AF447

[barit1]

Consider something here:

SAT is impossible to measure directly @ M0.8+ and FL350. The displayed SAT is a computed number, based on measured TAT, with the ram temperature rise (a function of Mach) "backed out".

If the pitot is blocked, then the computed ram rise is in error. Depending on altitude excursions, this error could be either plus or minus direction.

Ergo, the computed SAT could be in error either direction.

[SLFJB]

Briton Identified from AF447

Tiscali reporting that one of the bodies recovered from AF447, has been identified: See link below

Briton's body recovered from crash

My Question : The passengers would have been issued with seat numbers. Would Air France retain this information, and assuming that most people would be seated on a night flight, would this help to identify where the plane broke apart? (I.e. Near the wings as suggested , or at the VS)

[Dutch Bru]

Report on NSTB website

The site:
NTSB - Press Releases
The release:
Press Advisory

[DJ77]

From PJ2:

don't the GE installations use EPR?

As far as I know, GE-CF6 and GE90 series always used N1 as driving parameter.

Also, Nara is a medium size town abt 15 miles east of Osaka, jpn. This looks like a typo for Naha.

[SaturnV]

My Question : The passengers would have been issued with seat numbers. Would Air France retain this information, and assuming that most people would be seated on a night flight, would this help to identify where the plane broke apart? (I.e. Near the wings as suggested , or at the VS)

SLEJB,

In both AI182 and TW800, the recovered bodies of passengers were matched with seat assignments, as well as the results of the post-mortem autopsies. The information can be quite helpful in determining the cause or sequence of a mid-air break-up. Even if bodies aren;'t recovered, that can be telling. For example, in TW800, bodies were never recovered for several rows of passengers seated directly above the center tank.

[Hyperveloce]

Originally Posted by Dutch Bru

The site:
NTSB - Press Releases
The release:
Press Advisory

Ok, now it is official.
Could the NTSB impose mandatory airworthiness directives about Pitot probes although the AESA seems reluctant to do so ?
Jeff

[SaturnV]

Ok, now it is official.
Could the NTSB impose mandatory airworthiness directives about Pitot probes although the AESA seems reluctant to do so

Hyperveloce, the NTSB can investigate and make recommendations to a regulating agency that a certain action be taken. The FAA, as a regulatory agency, would issue any ADs.

The NTSB's recommendations are not binding on a regulatory agency.

In this instance, the NTSB has authority to investigate, as one flight originated in the United States, and the other flight was a U.S. airline

[PJ2]

DJ77;

As far as I know, GE-CF6 and GE90 series always used N1 as driving parameter.

Also, Nara is a medium size town abt 15 miles east of Osaka, jpn. This looks like a typo for Naha.

Yes, you're right, thank you, and the Trents are EPR, I recall now.

I found the town of Nara but was thinking of the name of the FIR. The "Naha" typo is immaterial now I suppose.

[YRP]

Originally posted by xcitation:

At the end of the day I can't understand how one sensor (ok 3 redundant pitots) can take out something as critical as an entire flight system. It appears that there are too many dependencies in the software that causes the pitot failure to initiate a chain of software errors that are not handled. Instead the errors cascade to secondary software modules which go into failed state. Good electronic/software design seeks to minimize coupling and un-necessary dependencies. Error exceptions are handled gracefully and trapped.

The problem with redundancy is when there are common fault cases. For example if (hypothetically - not implying this happened with AF447) there is a certain environmental condition where a pitot sensor is susceptible to icing, this might apply to all three.

With three failed pitot sensors, there ain't a whole bunch clever software can do. Even with two failures, you have a big problem - two incorrect sensors in agreement with each other, or possibly three all disagreeing.

But be careful about blaming automation for this. How much better would a human pilot do with three separate sensors connected to three separate airspeed indicators? If indeed there is a pitot icing problem here, the problem is a mechanical/physical pitot/static sensor design problem NOT one in the software/automation.

[FlyingOfficerKite]

Interesting that the body of the Captain has been found along with some of the passengers and cabin crew.

I wonder it he was in the cabin rather than the flightdeck at the time of the accident?

FOK

[Dutch Bru]

German "NSTB" report/SR on Airbus pitots from 1999

A report dating from January 1999 (nr. 5X002-0/98) by the German NSTB (the Bundesstelle fur Flugunfallundersuchung BFU) is a very interesting read vis-à-vis pitot icing on Airbus a/c.

The report concretely deals with a serious incident on an A320-200 in a holding pattern near Frankfurt (with PROBE/WINDOW HEAT on AUTO), when, in heavy icing, rain showers and turbulence, the speed indications on both Primary Flight Displays as well as on the stby-indicator disappeared. The A/P and A/T disconnected and several associated warning and failure messages were generated. If I understand things correctly, the a/c had gone into Direct Law, since the report mentions that the PIC stabilised the a/c at 10000 ft using horizon and engine power (PITCH and POWER). Shortly after stabilisation at that altitude all three speed indications returned. For the landing A/P and A/T were again available.

It is interesting that the report notes that the AIRPLANE OPERATING MANUAL states that when comparison between both ADR systems is impossible (ADR-DISAGREE) and the stby system isn't available either, as was the case here, the flight crew should act on their own discretion and experience.

Even more interesting the report (I repeat from 1999) notes that "disappearing speed indications in certain weather conditions have been notified by several A320 operators". And that "in 1993 Airbus had issued SB 34.10.00.011, addressing that specific problem", and had "informed all operators of A320, A321, A330 and A340". The report continues to note that "with the information note and with the experiences in daily operations it is a principle fact that all operators and maintenance organisations were bearing knowledge of weather related anomalies in the speed indication systems of certain Airbus types". However, the report also states that maintenance organisations were not given any indications for further measures.

The report concludes that because of the particular construction of the probes, unlimited flight of these a/c types in heavy rain and icing conditions is not possible. Reason why the BFU has issued the following safety recommendation:

01/99: The specification for the pitot probes should be modified in such a way that unlimited flight in heavy rain and icing conditions is possible. The application of the already developed improved pitot probes for all aircraft types (A320, A321, A330, A340) addressed in SIL 34-047 should be made mandatory.

[Graybeard]

"..How much better would a human pilot do with three separate sensors connected to three separate airspeed indicators? If indeed there is a pitot icing problem here, the problem is a mechanical/physical pitot/static sensor design problem NOT one in the software/automation."

I can visualize a PF saying, "Hey, my airspeed is falling off." The PNF replies, "Mine, too; and so is the standby, just like that recent advisory."

PF then disconnects the automatics and flies attitude.

Meanwhile, the automatics behave exactly as they were programmed 15 years ago. They didn't read the advisory..

GB

[jcarlosgon]

Qoute: "The report concretely deals with a serious incident on an A320-200 in a holding pattern near Frankfurt (with PROBE/WINDOW HEAT on AUTO), when, in heavy icing, rain showers and turbulence, the speed indications on both Primary Flight Displays as well as on the stby-indicator disappeared. The A/P and A/T disconnected and several associated warning and failure messages were generated. If I understand things correctly, the a/c had gone into Direct Law, since the report mentions that the PIC stabilised the a/c at 10000 ft using horizon and engine power (PITCH and POWER). Shortly after stabilisation at that altitude all three speed indications returned. For the landing A/P and A/T were again available."

If they were on direct law, they were lucky, IMHO. In alternate law, if IAS goes to a (right or wrong) low figure one has to pull the stick to avoid nose down automatic command (and the contrary for high - right or wrong - IAS reading). The problem seems to be on who decides what law to go to. It is the system’s decision, not the pilot’s. I wish I had a switch I could flip (no push button) into direct law in these cases, avoiding alternate law and its “protection” from too low/high right/wrong speed readings. Then it would be easier to maintain the required pitch.

Another subject, unreliable IAS procedure: I fly both GEs and PWs. GEs have N1 as the main thrust setting, PWs have EPR. For both types the Procedure shows only N1 settings for unreliable IAS, which is good. EPR is unstable and dependant on ice on probes, as well.

[PJ2]

YRP;

But be careful about blaming automation for this. How much better would a human pilot do with three separate sensors connected to three separate airspeed indicators? If indeed there is a pitot icing problem here, the problem is a mechanical/physical pitot/static sensor design problem NOT one in the software/automation.

[my italicizing/bolding]:

Very well stated. In fact, (if I may beg the indulgence of the mods for a moment), from designer to pilot, the notion you have expressed is fundamental to any comprehension of what is at the heart of "automated" flight. Software that can exercise human judgement (as when a pilot must make a judgement based upon 'what is reasonable'), has yet to be written not because it is difficult technically but because it is difficult philosophically as anyone fascinated by Turing and the notion of AI will attest.

The notion, "philosophically", is not meant in any ethical sense - it is not a "should we or shouldn't we?" question, it is a question of 'understanding what human understanding and judgement means'. Further, the act of understanding "understanding" is itself, a philosophical act.

And when we "understand" (or 'see'), we do so in a particular way and not just any way. Our "seeing" is a particular template and not a mere objective act from which we can then derive "objective" data or perceptions.

This is an implicit limitation on knowing what "human judgement" is, in the same way that Heisenberg thought of uncertainty when he posited the notion that the act of measurement itself, affects that which is measured.

In other words, our perceptions or, the "way" we perceive, are not innocent nor are they given - thus understanding judgement is an elusive project which demands not a technical understanding but a philosophical one.

Hopefully that circuitous discussion of your point regarding automation will add to understanding why designing and writing software to "mimic" human responses (ie, judgement) is a different-order problem and not an "automation" problem per se.

Graybeard;

PF saying, "Hey, my airspeed is falling off." The PNF replies, "Mine, too; and so is the standby, just like that recent advisory."

PF then disconnects the automatics and flies attitude.

Meanwhile, the automatics behave exactly as they were programmed 15 years ago. They didn't read the advisory..

Precisely the point, said in much clearer terms!

PJ2

[EGMA]

Excellent point made by Graybeard.

Have we not gone full circle?

Before CRM we had the FO content to watch the Captain fly the aircraft into the deck because he thought the Captain was perfection personified.

Now we have an entire crew trusting a perfect flight computer, unable to hit the 'manual' switch because it's not in the SOPs.

[PJ2]

EGMA;

Now we have an entire crew trusting a perfect flight computer, unable to hit the 'manual' switch because it's not in the SOPs.

Well, trusting an airplane is what a pilot must do - that isn't the issue, but I think I understand what you mean.

The issue with automation is a two-pronged fork which the industry must deal with. Many of us saw this coming and said this more than twenty years ago when the 320 was first introduced - reliance on automation is as much a management/cost issue as it is a training issue. Automation is a bean-counter's dream because they have been sold the bill of goods that "these airplanes fly themselves" and so pilot costs can be reduced as can training costs. I know many (including myself) who never agreed with this approach, could see it happening, and wrote then as now, about the issue. We have now arrived at the point first seen back then and the industry must deal with it.

PJ2

[UNCTUOUS]

What Physical Phenomenon ties all the developments?

NTSB release:

two recent incidents in which airspeed and altitude indications in the cockpits of Airbus A-330 aircraft may have malfunctioned. The first incident occurred May 21, 2009, when TAM Airlines flight 8091 (Brazilian registration PT-MVB) flying from Miami, Florida to Sao Paulo, Brazil, experienced a loss of primary speed and altitude information while in cruise flight. Initial reports indicate that the flight crew noted an abrupt drop in indicated outside air temperature, followed by the loss of the Air Data Reference System and disconnections of the autopilot and auto-thrust, along with the loss of speed and altitude information.

Just theorising here (and whilst ignorant of the actual A330 system design); I asked myself "what physical phenomenon could cause a sudden loss of both airspeed and altitude info in a system reliant upon Air Data Reference Units for so many inter-dependent flight control functions?". and "Why the abrupt drop in IOAT?"
.
I surmised that it would not be solely due to water ingress into (or freezing within) the pitot pressure lines and could only be due to a more insidious pneumatic error, most likely one within the aircraft's static pressure lines. So, a leak or locked pressure?
.
Assume that one (only) static line becomes contaminated by water and that this water contaminant freezes and expands in the high-level cruise - at a low point "water-trap" in the static system. The ADIRU comparator function now has a growing problem integrating and resolving ALL its static inputs. Because it must assimilate inherent minor differences in the various sensed statics, the system won’t fuss over minor errors in quantum, particularly not over one that arises in the stable cruise. However it would if a line froze up in a climb, because the error rate would be much higher. The static pressure that feeds the barometric height function of the autopilot is now compromised by the error but, in the level cruise, the error is latent and the error-rate isn't rapid enough or yet large enough to excite an ADIRU disagree trip and the aircraft will either climb or descend (very gradually at first).
.
Why will it do that? and why will the flight-crew not notice? Will it climb or descend?
.
During the finite period that the error is compounding, the autopilot's barometric function will maintain (what it interprets to be) the FL350 pressure surface and FL350 will be consistently displayed. However, because of the locked static pressure, that "pseudo level" will not accord to a "real" FL350 at the actual standard pressure datum (QNE: 1013Hpa). The aircraft will actually be climbing due to fuel burn-off, but all the pilots will see over this period is an infinitesimal progressive change in pitch attitude and they'll not notice the minuscule movements (if any) in elevator auto-trim. Postulating here a gradual and insidious error due to a compromised static feed to at least one ADIRU, an error magnitude that's initially below the threshold at which an ADIRU comparator disagree will trigger. Speed would bleed off gradually (see post 2265for the reason) - except that the auto-throttle will be gradually increasing thrust, (albeit minimally) to maintain the stipulated IAS. The actual IAS will however be higher (and will climb gradually).
.
Ultimately however, once a disagreement trip is triggered by error size, what could happen?
.
The ADIRU affected by the compromised static line will trip off as the error size becomes untenable.... and the others will then reject its distorting static pressure contributions.
.
What will suddenly happen to the displayed IAS/MACH, altitude, IOAT and autopilot functions at that point?
.
The speed will change drastically (i.e. increase suddenly to a correct value), the altimeters will jump to the actual increased altitude and the autopilot will disconnect. Auto-thrust will quit and the IOAT will suddenly drop to the true value at the increased altitude (assuming here that it's sensed within the static system).
.
What could trigger a loss of control? Firstly the aircraft's actual altitude (and speed), when the ADIRU "fails", will be nearer to, or even above, coffin corner and sudden manual flight (particularly in Alternate Law) can be quite marginal at that point, particularly if the pilots see the sudden change of IAS/Mach and altitude as an error (rather than what it is, a leap back to reality)..... and inadvertently react with large or uncoordinated manual inputs.
.
Why wouldn't this malfunction scenario be later detected in the aircraft that have survived such encounters?
.
Firstly there's a presumption of pitot probe misbehavior.
Secondly, unless trapped water accumulates such that it can be drained from a limited number of static system low point traps, the water may or may not be detected nor drained (or even if it is, will not be seen as the real villain in the piece).
.
To accept this theory you need to suspend disbelief in the incorruptible infallibility of displays not annunciating errors, have some knowledge of what an ice-locked static pressure can insidiously do to an automated system, believe that computer software programmers just would not have accommodated such a scenario - and do quite a bit of lateral thinking about a cumulative growing error due to burn-off. In respect of inherent software errors, we have the example of the Turkish 737 at Amsterdam for a precursor (i.e. it shows what a RADALT indication glitch, not an annunciated RADALT failure, can do to a flight control system).
.
How does water get into static lines? Does it happen inflight or on the ground?
.
Water ingress is most likely to occur on the ground. In heavy rain water running down fuselage sides can be "sucked in" - in an atmospheric pressure-dropping environment. In some aircraft the static port "bungs" used to plug upwards into the static port but they were also hollow to equalize the pressure. Rain-water running down these hollow bungs would still be sucked up into the system. Nowadays the static ports are left "unbunged" and are usually a number of small holes (about 5 to 15) grouped together, to defeat insect entry. Small diameter holes won't defeat rainwater ingress however.
Why would this process be happening nowadays and not earlier in the A330/A340's life? Minor changes in hardware, software, maintenance routines or operating procedures can introduce unintended consequences. Changes in operating procedures? Perhaps crews are flying higher earlier with lighter load factors.
.
Posts 2406 and 2408 - are also pertinent to this theory.
.
Don’t shoot the speculator. If you don’t like the premise, postulate a more credible scenario…. preferably one that explains how an iced pitot can affect altitude.
.

[24victor]

Getting back to the weather

Orestes wrote

The alleged NTSB e-mail mentions a sudden "drop" in air temperature - meaning it suddenly got colder? That seems counterintuitive - I had thought the more likely danger was the sudden increase in air temperature associated with these weather systems. I'm confused....

Conventional wisdom has it that you can't pick up heavy icing in the upper flight levels. Practical experience would suggest you can. If the NW incident reported by the NTSB is factually accurate they lost static as well, which I find hard to fathom. Does that mean they lost rudder limiting as well as the indicated speeds rolled back?

I had this recently from a mate of mine flying BBJs but with a lot of tropical heavy experience. He wrote;

Came back from SIN the other day and had ISA +28 at FL 360 going through some weather system. It must have been the beginning of a cyclone.We were 2000ft below Max Alt before we entered.Once in cloud Max alt came down to FL 360.Never seen a temp like that before.

High temps. Low temps. Ice where it shouldn't be. Rain at FL390! Someone remarked at the start of the first thread about Al Gore claiming responsibility under the guise of climate change. I grinned, but maybe there's something to it. We know a lot more now about windshear and microbursts and we're learning more about SCD. Maybe, like these phenomena, icing, rain and rapidly varying temps in the upper Flight Levels have always been there and we just didn't (collectively) know it.

Rgds.

24V

[Squawk_ident]

Signal located (?)

According to the website of the French newspaper "Le Figaro" dated yesterday evening

Le Figaro - France : AF 447*: les corps du piloteet d'un steward identifiés

.../...
Partial translation


The BEA in charge of the inquiry announced that it would publish its first report Tuesday.
No track is officially turned down.
The Brazilian authorities identified Tuesday the body of the commander of flight AF 447, Marc Dubois, and the one of a steward. The information was confirmed Thursday by the direction of Air France.
These two new identifications thus change the number of victims identified from 11 to 13 out of the 50 bodies retrieved from the South Atlantic. The confirmation of this information by Air France and not by the BEA in charge of the inquiry, confirms a certain uneasiness with the Brazilian authorities. Those indeed have not accepted within their teams the French doctor sent by the BEA who came back without informations.
The BEA on the other hand announced that it would publish officially its report on the accident on Tuesday. Until there the investigation could accelerate if the black boxes of the plane were located and found.
According to several sources contacted Thursday, the French Forces have indeed picked up a signal, “they are turning around it for a few days to locate it”, explains a military source. “The question is now rather to know if the transmitter is still interdependent of the black box considering the conditions of the accident”, explains a source close to the file.

.../...

Things started to get confusing I would say. It would mean that the previous information published by "Le Monde" on the 23th and rapidly denied by the BEA was correct after all?

[cribbagepeg]

Water at subzero temps

Here's where the much earlier comment about supercooled water could come into play. As I understand it, the term refers to water that has not passed from the liquid to solid state despite being below the freezing point. This can happen when cooling takes place very quickly, and in the absence of both mechanical forces AND particulate matter such as dust, around which crystals would normally begin to form. I've long since forgotten the physics, maybe something to do with the expansion and subsequent contraction of the droplet which would normally happen during "graceful" cooling.

BAM, the supercooled droplet undergoes mechanical stress, upsetting the delicate supercooled state, and it solidifies upon contact with the AC.
From here, I get lost in the thermodynamics of sublimation, melting and
eventual equilibrium. It strikes me, however, that a relatively large mass
of supercooled water (now ice) forming VERY suddenly might take a few
or many seconds to clear, even with fairly powerful heaters - remembering that latent heat of fusion thing....