Hi,

It was already some discussions a bout the Bonin license (and I was part of it)
That Bonin had the correct license or not is an interesting point to be argued in a court of justice
However about the accident itself this point seems futile since in addition to Bonin were present two other pilots whose license is not challenged and despites their presence .. we all known how this flight ends ....

So what is being said is that there is a doubt that the co-pilot flying the leg as acting Captain may not have had the correct licence or qualifications to fly this aircraft?
Is it possible that the PIC (Bonin) at the time of the accident was not qualified?

In the aviation business all is possible .. even faked licenses (that's few discussions about in this forum)

As far as I know, there is no question that Bonin was suitably qualified. It is covered by both the BEA report and by various contributors to these threads who included both present and retired airline captains.

What I and others have wondered, to which it is unlikely that any answer will ever be forthcoming, is why the commander left it until he was about to depart from the cockpit for his rest period before he sought confirmation from Bonin that the latter was indeed so qualified.

Unless somebody has evidence to directly challenge the investigation into licensing that BEA conducted, there appears to be really no reason to continue to cast doubt on this.

We are approaching the day of automation where aircraft will not have qualified pilots in the cockpit. Unfortunately AF447 is the best example of this so far.

Automation and the implications in real (rich) world
 

We are approaching the day of automation where aircraft will not have qualified pilots in the cockpit.

We need to put automation CAREFULLY in our designs. Keeping in mind the K.I.S.S. principle on the man-machine interface.

IMHO the current approach (and the trend) is dangerous and require much more from the pilots than before.

The timely "understanding" of certain situations is uncertain and this must orientate us seriously.

The "stuck" THS in this case is an example of the complexity of the issue.

I don't think it is the captains responsibility to verify the other pilots are qualified. That is the airlines responsibility knowing all three crew members have to leave the cockpit to have legal rest.

Has anyone on this forum ever asked another pilot to verify they are qualified? Didn't think so.

Also saying he was fully qualified but still pulled the SS up into a deep stall because he lost airspeed verifies he wasn't qualified except on paper. A few copilots I flew with were legal but not safe so I never left the cockpit. He sounds like one of them.

We are not in a pilot shortage quite yet so why do we hire incompetent pilots when we have competent ones available? Maybe the pay scale? I knew on my first day on the job I could single pilot the 737 if needed with no problem. Now all that seems important is the bottom line of company profits so pilot ability is not important, just pass the minimum standards.

I have a friend in the San Francisco Bay area that is in tune with this that will agree with the need to use pilotage, not automation, to safely fly an airliner.

Sully was a good example of how the Hudson incident was an example of how piloting skills are very important and the automation that day was worthless because it couldn't handle it.

Sully did what all of us should be able to do but if the trend goes to automation and we forget how to really fly the next time will be a disaster.

I fully agree with what you are saying. I also believe its not the Captains responsibility to ask if his flight crew are qualified.

I do believe however that Air France and all other airlines have a responsibility and a duty of care to all passengers and their crews to ensure that all personnel flying for them are fully qualified and trained to operate their respective aircraft in all conditions of the flight envelope.

How Air France get this information concerning qualifications of other crew members to their Flight Captains prior to pre-flight briefings is another matter.

First of all, the autopilot on US1549 was never engaged, so saying "the automation that day was worthless because it couldn't handle it" is pretty disingenuous.

Secondly, Sully's A320 remained in Normal Law (powered by the APU and RAT) throughout the ditching, with all flight control protections active until splashdown.

Great piloting yes, but he had some help.

thehighlander
 

You are right, but there's also a certification issue. No-one seems to think it odd that it's legal (certified) to have an autopilot that under certain conditions in cruise gives back control of the plane to the pilots. The same conditions in which the pilots are not trained to fly the plane.

It was made worse (probably made fatal) by the crew making no attempt to follow AF's clearly set out procedure for 'unreliable airspeed'.

Are you saying that you think that an autopilot should be designed to remain connected when a key piece of data that it uses to fly the aircraft is no longer available? I sure hope not. Why not suggest that it's okay for the autopillot to continue flying a CAT III approach even when the ILS signal is missing ... :confused:

JO, may I suggest that, in response to a problem, you are putting up an Aunt Sally of a wrong solution?


As many have suggested (myself and several better qualified people), what was needed was better training for the eventuality of “Otto” handing it over as it could not cope.


If pilots have no training or practice in hand flying at mach 0.8+, and FL 350+, and none in UAS at such levels, it is not surprising that it can go wrong.


We have also pointed out the poor CRM, lack of adherence to SOP and/or QRF, which to my mind also suggest inadequate training and testing.


AF, and/or the industry, and/or the regulators/ICAO, need to look at these issues and change things for the better.

Almost 3 years after the accident AF447 .. what are the real changes implemented for cope with those issues ?

I don't have an aunt Sally so I'm not sure what you mean. In any case, I was not disagreeing with the need to ensure that pilots are adequately trained and prepared to handle a situation like that faced by the AF447 crew. But the devil is in the details and to suggest that the autopilot going off was somehow a bad idea is leading to the wrong conclusion, IMHO. If I used a bad example to support my position, then it is what it is.

This entire discussion raises the very complex issue of knowing where your next accident is hiding. For all the talk about SMS and proactive risk management, what is not being addressed adequately is the fact that accidents such as AF447 are difficult to predict, and therefore difficult to prevent proactively. I've been in this business a long time and around Airbus FBW aircraft for close to 20 years, and I would never have envisioned that a pilot would respond to those specific circumstances in the way that he did, nor that once it happened, a crew would be unable to get out of it with over 3 minutes to react. If the investigation doesn't help to determine why this is the case and why it made sense to them at the time, then the outcome will be wholly unsatisfying.

Alas, I fear that we will be left unsatisfied. Instead, what we may be left with is the same problem we've faced for decades, that of having to make "tombstone changes" to the system in response to a bad event. This thread is littered with accusations that this accident was entirely preventable, but I am just not sure that this is a fair analogy.

Regardless of the "laws" he did not splash down with 13 degree ANU either did he? Do you think if that experience was on board AF447 it may have turned out a bit differently? I like to think so. Help or no help, he knows how to fly.

Comparing this to Sully's flight
 

I think it does no good to compare these two flights, even though Cap't. Sullenburger did display great airmanship in the face of no-time-to-think.

Besides the aforementioned A/P issue, Sully had perfect visibility in the daytime, so situational awareness was very good. On AF447, nobody could see nuttin', much less the horizon. Not to say that training wasn't deficient; it clearly was. :\

So far, there are no serious suggestions that the IR data 'went topsy-turvy', apart from a few conspiracy maniacs.

So yes, they could see the horizon... it was right in front of them on the instruments.

Suggesting, indirectly, one can't fly at night through the ITCZ unless being VFR, is slighly fatuous, no?

ChristiaanJ typed:
Well sir, though I take great pride in being fatuous at times, in fact I am a master at it, I say, "No!" I only mentioned this fact as a point of comparison between the two flights, AND...it has been oft-mentioned here by people with much greater qualifications than I, that the AF pilots' inability to see the real horizon may have contributed to the lack of situational awareness in the cockpit. Wasn't MY idea! Certainly, they weren't able to "fly at night through the ITCZ unless being VFR." I'll be online all morning awaiting your groveling apology. :8

By "people with much greater qualifications " I suppose you mean people who have qualifications.

In the interest of keeping the repititous character of these threads to a minimum, it might be worth letting those who have qualifications keep their opinions as their own, rather than having those of lesser or no qualifications co-opt and regurgiate them at far too frequent intervals.

That's the second tragedy of AF447. The first was the crash itself. The second would be if the lessons that need learning, and then applying, are not. Which would lead to a third tragedy, namely, a similar and preventable crash sometime in the future.

But based on some of the input in four of five extended threads on this topic, crews (in some airlines) seem to be on the receiving end of systems and situation awareness, and even some practical (simulator?) training.

That's a good first step, don't you think?

Just think of it as a 'dig' along the lines that "maybe the solution reached was logically flawed".

Comes from a maths mnemonic "Please Excuse My Dear Aunt Sally"; Parentheses, Exponents, Multiplication, Division, Addition (and) Subtraction.

Put another way, it could be said that those who remember it have been around too long!

at easy we are practicing what we think happened in our simulator check. Having had it explained on the white board, demo'd in slow time, knowing what was about to happen, I have to say that it is the most confusing set of indications I have ever seen.

How many of us Airbus pilots realised that the STALL STALL STALL silenced below 60kts? (and why would we).

I understand the theory of what happened now, but there is no way I could of comprehended it happening in real time the way it did to the poor souls on 447.

The understanding is coming out, but of course a lot is best guess.

Who would think any pilots could find a way to put an Airbus below 60 knots airborn? I don't think any aircraft is idiot proof.

Nothing indicates the RAT was deployed during the flight, either automatically or manually, and had no reason to be.

Do you think that experienced pilot needed help to not stall ?
Remaining in Normal Law prevented that experienced pilot to manage the flare at his convenience.

But maybe better to go there A320 and the Miracle on the Hudson to discuss further.

The RAT doesn't deploy unless you lose both engine generators in all aircraft I have flown and the APU would not have been on for that short flight because they could bleed air off the engines. I didn't bother to respond because I knew most people knew.

Also the autopilot being worthless meant how could it have had anything to do with the outcome of the Hudson landing. He was a highly skilled pilot and the autopilot did nothing to help him. He didn't need stall protection with his skill landing in the Hudson. None of us in any Boeing would have stalled so why did it help him? It didn't. Why do Airbus pilots always think the only way to fly is by automation? I don't understand it. Now I read an article about letting automation rotate the airplanes to prevent tail strikes. Have you ever had a tail strike? 23,000 hrs and zero.

You obviously "don't know" because the NTSB report very clearly states that the APU was started soon after the bird strike. Section 2.3, [Analysis of] Flight Crew Performance, 2.3.1 Decision to Use Engine Dual Failure Checklist:

My point is that "the automation that day was worthless because it couldn't handle it" is not a correct statement - the autopilot was never engaged, obviously, but the 'automation' in the sense of the flight control laws, remained active and assisted a superior pilot in handling the aircraft in a superior way (as is how an Airbus should be operated: by excellent pilots who know their aircraft...not without pilots, or by pilots without excellence.)

I recall them starting the APU but since the engine generators never went off line it never came on line but it was a good back up. The engines never totally failed, they just couldn't produce thrust, the engine generators were the only power supplying electrical power when they ditched. The APU would have picked it up if they had failed however. It was a good decision.

The aircraft (Airbus 330) itself dynamically never got 'below 60 knots', only the sensors associated with the air data system.

With minimal 'handling' demarcation between stalled & un-stalled other than the turbulence masked buffet, the speed at which it 'gently' mushed into the stalled condition was somewhere around 185KIAS, at a point in time where the KIAS on one or more displays may very well have been temporarily available to them, as valid air data was intermittently recovered at numerous points.

Without re-reading Interim #3, I believe it settled in somewhere around what would have been equivalent to 150KIAS on the displays had the air data sensors not been corrupted essentially by actual flight path versus body angle.

This is the reason for pages & pages of discussion about full time stand alone AOA availability, not connected to the ADR's and their 60 knot threshold for validity.

(Correct me if I've stated something in error here.)

The a/c stalled when AoA exceeded 10° at 2h10m57s. At that time the F/O was seeing IAS from ADR3 which was still UAS, while the Captain PFD was on ADR1 which had returned to normal some 20s earlier and was showing 207 kCAS at 2h10m58s.That's not in Interim #3, but is probably based on a graph I posted some time ago that showed CAS calculated from the recorded groundspeed.

What can be changed assist in avoiding this kind of accident?
 

This is a very long thread now. I am perhaps foolish to add to it further, but here goes...

It is clear that the pilots on the day (or night) were indeed idiots. However, I think it is fair to say that we are all capable of being idiots on our day also. It is desirable but impossible to assure that every person has the complete skill set or training to deal with every eventuallity.

Therefore, I think that even though working on better training and weeding out bad pilots is of some assistance it will never be able to ensure all pilots are adequate, properly trained, or will not have lapses in there handling of situations.

Notwithstanding the above, I think there are two things that could and should be changed that may well have given the crew assistance.

First, when the autopilot disengaged there should have been an explicit message that it did so because of air speed indication anomalies.

Second, clearly the stall warning should continue regardless of airspeed except in specific circumstances ie if you are airborn at significant altitude there really is no reason for the stall warning to be discontinued...

These matters are easily corrected and are clear failures in logic. No one will ever know if these changes would have saved the fateful flight but I am sure that they would have been of some assistance in overcoming the confused state that the pilots had fallen into.

In addition, "Pull up...Pull up" was perhaps not as helpfull as it could have been in that it was exorting the pilots to do exactly the opposite to that which they needed to do. It was in the event less helpful than "don't crash...don't crash" would have been. What it was really trying to say was that they were loosing altitude and getting close to the ground.

A careful review of the messaging that pilot get is needed in order to at least avoid ones that are not helpful and ensure that helpful ones are not disabled.

HN39: Thanks for the correction. I personally believe your graphs are probably very close to the event as it unfolded.

edit: BTW on page 92, it shows max altitude of 37,924 reached at 2:11:10 with ISIS CAS of 184 recovered 3 seconds earlier, so they were still climbing, at least ballistically, from 207 down to 184. The benign 'mush' started at this time??

Report: Cobham B712 at Kalgoorlie on Oct 13th 2010, stick shaker on two approaches

Humans behind the plane instead of ahead. :(:confused::sad:

Cheers.

Yes, at 2:11:10 the AoA is well past the maximum lift and the LF has dropped to 0.75 g. One could say the "mush" starts as early as 02:10:52. Up to this time the normal load factor (=lift, essentially) rises commensurately with AoA (page 43), then starts to flatten off while AoA continues to increase.

AF 447
 

I find it very disturbing that so much reliance is placed on computerised flight systems which apparently are not fully capable of informing the ‘human’ pilots of the aircrafts actual status in flight.

Yes, there were audio stall warnings on the flight deck of AF 447, but did any of these warnings advise the pilots to decrease AOA?

No.

Why not?

Multiple audio warnings and multiple ECAM messages on the flight deck only serve to distract the pilots from the job in hand. Passenger aircraft should be designed to give the pilots the option of disabling all automated systems and of taking full manual control when required.

Pilots should be trained and tested to ensure that they are skilled enough to do this.

As I understand it, this is not currently the case..

When I and my family fly as pax, I want to be confident that the flight crew can actually fly the aircraft manually and are not totally reliant on it’s automated systems.

Can anyone assure me that this is the case?

Carjockey

That is what the STALL STALL audio warning is for telling the "pilot", reduce the AOA NOW!!!!! Why the "pilot" in AF did not do so we will never know.

We are able to do so! Why do you think there are two pilots, one is pilot flying and one is pilot monitoring who runs the ECAM. The PF only has to crosscheck Thrust lever / Fuel Cut off switch / Fire switch and any guarded switch that the PM has to operate. So there should be NO major distraction. As PF you have to FLY the A/C first and unfortunately the PF in this case did not appear to do so!

Most pilots are and most airlines ensure this is so, again we will never KNOW why the PF did what he did and why the PM never took control!!!

My only concern is the MPL, which is a whole new topic, this "training" will be fine and dandy for box ticking, but does not instill the life preservation skills that each pilot needs.

Thanks. And to quote that report:Birgenair, Colgan, AF 447, Ethiopian at Beirut, Perpignan.... and more and it'll carry on until someone figures out that it is not a "minor safety issue". :ugh:

[ and it's not a sidestick vs. yoke or control feedback issue either ].

Maybe not so much sidestick vs yoke, but FBW vs Direct Control. This FBW CR*P is nothing but an "arcade" game instead of having FULL control of your aircraft. Yes.....cable-operated. Old-fashioned...yes...but 99.9% reliable.

Birgenair was "direct control", as was Colgan Air, Aero Peru, Ethiopian and many more. But let's not allow the facts to get in the way of a good rant.

Yes.....BUT.....in this particular situation, I think they would have survived with Direct Control of their aircraft rather than some computer.