RetiredF4

Although you are right for this case in following PJ2' s early recomendation to do nothing, the conclusion that training in high altitude manual handling and stall avoidance or recovery training for such events is not necessary is premature.
AF 447 could have dropped into direct law as any aircraft in the future could do as well where control in all three axis has to be applied immidiately. Another day, another situation, anouher fault and we go again.
A pilot has to be trained to be able to grab the stick and maintain the necessary attitude througout the complete flight envelope and be able to recover from situations, when this flightenvelope is exceeded by whatever cause it may be.

In the video Confiture posted just some posts before, watch the loadfactor. It is below the normal 1g for 25 sec prior the stall entry and it is nearly consecutive 60 sec below 1 g after the stall enty. Even when pulling the loadfactor was below 1 g. The human body is unfamiliar and very succeptable to reduced g loading, and it is not known to be accompanied by stall. Without physical training expierience of such a situation, with unreliable airspeed, no AOA and no loadfactor indication, how should they have been fully aware of their situation? We are talking in hindsight here, we have analysed and studied all available material, have looked at the FDR second by second, studied books and all available sources and have now a totally different knowledge base then they had during that night.
We know that they did not react to the stall like they should have, but besides the possibility that they were not able to identify in being in a stall they may have reacted on their knowledge base, but didn't know how much nose down stick input was required, how long it was required and how 0 g really feel. "I have no control......" were Bonins words.

Goldenrivett

vilas,
I agree. But how do you detect it early if you don't have active side stick controls fitted? See Page 4.
http://icatee.org/wp-content/uploads...zu110722en.pdf

"At 1 h 00 min 27 (point n°2), the PNF manually disengaged the AP by pressing the takeover pushbutton on the side-stick. A pitch-up input on the PNF’s side-stick going as far as 3⁄4 to stop was recorded for 6 seconds. This input was accompanied by an input to bank to the right then left. The PNF stated that he did not remember these inputs."

vilas

Goldenrivett
I am not against active side stick. I remember having read 7 or eight years back airbus gave very serious consideration to this but never implemented it. But the side stick is on one side so unless you kept your hand there you may still miss it. In AF447 they were flying without looking at PFD for most part and that I find very strange. It was a clear case of mental confusion may be caused by fear because they knew something was terribly wrong and were unable to figure it out. As you know average inputs into side stick are very small and a startled response in pitch has the possibility of causing excessive destabilisation. Nothing happens if you don't do anything for a few seconds to investigate.

PJ2

It's very old ground now, but to re-introduce the counterpoint, there is no evidence whatsoever that interconnected CCs are more likely to aid stall recovery and independent SSs.

In fact there are more fatal stall accidents in aircraft with CCs than with SSs.

A CC that is full-back-in-the-PM's-stomach is not a factor in recovery. I can't think of any such accidents where one crew member took action and pushed the stick away from his/her stomach and fully-forward to get the nose down in order to unload then unstall the wing.

In the few accidents I am aware of, the CC remained near/at full aft position until impact.

The problem is not "CC vs. SS" and therefore the solution is not the interconnection of the latter. The data just isn't there, whereas both a working stall warning throughout the entire flight regime and a THS that stops moving with the first stall warning do have data and therefore cause for further examination as does the installation of an AoA indicator.

seagull967

PJ2,

Absolutely. People tend not to realize the lack of salience the cc position has in a situation like this.

Clandestino

I have lost count how many times I've quoted preamble of Airbus manuals on PPRuNe; they are not intended to teach basic flying concepts. Pilot who is training on Airbus will easily relate what is written with what he experienced in sim or aeroplane. PPRuNer who has no idea about basic instrument flying, airline training and operations, after reading FCOM usually finds excuses to let his imagination run wild, resulting in spectacularly flawed and somewhat hilarious theories on how Airbus FBW works.

So? Unlike PPRuNers disconcerted with Airbus flight system architecture, extreme majority Airbus pilots do understand and operate their machines correctly. If all the BS about Airbus FBW written here were true, there wold be no need for BSers to flog dead AF447 horse as there would be dozens of fresh accidents to pontificate about.

There is no voting procedure for speed displays. Vsw is displayed in ALT2. By the time of continuous stall warning, ADR1 was measuring correct speed for about 14 seconds. By the apogee, pitot3 was unblocked too.

It's a moot point anyway, stall warning and buffeting were very well present.

It's self explanatory for anyone who has a slightest clue about flying but I guess I have to bring it down to level understandable to average PPRuNer: if one is below desired altitude, one has to climb to it. If one is above desired altitude, one has to descend to it.

So either the dummyness of pilots who no more understand their aeroplanes is bringing total number of fatal accidents (despite increasing number of flights) every year to record low or your comment is totally ignorant and somewhat offensive.

Report has made it pretty clear that left to her own devices, F-GZCP would have done nothing significant.

No search engine can change the fact there was none as described in US.

Issue with approach to stall recovery that emphasizes minimum altitude loss is not that it's wrong (it isn't) but that it is too narrow in scope. It does not prepare one for high altitude event, full stall, or low altitude event with mistrimmed stabilizer. So my type now has universal procedure that lays burden on the pilot; thrust - as required, trim - as required. Better be sure you make it right first time.

Exactly. Despite forcing aeroplane to climb at RoC and pitch dangerously unsuitable for cruise. Despite climbing high above optimum and maximum altitude. Despite losing control of the aeroplane. Despite warning systems shouting "STALL STALL" at them.

Say, what was the right thing done that enabled other 36 crews to survive UAS in 330/340s? They didn't have AoA gauges.

Maybe it would be useful to define "situation".

So, avoid 'em.

True but TAM A330 incident on 12 11 2003, described on the page 87 of final report proves that even extreme ham fistedness, on A330, flying at cruise level in alternate law is not necessarily lethal.


For Finnegan's sake, it reaches 1.5 g in initial pullup, gets to +0.6 G minimum and then mushes between +0.75 and + 1.25 G for the rest of the flight. It's nothing unusual even for airliner!

Like this:
Lufthansa A320 incident at Frankfurt.
Questions?

No! Both pilots verbalized loss of airspeed indication and CM2 kept roll very well under control so he must have been looking at PFD.

PJ2

On salience of position...

The Airbus SS movement is measured in cm's so visual displacement cannot be accurately judged. In terms of tactile sense, if an SS is interconnected, one cannot know by feel or sight where the "null" point is. Small SS movements can result in significant changes in flight path. The initial SS displacements which occurred in AF447 to pitch the aircraft up eventually to 15deg would have been momentary, (meaning 'not steadily applied' but varying), and perhaps 2 cm's, 4 at the most. CC movement for the same result is as you say, small...may be the same numbers but in inches. Approximately...

It is only during the stalled-descent that one sees full-aft CC displacement in the accidents I was thinking of and by then if full-forward CC or SS isn't held in until the wing begins flying again, the airplane does not recover. Some call that a deep stall, but I'm old-fashioned because these stalls are recoverable whereas a deep stall is not; AF447 was recoverable; some say (and have calculated) down to FL100.

CONF iture

Helicopter pilots say otherwise where flight control command displacements are known to be minimal ...

Interconnected CCs provide a first hand information to all crew members on the flight deck that Independent SSs just unfortunately suppress.

vilas

Clandestino
Pilot pulls back on side stick and instead of seeing the result on attitude only monitors bank. Isn't that strange instrument flying? Sure in alternate law he had to do only that provided he had not given any pitch input. The actions were not logical but panic stricken, perhaps not coming from rational part of the brain but primal.

seagull967

Ah, so easy to sit in your chair in front of your computer and think "that would not have happened to me..." Maybe you're right. Maybe. I know for a fact I could get a very high percentage to do the same thing this crew did with the right set up. Nothing like the wisdom of hindsight bias to make it all seem so obvious.

Machinbird

Vilas, that is what a roll PIO will do to a pilot. Destroy his faith in his flight control system. Mentally exhaust him. Terrify him.

All circumstantial except for one thing. Just why was AF447 oscillating in roll during those first 30 seconds?



Who here has had experience with any form of PIO?
Virtually no one

RetiredF4

When your management get's notice that you fly your airliner like that in daily flying, or even that it is usual for you to expose your passengers to such flying, then you get fired. If you do it in stalled situation, you might get killed.

Expieriencing continuous below one g for first consecutive 25 seconds and later for uninterrupted 60 seconds is unusual, whatever you say otherwise. And for this crew it might have been more unusual to decrease the g to 0g as needed for recovery for even longer time, as they never had been exposed to such a situation.

vilas

Machinbird
PIO as the word suggests doesn't happen. It is made to happen. Again we come back to the same point. First you make irrational pitch input then create PIO in roll and forget about the pitch totally and a training programme that caters for all this, You may have to select commercial pilots from aerobatic teams.

Goldenrivett

"However, the bank angle increased continuously up to approx. 22°. With the commander’s call out :“I can’t do anything more“ the first officer took over the controls with the words “I have control“ and pressed the TAKE OVER PUSH BUTTON."

If I understand your brief explanation correctly then you are suggesting that PF keeps the PM informed of his SS movements verbally, in order to enhance their mutual SA of the control input?

If they had active SS fitted, then the cross wired controller's input would have been cancelled out by the other side's controller and the problem recognised during taxi checks with Zero roll commands. (More easily recognised than Left / Right confusion).
http://aviationweek.com/technology/a...mmercial-debut

But you're right, why should we discuss ideas about how to improve such a perfect system.

Machinbird

Vilas, unless you get hold of a defective aircraft, you yourself are never going to experience a roll PIO (also known as Aircraft Pilot Coupling) in your career, you already have the Low Gain touch.

BTW, I think your statement about the irrational pitch input is backwards with the creation of roll PIO. In AF447, The roll PIO started immediately. The pitch excursion followed in the next few seconds.

There have been videos of other Airbus pilots who have a large amplitude (High Gain) control technique that may well be a hazard when suddenly confronted with an unexpected drop to a roll direct condition-particularly if it happens in cruise. I do not know for a fact that this is an actual problem, but logic tells me that it should cause those people problems in that circumstance.

To suggest that a Pilot In the loop Oscillation (PIO) is made to happen is deceptive. Yes the pilot is driving the oscillation due to his control inputs.
No he is not doing it deliberately. He is merely using the controls in the manner that he has been accustomed to and is getting an unexpected and undesired result. The unusual behavior grabs your full attention. This attention grab pulls the pilot out of his trained response pattern. It is as if a poisonous snake has suddenly been dropped onto your flight deck-it demands your full attention.

If you have never experienced a PIO, count yourself fortunate. They are rare but dramatic. The A320 has experienced roll PIO resulting in a missed approach on several occasions. I would not be surprised if the FCS has subsequently had its software modified in subtle ways to reduce the likelihood of future occurences.

Stopping a PIO is usually easy-just get out of the loop-let go of the controls. Then when the aircraft stabilizes, gently apply only the minimum control input necessary to achieve your objective. Has anyone reading this actually received training in this subject?

From what I am able to observe. the subject of PIO/APC is an aviation myth to most pilots. Something they will never confront and do not need to know about. Trying to explain PIO to professional aviators who have never seen it is like talking in a vacuum.

Clandestino

It's not perfect. As any aeronautical system, it has its compromises and drawbacks. Yet, despite many PPRuNers claiming otherwise, it is well understood and service proven.

What is the point of the discussion? If those who are making bold statements about it have no idea about: aerodynamics, aeroplane systems, aeroplane performance, instrument flying, multi-crew operations, state of the pilot training today, that moderate turbulence starts at 0.5G so 0.25 is light or that ten degree roll is really insignificant even in transport aeroplane terms, then it really says a lot about the participants in the discussion but not much about the subject matter.

Of course you don't. Point is that LH captain clearly stated he lost control and FO has promptly and correctly taken over. Completely unlike AF447 where airspeed indication failure was quickly recognized but nothing was done about it, right copilot stated he lost control but would not let one in the LHS to try to regain it and CM1 anyway instead of taking over and returning the aeroplane to its level, couldn't think of anything better than calling the captain from the bunk to sort out the mess. Nobody said anything about any warning. No "disregard" or "non standard" could be heard. As very wise man once said:

Another point that in modern MRJT or LRJT, what your significant cockpit other as PF, AFCS or even you yourself are doing with controls is of no interest to you as long as the aeroplane follows the desired flight path. Whether control input is appropriate is checked on PFD, not by looking at or touching the yoke.

SilkAir used to employ former Black Knights pilot.

Correct, but this strangeness is not limited to instrument flying. A pilot that forgets the basics, tries to make the aeroplane climb where it just can't and holds the yoke/stick into the stall which is only interrupted by ground collision was described by Wolfgang Langewiesche in 1944.

Absolutely! Asimov has called it "reptilian brain", Kahneman "system one" but they agree when it takes over, any rational thought gets sidelined.

It is easy and absolutely wrong, that's why I did not and will not say that. Experience of the latest confirmed victims of the impossible climb syndrome among ATPLs is as follows:

RHS: 6900 hrs, 6064 on type
LHS: 14 268 hrs, 6161 as captain on type, type rating examiner.

What little we know so far is that it can happen to anyone.

wiggy

Well there's at least one here who has, on a FBW type, on finals..and the big clue as to WTF was going on and what to do about it was the position of the control wheel verses the aircraft's behaviour.....

Not sure if that's going to further the debate or not......

Machinbird

Care to elaborate a bit?
Were you the oscillator or the oscillatee?

seagull967

Machine,

RE APC, well put. It has even caught low gain pilots off guard, particularly when coupled with a startle situation or a sudden ramp up in a tight tracking task. Flight test pilots have seen it, and the NRC even stated that there has not been one FBW or augmented hq aircraft that has not had at least one APC event during either flight test or in operation.

Goldenrivett

Better tell BAE they are wasting their time then.

"The biggest drawback of the “passive” sidesticks now used in civil aircraft is the lack of control feedback from the aircraft or the other pilot."
"This ensures both sidesticks move together in response to both pilot and autopilot commands, providing crew situational awareness equivalent to conventional pilot controls, says BAE."
"But the biggest advantage claimed for active over passive sidesticks is that both pilot and copilot controls track each other at all times. “This provides good situational awareness between the pilots—they can both see and feel their sidesticks moving,” says Taylor."

Active Sidestick Controls Make Commercial Debut | Technology content from Aviation Week

Do you have any idea about multi-crew operations?