 20:22

PJ2 I do not wish to put you on the spot.

Can you direct me to information I can assimilate that will answer a few basic questions re: AB a/p disengage and degrade to Alt Law, and the parallel and exquisitely important well studied and trained for reaction by human crew who, after disc a/p must fly the a/c at a higher skill level than the just abandoned automatics, and do so without aileron, a taste of Pitch trim, and a Rudder that deflects to a predetermined value, however appropriate?

Bear with me. I am understanding that with autos on in cruise, and a programmed set of predetermined limits relative to G, roll, (rate and limit) and protections, there is a point at which the a/c logic abandons a challenging controlled flight regime to pilots whose skills are doubted enough to recommend that a/p not be disconnected, and the computer disconnects due to unreliable airspeed data? It may seem untoward, but I get the distinct feeling that at auto disconnect, the a/c is trying to escape some kind of Blame for what may ensue.

ernst_mulder

15th Jun 2009, 20:53

Okay, before you read any further: I'm not a pilot. Nor an air crash investigator. Nor is English my native tongue. And this is my first posting here. Feel free to skip.

After reading this forum for some time now (and learning a lot) and since whatever any of us offers as an explanation no one knows the true chain of events, I venture to vent mine. Pure speculation. I would like answers to my four questions though.

Some questions I would like to ask first:

a) What happens when two of three systems monitoring flight parameters break down simultaneously and in a similar manner? What would the flight computers decide? Would the one remaining system deemed defective?

b) How severe can a lightning discharge be? Could it damage (melt) parts?

c) Could severe movements of the plane cause the ISIS gyro to produce an error condition?

d) Could a lightning discharge through the plane cause confusion in the TCAS?

I realise that the ACARS messages are not necessarily in exact chronological order. But what about the following sequence. Only logical if the answers to these questions is YES.

1) Auto Pilot switches off because of severe turbulence conditions caused by the lightning storm. More turbulence causes ATHR to switch off as well. Since both are 0210 time stamped I find it hard to believe that the pilot switched ATHR off (unless the pilot switched both AP and ATHR off intentionally).

2) A huge horizontal lightning discharge enters pitot tube on one side of the plane and leaves the plane at the other side by the pitot tube on the other side. Both are fused (partly) closed. Is this possible? Lightning does like sharp points.

3) The pilot flying is startled so much by the lightning discharge that by reflex he puts his feet down stepping hard on one of the rudder pedals.

4) The lightning discharge's induced heat confuses the outside air temperature sensor.

What happens next?

Joss

15th Jun 2009, 20:57

You keep throwing theories like crazy... Hoping that eventually someone will nail it provided you keep trying long enough.

It won't work! Dont you understand?

You build a theory and then try to match to facts to it and if there are no facts you invent them. Science works the other way, letting the facts tell the story and then explaining why, how, and so on.

This post has reached a point where it is self-sustained. The facts dont matter anymore, just with speculation and debunking speculation there is enough to keep it growing.

Even if no new evidence is ever found and therefore no new fact can be extracted this post will keep growing and growing out of shear speculation.

:ugh:

ACLS65

15th Jun 2009, 21:11

To those bemoaning FBW your ancestors probably expressed the following:

Why would I want to abandon wing warping for those flappy new hinged control surfaces, what if they fall off?

Retractable gear! No way, what if those darn things won't come down?

Enclosed canopy, what if I need to jump out?

Hydraulics, there is no feel, give me cable all the way!

Like it or not computers are progress and likely inescapable. Ask the B-2 or F-117A pilot who rely upon them even more.

Like any "new" technology there will be bugs, but in the end future pilots will just see them as the way things are, and be busily complaining about wing mounted solar panels, or fusion reactor heat issues. :)

Jetdoc

15th Jun 2009, 21:18

http://i658.photobucket.com/albums/uu308/Jetdoc1/Slidetableandfootrests.png

http://i658.photobucket.com/albums/uu308/Jetdoc1/Rudderpedalcoverandrest.jpg

Jetdoc

15th Jun 2009, 21:23

I think what you have got there is the AMM references for the maintenance of those components, not fault codes.

Mike-Bracknell

15th Jun 2009, 21:25

Quite true. But, I have read enough from those that HAVE flown these types to know that I'm glad I never had to. It is utterly amazing to read the differences expressed by those that fly these things as to how various systems operate. There are disagreements everywhere. The manuals are simply too full of information that not even a computer scientist could possibly remember/recall in the time of need. Who ever heard of an aircraft having THREE different flight control laws?!

Whatever happend to AF447, apparently happend so fast, the crew had no chance to sort out all the warnings being displayed on their panel. They probably had no chance to simply FLY their aircraft.

Whenever I read these loooong threads about crashes, and the myriad of people complaining about non-professionals putting in their 2p worth, previously i've wondered if people have too short a fuse. Now however you're talking my specialist subject, and i'm biting my tongue a bit on the relative opinions expressed on computer-related flight systems.

What I will say however is that I agree with this quote - i.e. despite the slow creep of computer systems into flight, it's obvious that the knowledge and workload on the average Captain & F/O are at times still way too high to be able to "just fly the aircraft" when needed.

I will stop short of suggesting that it's human preconception that is stunting the introduction of more computerisation into systems, but i'm willing to bet that those people thinking "computers" are thinking "MS Windows and blue screen of death", and in that sense you couldn't be comparing apples and oranges much harder if you tried.

Granted it's not all there yet, but i'd state it's not far off. For instance, in this day and age it shouldn't be beyond the wit of man to be able to reliably plot where you are on the planet in the X, Y, and Z plane, and given the satellite height data of the globe it's therefore almost inexcusable (IMHO) that we still have interfaces with the terrain. That's a pretty facile example, but you get where i'm coming from.

Maybe in a forum of pilots it's a pretty unpopular stance to take, arguing that computer control could probably do the job in the long run, but step back 50 years and see where we've come from there. Now step forward 20 years and you're probably going to find more and more of your beloved flight controlled by computers. It's something you might hate, but it's a trend that's only going one way IMHO. Thankfully, it's going the same way as the safety figures, and I for one am pretty sure there's more than just a vague relationship between the two.

Anyhow, this is drifting about as off-topic as it could get, and so i'll end it here as it's probably a thread in it's own right and I really wouldn't want this to turn into a s**tfight either. In fact, if you want to reply to this one, take it to a new thread - I encourage it :ok:

DC-ATE

15th Jun 2009, 21:28

ACLS65 -

One of the problems with this "progress" is that the only ones benefiting from it are the manufacturers. The airlines are in hock up to their ears buying all this new equipment trying to convince their passengers that they are equipped with the latest. The airlines have to cut costs to pay for all this. Whereas you could pull up to the gate in a DC-3 and people would still get on. FBW has its place, no doubt. But it was NEEDED to control high performance aircraft, not airliners.

As to: "Like any "new" technology there will be bugs..." The 'bugs' were pretty much taken care of with the "old" technology. Why put passengers through yet another learning process?

Whether or not new or old, we still have to get around the "pilot error" issue somehow, don't we? I'm not saying that was an issue with AF447, but there WERE other aircraft in or near that area in the same time frame that seemingly got through with no trouble. So.....was the crew at fault, or their systems, or the weather? Sadly, we will probably never know, as even if the recorders are recovered from, what is it, 12,000 feet?!, the data might not be useble.

Towhee

15th Jun 2009, 21:33

And these files are surprisingly small. William Cecil, marketing manager of data acquisition and wireless solutions at Teledyne Controls, estimates that the flight recorder on an Airbus A320 or Boeing 737NG accumulates 1.8 MB of unformatted data per flight hour.

Packets of raw engine data amount to as much for an entire flight. Even so, potentially useful flight data is frequently minimized, postponed or forgotten.

Why are operators so miserly with MRO data? Because ACARS is narrow-band ("an extremely slow, skinny pipe," Teledyne's Schmitz called it) and pricey. Messages are limited to blocks of 220 characters, one block at a time, up to a maximum of 16.

Throughput in VHF mode is 2.4 kilobits per second (kbps), the rate of an antique modem. Even VDL Mode-2, a recently introduced VHF variant with 10 times the bandwidth, clocks in at only 30 to 40 kbps.

Usage charges vary with a variety of considerations, including message size and communications channel, and prices have been falling under pressure from VDL-2 and Iridium's new satcom option. Still, the range is roughly five cents a message (VHF) to between $1.50 and $20 per minute (satcom).

In addition, Bob Bouchard, vice president of technology solutions at The Longbow Group, questioned ACARS's reliability: "A Wi-Fi or cellular connection uses standard TCP/IP. If a file is skipped, it resends it.

If it's missing, you'll know. With ACARS, you send a message hoping there's a ground receiver and really no acknowledgement it was received."
Given these technical limits, the possibility of using ACARS for sizable or continuous MRO data transfers is, William Cecil said, "a dream."

...
In-Flight Datalinks

Data rate per second, Range

ACARS (Satcom)9.6kglobal
ACARS (VHF)2.4k240 mi
ACARS (HF)0.18k5,000 mi
ACARS VDL Mode-2 (VHF)31.5k240 mi

from:

OverHaul & Maintenance, Jan 1, 2008

SLFinAZ

15th Jun 2009, 21:34

Like it or not computers are progress and likely inescapable. Ask the B-2 or F-117A pilot who rely upon them even more

The difference is that the FBW software in the above isnt trying to "protect" the airplane from the pilot. I think that one inescapable question in this incident is what role the complexity of the software contributed to the end result. In some ways this reminds me of the conflict between the original astronauts and the NASA. My humble opinion is when in doubt you need better pilots, not more complicated software. When it all goes south the guys in the pointy end still need "the right stuff"...not the "right OS":ok:

einhverfr

15th Jun 2009, 21:40

@DC-ATE:

I actually see FBW as being driven by something else entirely: need to save weight and reduce operating expenses.

I don't know of anyone who prefers to fly on an A330 vs a B747 just because the A330 is a more recent model. People choose the flights. Then the airline chooses the plane to match the flight.

One funny experience though flying from Denpasar to Jakarta a couple years ago on Merpati Airlines (shortly before Indonesia tightened age limits on commercial airplanes)... The plane B737) was really late. We got on the plane and I looked out the window. Then I said to my wife, "This plane is older than I am." I knew this because the engines were simple turbojet engines with no fan bypass. It was a different experience than I was used to also. Much more smooth acceleration (probably because turbofan engines provide much more thrust at slow speeds).

FBW is the future. That is fine. But I don't think anyone is in a hurry to retire serviceable, airworthy, and legal airplanes just because a new technology comes out.

Sparelung

15th Jun 2009, 21:41

Here's a extract from a scientific reference to show that teeth discolouration in the dead is normally found after prolonged water immersion and is not normally a conclusive indicator of cause of death.

Full abstract is here - Medico-legal aspects of postmortem pink teeth. [Int J Legal Med. 1994] - PubMed Result (http://www.ncbi.nlm.nih.gov/pubmed/8068567)

Borrman H, Du Chesne A, Brinkmann B. Medico-legal aspects of postmortem pink teeth. Int J Legal Med 1994;106(5):225-31.

"Except for very few and poorly documented exceptions, [pink teeth] develop earliest after 1 to 2 weeks postmortem.

...they are most often associated with water immersion.

The phenomenon is very often seen in victims of drowning where the head usually lies in a head-down position.

Since, there is no obvious connection between the occurrence of pink teeth and the cause of death, it may be concluded that pink teeth are not pathognomonic for a specific cause of death and this is therefore an unspecific phenomenon."

Art-Deco

15th Jun 2009, 21:54

Not a pilot, but I checked all pages and nobody has looked into this before:

Does anyone here know the timing-specs of the temp sensors on an A330 ?

As all temp sensors has an delay-time of x seconds (average 10-40 sec if change is over 20 Celcius directly) before indicating the correct temp when major changes occur in temp, it may could be one more source to the trouble for this flight with all the previous details of this flight such as coffin-corner and flying into a cb area with possible warmer zones.

Edit: Enclosing footprint map of Inmarsat/Satcom were ACARS is used, have a look on the AOR-area, it may be possible that the flight also entered a poor-reception area in it´s final stages.

http://www.alphatelecom.ru/inmarsat/images/engmap.jpg

Alwaysairbus

15th Jun 2009, 22:03

Sorry lost track after 1600+ posts,

So could someone please just outline the actual facts that are known at this point of time please.... what ACARS ECAM/Fault messages have been published (and from the experience i have on Airbuses, don't mean a lot with respect to what systems are actually doing), what remains of the airframe/engines (presumably none for the GE's) have been found, how the location finding is going with respect to the DFDR/CVR and what Airbus/ the Airworthiness authorities have relayed to operators so all the guessing / expert analysis can be put in it's rightfull place....

ELAC

15th Jun 2009, 22:14

PJ2,

A great explanation, but a wasted effort I think.

This thread has degraded completely and become a contest between those with a long held fixation on finding fault with FBW/Automation/Airbus in general and those that come here without the slightest bit of practical knowledge and post their commentary based on their in depth experience working in scrapping yards or IT departments or what have you. It's even gone so far down hill that I see someone has recently posited that it all may have come down to to X-Rays emitted by a Solar Proton Event and someone else suggests that it's the result of an uncontained engine failure apparently resulting in a failure of the entire aircraft structure. Throw in the previous suggestion that a meteor did it and it all becomes just too much!

What I hope that the many who read this thread looking for facts, informed insight and considered speculation will have realized is that, with the exception of a few noble efforts such as PJ2's, the thread has been taken over by individuals who for the most part don't know one end of an A330 from the other but seem intent on sleuthing their way to an answer to this accident that fits their preconceived notions about the aircraft. With so few facts available and just enough "clues" trickling in to power any form of fantastic supposition they are having a field day.

There is a valuable discussion that we could be having about the few things we do know, the implications they suggest about what this flight might have experienced and how those with experience have overcome similar circumstances in the past. We're not having that discussion and those who've followed enough PPRuNe threads will recognize that most of those who have contributed knowledgably in the past are absent from this thread. The reason for this is that it's become basically impossible to carry on any degree of informed discussion when it gets lost under the random speculations of anyone capable of registering a user name, or becomes fodder for the various "Captains" and other self-annointed experts who sieze upon any crumb of actual technical knowledge as a lever to forward their own pet theory about how the automatics, fly-by-wire, composites or just the French in general are responsible for this accident and the ensuing cover-up which they presume is under way.

There was a time when PPRuNe was a valuable venue for discussion of professional issues, particularly when events such as AF447 occurred. Unfortunately that time has passed, and the informed discussion is now more likely to occur elsewhere.

ELAC

With apologies to our hosts who I know are trying their best.

ACLS65

15th Jun 2009, 22:16

DC-ATE
One of the problems with this "progress" is that the only ones benefiting from it are the manufacturers. ... Whereas you could pull up to the gate in a DC-3 and people would still get on. FBW has its place, no doubt. But it was NEEDED to control high performance aircraft, not airliners.

The DC-3 was a great plane in its day, if we were all using it today we would be doing 130kn at FL24, and taking the China Clipper overseas. Not sure if that would "fly" with the public today.

Give it some time and airline pilots may end up working from home, as today's Predator and Global Hawk pilots are getting closer to doing now.

In terms of AF447, I suspect there will be an old bug (like pitot icing), a current bug (like how the flight controls handle multiple lost inputs), some human error or confusion (information overload), an act of god (the storm), and quite possibly something that none of us have even imagined yet. There could even be a spilled cup of coffee involved.

The key is that we learn the most we can from this very expensive lesson, and I am not talking dollars.

captainflame

15th Jun 2009, 22:32

Graybeard:

"So why wasn't there an ADR Fault transmitted at 0210, if that were the source of the TCAS Fault as you say?"

First, we are not sure of the sequencing logic of the faults on the ACARS messages. Is it really time sequence? or perhaps does it reflect the ECAM logic ?

Second, the sequencing logic of the ECAM is not a time sequence but an order of priority.

It seems to me that it is very possible that the ACARS sequencing logic follows the ECAM logic (to be confirmed by an engineer)
AP OFF, the first message one at 2:10 is a Master warning, and would be on top of the list on the ECAM !

Flags on PFDs would show up early on.

In other words, if the ACARS messages are listed as on the ECAM, at 2:10 all the secondary faults show up, the NAV ADR DISAGREE would be on top of the master cautions, as it is the "deepest fault", reporting on the ACARS message in the end !

It's a bit puzzling but the whole thing points to a pitot failure of all pitots at the same time practically. (which makes sense if they are all of the same making, all with the same design problem, all fail within the same conditions)
Including STBY for the ISIS !

ALT Law requires failures of 2 ADRs That's also when you get flags on PFDs.

Lastly, only a failure of ALL ADRs will trigger the associated TCAS inop. !!

On this: ADRs feed ATC box "w/ airspeed, Mach and V/S upon ground request" (from FCOM)

What's missing is the "WINSHEAR Prediction system Fault" and GPWS TERR Fault!! what happend to those ?? Because they should be right there along with the other faulty stuff associated....

Anyway, as you can understand by now, I don't buy any stuff about TCAS antenna broken and so on.

Cheers.

augustusjeremy

15th Jun 2009, 22:41

The new pitots - Those that now are replacing the older ones.
Why are they better, specifically ? Larger Diameter, length, more effective anti-icing system ?

Broomstick Flier

15th Jun 2009, 22:42

Hi,

I just received this from a friend who happens to be a capt. on the A320 and flight time on the RHS of the A330 as well. I don't know the source and when/where the simulator flight was done

Dear Ladies and Gentlemen,

The following information that is posted in this reply is something that may interest you all.
The scenario was conducted several times and the results at the end of each scenario produced consistent findings.

In an A330 simulator at FL 350 with a gross weight of 210 tonnes in ISA+10, with icing selected, the aircraft approaches a thunderstorm with a high intensity of turbulence. Due to the extreme turbulence, the autopilot disengages. Shortly thereafter a malfunction is selected to block both captain and first officer's pitot tubes to replicate extreme ice formation.

The airplane reverts to alternate law with protection lost. There is a speed flag on both the captain and fo's PFD. The severe turbulence activates repeated stall warnings. Manual thrust is being used at this time. The speed on the standby altimeter is reading 240kts or thereabouts with MACH .72. From the GPS the ground speed is 350 kts or thereabouts. It is very difficult to read the instruments and ECAM warnings.

Updrafts take the aircraft up to FL 370 and produces a negative G of .2. The aircraft then enters severe downdrafts and the rate of descent averages more than 19,000 fpm. The instinctive reaction is to pull on the stick to arrest the rate of descent. The aircraft shakes and buffets violently. The G force on the SD reads +5 but the instructor's panel shows +8. The aircraft breaks up in flight around 20,000 ft.

After several attempts at this with all results being equal one could not see AF447 sending out any distress signals if this is what happened to them. Applying an unreliable airspeed memory item would have proven to be very difficult because of the violent shaking and opening a QRH for an ADR check procedure even less likely.

Food for thought...

HarryMann

15th Jun 2009, 22:49

What's missing is the "WINSHEAR Prediction system Fault" and GPWS TERR Fault!! what happend to those ?? Because they should be right there along with the other faulty stuff associated....

Could it be those are perhaps inhibited in the cruise ?

JD-EE

15th Jun 2009, 22:50

Broomstick Flier

Is your report perhaps a recycled version of Poison's report, http://www.pprune.org/rumours-news/376433-af447-53.html#post4987398

Will Fraser

15th Jun 2009, 22:52

syseng68k

15th Jun 2009, 23:18

"The speed on the standby altimeter is reading 240kts or thereabouts with MACH .72."

Sorry to be picky, but ?...

ClippedCub

15th Jun 2009, 23:23

Updrafts take the aircraft up to FL 370 and produces a negative G of .2

Don't generally think of airlines under negative g, and g of -0.2 means the aircraft went from 1.0g, or more, giving a -1.2g or higher increment. When they developed the F-16 side stick controller, they found the pilot, (arm mass?), over controlled under negative g. To alleviate this human factors, (arm mass dynamic?), they limited forward travel of the side stick controller to a tiny amount, 1/100th of an inch if I remember. All other motions were in the x/16th scale, i.e., 5/16 inch travel.

Don't know if the sim replicates that, but if it doesn't then the 'push' could have been greater, setting up a steeper descent.

ClippedCub

15th Jun 2009, 23:33

One of those forms would be overloading the pilots with "useless" (under the circumstances) information.

In modern fighters, the weapons systems are very complex. The flight control/management computers are trained to only bug the pilot with sequenced, pilot action only, messages during combat engagement, since he's busy targeting, etc.. For example, if he gets an engine hit, the computer extinguishes the fire, shuts down the engine, and tells him he lost one. No pilot action required. Would have thought some of that logic structure would be present in civilian airliners.

HarryMann

15th Jun 2009, 23:45

OK, three pitots fail, (it happens). should the a/p disqualify itself, drop out and demand by SOP that the pilots fly pitch and power (not a memory item, here. Well a sensible SOP then would be to lookup equiv. attitude/power when first established in new cruise condition? Effectively a refresh memory item.

Will Fraser

15th Jun 2009, 23:52

Harry, it's bumpy, dark, noisy, and the glass is downtown Las Vegas on tilt, let's read from the Bible?? Darn, where is that mfer.

Let me admit at this point to my brethren that I have strayed, I will take a time out.

ClippedCub

15th Jun 2009, 23:55

Both the F-16 and B-2 are statically unstable without the artificial stability provided by computers. (Some of the WWI fighters were unstable too, but the pilot could figure it out because of the low speeds.) The F-16 is unstable so it it can pitch faster in combat. Somebody figured out that it didn't need to be that way years later, what I suspected, and that designing for pitch instability was not necessary since you're at M=0.7 and above. Everybody just got excited about something new since they could do it, and advanced the technology.

HarryMann

15th Jun 2009, 23:56

OK, 'when first establishing new cruise condition'... that'd be way back when life was simple and the lights were on.

BTW. Does anyone know the undamped dutch-roll-yaw amplitudes or frequency?

bsieker

16th Jun 2009, 00:03

[...]
This with an automatic penalty in controllability, to wit, cancel roll, no protex, stab trim only, and 8 degrees of Rudder travel, still enough if cycled to rip off the VS. ????

Glad to see PJ2, ELAC and others still taking the time to write long and knowledgeable replies and explanations. Let me add perhaps a few comments, though not quite as eloquently.

Will Fraser, would you please, like so many others, get your facts straight before asking such questions.

Again, what is Alternate Law (Protections lost)?

It is a load-factor demand control in pitch, and roll is controlled as in any conventional aircraft, direct stick-to-surface deflection.

Where does the idea come from that in this control law aileron control was lost, and only manual stabiliser trim and rudder was available? What you are talking about is the last reversion step, still far removed from Alternate Law. It is only in effect in case of loss of all Flight Control Computers.

The rudder travel limiter is independent of Normal or Alternate Law, although it is obviously not independnt of reliable airspeed information. So that Alternate Law and RTL fault co-incide is not surprising here, but Alternate Law does not automatically disable the RTL.

There is seemingly even more confusion about what the computers do, can do, don't and cannot do, both in Airbus and in Boeings.

What people talk about regarding "the computer flies the aircraft" is true for most of the time in most flights in all airliners. The computer doing this is called an Autopilot. Boeing and Airbus are no different.

What the fly-by-wire system does is steer the aircraft precisely and deterministically in the way the pilot commands it to. It rolls at a commanded rate and gives a commanded g-load factor. These are some very basic feedback-control loops, nothing could be farther from "Artificial Intelligence". In no situation is there anything remotely resembling a computer "trying to outsmart the pilot". Whether or not over-reliance on automation leads to an erosion of basic flying skills is another matter, but looking at recent accidents with Airbuses, Boeings and Bombardiers I think that problem is not specific to Airbus crews.

The most prominent exclamation cited to show how confusing computerised aircraft are is the "What's it doing now?" Again, this may be true, but it always relates to mode confusion, which is a problem with modern, highly complex autoflight systems ("Autopilot/Autothrust"). It is common to Airbus, Boeing, Embraer, Bombardier, and has absolutely nothing to do with the fly-by-wire system.

DC-ATE,

During Flight-Law-degradation in Airbus, the pilot does not have to figure out what is going on, and does not need to remember how to control it entirely differently in Normal, Alternate (red. prot), Alternate (no prot) or Direct Law. It's always the same: push the stick: nose goes down, pull the stick: nose comes up, push it left: plane banks left, push it right: plane banks right. It is an airplane, and it is flown like an airplane. The difference is the level of protection (which in this form, non-FBW aircraft don't have in the first place), and the amount of "smoothing" the computers do to achieve the desired trajectory.

Even if it crops up again and again in this thread, there is absolutely no indication that flight control had reverted to manual reversion, i. e. rudder, pitch trim and thrust only.

but the software will not let the plane be flown until the warnings are cleared.

Where does this come from? Are you seriously implying that, like a modal dialog in Windows Software, all operation of the flight controls is blocked or suspended until all ECAM procedures have been finished?

The aircraft can always be flown and will always be flown if that is necessary before any troubleshooting and other procedures can begin.

Again, that this is not always the case is well-known, but by no means an Airbus-only problem. An Adam Air B737 was lost not too long ago, because both pilots were troubleshooting the ADIRUs, and no-one was flying the airplane. It broke up in mid-air when recovery was finally attempted after the bank angle had reached 100 degrees, because in trying to arrest the steep dive the ultimate load limit was exceeded. Even in Alternate Law, Airbus would have prevented excessive load; if I remember correctly, the Adam Air flight reached some 3.7G before it disintegrated.

It is interesting to note that this flight broke up because it had no load limiting protection, and in the A320 accident at Perpignan the crwe might have wanted a bit more than the limited 2.5G. Whether or not 3G would have been enough to stop the descend before the water surface I cannot say.

As to flying "Pitch & Power": As has been mentioned before, the memory items are: disocnnect A/P and A/THR, set CL thrust (will be set as soon as A/THR is disconnected in the "normal" way), and 5 degrees pitch up (depending on altitude and flight phase).

There are then longish tables to be consulted which exact N1 value and pitch angle is to be set at which altitude. This may indeed be nigh impossible in severe turbulence at night.

ClippedCub,

Sometimes an engine that is on fire is still producing thrust that may be needed. It is the pilot's decision (on an airliner usually not engaged in a dogfight ...) when to shut it down. It should not be left to the computer. And on an airliner flight deck there are two pilots, unlike most modern fighters.

Bernd

ClippedCub

16th Jun 2009, 00:11

Engine fire point taken. Just an example. If the software is saturating the crew at the worse time with warnings and paper look-up requirments, perhaps some priority setting criteria would be useful. The military house doesn't always talk to the civil side.

ELAC

16th Jun 2009, 00:25

ELAC Put on your RayBans and read my hostile question re: AirBus, the a/c that is fully automated, almost flies itself, recommends no one touch the a/p in turbulence, but as it deftly maneuvers through hostile weather, it quits when it can't take the level of weather/dataloss/ etc. and drops it back in the laps of the pilots it generally disdains, but thinks nothing of quitting when the 's' hits N1.

Will,

You are so seriously barking up the wrong tree that it's hard to know where to begin. For starters I'd have to guess that you've never flown any other jet transport aircraft with an autopilot then? If you did you'd know that each of them has limits to their range of operation after which they must disconnect.

For example, ever try doing a go-around on the autopilot in a 757 or 767 with an engine failed? If you did you'd know that there was always a point where the autoflight went from triple to single channel at which you could expect the autopilot to kick-off due to insufficient remaining rudder control at certain trims, weights and configurations. By contrast I never once witnessed a similar event in an A320/A330/A340.

Any autoflight system can only work within its designed and certified limitations and when those limitations are exceeded the autoflight system must return control to the pilot. From my experience with both Boeing and Airbus the usable range of the Airbus autopilot is every bit as broad, and probably broader than that of any other jet transport in current use.

In a situation as extreme as the one you suggest (which is simply a hypothesis, hardly a certainty) pretty much any autopilot would have packed it in. So, if that's what occurred in this case, it had very little to do with the manufacturer and a whole lot to do with the extreme nature of the conditions encountered.

This with an automatic penalty in controllability, to wit, cancel roll, no protex, stab trim only, and 8 degrees of Rudder travel, still enough if cycled to rip off the VS. ????

You really have no idea what you're talking about do you? You are mixing up flight control laws and autoflight, two entirely different matters. Further it seems that you believe that all flight control computation was lost ("stab trim only") when there isn't even the slightest evidence suggesting that. The only thing we know with respect to the aircraft's flight control laws is that they changed to Alternate Law (Protections Lost), which in itself is hardly a crisis. I should know having flown and landed Airbuses in Alternate and Direct law.

Regarding the rudder on the A330, rudder deflection is limited as a function of speed. After a rudder travel limiter fault "the maximum rudder deflection remains at the value reached before failure". That value is dependent on speed, which according to my manuals at M.80 and FL350 (272 KIAS) is approx. 6 degrees (not 8). By contrast, if a similar failure (Rudder Ratio) occured on a B757, do you know what you'd get? The answer is that on the B757 the only thing that happens is that 1 of 3 hydraulic actuators is depowered and the pilot is given a warning to "Use rudder with care above 160 kts." So, you tell me, which rudder control system would provide the greater margin of safety in the conditions you described? One that was continuously calculating a safe limit and imposed it at failure of the normal system or one that simply reduced rudder control power by a flat 1/3rd and said "Take Care!"

The point of all this, Will, is that you are chasing after your own personal bogeyman without knowing in detail how the A330 or other similar aircraft would have handled the scenario which you suppose may have caused the accident. You have no facts that support your supposition so instead you start prevailing on PJ2 or others to provide you with what you consider a smoking gun, but which anyone who is professionally familiar with the aircraft would tell you is either inconsequential or inconclusive.

ELAC

DC-ATE

16th Jun 2009, 00:36

bsieker -
The difference is the level of protection (which in this form, non-FBW aircraft don't have in the first place), and the amount of "smoothing" the computers do to achieve the desired trajectory.

Ah.....but my "old" airplane, along with others with 'conventional' controls, HAD this protection: it's called Design Maneuvering Speed; the speed at which application of FULL available aileron, rudder or elevator will NOT overstress the airplane. It approximates Rough Air Speed.

ELAC

16th Jun 2009, 00:40

Quote: DC-ATE
Ah.....but my "old" airplane, along with others with 'conventional' controls, HAD this protection: it's called Design Maneuvering Speed; the speed at which application of FULL available aileron, rudder or elevator will NOT overstress the airplane. It approximates Rough Air Speed.

And you figure that's not available with "new" airplanes? Enough said and my point made about why this thread has degraded to catering mostly to the rants of the uninformed.

ELAC

EmBee

16th Jun 2009, 00:51

When should the a/p disconnect? I guess that is less about raw technology as the nerve of manufacturers safety engineers, pilots unions or the press (pilot error, particularly in difficult circumstances, is usually graciously forgiven, computers never!). For a peep at the future of the a/p and FBW try:-

Rockwell Shows Off Self-Healing UAV | AVIATION WEEK (http://www.aviationweek.com/aw/generic/story_channel.jsp?channel=defense&id=news/HEALING050509.xml)

or

Flying on a Wing and a Half - Defense News (http://www.defensenews.com/story.php?i=3609855)

…. although I guess that software would have had a hard time without attitude data or perhaps airspeed data!

selfin

16th Jun 2009, 00:52

Ah.....but my "old" airplane, along with others with 'conventional' controls, HAD this
protection: it's called Design Maneuvering Speed; the speed at which application of
FULL available aileron, rudder or elevator will NOT overstress the airplane. It
approximates Rough Air Speed.

There is a widespread misunderstanding among pilots about the degree of
structural protection that exists when full or abrupt flight control inputs are
made at airspeeds below the maneuvering speed.

NTSB/AAR-04/04 (http://www.ntsb.gov/publictn/2004/AAR0404.htm), p. 160, Conclusion #16.

BarbiesBoyfriend

16th Jun 2009, 00:56

Basic hand flying skills must be more difficult to maintain on the Airbus series than on any other type.

When it all goes 'tits up', ANY pilot should be able to fall back on the basic skill-set he learned before his first solo.

We all understand the basic utility of "automatics".

They are there to allow you to eat your breakfast or read the paper or look out the window (!).

They are NOT a substitute for that stuff you learned at your flight school!

I see young FOs these days that can barely fly the a/c.

The real worry tho' is that they think they are completely 'bulletproof'.

I'll bet a quid that this a/c was being flown by the F/Os.

If I'm right, and they were typical 'autos ON' F/Os, they might have found their lack of recency in holding the dog**** controller less than helpful.

Chu Chu

16th Jun 2009, 01:24

I always figure the simplest explanation's the most likely, and the simplest explanation of the VS breaking off is impact with the water. But if it came off early on, could the resutling yaw have caused the inconsistent pitot readings?

BlackBird

16th Jun 2009, 01:43

Since I haven't read through all 85 pages of posts; I was wondering if any airbus gurus know if the A-330 has an AoA indication on the PFD, and if so, where it gets it's information.

My company use to train flying an air transport category aircraft with just the AoA.

My apologizes if this has already been discussed.

augustusjeremy

16th Jun 2009, 02:04

Sorry. Too many posts... I have found it at last:

SB A330-34-3071 (first published in September 2007) proposes to change Thales Avionics pitot probes P/N C16195AA by pitot probes P/N C16195BA.QUOTE
REASON/DESCRIPTION/OPERATIONAL CONSEQUENCES A320 aircraft family operators have reported airspeed discrepancies while flying under heavy precipitations. A new Pitot probe has been designed to improve A320 aircraft airspeed behavior with these specific weather conditions. A few similar discrepancies had also been reported by A330/A340 aircraft operators and AIRBUS now proposes this Pitot probes improvement for Long Range (LR) aircraft. This Service Bulletin proposes the replacement of the three Pitot probes PN C16195AA (FINs 9DA1, 9DA2 and 9DA3) by new probes PN C16195BA. Accomplishment of this Service Bulletin will improve the resistance against water ingress under severe conditions resulting in improved airspeed behavior. In addition, this new Pitot probe introduces a new external protection layer to prevent corrosion.UNQUOTE
Please note that both of these SBs are RECOMMENDED only, meaning that it's up to the operators to decide upon the embodiement on their fleets.

To conclude, P/N C16195 has been improved by Thales Avionics to cope with some minor (because no AD was raised to cover those two SBs) in-service issues.

Therefore, the various retrofit campains that have been mentioned in the previous pages are led on a volontary basis by the operators.
No TFU on A330 airspeed discrepancies has been published by Airbus meaning that, even if some incidents have occured (see Air Caraïbes above), it was not on a widespread basis and/or no major safety issue was identified.

galaxy flyer

16th Jun 2009, 02:11

HAD this protection: it's called Design Maneuvering Speed; the speed at which application of FULL available aileron, rudder or elevator will NOT overstress the airplane. It approximates Rough Air Speed.

I believe the investigation into the AA A-300 crash in NYC put lie to that assertion. Design Maneuvering Speed provides only protection in the sense the wing will stall at or before reaching design "G' loading, it is not protection against gross mishandling or undesigned loads, like full and reversed rudder deflection or rolling "g"s.

GF

PJ2

16th Jun 2009, 03:05

DC-ATE;
But, I have read enough from those that HAVE flown these types to know that I'm glad I never had to.
Reading limited commentary here and being "glad I never had to. [fly the Airbus]" are personal statement of opinion which do not relate to the accident and do not testify to anything regarding an Airbus aircraft.
It is utterly amazing to read the differences expressed by those that fly these things as to how various systems operate. There are disagreements everywhere. The manuals are simply too full of information that not even a computer scientist could possibly remember/recall in the time of need. Who ever heard of an aircraft having THREE different flight control laws?!
Well DC-ATE that's the very same as what the king said to Mozart regarding his latest opera, "There's just too many notes!". In DC8 terms, it's like say, "there's far too much piping and too many pumps for my liking".

But in the larger issue, citing disagreements over discussions between pilots here on fbw systems as "cases in point" is simply a personal opinion which advances nothing, proves nothing and teaches nothing. Rather, the views express a prejudged opinion solidified without the benefit of knowledge or experience.

Anyone who says "the computer" is flying the airplane against the pilot's wishes/demands/requirements has never flown the aircraft. I've seen a lot of screw-ups on the 'bus and they are generally the result of a lack in some of the basic requirements like good training, experience and attitude - not always, but almost always.

Whatever happend to AF447, apparently happend so fast, the crew had no chance to sort out all the warnings being displayed on their panel. They probably had no chance to simply FLY their aircraft.
States the obvious. I'm sure you've seen more than one toppled gyro, iced-over windshield, unreliable speed indicator, difficult-to-read altimeter and a cockpit tossing so badly you can't read a thing in your day and I'm sure you know colleagues who had superb hands who flew a perfectly serviceable aircraft and it's passengers into a mountain or the ground. Did they have any more of a chance? If these guys lost everything for whatever reason (and we do NOT know those reasons or that they lost everything), and were being tossed like a cork in white water, what chance do any "hands" have?

Like it or not, his (Desertia) reference to having a "...piece of wire or hydraulics controlling..." says alot, along with those of us, yourself included, who learned how to "fly" airplanes and not manipulate a computer to do it for us.
Well, I don't mean to be unkind, but it really says nothing. The Fleet 80 I learned on is in principle no different than the A340. If one understands one's airplane and practises to learn more, one can fly the 340 like any aircraft including Boeing's 787.

My overriding point is, as always, one respected the limits of the machine one was in no matter when one flew, then or now. Respect means one learns about one's craft and one's airplane.

From your writing I know you knew your machine and cared. The 320/340 series is no different, nor is the 777 nor will be the 787 nor are those who fly present fbw aircraft.

An accident such as we now have does not negate over two decades of experience with the type, no more than the Amsterdam accident negates the brilliant 3 (almost 4)-decade experience with the 737.

Changing your mind isn't the goal. One goal is challenging ignorant comments about automatic flight. We are all free to have a say of course but that doesn't make what is said, right. Nor does that mean that autoflight is "innocent" by any means, but it takes some time to know the traps and corners and experience to discuss them legitimately.

The deeper goal is to recapture what has been lost and is becoming more elusive as "internet knowledge" makes "laptop experts" of more and more people, who cruise, scan, quote and freely shove an oar into a complex, specialist discussion without understanding; I am looking to recapture an abiding respect for those who actually have done, or still do the work on a professional, daily basis and who offer their time and expertise freely to help others who are genuinely keen to learn about these aircraft and what it really means to investigate such a tragic fatal accident.

best,
PJ2

PJ2

16th Jun 2009, 03:21

Will;

You are asking a fundamental design-philosophy which took if I recall ET's story-telling took many years for Airbus to develop. It's available in a paper he wrote (if I recall), entitled "Why we designed them like we did".

What you are essentially asking is, "why isn't there an autopilot system which can fly an airliner in every possibly-conceivable flight regime?"

The A330 autopilot disconnects at 45deg bank angle, 25deg pitch up, 13deg pitch down. The autoflight system reverts to alternate law (and cannot be recovered to normal law) in bank angles greater than 125deg.

In any airplane, it is unreasonable for a designer to not expect that a trained flight crew can take over the aircraft with sufficient competency to fly/land safely. If that is not the case, we are clearly into a very different zone of autoflight development which has yet to be fleshed out.

I don't think you understand fbw very well.

Bullethead

16th Jun 2009, 04:33

Some interesting comment gents.

A330 Accident: Pitot Tubes Not to Blame?: AINonline (http://www.ainonline.com/news/single-news-page/article/a330-accident-pitot-tubes-not-to-blame/)

Regards,
BH.

ClippedCub

16th Jun 2009, 04:45

From article
but the problem was with water on takeoff and landing, not a problem faced by aircraft flying at 25,000 feet.

Probably meant 35k feet which is normally above icing altitudes. So they're saying the problem is with water ingestion into the pitot, or the water forming ice. Does ingesting ice crystals, turned into water from the heat, count as water ingestion.

Me Myself

16th Jun 2009, 07:05

An AoA indicator is an available option for all Airbus FBW aircraft. I'm not aware of an airline that was willing to pay for it.

ELAC

Part of the Air France A320 fleet is equiped with it. These aircrafts were part of the Air Inter fleet that was merged to AF 10 years ago.

They also had head up displays which were removed after a while although they were very handy on CAT3 approaches.

This story is taking a nasty turn for AF in France.
Here's a link to Radio France national radio where a comedian name Stéphane Guillon, has a daily editorial and this what he had to say today.
France Inter - Vidéo (http://sites.radiofrance.fr/franceinter/video/humour.php)
No very good considering the large morning audience this guy has.

RWA

16th Jun 2009, 07:47

ClippedCub, in the SB for the A320s, the problem was said to be 'water ingress.' Apparently the problem was that the drainholes on that particular design of tube were getting clogged up and failing to let the water drain out. Whether the water then froze, I don't know; but presumably standing water would have a similar effect anyway.

JuggleDan

16th Jun 2009, 07:49

ClippedCub,

In modern fighters, the weapons systems are very complex. The flight control/management computers are trained to only bug the pilot with sequenced, pilot action only, messages during combat engagement, since he's busy targeting, etc.. [SNIP] Would have thought some of that logic structure would be present in civilian airliners.

When you increase "software intelligence", you exponentially increase incorrect specification risks, code complexity, testing costs... If you want a cheap, testable, reliable software, better make it as simple as feasible. You can go for more complex softwares only if you have the extra-money... and are willing to take the extra-risks that come with them...

From a quick glance at the RISK database (http://catless.ncl.ac.uk/Risks/index.1.html), there may be rather more fighter software bugs that pass undetetected. I guess the most famous example was the F-22A software bug (http://www.defenseindustrydaily.com/f22-squadron-shot-down-by-the-international-date-line-03087/):
At the international date line, whoops, all systems dumped and when I say all systems, I mean all systems, their navigation, part of their communications, their fuel systems.

I doubt that such an error could have made it undetected up to real-life flights on a civil plane...

Squawk7777

16th Jun 2009, 09:25

They are there to allow you to eat your breakfast or read the paper or look out the window (!).

They are NOT a substitute for that stuff you learned at your flight school!

I see young FOs these days that can barely fly the a/c.

The real worry tho' is that they think they are completely 'bulletproof'.

You took the words out of my mouth. The way flight training is headed (especially in Europe) I don't see much improvement. At least the FAA (still) emphasises on stick-and-rudder skills.

BOAC

16th Jun 2009, 09:58

Deep breath time again. Can I request a summary from 'the knowledgeable'?

1) Do we have ANY evidence to rule out some sort of mechanical disruptive event or fire around the cockpit/electronics bays?

2) Is it established that we have no ACARS message recording LOSS of cabin pressure AND is it confirmed that the ACARS message we have refers to excessive cabin DESCENT rate?

3) There have been several (I think unanswered) requests here for positions of the groups of bodies found. Do we have any, and has anyone back-plotted the position of the second recovery vis a vis ocean currents to see where they would have been in relation to the first, and likewise for the first to the POSSIBLE crash site?

4) All the ACARS messages COULD have related to software interpretations of events. Amongst them, is there any hard evidence of ACTUAL failures?

Apologies if I have missed a post or two - usual excuses.

DozyWannabe

16th Jun 2009, 10:04

When you increase "software intelligence", you exponentially increase incorrect specification risks, code complexity, testing costs...

For those of a more mechanical bent, a reasonable analogy would be the DC-10 cargo door situation in the 1970s. The system as specified around a hydraulic mechanism was fairly simple and the failure mode was non-catastrophic. Customer specified a change to electrical actuators to save weight, but this increased the complexity of the design significantly, and had the unintended knock-on effect of making the failure mode of the design much less safe.

Dave Gittins

16th Jun 2009, 10:21

The '10's cargo doors certainly didn't fail non catastrophically (leading to - I think - 2 hull losses (1 near Paris)) when they completely departed the aeroplane but the issue was due mainly to the failures of the over centre latching mechanism and no positive indication when the latches were secure / insecure not the power of the actuator. What they did reveal was some serious weakness in the cabin floor when, following an explosive decompression of the cargo hold, air couldn't vent fast enough from the passenger cabin and the cabin floor couldn't stand the pressure differential. The elevator and engine control runs in the floor were suddenly yanked which led to loss of control and severe unrecoverable pitch up (who says cables are better than wires ?).

Something of the swiss cheese effect that took a lot to sort out - accept that there was a problem - and subsequently fix. If such a compehensive fault is found with AF447, I hope the industry accepts what it is and resolves it quicker than 34 years ago.

Personally I doubt very much it will be the case and these swiss cheese holes will be far more obscure and harder to find.

Interflug

16th Jun 2009, 11:00

Re several posts concerning the pilot making inadvertent inputs on the sidestick in turbulence:

My instructor in Toulouse urged us to treat the sidestick as you would a dog s***, ie, only touch it if you have to, as theoretically the fbw will hold its current trajectory. But easier said than done, most especially in AF's predicament....Not everybody was in Toulouse for instruction I guess.
Or this particular pilot LOVES dog s***,
Or in theory the instructor is right, but... :uhoh:

LRilx-a3Hro

SaturnV

16th Jun 2009, 11:56

BOAC, with respect to your paragraph #3, the better indicator would be to learn the coordinates for the search grid for the pingers. One would expect that the French Navy would first target the most likely area, based on the extrapolated flight path, and the plot of the recovered wreckage and remains as matched against currents and surface winds over time. If repeated sweeps over an area produce nothing, the grid would be expanded.

I have not read if they have deployed side scan sonar to search for wreckage on the sea floor. Knowing the grids for that would also be a clue into where the French Navy thinks major portions of the wreckage are located.

As the pingers are about half way through their battery life, time is becoming of the essence.

In the instance of Air India 182, the recorders were recovered from 2,000 meters and read within about three weeks of the crash. Air India's wreckage was scattered for about five NM along the route of flight axis.

BOAC

16th Jun 2009, 12:13

based on the extrapolated flight path, - the problem is that there is NO WAY of knowing what the flight path was. We know the planned route, but not whether it diverted around cells, turned back or what. I would still think there is merit in oceanographic back-plotting from located debris? I would suggest that with all the troughs and peaks around there, the side-scan would be pushed to locate anything. It is pingers or nothing, I fear.

Rananim

16th Jun 2009, 12:52

I think the thread may be doomed unless someone like bsieker can re-focus the discussion with analysis of ACARS messages.Its all we've got to go on.What do the messages mean individually?Do they collectively indicate strictly an air data event and not an IR event?Air Caraibe/Qantas valid parallels or not?Presumably the ACARS messages indicate what ECAM messages the crew faced?Which ECAM messages were seen in Air Caraibe and Qantas?Commonality to AF447?ISIS fault interpretation?Air data only or possibly gyro-related,ie did that crew have attitude data?Robustness of RLG in severe turbulence?Power supply of ISIS?ESS bus or HOT BAT?Rudder travel limit controlled by airspeed but what if CAS was erroneous.Could RTL theoretically give FULL deflection at cruise altitude?Sensitivity of A330 rudder control system?Side-stick issues in severe turbulence..inadvertent input,conflicting dual pilot input.

DozyWannabe

16th Jun 2009, 13:09

But with the ACARS messages themselves being subject to scrutiny and the picture they paint being incomplete, maybe it's best we keep our powder dry until we have more evidence - otherwise we'll keep going round in circles.

ClippedCub

16th Jun 2009, 13:11

For the purposes of his comments, depends whether those melted ice crystals then re-freeze in a non-heated area of the pitot system and cause a blockage further in.

The ice crystals won't turn the corner, they'd just impinge on the back internally, would think that'd be heated. Also, the pitot doesn't flow, so 'cooling air' isn't flowing through the plumbing from the pitot.

Lemurian

16th Jun 2009, 13:13

BOAC :
1) Do we have ANY evidence to rule out some sort of mechanical disruptive event or fire around the cockpit/electronics bays?
No trace of fire in the debris , so far... and we're arguing about inflight break up or ocean surface impact. We don't know enough.
On this subject, there is the Crew Rest Container : two of its parts have been found on separate occasions ; one with the labelling, the other with that arrow pointing FWD. I personally have tried to find out whether they were part of the same debris trail or on very far apart locations. Guess we will eventually be told.
Another mystery is the left #1 spoiler, found by a tanker. Was it surrounded by other debris or was it all by itself away from everything else ?

2) Is it established that we have no ACARS message recording LOSS of cabin pressure AND is it confirmed that the ACARS message we have refers to excessive cabin DESCENT rate?
Yes, apparently. There is an advisory linked to a "cabin vertical speed", related to the ECAM presentation... It indicates a cabin vertcal speed in excess of 1800 ft/min up or down, translated on the ECAM screen as a very small green-pulsing bar.
A depressurisation would have been indicated by an "EXCESS CABIN ALT" warning, telling of a cabin above 9500 ft....That warning is not part of the received ACARS messages.

3) There have been several (I think unanswered) requests here for positions of the groups of bodies found. Do we have any, and has anyone back-plotted the position of the second recovery vis a vis ocean currents to see where they would have been in relation to the first, and likewise for the first to the POSSIBLE crash site?
I, for one, would be interested in that kind of info... See above.

4) All the ACARS messages COULD have related to software interpretations of events. Amongst them, is there any hard evidence of ACTUAL failures?
In this case it's at a very low level of interpretation. The messages are akin to two kinds of reporting :
1/-Hi ! I'm system X. I've diagnosed that I'm too sick to keep on working. Something is wrong with me and I've informed the crew through a warning or an advisory message on their ECAM.
2/-Hello ! This is X and Y. We've determined that Z is going nuts and we don't trust him anymore and we've put him in a straight jacket. We'd like the crew to look into the matter...
We seem to have here a convergence of related failures, detected by the computers which need air data inputs to function properly. In all fairness, that's about the extent of our knowledge.
On the other hand, we've already seen similar events, with very similar messages, in similar orders pointing a finger at the ADRs total pressure channels...

Graybeard

16th Jun 2009, 13:18

From what I read here, way back when, the pitot problem is one of poor quality control of the bleed hole. I have not read why they could not be reworked. Does anybody know the price of the new tubes? AF had them sheduled for the next C check, when a full pitot/static check would be performed anyhow.

I also read somewhere in all this that AF was not convinced the new pitot tubes were really an improvement.

GB

swish266

16th Jun 2009, 13:44

On AF SOP
Was the Captain in the cockpit at the time or was he taking his rest? Who was in his seat - the guy with 800h or the guy with 2600? Not that it matters so much, but in an extremely stressful sit, it will won't it?(post #383 - Air France A330-200 missing thread from Tech Log)

Pardon me for repeating myself, but I would like to put this question again, after 2 weeks to our grieving AF colleagues - what are the AF In-flight Relief procedures - this can give us a lot of easy food for thought...
Merci BCP

Lost in Saigon

16th Jun 2009, 13:56

Deep breath time again. Can I request a summary from 'the knowledgeable'?

2) Is it established that we have no ACARS message recording LOSS of cabin pressure AND is it confirmed that the ACARS message we have refers to excessive cabin DESCENT rate?

I don't think it has been established that it was a DESCENT rate. I believe this is simply an advisory of a rate of change either up or down that exceeds a set parameter. It has been said that this parameter is 1800 fpm.

Some are suggesting that due to de-pressurization the cabin vertical rate showed an excessive rate of climb, and others are suggesting a possible excessive rate of descent as the aircraft descended to the sea. Either scenario is possible.

It appears that some people are confused by the fact that the cabin pressure PSI will decrease as the cabin climbs or that the cabin pressure will increase as the cabin descends.

In reality the message is only about a simple rate of change in cabin altitude (vertical rate) and at this point it tells us very little.

The cabin altitude could be changing due pressurization at altitude (resulting in a climbing cabin) or it could also mean the cabin was descending below the normal cabin altitude of about 8000' when at cruise. (If the aircraft was in any sort of a dive, the cabin would at some point have to descend below 8,000 feet)

ClippedCub

16th Jun 2009, 13:57

Seeing the motion available in the joystick video and knowing that;

- fighter joysticks have limited forward travel due to inadvertant correction under negative g.

- Civil transport manufacturers wouldn't test to the neg 1 g limit outside of the simulator and that wouldn't simulate the pilots arm coming off the rest.

- sitting in your chair with arm on armrest, if you pivot your forearm upward, raise the elbow but keep the hand in place, the hand, (handle), pivots forward.

Then perhaps introducting even more negative g. Who knows. Depends on the ergonomics.

Anybody have access to a full motion A330 sym that wants to check this out. Set moderate/severe turbulence at high alt cruise. Grip sidestick controller in manner an experienced pilot would under such conditions with the autopilot off and in alternate law. Lift yourself up against the lap belt, raising the elbow an inch or two. Anything happen. In the interest of safety, worth a quick check at least.

Or does nomal everday flight in turbulence rule out this concern, i.e., already been demonstrated in the real world.

RWA

16th Jun 2009, 14:05

"I also read somewhere in all this that AF was not convinced the new pitot tubes were really an improvement."

Other way round, according to what I've read, Graybeard. Airbus said that there was no proven need to replace the pitot tubes, Air France (an airline for which I personally have enormous respect) reckoned different and replaced them anyway.

A bit of what I personally consider to be 'good news.' My daughter is a prosperous international lawyer living in Hong Kong. She often makes short visits home and I've quite often 'advised' her to choose Boeings, rather than A330s/A340s, whenever possible. Without going into too much technical detail.

She just emailed me this arvo to say, "Thanks, Dad. Reading about the Air France thing, guess I'll try to follow your advice from now on........."

Anyone on here think that a basically-thick, but luckily 'cunning,' guy like me was over-reacting? Or that my (highly-intelligent, 'to a fault', honours-law-degree and Masters' ticket) daughter, who knows nothing at ALL about aviation, is over-reacting too?

Five Livers

16th Jun 2009, 14:35

No it doesn't - it needs some answers!

Who was in the the flight deck?

Why did they did get into the weather?

With all the ACARs messages, why hasn't a replay of the the last few minutes of flight been reproduced?

Why has there been a deafening silence from Airbus?

Mercenary Pilot

16th Jun 2009, 14:46

This isnt an "Armchair crash vulture and enthusiasts" Forum. When BEA decide to release those details, then maybe it is worth further discussion but until then this thread has run its course especially by such purile posters as BigFootDriver and Barbies boyfriend.

Why has there been a deafening silence from Airbus? Probably because they are awaiting details and facts from the investigation team. They are not going to respond to flightsim wonder kids, spotters and conspiracy theorists who masquerade on this website like some sort of experts.

BOAC

16th Jun 2009, 14:59

2) Is it established that we have no ACARS message recording LOSS of cabin pressure AND is it confirmed that the ACARS message we have refers to excessive cabin DESCENT rate?
Yes, apparently. There is an advisory linked to a "cabin vertical speed", related to the ECAM presentation... It indicates a cabin vertical speed in excess of 1800 ft/min up or down, translated on the ECAM screen as a very small green-pulsing bar.
A depressurisation would have been indicated by an "EXCESS CABIN ALT" warning, telling of a cabin above 9500 ft....That warning is not part of the received ACARS messages.- picking up on just this one part of your reply - if only we could be sure we have ALL the ACARS messages in sight, then there is a very good chance we can rule out a high-level break-up in view of the lack of an "EXCESS CABIN ALT" warning as you say. Without such a signalled disruption to the cabin pressure, the only subsequent ACARS message must surely refer to a descent? Dare we surmise? PJ? Bsieker?

Lost in Saigon

16th Jun 2009, 14:59

No it doesn't - it needs some answers!

1) Who was in the the flight deck?

2) Why did they did get into the weather?

3) With all the ACARs messages, why hasn't a replay of the the last few minutes of flight been reproduced?

4) Why has there been a deafening silence from Airbus?

Your questions are answerable. I think they border on being ridiculous.

1) How do you propose to find out who was in the flight deck with the limited information now available?

2) There is no proof they went "into the weather". Only speculation. They may have diverted many miles around the nasty weather.

3) The ACARS messages can not replay anything. They are only very simple automated notices designed to give a "heads up" to maintenance.

4) Considering the lack of any concrete information at this point, what do you want them to say?

john clements

16th Jun 2009, 15:07

tomorrow at 1000-1100 bea have a press conference.........

einhverfr

16th Jun 2009, 15:10

1)You've heard of sublimation but at your low elevations probably haven't witnessed it. It's a common process in Colorado and I would expect melting ice on the exterior of pitot's to sublimate. An unknown question becomes whether heated ice crystals within the pitot system sublimates or melts into water.

I have actually been looking into this after reading of a number of pitot icing issues where I thought supercooled liquid water could be ruled out. Meteorological survey reports were a major source of information in that research.

I have concluded that there must be a secondary process by which ice forms in pitot tubes at high altitudes where there is a very low temperature and a high density of ice crystals. My guess is that sublimation is involved in this process, but I can't be sure. I can see two hypotheses for this icing:

The first is that the ice crystals deposit but don't form a hard solid mass. Since the ice will have greater mass than the air it displaces, it will also have more inertia, and thus a partial obstruction of the pitot tube with loose ice crystals would dampen pressure changes (venturi effect). The problem with this hypothesis in my view is that the small ice crystals are small enough to be dispersed more evenly and even occasionaly ejected by pressure changes which occur in turbulance. Thus I have trouble imagining this process causing problems over ten or twenty minutes.

The second hypothesis is that the ice crystals, once deposited, may sublimate and condence, fusing them together into a hard mass (similar to the way wind-crust develops on ski slopes). An obstruction of this sort would also reduce pressure via the venturi effect, but would be less prone to being ejected from the tube by changes in pressure.

There have been enough reports of pitot tube icing in met. surveys of MCS's outside of strong convection (in one of the main cases I came across, in a DC-8 ;-) ) that I think we can understand that there is a process that the pitot tubes can ice up in the absence of liquid water. What that process is exactly may be harder to verify.

The absence of supercooled liquid water does not mean that pitot tubes were unlikely to be iced up in AF447. The parallels I have been able to find suggest that the pitot tubes would have almost certainly iced up.

Mike-Bracknell

16th Jun 2009, 15:22

I have a question:

When you get pitot icing, how quickly does it occur? i.e. is the state of the pitot from reliable to unreliable a binary one? or does it show some degradation of service which allows you to consider icing as an issue?

Also, does pitot heating always work or is it sometimes overwhelmed by the icing conditions?

Lemurian

16th Jun 2009, 15:28

BOAC,
if only we could be sure we have ALL the ACARS messages in sight,
There is absolutely NO way of confirming that, apart from the declarations from the BEA and AF.
Secondly, these messages concern ONLY the ACARS sent to the maintenance department.
Missing are all those transmitted to OPS /CDG station / GIG plus all the routine queries for en-route alternate weather, routine messages to base, the automated initiation message, flight plan, loadsheet, the OUT and OFF....
then there is a very good chance we can rule out a high-level break-up in view of the lack of an "EXCESS CABIN ALT" warning as you say.
I don't think so. Look, these messages are just the picture until 0214Z. What happened next is pure speculation. Is the last message the very last transmitted ? Was there a power failure after the last transmission.?...Conjectures...

BOAC

16th Jun 2009, 15:40

Look, these messages are just the picture until 0214Z. What happened next is pure speculation. - maybe I'm not expressing myself clearly. I'm looking UP TO 0214Z, not 'next'. The point I am making is that there MAY be a case for saying the cabin was INTACT until the descent warning, which would cast doubt on the "the fin fell off and tore open the fuselage" line. In fact, how long would it take for the cabin to achieve 1800fpm? Could they even have been descending before the 0210Z message stream we have?

Me Myself

16th Jun 2009, 15:41

No it doesn't - it needs some answers!

Who was in the the flight deck?

Why did they did get into the weather?

No one knows !! but if the captain was in his seat, then another F/O would have been in the right hand one.
If the captain was on his break, then, the most experienced F/O would have been in the right hand seat. That's AF SOP for long haul crew relief so that the one in charge of let's say an emergency descent does it from the seat he is used to.
To tell you the truth, given the catastrophic system meltdowns, I can't see myself being able to fly the aircraft in a such a weather and no attitude indicator.

Why ???? When you answer that one give me a call.

DC-ATE

16th Jun 2009, 16:13

Dysag -

Curious; are you saying Mr. Larson's Report is false? Seems fairly credible to me. Afterall, this a/c was in service for how long, and how many cycles?

cribbagepeg

16th Jun 2009, 16:16

sublimation is the direct passage between the solid state to the gaseous state, without spending time in the liquid state. This is the normal method by which dry ice dissipates, and is common in many situations involving water.

A much earlier message from someone stated that energy input is needed to convert water to ice. In fact, the opposite is true - you remove energy to convert water to ice. In the reverse case, there is a parameter called the energy of fusion, easily demonstrated by applying heat to a mixture of water and ice. The temp remains right at the freezing point, no matter how much heat you apply,, until the ice is all melted, at which point the temp begins to rise.

wikipedia should have more appropriate definitions...

PJ2

16th Jun 2009, 16:21

Jetdoc;
I think what you have got there is the AMM references for the maintenance of those components, not fault codes.
Thanks - its very possible - the document is ancient and very difficult to navigate within - finding fault-failure codes is difficult - also I'm unaccustomed to using the document. The point of course is to convey complexity - these may not be the fault codes so that argument appears compromised somewhat until one understands that this applies to all systems including fault/failure analysis; I strongly suspect there are "more numbers" which may or may not be available and which will further determine which sensors were providing "bad" data. I appreciate the comment, and hope others pick it up as well.

TerraHertz

16th Jun 2009, 16:43

Being curious about the source of the two pages of ACARS messages, which are labelled 28/256 and 29/256, I looked around to see where they originated. Amazingly, this seems to be it:
AF447 accident - icing, pitot tubes and radar in the frame - Unusual Attitude (http://www.flightglobal.com/blogs/unusual-attitude/2009/06/af447-accident---icing-pitot-t.html)
So the crucial text on which everyone is basing analysis, was hand copied from a screen visible during a TV interview! Furthermore I gather from other comments here, that this is just one ACARS stream (maintenance messages?) of several. Also that it isn't publicly known for certain that these two pages are even all the relevent messages from that stream.

I'm new to this forum, and have no aviation-related expertise whatsoever. (I'm an electronics and computer engineer.) I greatly respect the professional experience present on this forum and others. But... I'm astonished that there doesn't seem to be any indignation that Air France and the French aviation authorities have not seen fit to release *all* the ACARS messages from AF447. Are there legal reasons why they wouldn't? Liability? Insurance? Given that what appear to be at least a large proportion of the critical ACARS messages are now public, why not either officially release them all, or at least make some effective demonstration that there were no more?

If this topic has already been dealt with, my apology. 87 pages...

For reference, here are some other links to the ACARS messages and commentary:
http://www.pprune.org/4975386-post45.html single post - ACARS msgs
http://img197.imageshack.us/img197/3410/acarsaf447e.png ACARS msgs

Engineer decodes Air France Flight 447 emergency messages (http://blog.seattlepi.com/aerospace/archives/170669.asp)
20090608 Engineer decodes Air France Flight 447 emergency messages

http://www.iag-inc.com/premium/acars2.pdf
Line by line commenting on the ACARS msgs.

http://www.nytimes.com/2009/06/13/world/13plane.html?_r=1
Error Messages From Air France Jet Offer Details but Little Insight

The New York Times > World > Image > The Last Minutes of Flight 447 (http://www.nytimes.com/imagepages/2009/06/13/world/13plane.grafic.ready.html)
http://graphics8.nytimes.com/images/2009/06/13/world/13plane-graf01.jpg

http://www.eurocockpit.com/images/acars447.png

PJ2

16th Jun 2009, 16:46

JuggleDan;
When you increase "software intelligence", you exponentially increase incorrect specification risks, code complexity, testing costs... If you want a cheap, testable, reliable software, better make it as simple as feasible. You can go for more complex softwares only if you have the extra-money... and are willing to take the extra-risks that come with them...
Superb post - said far better than the point I was trying to make about expense in design. Unfortunately, your wisdom and experience as an engineer commenting on this thread will be ignored so ignorance can prevail. (I haven't had a good laugh in a while until I read some stupid comment about the software not letting the pilot fly the airplane until the ECAM is complete. Someone's stuck in childhood understandings of the world where mom says no desert until you finish your vegetables...

Quite unbelievable. ELAC's post a few pages ago is spot on. The discussion is agenda-driven and some other venue will have to be found for serious, knowledgeable enquiry.

Thanks again for your contribution.

PJ2

FE Hoppy

16th Jun 2009, 16:48

EASA Safety Information Bulletin
SIB No.: 2009 - 17
Issued: 09 June 2009
Subject: Unreliable Airspeed indication
Background: During the recent accident of an A330 into the Atlantic Ocean
on 01 June 2009, and without prejudging the final outcome of
the investigation, a discrepancy between the different
measured airspeeds was reported. There have been a number
of occurrences of unreliable airspeed indications or misleading
air data information. The root cause of this may be due to, but
is not limited to, inappropriate maintenance, contamination by
small objects or materials on the ground or in the air, extreme
environmental conditions producing icing outside the
certification envelope of the probes or large amount of water
ingestion.
Description: The primary purpose of the pitot-static system is to provide the
flight crew with airspeed information, required to safely control
the aircraft. As noted above, the origins of potential pitot-static
system malfunctions are numerous and cannot be totally
excluded in the operational context. The Aircraft Flight
Manuals and/or Flight Crew Operating Manuals include
procedures for unreliable airspeed indication (Air data system
misleading information) and these should be well known by
flight crews. Correct application of these procedures by flight
crews may be crucial for assuring the safety of the aircraft
when such Pitot-static malfunctions occur.
Recommendations: Operators should ensure that flight crews have proper
knowledge and proficiency:
- To detect and to identify unreliable airspeed indication.
- To apply immediate and conservative actions for ensuring
short term safe flight control, in accordance with the
manufacturer procedures developed for the specific aircraft;
the use of memory items should be considered.
- To apply procedures for the safe continuation of a flight with
unreliable airspeed indication up to a safe landing.
EASA SIB No: 2009 - 17
This is information only. Recommendations are not mandatory.
EASA Form 117 Page 2/2
Familiarisation of flight crews with unreliable airspeed
indication procedures should be provided through adequate
training. Flight crew knowledge and proficiency should be
checked on a regular basis.
Applicability: All aeroplanes operating in commercial air transport.
Contact: For further information contact the Airworthiness Directives,
Safety Management & Research Section, Certification
Directorate, EASA. E-mail: [email protected].

I think EASA know what happened don't you?

HarryMann

16th Jun 2009, 17:16

The A320 elevators and rudder hinges whose actuators had been
removed shattered and the rudder and elevators came off.

That alone negates any other of his opinions as meaningless...
+ the rudder from F-GZCP seems still fixed to the fin itself.

ClippedCub

16th Jun 2009, 17:30

When you increase "software intelligence", you exponentially increase incorrect specification risks, code complexity, testing costs... If you want a cheap, testable, reliable software, better make it as simple as feasible. You can go for more complex softwares only if you have the extra-money... and are willing to take the extra-risks that come with them...

Naturally.

There'll have to be a demonstrated need, i.e., accident reports, pilot concerns, claiming there becomes too much for the crew to handle under a certain set of conditions. If true, then that'll become reflected in requirements to fix it. You can want a cheap airplane, but the requirements will tell you how cheap it can be.

Pontius Navigator

16th Jun 2009, 17:37

First, I had a Pprune database error this message was posted at 1840 GMT+1hr.

I think EASA know what happened don't you?

No.

Recommendations: Operators should ensure that flight crews have proper knowledge and proficiency:
- To detect and to identify unreliable airspeed indication.
- To apply immediate and conservative actions for ensuring
short term safe flight control, in accordance with the
manufacturer procedures developed for the specific aircraft;
the use of memory items should be considered.
- To apply procedures for the safe continuation of a flight with
unreliable airspeed indication up to a safe landing.

I think you might conclude that the air data system might have been a cause. They also say that flight crews (any flight crew) needs to have proper knowledge and proficiency in circumstances where there are air data errors.

You cannot infer that the crew on AF447 were not knowledgable or proficient nor that the air data system cause the crash. You might tentatively conclude that one or both were contributory but nothing else.

lomapaseo

16th Jun 2009, 17:49

EASA Safety Information Bulletin
SIB No.: 2009 - 17
Issued: 09 June 2009
Subject: Unreliable Airspeed indication
Background: During the recent accident of an A330 into the Atlantic Ocean
on 01 June 2009, and without prejudging the final outcome of
the investigation, a discrepancy between the different
measured airspeeds was reported. There have been a number
of occurrences of unreliable airspeed indications or misleading
air data information. The root cause of this may be due to, but
is not limited to, inappropriate maintenance, contamination by
small objects or materials on the ground or in the air, extreme
environmental conditions producing icing outside the
certification envelope of the probes or large amount of water
ingestion.
Description: The primary purpose of the pitot-static system is to provide the
flight crew with airspeed information, required to safely control
the aircraft. As noted above, the origins of potential pitot-static
system malfunctions are numerous and cannot be totally
excluded in the operational context. The Aircraft Flight
Manuals and/or Flight Crew Operating Manuals include
procedures for unreliable airspeed indication (Air data system
misleading information) and these should be well known by
flight crews. Correct application of these procedures by flight
crews may be crucial for assuring the safety of the aircraft
when such Pitot-static malfunctions occur.
Recommendations: Operators should ensure that flight crews have proper
knowledge and proficiency:
- To detect and to identify unreliable airspeed indication.
- To apply immediate and conservative actions for ensuring
short term safe flight control, in accordance with the
manufacturer procedures developed for the specific aircraft;
the use of memory items should be considered.
- To apply procedures for the safe continuation of a flight with
unreliable airspeed indication up to a safe landing.
EASA SIB No: 2009 - 17
This is information only. Recommendations are not mandatory.
EASA Form 117 Page 2/2
Familiarisation of flight crews with unreliable airspeed
indication procedures should be provided through adequate
training. Flight crew knowledge and proficiency should be
checked on a regular basis.
Applicability: All aeroplanes operating in commercial air transport.
Contact: For further information contact the Airworthiness Directives,
Safety Management & Research Section, Certification
Directorate, EASA. E-mail: .

I think EASA know what happened don't you?

Yes. This was known from the earliest by the investigation. What is puzzling is why did it happen?

Now we are at the stage what are we going to do about it.?

On the surface it seems like the pitot tube issue is only a "minor malfunction". However I have to ask is it a conditional malfunction in that it is lilely to be combined with other serious workloads on the crew?

If so we either have to minimize the heck out the pitot tube combinations or train the pilots for this malfunction in turbulence. (I won't use the word "severe" since we still don't know)

DC-ATE

16th Jun 2009, 18:22

HarryMann -
That alone negates any other of his opinions as meaningless...
+ the rudder from F-GZCP seems still fixed to the fin itself.

Ah.....if you'll read his report, he was NOT writing about the AF 447 tail assembly.

DespairingTraveller

16th Jun 2009, 18:35

RWA:

I've quite often 'advised' her to choose Boeings, rather than A330s/A340s, whenever possible. Without going into too much technical detail.

She just emailed me this arvo to say, "Thanks, Dad. Reading about the Air France thing, guess I'll try to follow your advice from now on........."

Anyone on here think that a basically-thick, but luckily 'cunning,' guy like me was over-reacting? Or that my (highly-intelligent, 'to a fault', honours-law-degree and Masters' ticket) daughter, who knows nothing at ALL about aviation, is over-reacting too?

I'm just PPL/SLF, with a similar professional background to your daughter, but I've no doubt you're both over-reacting, and that there are many, many aviation professionals on this site who could tell you the same.

But really you should get it from the horse's mouth: Boeing themselves. Have a look at their 2008 Statistical Summary of Commercial Jet Airplane Accidents (http://www.boeing.com/news/techissues/pdf/statsum.pdf). Pay particular attention to page 19, the summary of hull loss rates by aircraft type. I see nothing on that page to favour Boeing over Airbus. A casual inspection very much suggests the opposite in fact, although the period of time covered by the data undoubtedly muddies the waters.

But whether or not one type could be demonstrated to be superior to another is largely irrelevant. The rate of fatal hull losses per million departures is so fantastically low (~1) that any underlying systematic difference between manufacturers is likely to be a small change in an already tiny risk. Anyone seriously worried by such a marginal change should give up air travel for good. Not because air travel itself is dangerous, but because anyone so sensitive to such tiny risks must surely find getting in the car for the trip to the airport to be an activity so foolishly risky that they would never countenance it... ;)

PEI_3721

16th Jun 2009, 18:36

Re pitot icing; the principles of pitot operation is that air does not flow through the tubing, the system captures a pressure. There are some types (probably used by Airbus) which allow a small calibrated leak, but downstream of the ‘leak’ the systems’ operation are essentially identical.
A change in pitot pressure (erratic IAS) could occur quite quickly if an ice/water accumulation reduces the effective diameter of the tube, particularly with the ‘leaky’ systems. Diagrams showing the types of pitot and a range of disruptive mechanisms are shown in www.sae.org/events/icing/presentations/2007s30duvivier.pdf
An indication of the atmospheric conditions, icing type, and rate of icing can be seen by comparing the events of unusual engine icing and TAT probe icing – http://airs-icing.org/AIRS_II/AIAAReno2006/AIAA-2006-206-739.pdf

Exnomad

16th Jun 2009, 18:42

Having worked for a company manufacturing heated pitots that was bought by Goodrich, I accept that pitot heaters can fail, but what are the chances of three failing in a short period, all three would have been working at takeoff, it seem to me to be extremely unlikely that three would fail at once, they would likely be from different manufacturing batches.

DorianB

16th Jun 2009, 18:51

ClippedCub

16th Jun 2009, 19:13

I've written some software specifications. Computer reboot time, just in debugging exceeds response time available during an incident, before it becomes an accident. Also, the pilot can't be trained in what to expect during computer problems if there is a human, besides the flight crew, in the mix. The computer geek would be a wild card that the crew shouldn't have to deal with. Plus, you'd have to add weight back in for himself and his provisions, in addition to the computer weight.

Besides, if you introduce the third crew member back into the mix, one could make the case to get rid of the heavy and complex computers. All you'd really want in that case is FADEC to trim the engines for best fuel flow.

swish266

16th Jun 2009, 19:15

If the captain was on his break, then, the most experienced F/O would have been in the right hand seat. That's AF SOP for long haul crew relief so that the one in charge of let's say an emergency descent does it from the seat he is used to.
To tell you the truth, given the catastrophic system meltdowns, I can't see myself being able to fly the aircraft in a such a weather and no attitude indicator.

I am not sure about this. What if AF nominates IFR? In the worst case scenario the 600h on type F/O would have been in the RHS and the 2600h one in the LHS... What if AF stipulates that only the RH seated F/O can be in command...
The cheese holes could have been very much aligned even before the systems started to fail...
I would appreciate if an AF colleague can enlighten us about what actually are the SOPs and how stringently are they applied
:confused:

chefrp

16th Jun 2009, 19:35

Dutch Bru

16th Jun 2009, 19:40

Hey, Interflug (post 1701), there is a bit of difference between stick input in a gusty landing at approach stages and when cruising through CB at 35 kf. I'm confident that Airbus IT caters for the different circumstances.

PJ2, with respect for your posts in general, your questions about the mobile crew rest area on the 330 has already been dealt with elsewhere in this thread. Post nr 773 and subsequently (perhaps deleted inputs) from colleagues to confirm that A330 mobile rest area come in the form of a LD6 container is normally located below deck on the hight of the 3L/R doors, aft of the wings. I thought you said somewhere you were current on the 330 ?

augustusjeremy

16th Jun 2009, 19:46

it seem to me to be extremely unlikely that three would fail at once, they would likely be from different manufacturing batches.

Actually it seems that the ACARS messages are indicating the failure of just one of the sensors.

That's what is probably still missing in the "clogged pitot tube as a cause" hypothesis.

Correcting:

If the message 341115 – PROBE – PITOT actually refers to all probes disagreeing 1+2/2+3/1+3 instead of only two of them in agreement then I stand corrected that all three probes failed.

ClippedCub

16th Jun 2009, 19:52

News - more wreckage sighted;

PRESS RELEASE 32

INFORMATION ON THE SEARCH FOR AIR FRANCE FLIGHT 447

The Brazilian Navy Command and Aeronautical Command inform that, in the search today, military personnel aboard aircraft sighted wreckage 950 kilometers from Fernando de Noronha, in an area close to where debris was found before. No bodies were sighted.

At this moment, more than a thousand Brazilian military are directly involved in the searches, 761 belong to the Navy and 250 to the Air Force.

The Frigate Bosísio is heading back to Fernando de Noronha with the six bodies found before by the French Navy and transferred to the Brazilian ship. The arrival estimate to the island is tomorrow, Tuesday, June 16th.

NAVAL SOCIAL COMMUNICATION CENTER
AERONAUTICAL SOCIAL COMMUNICATION CENTER

einhverfr

16th Jun 2009, 19:55

chrefrp:

I don't think it would make sense to put a flight engineer in the cockpit. The systems in question have extremely limited inputs. In reality they should be simpler to troubleshoot than even the most simplistic systems one can think of. Additionally most of the real troubleshooting can only take place on the ground anyway.

Situation awareness is something that may need some additional work when looking at self-healing systems (this isn't really an FBW issue) but that is a separate issue.

ELAC

16th Jun 2009, 19:56

I am not sure about this. What if AF nominates IFR? In the worst case scenario the 600h on type F/O would have been in the RHS and the 2600h one in the LHS... What if AF stipulates that only the RH seated F/O can be in command...
The cheese holes could have been very much aligned even before the systems started to fail...
I would appreciate if an AF colleague can enlighten us about what actually are the SOPs and how stringently are they applied

Swish,

I can't state exactly what AF's procedure is, but there's no carrier that I'm aware of that would permit the scenario you describe. In a heavy crew, aside from the PIC one of the other pilots will be trained and designated as a "Relief" or "Cruise" captain. This pilot may be either another qualified captain or a senior F/O trained to hold a cruise captain qualification. In some carriers all pilots holding the F/O position are trained to this standard with the augmenting crew members coming from a lower seniority group specifically tasked to fly as relief pilots.

In the AF case it would appear that it was a senior F/O who held a cruise captain type of qualification. In any event, it would be this pilot, not the 600 hr. F/O who would be in charge during the PIC's absence. Decision making authority in the cockpit is not handed to the least qualified crewmember simply on the basis of the seat he is occupying.

ELAC

And I might mention further that I know a good number of relief pilots who are in fact very well qualified and capable pilots with significant command experience. The position a pilot holds on the flight is not in itself an indication of his experience or abilities.

ELAC

16th Jun 2009, 20:07

PJ2, with respect for your posts in general, your questions about the mobile crew rest area on the 330 has already been dealt with elsewhere in this thread. Post nr 773 and subsequently (perhaps deleted inputs) from colleagues to confirm that A330 mobile rest area come in the form of a LD6 container is normally located below deck on the hight of the 3L/R doors, aft of the wings. I thought you said somewhere you were current on the 330 ?

Dutch Bru,

Not all A330s are equipped the same. The container you are referring to is an LDMCR (Lower Deck Mobile Crew Rest), which in most carriers is used as a cabin crew rest facility. Many carriers equip the A330 with a seperate pilot crew rest bunk, often located immediately behind the cockpit opposite the forward lavatory. Some carriers, however, have placed this bunk in other locations and to my knowledge there has been no confirmation of whether AF 447 had a bunk, and if so where it was located. Being current on one carrier's aircraft won't tell you how another carrier has chosen to configure such items.

ELAC

Dutch Bru

16th Jun 2009, 20:09

Its obvious that ACARS maintenance messages are not to be dealt with by the base station in real time, since the're aimed at informing maintenance with a view to future deployability of company aircraft in general.

However, ACARS is also used, mainly by long distance operators, to exchange the latest info on weather along long distance routes.

Failing of publishing this type of info for the AF447 flght keeps a cloud over issues relevant in this cae, period.

chefrp

16th Jun 2009, 20:11

Pointus Navigator

They could be on shifts in different time zones....i.e. like my bank, never know who or where I'm talking to.

And yes your right it would be boreing. I guess it wouldnt be someones only job, sitting there waiting for bad news. But it could be part of an intergrated network to help maintain safety.

It just seems like this might be an area where progress could made. Only because these are the only messages that we know of at this time.

thanks for the feed back

RP

Svarin

16th Jun 2009, 20:16

1 – FACT : a careful reading of the first 3 pages from the (probably leaked) ACA report (available at Eurocockpit - Accueil (http://www.eurocockpit.com), direct link is http://www.eurocockpit.com/docs/ACA.pdf). Everything is in these first 3 pages, although the rest is well worth reading too. In short, one of two incidents on A330s with loss/discrepancy in all airspeed indications.

2 - HYPOTHESIS : from the available sequence of ACARS messages, let us postulate that the pilots in AF447 faced the same situation as did the ACA flights mentioned above.

3 - FACT : in the last paragraph on page 3 of said ACA report, the ACA flight Pilot Flying deliberately ignored all STALL warnings, contrary to what is requested by TECHNICAL RECOMMENDATIONS from the UNRELIABLE SPEED INDICATION C/L.

4 - HYPOTHESIS : following the same C/L, facing the same situation caused by the same conditions, let us postulate that the AF447 pilots do indeed heed the STALL warnings, as requested by C/L. In so doing, while having a mental picture of trying to fly away from a low speed stall condition, they actually fly towards a high speed buffet and beyond.

The ACA incident lasted 1 min 26 sec. It is not unreasonable to assume an approximately 4 minutes duration for the AF flight, from 0210Z to 0214Z.
From the initial condition : loss of all 3 airspeed indications, the ACARS messages pertaining to flight controls and other instrument failures are aggravating circumstances, but they are not necessary to explain the scenario at point 4 above.

Another report that can be found on the internet, regarding one AF 908 CDG TNR, describes a very similar situation. Again, the pilots disregarded the STALL warnings.

I remain ignorant of the actual circumstances of the AF447 accident. The reasoning above is, by necessity, a conjecture. It is however founded on very simple hypotheses, making it a good candidate to confront a sharp Occam's razor.

I find it necessary to point out the fact in item 3 above, regarding STALL warnings and how the C/L deals with these.

Dutch Bru

16th Jun 2009, 20:19

Appreciate your views, but the AF447 photo material as discussed in previous posts confirm that the flight in question was carrying a mobile rest area below deck aft of the wing section

Lost in Saigon

16th Jun 2009, 20:22

ClippedCub....

Understood.....thats why I asked.

Is it possible to have support from the ground when these ACARS messages come streaming in? That would make more sense.

The messages were not looked at till later, after the fact. And I assume this is the first time they have been so scrutanized.

I guess my faith in modern technology has taken a hit, and I assume that there has to be a way in this day and age to make these systems even safer than they already are.....

In addition to the auto messages to maintenance, ACARS can be used to send and receive messages to and from Flight Dispatch, Maintenance, Crew Scheduling, etc. They can ask for help and receive it any time they want.

In this case there was probably no request by the pilots because there would not have been enough time.

Lost in Saigon

16th Jun 2009, 20:25

Dutch Bru,

Not all A330s are equipped the same. The container you are referring to is an LDMCR (Lower Deck Mobile Crew Rest), which in most carriers is used as a cabin crew rest facility. Many carriers equip the A330 with a seperate pilot crew rest bunk, often located immediately behind the cockpit opposite the forward lavatory. Some carriers, however, have placed this bunk in other locations and to my knowledge there has been no confirmation of whether AF 447 had a bunk, and if so where it was located. Being current on one carrier's aircraft won't tell you how another carrier has chosen to configure such items.

ELAC

Some A330 operators use a regular J class seat for crew rest with 3 pilots operations.

Appreciate your views, but the AF447 photo material as discussed in previous posts confirm that the flight in question was carrying a mobile rest area below deck aft of the wing section

That may have been for Flight Attendants only

chefrp

16th Jun 2009, 20:30

Lost in Saigon

thanks for the info.

I guess in certain situations there is not much you can do but try your best.

It's nice to know that the crew can recieve info from the ground.
Can they recieve weather reports/warnings as well, or do they rely on thier on-board radar for this?

Thanks......

Svarin

16th Jun 2009, 20:31

Gentlemen,

I have tried very hard to refrain from commenting on the 'who was seated where' issue. Too many uninformed comments and speculation casting doubts on ability or seriousness have been spread here.

Gentlemen, I know as HARD FACTS the following two points, directly relevant to the discussion about any AF flight:

1- the Captain is responsible for organizing rest shifts on any flight with more than basic pilots complement. He takes into account all parameters, including fatigue and experience.

2- all pilots in AF who are not Captains are First Officers, there are no Second Officers, nor any kind of 'cruise pilot' or 'relief pilot'. All long haul aircraft rated F/Os are trained specifically up to what some refer to as 'Cruise Captain standards'.

chefrp

16th Jun 2009, 20:36

Svarin....

Would it really make a difference who was in the right seat that night?

Svarin

16th Jun 2009, 20:39

Precisely : no.

chefrp

16th Jun 2009, 20:40

Unless maybe if it was

"John Wayne"!

PJ2

16th Jun 2009, 20:43

Dutch Bru;
You said:PJ2, with respect for your posts in general, your questions about the mobile crew rest area on the 330 has already been dealt with elsewhere in this thread. Post nr 773 and subsequently (perhaps deleted inputs) from colleagues to confirm that A330 mobile rest area come in the form of a LD6 container is normally located below deck on the hight of the 3L/R doors, aft of the wings.
Re " your posts in general"...what about them? I comment on a lot of items and raised the crew rest location once, so not sure what "my posts in general" means. Here's what I posted on June 09, which includes the information you repeat above:
I might observe that the "CREW REST" wreckage may likely be from the F/A crew rest module, not the cockpit crew rest area. I say this because the cockpit crew area is not labeled, it generally being near the cockpit. But the F/A module is, in some cases, removable and as such, labeling makes a bit more sense, (when stored...not when 'in use'). The location of this module varies with airline - some are mid-wing, others near the tail section; to the best of my knowledge, all such modules are below the main cabin floor.
I thought you said somewhere you were current on the 330
You clearly haven't read many of my posts so let me help you out:
I have flown 18,000+hrs since 1967 and have flown as captain and as an A320 instructor, 319s, 320s, 330s and 340s since 1992, (I'm now retired) and even with that background I do not consider myself beyond learning, (fearing to go where angels tread), am not an IT or structural engineer, but I know a thing or two about flight safety work and about the Airbus (and other types, B767, B727, L1011, DC8, DC9).

I am indeed recently retired, not current on anything at the moment. Up until a short time ago I was active in the areas under discussion, (flight data analysis, writing Airbus A330 CBT training modules). I make the fact that I am retired clear on a regular basis so that people understand I am not writing from current experience but still have something to contribute to the dialogue. And finally,

Appreciate your views, but the AF447 photo material as discussed in previous posts confirm that the flight in question was carrying a mobile rest area below deck aft of the wing section
Maybe, maybe not - where'd you get the info from? BTW, are we to get stuck on this minutae as well?

ELAC;
Being current on one carrier's aircraft won't tell you how another carrier has chosen to configure such items.
Precisely - shoulda said that but didn't think it necessary at the time - guess it was.

LiSaigon;
That may have been for Flight Attendants only
Yes, almost certainly.

Cockpit crew rest is described in another post here in this discussion; a J-seat with a velcroed-curtain instead of a dedicated rest cubicle - it's all about cost and real-estate.

I know of no cockpit crew rest installations which are not near the cockpit. I have seen one option when in the cutting-metal stage for cockpit crew rest facilities below the tail where the F/A's facility was also located and it was rejected out of hand by the pilots association, for what ought to be, (but I take nothing for granted on this thread), obvious reasons.

jungle drums

16th Jun 2009, 20:47

1. Has there been any more recent information from the naval resources trying to listen to the signals coming from the AF 447 ELT/CVR/FDR? Should not a reasonable point of impact (within 100nm box) be calculable from the debris field locations? I hope so at 100hrs/4 days to scan each 100sq/nm? A rough guess at roughly 10kts...

2. Is there equipment out there which is capable of scanning the sea bed for relatively small metallic objects (magnetic anomolies such as an aircraft's airframe or engines)? Or does it have to be Titanic size?
And if so, at what sort of range?
Perhaps these sort of questions can't be answered directly for military reasons. A pity that SOSUS has already been negated.

3. Why is stated in this thread that there was possibly no Standby Attitude available (is it called ISUS in the Airbus A330?)? A Standby AI has saved my life after a lightning strike in a TS. I would think that if the airframe is intact and any attitude indication is available, then an aircraft should be basically controllable irrespective of erroneous airspeed indication. Yes, in any case of an ADC failure it takes a cool head to ignore all the BS (overspeed and underspeed/stall warnings) and set a reasonable attitude and power. But as we are all trained - attitude + power = performance.

That leads me to ask - is the A330 Standby AI display a mechanical/analogue or digital/electrical one? (With my limited, but quite frightening experience in TS and lightning, I guess that I have now have a psychological preference for the non-electrical mechanical/analogue standby AI).

Lost in Saigon

16th Jun 2009, 20:51

Lost in Saigon

Can they receive weather reports/warnings as well, or do they rely on their on-board radar for this?

Thanks......

They can only send or receive text messages. Flight Dispatch can warn them of hazardous weather or even suggest a re-route. Sometimes warnings can be in the form of SIGMETS (Significant Meteorological Information weather advisory)

SIGMETS are issued to warn pilots of un-forecast or changing hazardous weather conditions. SIGMET: Information from Answers.com (http://www.answers.com/topic/sigmet-1)

But while en-route, Airborne Radar is the primary way to avoid thunderstorms.

Pontius Navigator

16th Jun 2009, 20:58

2. Is there equipment out there which is capable of scanning the sea bed for relatively small metallic objects (magnetic anomolies such as an aircraft's airframe or engines)? Or does it have to be Titanic size?
And if so, at what sort of range?
Perhaps these sort of questions can't be answered directly for military reasons. A pity that SOSUS has already been negated..

I don't believe Sosus would have helped in this instance.

Magnetic anomoly detection is essentially short range. I don't know but would guess that best system might be a towed MAD sensor which could operate near the sea bed. If the range was greater than 5000 feet I would be surprised. Sidescan sonar offers a much better chance of painting something but how soon it might be recognised is entirely different.

Lost in Saigon

16th Jun 2009, 21:01

Seems two possibilities:

One is for an operations officer with a modicum of training to act as a brains trust. The second is for a pukka engineer from the shop operating on a rooster as an extra duty.

Either way there would be no question of being 'on call' as this incident appears to have been all over between 4 and 14 minutes. I suspect that this would provide a conclusive arguement why it is not reasonable to have an engineer on duty.

Most major airlines already have this in place. (at least mine does). Major Airlines are 24 hour operations and there is always some one to talk to.

In a matter of minutes we can be patched into Maintenance Control, Duty Pilot, Chief Pilot, or anybody else we desire to speak to using SATCOM and/or ACARS.

Pontius Navigator

16th Jun 2009, 21:06

Most major airlines already have this in place. (at least mine does). Major Airlines are 24 hour operations and there is always some one to talk to.

In a matter of minutes we can be patched into Maintenance Control, Duty Pilot, Chief Pilot, or anybody else we desire to speak to using SATCOM and/or ACARS.

Excellent. But matter of minutes :(

JD-EE

16th Jun 2009, 21:26

Pontius, magnetic detectors are "another square law" device. The closer you are the smaller an object you can detect. And since you're expanding much like a radio wave as your "surface" gets twice as far from the detector the signal is reduce by twice squared or four times.

And with this plane I get the impression there are some largish metallic items to find such as the engines. The rest is basically non-metallic or small. But, of course, aluminum is not particularly a live item for magnetic detection.

All that can really be said is, "It's been done before. If it is important enough this time it'll likely be done again."

{^_-}

ELAC

16th Jun 2009, 21:38

Quote:
Today 21:19 Dutch Bru

ELAC

Appreciate your views, but the AF447 photo material as discussed in previous posts confirm that the flight in question was carrying a mobile rest area below deck aft of the wing section
Dutch Bru,

Yes, that's obvious to all of us who have looked at the photos, but it's not particularly relevant. I didn't suggest that AF447 didn't have one, and I don't think that PJ2's comment was intended to infer that either. What I tried to impart is that the parts recovered don't tell us anything about where the flight deck crew member might have been resting or if indeed he was resting at all.

That a portion of the LDMCR has been recovered is, in itself, about as relevant as the fact that a spoiler panel has been recovered (actually, less so).

ELAC

lomapaseo

16th Jun 2009, 21:51

That a portion of the LDMCR has been recovered is, in itself, about as relevant as the fact that a spoiler panel has been recovered (actually, less so).

I thought that the inference from the photo was whether it was occupied and whether this was indicative of no specfic CC warning about turbulence.

I don't know the answers to this but I had sensed in the thread that is why it was being discussed.

DJ77

16th Jun 2009, 21:53

I'm new to this forum, so hello to all gentlemen.
Some tiny details in the ACARS message list presented by Air France suggest that at least part of it may have been manually edited, a process prone to human error. Looking closely at it, I found a discrepancy between the warning labelled "NAV TCAS FAULT" and the associated JASC code. This code, 3443, stands for NAV/DOPPLER SYSTEM in the JASC classification. However, the a/c was not Doppler equipped. TCAS should be coded 3445.
All other codes appear correctly related to their labels.
Whatever explains this discrepancy, including a possible use of a slightly different classification by Airbus, I believe it should be considered doubtful that a TCAS FAULT warning ever happened.

PJ2

16th Jun 2009, 21:58

JD-EE;
So both water and ice will work.

. . .

So one major postulate for answering the question above is whether there was some form of filtering between the radar return and the crew that made them unaware of the magnitude of the problem they faced.
Re water and ice, yes, ice crystals less so.
Filtering/attenuation/antenna tilt/gain...all are areas which modify the return signal being displayed. The 777 radar does a lot of this work but I am told but I don't know anything about it other than what can be read in the B777 AOM.

Dutch Bru

16th Jun 2009, 22:01

Although your specific post has diappeared: good points. Well taken. At this side full of good intends a well. Regards, CV

Lost in Saigon

16th Jun 2009, 22:12

I thought that the inference from the photo was whether it was occupied and whether this was indicative of no specfic CC warning about turbulence.

I don't know the answers to this but I had sensed in the thread that is why it was being discussed.

Are you confusing the photo of the F/A's jump seats with the belts stowed to that of the Crew Rest area?

In any case, none of the photos prove anything about turbulence or crew awareness.

venux

16th Jun 2009, 22:14

Quote:
Originally Posted by Lemurian
Look, these messages are just the picture until 0214Z. What happened next is pure speculation.

BOAC replies: - maybe I'm not expressing myself clearly. I'm looking UP TO 0214Z, not 'next'. The point I am making is that there MAY be a case for saying the cabin was INTACT until the descent warning, which would cast doubt on the "the fin fell off and tore open the fuselage" line. In fact, how long would it take for the cabin to achieve 1800fpm? Could they even have been descending before the 0210Z message stream we have?

There have been contradictory, vague and confusing descriptions of this 0214z warning and it would be helpful if one of the experts would clarify exactly what it means.

A long, long way back in this thread, somebody suggested that this warning indicated a rate of descent of the a/c great enough that it would OVERTAKE cabin altitude before reaching ground level. In other words, if the a/c continued descending at the same rate and unless cabin altitude fell at >1800fpm, pressure outside would end up greater than pressure inside.

IF that is the true meaning of this warning, it indicates that two things were true at 0214z: (1) the cabin was still intact, and (2) the a/c had just started to descend at a rate sufficient to trigger the warning.

Could one of the experts please confirm the meaning of this message?