AF447 [Archive] - Page 9

PJ2

20th Jun 2009, 19:39

ACLS6;

Thanks for your trouble in finding the galley photos. I did have these two galley photos in doing my own searching and you're right, the first one is the rear galley; the second one doesn't show enough to confirm it's position. The quoted note in takata's post is very helpful however, so we know that the recovered galley section almost certainly came from the front of the aircraft as Been Accounting has said.

Takata's quoted post (from Airliners.net) if very helpful as it confirms what I had earlier posted regarding the defibrilator case - the orange case in photographs posted earlier is likely the Doctors Kit then.

ttcse;
In the recovered gallery section, photo page 92, I assumed the storage bins for metal bins were on the lower part. Now it seems that section would be on top as installed in the aircraft. And then those bent support tabs are attached somewhere near the ceiling. That makes much more sense.
Yes, correct of course, my mistake, (that's my own hindsight bias going from memory. We have to be so careful). The trolleys are stored in the larger section which is the open area without the bottom in this recovered section. The supports are on the top and I have confirmed this in the AMM.

captainflame;
Looks like the RAT door to me.
I thought so too and this drawing from the AMM seems to confirm the shape of the "black line" which we are assuming is the RAT door and not just a manufacturing join in one of the canoes. The RAT is stored in the #4 flap-track canoe on the right wing. Still, I think the photo is vague enough that we cant be sure:

http://i277.photobucket.com/albums/kk76/batcave777/RATDoor_2009-06-19_223514.png

Though AF may be different there are typically only two locations where dual seating arrangements are installed - at the rear galley on both port and starboard sides at R4 and L4, facing forward, (possibly making the seats recovered the starboard set), and forward at L1 (port side) at the main entry just behind the cockpit:

Machaca, ELAC;

Machaca, thanks for your comments - On which flight control we are seeing in the collected wreckage, the elevator, aileron and flap surfaces have been suggested. Both size, (about a meter in width) and the two slots on the right side of the photo may make the most likely control the inboard aileron. As you point out, the 'D' leading edge structure behind the slat is perhaps a bit small for a wing leading edge.

In the AMM I've examined the attach points for the flaps and these attach points are on the bottom leading edge of both inboard and outboard flaps which are then attached to the flap linkage on the flap track itself. I cannot say for sure but there does not appear to be a slot through which the flap and rod attach lug protrudes. Also, with the aileron, there are two distinct slots, and although there is damage in the referenced area in the photo, there appears to be the beginnings of a slot on the right-hand side.

All speculative of course but with data in hand:

http://i277.photobucket.com/albums/kk76/batcave777/A330FlapTracks3-4-5_Ouboard-2009-06.png

and,

http://i277.photobucket.com/albums/kk76/batcave777/A330_inbdFlapTrk_2009-06-20_104905.png

whereas the slots in the visible structure recovered suggest an aileron structure.

http://i277.photobucket.com/albums/kk76/batcave777/A330_Inbd_Ail_2009-06-19_230233.jpg

What appears to be one end of the structure can be seen neaer the end of the oxygen bottle in this photograph thus, given the location of the slots, this may suggest an aileron from the right wing:

http://i277.photobucket.com/albums/kk76/batcave777/A330_AileronDetailsmaller.jpg

ELAC

20th Jun 2009, 19:40

I disagree, as leading edge is of much greater arc and joins to front of wing box. Seeing as the remnant in question appeared to have a complete airfoil (not a Nike like swoosh of the slats), this is much more likely a trailing edge flap. Perhaps it and the canoe were proximal to the recovered spoiler?

Flaps do have rods and notches - they are difficult to see even when fully extended.

Machaca,

I did in fact consider it as being a portion of the flaps initially. What made me reject that identification was the location of the notch for the actuator rod which, as your 3rd photo shows is below the leading edge of the flap, not at it as is seen in the recovered piece. Another consideration was that the flap actuator rod attachment point extrudes significantly and what is visible from the photo I saw (your second one) didn't match that.

Thanks for the first photo. This is one I hadn't seen. My thinking was that the segment was the portion of leading edge underlying the outermost slat from the forward spar forwards. At that point the section is quite a bit narrower than the slat makes the whole wing appear. But, looking at it from this other angle I'm no longer sold on that identification. Perhaps it is the aileron with a notch I've not seen during walk-arounds. If my perception of taper in this other picture is correct I think it would be too small overall to be a segment of the flaps.

Cheers,

ELAC

Golf Charlie Charlie

20th Jun 2009, 19:42

Regarding search time for boxes in previous accidents, the New York Times (for what that may be worth) says today that in the case of the South African 747 one box was found no less than 14 months after the crash, while in the Adam Air 737 case they were found after 6 months.

deSitter

20th Jun 2009, 19:52

Nice work PJ2, do you have diagrams of the elevator panels?

-drl

tquehl

20th Jun 2009, 19:59

All,

We are now talking about flat spins in chimpmunks, (sic), Pitts, etc. What a concept and I truly believe this has absolutely no bearing on this accident investigation. Why not go to inverted flat spins and carry this out to an additional ridiculous tangent? Spin/flat spin/inverted flat spin in an A330?

There are many posts on this thread. Some good and some bad. Many are intriguing and cause thought and research. It is extremely hard for me to believe that the aircraft broke up inflight unless it suffered the loss of a primary flight control or flight control supporting structure. If the VS did leave the airframe due to ultimate load limit being exceded, then other factors must have been contributing. If the crew in this tragedy experienced an overspeed past MMO/VMO, or slowed intentionally or unintentionally to below buffet boundry 1.5/1.3 this could be a contributing cause.

228 people lost their lives and this is not about conspiricy, national pride, Boeing vs Airbus or any other insignificant arguement. The investigation will determine the cause and hopefully direct the modification and or re-design of operational parameters for the airframe. This has been sucessfully accomplished after the accidents of most, if not all modern aircraft.

Over

Mr Optimistic

20th Jun 2009, 20:02

"Grunf, I think you are referring to the design criteria of a failure per 10^9 operational hours. You also mention probability. The probability of all three ADIRU's failing at once is more like 10^9*10^9*10^9!"

in the analysis of failure probability for redundant systems the concept of common mode failure is used (in the nuclear industry at least) to invalidate the chaining of probabilities such as suggested here. This recognises that anaysis is based on a 'model' of reality, not reality itself. I gather reading this forum that such failures have been observed rarely but often enough that 1 in 10^9 itself seems 'optimistic'.

Incidentally, one common mode failure path could be disruption through excessive vibration, I seem to remember reading that another version of similar equipment (different make) had problems in this respect before (hitting a mounting shelf ?) for which the existence of ISIS was a mitigating factor.

With all the systems which rely on inertial sensors, I would think that it is not possible to reset these in conditions where the aircraft is experiencing angular or linear accelerations above quite small limits owing to the need to find the direction of true 'g'.

Graybeard

20th Jun 2009, 20:11

".. ridiculous tangent? Spin/flat spin/inverted flat spin in an A330? ...

I would humbly suggest everyone take their training/experience/prejudice on the topic to Techlog if you don't like reading it here.

GB

PJ2

20th Jun 2009, 20:14

The elevator has 7 attach points:

http://i277.photobucket.com/albums/kk76/batcave777/A330ElevatorAttachPoints_2009-06-20.jpg

PJ2

20th Jun 2009, 20:31

tquehl;
It is extremely hard for me to believe that the aircraft broke up inflight
Yes, I know.

I have been wrestling with experience vice engineering data - the airplane is extremely strong; is the center of a fully-developed ITCZ CuNim cell capable of breaking an airplane? I think it is fully capable of overstressing the structure to the breaking point regardless of what the crew does with the airplane but I am not an engineer so cannot show or discuss the forces involved.

The available evidence (photos of the wreckage collected thus far) indicates that very large sections of the interior cabin structure were ejected/broke loose and were not deformed by collision with or otherwise inhibited in their trajectories by the fuselage structure. That means either very large sections of fuselage were missing in an otherwise somewhat-intact main structure (like the BOAC B707 over Fuji), or a more complete in-flight failure of all major structures.

The alternative of a low-speed ditching has been suggested but if we think about that, that means the aircraft would have had to have been under control until that point - we could expect at least further ACARS messages in such circumstances (assuming electrical power available and the pitot/ADIRU issues resolved somehow - very doubtful but possible).

The alternative of a "flat" spin suggesting a "pancake" impact has also been made, (the theory which has suggested that that is how the VS broke loose - upon impact, "tilting back" and fracturing the rudder structure - I submit that the tail-structure in that area is far more frangible and would do little to no damage to a vertical structure striking it from the top), but such an impact would result in very high 'g' loading so one would expect far more deformation of the larger parts seen in the photographs than is evident.

There is a common aspect to everything seen in the photographs; the structures are all relatively light in comparison to wings/engines/horizontal stabs/landing gear. Perhaps, like Challenger, (and as has been suggested here by various posters before), these lighter parts fell out of the main, heavier structures which had already failed into a number of open sections and continued to fall for many minutes with a comparatively, (and obviously) benign water-impact.

It has also been suggested that the ACARS trace occurred during the initial accident sequence and perhaps bears no relationship to the Air Caraibe incident - again, we must be cautious of hindsight bias while remaining open to the possibility of the more likely events.

I don't know what part of this is my own hindsight bias and which part is reasonable, so criticism of these kinds of notions is most important.

PJ2

deSitter

20th Jun 2009, 20:49

Thanks PJ2,

On the structure we see an actuator fitting within its recess, and then immediately to the right of it, half the remaining recess with a hinge fitting (failure occurred at the recess itself), then open structure. According to the elevator diagram posted, there are two such adjacent actuator/hinge fittings on each elevator. That, and its size, would seem to indicate that the structure is part of the elevator.

-drl

J.S.

20th Jun 2009, 20:56

What happens to the ISIS before/when reaching its determinated MTBF > 15.000h (usual MTBF for ISIS) ?

-Is it replaced ?
-Is made only a PM ?

Thank you for answers.

ACLS65

20th Jun 2009, 20:59

Very good points PJ2.

Another way of looking at this is from the standpoint of what has not been found. (I know, a lot!)

Engines, not something that will float up, same with CVR, FDR.

But we have also not see obvious cockpit pieces, tail structure other than VS, major wing structure other than slats or spoilers that appear to have been ripped loose.

Also we seem to have a cluster of debris from the forward galley area, but not necessarily other cabin debris in quantity.

I also remember some early descriptions of bodies that sounded like they were for the most part intact, and the later descriptions or almost unidentifiable remains. If the bodies are identified their seating location combined with their condition may prove to be very valuable information.

About all I have been able to conclude is that if I am ever on an A/C in distress I will hold on to the lav door, clutch the defibrillator case, and try to post a msg here as to what happened so you all don't have to wonder :)

PJ2

20th Jun 2009, 20:59

deSitter;
That, and its size, would seem to indicate that the structure is part of the elevator.
Except the attach points on the elevator are more evenly-spaced, while the aileron attach points are close together.

I can't judge the size because I don't know how large the elevator is compared to the aileron, at that point in either structure. An oxygen bottle is about a meter long and that's about how wide the flight control we're examining is.

I don't think it's flap, (because of the two attach points) but further photos are needed before we can determine origin.

I think it is important (especially with so little real evidence) because one outcome is "wing" and the other is "tail feathers".

regards,
PJ2

Been Accounting

20th Jun 2009, 21:11

#2024 PJ2

Can you say whether the galley faces, (bins and trollies), aft or forward?

This galley is forward of Door 2 and opens towards the rear of the aircraft

The photo in #2025 is the rear galley aft of door 4

RetiredF4

20th Jun 2009, 21:17

I´m following the various discussions with great interest.
Concerning the original location of the recovered items of the cabin i suggest it would be very helpful to use a seatplan.

Maybe it´s possible to continously edit the picture by posters with coresponding airctaft knowledge.

Hope the link works!

http://i.slimg.com/seatguru/airlines/Air_France/Air_France_Airbus_A330.gif

http://i.slimg.com/seatguru/airlines/Air_France/Air_France_Airbus_A330.gif

deSitter

20th Jun 2009, 21:21

In your linked diagram, those appear to be hinge points on the outboard side of the outboard aileron. The actuators are on the inboard (wide) side, see the planview photo here:

Airbus A330 - Wikipedia, the free encyclopedia (http://en.wikipedia.org/wiki/Airbus_A330)

http://upload.wikimedia.org/wikipedia/commons/7/74/Cyprus_airways_a330-200_5b-dbs_arp.jpg

They are also external to the structure and reside in aerodynamic nacelles, because one needs a greater moment arm from the actuator than with the elevator, I would imagine.

Barnyard math based on this photo yields a length for the outboard aileron of 16 ft (~5m) while the elevator is about 30 ft (~10m) - this does tend to indicate the aileron but I don't know how much structure is missing in the debris.

The photo mentioned shows what a magnificently beautiful airplane this beast is - sigh :(

-drl

Will Fraser

20th Jun 2009, 21:36

Please note the lack of camber on the piece in the photograph. Elevator needs no 'airfoil', it's work is done by deflecting airflow, as a trim tab would, and 'lift' isn't part of the purpose. It works on both sides of its companion, as the Rudder does with the VS, 'lift' ("airfoil" would inhibit one side, useless for what it needs to do). I will admit scale and perspective are difficult with these photos, but the unit appears to be symmetrical in cross section.

I'll repeat, as I see it, it is the outer and underside tip of the port elevator.

rgds all
Will

JuggleDan

20th Jun 2009, 21:59

ACLS65,

As a matter of fact, it seems that the French BEA is investigating that exact scenario: a major wastewater leak that would have first permeated the a/c structure, then its freezing would have allowed the a/c in-flight breakout.

Article is available in French here:
Vol AF447: une nouvelle piste sur les causes du crash? - Yahoo! Actualités (http://fr.news.yahoo.com/63/20090620/tfr-vol-af447-une-nouvelle-piste-sur-les-019dcf9.html)
(Sorry, had it just translated into English, but it was deleted and I'm too lazy to do it again...)

ACLS65

20th Jun 2009, 22:10

Yep JuggleDan I was replying to your previously deleted post.

Here is a link to the Google translation of that page.

Google Translate (http://translate.google.com/translate?prev=hp&hl=en&js=n&u=http%3A%2F%2Ffr.news.yahoo.com%2F63%2F20090620%2Ftfr-vol-af447-une-nouvelle-piste-sur-les-019dcf9.html&sl=fr&tl=en&history_state0=)

(Perhaps a French speaker can provide a better translation?)

One interesting quote...

"According to "informed sources" quoted in the first edition of the Journal du Dimanche, released Saturday, the Office of Investigations and Analysis (BEA) consider "the track of a leak on a circuit of the wastewater device, reported shortly after take-off by a first automated message warning of the Airbus A330. ""

Perhaps there are ACARS messages prior to what we have seen that may be significant and/or that lav ACARS message might be more significant.

BOAC

20th Jun 2009, 22:32

PJ2

20th Jun 2009, 22:44

Hi BOAC;

Re, "about 200 posts ago", yes, understand what you're saying but in the absence of data from the recorders and the cryptic messages from the ACARS, we are collectively "kicking tin", solving an aircrash the old-fashioned way.

I think most appreciate that there are very good people who are similarly kicking tin in South America from the actual parts and not the photos and could tell us in a few moments what it has taken us 200 posts to explore.

But neither that has occured nor have more definitive photos appeared; I discount the importance of the discussion from the point of view that "we" are not about to solve this mystery because we can't, and if we do it would be by serendipity and not finding "the" 'smoking gun', (to use an atrocious media metaphor). But we can posit what is reasonable to each and let the emerging evidence sort out theory.

Aileron, elevator, flap, leading edge behind the slat?...it doesn't matter. Mid-air breakup or impact of an intact structure with the sea and subsequent breakup - there are arguments to be made everywhere in the absence of experience on this side, and substantive evidence on the other. We're just kicking tin.

Green-dot

20th Jun 2009, 22:44

The alternative of a "flat" spin suggesting a "pancake" impact has also been made, (the theory which has suggested that that is how the VS broke loose - upon impact, "tilting back" and fracturing the rudder structure - I submit that the tail-structure in that area is far more frangible and would do little to no damage to a vertical structure striking it from the top), but such an impact would result in very high 'g' loading so one would expect far more deformation of the larger parts seen in the photographs than is evident.

There is a common aspect to everything seen in the photographs; the structures are all relatively light in comparison to wings/engines/horizontal stabs/landing gear. Perhaps, like Challenger, (and as has been suggested here by various posters before), these lighter parts fell out of the main, heavier structures which had already failed into a number of open sections and continued to fall for many minutes with a comparatively, (and obviously) benign water-impact.

If the aircraft impacted the ocean in a flat spin, just to compare, this is what wreckage looked like in an accident involving a large (500.000lbs) aircraft when losing its vertical stabilizers in a collision. This resulted in a flat spin 16 seconds after losing its verticals. The first link shows the sequence of events in that accident, the second link is a series of images of the wreckage, confined to a relatively small area:

Crash Sequence of the XB-70 (http://area51specialprojects.com/xb-70_crash_sequence.html)

LIFE: Xb-70 Wreckage - Hosted by Google (http://images.google.com/hosted/life/l?q=xb-70,+wreckage&prev=/images%3Fq%3Dxb-70,%2Bwreckage%26ndsp%3D20%26hl%3Dnl%26sa%3DN%26um%3D1&imgurl=e9ccd98b2973a21c)

Subject aircraft impacted the ground relatively flat. Its structure was mainly titanium semimonocoque and stainless steel honeycomb sandwich panels. Therefore, materials can not really be compared with the wreckage of AF447 but it does give an impression of the damage done to an aircraft destroyed in a flat spin.

Green-dot

JuggleDan

20th Jun 2009, 22:52

ACLS65,
I'll give a 2nd pass at French-aided Google translation, and see if the mods let this one live, who knows... :)

A new possibility is under investigation regarding the causes of the crash.
According to an "informed source" quoted in the first edition of this Saturday's Journal du Dimanche (JDD), the French BEA consider "the possibility of a leak in the wastewater conduits, reported shortly after take-off by a first automated message warning from the Airbus A330."
"The leak, undetectable to the crew" may have "caused a prolonged water flow," says Le Journal du Dimanche.Then, when the aircraft hit temperatures below -50°C, the water flow would have produced "a major frost outburst within the composite membrane (half metal, half-composite carbon-type)", explains the weekly newspaper.
This composite membrane is located inside the fuselage at the rear of the aircraft. According to the "informed source" quoted by Le JDD, it would be "frost which caused a sudden in-flight break-out of the aircraft structure."BOAC,
I thought of passing on this piece of information after reading a recent exchange between PJ2 and tquehl: "It is extremely hard for me to believe that the aircraft broke up inflight".
Maybe the water leak could answer that question, maybe not. I'm just the messenger, you guys are the experts... I'll leave it to you from there!

HarryMann

20th Jun 2009, 23:23

Elevator needs no 'airfoil', it's work is done by deflecting airflow, as a trim tab would, and 'lift' isn't part of the purpose.

Not always true Will...

Tailplanes very often have non-symmetrical (cambered) sections. Frequently they have negative cambered foils, as they are optimised for nominal downward trimming loads (c.g ahead of a.c) for stability.

The elevator would be a continuation of this section, containing a modicum of the camberline.

These are supercritical sections, and their aerofoil section also influenced at the root by local flow (note the main wing's section on modern airliner's has a camberline that also looks negative - at the root, due to wing-body interaction)

So whilst I've been out of this business for some years, I'd still bet that tail section is unlikely to be symmetrical. It wouldn't though, at the mid to tip sections, be heavily cambered (negative or positive).

Rgds,

HM

PJ2

20th Jun 2009, 23:31

Green-dot;

Re the damage pattern, yes exactly and that has been my point all along.

My thinking is, the only way in which a large, relatively fragile structure such as the galley that we see is for it to be ejected cleanly, without impediment or collision, from the interior of the cabin and for that to happen, the fuselage itself has to have already broken up at least in that area.

An uncontrolled impact with the water regardless of bank angle, would shatter/fragment most of the interior structure or at least leave substantial tell-tale marks or damage from collision with other parts s they too were ejected during the impact sequence and not simply eject it intact, with all the bins remaining in place and no damage to the support structure.

The galley structure along with many other interior pieces, are what in what we might call "pristine" condition, given the substantial forces involved under the circumstances which all but excludes both a controlled ditching (a highly unlikely scenario) and a flat spin impact with the sea.

I'm not arguing for this or that theory of course - I'm observing and thinking about what is evident and what fits the theory of a substantial mid-air breakup. Such a notion is clearly open for other interpretation providing what we see also fits and can explain both a controlled ditching for example and a galley, floating freely in the ocean. It's possible, just not probable.

ACLS65;
Re PHCs - yes there are 3 PHCs which control the probes and ports heating. The Captain's pitot is powered by the AC ESS bus; the F/O pitot is powered by the AC2 bus and the Standby pitot is powered by the AC1 bus. If the AC1 bus loses power, it can be switched to the AC ESS using controls in the cockpit.

"The PHC tramsmits one ARINC 429 low-speed data bus for fault message to the CMC."

I believe these messages are in turn picked up by the ACARS. I am assuming too that all 3 PHCs would send this kind of message should faults occur within the system or the PHC itself.

On the theory concerning water in the EE Bay from a forward Galley or lavatory overflow, this kind of a problem has occurred - I've seen it - in our case there were no problems. It may cause problems if sufficient water/fluid enters the EE bay (underneath the cockpit) but generally cabin crrew have such a "flood" under control long before much fluid escapes. I can't see water escaping on the ground, unnoticed, freezing, and then causing a problem, (if that is what the poster meant).

Regarding not having "objects from the cockpit", actually the green Oxygen bottle seen in the photographs is from the EE compartment area below the cockpit. There are no other such cylinders on the 330/340:

http://i277.photobucket.com/albums/kk76/batcave777/A340_CockpitCrew_OxygenStorage_2009.jpg

HarryMann

21st Jun 2009, 00:23

Regarding not having "objects from the cockpit", actually the green Oxygen bottle seen in the photographs is from the EE compartment area below the cockpit. There are no other such cylinders on the 330/340:

Mmm...

Do we know how that is fixed in the EE bay, can anyone 'speculate' then, on how that came to be floating about by itself, I presume empty?

augustusjeremy

21st Jun 2009, 00:39

Maybe the interim report will inform in details what exactly all those ACARS messages mean with the corresponding docs, including the lav (which lav ?) message.

I still would like to know what could generate that ISIS fault (comparison with ground speed data from the ADIRUs, and so on) and at what moment (realignment, initialization ?)

I also would like to know if discrepancy between pitots is detected directly or through the ADRUs

kilomikedelta

21st Jun 2009, 00:40

I suspect that if it was still charged with gas to the pressure limit of the cylinder and valve, it would probably still float. Pressurized gas doesn't have a whole lot of mass.

vapilot2004

21st Jun 2009, 01:18

Frozen waste water in the rear of the aircraft caused a loss of Air Data and subsequent maladies?

Sounds rather half baked if you ask me, like something cooked up at The Sun, The National Enquirer (US) or a certain silly online aviation enthusiast's forum (Not PPRuNe) which shall remain nameless.

ZeeDoktor

21st Jun 2009, 01:51

To put numbers on this:

Typical emergency O2 tank pressure is 1850 psi. From

rho = P / (R*T)

follows the density of the pressurized O2 is about 168 kg/m^3, less than a fifth that of water. Add about 30kg of steel for an assumed 40cm diameter by 1m length oxygen cylinder and the combined density of the O2 tank is about 450kg/m^3, or about half the density of water (and even less than that for salt water).

Thus, an O2 tank floating in 30 degree water will float, regardless of whether it is full or empty of pressurized O2.

Cheers

Doc

p51guy

21st Jun 2009, 02:41

I agree, Any leakage in the back would have no effect on pitot, static ports and probes. With no flight data why would they publish this nonsense. I sense they are trying to take attention away from the Airbus and put it on sensors.

ELAC

21st Jun 2009, 02:42

Except the attach points on the elevator are more evenly-spaced, while the aileron attach points are close together.

I can't judge the size because I don't know how large the elevator is compared to the aileron, at that point in either structure. An oxygen bottle is about a meter long and that's about how wide the flight control we're examining is.

I don't think it's flap, (because of the two attach points) but further photos are needed before we can determine origin.

I think it is important (especially with so little real evidence) because one outcome is "wing" and the other is "tail feathers".

regards,
PJ2

PJ2,

Having finally had a chance to look through Dorian's very complete album, there is another image that provides a different angle and much better sense of scale regarding this piece of wreckage.

dd5b67ca-ac0d-4440-ae29-e90158f179b.jpg picture by DorianBanks - Photobucket (http://s635.photobucket.com/albums/uu79/DorianBanks/?action=view&current=dd5b67ca-ac0d-4440-ae29-e90158f179b.jpg)

It is clearly evident in this picture (as was not the case in the other two) that the piece is from one end of a flight control. There is also an object protruding outwards from the piece just below the gentleman's right hand. One can't be certain but it does look like it could be a hinge attachment fitting. If so, this would tend to validate your identification of the two slots and suggest very strongly that this is the outboard portion of an aileron.

If that is what it is then I would I think that it's highly probable that it's from the left wing as opposed to the right. My reasoning for this goes back to the grease mark. In my experience I have only seen such marks on the underside of the airplane (which, granted, is the portion I see far more of on walk arounds). A plausible explanation for this mark would be a small glob of grease falling from one of the slat tracks on the slat forward of the aileron. I'm no expert at how such a glob might move in the airflow but my semi-educated guess is that it is far more likely to strike the under side than the upper side of an aileron. So, if what we are seeing is the bottom of the aileron then the hinge points would be at the left extremity looking forwards and hence on the left wing.

I would also agree that the piece of wreckage besides the flight control is quite likely a portion of a canoe. I don't think that it is a forward portion where the RAT is stowed, however. The size and taper are more in agreement with the aft portion of an outboard canoe. Additionally, the "door" portion of the canoe that houses the RAT has a rounded as opposed to squared off appearance.

Knowing which sides of the aircraft that the canoe and aileron (if that's what they are) come from along with where they were picked up could provide significant information about the proximity of the major components at the time the aircraft hit the water. Unfortunately what we see is insufficient to draw a conclusion, but it's a fair bet that those on the scene do know and that these particular bits will help forward the investigation.

ELAC

PJ2

21st Jun 2009, 03:37

ELAC;

Thanks, I hadn't seen that photo either, so that helps with "scale". I think your assessment of the grease marks is as reasonable as anything we can put forward and would agree.

On the canoe, re not being the RAT door, yes, concur - I think the black line is a sealant not a door.

There is a photo that came out right at the beginning when the photo of the galley showed up (around the 7th) that showed a lot of wiring and air conditioning ducting. I think that's from the cockpit area as well - that's where a lot of complex airconditioning ducting is found, in the overhead ceiling area. Again, entirely a guess; there's a honeycomb hatch-structure in the background upon which all this material is either resting or attached. There also appears to be a "stringer" structure in the darkened background but this may just be the ground cloth etc the material is lying upon - imagination is okay so long as its controlled.

The exercise of identifying parts is worthwhile even here; we just never know.

HarryMan;
Do we know how that is fixed in the EE bay,
Yes, I have the AMM drawings of the actual installation but didn't want to put too much detail here. The schematic I provided is quite specific on where it is mounted (right below the F/O's seat location) but not how - it's strapped down, longitudinal axis perpendicular to the longitudinal axis of the aircraft, with the valve/regulator facing starboard.

ELAC

21st Jun 2009, 04:01

Folks - since the last pics were posted by Damian (excellent work, by the way) we have had a frenzy of "it's this bit!"/"no it isn't it's this bit"/"it broke off this way" - coming up now to 200 posts.

May I suggest:

There are investigators in S America who can readily determine what and how - far better than anyone here from a photo

All this incredible knowledge and intellect (I am NOT being facetious) SHOULD surely be directed at what started all this?.

Greetings BOAC,

Both true points, however, the problem is that there is just not sufficient information in the public realm to build any single supportable hypothesis about what started this yet. I can easily envision at least 3 different scenarios all with different implications regarding the aircraft, crew and meteorological conditions. I'm sure others could imagine more, many without any foundation in the known facts. Until some fact leading to a more supportable conclusion than that a VSC message implies that frozen water was the culprit (for Pete's sake!) shows up I think we're at a bit of a standstill. We can posit til we're blue in the face but we can prove almost nothing.

It appears that for at least 4 minutes it was phoning home to mother about all the things that were going wrong. No 'sudden' breakup/explosive lightning bolt. The apparent loss of successive bits of important flight information must surely be a one-way street to loss of control with an inevitable consequence, given the time, altitude and weather conditions.

To my eye these would be assumptions we cannot prove. The flight data (airspeed) that we are reasonably certain was lost is that which would have been the most erratic and thus least useful in the conditions that we assume existed at the time. If the conditions were more stable and hence a loss of airspeed information more immediately meaningful then the likliehood of a loss of control would also have been diminished. There is not as yet anything to suggest that all attitude information was lost, which would be the information without which a loss of control would most likely be precipitated. So, if we suppose, as seems likely, that a loss of control did occur, then the big question relating to the ACARS messages is which came first, the instrument failures or the loss of control? Did the autoflight disconnect from an unusual attitude or erroneous airspeed data? Did one precede the other while trying to avoid CBs or did they both happen coincidentally after accidentally penetrating a CB? Without some facts in support how can we engage in any informed speculation?

Where 'this bit and that bit came from and how' FROM HERE - after 20+ days of wreckage release and drift is not actually going to do anything for the investigation. I believe attention to systems, software and pilot training/drills is where we should focus. A thorough understanding of the way these apparent failures interact and are handled would be useful.

Not entirely true. For those of us without degrees in ocean current analysis, or what have you, I'd agree that there's little we might conclude from this having been found here and that being found over there. But, the reverse is not true. Suppose that in the recovered fragments you found significant components of the left and right wings still floating in close proximity? And/or with elements of the tail and vertical stab? Granted almost none of us are trained in accident investgation, but co-location of such components would be strong evidence that the relative sections of the aircraft were close to each other, or possibly attached to each other at impact. So the discussion is worth pursuing in that lacking other information it might give some clue as to what sort of break-up occurred. And, unlike the as yet unprovable theories of how the event originated, the identification of components from the released pictures is something where known facts can be found to support or disprove what is theorized.

None of this discussion is going to solve this sad mystery in itself, but at least when we start talking about things that we can attempt to prove or disprove with available information we are using this forum more as it was intended as opposed to pushing out various personal theories that have no supporting facts.

Cheers,

ELAC

Minimbah

21st Jun 2009, 05:12

ABC Australia reporting that the search for parts and bodies has been scaled back.
Air France search scaled back - ABC News (Australian Broadcasting Corporation) (http://www.abc.net.au/news/stories/2009/06/21/2604008.htm)

Peter-1959

21st Jun 2009, 05:57

The ATSB report of VH-QPA, 7 oct 2008 describes the intermittent failure of ADIRU 1 during cruise, leading to sudden elevator movement of 10 degs, initiating a pitch down of 8 degs.
Could there be similarities, purely speculative, a possible impact by EMI because of earlier incidents in approx the same geographic area, near Learmonth, which lead the investigators to consider electromagnetic interference, so now coming back to this accident, heavy thunderstorms, nearby lightning, or even lightning strike. All components should comply with RTCA DO160-D, but if e.g. the ADIRU itself or interfacing wiring doesn't, there's a possibility for inadvertent flight control movement.
And further; there seems to be lack of comparison-check between the 3 different ADIRU's for the flightcontrol systems, FCPC's to base its decisions on. And last ; an ADIRU to cause an elevator deflection during cruise almost beyond the airplane's design limits , I wonder how most more elev deflection the airplane could sustain and where is the PRIM's protection against that.

rgds Peter

takata

21st Jun 2009, 06:35

Hello,

I was waiting for several days that someone more qualified than me would post something about the SAR but, as nothing is comming, I'll post a few maps to illustrate my findings. Beforehand, I would like to point out that the Press releases so far are fairly inacurate or contradictory about what actually happened to AF 447. If you want to cross-check my data below, please, go to the BEA:
Press Releases (http://www.bea.aero/anglaise/actualite/af447/press.releases.html)
Information on Investigation (http://www.bea.aero/anglaise/actualite/af447/information.on.investigation.html)
and FAB
FORÇA AÉREA BRASILEIRA - Asas de um povo soberano (http://www.fab.mil.br/portal/voo447/infograficos.php)

So, we'll start with the few (real) facts we already had on hand and everybody should be able to reconstruct the maps as I did. For doing so, we need to place a few points on Google Earth (coordinates below are in decimal degrees):

1. Waypoints & Flight Plan
During the first hours of June 1st, AF 447 flight on its way between the waypoints INTOL and TASIL:
- INTOL (-1.362, -32.832)
- TASIL (4.005, -29.990)

AF 447 automaticaly reported its position to Air France HQ (ACARS) every ten minutes. This is shown in this BEA Reconstituted flight path which use the same AF ACARS source: http://www.bea.aero/anglaise/actualite/af447/reconstitued.flight.path.jpg
http://takata1940.free.fr/images1/Crashsite0.jpg

2. Last Position, Turbulences & ACARS
The last auto-report was sent at 0210Z and revealed that AF 447 was in cruise (Mach 0.82) at FL350 (BEA).
- Last Position Reported was (2.98, -30.59)

I found this position, page 2, in the 6 June SHOM pdf (Services Hydrographiques et Océanographiques de la Marine,- SHOM, 6 juin 2009) from the French BEA site which is is missing in the English part:
http://www.bea.aero/fr/enquetes/vol.af.447/shom.050609.pdf

I cross-checked this position with the BEA presentations and it is always the same spot matching those coordinates. For unknow reason, the Brazilian "Ultimo reporte" is off 20 km (longitude) in their powerpoints.

Flight AF 447 was on track (slighty west 3NM) at 0210Z heading 30 to TASIL. The throttle settings (AUTO) and speed are showing that it was not in "turbulence" mode (A/THR OFF, Mach 0.80) at 0210Z. It might have encountered turbulences @ 0200Z but it cannot be acertained at the moment: basicaly this point is extrapolated from AF 1st press release:
Press Releases (http://alphasite.airfrance.com/en/s01/press-releases/#communique2537)
The Airbus A330-200, registration F-GZCP, left Rio on 31 May at 7:03pm local time (12:03am in Paris). The aircraft hit a zone of stormy weather with strong turbulence at 2am this morning (universal time), i.e. 4am in Paris. An automatic message was received from the aircraft at 2:14am (4:14am in Paris) indicating a failure in the electric circuit a long way from the coast.From the above press release, it is not clear if the 0200Z situation was:
a) an AF assumption because of the meteo reports;
b) an AF assumption extracted from the automated ACARS which included also the aircraft performances beside the positional report @ 0200Z.

So far, I tend to believe that no direct report was made by the crew about the so-called "fortes turbulences". Never, so far, the BEA did imply such a fact. Beside, this AF press release is counterfactual about the "electrical failure" pointed by the ACARS as this problem wasn't reported.

Between 0210Z and 0214Z, 24 maintenance ACARS were automaticaly sent by AF447. They are time stamped but unsorted in the listing. Those maintenance ACARS do not transmit the position of the aircraft => The real position of F-GZCP @ 0214Z is unknow. By flightplan extrapolation, She should have been about 31 NM further on its way to TASIL (3.432, -30.328).

http://takata1940.free.fr/images1/Crashsite1.jpg

5. Search & Rescue, recovery:
The BEA pdf "Sea search operations" page 10: http://www.bea.aero/anglaise/actualite/af447/sea.search.ops.pdf
is showing a clear partern of the body distribution (red dots) along a South-North axis. By Picking up the northern red dot each day, it would give roughly the daily speed of the sea drift.
- 6 June (3.566, -30.458) -> dot 06
- 7 June (3.810, -30.485) -> dot 07
- 8 June (4.045, -30.460) -> dot 08
- 9 June (4.415, -30.525) -> dot 09
- 10 June (4.730, -30.505) -> dot 10

There is no need [for this presentation purpose] to consider the other parts of the aircraft as the bodies are very unlikely to be affected by the wind like the other materials from the airframe. From d06 to d10, the distance is about 128 km covered in four days, then an average drift of 0.37 m/sec, with an heading of 357.

A quick Verification using the SHOM pdf linked above, there is a map, page 8, of the surface drift forecast (right map) in this area (red circle) for the period 6-12 June. The right gauge indicate that a 0.40+ m/s drift is pretty close and the heading is also North. Then, this rounded measure is correct enough for an estimation.

7. Probable Crash Zone
By doing an extrapolation of the surface sea drift allocated to the period following the crash we'll find that the recovery of d06 is dated ~0900Z on 6 June. Hence, 126.5 hrs after the crash. During this period, assuming that the drift speed (0.40 m/s) and heading (357) was the same, the bodies would have covered 182 km, which point to about 100 NM south of d06.
- Estimated Crash Area (1.915, -30.390)

http://takata1940.free.fr/images1/Crashsite2.jpg

8. Hypothesis and course post 0214Z
Considering that d06 was recovered ~127 hrs after the supposed crash hour at only 11 nautic miles North-West of the planned AF 447 position @ 0214Z, it seems very unlikely that F-GZCP crashed anywhere near this place if one is taking into account the surface speed of the drift and its heading to the North. The pattern of the recovered bodies seems very regular and is pointing directly to the crash zone and its surroundings.

The only possibility for F-GZCP to reach such a remote place, considering her flightplan, was to make a 180 turn East. After having suffered whaterver crippling instrumentation/airframe damages due to the weather or other cause. Her crew certainly attempted to divert for an emergency landing solution in Fernando de Noronha, at any time past 0210Z (last know position).

Depending at what time the Captain decided to turn back (even after 0214Z as, from the ACARS, no catastrophic failures occured until then), the flight might have lasted much longer than we thought, involving a cripled aircraft unable to keep its cruise speed and certainly also its altitude.

Therefore, the lack of subsequent ACARS past 0214Z should be considered as the consequence of the failure/destruction of the SATCOM system (i.e. ice/hail) rather than the consequence of an immediate fatal dive/break up. Moreover, F-GZCP had a COM3 inop when she took off from Rio. This was revealed by Eurocockpit.com yesterday.

Of course, this is a rough estimation of either the crash zone and the real weather situation. A much more detailed analysis of the actual surface sea drift between 1-6 june would be very usefull to confirm this estimation but, even if the drift was actually half my figure (it may also be twice), AFF 447 would have turned back and not crashed close to 2014Z. I tend to believe also that such a move concured a lot to the delays before the SAR could find any wreckage close to the supposed crash zone around TASIL as they reached it only 5 days later.

S~
Olivier

Pontius Navigator

21st Jun 2009, 07:47

There is no need to consider the other parts of the aircraft as the bodies are very unlikely to be affected by the wind like the other materials from the airframe.

. . .

The only possibility for F-GZCP to reach such a remote place, considering her flightplan, was to make a 180 turn East. [/QUOTE]

I agree.

I made a similar assumption some days ago, either in a PM or on the board. The only thing I would add is:

a second plot analysis for wind affected debris might give an additional coroborating position.

FlexibleResponse

21st Jun 2009, 07:49

takata,

Very good investigative work! I think you also need to take into account that the human body does not float in seawater until about 72 hours after death. Then there should be some allowance for the time from establishing positive buoyancy to float from the wreckage at the bottom of the ocean up to the surface. Again some allowance should be made for the average sea current during the ascent time (unlikely to be the same as the surface current). So finally it could be possible that the surface drift time before discovery of the first body could have been less than 24 hours...this might put the wreckage very close to the flight plan track?

BOAC

21st Jun 2009, 07:59

takata - congratulations on some excellent and thorough work. I did ask 'the experts' a day or so ago which choice - continuing ahead or turning back through the weather - would be likely vis a vis ETOPS/fuel etc. and have had nothing back. Instinctively I think your 'turn-back' looks like a very likely scenario. Even your projected radius of turn is plausible.

I'd still like to focus on what was LEFT on the instrument panel for them following the known ACARS messages; whether there is some common electrical supply issue that could have caused these failures and whether there is a common 'software junction' in it all. What did the ?Air Caribe? have left in their incident? I cannot find any detailed report.

DingerX

21st Jun 2009, 08:09

Very impressive Takata, but your plot assumes that bodies float on the surface. My understanding is that the buoyancy of bodies is highly variable, depending on the quantity of air in the lungs, and the ambient temperature. Often we sink, then, as our bacteria in the digestive tract and elsewhere continue their work, and generate gases, we float. So bodies aren't necessarily a faithful reflection of the surface current either.

--Edit: Flexible Response beat me to it!--

FlexibleResponse

21st Jun 2009, 09:15

ttcse

FlexibleResponse
Quote:
Then there should be some allowance for the time from establishing positive buoyancy to float from the wreckage at the bottom of the ocean up to the surface.
About 'the bottom of the ocean' part... given the extreme depths in that area and how much air cavities (& their gasses) compress and crush going down there, I would wonder whether those victims who actually went to the bottom could be refloated even after 72hours. For what it's worth.

I think that ttcse is correct. Beyond a certain depth, a body is unlikely to ever return to the surface, due to the very high pressure exerted by the depth of water.

RatherBeFlying

21st Jun 2009, 11:47

Here in Ontario, Lk. Superior is known to keep its bodies because cold water temperature prevents gas formation.

Bodies do tend to come up in Lk. Ontario when drowned near shore, but those lost well offshore may not. One sad Spring night several youths took a small overloaded boat out of a marina for a joyride. Not a single body ever came up -- even though the Ontario and New York shorelines were thoroughly searched and several items from the lost boat were washed to the NY shore.

As to the present case, people strapped in in fuselage sections that went to the bottom would be in water too cold for gases to form.

In the cases of bodies that have appeared on the surface, I suspect that they initially sank to an intermediate depth where the temperature was warm enough for gases to form.

lhp

21st Jun 2009, 13:42

@takata

Nice work putting together the graphics. In your analysis you discounted the effects of wind condition on the drift of the bodies. I think this is a miscalculation. While wind conditions won't affect flat objects in the sea, like it would a floating galley for example, the wind does influence the wave action. Wave action, in addition to current, would determine direction and distance of floating objects. Thus, your conclusion of probable crash site can't really by determined from the floating bodies position at time of recovery. Theoretically, a computer program might be able to calculate float direction and speed from many samples of wind direction/force and current condition over 5 days. Often the wind direction is opposite of current, creating different wave forces then when wind and current are in the same direction. It's too complicated a model to figure out without the help of supercomputers.

lomapaseo

21st Jun 2009, 13:49

ELAC

but it's a fair bet that those on the scene do know and that these particular bits will help forward the investigation.

ELAC

Amen to that.

That's the fullfillment of being on scene. Most of the questions raised in two days of posts here (and stilll unanswered) could be answered in an hour on scene

None of this discussion is going to solve this sad mystery in itself, but at least when we start talking about things that we can attempt to prove or disprove with available information we are using this forum more as it was intended as opposed to pushing out various personal theories that have no supporting facts.

Cheers,

ELAC

Spot on, but it probably helps us to reject ignorant speculation from others that doesn't fit the little facts we can discern. Ask yourself how many times have we been asked by our friends if this or that they heard or read is true. Do we just brush them off even as friends or do we throw a fact or two at them to politely end their speculation.

ttcse

One aspect of this I can't get over, and others have also noted, is how the debris seems to come from a very-low speed impact. Perhaps if your analysis bears truth it can ultimately be tied to why the debris is in the condition its in.

So far it only suggests that the pieces cited impacted the ocean in relatively free fall.

Most of the discussions in the last days have been about the tertiary events and understandably not about the missing links.

Pontius Navigator

21st Jun 2009, 13:51

lhp, initially I agreed with you. Then I realised that takata was actually comparing like with like. He has a series of plots all similarly affected by current and modified by wind driven surface motion.

By taking one common type of object then he was able to deduce the combined drift rate. He parts of the wreckage been included then the plot would not have been as pure. It would be possible to do a similar exercise with debris alone and compare the two plots.

BOAC

21st Jun 2009, 15:01

RECAP!! Do we ACTUALLY have any 'ACARS message' about waste water other than a possible vaccuum flush fault?

wilyflier

21st Jun 2009, 15:08

My previoous analysis was that the drift increased from day 6 to day 16 thus the trend was much less in the first few days leaving impact somewhere between last report point and 50 odd km north north east of it, (if there was no turn to reciprocal followed by ditching)

SaturnV

21st Jun 2009, 15:26

Takata, in your analysis of the sequence and date of body recoveries, and the drift of the bodies over time (from ocean currents and/or prevailing surface winds) how do you account for the significant amount of wreckage found on June 7 that is well to the west and south of the bodies recovered that day?

http://i136.photobucket.com/albums/q199/tahitiplage/SNAG-0289.jpg

The bodies (and wreckage) recovered on subsequent days suggest a northward drift, which then became NW or W last week.

Of course, if one had the index to which parts of the plane (or passenger items) the white numbered dots represent, that would probably help. To date, neither BEA nor Brazil have released even the approximate coordinates for where significant parts of the wreckage were retrieved. What was retrieved at dots numbered 9 and 11 may be key. There are no items numbered 1-8, or 10, that I have seen on the BEA charts. There is a 12. It may be that the early numbers were assigned to flotsam mistakenly identified as being from AF447 in the earliest days of the search. Either that, or only the larger items are shown on the maps that were publicly released.

Dutch Bru

21st Jun 2009, 15:55

According to my info fixed oxygen cylinders on the A330 are to be found in one other place than the one below the flightdeck reffered to: for cabin oxigen supply there are 5 + up to another 5 cylinders situated in the r/h sidewall lining in the fwrd cargo hold (source FCOM 1.35.20)

jimpy1979uk

21st Jun 2009, 15:57

PJ2,

Just to let you know gaseous oxygen is an option for the passenger system on the A340/340 aircraft. If it is fitted the oxygen bottles are fitted aft of the forward cargo door behind the cargo bay lining panels. I have worked on aircraft with generators and the gaseous systems fitted. I am not aware of what option Air France have on their longhaul airbus fleet.

Jimpy

Dutch Bru

21st Jun 2009, 16:15

AF uses gas cylinders for passenger oxigen. See also the accident report on AF's A340 F-GLZQ: http://www.tsb.gc.ca/eng/rapports-reports/aviation/2005/a05h0002/a05h0002.pdf

"The forward cargo door was found in place and closed. Next to the cargo
door, coincident with the location of the passenger oxygen cylinders, was a large hole in the fuselage wall that bore signs of explosive force."

PJ2

21st Jun 2009, 17:15

BOAC;

RECAP!! Do we ACTUALLY have any 'ACARS message' about waste water other than a possible vaccuum flush fault?
No, we do not.

The early, Class 3 maintenance message "LAV" would have nothing to do with any water/overflow issues. The "waste-water" theory is emerging in the absence of evidence. It is almost certainly a non-starter.
I did ask 'the experts' a day or so ago which choice - continuing ahead or turning back through the weather - would be likely vis a vis ETOPS/fuel etc. and have had nothing back. . . . if only we could be sure we have ALL the ACARS messages in sight, then there is a very good chance we can rule out a high-level break-up in view of the lack of an "EXCESS CABIN ALT" warning as you say. Without such a signalled disruption to the cabin pressure, the only subsequent ACARS message must surely refer to a descent? Dare we surmise? PJ? Bsieker?
If that's the post, I do have a few comments:

Your post doesn't ask about a turn-back, it only suggests the possibility of a descent as may be indicated by other, associated messages (which, as you say, may or may not exist).

First, a bit of background on the ACARS ADVISORY message:

When a system requires the crews' attention but isn't serious enough to warrant formal master caution/master warning indications plus other associated ECAM actions, the affected system's page is brought up on the lower ECAM, displaying a pulsing green 'advisory' message of the relevant parameter, in this case, the "CAB PRESS" text display on the pressurization page.

An advisory message to the crew does not necessarily require immediate crew action but advises that system parameters have been exceeded, not by a lot but crew attention/advice is required.

The ACARS cabin "ADVISORY" message refers to a pulsing, advisory 'CAB PRESS' message displayed on the pressurization system page on the lower ECAM display.

From the QRH, the "CAB PRESS" pulsing green ADVISORY message is displayed when:

- the cabin vertical speed is greater than, or less than, 1800fpm, or
- the cabin altitude is equal to/greater than 8800ft, or
- the cabin differential pressure (delta 'p') is equal to/greater than 1.5psi in flight phase 7.

Flight phase 7 refers to the approach phase below 800ft AGL to touchdown.

In each case of an advisory message there are recommended crew actions listed in the QRH. These action items are, as described, not of an emergency nature.

That doesn't mean that there was no emergency - it just means we don't have further ACARS messages.

This in and of itself wouldn't indicate either an aircraft descent or a turn-back. There would not likely be any ACARS messages associated with a descent per se, (and certainly none automatically associated with a turn-back), although a "top of descent" message may be automatically sent as a matter of individual airline ACARS design requirements - we don't know.

I understand the theory behind takata's fine sleuthing work and hope it bears fruit but as we can see, there are other considerations emerging which may place the aircraft differently. Very frustrating for all, as we know.

It is impossible to say whether the crew turned back and it is equally impossible to say whether such a decision would have been made or by who, (depending upon who was on the flight deck - we may surmise it would "definitely" be the captain, but we do not know this).

A turn-back at that point in the flight is a highly unusual response - it is essentially an emergency response made out of dire need. I can see a 90deg turn to the east to avoid Cu but that is not the maneuver contemplated by takata's work.

Clearly, there is a pattern, though "the data" is granular in nature, to where the wreckage and most bodies were found. I think takata's work is on the right track and (as I believe s/he him/herself would say), may require fine-tuning. I wouldn't be prepared to say there was a turn-back but such a notion teased as it is out of what little is known, is now a factor which must be proven/discounted.

PJ2

21st Jun 2009, 18:26

Pontius Navigator;
As I suggested earlier, back plotting these with known surface wind correction would be interesting.
The key is what takata and many other are attempting to do: pinpoint the actual position of the aircraft at 02:10. All the rest flows from that one piece of information.

A normal descent for a normal approach/landing will take, at the most, 160nm, with an average of 130nm. That would be the greatest "radius of action" in terms of the location of the larger sections.

The winds in the ITCZ are almost always light and variable, 20, 30kts or so, at cruise altitudes; the "polar jet" and "subtropical jet" are further north and south.

Somewhere, someone has these winds for the time of the accident.

Falling (not in a high speed dive) parts would be exposed to such winds for between roughly three and ten minutes depending upon weight and shape Even assuming the winds remain at, say, thirty knots all the way to the surface, (not likely), that translates to very little lateral distance covered in relation to the position of the aircraft at 02:10 - less than 20nm.

Ocean currents, as has been posited, will account for the rest of the displacement from original impact point. I don't think it takes a supercomputer to work out a reasonable radius of action - it takes the assumptions as demonstrated by takata's methods, which may then be fine-tuned by the availability of more accurate data, (as is stated).

Dutch Bru, jimpy1979uk;

Thanks, I wasn't aware there is an oxygen-tank option for the 340/330 - it isn't in the AMM - only the O2 generation system and the F/A portable O2 installations are.
AF uses gas cylinders for passenger oxigen.
To be completely accurate, we could reasonably assume from the quoted report, (thank you) that AF 340's have the cylinders installed. We need to confirm that the same installation has been made for the 330's.

In any case, the statement that the oxygen cylinder in the photograph IS from the cockpit, is in question.

ACLS65;
Can anyone confirm that there was a lav on AF447 port side directly aft of the cockpit? This could also refer to more than just a clogged lav, since it supposedly happened shortly after takeoff it might even be from before the lavs were in use.

. . . .

Also per the FCOM the vacuum generators for removing the wastewater are not used at altitudes greater than 16,000 feet. Could the lav ACARS msg we do have mean that this system "thought" it was at a lower altitude and tried to use the vacuum generators? Again maybe just another indicator that however this system receives altitude info was compromised.

Again all this assumes there is some validity to the original article and as of last night I had not found that same information anywhere else
The key in this is the statement, "all this assumes"; this is precisely the situation where a little knowledge (and no experience on the airplane), is a dangerous thing. I think it is wise to suspend judgement on this risk and wait.

As I said a few pages ago, backups/flooding can and do occur on rare occasion but the F/As are very swift at ensuring any such overflow of water is mopped up. The concern is that the forward lavs or galleys may leak into the EE Bay under the cockpit. The concern is legitimate in the sense that it is good, cautious thinking on board an aircraft but is not legitimate in the sense that there is a problem with flooding the EE Bay.

The "LAV" ACARS message is not an indication of an abnormality; it is a Class 3 maintenance message which appears quite often, usually in climb.

The comments about "16,000ft altitude" and vacuum generators being driven by a system that "thought it was below 16,000ft", while possible, is highly improbable and would not result in flooding anyway. A galley tap which can't be shut off is more likely a source of local (I emphasize "local" to forestall further launching in new directions!), flooding.

rer47

21st Jun 2009, 18:33

Very nice work, takata.

I was unaware of the SHOM PowerPoint and had not seen the last known position coordinates in print before - thanks. I have spent a bit of time re-plotting the BEA reconstructed flight path, and it is reassuring that the locations derived from reverse engineering the BEA flight path points in Google Earth are a good match for the GPS coordinates obtained from the ACARS 0210Z position report. My earlier analysis is in the Tim Vasquez forum:

Jet Crash forum • View topic - Crash location (http://www.jetcrashforum.com/viewtopic.php?f=1&t=3)

The BEA flight path points are labeled to the second (e.g. 02:00:00, probably the GPS time stamp), so that by reproducing the coordinates of the points in Google Earth by moving them around until they are a good match for details in the sea floor bathymetry in BEA plot it is possible to calculate speed and distance covered. Here is the result:

0120Z - 2° 48.000'S, 33° 35.000'W

77.0 nm, 462 kt

0130Z - 1° 39.500'S, 32° 59.000'W

78.9 nm, 473 kt

0140Z - 0° 29.421'S, 32° 22.002'W

78.0 nm, 468 kt

0150Z - 0° 39.943'N, 31° 45.392'W

77.4 nm, 464 kt

0200Z - 1° 48.654'N, 31° 9.109'W

77.2 nm, 463 kt

0210Z - last known position - 2° 58.700'N, 30° 35.800'W

This last known position derived independently by reverse engineering the BEA plot is 2.978°N, 30.597°W in decimal degrees, a very close match to the location published by SHOM of 2.98°N, 30.59° W, which gives me some confidence in the others. Incidentally, this position is also a good match for the last reported position plotted on one of the actual FAB SAR maps (the red airplane symbol) used by the search teams, as I posted earlier:

http://www.pprune.org/rumours-news/376433-af447-32.html#post4982877

Variation in the locations, and therefore the speeds and distances, could well be due to small errors in reproducing the BEA flight path positions in Google Earth. It will be interesting to see how close these are to the actual GPS coordinates that hopefully will be in the BEA accident report. The calculated speed between 0120Z and 0140Z is 468 kt, a match for the ground speed of 467 kt calculated by Barry Carlson from the flight plan speed of Mach 0.82:

Final Route of AF447 - an Analysis (http://countjustonce.com/af447/)

One key result of my analysis is that there evidently was no change in speed prior to 0210Z.

Much of takata's analysis concerns the speed and direction of the drifting debris. There are actually some data on drift in this area that cover the period 1 June to 6 June from oceanographic drifting buoys, although they are not all that close to the last known position. Buoy 31919 NE of TASIL drifted 46 nm to the WSW during this period, and Buoy 31256 SE of TASIL drifted 64 nm to the NE. These are more or less in opposite directions, so I suspect the actual drift of the debris during this period was complex.

Attachments do not seem to be enabled, or I would post a map of the flight path and drifter buoy paths. You can view it in the Sun Jun 21, 2009 6:53 pm post at:

Jet Crash forum • View topic - Crash location (http://www.jetcrashforum.com/viewtopic.php?f=1&t=3&start=0)

-rer47

augustusjeremy

21st Jun 2009, 18:34

Hyperveloce

21st Jun 2009, 18:54

From the plots in the qantas preliminary report

http://www.atsb.gov.au/publications/investigation_reports/2008/AAIR/pdf/AO2008070_prelim.pdf

a) IR1 fault is detected but the A/C continues in normal law even after master PRIM becomes PRIM2;

b) When master PRIM is switched back to PRIM1 then the a/c gets into Alternate Law and the crew regains control.

yes and the IR1 fault was 10 sec. before the ADR1 failure for Qantas, the two faults maybe within the same minute for the AF447. why this IR1 is declared faulty whereas it is its ADR part which is providing erroneous AoA measurements ? where there identified problems with the gyrometers/accelerometers afterward ? It does not appear so.
Jeff

mercurydancer

21st Jun 2009, 19:32

A body may well float in such circumstances from the first moment of immersion. The ocean temperature will be a factor. The sea was warm (in the region of the high 20s C) so flotation would be possible. A cold body will generally sink and stay down, often in the same area where it was immersed.

Assistance of flotation support, whether intentional or not, may be a factor in the bodies remaining on the surface.

aguadalte

21st Jun 2009, 19:37

yes and the IR1 fault was 10 sec. before the ADR1 failure for Qantas, the two faults maybe within the same minute for the AF447. why this IR1 is declared faulty whereas it is its ADR part which is providing erroneous AoA measurements ? where there identified problems with the gyrometers/accelerometers afterward ? It does not appear so.
Jeff

Part Of OEB nº 74/4:

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

wes_wall

21st Jun 2009, 20:10

The comments on what has been recovered is interesting, but what is more interesting is – what has not been found. I have not seen many seats, seat cushions, personal effects, clothing, carry on items, cargo items and/or baggage, the myriad of paper items which are carried on the airplane, galley stores, and the list goes on. All this leads me to believe that the airplane entered the water mostly intact. This is further supported by the lack of a large debris field. Had there been a catastrophic failure at altitude, one could expect a long trail of items floating on the surface. To my knowledge, this has not been the case.

The discovery of the L2_R2 galley in good condition, along with the lav door, the FA seats, the defibrillator, could indicate a failure of the fuselage upon contact with the water. Other than the VS, it seems most of the items recovered have come from around the L2 interior and exterior section. The airplane could be in two separate parts which could account for some of these items retrieved in good condition.

BOAC

21st Jun 2009, 20:30

is this the post you're referring to,Actually post #1949, PJ:A different question for those familiar with that route. From what I can make out in terms of where this might have occurred in the actual area of CB activity, carrying on on track after 'recovery' from whatever happened would appear to be a better weather solution than turning back to the mainland. How would this fit with a possible 'worse case' ETOPS fuel scenario ie low level 2 engines vis a vis ETPs and excess fuel, and where would ETP 1 be likely to be in relation to TASIL?

quaeler

21st Jun 2009, 20:31

The discovery of the L2_R2 galley in good condition, along with the lav door, the FA seats, the defibrillator, could indicate a failure of the fuselage upon contact with the water. Other than the VS, it seems most of the items recovered have come from around the L2 interior and exterior section. The airplane could be in two separate parts which could account for some of these items retrieved in good condition.

They recovered the overhead baggage facade for the right rear of the cabin, as well; i haven't seen any photos of something i could recognize from the middle space of the cabin.

hautemude

21st Jun 2009, 20:52

There have been one or two comments about the containers and the condition of the galley being in good order.

I'm sure that many of you know how relatively feeble is the retaining mechanism for the galley containers, how often they come adrift even after a gentle landing. Imagine then, an aircraft breaking up in flight, various parts spinning and spiralling into the sea, almost inevitably with considerable force; would those containers really still be there, in their stowed position? It would also be interesting to know how much water they contained as it might be indicative of how deep they went before surfacing, and whether the buoyancy of the structure to which they were secured was sufficient to bring them to the surface if they were water logged.

To me it is more suggestive of an aircraft breaking up on contact with the sea.

augustusjeremy

21st Jun 2009, 20:53

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 Aguadalte,

I think I know what you mean.
There are some differences from the qantas 72 case, though.

They kept at normal law almost to the end of the flight. AF447 switched to alternate law.

I still don't understand why the qantas a/c kept in normal law and at the same time it kept using the faulty IR1/ADR1 data.

In alternate law there is still (it seems) some protection regarding AoA but it can be overriden by the crew.

Anyway if AF447 was wrongly informed about their speed the situation would still be dangerous (I suppose).

Can't be sure about the pitots message... Maybe one pitot was already blocked... I don't know how this problem is detected.... Nothing from the Pitot Heat Computer specification to get any valuable hint.

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.

The faulty IR was IR2... Thus even if ADR2 was still providing wrong data, there would be ADR1 and ADR3 in aggreement.

This is complicated stuff...

Hyperveloce

21st Jun 2009, 21:30

if two ADR disagree, as it was the case for the Air Caraïbe and the AF 447 flights, normal law should irreversibly switch to alternate law 2, if one ADR fails, it switches to alternate law 1 with possibility to revert to the normal law, as it was the was for the Qantas flight, 3 seconds after the 2nd pitch down.
this does not explain to me why a faulty ADR (how could it detect itself that it is generating erroneous measures ?) give rise to an IR failure when nothing anomalous is seen with it (does the post flight analysis show any defect with the IRU part ?). but this is nother problem.
it simply suggests that faulty/disagreeing ADRs can trigger IRU fault reports and alternate law 2 (plus a strange behaviour of the autopilot)
Jeff

PJ2

21st Jun 2009, 21:35

quaeler;
They recovered the overhead baggage facade for the right rear of the cabin, as well; i haven't seen any photos of something i could recognize from the middle space of the cabin.
Not sure if this is what you're meaning by "middle space", but the overhead bin structure in this photo is the middle-of-the-cabin structure. There's also the galley which we know about:

http://i277.photobucket.com/albums/kk76/batcave777/WreckageOnBoat3-1.jpg

and,

http://i277.photobucket.com/albums/kk76/batcave777/WreckageOnBoat4pg-1.jpg

ELAC

21st Jun 2009, 21:35

An inference can be drawn from any piece of evidence be it physical, electrronic or witnessed, and I feel that the inference must be compared to the benchmark of "consistency" in that any inference drawn must be consistent with the other inferences from facts already known.

Speculation is fine, (even essesntial) if we maintain scrupulous adherence to consistency then a clearer view will reveal itself.

Mercurydancer,

Perhaps you've misunderstood the thrust of my post, or maybe I simply haven't stated it well enough.

I would agree with you entirely that there are some inferences that can be drawn from each piece of evidence, and I am certainly not suggesting that these should be discounted. However, I would say that the inference drawn has something to do with the level of expertise one has with the particular element of evidence, and this is where different contributors here each bring different strengths (and weaknesses).

Personally I know very little about the failure modes of vertical stabilizers/attach points and know nothing about rates and directions debris spread based on current movement, so I'll leave the inference drawing on such matters to others better educated to do so. When they do, with consistency to known facts, it makes for a valuable contribution to the discussion and that's certainly to be encouraged. It may also provide the rest of us a link back to another area where we can provide informed speculation that comes to make more sense in context with that contribution.

Where things have occasionally gone wrong is with incorrect inferences being elucidated by some without particular expertise and those inferences then being extrapolated into conclusions of what happened and/or why without there being any actual facts to support either the inference or the conclusions derived from it. I'm not sure if you've followed the entire thread but, at times, there have been some pretty strongly worded conclusions about what type of design/system/software/crew error must have been responsible for this accident and how this represents a particular failure of the manufacturer/operator/crew. It's these sort of speculations, not reasoned, informed and factually supportable ones such as you are suggesting, that I have been referring to.

In any event, what I think we all hope to achieve from the discussion is a better understanding of all the circumstances surrounding this accident and how that greater knowledge might improve our own procedures and decisions when confronted with similar circumstances as a pilot/passenger/dispatcher/engineer/designer or even accident investigator.

Cheers,

ELAC

takata

21st Jun 2009, 21:51

Very good investigative work! I think you also need to take into account that the human body does not float in seawater until about 72 hours after death. Then there should be some allowance for the time from establishing positive buoyancy to float from the wreckage at the bottom of the ocean up to the surface. Again some allowance should be made for the average sea current during the ascent time (unlikely to be the same as the surface current). So finally it could be possible that the surface drift time before discovery of the first body could have been less than 24 hours...this might put the wreckage very close to the flight plan track?

Hi and thank you,
This is an interesting point which may indeed affect my estimation but, as a matter of fact, the human body, with a gravity below 1 (clear water), is naturaly buoyant. In salted sea water it is even more buoyant but temperature, etc. are other factors changing sea water density. On the other hand, there will be also some variation with the sex, age, clothes, mutilations, etc. of the victims. So several paramaters will affect how much of the body will surface (very few percent indeed until the decomposition process is well advanced). Next, decomposition at sea will start to increase buoyancy immediately following death. There is no need to wait for 72 hrs, it is just a matter of equilibrum between the body state and the sea density.

S~
Olivier

(Note: I already tried to post twice an answer but my posts didn't show up after many hours. I hope this one will go thru.)

mercurydancer

21st Jun 2009, 21:53

ELAC

Thank you for your comments. Its clear that we agree on many things. It ws my intention to add detail to your comments and not to decry them. My knowledge of such things as vertical stabilisers could be written on the back of a postage stamp, and yes, you are right that some know an awful lot about them. In the main Ive found that the posters who do know what they are talking about have presented things in a clear and unambigous manner.

There have been some very confident explanations as to why AF447 got into difficulties. What seperates the wheat from the chaff in these opinions is whether they are consistent with known facts. The problem being that known facts in this case are very few in number.

Chu Chu

21st Jun 2009, 22:05

For what it's worth, fuselage failure on contact with the water makes more sense to me too. I'd have expected more impact damage to the galley if it was ejected in an in-flight breakup and hit the water in free fall. Maybe someone with real knowledge will explain that it's strong and light enough to survive such an event undamaged, but that doesn't seem too likely on its face.

takata

21st Jun 2009, 22:31

Sorry everybody,
I already answered three times the questions about my maps but the system fails to show my post. Is quoting someone delaying the post?

Thx,
Olivier

(Edit: I guess as this post appeared immediately!)

SaturnV

21st Jun 2009, 22:59

The Brazilian Air Force's site has this plot for retrieval of wreckage on June 3. The Brazilian AF charts for search and recovery effort begin with two maps for early wreckage recovery, one is undated. (The undated chart has wreckage recovery 160 km WNW of Sao Pedro and Sao Paulo.) Both maps look to be overlays on aeronautical charts.

See very large image here:
http://www.fab.mil.br/portal/voo447/INFOGRAFICOS/rota_voo447_2.jpg

However, it seems likely that that the 'wreckage' retrieved on June 3 was ultimately identified as not coming from AF447.

On June 6, the first wreckage definitively identified as coming from AF447 was located at 03.34.08' N and 030.27.30' W. It is a seat cushion (poltrona) with a serial number. A male corpse is also recovered (apparently from the same area as the seat cushion). The position is WNW of the last reported position. Another male body is recovered later in the day from this area.

See Powerpoint slides at:
http://www.fab.mil.br/portal/voo447/INFOGRAFICOS/coletiva_06_06_09.ppt

The Brazilian Air Force site also has a plot for the recovery of the 50th body showing location from where 19 bodies were recovered. On June 20th, ENE of the 50th body, and NNE of Tasil, there was recovery of "Despojo". (Despojo means spoils, might it mean body psrts?)

For plots of the early search and rescue patterns, and weather in the general area from June 6 onward, and for various charts including those cited above, see:
FORÇA AÉREA BRASILEIRA - Asas de um povo soberano (http://www.fab.mil.br/portal/voo447/infograficos.php)

Thanks to takata for finding the link.
____________________________________________
Edited to add this image of a section of a Brazilian Air Force Powerpoint slide indicating where 'wreckage' was located on June 2. I think none of this wreckage was ultimately identified as being from AF447. But it does show what was located in an area takata thinks the plane might have crashed if a 180 turn toward Sao Pedro and Sao Paulo was attempted.

http://i136.photobucket.com/albums/q199/tahitiplage/SNAG-0290.jpg
___________________________________________
Second edit to add the search grids by date. They were looking mostly to the east of the track.

June 1 search grids:
http://i136.photobucket.com/albums/q199/tahitiplage/SNAG-0292.jpg

June 2 search grids
http://i136.photobucket.com/albums/q199/tahitiplage/SNAG-0294.jpg

June 3 search grids
http://i136.photobucket.com/albums/q199/tahitiplage/SNAG-0295.jpg

June 4 search grids
http://i136.photobucket.com/albums/q199/tahitiplage/SNAG-0296.jpg

June 5 search grids
http://i136.photobucket.com/albums/q199/tahitiplage/SNAG-0297.jpg

June 6 search grid, showing grid that first located definite AF447 wreckage and bodies.
http://i136.photobucket.com/albums/q199/tahitiplage/SNAG-0298.jpg

The early, unproductive searches to the east of the track suggest that AF447 did not turn east and south toward Sao Pedro and Sao Paulo. Unknown why there was no early searching west of the track or northwest of the last reported position. I don't believe the green circle in the June 4 and June 5 maps is a search grid. And the grids for June 2 -- June 5 look to be cumulative, not repetitive.

augustusjeremy

21st Jun 2009, 23:00

as it was the was for the Qantas flight, 3 seconds after the 2nd pitch down.

Hiperveloce,

What I see in the qantas plots is an a/c leaving normal law only when the PRIM1 is set to master. Looks like an abnormal behaviour with a faulty IR1/ADR1 input.

Either it should be in normal law with ADIRUS2 and 3 or it should leave it to alternate law

Rananim

21st Jun 2009, 23:15

http://www.dcs.gla.ac.uk/~johnson/papers/ISSC09/ADIRU_Accident_Submission.pdf

:confused:

Rananim

21st Jun 2009, 23:41

The "Human factor"(ie.correct diagnosis followed by correct transfer switching,inadvertent shutdown of a good unit etc) theory following ADIRU event(Aguadalte) needs much closer scrutiny IMHO.What those Qantas pilots had to go through is above and beyond the normal scope of "airline pilot".

As can be seen, the Captain’s
Primary Flight Display usually presented data from ADIRU 1, following the presentation of the NAV IR1 FAULT,
the source of data was switched to Inertial Reference unit 3 on ADIRU 3. However, this did not automatically
switch the source for Air Data References, which continued to be ADIRU 1. This illustrates the complexity of
interaction with redundant systems as crews struggle to ensure that they receive data from a reliable source without
knowing for sure which of the alternate ADIRU’s is providing reliable information. The problems with the Captain’s
Primary Flight Display could not be resolved before a second uncommanded pitch down. The uncertainty created by
crew interaction with their redundant systems was exacerbated by the way in which the master flight control primary
computer was switched from PRIM1 to PRIM 2 following the first pitch down event. The subsequent indication of
a fault on PRIM 3 then triggered a further change in the master from PRIM 2 back to PRIM 1 and it was only in
subsequent discussions with the operators maintenance watch unit in Sydney, while the flight was still in the air that
the crew decided to switch off PRIM 3.

KISS

ACLS65

21st Jun 2009, 23:42

Good link Rananim

The last line is almost prophetic.

"It would appear that the crew were able to benefit from the lessons learned in the previous incident; however, it is also clear that we
have further lessons to learn in the application of advanced redundancy techniques for safety-critical software."

AF447 may turn out to be the next lesson.

quaeler

22nd Jun 2009, 00:48

PJ2 -
Not sure if this is what you're meaning by "middle space", but the overhead bin structure in this photo is the middle-of-the-cabin structure. There's also the galley which we know about:Thanks for the reply; i was meaning something that we could verify was from roughly row 7 (which i understand the galley was 2 rows in front of) to row 30-something. I'm wondering ~aloud that since we know we have debris from in front of row 5 and from row 42, whether we know if we have cabin contents from a row/area in the middle of the plane.

In the picture you cited, do we know that the overhead bin structure was from rows more towards the center-length-wise of the plane, or is that a different angle on the 42-J,K overhead bin structure?

augustusjeremy

22nd Jun 2009, 01:17

from the preliminary report on qantas 72 flight:

Based on an examination of the FDR data, the aircraft manufacturer recommended removing ADIRU 1 and the number-1 probe heat computer (PHC)21 before conducting any data downloads or testing of the aircraft’s systems.

21 A PHC fault was recorded on an initial download of maintenance data. On further analysis of the available information, this indicated fault was considered spurious.

ACLS65

22nd Jun 2009, 02:28

"Probe heat computer
Some of the PFR messages indicated a potential fault with the number-1 probe heat computer (PHC). Those messages could be generated by either a PHC fault or by an ADIRU fault. The PHC (serial number 2083) was tested by an authorised agency and no fault was found. Based on a review of available information, the messages related to the PHC were considered to be spurious."

http://www.atsb.gov.au/publications/investigation_reports/2008/AAIR/pdf/AO2008070_interim.pdf

PJ2

22nd Jun 2009, 03:29

quaeler;
AFAIK, the "middle" overhead bin structure is mounted in all cabins, J through Y, "5 through 42" so we don't know which cabin it's from, from the photographs. Seat/row numbers are on the structure just below the bins (in the small lights that extinguish when the cabin doors are all closed in the installation I'm familiar with). That piece of wreckage on board ship, is by the galley bins (O2 masks etc), separated from the overhead bins so we don't know if they are associated.

I understand that the search for pieces etc is being wound down so we won't likely see much more than what we have already. I understand that the underwater search continues.

ELAC - absolutely superb summary of the nature of assessing, and of assesed, evidence.

takata

22nd Jun 2009, 05:44

Thank you SaturnV for the posting and comments.

I first did not take into account the findings of the Brazilian SAR before they recovered AF 447's first body and wreckage because this sequence of events was badly documented and I could not rely on clear reports. Too many contradictory news were published about the first findings. So, my presentation was only based on post 06 June facts without taking into account anything before.

But now that you are showing us the 06-06-2009 reports of Brazilian SAR,
a much closer look at the 02 June Brazilian researches become obvious if we are taking into account where the crash location could be roughly infered by the 06-10 June surface sea drift.

Here is the picture again without SAR zones of 02 June:

http://takata1940.free.fr/images1/braz0.jpg (http://takata1940.free.fr/images1/Braz0.jpg)

The three items discovered were:
1. Poltrona = seat
2. Pedaços de metal = piece of metal
3. Mancha de combustível = strain of gazoline

They were spotted from the sky and, as no ship was close enough, they could not be recovered. Then, the following days, the bad weather made the SAR difficult and this specific stuff could not be located again. Other 'wreckage' was spotted and identified as sea garbage (orange buoy, fuel traces, wood palet, etc) but we can see also that it was far away from this place on the other pictures. I remember well the contradictory declarations from the authorities about the first findings, some were positive (I guess the spotting crew), and others doubted about it because nothing was recovered (specifically the blue seat). So, maybe this air crew was mocked for no reasons as this map is showing now:

http://takata1940.free.fr/images1/Crashsite3.jpg (http://takata1940.free.fr/images1/Crashsite3.jpg)

An estimation of the coordinates based on sea relief for those three items give:
- seat (2.077, -29.800)
- metal piece (1.895, -30.080)
- gazoline (1.745, -30.035)

I picked the seat:
a) this is the best described item unlikely to be something else if spoted from a recon aircraft. Beside, there was several 'fuel strain' and 'piece of metal' spoted in several other places which were not from AF 447.
b) If other items would belong to AF 447, their specifics won't match very well the drift of the bodies.
c) An aircraft seat will match pretty close a body in water as it will be nearly fully submerged and won't be affected that much by the winds.

The distance from d06 to the seat is 180 km. As an approximation, (no spotting hour is available) four days between the spotting and d06 recovery which make an average 0.52 m/s drift. This is in the range of possible following the SHOM tables:
http://takata1940.free.fr/images1/shom0.bmp
This map is showing a slice of the drift at 30.5W 2.5N to 30.5W 3.5N on 04-06-2009 from 1.000 meters deep to the surface (vert. axis) and the distance in km is on the horizontal axis. It is a slice of the sea at this point in the red circle shown on the second map. This one is maping the surface drift. As we can see, the 'surface' drift is fairly deep, it is not only a few meters deep layer.

http://takata1940.free.fr/images1/Shom1.jpg

Using the same method as before, it is now possible to roughly estimate a new possible crash zone 24NM farther (24 hrs @ 0.52 m/s). The heading will be @ 336 to d06.
- second hypothesis crash zone coordinates (1.708, -29.630)

http://takata1940.free.fr/images1/Crashsite4.jpg

Now, adding a new course (yellow) from S. Peter Rocks crossing the crash zone, it bring us directly to the 0214Z flightplan spot. I didn't change the U-turn as nobody knows actually what was the real course after 0210Z and the turn back could have started from this point or later, so its shape is fairly unknow and all this above is purely hypothetical, only based on the wreckage spots, and not on what happened really on board. It would be too early to speculate with the little information we have.

A general view of the wreckage is more telling and of course, the drift is vectorized as the flow may be much more turbulent in reality than pictured by this:

http://takata1940.free.fr/images1/Crashsite6.jpg

I'll answer other questions later about the bodies sinking or not. Beside, sorry for the typos and grammar (I need to add a 'h' to hypothesis and 't' to spotted on my maps!) due to my poor English skills.

S~
Olivier

cats_five

22nd Jun 2009, 05:47

"Officials in Brazil have identified the first 11 of 50 bodies recovered from the Air France disaster in which 228 people died three weeks ago."

Full story at:

BBC NEWS | Americas | Atlantic crash bodies identified (http://news.bbc.co.uk/1/hi/world/americas/8112081.stm)

Dysag

22nd Jun 2009, 05:49

It's not from the middle of the cabin, I admit, but they do have part of the crew rest container, which is located just aft of the wing centre section.

augustusjeremy

22nd Jun 2009, 06:25

"Probe heat computer
Some of the PFR messages indicated a potential fault with the number-1 probe heat computer (PHC). Those messages could be generated by either a PHC fault or by an ADIRU fault. The PHC (serial number 2083) was tested by an authorised agency and no fault was found. Based on a review of available information, the messages related to the PHC were considered to be spurious."

Is there any relationship between those messages and anything regarding pitot malfunctions ? Does it prove that pitots status is INDIRECTLY verified by ADR data ?

Pontius Navigator

22nd Jun 2009, 06:26

The light blue/azure colour in tataka's current map also shows the thermocline.

The deepest part is about 600 feet and the shallower at about 350. Whilst the thermocline marks the difference between temperature profile is is also almost a 'hard' surface for buoyant objects. It is also the layer most affected by wind and wave action. Below that things remain faily constant.

IIRC sink rates for heavy objects would be about 40 seconds per thousand feet so for 10k feet it would take some 6-7 minutes. There would be sub-surface drift but it would be insignficant for an object sinking to the bottom.

puff m'call

22nd Jun 2009, 06:55

Has there been any news yes on the recovary of the two "Black Boxes"

It's all gone very quiet from a news piont of view.

wilyflier

22nd Jun 2009, 08:08

I seem to recall that a freighter picked up a spoiler about 45 km NW of StPeter & St Pauls. This matches Takatas latest Hypothesis but doesnt show on his plot.
It also seems to support a more consistent and constant rate of drift from the discovery point of the first most southerly items

(The report came on PPrune after the VS was first picked up, but I havent been find it . There is a post #1325 June 13 which has a pic of freighter and spoiler, but I cant find any seperate report text now,)

funfly

22nd Jun 2009, 09:39

Now it's quietened down I must observe that the contribution of takata seems to be pulling it all together. My only obsevation is that the tragectory he shows surely involves enough time for a life jacket request from the crew and there seems no evidence of their use yet. It also does not prove that the aircraft was under the control of the crew if it 'flew' this direction. Can I ask if the fact that the recovered bodies were intact indicates that they did not have seat belts fastened?

Me Myself

22nd Jun 2009, 09:48

Olivier / Takata

You bear all the signs of an Airbus insider and since you seem to know what you're talking about I'd like to know how many incidents DIRECTLY related to the probes happened with other A330 operators.
Surely, it can't be a FRENCH curse, can it ? What are the, let's say 20 AF A330 compared to the rest of the world's fleet ?

One thing I can tell about Airbus is that in a way it sucks !!
I flew the 320 for 4 years, really enjoying it but totally unaware of the consequences of pitot problems. Yet, that's where it all started, on the 320 fleet.

I'm now on the 777 and I can tell you it is a very comforting thought to know the Primary flight computer switch off button is right above my head. A flick of a switch and I fly a " basic " airplane again.
This probably saved Malaysian over Australia in 2005 and could have allowed Qantas 072 their second dive.
Do you think Airbus would / will come up with such a feature ??? That would imply tons of humility, quality they've proved to be rather short of.
On the Airbus, you have to cope with whatever the aircraft decides to give you. That sucks big time.

Dysag

22nd Jun 2009, 10:07

RealQuax

22nd Jun 2009, 10:18

Takata's analysis is very thorough and looks conclusive.

However, being a skipper on the A330 myself, I have some doubts for operational reasons:

deciding to leave the track for a turnback is a serious matter and would be coordinated, if ever possible, with ATC.
If unable due to urgency or no contact, broadcasting the intention on 121.5 is certainly a high priority.
Assuming, the reason was for a crippled aircraft: VHF #1 is connected to the DC-essential bus and available in ELECTRICAL EMERGENCY CONFIG. Assuming there was no complete electrical loss at the time of a turnback-decision (which would render a turnback almost impossible because of flight in MECHANICAL BACKUP), another aircraft should have catched such a transmission.

I would also like to hint to funfly's conclusion with no signs of an ongoing preparation for an emergency in the cabin (e.g. deployed lifevests)

Thomas

OleOle

22nd Jun 2009, 10:56

There seems to have been a drifting buoy dumped into the crash area on 2009-Jun-15 1719. The area where it has been dumped seems to be more or less the same area that was identified by takata.

Ship Status Report: callsign 31998 (http://www.sailwx.info/shiptrack/shipposition.phtml?call=31998)

Even though in the screen shot the track history is stated as 530 hours the track history actually begins on the above date. You can verify that by dumping the complete track history on the website. The depicted track represents 7 days worth of drift.

The orange triangles representing the last known position and TASIL where inserted by me.http://www.bilderload.com/bild/7156/bild3X9YRJ.jpg

funfly

22nd Jun 2009, 11:10

Not facts just conjecture and a few questions....
Reclaimed bodies intact - no seatbelts?
No lifejackets - no time or no announcement?
Not many found - most remain within fuselage at time of impact?
Bodies found were generally unclothed - does this indicate leaving the aircraft at altitude or does it deny this?
Few pieces on the surface - Would an object the size and construction of the aircraft hit the surface at very high speed and not disintegrate.
Add to this the condition of the gallery component (it doesn't have the apearance of something that dropped a few thousand feet) and it is difficult to escape the feeling that the aircraft broke apart relatively near the surface of the sea and at a relatively low airspeed.

Orestes

22nd Jun 2009, 11:16

ttcse:

Your remark about the fuel tanks just made me notice something peculiar....very early in the search for this plane, some wooden pallets and an oil slick were located, but these were quickly determined to be not from the plane but likely dumped (illegally?) from passing cargo ships.

So, does this mean that no trace of a fuel spill from the aircraft itself has been found? This plane was in the early portion of a long transatlantic flight, so surely it was carrying a significant fuel load. Yet (correct me if I'm wrong) I'm not aware of any reported sighting of a fuel slick from the plane along with the wreckage.

A question to anyone out there in the know: Could jet fuel dissipate so quickly in the ocean to have already disappeared?

Wader2

22nd Jun 2009, 11:20

Can I ask if the fact that the recovered bodies were intact indicates that they did not have seat belts fastened?

This was answered in some detail earlier on in this thread. I believe it stated that injuries were indicative of seated passengers strapped in to their seats.

I would speculate that some at least those that had their seat belts fastened had them fastened on a relaxed manner as you might when trying to sleep. Others may have had them fastened more securely and this might account for the proportion of bodies recovered.

If that is true then it would acocunt for the lack of life vests.

Edited to accept harry mann's point about severe turbulence.

It is conceivable that although the seat belt light may have illumiated the FAs were in the process of waking passengers to ensure that seat belts were properly fastened. If they had not completed their checks it would account for FAs not being strapped in, for some passengers not being strapped in and for some to have slack seat belts.

One expects that there may or may not be evidence of seat belt injuries to supprt these possibilities.

Wader2

22nd Jun 2009, 11:33

Could jet fuel dissipate so quickly in the ocean to have already disappeared?

Yes.

You will recall that the early search period was hampered by bad weather. aviation fuel is considerably lighter and more volatile than furnace fuel oil or bunker oil from a ship. It would be dispersed quite quickly in heavy seas.

It is also conceivable that fuel in unruptured tanks would not be released and could still be trapped in the tanks at depth. Even if later released at depth it is unlikely to be visible on the surface.

HarryMann

22nd Jun 2009, 12:46

Reclaimed bodies intact - no seatbelts?

Irrespective of speculation on mode of ejection, injuries etc.... it would be expected to a high probability that most/all would be seat-belted in 'strong turbulence', as reported shortly before.

We await details of whether many were FA's and where they all were seated, to draw any really valid conclusions?

SaturnV

22nd Jun 2009, 12:48

Wily flier,
I seem to recall that a freighter picked up a spoiler about 45 km NW of StPeter & St Pauls. This matches Takatas latest Hypothesis but doesnt show on his plot.

It also seems to support a more consistent and constant rate of drift from the discovery point of the first most southerly items

(The report came on PPrune after the VS was first picked up, but I havent been find it . There is a post #1325 June 13 which has a pic of freighter and spoiler, but I cant find any seperate report text now,

.....the merchant ship picked up that piece of the plane north of Tasil on June 13.

See the Powetpoint slide for June 13 on the Brazilian Air Force site:
FORÇA AÉREA BRASILEIRA - Asas de um povo soberano (http://www.fab.mil.br/portal/voo447/infograficos.php)

Poit

22nd Jun 2009, 13:06

I've noted an increasing amount of speculation that the aircraft might have hit the ocean in-tact. I think all the evidence we have points to the contrary.

I think an excellent example, in relation to mid-air break up, is the Air India Flight 182 in the mid 80's. Do a search on 'you tube' for Air India 182, and watch the documentaries relating to that incident.

Note the condition of the bodies found in that case: no clothes on the majority of bodies found; fractures to hips, shoulders and other joints (known a 'flail' injuries, resulting from violent tumbling through the air); and no sign of drowning (ie they were not breathing when they hit the water). Also note the conclusion that the pathologist and investigators drew based on those facts: "based on the injuries, we could only say the plane had broken up at 31,000ft". I think the similarities between the bodies found in this instance, and the Air India accident, are too common to be ingnored. It seems quite obvious, then, that the unfortunate people on AF447 suffered a similar fate to the passengers on the Air India jet. What remains to be answered is 'what caused it to break apart?'

The other interesting fact from the Air India case (which involved a 747) is the size of the debris field: 16 x 6km. Speculation that various elements of AF447 could be hundreds of miles apart seems unlikely in light of that fact.

Hyperveloce

22nd Jun 2009, 13:16

For the U-turn hypothesis and the infered crash area, is it possible that an airliner (at the assumed flight point, airspeed and altitude) bank with such a short radius curve ? (as plotted on the map)
Just remembering the order for the recovery of the debris and bodies: the first piece collected (on june 6) was the crew rest cabin remain (where is the mobile crew rest deck in the A330 ?), less than 2 days later, the vertical stabilizer was found, and latter other groups of bodies/debris (the whole with a large dispersion)
Doesn't it suggest that these people and pieces were first lost in flight after the last ACARS, that the VS was lost and the rear part of the A330 fuselage may have been severely damaged in mid air ? (In this case, the airplane didn't try to make a U-turn but simply tried to control its attitude and its altitude)
Apart from a stall, how can an airliner break up in mid-air in cunjunction with all the reported avionics faults ? (a cumulonimbus breaking a plane does not need the ACARS faults: would-it explain them ? whereas the stall possibility can stem from these).
Why isn't it interesting to consider the early leakage from the LAV L54 (which had already problems on the 10th of May) toward the rear of the plane: isn't the BEA right when it suggests that combined to the very low temperatures encountered may have frozen a part of the composite structures, weakening them, making them more prone to a structural failure under heavy stress ? What is the volume of the fluids that may have leaked ? (<100 l, 500 l ?), from which part of the wasted fluids circuit, and in which areas would it flow in more than 3 hours ?
Can't we assume that the A330 tail and other control surfaces were under heavy constraints (no longuer protected by the normal law) while trying to recover from assumed situation of emergency ? (possibly creating a huge torgue of the rear part of the fuselage)
Is an airliner without its VS still in control in a turbulent atmosphere ? would it be in ideal flight conditions ?
To conclude, can't there be a large initial dispersion that was increased by the several days drift ? The A330 airspeed gradually decreasing between 02:10 and 02:14, then a final trajectory initiated by a stall with a right bank, loss of VS/rear part of the fuselage, gradually loosing passengers and pieces, ending 10-20 NM south east of TASIL. Then a few days NW drift of this initial distribution)
Jeff
PS) my apologies if these questions were already answered.

SaturnV

22nd Jun 2009, 13:38

takata,

thank you for your thorough and very interesting analysis.

I would make several points.

a.) From looking at the plots, the Brazilian Air Force's search grids are systematic on June 1, and between June 2 and 5 are systematic only to the east, south, and southwest of the last reported position. Much of the searching between June 2 and June 5 seems centered on various targets observed on the sea surface than on a systematic grid search.

b.) Why the Brazilian Air Force did not look immediately west or north of the last reported position before the June 6 search grid remains a question. Was this due to the operational limits of the recon aircraft? Did the Air Force immediately focus on the first sightings of possible wreckage?

c.) If the prevailing current was from south to north, why did the Air Force initially expand the search grid to the south, rather than to the north?

d.) Was the Air Force using the meteorology for June 1 and June 2 to infer that the thunderstorm complex encountered by AF447 moved south or east, and that winds from the complex probably pushed the lighter debris in that direction?
________________
Where did you get the chart showing the surface current speed and direction on June 5 2009?

Squawk_ident

22nd Jun 2009, 14:02

This is the English traduction of an Air France internal document addressed to Cockpit Crew dated 18/06/09 that you can find here : (may be elsewhere I don't know)

Les dossiers noirs du transport aérien (http://henrimarnetcornus.20minutes-blogs.fr/)

Scroll down to 21.06.2009
or direct link :

http://henrimarnetcornus.20minutes-blogs.fr/media/01/00/1453557506.pdf

It's an automated translation plus some personal adjustments.

HEAD OFFICE OPERATIONS AND QUALITY
DIRECTION OF SAFETY
DIRECTION OF THE AIR OPERATIONS
ALL PNT
Roissy, On June 18, 2009
INFORMATION ALL PNT N° 5

On June 15, a point on the inquiry was made by the Management of the safety of the company in front of executives PN of the Air Operations. This "N°5" information" recapitulates all the points approached at the time of this presentation.

INVESTIGATIONS
The accident of AF447 took place in the international water; this is thus the State where the aircraft is registered that leads the technical inquiry. Four working groups were made to this end by the Bureau d'Enquêtes et d'Analyses. Air France contributes to this investigation in bringing its expertise to work with these groups of which each members is engaged to respect the confidentiality of information.
Only the BEA is authorized to communicate on the progress of the inquiry. It will publish a preliminary report before the the 30th of June.
In parallel to the technical inquiry and in accordance with the French Right, a judicial enquiry is carried out by the GTA (Gendarmerie du Transport Aérien) under control of an instruction judge of the Court of Paris.
The rules specific to the company envisages in addition the installation of an internal inquiry. For this reason the Head office designated 5 permanent members including two flight crew staff representatives.
Finally 2 other investigations will be led at the initiative of the CHSCT PNT and PNC.

FACTS AVAILABLE AND THE PRELIMINARY REPORT OF THE BEA.

At the time of the accident the plane was leaving a zone of convection related to intertropical convergence. The satellite photos diffused to date show clouds spreading out at high-altitude, but these photographies do not indicate directly the position of the stormy cells that the crews of different companies flying on this road met. A complementary work of analyses is in progress.
The main informations that we have about the plane result from automatic maintenance messages.
These messages are embodied in the information transmitted by the plane without intervention of the crew to alow maintenance teams to prepare the interventions upon the arrival of the plane. These messages are not easily exploitable for an investigation and cannot replace data from CVR (Cockpit Voice Recorder) and DFDR (Digital Flight Dated Recorder).
More particularly, the format of the messages and logics of transmission do not allow to reconstitute with certainty the chronology of appearance of the anomalies in the cockpit.

The majority of these messages are the consequence of the anomalies and the inconsistencies of aerodynamic speeds; they were detected by the flight control calculators and the automatic pilot. Among these one can be quoted:
The loss of characteristic speeds
The passage in secondary law of flight (Alternate Law)
These associated anomalies and their consequences are tested during the certification of the plane; this allows to guarantee that the plane remains controllable in this situation.
All of these messages and their significance will be very certaily published in the preliminary report of the BEA.
At this stage, nothing in these messages allows to establish:
- A loss of power supply,
- A loss of the screens of piloting (PFD and Stand-by horizon),
- A faulty operation of the ADIRU which could have involved an incident of the type of which Qantas company knew recently.

Only one certainty: the sequence of the messages does not allow to explain the accident by itself.

PROBES OF PITOT

Many erroneous assertions were advanced in connection with the anemometric probes equipping the Airbus fleet. We present to you in chronological order a summary of the reports and especially of the actions carried out by the company since 2001.

August 2001:

Following fluctuations and/or losses of the indications of the aerodynamic speed on A330 and A340 reported by certain companies, the DGAC publishes an “Airworthiness Directive” (AD) by plane type, to impose the replacement of the probes of Pitot ROSEMOUNT P/N 0851GR, either by probes GOODRICH P/N 0851HL, or by SEXTANT (THALÈS) P/N C16195-AA; this operation having to be finished before the 31 December 2003.
Indeed, the official services allot these events to the presence of crystals of ice and/or of quantities of water which exceeds the specifications of the probes Pitot ROSEMOUNT P/N 0851GR. In accordance with the “Airworthiness Directive”, the model SEXTANT (THALÈS) P/N C16195-AA is installed on the Air France A340 fleet ; as from December 2001, Air France receives its first A330 which is origin-equipped of the probes SEXTANT (THALÈS) P/N C16195-AA. No event of this type will be reported on Air France A330 and A340 until May 2008.

September 2007:

Airbus emits technical notes (Service bulletins or SB) which recommend, without being mandatory and outside all context affecting the navigability of the planes, the replacement of the probes THALÈS P/N C16195-AA installed on fleets A320/A330/A340 by new model probes THALÈS P/N C16195-BA.
It is indicated that this model improves the behaviour of the probe by limiting the consequences of water ingestion by strong rains and by reducing the risk of icing. After analysis, the Air France technical teams decide to launch this modification on the A320 fleet which has incidents with losses of speed indications at low altitude in case of strong rains. They decide to replace probes on A330/A340 by the new models only in the event of failure, the A330/A340 fleet having then no incidents with loss of speed informations.

May - August 2008:

A first incident of icing of the probes occurs on a Air France A340 with temporary loss of speed indications; followed by a second incident in July 2008 and then of 3 incidents in August 2008, all on A340. Airbus is immediately questioned on the origin of these incidents and to the measures allowing to cure it.

September and October 2008:

Many exchanges with the Airbus technical teams take place. An exploitation incident is recorded in September and one in October 2008. In six months, 7 incidents were thus recorded, when no incident of this kind had been reported before.
Airbus answers that:
o The supposed origin of these incidents is a icing by crystal formation in the probes of velocity measurement,
o The new model THALÈS P/N C16195-BA was not conceived to answer the problem of icing and thus should not bring significant improvement with this problem,
o The probes installed are in conformity and even exceed the lawful requirements in term of airworthiness and of safety of the flights.

November 2008:

Following various follow-up from Air France technical services, Airbus corrects its September 2007 technical notes in a dated November 12, 2008 edition that withdraws any mention of a contribution of the probe THALÈS P/N C16195-BA improved resistance to icing.
On 24 November2008 a meeting between the technical directions of Air France and Airbus discusses lengthly of the incidents with loss of speed informations. Air France requires that a technical solution be quickly brought to solve these incidents. Airbus confirms again that the origin of these incidents is an icing of the probes, that the last model THALÈS P/N C16195-BA does not treat of the subject and that the probes installed are in conformity with the airworthiness requirements and safety of the flights.

February 2009:

Facing our insistence to find a solution, wind-tunnel tests are undertaken by Thalès and Airbus on the behavior of the probe THALÈS P/N C16195-BA.

March 2009:

At the end of March 2009, two new incidents of exploitation are recorded of which a first on A330. That brings the total number of incident to 9, including 8 on A340 and one on A330.
Airbus, again requested on several occasions, answers by confirming the presumption of icing of probes and refers to a procedure of maintenance and checking of the probes.

April 2009:

In a letter of April 15, 2009, Airbus informs of a new element: the probe THALÈS P/N C16195-BA does not have vocation to answer the problem of icing of the probes, but the tests carried out by Thalès show a behavior definitely better than that of the former model. Taking into account the limitations of the wind tunnel tests, Airbus suggests an experimentation on Air France planes to check if an improvement is confirmed in real situation. Without awaiting this experimentation, Air France decides to immediately extend this measure to its entire long-distance Airbus A330/A340 fleet and to replace the totality of the probes speed. One internal technical document launching the modification is established dated April 27, 2009. The beginning of the planes modification is planned as of reception of the parts, at a rate of several planes per week, from June 1.

May 2009:

Air France requires THALÈS to accelerate the delivery schedule of the probes. Those are delivered starting May 26, 2009 at a rate of 12 tubes of Pitot per week. The program thus could be accelerated.

Since the accident:

Without prejudging of a link between the anemometric probes and the inconsistencies in the indications of speeds presented to the pilots, Air France has decided to accelerate its plan of replacement of the probes THALÈS P/N C16195-AA on the fleet Airbus. Since last June 12, all A320,A330,A340 Airbus, in exploitation within Air France are equipped with the probes of last generation THALÈS P/N C16195-BA.
Nevertheless in an information bulletin published on June 8, Airbus confirms that the Airbus world fleet can be exploited with one of the 3 types of anemometric probes which equips world fleet, namely THALÈS P/N C16195-AA, THALÈS P/N C16195-BA and GOODRICH P/N 0851HL.

CONCLUSION

We intend to recall that any step of prevention imposes three requirements;
- a requirement for transparency,
- a requirement for reactivity,
- and a requirement of pro-activity.

Whatever the circumstances, it is advisable to proceed in this way.
This is why, we have decided to keep you regularly informed of the state of advance of the inquiry and this, in an agreement with the BEA that remains the only one in charge of the communication of the factual elements.

In addition, we named two staff representatives within the Internal Commission
- Mrs L G for Commercial flight crew
- Mr O R for the Technical Flight crew.
This commission reserves the right to raise recommendations constantly if it feels it need.

----------------------------------------------------------
2 names deleted.
PNT : Personnel Navigant Technique (Cockpit Crew)
PNC : Personnel Navigant de Cabine (Cabin Crew)
DGAC: Direction de l'Aviation Civile
CHSCT Comité d'Hygiène, de Sécurié et des Conditions de Travail
The members of the CHSCT are elected Unions members. CHSCT is competent for all matters related to the security, hygiene and work conditions of the employees in a given society.

-----------------------------
My own comments if you allow:
This document is an internal one but the fact that some terms such as CVR or DFDR are translated in "plain" language (or even the GTA acronym) indicates 1/ That the AF PNT is really stupid or 2/ May be this document is intended for the PNT but also for the entire world. And written knowing that it might leak out. In this case who is the Charming and who is the Villain?

max motor

22nd Jun 2009, 14:07

Just for my understanding, what is that "basic airplane" you're getting? I thought you would still be using the control column to provide inputs to the fly-by-wire computers. I didn't think the 777 had direct mechanical linkages to all control surfaces.

When the 777 is manually switched to "Direct" mode you are effectively removing the Primary Flight Computers and may control the aircraft via unmodified commands using ANALOGUE electrical signals that are passed directly to the control surface actuators.

Max

takata

22nd Jun 2009, 14:44

Hi,
Harry Mann: Irrespective of speculation on mode of ejection, injuries etc.... it would be expected to a high probability that most/all would be seat-belted in 'strong turbulence', as reported shortly before.

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.

In a note dated on 6 Nov. 2008, Air France informed its crews of speed discrepancy problems with its A330/340 Fleet: http://www.eurocockpit.com/docs/OSVAF.pdf
- loss of airspeed data;
- numerous ECAM messages displayed;
- configuration warnings (sometime)

The turbulence context is specified for 6 flights:
- Turbulences 'light to moderate'
- 2 cases with 'Fortes Turbulences' and 'MSS/Turbulence' applied.

So, if the 'fortes turbulences' occured (who know?), they would came at the same time of the pitots freezing or later.

S~
Olivier

takata

22nd Jun 2009, 15:11

SaturnV:
thank you for your thorough and very interesting analysis.
I would make several points.

a.) From looking at the plots, the Brazilian Air Force's search grids are systematic on June 1, and between June 2 and 5 are systematic only to the east, south, and southwest of the last reported position. Much of the searching between June 2 and June 5 seems centered on various targets observed on the sea surface than on a systematic grid search.

b.) Why the Brazilian Air Force did not look immediately west or north of the last reported position before the June 6 search grid remains a question. Was this due to the operational limits of the recon aircraft? Did the Air Force immediately focus on the first sightings of possible wreckage?

c.) If the prevailing current was from south to north, why did the Air Force initially expand the search grid to the south, rather than to the north?

d.) Was the Air Force using the meteorology for June 1 and June 2 to infer that the thunderstorm complex encountered by AF447 moved south or east, and that winds from the complex probably pushed the lighter debris in that direction?
________________
Where did you get the chart showing the surface current speed and direction on June 5 2009?

The charts are part of the SHOM pdf linked previously: Services Hydrographiques et Océanographiques de la Marine,- SHOM, 6 juin 2009:
http://www.bea.aero/fr/enquetes/vol.af.447/shom.050609.pdf

For questions a) to d), of course, I can't give you an answer in place of the FAB!
But I do remember that the bad weather somewhat disrupted the SAR and the aircraft were searching at one point mostly where they could make observations at sea level rather than following a rigid search pattern obstructed by the clouds.
:)
S~
Olivier

MATELO

22nd Jun 2009, 15:41

Just to clarify a point.

The bodies recovered from the Lockerbie Air Disaster, that were ejected at high altitude were fully clothed, partially clothed or naked.

I dont see any relevance in the bodies being either clothed or unclothed in this matter as you cant draw any real clues to the height they left the aircraft.

takata

22nd Jun 2009, 16:22

Hi Jeff,
Hyperveloce:
For the U-turn hypothesis and the infered crash area, is it possible that an airliner (at the assumed flight point, airspeed and altitude) bank with such a short radius curve ? (as plotted on the map)Certainly not as I have no clue of what happened between 0210Z and 0214Z (and later) - beside the ACARS which are revealing nothing about the course- up to the zone where the plane might have crashed.

The point (as it is specified) was not to reconstruct an accurate course between 0210Z and the hypothetical crash zone. The body drift infered a position where a U-Turn was mandatory for F-GZCP to end there. The first hypothesis then was an heading to Fernando de Mononha for an emergency landing, the second, taking into account the wreckage spotted on 02 June, pointed to another spot and a more desesperate alternative (crash landing near St Paul's Rocks?).

But the third logical hypothesis would be that they had no plan left at all with an aircraft uncontrolable which ended in this zone.

Just remembering the order for the recovery of the debris and bodies: the first piece collected (on june 6) was the crew rest cabin remain (where is the mobile crew rest deck in the A330 ?), less than 2 days later, the vertical stabilizer was found, and latter other groups of bodies/debris (the whole with a large dispersion)Yes I remember. The large rigid pieces of the airframe are much more exposed to the wind than nearly fully immersed bodies. The wind was turbulent and somewhat contrary to the drift at some point (this is ITCZ where the weather is fairly complex).

Doesn't it suggest that these people and pieces were first lost in flight after the last ACARS, that the VS was lost and the rear part of the A330 fuselage may have been severely damaged in mid air ? (In this case, the airplane didn't try to make a U-turn but simply tried to control its attitude and its altitude)It seems unlikely that the aircraft broke up just after 0214Z, loosing passengers, and end there (my only hypothesis is to show that, not to explain it). The distance is about 110+ NM back South-East and more than 12 mn at cruise speed @ FL350 in straight line.

Apart from a stall, how can an airliner break up in mid-air in cunjunction with all the reported avionics faults ? (a cumulonimbus breaking a plane does not need the ACARS faults: would-it explain them ? whereas the stall possibility can stem from these).The ACARS do not suggest the plane was breaking up in mid air. The end of the ACARS and the lack of further communications only suggest that the plane could not communicate any more, but not that she was unable to 'pilotate' and possibly 'navigate' for a while...

Why isn't it interesting to consider the early leakage from the LAV L54 (which had already problems on the 10th of May) toward the rear of the plane: isn't the BEA right when it suggests that combined to the very low temperatures encountered may have frozen a part of the composite structures, weakening them, making them more prone to a structural failure under heavy stress ?The BEA did not suggest anything like that. The press did. It is like those so-called 'autopsy' reports 'close to the investigation'. Let's see what will be actually released by the officials and we will compare the report to those 'leaks'. I have read two completely opposite versions of the 'autopsy leaked' published at about the same time

To conclude, can't there be a large initial dispersion that was increased by the several days drift ? The A330 airspeed gradually decreasing between 02:10 and 02:14, then a final trajectory initiated by a stall with a right bank, loss of VS/rear part of the fuselage, gradually loosing passengers and pieces, ending 10-20 NM south east of TASIL. Then a few days NW drift of this initial distribution)Look at the map scale and the sea drift speed. There is 11 NM between 2014Z and the first bodies recovered. Adding 20 NM is not going to make a good count for a nearly crash zone. Everything else is very hypothetical but this pattern of body and their drift speed is nearly certain as well as the utlimate positional report following by the ACARS sequence.

S~
Olivier

ACLS65

22nd Jun 2009, 16:44

Per the AF report above, The Rosemount/Thales pitot probes have been problematic for years. Since Airbus insisted the newer -AB probes do not claim improved performance in icing, the prudent course of action for AF would have been to replace all Thales probes with Goodrich.
GB

It sounds like the Rosemount/Goodrich ones have been redesigned because of icing as well. Also in both redesigns it has been to improve resistance to icing not necessarily prevent it completely.

I think we still need flight control systems and procedures (beyond the current) that can deal with this issue as well.

"FEATURES OF MODEL 0851HL
Improved Design Features
In order to meet the Airbus extreme icing conditions specification, the Model 0851HL Pitot Probe has been designed as a replacement for the Model 0851GR. These performance enhancements were accomplished by
increasing the power density in the tip region by 35% over the existing probe, and incorporating the high power density in the drain hole region to ensure proper drainage during severe icing conditions."
http://www.goodrich.com/portal/goodrich/images/Goodrich%20Content/Business%20Content/Sensors%20and%20Integrated%20Systems/Products/Literature%20Listing/4004%20Pitot%200851HL.pdf

Hyperveloce

22nd Jun 2009, 16:49

the two leaks about the autopsies are separated by several days: the first group of autopsies showed traumas consistant with a free fall and a high velocity impact with the sea surface, the second group of autopsies (and leakage) suggested that some passengers impacted the sea surface sitting in their chair. While we must certainly be cautious with these non official forensic data, these are not necessarily contradictory: some passengers may have been lost in mid-flight (with the VS and then the rear part of the aircraft) while the others might have continued their flight with the main part of the aircraft toward the "crash area". I was suggesting that multiple biases on the airspeed may have triggered a stall after ou around 02:14Z, a stall accompanied by a right bank/sideslip. to me, this is necessary to explain
-that the ACARS satcom link was severed: the attitude of the A330 was exceeding the steering angular range of the satcom beam,
-the exiting of the flight enveloppe and the bad situation that may have led to a destruction of the VS
-the drift of the bodies/debris and where they were collected (but there is here a high degree of uncertainty about the prevailing surface currents and surface winds)
As for the wasted fluids leakage, we are not definitive about the source of this hypothesis but is it possible to assess its plausibility as a contributing factor to a structural failure ?
Jeff

DorianB

22nd Jun 2009, 16:58

I know it was brought up forever ago but would be interesting to see if the satellite kept any logs about transmission strengths or other data regarding the uplink or attempted uplink(s). Might give some clues about location, speed, direction...etc.

DorianB

22nd Jun 2009, 17:08

Locations of bodies #1 and #50. 145km apart in 19-20 days.

http://i635.photobucket.com/albums/uu79/DorianBanks/th_1-50.jpg (http://s635.photobucket.com/albums/uu79/DorianBanks/?action=view&current=1-50.jpg)

funfly

22nd Jun 2009, 17:09

"The distance is about 110+ NM back South-East and more than 12 mn at cruise speed @ FL350 in straight line."

That, of course, only 3000fpm.

OleOle

22nd Jun 2009, 17:14

Actually a number of drifting buoys have been dropped in the area on June 15, possibly to keep track of debris.
From Weather observations and positions of ships at sea (http://www.sailwx.info/shiptrack/).
http://www.bilderload.com/bild/7159/bild6P4ZO1.jpg

takata

22nd Jun 2009, 17:23

The post above by ACLS65 about Goodrich Sensor System 0851HL Pitot Probe is describing in short the the Pitot functions in the second page; the end of it is interesting to draw a direct link between the pitot's failure and the cabin vertical speed control also. Then, the advisory (no warning, no check list) at 2014Z might be also linked to the unreliable Air data provided by the pitots as Thales and Goodrich probes are interchangeable.

The 0851HL pitot is also functionally
interchangeable with existing pitot probes certified on
A318/319/320/321/330/340 manufactured by Goodrich
competitors.

DESCRIPTION
Sensor Systems Model 0851HL Pitot Probe is an integral
part of the Air Data System of the aircraft. Sensor Systems
pitot probes provide vital information for aircraft flight
control. Pitot pressure measurements are used for
calculating flight parameters, which include pressure
altitude, airspeed and Mach number. Air data probes also
provide information for secondary purposes such as
engine control and cabin pressure differential.

SandyYoung

22nd Jun 2009, 17:41

I have no connection with the aircraft industry but have been following with interest some of the very technical discussions on this forum.

One thing I do not understand is why the pinger signal is so weak. I would have expected that there would have been a massive 'ping', perhaps every hour, which would allow those searching to at least obtain an approximate bearing. Limiting a higher output to each hour would make almost no difference to battery life.

Submarines have very sensitive listening gear and one would expect them to locate the recorders more readily. Or have I missed something?

Pontius Navigator

22nd Jun 2009, 17:53

Some interesting information on the FDR pinger here:

Black Box: Locating Flight Recorder of Air France Flight 447 in Atlantic Ocean | MarineBuzz.com (http://www.marinebuzz.com/2009/06/08/black-box-locating-flight-recorder-of-air-france-flight-447-in-atlantic-ocean/)

It talks of the range being one mile, say 6000 feet, and that might be explained by the high frequency in use. In turn this suggests that th eonly way of detection will be by a towed array at a depth of 10000 feet. A two-mile length tow is a lot of tow and a 2 mile sweep width means a lot of steaming to cover the area.

Graybeard

22nd Jun 2009, 17:54

Thanks for the correction, ALCS; I thought Rosemount had been bought by Thales, and not Goodrich. I just visited the Goodrich site, and see you are correct. I'm sure I had read somewhere in this thread that Thales had acquired Rosemount. Not so.

Whose probe business did Thales acquire?

I'll delete my post of Today: 08:40

GB

ArthurBorges

22nd Jun 2009, 18:03

Gosh a few dumb language comments:

"instruction judge" = "investigating magistrate": in France, the police collect evidence and a judge supervises the investigation, with a final say on how everything gets pieced together. This also makes life easier on the prosecutor.

The "judicial enquiry" is therefore the police investigation.

I haven't time now to pick through the rest of this translation.

However, I will note that French techies across a range of fields will coopt English acronyms and terms because that way everybody is sure of what you're talking about.

French equivalents tend to be unwieldy and confuse newbies. Translations of operating & maintenance manuals tend to be unsatisfactory (not the case at Airbus, I understand!) so that when something breaks down, the techies go to the English original to figure out how to cope. Moreover, by the time somebody fixes up the translation, the manufacturer has come out with revisions, upgrades and updates and the whole process starts all over again.

My bottomline is that the use of English acronyms, in itself, does not indicate the document was drafted with the intention of being leaked out.

Nitey nite from Zhengzhou (CGO)

grizzled

22nd Jun 2009, 18:08

poit . . .
re your comment (regarding comparisons between AF447 and Air India 182).

You wrote:
"Note the condition of the bodies found in that case: no clothes on the majority of bodies found; fractures to hips, shoulders and other joints (known as 'flail' injuries, resulting from violent tumbling through the air); and no sign of drowning (ie they were not breathing when they hit the water)."

Your source is incorrect. You mention using youtube to get information from documentaries, so the first thing I must say is this: The internet is a vast source of information but if you want factual information (about an aircraft accident for instance) search the sites of the official national investigative bodies, and reputable aviation-based sites, not Youtube or TV documentaries.

In the case of Air India 182, of the 131 bodies recovered; only 8 exhibited classic flail pattern injuries, 26 showed signs of hypoxia, 23 had signs of injuries from a vertical force, and 21 were found with little or no clothing.

Although all victims exhibited multiple injuries, the PM examinations indicated many different “primary causes”, including 2 from asphyxia and 3 from drowning. (Note, drowning does not infer consciousness at the time of death.)

grizz

JD-EE

22nd Jun 2009, 18:14

The first criticism may place the crash point for AF447 a little closer to the first location for recovered debris. Let's look at Takata's picture of the current profile in the area. Deep red is highest and deep blue is slowesthttp://takata1940.free.fr/images1/shom0.bmp
Please take note of the blue band at about 50 meters depth. From about 10 meters depth to 40 meters depth the speed of the current falls by about half.

Initially bodies sink, a bit. This depends on the water temperature per other observations here. Or they may have been released from a sinking broken fuselage and floated in that band. They may not have been at the surface all that time. They certainly were not seen even though the initial air searches did cover the areas they'd be expected. Thus would fit with the bodies being submerged below 20 meters for their seven days under water.

I'd also question any assumption that the second band of fast current, the one at 100 meters, is going in the same direction as the surface current. The region of slower current may be due to a sheer region between currents running different directions. Getting a direction profile for the surface, 50 meters depth, and 100 meters depth would certainly help prove the contentions made about final crash site.

The second criticism comes from presuming that the plane was still on course to 0214. The last reported position was 0210. Shortly thereafter the pilots could have instituted their turn, willing or not. The last position the plane was known to be controlled well enough to keep the SatCom antenna pointed was 0214. I would hazard the presumption that the plane became uncontrollable to the extent that the SatCom antenna software was no longer able to point the antenna.

So Takata might look into scenarios which might have a turn starting any time between 0210 and 0214 as well as have less time moving under control after 0214.

I do not expect a dramatic change from Takata's conclusions. But I do expect some change based upon examining both the above factors.

JD-EE

rer47

22nd Jun 2009, 18:38

Takata's analysis suggesting a turn back toward Fernando de Noronha is interesting and well thought out, but I think there is one factor that has not been considered. By 0210Z the aircraft was evidently just NW of a very large CB cell and the hypothesized turn would have taken the aircraft right through this cell complex, with tops exceeding 50,000 feet. See the BEA flight path points plotted on the probable radar image reconstructed by Tim Vasquez at:

Jet Crash forum • View topic - Crash location (http://www.jetcrashforum.com/viewtopic.php?f=1&t=3)

I suggest a different final location here. The now-discredited 0214Z "ACARS" location of N3.5777, W30.3744 that was attributed to FAB has always been intriguing because it is quite close to the location where bodies were recovered by FAB on 6 June (see the FAB cumulative PowerPoint presentation of 6-6-09). Now that it is clear that the aircraft transmitted position reports at ten minute intervals throughout the flight, we must ask ourselves what else was contained in those position reports.

Although I have no idea how the A330 system is configured, the aircraft GPS tracking devices with which I am familiar provide (in addition to latitude and longitude) ground speed, heading, and altitude. See for example the devices offered by:

EMS Sky Connect (http://www.skyconnect.aero/)

Thus, in addition to latitude and longitude, the heading of the aircraft at 0210Z was probably known to BEA by the morning of 1 June. These headings are calculated by the GPS receiver based on the azimuth between present and previous positions, and these positions are typically calculated at about 1 second intervals. My hypothesis: BEA actually calculated the N3.5777, W30.3744 position from the heading at 0210Z, and projected that heading forward to 0214Z when the last ACARS maintenance message was sent. But that position requires impossibly fast speeds to cover the 38.2 nm distance between 0210Z and the N3.5777, W30.3744 position (573 kt, ~Mach 0.98). Could it be that BEA simply used an erroneous assumed speed for the calculation?

If you use the last calculated ground speed of AF447 between 0200Z and 0210Z, 463 kt, and project it forward along the path between the 0210Z and the "0214Z" FAB location, the aircraft would have covered about 30.9 nm and the 0214Z location would be 3° 27.8'N, 30° 25.0'W, about 7.3 nm south of the FAB 0214Z position. Assuming the aircraft made only one turn to the left, projecting the path from 0210Z back to intersect the planned flight path would place the hypothesized turn at about 2° 34.5'N, 30° 44.8'W. This would place the crash location south of the recovery location of the first bodies on 6 June, as appears to be required by the drift direction of the debris field.

Alternatively, perhaps the 0214Z ACARS maintenance time stamp is truncated, and the actual time might have been as late as 02:14:59Z. If BEA made this assumption to make a best estimate of where the wreckage might have been to aid the SAR teams on the morning of 1 June, from the 0210Z position at 463 kt the aircraft would cover 38.6 nm in five minutes, placing it only 800 m north of the "0214Z ACARS-FAB N3.5777, W30.3744" position.

I am unable to post images in this forum for some reason, but you can see a map of this scenaio on page 2 of:

Jet Crash forum • View topic - Crash location (http://www.jetcrashforum.com/viewtopic.php?f=1&t=3)

-rer47

PJ2

22nd Jun 2009, 18:54

Hyperveloce - there is a lot of information here that will help you with your questions, and many of the questions you have, are ones most who are on the thread have as well - there's been a lot of hypothesizing by very good contributors but few answers because there is so little data. Questions which will arise out of the reading you do - when they do, they are welcome because they are informed.

Cheers,
PJ2

iakobos

22nd Jun 2009, 19:23

I would hazard the presumption that the plane became uncontrollable to the extent that the SatCom antenna software was no longer able to point the antenna

Inmarsat Aero-L used by ACARS makes use of a solid state antenna.
The "radiation" diagram of the antenna is omnidirectional (azimuth) and hemispheric (elevation).
Two satellites cover the area, one to the E (AOR-E sat), the other to the W.
As long as the antenna sees the sky the link is possible.

SaturnV

22nd Jun 2009, 20:13

To muddy the picture of the surface current, so to speak.

This is the surface current plot for June 5, the box is centered on 3N 30W.

Red / orange marks the strongest surface current.

The page with the chart in the pdf file is titled:
Previsions 3D de courant quotidiennes a J+1 et J+6

The chart is titled:
courants marins (0m) du 05/06/2009 echeance 24h

http://i136.photobucket.com/albums/q199/tahitiplage/SNAG-0300.jpg

The only charts included are those for June 5, though the page title indicates there are other dates not provided.

The current direction is SW to NE which would suggest that if AF447 crashed east of 30W, the debris and bodies would be carried to the east. Instead the bodies and wreckage have been found west of 30W.

It is possible the the charts for the surface currents on June1-4 -- if the current changed -- would allow for bodies and wreckage to be carried from east to west.

The other current chart included in the pdf is for 1000 meters; that current is very light and moving from east to west.

Source:
pdf p 5 of this document:
http://www.bea.aero/fr/enquetes/vol.af.447/shom.050609.pdf

ACLS65

22nd Jun 2009, 21:10

Takata:

Then, the advisory (no warning, no check list) at 2014Z might be also linked to the unreliable Air data provided by the pitots as Thales and Goodrich probes are interchangeable.

From pdf:

Air data probes also
provide information for secondary purposes such as
engine control and cabin pressure differential. I think Takata makes an interesting point and don't think I have seen it mentioned previously. That is that an ice clogged pitot might also be related to the cabin pressure ACARS msg as I assume the pressure differential is between cabin pressure and outside pressure. Previously I think the options discussed were either a climb or dive greater than 1800fps, or possibly some sort of cabin depressurization.

Can anyone with more probe knowledge comment on what readings in terms of AS, AOA, and ALT we would expect to see with an ice clogged pitot?

I guess then it gets more complex in terms of determining how the computers would have reacted to this sudden change.

syseng68k

22nd Jun 2009, 21:23

I think Takata makes an interesting point and don't think I have seen it mentioned previously. That is that an ice clogged pitot might also be related to the cabin pressure ACARS msg as I assume the pressure differential is between cabin pressure and outside pressure. Previously I think the options discussed were either a climb or dive greater than 1800fps, or possibly some sort of cabin depressurization.

Cabin pressure is related to outside pressure, but this is measured by the static ports, not pitot. Pitot and static lines are typically on the same probe assembly for smaller a/c, but not on A330.

The effect of a blocked pitot has been discussed elsewhere in this thread...

Chris

PJ2

22nd Jun 2009, 22:17

ACLS65;
Previously I think the options discussed were either a climb or dive greater than 1800fps, or possibly some sort of cabin depressurization.
Let's be clear about this piece of information as well. It is the cabin-rate-of-climb (or descent) that is being referenced here and not the aircraft rate of climb or descent.

The concept is perhaps not as easy to understand as one might expect - while the aircraft is climbing to cruise altitude, say, FL350, at a rate which is typically between 1500fpm and 2500fpm depending upon a lot of factors, the cabin is being pressurized such that it's pressure is equivalent to that found at between 6000 and 8000ft, approximately.

Briefly, air taken in by the engines is partially diverted (at the second and third stages of compression in the engine), through an air-conditioning "pack" into the cabin. The outflow valves in the aft fuselage close so that the air escaping the cabin through these valves is slightly less than the air being routed into the cabin through the air conditioning packs, thus pressurizing the cabin. The cabin "climbs" from it's starting pressure which is that of the departure airport, to a pressure that is equivalent to an altitude described above. Cabin "altitude" is stated in hundreds of feet, such as, "the cabin altitude is 7600ft". That means the passengers are breathing air (not oxygen, as many think) as if they were at 7600ft, or just a bit higher than Denver, Colorado or walking up the road to Machu Pichu, etc.

The "cabin-rate-of-climb" is thus the rate of pressurization from it's starting altitude to it's cruising altitude. Typically, if one departs near sea level, the cabin rate of climb is around 300 to 500fpm, controlled by the Cabin Pressure Controller(s). Part of the input to these computers would be the static pressured outside the aircraft so the CPC "knows" how high the aircraft is.

Now it makes sense to talk about a "cabin rate of climb/descent" and one can readily see now why a cabin rate of climb of "1800fpm" up or down, is not only excessive, it is an indication that the CPC is not able to control cabin pressure, for whatever reason.

Hyperveloce and others will now understand why a cabin rate of climb has nothing whatsoever to do with, nor is it an indicator of, an aircraft "stall" condition. It is an indication that not all is well with the CPC, the outflow valves, the cabin itself or...?

One can also now understand what a pilot means when a pilot says "we caught the cabin" - it means that the aircraft altitude is the same as the cabin, usually in descent, and to keep up, the cabin begins to follow the aircraft rate of descent.

Obviously if one thinks about it, there is a cabin climb schedule which matches the aircraft rate of climb so that "catching the cabin" does not occur. In climb this rarely if ever, happens but if one is descending very quickly, one can get to the altitude that the cabin is at and then the cabin will descend with the aircraft - very hard on the ears.

The ACARS advisory message, (see the earlier description of same) advises the crew of this rate of climb/descent. The cabin pressure warning (red master warning, steady chime, red ECAM message "CAB PRESS") do not occur until a cabin altitude of just under 14,000ft.

We cannot know the full meaning, (ie, was this a depressurization, a CPC fault or...?) of the advisory but we do know that at some point the fuselage/cabin structure was compromised/breaking up, at some point between 350 and sea level.

HarryMann

22nd Jun 2009, 23:13

Obviously if one thinks about it, there is a cabin climb schedule which matches the aircraft rate of climb so that "catching the cabin" does not occur. In climb this rarely if ever, happens but if one is descending very quickly, one can get to the altitude that the cabin is at and then the cabin will descend with the aircraft - very hard on the ears.

To add to that well drawn picture, there's a focus on cabin rate of descent because the human ear is more sensitive to increasing than decreasing pressure. Speeds down mines for instance being limited to about 20 ft/s (1200 ft/min) certainly for members of the public.
IIRC - a working maximum (increasing pressure) was about - 800 fpm

wes_wall

23rd Jun 2009, 00:15

PJ2

.......but we do know that at some point the fuselage/cabin structure was compromised/breaking up, at some point between 350 and sea level.

Excuse me, but how do we know this?

ACLS65

23rd Jun 2009, 00:30

PJ2 and HarryMann, thanks for both the quality explanation and patience.

If I follow this right then the CPC's receive inputs from the FMGS and ADIRS. I assume the altitude information from the ADIRS is determined in part from the static tubes.

Is the system normally flown in Auto mode?

Do you know if the ACARS msg corresponds to the Max Differential Pressure Limiter Function limit being hit or Cabin V/S 1800+-fpm?

I was mainly wondering of the cabin press ACARS msg was a sign of an actual problem (breakup, rapid altitude change, etc), or if bad altitude information either from ice buildup on the tubes or anomalous ADIRS info could have convinced the CPC that the A/C altitude had changed significantly beyond an amount it could compensate for, triggering the warnings and ACARS msg.

Also are the two pneumatic safety valves completely mechanical in nature? Everytime I turn around I run into another computer, so I thought it best to ask.

wozzo

23rd Jun 2009, 00:32

but we do know that at some point the fuselage/cabin structure was compromised/breaking up, at some point between 350 and sea level.

Excuse me, but how do we know this?

Because of all the debris floating around?

PJ2

23rd Jun 2009, 01:04

ACLS65;
Do you know if the ACARS msg corresponds to the Max Differential Pressure Limiter Function limit being hit or Cabin V/S 1800+-fpm?
Unless I have missed the interpretation of the ATA code somewhere, no, I don't think it corresponds to anything except the advisory message. Someone?

JD-EE

23rd Jun 2009, 01:39

wozzo (http://www.pprune.org/members/275610-wozzo)

I'd be pedantic and also note that the plane COULD have ascended a little in an updraft. Then I'd say the plane broke up somewhere between and including 37000' and sea level to satisfy the fussy folks.

It rather obviously broke up SOMEWHERE.

I'd also note the plethora of messages virtually minute by minute from 0210 to 0214 and nothing subsequently. Thus one can say the plane was (relatively) in control with things rapidly turning to slime through 0214 and probably broke up within two or at most three minutes.

That's not guaranteed - the chain of failures MIGHT have ended at 0214. But that pushes my "but coincidences like that don't happen" buttons.

JD-EE

HarryMann

23rd Jun 2009, 01:39

I feel compelled to take a step back for reflection at this stage, being minded that if Takata's track and crash site deductions proves remotely correct, then it's likely a quite titanic struggle to stay airborne took place that night...

wings1011

23rd Jun 2009, 01:46

As a reply to the above questions about wheather the aircraft did break up before it hit the surface of the water is at this stage impossible to say as said before . We simply do not know if it hit the surface in low or high speed such would explain why some pieces of seemed less damaged pieces were found .Even if it maybe would seem more logical that fuselage hit surface with high speed it must be proven first . Aircraft could have made a ditch or broken up in it's decend to the surface at some point.To my knowledge impossible to prove any of this scenarios just now. Both of them would almost give the same results in form of found debris at this point, or am I wrong ?
Regards
1011

Chu Chu

23rd Jun 2009, 02:03

To be absolutely 100% safe, you'd have to say the aircraft broke up somewhere between 37,000 feet and the bottom of the ocean. I don't believe for a minute it landed intact and broke up as it sank, but I'm not sure we can conclusively rule it out (anyway, ships often break up on the way to the bottom).

I suspect the recovered debris contain the information needed to determine where the break up occurred. But the pictures of the debris may not.

geobar

23rd Jun 2009, 03:06

Without trying to make any assumptions, I found some of the earlier maps relying on FAB PowerPoint presentations overlaid with BEA debris data points to be rather ambiguous.

Many of the FAB graphics appear to be designed for media consumption and minor? details such as the "Ultimo Reporte" position do not appear in the correct positions.

I know 'rer47' mentioned some pages back the link to:- Final Route of AF447 - an Analysis (http://countjustonce.com/af447/) and the map associated with it, along with some assumptions made could be worthy of consideration.

:=

lomapaseo

23rd Jun 2009, 03:12

[B]Aircraft could have made a ditch or broken up in it's decend to the surface at some point.To my knowledge impossible to prove any of this scenarios just now. Both of them would almost give the same results in form of found debris at this point, or am I wrong ?
Regards
1011

wings1011

For sure from just looking at the pictures and reading the news 'we' can't prove anything. That's up to the investigators.

However we do have inferences.

For one thig it has been reported that the investigators believe the aircraft broke up based on wreckage distribution and body and seat recovery.

Some of us read the recovered wreckage photos as mostly free fall damage (terminal velocities in a vertical direction). Some of which seems to have fallen out of the fuselage at altitude while other pieces were ejected from the fuselage sections when they free fell into the water.

Much of the conclusive stuff has sunk so here we are working with flotsam and jetsam. No harm in waiting for more conclusive evidence but meanwhile some of us are beginning to focus on fewer postulations based on our beliefs.

ClippedCub

23rd Jun 2009, 03:14

If there was a controlled ditching gone wrong, why would they ditch. We determined that they didn't encounter significant hail, but even hail taking out the engines would be a stretch. Like others, I'm leaning towards in-flight break-up from overload, spinning, etc, but there's nothing conclusively pointing that way, unless we can rule out a ditching.

Can anyone with more probe knowledge comment on what readings in terms of AS, AOA, and ALT we would expect to see with an ice clogged pitot?AOA is a separate probe or vane, don't think it uses static, Alt runs off static. AS is pitot and static.

Thought cabin pressure readings could be corrupted by static.

Can't remember the air data system schematic PJ2 posted, but the statics should be cross plumbed between left and right fuselage sides to neutralize effects of side slip and yaw rate, so they wouldn't be effected by yaw due to an engine out transient for example. Though something as severe as a spin probably would impact the static readings.

SaturnV

23rd Jun 2009, 03:22

takata, what if you flipped your analysis, so that the hypothesized impact points were west of 30W, and not east of it?

The Brazilian Air Force search grid for June 1 is limited to the projected track. On June 2, the Air Force does two rectangular grid searches that run eastward from the search grid of June 1. Searches on June 3 and 4 are in areas prediominately east of the track. To me that suggests that the June1-4 surface current was similar in velocity and direction (SW to NE) to what is shown in the chart for June 5, and the Air Force search pattern was based on the expected drift from the current.

If AF447 had turned west to divert around the CB complex, or even turned back toward Brazil when systems began to fail, and began breaking up around 31 or 32W, then the prevailing surface current may have carried parts of the wreckage NE to the point near 30W and 3N where the first bodies and confirmed parts of the plane were retrieved.

If the above hypothesis is correct, then I would expect that the first group of bodies that were recovered to be associated with the plane's impact with the water, and the second larger group of bodies which was subsequently found much further north to possibly have left the plane during the initial break-up phase. (There are enough bodies that the investigators should be able to match bodies with assigned seats and test the validity of the above hypothesis.) This clustering of the bodies generally into two distinct groups would be the reverse of what one might expect if the plane had remained on a NNE heading, and broken up.

ACLS65

23rd Jun 2009, 03:50

An interesting paper, The Evolution of Flight Data Analysis by Neil Campbell.

http://asasi.org/papers/2007/The_Evolution_of_Flight_Data_Analysis_Neil_Campbell.pdf

Passenger 389

23rd Jun 2009, 05:35

Re: debate on whether the aircraft may have been mostly intact until it reached the water (e.g., a low-speed controlled ditching, or "flat" spin).

Given estimated fuel aboard, and implied assumption that structural integrity was maintained until impact but almost certainly not beyond, what is the likelihood of a post-impact fire in that circumstance?

If the probability is high, then does the (reported) absence of charring on recovered debris or bodies tend to exclude a controlled ditching or similar hypothesis, and suggest structural integrity was lost prior to impact with the water? Or would debris and bodies initally sink (and have been protected from any fire), surfacing only later?

nyt

23rd Jun 2009, 05:47

CVR/FDR pinger heard:
Les boîtes noires de l'Airbus Rio-Paris repérées - Société - Le Monde.fr (http://www.lemonde.fr/societe/article/2009/06/23/les-boites-noires-de-l-airbus-rio-paris-reperees_1210102_3224.html)

alainthailande

23rd Jun 2009, 05:48

As "Le Monde" is generally regarded as the most reliable french newspaper, I thought I'd quote it here :

Les boîtes noires de l'Airbus Rio-Paris repérées - Société - Le Monde.fr (http://www.lemonde.fr/societe/article/2009/06/23/les-boites-noires-de-l-airbus-rio-paris-reperees_1210102_3224.html)

Quick and dirty translation : the "Nautile" submarine has beed diving on June 22nd in an attempt to locate the black boxes of the Air France Airbus that vanished on June 1st over the southern Atlantic. It was guided by a very faint signal picked up by one the the French Navy ships. Search is made very difficult due to very montainous nature of the seabed with depths around 5,000 meters. The batteries of the black boxes should last for another week or so.

Mike-Bracknell

23rd Jun 2009, 05:48

Given estimated fuel aboard, and implied assumption that structural integrity was maintained until impact but almost certainly not beyond, what is the likelihood of a post-impact fire in that circumstance?
ironically, the more fuel in the tanks the less chance of a fire due to the limited presence of vapour.

woodja51

23rd Jun 2009, 05:53

Okay - heres a long shot and if its jumping to conclusions before the full report sorry.

Take off 7pm, three pilot crew one captain, two f.o,s probably Air France cadet system pilots.( nothing wrong with that but bet they probably have never been upside down in an aircraft let alone a big one hence the usual degradation in handling skills that modern aviation teaches)

Captain goes back shortly after take- off to first class or what ever rest they use, F/Os left in charge.

jet hits some cb's , maybe painting on radar, maybe not.

For one reason or another - all of which have been covered, jet reverts to alternate law, unreliable airspeed actions required ( memory items ) as A/P and A/T have disconnected.

Still in bumps, PF - who I would assume should be in the RHS... but maybe not....very low buffet margins @ F350. Speed tape dancing up and back.. maybe ...bearing in mind that depending on the sort of protection remaining any thing is possible with the pitch /bank/altitude/speed and thrust levers.

In attempt to control aircraft, jet is overhandled ( have seen it before with pilots "wanting to fix something'...

exacerbate situation and loose control. Jet breaks up at some point.

Jet aircraft in spiral dive or simliar, ROD works out to be about 12 000 FPM ( based on B707 loss of control over East Sale - hit water from 6000 feet in about 25 secs after demo of VMCG loss of control. yaw rate enough to shear off inboard engine and take out #4.

They managed to get a may day out ... just ..

So may have these guys but who was listening? , and besides with the yaw and pitch rates I'm not so sure you might have thought more about getting back on the rails first.

All speculation but how come AIRBUS have just put out a bulletin related to unreliable airspeed flight and its importance.

Yes granted may have been environmental initial cause but if I thought that lightening and thunderstorm penetration would result in this occurence being any greater than 1 x10-9 then I would not still be sitting in a cockpit?

Which I have done on both B707/777 and Airbus 320/343.345 types for 13000 hours.

I have ultimate confidence in the machinery - it is the human side that bothers me now with the increasing reliance on automation to make ones job easy ... right up to when it checks out and says handing over....!!

Not to speak ill of the dead but this is becoming all too common.

Ready for all incoming now... If I am wrong I will retract all of this conjecture.. just seems plausible to me...
W

mingocr83

23rd Jun 2009, 05:56

Yup, it looks like that a weak pinger has been heard...

5000mts depth, at least is something! Hopefully the FDR/CVR can be recovered succesfully

Roy Bouchier

23rd Jun 2009, 06:05

woodja51 (http://www.pprune.org/members/136195-woodja51)

Did you bother to check the experience of the co-pilots?

PJ2

23rd Jun 2009, 06:12

JD-EE;
...to satisfy the fussy folks.

It rather obviously broke up SOMEWHERE.
Indeed it did. Somewhere between 40,000ft and sea level. What I meant seems clear to most.

Anxiously awaiting further news regarding the reports of hearing pings and anything else, signalling large metal structures, or not. Are there any further photos of wreckage found or are we at an end, more or less?

ACLS65

23rd Jun 2009, 06:18

Air France "black box" signals located
Tue Jun 23, 2009 2:06am EDT

PARIS (Reuters) - Signals from the flight data recorders of the Air France airliner that crashed into the Atlantic killing all 228 people on board have been located, Le Monde newspaper said on its website on Tuesday.

An Air France spokeswoman said she could not confirm the report.

Le Monde said French naval vessels had picked up a weak signal from the "black boxes" and that a mini submarine had been dispatched on Monday to try and find them on the bottom of the rugged ocean floor.

The "black boxes" may contain vital information that could help explain what happened when the Airbus A330 aircraft crashed into the sea en route from Rio de Janeiro to Paris on June 1.

(Writing by James Mackenzie, Editing by Ralph Gowling
Air France black box signals located: report | U.S. | Reuters (http://www.reuters.com/article/newsOne/idUSTRE55M0YX20090623)

FrequentSLF

23rd Jun 2009, 06:22

Take off 7pm, three pilot crew one captain, two f.o,s probably Air France cadet system pilots

Captain 11,000 hrs TT (1700 on Airbus A330/A340)
Copilot 3,000 hrs TT (800 on Airbus A330/A340)
Copilot 6,600 hrs TT (2600 on Airbus A330/A340)

greenslopes

23rd Jun 2009, 06:44

Article from: Reuters
SIGNALS from the flight data recorders of the Air France airliner that crashed into the Atlantic killing all 228 people on board have been located, Le Monde newspaper said on its website today.

An Air France spokeswoman said she could not confirm the report.

Le Monde said French naval vessels had picked up a weak signal from the "black boxes" and that a mini submarine had been dispatched yesterday to try and find them on the bottom of the rugged ocean floor.

The "black boxes" may contain vital information that could help explain what happened when the Airbus A330 aircraft crashed into the sea en route from Rio de Janeiro to Paris on June 1.

three eighty

23rd Jun 2009, 06:50

CNN reporting that the boxes have been found.

Left Coaster

23rd Jun 2009, 06:51

CNN reporting that a sub has been dispatched to the area after the "ELT's" from the recorder (s) have been heard...hope it's true...

Weapons_Hot

23rd Jun 2009, 07:01

Woodja51

As a long-haul heavy driver, I too have this notion. From experience, knowing the experience of the relief crew operating the airplane while I am "down the back", I "rest" very uneasily.

It is not about hours on type, it is about EXPERIENCE. Typically, a long-haul FO or SFO with about 5,000 hours, has, if all he/she has been flying is long-haul, about 10-15% of total time as experience. The other 85-90% is either sleeping "down the back", or watching the AP do the hard stuff.

The expression "clutched defeat from the jaws of victory" comes to mind (along with "if it ain't itchy, don't scratch it").

Time for the Kevlar flack jack.:ouch:

alainthailande

23rd Jun 2009, 07:11

FWIW the Ministry of Transport just made a statement denying the fact that the flight recorders have been located (as said in an article of Le Monde newspaper).
The statement says that signals picked up by the French Navy vessels were unrelated to the Airbus.

cats_five

23rd Jun 2009, 07:13

Are the 'pings' from the boxes so distinctive they cannot be confused with other noises? Hope not. And hope the mini-subs are able to retrieve the boxes.

Nightrider

23rd Jun 2009, 07:29

Weapons_hot and Woodja51

While most certainly a lot of colleagues may agree rather silently with your outspoken assumptions, this is not solely a longhaul problem. On short- and mediumhaul the situation is very similar. However, the nature of short- and mediumhaul allows for a more quicker build-up of "hands-on" experience.
There was, while aviation was definately something different as it is today, the rule in the major field players manuals that the way to longhauls office chairs has to start on shorthaul and with a visit for some time on the mediumhaul fleet.
Things have changed and may not everywhere the same anymore.

Comming to the point of this, with the unease you mentioned in your post, would anyone expect to find the captain in bunk while crossing the most challenging part of the journey?
Guess not!

Let's wait for the boxes to tell us more.

Pontius Navigator

23rd Jun 2009, 07:41

Guava Tree

23rd Jun 2009, 08:50

If the posts of Woodja51 (#2199) and Weapons Hot(#2205) turn out to be on the mark, then it would show a major flaw in the “Cruise Captain” concept for long haul flights. In fact, in many airlines, cruise captains have almost zero experience of manual flight from the left hand seat, and that manual flight experience would very likely be only in the simulator.
Perhaps at the time of dividing up and allocating “on seat” and “rest” periods among the pilots for the flight, that part of the flight which eventually proved to be most critical was not considered to be likely to be the most challenging.

captainflame

23rd Jun 2009, 08:56

But a spokeswoman from the BEA, the French air accident authority, noted that many sounds were picked up on the sea bed and investigators were not sure that what they had detected was from the flight recorders.

"It's not the first time sounds have been heard and we will be verifying this with all the equipment we have at our disposal," she said. "The search is continuing and we haven't found the recorders."

An Air France spokeswoman also said she could not confirm the newspaper report.

Le Monde said French naval vessels had picked up a weak signal from the flight recorders and that a mini submarine had been dispatched on Monday to try and find the "black boxes" on the bottom of the rugged ocean floor.

Not yet, not yet !

woodja51

23rd Jun 2009, 09:03

To those who threw me the info on the copilots experience, thanks - perhaps lost in translation but I was not casting aspersions merely observing what seems to be a common trend in modern aviation.

That is, the lack of hands on time when it really matters. Be it SOPS and automation, modern aircraft being so good etc etc.

I see that I may have some followers in what I was trying to say albeit poorly.

In as much as when something goes wrong it sometimes is better to 'wind the aircraft clock' as the memory item/checklist....then action the event... ( ooops can't wind the Breitling any more .. all GPS updated.. but you get the idea.)

Sorry for a bit of thread drift , By the way, here at EK they have just increased the experience levels for command. But as some of the F/Os have pointed out it seems that in our case we can have cadet pilots get command status ahead of other chaps based purely on hours with the company - and that experience gained from the school of hard knocks be it G/A or military is soon to count for very little. Some are excellent pilots, no mistake but have not had the benefit of lots of hands on/sectors etc.

It is this perception of hours actually meaning something that is the root of some of the problems aviation faces.

To the uninitiated, having 10 000 hours must mean that you are twice as good as someone with 5000 hours.

Guess reading 'fate is the hunter 'gives a better sense of what experience really means?

A pilot doing the same thing for 10 000 hours is not the same as doing a different thing for an hour 10 000 times .. you get my point.

As far as inflight rest goes, (at least in EK )the captain is not allowed to rest in first class when there is a bunk - located in the aft of the aircraft too....( yes I know captains prerogative and all)

But if I think that the F/Os are not up to speed to do an Atlantic crossing at night, avoid weather etc, it is time to chuck it in.

Anyway, I hope that some light is shed on the cause ( maybe B/B's found?) as otherwise nothing will be learned from the loss of so much life.

W

clearedtocross

23rd Jun 2009, 09:37

Please be aware that the pings are acoustic (sonic waves) and not electromagnetic (radio waves). Water is not an ideal isolator, any electromagnetic emission is absorbed within very short distance.

Sonic (audible) pings have a carrier frequency in the low kilohertz band, ultrasonic waves may go up to 100 kHz. Modulation of such a "carrier" is physically possible but technically not feasible, since the "carrier" signal is far too weak to carry any useful and detectable modulation. In the deep sea, sonic waves do not propagate easily, they get deflected and echoed by changes in water temperature and density anomalies. But even if that problem would could be overcome, any modulated data transmission rate is intrinsically limited to a fraction of the carrier frequency, in an underwater application to something like a morse or telex transmission.

That's why the black boxes cannot transmit any data.

pax2908

23rd Jun 2009, 09:39

Pt probes question. In this AF note we read [my own rough translation] "March 2009 [...] Airbus recalls the procedure for maintenance and verification of the probes". The note does not say what anomalies (if any) were ever found during these checks. Could someone explain the procedure? In particular, can the drain holes be inspected visually on the ground, and how often is this done? Somebody mentioned the possibility of obstruction caused by corrosion; so this problem - if present - should persist long enough that it is detectable on the ground.

takata

23rd Jun 2009, 09:45

Hi,
PJ2:
Now it makes sense to talk about a "cabin rate of climb/descent" and one can readily see now why a cabin rate of climb of "1800fpm" up or down, is not only excessive, it is an indication that the CPC is not able to control cabin pressure, for whatever reason.
[...]
We cannot know the full meaning, (ie, was this a depressurization, a CPC fault or...?) of the advisory but we do know that at some point the fuselage/cabin structure was compromised/breaking up, at some point between 350 and sea level.
[...]
ACLS65:
Do you know if the ACARS msg corresponds to the Max Differential Pressure Limiter Function limit being hit or Cabin V/S 1800+-fpm? Unless I have missed the interpretation of the ATA code somewhere, no, I don't think it corresponds to anything except the advisory message. Someone?

This item is described by Eurocockpit.com (they have the full ACARS headers and content) as being a CABIN VERTICAL SPEED advisory.
Check this page (scroll down for the English text):
Eurocockpit - Archives (http://www.eurocockpit.com/archives/indiv/E009438.php)

I don't fully understand why Goodrich would say that the pitots probes are also sending data for the cabin pressure controler (and for engine monitoring) when I understand that such system would use the static pressure instead of the total pressure in order to control the pressurisation (Cabin delta P). Moreover, the pressurization schema is pointing ADIR 1+2+3 for the controler data.

It looks like each pitot probe is coupled with two static ports, then if the pitot probe is faulty, the static ports data are considered faulty also?
Or, not only the pitots probes were frozen but also the static ports?
In fact, if the static ports were considered faulty, all the sequence would make much more sense. In the Air Caraibes case, the sequence start with the TAT probes freezing before the AIS, Altitude, ADRs went nuts.

S~
Olivier

Professorrah

23rd Jun 2009, 10:09

Propagation through water provides more attenuation horizontally than it does vertically, due to amongst things, the lensing effects and characteristics.
The Box's can be deeper than the horizontal range and this is why the search is aided by a submarine and has better chances than a surface vertical.

BOAC

23rd Jun 2009, 10:11

takata - not familiar with the AB system, but I suspect something is 'lost in translation' on this. I would expect the CPC receives a digital 'static' input from AD1/2/3/ and thus if they are either shut down by crew action or isolated by software the source may be lost? Whether there is a backup?

Incidentally, did I see someone claiming the ACARS message could also be cabin alt >8.8k? Is this a revision of the 'definite' claim that it was only an up or down rate signal?

UNCTUOUS

23rd Jun 2009, 10:26

Takata said (post 2219)I don't fully understand why Goodrich would say that the pitot probes are also sending data for the cabin pressure controler (and for engine monitoring) when I understand that such system would use the static pressure instead of the total pressure in order to control the pressurisation (Cabin delta P). Moreover, the pressurization schematic is pointing ADIR 1+2+3 for the controler data.

It looks like each pitot probe is coupled with two static ports, then if the pitot probe is faulty, the static ports' data are considered faulty also?
Or, not only the pitot probes were frozen but also the static ports?
In fact, if the static ports were considered faulty, all of the sequence would make much more sense. In the Air Caraibes case, the sequence starts with the TAT probes freezing before the AIS, Altitude, ADRs went nuts.

For a possible explanation/resolution of the confusion about static pressure sourcing, see this post by OVERTALK:
link (http://www.pprune.org/rumours-news/376433-af447-80.html#post4998196)
.

barstow

23rd Jun 2009, 10:32

"That's why the black boxes cannot transmit any data."

I know this is going to sound pedantic, but the black boxes are sending a datum, as in the ping. They could easily be modified as you intimated to morse a fleet ID, which under ideal conditions would be helpful, but under poor conditions could be next to useless. However, the idea of a morse modulation would parallel the SETI model of looking for intelligence amongst noise, which seems to be the situation ATM.

Squawk_ident

23rd Jun 2009, 10:35

"Aucun signal émanant des balises acoustiques des enregistreurs de vol n’a été validé à ce jour. Des travaux sont régulièrement entrepris dans le cadre des recherches en cours pour lever tout doute sur des bruits perçus et tout fait établi sera rendu public."

No emanating signal of the acoustic beacons of the flight recorders was validated to date. Works are regularly undertaken within the framework of researches in progress to raise any doubt about perceived noises and all established fact will be made public.

Bureau d'Enquêtes et d'Analyses (http://www.bea.aero/fr/enquetes/vol.af.447/com23juin2009.php)

forget

23rd Jun 2009, 10:44

Tech Data on Dukane Underwater Locator Beacon.

Dukane model DK120 and DK100 Underwater acoustic locating beacons (http://www.dukcorp.com/seacom/Products/Beacons.htm)

fullforward

23rd Jun 2009, 10:58

Congrats on the excellent, no BS posts.
After 16k hrs including 13k plus hrs PIC on 737/67/320 and 777 make us see things clearer.

clearedtocross

23rd Jun 2009, 11:35

barstow, fully agreed, any 'intelligent' signature that can be correlated would make detection of the pingers easier, as long as it does not weaken the signal strength significantly.

Mr Optimistic

23rd Jun 2009, 11:39

suspect you have in mind a correlation detector (like gps) which can pull a signal out of noise. still need bandwidth and unfortunately water is a very dispersive and stratified medium.

22/04

23rd Jun 2009, 12:04

DorianB

23rd Jun 2009, 12:05

Jun 23, 8:04 AM EDT

Official: No black box signals from Flight 447
BY ANGELA CHARLTON
Associated Press Writer

PARIS (AP) -- French military ships searching for the black boxes of Flight 447 have detected sounds in the Atlantic depths but they are not from the Air France plane's flight recorders, French officials said Tuesday.
The official and French investigators denied a report on the Web site of the French newspaper Le Monde that French ships had picked up a signal from the black boxes.
The two recorders, key to helping determine what happened to the plane, which plunged into the ocean May 31, will only continue to emit signals for another eight days or so.
French vessels in the search area have picked up noises regularly, but subsequent investigation has revealed no link to the black boxes, French military spokesman Christophe Prazuck told Associated Press Television News.
"The black boxes have not been found. The black boxes have not been located. We're still looking for the black boxes," Prazuck said in English.
"Regularly they have alerts. They hear noises that could be related to the black boxes so they have to investigate these noises," he added, saying the French authorities "never" have had confirmation that any of the sounds detected were related to the black boxes.
The Airbus A330 plane came down in the Atlantic after running into thunderstorms en route from Rio de Janeiro to Paris. All 228 people aboard were killed. The cause of the crash remains unclear.
The French air accident investigation agency, BEA, said in a statement Tuesday that "no signals transmitted by the flight recorders' locator beacons have been validated up to now."
The BEA said work is continuing "aimed at eliminating any doubts related to any sounds that may be heard, and any findings will be made public."
Last week, BEA director Paul-Louis Arslanian sternly warned against any unconfirmed leaks in the investigation, saying they could mislead the public and unnecessarily worry or encourage the families.
Le Monde said a mini research submarine, the Nautile, dived Monday to search for the boxes based on a "very weak signal" from the flight recorders picked up by the French ships.
Searchers from Brazil, France, the United States and other countries are methodically scanning the Atlantic for signs of the plane.
French-chartered ships are trolling a search area with a radius of 50 miles (80 kilometers), pulling U.S. Navy underwater listening devices attached to 19,700 feet (6,000 meters) of cable. A French submarine is also searching.
The black boxes send out an electronic tapping sound that can be heard up to 1.25 miles (2 kilometers) away.
Ten of 50 bodies recovered from the Air France flight have been identified as those of Brazilians, medical examiners said.
Dental records, fingerprints and DNA samples were used to identify the bodies. Investigators are reviewing all remains, debris and baggage at a base set up in Recife, Brazil.

warmkiter

23rd Jun 2009, 12:23

to woodja 51 and followers like Fullworward...

could you please stop this crap about poor handling/decision making of FO/SFO flying the plane while CPT is in his rest.

"lack of hands on time when it really mathers" do you think your FO is learning less on his leg being PF than you when you are PF?

So the AF FO/SFO spent thousands of hours learning nothing what really mathers? what do you really think about your fellow pilots? Only captains gain valuable experience when flying and somebody who sits in the same plane flying next sector doesnt?
If this is the case, you should maybe check why you dont give your collague chance to do so?

"But if I think that the F/Os are not up to speed to do an Atlantic crossing at night, avoid weather etc, it is time to chuck it in"

So what do you do? stay awake or have a nightmare in your rest that the youngsters cant handle the heavy at night past 10W. Yeah, i have been flying with these CPTs too, really seldom, but a few times. I can tell they were not the sharpest knife in the drawer. Usually lack of selfconfidence and below average pilot skills make them feel unconfortable in the cockpit. they fear loss of control and therefore over react to any situation what they cant totally foresee. this makes them believe, only them can handle the plane and generates a distrust towards younger collegues.

if you have doubts about the skill and experience level of your crew, feel free to contact your training department:}

poor copilots will become poor captains and poor captains were poor copilots before the upgrade. i have encountered both, cross your fingers they dont fly together when the **** hits the fan...

in the AF447 there were professional collagues trying their best, which for sure had enough experience but maybe not enough luck that night...

back to fact finding...

L

screwballburling

23rd Jun 2009, 12:35

I am having trouble understanding the fact the french or indeed any military cannot detect the signals from these boxes.

The Submariners, IIRC, spend most of their lives trying to track down the "enemy" that is hiding, yet they are having difficulty detecting something that "wants to be found", that is sending out a pinger signal for a over 3 weeks.

Come next month will we be greeted with the garlic shrug and "non"?

Please feel free to shoot me down in flames, should I need to be.

CptRegionalJet

23rd Jun 2009, 12:42

.....if they send out signals at all that is!

Poit

23rd Jun 2009, 12:57

Thanks for your reply. I thought the source was reputable enough, it's a very well regarded and generally well-researched series (Air Crash Investigation). The fact someone put it up on youtube shouldn't lessen it's reputation. But granted, it may not contain the amount of detail that the full report does.

I should probably not have said anything until more solid info becomes available, and I apologise for the conjecture. That said, I'm sure you can see where I was coming from.

ORAC

23rd Jun 2009, 13:02

I am having trouble understanding the fact the french or indeed any military cannot detect the signals from these boxes. The ocean has layers of different temperature and salinity. These can act, for sound waves, in the same way a prism or mirror acts for a light wave - refraction and reflection.

Some sounds can be reflected between layers and travel across thousands of miles. Others can be emitted from just a few hundred feet below and not heard within a few miles.

Now consider a black box under 6000+ feet of water with multiple layers between it and the surface. take a submarine with a crush depth of less than a thousand feet. take a couple of submersibles with a greater crush depth, but miniscule range and no sonar.

Now consider how lucky one must be to detect the signal, and if detected, to locate the source.

I would have thought their best chance of a detection was the SOSUS chain.

PJ2

23rd Jun 2009, 13:16

woodja51;

Your post very well-stated and balanced. Ninety-nine percent of the time it isn't an issue, but each flight is different and it is up to the captain to determine how the flight will be handled.

I don't think the issue requires the sweeping comments as some who disagree with you have provided - more subtlety, borne of experience in doing this kind of flying, is needed to understand what is being said. It isn't a matter of trust, or being over-bearing, it is a matter of assessing all factors within an environment in which a high degree of competence and skill already exists. A brand-new CRP (Cruise Relief Pilot - not permitted to occupy the front seats below cruise), who has just passed all the airline's requirements isn't as competent as a CRP who has flown extensively in these areas either in a previous life or as CRP for a few years, and as a captain I would (and quietly did) assess enroute weather and judge that accordingly, as I should.

In my experience, the proper use of airborne radar is not taught and not every pilot is equally "into the books". (I wrote the post on radar (http://www.pprune.org/rumours-news/375937-air-france-a330-200-missing-20.html#post4967741)on the other thread for this reason.) If there is weather enroute, I think that is an important bit of knowledge to establish and was one factor among many which I quietly assessed before taking my break. It was never as black-and-white as some have characterized.

'nuff thread drift. Re the pingers - Damn.

Pontius Navigator

23rd Jun 2009, 13:18

I would have thought their best chance of a detection was the SOSUS chain.

If it is still operational.

Back to the original post. A towed sonar array would possibly need to be within 6000 feet of the source. That gives a swept width of just 2 miles. Allowing for overlap you are probably looking at a track spacing of 1.5 miles. To detect a source using a towed array I would guess no more than 9 kts - let us say 6 kts then. It would take 10 hours for one 60 mile sweep. It would take 60 hours to clear a box just 10 miles x 60.

Then you must factor in water depth. No good with your sensor at more than 6000 feet off the bottom.

If you imagine the ditching or crash distance of 15 miles in relation to track you are looking at a box 30 x 60 or 180 hours. That is over one week to sweep one box with no better than 70% probability.

Orestes

23rd Jun 2009, 13:31

22/04:

I still like best the idea that was put forward earlier in this thread: simply put the recorders inside the vertical stabilizer. It seems that portion of many airplanes is usually still recognizable after a crash, and it appears that the VS already has an inherent "release & float" characteristic.

screwballburling

23rd Jun 2009, 13:41

Thanks for the information on the inherent difficulties of detecting the signals from these boxes.

Placing them in the vertical stabilizer seems like a good idea. I think the authorities may have to look at redesigning the whole Sonia detection signal, strength and relocation of the BB's etc.

There has to be a better way than the installation and location, than at present.

Are we all going to have to wait for the next prang, in order to try and find out what may have happened in this tragedy?

Backoffice

23rd Jun 2009, 13:42

FDR & CVR – Why don’t they float, they don’t have to but it would be really useful if the memory modules did and what’s more, think 16GB USB stick, wouldn’t it be useful if the data was ported to more than one memory module and both voice and flight data were recorded on the same module.
Now the memory is so small it should be able to fit in just about anywhere, even wireless is possible.

One thought on the current situation. If the aircraft broke up and the rear section is partially intact, perhaps with the rear bulkhead still in place, it may actually be floating sub-surface due to trapped air and the recorders may not be immersed and therefore, not pinging but perhaps transmitting.

One item mentioned earlier – the press in emergency button, which would ACARS and transmit an automatic position report/heading/speed/altitude and keep doing it until cancelled is a brilliant idea, should be one on every joystick.

Can you imagine what would have happened if they had managed a successful ditching but not been able to talk to ATC or Paris first.

Lastly, I wouldn't rule out those other sounds, unfortunately fish and probably squid will be attracted to the wreck and whales of course eat squid so definitely listen for them.

skadi

23rd Jun 2009, 13:44

Mrs 22/04 said to me last night; why don't the recorders float; they could then be recovered with the seats bodies etc.

Set me thinking- put them in a box that it triggerd to release or eject by presence of water and pressure (depth) - is this a viable idea?

Wouldnt be new, the famous Lockheed 104 Starfighter had such device, an airfoil shaped floatable container behind the cockpit on top of the fuselage...

skadi

Pontius Navigator

23rd Jun 2009, 14:01

If the FDR and CVR floated you would need to add positional information too so that the crash location was marked. At present, the FDR usually marks the crash site too.

Graybeard

23rd Jun 2009, 14:13

Fortunately, the need for deep water recovery of the boxes is very rare. Rather than more crash recorders, I would put more money into pitot probes that don't ice up, AOA sensors that don't put out noise spikes, and ADIRU that don't get fooled.

GB

Litebulbs

23rd Jun 2009, 14:15

Or a rethink of the currently fitted ELT's!

ArthurBorges

23rd Jun 2009, 15:27

The link was kindly posted on page 110 here. It was updated at 1150LT (Paris), five hours after the first version went online.

The bottomline is somebody heard a noise but no confirmation it was from any blackbox.

------

French naval vessels captured a very weak signal. It was thought it might be from the black boxes. The Nautile was sent down on June 22 to attempt a recovery, expected to be difficult because of their location in the aircraft.

A BEA spokesman said the blackboxes had not been precisely located.

Ifremer Captain Philippe Guillemet of the "Pourquoi Pas" search ship said: "We're still looking...We get acoustic signals but none have been validated. Unfortunately."

Asked: "You haven't located the blackboxes at all?", Guillemet replied: "Absolutely."

The article then notes the blackboxes can be heard only up to 2 km away and that the seabottom is rough terrain at 5,000m depth.

It adds the blackboxes will ping for about eight more days; the next paragraph moves on to the identification of 11 bodies.
--------

---

Captain Bob

23rd Jun 2009, 15:51

I can only add my hours as experience as a International Captain for a Major Airline. All our FO's at our airline have been here for quite a while. I would not have a problem taking a nap during my rest cycle. However that was not always the case.

Several years ago when we were hiring we were putting guys into the right seat of the 757 and 767 aircraft into the International Division. A fellow Captain had told me that he just got through flying to LGW from DFW with 2 newbie's in the 2nd and 3rd seat. He told me he took his break in the cockpit just to make sure.

Well it comes my turn and I get called out to fly to ZRH. Meeting the 2 FO's in operations I noted that when I signed in for the trip the FO had 139 hrs in the 767 and the FB (extra First Officer) only had about 50+. This was not their total flying time however, just since they started flying for us and the time that they had in the 757/767 Fleet.

They were young guys probably mid to late 20's straight from the commuters with no jet experience at all much less flying a heavy aircraft across the pond over to Europe and beyond. I would be by myself for the most part.

Needless to say I did not take my break in the back. I simply couldn't let these two young guys get in over their heads while trying to coast out and follow the North Atlantic Track Procedures. I let them do as much as they could but they really were in over there heads experience wise.

Sometimes airlines rush to put people into certain positions simply because they can. They rely a lot on experienced Captains and experienced First Officers to pass their knowledge along to the next guy. Sometimes you get this gut feeling and you just have to make the best decision that you can. Even if you step on someone's toes. You are responsible for everything and everyone is counting on you.

Who was in the Cockpit on the Air France Flight? No one knows at this time. Only when and if they recover the CVR will they know. And lets only hope that they find it soon. We need the answer to this tragedy as soon as we can.

Me Myself

23rd Jun 2009, 16:09

I would opin that the unease experienced by 'experienced' captains is not due to a perceived lack of experience in their FO but from their own lack of experience in temporarily handing over their responsibility.

Sorry but that s rub...h. I ve seen it time and time again, when you re in the bunk the most amazing things happen. Not all the time. I ve flown with really outstanding F/O s who had their heads bolted the right way and this, regardless of age and origin.
Unfortunatly, I ve also experienced the worst kind, type, I m a cool dude and who the :mad: are you to dare telling me what I need to do. By the way, what s keeping you from upgrading my wife and kids right now ????
They are not the majority but they are enough to poison the well. What will management do ??? Actually nothing since the word for the lat 10 years has been " Do not rock the boat old boy, or you re the one who is going to get hurt ".
This might be totally unrelated with what happened to 447 but it gets people talking about their own experience and I find this telling.
By the way, some F/O s are also very relieved when the skipper goes to sleep..................and for some very good reasons too.

DJ77

23rd Jun 2009, 16:50

If what Weopons Hot made you nervous, exercise your right of choice and only fly with carriers who don't use 2nd officers/cruise pilots.

French unions always rejected the notion of "cruise pilot" and you can be assured that the guys at the controls in AF447 were checked out satisfactorily against the same technical standards. Also, AF has a policy of not pairing low experienced pilots on the a/c type when scheduling crews.

However, I can't help feeling that the standards are no longer what they used to be.

RatherBeFlying

23rd Jun 2009, 16:53

Hoping that they can find this black box but not looking good is it?The usual way to find the CVR and FDR is to follow the pinger, but even in shallow water that can take several days as we saw last year with the a/c that hit the water on a handover flight.

Once the pinger battery packs it in, they will have to go to sidescan sonar. Given the irregular bottom profile and the likelihood that the section containing the CVR and FDR will be just one of many dispersed pieces, it's going to take a loooong time and a serious pile of megabucks.

But it's a doable.

Amazon.com: Ship of Gold in the Deep Blue Sea: Gary Kinder: Books (http://www.amazon.com/Ship-Gold-Deep-Blue-Sea/dp/0375703373/ref=ed_oe_p) documents the years of effort required to find a historic shipwreck on a flat seabottom.

FE Hoppy

23rd Jun 2009, 18:30

Take it from me. If they have a ping it won't take long to triangulate it. Now a very big ocean is much much smaller. My bet is recovery of the recorder(s) within 3 days.

DorianB

23rd Jun 2009, 18:49

Some very interesting new photos (once clicked on the images will open in a reduced size, click on again to see a very large version):

http://i635.photobucket.com/albums/uu79/DorianBanks/th_fox-1.jpg (http://s635.photobucket.com/albums/uu79/DorianBanks/?action=view&current=fox-1.jpg)

http://i635.photobucket.com/albums/uu79/DorianBanks/th_alfa_gastao_motta.jpg (http://s635.photobucket.com/albums/uu79/DorianBanks/?action=view&current=alfa_gastao_motta.jpg)

http://i635.photobucket.com/albums/uu79/DorianBanks/th_echo-1.jpg (http://s635.photobucket.com/albums/uu79/DorianBanks/?action=view&current=echo-1.jpg)

http://i635.photobucket.com/albums/uu79/DorianBanks/th_delta-1.jpg (http://s635.photobucket.com/albums/uu79/DorianBanks/?action=view&current=delta-1.jpg)

kuningan

23rd Jun 2009, 19:06

This: wreckage (http://s635.photobucket.com/albums/uu79/DorianBanks/?action=view&current=alfa_gastao_motta.jpg) (bottom right) looks like it has part of a Business Class seat (http://www.sevenstarsandstripes.com/content/airline/airfrance/2007/seat.gif).

grizzled

23rd Jun 2009, 19:07

DB:

As usual, thanks for posting these pics. And thanks again for organising the album.

Though there isn't much we can conclude about what happened (yet), the more we see photos of items from the aircraft, the more we can posit what didn't happen.

Grizz

jimjim1

23rd Jun 2009, 20:04

I have searched pprune and cannot find any reference to this paper. It does seem that the search is not working, but maybe I am not driving it properly?

Blank Design page (http://strumpfer.com/Papers/HelderbergSearch.htm)
DEEP OCEAN SEARCH PLANNING:
A CASE STUDY OF PROBLEM SOLVING
Johan Strümpfer (http://strumpfer.com/Contact.htm)[1]

In late 1987 a Boeing 747 of South African Airways crashed into the Indian Ocean after an onboard fire.

Seems to me a but self congratulary but also seems to have a lot of good material. I read it a few days ago but the circumstances seem very similar to the AF447 and they found one of the recorder after about a year. I am not sure what the bottom was like but the depth was similar.

Will Fraser

23rd Jun 2009, 20:19

Post # 2243

My Friend, Precisely!! Some cannot see the forest for the trees. I think there will be a great deal of rethinking and redirecting of technology and resources after this accident gets sorted out. Training, pilot/box interface, newer com, tx, mx, wx, all of it. We got complacent and lost our way a bit.

Regards, Will

ACLS65

23rd Jun 2009, 20:24

Thanks for the posted link jimjim1.

Just a couple of tidbits that jumped out at me...

The search for the wreckage was continued using side scan sonar, a system of using sound reflections to record the presence of objects on the sea bottom. To achieve this in the ocean depths in the accident area a cable of 9 kilometres long was required to tow the sonar near the ocean bottom. The equipment again is highly specialised and brought its own logistical problems. It was this equipment that located the sea bottom wreckage.The conditions under which the search was conducted were beyond the designed situation for this type of search. The consequence was a large number of false alarms. Of the total number of pinger like sources that were recorded, 32 had to be considered, of which 13 could not be rejected as possible actual pingers. They give you hope for even after the recorder pinger batteries fail, but also point out "false alarms" which were not something I had ever considered.

This was in 1987, I am hoping that the newer equipment for detecting the pings and advances in side scan sonar give the current searchers a bit more to work with.

DorianB

23rd Jun 2009, 21:29

For the region CSS Dakar/Senegal:

AFR 447 SAR OPERATIONS
END OF SAR OPERATIONS.

We inform you that we have stopped the search operations in our area. If you need supplementary means of assistant we could send you our research means.
We extend our great appreciations to all of you helpful support in fulfillment of our duties vis-à-vis the international SAR community.

Regards,
POUR LE COLONEL, CHEF D´ETAT-MAJOR DE L`ARMEE DE L`AIR ET DIRECTEUR DU CCS.

___________
FULL TEXT

23/06/2009 - Informações sobre as buscas do voo AF 447 da Air France - Nota nº 40

O Comando da Marinha e o Comando da Aeronáutica informam que, nesta terça-feira, dia 23, não foram avistados corpos na área de buscas, tendo havido apenas visualização de material diverso, de forma esparsa e em pequena quantidade.

O SALVAERO RECIFE recebeu hoje mensagem do Centre de Coordination de Sauvetage (CCS DAKAR/Senegal) informando sobre a suspensão em definitivo das atividades aéreas de busca sob coordenação daquele país.

O documento também destaca a ajuda prestada pelo Brasil, apoio esse fundamental para a comunidade internacional de busca e resgate.

Consequentemente, a partir de hoje, apenas aviões sob a coordenação do SALVAERO RECIFE permanecem nas buscas.

Leia, na íntegra, a mensagem do CCS DAKAR.

AFR 447 SAR OPERATIONS
END OF SAR OPERATIONS.
We inform you that we have stopped the search operations in our area. If you need supplementary means of assistant we could send you our research means.
We extend our great appreciations to all of you helpful support in fulfillment of our duties vis-à-vis the international SAR community.

Regards,
POUR LE COLONEL, CHEF D´ETAT-MAJOR DE L`ARMEE DE L`AIR ET DIRECTEUR DU CCS.

____________________________

I guess this graphic shows that region, seems kinda odd to me that they would stop searching in this region...considering the northerly current....

http://i635.photobucket.com/albums/uu79/DorianBanks/th_1-50.jpg (http://s635.photobucket.com/albums/uu79/DorianBanks/?action=view&current=1-50.jpg)

tubby linton

23rd Jun 2009, 21:35

Looking for the boxes.
I asked a friend who does analysis for various people to look at the problem of how to find the boxes and received the following reply:

For these sorts of searches, the best possible search pattern is to just "mow the lawn." If that's the case then the probability of detection is just the ratio of the area searched to the area to be searched. Generally, though that estimate is excessively optimistic. In practice few searchers actually conform to that plan for a myriad of reasons. Generally what actually happens is that they just randomly churn around places they haven't looked before until finally they find something. Statistically, the worst you can do is just by random chance. Is given by the cumulative exponential distribution.

Pd = 1 - exp(2*R*v*t/A)

where

Pd = the probability of detection
R = is the sensor range
v = the searcher speed
t = search time
A = size of the area to be searched.

Machaca

23rd Jun 2009, 22:16

Good eye Kuningan - parts of 1st Class seats:

http://i337.photobucket.com/albums/n385/motidog/AF447-seats01.jpg

ACLS65

23rd Jun 2009, 23:16

Figures, now I find a pic that probably shows the large galley structure from the previous set of pics in the upper right hand corner. :ugh:

Google Image Result for http://farm3.static.flickr.com/2184/2355546554_772bf0c5b7.jpg (http://images.google.com/imgres?imgurl=http://farm3.static.flickr.com/2184/2355546554_772bf0c5b7.jpg&imgrefurl=http://bizflight.*************/2008/10/air-india-a330-200-executive-class.html&usg=__Elh040LuM_5du9CCmo8ZO2r3Vsc=&h=375&w=500&sz=132&hl=en&start=72&um=1&tbnid=EVsSOadtkkk3RM:&tbnh=98&tbnw=130&prev=/images%3Fq%3Dair%2Bfrance%2Ba330%2Bfirst%2Bclass%2Bpictures% 26ndsp%3D18%26hl%3Den%26client%3Dfirefox-a%26rls%3Dorg.mozilla:en-US:official%26sa%3DN%26start%3D54%26um%3D1)

Any idea if the large rectangular object in the new pics is the coat closet like area in the upper left of this pic?

The piece in the wreckage looks like it has ridges for holding something like trays that would be slid in, think refrigerator shelves, but with only a couple inches between them.

ACLS65

23rd Jun 2009, 23:42

ornis:

My post back at #2194 has a link to an interesting paper about the past and future of the recorders.

lasers

24th Jun 2009, 01:07

Eurocopit www.eurocockpit.com/images/PFR447.php (http://www.eurocockpit.com/images/PFR447.php) has puplished updated and more complete ACARS messages:

http://www.eurocockpit.com/images/PFR447.png

ACLS65

24th Jun 2009, 01:14

In the first two pictures there is a piece with a circular hole about 6" in diameter.

Could this be the inspection window in the cabin door?

YouTube - A330 Cabin Door (http://www.youtube.com/watch?v=k59CbAdGFjE)

ClippedCub

24th Jun 2009, 02:05

A = size of the area to be searched.The area in this case will be projected area of the underwater mountains, and will be greater than a planform area.

Can't visualize much success from the side scan sonar seeking objects. Side scan outlines shapes when looking for objects on a flat sea floor.

OVERTALK

24th Jun 2009, 02:59

No matter how much you computerize and automate and digitize via pressure-sensing transducers, the pneumatically derived values of pitot pressure (dynamic plus static) come from the same sources. However, courtesy of automation, the engineers are free to determine whether or not they should integrate the port and starboard digitized values and compute an ongoing average value or also incorporate (and integrate) the "pitot-derived static" in order to "iron out" the minor errors at different speeds and attitudes/altitudes related to the static sources (aka position error correction).

As you'll no doubt be aware, the source of the speed display on an ASI is the result of "balancing" (or deducting) the "static source derived" static pressure against that (perhaps slightly different) static pressure that's part of the dynamic input to the pitot heads. That's why, when water in static lines freezes, the ASI will wind back to zero in a climb. It only takes about 2000ft in a 210kt climb as the pitot-derived component of static is overcome by the much greater trapped static line pressure. I've seen that on a number of occasions due to aircraft being parked in rain without static port bungs in place and the "sucked in" rain-water freezing as you pass freezing level in the climb (altimeter stops, VSI reads zero). You're then in POWER + ATTITUDE = PERFORMANCE country. If you're feeling destructive, you can depressurize and smash the glass face of the VSI to get a source of unfrozen static pressure and restore speed display, altimeters and VSI.

When you take into account that there are at least two (sometimes three) pitot heads (e.g. the A330/A340) and multiple static ports, then the engineers are "spoilt for choice" when it comes to sourcing and integrating their data flows to the ADIRU's.

It's never done as simply as you may think. Keep in mind also that normally there will be little difference between port and starboard sources (airplanes being symmetric and normally flying symmetrically). However an inadvertent spin entry from a coffin corner loss of control situation may change that port/starboard equivalence in both pitot and static - and be capable of providing the sort of speed disagreements between ADIRU's that would generate the messages sent by ACARS in the final 4 minute plummet from altitude in a spin/spiral.

Hello,

I was reading your post

1582 (http://www.pprune.org/rumours-news/376433-af447-80.html#post4998196) and

noticed this term:

"pitot derived static"

Am I to understand that it is possible to extract static pressure from the pitot line?

I'm a curious glider pilot.

Cheers,

Graybeard

24th Jun 2009, 03:23

May I suggest that general shortcomings and suggested improvements to the fright recorders be explored in Tech Log? They are less likely to be deleted there.

GB

deSitter

24th Jun 2009, 03:40

You know it doesn't do much good to go on and on about every last piece of debris. There is a lot of good information that comes out of control surfaces and so on, in order to determine why the airplane quit flying. There is almost no good in debating all these smashed seats and entertainment systems and oxygen bottles and on and on. That kind of speculation is too near the experience of the dead to countenance.

-drl

ClippedCub

24th Jun 2009, 03:47

too near the experience of the dead to countenance.

Respect your compassion. Had a tough time reading about China 611 and infants still being clutched by mothers. Also respect the investigators that can sequester that to another part of their mind and inspect/dissect all the evidence so this particular fault won't readily happen again.

takata

24th Jun 2009, 05:45

ClippedCub:
Affirmative. Saying that damage, and the damage to a perceived forward door, might be too severe to think a forward galley would survive intact. Hard to say either way.Actually this galley is from the business class, at left hand, boarding from door 2L. It is the only one in aircraft with 5 chariots. A good part of the wreckage identified is from this part of the aicraft.
See the post here:
http://www.pprune.org/rumours-news/376433-af447-102.html#post5009874

ClippedCub:
However an inadvertent spin entry from a coffin corner loss of control situation may change that port/starboard equivalence in both pitot and static - and be capable of providing the sort of speed disagreements between ADIRU's that would generate the messages sent by ACARS in the final 4 minute plummet from altitude in a spin/spiral.Posited that a few pages back. So far unchallenged.
Interesting explanation but, so far, about 35 other A330/340 pitots probe incidents occured (discounting AF447) as reported now, at cruise level, including 9 AF flights, most triggering the same sequence of ACARS and none were actually stalling/spinning but flying strait and level. So, obivously this theory is not only challenged but already debunked. This sequence is very well documented by Air Caraibes case and its obvious cause was probes freezing. If any stall/spin occured to AF447, it was after the probes problem, and wasn't the origin but could be the consequence.

S~
Olivier

takata

24th Jun 2009, 06:08

AMF:
It's a assumption to believe the autopilot did so as a result of equipment malfunction, and it was the first recorded ACARS message. If AP parameters (say, pitch, roll, trim limits due to altitude hold, etc) were exceeded and it thus disengaged, what were the forces acting on the aircraft that caused those exceedences?"first recorded ACARS"... no, they are unsorted and should be sorted in order of priority, plenty are triggered at the same time.
AP & A/THR are kicked off because the aircraft go to ALERNATE LAW2 as a consequence of the unreliable AIR DATA, etc. No exceedences of forces acting on the aircraft are needed to do that, but it certainly won't help for crossing a [undetected?] tropical CB.
See above post, the ACARS are now published in their integrality.

S~
Olivier

JuggleDan

24th Jun 2009, 06:25

Eurocockpit (http://www.eurocockpit.com/) reports another AF document pointing at a known problem with pitot probes:
French version: page 1 (http://forfaiture.freeservers.com/AF447_1.pdf), page 2 (http://forfaiture.freeservers.com/AF447_2.pdf)
English version: page 1 (http://forfaiture.freeservers.com/AF447_3.pdf), page 2 (http://forfaiture.freeservers.com/AF447_4.pdf)

Note that this AF maintenance from Aug. 2008 states that:
The purpose of this NT is to gather information in order to confirm the involvement of pitot probes in case of "NAV IAS discrepancy"
At the time of the creation of this NT, a case on THT and six cases on A340 AFR have been reportedFurthermore, if I'm not mistaken, several warnings reported in this note are found in the ACARS messages received from AF447

marchino61

24th Jun 2009, 07:57

The links in the post above don't work, but if you click on them, then copy and paste the address into the address bar and hit return, you will see the document. The address still looks the same, but it is not linking from a "foreign" site if you do this and hence is OK.

AMF

24th Jun 2009, 08:43

takata Quote:

"first recorded ACARS"... no, they are unsorted and should be sorted in order of priority, plenty are triggered at the same time.
AP & A/THR are kicked off because the aircraft go to ALERNATE LAW2 as a consequence of the unreliable AIR DATA, etc. No exceedences of forces acting on the aircraft are needed to do that, but it certainly won't help for crossing a [undetected?] tropical CB.
See above post, the ACARS are now published in their integrality.

S~
Olivier

My point being, crossing a (undetected?) tropical CB would, in addition to acting on all the aircraft's axis, also raise hell with the air data, especially in terms of miscompares. A 100mph vertical gust equals about 9,000 fpm. In a fast-building CB the measured-by-water droplet vertical velocities can approach twice that....18,000 fpm, and the real updraft speed (since droplets move more slowly) even higher.

There isn't an aircraft designed that can cooly transition though something like that even with all systems functioning. Inadvertent CB Penetration 101 includes not only disabling any altitude hold fuction and accepting altitude deviations, but also accepting radical airpspeed fluxuations, not only because the indications may or will become unreliable in relation to true aircraft speed, but also because the multiple Air Data systems may become unreliable in relation to each other due to unequal forces acting on the aircraft.

Meanwhile, with these unreliable and conflicting indications, a hand flying pilot, possibly in severe turbulence, must still maintain the reality as it exsists out on the wing...airflow somewhere between low-speed buffet and mach buffet, with the margins between them fairly close at high altitude. Overspeeding is a world of hurt just as big or bigger than stalling. And at high altitude and weight the pilot must stay within those close buffet margins while not executing anything near the rolling action he could at lower altitudes to deviate his course around more weather or counteracting turbulence and/or dutch roll. To do so at that high altitude risks a load factor-induced stall through a raised low-end buffet margin.

Worst of all, all those conditions can and do exist in the clear air above developing CBs, and inside those that don't contain enough moisture at that paint clearly or produce icing of any significance.

Dave Gittins

24th Jun 2009, 09:30

I start by announcing that I am just a 200 hr Cherokee driver.

Is the suggestion that severe weather/turbulence/updraft etc caused by a CB and with the aeroplane in "coffin corner" of the flight envelope caused an upset ?

Is there then postulation that the upset was of such a magnitude that even when the aeroplane got to (say 10,000 ft) a level where the available speed range was greater and manoeverability and controllability higher it could not recover ?

Is this hypothesised to be through speed / attitude ?

I thought only the "T" tails (discovered I think by G-ARPY) got into a locked in unrecoverable condition.

These are questions so be nice to me please.

Dave

Sober Lark

24th Jun 2009, 09:38

I agree with you desitter. Finding pieces and personal items from Air India to hand in to authorities was hard for me. Posting photos of seats etc from AF must be unbearable for relatives and doesn't add to any professional understanding.

HarryMann

24th Jun 2009, 09:47

Yes, there have been suggestions/speculation in these pages & elsewhere since day 1 that a high alt. upset might have occurred..

Aircraft of this type can remain out of control, in other modes than the 'T-tailed deep-stall'... e.g. the spinning characteristics are not explored fully during flight-test.
And of course, if IMC instruments are not all available, even a spiral dive can easily become critical.

Reading an assortment of the past posts should give you a good background to what might be possible or not

DJ77

24th Jun 2009, 10:16

AMF

Nice scenario ... for Hollywood.

AMG65

24th Jun 2009, 12:09

You have to chill mate. :=

DJ77

24th Jun 2009, 13:59

AMG65 and AMF

You have to chill mate.

No overheat here. And I do not mean anything personal.

But AMF's description should be supported by references to theoretical or experimental studies or even personal experience (if any). Otherwise, it looks like fiction.

See, for example, Tim Vasquez's analysis: no updraft higher than 40kt (about 4000 ft/mn), not "18000 ft/min, even higher" as mentioned, and no heat sink signature observed.

Of course, I don't intend to suggest that the pilots were not faced with a very severe situation flying in (or close to) bad weather without reliable airspeed indication.

Hyperveloce

24th Jun 2009, 14:31

Hi there.
The more I think to the debris location and to the last minutes of the AF 447, more confuse the whole picture becomes. Here is an agregated map from different sources (brezilian AF, NESDIS and OSCAR/NOAA, tim Vasquez analyze,...).
http://img199.imageshack.us/img199/6099/crashsite2k.jpg
OSCAR - Ocean Surface Currents Analyses - Realtime (http://www.oscar.noaa.gov/datadisplay/)
QuikSCAT (http://manati.orbit.nesdis.noaa.gov/quikscat/)
The surface current and wind fields are derived from satellite observation from June the 1st to June 12 (see Quickscat and NOAA databases): the currents rotate between 3°N and 4°N from a S-SW-->N-NE direction to a S-SE-->N-NW direction and remain <0.10m/s. The drifting buoys can give clues for the following days. The wind was oriented to the west, between 4-5 kts and 15 kts. You can get clues about the offshore swell via EVISAR/ASAR (S-SW-->N-NE direction).
Atlantic Ocean / Swell animation from ENVISAT ASAR instrument / Movies / View / Data / CERSAT - Le CERSAT (http://cersat.ifremer.fr/data/view/movies/swell_animation_from_envisat_asar_instrument/atlantic_ocean)
I don't get this:
If the AF 447 had been in nominal cruise speed between 02:10Z and 02:14Z (above mach 0.8 ?), it would have been too far to the north to explain the first body/debris location of June 6 ? (to be found where they were found, they would have to drift toward the west, but in this area, the currents and the swell were oriented toward the N-NE).
Would it suggest that between 02:10Z and 02:14Z, the AF 447 was already in big trouble, loosing rapidly altitude, that it never got out from the mesoscale convective system and crashed just after the last ACAR between the two red points ? (possibly just exiting the MSC system)
Under this hypothesis, the drift between June 1-June 6 (from crash area to first boby/debris) would have been largely smaller in 5 days than the drift between June 6-June 9 in 3 days (then from June 10 and June 15, at W-NW of TASIL in the search area, is oriented toward the NW. The satellite time series show that the current speed increased a bit between 6 and 12 but only to reach 0.12 m/s: it does not explain the different drifts by far): does it suggest that the bodies and debris (recovered between the 6 and 9 of June) were initially dispersed ?
Hope these are not silly/already answered questions.
Jeff