AF447 Thread No. 3 [Archive]

jcjeant

27th May 2011, 12:48

Hi,

A few posters have commented on the THS movement, and how one idea was the pilots not moving the THS following recovery of the stall.

A 2 h 10 min 51 , l’alarme de décrochage se déclenche à nouveau. Les manettes de commande de poussée sont placées sur le cran TO/GA et le PF maintient son ordre à cabrer. L’incidence enregistrée, de l’ordre de 6 degrés au déclenchement de l’alarme de décrochage, continue à augmenter. Le plan horizontal réglable (PHR) passe de 3 à 13 degrés à cabrer en 1 minute environ ; il restera dans cette dernière position jusqu'à la fin du vol.

At 2H10.51 this is again a stall alarm.the power levers are put on the position TO/GA and the PF keep the command to climb.
Incidence angle of 6° when sounded the stall alarm continue to rise.
The adjustable horizontal stabilizer go from 3° to 13° (climb position) in about 1 minute.
He will stay in this position to the end of the flight.A 2 h 12 min 02, le PF dit « je n’ai plus aucune indication », et le PNF « on n’a aucune indication qui soit valable ». A cet instant, les manettes de commande de poussée se trouvent sur le cran IDLE, les N1 des moteurs sont à 55 %. Une quinzaine de secondes plus tard, le PF fait des actions à piquer

At 2H12.02 the PF tell "I have no more indication" and the PNF "we don't have any accurate indication"
At this moment the power levers are on position IDLE and N1 engines indicate 35%
15 seconds after the PF perform a command to diveHow it's possible that the horizontal stabilizer stay in climb position when the PF perform a dive command ?

infrequentflyer789

27th May 2011, 12:57

From my knowledge this is an auto-trim aircraft. Aside from movement of the trim wheel is there any other action that could have been done to manually reset the THS. Unlike a 737 for instance where there is a separate trim switch.

It's auto-trim until it's not - that function/law can drop out automatically (when HAL/George gives up) and it did at Perpignan.

737 has auto-trim too - again, it can drop out automatically, after putting the trim full up - see Schipol (and at least one other incident).

geoff sutherland

27th May 2011, 12:57

Maybe the reason the PF persists with a nose-up command and then resets throttles to idle is that his instruments tell him he is diving and overspeeding????

Lonewolf_50

27th May 2011, 12:59

At 2H12.02 the PF tell "I have no more indication" and the PNF "we don't have any accurate indication" At this moment the power levers are on position IDLE and N1 engines indicate 55% {per the French original} 15 seconds after the PF perform a command to dive

Don't have any accurate indication of what parameter, what gage on the instrument panel? :confused:

takata

27th May 2011, 13:02

I'm posting here as the old thread disapeared.

Hi Svarin,

BEA wrote :

The airplane’s angle of attack increased progressively beyond 10 degrees and the plane started to climb. The PF made nose-down control inputs and alternately left and right roll inputs.Feels strange to me...

The trimmable horizontal stabilizer (THS) passed from 3 to 13 degrees nose-up in about 1 minute and remained in the latter position until the end of the flight.Suspiciously similar to Perpignan.
Well... suspiciously not surprising from you by the way you are quoting the report by removing the parts between both "nose down" orders and what happened when this THS trimming changed from +3 to +13 deg. Read it between 0210:51 and 0212:17.

For this part, the report says:
A 2 h 10 min 51 (5), l’alarme de décrochage se déclenche à nouveau. Les manettes de commande de poussée sont placées sur le cran TO/GA et le PF maintient son ordre à cabrer. L’incidence enregistrée, de l’ordre de 6 degrés au déclenchement de l’alarme de décrochage, continue à augmenter. Le plan horizontal réglable (PHR) passe de 3 à 13 degrés à cabrer en 1 minute environ ; il restera dans cette dernière position jusqu'à la fin du vol.L’altitude est alors d’environ 35 000 ft, l'incidence dépasse 40 degrés et la vitesse verticale est d’environ - 10 000 ft/min. L’assiette de l’avion ne dépasse pas 15 degrés et les N1 des moteurs sont proches de 100 %. L’avion subit des oscillations en roulis atteignant parfois 40 degrés. Le PF exerce une action sur le manche en butée à gauche et à cabrer, qui dure environ 30 secondes.0) Stall Warning
1) TOGA applied (pitch up reaction)
2) PF maintaining pitch up during 30 seconds
3) THS moving from +3 to +13 deg.
4) Pitch increase to about +30 deg (from +10 to +40).

A 2 h 12 min 02, le PF dit « je n’ai plus aucune indication », et le PNF « on n’a aucune indication qui soit valable ». A cet instant, les manettes de commande de poussée se trouvent sur le cran IDLE, les N1 des moteurs sont à 55 %. Une quinzaine de secondes plus tard, le PF fait des actions à piquer. Dans les instants qui suivent, on constate une diminution d’incidence,les vitesses redeviennent valides et l’alarme de décrochage se réactive.It looks like the PF did not order a nose down command until 0212:17 -- hence, 1 mn 26 sec without any nose down orders... no indication that he tried to override manually the THS either). The only corrective action against the positivie attitude was to select IDLE thrust. Those late nose down action started to reduce the pitch despite this +13 deg THS. (Now, why it stayed frozen here is nonetheless a disturbing issue when those nose down orders were finally applied, but it is the reason why I'm waiting for more info about any use of the manual trim wheel during the previous upset).

But, a VERY selective reading could always make one quite happy with his very "suspicious" findings!

jcjeant

27th May 2011, 13:06

Hi,

In the report chapter "new findings" I read:
The inputs made by the PF were mainly nose-up

I may ask to the pilots in this forum
Is this a normal behavior for recovery from a stall event ?

takata

27th May 2011, 13:14

Hi jcjeant,

In the report chapter "new findings" I read:
The inputs made by the PF were mainly nose-up

I may ask to the pilots in this forum
Is this a normal behavior for recovery from a stall event ?
No, unless one really want to stay between alpha prot and alpha max with an High AoA protection kicking in Normal Law.
That may be the confusion if one is only trained to low alt, low speed, Normal Law stall recovery, even if both Pilots aknowledged they were in Alternate Law.
Something that need a serious investigation here.

PENKO

27th May 2011, 13:19

Maybe the reason the PF persists with a nose-up command and then resets throttles to idle is that his instruments tell him he is diving and overspeeding????

Then why give TOGA power? The report clearly says that the pilot reacted to the stall with TOGA power and a pitch up command. That's the puzzling aspect.

Of course it crossed my mind that for some reason he thought he was diving or gaining too much speed. But then he would not add power.
It just makes no sense.

And it is not a momentary confusion either, it is a persistent pitch up command. There were two other pilots there who did not react to this inconsistent action by the PF which makes it all the more puzzling. I maintain that we are missing half the information.

jcjeant

27th May 2011, 13:22

Hi,

In one of the graph (point number 6) the aircraft made a turn to the right (and not a small turn!)
How that can possible as in the rapport the first PF action is a command to climb with a turn command to the left ...

blind pew

27th May 2011, 13:24

What I believe the report doesn't say is what the PF instruments were saying.

In the BEA BAC 111 incident in the 70s the crew had both overspeed and stall warnings at the same time - caused by ice in the pitot and static lines.

There was somewhere over Asia a pitot ice problem just after the AF with prang similar instrument indications to the 111.

I read the post from the crew on PPrune.

The skipper stated that if it had been at night they might not have got away with it.

Speaking from my own experience only - If the choice was overspeed or stall then I would lean towards stalling as with a gross overspeed one has no chance to recover.

On early jets - 707 etc - with an overspeed you opened the airbrakes immediately - did it once in wave over the alps after returning from the call of nature and not realizing the ball park had changed.

Assuming there was not a flight control system fault then it seems as though PF didn't understand what was going on.

Which leads to a post I made last week - why don't they fit AOA instruments?

Lost my best friend in a stall which was not identified nearly 40 years ago and eff all appears to have changed.

PENKO

27th May 2011, 13:27

Well, there is an AOA instrument on board: the stall warning

jcjeant

27th May 2011, 13:29

Hi,

both Pilots aknowledged they were in Alternate Law

In alternate law .. the trim is in "auto-trim" ?

AlphaZuluRomeo

27th May 2011, 13:34

@ silverline001 : Les Airbus A330 d’Air France sont équipés du radar Collins WXR700X. L’image radar est présentée sur les ND, superposée aux autres informations.
Source : BEA's interim #1 report, dated july 2009.

@ jcjeant : A roll command to the left is not a turn command to the left. Particularly when this roll command seems aimed to counter an aircraft roll to the right...
Are you already trying to find a reason to disregard the last BEA sayings ? :rolleyes:

spagiola

27th May 2011, 13:34

And it is not a momentary confusion either, it is a persistent pitch up command. There were two other pilots there who did not react to this inconsistent action by the PF which makes it all the more puzzling

My emphasis added.

We do not know that the others didn't react, as the report does not tell us either way. The report does note, however, that the PF and PNF both did something a 2:13:28 (1 minute before impact), at which point the PF handed over control to the other. So there does seem to have been some reaction.

Edited to add emphasis.

Lonewolf_50

27th May 2011, 13:36

infrequent flyer: this is a wirehead question in two parts, but I'll ask anyway.

1. What parameters inform (other than sidestick command, or in mechanical back up mode the trim wheels) how the THS moves up or down?

Are the inputs into Pri or Sec that lead to THS commands (which include autotrim commands)
-airspeed?
-AoA?
-a blend/sum of both?

(Are other intertial systems inputs summed in to create the final signal? the (dated) block diagram doesn't get into the wire and signal inputs and loops).

2. As I understand the (dated) block diagram, mechanical mode (trim wheels) does not command the elevators, it commands only the THS.
(If I have misunderstood this, please advise).

Interesting comment on the old training aid I am referring to:

Regardless of pilot inputs, the computers will prevent :
- excessive maneuvres
- exceedance of the safe flight envelope

exosphere

27th May 2011, 13:39

Another puzzling combination, in addition to the already mentioned ones,is that with the application of TOGA they gave a pitch Up input, and when they applied IDLE power they gave a pitch Down input...I don't get it, it's the inverse of what you might expect: TOGA application makes sense in a stalled condition together with a pitch down input, and IDLE application makes sense in a overspeeding condition where pitch Up might be required. Something is missing or I'm missing something.

infrequentflyer789

27th May 2011, 13:40

Hi,
How it's possible that the horizontal stabilizer stay in climb position when the PF perform a dive command ?

Don't know yet - very likely will be part of the detailed technical investigation.

Some possibilities (from my basic understanding of the engineering)

a) because auto-trim has been lost, it's not there in all laws (not clear this is what happened in this case)

or

b) because autotrim went wrong - possibly due to invalid speed input to gain

or

c) because maybe the pitch-down input was brief and/or could be satisfied without moving THS

Remember the pilots went hard pitch up (to the "stops"), then briefly pitch-down, quite possibly the attitude never reached level let alone "dive". BEA states that "angle of attack decreased" following the pitch-down - therefore the it appears the a/c was still responding to pitch controls (and in the right direction, but at what rate?). Shortly after, the pilots comment about reaching FL100, and there are further control inputs - not specified, but possibly pitch up again...

Another consideration:

BEA: "The airplane was subject to roll oscillations that sometimes reached 40 degrees."

From http://www.airbusdriver.net/airbus_fltlaws.htm:

No requirement to change pitch trim for changes in airspeed, configuration, or bank up to 33 degrees.

Not clear what those two taken together mean - maybe autotrim stopped temporaily due to the roll angles.

Note: there is still insufficient information to say the aircraft and flight control systems are in the clear, but equally not enough to show they are to blame. We do know, however, that the aircraft didn't pitch-up and stall on its own - pilot inputs were made. I'll leave those to the heavy pilots to discuss.

glenbrook

27th May 2011, 13:42

I think there is vital information not presented in the report. In particular the position of the thrust levers is only given at a couple of points. We know PF dialled in a lower speed because of turbulence and shortly afterwards AT disengaged. How long after this point did the PF touch the throttle levers to increase thrust? I don't think we are told. Later, after the first stall warning, the engine levers at at TOGA. It would be very useful to see a graph of engine N1 over the whole period.

takata

27th May 2011, 13:44

In one of the graph (point number 6) the aircraft made a turn to the right (and not a small turn!)
How that can possible as in the rapport the first PF action is a command to climb with a turn command to the left ...
An explanation was that the right wing stalled, due to lack of speed, and then the PF tried to counter it by rolling to the left, but it is useless or will make the things worse.
In alternate law .. the trim is in "auto-trim" ?
Until you are using the manual wheel which is overriding it.
Maybe the reason the PF persists with a nose-up command and then resets throttles to idle is that his instruments tell him he is diving and overspeeding?
Recorded (and displayed) speed dropped (LHS and stand-by) due to ice, down to 60 kt (false) during one minute and come back to 215 (real), then 185 (real). They heard those stall warnings... no overspeed.

Lazerdog

27th May 2011, 13:54

I don't think many have experienced a stall in a turbulent cell before. Recognizing that the aircraft had indeed stalled might have been very difficult at night (as Blind Pew referred to earlier) and with all the alarms going off. (And agreed, the stall warning should have been a primary indication but that was preceeded by other alarms.) Once established in the stalled state, the aircraft still had roll control and may have been fairly stable in that state with no further stall breaks, causing confusion as to the correct action to perform. Nowhere in that report do I see any overspeed reference so why forward stick was not applied is very odd unless the effects of vertigo were so convincing. THS up trim to 13 degrees up (whether done by the system or manually) definitely needs to be analyzed and understood.

infrequentflyer789

27th May 2011, 13:54

infrequent flyer: this is a wirehead question in two parts, but I'll ask anyway.

1. What parameters inform (other than sidestick command, or in mechanical back up mode the trim wheels) how the THS moves up or down?

Afraid you are crediting me with more detailed knowledge than I actually have.

Short answer - I don't know, but it is one of the questions on my mind too...
I strongly suspect that airspeed is in there somewhere, but what value is used when airspeed is known-invalid, I have no idea.

Regardless of pilot inputs, the computers will prevent :
- excessive maneuvres
- exceedance of the safe flight envelope

And the concierge can fly it too... :suspect:

All missing the big caveat "as long as everything works normally" (and actually, even then the concierge quote is rubbish). Drop out of normal law and all that no longer applies.

spagiola

27th May 2011, 14:00

Maybe the reason the PF persists with a nose-up command and then resets throttles to idle is that his instruments tell him he is diving and overspeeding????

There's a stall warning at 2:10:51, then the report notes that the stall warning horn stopped at 2:11:40, the clear implication being that the horn was sounding throughout the intervening period. So the instruments could not have been telling the PF he is overspeeding, as the instruments clearly thought he was stalling -- and saying so.

So it seems more likely he did not believe the instruments. Or is it possibile he misinterpreted the protections the current law was providing? If he thought he still had stall protection, would commanding pitch up be a way of trying to tell the aircraft: I don't want to lose altitude, and I'm relying on you not to let me stall?

Lonewolf_50

27th May 2011, 14:03

L@zerdog, what you described looks like a hybrid between a spin and a spiral.
Once established in the stalled state, the aircraft still had roll control and may have been fairly stable in that state with no further stall breaks, causing confusion as to the correct action to perform.
Spin in the goo, in a CB? :eek: Ouch. (EDIT: Capn Bloogs, thanks, not in a cb, but in the goo nonetheless)

takata

27th May 2011, 14:03

What parameters inform (other than sidestick command, or in mechanical back up mode the trim wheels) how the THS moves up or down?Load factor. This is only a load factor law supposed to maintain 1g flight by small adjustement of the horizontal stab trim (and it is only commanded by the PRIMs).

infrequentflyer789

27th May 2011, 14:04

The rubbish has started in the "press" - front of dailymail website (my bold):

Passengers and crew on board doomed Air France jet endured terrifying THREE MINUTE plunge into the ocean after engines failedThe pilot, Marc Dubois, had been taking a break when the plane hit heavy turbulence and never returned to his seat before the crash.So, no one (as there's only one pilot, singular) at the controls, heavy turbulence, engine failure. :ugh:

iskyfly

27th May 2011, 14:06

Can somebody correlate the ACARS messages with the recently released CVR/FDR information? It would be interesting to see some of the mysteriousness cleared up as to what triggered the ACARS messages and their meaning in light of this new information.

Thanks

glenbrook

27th May 2011, 14:06

hi takata

An explanation was that the right wing stalled, due to lack of speed, and then the PF tried to counter it by rolling to the left, but it is useless or will make the things worse.

Until you are using the manual wheel which is overriding it.

Recorded (and displayed) speed dropped (LHS and stand-by) due to ice, down to 60 kt (false) during one minute and come back to 215 (real), then 185 (real). They heard those stall warnings... no overspeed.

We don't know that the roll to the right happened because of the stall or just before it. The report mentions the roll, the pitch up and then the stall warning in that order. Perhaps the PF merely thought the a/c was rolling and and pitching down, having forgotten that the engine thrust was set at a lower level when AT disconnected.

Capn Bloggs

27th May 2011, 14:08

how they ended up in a CB IN THE FIRST PLACE,would most likely be ,the blame of the weather radar.To me 10 000 fpm descent, might have been aided with the CB downdraft.
There is no evidence to even slightly suggest that they were in a CB. In fact, based on the statements in the report (the call to the cabin, the 12° left turn, the decel to M0.8) they knew very well what was going on with the weather and were steering around the worst bit. IMO they were obviously not in a CB, but in cloud in bad icing conditions.

Lonewolf_50

27th May 2011, 14:09

Vertical stab isn't the THS, ;) but thank you for the load factor. :)

surplus1

27th May 2011, 14:10

I cannot access the BEA report - will wait until later and try again.

Meanwhile I have a question for anyone with actual technical knowledge of the A330:

If stablizer trim is full nose up (why is not important at this point) and TOGA is applied - does the elevator have the ability to overcome the consequences if full nose down input is made and held (without changing stabilizer trim)?

BOAC

27th May 2011, 14:12

Well, I'm no clearer now that I was! There are too many inexplicable events to make sense of this, including pilot input and THS angle. It is unfortunate we do not have the RHS IAS readings, and I'm not clear from that report which seat PF was in, nor which IAS readings he was reacting to. A 'zoom' from FL 350 to 375 with 7000fpm is pretty dramatic. It also appears that they were close to being unstalled at the 375 point.

The only observation I can make is that I have seen a full and maintained nose-up demand from a pilot in a different aircraft before, in reaction to a large (and unexpected) rate of descent while stalled.

Way to go yet.

Svarin

27th May 2011, 14:15

takata wrote :

That may be the confusion if one is only trained to low alt, low speed, Normal Law stall recovery, even if both Pilots aknowledged they were in Alternate Law. Something that need a serious investigation here. Agreed.

PENKO wrote :

And it is not a momentary confusion either, it is a persistent pitch up command. There were two other pilots there who did not react to this inconsistent action by the PF which makes it all the more puzzling. I maintain that we are missing half the information.Agreed, too.

One hypothesis is along the lines of "stall training" such as airline pilots usually get during recurrent simulator training. Since the aircraft is reputedly "protected" against stall, training for upset recovery, stall identification and recovery has been, at best, extremely poor. The only manoeuver which is considered is "approach to stall, minimum altitude loss". And as takata pointed out, in Normal Law, perhaps modified through landing configuration (such as was attempted in Perpignan).

In this specific sequence of "approach to stall, minimum altitude loss", the reaction is to power out of pre-stall condition using TOGA thrust, maintaining some kind of "average" pitch attitude. This is of course completely self-defeating in a high-altitude post-stalled condition. But this condition must be identified first...

Both major manufacturers produced together a video document about upset recovery a few years back. But that was likely the only "training" the accident flight pilots ever got about it. And of course, I am with those who will claim that a pilot should know how to get out of a stall (like someone mentioned from a beginner's PPL experience, very valid point).

Unfortunately, what we have here is the glaring example of why pilots should remain pilots for real, and not become system operators, or cockpit managers, or what have you. The design of any aircraft, especially one as revolutionary in its time as that which appeared in the late eighties, will have an influence on what the role, function, skillset, attitude of those on the flight deck will become. I maintain that these changes were intentional, but of course not from individual engineers or programmers, this is ludicrous. It is a political decision.

One does not live through the career of an airline pilot in isolation, working their skill at home with personal dedication. Any airline pilot is essentially the product of a whole system, regulator, airline and manufacturer included. The only personal thing that can be added to this is personal light aircraft training (as can or cannot be afforded depending on work and wages conditions...), homework with the FCOM and other documents, personal discipline regarding sleep, food, physical activity, and personal ethics.

But even if I was ready to pay for it in person (I cannot), I doubt I could get even a 12hrs simulator training syllabus to get me vaguely up to par about all the upset and stall conditions I would have perhaps to face one day in a large heavy jet airliner. But in all likelyhood, I will never face these conditions. How expensive would that statistically useless training make me ?

Finally, as the pilot's responsibility with the life or death of his charges justifies inquiring in detail into all his actions even post-mortem (and this occasionally includes very personal details), similarly, the authority claimed by the aircraft (design policy-makers, technical designers, head of project and so on) over life and death "protection" justifies that it be subjected to a similarly detailed scrutiny. There is enough evidence already to justify such scrutiny.

Gary Brown

27th May 2011, 14:15

Question.

Most of the new BEA info regarding instrument read-outs explicitly relates to the left-hand seat displays (with the right hand-seat displays not available in the FDR). But which seat was the PF - at the time of the initial upset - sitting in?

There was speculation in one of these various threads that he would almost certainly have been in the right-hand seat. That he handed over control in the final moments (presumably to the Commander, who had returned to the cockpit) seems to support that.

Could the initial PF have been seeing different values on his displays than on those recorded in the FDR?

[Obviously, everyone could hear the stall alarms wherever sitting / standing.]

paull

27th May 2011, 14:26

Can the captain in the jump seat, even SEE that PF was giving pitch up input, not pitch down? After all, they are impulses, not stick positions?

MurphyWasRight

27th May 2011, 14:27

spagiola comments:

Quote:

And it is not a momentary confusion either, it is a persistent pitch up command. There were two other pilots there who did not react to this inconsistent action by the PF which makes it all the more puzzling
My emphasis added
.

We do not know that the others didn't react, as the report does not tell us either way.

Agree that the report is very light on some details but that does bring up a question in my (obviously not heavy pilot) mind:

Given the lack of feedback/coupling between the 2 sets of controls what other immediate clues are available to the PNF on the action the PF is taking?

Note: This is a sincere question not an attempt to stir the A/B pot.

takata

27th May 2011, 14:28

Hi glenbrook,
We don't know that the roll to the right happened because of the stall or just before it. The report mentions the roll, the pitch up and then the stall warning in that order. Perhaps the PF merely thought the a/c was rolling and and pitching down, having forgotten that the engine thrust was set at a lower level when AT disconnected.
First, I fully agree that we don't know (nothing is said about it) and I may be totally wrong about that!
I just tried to answer the sharp turn to the right as seen on the BEA map:
- Red dots as legended:
(1) 0135:15 (last contact with ATC)
(2) 0159:32 (meteo briefing with Captain, light/moderate turbulences expected
....................Captain leaves flight deck)
(3) 0208:07 (12 deg offset manoeuver start, light turbulences, Mach 0.80)
(4) 0210:05 (icing: autopilot off, autothrust off, ALT law, stall alarm -twice-, right wing bank;
.............PF: pitch up, left roll ordered, speed decr. 275-> 60 kt)
(5) 0210:51 (stall alarm again. PF: TOGA, max pitch up and left during 30 sec., THS +10, ...
(6) 0211:40 (Captain is back in the cockpit...

http://takata1940.free.fr/LastFlight.jpg

Interested_Layman

27th May 2011, 14:28

Question from ignorant SLF, the BEA report quotes different values for AoA and pitch at the same point in time.
Are these not the same datum?
Or is pitch calculated from the laser ring gyro and AoA via an aerodyamic sensor?

Lonewolf_50

27th May 2011, 14:31

What we won't know is where the PF's eyes were. (Interesting question on if PF and PNF were seeing the same things on their instruments) Where his eyes were would inform what his hands were doing while he tried to regain control of the aircraft.

LYKA

27th May 2011, 14:31

There is no evidence to even slightly suggest that they were in a CB. In fact, based on the statements in the report (the call to the cabin, the 12° left turn, the decel to M0.8) they knew very well what was going on with the weather and were steering around the worst bit. IMO they were obviously not in a CB, but in cloud in bad icing conditions

But there was convective WX around and we don't know what the tilt setting was - they could have been dangerously overscanning the storms, leading to a false impression that the flight path choosen (10 left) was clear:

http://http://www.weathergraphics.com/tim/af447/

Svarin

27th May 2011, 14:41

jcjeant posted :
In alternate law .. the trim is in "auto-trim" ?
takata answered :
Until you are using the manual wheel which is overriding it.takata, have you ever flown an Airbus ? Have you ever trained for type rating on either 320, 330, 340 or 380 ? I will tell you a ghastly secret : using manual pitch trim is one of the very first things one has to unlearn in order to be accepted as a candidate for type rating. After which, during all operational flight hours, and during every simulator training and check, it is verboten to touch that wheel.

CPT : A330/340 experience -> 1700hrs (plus all hours on A320 not mentioned)
F/O 1 : A320/330/340 experience -> 6500hrs
F/O 2 : A320/330/340 experience -> 2900hrs

It would seem obvious that touching that verboten wheel would have perhaps been useful. But the pilots here had never touched it, even in benign conditions, for a total of more than 10000 hours...:{

JPI33600

27th May 2011, 14:44

Having read the BEA document, to the SLF I am, and with all due respect to people who lost their lives, this sad accident seems to result from :

1. loss of airspeed indication resulting from ice-clogged pitots

2. inappropriate reaction of crew that did not immediately apply the recommended unreliable airspeed procedure which prescribes 5° pitch and CLB thrust setting (see HyperVeloce post (http://www.pprune.org/tech-log/395105-af-447-search-resume-80.html#post5772481) in part #1 of this thread) until the situation has been sorted out

Yet another combination of technical failure and human error. Apart from R&D work about how to prevent UAS condition, this seems to call for a better training of pilots in unusual situations.

Professional opinions welcome of course.

takata

27th May 2011, 14:45

Question from ignorant SLF, the BEA report quotes different values for AoA and pitch at the same point in time.
Are these not the same datum?
Pitch is the aircraft attitude vs. the horizon, while Angle-of-attack is wing airfoil vs. airflow direction (an air vector is needed but may be calculated by Inertial Reference). When you have a pitch of 0 deg and an AoA of 90+ deg, you are in "ligne de vol", but you are falling down vertically. BEA says that [edited: pitch was 16.2 deg at impact and AoA was not valid].

takata

27th May 2011, 14:50

I will tell you a ghastly secret : using manual pitch trim is one of the very first things one has to unlearn in order to be accepted as a candidate for type rating. After which, during all operational flight hours, and during every simulator training and check, it is verboten to touch that wheel.
You are only a troll, Svarin. If Airbus didn't want someone to ever use the trim wheel, the very simple way was not to put one in the cockpit at the first place. They would even have made the electrical imputs to have a precedence over the pilot mechanical control which was, of course, made the other way.
Get a life.

PENKO

27th May 2011, 14:50

takata, the pitch angle was 16.2 degrees on impact, there was no valid AOA data at that point.

takata

27th May 2011, 14:54

takata, the pitch angle was 16.2 degrees on impact, there was no valid AOA data at that point.
Right, I stand corrected. wrongly remembered "incidence" instead of "assiette".

Lonewolf_50

27th May 2011, 14:55

For interested layman:

Angle of Attack Indicator (http://www.navyair.com/Angle%20of%20Attack%20Indicator.htm)

If I hold the pitch of my airfoil constant in a fixed wing aircraft (I'll do that by controlling the nose pitch attitude, either via visual reference to the horizon, or on the attitude indicator based on a gyro of some sort)

and then I vary my airspeed,

my AoA will change: increase as I go slower, decrease as I go faster.

Depending upon model of aircraft, when the fuselage is roughly level compared to the ground, you may have a few degrees of pitch up, relative to level to the ground on the airfoil. (There is also on some aircraft "twist" from wing root to wing tip, which would induce the wing root to stall before the wing tip at high angles of attack ... )

eireoflot82

27th May 2011, 14:55

Lonewolf_50

27th May 2011, 15:05

It can. 3-D disorientation is a common phenomnon in instrument flying.

Part of the purpose of instrument training is to get you to focus on your instruments, regardless of spatial disorientation, and to develop scan patterns that allow your brain to sort out what the aircraft is doing. It's a skill that is useful to practice in case you need it. If you have some instruments that are of doubtful accuracy, most scan training teaches you to cross check other instruments to see what the aircraft is doing. (I called this "partial panel scan" in another thread but that may not be the current term of art).

Now, if you are stalled, the flight controls don't respond as they normally do, getting unstalled on instrument scan (the first time you do it) is a bit tougher than simply flying maneuvers on instruments. If you are stalled and rotating, the degree of difficulty goes up an order of magnitude or so. (First time I ever did a training spin "under the bag" in instrument training was an eye opener. It took a few tries to get the hang of what I needed to do, and to actually do it. Some years ago in a single engine trainer).

takata

27th May 2011, 15:07

Vertical stab isn't the THS, http://images.ibsrv.net/ibsrv/res/src:www.pprune.org/get/images/smilies/wink2.gif but thank you for the load factor. http://images.ibsrv.net/ibsrv/res/src:www.pprune.org/get/images/smilies/smile.gif
Good catch!... I didn't see it while typing/posting that.
:ugh:

Checkboard

27th May 2011, 15:07

Which leads to a post I made last week - why don't they fit AOA instruments?

The difference between "the bird" (the flight vector symbol) and the pitch reference on the PFD gives you the angle of attack - if you are aware of it.

If stablizer trim is full nose up (why is not important at this point) and TOGA is applied - does the elevator have the ability to overcome the consequences if full nose down input is made and held (without changing stabilizer trim)?

No - for any aircraft with this configuration, the stabilizer is more effective than the elevator, and easily overpowers it, especially with the added pitch up moment from high thrust.

There's a stall warning at 2:10:51, then the report notes that the stall warning horn stopped at 2:11:40, the clear implication being that the horn was sounding throughout the intervening period. So the instruments could not have been telling the PF he is overspeeding, as the instruments clearly thought he was stalling -- and saying so.

Stall warning comes from the angle of attack sensors, not the airspeed indicators - so a stall warning with an indicated high speed is possible.

Or is pitch calculated from the l@ser ring gyro and AoA via an aerodyamic sensor?

That's right - in the pic below (this one from a Boeing, but airbus use the same systems) you see two pitot tubes which detect airspeed, and between them the angle of attack vane, which detects the direction of the airflow (and thus the angle of attack).

http://i665.photobucket.com/albums/vv20/Checkboard/angleofattack.jpg

The inertial reference system also supplies vertical speed information, along with ground speed - so the vertical speed is independent of any air sensing failure.

PENKO

27th May 2011, 15:09

What is also interesting is that the pilots discussed the wx ahead. They told the cabin crew that here would be an increase in turbulence. How did they know that? Were they planning to fly through some of the returns? Flying through echo's at level 350 is a bit iffy, not something I would do voluntarily and not something I can imagine the captain leaving to the two FO's.

So here is the retrospective catch-22. The wx could not have been that bad, otherwhise the captain would linger a bit longer in the flightdeck. However the wx was bad enough for the FO's to warn the cabin crew. Was it that bad that there was no space to properly avoid the echo's? In that case, why was the captain away? The PNF suggested tentatively a turn, but only a turn of 12 degrees was made. What were they avoiding?

I have never flown across the ITCZ, so maybe others can comment, but again I feel as if we are missing a lot of information.

gums

27th May 2011, 15:09

Hey, TK??? Or others, for that matter.

- Don't have the full manual, only some flight control law pages. So...

Is the THS manual trim wheel connected to the THS via a mechanical means versus simply another electric command to the servoactuators/jackscrew motor/whatever?

- Is the THS position displayed in the cockpit?

- Any mention of "runwaway trim" procedures?

Interested_Layman

27th May 2011, 15:12

Got it, thanks ( and tanaka too )

Hyperveloce

27th May 2011, 15:14

Hi there,
The BEA note states that
- most of the pilot actions were pich up commands
- @ 02:10:51: incidence around 6° (when the stall alarms sounded), TOGA, pich up commands. 15 sec later, incidence around 16°, FL380, pich up commands again
- @ 02:11:40: end of stall alarms, FL350, -10 000 fpm, pich angle 15°, N1=100%, large (40°) roll excursions, pilot commands: roll compensation & pich up (during 30 s)
- @ 02:12:02: FP says "no longer any reliable indications", N1=55 %, 15 s later: pich down commands, incidence is decreasing, IAS reliable again, stall alarms
- @ 02:13:32: FP says "we are going to get under the FL100"

the incidence has always exceeded 35°, and seen from the pilots, the altitude was decreasing rapidly (a bit more than -10 000 fpm) with an inertial piching attitude reaching 15°: how can we explain that most of the pilot commands were to pich up the plane except around 02:12:17 (15 s after 02:12:02) ? Did the crew apply the AF SOP in the early phase (around 02:10:51) by setting TOGA thrust and a pich up at high altitude ? It is only 1 min 11 sec after that THR is on IDLE, the N1 is on 55 %, and that pich down commands are applied.

the stall is not recovered and the terminal vertical velocity is around -10 000 fpm.

why a crew would respond to the early stall alarms by a nose up ? spatial disorientation ? confusion/stress due to multiple conflicting indicators ? or tricky/misliding/lengthy procedures ?

JamesT73J

27th May 2011, 15:15

The only observation I can make is that I have seen a full and maintained nose-up demand from a pilot in a different aircraft before, in reaction to a large (and unexpected) rate of descent while stalled.

You know what? That makes absolute sense.

Checkboard

27th May 2011, 15:16

Is the THS manual trim wheel connected to the THS via a mechanical means versus simply another electric command to the servoactuators/jackscrew motor/whatever?
Yes, cable connection to the hydraulic servo motor (in the A320 at least) with a placard position gage next to the wheel. If you have hydraulic power, you have trim.

NARVAL

27th May 2011, 15:16

A few thoughts after reading this prelimanary report, and the many questions it raises. I write in a modest way, as it has been some years since I flew those planes, and I have forgotten a lot.
My very sad and friendly thoughts to the pilots of this flight.And everyone aboard, of course.
The THS (trimmable hydraulic stabilizer) auto trims the plane, the PRIM 1 is normally in charge of that job. The report does not say if the PRIMs (there are three) were lost, which is possible after losing speed information, and it does not say (understandably, it will come later) if from normal law, then alternate, the plane ended in direct law, then changed back etc…(possible resettings, speed indications coming back…)
If in direct law, there is not automatic trim, the horizontal stabilizer stays where it is and you have to use manually the trim wheel which a warning on the PFD, among very many other warnings in that case, tells you to do.(I am reminded of the Perpignan accident where the captain fought with the plane but if I remember correctly, never trimmed manually, with a very UP THS, exactly as described here, though for other reasons of course).
The THS is very effective, and even with sidestick applied fully down, if it stays at 13 degrees up, I think lowering the nose will be difficult…
From what I understand, having been familiar with this company, the captain went to take a rest, asked the most experienced of the two copilots (who had the appropriate qualification) to take his seat (left one) and the copilot in the right seat was PF. This is quite normal. If, and I do not believe it for a second, the PF was « lost » and applying full aft sidestick, while on the instruments, all kinds of understandable things happened, with chimes, alarms, synthetic voices etc…the copilot in the left seat , and even the captain coming back and sitting between them, had no way at all of knowing this, and could believe that he was applying the correct inputs, unsuccessfully. In the old days (and I am a very old-timer), putting your hand on the trimwheel was a habit…Nowadays, you don’t even know you have one, except for checking its position at take-off, which is not the same as using it in manual flight. Rolling it manually far forward, and that is always possible, in all modes or laws (correct me if I’m wrong, active bus drivers) and applying cruise thrust might have solved the problem, but that is easy to say from where I sit. I am quite sure of one thing, as you all are, but it bears to be repeated : thos pitot probes were changed by an efficient management, many months before, in another company, after an icing incident. The management of Air France truly thinks, not having done anything, that they did everythin very well, since it was not mandatory. Food for thought. May they sleep in peace.

jcjeant

27th May 2011, 15:17

Hi,

The BEA has kept its promise
He submitted a series of factual facts in chronological order
He announced that this would be incomplete .. it does not disappoint.
Upon reading the end of the release .. Lambda can conclude that:
The aircraft responded to control of the pilots and it was structurally capable
That the pilots did not follow the guidelines of Air France in effect has the time for this kind of event.
As this release is incomplete and is filled with shadows .. full of scenarios can be envisaged.
I am sure members of this forum will describe possible scenarios
Conclusion:
Much work awaits the BEA and the court of Paris

atakacs

27th May 2011, 15:19

Few remarks come to mind

Apparently those 3 well trained pilots did not manage to recover from a high altitude stall with a fully functional and structurally sound aircraft. They even recovered reliable speed measurements about half way in their descent but still ended up in the water. This is really puzzling and I believe we might still be missing an important piece of the equation.

For the fairly high tech measurements and recording equipments on board it seems that there is in fact not that much information being saved. I would venture to say that with modern FBW planes we should have all inputs covered as well as some "trace / log" of what the software is actually doing.

Although rare pitot icing is not unheard of. As other have mentioned it would be desirable to add some different airspeed measurement devices, as FWB can really become confusing with unreliable airspeed.

Checkboard

27th May 2011, 15:20

why a crew would respond to the early stall alarms by a nose up ? spatial disorientation ? confusion/stress due to multiple conflicting indicators ? or tricky/misliding/lengthy procedures ?

... or simply being constantly told during training "You can't stall an Airbus." as an absolute statement, reinforced by full back & side stick climbs at the lowest possible speed in the simulator to demonstrate "How the aircraft won't let you stall."

GarageYears

27th May 2011, 15:21

Gums: take a look here > SmartCockpit - Airbus 330 (http://www.smartcockpit.com/plane/airbus/A330/)

I believe the flight controls section has what you might need.

My understanding is the trim wheel has mechanical linkage.

Svarin

27th May 2011, 15:21

takata wrote : If Airbus didn't want someone to ever use the trim wheel, the very simple way was not to put one in the cockpit at the first place. They would even have made the electrical imputs to have a precedence over the pilot mechanical control which was, of course, made the other way.In the Perpignan accident, manual pitch trim was not used.
In this accident, it would appear it was not used either.

The fact that the manufacturer left this manual pitch trim wheel in the cockpit does not exclude the fact that in real life, in real airlines, on board real airplanes, pilots are actually taught not to use it, and never use it, according to design.

The piloting quality one would get using for the first time in a highly dynamic, critical situation this item of equipment would likely be very poor, anyway.

takata

27th May 2011, 15:22

Hi gums,
Is the THS manual trim wheel connected to the THS via a mechanical means versus simply another electric command to the servoactuators/jackscrew motor/whatever?
Yes. mechanical control, it is all you have in direct law.

- Is the THS position displayed in the cockpit?
Yes.

- Any mention of "runwaway trim" procedures?
Yes. indications written on trim wheel are for take-off (not related to flight position). I'll sent you that part later (no time to dig in now).

ImbracableCrunk

27th May 2011, 15:24

Does anyone remember why the US Navy set up Top Gun? Maybe it's time airlines start renting Cessnas, Robins, and Pipers.

That, and fixing pitot tubes that ice up.

3holelover

27th May 2011, 15:25

:confused:.... While it's nice to have some "factual" information, for a change, I'm certainly none the wiser for it. :( .... I hope it's not ages before we get the rest of the story. Conversations in the flight deck could be somewhat illuminating...

Golf-Sierra

27th May 2011, 15:27

There is a F/CTL STAB CTL FAULT procedure.

It basically says to use the Pitch Trim wheel.

One of the subsections reads: If trim is locked above 8 degrees UP, pitch down authority may be insufficient for speed above 180 knots.

PENKO

27th May 2011, 15:30

... or simply being constantly told during training "You can't stall an Airbus." as an absolute statement, reinforced by full back & side stick climbs at the lowest possible speed in the simulator to demonstrate "How the aircraft won't let you stall

That might be an (far fetched) explanation for the initial pitch up. Maybe the PF had a reflex action and applied a kind of windshear/terrain avoidance maneuvre. But after a few seconds it should be clear that that is not working.

I wonder in what state the aircraft was when the AP dropped out. Was it pitching down, or up? That's crucial information.

surplus1

27th May 2011, 15:35

No - for any aircraft with this configuration, the stabilizer is more effective than the elevator, and easily overpowers it, especially with the added pitch up moment from high thrust.

Thanks for the reply. I agree with your basic answer - although I would not go so far as to say "any aircraft" .

I presume you are AB330 qualified therefore, second question:

The BEA release (which I have now read) seems to indicate that the position of the THS was +13 deg. (nose up).

Question: On the AB330, what is the maximum nose up deflection of the THS? Might it be +13 deg or anything close to that?

FE Hoppy

27th May 2011, 15:41

No indication that the airspeed on the PFD was inaccurate. Just that after it went out of range, the PFD came back about a minute before the standby. They were close to max cruise alt, got a stall warning and zoom climbed to 38000.
On first reading this doesn't look good!

Checkboard

27th May 2011, 15:43

Not A330, but A320 series qualified. On that type, the THS limits are 13.5º nose up and 4º nose down.

DozyWannabe

27th May 2011, 15:47

Dozy, I would not dare to say that yet, the report is very unclear. It feels as if we are presented only half the information.

Just to clarify - I wasn't suggesting that's what happened, I was responding to your question as to why a pilot approaching stall would pitch up suddenly in that manner. All I meant by the example was to show that pilots have done it before.

blind pew

27th May 2011, 15:48

Checkboard

Thanks for the AOA post - never flew the bus.

BOAC
regarding pilot pulling back on the stick with a stall
- seen it but not on an airliner - it is a natural reaction to a falling sensation as is sticking out a hand which leads to a broken wrist.

Re; simultaneous stall and overspeed warnings for the non ATPL holders.
Three pitots and two static sources - depends on timing and sequence of freezing up.

Similar to the Trident crash;
the crew would have faced many simultaneous aural and visual warnings.
it would be very confusing.

The philosophy at the time of the trident crash was that if the stick push fired then dump it as it would be a malfunction.

I can understand with the bus design philosophy of not being able to stall the bus that the crew disbelieved the stall warning.

One question that has not been addressed is whether the horizon on PF side was indicating correctly?

I doubt if we will ever know what the pilots were looking at.

ITCZ 12o heading change - quite normal - it's like flying through a bowl of tapioca with different sized lumps.

hopefully we will one day have 24/7 data links with all parameters including video - I suggested video for sim checks 20 yrs ago and was told why?

GarageYears

27th May 2011, 15:50

http://i1116.photobucket.com/albums/k579/GarageYears/A330_Pitch.jpg

paull

27th May 2011, 15:50

Simple question, how easy is it with the side stick to know what inputs the other guy is making?

Lonewolf_50

27th May 2011, 15:50

@Checkboard:
From a training briefing dated 1999 on Airbus 330. I don't know if Airbus has changed this. Looks similar to what you are familiar with.

"Pitch control is provided by two elevators and the THS :
- elevator deflections 30° nose up - 15° nose down
- THS deflections 14° nose up - 2° nose down."

Yellow Pen

27th May 2011, 15:54

Where on earth has this idea that you can't stall an Airbus come from? It's made perfectly clear during ground school before you even get near a simulator that you can stall the aircraft, and you will stall the simulator during training. What is more of a concern is the full left and nose up sidestick input, something which would not be evident to the other pilots.

takata

27th May 2011, 15:55

In the Perpignan accident, manual pitch trim was not used.
In this accident, it would appear it was not used either.
Perpignan was a test flight where they tried to kick "Alpha prot-Alpha max" at low altitude with defective AoA probes. They had no time to recover from FL380 like AF447 was. I think I do remember that those test pilots crashing in Toulouse (1994), very close from recovering from a low alt LOS with an engine stopped on test purpose, used manual ths trim. But what would be the point if they did? They would have been authorized and any other pilot forbidden.

The fact that the manufacturer left this manual pitch trim wheel in the cockpit does not exclude the fact that in real life, in real airlines, on board real airplanes, pilots are actually taught not to use it, and never use it, according to design.

Right, of course: it is like having a "direct law" with only manual THS trim available, with a message poping up and saying "Manual pitch trim only"... WTF!!! I won't ever use this cr@p or I'll be fired!
Your position doesn't make sense: it is not "per-design", it is not "forbidden": it is always better not to have to use it (like being in normal law instead of direct)... up to the point that you need it!
Now, how training is performed is a different issue.

PENKO

27th May 2011, 15:56

I wonder what caused the first stall warning. Was it loss of airspeed or was it G-induced by the pilot trying to recover after the AP dropped out? After that warning they still managed to climb with 7000 feet per minute to FL380...
This was recovered by a nose down command, followed by nose up for the second stall warning.

jcjeant

27th May 2011, 15:58

Hi,

For those who understand french .. it can be interesting ...
A graphical chronology of the events commented in the BEA communication.... (from a french forum)
It's more easy to see the gap in time of no comments at all (shadow zones)

http://i.imgur.com/2HeUa.jpg

Iceheart

27th May 2011, 16:01

One may want to compare this accident with Pulkovo Flight 612 accident:

- in both cases, aircraft stalled at high altitude when crossing severe turbulence;
- in both cases, pilots were perfectly aware of weather conditions and made attempts to avoid flying into the worst;
- in both cases, pilots were mainly pulling the stick (control column in case of Pulkovo 612) instead of applying proper stall recovery procedure until it was too late.

And yet:

- Pulkovo 612 was a Tu-154 aircraft, which is not FBW and is lagging almost 50 years behind A330 in terms of automation;
- One of the Pulkovo 612 pilots was a former aerobatic champion, so he probably knew something about stalls and stall recovery;
- Tu-154s have AoA indicators in cockpit, which provided meaningful and valid information throughout the accident. Also, pitots did not freeze and the airspeed indicators were also valid.

Yet it seems that in both cases pilots either did not recognize that the aircraft has entered fully-developed stall, or somehow failed to apply proper recovery procedure.

One can probably draw a conclusion that such proposed measures as "upgrading the pitots so that they never freeze" or "reducing level of automation" (whatever falls into that category) would not prevent such accidents from happening again.

Just my 2 cents.

surplus1

27th May 2011, 16:07

Thank you very much, gentlemen (Checkboard, Garage Years, Lonewolf 50).

From your replys and the BEA release, I gather that the THS was within 1 deg. (+13) of its maximum nose up position - and approximately 5 deg. beyond the point at which it could be overcome by full nose down elevator (without changing THS angle). [Please correct me if this is wrong.]

One final question: Is the thrust line of the AB330 such that the application of TOGA thrust will induce a significant nose up (pitch up)moment?

gums

27th May 2011, 16:07

Wow! Super answers, and the Garage guy's reference is outstanding. Hell, I'm almost ready to try to fly the beast after a few more hours of study, heh heh. As a TV ad says, " it's so easy that a caveman can do it".

I raised the runaway trim issue, as many of us were taught to roll the plane if we had runaway nose up trim. The technique was intended to maintain a reasonable pitch attitude. For nose down, the idea was to reduce power/slow down in order to overcome the elevator trim. Of course those procedures were for "elevators" and not the horizontal stab.

As we try to digest the data and maybe cry a bit, I'll ad lib and add a war story about horizontal stabs. JT Moderator can delete if appropriate, but it explains a few things about many planes' horizontal stab designs.
++++++++++++++++++++++++++++++

So first few months at Hill we had all kindsa celebrities drop by the first Viper unit in the world. One was Chuck Yeager. And we all assembled in the main briefing room and he sat on the stage, feet on the floor, and told us war stories and answered questions. Was a magic moment, I tellya.

He got to the part about the first supersonic flight and a bit of advice from a Bell technician/aero dude. He told Yeager that once supersonic or even close to the mach that they were worried about losing elevator control due to the shock waves. So they had a manual wheel to 'trim" the horizontal stab. He told Yeager that if all else failed, to use that sucker and he might gain pitch control.

Sure enough, above the mach the elevators didn't work very well, if at all. So Yeager cranks the wheel back and forth and regains pitch control. Back below the mach all was "normal". He told us that this discovery was why North American and other folks developing the new jets went to the one-piece horizontal stab. Moving the entire thing as one piece changes the shock wave pressures and the thing acts like a "normal" jet.

Gotta love it!
++++++++++++++++++++

We now return to our regular hypothesizing and second-guessing, heh heh

DozyWannabe

27th May 2011, 16:12

... or simply being constantly told during training "You can't stall an Airbus." as an absolute statement, reinforced by full back & side stick climbs at the lowest possible speed in the simulator to demonstrate "How the aircraft won't let you stall."

You're forgetting the caveat "as long as the relevant protections are working", which is a pretty important one.

Now, how training is performed is a different issue.

Agreed - there seems to be considerable confusion here between recommendations made by Airbus themselves versus how those recommendations are filtered down to airline training departments. If what Svarin says is true, then I'd say that's a pretty serious deficiency in that airline's training!

Someone asked me in the previous thread about what I meant by Airbus acknowledging some of the more lurid claims made about the safety aspects of the system in the early days were unhelpful, to which I'll say that their whole training and marketing syllabus changed tack dramatically in the 1990s. Rather than focusing on how easy the aircraft was to fly compared to more traditional aircraft, the focus changed to one of the systems "assisting" the pilot.

Peter H

27th May 2011, 16:14

JPI33600 (http://www.pprune.org/members/329150-jpi33600)
Apart from R&D work about how to prevent UAS condition, this seems to call for a better training of pilots in unusual situations.As someone with no aviation experience, can I ask how much a timely warning of [the likelihood of] UAS would have helped the pilots?

If it would have helped, perhaps more attention should be given to monitoring systems.

If cars can warn of hazardous road-temperatures, might not planes try to warn of ice-crystals.

Regards, Peter

PS The more I think about a "flight engineer's console", the more it seems to complement the strengths of existing FBW systems.
For example, the u/s (but not 'faulty') RA at Schiphol could have been identified by sanity and/or consistency checks, and [manually] taken out of service.

cats_five

27th May 2011, 16:14

Could the physical sensation of a rapid descent combined with doubt about the veracity of the flight information and the lack of outside visual references simply trick the brain into misreading the situation they were in?

You don't feel descent, only acceleration. You will feel the initial downwards acceleration but once travelling at a steady speed there is nothing to feel.

FE Hoppy

27th May 2011, 16:19

A question for the AB pilots:

Is there an elevator offload to the stab? By this I mean does the aircraft auto trim in the direction of input to centralise the elevator? and would a prolonged nose up pitch input lead to the stab auto trimming itself nose up?

Svarin

27th May 2011, 16:21

studi wrote :

Am I the only one here thinking that the climb was not induced by the pilots but by some sort of ill functioning protection?

No.

Interested_Layman

27th May 2011, 16:33

Neglecting for the moment how the aircraft entered the stall, do the A330 pilots here believe such a stall was recoverable, allowing for the THS trim (correctable? ), cruise aft CG, and aircraft weight and altitude?

Possibly recoverable, or routinely recoverable?

Is this ever practiced in sim? ( guessing not, hoping it is )

blind pew

27th May 2011, 16:41

Am I the only one here thinking that the climb was not induced by the pilots but by some sort of ill functioning protection?

moi aussi.

But the question is what were his instruments telling him?

If the static was blocked he would not have known that he was climbing.

If there was some pitot pressure and he climbed with blocked static then that would have indicated a reduction in IAS.

But it is still hard to believe with a correct horizon that he would pitch up.

Having declared a mayday with a hitherto unknown flight control malfunction on a classic aircraft I have a lot of sympathy for those who have to deal with modern aircraft - just spent two hours trying to get my computer to recognize my printer - unsuccessfully!

Despite AF resent history my money is on more than icing, pilot training and pilot error.

Svarin

27th May 2011, 16:49

takata wrote :

Right, of course: it is like having a "direct law" with only manual THS trim available, with a message poping up and saying "Manual pitch trim only"... WTF!!! I won't ever use this cr@p or I'll be fired! Your position doesn't make sense: it is not "per-design", it is not "forbidden": it is always better not to have to use it (like being in normal law instead of direct)... up to the point that you need it! Now, how training is performed is a different issue. Regarding manual pitch trim, the design and associated training is thus :

Normal law : auto-trim. Dont touch it or fail the test and get fired. 99.9999% of the time for airline pilots. Protections have ultimate authority on control surfaces (hard protections). Eliminated at the start of the accident event sequence.

Alternate law (1 & 2) : auto-trim, which means dont touch it either. No specific PFD, ECAM, whatever, indication regarding manual pitch trim. Rarely trained for. Some protections remain, with a less assertive authority (soft protections). According to factual report, this was how the airplane flew its last minutes. Requires at least one working PRIM.

Direct law : PFD amber warning USE MAN PITCH TRIM. Similar to conventional airplane, direct stick position to control surface position. According to factual information, this law was not triggered. No protection. Rarely if ever trained for, except initial training towards type rating (box-ticking mentality). Requires at least one working SEC.

Mechanical backup : PFD red warning MAN PITCH TRIM ONLY. Basically, sidestick becomes useless. Remaining only manual pitch trim wheel and mechanical rudder, which could give you a reasonable fighting chance depending on actual circumstances. According to factual information, not triggered. Trained once, upon type rating training process (box-ticking mentality). Main condition is all five FCCs lost (3 PRIMs, 2 SECs, all off. Did not happen.)

For the sake of almost completeness :
Other laws : abnormal attitude law, flare law, ground law, ground/air blending laws... These (apart from abnormal attitudes) are quiet transitions, appear every day, and would be perhaps considered subsets of Normal law. No cockpit indication that these sub-laws are triggered. Conditions that activate them can be found in the FCOM. Probably irrelevant to our current matter, apart perhaps from abnormal law but this was not reported.

Assuming that such a significant event as any additional flight law change would have been reported by BEA. Subject to revision as more information becomes available.

glenbrook

27th May 2011, 16:53

I was wrong. I can't understand how a pilot could pitch his a/c up 2500 feet without realising it. He knew he had UAS but surely he didn't think he had unreliable altitude? 7000 feet per minute upwards after you get a stall warning? What happened in the 11 seconds between the AP disconnect and the pilot's realisation that he had UAS? Did he pitch up, get a stall warning or did he get a stall warning and then pitch up? What was the engine speed at this time? What BEA has done here is worse than useless. They have tried to provide details to quash rumours but by providing less information than is necessary to properly understand what happened. I think they should have either given out everything or nothing.

DozyWannabe

27th May 2011, 16:59

Am I the only one here thinking that the climb was not induced by the pilots but by some sort of ill functioning protection?

The only protection I know of that would do anything like that is one that applies to the autopilot when the aircraft makes a sudden vertical maneouvre opposite the direction commanded. If the autopilot was already disconnected by that point then it's unlikely that protection mode would have been commanded.

Checkboard

27th May 2011, 17:12

Is there an elevator offload to the stab? By this I mean does the aircraft auto trim in the direction of input to centralise the elevator? and would a prolonged nose up pitch input lead to the stab auto trimming itself nose up?
Of course - that's how trim, and auto trimming works. A constant input on the stick (or from the autopilot) is trimmed out using the Trimmable Horizontal Stab.

Neglecting for the moment how the aircraft entered the stall, do the A330 pilots here believe such a stall was recoverable, allowing for the THS trim (correctable? ), cruise aft CG, and aircraft weight and altitude?

Possibly recoverable, or routinely recoverable?
The aircraft (with the possible exception of the pitot tubes) was fine - the situation should have been routinely recoverable.

If there was some pitot pressure and he climbed with blocked static then that would have indicated a reduction in IAS.

Climbing with blocked pitot results in the airspeed displayed increasing from the speed displayed when the blockage occurred, not decreasing.

In the Perpignon disaster, from the final report (http://www.bea.aero/docspa/2008/d-la081127.en/pdf/d-la081127.en.pdf):

When the aeroplane exceeds certain attitude thresholds, roll, angle of attack or speed, the system uses a specific law [abnormal attitudes law]. The display on the PFD is identical to that for alternate law. This abnormal attitudes law breaks down into two phases:

ˆ For the first phase, the law used in pitch corresponds to the alternate law without auto-trim and with only the load factor protection. In roll, a direct law with maximum authority is used. The yaw is controlled mechanically. This first phase should make it easier for the crew to return to more usual attitudes.

When the stall warning sounded, the Captain reacted by placing the thrust levers in the TO/GA detent and by pitching the aeroplane down, in accordance with procedures. The nose-down input was not however sufficient for the automatic compensation system to vary the position of the horizontal stabilizer, which had been progressively deflected to the pitch-up stop by this system during the deceleration. The Captain controlled a left roll movement, caused by the stall. The aeroplane’s high angle of attack and the roll movements generated asymmetry, and a speed variation between ADR 1 and 2 appeared. This increasing divergence caused a rejection of the three ADRs by the FAC then the ELAC. The flight control system then passed into direct law. [i.e. No auto trimming in Direct Law] It is likely that the crew did not notice this due to the emergency situation and the aural stall warning that covered the warning of a change of flight control laws. The Air New Zealand pilot, by saying “alpha floor, we’re in manual” likely considered that the alpha floor function had triggered and that in fact the autopilot had disconnected.

The aeroplane attitude increased sharply and its speed dropped to the point that rendered it practically uncontrollable, the flight control surfaces becoming ineffective due to the low speed and the high angle of attack. The aeroplane stalled again, this time irrecoverably, bearing in mind the aeroplane’s altitude and without any crew inputs on the trim wheel and the thrust levers.

The loss of control was thus caused by a thrust increase performed with a full pitch-up horizontal stabilizer position. This position and the engine thrust made pitch down control impossible. It should be noted that the PF made no inputs on the horizontal stabilizer nor reduced the thrust and that the PNF did not intervene. This seems to indicate that none of them were aware that the automatic trim system, which relieves the pilot of any actions to trim the aeroplane, was no longer available.

On approach to stall and taking into account the dynamic of the flight and of the complexity of the displays, the automatic changes in the control laws can fail to be perceived and their consequences can sometimes be misunderstood by pilots. In this case, the passage to direct law rendered the auto-trim function inoperative. Even if the amber USE MAN PITCH TRIM flag was displayed on the two PFD artificial horizons, the crew did not notice the position of the stabilizer and did not command the trim wheel manually during the twenty-five seconds in direct law between 15 h 45 min 15 s and 15 h 45 min 40 s. From this time on and for the rest off the flight, as a result of passing into abnormal attitudes law, the amber USE MAN PITCH TRIM flag was no longer displayed. The systems thus functioned in a degraded manner, without the real overall situation of the aeroplane being known by the crew.

[emphasis and notes mine]

GarageYears

27th May 2011, 17:13

Am I the only one here thinking that the climb was not induced by the pilots but by some sort of ill functioning protection?

No - you yourself provide the quote from the BEA that indicate it was initiated by the crew (my bold):

From 2 h 10 min 05, the autopilot then auto-thrust disengaged and the PF said "I have the controls". The airplane began to roll to the right and the PF made a left nose-up input. The stall warning sounded twice in a row. The recorded parameters show a sharp fall from about 275 kt to 60 kt in the speed displayed on the left primary flight display (PFD), then a few moments
later in the speed displayed on the integrated standby instrument system (ISIS).
Note 1: Only the speeds displayed on the left PFD and the ISIS are recorded on the FDR; the speed
displayed on the right side is not recorded.
Note 2: Autopilot and auto-thrust remained disengaged for the rest of the flight.
At 2 h 10 min 16, the PNF said "so, we’ve lost the speeds" then "alternate law […]".
Note 1: The angle of attack is the angle between the airflow and longitudinal axis of the airplane. This information is not presented to pilots.
Note 2 : In alternate or direct law, the angle-of-attack protections are no longer available but a stall warning is triggered when the greatest of the valid angle-of-attack values exceeds a certain threshold.
The airplane’s pitch attitude increased progressively beyond 10 degrees and the plane started to climb. The PF made nose-down control inputs and alternately left and right roll inputs. The vertical speed, which had reached 7,000 ft/min, dropped to 700 ft/min and the roll varied between 12 degrees right and 10 degrees left. The speed displayed on the left side increased
sharply to 215 kt (Mach 0.68). The airplane was then at an altitude of about 37,500 ft and the recorded angle of attack was around 4 degrees.
From 2 h 10 min 50, the PNF tried several times to call the Captain back.

From my read of the above, the pitch up command pre-dates the Stall Warning (though we do not have a well defined timeline to understand exactly the second by second sequencing).

We can see that the initial Pitch Up demand led to the climb, ultimately reading 7000ft/min, prior to the Pitch Down command which slowed the climb rate to 700ft/min, however, it seems by then significant airspeed was lost.

The text that follows in the report seems to be the most perplexing:

At 2 h 10 min 51, the stall warning was triggered again. The thrust levers were positioned in the TO/GA detent and the PF maintained nose-up inputs. The recorded angle of attack, of around 6 degrees at the triggering of the stall warning, continued to increase. The trimmable horizontal stabilizer (THS) passed from 3 to 13 degrees nose-up in about 1 minute and remained in the latter position until the end of the flight.

Around fifteen seconds later, the speed displayed on the ISIS increased sharply towards 185 kt; it was then consistent with the other recorded speed. The PF continued to make nose-up inputs. The airplane’s altitude reached its maximum of about 38,000 ft, its pitch attitude and angle of attack being 16 degrees.

At this point we've seen a half-heart attempt to use pitch-down, which may just have been the solution. But that appears to have been abandoned. Pitch up demands then remain, until very late in the sequence, at which point the vertical speed was already at -10,000ft/min with an angle of attack stated to be +40degrees - the stall was full developed and they ran out of altitude.

Re-reading the report is a sad and sobering task. I simply do not understand the initial pitch-up demand. All that seemed warranted was a correction due to the initial right roll? However there is no mention of any throttle demand at the point where the first pitch up command was provided, so presumably airspeed is bleeding off from this moment on. TO/GA thrust was eventually selected 46 seconds after AP and AT disengagement. Too late.

AlphaZuluRomeo

27th May 2011, 17:16

@ studi, Svarin, blind pew and those who think "that the climb was not induced by the pilots but by some sort of ill functioning protection":

May I kindly ask:
- what protection you refer to / could you elaborate on why/how it's been triggered (with AP&ATHR OFF since 2 h 10 min 05, and ALT2 law active since 2 h 10 min 16 at the latest)?
- why/how do you seem do discart the followings?
From 2 h 10 min 05, the autopilot then auto-thrust disengaged and the PF said "I have the controls". The airplane began to roll to the right and the PF made a left nose-up input.

FE Hoppy

27th May 2011, 17:17

Of course - that's how trim, and auto trimming works. A constant input on the stick (or from the autopilot) is trimmed out using the Trimmable Horizontal Stab.

I prefer not to guess how other aircraft systems work, hence the question.

From this we can deduce that the THS movement was in response to the pilots input and not a cause of the initial climb or final stall event.

chaz88z

27th May 2011, 17:20

Quote:
Originally Posted by eireoflot82
""Could the physical sensation of a rapid descent combined with doubt about the veracity of the flight information and the lack of outside visual references simply trick the brain into misreading the situation they were in?

You don't feel descent, only acceleration. You will feel the initial downwards acceleration but once travelling at a steady speed there is nothing to feel.""

You don't feel acceleration but might be sensing the lack of it :

Descending at ~10.000fpm isn't it an aproximate rate of free-fall?
Could you relate this very situation with a 0-G maneuver (on the A300-0G)?

The report says 40° of angle of attack at some point even if pilots are trained to recognize a stall situation (on light aircrafts)
Wouldn't that be disturbing when flying in the "goo" with no airspeed, a pitch up indication, possibly a left bank angle but heading increasing ?

One of the copilot even says to the captain :
"We do not have any valid information..."

With all the workcharge, failures and warnings and wrong infos (speeds) one might become sceptical and "untrusty" over his Attitude indicator.
Moreover, nobody knows what the RH PFD was showing. Even if I tend to think they were both showing more or less identical values

Checkboard

27th May 2011, 17:27

Descending at ~10.000fpm isn't it an aproximate rate of free-fall?
Could you relate this very situation with a 0-G maneuver (on the A300-0G).
Actually, it's a bit quicker than a parachutist at terminal velocity in free fall - but a parachutist at terminal velocity is experiencing 1 G, not 0 G, (they are effectively lying on a bed of air - but internally feeling the same as lying on a real bed.)

MurphyWasRight

27th May 2011, 17:28

An attempt at extracting a logical flow from the narrative. Ellipses (...)
indicate portions I removed for clarity, items such as inline explanations,times and displayed speeds. <comment> are my thoughts.
Bold indicate possible action/reaction chains.

...
the autopilot then auto-thrust disengaged and the PF said "I have the controls".
The airplane began to roll to the right and the PF made a left nose-up input.
The stall warning sounded twice in a row.
...
The airplane's pitch attitude <corrected from AOA> increased progressively beyond 10 degrees and the plane started to climb.

The PF made nose-down control inputs and alternately left and right roll inputs.
The vertical speed, which had reached 7,000 ft/min, dropped to 700 ft/min and the roll varied between 12 degrees right and 10 degrees left.
The airplane was then at an altitude of about 37,500 ft and the recorded angle of attack was around 4 degrees.
<almost regained control?>
...
the stall warning was triggered again. < by the AOA greater 6 degrees?, mentioned later>

The thrust levers were positioned in the TO/GA detent and the PF maintained nose-up inputs.
The recorded angle of attack, of around 6 degrees at the triggering of the stall warning, continued to increase.

The trimmable horizontal stabilizer (THS) passed from 3 to 13 degrees nose-up in about 1 minute and remained in the latter position until the end of the flight. < as result of sustanined nose up input?>
...
The PF continued to make nose-up inputs. The airplane's altitude reached its maximum of about 38,000 ft, its pitch attitude and
angle of attack being 16 degrees.
...
During the following seconds, all of the recorded speeds became invalid and the stall warning stopped.

<possible "well that worked" reaction by PF who was ignoring displayed speeds at this point anyway? >
...
The altitude was then about 35,000 ft, the angle of attack exceeded 40 degrees and the vertical speed was about -10,000 ft/min.
...
The airplane was subject to roll oscillations that sometimes reached 40 degrees.
<falling like a leaf but only sensing side to side not down since steady state,not trusting displayed vertical rate?>

The PF made an input on the sidestick to the left and nose-up stops, which lasted about 30 seconds.
...
(both said) "we have no valid indications". At that moment, the thrust levers were in the IDLE detent
...
Around fifteen seconds later, the PF made pitch-down inputs. Inthe following moments, the angle of attack decreased, the speeds became valid again and the stall warning sounded again.

< return of stall warning was confusing since it coincided with actual improvement? Perhap led to nose up command? >
...
About fifteen seconds later, simultaneous inputs by both pilots on the sidesticks were recorded
<would be usefull to know the inputs and if they matched>
and the PF said "go ahead you have the controls".
The angle of attack, when it was valid, always remained above 35 degrees.

spagiola

27th May 2011, 17:33

From the BEA report: about 15 sec after 2:10:51 (ie about 2:11:06), the two recorded speeds (LHS and ISIS) are once again the same, and remain the same for the remainder of the flight. To me, this sounds like the pitot probe problem cleared (ice melted or was dislodged?) at that point, and did not recur. The speeds then are noted as becoming invalid again (eg at 2:11:40 just after the captain arrives, until 2:12:02 + about 15 sec), but the report pointedly notes that they would do so whenever they fall below 60KT.

That's over three whole minutes in which they apparently had functioning speed indicators (a little after 2:11 up to impact at 2:14:28). They also had plenty of altitude, as this is about the same time they reached their maximum altitude of 38000.

By then the THS is trimmed nose up 13 degrees. Between 2:11:40 (when we are told N1 is almost 100%) and 2:12:02 (when we are told throttles are already at idle and N1 is 55%), the throttles were retarded.

At about 2:12:02+15, the PF commands pitch down, pitch is reduced, and speeds become valid again (ie climbing above 30KT).

Impact didn't occur until over whole two minutes after pitch down was commanded. Apparently not enough pitch down as the aircraft remained nose high (AOA at least 35 degrees), descending very rapidly. Did the PF stop pitching down? Or was he continuing to pitch down, but failing to drop the nose because of the up trim on the THS? The report says nothing about pitch commands after 2:12:02+15 except to note that, in general, pitch commands were "mostly up" after AP disengagement.

To my mind, it's those last three minutes, and particularly the last two, that need to be explained. There was an initial upset, and a reaction to that which led to a stall, but by about 2:11 the speed indications were apparently valid.

Was the crew confused or fixated on the wrong problem? If we knew more about what they were saying, it might be possible to figure it out. I can't imagine that nothing was said in these whole of the last two minutes beyond the brief exchange noted at 2:13:32.

Did the resumption of the stall horn once the speeds once again become valid after commanding pitch down (at 2:12:02+15) confuse the PF into thinking he was doing the wrong thing?

Perhaps it might also be worth noting that by 2:11:40, the aircraft was already dropping at -10000fpm, and that it would have almost exactly this vertical velocity when it hit, some three minutes later. This suggests there may have been little vertical acceleration during this period, perhaps lulling the pilots into thinking they were in stable flight? Yet they did note that they were dropping through FL100.

Lots of things are just not adding up.

[edited for grammar and to correct error on min speed at which speed is considered invalid; 60KT, not 30KT (see BEA report, note at bottom of p.2 of French version]

Lonewolf_50

27th May 2011, 17:35

Murphy, as you broke that down, you end up part way through the sequence with this progression:

"I push the stick forward, I am now stalled again, I need to pull the stick back" which is a valid response in inverted flight.

:confused:

In re roll oscillations: is that characteristic of an AB in a stall, or a symptom of the usual auto trim/auto correct dampening functions dropping off, or a combination of the latter and a pilot seeking a stable condition?

(Recalling SAS off flying and how sensitive it is to inputs ... )

wozzo

27th May 2011, 17:40

(...)The airplane's angle of attack increased progressively beyond 10 degrees and the plane started to climb.
Corrected by BEA to "The airplane’s pitch attitude increased progressively beyond 10 degrees (...)" (See current PDF)

MurphyWasRight

27th May 2011, 17:42

LW50:

Murphy, as you broke that down, you end up part way through the sequence with this progression:

"I put the nose down, I am now stalled again, I need to bring the nose back up" which is a valid response in inverted flight.

Interesting point, could the pilots have been -that- disoriented?
(I have no knowledge/opinion either way.)

If I parsed the narrative correctly (would much rather look at the DFR data) the stall warning returned when the speeds were again deemed valid by the system.

SeenItAll

27th May 2011, 17:43

I am not a pilot, but have some knowledge of vectors and physics.

It seems to me that the critical fact is that the plane was pitched up, but the pilot input was to command nose up. It it possible that this incongruous situation was the result of a huge discrepancy between pitch (which is relative to the horizon) and AoA (which is relative to the apparent wind vector)?

If, for example, the plane was caught in a draft of wind from below (which might explain some of the altitude gain), wouldn't the AoA indication be negative -- and the pilot's response be to command nose down -- even while the plane's pitch was positive? Or is my thinking garbage? Don't worry about being harsh with me, I don't claim to be an expert. :)

atakacs

27th May 2011, 17:49

Quick question: on what basis does the BEA reports that the speed measurement became valid again at 2:12:21 ?

blind pew

27th May 2011, 17:49

Roll oscillation sounds like dutch roll!

BOAC VC 10 got a high fly wrong and rolled through the vertical - day time with instructor and a horizon.

checkbord - posted blocked static - increase alt = ias decreases.

PENKO

27th May 2011, 17:52

Maybe this question has been asked before in the previous topics but I dismissed those as way too much speculation:

Is it common to have airframe icing at FL350 outside a CB in those areas?

SeenItAll

27th May 2011, 17:59

Official AF reaction to today's news -- the aircraft's instruments failed and the pilots reacted completely professionally.
Air France - Corporate : AF 447 - Air France (http://corporate.airfrance.com/en/press/news/article/item/af-447-reaction-dair-france-a-la-note-dinformation-du-bea/)

mitrosft

27th May 2011, 18:04

Well BEA made the case too easy for an Airbus and too complicated for general public by just saying there was constant pitch up input from PF without actually saying the reason for it.

Maybe this will be revealed in the course of investigation maybe not. However all later findings would be biased by this "pilot error" suspicion.

Hope that French court would be as demanding as PPruNe society.

Coming from very centralised country (France is very alike) I think that Airbus will be set free and pilots will be to blame. :ugh:

Sad.

spagiola

27th May 2011, 18:04

Quick question: on what basis does the BEA reports that the speed measurement became valid again at 2:12:21 ?

2:12:21? No, not then.

The report states that about 15 seconds after 2:10:51, the ISIS speed measurement abruptly jumps to 185KT, at which point it is consistent with the other recorded speed (ie the LH speed, as the RH speed is not recorded). It notes no further disagreements between the two recorded speeds for the remainder of the flight; rather, it notes that at various points that "all recorded speeds became invalid" (presumably when they fell below 30KT) or became valid (climbing above 60KT), indicating that they remained consistent with each other.

If I'm not mistaken, with at least two consistent speeds, they would be considered reliable even if the third (which was not recorded, and so we don't know if it differed) was different.

[edited to correct error on min speed at which speed is considered invalid; 60KT, not 30KT (see BEA report, note at bottom of p.2 of French version)]

Checkboard

27th May 2011, 18:12

checkbord - posted blocked static - increase alt = ias decreases
Actually checkbord - posted blocked static - increase alt = ias increases.

Pitot tube has Dynamic Pressure (airspeed information) AND static pressure in it. The Air Data Unit (ADU) subtracts the sensed static pressure (from the static port) from the pitot total pressure to determine the dynamic pressure alone, and thus calculate the airspeed.

If the pitot becomes blocked, the dynamic pressure & static pressure at the time of the blockage remains the same. As the aircraft climbs, the sensed static pressure reduces (with the reducing altitude) so the ADU subtracts a smaller static pressure from the pitot total pressure, and thus comes up with an increasing airspeed. The airspeed guage is now working something like an altimeter.

... Now that's the theory - it actually depends on the nature of the blockage and the leakage within the "blocked" pitot system.

henra

27th May 2011, 18:17

Am I the only one here thinking that the climb was not induced by the pilots but by some sort of ill functioning protection?

Sorry if I have to be rude to you:
Did you care to read the report of BEA ???

What do you think 'Nose-up Input maintained by the PF' means ??

What is ambiguous about this statement ?

I'm completely at a loss how the first reaction of our conspiracy brigade (not only you) can be total denial of the most basic and easily determinable findings of the FDR i.e. sidestick position???

Yes it is difficult to understand why a pilot would pitch up signifcantly after loss of airspeed indications but the sensation of turbulence can play nasty tricks on you. Maybe they hit some bumps or holes in the air during AP disconnect and lost some altitude exactly at that moment. Such things can be badly misleading....

robertbartsch

27th May 2011, 18:23

Does the new report shed any light on the structural condition of the air frame? Early on, many believed the VS separated before the impact with the water.

Microburst2002

27th May 2011, 18:24

The non-recorded speed is an anoying thing.

Who was PF. the left hand or the right hand?

If the PF was the one flying on the non recorded speed, is it possible that he was pitching up because he saw speed too high and increasing?

AmericanFlyer

27th May 2011, 18:24

I am not a pilot, but many knowledgeable posters feel that a factor has been overlooked which would help in understanding why the pilots reacted as they did.

Could that factor have been the local weather? Could there have been a large downdraft or updraft that caused the pilots to misunderstand what was happening to their aircraft? At night, without external references, it may have been very confusing.

Checkboard

27th May 2011, 18:26

I'm completely at a loss how the first reaction of our conspiracy brigade (not only you) can be total denial of the most basic and easily determinable findings of the FDR i.e. sidestick position???

Well, The FDR really tells us nothing about sidestick position - you need a camera for that - it only tells us about the sidestick command. A sidestick failure, giving pitch up commands regardless of stick position is a (slim) possibility. The actual FDR graph of stick command would show if it is a possible culprit though - and as it has not been mentioned (and it would by now) you can discount it.

grity

27th May 2011, 18:30

Am I the only one here thinking that the climb was not induced by the pilots but by some sort of ill functioning protection?

pull zoomclimb stall chrash....

A340 TC-JDN:
the first elevator moving to +4 deg, 4 sec after the autopilot self-disengaged as the aircraft exceeded the speed limit, +4 deg only for 2 sec was enough to start the zoom climb for the A340 TC-JDN (en-route from Istanbul to New York) and was defenitiv not triggerd by the sidestick of the pilot (the first sidestick pitch was 20 sec later).....

AF447:
2 h 10 min 05 autopilot self-disengaged
2 h 10 min 15 "alternate law"

what is with the g-protection in the time between ???

GarageYears

27th May 2011, 18:31

Official AF reaction to today's news -- the aircraft's instruments failed and the pilots reacted completely professionally.

Well, the exact words are:

It appears that the flight deck crew was monitoring the changing weather conditions and thus altered the flight path, that the initial problem was the failure of the speed probes which led to the disconnection of the autopilot and the loss of the associated piloting protection systems, and that the aircraft stalled at high altitude.

This seems to be an interpretive version of the events - what protections were 'lost'? Protections still exist in Alt - however they are not the same as in Normal.

Rob21

27th May 2011, 18:41

The PF reaction is typical of a pilot flying IMC with no instruments at all, or no reliable basic instruments at all.

How come a huge airline transport aircraft can come down just because the computers gave up after a false airspeed indication?

I have the impression the pilots lost all the information they needed to maintain attitude. Worse, they had their displays going crazy on them...

BEA is clearly "inducing" public opinion to believe the pilot did wrong.
We only see on the report parts that makes us believe the aircraft responded accordingly. They insisted on the "PF kept commanding pitch up" thing.

They (BEA) are trying to make us beleive that the only thing that failled were the pitot probes. The pilots "messed up" after the perfect AP disengaged...

Not what I see. I see pilots struggling to fly an uncontrolable airplane, with no attitude indication, at night with no horizon.

I don't believe the pilot would command aft stick knowing his nose was already up. Not even a student pilot would do this.

The aircraft went crazy on them.
Why?

dcpax

27th May 2011, 18:43

Passenger here with a question for the professionals.

The BEA report notes that TOGA thrust and pitch up were selected at least some of the time.

Is it at all possible that this could have resulted from following an Unreliable Airspeed or Stall Warning Procedure that turned out not to be optimal for the circumstances?

I ask because there appear to be several different power and pitch recommendations depending on the phase of flight in which the problem occurs. For example, the unreliable airspeed procedure recommends TOGA and 15 degrees if the problem occurs before reducing thrust, but lower thrust and pitch at higher altitudes.

I am in no way trying to place blame here. I am certain many factors will prove to be involved. I ask because some have suggested that the procedures were confusing and wondered whether they could have led the crew astray. For example, was it clear whether a stall warning should be respected (and that procedure followed) or not?

JCviggen

27th May 2011, 18:49

BEA is clearly "inducing" public opinion to believe the pilot did wrong.

And you know that how? What a nice world it would be if a thread could stick to facts for at least 10 pages.

Lonewolf_50

27th May 2011, 18:54

checkboard, this is intriguing

If the pitot becomes blocked, the dynamic pressure & static pressure at the time of the blockage remains the same. As the aircraft climbs, the sensed static pressure reduces (with the reducing altitude) so the ADU subtracts a smaller static pressure from the pitot total pressure, and thus comes up with an increasing airspeed. The airspeed altimeter is now working something like an altimeter.
From that, here is how it could play out. (Possible, not probable)

pitot blocked -> assume AP off and pilot flying -> (add trigger like possible turb/updraft that creates alt incr?) -> AS apparent incr -> back stick command -> alt incr -> AS apparent increase -> backstick command -> alt incr -> AS apparent incr -> backstick command -> rinse and repeat for x interations -> pilot's cross check on alt and / or att indicators alerts him "whoa, I'm climbing" -> PF reaction seen in change 7000 fpm climb to 700 fpm climb (or was that due to power redux???) -> pilot probably slightly behind aircraft at this point -> stall warning or stall indication-> then loss of stall warning -> AS unreliable for X amount of time -> AS comes back -> stall warning returns -> pilot now likely on back up inst scan or partial panel scan on his own display ...

How often did anyone practice this one in the sim? :confused:

Not sure how well that picture conforms to reality, perhaps not at all. When a pilot gets behind acft, tends not to say much until back with or ahead of aircraft ... though I've heard a few talk themselves back into a scan on hot mic ...

GarageYears

27th May 2011, 18:55

They (BEA) are trying to make us beleive that the only thing that failled were the pitot probes. The pilots "messed up" after the perfect AP disengaged...

Not what I see. I see pilots struggling to fly an uncontrolable airplane, with no attitude indication, at night with no horizon.

I don't believe the pilot would command aft stick knowing his nose was already up. Not even a student pilot would do this.

The aircraft went crazy on them.
Why?

Where did you dig that from?

The FDR and CVR data will eventually come out into the open, so it would be ridiculous to suggest the BEA are reporting anything other than what happened.

At a minimum stick to interpreting what is known, not what you want to believe - those are very different. :ugh:

Graybeard

27th May 2011, 18:56

Lonewolf_50

27th May 2011, 19:00

To clarify ... if I am in RH seat, I move side stick, LH seat side stick does not move. Do I have that right, or wrong? Block diagram (old) suggests to me that would be the case. If the sticks inputs are summed and PNF's stick moves when PF stick deflects, you'd get twice the input you were looking for, right? :confused:

MurphyWasRight

27th May 2011, 19:02

Sapgiola:

It notes no further disagreements between the two recorded speeds for the remainder of the flight; rather, it notes that at various points that "all recorded speeds became invalid" (presumably when they fell below 30KT) or became valid (climbing above 30KT), indicating that they remained consistent with each other.

BEA states that the stall warning system AOA sensor is considered invalid below 60KT.

When stall warning sounds there are 2 actions that can cause the warning to stop:

A: Nose down, gain speed apply power warning stops

B: Nose up, bleed speed to below 60KT warning stops.

Given what has been written about low altitude (approach to?) stall Airbus training (must loose minumum altitude) ending up with option B is not as strange as it might appear on paper.

Especially so since the indicated speeds had very recently been unreliable the stall warning may have been suspect as well.

henra

27th May 2011, 19:05

Quick question: on what basis does the BEA reports that the speed measurement became valid again at 2:12:21 ?

The reply to that question is probably given in that part of the BEA report:

Note: When the measured speeds are below 60 kt, the measured angle of attack values are considered
invalid and are not taken into account by the systems. When they are below 30 kt, the speed values
themselves are considered invalid.

My explanation:
After the AoA exceeded 40° the airstream to the pitot probably totally collapsed/stalled, leading to an IAS below 30kts.
After reducing AoA by the PF the airstream probably partly unstalled and IAS in the pitot exceeded that threshold again.

That is what probably led to the confusing result of pitch-down leading to the stall warning going off again.
IMHO that is not a good solution. AI should seriously rethink that.

The only strange thing is that speed being declared invalid and stall warning stopping suppoesedly happened at the same time as the stall warning should disappear below 60kts and the speed indication below 30kts.

/Edit: parallel post to MurphyWasRight

Checkboard

27th May 2011, 19:06

In all the shaking, rolling, whistles and gongs, could the Pilot Monitoring possibly not realize the PF was holding aft stick? Could the PM, if in the right seat, override or at least neutralize the PF holding aft stick if he did know what the PF was doing?

To clarify ... if I am in RH seat, I move side stick, LH seat side stick does not move. Do I have that right, or wrong? Block diagram (old) suggests to me that would be the case.

That is correct. In the aircraft you cannot see what stick command the other pilot is giving, and the other stick does not move in any way (unless you move it yourself). The only way to cross check is by monitoring the performance of the aircraft.

GarageYears

27th May 2011, 19:08

Upset Recovery Training - Pitch Angle versus Angle of Attack when Thrust-limited | APS Emergency Maneuver Training (http://www.apstraining.com/2009/three-critical-angles/)

Found this explanation, which would seem to be useful to those less well versed with the basic mechanics being described in the thread.

aa73

27th May 2011, 19:12

And this is one of the reasons why Airbus cockpit philosophy has, and will continue, to be subject to criticism from pilots worldwide:

The lack of cues that we are so used to having since day one of pilot training. Yokes (or control sticks) that MOVE when the other pilot moves them. Throttles that MOVE when the engine power increases/decreases. Systems that allow the PILOT to be in control, and not 385 bloody on board computers!!

Not saying that the pilots would have possibly been able to maintain control had this been a Boeing - an airspeed/PFD disagree/unreliable at altitude, in IMC, around Tstorms, at night is a handful no matter what aircraft you are flying. And we have seen two tragedies in the past - Aero Peru 603 and Birgenair 301, where unreliable airspeed led to loss of control in a conventional airliner, the 757.

But the lack of traditional cues in the A330 probably did not help them one bit.

PEI_3721

27th May 2011, 19:13

The aircraft encountered unusual weather conditions (not in a Cb) which resulted in erroneous airspeed.
The crew instruments used this data (thus the abnormal indications), and other systems including the flight control computers, which interpreted some values as invalid (out of tolerance), and thus reverted to an alternative control law.

The FDR recorded the erroneous values of airspeed, which were not the actual aircraft speed, nor necessarily even after the ‘icing’ encounter with apparently ‘real’ airspeed values – ice blockage still clearing etc.
It is assumed that these same erroneous data were displayed on the PFD / ISIS, if at all, due to out of tolerance blanking.

In alternate law, the aircraft commenced a climb; probably a combination of trimmed state – residual A/P trim before disconnect and ‘inadvertent’ crew input – lack of control force feedback.
Question; does the THS automatically follow up a nose up command in alternate law?

In alternate law, there appears to be some high speed protection. Thus even with increasing AOA (and subsequent stall warning), if the ‘sensed’ erroneous airspeed was high ‘over-speed’ (“sharp rise in ‘recorded’ airspeed 215 kts”), then an automatic nose up command would be provided to reduce speed. ( In the small print / notes – alternate law. (http://www.airbusdriver.net/airbus_fltlaws.htm))
Question; does the THS automatically follow up in this condition?

A change in yaw / roll control law and/or loss of yaw damping in abnormal alternate law could have contributed to the apparent roll control activity. Not discounting turbulence.

The aircraft decelerated in the climb, approaching the stall, but not necessarily stalling immediately (note nose down pitch control inputs, 7000 ft/min – 700 ft/min). The crew, without airspeed indication (or at least viable display values), experienced conflicting information; stall warning from the AOA vanes, but nose up movement (and trim?) from a potential over-speed condition (erroneous airspeed – as above).

Nose down control corrective action had minimum effect due to nose up THS position (and TOGA contribution). The aircraft apparently stalled with nose up trim, but subsequently, the crew at times had apparently ‘reasonable’ airspeed indications and no stall warning (AOA inhibited at very low speed).

The aircraft remained stalled, the crew had conflicting / minimum information, the trim remained nose up, and the crew had no control force feedback of the aircraft’s trimmed state.

Question; does the stall recovery action, or any other low speed flight, require the crew to check the trim (THS) position?

Question: at any time in this scenario, did the display of aircraft attitude fail?

Lonewolf_50

27th May 2011, 19:17

aa73, on the other hand, with the system as it is in A330, you don't get the other pilot "riding the controls" on you. :cool: (Haven't we all run into that once or twice in our lives? ;) )

PEI:

You raise an interesting question beyond the three questions in re THS. :

At what point, after the interval (of unknown duration) during which airspeed wasn't reliable -- a period during which FDR recording is of dubious accuracy -- would the airspeed return to reliable state and so that one can again be confident of the FDR's recorded airspeed data?

Aside: It seems that at some point in the descent, the airspeed indication returned to reliable state. It might be dangerous to assume that when the stall indication returned, A/S was once again reliable. What appears to be true is that the airspeed data, erroneous or true, met the threshold of acceptance in the system. That is not necessarily identical to the airspeed measured by the system being what the airframe was experiencing in the airmass.

The crux of some tough analysis BEA will be doing in the near future seems to me to hinge upon the answer to that question.

Checkboard

27th May 2011, 19:19

Question; does the stall recovery action, or any other low speed flight, require the crew to check the trim (THS) position?
Not in any of the aircraft that I have flown - Boeing, Airbus, BAe etc etc.

kilomikedelta

27th May 2011, 19:22

Why would the software maintain the THS at its maximum nose-up position for the last 3.5 minutes of the flight?

max_torque

27th May 2011, 19:23

Firstly, hello! from a newbie. I have been following this thread with interest as a controls engineer, but in the automotive world rather than aero (more 2d than 3d, but many more things to hit.......)

So, please forgive me if this is made up rubbish, but:

assumption 1) no spatial awareness by the pilots (nighttime, possbly in turbulence etc)

assumption 2) they are aware of unreliable airspeed and possibly other instrumentation faults

assumption 3) They have no reference to quickly decide if the stall warning gongs are valid, or just due to unreliable airspeed

assumption 4) they have no way of knowing that in fact, the aircraft has not suffered some major structural or systems failure

assumption 5) Scaning the flight displays, the one number that might stand out, would be the altimeter winding down at a massive rate, and maybe with other signals, such as cabin pressure changes, and possible abnormal wind noise etc (due to the extreme AOA), helping push that "fact" as "True"

assumption 6) Without other stimulus, i could see how one could assume that in order to be loosing 10k f/min altitude, the plane "must" be in a steep dive (negative pitch, nose down) (I'm not sure if the thought of a "deep stall" would have even occured?, esp in an aircraft that is "impossible to stall" (not true obviously, but i can again imagine that after 10k hrs stick time, you might formed an option of the aircrafts correct and accurate control response to most unusual situations

assumption 7) Pilots cannot trust the autopilot with degraded information (i.e. the phantom steep dive "could" have been caused by the automatic flight controls pushing the nose down to maintain speed, except the pilots know they can't trust that speed, so by default, the actions of the autopilot might not be trusted either)

I these cases (and i admit it is a complex series of scenarios) then throttling back, and pulling "up" could possibly seem like a valid response. (especially if the lack of pitch authority might have seemed like the result of a mechanical systems failure (instead of the deep stall) ??

Unfortunately, in the limited time availible, they were unable to work out what was actually happening to the aircraft. (it would be a brave move to go full throttle and stick forward if you genuinely "believed" you were in a steep fast dive would it not)

Appologies for the lack of technical jargon, but if you assume that the pilots were trying to "pull up" the entire time, then something must have caused them to carry out that action?

Khashoggi

27th May 2011, 19:25

henra

27th May 2011, 19:26

Not what I see. I see pilots struggling to fly an uncontrolable airplane, with no attitude indication, at night with no horizon.

Ohhh, nooooo!
Not again!!

Where oh where does this rubbish with the lost horizon come from????

It has been mentioned time and again that there is not the slightest indication that happened.
And the fact that they always managed to keep the wings reasonably level supports the fact they had attitude information.
In IMC without attitude information your landing gear will point straight up at one point or the other.

deSitter

27th May 2011, 19:26

How would the airplane behave if its fin came off in the zoom? Wouldn't it start Dutch rolling?

Ashling

27th May 2011, 19:32

PEI 3721

There is no requirement to check the THS.

Airbus have changed their stall recovery procedures to seperate the nose down input from the power up. You now only smoothly increase power when the aircraft is unstalled so the nose down input comes first. There is a note that you may even have to reduce power for this nose down input to be effective.

It also mentions that, below 20,000', if your clean deploy Flap 1.

One thought. As they lost the stall warnings due to low IAS, 30/60 kts when the aircraft stops them, maybe they got confused when they lowered the nose and the warnings returned when the speed increased. I.E. back pressure and the warnings go forward pressure and they return. As I said just a thought and I may have misread the report.

takata

27th May 2011, 19:33

Hi grity,
pull zoomclimb stall chrash....
A340 TC-JDN:
the first elevator moving to +4 deg, 4 sec after the autopilot self-disengaged as the aircraft exceeded the speed limit, +4 deg only for 2 sec was enough to start the zoom climb for the A340 TC-JDN (en-route from Istanbul to New York) and was defenitiv not triggerd by the sidestick of the pilot (the first sidestick pitch was 20 sec later).....
AP/ATHR kicking off in this case was obviously the direct consequence of an high speed prot, but without switching her to ALT. Those few seconds lagg seems to be due to turbulence action at the same point. But those A340 pilots applied also full thrust when autothrust disengaged, right after its auto-reduction from turbulences... and then, she goes from less than 70% N1 to 100%. Guess what, she climbed well! (with four engines TOGA). But, here also, nobody applied any stick imput until the top of it (about 30 seconds without a single imput).

AF447:
2 h 10 min 05 autopilot self-disengaged
2 h 10 min 15 "alternate law"
what is with the g-protection in the time between ???
Pilots are quoted talking about Alternate law less than 10 sec. after AP & ATHR kicked off... where is the lagg?
This is not like recorded DFDR data that will confirm that it was simultaneous with those airspeed rejection (and, at least, the BEA should know it without any doubt at this point).

vapilot2004

27th May 2011, 19:33

Per BEA initial report - in structured form for clarity.

01:35:15 Radio:TO ATLANTICO - @ INTOL
01:35:46 Radio:FROM ATLANTICO - Maintain FL350...eta TASIL?
01:55:xx Captain awakens #2 co-pilot
01:59:32-02:01:46 Turbulence & WX Briefing
02:06:04 PF notifies CC turbulence ahead
02:08:07 PNF "Go left a little"

11 Seconds elapsed:
=================================================
02:10:05
AP then AT disconnects.
PF I have the controls"
A/C rolls to right.
PF inputs left and nose-up
Stall warning 2X
A/S "sharp fall" from 275 > 60kts Captain's PFD
Same indications on ISIS "few moments later"
=================================================

34 Seconds elapsed:
=================================================
02:10:16
PNF "...we've lost the speeds" "...alternate law"
A/C attitude goes beyond 10 pitch up
A/C V/S indicates positive rate of climb
PF nose down inputs. L/R inputs.
V/S goes from 7000 fpm to 700 fpm
Roll angle recorded between 12 deg R and 10 deg L
Captain's A/S rises sharply to 215kts
ALT 37,500.
AOA 4 degrees
=================================================

02:10:50 PNF calls Captain to cockpit

49 Seconds elapsed:
=================================================
2:10:51
Stall warning
TOGA
AOA 6 degrees
PF "maintained" pitch up commands
AOA 6 degrees and rising
THS 3 degrees to 13 degrees nose-up in ~1 minute and remained there
"15 seconds later" - ISIS records sharp increase to 185kts and agrees with captain's A/S afterwards
PF nose-up "inputs" continues
ALT 38,000
AOA 16 degrees
=================================================

22 Seconds elapsed:
=================================================
02:11:40
Captain returns
Speeds invalid
Stall warning stops
ALT 35,000
AOA > 40 degrees
V/S ~ -10,000 fpm
Pitch maxed @ 15 degrees
N1 ~100%
Roll angles approached 40 degrees
PF input: full left and full nose-up 30 seconds
=================================================

15 Seconds elapsed:
=================================================
02:12:02
PF "I don't have any more indications"
PNF "We have no valid indications"
TLA @ idle
N1 55%
=================================================

75 Seconds elapsed:
=================================================
02:12:17 (approx from BEA)
PF commanded pitch down
AOA decreased
Stall warning (upon valid speed inputs-BEA)
=================================================

56 Seconds elapsed:
=================================================
02:13:32
PF "...arrive at level 100" (10,000ft)
02:13:47 (approx from BEA)
Both sidesticks recieved "inputs"
AOA "when valid" remained above 35 degrees
=================================================

02:14:28 End of recording
V/S -10,912fpm
GS 107kts
Pitch 16.2 degrees nose-up
Roll angle 5.3 degrees L
Heading 270 degrees

henra

27th May 2011, 19:35

Why would the software maintain the THS at its maximum nose-up position for the last 3.5 minutes of the flight?

That would be the natural reaction of auto-trim to a nose-up command by the PF.
That seems to match the description by BEA that stick-command was predominantly nose-up. Auto-trim wiould try to follow suit in order to allow centralising of stick.

Had the stick command been constantly nose-down, Auto-trim would have reduced THS angle consequently.
Sadly that continuous Nose-down command was apparently not apllied.

CVR might have some cues as to the why.

GarageYears

27th May 2011, 19:36

I would like to know what was going on between 2 h 12 min 02 - 2 h 13 min 32.

The BEA release is quite devoid of any facts for that long 90 seconds or so where AF447 went from FL3xx to FL100.

No cvr information, no sidestick input information, no flight parameter information nothing

Vertical speed -10,000 ft/min...

You're not quite right though, there is some info about that time period:

At around 2 h 11 min 40 <snip>
The altitude was then about 35,000 ft, the angle of attack exceeded 40 degrees and the vertical speed was about -10,000 ft/min. The airplane’s pitch attitude did not exceed 15 degrees and the engines’ N1’s were close to 100%. The airplane was subject to roll oscillations that sometimes reached 40 degrees. The PF made an input on the sidestick to the left and nose-up stops, which lasted about 30 seconds.
At 2 h 12 min 02, the PF said "I don’t have any more indications", and the PNF said "we have no valid indications". At that moment, the thrust levers were in the IDLE detent and the engines’ N1’s were at 55%. Around fifteen seconds later, the PF made pitch-down inputs. In the following moments, the angle of attack decreased, the speeds became valid again and the stall warning sounded again.
At 2 h 13 min 32, the PF said "we’re going to arrive at level one hundred". About fifteen seconds later, simultaneous inputs by both pilots on the sidesticks were recorded and the PF said "go ahead you have the controls".

The initial altitude (FL350) and 2 minutes @ 10K/min gets you close to FL100

Ashling

27th May 2011, 19:48

Our training is always to honour the stall warning with unreliable airspeed indications as it is based on AoA.

Rob21

27th May 2011, 19:53

Henra, I am talking about pitch, not only roll.
As far as I know, an attitude indicator indicates (or should) pitch and roll.

This is not the primary instrument you look when flying instruments by hand?
AI, airspeed, altitude, climb, AI, lost airspeed, altitude, climb, AI, lost airspeed, altitude, climb, AI, AI, AI (attitude indicator).

Ok, I'm sorry.
BEA did not say clearly that the pilots did wrong. But the lack of FACTS on this report, like the absence of dialogs between 2:10:16 until 2:12:02 is kind of strange. Why only open to the general public the parts that may incriminate the pilots?

P.S. Where is PJ?

kilomikedelta

27th May 2011, 19:58

DozyWannabe

27th May 2011, 20:00

@Rob21 - they've released this note as a counterweight to apparent press leaks. It's not even an interim report, just a collection of facts they've managed to prove so far. The reason there isn't more *right this second* is because that is literally all they have.

As was stated earlier, PJ2 was in the air - odds-on he'll be back sometime in the next 24-48 hours.

GarageYears

27th May 2011, 20:10

I guess I have difficulty understanding why the software would find 3.5 minutes of maximum nose-up for the THS perfectly acceptable while all other inputs were going to hell in a hand basket. Was the processor too busy with other interrupts to check the THS or was this combination of input anomalies not part of the decision tree? :eek: So presumably you'd have preferred the software to decide this wasn't allowed and roll off the trim - despite the input from the pilot demanding nose up? You really can't have it both ways now, can you? On the one hand half the crowd here are screaming for the head(s) of the software programing team for removing so much 'authority' from the crew, and here we have a request for the software to "decide" that the crew demand was unhealthy - even if it *was* unhealthy, are you sure you want additional laws and protections? Given that the aircraft has an auto-trim system, it appears to have been operating as expected, given the commands from the crew?

Chris Scott

27th May 2011, 20:11

Presumably, this afternoon, you guys have been discussing the BEA "Update on Investigation". For what it's worth, I've been in voluntary quarantine since just before it came out. As soon as it appeared on the BEA website, I cut myself off completely from PPRuNe, all other media, and any contact with aviation-minded people; and that situation still pertains as I write. I deliberately progressed very slowly through the paper, taking notes and some very long stretch-breaks. There were a number of surprises. Since then, I've been mulling it over, and drafting this.

So what you are about (not/partially) to read are questions and comments arising from my own interpretation of the BEA account, uncoloured since its publication by the wisdom of third parties such as your good selves, and burdened by my limitations. No doubt, I'll have missed many points that, to you, are already "old hat". Nevertheless, here are my thoughts.

QUESTIONS & COMMENTS ARISING

(1) Why no R/H ASI parameter in DFDR? Would it be in QAR? The two recorded ASIs use pitots on the L/H side. (The PF was in R/H seat, presumably.)

(2) "From" 02:10:05, at what stage did the L/H (PNF's) ASI and ISIS ASI readings fall sharply from 275 to 60? After the two stall warnings, or before?

(3) Why did PF allow/encourage/command A/C to climb 3000ft and up to +16 pitch after receipt of stall warnings? One of the pilots had selected TOGA thrust, presumably in response to them.

(4) Why did/does the FBW system in Pitch-Alternate Law continue nose-up trimming of the THS past the stall-warning level of AoA?

(5) As the THS remained at 13 deg UP throughout the descent, is there any possibility that the THS motor stalled during down-elevator inputs? (The BEA "finding" is that "the inputs by the PF were mainly nose-up"[my emphasis of "mainly"].

(6) At about 02:11:04 (at the apogee of FL380), the L/H ASI and ISIS ASI showed 185kts. Was the R/H ASI showing a higher value – possibly due to a blocked drain-hole PLUS the climb of 3000ft?

(7) Quote from BEA: "At around 2 h 11 min 40 [6], the Captain re-entered the cockpit. During the following seconds, all of the recorded speeds became invalid and the stall warning stopped.
Note: When the measured speeds are below 60 kt, the measured angle of attack values are considered invalid and are not taken into account by the systems. When they are below 30 kt, the speed values themselves are considered invalid."
Why are the AoA values considered invalid below 60kts IAS when the A/C is not on the ground?
(Looks at this stage that: first the PF has mis-controlled the A/C into an inappropriate, steep climb, despite a stall warning; but now the systems are failing to inform him that he has stalled the A/C.)

(8) In the descent, why did neither pilot seem to recognise that the A/C was even in a stall, let alone a deep/super stall? Why was idle thrust selected?

(9) At 02:12:02, several thousand feet below FL350, does the PF's comment that he had no more indications indicate that his ASI reading had only just dropped to a value that he no longer believed? Is it remotely possible that, perhaps unlike the others, the drain hole of his pitot tube had blocked at the same time as the intake? During the initial climb, could he have thought that his ASI was still valid, and that he was trying to avoid an overspeed? [See (6) & (7), above.]

[SIZE=2][FONT=Verdana]SUMMARY OF KNOWN ENERGY ASPECTS (from BEA text and diagram)

02:08:07 – 02:10:05 (~15nm)
FL350 (about 37000ft amsl). Average GS ~450kts.

02:10:05 [FONT=Verdana]– 02:10:51 (~5.4nm)
FL350, then climbing steeply. Average GS ~420kts.

02:10:51 – 02:11:40 (~4.3nm)
Climbing to apogee of FL380 (about 40000ft amsl), then descending steeply through FL350, all with TOGA thrust. Average GS ~320kts.

02:11:40 – 02:14:28 (Tear-drop track-distance unknown)
Descent from (passing) FL350 (about 37000ft amsl) to sea-level, Thrust reduced from TOGA to ~IDLE by 02:12:02.

Footnote
During the couple of hours I spent reading this short report (including going back and forth), making notes, and referring to system diagrams, I made several unnecessary errors with my printer. Some sheets had to be discarded and fresh attempts made. The flight crew had less than three minutes to diagnose the problem, take action, and recover from any mistakes.

bearfoil

27th May 2011, 20:12

Garage Years

Quote:
At around 2 h 11 min 40 <snip>
The altitude was then about 35,000 ft, the angle of attack exceeded 40 degrees and the vertical speed was about -10,000 ft/min. The airplane’s pitch attitude did not exceed 15 degrees and the engines’ N1’s were close to 100%. The airplane was subject to roll oscillations that sometimes reached 40 degrees. The PF made an input on the sidestick to the left and nose-up stops, which lasted about 30 seconds.
At 2 h 12 min 02, the PF said "I don’t have any more indications", and the PNF said "we have no valid indications". At that moment, the thrust levers were in the IDLE detent and the engines’ N1’s were at 55%. Around fifteen seconds later, the PF made pitch-down inputs. In the following moments, the angle of attack decreased, the speeds became valid again and the stall warning sounded again.
At 2 h 13 min 32, the PF said "we’re going to arrive at level one hundred". About fifteen seconds later, simultaneous inputs by both pilots on the sidesticks were recorded and the PF said "go ahead you have the controls".
The initial altitude (FL350) and 2 minutes @ 10K/min gets you close to FL100

************************************************************ **

(My new bold) Between 2h13m32s and eor, 2h14m38s, the a/c descended ~two miles.

Just before, with ND, the AoA lessened, the pitots stopped blanking, and an a/s was presented. Still at Stall, the vertical speed was 107 knots, square with horizontal, also 107 knots. This is an AoD of 45 degrees, with full power seemingly available. So at FL100, and these numbers......If this is true, they did not "miss" by much, a recovery...?

henra

27th May 2011, 20:13

Henra, I am talking about pitch, not only roll.
As far as I know, an attitude indicator indicates (or should) pitch and roll.

Hmmm, maybe I misinterpreted your post. If that is the case, I do apologise.
I understood your post pointing to a non-functioning of attitude indication.
And that is where I get a bit sick and tired reading it time and again, although there is not the slightest indication that IR and thereby attitude information was lost.

And the description of the attitude during descent aboslutely points in the same direction. The attitude was always rather level. In all axis btw.
Ok there was a 40° bank at one point which the pilots tried to correct immediately. This confirms they had suitable attitude information.

Aircraft in IMC where Attitude information is lost arrive in a totally different way.
Absolutely totally different....

My feeling is the thing was seemingly somehow under control, seemingly missing only one piece in the puzzle, the trajectory.
I'm afraid these poor guys never really figured out their effective flight path and speed through the air. Being lost in warnings, seemingly benign attitude and physical sensations and strange instrument indications.

Pawlos

27th May 2011, 20:17

Hello all PPRuNers, I have been following this thread with great interest since tragic loss of AF447 two years ago (and other PPRuNe threads also :cool:). I am not a pilot nor I do have any connection to aviation industry whatsoever, except great passion for airplanes and flying but only as SLF. I must thank to you all as over time I did learn a lot from you and I am very grateful for that.

As I read BEA I was puzzled by THS being in almost full up position for about 3,5 minutes and not changing its position although PF made pitch-down input. Unfortunately report doesn't mention for how long pitch-down input lasted. I took my copy of A330 FCOM (not current one) and found out, please correct if I am wrong, that flight control law could have changed from ALT 2 to Abnormal attitude as angle of attack was greater than 30 deg causing no auto trim available until angle of attack decreases below 30 deg, which according to BEA report did not happen.

Is there any indication in the cockpit when this happens?

CONF iture

27th May 2011, 20:18

1- AoA at 6 degrees and increasing + STALL warning
How is it possible the THS was still trimming up ?
What's wrong here ?

2- As both recorded IAS became consistent again, did the FDs reappear ?

3- BEA, give us all but not that so little !

GarageYears

27th May 2011, 20:19

Chris:

(4) Why did/does the FBW system in Pitch-Alternate Law continue nose-up trimming of the THS past the stall-warning level of AoA?I believe Pitch Attitude protections are lost in Alt1 law and I remember a note about stall protections being 'override-able' in Alt law.

See: A340 - A330 Control: Flight & Laws (http://countjustonce.com/a330/a330-flight-laws.html)

Mr Optimistic

27th May 2011, 20:21

CogSim

27th May 2011, 20:25

(9) At 02:12:02, several thousand feet below FL350, does the PF's comment that he had no more indications indicate that his ASI reading had only just dropped to a value that he no longer believed? Is it remotely possible that, perhaps unlike the others, the drain hole of his pitot tube had blocked at the same time as the intake? During the initial climb, could he have thought that his ASI was still valid, and that he was trying to avoid an overspeed? [See (6) & (7), above.]

02:10:16
PNF "...we've lost the speeds" "...alternate law"

No acknowledgement from PF.

02:12:02
PF "I don't have any more indications"
PNF "We have no valid indications"

First sign of PF identifying problem with indications.
PNF acknowledges.

paull

27th May 2011, 20:28

Surely the point in having more than one crew member is so that the PNF can say either:
1- What the fxxx are you doing or;
2- What the fxxx is IT doing depending on whether your are requesting the right or wrong inputs and getting the right or wrong responses from the system.

So, in the Bus, can the PNF or jump seat occupant see what the PFcommanded with the stick? If not, do not expect them to be able to call you on it.

Could someone who flies this confirm the simple question
"Can you tell what the other is doing?"

Frankly I am impressed with the self restraint of a PNF who waited 27k feet before deciding to put hands on. I stop being a team player way before that:confused:

Mr Optimistic

27th May 2011, 20:29

kilomikedelta

27th May 2011, 20:29

GarageYears; Perhaps you misunderstood my question. I was inquiring whether maximum nose-up for 3.5 minutes was considered acceptable by the software and whether or not the PF should have been made aware of that fact (although he probably had more than enough software output to handle). The pilot should be making the final decision but when a computer puts a flight control in an extreme position for a considerable time (not directly by his command) it might be helpful for him to know what he's fighting.

morbos

27th May 2011, 20:34

The data from the DFDR will undoubtedly find its way to a sim in the future. What might the outcome had been if the PF had not touched the sidestick at all and allowed the a/c to ride out the uas. Presumably in that night, the movement of the trim wheel and subsequent lack of movement back would not have been a high priority.

grizzled

27th May 2011, 20:35

Mr. Optimistic...

Is there some dialogue we are missing ?

The answer is yes. Lots. Without any doubt.

Yellow Pen

27th May 2011, 20:37

KMD - Do you know of any aircraft in which a computer will override the pilots trim command input in that manner? The computer did not put the flight control in an extreme position, the pilot put the THS there through the continued application of nose up stick. The auto-trim did exactly what it was designed to do and which any Airbus pilot would be expecting.

CogSim

27th May 2011, 20:37

That's a very calm sounding FD considering what's going on. The report of the CPT 'shouting' sounds more like it. Think the FD communications may be a bit too dramatic to release without other context and explanation.

That begs the question what were the criteria used in the editing process. Clearly they must believe whatever was released presents the event without too much distortion.

GarageYears

27th May 2011, 20:37

kmd:

Yes, I guess I was a little spicy there - apologies for barking! I understand where you're coming from now - I suppose the question is "how does the crew know AT ALL where the THS is pointing?". And as a supplemental to that, should there be some (obvious) indication when the THS falls outside of some nominally "normal" range? Presumably 13 degrees up is not a common demand....?

DozyWannabe

27th May 2011, 20:39

About fifteen seconds later, simultaneous inputs by both pilots on the sidesticks were recorded and the PF said "go ahead you have the controls".

Why would he relinquish control ? Is there some dialogue we are missing ?

The thought that occurs to me - and take it as it is, a wild-arsed guess - is whether the Captain managed to regain his seat at that point. We know there were some large rolls on the way down, but whether that would have presented a major obstacle to him doing so either because of excess deck angle or G is unknown at this point.

It also rather scotches the idea that the PF was unaware of any control inputs made on the other side, because whatever he saw, whether it was the occupant of the LHS moving the stick or the "DUAL INPUT" warning, he was immediately aware that the stick was being moved and relinquished control via the switch at the base of the LHS sidestick, just as protocol states.

Capi_Cafre'

27th May 2011, 20:42

The level of arcane systems knowledge this crew needed to dig up in the heat of battle in order to solve the riddle it was faced with is frightening. Having the return of the stall warning triggered by an appropriate control input would have been fiendishly difficult to comprehend under the circumstances. So sad that at this late date new knowledge of our craft is still being written in blood.

TLB

27th May 2011, 20:46

Folks,

There has been much discussion about what was released today. What is more important, IMHO, is what has NOT been released today. Specifically, the full record of the CVR. There is no question in my mind that there is a whole bunch of communications between the PF, PNF and Capt (when he got back to the flight deck) during that 4 minutes of total confusion.

When I read this report today (and as I reread it again), I did not think for a moment that it includes anywhere near the total cockpit dialog.

cuddieheadrigg

27th May 2011, 20:47

More fascinating than ever. There have been some guesses here that are now proven to be very close to what actually happened.

Without attempting or wishing to accuse/blame the pilots - I wonder if what we have here is a tragic misinterpretation of events?

The Captain returned to the cockpit, but did not feel it vital to assume control - this surely suggests that as far as they believed, the situation was under control.?

Could the pitch up commands have SIMPLY been in response to a belief that the descent was due to being nose down - as others have said - to pitch up BELIEVING that the aircraft is stalled is surely the last thing that you would wish to do - conversely if you believe airspeed is OK then you would naturally 'pull up' to arrest descent.

This theory does not take account of what, if any indications the pilots were (or not) receiving - they either beieved the stall warnings (when they came) or did not. It is hard to believe that any pilot would make a control input (even if it is wrong) UNLESS they believed it was the correct course of action.

So the nub is: simple mistake?

vapilot2004

27th May 2011, 20:48

Mr. Optimistic...

Is there some dialogue we are missing ?

The answer is yes. Lots. Without any doubt.

Dialog and data missing in copious amounts.

Owing to the lack of meaningful disclosure by the BEA, speculation is fairly useless at this time.

Interflug

27th May 2011, 20:53

(3) Why did PF allow/encourage/command A/C to climb 3000ft and up to +16 pitch after receipt of stall warnings? One of the pilots had selected TOGA thrust, presumably in response to them.

Interflug A310 Incident approaching Moscow in 1991 comes to mind.

ASN Aircraft accident Airbus A310-304 D-AOAC Moskva (http://aviation-safety.net/database/record.php?id=19910211-0)

henra

27th May 2011, 20:59

1- AoA at 6 degrees and increasing + STALL warning
How is it possible the THS was still trimming up ?
What's wrong here ?

From GarageYears Link:

In Alternate Law, Valpha prot and Valpha max are replaced by VSW. Note that VSW is a stall speed and so is EAS sensitive unlike Valpha prot/max. Stick free aircraft aims to control alpha by limited authority stability order to less than VSW. If the stick input forces VSW exceedance, alpha floor activation of TOGA occurs. However, it is possible to stall the aircraft as the protection may be overridden.

and

Protections are as in Alternate 1 except that there is no bank angle protection in ALT 2 and in the case of failure of 2 ADRs, no VSW prot

So with the double ADR fail even Vsw prot would be lost thereby effectively losing basically all stall protection.

It has to be noted that Alt1+2 are Load Factor Demand laws with Auto trim.
So pulling Nose- Up sidestick command would cause Auto trim to trim Nose-up even above the stall.
I do not see anything indicating in the description of the laws preventing that in the situation AF was in.

Khashoggi

27th May 2011, 21:01

Owing to the lack of meaningful disclosure by the BEA, speculation is fairly useless at this time.

+1

:ok:

It sure is interesting how the detail drops off after AP/AT disconnect and the zoom climb.

DozyWannabe

27th May 2011, 21:05

Interflug:

Different type (advanced autopilot, but no envelope protection), different set of inputs entirely.

If I remember correctly, what happened there was that immediately after selecting TOGA, the PF in that case felt that the climb gradient was too steep, so he instinctively pushed forward on the yoke to get the nose down. This led the autopilot to apply up elevator and trim to counteract the yoke force. The PF then re-applied forward yoke pressure and the AP responded again with opposite (nose-up) inputs. The oscillations this caused led to the aircraft pitching up almost to the vertical before control was regained.

What we have in this case appears to be deliberate nose-up commands applied by the PF in response to the loss of speed information - for reasons as yet uncertain. There was no autopilot to counteract those inputs as it had tripped out due to loss of speed data. Everything else that happened on the flight deck at this point is unknown, so all we know for certain is the trajectory the aircraft took. It's obvious that the BEA want to get this right rather than rush a report, so I for one don't see why we shouldn't let them get on with it - those that are complaining about lack of information at this juncture seem to be almost childishly impatient.

jcjeant

27th May 2011, 21:06

Hi,

So presumably you'd have preferred the software to decide this wasn't allowed and roll off the trim - despite the input from the pilot demanding nose up? You really can't have it both ways now, can you? On the one hand half the crowd here are screaming for the head(s) of the software programing team for removing so much 'authority' from the crew, and here we have a request for the software to "decide" that the crew demand was unhealthy - even if it *was* unhealthy, are you sure you want additional laws and protections? Given that the aircraft has an auto-trim system, it appears to have been operating as expected, given the commands from the crew?

I understand that
Nevertheless the question was not about the first input .. but about a timing (3.5 minute)
3.5 minutes at 13° climb (about the max obtainable) position for the horizontal stabilizer with many other parameters complety jammed
How manage this by software (protection) is another story
Just a tought.

good spark

27th May 2011, 21:07

hello gents
a small question, does anyone know if any of the crew on this unfortunate had any time on older generation types? i dont want to start some sort of pissing contest but i do think the type of experience is relevent- lets not forget 5000hrs on one type is exactly that but 5k on 4 or 5 types must be worth 10k on on type yes or no?

gs

Rob21

27th May 2011, 21:09

Henra, maybe I got it wrong but I understood from the report that the aircraft was in a pitch up attitude of 16º when it reached 38.000 ft.

Level attitude would not be something around 4 or 5º ?

I understand that a good attitude indicator should indicate pitch variations of, at least, one degree. Sixteen degrees on the A-330 attitude indidator is not very noticeable?

kilomikedelta

27th May 2011, 21:10

MurphyWasRight

27th May 2011, 21:11

CapCafre:
The level of arcane systems knowledge this crew needed to dig up in the heat of battle in order to solve the riddle it was faced with is frightening. Having the return of the stall warning triggered by an appropriate control input would have been fiendishly difficult to comprehend under the circumstances. So sad that at this late date new knowledge of our craft is still being written in blood.

A simple question for Airbus pilots:

Until today were you aware that the stall warning could go away if a stall developed to the point that indicated airspeed dropped below 60KT?

And that it would return as you recovered?

JPI33600

27th May 2011, 21:14

Peter H
As someone with no aviation experience, can I ask how much a timely warning of [the likelihood of] UAS would have helped the pilots?
If it would have helped, perhaps more attention should be given to monitoring systems.
If cars can warn of hazardous road-temperatures, might not planes try to warn of ice-crystals. As far as I can say, the requirements are very different: your car warns you about the external temperature, which would be useless for an airplane flying in an air mass at -50°C and below. The problem seems to be associated with ice particles having a very specific size; if they are smaller, they will be easily processed by the pitot heating system; if they are too big, they will not enter the pitot tube. Thus a suitable warning device would have to accurately predict ice crystals size in a timely fashion, say 20 seconds. With a plane flying at 300 m/s, that requires being able to monitor the size of ice particles having a diameter of a few millimeters as far as 6 km away, possibly through several layers of ice crystals having a different size. I can't imagine a way to do that with the technology currently available.

AlphaZuluRomeo

27th May 2011, 21:18

@ studi (#117 page 6) : Thanks for answering :)

I was thinking of speed stability in Alternate Law.
OK so this one, I presume (*) :
"A nose up command is introduced any time the airplane exceeds VMO/MMO to keep the speed from increasing further, which CAN be overridden by the sidestick." (source (http://www.airbusdriver.net/airbus_fltlaws.htm))
My problem with accepting this hypothesis is that as far as we know, the indicated speed was not exceeding VMO/MMO. In fact, it was the contrary if the last BEA note is to be trusted.

(*) another (low) speed stability function exists, but this one commands a nose down input, not the case here.

If we look at it from the other side, IF the plane was functioning normally, we would have a case of 3 pilots being TOTALLY inapt for their job, which I honestly can not even believe for Air France being the case.
Uh oh, I don't agree. They may have made a (fatal) mistake, i.e. not recognizing the situation. Are they to be called "totally inapt" for that? I'm not sure.
What were the conditions ? Why did they seem to loose all confidence in all of their "indications" (after having firstly correctly recognized a problem with the -sole- speed) ?
I don't know and some parts of the story are still missing.

MurphyWasRight's post #119 is for example very interesting with the stall warning on/off/on and the possible consequences of that on the PF's mind.

Yellow Pen

27th May 2011, 21:26

KMD - Apologies, I thought you were a pilot. I would speculate the software allows the THS to remain at it's upper limit because thats precisely what the pilot was commanding. There's got to be an approved upper limit to THS position somewhere, and if it's approved then having the THS at that position is not necessarily unsafe. There are numerous failsafes to counteract THS runaway or uncommanded movement, but if none of these failsafe triggers have been activated and the pilot is commanding nose up THS movement then as far as the machine is concerned it's situation normal. When an Airbus enters alternate law it's because the computers don't have sufficient information to reliably keep the aircraft within the normal flight envelope protections. I would suggest the more sensible option in that situation is to err towards greater control authority for the pilot.

augustusjeremyreborn

27th May 2011, 21:27

Would that stall happen (the plane having no aft-forward velocity before a "natural nose down") with a "more forward" CG ?

I mean: Wouldn´t the plane go nose down before the a/c aft-forward speed got so low even with the HS trying a "pitch-up" if the cg was properly set ?

Checkboard

27th May 2011, 21:29

China Airlines Airbus Industrie A300B4-622R, Nagoya Airport April 26, 1994
Bournemouth Airport, U.K. - A Thomsonfly (now known as Thomson Airways) Boeing 737-300 with 132 passengers on a flight from Faro in Portugal to Bournemouth in England, September 2007
An Air New Zealand Airbus A320-200, registration D-AXLA (ZK-OJL), test flight GXL888T from Perpignan (France), Nov 2008
Turkish Airlines Boeing 737-800 at Amsterdam's Schiphol Airport, February 2009

etc etc All aircraft which crashed (or very nearly so) because the pilots failed to realise that the trim had been set full nose up, and couldn't understand why they lost pitch control.

Generally airline stall recovery training doesn't use or show a set up with the trim set this way - it is usually an artificial "OK, we are going to stall ... disengage the autopilot and slow the aircraft down, and stop trimming to allow for the recovery..."

JPI33600

27th May 2011, 21:30

Khashoggi
It sure is interesting how the detail drops off after AP/AT disconnect and the zoom climb. Don't forget there is a judicial enquiry engaged: maybe communicating too much info about the cockpit conversations could be considered a violation de l'instruction (early disclosure of judicial information) ?

Swedishflyingkiwi

27th May 2011, 21:34

Who knows what was happening in the office (cockpit). Unreliable airspeed indications, stall warnings and watching the altimeter winding down.

My feel is the overwhelming urge to pull up to address the reality of loss of height.... being led to believe they must be in a dive, not dropping like a brick due to being held in a stall.

No time, confused data and three pilots possibly adding to the confusion.
We grieve for all on board.:{

DozyWannabe

27th May 2011, 21:36

Yellow Pen; I'm not in the aviation software business so I'm not familiar with the details of any particular aircraft system. I have some experience in software writing and crisis management. I'm just asking why the software would accept the THS being at its limit for so long.

KMD - I suggest you read through the previous thread on the subject, where a lot of people with experience of how real-time software works patiently explained what the software does and does not do.

The software is designed to give the pilot exactly what he or she asks for, except in certain circumstances where a dangerous situation is developing (extreme nose-up command without attendant power increase, or an incipient spiral dive). It does this by monitoring several parameters at once, and requires that those parameters be valid and the protections enabled in order to do so.

At no point will the computer override a pilot's command completely, it will simply mitigate the response of the flight surfaces to carry out what the pilot is asking of it as safely as possible. In the case of an incipient stall due to full stick-back, it will increase thrust to maximum power. In the case of an incipient spiral dive it will limit the angle of bank to an absolute maximum of 67 degrees.

The initial situation we have here is the flight protections falling back to Alternate Law in response to the loss of speed data. This removes some protections and at this point it becomes easier for a pilot to inadvertently put the aircraft into a dangeous attitude. The PF is appearing to command vigorous nose-up by pulling back on the stick in response to these indications, but because of the degraded control laws and lack of speed data, the computer is unable to determine whether such a command is in fact endangering the aircraft. The design pattern suggests that at this point the computer is not best-placed to determine whether the demand is unreasonable or not, and the control logic simply follows the pilot's commands without intervening.

Based on the information we have so far (which is admitedly sparse), it appears that the system behaved exactly as it was designed, and the crew was faced with an unenviable situation - unreliable instrument readings at cruise altitude in unfavourable weather in the middle of the night. Even with everything working I'm pretty sure that many pilots have many places they'd rather be.

JPI33600

27th May 2011, 21:39

kilomikedelta
I'm just asking why the software would accept the THS being at its limit for so long. I suppose we can imagine a situation where a deteriorated airframe (e.g. DHL plane hit by missile) would induce a pitch down moment that would need to be compensated by THS being permanently held close to or at its pitch-up limit.

kilomikedelta

27th May 2011, 21:40

Yellow Pen; We understand each other then. I suppose it would have been helpful to notify the pilot that all his pitch control efforts via the sidestick were for naught and that he should have considered using the verboten trim wheel.

HarryMann

27th May 2011, 21:43

And that it would return as you recovered?

..implying that you might have 'done' the 'wrong' thing :ugh:

WIH is wrong with displaying AoA; WHY this 'committee' level decison making on filtering vital flight data using simplistic 'buzzer of doom' logic (NB: it used to be a yaw string, or a simple visible vane, on a hang-glider it's 'wind on the face', in many a/c its stick force... )

In an A330, it's, er... a heavily filtered 0-1 logic state :hmm:

Checkboard

27th May 2011, 21:44

I suppose it would have been helpful to notify the pilot that all his pitch control efforts via the sidestick were for naught and that he should have considered using the verboten trim wheel.

That wouldn't help - as the pilot's sidestick efforts were working perfectly. If he thought of the pitch trim wheel, he would have probably attempted to trim nose up, given that was the input he was applying to the stick.

deSitter

27th May 2011, 21:45

No time, confused data and three pilots possibly adding to the confusion.
We grieve for all on board.http://images.ibsrv.net/ibsrv/res/src:www.pprune.org/get/images/smilies/boohoo.gif

Oh boy this is an understatement. There aren't many chapters in aviation history as sad as this one. One finds oneself grasping at straws to understand it :(

Old Engineer

27th May 2011, 21:47

@MurphyWasRight

A simple question for Airbus pilots:

Until today were you aware that the stall warning could go away if a stall developed to the point that indicated airspeed dropped below 60KT?

And that it would return as you recovered? 27th May 2011 16:10

It would be interesting to know if PJ2 is aware of this behavior of the stall warning. If he is not, that would seem to confirm it as being an exceedingly arcane behavior. OTOH, AB has said the FBW was working as intended. Was it so, or is this behavior an oversight... an unintended consequence of a situation not considered? --OE

Ashling

27th May 2011, 21:48

Murphy,

No I wasn't, maybe vaguely somewhere in the grey matter, and even if I was could I recall that info under pressure?

I make the same point that you are re the confusion that might have caused in post 159. Its a thought.

That said if they had responded correctly to the initial stall warning would they have found themselves in that situationand having found themselves there they had other indications, attitude, to help with the diagnosis. A pilot well versed in UPs/Upsets should have been able to recover the aircraft safely.

henra

27th May 2011, 21:49

Henra, maybe I got it wrong but I understood from the report that the aircraft was in a pitch up attitude of 16º when it reached 38.000 ft.

Agreed !
That is indeed a bit mysterious why the 16° NU at that altitude was allowed to develop.
Probably it is linked to the general question why NU was commanded.
Fear of Overspeed after some turbulent up and down compromising feeling for energy state?

kilomikedelta

27th May 2011, 21:54

Checkboard; Would the cockpit displays tell the pilot that the THS was at maximum nose-up so that stick aft wasn't helping?

mm43

27th May 2011, 22:05

Rob21; Old Engineer;

About 11 hours ago, very near the end of the Part2 thread, I noted that PJ2 was in the air, and therefore no comment from him in the short term.

I suspect he is by now absorbing the confusing data. I have slept on it, and am still confused by many things. Not least is the lack of coherency in what has been reported by the BEA, which I believe is possibly a deliberate attempt not to put any party in a bad light, and also to placate while side-stepping any judicial implications.

GarageYears has recently posted the Flight Laws link, but here it is again:-

A340 / A330 Control: flight & laws (http://countjustonce.com/a330/a330-flight-laws.html)

.... which I have tidied up during the past couple of days.

unmanned transport

27th May 2011, 22:09

Unmanned/unwomanned transportation in the pointy end is the way to go for the future. Just think of all the carnage humans have created in past wrecks.

Alber Ratman

27th May 2011, 22:09

Surely their horizions, PFD and standby would not be affected by AD errors? They corrected roll deviations that must have been indicated by those instruments. If they were getting accurate roll info, surely the instrument would be indicating high nose up angles as well?

Low speed / high alpha / rapid descent = Stall.. Not a dive. So why continue to hold a nose up attitude? We will never know unfortunately.

RetiredF4

27th May 2011, 22:10

As i understand the previus posts concerning the nose up THS trim, this is said to be a normal function of Pilot nose up stick input and the computers.

Can somebody explain, how tis is done in relation to the airspeed? I mean, at high airspeed the change of the trim for a given stick input would be slower and at low airspeed it would be higher?

What kind of speed input does the system use, the wrong one from the iced up pitots? If that is the case, in the beginning the actual speed was still high, but the sensed speed was wrong and already much lower.

PNF recogniozes abnormal speed and sets pitch and power, in this case correcting also for the drop of the right wing while trying to get the nose to 5° pitch. The system generates a much greater pitch imput due to the wrong airspeed, which the PF does not recognize right away due to other distractions.

Or does the system generate the same input regardless of airspeed (can´t imageine that would work)?

It would explain the initial maneuvre.

TyroPicard

27th May 2011, 22:11

Chris Scott
Why are the AoA values considered invalid below 60kts IAS when the A/C is not on the ground?
Because without much airflow at really low IAS gravity affects the position of the AoA vane just as it does on the ground? The designers had to choose a speed - they chose 60kts.

Checkboard

27th May 2011, 22:18

Checkboard; Would the cockpit displays tell the pilot that the THS was at maximum nose-up so that stick aft wasn't helping?

The pitch trim position is shown on the flight control page of one of the displays - but you would have to select the correct page on the screen to see it. It is also constantly shown on the physical trim wheel in the cockpit.

Alber Ratman

27th May 2011, 22:21

Is AOA visible on all flt Control screens as well?

Peter Fox

27th May 2011, 22:21

Ashling,
I agree with you. Applying the UAS QRH memory items "should" have saved the day.

I was also _not_ aware of the fact re stall indications during the recovery with the airspeed decayed to such low values. This is certainly something to put away "just in case" and to pass on to others.....

Regarding selecting Flaps below 20000ft, "our generic" QRH has some interesting indentation wrt the "If in clean configuration and below 20 000ft":
The way I read this, literally, is: Only if you are out of the stall you select Flap 1. Clearly that is wrong, but if I was a lawyer.......

Incidentally, it is the same in the QRH "on board"!

Cheers

Yellow Pen

27th May 2011, 22:22

Until today were you aware that the stall warning could go away if a stall developed to the point that indicated airspeed dropped below 60KT?

I can't remember if I did as I'm not current on type, but I do remember that in the event of unreliable airspeed you'll could get a whole range of varied and contradictory warning signals and I'd place no faith in any of them. Pitch and power are the only things you can rely on until normality is resumed and getting a further stall warning in response to doing the right thing, whilst damn confusing, shouldn't be unexpected.

Checkboard

27th May 2011, 22:26

Is AOA visible on all flt Control screens as well?

AoA isn't displayed directly anywhere - however if you select the flight path vector (FPV) on the Primary Flight Display (PFD), then the gap between the FPV and the pitch reference is the AoA.

http://www.airsimmer.com/snap/ags/PFD-Attitude.jpghttp://www.airsimmer.com/snap/ags/PFD-FPV.jpg

The image on the right is displaying an Angle of Attack of around 5º (Normal cruise is about 3.8º - 4º, stall about 16º )

Peter Fox

27th May 2011, 22:31

Very true Checkboard,

however, have a look at our QRH (A319/320), page 2.16:
"If remaining altitude indication is unreliable:
- Do not use FPV and/or V/S which are affected.
etc....

I read this as follows:
You can not rely on the FPV all the time.

In case of erroneous Air Data I would, personally, go for the Pitch / Power settings given/memorized.
Then GPS alt.
If the FPV matches all of those, good. I _might_ consider using it.

Cheers

mm43

27th May 2011, 22:32

Originally posted by Alber Ratman ...
Is AOA visible on all flt Control screens as well? Checkboard has shown you the display, but the crew selected the appropriate page(s) and a not available warning was displayed. This information became the subject of 2 x ACARS messages timestamped 0211 and received 2:12:10 and 2:12:16 respectively.

EDIT :: List of ACARS Messages (http://countjustonce.com/a330/acars.html)

Alber Ratman

27th May 2011, 22:38

Thanks checkerboard.. Would the crew have changed screens in their situation if the flight vector had not been selected / knocked out?

True, the AOA data would have also been false due to lack of airspeed on the vane and may have been another inject to make a bad situation even worse.. However it might have done the opposite to confirm what state the aircraft really was in..

EDITED.. Other posters have stated that such infomation was knocked out with the first ACARS messages.

gums

27th May 2011, 22:39

I don't think the thing works like we expect, RetiredF4. It doesn't even work like the Viper or Shuttle FBW.

After going thru the manual once again, looks like the trim is not set for a gee command from the stick. It's set for 1 gee at bank angle zero and a bit more up to 30 +/- degrees of bank.. In other words, if you hold the stick back "x" degrees for "x" gees, you don't gradually move the stick to neutral and continue at the original "x" gees you commanded.

So for you 'bus drivers!! Would love you to try it for real, but let's use the sim. Command 1.5 gees with aft stick and let us know if you can gradually let off the stick and maintain 1.5 gees. I don't think so. From what I read, as soon as you let go of the stick the plane tries to get to 1 gee.

The Viper pitch trim was gee, and stick commanded gee like the .bus. We had a roller trim doofer and could trim for a gee or so negative up to a max of 3.5 gees positive. The stick added or subtracted from this. So we could roll it ( trim wheel) full aft and do a really neat hands off loop that started at 3.5 gees, then let off once AoA got above 15 degrees, then continued over the top at 25 deg AoA until back to 15 deg AoA ( due to increasing speed), then come out the bottom at 3,5 gees. kinda neat, huh?

So the so-called auto-trim means you don't have to crank a wheel or use a trim hat switch when speeding up or slowing down or in a turn at 30 deg of bank +/-. Then, you have the plane (Airbus) trying to maintain a speed/mach according to the AP mode and also trying to "protect" you from bad conditions. The Viper had the gee command for pitch ( body rates and AoA blended), and so we had pee poor speed stability, but we stayed at 1 gee or the trim setting. No auto throttle, ya see?

One of the laws in the 'bus will crank the THS nose up if the computers think you are overspeeding. So with all the sorry data being fed the AF447 computers, I can see the plane cranking the THS all the way up.

CONF iture

27th May 2011, 22:45

Could someone who flies this confirm the simple question
"Can you tell what the other is doing?"

No Paul :{ and Airbus thinks it is just fine ...

BarbiesBoyfriend

27th May 2011, 22:48

The Colgan crew were near the stall.....but pulled.

This crew seem to have had stall warnings and also pulled.

Surely to goodness, faced with unreliable speed, the LAST thing you'd do is pull?

DozyWannabe

27th May 2011, 22:49

Oh boy this is an understatement. There aren't many chapters in aviation history as sad as this one. One finds oneself grasping at straws to understand it :(

Based on the information we've had over recent years, it's not that uncommon an accident type (now that more basic errors and systems failures have reduced in number). Night operations are the worst possible time and place to have unreliable instrument indications - being in the vicinity of convective activity only adds to the risk - certainly something that would weigh heavily on the mind of any pilot in that situation, regardless of experience, age, type rating or temperament.

That said if they had responded correctly to the initial stall warning would they have found themselves in that situationand having found themselves there they had other indications, attitude, to help with the diagnosis. A pilot well versed in UPs/Upsets should have been able to recover the aircraft safely.

Checkboard; Would the cockpit displays tell the pilot that the THS was at maximum nose-up so that stick aft wasn't helping?

The attitude indicator/"artificial horizon" would have been telling them exactly what was going on in that regard. However, as has been shown by accidents over the last 20 years, how pilots react when instruments become unreliable is heavily dependent on their aeronautical knowledge and temperament. In the case of the Birgenair 757, only one pitot tube (the Captain's) was blocked, and all other instruments including the F/O's ASI were indicating correctly. Despite this, the Captain seems to have almost shut down, focusing on his erroneous speed indications despite the F/O clearly warning him that his own instruments indicated that they were stalling.

The question is - given that you know at least one instrument has failed, how much are you willing to trust the others? In this case you're in the middle of threading your way through some dicey weather conditions in the middle of the night, so you have no outside reference, and as has already been pointed out it's around 3 minutes from the onset of stall to impact with the ocean (it took me longer to write this paragraph).

Not least is the lack of coherency in what has been reported by the BEA, which I believe is possibly a deliberate attempt not to put any party in a bad light, and also to placate while side-stepping any judicial implications.

Or possibly, you know, just releasing information that they can confirm completely at this time rather than releasing information that needs time to be double-checked and re-checked. In their position would you rather go with what you know and can prove, or go off half-cocked and possibly have to issue an embarrassing retraction at a later date?

That said, thanks for the link - this seems to have some relevance by my reading (emphasis mine):

In Alternate Law, Valpha prot and Valpha max are replaced by VSW. Note that VSW is a stall speed and so is EAS sensitive unlike Valpha prot/max. Stick free aircraft aims to control alpha by limited authority stability order to less than VSW. If the stick input forces VSW exceedance, alpha floor activation of TOGA occurs. However, it is possible to stall the aircraft as the protection may be overridden. Above V/Mmo an over-rideable nose up demand is introduced. The VSW symbol disappears but Vmo+4 and Vmo+.006 warnings remain. V/Mmo is reduced to 305/330/.82.

Alternate 2 : Pitch control laws are identical to Alternate 1

Can somebody explain, how tis is done in relation to the airspeed? I mean, at high airspeed the change of the trim for a given stick input would be slower and at low airspeed it would be higher?

What kind of speed input does the system use, the wrong one from the iced up pitots? If that is the case, in the beginning the actual speed was still high, but the sensed speed was wrong and already much lower.

If I've got this right - the system will detect the unreliable speed indications and simply give the pilot the authority requested (in this case full nose-up or thereabouts). Loss of airspeed indication basically implies stall protection is minimal, if not lost entirely in this failure mode.

[EDIT: Just checked - autothrust was disengaged early in the accident sequence. With autothrust disengaged, stall protection in this mode is indeed non-functional. At Habsheim, we saw a demonstration of what Normal Law will do to prevent stall in the event of autothrust disconnect/disable - i.e. command elevator inputs to maintain incidence even if the PF attempts to command nose-up. It would appear that Alternate Law in combination with autothrust disconnect defers to pilot input (as designed).]

No Paul :{ and Airbus thinks it is just fine ...

Really, CONF?

At 2 h 13 min 32, the PF said "we’re going to arrive at level one hundred". About fifteen seconds later, simultaneous inputs by both pilots on the sidesticks were recorded and the PF said "go ahead you have the controls".

Looks like he was perfectly aware from where I'm sitting...

Des Dimona

27th May 2011, 22:53

If both PFD's and the ISIS had correct ATTITUDE information, then application of significant forward S/S pressure to place the pitch attitude well below the horizon, irrespective of the IAS indications across the cockpit, would probably have changed the outcome.

Unfortunately, this comes with hindsight and a wider knowledge of the changed Airbus procedure for stall recovery. (which incidentally, is the same as the stall recovery I was taught in light aircraft in the late 1960's! - ie reduce the angle of attack).

Development of the Airbus flight control systems during the 1980's focused so much on the level of protections, that there was no real thought by Airbus engineers that a line crew would ever be confronted with the events that AF447 ultimately experienced.

Stall recovery at that time was based on TOGA thrust, perhaps along the lines of propeller aircraft that developed significant airflow over the wings due to prop wash when power was increased? This was not unique to Airbus.

Habsheim (Idle thrust into the trees) and Air INTER (incorrect FPA set) created the first clues that modifications needed to be made to the AB flight control and FCU systems - and it's been a steady stream of changes since then.

The latest incarnation of the UNRELIABLE AIRSPEED reinforces attitude flying + thrust setting = performance.

This works well at lower levels provided it's followed. Around the FL 350 area in a very heavy 320 / 321 / 330 / 340 , I don't think setting 5 degrees nose up and CLIMB thrust would be such a good idea ?? At these levels, there is virtually no margin between cruise and climb thrust and all that's going to happen is a speed reduction. At standard ECON speeds, it won't take long for the speed decay to reach VLS and below.

This checklist needs to be modified to look at higher flight level recovery.

Up until now, triplicated computer systems as installed in Airbus products can provide problems. If 2 of the three say one thing, the voting system will assume they are correct and "outvote" the third and possibly only correct system. This is one very important reason to get pitch attitude and thrust settings from the QRH to check what is really going on.

At heavy weights and high levels, I have the descent numbers from the QRH U/Airspeed checklist (IDLE thrust and the appropriate pitch attitude) handy and that's what I would be using (after recovering from any STALL warning) , until we can establish which instruments are correct.

Much more basic training needs to be provided by manufacturers during type ratings so that we have a much bigger awareness of the art of attitude flying.

Far too much reliance on automated systems is causing as many issues as these systems were designed to address.

In many years of Airbus experience, I have had two events that required manual flight and attitude intervention due to IAS irregularities. Both involved icing events.

I wonder what the real number of these occurrences has been over the last 20 years ?

Checkboard

27th May 2011, 23:02

- Do not use FPV and/or V/S which are affected.
I don't know why this would be, as the FPV and the VS are derived purely from the Inertial Reference System (IRS). :confused:

So for you 'bus drivers!! Would love you to try it for real, but let's use the sim. Command 1.5 gees with aft stick and let us know if you can gradually let off the stick and maintain 1.5 gees. I don't think so. From what I read, as soon as you let go of the stick the plane tries to get to 1 gee.

That's right - the stick (usually) commands a G acceleration - pull back for 1.5G pitch up, and that's what you get as long as you hold the stick back. The nose raises - then when you return the stick to neutral, the nose stops rising, the G returns to 1G and the aircraft trims to that position, so the nose stays where you put it.

True, the AOA data would have also been false due to lack of airspeed on the vane
The AoA data from the vane is fed to the Air Data Units for stall calculation (and some other stuff). The PFD indications and the FPV are from the Inertial Reference Units - nothing to do with air sensing.

SandyYoung

27th May 2011, 23:04

I know nothing about flying but having spent most of my professional life writing software am well aware of the difficulty of the 'user interface' between man and machine.

One thing is crystal-clear here - the current system is far, far too complicated for any normal mortal to deal with in an emergency. If professionals can't agree on what works what, whilst sat at a comfortable desk with all the time in the world to think about it, pity the poor souls at the sharp end with two or three minutes to sort out what can only be described as a mess of meaningless abbreviations, dials, guages, bells and whistles. And in darkness.

If he or she doesn't know what speed the plane is travelling at and whether it's pointing up or down perhaps the time has come for a major re-think.

These systems are fine when they are working but if something goes adrift we are asking far too much of the pilots.

Shovelling in yet more software is not the answer. Less might be better.

Peter Fox

27th May 2011, 23:15

SandyYoung:

Lets have an analogy:
You walk along a straight road. At a constant pace. Suddenly you close your eyes.
Do you:
a) Try to keep walking straight. At a constant speed.
b) turn left and start running, then slowing down, then turn right, maybe even backwards?

In pilots terms:
Set a pitch attitude. Set Power. You get performance.

Then open your eyes again.

Read the QRH.

Cheers

Peter Fox

27th May 2011, 23:19

Checkboard,

I do not know either.

But it is there, black on white regarding the FPV and V/S.

What it also says is:

RESPECT STALL WARNING!

Something to remember!

Cheers.

Nekkie

27th May 2011, 23:20

It all started with an unreliable speed, so they didn't trust the speed indicators. In the turbulence they lost all sense of what is up or down. They thought they were not stalled because the stall warning did not sound. When they pushed the nose the stall warnign came up, so that must have made them think that could not be trusted either. Adrealine is rushing and you see the altimeters showing high rate of descent. You think you are diving and pull up. If that does not help panic brakes out and clear thinking stops...

DozyWannabe

27th May 2011, 23:29

@SandyYoung:

It's not really that complicated, it's just a case of applying common sense. The computers are driven by the same sensors that drive the flight deck displays in front of the pilots. If the pilots can see that that their airspeed indications have failed, then it's not a major logical jump to know that the computers have detected that airspeed indication has failed, and it can be reasonably assumed that any protections requiring airspeed information (chief among these being stall protection) will no longer function.

At the risk of sounding like a broken record myself, from this point onwards it doesn't matter if the aircraft you are flying has FBW or conventional controls, glass cockpit or "steam gauges", the onus is on the pilot to diagnose what *is* in fact working and then use that information with their knowledge and training to try to rectify the situation. Sadly, the odds are heavily stacked against even the most experienced and able pilots when they have no outside frame of reference and such a limited time to perform said diagnosis. It's the one situation where the knowledge that one has to do something, coupled with the knowledge that to do the wrong thing could doom you and everyone behind you would put an unbearable amount of mental pressure on even the most unflappable individual or crew.

@Nekkie:
Adrealine is rushing and you see the altimeters showing high rate of descent.

You've also got an attitude indicator telling you you're nose-high. If you see nose-low in your AI and a high rate of descent then you can conclude you're in a dive and pull up. Nose-high with a high rate of descent and the only reasonable conclusion is that you're in a stall and falling. Once you've been able to work that out, you put the nose down and cram on thrust until your stick inputs start to make sense on the AI, you then level out and fly pitch and power until you're confident you're fully in control. The problem, as lots have pointed out, is that it's pitch black outside, so you have no confirmation that your AI is telling you the truth and you've got 3 minutes to make that call with a cacophony of lights, messages and warnings going off around you. Your boss has just entered the cockpit, maybe he can help? But the clock is already ticking...

Svarin

27th May 2011, 23:51

Gentlemen,

after this one, I will take an extended leave from pprune. The hairy troll gets hungry and will go out to pick mushrooms. I have worked way too hard on this, and my life (yes I do have one) is calling me.

For those who are ready to read and try to understand with an open mind, please bear with me.

I have come to study very carefully the flight control system mounted on this type of aircraft. Whatever modest findings I came up with do not exclude any other factor added to the accident sequence. Yes, I may even be wrong ! At this stage, however, findings provide material for crucial questions.

What prompts me to question the Flight Controls System in this instance has two aspects :

- a philosophical aspect, as I explained in earlier posts : since the system is built in such a way as to claim final authority (under certain circumstances) over the aircraft control surfaces, it should be questioned first, regardless of pilot actions, and regardless of which law it is supposed to have been working in, or what the aircraft FBW system advertised capabilities are. Its central position in the design and its criticality in a Loss Of Control accident should make it the subject of extremely careful scrutiny as a matter of principle.

- a technical aspect, the core substance of which I will now address.

As an introduction, it must be stressed that "protections" are provided in many forms for this aircraft, and the really assertive forms of protection do not come from the autopilot but directly from the main Flight Control Computers, the PRIMs, even in manual flight, and even in alternate 2 law, albeit in a less severe version. Only Direct law and worse remove all protection.

Three out of five components of the Flight Controls System have suffered anomalies :

- PRIM2 suffered a connectivity problem with ADR1, such problem is deemed HARD, which means permanent, not transient.
- PRIM1 stopped operating.
- SEC1 stopped operating.

These anomalies have consequences which can be verified using the Flight Controls Reconfiguration Schematics from FCOM chapter 27 :

- PRIM2 is tasked with THS motor because PRIM1 quit.
- PRIM2 is tasked with both elevator halves because both PRIM1 & SEC1 quit.

Additionally, PRIM2 operation is polluted because it lost connectivity with ADR1, thus reducing its discriminating capability while simultaneously getting fed multiple erroneous ADR data.

This simultaneous double failure condition : connectivity + erroneous data, is an extremely serious issue. I very much doubt that such apparently unrelated, simultaneous, totally different failures (external cause -> icing + internal cause -> wiring) were ever considered together during the design phase of the Flight Control System.

This extraordinary combination precisely affected the one computer which ended up being tasked with :
- interpreting sidestick commands on the pitch axis.
- sending orders to the hydraulic servo jacks located on the elevator moving parts.
- sending orders to the electric THS motor.
- providing whatever "protection" it deemed necessary to provide while in its undefined state (unreliable ADR data + lost connectivity with one ADR)

The lethal question exists in either of these modes :

- the researcher will ask : what is the exact state of PRIM2 throughout the whole event and what is its exact role in its unfolding ?
- the pilot will ask : what the heck is happening ?

It took me two years to simply get an adequate understanding of the core issues. They had two minutes before it was doomed.

Rest In Peace.

Svarin

PS : Post Scriptum / Pure Speculation

PURE SPECULATION BEGINS
"Je ne comprends rien."
Total confusion. Unimaginable betrayal. Such as Flight Controls twisting on you.

I have been asked to describe the protections I thought interfered with the pilot's actions. Contrary to all that I have posted before, what follows is pure speculation. The relationship with any real accident would be coincidental. This does not mention any aircraft-specific data. We, I do not know what happened yet. This is provided only as a thought-provoking element, which, although speculative, is not contrary to such kind of system's description under the condition of erroneous data being fed to a computer which did not correctly assess its degraded state. In other words, an example of a Byzantine fault made manifest :

-imagine two concurrent protections triggering alternatively upon erroneous indications. Imagine the Flight Controls System erroneously believing itself deep into coffin corner, overspeed and low stall at the same time.
-Stall protection has first priority and pushes the nose down without the pilot knowing why, and when the pilot pushes the control down, stall protection deactivates (the pilot did the right thing).
-Then overspeed protection undesiredly kicks in (second on protection priority list) which pulls the nose up, again without the pilot knowing why.
-The time lag between the two is such that, as the pilot pushes and waits for a result, the protections switch to overspeed and the aircraft noses up, and as the pilot pulls up, the protections again switch to anti-stall and the aircraft noses down until stall protection deactivates.
-And the cycle continues. One way to try and break it is to deliberately apply one clear, full deflection control command for a significant amount of time, and see what happens.
-Both protections acting undesiredly in this kind of deadly ballet would maintain AoA more or less near its maximum until real stall occurs from total energy depletion. Deep stall follows.

Too complex, eh ? No way this can be true, of course. But it sure makes one think twice about "protections", especially multiple, contradictory ones.
PURE SPECULATION ENDS

PPS :

Can any protection backfire on you ?
Mother said you would be safe from harm
A child can dream of becoming a pilot
But a pilot is not a child anymore

takata

27th May 2011, 23:52

Hi gums,
One of the laws in the 'bus will crank the THS nose up if the computers think you are overspeeding. So with all the sorry data being fed the AF447 computers, I can see the plane cranking the THS all the way up.
You should read again what flight envelope protections are left under ALTERNATE LAW 2, also displayed as ALTERNATE LAW (PROT LOST).
Difference between ALT1 and ALT2 is due to loss of Air Data, it just can't work with unreliable Air Data (at least 2 reliable ADRs needed).
ALT2= no high speed protection, no high angle-of-attack protection, because no valid airspeed can be computed by the PRIMs to display VMO/MMO or Valpha-prot and Valpha-max = Loss of the function enabling calculation of limit and characteristic speeds.
In fact, this may explain why the PF was pitching up before realising that it was not working. He was not stable due to oscillations (turbulences?) and wanted to slow down to better manoeuvering speed and to catch up Alpha-prot... which was never triggered.

BarbiesBoyfriend

28th May 2011, 00:11

Surely to Goodness, if it all goes to hell, fly attitude.

The EADI will (most) always tell trhe truth.

In a big ol' Cb, you cant hold height or speed anyway, but you CAN hold attitude.....and as long as you do so you're in with a good chance.

I think these guys may have been poorly served by their (A/P loving) training.

john_tullamarine

28th May 2011, 00:24

I don't think I've ever seen a thread put on 12 pages of posts overnight.

For those who wish to review the earlier threads, a link to thread #2 can be found in post #1 of this thread and similarly in thread #2 to get back to thread #1. Both of the earlier threads will be deep in the archives so the links are the convenient means to step between threads.

I have to agree with several posters - BEA has added to the confusion by giving up so little to mull over in its latest release.

DozyWannabe

28th May 2011, 00:32

OK, once more and then I'm signing off for a bit - I've noticed my handle cropping up far too often in the last few pages for it to be healthy.

- a philosophical aspect, as I explained in earlier posts : since the system is built in such a way as to claim final authority (under certain circumstances) over the aircraft control surfaces, it should be questioned first ... Its central position in the design and its criticality in a Loss Of Control accident should make it the subject of extremely careful scrutiny as a matter of principle.

And you can bet it is, right now, the subject of lab tests and simulator runs that will take months to perform and collate the data from.

As an introduction, it must be stressed that "protections" are provided in many forms for this aircraft, and the really assertive forms of protection ... come ... directly from the main Flight Control Computers, the PRIMs, even in manual flight, and even in alternate 2 law, albeit in a less severe version. Only Direct law and worse remove all protection.

Their design, however, is very clear on the fact that loss of speed information -> no more autothrust -> practically no stall protection.

I very much doubt that such apparently unrelated, simultaneous, totally different failures (external cause -> icing + internal cause -> wiring) were ever considered together during the design phase of the Flight Control System.

Actually, what you're describing would be relatively trivial to set up on the test bench, and is the very epitome of what software engineers refer to as an "edge case"

This extraordinary combination precisely affected the one computer which ended up being tasked with :
- interpreting sidestick commands on the pitch axis.
- sending orders to the hydraulic servo jacks located on the elevator moving parts.
- sending orders to the electric THS motor.

A series of tasks for which it is eminently capable

- providing whatever "protection" it deemed necessary to provide while in its undefined state (unreliable ADR data + lost connectivity with one ADR)

But it knows that no speed information -> no stall/overspeed protection.

It is a "baked in" assumption in the design of the system that if it is ever incapable of making a judgment, then ultimate control authority is given to the pilot, who has an array of instruments and, in daylight, can see outside and use external references if necessary, something the system cannot do.

PURE SPECULATION BEGINS
...
-Then overspeed protection undesiredly kicks in (second on protection priority list) which pulls the nose up, again without the pilot knowing why.

Problem being, there is no speed indication - the flight management computer knows this, and is therefore unable to command either overspeed or stall protection. It's all there in Alternate 2 (aka Alternate "NO PROT" - is it becoming clear yet?).

But it sure makes one think twice about "protections", especially multiple, contradictory ones.

The question is, how can a protection that has been disabled by design cause the scenario you describe to unfold?

I think these guys may have been poorly served by their (A/P loving) training.

Instinctively I'm inclined to agree with you. On the other hand my mind goes back to the Birgenair 757 incident, where a long-serving pilot who came out of the Turkish military was just as flummoxed by a similar situation. Which leads me to wonder - regardless of whether training turns out to be a factor, what part does temperament have to play?

Graybeard

28th May 2011, 01:06

From series II: (last night)

26th May 2011, 22:21 #2518 (permalink)
llagonne66

GB
I beg to differ on your comment.
QUOTE
Just maybe Airbus squeezed the specs just a little too tight, and ended up with pitot that are fine nearly all the time, but without enough reserve
heating energy for every case of high altitude icing conditions.
UNQUOTE

Please go back to BEA interim report #2 page 63 :
QUOTE
The set of icing tests to be performed to meet the Airbus specification includes 26 test points in all (10 for covering appendix C and 16 additional tests), thus covering a wider envelope than that defined by the JAR25 regulations.

The Airbus specifications used for the certification of the probes are therefore stricter than those of JAR 25 (annex 4).
UNQUOTE

We don't know by how much margin they passed the tests, or whether deterioration of either the probes or the power supply or wiring have been a factor. We do know the pitot on the A330 have not always been robust enough for real life icing.

We also know that the computers were not robust enough to accommodate erroneous airspeed.

GB

jcjeant

28th May 2011, 01:08

Hi,

It's just my opinion .. but I always thought that the independence between the two sticks was a mistake or at least a contributing factor of misunderstanding
The Boeing system seems more relevant from a security standpoint and offer better communication between the two pilots (visual and sensitive)
Was this a contributor factor in this case ? ... I can't say....

DozyWannabe

28th May 2011, 01:23

We also know that the computers were not robust enough to accommodate erroneous airspeed.

OK - one very-last post then... ;)

You can't make a computer perform tasks with data that it known to be erroneous, it's just a logical impossibility. As such, the design of the system - on paper, before a single line of code was written - clearly defined that in a case where it is unable to do its job, the best hope lies with the pilots. So issue a "stop" command to the relevant module (beyond which nothing other than a full reset can bring it back up again), notify the pilots of the failure and leave the flying to them.

As soon as "Alternate (NO PROT)" mode was triggered, the pilots were effectively in control of a regular airliner with no protections and full control authority in the pitch axis. What isn't clear at this point (and what I suspect the BEA will be spending months trying to determine) is how the crew reacted to the situation they found themselves in, and what their intent was.

Just so we're clear - one last time. Based on the evidence we have so far, flight control protections and the software that defines them are unlikely to be directly relevant in this accident sequence, because they were disabled very early on in that sequence, precisely as they were designed to.

alph2z

28th May 2011, 01:42

In today's report (and in this thread, yet) there is no mention of overspeed and especially its alarms (if any).

Before this report I always assumed that they might have had both stall and overspeed alarms. Similar to what happened to the 2 crashes due to clogged air-pressure instruments due to paint and insects (i.e. near Peru, & Dominican Republic).

This prel. report gives the impression (possibly wrongly) that the main alarms they got were stall warnings (amongst others) and not overspeed ones. And yet the pilots may have thought they were overspeeding. Oh my.

Anybody have any ideas about overspeed ?
.

grizzled

28th May 2011, 01:42

Alber Ratman wrote:

Low speed / high alpha / rapid descent = Stall.. Not a dive. So why continue to hold a nose up attitude? We will never know unfortunately. (my italics)

I don't agree. As is most often the case, the reasons for the actions of the participants will very likely be made clear though the CVR (possibly even aided by the DFDR). The info released by BEA is extremely sparse (for many reasons) but one can be quite certain that a great deal of discussion and interaction was occurring throughout those final minutes.

grizz

grizzled

28th May 2011, 01:57

Dozy...

Looks like he was perfectly aware from where I'm sitting...

I believe your conclusion is not justified by the facts as we know them.

There is nothing we know yet that tells us whether the "dual input" was complementary or contradictory. Nor do we know what other discussion surrounded that action. Nor even, at this point, which two crew members were invloved.

And Conf's point stands: Sidestick inputs by one crew member may not be obvious or clear to another crew member. Whether you consider that is a "good thing" or not, it is something that AI is (so far) satisfied with.

grizz

bearfoil

28th May 2011, 01:59

Dozy Wannabe

...Just so we're clear - one last time. Based on the evidence we have so far, flight control protections and the software that defines them are unlikely to be directly relevant in this accident sequence, because they were disabled very early on in that sequence, precisely as they were designed to..."

I finally see your point of view. Some one needs to explain your Faith to some few qualified pilots, who are obviously of a different Faith.

No harm no foul, there is always room for your opinion well within whatever the experts decide...

See? Foolproof......

bear

Lazerdog

28th May 2011, 02:23

A lot of posts expressing surprise that they didn't realize they were in a stall. A lot of aircraft will stall and have no phugoid oscillations, so once you are in the stall there are really no "seat-of-the-pants" feelings of being in the stall. (You only have a falling sensation at the stall when you go from level flight to descending in the stall.) You have to be aware of AOA, airspeed, or reserve lift. Remember, this was at night in likely bumpy conditions so relying on the instruments was the only way out. You can bet that someone has already run a simulator with this accident scenario to see what a simple push would have accomplished.

Graybeard

28th May 2011, 02:29

Dozy Wannabe:
You can't make a computer perform tasks with data that it known to be erroneous, it's just a logical impossibility...

Just so we're clear - one last time. Based on the evidence we have so far, flight control protections and the software that defines them are unlikely to be directly relevant in this accident sequence, because they were disabled very early on in that sequence, precisely as they were designed to.

It's PRECISELY the DESIGN that needs a review. The AP/AT were no doubt in Alt Hold and Spd Hold modes. The A/P has inputs from 3 A/S sources for good reason: they vote. If two or more go wonky, the AP should have enough intelligence to reject them all and go into flywheel mode, holding the last power and pitch setting. It should then send an SOS to the pilots, giving them the decision to take over, or let the flywheel run awhile.

See, a really capable AP could do that smoothly without going off in a pout.

GB

Chris Scott

28th May 2011, 02:33

spagiola post (http://www.pprune.org/tech-log/452836-af447-thread-no-3-a-6.html#post6477199) #117 (27/1733z):
"Did the resumption of the stall horn once the speeds once again become valid after commanding pitch down (at 2:12:02+15) confuse the PF into thinking he was doing the wrong thing?" [my emphasis]
Quote from MurphyWasRight:
"Until today were you aware that the stall warning could go away if a stall developed to the point that indicated airspeed dropped below 60KT?
And that it would return as you recovered?"

I think it may well have done, particularly considering the situation he had found himself in. The irony is that the AoA-based stall warning had been inhibited, indirectly, by the system perceiving an IAS below 60kts. On the occasions when a reduction of AoA was achieved, this may have cruelly coincided with a recovery in the (false) IAS to have enabled the (warbling, audio) stall warning to resume.

(In any case, because of the under-reading of IAS, the threshold of stall warning would have probably been too high for the altitude, as has been discussed by HazelNuts39 and others on previous threads [see Page 46 Of BEA Interim Report No 2]. So it is arguable that the stall warning, even when not inhibited, may have been inactive at times when it should have been active.)

This accident seems to demonstrate an Achilles-heel (weakness) in the stall warning system for Pitch-Alternate Law and Direct Law that needs to be addressed. As I wrote in my previous post, why is it considered necessary to inhibit AoA-probe data IN FLIGHT when the IAS is measured below 60 kts? (On the GROUND, it must be.) It seems to me that there are sufficient indications available to establish GROUND/FLIGHT status automatically, as used by numerous other systems on the aircraft.

Quote from MurphyWasRight:
"Until today were you aware that the stall warning could go away if a stall developed to the point that indicated airspeed dropped below 60KT?
And that it would return as you recovered?"
Although what you say in your last sentence may have happened, it seems to have been by (tragic) coincidence of the vagaries of the UAS. If the erroneous IAS reading had remained near zero, the stall warning would not have returned.

Checkbard, post (http://www.pprune.org/tech-log/452836-af447-thread-no-3-a-10.html#post6477599) at 27/2129z, quote:
"[....] aircraft which crashed (or very nearly so) because the pilots failed to realise that the trim had been set full nose up, and couldn't understand why they lost pitch control."
This seems possible, if unlikely. The limited pitch-down commands from the PF may have not been sufficient to cause the THS to run to a less nose-up trim-state than the 13 degrees. In my post (http://www.pprune.org/tech-log/452836-af447-thread-no-3-a-9.html#post6477446) at 27/2011z, however, I asked:
"is there any possibility that the THS motor stalled during down-elevator inputs?"
This has been known to happen in older types, but seems improbable in this case.

Quote from TyroPicard, re my question as to why the AoA data is inhibited IN FLIGHT below 60kts IAS:
"...without much airflow at really low IAS gravity affects the position of the AoA vane just as it does on the ground? The designers had to choose a speed - they chose 60kts."
You may be right, but, if so, surely they could be balanced and damped. G-forces could be the problem? (The VC10 has non-mechanical AoA probes, as you may remember.)

Yellow_Pen (27/2222z),
Agree that pitch and thrust can be relied on in fairly level flight. Once you are climbing or descending steeply, particularly in a stall, I guess it gets tough as a recovery tool? Maybe AoA is the only answer, but it would be a very radical step in airline ops.

Checkboard (27/2226z and 2302z).
Think it was me who started a discussion on use of FPV as an indication of AoA last year. It's not that easy to use in changing bank, and inaccurate when wind-to-TAS ratio is significant. But it's much better than nothing. Unfortunately, AF447 seems to have lost FPV-capability during 0211z.
Re the FPV VS data, I also used to think it was inertial, but remember being corrected by one of my copilots! Too near bedtime to check: sorry.

Svarin ("The Last (http://www.pprune.org/tech-log/452836-af447-thread-no-3-a-12.html#post6477847) Effort"?),
Your posts have always given much food for thought. Don't go off mushroom-picking at this stage, please, just because you want a life...

Chris