These weren't the first airline crews to go all the way to the ground with the stick back. I just gave two briefings on Loss-of-Control Accidents and it seems to be a common theme.

I've also been researching cascading failures and multiple apparently random annunciations. We may have a bigger problem than it appears at first glance.



Goldfish

Hi,

TheShadow

Interesting .. but I read

The aircraft (announced by one of the pilot) go in alternate law and it's big difference when compare with direct law.

Sorry but I disagree, 5° CLB is the pitch and power data to get you away from troubles (if the safe conduct of the flight is impacted), then you set the pitch and power from the table off the cecklist...

If you are at 39000 and loose the airspeed indication all of sudden but still level, no change of airframe noise and altitude then you go for the table without setting 5° and CLB, therefore I cannot agree with you when you say that there is no airbus checklist for unreliable speed at cruise level, there is, and to me it's quite clear.

Thank you, TheShadow, very informative. I guess I come from a generation when "having to manually hand-fly an aircraft" was the preferred option.
If there's to be any real progress, I guess we need to examine what training led to the 'human factors' error they made. From my humble flight safety background, it usually turns out to be human factors rather than pilot error.
Next time I fly the pond, I'm tempted to fly it myself in a puddlejumper(again).
p.s. there will only ever be one 'procedure' for everything; it's called "airmanship".

Stabilizer trim to 13 degrees up?
 

Sorry if this has already been brought up, but why would THS go from 3 to 13 degrees during the interval 2:10:51 (red ball 5) and the next entry 50 seconds later, and stay there for the next 4 minutes until the data stops?

The way I read it is that the AP had disconnected by this time, so was some other system automatially increasing THS in the 2:10:51 interval?

Well, as an example of how confusing it is, I happen to have my Jeppesen basic IFR book here, and it seems to say the opposite:

As an aside, what's wong with a good, old-fashioned idea of having a small red or yellow ribbon on a small stick which is bolted just outside cockpit window? Shine your flashlight out at it and it will give you a very fast indication of what direction the air is going past the airplane.

Deep stall recovery
 

So it seems this plane was into a deep stall with nose up 16 degrees. Angle of attack 40 degrees. Vertical speed 110 knots. Horisontal speed low(60knots) Seems to me that the only option to get the plane out of this condition would be to get the nose of the plane into the direction of movment. Even full trottle would not be able to make a horisontal speed high enough to give a flyable angle of attack. So the only option would be to get the nose down in a steep dive.

I suggest using the rudders hard to one side to completely stall one wing get the plane to flip over in a spin would acctually help to get the nose down fast and then get the correct angle of attack back. An other option to get the plane to flip over would perhaps to give a high trust on only one of the engines - to create the nessesary rotation to loose all lift in one wing and get the nose.

I suggest even taking trottle to idle would not harm at all (once the spin is there - at suficcient altitude of course) - It might simplify recovery not have to deal with those nose-lifting forces and to rapid increasing speed diving with the nose to the ground. Simply get the nose of the plane into the direction of movementwould exchange height for flying speed rapidly and solve the problem. .

Also - I would like to know if comercical pilots ever practise stall recovery in a real plane in real air. I have practiced stall/spinn from 3000 feet in a sail-plane several times - and suggest practice of this method in a sailplane could be valuable even to comercical pilots.

The initial stall-spinn does not involve high G-forces. The G-forces will be felt when at high flying speed with the nose to the ground will slowly lift the nose back to horisontal (and then add some trottle to keep a normaliced flight) In my sailplane typically 2-3 G.

I'm not talking about "what to do when you get the stall-shaker alarm" but - what to do when your allredy deep into a stall - but horisontal wings - and nose up.

I also realize that there may be some flaws to my argument here- Im not an expert in this area.

Bottle of Water = Attitude Indicator
 

If the pilots did believed "all indicators are down" then they could have used a bottle of water to get a rough idea of horizon. It would give an instant sanity check on the pitch and bank.
If you think you are -45 deg but you look at the water level shows +45 then you are likely to change from pull stick try to push stick. When you do so and the water level instantly responds gradually returning to level. The pilot then has a feel for the spacial orientation of the a/c in the absence of a visual horizon out of the window.
The jostling of the airframe would move the water around however I believe it would be enough to approximate in the absence of any other indicators.
It really sounds like they had no instrumentation or at least did not trust the stall warnings. Otherwise they would follow the stalled checklist.
GPS gives the ground speed and direction. Again not ideal but better than nothing. Clearly the pilots were way out of the normal trained situation and flight envelope.

A few comments refer to "seat of the pants" flying. This has itself caused incidents when pilots don't look at the instruments and instead rely on gut feel. The mind plays nasty tricks when you experience acceleration without visual reference points.

Speevy,

If you get unreliable airspeed in the cruise suddenly with smooth flying conditions and everything normal then 5 degrees climb power may not kill you straight away. However, in a realistic scenario, with moderate turbulence, by the time you realise the aircraft has unreliable airspeed it's going to be possibly over 20 seconds from when it occurred. As a result your airspeed may already be far too high/low. By then pitching UP to 5 degrees (it will be a pitch up as you don't have 5 degrees pitch in the cruise normally) you are potentially worsening the situation. A slight overspeed is better than a low speed. Yet again, there is no procedure from Airbus for this situation. Your point of 5 degrees CLB was not designed for this scenario, let alone the fact that if your at your ceiling this is potentially disastrous.

stall warning stops when you are in a stall
 

Contrary to what is mainly reported, it seems that the pilot recognized a stall but changed his mind because the stall warning stops when in a stall :

"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. "

=> One can understand : the PF is the cause of the stall.

"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 "

=> The pilot recognized a stall.

"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 PF is responsible for a new stall

"the Captain re-entered the cockpit. During the following seconds, all of the recorded speeds became invalid and the stall warning stopped. "

=> when the captain re-entered, no information was available to him

"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."

=> the stall warning sounded when he pitch-down then his reaction is to come back to the last situation when the stall warning didn't sounded i.e. with the nose up.

It is just a human reaction and the captain had no clue to help him.

QUESTION : since the stall warning stops when you are in a stall ( speed < 60 kt) but sounds when you are recovering, what has to be changed ?

Idiots (and pilots) summary of WTHIH?
 

Hamrah post #530 , wiggy post # 569, think your two posts are a good summary of the unfortunate situation the crew were handed by their aircraft.

I am, quite unashamedly, anti Airbus FBW.

22 years on a 737 perhaps labels me somewhat Jurassic, but the same scenario (more or less) on my prof check 48hrs ago (my instructor was psychic? ) was difficult but eminently doable.
Not entirely sure if the wonders of Airbus (uncrashable! ! as I cringe yet again remembering Bernard "cablecar "Z emphasising on the Beebs coverage of Farnborough airshow all those years ago) actually help or diminish the human ability to alleviate sh1t happening, my personal opinion is very much the latter.

Edited to say The Shadow post #590, I think you deserve "post of the thread award" & I wish I had the influence/contacts to secure you a secondment to the BEA for a few months.
I sincerely hope they have, and are allowed to utilise, as much imagination/lateral thinking/insight, as you have shown.
Illuminating post :ok:

2nd edit jcjeant post #598, interesting link. If indeed BUSS was a good thing, yet again, like AF skimping costs by not installing GPWS and the resulting A320 crash in Strasbourg, the bean counters should be ahead of the pilots, when we attribute blame.

Oh BTW, only very slightly off topic, have all you (fellow ) Boeing drivers read the recent bulletin re ice crystals ? ?

Bottle of Water = slip/skid ball. Water doesn't know where the horizon is any better than you do. :rolleyes:

Satellite Uplink To Tech Support
 

Perhaps a big red button in the flight deck that instantly uplinks a live data feed from the flight computers to boffin headquaters at airbus would help. The boffins can analyse and have a chief test pilot tell you in real time what the heck is going on and the optimal recovery procedure. Meanwhile you can concentrate on flying. Low cost as you only get billed when you activate the uplink

Why ??
 

My question is why did the crew fly through the storm in the first place. Other aircraft diverted up to 90 nm to go around the storm but not this crew. WHY ??
Was the radar working ? Did they leave Rio with inop radar? Did they simply ignore the radar?
It seems quite stupid to me to fly a passenger jet at hight altitude close to the coffin corner through a CB. What were they thinking? Had they diverted, they would be alive today.

There is absolutely NO evidence the crew flew through or near a storm.

The first step in solving a problem is the acquisition of the correct information.

From the reports so far it seems obvious that this crew were faced with a considerable amount of conflicting information and distractions.

It is all too easy to look at an accident and think how an earth did that crew allow themselves to get into that situation? They were doing the best they could with the resources that were available to them at the time.

I think TheShadow has interpreted the available facts and portrayed an extremely plausible sequence of events. That scenario is horribly plausible. That it was erudite and technically accurate certainly add validity. As a current 330-200 pilot I can envisage just such a sequence and can now perhaps understand the confusion and fear that must have reigned. Bravo!

The captain left the cabin 10 minutes before the encounter so must have looked at the radar 100 miles ahead and been happy with it. Sounds like the turbulence wasn't a factor according to the report but the handling of the erronious airspeeds and autopilot and autothrottle disconnect were. I have never flown an airplane that you pull back on the controls with stall warnings but my Airbus friends say you can and the computer won't let you stall. I guess that depends on whether it is in normal, alternate or manual law. I think that is why I refused to fly the Airbus. My 4 airliners and 70 other types said if you stall lower the nose and add power. Not one said pull back. That was our spin entry when I was teaching aerobatics.

All the airlines I flew at had charts for unreliable airspeed showing power and attitude for your weight and altitude which would have taken care of this situation. We also had very qualified pilots in both seats so all had basic hand flying skills which were tested before being hired.

Now a lot of airliners are offering 18,000 per yr for a 250 hr pilot because the airplanes are automatic and don't require pilot skills, according to the manufacturer. See where we are going? I know these guys had some experience when the captain left but how much actual hands on time?

We were paid for our experience.

Ribbon
 

That ribbon outside cocpit window would do the job better than any digital instrument to indicate direction of air compared to the plane.

Also:

What would be the difference, if any -aerodynamically - between a "deep stall" and a "flat spin". Correct me if I'm wrong; but wings and stabilizers are completely stalled. You have little if any elevator authority. Increasing throttle will push the nose up increasing the AoA further. No ailron authority. I guess the difference is in the rotation.

So: How to get out of it?

Xcitation

Bottle of Water!!!

Isn't it embarrassing when you realize how rediculous was your suggestion? Trouble is someone might just believe it.

Milt, it makes as much sense as anything else on this thread.

Flat Spin vs Deep Stall
 

Ask21:

A flat spin is a spin where the AOA is almost 90 degrees, so you are headed virtually straight down but with very little pitch.

Recovery is as follows:

Have all of the pax crowd into the front. Then have them all return to their seats as you level off with airspeed, if they can do it under the G's. :8

A deep stall can occur in some in some designs, generally T-tailed aircraft, and unlike a flat spin, the aircraft is steeply nose down. In this condition the wings block airflow over the tail, so you have no pitch or yaw control.

They say the chances of recovering from a deep stall are greater than from a flat spin, but I've never flown an aircraft with deep stall characteristics, so I don't know the recovery procedure for that one. I imagine it varies by aircraft.

Hi,

Indeed .. high viscosity oil will made better .. but hey .. in emergency you take what at hand .....

I was in the crowd time ago .. in a Super Puma ...
10 people rushed forward on the pilot order .... some seconds after take off (bad loading cargo at the rear (was a !!!!load visible rear the net !)
Moving cargo during flight and no more prob !
That was in Libya ........

White Knight.

Thanks for stating the bleeding obvious truth.

Keep saying it.

These 'button pushing' pilots are easily overwhelmed and if I do not watch it , I will soon join them in their fantasy button-pressing '**** all to fall back on' world.

The de-skilling of flying is NOT well intentioned!

The de-skilling of Flying is intended solely to cut costs, thus allowing airlines to make more money.

The PPL should advise pilots about basic stuff but many pilots, who have thousands of hours (doing sudoku & eating) can barely remember how to pilot an aircraft.

Time for a big-Time wake up call I think.

Perhaps not. Think of the cost!

Hi,

Yes .. bean counters ...
To notice that the "BUSS" from AF statement not work well over level 250
Well ... the AF447 had about two minutes before reaching sea water when passing level 250
Methink it was maybe enough time to apply useful corrections .. if AC had a BUSS
Results may varies ....

As a mere SLF this is all very interesting but I don't understand why, if faced with an altimeter winding down, the instinct isnt to put the nose of the a/c into the windstream. If control authority isn't there well unwind the THS ( as per Schipol, not) and if that doesn't work put it to idle, and if that doesn't work deploy the gear. Did these guys not know they were fighting for their lives ?

FBW and unreliable IAS
 

On the A330 FBW, with the IAS indicating excessively low speed, the pilot's correct action of pushing the nose down and pushing the thrust to max may, in my opinion, cause the auto trim to trim "backwards" causing a pitch up. In a normally indicating IAS, the pitch trim should trim forwards, but because of the false low speed indication, the forward application of the side stick will be interpreted by the FBW as a pitch forward and therefore countered with the pitch trim going backwards.

If I am not mistaken, a similar situation developed in a non-FBW A300 doing a botched-up go-around in Nagoya, Japan, many years ago. In that incident, the auto pilot interpreted the pitch forward command of the pilot through the control column as a situation that required the auto trim to trim backwards, causing the aircraft to stall, and eventually crash.

In FBW, with the auto pilot off, the pitch trim still acts on its own, as if the auto pilot is on (in a conventional aircraft). You must remember that there is no trim switch on the control column wheel on the A330 for the pilots to trim forwards or backwards as Normal and Alternate Laws do all the trimming for you. In fact, most FBW pilots will not notice the position of the pitch trim at any given moment, so it would not surprise me if the pilots on the ill fated AF447 had not noticed the extreme pitch up position the auto trim had placed the aircraft in... and so they would naturally react to the stall situation by pushing the side stick forward more, further aggravating the situation.

As I see it, with hindsight, the correct action to recover from the stall in this case would be to release the side stick, and manually roll the pitch trim forwards until it is out of the extreme pitch up position the aircraft was in.

What Pish some folk post!

To fly an aircraft in IMC, one merely has to fly a certain attitude and a certain power.

Given that (probably) these guys had good attitude info, that only left them to sort out the power and fly.

The fact that they could not do 'straight and level' tells badly both about the computerised Airbus and the pilots easy reliance on its systems.

In rotten wx, autos often go for a wee lie down. Who then shall fly the aircraft?

Surely, the pilots?

In other words: Yet another 'loss of control' accident.

We need to get away from the automated flight regime that we are in today.

The writing is on the wall for all who care to read it and 228 people and their relatives know it.

Pilots must be able to fly.

And to a better standard than the autopilot!

To be brutally honest, a great many of my co-pilot colleagues could NOT manage their flying day without the autopilot. They would be sorely taxed.

These same co-pilots are now being promoted, having accrued about 2000 hrs.

To be even more brutally honest, I'll soon join them in their inadaquacy unless I make more of an effort to fly. (I have 10,000 hrs on regional ops)

It wil cost a lot of money to retrain these 'button pushers' to fly again, but they did it before (briefly) on their IR!

How much did recovering AF447 cost?

just an edit to say, there but for the Grace of God.......:sad:

The Captain's Response
 

I am definitely SLF.

The Captain re-entered the flightdeck in the middle of the developing situation. He had to gain situational awareness. He either gained it from the other pilots, in which case there would be a recording of what they thought was happening to the aircraft, or he gained it from his own senses i.e. whatever the instruments were then saying and what he could hear and feel.

It is fairly easy to assume the other pilots misinterpreting the information available to them and having formed an incorrect interpretation, continuing to apply the erroneous interpretion, disregarding evidence that conflicted with it. Unless the newcomer gained his understanding of the position by receiving an incorrect briefing, it is more difficult to accept that the Captain joined his fellow pilots' erroneous view of the situation without there being some pretty compelling, but ultimately wrong, data as to what the aircraft was doing.

The bare facts will show that they had 4 minutes plus to save the plane. Surely that should have been enough.

Mr Optomistic

That would be loads of time if you knew how to pilot an aircraft.

To be fair, when did anyone last do stall/ spin training on an A330?

"You can't stall it!"

:rolleyes:

You clearly know that I don't know. But how can you ride that thing down from 35k feet like that?

When the aircraft reacts opposite to what you were trained to do, and did.

As someone who has witnessed an A330 stall in the simulator I can assure you it is a 'pig' of an aircraft to recover if you get it fully developed. Also don't forget that you will only have the benefit of the BUSS if you have the presence of mind to switch off all the ADR's below FL250 in the middle of a very confusing situation and then it is as useless as a chocolate teapot if the aircraft is still in a stall. Sounds similar to the AIRNZ airbus in France, the aircraft should have been totally recoverable if the initial situation had been handled properly.

SLF question here: Given the weather conditions at the time, is it possible they thought that they were caught in a violent sinking air column in the storm (ignoring the stall warning and paying more attention to erratic airspeed and the altimeter), and tried to power and climb their way out of it?

Of course something like that should be evident from the CVR, so I don't know what to think otherwise.

Direct Law
 

What if they had chosen Direct instead of Alternate Law?

We've had so many theories since the crash about what may have happened... from lightening to new composite materials on the plane delaminating in flight to a storm so bad that it broke the airplane apart.

In fact, it crashed just because the pilots couldn't fly the plane in a way that anyone with a PPL would know how to do (when stall warning occurs nose down full throttle!).

So basically the mystery of AF447 crash is that... the pilots crashed an aircraft that was in perfect working order. Yes, it was in perfect working order, failed pitot tubes doesn't mean the plane was not in PERFECT working order.

The pilots incompetentness are to blame for the death of all those poor helpless passengers.

stall recovery
 

I'm a light airplane guy and don't know much about this, so forgive me if this is nonsense, but it strikes me that the A330 displayed remarkably good post-stall behavior -- no departure, no spin. It seems to have plowed along in a steep glide at 250 ktas or so (100 knots vertical speed, 20-25 degree flight path) and continued to respond to roll commands. But the advice to fly attitude confuses me. If you put the airplane symbol on the horizon, the airplane itself would still be stalled; it would be necessary to point it 20 degrees or so below the horizon to unstall it. It appears as if the sheer complexity of the systems masked the simplicity of what was really going on.

thcrozier, you cannot select Direct Law just like that on the A330. Alternate Laws 1 & 2, and Direct Law are reversion modes... which means that they come in only after a higher law cannot be maintained (usually because of flight control computer(s) failures).

Theoretically in Normal Law, the FBW A330 cannot stall, but as we all know now, did so on AF447. Whether the aircraft reverted to Alternate Law or not I do not know, but even if it did (to help recover from the stall), pushing the side stick forward could very well have caused the pitch trim to go backwards because of the false low speed sensed (outside of the flight envelope), thus causing aggravation to the stall situation.