Gee Capn Bloggs, I'm a gurl and I can do math and estimate better than that. 100kt horizontal and 107kt vertical would be less than 45 degrees from horizontal by a small amount.

10912' per minute is 107.75305 knots if you work it out to an unwarranted degree of precision. So the velocity was about 151 knots at very nearly a 45° angle with horizontal.

(It's handy with tabbed browsers to keep this link open when discussing this accident. It really helps avoid foot in mouth disease. It's not impossible, as I probably prove. But it does help. The other two good links are here and here.)

The velocity along the flight path (FPA -45), as you say, was 151 knots. The KTAS if sensed at the aircraft would have been around 89 knots, as MM43's geometry clearly shows.

It ain't "simply"
 

from a recent contributor:

First of all, I do not know of a single FBW system that acts in the same manner as the "old" hydraulic control systems we got used to starting in the 1950's. No mechanical connection to control surfaces, and the only feedback was body rates ( your own skinny butt, not an inertial or rate gyro). So the manufacturers did their best to give we lowly pilots a sense of feedback via springs, pneumatic bellows, etc. They also added things between we pilots and the actual control surfaces such as pitch and roll and yaw dampers, as we were moving from the old days to new designs operating at high speeds and aero effects the older planes dealt with, like Gooney Bird, P-51, Aeronica Champ, et al. Those in-between things helped to keep the pointy end forward and really helped in turbulence or near-stall conditions.

The big difference with the Airbus and the Viper FBW system compared to "simplistic" descriptions of FBW is you can never get "direct" control of the control surfaces via the computer. Well, not so fast, Gums. In the Viper, we DID HAVE A WAY to bypass the computer corrections and gains via a manual pitch override function. That thing acted like the "old" hydraulic control stick/rudder. Not recommended for the heavies, and only reason we got it was we found out we had an unusual flight condition that the computers couldn't handle. We still were at the mercy of the computers for directional stability and lateral stability. And for the "faithful" believer in the reliability of a well-designed computer system, how come they did not allow for this in the beginning?

The Airbus FBW, as in the F-16, F-22 and F-35 do not "simply" provide electrons to command the hydraulic actuators. The computers use air data, body rates, known aero characteristics, etc. to "tailor" the control surface movement in both rate and amount of movement. So this is vastly different than the "old" systems most pilots are used to and the skills they mastered.

What is this fascination with "protection"?

Beats the hell outta me, but I would rather educate the pilots that the jet "allows" this and that, and it will limit your inputs according to "its" idea of what you should be commanding.

The Airbus reversion sequences and all the "protection" and various control laws could be appropriate for a pinball wizard or drone operator. But they need better human engineering and human education as to what happens when this system goes south and what aerodynamic conditions require less "protection".

This discussion prolly needs to be moved to a separate thread in order to allow others to question who was in what seat, what pilot actions could have resulted in roll or pitch angles, what time the ACAR's transmitted a message, the bad weather in the area, etc.

Possible reasons for roll when AP disconnects in ALT 2....

Fuel imbalance
Asymmetric thrust
Rudder position
Roll demand
Aileron or spoiler position
Icing

Any more, anyone?

And responding to ventus45....

The autopilot uses bank angle to control heading. It applies rudder to maintain balanced flight. If this is done as yaw damping the rudder pedals do not move.

No.

But once PF was flying manually in ALT 2 he would have his feet on the pedals.

I must have missed this "emphasis" - can you point me at a quote from the BEA?

If the pilot needs to adjust the rudder trim he uses the rudder trim switch. When the AP disconnects it is not a requirement to press the "zeroing" button.

We've been making rather little note of the important FDR detail, RHS is not recorded. The plane was being flown from RHS, apparently. So that does conjure the notion that RHS, PF, was seeing something vastly different from what was recorded.

I am not sure probes will ice the same way. If the drain hole on RHS clogged PF would have seen a significant (17%) increase in speed. That could be the reason for the initial pull-up. After that the RHS speed indications may have been too highly varied to be used. If the AoA got very large then sensor_validation's note about pitots being straight ahead sort of devices suggests the real airspeed was unknowable in the RHS of the cockpit.

I am wondering if there is a scenario that experienced pilots can concoct that could explain the PF's actions presuming he was a fully qualified experienced pilot, with or without any valid airspeed indication.

Should, ;) and one would hope that is a reflexive and trained reaction.

(To play devil's advocate ... one would presume that pitch and power is the response ... but that does not appear to be the case. So it's hard to be sure, puzzle pieces missing ... )

OKC:At what altitude? 35K? 25K? 10K? Within 1000 feet of the surface?

Terminal phase seems to have been where the estimates are coming from, which is down low. (FWIW, as we play with our CR2's or other wiz wheels! :) )

rrr, the PF's first reaction was a left nose-up stick input at 2:10:05.

And after PickyPerkins' most informative post I wonder if the rolls were an attempt to diagnose if the plane was in a stall or not. Lack of roll authority is one of the "features" of a stall mentioned.

Litebulbs, we know the plane traded a lot of velocity for altitude. I worked out the simple math on that one going back to mgh and 1/2 mv^2 basics. It lost around 40% to 50% (I'm not sure of the knots to KCAS conversion here) if we presume 100% efficient conversion. Another 10% for aerodynamics would not surprise me.

(It takes a LOT of pushing these days but when pushed that hard I do have a nasty habit of "doing the math." It happens in political discussions, too.)

Flex, Picky's post is why I am starting to seriously question what PF presumably in the RHS saw on the RHS indicators and suspecting PF used roll as a stall diagnostic after the training of the day to possibly ignore spurious stall indications. It might be that visual inputs to PF were different from what we assume based on LHS only data leading to some of his apparently bizarre behavior.

It is interesting to ponder one of the themes in Perky's offering of how the two year process was kicked off, with the airline's position being ...

I am reminded of an old adage: complex problems frequently have offered simple, and wrong, solutions.

Granted, the two year process hopefully hammered out most of the wrongness, but you can see a trace of dangerous group think that needed addressing from the outset. And the problem is so noted ...

What are computers? Rule based machines. :ugh: That's a source of tension, and friction, that isn't going to go away anytime soon. I concur with the knowledge based approach, and have long been an advocate of both education and training.

Might have made a lousy executive.

That said, techniques can be taught, but the opportunity and structure (to keep things standardized) has to be there.


By the way, a few years back (Within the past ten years) I ran into a USAF field grade officer who was not novice at training. He'd flown various planes, and had instructed in T-38's, and in less awesome training aircraft. His attitude was, for military pilots nowadays ... "We don't need to teach them stalls and spins anymore, we need to teach them how to avoid stalls and spins." :eek: (Mind you, when he and I were both young'ns, we'd both been taught both) He and I disagreed, as I took the position that one needs to teach both. (Whoever wants a bit of extra credit can look up the T-6A Texan II, and see if it spins or stalls ... :} )

It appears that the "corporate front office bean counter" attitude about training can infiltrate into the most unexpected places ...

No problem - I don't consider it off-topic because in this game, context is everything and it is important to understand the context in which these systems were designed. With an issue that is as emotive as this to pilots, on a forum full of pilots, it's sometimes necessary to pin down what the technology should be doing to tame the wilder theories. :)

Not cantankerous at all - prudent, I'd say. I suspect the engineers that designed and built the Airbus systems would agree.

And that's exactly the kind of experience that I and others on here need to build as full a picture as possible - keep it up!


They certainly didn't do themselves any favours in the early days when it came to engaging with end-users by the sound of things. I'm glad it improved - about how long would you say it was before they did?

Will do so.

I note the phrasing there. It is one of my pet hobby horses. Sometimes "bad stuff" happens. Sometimes there is no person to blame in any sane world. But trial lawyers always insist there is an entity (with money) to blame and that entity is other than the entity they are representing. "Somebody is always at fault and that somebody isn't me." is a most pernicious philosophy.

The more I see of this it looks like "best practices" of the time may have been wrong. Knowledge is not static. It grows. Best practices of 2009 already appear to be different from best practices of today. Holding Air anybody to blame for not following today's best practices 2 years ago is repugnant hubris. It presumes the mere men should be gods, always right and perfect.

(I suspect a good read of history will reveal that societies that live by the lawyer, die by the lawyer. Societies that live by the law (with few lawyers) survive by the law.)

CogSim, on the incredible coincidence of geometry somebody else has already mentioned airmass speed as a factor not in evidence in the equations. A fairly stiff headwind MIGHT fit better with the initial presumptions of the impact with the water.

JD, who writes the law? ;)

Lawyers ... except in an AB (or a Boeing) where control law is written by ... :E

Do pilots have options?
 

Alex

Many good questions. Re the Pilots Authority:
The PIC always has the ultimate say in whatever he does. It has been hinted that there may be subtle pressures, or "frowns" regarding Staying the Course for fuel or time concerns. There are always alternate fields available in case of necessity; engine failure, running low on fuel etc.
The pilot can always plan to deviate around storms and if necessary unload cargo or pax to take on more fuel. He can always go or refuse to go if he considers it unsafe to do so. He'll always have to explain his actions and justify them.
He may try to clear a deviation with ATC but if out of radio contact he can always just announce his actions on "Guard" channel, which planes in the vicinity are supposed to monitor.
Or he can declare "Mayday" declare an emergency and go wherever he feels safe.
But, to charge head on into massive build-ups up to 60,000 feet is sheer madness to me.
Everything I've ever seen about thunderstorms says: "Avoid them!"

There is no question in my mind that the violence of the thunderstorm caused the ice buildups, strong updrafts to an out of control situation, that caused all of the failures, and left the pilots incapable of dealing with it.

New Thread "The future"
 

My piloting is 4hrs , Auster, age 16, 1975 - so that is clear... Bitten, hooked, but physically disqualified to join your ranks. So here are my unbiased thoughts on
"Manual versus Automated".
Some suggest that it would make sense to let people hand fly from time to time, and that the bean counters are against it. Well, IF at any random time hand-flying (whatever that means) was better than automation, then whoever was selling automation would be out of business, and they are not, so I assume that (sorry), more often than not, Hal saves your average arse, more often than he kills you. I know, we individually accept our own mistakes and get pissed as hell when we are right and that damn computer is wrong but the passenger down the back prefer to go with the odds.

Of course, there are times when the bean counters will say it is ok to hand fly, probably when you are well rested, traffic is quiet, visibility is good, so, SO no pain no gain, what is the benefit, I have an open mind on whether hand flying in good times is enough to help you in bad times- you tell me.
Now clearly, when things get really interesting, Hal will be challenged, and with the benefit of hindsight, still being alive (cause he saved the average arse last week), we can spot the faults.

I propose the following scheme.
Flight regime is Normal or "10 years ago". Cessna produce a plane that flies like a 10 year old Airbus and every pilot gets free flights on that, hand fly all you like, push the envelope, but leave the SLF on the ground. IF you can fly better than the current Airbus then you get a job with Airbus.

Of course, there is a wrinkle, your bean counters might not let you fly the 10yr old technology, you might have to stand your own life insurance. At Paull (airfield) we had an RAF guy (navigator) who wanted to get his PPL, the RAF decided they had invested too much in him to allow him to do something as stupid as hand fly a C150!

If Automation was not better than Mr Average they would be out of business, of course there is the argument that they have moved the average down with their marketing but the statistics say that we are progressing - get on board! It is WIP, it is not perfect, it is progress.

If you are below average, say thanks to the automation, if you are above average (as are all Pprune members) then contribute your experience to the continous improvement!

In this particular crash , I think we are going to see shock waves everywhere. At these levels of reliability is is rather expensive to do (true) preventive, the most cost effective is fast reactive, I think in an ideal world we would like to think that the pitots and training had been changed after the first 10-15 cases without waiting for the big one.:(

Thought my graphic clearly stated in the title - Final Data -, i.e. FL0.

The BEA have indicated that at TOC (FL380) the pitch attitude and AoA were both 16° and the stall was established at that point. As far as I can tell the TE was traded off (as power was reduced) for increased AoA, and the RoD was initially established at around 10K ft/min and only changed slowly to a higher value as air density changed and the AoA became greater.

LW 50:

Those are just the near end game values.

I prefer think of the 151 knots as KTS (knots true speed) as opposed to KTAS. There probably wasn't much in the way of air involved at the approximate 60 degree AOA, even unblocked.

The 151 knot speed is a purely trigonometric calculation based on a report which furnished near end game ground speed and vertical speed. Nobody is blowing air on the trig tables. (Maybe some smoke however)

The 89 knots along the aircraft longitudinal axis would approximate a no wind a KTAS via KCAS inputs if the systems were functional to sense it, which corresponds to a fairly low KCAS down low. It could be somewhat more or less, however, depending on actual wind. The cosine of 16 degrees is .96 so its horizontal component is not reduced much.

I suspect the KCAS was higher with a headwind component in effect.

My only point was that the 151 knots was not necessarily Knots True AIR Speed. I am not trying to offer anything remotely relating to cause.

Math is not hard…Math is not hard…Math is…

Absent pilot input would it remain deflected or would it zero itself? I think you answered the question implicitly below as "yes".The BEA report noted the PNF commenting about ALT law. It did not say the PF indicated he was aware of this. He should have been. But other crashes indicate sometimes the PF fixates on one problem and starts dangerously ignoring the world.

I think the following has to be said again :

When the flight control inputs were inappropriate, the stall warning ceased, and when they were appropriate, the stall warning reengaged …


DozyWannabe, the way you 'simplify' the Laws is not correct :
Neither of the Alternate Laws give full authority to the pilot.
Except from Alpha-Floor, thrust authority is not Law dependant.

Too many Laws which also overload Airbus accident reports ...

Lone - politicians tell the lawyers what they want in a law in response to their electorate's demands and trial lawyer campaign contributions.

- Enough OT....

What made him pull up?
 

MartinM

"I was not trying to blame the PF at this stage. I did not say hang him. All I try is to figure out what made him pull after disconnecting the AP. But obviously there is no real data around to get a conclusion on this.

Yes, maybe BEA will release at some point further details of the FDR and CVR. A reconstruction of the instrument situation would be interesting."

Everyone seems to be saying that the PF pulled the plane up. The only reference to that is that the elevator increased into a nose up trim condition.

What if a strong updraft pushed the plane up to 38,000 while at the same time the nose dropped. Then trim would autmatically apply in the opposite direction, to trim up.

We're not getting precise information on exactly when the icing occurred, and when the auto systems kicked off or on again.

Or what the pilots were saying at that same time.

Wild turbulent excursions create a lot of strange inputs that I question the auto-systems can cope with. Which is why they kicked off.

And this is not specific to Airbus aircraft CONF. The fact is that when you have pitot/static failure, all bets are off when it comes to warnings/indications and the crew must concentrate extremely hard to work out what's going on.

Specific technical questions.
 

From a simplified description of the residual high-speed protection in alternate law:- “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.”

Will the protection-generated nose up command result in (time delayed) trim movement, or is trim only activated by a sidestick input?

Is a fly-level stick command interpreted as an override of the protection command? I suspect not as how else will the protection work; thus does the sidestick have to be moved forward to effect the override?

Would the FDR differentiate between a protection-generated nose up command and that of the sidestick?

Are there any default values for VMO / MMO for failure or invalid airspeed / Mach inputs to the limit computations?

CONFiture, I tend to agree with you. Too many laws. Unless the control laws decay nicely with a tactile feedback signal via the stick it's really hard to keep all the "if" clauses together in your mind in a crisis.

History of trim recommendations
 

I don't think Wainwright's talk mentioned AoA.

My main reason for thinking that Wainwright's talk might be of interest on this thread was (and is) his reference to the use of trim in upset recovery.

--- START EXTRACTS -----
Thus, for us (manufacturer's test pilots), an aircraft that is out-of-trim, for whatever reason, human or mechanical failure, should be re-trimmed. Whereas the airline instructors were against the use of trim .............

We spent a lot of time discussing the use of elevator trim, and we never reached agreement.

All the major US airlines were adamant on their policy to recover first using “primary controls” which excluded any reference to trimming. ...........

Again, a compromise was necessary ...............

And, the use of trim is not mentioned in the simplified lists of actions to be taken. .............
--- END EXTRACTS -------

So after two years of debate, the test pilots were forced into a "compromise" where although they considered the recovery sequence shoukd be:
first, trim the a/c --> then, regain flying speed --> finally, start recovering, in that order, the airline representitives refused to allow trim to appear at all in the short list of actions to be taken.

From some comments on this thread I gather that trim is still often not a priority action in airline upset recovery training.

Perhaps someone with recent experience can comment?

And the cosine of 61 degrees is 0.4848, so the longitudinal component of 151 kt is 73 knots?

Total energy
 

In that case the airspeed would increase, and the total energy. That did not occur. Total energy decreased slightly until the stall, then dropped rapidly.

Hi speed protection (Normal Law) cannot be overridden. The pitch up can be reduced, possably delaying recovery, but not overridden even with full forward stick. If you leave it alone to do its thing the altitude deviation will be no more than 700'. This is why Airbus advise against making pitch up commands during hi speed protection.

Importantly the autopilot disconnects on activation and if you pitch up strongly you can end up in AoA protection. Source Airbus Safety First Magazine.

Alternate Laws
 

Question 1:
What are the differences between Alternate 1 and alternate 2 laws regarding pitch? (Info below taken from an A330 Vacbi.)

Alternate 1:
The pitch attitude (Theta) protection is lost. High speed and stall protection are alternate.

Alternate 1A:
In addition the stall protection is lost.

Alternate 2:
The pitch attitude is lost and high speed and stall protection are alternate.
In addition the lateral normal law is lost and replaced by the lateral alternate law (roll direct and yaw alternate).

Alternate 2A:
In additition stall protection is lost.

Alternate 2B:
Pitch attitude, high speed, and stall protection are lost and the lateral normal law is replaced by lateral alternate law (bank angle protection lost).

With regards to "Pitch attitude is lost" in Alternate 2: Does this aggravate the situation perhaps in combination with a loss of- or unreliable pitch indication on the PFD (ADI portion) due to faulty/rejected ADIRU's? With the PF commanding a/c nose up, could he have been looking at an ADI telling him (erroneously) the a/c nose was pointing down?

Question 2:
Does the A330 FBW have a Flight Control System Self-test?

I remember way back in the 80's when I was a crewchief on the Viper. During launch, the pilot did pre-flight checks as I monitored launch procedures in concert with the pilot. One of them was the flight control self-test which was performed with assistance of the crewchief to monitor flight control surface deflections.

On one occasion the aircraft passed the flight control self-test as far as cockpit indications were concerned but I observed a jittering LH flaperon. I informed the pilot and he did a second self-test with the same results, the system passed the test but the jittering flaperon returned. After the sef-tests were completed, when the pilot moved the stick manually, the jittering was also observed as soon as the flaperon moved up or down from the neutral position. Obviously the flight was aborted and subsequent trouble shooting revealed a bad ISA (Integrated Servo Actuator). If I remember correctly a voting spool was the culprit.


Could the A330 FBW system have similar failures go undetected?

Worst case is alarms that don't help.
May I add this to the debate:
14 CFR 25.207 (extract): "Once initiated, stall warning must continue until the angle of attack is reduced to approximately that at which stall warning began."

DJ77;

EADS A330 TCDS A.004 mentions two Special Conditions which may be relevant: SC F-1 Stalling and scheduled operating speeds, and SC F-5 Flight envelope protections.

The wording of the FCOM suggests that there is no stall warning in normal law. Alternate law is a low-probability failure condition in which an airplane is not required to meet all requirements that apply normally.

update on stall procedure airbus
 

17.th Performance and Operations Conference in Dubai May 2011

Interesting
- the emphasize on AOA in relation to stall
- the problems concerning destiction of approach to stall and already stalled
- the date of publishing May 2011 !!

Question? The title sounds funny. Would have expected something like

"update on stall avoidance & recovery procedure airbus"


Edit: Jusdt saw, that it is on the other thread in R+N, its really getting hard to keep track on two different threads. Sorry for that.

HN39 My understanding is that Stall Warning is available in Normal Law - however, AoA protection means that you should not get to the AoA required to trigger it. Perpignan showed this where there was a Stall Warning in Normal Law (2(?) AoA probes frozen prevented Alpha protection, 3rd [unfrozen] AoA probe triggered the Stall Warning IIRC.

I would say that even if I am correct, this does not alter the substance of your post ;)

Hi retiredF4,

Thanks for the link to Stall Procedures Airbus May 11.

It is interesting to note that there is now a reference to stab trim.

"2 b) Nose down pitch trim…….………………………………………..As needed"

When the flight control inputs were inappropriate, the stall warning ceased, and when they were appropriate, the stall warning reengaged …

Is there any circumstance were pushing the nose down can initiate or exacerbate a stall ?

I had a long post but after catching up on the huge backlog of posts that accrued while I was busy, much has been covered so will make some numbered points instead:

1. I've read this as a 2 stage process - AP off, an initial upset but the crew possibly thinks they have recovered it by 2:10:50 (when one calls for the captain). By this stage, PF has bled off a large amount of speed. Then a second event around 2:10:51 which causes a stall, apparently unrecovered until impact

2. At FL380, assuming actual airspeed was 215kt, and AoA 4 degrees, how far off stall were they as a function of airspeed? Could it be as simple as decaying airspeed at that point, hidden by UAS problems? (With underlying original cause, an unintentional climb, when trying to control roll, that lost them a large amount of airspeed?)

3. If the first two stall warnings were invalid, but the third one (which lasted a minute, and then reappeared briefly when nose was put down) was valid, could it have been disregarded initially because it followed two invalid ones? That is, crying wolf twice, followed by a true warning that was then ignored.

4. Could at least some of the nose-up commands (after the initial left/up) be due to the large divergence between AoA and pitch? If the plane started dropping hard vertically, causing very large AoA, but the ride was relatively smooth (see gums' posts a long while back), and pitch did not appear to be too high, then wouldn't nose-up commands appear logical? ie the pilot sees altitude dropping with nose apparently not too high, does not believe they are in a stall, and tries to lift it? Instead, consolidating the stall? All this even harder to detect at night, inside bad weather, no moon anyway. And with the effect of moving the THS back so far.

5. During the minute that followed the stall, ISIS airspeed recovers. Then after that minute, airspeeds go "invalid" and the third stall warning stops (having been going for a minute). While considered "invalid", was the airspeed either accurately measured <30 kts (or was it low, but further affected by the angle of incidence for pitots) and merely considered to be invalid because it was so low -- and in that case, we have a true (or nearly true) "invalid" airspeed following, and possibly confused by, original false reports caused by pitot problems? (ie again, the low airspeed was initially pitots crying wolf, but the next time it was for real because so much forward airspeed had been lost)

6. A design error that the (third) stall warning deactivated at unreliable airspeeds, only to reactivate once control started being reestablished (nose down, pushing airspeeds up above the "invalid" threshold, and leading to valid computations again) -- but with the net effect of making the correct input appear to be the wrong one.

(Apologies if anything above is self-evidently wrong to a real pilot - happy to be corrected if that's the case)

With reference to somatogravic illusion, without visual cues:

- Can a steadily increasing pitch-up, of the degree experienced by 447, be misinterpreted as acceleration?

- Can the deceleration of the aircraft (either by the initial speed reduction to M0.8, or the slowing resulting from the climb) be misinterpreted as pitch-down motion?

- If so, can the combination of the two - when taken with rapid vertical speed and altitude loss - be sensed as a steep dive instead of a pitched-up fall?

I completely concur. The Airbus philosophy basically changed the game during a cricial phase of the recovery by altering the system and handling behaviour and produced inconsistent aircraft annunciations just when the crew were desperate for reliable information.

I sinerely hope for a proper objective professional comentary on this story. The Flight Global article and David Learmont blog contain spin, is not factual compared to the BEA report and mixes up the chronology implying the cause lies with the crew when clearly such an assumption is neither fair nor reasonable at this stage. The hand of Airbus Industries at work already?

@auraflyer

My reading of the BEA document is that at least 2 out of 3 Pitot tubes were actually working correctly just before the real stall, but wouldn't have been trusted. We don't know what the 3rd pitot derived airspeed was displaying - it could have been reading high at this altitude if total pressure locked in from FL350.