Turkish airliner crashes at Schiphol

[PLovett]

Just to play devils advocate here in relation to the THY crash.

Much has been made here about the length of time that the throttles were at idle and that the crew should have been aware of the fact. Wasn't it the case that the GS intercept was from above and consequently it was likely that the throttles would have been at idle during that time?

Wasn't it also possible that the speed was decaying, initially from above Vref so again the crew would not have been surprised to see that situation? Yes, they should have noticed that it was falling below Vref and IIRC the FO did as the throttles were manually pushed up but he then let them go and they retarded to idle yet again.

I don't disagree that there was a lot that was wrong in the cockpit that day (systems knowledge; cockpit gradient etc.) but I think the time that the crew had to realise something wasn't right was a lot shorter than most have thought. However, I don't fly a 738 and quite happy to be proven wrong.

[DozyWannabe]

Originally Posted by hetfield

"Don't touch anything"
"Fully automatique"
"All by Puesh buetton"

With all due respect, I'd be interested to know which Airbus document contains such a philosophy, because it's not in any I've read (and for a non-pilot I've read a *lot*).

The push for as much automation as possible came from *airlines*, not manufacturers.

Originally Posted by RAT 5

One wonders if the policy of many airlines to use relatively (2 years experience) inexperienced F/O's as SFI's is correct.

A fair concern, but in this case the check Captain was very experienced in terms of hours.

To an extent I'm with PLovett - while it's understandably tempting to want to blame lack of experience or overreliance on automation, in this case we have a situation which is more complex. What piques my interest is the fact that at the precise point the autothrottle rolled back, the crew were expecting a reduction in thrust as part of the approach. They may have seen the levers moving in their peripheral vision and assumed that was what was happening, but none of them double-checked just how much power was being rolled off...

[Giolla]

These are political statements and have been made in order not to inflame and insult the feelings of the large Turkish population in Holland. No more, no less.

For the very same reason the Dutch prosecutor did not take measurements against the Turkish Airline. All political. The same counts for the Dekker Report, also here the utmost consideration for the extremely sensible pride of Turkish nationals.

[Pepperseed]

Pilots at that stage of the flight should have had theirs hands on the controls for quite a while to have gotten a feel of what she was "doing". This, done as a routine practice, would have educated the pilot at what normally happens in that particular phase of the flight in terms of "controls behavior". A prolonged idle position of the throttles at such a distance of the runway would have been automatically detected and would have triggered a reaction without needing any kind of intellectual stage of system analysis. Basic airmanship. So yes Plovett theory is acceptable in theory, but has to be totally dismissed as an acceptable excuse for the cock up. A pilot has to "feel" what his/her aeroplane is doing and react when things don't go according to plan. Click click click.... End of.
At the end of the day pilots have to be taught the right methods and not be systematically "replaced" by machines or have their behaviors "mechanised" by "covering arse" driven SOP's
ffs...

[RAT 5]

There were, I suspect, many visual clues that something was as it should be. OK, the T.L's were at idle, but think about the attitude to maintain the G/S at such a low speed. I'm surprised they could even see the rwy. The attitude should have been 0 - 2 degrees. It must have been much higher than that. This was masked by the F.D's being centred; and if the F.D's are centred then everything MUST be ALL OK; so said my instructor, I think? Same was true of the Airbus Air Inter crash at Strasbourg. The one where they dialled in 3300ROD instead of 3.3 FPA. They had 5 degrees nose down, in the brown, yet no alarm bells. Both accidents were on the automatics, so it's not a flying/handling problem, it's a lack of appreciation of performance parameters in any given scenario. That could be enhanced by less FD flying, or better training in knowing what the attitude should be and looking THROUGH the FD instead of looking AT the FD. If this has not been instilled at the TQ stage it never will be. I see this every day in the sim and my students are hammered with it. We've seen it will save your life, and when they question this philosophy I quote some the relevant crashes. They catch on quick afterwards.

[Pepperseed]

Rat5

Excellent point about the pitch.

Not a handling related problem ? I disagree to some extent.

A pilot used to fly manually/raw data develops an instinctive scanning reflex and he/she instinctively continues to use it even when he/she kindly lets the autopilot have fun with his/her aeroplane. [agreed, in a less agressive way, but still...]. So the detection of a wrong pitch at an early enough stage would have occurred, had the pilot been of the "flying-type" category. That's another angle whereby a reasonably good handler would have saved the day.

[fireflybob]

Whilst I agree that pilots should be aware of raw data attitudes would it be beyond the bound of possibility to have the FMC (or other device) to compare actual a/c performance against the desired theoretical performance?

We accept that a/c have stall warning systems so why not a system that auto calls "attitude" after glide slope capture when pitch attitude exceeds a certain threshold?

[RAT 5]

"We accept that a/c have stall warning systems so why not a system that auto calls "attitude" after glide slope capture when pitch attitude exceeds a certain threshold?"

There is; it's called 'the pilot monitoring, or PNF'.

[fireflybob]

There is; it's called 'the pilot monitoring, or PNF'.

RAT5, yes I am with you all the way there but if we accept there are many children of the magenta line flying around these days it might be one way, as well as raw data training and emphasis etc., of ameliorating the hazard.

[CDRW]

Many many moons ago, when I learnt to fly one of the many "sayings" being thrown about by the old boys was:
" Laddy, keep thine airspeed up - lest the ground rise up and smite thee".
How apt

[fdr]

Cockpit crisis

In five years, over 50 commercial airplanes crashed in loss-of-control accidents. What’s going on?

MT

Agreed, but OTOH, (looking on the bright side, there is a Darwinian process at play... )

The failure is a loss of direction that has resulted in the capitulation of the standards based flight operations to the acquiescence to the beancounters and politically correct crowd that pervades flight operations management. Not the flt ops guys fault per se, they are the result of the selection process that has developed from the invasion into a critical process by those of the MBA set, in a world of deregulation and the associated arbitrage of standards that occurs.

Instead of concentrating on quality by maintaining the training standards and using the technology enhancements to improve margins, the standards have been allowed to decline on the hope that technology may save the day. As you generally are dead for an extended period, I would prefer the use of technology to enhance standards by maintaining the core skills, but the bean counters, and by inference, the passengers, would prefer to play Russian Roulette with their lives to save a few pennies/shekels/Bhat/Rupee as may be applicable. Reading the available chicken entrails, it still appears that the issue has not got to the point of pain in the publics perception that is necessary to turn the SS Titanic around, so there are still icebergs ahead. The irrational state of the industry in respect to capacity expansion/yield decay remains in ascendency.

Re Rat5/attitude is a give away of speed... true enough, but the downside is that you actually have to know enough in the first place to recognise the changes, and more problematically, the initial speed decay/attitude change is not large order, but diverges from normal rapidly as it is an unstable process in the case of TK1951... it accelerates rapidly once it has been given enough time to get out of sorts. Little deviations are triggers to bigger and wusser things portended.

[PJ2]

Any medium transport aircraft (B727, MD80, A320 etc) on a routine domestic flight of around 4hrs will almost never have a Vref lower than about 110kts. IIRC, the Vref for this approach was around 140kts. The speed was unmonitored for about 100 seconds as it dropped below 100kts. So along with pitch, the "space" on the airspeed tape between the bugged speed or Vref and the actual speed would also provide a warning.

Great post, fdr.

[ATC Watcher]

Great text fdr, very true, and not only in the flight ops area. Everywhere in fact.
I would like to be optimistic believing this is only a pendulum effect, and that the common sense will return the pendulum back.

The AF447 training issues, FR internal policies, and now 787 batteries might be some of the signals that the pendulum has reached its peak one way.

[RAT 5]

The old rule of thumb still applies:

Push forward, houses get bigger;
Pull back, houses get smaller;
Keep pulling back, houses get bigger again.

Same for the whizz-bang FBW thingies if the safety nets don't work for some reason. It brings a whole new chapter to the "what's it doing now?" pilots' handbook.

[SPA83]

Interesting to read

TK1951 SYSTEM FAILURE CASE STUDIES by NASA
System Failure Case Studies

Read also "What's Happening?" The Loss of Air France Flight 447 by NASA too

[AtomKraft]

Pepperseed.

Hands on the controls 'monitoring' the actions of the AP or hands on the controls flying the aircraft. One or the other!
When the Turkish was no. 1, I was no. 3. (a Transavia was no.2). I was hand-flying my aircraft- mainly because I thought we might get a bit of wake but also through preference.

CDRW.

So true. Never forget it. I never do.

[averow]

Well said...

Kudos on a succinct summary of what is creeping into aviation, engineering and my own field of medicine.

Young trainees are getting less familiarity with raw data/feel of controls in ones hands, scribbled equations, etc are are now much more reliant on what the computer screen or the magenta line is telling them.

When I am in my little Cessna looking out the window helps reinforce what my instrument scan tells me. There is a beautiful synergy at work that is sometimes missing in very complex systems.

I sometimes have trainees tell me that my patient is fine based on what the screen is showing them without looking once at the patient themselves.

I wonder if this is creeping into engineering design as well: if the sim or model tells us that all will work out then ones design must be OK.

[RAT 5]

"The children of the magenta line" should be compulsory reading for all training managers: and after that it should be deep learning for all students. There needs to be a club, "bring back the basics." I train so many cadets that joined because they like flying. They then embark on a TQ course and they realise that 'flying' is the least of their tasks. I then fly the line with them 6 months later and they are so bored and disillusioned. My god, it's a sad profession. It is NOT about flying IT IS A WAY OF LIFE. It depends on who you work for and what a/c you fly. IT IS NOT ABOUT FLYING!. Please, let the selection and interview process be honest.

[captjns]

New pilots can almost cite chapter and verse all their manuals. Pity they can't fly the same way. Can't really blame the newly released puppys however. It's all in the hands of the line trainers and how far they are willing to go to ensure their charges have a clue of the four fundementals of flight without utilizing the automatics.

[BARKINGMAD]

BAD ATTITUDE.

Isn't it amazing how far apart the THY and the AF447 accidents are in most respects, yet the same apparent cause?

CFS Stalling 1 exercise: Look Bloggs, here we have the symptoms of the approaching stall.

1 Low decreasing airspeed.

2 Higher than normal nose attitude.

3 Sloppy controls (sorry Airbus, forget that part, refer to threads on Sidesticks!)

4.And so on into the light buffet or stall warning.

What's happened to the industry that such BASIC FLYING TRAINING is being ignored in our shiny computerised people/cargo movers?

Come out of the office, management, and defend your obsession with automatics at the expense of loss of control events?