AF447

[BryceM]

Me Myself:
This is really far streched mate !! The last accident was in 1969 and nothing happened until 1988 which makes it a good 19 years.
The question that needs to be asked is :
How come we went from the horror show of the 1960's to the pretty good track record of the 70's and 80's.
I can partly answer that one. 2 of these 1960's crashes were the direct result of very poor CRM. I can remember flying with skippers who were F/O's during that period of time and you wouldn't believe the stories they had to tell.
Obviously, things were done during these 19 years and then..........well, the plot was lost.
Fast growth, if a factor, is not acceptable. AF isn't the only airline that grew fast in the last 20 years.

I don't want to get drawn into a prolonged discussion on this, but all I'm saying is that I don't believe that your personal feelings about a declining safety culture are necessarily backed up by the evidence of accident stastics. The point I was making is that by selecting arbitrary start dates and periods for the accident statistics quoted, very different pictures can be painted of the 'change' in accident rate. (Why not start in 1960 and divide into 10 year periods? - again, a completely different 'trend' might emerge; the periods selected were clearly chosen to give the worst possible impression).

It is very likely that in a strict statistical sense, the number of crashes in recent years tells you nothing meaningful about AFs safety record. (As in 'the null hypothesis that AFs safety record has got worse was not proved to a confidence interval of 95%' [or whatever]).

You can have your own anecdotal evidence about a decline in safety culture, and that is what you should probably base your arguments on. After all, that user feedback is what the Flight Safety dept of any airline should focus on, surely? (I mean, they don't work harder only if they've had an accident, do they?)

As I say, I don't have an axe to grind about this - just pointing out that the statistical 'evidence' being touted for a decline in AF safety isn't very
meaningful.

We've wandered away from AF447 a bit. I'm done on this subject, anyway.

[txl]

Just to make that clear, in my last post I was just paraphrasing what a German language news story said, not my opinion, as I don't have one regarding this particular topic here.

As for the report itself: I wouldn't say the "Spiegel" report shows anti-French bias. It suffers from bad reporting. Pulling a three year old report out of the file and adding it to the mix of seemingly arbitrary and/or completely unrelated numbers for some "perspective" (even if there might not be any) is typical of cheap, sloppy and sensationalist reporting.

[Me Myself]

I don't believe that your personal feelings about a declining safety culture are necessarily backed up by the evidence of accident stastics.

7 crash in a little more than 20 years ???? What is a statistical evidence then ??? You can't be more " evident " than a pile of rubles can you ?

[Art-Deco]

On page 47 it is quoted:

The company that operates the satellite used by AF447 has provided the traces of the messages transmitted to the aircraft and seen by the satellite.

Why isn´t the log attached ? Why didn´t BEA take this into their calculation ?
The NOC-log (telemetry-transponder data) from Inmarsat would also include crucial data type strength of transmission signal an timing of packages etc.

As the infamous gap of 31 seconds EXISTS in the report, it should ALSO included the telemetry of all these ACARS. If, for example the signalstrength is fixed all the way with a small variable, we could directly calculate that the a/c kept is fligth level in bad weather, but if signal was degrading all the time we could calculate from what timing the degradation occurs and pinpoint approx at what ACARS-message that the a/c starts to divert.

My impression is that BEA has a copy of the telemetry-read out for this date from Inmarsat Noc, otherwise they should ask for a copy a.s.a.p.

[Robin42]

My impression is that BEA has a copy of the telemetry-read out for this date from Inmarsat Noc, otherwise they should ask for a copy a.s.a.p.



l’écart observé entre le message de 2 h 13 min 14 et celui de 2 h 13 min 45
est dû, au moins en partie, à une interruption temporaire du lien de communication entre l’avion et le satellite,


A follow up to my post "Mobile OnAir GSM service on board?", which provided an Air France source got deleted for elliptical reasons. May be it wasn't the time for it, as there might be sound tactical reasons not to publish such things too early into the investigation. Now that more than a month passed by, nobody can hinder anymore the investigation by falsy claiming to have received calls or SMS from a passenger during that sad night.

Yet my assumption proved reasonable, as the French version states on p. 71:
il n'y a pas eu de communication téléphoniques par satellite entre l'avion et le sol.Source: http://www.bea.aero/docspa/2009/f-cp090601e1/pdf/f-cp090601e1.pdf

So they probably had Mobile OnAir installed, which is a satelite linked GSM picocell, allowing for inflight use of a cell phone for calls, SMS and internet via GPRS. Source: http://corporate.airfrance.com/index.php?id=communiques_detail&tx_ttnews[tt_news]=2699&L=1&no_cache=1 (http://corporate.airfrance.com/index.php?id=communiques_detail&tx_ttnews%5Btt_news%5D=2699&L=1&no_cache=1)

But even if there actually were no phone calls (possible, during night), it would be interesting when the last SMS got transmitted or received, when the last internet packet got transmitted, and exactly when INMARSAT saw the picocell on board the very last time. This would allow to differenciate the case of more ACARS messages not being sent because there weren't anymore, and the case that the satellite link broke down.

[Art-Deco]

re: Robin42.

Another detail is the hf communication.
Around the Intol area, Stockholm radio has a class-2 area of reception, and AF OCC used them the same morning but several hours later in a atempt to reach 447. All statements of poor hf is only with Dakar to my knowledge, and Dakar also didn´t have the flightplan, it could pinpoint that Dakar had a comms-breakdown at a real bad timing as well.

Service-area:
Stockholmradio aero (http://www.stockholmradio.se/aero/show.php?id=1225563)

Propagation forecast:
http://www.stockholmradio.se/aero/source.php/1240890/HF-Propag-forecast_MAY-JUL09.pdf

[SaturnV]

FLY400, thanks for the clarification and the link. As I read the description, the first step is that the crew of the plane must first logon to the system, and after logon is successful, then contact is initiated by the ground station, which can query the plane periodically without further involvement of the crew.

If so, my sense is that the crew tried to logon with Dakar three times, and each time the logon failed because Dakar had no flight plan filed for AF447. I would not think the system would allow Dakar to initiate these three contacts if Dakar had no AF447 flight plan within its system. (Though I would expect that the system at Dakar would record logon failures, and that is where the record originates.)

[Me Myself]

But even if there actually were no phone calls (possible, during night), it would be interesting when the last SMS got transmitted or received

There is no GSM system on board AF long haul aircrafts. The only system available to passengers is the satellite phone only accessible to business class. That kills the SMS story.

[CONF iture]

BEA never mentioned AF447 was 3 NM off track ... WHY ?

[Art-Deco]

Could someone here explain why the 3rd ADS-C attempt 2:01 didn´t give a GPS-position, and why BEA has removed all 3 attempts from the ACARS-list?
Clearly the 02:01Z was via ACARS as they were out of VHF-range.

Copied from BEA:s intrerim-report:

1.9.2 Coordination between the control centres

At 1 h 46, the DAKAR controller asked the ATLANTICO controller for further information regarding flight AF447 since he had no flight plan. The ATLANTICO controller provided the following elements: A332, from SBGL to LFPG, SELCAL: CPHQ.

The DAKAR OCEANIC Regional Control Centre created the flight plan and activated it. The result of this was to generate a virtual flight following the planned trajectory in the DAKAR FIR between TASIL and POMAT. There was no radio contact between AF447 and DAKAR, nor any ADS-C connection. The flight remained virtual.


1.16.2.1 ATC messages

No ATC messages were received or transmitted by F-GZCP. Only three attempts were made to connect up to the Dakar centre ADS-C system and were recorded on 1st June at 1 h 33, 1 h 35 and 2 h 01. The three requests were refused with a FAK4 code, meaning that the control system had detected the absence of a flight plan for this aircraft or that there was a mismatch between the flight plan filed for this registration number, the flight number and the reported position.

[WojtekSz]

BryceM: .... After all, that user feedback is what the Flight Safety dept of any airline should focus on, surely? (I mean, they don't work harder only if they've had an accident, do they?)definitely they do work harder! Incidents and accidents are the moments of the truth for any security system - it is only then that all assumptions and SOP have met the real-life, real-time combinations of inputs that has proven to be not accounted for and has to be carefully scrutinized and used as guidance for new set of SOPs. This is called continuous improvement process.

BryceM: ... just pointing out that the statistical 'evidence' being touted for a decline in AF safety isn't very meaningful.statistics is sometimes misleading as for money it can show different face of the same reality. The arguments put up by Me_Myself (years between accidents) seems to be more objective - definitely for passengers who prefere to not to be on the wrong side of the statistics.


I am observing this thread for the last 4 weeks. I have learned a lot on intricacies of the planes and procedures. Many hypothesies were proposed and some seem to look more probable than the others. Yet without more hard facts from FDR/CVR we are left in the dark.

At the same time the thread has reached a point where it started scratching the surface of the delicate subject of the pilots being humans spending part of their time flying the planes and the other part being a part of the big commercial corporation. These two lives do mingle and this may influence safety.
And without any prejudice what or who has played an important role in critical sequence this may be important factor towards understanding what has caused AF477 down.

For all the AB automation it is the PF who is called for when there is a problem. And for any foreseeable future passengers will prefer to fly with qualified pilots onboard :)

Wojtek

[mm43]

@Art-Deco

Could someone here explain why the 3rd ADS-C attempt 2:01 didn´t give a GPS-position, and why BEA has removed all 3 attempts from the ACARS-list?
Clearly the 02:01Z was via ACARS as they were out of VHF-range.1.. They are not ACARS messages. Those ACARS messages were transmitted via SATCOM to Air France Operations/Maintenance. The crew also rec'd METAR / TAF info ex Paris by the same means.

2.. The ADS-CPDLC uses SATCOM also, but the initial contact with an ATS is made by the crew. Once log on is established the ATC select the info they require etc..

Here are DAKAR's requirements to establish ADS-CPDLC:-

Connection procedure in the DAKAR FIR (DAKAR land and DAKAR ocean)
The first connection with the system is made by the crew. For flights entering the DAKAR control region from an FIR not equipped with CPDLC, the DAKAR control centre demands the connection at least twenty minutes before entry into the DAKAR FIR. For flights from a FIR equipped with CPDLC, the first connection must occur five minutes before entry into the DAKAR FIR.NOTE: AF447 was not ADS-CPDLC with Brazil ATC and the 20 minute Log On requirement was in effect.

It is obvious that DAKAR ATC failed to make provision in their ADS-C system for an a/c they had accepted on a Virtual Flight Plan - hence the Log On failures. The interaction between ATLANTICO and DAKAR (or lack of it) may ultimately have a significant bearing on where the finger points.

mm43

[PJ2]

With the entry of a number of new posters, the thread is now chasing it's own tail with all these comments about ACARS, SATCOM and personal opinions about cultural differences and flight safety and micro-interpretations regarding tiny nuances of the BEA report and the safety records of various airlines. These issues were either dealt with a thousand posts ago, or are irrelevant and inconclusive. Given that those who were contributing good, creative stuff that got us somewhere have gone silent, I suspect for most these fly****-in-pepper mini-me points are mind-numbingly trivial. Can we not do better than this?

[singpilot]

+100.

Thanks PJ2

[JD-EE]

WhyIsThereAir

I read the report of the IB crew as saying they had decided to take on the additional 2000 kg of fuel. AF of course took on about half that.


That raises another can of worms. It is reported that the plane was over MTOW as it left the gate and was just under it as it lifted off. Does that suggest that the aircraft was overloaded and the crew did its best to handle the situation despite that fact? The fuel situation may well have colored the crew's election to plow on roughly straight line into the mess.

I hope the hole in my conjecture does not mean the pilot lacks the authority to cut some of the load from the plane to get more fuel aboard. That sounds absurd.

JD-EE

[LeandroSecundo]

Hi,

BEA said that "nothing" so far can link the disaster to failure of the Pitot probes.

If we was able to return backward .. and have not the pitot problem (a wizzard keeping them working) ... can the AF447 disaster occur so certainly ?

Bye.

[WojtekSz]

PJ2: ...With the entry of a number of new posters, the thread is now chasing it's own tail with all these comments about [...] flight safety and [...] the safety records of various airlines. These issues were either dealt with a thousand posts ago, or are irrelevant and inconclusive. PJ2:
Sir, i do have highest respect for your extensive knowledge on the subject yet i can not concur with your statement that comments on flying safety are ' irrelevant and inconclusive '. One can talk about conclusive when discussing technical points which can be proven by referencing to other identical or similar planes/solutions. Yet there are some inconclusive subjects which are as important as the technical ones.

In principle planes crash because of :
plane (technical) problems - many possibilities have been discussed earlier, also with your, highly respectable, input. Seems that the F-GZCP plane was fit for the flight and whatever technical problems encountered the plane itself was not a direct cause for the crash. Still there are open issues about AI approach to flight automation which were thoroughly discussed with many improvements proposed
environment (weather) problems - discussed earlier in the thread with excellent Tim's & al input. The weather was bad yet typical for the time and area. Planes do fly the same route every day.
pilots error - just a possibility but as it is extremely delicate matter to be discussed among pilots this part was not so carefully discussed. As tough it is, still it has to be analysed because it MAY reveal aspects that MAY have lead to the tragic crash. Especially that those aspects are sometimes referred to as soft or indirect ('economy' training being already discussed, corporate culture discussed a little) - still having crucial influence on many things. The water also can drill the stone if given enough time. Last few days has brought some light on this subject and even more input is needed to grasp the problem. Even the captain is reporting to someone... If flying crew made mistakes lets identify them and teach others to avoid them. If roots of the mistakes are in the corporate politics lets identify them and imply change to the corporate ethics/policies before some other pilot fill find himself and his plane in problemsPJ2: ...Can we not do better than this?yes, we have to. In memory for the 3 flying crew and 9 cabin crew members and all 216 passengers. And all other people that are bording the planes every day, in every part or the world.

[p51guy]

With no new data rehashing old information seems pretty meaningless at this point. Hopefully the black boxes will be found or this will be all we will ever know about what happened.

[lomapaseo]

WojtekSz

One can talk about conclusive when discussing technical points which can be proven by referencing to other identical or similar planes/solutions. Yet there are some inconclusive subjects which are as important as the technical ones.


We tried this out in the Techical forum and it seemed to work best there.

If you place this kind of discussion within a specific accident thread it begins to smell like the blame game before the wreckage is even recovered and they should have seen it coming the word (they) being a fill-in-the-blank group

[Dysag]

If AF447 took extra fuel it would at least give a clue that they were prepared to deviate around the weather.

If a flight didn't take more fuel it tells us nothing. The extra fuel required to pass 50nm to one side and then rejoin the original track is almost negligable: just a fraction of the en-route contingency already on board.

[FLY400]

If I (FLY400) am one of those new posters to which you refer, I will be very happy to continue the ADS-C discussion on a new thread, maybe in Tech Log. Or, alternatively I can butt out completely and just continue the private email disussions that I have been having with Lemurian and PK-KAR.

I was mererly answering a question in this thread. In this case, as an expert witness.

FYI, I was one of the original team at ISPACG that put together the protocols for FANS 1/A, as was the amazing guy that maintains the website at ATC DL News Home (http://members.optusnet.com.au/~cjr/index.html) which he does gratis, in his spare time. Craig is one of the most dedicated, professional and practical Air Traffickers that I have ever had the privilege to meet.

Patience my friend.This is a very complex puzzle and is going to take a very long time to sort out. I admire your inputs in this forum and understand your frustration.

To sort it out is paramount for for the sake of those who lost their lives and our industry! Please, the searchers will find the wreckage and the CVR/FDR.

David

[Lightning6]

If AF447 took additional fuel it would at least give a clue that they were prepared to deviate around the weather.

If they didn't take more fuel, it tells us nothing. The extra fuel required to pass 50nm to one side and then rejoin the original track is almost negligable: just a fraction of the en-route contingency already on board.

IMO, if they didn't have enough fuel for a diversion around the CB (Which I doubt), they should have diverted back, probably posted on here before, but this thread seems to be going round in circles and getting nowhere!!

Having said that, it's been interesting reading this thread, if only for the technical information learnt.

[BOAC]

That raises another can of worms. - can I kill that one straight away please? Quite 'normal' to 'leave the ramp' over Max take-off mass - there is a 'max ramp mass' published for aircraft. They took off BELOW max take-off mass, therefore there is anything wrong there.

It appears that they fuelled to max. With ETOPS there are always options at all points along the route for diversion etc if necessary. 940kg is as much as they could get on and was apparently ABOVE the 'extra' required for the ETOPS planning. Cannot see a problem there. A well-planned 'detour' around weather does not burn that much extra.

[PJ2]

To those who offered their thoughts on my last post above, thank you.

I didn't communicate my thoughts very well but my contribution was motivated by a slight frustration with the minutae - a half-mile here or there with regard to the last position, the finely-honed discussion on intrepreting the report and so on - clearly such discussion has meaning for others but didn't seem to be advancing what we need to understand most which is, what caused the accident. I think lomapaseo has made a good observation and suggestion. As for me, I'll adopt a more patient stance. Again, thanks for the feedback.

[CONF iture]

The coordinates for the last position received are 3 NM abeam the Airway UN873.

There is absolutely no justification for the Interim Report not to explicitly mention that information.
To limit the analysis to the following comment is not satisfactory :
Originally Posted by english copy p. 68
up to the last automatic position point, received at 2 h 10 min 35 s, the flight had followed the route indicated in the flight plan

[Will Fraser]

CONF iture

I think there is a fair amount to criticize in the Prelim. I for one do not understand the apparent anger it provokes in some. No document is perfect, and if notice of some discrepancies is ok as regards AB or NavAids, I question the 'protectionist' stance of some for the BEA. We are all human here.

[Captain Bob]

BEA never mentioned AF447 was 3 NM off track ... WHY ?

It doesn't matter. Everyone does it. Parallel Track. It has nothing to do with
what happened.

[John47]

USA Today has a round up of AF447 so far - nothing new but interesting comment from John Fish
Even though ships have prowled the Atlantic in vain for several weeks in search of the jet's two crash-proof recorders, chances are still good that the recorders and the wreckage will be found, according to investigators and deep water salvage experts.
John Fish, a sonar expert who has helped recover underwater aircraft wreckage, said he's optimistic that a thorough search of the ocean bottom will help uncover the jet's remains.
"They are going to find it," Fish said. "It's just going to take longer because it is in deeper water."
The search is similar to what occurred after a South African Airways jet crashed in the Indian Ocean on Nov. 28, 1987, Fish said. Initial attempts to listen for the pingers on the recorders, which were nearly 15,000 feet deep, were futile.
But a U.S. team using sonar located the wreckage two months later. More than a year after the accident, one of the two recorders was found. The cockpit recording confirmed that the jet had been brought down by a fire.
Air France crash still a mystery - USATODAY.com (http://www.usatoday.com/news/world/2009-07-12-csiplane_N.htm)

[Dysag]

"They are going to find it, fish said"

[HazelNuts39]

RE: Captain Bob (#3536)

Quote:
It doesn't matter. Everyone does it. Parallel Track. It has nothing to do with
what happened.
/Unquote

I thought so too, but after reading about SLOP (Strategic Lateral Offset Procedure - Wikipedia, the free encyclopedia (http://en.wikipedia.org/wiki/Strategic_Lateral_Offset_Procedure)) I felt that maybe 3 NM left of track is possibly significant. With as little evidence as we have, it would seem justified to examine even this detail.

Quote from BEA Interim Rept:
Note: A position report message (AOC type) was received at 2 h 10 min 34 s, between two maintenance
messages. This can be explained by the fact that AOC messages take priority over maintenance messages
/Unquote

Considering that the first WRN message transmitted by ACARS was received at 02:10:10Z and the last position report at 02:10:34Z - I wanted to have a feel for how much time it would take to get 3 NM off track, if one assumes that the airplane is initially on track at 468 kt TAS. For example, in a relatively shallow turn (10 degrees of bank) it would take approx. 80 seconds, or 10 NM along track, and a change of heading of 33 degrees. Similar data also for steeper turns are shown in the following table:

Track deviation [NM] . 3
TAS .[kt] ............ 468
Bankangle .[deg] ...... 10 ..... 15 .... 20 ... 25 .... 30

Loadfactor .......... = 1,015 . 1,035 . 1,064 . 1,103 . 1,155
TurnRadius [NM] ......= 18,1 . 11,9 ... 8,8 ... 6,8 ... 5,5
TurnRate [deg/sec] ...= 0,412 . 0,625 . 0,850 . 1,088 . 1,348
Heading Chg [deg] ... = 33,5 .. 41,6 .. 48,9 .. 55,8 .. 62,8
Time [sec] .......... = 81,3 .. 66,5 .. 57,5 .. 51,3 .. 46,6
Dist along track [NM] = 10,0 .. 7,9 ... 6,6 ... 5,7 ... 4,9

P.S. As you might guess, I'm an engineer, not a pilot, and a newcomer to this forum. If it's irrelevant, don't get angry with me, just tell me and I'll be happy to hit the EDIT button to delete it.

regards,
HN39

[rgbrock1]

Will Fraser wrote:
Quote:
The two pilots noted turbulence in real time, not as a tx. Not that they 'reported' it, they made note of it to eachother, or FP communicated it to PNF.

Thanks for clearing that up for me. I missed that bit.nh


What? I don't understand. If the pilots noted turbulence in real time and not as a transmission, and they did not "report" it, yet they made note of it to each other.... how do we know this?

[DJ77]

from Captain Bob about offset:
It doesn't matter. Everyone does it. Parallel Track. It has nothing to do with what happened.


You seem to be unfamiliar with oceanic airspace navigation rules. Prior to the introduction of Strategic Lateral Offset Procedures (SLOP) the only authorised navigation was strictly on the cleared track. Without describing history and justification of SLOP which would be a bit out of topic, navigation is now permitted either on track, or 1 nm right of track, or 2 nm right of track. Application of SLOP does not require ATC clearance. Other rules apply for emergency situations and weather avoidance to cope with the fact that HF communications generally do not allow immediate delivery of ATC clearances.

Has the 3 nm left cross track error (XTK) reported by ACARS at 0210 something to do with what happened? Well, I would say that it depends on what happened exactly, what caused the crash.

Allowing for FM / GPS difference, rounding, ACARS transmission delays etc ... this XTK shows with certainty that AF 447 was steering off track when the AP disconnected and this was not in application of SLOP. Since I do not see why they would be doing so unintentionally or for no reason, I believe they were performing weather deviations. Whether these deviations were pertinent and appropriate I don't know at all.

Among possible hypothesis, perhaps the crew shifted their focus from weather avoidance to unreliable airspeed procedure/alarms and then inadvertently hit a convective cell and lost control. By the way, what would you qualify as the main cause of the crash in this case?

[Will Fraser]

My post has been edited to include the word 'hypotheticals'. There are more than one possible ways to interpret the crews 'decision' to allow autopilot to continue prior to auto disconnect. The release of control by auto was either 'anticipated' or a 'surprise', or something in between, and bears very heavily on what happened next, (or had happened 'already'). eg. The cause of disconnect by exceeded control limits of a/p? The cause of disconnect Unreliable a/s? If a/c was deviating for weather, was it because of turbulence at the moment, was it Radar data? Was it both escape and avoid? One strong possibility is that there was not turbulence that was a factor, and the crew was merely deviating having monitored LH, and/or Radar. That is supported by the continued use of a/p? This may also serve to focus more on an Unreliable a/s problem causing release of control by a/p?

[rgbrock1]

Surplus one wrote:

Germans? Well we all know there is just nothing arrogant about a German, now is there? Germans are more perfect than anyone. Lufthansa would never lose an airplane due to its "safety culture". That's impossible!


Surplus One: i take exception with your implied remark. I do not think Germans, as a people, are any more or less 'arrogant' than any other peoples. To subscribe to such a statement based on a superfluous media report shows some serious narrow-mindedness on your part. :*:*:*

[Hyperveloce]

02:14:20 - .1/FLR/FR0906010213 22833406AFS 1,,,,,,,FMGEC1(1CA1),INTERMITTENT
FMGEC1 (1CA1) (2 h 13)
ATA : 228334
Source : AFS
Identifiants : -
Classe 1, INTERMITTENT
Hi there.
This last FLR has not been fully explained by the BEA (it is said that this would be a transient/intermittent problem with the FMGEC, raised by the AFS, for an anomaly of less than 2.5 s).
This occurred 3 minutes and 10 sec. after the A/THR off, presumably linked to a Pitot problem, and this Pitot freezing problems did not last more than 2 or 3 min in the other known occurrences of Pitot problems: could this unexplained FLR be linked to airspeeds fluctuating back to their nominal value ?
Would it be possible to look into the Airbus TSM to get all the possible problems this FLR might point to ? (to see whether some of these are linked in any manner to airspeeds, to airspeeds out of the flight enveloppe, etc...)
Jeff

[CONF iture]

It doesn't matter. Everyone does it. Parallel Track. It has nothing to do with what happened
To state "it doesn't matter" belongs to you.
To present the simple data in a 100+ pages Preliminary Report is a duty for the official investigating team.

Now, would you present the document that allows you to adopt a Parallel Track 3 NM on the left ?

[PJ2]

GB;
The Collins radar on the AF330 fleet is highly monitored, and since it's an essential system, any detected failure would be transmitted via ACARS so that maintenance would be prepared to fix it.

Possible undetected failures would include:
*Heavy coating of ice on the radome - highly unlikely.
*Antenna plate coming detached from its pedestal. It is attached by clamps held by two small bolts that are safety-wired. That one is pretty rare, too.
All other functions, including antenna drive position, are monitored, and hence any failure would be reported.

Water ingestion into the honeycomb structure of the radome may cause false targets, but will not get large enough to matter at altitude, once frozen.
It's been asked but I don't think answered in the thread - the radar would fail with the loss of all IRS data, (which had occurred by 02:13Z). There are two transceivers. ADIRS 1 supplies tcvr 1 with ADIRS 3 as backup, and ADIRS 2 supplies tcvr 2, with ADIRS 3 as backup.

The failure messages would be:


WXR ATT (amber) : Attitude control failure.
WXR STAB (amber) : Antenna stabilization failure.

According to the AOM, these particular messages are displayed on the ND, (nav data display). According to the A340 AMM, (I am assuming a similar installation on the A330):

Re "heavily monitored", yes it is. The BITE montors the radar system itself as well as the periherals, (EFIS CTL PNL, ADIRUs & CMC - Central Maintenance Computer). All failure messages and BITE results are immediately transmitted to the CMC. The ATA failure codes are:

RADAR Transceiver....................................34-41-33
RADAR Antenna.........................................34-41-11
RADAR Control Unit....................................34-41-12
RADAR No Data From ADIRU......................34-12-34
RADAR No Data from EFIS 1 Panel..............22-81-12
RADAR No Data from EFIS 2 Panel..............22-81-12
RADAR No Data from CFDIU........................31-32-34

I think we can assume that the radar could function up until the loss of the IRU data.

PJ2

[fantom]

Now, would you present the document that allows you to adopt a Parallel Track 3 NM on the left ?

CONFijam, we have all used offset to avoid a nasty looking mass of CBs. If the AF crew were faced with nastiness to the right and took the view that an offset to the left was the better option, would you have asked them to produce the doc allowing it? Who are they going to hazard on an oceanic track by going 3nm left instead of two right?

[Will Fraser]

I think CONF iture's point is that any deviation from 'en route de vol' should have been mentioned in the report. The motivation is not known, but isn't it established deviation is PIC's discretion? Also, 3nm in 119 appears to be a straight line and gentle course, who is to say the 3nm does not represent a 60degree left turn, made in urgent fashion?

Also, my understanding is that 'present the document' means would Capt Bob produce a report of accident without the notation of course discrepancy?

[BOAC]

I think we can assume that the radar could function up until the loss of the IRU data.- I must have missed this in 3500 posts - has the loss of IR information been confirmed, then?

[Will Fraser]

Without appearance on ACARS of Wx Radar, I think it means that Radar is IRU dependent, not that IRU(s) failed.

[BOAC]

the radar would fail with the loss of all IRS data, (which had occurred by 02:13Z) - Will - I should have included this quote.

[CONF iture]

Thanks fantom, what you describe here is called weather deviation and has nothing to do with Strategic Lateral Offset Procedure SLOP to which Captain Bob was making reference but not knowing how strictly it applies.

Finally someone to consider that the AF447 crew was maybe not that blind or careless, but trying its best to avoid weather, like already did LH507 20 minutes before them, or IB6024 and AF459 12 and 37 minutes after them.

I really would have loved seeing the BEA stating that critical point instead of writing the erroneous following comment :
up to the last automatic position point, received at 2 h 10 min 35 s, the flight had followed the route indicated in the flight plan

[PJ2]

BOAC, Will;

I'm examining the fault message at 02:13:14:

"- .1/FLR/FR0906010211 34123406IR2 1,EFCS1X,IR1,IR3,,,,ADIRU2 (1FP2),HARD"

I know that the "IR1,IR3" messages are "identifiers" and that "IR2", a Class 1 fault, is the "source" of the original message. What I am searching for information on is the meaning of the IR1 & IR3 messages. I know there weren't "radar" messages as part of the 25 messages received but we know that there is a possibility of missing messages. For example, the CMC records only six "identifiers", (related systems); if there are more, they are lost. The BEA report also states that the messages are not necessarily received or displayed in the order of the equipment faults. I am pausing to consider the loss of 3 IRUs.

Will, yes, the radar is IRU-stabilized.

[WhyIsThereAir]

I really would have loved seeing the BEA stating that critical point instead of writing the erroneous following comment :


up to the last automatic position point, received at 2 h 10 min 35 s, the flight had followed the route indicated in the flight plan


The statement is not erroneous. Up to (but not including) the the last point there is no sign of deviation. At the last point there is a sign of deviation.

There, how's that for showing that the BEA is inherently evil? :ok:

[WhyIsThereAir]

"- .1/FLR/FR0906010211 34123406IR2 1,EFCS1X,IR1,IR3,,,,ADIRU2 (1FP2),HARD"

As I read the section on analyzing ACARS messages in the BEA preliminary report, I read it that "identifiers" indicate additional units that submitted messages during the 60 second fault window. These messages will have had the leading 3/4 (or was it 2/3?) digits of the message the same as the message reported. Therefore we know(?) that IR1 and IR3 also reported something that began with 34, or 341, or 3412, depending on how many leading digits were matched.

The above "analysis" is based on my memory of the pages in the BEA report which I read a couple weeks ago, so the details are fuzzy. It is also based on the assumption the BEA description is accurate; and that I interpreted it accurately. I'd encourage others to reread the BEA description, or even better, some original description in a maint. manual if you have one.

[augustusjeremy]

I'm examining the fault message at 02:13:14:

"- .1/FLR/FR0906010211 34123406IR2 1,EFCS1X,IR1,IR3,,,,ADIRU2 (1FP2),HARD"

this fault is "timestamped" at 2:11 - was received (or transmited) at 2:13

Maybe it was triggered at 2:10 and received by the CMC at 2:11.

I suppose it means that only IR2 is "faulty"... But it wasn't followed by a message like 'NAV IR2 FAULT'.

What puzzles me is that almost at the same time - or a short time before - a fault relating to ADR1 not being recognized by PRIM2 - only - was also triggered

quite a concidence - or not.

[falconer1]

we get "back to basics"...

"Air France is reviewing procedures and use of weather radar.."

based on preliminary findings..

I'd say, a good start..

Air France reviewing weather-radar use after AF447 crash: CEO (http://www.flightglobal.com/articles/2009/07/13/329562/air-france-reviewing-weather-radar-use-after-af447-crash.html)

[BOAC]

I am pausing to consider the loss of 3 IRUs. - which returns me to the (?unanswered?) question as to what effect all this MIGHT have had on ISIS attitude.

[augustusjeremy]

To be honest, if no major upset is demonstrated getting the root cause of this accident will be mission impossible - Maybe even if the CVR/FDR are found.

You would probably need the readings for each ADR and IR channel to each of the PRIMs at least to identify an electronic fault if any - something that probably isn't availabe in the A330 FDR.

I suppose that it is time to give more attention to computer systems - and specialists - when dealing with aircraft accidents - be it Airbus, Boeing or whatever a/c involved.

[Graybeard]

Right, PJ2; the WXR would annunciate failure of its attitude input with WXR ATT. SAT annunciates when the combination of pitch and roll exceed (saturate) the mechanical limits of the antenna to scan with respect to the horizon. That takes at least 45 degrees bank.

STAB, IIRC, is pilot selectable: de-selecting stabilization input.

Regardless, with no attitude input, the WXR will scan wings level, as if the plane were at 0 pitch and 0 bank.

While not annunciated, the unlikely case of the flat plate falling off the antenna pedestal will result in zero return, which would be obvious to the pilot.

As for ice on the radome, the only case I have heard was an MD-80 on approach to KLAS. The display went full red. Once at the gate, they took a half hour delay for chipping ice off the radome.

GB

[Graybeard]

Hmm, if your IRUs tumbled, and you weren't yet out of control, you could put 5 degrees downtilt, 80 mile range on the WXR, and fly by the ground return. Rough sea is a good reflector.

GB

[Dutch Bru]

There are at least some suggestions that if they would have been as versatile as that in using the wx radar they would not have been in trouble in the first place.

[mm43]

Everyone has an opinion on this and every subject under the sun. How about we take a step back from the known fact that 228 souls have been lost and a broken aircraft lies in a deep and rather inaccessible location of the North Atlantic.

Also remind ourselves that the BEA is the ultimate authority responsible for determining why this happened and how to avoid such a mishap in the future. In this regard, it is assisted by the AAIB and the NTSB, which means that there is little point in saying that the Preliminary Report or the Final Report will be other than objective. Also remember that the BEA has to deal with the jurisdictions of Brazil, Senegal and France while it attempts to objectively gather evidence. To ignore the political implications of not being diplomatic could well jeopardise its access to vital information - whether this is provided by international treaty or not.

The Preliminary Report has provided us with a number of confusing facts, but also remind ourselves that without the ACARS messages, 10 minute automatic positions and ADS-CPDLC Log On failures with DAKAR ATS - all via SATCOM, where would the speculation obvious in this thread have gone.

AVIATE - meaning to operate the aircraft with due diligence and within the parameters dictated by the current environment.
NAVIGATE - meaning to have positional awareness with respect to other aircraft, terrain, way points, navigational aids and weather, while at all times using the aids provided.
COMMUNICATE - meaning to use the radio communications facilities provided on the aircraft to receive and acknowledge ATC instructions. To pass essential navigation details to ATC as and when requested, get METAR / TAF updates, and monitor the 121.5 / 123.45 guard frequencies at all times when traversing Oceanic FIR's.

It is worth noting that the smaller word comes first and the largest last. So while AVIATE is at the apex of the triangle and COMMUNICATE at the base, I contend that without a solid base the apex of the triangle will become more acute and room for error decreases accordingly. In this regard, I believe that events missing from the base will have a very direct bearing on the least controllable events at the apex.

By all means speculate, but the result always has a beginning.

mm43

[PJ2]

GB;
Hmm, if your IRUs tumbled, and you weren't yet out of control, you could put 5 degrees downtilt, 80 mile range on the WXR, and fly by the ground return. Rough sea is a good reflector.
Yep - great idea, alongside using the GPS for speed and altitude info as some have mentioned. Buddy of mine purchased the Collins 700 radar manual last week. There's some bread-and-butter info in one of our FCOMs and it's better than it was (which was impossible not to accomplish given that there was nothing) but it's still real thin.

The "how" of this accident is the thing...Why? Loss of speed data doesnt' cause loss of control. I think there are some good notions here but I hope we'll have a more definitive answer some day - sooner, than later.

[EMIT]

On the idea of using 5 degrees radar downtilt if all IRU's are lost:

From where does the radar get information of the whereabouts of the horizon?
(info that is needed to determine where 5 degrees below the horizon is).

Uhm, from an IRU perhaps?

[Razoray]

mm43

By all means speculate, but the result always has a beginning.

Excellent post.
Getting to the beginning and where and why this happend is the key. The clues, I believe are in the ACARS messages. One thing that makes them so confusing is that they are not in sequential order.

This is an intense mystery that must be solved. And it seems it will have to be solved w/o the CVR/FDR.........

[funfly]

As has been said before, GPS cannot give you airspeed - it gives you ground speed. It's airspeed that makes aeroplanes fly.
KISS - there can be nothing more basic and therefore reliable for measuring airspeed than an air pressure monitor sticking out into the air itself (pitot tube). The only thing that could affect this would be to obstruct the air to it. Apart from a physical object (bird?) which would possibly only effect one pitot then ice seems to be the only other thing that can cause obstruction.
As a pilot of smaller aircraft, I cannot believe that in this day and age it is beyond the wit of engineers to be able to eliminate any chance of ice build up on a pitot head of a large aircraft and for there to be a doubt about a pitot being able to resist ice (as seems to be the case) seems unbelievable considering the importance of the information from this item.

[Graybeard]

Emit On the idea of using 5 degrees radar downtilt if all IRU's are lost:

From where does the radar get information of the whereabouts of the horizon?
(info that is needed to determine where 5 degrees below the horizon is).

My prior post, apparently unclearly, stated if IRU input to the WXR is lost, the WXR will continue to paint, but with reference to the fuselage: 0 pitch and 0 bank. The 5 degree downtilt I suggested would be in reference to the longitudinal and horizontal axes of the plane, rather than the IRU reported horizon.

Bank right, and you get more ground return on the right, and less on the left. Same for nose up/down.

I realize now I made a mistake in terminology on that post, unrelated to this question, so I will edit.

GB

[mm43]

@ JuggleDan

Thus, I certainly understand the BEA report as saying that all position reports were sent from the a/c nominal route, up to and including the last message at 0210.I am not about to dispute your technical French definition of the meaning attributed by the BEA to the 02:10:34 final position, but I'd rather suggest the BEA has further data surrounding the basic position, e.g. flight level etc.. which they have deemed not relevant or desirable to be released in the Preliminary Report. In this respect, I submit that the AAIB and NTSB concurred with the wording used both in the prime report and the translated English version.

The 10 minute position reports (irrespective of interpretations around the accuracy associated with them) placed the a/c on the UN873 center line and the GS over the last 10 minutes (463KT) was only 1 knot slower than the earlier period. Without the FDR/CVR, 'when' the deviation occurred is purely speculative.

mm43

[DJ77]

"- .1/FLR/FR0906010211 34123406IR2 1,EFCS1X,IR1,IR3,,,,ADIRU2 (1FP2),HARD"


From the description of FLR messages in the BEA report, 4 systems identified an IR2 fault during the 1 min correlation window that opened when the first system detected the fault.

IR2 got priority in the message construct because it detected an internal fault.

Because of the X after EFCS1 in the message, EFCS1 only detected a class 2 fault (crew awareness not required).

IR1 and IR3 also detected the IR2 class 1 fault (they were not faulty themselves).

The exact nature of the fault is unknown.

No other system reported an IR2 fault during the correlation window, since the remaining 3 identifier fields separated by comas are empty. If an IR2 fault was detected afer the end of the correlation window the CMC would not have it reported.

We cannot deduce from the message what system detected the fault first.

Incidentally, since IR1 and IR3 were able to detect an IR2 fault, it appears there is some cross talk between ADIRUs undocumented in the FCOM. A few details related to the QF072 accident also suggest this.

[PJ2]

funfly;
As has been said before, GPS cannot give you airspeed - it gives you ground speed. It's airspeed that makes aeroplanes fly.
Yes, thank you.

However, should there be no speed information at all due to these failures, (I have had one such failure on the 767), with a stall speed around 200kts for a 330 of around 200k kgs at 350, (that's a guess - my charts only go to FL200 - anyone have anything higher?), keeping 350 to 500kts on the GPS (depending of course on wind !), will provide a rough basis upon which to fine-tune the power and attitude and should keep us safely in the air until we can sort it out.

For those aircraft without GPS, the Flight With Unreliable Airspeed checklist, of which there are several memory items regarding pitch and power, will tide one over until better times. As the autopilot and autothrust disconnect, the key is to leave everything alone. So often, we feel we must "do" something, when the best action in some cases is to do nothing, for a short period of time. The airplane was flying fine before the loss so the current pitch and power would be a good starting point from which to fine tune things. The challenge in heavy turbulence, mountain wave or if one is caught inside a thunderstorm is to let the airplane ride vertically while maintaining pitch and power - too fast is far better than too slow- pitch is key.

Sure is easy to say it and type it in the comfort of one's home, isn't it?....

DJ77;
Thank you for the information. Quick question - is it the "x" that designates a Class 2 message? My AMM and the BEA report suggest it's an asterisk, or are we talking apples vs oranges? Tx.

[24victor]

GB
With your radar tilt and a whiskey compass they'd have had primary and secondary horizons :8 ....... Oh well.

On a different subject, I note the NTSB prelim on the NorthWest incident came and went without much in the way of comment.

http://www.ntsb.gov/ntsb/GenPDF.asp?id=DCA09IA064&rpt=p

A while back someone flamed the crew's statement that the cockpit temperature and humidity had suddenly risen although I've heard anecdotal tales over the years that reference this phenomenon in the tropics in the upper FL's, albeit in aircraft without the layered automation to make it a significant issue.

Rgds.

24V

[JD-EE]

You are in an aircraft just currently in a really bad patch. (At about 0209, say.)

You have lost all three pitot tube inputs leading to all three IRU's declaring sick. You are following the manual and flying attitude and throttle. And you are significantly but not dramatically closer to landing in Brazil than any other place.

Does the pilot try to fly on to Dakar going through the rest of the bad patch essentially running ALT 2 all the way or does the pilot try to fly through the storm, again, to get to Brazil running ALT 2 all the way?

Either way would you want to get down to warmer air to get the pitot tube function back?

Is there a standard procedure or is it the pilot's call? Might they have elected for the devil they knew than the devil they didn't know?

JD-EE

[LeandroSecundo]

Hi,

They had to do that all, then produce a report and eventually have it validated by any number of politics!

Are you insinuating the BEA board is not independent ?
BTW ... I don't too much worry about what is in this preliminary report .. but instead I worry about what is not in .. EG .. some recommandations as required by the ICAO.

Bye.

[WhyIsThereAir]

- If you don't give any further precision, as it is the case here, it most often means that the end-point is included
- Thus, I certainly understand the BEA report as saying that all position reports were sent from the a/c nominal route, up to and including the last message at 0210 .


Yes. And the same holds in English: it is generally assumed that the final point is included if it has not been specified one way or the other. So those that want to lambast the BEA for 'lying about the deviation' have convention on their side in both languages.

But note that this is only convention, and possibly that paragraph was originally written by some math professor who is used to thinking in terms of excluded endpoints (mathematically it is often common to specify a range as including the initial value and NOT including the final value). So the person writing it might not realize that it is ambiguous.

I think more likely though, from the very sketchy coverage of almost everything in the report, that we should consider this to be a preliminary report, and that the BEA did not have all of its ducks in a row when it wrote it; indeed, they didn't even have a count of how many ducks they had to line up. There are other places in the report where there is bad grammar in both French and English, and the best explanation for the bad grammar was that the sentence was changed at the last minute, probably after the report had already been translated.

[crippen]

From SLF newbie. Having read this whole thread twice and being impressed by all the thoughts and skills brought to bear on this terrible tragedy,it appears not to have reached any serious conclusion. However,reading up on GPS in Wikepedia in the 'Basic Concept of GPS' paragraph the article states "Some GPS Receivers may use Additional clues or assumptions (such as using the last known altitude,dead reckoning,inertial navigation,.......)to give a degraded position when fewer than four satellites are visible.(see notes) end of quote. Given that the plane might have been in trouble and descending in various attitudes at the last ACAR transmission,could this position be degraded,and if so does the loss of height vary the position calculated by the GPS unit./ Sorry for the lack of technical termanology,I am not a flier only a passenger.

[augustusjeremy]

You have lost all three pitot tube inputs leading to all three IRU's declaring sick.

Apologies If I am wrong, but pitot tubes provide ADR data, not Inertial Reference Data.

Unless a very strange comparison algorithm validates Inertial Data from Air Data - and not just the opposed - blocked pitot tubes cannot generate IR faults by themselves.

There is no clear indication of loss of all IRUs in the AF447 case.

[BOAC]

You have lost all three pitot tube inputs leading to all three IRU's declaring sick. - where do you get this info from?

Regarding route decisions, I asked the same at #1926 but received no answer.

[IcePack]

"rolled back to 60kts. The auto pilot and auto *
throttles disengaged. The Master Warning and Master Caution flashed, *
and the sounds of chirps and clicks letting us know these things *
were happening."

Sorry guys/gals it would not be ethical for me to post the e/mail, but the above is a quote from a very recent airspeed roll back on a 330. (Quoted previously on this thread)
So it would seem that airspeed failures at height are still happening. What I'd like to know is what Airbus are doing about it! In this case the failure messages were remarkedly similar to AF.

[DJ77]

PJ2

CFR messages seem to be built using a limited character set that does not include an asterisk. Instead they use an "X" after the system acronym. Same messages in a PFR appear modified with a "*" before the acronym.

[HazelNuts39]

RE: PJ2 (#3572) : agreed also

RE: funfly (P.178 #3567)

Quote:
GPS cannot give you airspeed - it gives you ground speed. It's airspeed that makes aeroplanes fly.
KISS - there can be nothing more basic and therefore reliable for measuring airspeed than an air pressure monitor sticking out into the air itself (pitot tube).
/Unquote

Agreed entirely. However, I would like to add that there are other means of measuring airspeed. Before going into that, let me expose my understanding of the problem with pitot tubes. They need to be ventilated by drain holes to get rid of water or (melted) ice particles. The problem arises when the drain holes are blocked. Let me therefore explain two schemes which don't need drain holes.

As an ex-glider pilot I recall that sailplanes used to have a venturi probe. Whereas the pitot pressure is total pressure equal to static pressure plus dynamic pressure, the pressure measured with such a venturi is static MINUS dynamic pressure. In sailplanes that pressure is fed to a pneumatic rate-of-climb indicator (variometer) which then indicates variation of total energy, that is altitude corrected for airspeed changes. Since the air passes right through the venturi, it doesn't need drains.

Another scheme which I have come across was envisaged to replace the AoA vane (I don't recall why someone would want to do that, except that it had to do with icing and/or ice detectors). It consisted of a small cylinder projecting into the airstream with 2 pressure holes 120 degrees apart at the 1 o'clock and 5 o'clock positions (airflow coming from 9 o'clock). A servomotor would rotate the cylinder about its axis until the pressures at the 1 o'clock and 5 o'clock positions are equal, which means that the 9 o'clock position is the stagnation point indicating the direction of the local airflow, just as an AoA vane would do. I am writing this from distant memory so that details could well be different. The point is that one could measure a pressure that is representative of dynamic pressure on the surface of a suitably shaped body at a pressure hole where the airflow is flowing past it rather than into the hole.

However, just keep in mind that any scheme to replace pitots may well import new problems that are not easy to solve.

regards,
HN39

[HazelNuts39]

RE: augustusjeremy (#3578)

Quote:
Unless a very strange comparison algorithm validates Inertial Data from Air Data
/Unquote

A functional diagram showing the relations between ADIRUs and PRIMs on this forum many pages ago showed an arrow within each of the three ADIRUs from ADR to IRU, indicating that the IRU receives data from the ADR. I've been puzzled by this ever since. Does anyone know why this is so?

regards,
HN39

[vapilot2004]

CFR messages seem to be built using a limited character set that does not include an asterisk. Instead they use an "X" after the system acronym. Same messages in a PFR appear modified with a "*" before the acronym.

The limitation is imposed by ACARS.

[rgbrock1]

Lurking SLF here again. Is my understanding correct in that when an aircraft such as the Airbus 3xx-series enters alternate, or direct, law that in so doing the aircraft is being "handed over" to the flying crew in a less than desirable state? I know I'm probably simplifying it but I think my question should be obvious in intent.
Thanks.

[rgbrock1]

Quote:
Are you insinuating the BEA board is not independent ?
/unquote

Probably no more or no less independent than any other government entity anywhere else in the world. They all have an agenda.

[Will Fraser]

Given that the recurring problem vis a vis Unreliable a/s is a suspect in AF447, how is the pitot problem continuing to carry weight? Each occurrence of anomalous readings means frozen pitots? I suggest there is more to the story besides rampant heated pitot failure?? Other types don't present the same Pitot issues. Pitot/airframe mis-match?? Power supply issues? It is getting mathematically difficult to sustain the three frozen pitot postulate.

There was an AD late last year that addressed Prim reset after engine failure test in sim. At failure, a/c pitched to zero, disallowing pilot input, a separate power supply was installed on separate bus bar, as the a/c was not pitch responsive while computer reset?

Will

[Peter H]

rgbrock1 (http://www.pprune.org/members/20268-rgbrock1)

I'm a software engineer rather than a pilot, so a s/w based answer.

On handover the pilot may be provided with a less functional plane than normal.

However, this is because the s/w has 'seen' events that [its designers feel] prevents it from reliably providing the 'lost' functionality.

So your wording of the question was very pertinent. Is it 'more desirable' to have full [but untrustworthy] functionality or a more limited [but trustworthy] functionality? AB have naturally gone for the latter.

Regards, Peter

[AstraMike]

I dont suppose they were in the midst of negotiating some very impressive looking weather at the time?

[AstraMike]

Yes, but both schemes are subject to icing, particularly a venturi and thus no great improvement.

[wes_wall]

rolled back to 60kts. The auto pilot and auto *
throttles disengaged. The Master Warning and Master Caution flashed, *
and the sounds of chirps and clicks letting us know these things *
were happening

Can you provide any other details? When, where, source, under what circumstances?

[DozyWannabe]

rgbrock1:
Is my understanding correct in that when an aircraft such as the Airbus 3xx-series enters alternate, or direct, law that in so doing the aircraft is being "handed over" to the flying crew in a less than desirable state?
In fact, you've got slightly muddled here.

As I understand it, when an Airbus A3(2|3|4|8)0 degrades from Normal to either Alternate, thence to Direct Law you're talking about the FBW flight envelope protections.

When you're talking about an aircraft "handing over" control to the crew you're talking about the FMS, aka the autopilot, which is a completely different beast and one which is not restricted to Airbus - incidents and accidents involving an autopilot reaching its limits and handing a barely-controllable aircraft back to the crew include the China Airlines 747SP over San Francisco, and the American Eagle ATR at Roselawn.

Will Fraser:
There was an AD late last year that addressed Prim reset after engine failure test in sim. At failure, a/c pitched to zero, disallowing pilot input, a separate power supply was installed on separate bus bar, as the a/c was not pitch responsive while computer reset?
Not sure what you're driving at here... Was this problem specific to the sim or an issue with the actual aircraft?

[Will Fraser]

DozyWannabe

AD 2008-0010R1 07/October/2008

Actual malfunction reproduced in Sim, followed by AD by EASA.

Not necessarily pertaining to Unreliable a/s, but a power supply/controls issue.

[HazelNuts39]

RE: AstraMike (#3590)
Quot:
Yes, but both schemes are subject to icing
/unquote

Thank you, I am well aware of that. Any probe needs heating for ice protection. What I don't understand (can anybody help?) that the heating of the Thales pitots does apparently not prevent icing of the drains?

regards,
HN39

[Rockhound]

No one has responded to Rgbrock1's question (post 3539), which seems very valid to me. Where did this idea that the crew reported experiencing turbulence originate? I cannot find any reference to such an observation in the BEA preliminary report.
Rockhound

[TeachMe]

In response to Crippen on needing four satellites and the plane's position be degraded without that.

It would be highly unlikely to impossible to not get four satellites in sight from a plane in the middle of the ocean. They only time that tends to happen is in mountainous areas where much of the sky is blocked by the mountains.

TME

[Will Fraser]

Rockhound

Read 3541. "hypotheticals".

Will

[DJ77]

HN39:

What I don't understand (can anybody help?) that the heating of the Thales pitots does apparently not prevent icing of the drains?



I think the problem with these pitot probes is not icing of the drainage path but pollution by either corrosion or dust reducing drainage efficiency or perhaps a design which satisfies current certification tests but still cannot drain fast enough the water or melted ice ingested in some high Total Water Content clouds.

DJ

[Will Fraser]

Relative to BEA "independence". All Govt. agencies are interdependent.

What is crucial is NOT independence, but objectivity. Major work is done by AB and others here. Analysis, both engineering and operational, is conducted by interests who have an abiding interest in the outcome of the investigation. This is a given, and not disqualifying; they have the records and the expertise. What is very important is the final product produced by the Agency.

DJ77 re: your last post. Any Pitot system that melted ice at the orifice only to send it to a drain to refreeze would be laughed at, it is as you say.
I don't think the Pitots are framed correctly here, as I have said, to pin all Ua/s on freezing pitots (three at a time), stretches credulity.

Will

[crippen]

Thank you Teachme for your reply about the GPS 'seeing' at least 4 Sats.Would the GPS cope with a very unsteady platform,ie spinning and pitching hence losing sats and gaining others?

[Hyperveloce]

Yes, if the GPS reception antenna is accelerating relatively to the line of sight of the emitting satellite (being tracked, with others), the demodulation of the GPS spread spectrum signal (hence the SNR) can be further degraded.
Jeff

[rgbrock1]

All,

I fully realize now after reading this board that the pitot issues experienced by AF447 - alluded to by the ACARS messages, statements by the BEA and, if I'm not mistaken, in the BEA prelim. I also realize that pitot issues alone may, or may not have been the initiator of the disaster which obviously befell this flight.

I am not an aviator of any type. I am however a systems engineer by profession so I do have intimate knowledge of various computer systems.
(None to do with aviation though.)

I have seen pitot issues being raised as the cause of some unpleasantness in several other flights, one or two of which have been alluded to on this board.

It seems to me that in this day and age of rapid advances in so many aspects of technology (some of which are astounding) that a more efficient pitot system, or even replacement by new technology, can not be implemented. From my understanding, pitot tube technology dates back to the 19th century and were used on ships? So in two centuries we have not been able to bring this specific technology any further? I find that rather difficult to accept. I understand that to date no commercial airliner has been lost solely due to pitot issues. But taking into account some of the more harrowing reports of these issues isn't it about time some pro-activity be undertaken in perfecting these vital instruments, or even replace them with new technology?????

[Graybeard]

This topic is covered in Techlog, where this question belongs, I believe.

[rgbrock1]

You are correct Greybeard. I'm looking at the pitot system alternative thread now. I should have looked first instead of posting. Mea culpa.
I'm sure the mods will see to it that my posting is appropriately deep-sixed!
Thanks.

[Graybeard]

GPS reception is not a factor in this accident. Dual or triple GPS are fed into the Fright Management System, along with triple IRU acceleration derived data. Planes crossed the oceans long before GPS with only Inertial Navigation Systems, and can still do it. In fact, the IRU provides a flywheel effect at critical times, as it is based on accelerations, whereas GPS is positionally derived, subtracting where you are from where you were.

Juneau Alaska has a particularly difficult approach and go around requirement. GPS and ground based navigation sensors are supplemented by IRU to provide that flywheel for navigating safely in the go around.

GB

[JD-EE]

hyperveloce,

GPS reception
Yes, if the GPS reception antenna is accelerating relatively to the line of sight of the emitting satellite (being tracked, with others), the demodulation of the GPS spread spectrum signal (hence the SNR) can be further degraded.
Jeff
Not on any GPS set I worked on. That level of tracking, particularly with an aided GPS, is built into the beasts. An extreme example of aiding performance is in dog fights. The GPS sets lose lock only briefly and bring themselves right back in using the aiding as a clue for likely loop errors.

Military sets do have a slight advantage due to tracking the higher rate P (or Y) code channel. But they need it for jets that can pull 7 G moves. A relatively normal commercial plane's accelerations are not going to give any decent GPS set any particular problem. This is what high order tracking loops are for.

Of course, if the IRU aiding is lost then GPS can become sketchy until the aircraft dynamics settle down. Then it becomes as accurate as before.

Are you, perhaps, referring to a different problem GPS has? It tells you where you were and when you were there. It does not tell you precisely where you are now if you are moving or accelerating. Of course, in usual cases this error is smaller than you need to consider.

JD-EE

[Robin42]

AD 2008-0010R1 07/October/2008
Actual malfunction reproduced in Sim, followed by AD by EASA.
Not necessarily pertaining to Unreliable a/s, but a power supply/controls issue. (Bold mine)

Will, I too saw this record when sifting through the Airworthiness Directives re A330 in 2009 and 2008. It's amazing what remote constellations else actually do happen. E.g. an untight hydraulic line spraying the Avionic bay, leading (during approach) to ECAM warnings due to rotten contacts, and even the risk of fire.


On A330/A340 aircraft, the Flight Control Primary Computer 2 (FCPC2 and FCPC3 are supplied with power from the 2PP bus bar. Electrical transients on the 2PP bus bar occur, in particular during engine n.2 failure on A330 aircraft or engine n.3 failure on A340 aircraft. Such electrical transients lead to a FCPC2 reset. FCPC3 reset does not occur thanks to the introduction of second electrical power supply to FCPC3 from 1PP bus bar associated to Electrical Contactor Management Unit (ECMU) standard 5.

During the evaluation of specific engine failure cases at take-off on Airbus flight simulators, it has been evidenced that with FCPC1 inoperative, in the worst case, when FCPC2 and FCPC3 resets occur during rotation at take off, a transient loss of elevator control associated with a temporary incorrect flight control law reconfiguration could occur. It leads to a movement of the elevators to the zero position, which induces a pitch down movement instead of a pitch up movement needed to lift off. In addition, it leads to a limitation of the pilot authority in pitch axis and limits the capacity to counter the pitch down movement during this flight phase Source: http://www2.lba.de/dokumente/lta/2008/20080091.pdf

This made me to think about why Air France proposed that a lightning strike might have brought down the plane in the first place. Shortly afterwards they even speculated about a general power failure. I still think this was a honest statement originating from their maintenance engineers pondering the ACARS pattern.

The magnetic pulse of a lightning current certainly would have induced spikes and glitches on all sort of electrical cables. But, as it appears, Airbus also thinks about other, yet unexplained possibilities for transients on power lines: "Electrical transients on the 2PP bus bar occur, in particular during engine n.2 failure on A330" (my italics)

The bunch of still unexplained ACARS messages like triple pitot, IR and transient FMGEC1 faults triggered these thoughts: A power down on a specific bus would have triggered specific ECAM/ACARS messages in the first place. Probably not so for spikes and glitches. Due to their HF nature, power spikes do spread, and are also very difficult to track down in the field. NOTE: I'm NOT saying this is what happened. But it could also explain these nose-down events tied to ADIRUs manufactured by a different company -- if you assume the device would have been operated temporarily outside it's electrical specification.

[CONF iture]

WhyIsThereAir,

When the BEA is claiming transparency, it has to clearly put some basic data on paper :
“The last automatic report was 3 NM left of the airway”
One single line ... but not playing a childish game on the choice of words.

By the way, there is no doubt in the original version, which is in French, that the BEA had no intention whatsoever to bring forward any kind of deviation for that last automatic position report !
jusqu’au dernier point de position automatique, reçu à 2 h 10 min 34, le vol s’est déroulé sur la route prévue dans le plan de vol

[Hyperveloce]

JD-EE,
You are probably refering to GPS tracking loops aided by an inertial reference (higher resistance to acceleration, jamming, better/faster reacquisition) like for military applications. Is it the case on the Airbus A330 ?
Jeff

[Graybeard]

Robin42, there is also the TCAS Fail report, which BEA could not explain.

GB

[Robin42]

Robin42, there is also the TCAS Fail report, which BEA could not explain.


Yes, and the rudder limiter case, which otherwise had been tied to the missing airspeed, as it would be reasonable also for TCAS, assuming it needs at least altitude information, normally both provided by the ADIRUs.

But, as I understand your point, there are several electrical busses, and without a circuit diagram at hands you would not assume you had more HF than power on all these different circuits.

[mm43]

CONF iture,

By the way, there is no doubt in the original version, which is in French, that the BEA had no intention whatsoever to bring forward any kind of deviation for that last automatic position report !In an earlier post directed at JuggleDan I said,

I am not about to dispute your technical French definition of the meaning attributed by the BEA to the 02:10:34 final position, but I'd rather suggest the BEA has further data surrounding the basic position, e.g. flight level etc.. which they have deemed not relevant or desirable to be released in the Preliminary Report. In this respect, I submit that the AAIB and NTSB concurred with the wording used both in the prime report and the translated English version.

The position ultimately needs to be in 3D, and in this respect it would be reasonable to assume that AF447's flight level was other than FL350 at 02:10:34. As mentioned in the above quote, I am sure the BEA had the concurrence of both the AAIB and NTSB when using the "until" wording. Open to interpretation either way and leaving room to manoeuver. Or did no one actually see the potential implications of what was published?

mm43

[Graybeard]

Sorry, loss of airspeed does not kill TCAS. Altitude is a separate ADR output from airspeed.

A lightning strike could damage the upper TCAS antenna. I've been involved with a few lightning strike events, and don't remember any causing a power surge on the electrical busses. They will damage poorly bonded antennas, however.

[Will Fraser]

Given the deviation, what was the reported altitude at 0210:34??

If Prim1 was inop at some stage in the final minutes would the HS "lockout" obtain, given EASA's conclusion that reset involved no Pitch authority, irrespective of 'Load' prot.??? Had the a/c entered a dive at 0210, established descent, then encountered only stab trim to pullout, to find a serious lack of Pitch authority while waiting for reset??? I.E. "0" HS 'set'.

(This may not derive if AD was complied with, unless PRIM1 was inop on launch, MEL). Either way, if PRIM1 was inop and what followed was a second computer trip, this HS lock explains a possible unrecoverable descent, perhaps with parts leaving the airframe closer to 0214:30.

[Hyperveloce]

How did it occur since this plane can perfectly be flown using pich & thrust ?

I wonder whether we should add hypothesis to the events already known like brutal turbulences (see the wind measures of the AMDAR flight at FL325), up/downdrafts, engines loss, inertial/artificial horizon loss, mid air break up, etc...
I see that a sudden AP disengagement as a consequence of problems already unfolding (Pitot freezing, icing of the airframe,...) can initiate a drift (altitude, roll/bank,...) that you have to notice/detect & control switfly at high altitude. It may be in a context of stress/frustration/tasks overload due to instrumental ambiguities/failures hard to understand and to isolate, multiples sound/visual alarms, new tasks and procedures to implement, loss of flight assistances possibly for the rest of the flight... In a context of very poor visibility & possible spatial disorientation (despite the artificial horizon), shared/split attention, etc...
I see that all these Pitot symptoms may go hand in hand with stall alarms like in the Air Caraïbe case, 30 sec after the AP disengagement (possibly overspeed alarms in other cases) and that pilots are instructed to take them in account and react accordingly. Meaning that this plane was not piloted according to a normal pich a thrust procedure that would otherwise have done it ?
It may also be that the first copilot (37 yrs) was taking his duty, possibly replacing the Captain in this time frame (around 02:00Z).
If you cumulate these adverse elements, wouldn't it be already enough to put a crew in a dire strait ? Does a high altitude loss of control require more hypothesis ?
Jeff

[Will Fraser]

Jeff

My guess is that at no time would copilot2 be PIC, which means PIC would either be Captain, or F/O, both flying in their respective assigned seats, LH (Capt.) or RH (F/O). Is this correct?

[Hyperveloce]

Will,
Here is a contribution by Pihero on the Airliner.net forum

The Case of the Cruise Crew

Task attributions are generally well defined in a modern flight-deck: One pilot flies and communicates, the other manages the systems and the check-lists..
There is some uncertainty, though, as to who the “Pilot Flying” was. With the Captain in the LH seat, either one could have been it. During the Captain’s rest, only the most experienced co-pilot could have been PF and only from the RH seat, the second co-pilot being in charge of the systems, radio-com and navigation from the left-hand seat.
It was then up to him to keep an update of the weather at alternates, keep a sharp eye on the ETPs between the alternates in an ETOPS environment, fuel siytuation, including the aircraft balance, help with the radar… It’s also his job to compute at every reporting waypoint engine and speed parameters for turbulence penetration, altitude capability and possible step climb…etc.,..
This is how AF captains generally manage their crews : to establish a roster, take off ½ an hour for climb and ½ hour for descent ; then divide the remainin time by three  that’s the times of rest.
So we have (hours rounded up) as per the flight plan :
22.30 Takeoff
23.00 end of climb
08.30 Descent
09.00 Landing.
So remain 9 hrs and 30 mins for rest, divided by 3 = 3 hrs and 10 mins each., minus 5 to 10 minutes for hand over, control transfer and instructions :
So we have this roster :

From T/O to 23.00Z All three in Flight Deck
From 23.00Z to 0210Z, One co-pilot resting
From 0210Z to 0520Z, F/O and SF/O on duty, Captain resting
From 0520Z to 0830Z, Second cop-pilot resting
From 0830Z to landing (ETA 0900) All three in Flight Deck.But looking at the above, that 0210Z time is striking… and you know why as it is the moment of the first of the ACARS failure messages…
Captains in general chose the middle bunk-slot as they want to be in the cockpit during the most work-loaded phases of the flight : Dense ATC, multiple clearances for level changes, direct routings, Departure and arrival procedures...etc...leaving the - comparatively - quieter and more routine phase of the cruise to his young aides.

[Hyperveloce]

The debris are currently been transfered to the Toulouse Aeronautical Test Centre (CEAT), a test center under the aegis of D.G.A. (French Weapon Procurement Agency), where a deeper analysis will hopefully provide new facts, along with the autopsy results.
Jeff

The CEAT was established in 1949. Located in the vicinity of Toulouse at Balma, CEAT occupies a site of 40 hectares. CEAT is the main European centre for ground testing of military and civil aircraft. It is gradually offering tests for other domains: land and naval systems and missiles.
DGA - Aeronautical tests (http://www.defense.gouv.fr/dga/dossiers/aeronautical_tests)

[Will Fraser]

If I may, I'll suggest a finer schedule. At 0200, all three pilots would be on the FD. It is 'changeover' and at 0200, the copilot2 is 'getting the panel' to relieve Captain. At 0210, exactly, copilot2 is taking the LS, Captain would probably hang around, to satisfy himself all was smooth, before heading aft to rest. This is no doubt on the CVR, but it isn't absolutely what we have to go on here, obviously.

Will

[JD-EE]

hyperveloce re aided GPS...

<blink><blink> They AREN'T! :uhoh:

JD-EE

[CONF iture]

Mm43,
Strongly disagree here : Would the FL be different from 350 at that last position, it is an absolute crucial information that would have to be mentioned.
We are not talking here of a mere 5 pages preliminary report but 70 pages and almost as many in annexes, where such raw data must have priority.
I really don’t see that kind of info as "irrelevant"
Also, what do you mean by "not desirable to be released" ?
Would you give an example ?

[Will Fraser]

Where some things are excluded but can be supported (deviation)
And some things are included but can't be supported (intact @ impact)
It becomes unclear what means what, and to whom.

A lower alt. at final report would indicate a descending deviation, or any of a number of other possibilities. Oh what a tangled web.....

Will

[Lightning6]

With little facts available together with a lot of speculation a "Tangled Web" is what you're going to get.

[mm43]

CONF iture;

I really don’t see that kind of info as "irrelevant"
Also, what do you mean by "not desirable to be released" ?
Would you give an example ?My reference to "not relevant or desirable" was my assumption that the BEA had not determined why the a/c was at some level other than the cleared level of FL350. If so, it could have been the reason they have chosen not to mention it. On the other-hand the a/c could have still been at FL350, and until that information is released, we will not know one way or the other. My only concern, is that it is sometimes not in the best interest of an investigation to promote something as fact, when there could be another reason for the position/FL being transmitted in the form that it presumably was, i.e. the data validity could be in questionable.

Hopefully, time will tell, but in the meantime anything said in this thread regarding the 02:10:34 position is speculation - nothing more.

mm43

[Will Fraser]

mm43

Bear with me please. The plot of the last reported position is connected to the previous reported position with a straight line. Is such a line from BEA?
Because that would mean they had determined the course between positions was straight. Fair enough, but what if the 3nm discrepancy West was a sharpish turn made in the last few miles before LRP? Do we know this? Also because I don't see how anyone would know the route, the points yes, but not the route. The originator of the straight course line has some explaining to do, whomever it was. Yes?? In other words, isn't the straight course line speculation? If not, and it is accurate, it means the crew started diverting at the previous course report.

Will

[JD-EE]

CONF iture,

Would the FL be different from 350 at that last position, it is an absolute crucial information that would have to be mentioned.

Irrelevant or not it appears there is another serious lack in the ACARS messages reports that I mentioned before. If you look at page 46 (adobe acrobat page 47) of the English report you will see the table of ACARS messages WITH the position report at 02:10:34 interspersed with the other messages. It shows latitude and longitude and no altitude.

This simple minded crazed cyberunit, therefore, draws the conclusion that for some reason unknown to rational people the altitude was left out of the ACARS messages (or somebody is hiding something.)

JD-EE

[MG23]

This simple minded crazed cyberunit, therefore, draws the conclusion that for some reason unknown to rational people the altitude was left out of the ACARS messages (or somebody is hiding something.) Why would a rational person put altitude in an ACARS message when they're paying by the byte? In normal circumstances the airline wants to know where their planes are, not how high.

[DJ77]

Will,
I don't think the Pitots are framed correctly here, as I have said, to pin all Ua/s on freezing pitots (three at a time), stretches credulity.

Fact is design probability targets and statistics don't automatically agree.

[HazelNuts39]

RE: DJ77 (#3589)

Thank you for your explanation. So its not icing, but still the only known problem with these probes (as far as I know) is the drainage, and apparently its not easy to solve.

The BEA Interim report mentions that the probes are inspected and cleaned at C-checks. Does anyone know the frequency of C-checks?

regards,
HN39

[MATELO]

For Robin42 & GB

"This made me to think about why Air France proposed that a lightning strike might have brought down the plane in the first place. Shortly afterwards they even speculated about a general power failure. I still think this was a honest statement originating from their maintenance engineers pondering the ACARS pattern"

& "
Robin42, there is also the TCAS Fail report, which BEA could not explain.

GB""

Can I just ask you guys what would happen if 447 took a lighting strike on the left had side on or neat the pitot tubes. All being well, the strike should dissipate itself around the aircraft, but could the electricty travel around the circuits in the nose rendering a few of the to reset themselves or blowing fuses. Just asking to clarify any situation that may have arisen.

[LeandroSecundo]

Hi,

C-checks. Does anyone know the frequency of C-checks?
+/- 18 months :sad:

Bye.

[takata]

@ LeandroSecundo
Check-C is every 3,000 flight hours. At A330/340 flying rate, it is flown in less than 9 months of commercial service.

S~
Olivier

[LeandroSecundo]

Hi,

Check C more:

SERVICE INFORMATION LETTER : (http://209.85.135.132/search?q=cache:VHZCGDy3m-EJ:www.iag-inc.com/premium/sil_34_084.pdf+%22SIL+34-084%22&cd=1&hl=fr&ct=clnk&gl=be&lr=lang_en)

Bye.

[Peter H]

HazelNuts39 (http://www.pprune.org/members/305001-hazelnuts39) The BEA Interim report mentions that the probes are inspected and cleaned at C-checks. Does anyone know the frequency of C-checks?

LeandroSecundo (http://www.pprune.org/members/304207-leandrosecundo) +/- 18 months http://images.ibsrv.net/ibsrv/res/src:www.pprune.org/get/images/smilies/puppy_dog_eyes.gif

There is a big difference between checking the hole in an external pitot isn't blocked and a full check of the probe system. Is there a schedule for doing the former?

Performing a full probe check can require access to inaccessible places (see attached entry from the pitot tube thread). Leaving this to C-check time seems understandable.

Regards, Peter

(permalink (http://www.pprune.org/tech-log/376881-technical-alternatives-pitot-tubes-3.html#post4993487)) JRBarrett (http://www.pprune.org/members/98247-jrbarrett)
I recently had a G-IV which had developed a serious in-flight data miscompare between the L/H and R/H air data computers. Ground testing quickly revealed a major leak in the static line feeding the #2 ADC.

We eventually found the leak to be caused by a cracked plastic B nut at the point where the static line feeding the #2 ADC connected to the R/H static port.

Fixing the actual problem took less than 30 minutes. Getting access to the point of failure however was another matter entirely. Six man-hours to remove the overhead ceiling panels in the entranceway to gain access to the tee fitting where the L/H and R/H static port lines joined together, (in order to isolate which side was leaking) - then an additional 35 man-hours of labor time to disassemble and remove the forward lav and galley, and then put it all back together afterward.

[funfly]

From this thread, one is bound to consider that the investigation has as yet not produced enough evidence to fixate on any initial cause.

However one is bound to consider two scenario's as most relevent (again from stuff on here, not the official report)

1. Ice blockage of all pitot tubes rendering the aircraft impossible to fly.
2. Aircraft entered a weather situation where it was unable to maintain flight. Pilot or Pitot?
1. is of course a technical fault, however the concensus here seems that failure of the pitot tubes could only be as a result of the aircraft being in bad weather.
Although this thread brings up some fascinating and very relevent technical information, it seems pretty obvious that the aircraft crashed because, for whatever reason, a decision was made to enter a weather system that it was not capable of dealing with - it's got to be as simple as that

[pax2908]

Not having access to the AMMs, so just a couple of questions here: flushing really means flushing of the whole pneumatic line all the way to the pressure sensor itself? So this procedure will necessarily call for a leak test, at the end?

[MROJunkie]

Hi,
intervals vary from airline to airline. As LeandroSecundo already stated for what is called simplified C-Check the interval is +/ - 18 month.

see below extract from a A330 maintenance requirement

CHECK MAX. INTERVAL
MPC
PFC Before each departure
W 14 calendar days
A 800 FH since
C 1C = 18 months
2C = 36 months
4C = 72 months
8C = 144 months
IV 1IV = 6Y = 72 month
D 1D =12Y =120month

[DJ77]

So its not icing, but still the only known problem with these probes (as far as I know) is the drainage, and apparently its not easy to solve.

HN39,

It's not icing at the drain end of the Pitot but eventually, I think the front end ice up.

The following is a surmise entirely (I'm not an engineer) but it is the way I think things happen.

When drain flow can no longer evacuate all the incoming water, the tube starts filling. At this stage, total air pressure is still transmitted through liquid water with little error and possible instability. However, while the tube is filling up, the heat flow from the de-icing elements is modified by the increasing water volume and becomes gradually unable to efficiently deice the forward tip of the tube. When the tube is clogged, pressure falls rapidly through the drain at the rear and air speed indication follows. Since the liquid water is still heated it could start boiling close to the drain and cause intermittent overspeed alarms.

[rgbrock1]

I've noted some discussion of weather-related issues (turbulence) as being a possible initiator of AF447's loss. (please note the use of possible)
I do not, in any sense, wish to start a flame war concerning the topic of
climate change/global warming. (I prefer to call it global warning.)
However, and I'm sure those who are currently still flying would be able
to support, or refute this, perhaps AF447 is just the beginning of some rather
unpleasant nastiness brought on by the forces of nature, run amok.

In the back of my mind, anyway, I continue to wonder if AF447 entered a "freak" weather system which would have been "unsurvivable" by any aircraft ie, loss of control.
I know this is all conjecture on my part and I have seen no evidence of my
supposition. However, the images I have seen of the weather systems present at around the time of AF447's flight do show some degree of nastiness. But, perhaps, these images just don't tell the whole story. Would any satellite be able to show exactly what is contained in any weather system?

In essence what I'm thinking is that perhaps AF447 is just the beginning
of other weather-related phenomena heretofore unseen, and unexperienced, because of climate change-related issues?

I'm ready for the daggers now!

[BOAC]

No 'daggers', rgb, but remember several other a/c flew through the same ITCZ that night and came out the other side intact PLUS the sat weather pictures are colour enhanced. It appears to be have been a pretty 'standard' ITCZ by all accounts.

[funfly]

lomapaseo, I appreciate your comments and have ammended as you suggested.
Of course we won't know the 'official' verdict until the final report emerges - it's been said on here many times before.
However, this is a pilot's 'rumours and news' thread thread and a lot of the contributors are very knowledgable people (don't include me in this!)

[Hyperveloce]

Since the liquid water is still heated it could start boiling close to the drain and cause intermittent overspeed alarms.

thank you for that ! this is the kind of phenomenology I was looking for, to generate overestimated "airspeeds" and false overspeed alarms without any altitude change (and/or static ports failure). Now, does it mean that under this phenomenology, airspeeds should be underestimated in a first time and then possibly overestimated (bubbles overpressure build up) if it remains in icy conditions, before it recover its normal values ?
Jeff

[rgbrock1]

BOAC. I understand. But those other aircraft did not traverse the ITCZ
at exactly the same point in time and probably not in the same point
of space. Yes, the images I have seen appear to be "normal" for weather systems in the ITCZ. But, again, do these images tell the entire story? And, can they? (I'm thinking: heretofore unseen and extremely violent, updrafts and downdrafts buffeting the aircraft beyond design limits causing, at a minimum, loss of control. Something beyond the design limits of ANY aircraft.)
I know some of my supposition may appear to be "out there" but I truly do not think it is out of the realm of possiblity as to an initiator of AF447's loss.
Again, most of this supposition is based on my thoughts on climate change.
I believe it's real and I believe it's going to get worse. And I think that AF447 is just a sign. I hope I'm wrong.

[thermalsniffer]

No one has answered the question regarding the purported crew report of "fortes turbulances" (see posts 574 and 628 and underlying source material), and that it is not mentioned in the BEA report.

A second question is if the report by the crew did happen at 2:00, why would the A/THR be on at 2:10:46??

[Razoray]

I believe it's real and I believe it's going to get worse. And I think that AF447 is just a sign. I hope I'm wrong


Global Warming is real and it will effect areas already known for turbulent weather to produce even more severe conditions. The biggest issue would be the ocean temperture increasing, hence causing storms to intensify to levels we may never have seen before. The ocean temperture in the ITCZ was about 80f, I believe, which is very warm.

With Global Warming becoming a issue, traditional routes may need to alter to avoid these weather patterns. Just my opinion......

[mm43]

Will Fraser

In other words, isn't the straight course line speculation? If not, and it is accurate, it means the crew started diverting at the previous course report.When I originally took a closer look at the BEA plot, I became aware when I blew the graphic up that the line drawn from the 0200z SATCOM position was not a continuation of the 028°T track to TASIL. As we all know now, the 02:10:34 position is actually 2.94NM left of track. We have no way of determining when this deviation took place, though one would suspect that it is related to events that commenced about a minute earlier - whether by design or as a result of is of course speculation.

I've prepared a graphic which shows the ETO times SALPU, ORARO and TASIL calculated from the 0150z ~ 0200z GS of 464KT. On passing SALPU the SBFN-SSR would have received the last squawk from AF447 - at about 254NM from the radar. The 0214z Ultimo Reporté and the now discredited 0214z ACARS position are also shown.

Somewhere in the Preliminary Report I remember seeing mention that the "Fortes Turbulence" report at 0200z didn't happen - seems it must have been media inspired.

A further update to the graphic is planned shortly which will show the circulation and likely wind shear encountered between a convective cell to the left of the track and the major cell the a/c apparently penetrated around 0209z with its center slightly to the right of track.

mm43

http://countjustonce.com/af447-lkp.jpg

[BOAC]

No one has answered the question regarding the purported crew report - I believe that posts here (certainly those I have seen in the last couple of weeks) state there is no indication that a report was actually made and the BEA interim does not mention any.

rgb - I don't think we need to worry about rogue cells in fronts at the moment. Yes the weather is going to change. Yes storms will be more intense. This front seemed pretty 'average' to me, and there is no reason (yet) to believe AF447 encountered any unusually strong turbulence. Maybe the FDR would show differently, but.................

ITCZs like this are crossed daily by 10's if not 100's of aircraft. Active cells are planned to be avoided. I remind you again that the 'colours' on the sat weather pictures are not an indication of what the weather radar would see. I believe they indicate cloud top temperatures?

[promani]

Razoray, that would be the surface temperature of the sea, I believe.
I am another newbie to post on this thread, sorry, but I have followed it daily since June 2 with deep interest. I am quiet frustrated in the fact that the search for the FDR/CVR has so far been fruitless, and also the lack of information from the investigators.
I would like to ask you guys if any of you would have any idea what structural damage would have Af447 incurred upon impact with the ocean. As it is made of composite materials, would it have disintegrated into small pieces like a glass, or would it have broken into large segments? The reason for this question is for me to try to visualise the chances of the search finding the BBs. If they are still enclosed inside the tail section, then the chances of detection would be better, at least that is my inexperienced opinion. Thanks guys, and stay and fly safe.

[Hyperveloce]

Will Fraser
I've prepared a graphic which shows the ETO times SALPU, ORARO and TASIL calculated from the 0150z ~ 0200z GS of 464KT. On passing SALPU the SBFN-SSR would have received the last squawk from AF447 - at about 254NM from the radar. The 0214z Ultimo Reporté and the now discredited 0214z ACARS position are also shown.

Strange, the speed between ORARO and the last known position at 02:10Z is much lower than between the other markers ! (435 kts versus 470 kts ?!).
Jeff

[BJ-ENG]

Bourse - French Submarines To Resume Search For Brazil Flight Boxes (http://www.easybourse.com/bourse/actualite/french-submarines-to-resume-search-for-brazil-flight-boxes-FR0000031122-701742)

Quote: "Starting next week, French submarines will attempt to track down the devices, in a second search phase lasting around a month, said a statement from the French bureau leading the crash investigation, the BEA."


I wonder why only a month?

In Johan Strumpfer's account of the search for the SA 747 in 1987, it was estimated that the total area of search would have stretched in excess of 800 square kilometres, an area that would have required in excess of six months. Ocean depth was in places over 5 km deep and very mountainous.

[Jig Peter]

@ Romani

Most of the A330 is of conventional metal construction, with composite vertical and horizontal tail surfaces, so behaviour on contact with the sea would probably (sea states and other factors unknown, of course) be similar to the A320 "Sully" event ...

[takata]

@thermalsniffer
No one has answered the question regarding the purported crew report of "fortes turbulances" (see posts 574 and 628 and underlying source material), and that it is not mentioned in the BEA report.1. no manual crew report was issued
2. an automatic position report triggered by the FMGS was issued at 0200, which should include:
- from waypoint xxxx, time and FL
- current position (lat, lon), time and FL
- to waypoint yyyy, time and FL
- next waypoint zzzz, time and FL
- Wind (heading, speed)
- SAT (deg C)
- Icing (manual field?)
- Turbulences (manual field?)
- Airspeed (kts/mach)
- Ground speed (kts)
- Vertical speed (ft/mn)
- Heading
- Track angle
3. press reports are to be considered unreliable until cross-checked with official investigations sources.

A second question is if the report by the crew did happen at 2:00, why would the A/THR be on at 2:10:46?? 1. no turbulences reported at 0200, or
2. turbulences automatically reported at 0200 but none at 0210...

@ LeandroSecundo
Sorry, you were right, I wrongly noted 2C check at 6,000 hrs when it was actually 1C (beside, this document you linked doesn't give the answer for A330 but it may be infered from A320s checks). Thx MROJunkie for the correct table.

S~
Olivier

[HazelNuts39]

RE: rgbrock1 (#3634) and BOAC (#3631)

Quote:
But, again, do these images tell the entire story?
/Unquote

No they do not. But there is verbal explanation that goes with it. While the meteorologist specialist stuff is at places difficult to follow for a layman like me, they did more than just look at those images. From what I understand, there are two sets of data that permit them to say something about the probability of extreme updrafts at the time of AF447.

The first is the vertical temperature profile of the atmosphere as obtained from radiosonde ascents on that day in the area, combined with the temperature of the seawater. Convective air currents derive their energy from the fact that the rising air is warmer and hence lighter than the surrounding air. The greater this temperature difference, the greater the speeds that upward moving airmasses are likely to attain (probably a gross oversimplification on my part). An analysis by Tim Vasques available on the Web compares these temperatures and concludes from them that extreme updrafts are unlikely.

The second is the apparent "overshoot" in the convective cells as can be seen in the satellite infrared images, as discussed in the BEA report. The rising air, after penetrating the tropopause, will reach an altitude at which its temperature is equal to that of the surrounding atmosphere. At that (equilibrium) level it no longer experiences an upward bouyancy but since it is moving at a certain speed it will not suddenly stop there, but will continue upwards, gradually losing momentum, until it stops. At that point it will be cooler than its surroundings, and that temperature can be seen in the satellite IR images looking at cloud tops. The greater the upward speed, the greater that overshoot beyond the equilibrium level, and hence the lower the temperature where it stops. That analysis, done by Meteofrance in the BEA report, also shows that there is no evidence pointing towards extreme updrafts.

I repeat that this is a layman's account, trying to help fellow-laymen understand the meteo lingo. Hope it helps.

regards,
HN39

[HarryMann]

No, unlikely a composite component would shatter-like-glass, as it's a matrix of fibres within a substrate.
Although a lot of strain-energy would be released when a component does break, that then would release the strain throughout much of the rest of the component.

The fin was not found in a thousand pieces, nor would that be expected !
One hypothesis, is that it failed due to inertial loading, when the fuse entered the water (in line of flight - read axially in this case), decelerating at a very high rate, as it's(the fin's) supporting structure progressively buried itself under water, with the fin still in free-air (little balancing drag)

No doubt BEA have already calculated the deceleration required, in round terms, to fail the fixings through forward bending in this manner... from its mass and vertical mass centroid. I would imagine this would be many tens, maybe 20 ~50g or more (but don't shoot me unless you've done some sums as well :)

[John47]

rgbrock1

Tim Vasques detailed analysis of the weather for AF447 can be found here.

Air France 447 - AFR447 - A detailed meteorological analysis - Satellite and weather data (http://www.weathergraphics.com/tim/af447/)

Crudely speaking he concludes that the weather system was not unusual for that area.

But best to read his site yourself.

[takata]

mm43:
I've prepared a graphic which shows the ETO times SALPU, ORARO and TASIL calculated from the 0150z ~ 0200z GS of 464KT. On passing SALPU the SBFN-SSR would have received the last squawk from AF447 - at about 254NM from the radar. The 0214z Ultimo Reporté and the now discredited 0214z ACARS position are also shown.

Hi,
It is nice, but:
- 0214 position doesn't exist.
- 0150 & 0200 exact positions are still yet unpublished,
If extracted from the BEA flightplan reconstitued, due to its scale, it is not possible to make any precise position within a margin of error of 3-5 NM.

So, it is actually not possible (for any one of us) to figure out if any flight plan deviation occured already before 0210, nor to compute the aircraft airspeed at all these points, including the last one.

On the other hand, the BEA should be able to do it from the vectors recieved at 0150, 0200 and 0210 (see my post above) and it should be able to verify precisely the distance actually flown (with airspeed, temp, wind, and heading) between those points, including the final vector after 0210.
This is what I found seriously missing in the preliminary report, as they only published the last position, and previously also the altitude (FL350, during June 6th conference). My opinion is that F-GZCP airspeed was already reported erroneous in this last position report.

S~
Olivier

[DJ77]

press reports are to be considered unreliable until cross-checked with official investigations sources


takata, with all due restect, it may also apply to forum reports. Frankly, I doubt the auto position report at 0200 contained all this info and the BEA did not mention it. Is it not possible that what you describe is an ADS full position report, not a simplified company one ?

[thermalsniffer]

Takata, you already answered the questions at Post 2125. Sorry for the repetition, but I missed it in my thread search.

The BEA pointed that she was flying at cruise settings Mach 0.82 @ 0210Z and autothrust was ON (self-disconnected at 0210Z); unlikely a 'fortes turbulences' mode which would be at reduced speed to Mach 0.80 and autothrust OFF (like in Air Caraibes case). See my first hypothesis post about the so-called 'fortes turbulences' report issued at 0200Z.

[takata]

DJ777:
takata, with all due restect, it may also apply to forum reports. Frankly, I doubt the auto position report at 0200 contained all this info and the BEA did not mention it. Is it not possible that what you describe is an ADS full position report, not a simplified company one ?
All those fields are automatically calculated by the FMGS and within the minimum ACARS allocated size, it is possible to send a lot more than just an altitude, position and speed. By the way, the BEA did not develop anything about the last leg auto-reports beside the maintenance ACARS. This aircraft was sending also a real time engine monitoring report, an operational report (including fuel left, etc.).
So, wait and see, I'm pretty sure they are one step beyond us in this case because they have much more data on hand than published yet. I asked someone to get the exact format used by Air France for its long haul position reports and I'll get his answer soon.

S~
Olivier

[wilyflier]

Is it possible the reduced airspeed ( calculated and shown a few posts ago) after 0200 is really due to a dogleg having been performed and track nearly regained??

[BOAC]

Wily - at this stage just about ALL things are 'possible'. Let's not start off on a wild goose-chase about groundspeeds? Please re-read takata's #3647

[wilyflier]

Sorry I was still answering Hyperveloce at# 3640 It accounts for running late

[promani]

Jig Peter and HarryMann, Tx for your input. I now feel more confident that the bbs may be found. Let's hope so. Harry fantastic calculations. I accept them as fact. Tx :)

[Will Fraser]

Hyperveloce

You cannot know the "speed" of the a/c point to point without knowing its route (Miles covered). This is where speculation gets out of control. My point is that the "line" between the last two reporting points is "straight", it may not have been, it hasn't been published if it is known. The line is itself speculation, onto which you have heaped additional (speed) speculation. What is "known" is the 3nm separation a/c from "flight path"
(per BEA).

[mm43]

takata

It is nice, but:
- 0214 position doesn't exist.
- 0150 & 0200 exact positions are still yet unpublished,
If extracted from the BEA flightplan reconstitued, due to its scale, it is not possible to make any precise position within a margin of error of 3-5 NM.As I pointed out in the accompanying text, the original 0214z so-called ACARS position first reported by the media was discredited.

I accept that the 0150 and 0200 positions have not been published, but with diligence it is possible to derive them to a reasonable degree of accuracy which assumes that the BEA used Google Earth to construct the graphic in the first place. The BEA graphic provided the first evidence that the 02:10:34 position was not on the ORARO ~ TASIL track, even before I found the French SHOM report and 2.98°N 30.59°W.

Similarly, the 0130 and 0140 positions place the a/c overhead INTOL at 01:32:32.

All that can be deduced so far is that the a/c essentially flew the UN873 center-line plus or minus the plotting error, but that irrespective of that error the 02:10:34 position was 3NM left of track. When the deviation occurred is unknown. It certainly isn't a SLOPS offset (left?).

mm43

[DJ77]

You cannot know the "speed" of the a/c point to point without knowing its route (Miles covered). This is where speculation gets out of control. My point is that the "line" between the last two reporting points is "straight", it may not have been, it hasn't been published if it is known. The line is itself speculation, onto which you have heaped additional (speed) speculation. What is "known" is the 3nm separation a/c from "flight path"
(per BEA).


Will, I fully agree with your post.:)

[ttcse]

Sorry I didn't have the time to read all the thousands of posts prior to commenting. You folks are a bit testy.Sadly odds are someone else will waltz in here in another couple weeks with the same opening assertion.

[Graybeard]

Anut and other newbies to this thread: If you can't find the time to slog though all the posts on this thread as many of us have, then please do a search of posts with a keyword that interests you. You, for example, might do searches on Vertical Stabilizer, VS and Rudder.

GB

[PJ2]

AircraftNut;

Thanks. I don't feel very welcome among all of these "experts".
Well then, ask yourself why you feel that way after just one post. A lot of people here have done a lot of work with a great deal of varied expertise and thought behind their contributions. They are no more special than anyone else except they have put in some time reading what others have said. You put the word "experts" in quotation marks. That is disrespectful to those here who, like yourself, have both a long list of qualifications and lots of experience to speak from. The contributors here who post the most useful and helpful information do their research first or they are reminded, sometimes politely even, that this is a serious discussion with some 3600 contributions which just may have touched upon the obvious items you posted. I hope this is of some help. The place is definitely worth the effort.

[thermalsniffer]

Excellent advice Graybeard and my addition to this advice is to review the thread very carefully. Even after careful searches of "fortes turbulences," I missed that my question had been asked and answered. I had a lingering concern in my lurking mind regarding the purported report by the crew of strong turbulence (not mentioned in the BEA report) and that the auto thrust was disengaged. After Takata's response, I went back and indeed I had assumed the crew report was from an original source---it was a media quote allegedly made by a BEA official.

[mm43]

Hyperveloce


Strange, the speed between ORARO and the last known position at 02:10Z is much lower than between the other markers ! (435 kts versus 470 kts ?!).The positions plotted on Google Earth are as follows:-

TASIL 4°00'18"N 29°59'24"W
LKP 2°58'48"N 30°35'24"W
ORARO 2°14'48"N 30°55'24"W
0200 1°48'36"N 31°09'06"W
0150 0°39'54"N 31°45'24"W
SALPU 0°26'36"N 31°52'36"W

Even though the Last Known Position (LKP) is timed at 02:10:34, that time is the logged time of the Ground Station / Aircraft 'handshake'. The nominal time is assumed to be 02:10:00z. AOC reports take precedence over maintenance traffic, and I presume the 0210 one was slotted into the sequence immediately following the receipt of a message being transmitted.

The positions are shown with a precision of 0.1', though it should be noted that the LKP is only accurate to 0.01° - or +/- 1824 feet (556m).

ORARO to LKP is 024.4°T x 48.3NM
02:03:50 ORARO ~ 02:10:00 LKP = 48.3[1/(6.1667 / 60)] = 469.9KT
02:03:50 ORARO ~ 02:10:34 LKP = 48.3[1/(6.7333 / 60)] = 430.4KT

I'm happy to stand corrected if anyone is able to prove the LKP time-stamp is different. The original BEA graphic shows the LKP as 0210z.

mm43

[Will Fraser]

mm43
Just two minor things, one a bit picky.

There is a last position, it is reported, not known.

You assume the route between penultimate position and ultimate position to be a straight line, that is not at all known. If there were other course "changes" between the last two points, the speed could be a great deal higher. IMO. Which is to say, not constant. Fluctuating.

Will

[mm43]

Will Fraser

You assume the route between penultimate position and ultimate position to be a straight line, that is not at all known. If there were other course "changes" between the last two points, the speed could be a great deal higher. IMONot the case at all. I was responding to a point made by Hyperveloce, and it involved the speeds he had deduced, ie. time v distances. The times were the mute point. Your point of course is valid!

You will remember that I have previously said that we know nothing about the 0210 Last Known Position - AF and the BEA know a lot more. When or if a deviation occurred and all the other facts that go to make up that 3D position have been with-held.

I am more inclined to believe it happened as a result of an upset - but even that is speculation.

mm43

[JuggleDan]

Why has crew subtle incapacitation or a deliberate malevolent act seemingly been ruled out... when we have no idea at all what was taking place in that cabin, or cockpit.Most of the reflexion going on is based on what we have: ACARS messages. I guess "crew subtle incapacitation" looks like a far fetched explanation for these messages.

[Razoray]

Actually it was speculated early on that it was a terrorist attack......famous terrorist on board or something or other.......crime fighter?

spooky spooky...some one slips into flight deck during change over..........done deal!

I dont know???

:*

[augustusjeremy]

All those warnings begin

The crew guesses by turning ADR1 off. Only PRIM2 detects the move.

The ISIS, which is fed by ADR1 input, complains.

But the culprit was IR2 which generated wrong ADR2 output.

Then... (fill in the blank)

edit: the crew guesses by turning ADR1 and IR1 off... So that there is no more comparison to generate a NAV IR2 FAULT ECAM warning, as only IR3 remains "healthy"..

[JD-EE]

GB,
Way early, they ruled out suspected terrorists, after mentioning two possible.

Did they explicitly rule out all acts of terrorism or only a terrorist bomb? I can see ruling out bomb. I am not sure ruling out terrorist is sensible unless there is an ACARS sort of message that goes out with a really simply button press by anybody in the cockpit.

JD-EE

[RetiredF4]

Quote:
Why has crew subtle incapacitation or a deliberate malevolent act seemingly been ruled out... when we have no idea at all what was taking place in that cabin, or cockpit.

Quote:
Crew incapacitation could explain why AF447 flew into the CB cell on AP rather than deviating around it, and maybe explain why they didn't respond to HF Comms.

I raised this question before , but it was probably not the apropriate time, so it got deleted.

"At what time can we confirm the last positive action of the crew in the cockpit"?

regards

[auraflyer]

I raised this question before , but it was probably not the apropriate time, so it got deleted.

"At what time can we confirm the last positive action of the crew in the cockpit"?

regards

2:01.

According to sec 1.16.2.1 of the BEA report, AF447 made 3 attempts to connect to Dakar ADS-C: at 1:33, 1:35 and 2:01. (These were rejected because of a flight plan absence or mismatch.)

[Me Myself]

Why has crew subtle incapacitation or a deliberate malevolent act seemingly been ruled out... when we have no idea at all what was taking place in that cabin, or cockpit.


Deliberate malevolent act ???????????? Like they were muslim fundamuntalists ??????????? This is Air France mate, or is it that France and Egypt are the same to you ???
As to incapacitation, you've got your answer on the post above.
Are you for real ??

[Hyperveloce]

"At what time can we confirm the last positive action of the crew in the cockpit"?


Hasn't the plane been piloted till the end ? what is the probability for a non piloted/non pilotable plane to impact the surface, on his belly, "en ligne de vol", after a 35 000 ft rapid descent ?
Jeff

[RetiredF4]

Quote:
Originally Posted by RetiredF4 http://images.ibsrv.net/ibsrv/res/src:www.pprune.org/get/images/buttons/viewpost.gif (http://www.pprune.org/rumours-news/376433-af447-now-added-preliminary-report-post5063862.html#post5063862)
"At what time can we confirm the last positive action of the crew in the cockpit"?

Hasn't the plane been piloted till the end ? what is the probability for a non piloted/non pilotable plane to impact the surface, on his belly, "en ligne de vol", after a 35 000 ft rapid descent ?
Jeff


The Question ist not what we know, but what we do not know. And we do not know yet why it happened , only how it ended. There are a lot of probabilities and possible reasons, and no one has prooved itself yet. And why should an aircraft falling out of the sky not hit with the belly first? I don´t think that from the mode of impact we can conclude the aircraft being piloted until the point of impact.

So if i understand correctly, the attempt to contact ATC at 1:33, 1:35 and 2:01 needed deliberate action of the crew at those given times, it has no automatic repeat function in case of a failed contact attempt?

Anything else know after that given times?

FR

[augustusjeremy]

Yes, Aguadalte...
You were right since the beginning.

The question is: Did they turn on ADIRU1 again ? Wouldn't it take some time to realign ? And how would it realign without trustful IR data (as the other two IRs were probably disagreeing ?)

The text you posted earlier in this forum:

RED OEB – RED OEB – RED OEB – RED OEB – RED OEB – RED OEB – RED OEB
Issued by STL
File in FCOM Vol 3
OEB N°: 74/4 DEC 08
Associated with QRH OEB PROC N°: 74/4
- This OEB covers a significant operational issue. Non-compliance with this OEB should have a significant impact on the safe operations of the aircraft. The Operators shall distribute its content to all flight crews without delay. An extract of this OEB is provided for insertion in the QRH.
- It is strongly recommended that all Operators accelerate the incorporation of all corrective Service Bulletins as soon as they are available.
SUBJECT:
IR FAILURE OR ATT FLAG ON PFD
APPLICABLE TO:
All A330 aircraft fitted with NORTHROP GRUMMAN – LITTON ADIRU
CANCELLED BY:
TBD
R
REASON FOR ISSUE 4:
RRRR
The previous OEB revision requested to de-energize the affected ADIRU if the IR and/or ADR OFF lights did not illuminate. The OEB procedure is now revised in order to recommend that the IR mode rotary selector be set to OFF in all cases in order to address all identified failure cases.
Page 1 of 5
Operations Engineering Bulletins are issued by Airbus as the need arises to quickly transmit technical and procedural information. They are distributed to all FCOM holders and to others who need advice of changes to operational information.
Information in this bulletin is recommended by Airbus but may not be approved by Airworthiness Authorities.
If the procedures contained in this OEB differ from the procedures in the AFM, the AFM remains the reference.
OEB N° Page 2 of 5
74/4
REASON FOR ISSUE:
This OEB is issued in order to provide a procedure enabling to mitigate the probability of occurrence of a sudden nose down order.
EXPLANATION:
An A330 aircraft experienced a sudden nose down order during cruise. This order was preceded by an automatic autopilot disconnection and triggering of the NAV IR 1 FAULT ECAM caution.
Investigations highlighted that at the moment of the event the ADR 1 was providing erroneous and temporary wrong parameters in a random manner. This abnormal behavior of the ADR 1 led to several consequences such as unjustified stall and over speed warnings, loss of attitude information on the Primary Flight Display, several ECAM warnings.
Among the abnormal parameters, the provided Angle of Attack value was such that the flight control computers commanded nose down movement.
The aim of the following procedure is to isolate both the IR and ADR when an IR is detected faulty in order to prevent the ADR from providing erroneous data to other aircraft systems.
TO BE CONTINUED ON NEXT PAGE
PROCEDURE:
RRRR
OEB N° Page 3 of 5
74/4
RRRR
• If all ADIRU operative before failure:
If one IR is self detected faulty or if the ATT red flag is displayed on the Captain or First Officer PFD, the affected IR and the corresponding ADR must be disconnected.
NAV IR 1(2)(3) FAULT, or ATT flag displayed on CAPT (F/O) PFD
-IR (affected) pb …………………………..…… OFF
The affected IR is the one self-detected faulty or supplying the PFD displaying the ATT red Flag.
The OFF light may not illuminate
-ADR (corresponding) pb ………..…………… OFF
ADR 1 corresponds to IR 1.
ADR 2 corresponds to IR 2.
ADR 3 corresponds to IR 3.
The OFF light may not illuminate
-IR (affected) MODE rotary selector ……….. OFF
The IR mode rotary selector is set to OFF in order to de-energize the ADIRU. The OFF lights of the IR and ADR pushbutton will go off.
Setting the IR mode rotary selector to OFF is an irreversible action. Therefore, both PF and PNF must clearly identify the corresponding selector before setting it to OFF.
• If IR 1(2) affected:
In order to prevent possible failure of ADIRU 3, the flight crew must switch the AIR DATA and the ATT HDG in the following order:
-AIR DATA SWTG ………..………..……… CAPT (F/O) ON 3
-ATT HDG SWTG ..………….…………….. CAPT (F/O) ON 3
TO BE CONTINUED ON NEXT PAGE
CAUTION
RRRRR RRRR
• If one ADIRU already disconnected before failure:
In case of dispatch with one ADIRU under MMEL or one ADIRU already disconnected in flight, and an IR failure occurs, either detected by an IR 1+2 (1+3)(2+3) FAULT or with ATT red flag displayed on CAPT or F/O PFD, the affected IR and the corresponding ADR must be disconnected.
NAV IR 1+2(1+3)(2+3) FAULT, or ATT flag displayed on CAPT (F/O) PFD
-IR (affected) pb ………………..…….………… OFF
The affected IR is the one self-detected faulty or supplying the PFD displaying the ATT red Flag.
The OFF light may not illuminate
-ADR (corresponding) pb …………………….. OFF
ADR 1 corresponds to IR 1.
ADR 2 corresponds to IR 2.
ADR 3 corresponds to IR 3.
The OFF light may not illuminate
-IR (affected) MODE rotary selector ……….. OFF
The IR mode rotary selector is set to OFF in order to de-energize the ADIRU. The OFF lights of the IR and ADR pushbutton will go off.
Setting the IR mode rotary selector to OFF is an irreversible action. Therefore, both PF and PNF must clearly identify the corresponding selector before setting it to OFF.
CAUTION
• If IR 1+2 affected:
In order to prevent possible failure of ADIRU 3, the flight crew must switch the AIR DATA and the ATT HDG in the following order:
-AIR DATA SWTG ……….………………….CAPT ON 3
-ATT HDG SWTG …….……………..………CAPT ON 3
Note: First officer can recover IR information, by using the EFIS DMC selector (copy of the opposite side).
SPD BRK ………………………………….…….. DO NOT USE
• IF CG AFT 32%:
-T TANK MODE………………….……….. FWD
F/CTL ALTN LAW (PROT LOST)
MAX SPEED ………………………..………..…. 330/.82
OEB N° Page 4 of 5
74/4
OEB N° Page 5 of 5
74/4
OEB REMINDER:
On aircraft that have the OEB reminder function, the procedures of NAV IR 1(2)(3) and NAV IR 1+2(1+3)(2+3) FAULT ECAM cautions may be flagged.
The “refer to QRH PROC” line will then be displayed instead of the procedure itself.
To flag those procedures, the following codes should be entered in the FWC OEB database.
Code WARN STS
NAV IR 1 FAULT YES NO
34/10/050/061
NAV IR 2 FAULT YES NO
34/10/060/063
NAV IR 3 FAULT YES NO
34/10/070/065
NAV IR 1+2 FAULT YES NO
34/10/020/055
NAV IR 1+3 FAULT YES NO
34/10/030/057
NAV IR 2+3 FAULT YES NO
34/10/040/059
CORRECTIVE ACTION:
Under investigation
Note: The interchangeability code, given in the Illustrated Part Catalog (IPC), indicates the conditions for interchangeability of equipment. After installation of corrective modification(s)/SB(s), if an Operator reinstalls any equipment affected by this OEB it is the Operator’s responsibility to ensure that the recommendations given in this OEB are applied again for the applicable aircraft.

Edit:

Is it really a problem with the ADIRU equipment itself ?

[takata]

augustusjeremy:
Yes, Aguadalte...
You were right since the beginning.
[...]
APPLICABLE TO:
All A330 aircraft fitted with NORTHROP GRUMMAN – LITTON ADIRU
I just wonder what in this case is convincing you to believe that an A330-200, fitted with HONEYWELL ADIRUs, and showing problems with unreliable airspeed probes freezing, would be comparable with an A330-300, fitted with NORTHROP GRUMMAN-LITTON ADIRUs, and showing AOAs spikes unfiltered?

What in AF447 ACARS sequence is similar to QF72 case?
- answer: nothing.
So, maybe its time to drop this claim for lack of any substantial or relevant fact to come with.

S~
Olivier

[augustusjeremy]

What in AF447 ACARS sequence is similar to QF72 case?
- answer: nothing.
So, maybe its time to drop this claim for lack of any substantial or relevant fact to come with.

Takata,

I cannot answer this question because I don't have QF72 ACARS sequence.

What I know is that a single IR failure in QF72 (not two or three) made the A/P disconnect and so on, similarly to the AF447 case.

And a sudden pitch down might have made them decide to turn ADIRU1 off in association with the QF72 incident.

Nothing to Blame specifically. I just wanted to discuss a possibility and was a little provocative to get some feedback. :ok:

[Will Fraser]

augustusjeremy/takata/PJ2

AD 2008-0010R1 07/October/2008 EASA

Be careful what you turn off, lest nose drop.

[PJ2]

augustusjeremy;
The question is: Did they turn on ADIRU1 again ? Wouldn't it take some time to realign ? And how would it realign without trustful IR data (as the other two IRs were probably disagreeing ?)[my bolding]

It should be clearly understood by anyone examining this course of action, that once the rotary switch is turned OFF, the IRS cannot be re-aligned without landing the aircraft. That fact has been understood since 1972 when INS first entered airline service.

The irreversibility of this action is noted in the OEB thus:
• If all ADIRU operative before failure:
If one IR is self detected faulty or if the ATT red flag is displayed on the Captain or First Officer PFD, the affected IR and the corresponding ADR must be disconnected.
NAV IR 1(2)(3) FAULT, or ATT flag displayed on CAPT (F/O) PFD
-IR (affected) pb …………………………..…… OFF
The affected IR is the one self-detected faulty or supplying the PFD displaying the ATT red Flag.
The OFF light may not illuminate
-ADR (corresponding) pb ………..…………… OFF
ADR 1 corresponds to IR 1.
ADR 2 corresponds to IR 2.
ADR 3 corresponds to IR 3.
The OFF light may not illuminate
-IR (affected) MODE rotary selector ……….. OFF
The IR mode rotary selector is set to OFF in order to de-energize the ADIRU. The OFF lights of the IR and ADR pushbutton will go off.
Setting the IR mode rotary selector to OFF is an irreversible action. Therefore, both PF and PNF must clearly identify the corresponding selector before setting it to OFF.
• If IR 1(2) affected:
In order to prevent possible failure of ADIRU 3, the flight crew must switch the AIR DATA and the ATT HDG in the following order:
-AIR DATA SWTG ………..………..……… CAPT (F/O) ON 3
-ATT HDG SWTG ..………….…………….. CAPT (F/O) ON 3
Once the rotary switch is turned OFF, all attitude information from that ADIRU is lost. Depending upon faults, one may lose all attitude information (as per the cautions expressed in the OEB) if one does not strictly follow this OEB.

Needless to say, one is extremely careful and deliberate in taking these actions. One does not execute these actions swiftly or without confirmation with the other crew member. One very obvious reason for caution is, the ADIRUs are not numbered sequentially from left to right, (1,2,3) but are instead numbered 1,3,2 - to represent the (schematic) correspondence of the ADIRU to the left (captain) and right (First Officer) systems and the #3 backup for either system depending upon the data switching. Among other reasons, when on IRS-ONLY or GPS Navigation, (beyond land) one never messes with the ADIRUs unless absolutely necessary.

Some will think and perhaps argue that the situation faced by the QANTAS crew, if faced by the AF447 crew (which is the case in fact being argued here), that swiftness is the order of the day due to rapidly unfolding/deteriorating circumstances. I submit that that is not the correct approach as mistakes can and will be made any emergency procedure is rushed. These are "mission-critical" systems the loss of which is a threat to the safety of the flight. At the same time, one does not dawdle but gets on with it directly - a well-trained crew that is familiar with both the system and the procedure can execute this swiftly and accurately. A crew that is hesitant because they do not know the system thoroughly may be at a disadvantage depending upon other circumstances. The QANTAS crew could not have been expected to comprehend what was going on with their aircraft.

I frankly doubt if this scenario unfolded but it is, as other scenarios are, a possibility.

takata;
What in AF447 ACARS sequence is similar to QF72 case?
- answer: nothing.
So, maybe its time to drop this claim for lack of any substantial or relevant fact to come with.
Precisely.

augustusjeremy;
I think you have the feedback you were seeking...

For orientation and understanding, the respective panels are shown below:

http://i277.photobucket.com/albums/kk76/batcave777/ADIRS_Schematic2_2009-07-16_0757-1.jpg

ADIRU control panel - Overhead Panel, top left

http://i277.photobucket.com/albums/kk76/batcave777/ADIRUpanel_2_2009-07-16_075002.jpg

Switching Panel - forward Pedestal

http://i277.photobucket.com/albums/kk76/batcave777/Switching_2009-07-16_075111-1-1.jpg

[takata]

Will Fraser:
AD 2008-0010R1 07/October/2008 EASA
Be careful what you turn off, lest nose drop. I'm not turning off anything until facts are making some leads very very unlikely. You are pointing an EASA AD concerning:
1. very specific models (without rev.), about which we don't know if F-GZCP belongs to any.
2. very specific in flight Law (re)configurations (low level go around), when FCPC1 is inop and FCPC3 not fully powered, with an engine stopped.
3. a transient pitch contol problem when pitch reset to zero.

On the other hand, (only if F-GZCP could initially develop such an issue):
a) she was at FL350;
b) she had both engine running;
c) she had AP, A/TRH, PFD off;
d) she was in Alternate Law 2;
e) she had an FCPC1 reset when a-d conditions were still valid.

Considering 1-3 vs a-e, it makes it very stretched to consider that what could have been a very serious issue at low level would precipit F-GZCP out of control from FL350 when almost none of these documented "upset" conditions were never met by her at any time.

Consequently, this "nose drop" theory is mostly an attempt to bend the facts in order to fit it, beside discarding all we know that doesn't fit very well, as plenty of "probe icing" experiences just proved that no aircraft ever had such a tendency to upset (and drop nose) in this case.

S~
Olivier

[Will Fraser]

takata

What I thought initially as well when the AD was forwarded to me.

Since it is a flight controls issue relative to the computer, and involved similarities with ACARS from 447, I'm not convinced it doesn't obtain.

See "PRIM1" specifically. Inop or selected off, the chain is interesting relative to HS movement and the possibility of overspeed leading to upset.
You are the authority here, along with PJ2. I defer.

I still await some info that proves the pitots iced. On any flight. The Thales R&R only mentions "improved" performance, not a "fix".

Will Fraser

[augustusjeremy]

e) she had an FCPC1 reset when a-d conditions were still valid.Takata,

I beg to differ. I think this is at least controversial.

And I remeber you saying something about failed engines. Maybe that's why Will Fraser is coming up with that AD now.

PJ2,

I understand your point. I thought realignment could be done in flight if you had the other two IRs ok.

Of course if you don't have any NAV information to realign from you will have to land and do it on the ground anyway.

But remember that if a pitch down ever occurred - something equally controversial - the crew would be put in extreme stress - not understanding what really was going on.

Let's also recall that between 2:11 and 2:12 (2:13 ?) zulu that night something happened that caused a brief interruption in the ACARS message transmission.

[PJ2]

Will;
See "PRIM1" specifically. Inop or selected off, the chain is interesting relative to HS movement and the possibility of overspeed leading to upset.Though some will know this, I think just to be absolutely clear for all, it is the elevator movement, not the "HS movement" which was commanded in the QANTAS upset. Although the horizontal stabilizer movement is controlled by PRIM1, it is a trim function not a flight control function. It is possible that ND trim motion occurred but not only have we no evidence to believe that, but we do not have evidence of a high-forward-speed, high-rate-of-descent impact with the sea, which an uncontained/uncorrected ND HS trim condition would almost certainly cause. (Note: HS trim does have manual reversion backup).

augustusjeremy;

Just so all are clear on IRS alignment procedures, an IRS cannot be aligned when in motion. Its alignment process uses the angular acceleration sensed by the laser gyros and a mathematical Euler pendulum (theoretical center of the earth perpendicular to the aircraft attitude at rest, if I recall - someone can correct me here, it's been a while), to "null" the only acceleration affecting attitude. The accelerations are "resolved" to a position from which the IRU calculates latitude.

I'm only trying to convey a sense of the alignment process and not a blow-by-blow detailed description, (unnecessary), and why alignment can't be done if the aircraft is at all in motion. There is no such thing as "entering one's position" or taking present positions from the other IRSs - the potential for gross errors are obvious. In fact, a good stiff breeze pushing the tail around while the aircraft is parked at the ramp will upset or delay the alignment process as will an inadvertent towing of the aircraft from one gate to another, (it's been done...!). Forgive me for droning on, but I think this technical aspectd needs to be clear.

[takata]

Will Fraser:
Since it is a flight controls issue relative to the computer, and involved similarities with ACARS from 447, I'm not convinced it doesn't obtain.All we have in AF447 case is "F/CTL PRIM 1 FAULT" @ 0213
It's a bit fast to talk about "similarities" when at this point (ALT2), pitch was only protected to g-loads, and the flight control computer only ability was to correct any pitch imput in excess off structural limits.

See "PRIM1" specifically. Inop or selected off, the chain is interesting relative to HS movement and the possibility of overspeed leading to upset.I'm trying to give credit to the engineers here, and as well, I'm trying to give credit to the pilots: what would be the pitch setting at FL350? something like +2.5 deg and, even if (very unlikely) the pitch was unexpectidly reseted to 0 (whilch system would do that in Alternate 2?), this would hardly introduce a critical upset outside the pilots skills to recover from. The speed margin would be about Mach 0.04-0.06 under overspeed, and such an overspeed would still be far from the real structural limits (more than 50+ kts). By the way, if an overspeed really occured at one point, I guess that it will show some traces on the tailfin leading edge; then, the BEA will tell us about it but even so, it won't tell us if an overspeed was due to any computer or crew action.

Last, I'm hardly an authority you have to defer in any way. We are only exchanging personnal point of views in a public forum.

augustusjeremy:
I beg to differ. I think this is at least controversial.
And I remeber you saying something about failed engines. Maybe that's why Will Fraser is coming up with that AD now.Sorry if I may have induced some of us in error, but the possibility of a (double) engine failures seems totally excluded until the ACARS stopped. And by this point, everything would be in direct law (manual) due to the limited power supply. So, it is very difficult to add any computer mis-behavoir after a situation like that would have developped.
I just don't believe in the scenario: "fast fall from FL350 in one piece and four minutes" I may be wrong, but it looks to me like a false lead to follow until we'll get much more facts to explain it.

S~
Olivier

[Hyperveloce]

Takata,
But remember that if a pitch down ever occurred - something equally controversial - the crew would be put in extreme stress - not understanding what really was going on.
Let's also recall that between 2:11 and 2:12 (2:13 ?) zulu that night something happened that caused a brief interruption in the ACARS message transmission.

Are you refering to a pich down in the Qantas fashion (under AP control), or to a nose down induced by the flying pilot (under human control) ? The Qantas nose down has probably nothing to do with the AP¨447 case (except for possible weird reactions of the AP just before it throws the towel), but for a good model, I would rather refer to the Air Caraïbe report (of great safety interest). Take the same Air Caraïbe situation, manageable using pich and thrust (or by switching off the A/THR & fixing a N1 value by applying the "severe turbulence" proc. but it doesn't seem it was the case for AF 447) and concentrating on the trajectory, but just assume that the AF 447 took the opposite decision that was made when stall alarms sounded in the cockpit (the stall alarms were ignored by the Air Caraïbe CPT, although the procedure was to take them in account). It is just about making the assumption that the AF SOPs were applied strictly in this case (in a similar context to the Air Caraïbe incident, with stall alarms): what happens next, at high altitude, if the crew attention was also focussed on the failing nav, on the SOPs implementation, on failures very difficult to understand and isolate, and on the AP reanimation ?
Jeff

[Will Fraser]

takata/PJ2

Respectfully, a trim function would have more relevance to "upset".
At 2 +/- degrees, a down trim to "zero" would have a steady and insidious
effect on stability relative to airspeeds. With a fault as addressed in the pre-AD testing, trim wouldn't be correctable, and/or noticeable to pilots or computers wrestling with Unreliable a/s data to begin with. Only a few hundred feet (Qantas) can produce a substantial increase in speed. Any established descent, left alone, could be disastrous at flight limits experienced by 447: high, hot, and heavy.

It isn't necessary to impute any other entry to the Sea than BEA's, following an overspeed recovery. Seven miles of altitude to "recover" in, (Assumed). A high speed recovery in Direct Law wouldn't cost a bit of Elevator, VS or other control surface?

[PJ2]

Will;

Yes, I agree with your comment, (as stated in the earlier post) - trim would have an enormous and initialy insidious effect if left uncontained/uncorrected. The QANTAS accident occurred through elevator deflection, however. Caught quickly, it is controllable through pilot input. I think what takata stated regarding giving credit to the engineers and pilots is the correct view and coincides with my own experience in general - it is an easily controlled set of circumstances if caught quickly - and it's what pilots do - just like the QANTAS guys did. It's just that I don't think that's what occurred here.

[Hyperveloce]

There are known past cases of losses of control with very high vertical rates: 20 000 fpm, 30 000 fpm, even more (>40 000 fpm) before a mid air break up occured. In the Adams Air case, involved in a 60° high dive, the structural integrity of the airframe was compromised at ~475 kts, under a vertical load factor exceeding +3G (over 40 000 fpm). Those irrecoverable dives lead to short scenarii (<1min30sec: compatible with ACARS being emitted more than 3mn after the multiple FLRs and AP disengagement ?), very high impact speed and incidence angles, very small debris and no body to recover as such. Looking to the debris as shown by the BEA report (more spec. the big furnitures/Galley G2), knowing that the Airbus was in "ligne de vol" when it impacted on its belly (or on its tail, with a slight pich up), knowing that the 30 bodies recovered by the Ventose Frigate were all in a relatively good overall state (with their clothes), it seems difficult to think that the impact occured in such a way (with such high vertical rates). If we need a longuer scenario (>3mn), maybe we have to assume that the control initially lost at high altitude was somewhat regained but either not enough, or overcorrected ? and that the max. vertical rate at the impact did not exceed 10 000 fpm ?
Jeff

[takata]

Will Fraser:
Respectfully, a trim function would have more relevance to "upset".
At 2 +/- degrees, a down trim to "zero" would have a steady and insidious
effect on stability relative to airspeeds. With a fault as addressed in the pre-AD testing, trim wouldn't be correctable, and/or noticeable to pilots or computers wrestling with Unreliable a/s data to begin with. Only a few hundred feet (Qantas) can produce a substantial increase in speed. Any established descent, left alone, could be disastrous at flight limits experienced by 447: high, hot, and heavy.As I said before, this is going to be very stretched to fit this hypothesis:
- we don't know whilch system would act "by itself" to do that in Alternate 2;
- we don't know why trim and elevators control "wouldn't be correctable", why wouldn't they?
- we don't know why the pilots, trying to fly "pitch and thrust" wouldn't notice any pitch issue, being due to some weirdo computer trim settings?
- it doesn't look like F-GZCP was so "high, hot or heavy"; she still had a good margin in all her flight parameters at this point;
- everything (beside the technical root which is very undetermined) may be assumed only if the crew would be totally panicked.

It makes a lot of assumptions and a lot of unadressed root causes in a system already proven by more than 13 millions flight hours...

S~
Olivier

[Will Fraser]

PJ2

The engine failure causing power interruption in the reasearch pre-AD is not at issue here. Neither is the TO mode. The linkage is what happens when PRIM1 is inop/unselect, and the trim is set to zero, and an inability to correct it with any means, until the second computer resets. To me, this isn't more than just another serious problem to deal with on a FD with serious issues of mysterious severity presenting to pilots who have (ostensibly) had the a/c handed back to them (automatically)

I will keep in mind what ACARS has reported (demonstrably after the fact), and assess what the avenues of handling were, as available to the PIC. (Excuse the bolding, it isn't meant to appear confrontational).

takata

I'm not assuming, I use the Data in the AD. I am however, noting that ACARS is not the FDR. If pitots are involved (it is said they are), Ua/s in combination with a PRIMinop state, unknown a/s or attitude (IRU) a nose down increasing a/s would be unknown to the crew, but if on determining a nose up was necessary and initiated with the ELEV inop (see AD), an input that was correct could be enhanced by a returning ELEV function, to the value of as much as an additional 3 degrees NU notwithstanding a lack of Pitch authority at all. At an already seriously overspeed conditon, further pitchup (by a reentering FC) could cause serious damage even without an upset. This is actually suggested by ACARS. If in Alt Law 2 but no elevator function is available, try to transit to Direct Law, only to find that the HS trim (manual) is woefully insufficient to command recovery.

Will

[takata]

Will:
I'm not assuming, I use the Data in the AD

So lets see the data.
http://atis.casa.go.kr/asms_ad/file/EASA%202008-0010R1.pdf
IMHO, this doesn't match with your explanation of the failure:

Reason:
On A330/A340 aircraft, the Flight Control Primary Computer 2 (FCPC2)
and FCPC3 are supplied with power from the 2PP bus bar. Electrical
transients on the 2PP bus bar occur, in particular during engine n.2 failure
on A330 aircraft or engine n.3 failure on A340 aircraft. Such electrical
transients lead to a FCPC2 reset. FCPC3 reset does not occur thanks to
the introduction of second electrical power supply to FCPC3 from 1PP bus
bar associated to Electrical Contactor Management Unit (ECMU) standard
5.
During the evaluation of specific engine failure cases at take-off on Airbus
flight simulators, it has been evidenced that with FCPC1 inoperative, in the
worst case, when FCPC2 and FCPC3 resets occur during rotation at take
off, a transient loss of elevator control associated with a temporary
incorrect flight control law reconfiguration could occur. It leads to a
movement of the elevators to the zero position, which induces a pitch
down movement instead of a pitch up movement needed to lift off. In
addition, it leads to a limitation of the pilot authority in pitch axis and limits
the capacity to counter the pitch down movement during this flight phase,
which constitutes an unsafe condition.

[Will Fraser]

takata

I have corrected my post to read Elevator inop. Bonehead mistake.
The material is in the AD. I see it, honestly. Law degradation requiring reset, inop elevators, Pitch unavailable, etc. Due to inop FCPC1. A Pitch problem at TO is very dangerous, no less so at altitude near the regime of 447 at loss of control.

[lomapaseo]

Hyperveloce

Thank you for that reasoning. It adds an important boundary (if accepted) on our speculations):ok:

[takata]

Will:
I see it, honestly. Law degradation requiring reset, inop elevators, Pitch unavailable, etc. Due to inop FCPC1. A Pitch problem at TO is very dangerous, no less so at altitude near the regime of 447 at loss of control.
Well, if you see it "honestly", what can I add to make you change your point?
For me, the problem described is not relevant to any "Law degradation", nor specific to a PRIM1 reset, but only inside the above configuration described or the AD wouldn't be so specific with this issue. Moreover, it is involving both FCPC2/3 resets (in the worst case) in normal law situation (where FCPCs are managing "protections") with a transient issue concerning elevators reset position at the moment they are needed the other way with a larger deflection. It doesn't imply something which lasted and any "pitch unavailability", it is said that the zero pitch could not be countered by imputs, neither that it would happen in other configurations.
Again, it is taking a few bits from this AD for making something else, ignoring what doesn't fit with theory.

S~
Olivier

[augustusjeremy]

I just don't believe in the scenario: "fast fall from FL350 in one piece and four minutes" I may be wrong, but it looks to me like a false lead to follow until we'll get much more facts to explain it.
takata,

I would accept the blocked pitot theory without further questioning if it was not that IR2 fault.

I am just trying to find a rational relationship between those ACARS messages.

But I agree with you. We need much more facts to tell the whole story.

Anyway several simulator tests are probably being performed right now at AI to check all the possibilites that could lead to those ACARS messages.


Forgive me for droning on, but I think this technical aspectd needs to be clear.PJ2,

I just have to thank you for your technical explanation.

[aguadalte]

I just wonder what in this case is convincing you to believe that an A330-200, fitted with HONEYWELL ADIRUs, and showing problems with unreliable airspeed probes freezing, would be comparable with an A330-300, fitted with NORTHROP GRUMMAN-LITTON ADIRUs, and showing AOAs spikes unfiltered?

What in AF447 ACARS sequence is similar to QF72 case?
- answer: nothing.
So, maybe its time to drop this claim for lack of any substantial or relevant fact to come with.
Takata,

Apart from the fact that there are reports of allegedly unreliable speed indication problems with all sorts of equipment, there is no relationship between AF447 and QF72.
But regarding the speculation about what really happened to AF447, your and everyone's speculative ideas, have the same value of mine...
And having in consideration the statistics regarding causes for air accidents, I think human error is well ahead of equipment failure.
I'm convinced that the Air France pilots are just men, not Gods, and that they may have committed a mistake by disconnecting the wrong equipment while dealing with a dramatic situation.


Coming back to basics of the investigation, its easy for me, to understand why they didn't deviate from the storm. They have taken-off from SBGL limited by MTOW. The Captain could not have ordered any extra fuel without prejudice of payload due to the ETOW of 233,757Kg.
According to BEA prelim report page 17, fuel weight was 70,400Kg.
In the next page of the report:
The on-board fuel weight corresponded to forecast trip fuel of 63,900 kg, route factor fuel of 1460 kg, final reserve of 2200 kg, fuel to alternate airport reserve of 1900 kg and 940 kg additional fuel.
According to EU OPS 1:

(12) OPS 1.255 Fuel Policy:
(i) For A to A Flights — An operator shall specify the minimum fuel contents at which a flight must end. This
minimum, final reserve, fuel must not be less than the amount needed to fly for a period of 45 minutes.
(ii) For A to B Flights — An operator shall ensure that the pre-flight calculation of usable fuel required for a flight
includes;
(A) Taxi fuel — Fuel consumed before take-off, if significant; and
(B) Trip fuel (Fuel to reach the destination); and
(C) Reserve fuel –
(1) Contingency fuel —
Fuel that is not less than 5 % of the planned trip fuel or, in the event of in-flight replanning, 5 % of the trip fuel for the remainder of the flight; and
(2) Final reserve fuel —
Fuel to fly for an additional period of 45 minutes (piston engines) or 30 minutes (turbine engines);
and
(D) Alternate fuel —
Fuel to reach the destination alternate via the destination, if a destination alternate is required; and
(E) Extra fuel —
Fuel that the commander may require in addition to that required under subparagraphs (A)-(D) above.

The term Additional Fuel is used in EU OPS 1, when needed due to ETOPS especial fuel needs:
Additional fuel, if required by the type of operation (e.g. ETOPS);

I believe the term used by BEA, route factor fuel, meant contingency fuel, (a value of 5 % of the trip fuel (that may also be reduced to 3%, when a fuel alternate aerodrome is available under certain circumstances). It intrigues me that the value of contingency fuel of AF447 is only 1460 kg. Additional fuel is supposed to cope for ETOPS minimum fuel needs at ETPs and is added to Estimated Total Fuel that must already contemplate all the items of Part 1.255 of EU OPS 1.
I'm not saying that, because they were limited on fuel, they could not deviate, but I'm willing to accept the fact they didn't. What I don't understand is why, after almost 3700 replies on this second thread, I'm almost the only one here writing about human factors.

[PJ2]

aquadalte;
What I don't understand is why, after almost 3700 replies on this second thread, I'm almost the only one here writing about human factors
Well, if I can offer this view as a pilot (now retired), I think many colleagues here are only too keenly aware of the human factors possibilities but have remained essentially silent because, like the accident itself there is only so much one can say and it's all from the ACARS messages and what little the BEA adds to the thinking. At this point there is absolutely nothing to speculate meaningfully upon regarding human factors. We can delve tentatively into a few possibilities that you have touched upon but I'm not sure the discussion has anywhere to go at the moment.

You mention fuel load - most pilots who have contributed to this thread have faced the issue of offloading freight for more fuel or risking an enroute landing/diversion due to weather etc - those flight planning problems are all in a day's work. Whether the two guys up front didn't want to risk a weather diversion because they only had 900+kg of fuel and weren't aware how little it takes to go a hundred nm's off course and back, or whether they knew how to read and interpret their radar returns are all human factors and are all moot points for the time being.

Analyzing "human factors" at this point risks building a series of straw man arguments and then knocking them down - the arguments and their possibilities, mean nothing in the sense that they explain the accident in a way that other ways have not.

Yes, the crew could have run across an extremely rare triple failure - pitot sensors, ADIRU 1 or 2, and a PRIM1, these equipment failures being all independant of one another for the theory to work, the failure of the PRIM with subsequent ND bias requiring action under the AD, (assuming they knew that action off by heart, which I am almost 100% certain both would not know), - there is NO way they would have had time to look it up and run the checklist from the QRH OEB section.

And if they launched into shutting an ADR and/or IRU off "by memory", unless they were following an ECAM drill that got interrupted by a serious event, that takes us into extremely rare and difficult territory -in the realm of rogue actions outside memory and checklist territory in a last-ditch panic, again another human factors matter but in my view given the airline involved and the circumstances the flight was in, not very probable if not all but impossible. Also, we still dont' know what manufacturer's ADIRUs were installed.

[kilomikedelta]

3700+ posts (including the deleted ones) based on some photos, cryptic BEA reports and archival meteorological satellite data? I would say that 99% of the postings are egocentric masturbations by people with lots of hubris, too much time on their hands and regrets about their contribution to society. Let's wait a year or so while the experts accumulate the real data and sort out what's possible to deduce. Then we can revisit it. Time to close the thread and time for the frequent posters to discuss their behavior with their therapists. KMD M.D.

[Will Fraser]

And now yours makes 3,701, mine 3,702. You are wanting to join? You have authority issues? Your Munchhausen's is unresponsive to treatment? Lonely in Canada? I think by your definition you have wasted some of your valuable time. WF CPL

[HarryMann]

I would say that 99% of the postings are egocentric masturbations by people with lots of hubris, too much time on their hands and regrets about their contribution to society.

I would say about 50% might fit that bill, including yours...

[Lemurian]

Aguadalte :
Apart from the fact that there are reports of allegedly unreliable speed indication problems with all sorts of equipment, there is no relationship between AF447 and QF72.
I agree

Coming back to basics of the investigation, its easy for me, to understand why they didn't deviate from the storm.
Whether they did or not is a red herring. The position at 0210Z, reported on either the MeteoFrance map or Tim's 0215Z sat infrared picture puts the airplane out of the convection area. (something that most posters still haven't accepted, I wonder why ? ).
The on-board fuel weight corresponded to forecast trip fuel of 63,900 kg, route factor fuel of 1460 kg, final reserve of 2200 kg, fuel to alternate airport reserve of 1900 kg and 940 kg additional fuel
This is the flight plan that the Captain signed :

Trip fuel ...63940
Alternate fuel ORY 1900
Contingency 1460
Final reserve 2200
Extra fuel 900 (at captain's decision)

What additional fuel, for ETOPS / loss of pressurisation, ETPs is included in the route/flight study and is inside the fuel load. Please note that on this particular flight, the en-route alternates were Natal and Sal, which are on the planned flight track.
It intrigues me that the value of contingency fuel of AF447 is only 1460 kg.
It's because AF fuel policy takes into account the burn-off caused by the need to carry a given load. The fuel plan shows the quantities left at the end of the flight : Basically, the alternate / contingency / final reserve /...are carried to the landing at destination, hence they are part of the landing weight and the burn-off they cause is included in the trip fuel.
In this case, and considering that the K factor on this flight was 1.350, a contingency fuel of 1460 kg at destination is worth 1460 x 1.350 = 1970 kg which is in fact slightly more than the required 3% of the trip fuel (1918 kg ).

I'm not saying that, because they were limited on fuel, they could not deviate, but I'm willing to accept the fact they didn't.
You are not saying it....true....Just strongly suggesting...
I'm almost the only one here writing about human factors.
And what is the basis for your discussion ?

All aspects of the flight prior to 0210Z show a functioning aircrew, doing a serious job of keeping ahead of the aircraft, proposing some strict R/T procedures...
A crew who was kept abreast of any changes on the route weather (the despatch message confirming some active convection inside the ITCZ)

Against that, a lot of hot air about "other aircraft have deviated...bla..bla bla ..." and not one of these posters has taken the effort of plotting those deviations on the sat pictures. Most of these flights were still smack inside the red area, except maybe the IB flight on their 70 Nm east of track who would have flown between the two most active zones.
I find it quite arrogant to judge on a crew's action without a modicum 1/-of experience of flying through the ITCZ and 2/-in the comfort of an armchair.
These guys had one hour and six minutes of flight worth in fuel and you tell me that they were concerned about any deviation or diversion ?
As to what happened after 0210Z, we are into speculations and conjectures.
And if we start the game about how this crew reacted to a wholly abnormal situation, we need both to consider what happened as the ACARS messages don't tell the whole story, and who was in the flight deck at that moment, taking into account AF policy of "no left-hand seat qual for F/Os".
But that's the sort of speculation I would be very careful in discussing.

[JD-EE]

Coagie, can you reconcile your proposed scenario with the locations of the bodies and wreckage after 7 or more days of drift at sea? Somehow the plane apparently made some serious travel in the reverse direction of its flight plan from its last reported position. The distance may have been big enough that a significant portion of that travel took place after the last ACARS message. Some REALLY accurate back tracking on the drift of objects is needed to get some kind of estimate of where the plane went down. Appearances are that it did not go down at its 02:10:34 position or further along its flight plan.

So losing the tail in a scenario that requires it be blown off forward BUT with damage to its rudder's lower edges rather than tipping off.

There is a nice way to get a "feel" for what the BEA folks saw at the base of the VS if you're modestly handy with tools. Take a piece of paper and a small board, say 2.5cm by 2.5cm by 30cm. Cut a slot most of the way through the board lengthwise in the middle of one side. Slide the piece of paper into the slot. Get some short dowel rod. Cut off three pieces about 3 to 4 cm long. Find a drill a little larger than the diameter of the rod. Drill three holes through the board and paper in the "front" 2/3 of the paper and slot. Don't have the hole drilled in the paper actually break the edge of the paper. Slip in the pieces of dowel rod. Now grab the the paper with both hands in a "prayer" position and pull up and "forward". Notice the tears in the paper at the hole positions is not symmetrical. This allows you to figure out which way it was broken loose. It is on this basis the BEA made its pronouncement.

You can see this to a lesser degree pulling paper out unevenly from a three ring binder.

I believe they have a very good notion of that direction the VS was trying to move as it was pried loose. And I think I have a half a good guess on how this happened. It does not include breaking off in flight.

The theories are fun. But it's more interesting when you can fit the theory to the very limited set of facts present.

JD-EE

[llagonne66]

PJ2

As stated somewhere above (or maybe in the previous thread), Honeywell ADIRUs (P/N HG2030AE22) are installed on AF A330s.

[VicMel]

As there were a lot of ACARS messages within the 2h 10 'window', could the aircraft already have been in trouble (and starting to drop) sometime before, say between 2h 08 and 2h 09? Also if ACARS stopped because both engines had been lost (physical or flame-out)the impact could be sometime after 2h 14, say between 2h 15 and 2h 16. This would give a descent time of 6 to 8 minutes. If terminal velocity was reached quite quickly, this would give an impact velocity of about 5000 ft/min

[Jag6]

All aspects of the flight prior to 0210Z show a functioning aircrew, doing a serious job of keeping ahead of the aircraft, proposing some strict R/T procedures...


Actually, I think that this statement needs examination. From 1:35 until 2:14 the only crew action of which we are aware is their third attempt to establish the ADS-C link with Dakar ATC at 2:01. No other comms at all. Having made two attempts at 1:33 & 1:35, the crew then apparantly waited another 25 minutes, until the last possible minute to meet ADS-C requirements, to make their third try. The BEA report says that the minimum ADS-C requirement is reporting 20 mins ahead of entry into Dakar FIR; 2:01 is just about 20 mins ahead of their estimated time at TASIL (actually I believe 2:20 was Atlantico ATC's projection, not the crew's estimate).
Can someone here who has experience/knowledge of ADS-C answer this: once an initial connection has been attempted, could the system automatically generate further attempts - especially at the 20 mins threshhold when it detects that the connection has not been made? Or can the crew program the system to make further attempts? The BEA report says that the first connection has to be made by the crew - but could the BEA mean that the initial attempt at a connection has to be made by the crew?
I appreciate that there may well be a perfectly mundane explanation for the 25 minutes silence, but as this is the only comm I believe it is worth asking the question.
My first post - have been reading this from day 1, but it is easy to miss things/forget what you have read, so apologies if this has already been answered - but I don't believe that it has.

[VicMel]

augustusjeremy, re #3694
It is not just the 'IR2 fault' message that questions the 'blocked pitot theory'.
As the BEA report reiterates, the ACARS messages are maintenance messages for the ground crew, not warnings to the flight crew. A maintenance message that is generated by a system's BITE should be due to that system's BITE detecting an internal fault:-
1 NAV TCAS FAULT Message.
As indicated in the BEA report (page 50) the cause is unclear. I suspect that the most vital parameter derived by TCAS is the relative velocity with respect to other aircraft. As well as a BITE check on max and min, I would expect the rate of change of such a vital parameter to be checked. A rate of change of more than about 20 knots per second for say ten samples might be considered as out of spec and hence warrant a maintenance fault message.
2 ADR Fault.
It is not clear as to the level of inconsistency between the three ADRs; the statements within the BEA report seem to conflict. On page 50 regarding NAV ADR DISAGREE message, “Meaning: This message indicates that the EFCSs have rejected an ADR, and then identified an inconsistency between the two remaining ADRs on one of the monitored parameters.”, but on page 51 regarding the PROBE-PITOT message, “...triggered one of the speed monitoring processes: they have detected a decrease of more than 30 kt in one second of the 'polled' speed value. The three ADRs were considered valid by the EFCS2 at the time the monitoring was triggered... “.
3 ISIS SPEED OR MACH FUNCTION
Presumably this could due to the same cause as for the ADR Fault; what a shame that the CMC does not send the flag status, just a single character representing 'flags code' would give so much information at so little cost! Or if the info in post #1442 is correct (but #1757 is not), this could be a problem with gyro derived direction.

[Captain104]

3700+ posts (including the deleted ones) based on some photos, cryptic BEA reports and archival meteorological satellite data? I would say that 99% of the postings are egocentric masturbations by people with lots of hubris, too much time on their hands and regrets about their contribution to society.

Mast......ns is bit harsh but nevertheless true. Too much time on their hands? Must be also a kind of play. I asked in post #3367 for a closure. Danny did not hear me.:rolleyes:
Regards

[Robin42]

PJ2
As stated somewhere above (or maybe in the previous thread), Honeywell ADIRUs (P/N HG2030AE22) are installed on AF A330s.

In such a weird case, you would have expected such informations to show up in the preliminary report, along with alread applied modifications as scheduled by the applicable Airworthiness Directives. You would have expected all ACARS messages to be listed, not only those routed to the AF maintenance division. You would have expected the list of waypoints. You would have expected the INMARSAT signal quality data listed along with the messages. You would have expected the annex of the French version to appear in the English version. All this would have been very easy to do, completely independent from Brazil's stance.

[Graybeard]

VicMel: 1 NAV TCAS FAULT Message.
As indicated in the BEA report (page 50) the cause is unclear. I suspect that the most vital parameter derived by TCAS is the relative velocity with respect to other aircraft. As well as a BITE check on max and min, I would expect the rate of change of such a vital parameter to be checked. A rate of change of more than about 20 knots per second for say ten samples might be considered as out of spec and hence warrant a maintenance fault message.

What other aircraft? The one 80 miles in trail? TCAS has no airspeed input. Relative velocity with targets depends on tracking them via TCAS/transponder link, and not onboard data. Altitude difference, rate of closure, and time to CPA, closest point of approach, are the TCAS triggers.

I'm with BEA that this fail remains unanswered, and unrelated to the other faults, which can be tied back to faulty airspeed.

GB

[aguadalte]

All aspects of the flight prior to 0210Z show a functioning aircrew, doing a serious job of keeping ahead of the aircraft, proposing some strict R/T procedures... Actually, I think that this statement needs examination. From 1:35 until 2:14 the only crew action of which we are aware is their third attempt to establish the ADS-C link with Dakar ATC at 2:01. No other comms at all. Having made two attempts at 1:33 & 1:35, the crew then apparantly waited another 25 minutes, until the last possible minute to meet ADS-C requirements, to make their third try. The BEA report says that the minimum ADS-C requirement is reporting 20 mins ahead of entry into Dakar FIR; 2:01 is just about 20 mins ahead of their estimated time at TASIL (actually I believe 2:20 was Atlantico ATC's projection, not the crew's estimate).
Can someone here who has experience/knowledge of ADS-C answer this: once an initial connection has been attempted, could the system automatically generate further attempts - especially at the 20 mins threshhold when it detects that the connection has not been made? Or can the crew program the system to make further attempts? The BEA report says that the first connection has to be made by the crew - but could the BEA mean that the initial attempt at a connection has to be made by the crew?
I appreciate that there may well be a perfectly mundane explanation for the 25 minutes silence, but as this is the only comm I believe it is worth asking the question.
My first post - have been reading this from day 1, but it is easy to miss things/forget what you have read, so apologies if this has already been answered - but I don't believe that it has.As far as I know, (at least with the systems we use) I don't think connections may be "programmed by the pilot to automatically generate further attempts". CPDLC-ADS was on testing in the Atlantico and Dakar areas. Connections with Atlantico were, until last month, completely impossible to make, but flights could in fact connect with Dakar by prior talking with them via HF (about 40 minutes prior to enter de FIR), stating that they wanted to establish an ADS connection and giving them a complete radio report with the estimates for all reporting points, altitude, speed, A/C type, tail number and selcal. This would allow the Dakar operator to insert the data into the system. After 10 to 15 minutes, they would call you back saying that you could connect. I've done this a lot of times and it worked fine. Of course you have to fly to Africa and to the Latin Americas to "understand the system". How they work. The use of CPDLC-ADS in those Continents is completely different from the one in the North Atlantic...

But then, again...what do I know?
I'm just an...
arrogant to judge on a crew's action without a modicum 1/-of experience of flying through the ITCZ and 2/-in the comfort of an armchair.Lemurian: I wonder what you meant by: These guys had one hour and six minutes of flight worth in fuel I think they had more...if they were available for a refueling technical landing to complete the flight, or had less, having in mind the completeness of the flight, provided they would have to re plan, having in mind (EU OPS 1 (again...)
(...) (1) Contingency fuel —
Fuel that is not less than 5 % of the planned trip fuel or, in the event of in-flight replanning, 5 %
of the trip fuel for the remainder of the flight; and (...)

Further, I understand your explanation of the fuel figures used by AF. I just don't think they are in the spirit of the law. EU OPS 1 says, x% of the trip fuel, it doesn't say, x% of the trip fuel, deducted by the fuel burned by the fact that you have to carry it, that should be added to trip fuel. And the reason why, is because pilots do need to have clear figures to administrate. How does the contingency fuel shows to the pilot on the FMGC? Is it shown, one part of it on "trip", and the other on a "percentage of the trip"? Because, as far as I know, 3% of the trip fuel, shown on FMGC, before take-off should read 1910 kg, not 1460 kg. And those 940 kg, would have been showing up on the extra fuel prompt.

[augustusjeremy]

TCAS has no airspeed input

Maybe it has groundspeed as input.

I suppose - please correct me if i am wrong - that groundspeed is provided by the Inertial Reference Units only (not ADRs).

[astir 8]

Guyana newspapers report a large section of AF 447 and other items have been washed up there. Clearly the search wasn't 100%

[Graybeard]

TCAS has no speed input at all. Honeywell uses the attitude input to select which TCAS antenna to use for target tracking. Collins uses relative signal strength instead. TCAS is interested in rate of closure.

GB

[augustusjeremy]

TCAS has no speed input at all. Honeywell uses the attitude input to select which TCAS antenna to use for target tracking. Collins uses relative signal strength instead. TCAS is interested in rate of closure.

So it would depend on the TCAs model - I don't remember if someone wrote about it in this thread - in this particular a/c the need of a direct attitude input ?

I wonder how TCAs gets altitude information... GPS only, GPS with IRS, IRS only, baro/altitude input from ADRs (with or without eventually computing airspeed (pitots) for adjustements ?)

[LeandroSecundo]

Hi,

Guyana newspapers report a large section of AF 447 and other items have been washed up there. Clearly the search wasn't 100%

All will be happy if you publish your source(s)

Bye.

[rgbrock1]

Quote:
Guyana newspapers report a large section of AF 447 and other items have been washed up there. Clearly the search wasn't 100%

Yes indeed. I've looked at 6 online Guyanan newspapers and not one
has any mention of this.

[TyroPicard]

VicMel
It is not just the 'IR2 fault' message that questions the 'blocked pitot theory'.As I understand it, you cannot be certain there is an IR2 fault - IR2 is the first of the systems that is reporting the ADIRU2 (1FP2), HARD fault. All explained on page 48 of the English translation. So the fault may not be the IR but some other portion of the ADIRU. I stand to be corrected if the (1FP2) bit gives that information, as I don't have a clue what it means.

[Will Fraser]

Graybeard

Given that the sequence of ACARS reporting is indistinct, and attitude is a parameter for TCAS antenna sequestering, what of IRU moaning has to do with its (TCAS) fault? Loss of attitude management could even explain the a/p trip. Turbulence? FCPC? TCAS fault is not inexplicable.

Holding ACARS loosely, which I do, a genesis of fault reporting could begin while 447 was "on flight path" and the "last reported" position may have been an accurate record of several minutes of heading, altitude, airspeed discrepancies contributing to an upset.

Why hold ACARS loosely? BEA included only the data originally leaked to France2. Essentially, granting it provenance.

Fair enough, but why then withold the "remaining" data?? (FromRobin42, above).

If one takes ACARS (the system and this specific data) with a grain of salt, there would be fewer "theories" to be bolstered by "FDR style" "official" data that comes from a maintenance tool, ACARS.

[Will Fraser]

Further, if 447 was lost at 0214+, that would time the messages at ~four minutes, and its prelude possibly ~4 minutes. If, as above, the a/c was on published route, perhaps until 0206, when attitude and turbulence induced handling qualities diminished, the time span of accident behaviour would be ~8 minutes, an average ROD of 4,500fpm. Consistent with BEA, its speed at impact would be somewhat higher.

The four minute reporting frame supports a flight path without large heading and altitude excursions (no "spin", no "disintegration"). Not without its challenges, this mode may have then deteriorated (rather than recovered as in Q and NW), to produce the flight aspects at entry suggested by BEA.

Will

[Pro-Heavy]

Air France Airbus A330-200 crash: Wreckage, sensory equipment recovered on local shores

July 17, 2009 | By Christopher (http://www.kaieteurnewsonline.com/author/christopher/) | Filed Under News (http://www.kaieteurnewsonline.com/category/news/)
By Gary Eleazar
Dennis Baksh, a farmer from Ann’s Grove, East Coast Demerara, knew that he had stumbled on to something valuable last Friday whilst on the Hope Beach.
Baksh had found what is believed to be a part of the fuselage and a component believed to be the first piece of sensory equipment recovered from Air France Airbus A330-200 that crashed into the Atlantic Ocean after leaving neighbouring Brazil en route to France on June 1.
According to Baksh, he immediately realized that what he had found was not a part of a boat and as such made arrangements to have a tractor haul the piece of debris to his home, some distance inland.
“I see it pun deh seawall and when a go and check I say that this got to be a part of a plane,” said Baksh, yesterday.
http://www.kaieteurnewsonline.com/images/2009/07/plane-1-copy1.jpg (http://www.kaieteurnewsonline.com/images/2009/07/plane-1-copy1.jpg) A close up of the sensory equipment

The man who made the important discovery said that he did not see any other piece of debris in the area.
The sensory component that was attached to the fuselage is a three-axis Inertial Measurement Unit (IMU) which is cylindrical in shape and is mounted on a flat base with electrical wiring protruding from within it. It bears the trademark ‘Sensorex.’
Baksh attempted to open the device but only managed to break into the cylindrical case; the cube within was impenetrable to Baksh.
A GCAA official said that he noted the numbers on the fuselage and on the equipment marked Sensorex and that he would be forwarding these to the relevant authorities.
Another aviation official explained that the piece of equipment marked Sensorex might be the first piece of sensory equipment retrieved from the crashed plane. He explained that it sends signals to the flight panel and this might be the piece of malfunctioning equipment that caused the crash.
An Inertial Measurement Unit (IMU) is the main component of inertial guidance systems used in air, space, and watercraft, including guided missiles.
An IMU works by sensing motion including the type, rate, and direction of that motion using a combination of accelerometers and gyroscopes.
The data collected from these sensors allows a computer to track a craft’s position, using a method known as dead reckoning.
The IMU works by detecting the current rate of acceleration, as well as changes in rotational attributes, including pitch, roll and yaw.
The data is then fed into a computer, which calculates the current speed and position, given a known initial speed and position.
For example, if an IMU installed in an airplane were to detect that the craft accelerated westward, resulting in a calculated, constant speed of 500 miles per hour, and detected no other accelerations for one hour, then the guidance computer would deduce that the plane must be 500 miles west of its initial position.
When combined with a computerized system of maps, the guidance system could use this method to show a pilot where the plane is located geographically, similar to a GPS navigation system but without the need to communicate with any outside components, such as satellites.
The term IMU is widely used to refer to a box containing three accelerometers and three gyroscopes. The accelerometers are placed such that their measuring axes are orthogonal to each other. They measure inertial acceleration, also known as G-forces.
Three gyroscopes are placed in a similar orthogonal pattern, measuring rotational position in reference to an arbitrarily chosen coordinate system.
IMUs are primarily used in vehicle-installed inertial guidance systems. Today almost every commercial or military water-going vessel has one.
Most aircraft are also equipped with IMUs.
http://www.kaieteurnewsonline.com/images/2009/07/plane-copy1.jpg (http://www.kaieteurnewsonline.com/images/2009/07/plane-copy1.jpg) Dennis Baksh displays the sensory component alongside the fuselage believed to be that from the Air France Airbus A330-200 that crashed in the Atlantic Ocean on June 1 last.

When Baksh found the piece of fuselage, he said that there was also a thick foam-like material attached to the inside of the piece of fuselage but this he removed given that while it was water soaked it made the already heavy piece of debris almost immovable.
The Police and the Guyana Civil Aviation Authority were subsequently contacted with the latter confirming that the piece of debris was definitely from a large aircraft and that the component bearing the trademark Sensorex was indeed a piece of sensory equipment used on aircrafts.
The Brazilian Embassy was also contacted.
At present, it is still unclear what caused the crash that killed 228 persons aboard.
The aircraft was flying in turbulent weather at the time but some international analysts have said it is unlikely that the weather contributed to the crash but are not ruling it out completely.
Chief Air France spokesman Francois Brousse had suggested the plane could have been struck by lightning but most experts say lightning doesn’t usually bring down a modern airliner, unless it coincides with other factors that contribute to the accident.
“Planes are built with lightning strikes in mind and are struck reasonably frequently,” according to Patrick Smith, a U.S. commercial pilot and aviation writer.
Aviation safety statistics indicate that each large passenger jet such as the Airbus A330 is struck by lightning about once every three years on average.
http://www.kaieteurnewsonline.com/images/2009/07/plane-2-copy1.jpg (http://www.kaieteurnewsonline.com/images/2009/07/plane-2-copy1.jpg) Searching for any markings on the fuselage.

Regional aircraft, however, which fly at lower altitudes, are hit more frequently, about once a year.
About seven hours after taking off and flying through the night over the mid-Atlantic, the pilots of the Air France Airbus reported that they had encountered an area of intense cumulonimbus activity, part of the massive thunderstorms that regularly batter the region
Air France reported that the aircraft’s ACARS (Aircraft Communications and Addressing System) a digital datalink that automatically transmits service messages from the aircraft to ground stations messaged the company’s headquarters regarding a problem with the aircraft’s electrical and pressurization systems.
The French aircraft accident investigation agency, the Bureau d’Enquetes et d’Analyses (BEA) had announced that an examination of structural components of the aircraft recovered from the surface of the ocean revealed that they were deformed from the bottom to the top.
This according BEA suggests that the aircraft hit the sea “in the direction of flight and with a sharp vertical acceleration”. This, in turn, indicates that the aircraft was “not destroyed in flight”. The aeroplane’s speed sensors were “a factor but not the cause” of the disaster.
The investigators had also determined that no distress call was made, either to air traffic control or to other aircraft.
The most direct evidence of the cause of the accident is the presence of suspected aircraft debris on the ocean surface and a series of 24 automatically generated maintenance messages that were transmitted from the plane shortly before the crash.
Investigators also said that messages indicated that cabin pressure was lost and there was some kind of electrical system failure.

Air France Airbus A330-200 crash: Wreckage, sensory equipment recovered on local shores : Kaieteur News (http://www.kaieteurnewsonline.com/2009/07/17/air-france-airbus-a330-200-crash-wreckage-sensory-equipment-recovered-on-local-shores/)

[Pro-Heavy]

SENSOREX,
a MEGGITT group company
As the European leader in LVDTs for flight control and
inertial systems, Sensorex offers a wide range of
technical solutions for measuring linear and angular
displacement, tilt, acceleration and angular speed.
Based on inductive principles and MEMS, our
sensors are particularly suited for use in severe
environments. As there is no need for mechanical
contact, our products guarantee performance and
reliability.
We also offer integrated electronic solutions and
smart sensors using our own micro-electronic technology.
In increasingly complex technical environments, we
are confident that our long experience, drive to
innovate and advanced know-how can deliver
the onboard systems that will comply with the
most stringent requirements.

[Will Fraser]

Looks like a portside ventral wing root fairing.

[HarryMann]

Could that 'sensor' holds any stored information or secrets internally? Perhaps, how it came to be hanging in there, of importance...

The simple fact that 'skin' parts and possibly further appendages are still floating about in the great ocean could however, be of some considerable importance.

[HazelNuts39]

Re: Will Fraser (3721)
I doubt that an IMU would be fitted to a wing fairing. Could it be part of an Ariane booster?

During the search wreckage moved northwards. Georgetown is approx 1600 NM NW of Last Rept Psn.

regards,
HN39

[BOAC]

Has anyone calculated the likely ocean drift of the piece? Is that distance possible?

[ttcse]

It's not part of the wing

[DJ77]

Looks like they found a SX43030

Sensorex (http://www.sensorex.fr/en/catalogues/product-inertial-measurement-units-19.html)

[Will Fraser]

A fairing is not part of the wing, that is correct. It also could be a cousin, the fairing under the Horizontal Stabilizer. From its shape and the contour of the "leading edge" negative, it too could have been oriented as above.
If near the HS, it could be associated with the "trimmable" Horizontal Stabilizer piece found as described by BEA. The "cutout" (negative) in the Aluminum is hard to judge, but I do believe it was near an airfoil.

[thermalsniffer]

Inertial systems - Sensorex - Displacement transducer, signal conditioner, electronic interface, sensor, conditionner, intrinsically safe, shock recorder, inclinometer, accelerometer, rate gyros (http://www.sensorex.fr/en/catalogues/02_inertiels_centinert.php4)

[ttcse]

Will FraserIt also could be a cousin, the fairing under the Horizontal Stabilizer. From its shape and the contour of the "leading edge" negative, it too could have been oriented as above.
If near the HS, it could be associated with the "trimmable" Horizontal Stabilizer piece found as described by BEA. The "cutout" (negative) in the Aluminum is hard to judge, but I do believe it was near an airfoil.

You're not even close

[Razoray]

Could anyone please explain the signifigance of finding this IMU.
How it can help in the investigation?????

[rgbrock1]

ttcse:

So far you've written that "it's not part of the wing" and "you're not even close". So, pray tell, what do YOU think it is?

[Will Fraser]

HazelNuts39

I would locate an IMU where the motion is least susceptible to "trim".
The wing, when it moves, as it is located nearer the cg, is more likely to be an actual rather than a trimmable, movement. Less susceptible to transient, higher value accelerations as well. Just guessing.

The piece might fit around the VS also. That would make it part of the Fuselage.

[ttcse]

rgbrock1So, pray tell, what do YOU think it is?It's not part of a flight surface nor a fairing. It's part of the fuselage. I 'think' it's part of the forward fuselage.

That's my call from simply looking at the photo. If you read some of the text there are further 'clues' that supports my observation.

[Hyperveloce]

If the big debris is a part of the fuselage, where are the IRUs located exactly ? An avionic bay below the cockpit ?
Jeff

[CONF iture]

The QANTAS accident occurred through elevator deflection, however. Caught quickly, it is controllable through pilot input. I think what takata stated regarding giving credit to the engineers and pilots is the correct view and coincides with my own experience in general - it is an easily controlled set of circumstances if caught quickly - and it's what pilots do - just like the QANTAS guys did.
Quick or not, regarding QF72, there was nothing to be done as long as the protections were in control ...

[jimbeetle]

First post after finding this thread about 150 or so pages ago, so be gentle.

From everything I've read here and other sources, the IR part of the ADIRUs only rely on gyros -- and do not include accelerometers -- whereas IMUs use both. Also from what I've read, it appears that for the ADIRU the gyros are contained within the IR part of the box in the avionics bay, not externally (and there really is no reason for them to be).

So, as this has not yet been confirmed as part of AF 447 I have to go with HazelNuts' suggestion and for now assume it might be a part of an Ariane (there just happened to be a launch that included two boosters on July 1st).

[rgbrock1]

jimbeetle:

So, as this has not yet been confirmed as part of AF 447 I have to go with HazelNuts' suggestion and for now assume it might be a part of an Ariane (there just happened to be a launch that included two boosters on July 1st).


The only thing I would question about this is: how many Ariane launches
have occured in the area? And why, until now, have no other discarded parts of this vessel ever wound up on a beach in Guyana?

[Hyperveloce]

So, as this has not yet been confirmed as part of AF 447 I have to go with HazelNuts' suggestion and for now assume it might be a part of an Ariane (there just happened to be a launch that included two boosters on July 1st).

HazelNuts & ThermalSniffer may be right, these model of IMU could be a part of the Ariane 5 boosters upper stage: Centrales inertielles - Sensorex - Capteur de déplacement, enregistreur de choc, électronique, micro électronique, conditionneur de signal, sécurité intrinsèque, inclinomètre, accéléromètre, gyromètre (http://www.sensorex.fr/fr/catalogues/02_inertiels_centinert.php4)
Jeff
Correction: it seems that there are no IMU in the Ariane boosters, only in the equipment bay of the main stage that goes to orbit with the payload.

[Will Fraser]

Keeping an open mind, of course, but authorities who have seen the debris call it 447. White outside, etc. The lamination of foam and two layers of Al is almost assuredly AB, and since the IMU was attached to the panel, well, there is a problem.

Accelerometers are found on AB, else there would be no aerodynamic load protections?

[wes_wall]

Seems a no brainer to me. Who ever knows, if the A320 has an IMU, where is it installed. Surely this would not be a difficult question to put to bed.

[LeandroSecundo]

Hi,

Quick or not, regarding QF72, there was nothing to be done as long as the protections were in control ...

So true ... so sad ...

Bye.

[BJ-ENG]

The transducer is a 2 axis inclinometer (G1 & G3), and the large piece may well be part an Arian nose fairing.

See blue section " Herschel Plank" with lump at top plus gray stripe above.

http://rascmontreal.org/IYA-2009-InMontreal/wp-content/uploads/2009/05/iya_logo_fairing.jpg



Now look at the dark section on the wreckage near the man's head, plus a similar gray ring.

http://www.kaieteurnewsonline.com/images/2009/07/plane-2-copy1.jpg

[vovachan]

Could anyone please explain the signifigance of finding this IMU.
How it can help in the investigation?????

Absolutely none, most likely.

[HazelNuts39]

RE: Will Fraser (#3739)

Will,
my reasoning was simply: the airplane has three ADIRU's, each containing an ADR and an IRU, what would it need an IMU for? (by the way, my understanding is that the IRU, as any INS, comprises laser-gyros and accelerometers for all three axes, it cannot derive speeds and position from gyros alone).

In addition, units like this in an airplane would have to be line-replaceable, easily accessible for maintenance, the pictures didn't looke like that.

regards,
HN39

[LifterShifter]

Talking to a Guyanese pilot friend, many parts of Ariane wreckage have been found over the years in Guyana - usually smaller that the current piece but typically an aluminum/foam/aluminum sandwich.
There have 190+ launches of Ariane rockets since 1979.
LS

[Hyperveloce]

Could anyone please explain the signifigance of finding this IMU.
How it can help in the investigation?????

Maybe the MEMS holds the physical traces of the impact severity and could allow a finer estimation than the other debris ?
Maybe it can give clue on how the airframe failed at the impact ? (why these two pieces are retrieved together)
Maybe it can be proven that this instrument failed in flight ?
... ?
Jeff

[Coagie]

JD-EE, Thanks for a good description of how they might figure out what happened to the vstab. I hope the investigators are as clever and driven as the people in this and other forums. I still think it's a good idea to simulate different failure scenarios on the ground with an ACARS configured in the same way as on AF447, just to see which ones match the alerts received from AF447.

[VicMel]

According to #1442: at 0210, a 3443 event, a TCAS problem (34 is navigation; 3443 is the Doppler system. The Doppler system here is used to measure relative motion of another body, in this case another aircraft, for TCAS).

However, if as Graybeard states (#3708) no other aircraft were in transponder range, an out of spec TCAS derived parameter could not have been the reason for the message.

From the FAA booklet, 'Introduction to TCAS ii, Version 7'
Page 7: TCAS is designed to work autonomously of the aircraft navigation equipment and

Page 33: The performance monitor validates many of the inputs received from other aircraft systems and validates the performance of the TCAS processor. These include the own aircraft pressure altitude input and the connection of TCAS to the aircraft suppression bus.
.

This possibly suggests pressure alt as being the most likely cause, but (back to my original point) the existence of this message still questions the 'blocked pitot theory'.

[Hyperveloce]

I am told that there are other IMUs than those of the ADIRUs, located near the CoG of the plane, for the Manoeuver Load Allevation and Turbulence Dumping functions ?
These recovered IMUs do not look like the box shaped ADIRUs ?
Jeff
PS) Well... the following article says that Sensorex IMUs are used both on Airbuses and Ariane's boosters (they are not autopiloted, but it is for trajectographic purposes)
Sensorex a développé un capteur d’une précision extrême déjà adopté par Airbus et Ariane:
Sensorex se positionne sur des marchés pointus. La moitié de son activité est dédiée aux capteurs de déplacement linéaire ; installés avec des applications dans les commandes de vol d’Airbus, dans des machines-outils, des équipements de métrologie ou des installations nucléaires. Elle produit également des inclinomètres, accéléromètres, et autres gyromètres pour l’aéronautique ou l’industrie : afin de suivre la trajectoire d’un booster d’Ariane.
http://www.brefonline.com/numeroERA_affichearticle.asp?idA=2973 (http://www.brefonline.com/numeroERA_affichearticle.asp?idA=2973)

[ttcse]

BJ-ENGNow look at the dark section on the wreckage near the man's head, plus a similar gray ring.

I did try to match the dark colors of the wreckage, as if it were paint, with photo's of AF 330's but could not see how they could match.

I wouldn't think all Ariane's get the same paint job.

I did notice the compound curve of the wreckage which says nose or tail area(of something).

HazelNuts39Could it be part of an Ariane booster?Good thinking.

[24victor]

@BJ

The paint job is a direct match. Look at the blue line and underlying join on the news photo and then the dark kick-up to the left and compare with the top ring of your picture and the kick-up on the logo. Well spotted.

Rgds.

24V