RAT 5 : No lack of that on this forum.

Asleep at the wheel
 

How asleep do you have to be to get all the way to Vref-40 on finals? I appreciate the workload is high, but seriously, all the clues are there, right?

BHDH.

Glass houses/ stones and all that, but . . . . well, Yep, difficult to justify.
I think EVERYONE who flies any aircraft for a living, is waiting to hear some one come up with an explanation of how they ended up so far out of the ballpark :confused:

What time was it after us old guys retired did procedures not require someone to be monitoring the automation? I never trusted an autoland approach enough to take my hands off of the controls and disconnected at the first indication of deviance from what was desired. Have the new procedures changed? The manufacturer should not be held responsible for this accident because the #1 RA wasn't working properly. Two people were asleep at the wheel.

Re “…monitoring the automation…” (#2225)
Part of the problem is that current teaching involves monitoring automation, whereas monitoring the aircraft flight path as if it were being hand flown by another pilot should provide a clearer understanding of the situation.
It is necessary to check FMA, RA displays etc, but in sensible priority to the larger picture of what the aircraft is doing;- attitude, airspeed, altitude, hdg/trk – a robust basic instrument scan.
With time available, where the scan includes subsidiary items, it may be possible to project deviations or failure from faults (thinking ahead), but with time constraint – high workload, training demand, inexperience, then we have to rely on a basic scan pattern to detect aircraft deviation from prescribed limits. If automation is the cause, correct or disconnect it. If evaluation time is short, don’t let the situation determine the future – you do that, intervene in the situation with manual flight.

Stop speculating, please!
 

"There is no blame to the automatic system, this accident is down to a deep, deep failing of the human factors element- that is where the serious failure lay. We rely on human situational awareness to keep us alive- the way that failed here is of despair to all pilots." - Rainboe

Yes, maybe if we blame and criminalise pilots we can begin to do someting meaningful about safety! really? And what actually is situational awareness? People throw that term around so much it looses all meaning, peoples interpretation of it becomes so subjective. You are so caught with the view of hindsight and you fail to consider how the situation may have unfolded from the perspective of the crew, at that time, in their environment, what were they aware of? It is too early to tell, so why speculate further? The HF experts have only just begun to look into all the factors, trying to paste all the messey details together.

But carry on ranting, go ahead, maybe if we scare enough pilots they will stop making UNINTENTIONAL mistakes. How will this encourage a fair / just culture or open reporting system? If you carry on voicing these views,we will all very soon all end up in a smoking hole in the ground.

I've known very few pilots who make INTENTIONAL mistakes.

Yes, a single driver can be distracted for three seconds, perhaps fatally.

But we are talking about a team of three drivers - all seemingly distracted for one hundred seconds - and no mitigating reason has yet been officially proposed.

Case closed? Of course not. But the official silence is disquieting.

Rainboe:
I agree with most of what you said.

But my opinion Is that nobody will do such mistakes intentionally, not a 14.000+ Captain, nor a 70.000-combined crew involved in a maintenance check flight, but they happened. Incompetence ? Maybe, but those accident happened unintentionally ! Accidents happen also to truck drivers and bus drivers (professional drivers), and also in that case they should not happen, and this is what a bus passengers expect from them : to safely arrive at the destination. Survived professional driver/pilots will probably undergo a trial, and I agree with that, if they are at fault , they deserve punishment, as everybody else. But my point is that for most of those accident neglicence (culpable neglicence ? negligence ‘in vigilando’, a lawyer would say ? ) probably is the main reason, and I agree with you that this type of mistake should not be made by professionals – whichever type of professional.

You push a bit here ... they didn't wait for that one to take that direction.

Then, if human intervention or (non) intervention is not done from the cockpit, it will be done somehow somewhere from the ground ...
Much better to have someone at the front who does anything to save its own life than someone on the ground ...

A few post back I gave a fairly accurate descrition of the accident sequence.

I too do not believe the automatics can be blamed. We train on a regular basis for the most horrible scenarios, engine fires/failures etc. It can not be that a simple failure(I know it didn't fail but it wasn't working correctly) of the RA could lead to an accident.

However I do try to learn from it, could I have made the same mistake?

Once again, the only thing wrong for the first 70 seconds was the RETARD annunciation instead of MCP speed, other than that the airplane was behaving exactly as it was supposed to. How many guys would have caught that? And, once again the FO might even have noticed but didn't realise it was not appropiate for that stage of flight. I have once descended in VNAV wondering why the A/C was not following the profile, only later to realise the A/C was descending in VNAV SPD instead of VNAV PATH.

It's gonna be interesting when the final report comes out and a transcript of the CVR is released if the FP or the PM said "retard" and the other pilot said "checked". I'm gonna go out on a limb and expose myself to the ridicule of the pilots on here, but I in that stage of my training I possibly could have not realised RETARD was not appropriate for that stage of flight.

The speed decay from FAS to stick shaker is beyond me, really!

But the recovery, the P/L's going back to idle, hmmm, however it won't happen to me now!

In the end, it's gonna be pilot error though, no doubt, but why??

It has been my experience that whenever there is a large contribution of crew error in an accident that official comment is quiet for a very long time if not forever. In spite of this the lessons are available to be learned by other crews if only by inference (from the Boeing or Airbus info releases)

I believe that safety is still being served.

The only weakness is that those who want to believe that "it can't happen to me"

I have not read Learmount's article, but I still believe that we should rather be designing pilots INTO the system. Automation is a wonderful thing, and as some of us know, means we can nod off on along sector leaving 't'other to monitor the 'perfect' systems:)

However, I have never been so 'in the loop' as I was at 250'/420kts with a map in my hand or navigating an ageing 737-200 across the Atlantic to the Canaries on DR. Nowadays, we sail across the world with little thought as to how.

I have constantly argued for LESS automation and making the pilots do more. Yes, I love the modern FMS/GPS/magenta line stuff, but I know that lots of modern youngsters do not have a clue where they are and where they are actually going a lot of the time - indeed why should they? Why should they worry about speed when the aircraft will look after it for them? Why worry about stall recovery because the aircraft will NOT let you stall?

I firmly maintain that with the current state of the art in avionics, software writing skills and equipment we are going the wrong way in both design and training, and Schipol and Perpignan sadly re-inforce that view.

BOAC, I quite agree! To me (and I am sure to you and all others) a speed 10kts below Vref and reducing is not only "a bit unusual" but highly alarming to the the extent that I would be taking some pretty decisive action, I hope.

Rainboe, I agree with you too - whichever way you dress it up, your comments highlight the real cause.

No excuse for stalling an airworthy airplane on final approach . . .
 

The operating CVR will explain the crew's distraction from monitoring the instruments and from staying focused during the final approach in day VMC.

This shocking, elementary breakdown in cockpit discipline during a critical phase of flight suggests that Türk Hava Yolları has to revamp its training program, with emphasis on strict adherence to SOPs. :ooh:

Rainboe;
And Egyptair 990...

Your remarks are appropriately blunt and spot on the mark.

I am firmly in the camp of using a systems approach to flight safety where organizational factors are often ignored as first causes because these days that is an appropriate area of investigation given that human factors is by far the #1 cause of accidents. But at some point it has to be acknowledged that there is no other reasonable explanation for this accident than the captain, and the crew, did not do their job, and instead stalled their aircraft resulting in the deaths of passengers. There is simply no excuse available to an experienced professional crew for ignoring a 40kt reduction in airspeed for over a minute and fourty seconds on final. And I agree too, that Boeing has soft-pedaled this far too much. Why many are still dancing instead of looking at this for what it is, is disturbing for the very reason they give for dancing - "perhaps there is something deeper to learn here?" There is, and the sooner it is acknowledged the better: Why did the captain, and the two other pilots stall their aircraft?, and I am not pointing back to anything other than SA - what permitted the loss/absence of SA? Certainly not a radio altimeter fault, nor the warnings which resulted from same. The only possibility is a non-sterile cockpit environment, (which is also the subject of interest in the Colgan accident). The loss of SA is precisely where the lesson can be learned, if there are any lessons here at all. Again Rainboe - well focussed, well stated.

Well thanks but I pretty much disagree with most of your points.
1,2,3 and 3

Once they intercepted the GS the airplsne would start to reduce speed as one would normally do. In line operations you use a decelerating rating type of approach. the pl's would go to the idle detent, the speed would start to reduce and the angle of attack would increase. The engine parameters would also indicate the engines being at idle.

read the boeing publication. It took about 70 seconds to decelerate to FAS, during this time the pl would have been at idle just as they would have been during a normal approach. It was the next 30 seconds where things went wrong. that's were the 40 kts speed reduction below FAS happened(still inexcusable) as the previous poster claimes, makes it more dramatic though,I agree.

I'm done with this thread. It is starting to get to the point were people feel the need to post just simple oneliners to prove how good they are. Great, I don't have a need to prove anything here though. Some interesting things where to be learned but not anymore.

good luck with the discussion.

p.s. reluctantly I'm going to agree with rainboe, a lot of people seem to have a reading disorder.

Rainboe in response to your post about Turkey.
They are proud people, of their country and culture, it is also well know to be very corrupt when it comes to anything such as this, money talks, if it goes into the pockets of the politicians, government agencies or police.
I lived there for a while and was really surprised at some things.
What really shocks me is that some ambulance chasing lawyers is trying to blame the manufacturer for a flaw, never mind that 3 crew with 3 sets of eyes allowed the airspeed to fall 40 kts below VREF.
Then the press in this crash as others in Turkey are labeling the crew as some kind of martyrs.
People everywhere make mistakes, we all learn from them, but to take a mistake and try and blame others just to collect money wrongfully due is beyond me.
The R/A was a minor fault, not monitoring airspeed and altitude was a major failure which cannot be blamed on the aircraft or manufacturer.

Just ask some of the westerners with Orex Air cargo out of IST with the 747.
One Captain was fired for having the aircraft de-iced.
Several others were fired for asking for their pay after it was 2 months late.
They seem to have there own little world there with aviation.
Problem is that its does not comply with the way the rest of us are required to operate.
Maybe Turkish airlines is the best of the rest, but still something is wrong there.
http://www.pprune.org/freight-dogs/3...x-cargo-2.html
I do know some of these guys, everything posted is how most of the operations work there.
If you don't fly against the MEL or bust FDT limitations you can go home.
But this is Turkey!!!!!
If you can get away with something you are not considered to be wrong, just clever!

While more or less all agree that a lapse in basic airmanship allowed the unnoticed left RA problem to trigger a series of issues, it does seem that it would be a good idea - and easy to achieve - to incorporate a system to flag up and announce a mismatch between the two RAs. It went unanswered, very early in this thread, that the left RA could only flag a failure but not an error while the right RA was sitting there waiting to have its measurement compared. Being electronic instruments, rather than pressure-operated, should make a comparison system relatively easy and cheap - both as original equipment but also, now, as a modification.

flyburg

Good point – got myself totally confused there between events pre- and post-FAS - so now your point makes complete sense! My apologies!

Was just having a break while Rotavating the lawn! Exhaustion led to complete brain failure! That’s my excuse. :O

Anyway, my Post now deleted as it is clearly bo**ox. :rolleyes:

As you said, from FAS down to the shaker – that’s where it all did not add up!

Note to self - do not Post when shattered. :=

Off for a bath and a beer! :ok:

H ‘n’ H

No worries,

Been there, done that.

:ok:

Automation failed the pilots.
 

I feel ,like the naked emperor but you proffessional pilots insist on hair shirts and blaming the aircrew.

As a proffessional control engineer (industrial) I have no hesitation in saying that the automation concept was prima facie to blame, and I am ashamed that such a tragic system was designed.

The automation FAILED, the pilots were RELYING on it and you must always be able to rely on it,until it detects a problem,announces it and relinquishes control. Any system which does not do this shoud never be installed in anything serious.

If the trim in my puddle jumper motors uninvited to full extent while I'm on a bumpy crosswind sun in eyes final over turb ridden sheds am I to blame for losing attitude control?
NO, I blame the b faulty switch and miniscule indicator,and more importantly the CONCEPT which motorised away my absolute authority over pitch,while I was otherwise occupied.

The architecture of this 'automatic' control is pathetic.
One RA should be in control of all parameters,any deviations would then affect all controlled functions,and there would be no need to spend invaluable time and distraction in trying to puzzle why some functions are OK and others incorrect,spurious alarms going off etc due to RAs inputting to different elements of control,and the pilots understandably unaware of what controls what,probably 'cos no one told them.

The other RA should input check data and when the two RAs differ beyond a certain amount,zowie,announce problem,hand back control.

atceng (retd)

:ugh::ugh::ugh:

This whole RA thing obviously offends so many electronic design types who are posting.

The serious failure case is covered by Boeing - you cannot engage both autopilots for a Cat 3 approch and autoland if the RA system is not operating as it should. Any fault in the RA system at any other time is, to use Rainboe's word, trivial. Please understand that. There is nothing wrong with the system.

There are countless, possible trivial failures that can inconvenience an approach. Crew deal with them every day - it is part of the job. Maybe the windshield wiper packs up in heavy rain, maybe (God forbid) the autopilot drops out on intercepting the glideslope and you actually have to fly it yourself. In this particular case, disconnecting the autothrottle would have done it. It was, after all, telling them with a RETARD caption and by not applying power to maintain airspeed. What more warning do you need? The autopilot would have stayed connected. There was no technical problem. The problem was the crew not monitoring the approach - full stop.

Hopefully we shall learn why they did not monitor the approach but to criticise the RA system is not helpful.

atceng;
Well, you need to do just that: hesitate and think. Before launching, you need to understand the nature of the failure here. Radio Altimeter "failures", while not a regular occurence, are, and should be, a complete non-issue to a professional airline crew. It is, please understand, simply a non-event.

What is serious and has not been addressed by engineers but has been addressed by airlines, (in terms of emergency SOPs), is blocked pitot tubes and/or static ports. That issue will cause and has caused accidents; this very issue wrote off a perfectly serviceable $2b B2 bomber if you'll recall. That is an issue worth discussing; this one is most certainly not except for why this crew didn't fly their aircraft.

The plain fact is, this crew did not do their job; they abdicated their professional responsibilities and killed some of their passengers, because they stalled their aircraft, because, for whatever reason, they did nothing about a degrading airspeed. As someone else said, the CVR will tell the story but most bet we'll never read a transcript.

This crews' failure is in the same categorical nature and quality of professional failure that an engineer would be if s/he were to ignore/forget/mistake material strengths or span fracture mechanics. It is truly no more complex than that.

If you're referring to the Birgenair, that accident has an interesting twist/component to it as well which is left to others to discover.

Yes, basic airmanship was absent in that accident. The Aeroperu accident was more difficult to handle in terms of basic airmanship. That said, there were ways out for both, but both were far longer shots at success than this crew gave themselves and their passengers.

Quite the opposite Rainboe !

No crew in that aircraft and automation brings it down all the same.
That crew also failed but how many crews would have properly intervened.
I believe every single day pilots take over automation but as they do nobody hear about it.

The answer is not by keeping the pilots out the cockpit but by limiting the level of automation and by proper and regular training.

Glueball said:

That one statement pretty much explains why these type of accidents occur, someone screws up and we try to go back or invent a procedure. Procedures will never substitute for awareness. Another cry that is heard is 'monitor the automation, read the FMA'. They are both inherently wrong statements too. What should be done is to check the automation mode desired is selected and then 'monitor the performance instruments' to make sure the aircraft is doing what you wish. Whatever happened to 'selective radial scannning'? Basically the introduction of the PFD!

sticky: Turkish Crash Forum
 

I am having trouble keeping up with his Turkship, and his incidents and accidents, is it now time for a Turkish sticky...........

@rainboe

As you have (I think) previously suggested, there may be no helpful answer to "Why?".

It may well be there was simply a 10^-9 simultaneous and sufficiently protracted failure by all three pilots to remember what they were there for.

Why would that shock you? That is exactly what "ambulence chasing lawers" do - go after the deep pockets.

While I broadly agree with the feeling of shock that an accident like this can happen, doesn't the history of such accidents show us that we will always react like this? (In judging the crew harshly). "How could they do this" rather than "there but for the grace of god...." Until the next one. Or even worse, until we ourselves make a 'human error'?

Isn't this the whole point of human factors - that we are all fallible? When the holes in the cheese line up.............

If I've read the Boeing publication correctly, the speed decayed from normal to Vref-40 in 30 secs. We don't know what distractions occured during those 30 secs.

This is not to excuse negligence if that is the case, but on the other hand, can any of us truly say that in our whole flying career we have never been distracted in such a way? For 30 secs?

I agree that a rad alt failure is a total non-event - I'm sure I've had loads of erroneous displays over the years.

But like all similar accidents, it's the subtle interplay of all the factors that bring us inexorably to the disastrous consequences.

At the beginning of this thread I felt more judgemental on the crew. Now I feel 'who knows'? We need the specifics of the distraction in those 30 secs when the last holes in the cheese lined up to truly learn from this accident.

I'm sure the fact of it being a training flight will play a big part.

Yes. But where does fallibility cross into negligence? If you drive down the road at 100 mph on the 'phone while looking for a track on your iPod and have an accident, is that 'human factors'? Both drivers and pilots undergo instruction/training and are expected to reach a certain level of basic competence like: "don't let go of the steering wheel" and "always fly the airplane".

At 100W, possibly. On short finals to a busy airport in IMC? No, I haven't and I expect the majority share this view. Close to the ground the flightpath of the aeroplane needs continual monitoring, which there appears to have been a complete absence of in this case.

When I read incident/accident reports, I almost always learn something useful that hopefully stops me following in their footsteps. Going with what information has been released so far, this accident appears that it will contribute precious little to the sum of aviation knowledge, apart from the extremely obvious.

It is pointless discussing whether or not we 'super-heroes' have or have not been distracted for 29 or 31 seconds on finals (borrowed '!'). We are all human (well. most of us) and surely the lesson here is that IF there is such 'distraction', we must not allow it to eliminate the monitoring of the 'vital signs' of aviation life. If it does so, the only remaining course of action is to break off the particular manouevre which we are attempting.

In the case of Schipol, if there were such a distraction, it must have been significant to stop 3 pairs of eyes from seeing, in which case one reasonably expects a TC to grasp the nettle and take positive action.

You are generalizing.
There are only two countries with a Muslim majority

Where I believe it will contribute however, is to the sense of how much distraction you allow. We all get distracted and as many on this thread have said it's a question of how long before we come back to the duties at hand.

After each such episodes of distraction in the future you can now think about how long it took you to come back to the critical tasks and judge your own performance thinking about this accident.

Let me venture into the lions den. Automation has made all our lives safer and easier in many fields and is here to stay.

Rainboe is of course right, there are certain things that are so basic that they should never be lost sight of. There can never really be an excuse for this.

On the other hand it does seem, from all the comments here that in highly automated multimode systems, mode confusion can be a serious problem for the operator, particularly in unusual situations.

In mode A then X, Y and Z are attended to, in mode B, X and Z but not Y, in mode C, X alone and so on.

This demands constant full awareness of the current mode, what it does, when it will transition to another mode and why, and what will happen when assorted inputs are degraded either individually or together.

I would venture that such constant awareness while monitoring the automatics is less "natural" than the immediate focus of hand flying where matters may be more predictable and immediate.

But obviously this is part of a modern ATPL pilot's training.

Did the aircraft meet the stabilized approach criterion at 1000 ft? I think this crew "lost it" well before those 30 secs to Vref-40.

The only stand-out item would probably be the TL's being at idle annunciated only by the RETARD caption, speed and configuration would likely have been within limits at that point.

If bitchin betty could be programmed to issue a stabilized approach criterion warning - "UNSTABLE APPROACH - POWER LEVERS IDLE", that could have been enough to wake them up and save the flight and some lives. But the suggestion has been shot down already on this thread by the purists who want to see the crew swing and lowly Boeing shielded from any implication that their Auto Thrust design on the 737NG might be less than optimal.

Ok, lets put a spin on this accident!
If the Turks, Lawyers etc are saying its a Boeing design defect and not pilot error.
What will happen to the previous Turkish Airlines crews that did not report this problem as the FDR showed, and nothing entered in the maint logbooks?
Two sides to every story then there is the truth !!!
Points right back at them no matter how its seen.
Bad Airman ship, little regard for the basics, sleep at the wheel.
A good Boeing lawyer will eat them for lunch.