Diversification

How shure are we that the "nose-up" was made by the pilot flying. I have been thinking about a possible sudden malfuntion of his side-stick, like a broken spring or failing position transducer.
Could this be a real possibility of would it have generated another ACARS mgs?

A-3TWENTY

Did they loose the IR`s or only ADR`S?

spagiola

Hmmm, no. If the AP kicked out because of unreliable airspeed, he was already in Alt law -- even though the PNF didn't verbally state this until 11 seconds later.

Teddy Robinson

I'm not qualified, nor interested in the level of speculation, informed or otherwise on this thread, next report issued, we can look again perhaps.

My parting shot is this : leave it alone it will still give 1g / zero roll rate .. hold it in ALT law, the THS will trim back to support the input, to reverse this requires a sustained push as the THS re-adjusts .. so without over complicating matters, and without having access to the full picture .. there we, or at least I have to leave it until we have further.

The aircraft does not (in my recollection) have a true AP / OUT mode, call it CWS if you prefer, until you are at the absolute level of reversion, it says what it will do on the packet. Aggravated stalled condition perhaps .. deep stall no.

all yours.

mm43

Something that is apparent to me, is that once the aircraft was in the fully developed stall, the right roll was effectively a side-slip/bank as the direction of flight changed and became stabilized once the wings were leveled.

No mention has been made of any rudder inputs, and it looks as if there was no spin element in the descent. The final roll to the right may have been in the process of being corrected at the time of impact, and hence the left wing low and the tail yawing to port. The same could have happened if someone planted their foot on the right pedal shortly before impact, i.e causing the yaw and left wing low.

What was causing the roll to starboard? Some symmetry problem, or payload/fuel out of balance and no flyable airspeed to compensate?

milsabords

@Minorité Invisible
Cher cousin, it does not sound ambiguous to me: "en butée" (full stop) applies to "gauche" (left) and "à cabrer" (nose up).
Otherwise it would be: "à cabrer et en butée à gauche", or "à gauche et en butée à cabrer".

Machinbird

I have been looking at statements by posters from various parts of the world regarding use of the manual pitch trim and there appears to be a dichotomy of views.

Posters from Europe seem to believe that you will never touch the manual trim wheel while airborne. And if you do use the trim wheel during a training session, you will not pass.
http://www.pprune.org/6476909-post77.html by Svarin and http://www.pprune.org/6477104-post106.html by Svarin
http://www.pprune.org/6476888-post72.html by Narval

A poster from Hong Kong says, yes you don't use it often, but there are times when its use is highly appropriate, and he remembers using the manual pitch trim during training.
http://www.pprune.org/6479397-post419.html by cxhk

There is one of those little rules that we all know to be true. It is use it or lose it.
It appears that non-use of the manual pitch trim is dogma in European training centers. The Hong Kong poster's statement highlights a more pragmatic view.

We have now planted two European Crewed Airbus aircraft in stall related situations where timely use of manual pitch trim would have been the fist step in saving the day.

Under the pressures of the moment, it never occured to the crews to lay hands on the manual pitch trim, or apparently even to look down at it to see where the trim was.

It would seem that Airbus drivers everywhere need some training reminders on the use of manual pitch trim.
The old training dogmas regarding manual pitch trim use need to be thrown into the dustbin.

JD-EE

TLB - TO/GA was not asserted until the second stall warning. At about the time of the first stall warning, perhaps before it upon AP/AT disconnect, the PF simply raised the nose a little and left the CLB default throttle setting.

JD-EE

A-3TWENTY, all the lost was the ADRs. The IRUs did not have any reason to drop out. They're more like a reference for everything else with some tweaking via the navigation system Kalman filter, presuming there is one.

DJ77

TurbineD, in your post #596 you asked about possibile similarities betwwen AF447 and the airprox incident over North Atlantic (AAIB bulletin 6/2001).

Looking at appendix B of the report showing P1 sidestick controller inputs it is a bit puzzling that the majority of the inputs are upwards during the zoom climb (while TCAS was calling "descend") and even during part of the recovery phase of the event.

I have no idea what to conclude.

TLB

> TLB - TO/GA was not asserted until the second stall warning. At about the
> time of the first stall warning, perhaps before it upon AP/AT disconnect, the
> PF simply raised the nose a little and left the CLB default throttle setting.

JD-EE,

Thanks. But this is exactly the point I am trying to raise. If the PF initially thought his main problem was an unreliable airspeed situation, then his initial response (raised the nose a little and left the CLB default throttle setting - your words) was exactly the right thing to do, according to the C/L.

Centrosphere

JD-EE,

With due respect, but I fail to see how having a bad day can be an excuse. If a pilot has a so bad day that it can impairs his/her judgement, he/she has the option of staying home, sparing the lives of a lot of people. In fact it´s not an "option", is a moral obligation.

And in fact it seems that the less experienced pilot was commanding the airplane. This raises some interesting questions beyond his own hipotetycal culpability: for example, is it considered to be good practice for a captain to let the less experienced pilot drive the thing during the crossing of the ITCZ? This is a fair question, you see, even if in the end BEA investigation proofs that the pilot received erroneous information from the system.

takata

Hi,
Yep, I concur with your explanation, hence also this English translation.
Alternate law should pop up on each pilot's PFD at 0210:05, with several amber flags (3). The PNF talking about the "speed lost" and "alternate" are more sounding like a cold statement than a timely warning. The THS certainly did close to nothing at and right after this point as it was still at +3 deg (close to normal) 46 seconds later until much more sustained stick NU. It took about one minute to reach +13 deg but it was between 0210:51 and 0211:51 (more likely 0211:40). More than very likely, this trim rate was caused by pilots NU imputs alone.

From 0211:40, she would not be anymore in Alternate law as all speeds and AoAs were invalidated => abnormal law: THS is frozen at +13 deg and autotrim become inactive (like the RTLU was at 0210:05, remember? we told you that it was its last valid airspeed before ADRs faults).

About the roll, it looks like an oscillation (left and right) and I'm wondering also about what caused it as she seems unbalanced at its right from the time AP went off. From 0210:05, yaw and roll axis were in direct law without any PRIM control and to not overshoot bank corrections might be tricky even in light turbulence.

Some questions I'm asking myself:
1. was the first stall warning "real" when AP went off?
2. what caused this roll just before AP went off?

cwatters

Would be nice to see a plot of the aircraft attitude rather than just a few AoA data points.

At one point they say the AoA reaches 40 degrees. It's tempting to think this means the nose is pointing at the sky but that's probably not the case. I mean if they are descending at 100kts vertically and flying horizontally at a similar speed what would/could the attitude be to achieve a 40 AoA ? Could the attitude be close to level?

Machinbird

Since this was an A330, there are two Alternate Law reversions possible.
Per BEA, the aircraft reverted to Alt 2 Law.

Per my understanding of the flight control system:

Alt 2 is a roll direct law. This means that the aircraft is in direct law on the roll channel and Alternate Law in pitch.
Direct law is a 'you fly it' law. Control displacement causes proportional control surface displacement very much like any non-FBW aircraft.

This means to me that keeping hands on stick was not optional, particularly if the aircraft was a bit right wing heavy. There is no lateral trim on Airbus. About all you might hope to do is throw in a bit of rudder trim to compensate and move fuel around to adjust lateral balance. The PF could not have let go of the stick with a wing heavy aircraft. Under this situation, he could very easily tense up and apply unintentional nose up input.
Now, would some real A330 pilot tell me if I am reading the Flight Control FCOM properly.

wes_wall

What position was the rudder in? Or, where was the vert. stab. Was it still on the airplane. Just wondering. A lot of indications the sitck was moving from up to down, and left to right. What were their feet doing?

infrequentflyer789

Certainly not in the real aircraft. The sim is likely accurate enough that you will need the stick forward to recover from stall (not approach to stall).

See post 498, (http://www.pprune.org/tech-log/45283...ml#post6479876) by jcarlosgon for someone who appears to have actually done such training.

@jcarlosgon
thanks for the informative post,I, for one, would be fascinating to know if there was any debrief on why it took so long and whether any reason for slow control response was discussed.

Clandestino

They are freakingly correct! Under no circumstances is the trimwheel to be used while the autotrim is operative. It's not a dogma, manual trim in normal and altn law is useless as autotrim will return it to what it thinks is correct setting and 99.9999% of the time, it is right too.

What the poster from HK was referring to is loss of autotrim in alternate abnormal and direct laws or in mechanical backup. Then you have to use manual trim. As A320 was my first jet aeroplane ("You reek of propellers" was one of milder estimates of my capabilities I received from established jet jockeys), powers that be decided I should be given 20 hrs of sim time in direct law to get used to oh-so-enormous-speed-of-320. To my utter disappointment, 320 turned to be much easier and more comfortable to fly than ATR-42 even in direct law with all protections off and no autotrim.

I've never, ever heard of sidestick going haywire in real life, except for DLH's roll polarity mess up - courtesy of maintenance.

There are hundreds, if not thousands, flights through ITCZ every day. Somehow, they manage to get to their destinations safely. Go figure.

CONFiture has made an excellent question: why does autotrim doesn't stop when stall warning is activated? I hope that final report will have equally good answer to that.

As for the rest of speculation around, it seems that my neural centre for disgust over completely unfounded conjecture has overloaded and popped the breaker.

takata

Hi Machinbird,
Well, I think that's a bit jumping too early at those trim wheels.
I do believe that posters from Europe and Hong Kong would agree that:
1. You need first to aknowledge the stall situation;
2. You need next to apply full nose down before starting to think about moving the wheels, reduce thrust, or rolling to nose down pitch.

Airbus seems pretty confident about its elevators authority, even with a jammed THS between +8 and +14 degrees, as it is said to be an issue mainly above 180 kts.

infrequentflyer789

The way I read the BEA info, the nose was raised more than "a little" and pitch much more than 5 degrees.

But on the C/L, buried in a postscript to a post a few pages back by Hyperveloce is this little bombshell:

Now, is that a "reminder to follow the procedures", or an urgent message that the previously advised procedures were wrong ???

If it is the latter, then AF found probable cause in training / SOPs only 5 days after the crash - who did they tell ?