AF447

[wilyflier]

rollbacks

Extreme "outside normally expected maxima" of atmospheric high altitude water/fine ice content has been quantified elsewhere recently . I recall 4 times previous accepted figures
.Does a double engine ice event and subsequent controlled descent actually fit all the ACAS signals without further complexity.?
(Silly me, didnt note the web source)

[WhyIsThereAir]

Does a double engine ice event and subsequent controlled descent actually fit all the ACAS signals without further complexity.?

Does an engine flameout from icing make an ACARS message of any type? If so, is it a failure or a warning message? Is there a cockpit flag or indicator if an engine flames out? If so, it might make an ACARS warning message.

We know that ACARS can't send after both engines are down. But it appears that it should be able to send if only one engine is down.

If an engine down creates a failure message (and only a failure message), and both engines go down within less than 60 seconds of each other, then we would not see the failure because failures are held for 60 seconds, and the AC1 bus might have gone down before the close of the window.

On the other hand, if an engine down causes a warning message, or causes a cockpit flag or indicator that would be reported with a warning message, then since warning messages are sent 'immediately' (within a few seconds, normally) then I'd expect to see something about one or both engines going down before power to the ACARS data chain was lost.

[JenCluse]

If only . . .

. . . this was up and running back then . . .

Satellites get the jump on storm prediction

. . . but hopefully it's not too far off.

[DJ77]

Spend $200 Million for a new airplane, but not $200.1 Million for one with the latest and greatest Wx Radar. .

Is that aimed at Air France ? Do you know something that we don't ?

And, Multiscan radar is not the panacea that you seem to believe. I often had a much better picture scanning manually, playing with gain when needed. The reason appears to be that auto scan mode systematically uses ground clutter suppression (GCS). GCS is effective, but it also supresses a lot of real weather returns.

[NARVAL]

Narval, when Mr. Spinetta was appointed to head Air France over a decade ago, nobody there or at Air Inter, the domestic carrier, expected him to last very long. The job was a political plum that came on plush one- or two-year ejection seat and he turned it into a real job. From your pseudonym, I infer you already know this but it might be news to at least some readers here.

Quite right ArthurBorges. It happens that I know Mr Spinetta personally, for reasons outside the aviation subjects. He is at the very least,honest. And he attended too many funerals not to be truly shaken by this accident. Mr Gourgeon has been for some time with Air France, but has a difficult position as he was a former chief of the DGAC and among other decisions was the certification authority for the A330. He often says to journalists "I am not an expert in technical matters..."That deserves to be taken as what the British call an understatement.

[ArthurBorges]

Dear Narval

I think we know the same set of interlocking networks. We crossed pedestrian paths at GARD, which he presided before rising to his current level of eminence although even then my face was worth only his nod, a smile and a handshake.

Still, I am sure he was shaken.

I am equally sure that Mr. Gourgeon was equally shaken.

Is there ego in it?

Yeah.

But it's still sincere.

As pedestrians, we do what we can with what and who we are.

(To moddies: this is OT and I apologize -- please make an exception just this once)

[Dutch Bru]

wx training + missing SIGMETs in AF prelim report

It is at least interesting that despite the BEA report of August 2008 (a study on turbulence/weather related upsets, including with AF long distance a/c) there was apparently so far no hightened awaress on the use of weather radar. Link to the report in post # 800. This study pinpoints amongst others to weaknesses in the operation of wx radar by f/c's.

It is BTW interesting that the report mentions one (AF) accident with an A340, where the captain was not on the f/d when a serious upset happened (overspeed, A/P disengage) and where the PF was unable to open the door for the captain to re-enter until after the upset, since the PF needed to regain control over the flight.

The report also states that it would make sense that ACARS is used more for companies and crew in flight to communicate weather updates. In the case of AF447 the f/c nor AF did make use of this possibility.

Fact is that there would have been quite some reason for that. The TEMSI file on board showed ISOL and EMBD thunderstorms en route and so did the SIGMET 7 for FIR ATLANTICO, valid from 1800-2200, as mentioned in the prelim report indicating EMBD TS.

What is interesting of the prelim report is that it seems to mention SIGMET 10 for FIR ATLANTICO as the next SIGMET. While this is true in terms of period of coverage (2200-0200), the prelim report does not mention SIGMET 8 and 9.

SBAO SIGMET 8 VALID 311800/312200 SBRE- ATLANTIC FIR EMBD TS FCST SW OF CIDER PSN/ S35W034 AREA TOP FL400 STNR NC=

SBAO SIGMET 9 VALID 312200/010200 SBRE- ATLANTIC FIR EMBD TS FCST SW OF REGIS PSN/ERETU PSN/ N02W046 AREA TOP FL370 STNR NC=

Combined in a series 7 to 10, the SIGMETs show highly dynamic, intense and widespread EMBD TS activity in the ATLANTICO FIR on the foreseen route. Had the AF447 f/c availed about the full series through ACARS, they could probably have been on higher alert as to penetrate the zone concerned and perhaps used the wx radar more optimally.

[Rananim]

loss of both engines does not cause an immediate loss of pressurization, nor does it even result in a high cabin rate of climb. By the time any significant cabin rate of climb obtained, the capability of sending ACARS messages would have long been lost).

PJ2(page 143)

What if rollback or stall preceded flameout?GCB still closed but bleed flow to packs disrupted to trigger the last ACARS.

As to the "what then"...well,I dont think the lack of any distress call is too big an issue considering where they were.The lack of LJ's?Failure to re-light?Bodies were found with no clothing.If bodies had been found clothed but with no LJ,yes.Re-light requires flt envelope check which requires reliable instrumentation.Dark night,bad weather etc.Relights folllowing crystalline icing are 100% from what Ive read so yes its still a tenuous theory.

[PJ2]

Dutch Bru;

Combined in a series 7 to 10, the SIGMETs show highly dynamic, intense and widespread EMBD TS activity in the ATLANTICO FIR on the foreseen route. Had the AF447 f/c availed about the full series through ACARS, they could probably have been on higher alert as to penetrate the zone concerned and perhaps used the wx radar more optimally.

Flight crews avail themselves of ACARS met reports all the time. In my experience, company dispatch provided timely and up-to-date information on SIGMETS as a matter of flight following, a legal and operational requirement in many states. Most crews would plot the SIGMET'd area on their planning charts then keep a wary eye using all available techniques including radar. That is not where the problem lies. The problem is in teaching how to use the radar "optimally". Likely, most pilots pick up the use of their radar through "osmosis", which does not filter misunderstandings and shortcomings in knowing one's equipment.

Rananim;

What if rollback or stall preceded flameout?GCB still closed but bleed flow to packs disrupted to trigger the last ACARS.

Sorry, do you mean "stall" of the engine or of the aircraft? I'll assume engine for now, . A stall of the aircraft at that point, (powered), would generate ACARS messages, and we see none.

Logically, this scenario, (engine rollback/stall, then cabin pressure message) can't work because there would (likely, but not necessarily) have been messages indicating engine "failure". It depends - the word "rollback" means the inability to accelerate the engine but it's still turning, burning and hydraulic-ing - so the possibility, (but not the probability) is there.

As to the "what then"...well,I dont think the lack of any distress call is too big an issue considering where they were.

It is not an issue at all. Those who are either confused by the absence of such a call or are mildly chiding/merely "observing" the absence of a call don't understand the swiftness of what happened in the cockpit and what ensued. The crew would have been struggling to maintain situational awareness and discipline in a very serious, rapidly degrading circumstances. The judging of this absence is therefore wholly inappropriate; even the notion that an "automatic" distress call could be sent is naive.

The lack of LJ's?Failure to re-light?Bodies were found with no clothing.If bodies had been found clothed but with no LJ,yes.Re-light requires flt envelope check which requires reliable instrumentation.Dark night,bad weather etc.

Absence of any life-jackets in use is evidence for swiftly overwhelming events in which there was no time to prepare for the emergency.

Relights folllowing crystalline icing are 100% from what Ive read so yes its still a tenuous theory.

I fully agree. Just to be clear, it is not my theory, nor do I support the view that the engines flamed out due to ice crystals. The 330 has flown thousands of hours through "ice crystals" in high altitude layers without result. I've flown in "ice crystals" throughout the career without result in terms of engine roll-back or upset.

To recall the point, if the aircraft lost sufficient speed to enter a stall and the BEA has concluded that impact was of a flat aircraft attitude, high-velocity, vertical trajectory nature, then I submit that due to the very high angle of incidence in such a trajectory, the engines would have flamed out. I submit that the ACARS cabin advisory message was due to the effects of degrading upstream monitoring processes rather than a compromise of the cabin structure at that point or the loss of bleed-air input into the cabin to maintain pressurization. The BEA report does not comment on this message and so it is entirely conjecture.

[SaturnV]

AP RIO DE JANEIRO – An American colonel says two ships searching for the black boxes of Air France Flight 447 are ending their hunt.

A French nuclear submarine is continuing to look.

U.S. Air Force Col. Willie Berges is the Brazil-based commander of American military forces supporting the effort. He says one ship towing a U.S. Navy listening device stopped searching Friday.

He says the other ship towing a device will end its search within hours.

Berges says the ships had "no success."

Time for the side-scan sonar.

[xcitation]

Flight Plan

Anyone know what the major airline policies/guidance are for deviation due to weather?
The IB flight leaving 10 minutes later presumably had the same pre-flight weather data as AF447. They chose to load extra fuel (I recall 4000 gals) and a wide deviation around the large multi-CB weather system.
Curiously a LH flight and another AF flight the captains chose to try to pick a hole through the weather system.
Is it a question of the captain filing a flight plan and it is never reviewed or questioned? Or is there some feedback/friendly advice from the airline.
I am trying to understand the factors that influenced IB captain to opt for a very safe wide deviation pre-flight whereas other flights chose a path through it. Perhaps the flight plan process could be improved by cross checking that the best flight plan has been selected with a central office. Centralizing this function would facilitate adherence to standards set by the airline.
For passenger/crew comfort and reduced wear I would say large deviations are a no brainer given the relatively low cost.

[SaturnV]

DutchBru, Dispatch was not asleep.

At 0 h 31 dispatch sent the following message:
“BONJOUR AF447
METEO EN ROUTE SAILOR :
o PHOTO SAT DE 0000Z : CONVECTION ZCIT SALPU/TASIL
o PREVI CAT : NIL
SLTS DISPATCH”,

The BEA report notes that AF447 did not request copies of any SIGMETs through ACARS, ... "that functionality was not used by the crew."

Cloud tops at -60C about 0200 June 1



Tops at -70C



Tops at -75C



While I have seen the information on experience and hours on type, etc., I've not seen any information on how many times the three pilots have flown on this route. Has anyone?

[AstraMike]

Weather and Aerodynamics

Some questions:
1. Would the A330’s engines continue to run in the event of a Flat Spin, considering the intake angle to relative wind would be upwards of 60° and as much as 90°? Even though Relative Wind in severe convective activity with tops in excess of 50,000 feet, with potential for both vertical and horizontal air currents, may be a little hard to define… I would suggest that although the potential of ice may be a contributing factor, it is not required to lose the engines.
2. If the engines did fail, would the ACARS system still have the power necessary to transmit and receive and would data still be available for download to the ACARS system? I suspect not.
3. Given the size and condition of some of the recovered bits and pieces, with particular reference to the Galley, what logical solution is there apart from the Flat Spin that will bring the aircraft from where it was, to where it ended up, with just the right momentum to break its back, yet not destroy its contents beyond what we see? I suspect that in practical terms there is none.
4. Does tracking the transmitted 10 minute position reports and the last position transmitted take the aircraft through the convective activity displayed on the satellite graphics as submitted by Tim Vazquez, as it appears it does, and if so, what activity would you expect within such weather, especially as it seems to have been building at the time? Moderate or Extreme turbulence? Vertical air currents? Horizontal air currents? I would suggest all of the above
5. I understand, from reading this thread, that AF447 likely had limited rudder authority. Can it then be assumed that there was no rudder, or not enough rudder, to keep the wings level during the flight complexity of CB penetration, especially considering the effects of adverse yaw, should aileron be used at unintentionally high angles of attack? In other words is limited rudder authority, necessary in cruise, a detriment that might even contribute to an accident in this case?
6. Is it reasonable to suppose that given entry into the convective activity, apparently present in the area, as would appear likely to most, the crew would be faced with an impossible situation with an aircraft dancing around at the mercy of the storm (for want of a better description) small room for error with VS on one hand, VD on the other and apparent loss of rather important information (airspeed)? Not that it would make that much difference as it would likely be impossible to read the tape anyway, nor is it too likely that they could keep the aircraft level either in pitch or roll as nature played with it; you just cannot underestimate the power unleashed in these things and you simply cannot predict what will happen next.
7. Which pilot, hand on heart, would ever think of turning in a CB, let alone turning back to go through it again once it has caused difficulty? Nobody I know!!
8. Why is it so difficult to believe that loss of control due to weather assisted by pitot ice resulting in a spin is the most likely outcome?
9. You can be fairly certain that a Flat Spin does not degrade into a “full” Steep Spin, or a Spiral Dive, which might be considered a part of a recovery (getting the nose down) – if you are fortunate enough to do so, which is doubtful unless the weather that got you there also influences a recovery assist. With the wrong (or right, depending on your point of view) control inputs a Steep Spin will turn into a Flat Spin, even without the effect of turbulence which is quite likely to do the job for you.
10. I say again, do not underestimate the forces at work in a building 50,000+ foot convective build-up, I am sure most pilots respect and avoid them, so to suggest a crew, any crew, would turn back into the weather is frankly unbelievable – in fact even attempting a turn in the condition one can suspect they had on their hands would be frankly foolhardy and does not allow the crew the credit they deserve.
11. Although subject to considerable yaw force, it should not be assumed that a flat spin will cause the Vertical Stabilizer to part company, (it might actually be blanked out by the wing) or that such a departure is required to establish a flat spin which has a rotational rate that is less than a Full Spin and a rate of descent that is also about 50% less than a Full Spin all of which makes it the more logical manoeuvre for getting from A to B given the apparent arrival impact. I suspect it is far more likely that the Vertical Stabilizer failed as a result of the forces in play at impact, but would not insist on it.
12. Center of Gravity in a flat spin will really only dictate the relative position of the nose to the horizon – which need not be constant initially. It is easier to Flat Spin an aircraft with an aft (less stable) C of G and if cruise policy is to operate this way, it only adds to the Flat Spin suspicion. An aircraft with a more forward C of G will simply not look as flat.
13. I asked before if anyone could recognize a spin on instruments – no one seems to have that confidence. When I was in a Flat Spin I was VFR and could see the horizon going round, and round and round. I wonder how electronic instruments would depict this, would they become non-functional as the old steam driven instruments did. I don’t expect you could tell what was going on unless you had been there before, in the not too distant past
14. What was the sea state at impact? Anything but smooth, I suspect, yet the talk is as if the aircraft arrived out of the overcast in a glide, or at worst in a spiral and didn’t quite manage to pull out. This sort of thing typically ends up with what was an aircraft wrapped into a ball of wreckage by the forces of impact and sinking in tiny pieces. The dead stick landing in the river recently was impressive because of the unusually minimal damage caused. All the indications we have seem to refute this sort of scenario. But the sea and changing winds under the convective skies must surely have played a part that “normal” can't explain.
15. I have never flown trans-oceanic without guarding (listening to) 121.5. In fact I seldom fly anywhere without having it quietly selected. Other aircraft in the sky that night were doing just that – of course they were, they were diverting off course and announcing their intentions, as a courtesy, no doubt – yet none heard a word from AF447, no courtesy advice of a diversion, no mayday, nothing. That, and the fact that life vests do not seem to have been deployed, or should I say the cabin prepared, implies sudden onset of something totally unexpected that, simply took over.


In the end, it doesn’t really matter how the aircraft descended. I think one of the few things agreed is that it did, and no one can turn back the clock; it’s all history now. What matters is; what happened to cause the series of events that defeated the aircraft and its crew? There are not that many candidates and I would suggest pitot ice alone would not be one. From many of the things stated on this thread, there would appear to be some dark corners in airliner cockpits where light needs to shine and lessons be learned, other than the more obvious avoidance of weather.

[L337]

Anyone know what the major airline policies/guidance are for deviation due to weather?

From my FCOM:

Monitor the weather at both long and short ranges in order to plan efficient
course changes. When avoiding storm activity stay upwind of the storm, if
possible. If thunderstorms exist over the landing airfield or landing reports
indicate severe turbulence, consider delaying landing or diverting.
Plan ahead to avoid large storms:
Avoid magenta and red radar returns and fringes by at least 20 nm.
Avoid single magenta areas of turbulence (not associated with heavy
precipitation) by at least 5 nm.
Avoid penetrating a cell or clear it’s top by less than 5000 ft, otherwise
severe turbulence may be encountered.
Do not overfly cells if the top of the cell is at or above 25000 ft to avoid
the possibility of encountering turbulence stronger than expected.
Avoid flying under a thunderstorm due to possible windshear, microbursts,
severe turbulence or hail.
If severe turbulence cannot be avoided, an increased buffet margin is
recommended. This can be achieved by descending up to 4,000 ft below
the FMS optimum altitude.
Adjust the tilt frequently in order to monitor storm development and to get
the best cell echo.

Curiously a LH flight and another AF flight the captains chose to try to pick a hole through the weather system.

I assume as they must still be alive, they did not "try to" they did pick a hole.

Is it a question of the captain filing a flight plan and it is never reviewed or questioned?

It does not happen like that. A large computer talks to another large computer and extracts all the weather, the winds, notices, navigation charges, fuel Price, and a hundred other items, and it produces an optimum route for the flight. The company file the flight plan. The controlling item is almost always cost. Having received all the paperwork for the flight I and the crew then review it all. If I think we need more fuel for the route, that is my decision. So if I see some nasty weather on my route that might cost me extra fuel, and I want to take extra fuel, I do. If I want another route, I phone the office up and ask for one. As Captain the buck stops with me.

For passenger/crew comfort and reduced wear I would say large deviations are a no brainer given the relatively low cost.

The 747-400 burns roughly 10,000kg of fuel an hour. It all adds up.

Safe and Commercial is the plan. Safety comes first always, but it needs to be commercial as well. After all the safest thing is just to stay in bed...

[Mudman]

Engine failure & Flight Recorders

If the engines flamed out as some folks have theorized would no electrical power from the engines mean that the FDR and CVR would stop working? Is there an alternate power source for them? I think this happened on Swiss Air 111, Eygpt Air 990 and American 587, the recorders stopped before impact.

thanks.

(if this is irrelevant or redundant please delete)

[All-Ex]

SaturnV: While I have seen the information on experience and hours on type, etc., I've not seen any information on how many times the three pilots have flown on this route. Has anyone?

From the BEA interim report page 12ff:

Captain:
The captain had carried out sixteen rotations in the South America sector since he arrived in the A330/A340 division in 2007.

Co-Pilot 1:
This pilot had performed 39 rotations on the South America sector since arriving in the A330/A340 division in 2002.

Co-Pilot 2:
This pilot had performed five rotations in the South America sector since arriving in the A330/A340 division in 2008, including one to Rio de Janeiro.

[LeandroSecundo]

Deviation due to weather?

Hi,

Some answer about deviation (supposed) there

AF 447 : comme Lufthansa, l'avion a dévié à l'Ouest de sa route prévue

Eurocockpit - Accueil

Bye.

[Razoray]

Mudman.....

On Swiss Air 111 the FDR stopped working because the cockpit burned up...

[spagiola]

Some answer about deviation (supposed) there

AF447 : comme Lufthansa, l'avion a dévié à l'Ouest de sa route prévue

A quick summary of what eurocockpit is claiming, for non-French readers:

The last reported position of AF447, at 0210Z, is actually about 3nm west of track, so it is likely that AF447 was in fact deviating to avoid weather. Given GPS etc, this discrepancy is unlikely to be due to poor navigation; nor is it an offset, as that would have placed AF447 1 nm EAST of track.

Thus they place in doubt BEA's finding that AF447 was on track, and AF CEO Gourgeon's suggestions that perhaps the crew of AF447 didn't use their wx radar appropriately.

(Before someone asks, I grew up in France and the French-speaking part of Switzerland, and I'm a pilot (albeit only PPL) and so understand aeronautical French as well as normal French)

[wilyflier]

flat attitude

Flat attitude
A lot of time with semantics was spent on the exact attitude
"en ligne d`vol ",flat horizontal,with very little forward speed...
With regard to the VS possibly being ripped off to tumble forward.
How stormy was it at sea level? How big the swells?
I envisage a tail down approach (not into wind, cross swell), at high attitude and hitting the back face of a swell smack on, sufficiently to break the belly in.
(Think 737 on the Motorway embankment at Derby)
With for good measure the tail cone clipping a previous swell and breaking UP
Thus striking and crushing the rudder base and catapulting it forward free of its compromised hull structure
This explains the ability of the hull to hit the water surface forcefully wth mainly 90 degree impact while still having a much more credible forward airspeed.
Of course it would stop in the water very quickly with the following swell "pooping" and washing out the now opened hull, and smashing into the back of the vulnerable spoiler(s)