Turkish airliner crashes at Schiphol

[Ralph Cramden]

Where have all the pilots gone? A sad lament indeed. They have been automated out of existence. I'm afraid there are not many Sullys left in our industry. A man who took an automatic A/C and flew it. There still are a few out there who would have produced the same result given the same circumstances; but not many.
Most of the discussion on this tread has been on how to turn a fool proof A/C into an idiot proof one. Good luck with that.
Airlines love these A/C because they no longer need skilled pilots to fly them. They need operators who can program and monitor the systems that actually fly the A/C. Of course they have to be able to taxi, take-off and sometimes land. The latter skill often botched. Hopefully they will also be able to keep the blue side up when the gremlins show up.
There has been a lot of discussion about how hand flying is difficult, less safe and increases cockpit work load. Well I quess if you don't feel comfortable with it, these things would be true. It takes a lot of practise to aquire and maintain these skills. Someone suggested here that four sim sessions per year would be adequate!! Can you believe that! Sixty years ago (60) pilots were hand flying their A/C into Tempelhof in all kinds of WX using GCA. I have hand flown into most major US airports and a good number in the rest of the world inluding ORD, LGA, LAX, DFW, MIA, JFK, SFO, well you get the picture. Perfectly normal in my day. You were 100% in the loop and connected to your A/C.
Automation is here to stay. It will increase. The trick will be to involve the operator in a more active role than simply as a monitor. IMO the most simple and effective way to do this would be to limit the use of A/T. If the pilot is in charge of thrust he is forced into the loop. There have been suggestions here that all A/P's should use A/T all the time. Nonsense! Just as an aside, when flying in a mountain wave, I used A/B to control A/S rather than thrust. I found this produced a smoother, quieter flight.

Ralph

[TheKabaka]

Just for completeness (ref the airbus accident above). When doing a man land the airbus does not retard the thrust automatically, it has to be done manually!

If 1 TL is brought to idle it causes the AT to disconnect, the other TL was in the climb position and the thrust on that engine increased up to climb power.

[Nicholas49]

AnthonyGA - I feel compelled to respond to your ridiculous post. Please tell me exactly what level of experience you have of "running an airline"? Making sweeping statements like: "they [airlines] don't care whether pilots live or die, and they don't even care about the long-term health of the airline" really do not show you up in a favourable
light. A pessimistic view? Yes, absolutely, but also one of someone who clearly has no understanding whatsoever of what is involved in running a successful company. Back you go to MS Flight Sim.

[MU3001A]

This implies that if the Airbus was fitted with a RA-operated automatic thrust retard mechanism at 27', the Congonhas A319 would not have retained power on No 2 after landing?

This implies that RETARD is a necessary function of any A/T system, which itself is deemed to be a necessity presumably because it is safer than simply allowing pilots to control thrust directly.

[bsieker]

Originally Posted by Rainboe

This implies that if the Airbus was fitted with a RA-operated automatic thrust retard mechanism at 27', the Congonhas A319 would not have retained power on No 2 after landing? This is exactly how all those opinions here wanted the 737 to be redesigned! It wouldn't have run off the end. Unfortunately it might have crashed on final approach instead if the crew had gone to sleep.

Which serves to illustrate the point that automation cannot be made foolproof and is notoriously hard, if not impossible to get "right". After every accident self-proclaimed experts know how to make the aircraft better and safer. But every change has its traps. A320 (a "normal" A320 at Congonhas, not an A319, btw) and B737 accident records are comparable, and both are very low. Both incorporate different design decisions, but apparently both work very well. Changing something because of a single accident is rarely a good idea. Who knows how many accidents the design decision in question has prevented?

A320 has automatic thrust retardation in autoland. Pilot still has to pull the levers back and is reminded to do such by the infamous "RETARD!" callout, which is generated in both manual and automatic landings.


Bernd

[PJ2]

Flyinheavy;

Could it show a lack of training for this particular situation (1 T/R inop) or missinterpretation of eng contrl system, which btw is rather complex and as such providing traps.

This was covered extensively in the TAM accident thread. Neither manufacturer's thrust lever design is immune to technical or human error. Non-moving thrust levers are not an issue either. Before autothrust all of us flew non-moving throttles that we actually had to physically close at touchdown. There are incidents of not doing that for all types and not just the 320. They are rare, but they have occurred.

All automated systems are complex and are therefore open to technical or human error. I think that stating that such "provides traps" is like saying airplanes provide traps - both statements are trivially true and do not point either to problems or solutions.

Training is a key as I have said a number of times on this and other threads. I have said many pilots are afraid to disconnect the autothrust on the 320 - that is a training issue, not a complexity matter. The airplane flies beautifully in fully-manual flight - it is no more difficult to fly than any other type and no busier - if it is, there is a training/practise problem occuring not a complexity/technical matter.

The other issue is, as Bernd has eloquently stated, no automation can handle all human error. That is indeed where it can get complex. How to determine failure and arbitrate failure in two systems is a case in point. Every solution is attended by other, perhaps less obvious problems. I should think that all those suggested design changes which have been suggested here regarding various perceived system/automation faults have already been vetted by those who do this for a living and rejected for one or another good reason.

In other places I have suggested, not uniquely of course, that because humans are so poor at monitoring over long periods of idleness and that computers never tire, that there are some, not all, areas of normal operation which could be passively monitored with increasing intervention as "normal" degrades to abnormal. I know that there are significant problems with this approach and know also that the Airbus has already taken this approach in some, but not all, areas of aircraft operation.

[Maximum]

Oh for goodness sake people. This thread long ago became a parody of itself and is now about to disappear up its own proverbial.

Ralph Cramden, your post takes the biscuit.

I have hand flown into most major US airports and a good number in the rest of the world inluding ORD, LGA, LAX, DFW, MIA, JFK, SFO, well you get the picture. Perfectly normal in my day. You were 100% in the loop and connected to your A/C.

erm, yea, for most of us this remains true. We're still poling it down to the ground on those stormy nights when the autopilot gives up, and most of the rest of the time too. Those of us who fly into the more interesting places (and close to the UK that includes the Greek islands) are more than familiar with NDB's, circling to minimums etc.

So we take one unfortunate accident about which we know relatively little, and from this conclude that most airline pilots can no longer fly and simply push buttons.

In the words of our American cousins, do me a favour will ya?

[Rananim]

Maximum,
Hope you're right.But look once again at what RC said:

There has been a lot of discussion about how hand flying is difficult, less safe and increases cockpit work load. Well I quess if you don't feel comfortable with it, these things would be true. It takes a lot of practise to aquire and maintain these skills.Someone suggested here that four sim sessions per year would be adequate!! Can you believe that!

Over-reliance on automation degrades manual skills,breeds laziness and complacency and affects the pilots scan detrimentally.New pilots are being taught to operate their aircraft via the automatics.You hear of visual circuits being programmed into the FMS.Of pilots asking if they can take the AP out to fly an approach.You hear pilots saying if the vis drops below 3/4 mile,theyd better do an autoland.If the runway changes on them and theyre suddenly 4000'high on the VNAV do they "fly the plane" or reach for the FMC and start pressing buttons.How many can fly an arc procedure using raw data alone,controlling speed and profile at the same time.Or enter a hold without AP/FD/AT,turn the right way,begin the outbound timing at the correct point and fly the correct wind-adjusted heading w/o ref to track line.
Some pilots doggedly retain these skills but its entirely down to self-discipline.The problem is not with the pilots,but the system.We have too many spineless politically-correct corporate-minded DFO's who want their "pilots" to be something entirely different;I think the term these people use is "flight managers".Over-use of automation is mandated into the SOP's and the QAR's enforce it rigidly.If you dare fly an approach without the AFDS,you might get hauled before the safety committee...There are airlines today where pilots simply never disengage the AP/AT.Their hand goes from the MCP down to the FMC down to the radio panel and back again in a never-ending automatized stupor.At 500' they might just elect to fly the last 30 seconds of the flight.If automation wasnt so reliable,planes would fall out of the sky on a regular basis,I kid you not.
Automation is a nice tool and its correct use is an art in itself;but never use it because it can do what you cant do.When you feel like that due to inexperience/recency issues,that is exactly the time not to use it.You use it when you're damn sure that you can do everything you ask of it yourself.

Crashes due automation reliance/complacency/confusion:
Congonhas
Cali
Strasbourg
Bahrain 320
AeroPeru
Birgen
and many many more.It is a problem and we need to recognize that.

[Ralph Cramden]

Maximum

My my you are touchy. I don't think I mentioned The Schiphol crash at all in my post. We will have to wait for the final report to find out what went wrong there.
I was trying to point out that automation carried to the extreme lengths that it is today is detrimental to one's basic flying skills. I think that is a self evident fact, like it or not. I am sure however that you are a notable exception.

Ralph

[Maximum]

Hi Ralph, just because I didn't agree with everything you said makes me touchy??

Funnily enough, I agree that over-reliance on automatics can degrade certain manual flying skills. On the other hand, the automatics make it possible to operate more safely in today's complex and sometimes overcrowded airspace.

Surely the key is to make sure that pilots of our modern, sophisticated airliners are trained to operate them as they are intended, and that they are fully aware of the traps associated with this high degree of automation.

Yes, some of the skills are different, but they are still skills. I've seen some crusty kings of the sky make a real mess of things because they have refused to understand this philosophy.

I fully agree that the foundation of our skills must be a confident ability to 'take it all out' and simply fly the aeroplane. But I would also say that in the UK at least this has not changed for the vast majority of those in command or approaching it.

To say otherwise gives the layman or the PPL a very strange view of our professional standards which is just not true.

An example - I had a learner PPL telling me about how well his instructor thought he was getting on with crosswind landings. "Much harder than in the jet you fly" he said "where the automatics do it for you".

So c'mon guys, lets have some perspective here, that's all I'm saying. If I stick to what I know, the UK, then I can safely say that 99.9% of airliners out there are being flown by highly skilled and disciplined, extremely competent professional aviators.

[protectthehornet]

blending skills and automation

I do believe you can blend skills and automation, but the engineers haven't done it.

Displays, both visual and aural (sound) of information and warnings can be enhanced. Also navigational, weather, traffic and the like can all be helpful WHILE THE PILOT STILL FLYS THE PLANE.

We all seem to set BUGS for heading or airspeed or altitude. A gadget that says...HEY PAL, YOU ARE FIVE KNOTS BELOW VREF or YOU ARE FIFTY FEET TOO HIGH would demand pilot skill improvements.

Have you ever watched a mediocre pilot hand fly a landing/approach? No auto anything! He really never gets the engines to match up and is dancing on the rudders to keep going straight. I've seen pilots be at idle on one engine while using twice as much thrust on the other to make the approach, dancing on the rudders all the way down. NUTS.

You have to force your mind to do many things...take a look at DP Davies "handling the big jets" Those subconcsious things pilots do to make a landing...they all have to be consciously learned at one time or another.

[aguadalte]

Rananim:
Over-reliance on automation degrades manual skills,breeds laziness and complacency and affects the pilots scan detrimentally.New pilots are being taught to operate their aircraft via the automatics.You hear of visual circuits being programmed into the FMS.Of pilots asking if they can take the AP out to fly an approach.You hear pilots saying if the vis drops below 3/4 mile,theyd better do an autoland.If the runway changes on them and theyre suddenly 4000'high on the VNAV do they "fly the plane" or reach for the FMC and start pressing buttons.How many can fly an arc procedure using raw data alone,controlling speed and profile at the same time.Or enter a hold without AP/FD/AT,turn the right way,begin the outbound timing at the correct point and fly the correct wind-adjusted heading w/o ref to track line.
Some pilots doggedly retain these skills but its entirely down to self-discipline.The problem is not with the pilots,but the system.We have too many spineless politically-correct corporate-minded DFO's who want their "pilots" to be something entirely different;I think the term these people use is "flight managers".Over-use of automation is mandated into the SOP's and the QAR's enforce it rigidly.If you dare fly an approach without the AFDS,you might get hauled before the safety committee...There are airlines today where pilots simply never disengage the AP/AT.Their hand goes from the MCP down to the FMC down to the radio panel and back again in a never-ending automatized stupor.At 500' they might just elect to fly the last 30 seconds of the flight.If automation wasnt so reliable,planes would fall out of the sky on a regular basis,I kid you not.
Automation is a nice tool and its correct use is an art in itself;but never use it because it can do what you cant do.When you feel like that due to inexperience/recency issues,that is exactly the time not to use it.You use it when you're damn sure that you can do everything you ask of it yourself.

Rananim, I completely agree with you.
And the best way to maintain our skills and proficiency is to hand fly our aircraft in non stressing environments (at your home base or at low flow airports) on a regular basis. Sometimes your co-pilot only needs a word of incentive, because most of the times he might just be waiting for that opportunity to hand-fly the bird. Always include in your briefing the way you are going to fly your aircraft, so that your companion is not taken by surprise. (that doesn't preclude the pilot to switch of AP/FD/ATS in case of any malfunction or misbehavior - but clearly stating so).
Commonsense, plays a very important role here. If you're tired, in bad weather, or in a complex situation, just put your AP working for you. Otherwise, enjoy your hand-flying...Wasn't that the reason why you chose that profession?
(If you're a good, skilled and trained pilot...your pax will never notice!)

[Teddy Robinson]

another view.

This is not to detract from any of the opinions so far expressed.

This was a training scenario.

I was taught to teach a certain way, namely, demonstrate the systems and capabilities to educate and reassure, then gradually introduce the gotcha's to demonstrate the quirks of the systems backed up by accident/incident reports, the objective being to impart both knowledge and situational awareness.

In the analogue / clockwork machinery I used to fly, the limitations on autoflight did mean hand flying the aircraft a fair percentage of the time during the most critical phases.

Advancing onto a comparable type to that involved in this accident was an interesting experience, because the basis of early training was to monitor and let the systems do the work, a concept I was familiar with as an instructor but not as an operational philosophy.

A subtle shift in emphasis was that one was no longer announcing one's actions to be confirmed on the MCP (by the NHP), one was announcing what the MCP was now displaying as the handling pilot, to be confirmed by the NHP.

What we will not know until the report is properly issued is the background in aero / cultural terms of the mix in the cockpit, hours alone do not tell the full story, nor necessarily does the nationality, or the shared initial training background of those involved. Not yet at least.

By most civilian standards, all of the crew members were quite experienced in terms of hours, other operators field crews with far less combined experience on the same aircraft, and a similar incident has seemingly failed to occur.

Personal experience was that as hours on type accumulated, and situations permitted the hand flying side was re introduced and encouraged.
To me, this was a sound philosophy moving to the "new" glass cockpit environment.

There were times, however, when the workload was very high the active training had to stop and we had to concentrate on being a crew, as trained in the simulator, work to our combined strengths, and concentrate on flying the aircraft, calling deviations and sticking to company SOP !

After all a plane is just a plane whatever badge it carries.

As I say, not a statement as to what caused what, just an observation.

TR

[GlueBall]

Ralph Cramden: . . . "Where have all the pilots gone? A sad lament indeed. They have been automated out of existence. I'm afraid there are not many Sullys left in our industry. A man who took an automatic A/C and flew it."

Sorry Maximum, you may be flying with 99.5% sharp UK copilots whose manual flying skills are never in doubt. But I am mostly dumbfounded by the many [mostly non UK] 500hrs-on-type copilots who are either afraid or not motivated to ever want to hand fly the airplane. Not even in VMC!

Irrespective of the training culture and idiosyncracies with overall emphasis on automation, . . . For the life of me, I can't figure out why these chaps have an aversion to ever wanting to get a grip on the real airplane.

Instead, they hang on the A/P and do an autoland at destination; or at best, disconnect at 500 feet and then "fly" 30 seconds in VMC to touchdown.

I would often suggest to these chaps: "After 14 boring hours, don't you want to at least disconnect the A/P and A/T for 10 minutes . . . and do a manual intercept and try hand flying the ILS?" Duh.

[bullet190]

Those that say 'the motivation to hand fly stems from the top' are spot on.

At the moment the managers in the airline I fly for certainly do not encourage manual flying/visual approaches. When flight crew first convert to type they are not allowed to do a visual approach for the first 6 months. The end result of this is even after 6 months, most will not attempt a visual approach due to a lack of confidence.

In the last year alone I can only remember one occasion when an F/O hand flew a departure to the point of first level off - all the others have got the A/P in buy then.

What happens when it's u/s ? Indeed I have recently been informed of a Capt. who refused to fly an A/C because the A/T was u/s.

I have now been informed by the training department that the whole 6 monthly sim sessions can be done with the A/P in ! Most are obviously choosing to do this.

The AMS accident is a snippet of what's to come - it is the tip of the iceberg.

[Baron737]

How much manual hours are necessary ?

Maybe simulators should be replanned for manual flying only apart from autoland training- frankly if 4 sessions of hard manual flying a year don't do it, then you need to re-examine your flying skills or take up private flying to boost them.

I was really shocked when reading this. Am I such a bad, unskilled pilot, because I need 4 hours a week to keep my scan at its best ?
I do about 80 % of my approaches manual (...and that is of course raw data).
If i am tired I use the A/P. The best training i get is, when the weather is bad. In VMC it doesnīt really boost my skills, itīs just for fun in that case.

I really admire those super-skilled collegues, who donīt need that training.
But sometimes it shows, that they just misjudge their abilities

[Hiflyer1757]

What is the schedule for the next update

Have the Dutch announced any future briefings/meetings/hearings in regards to this accident...or has it gone 'behind closed doors' for the time being?

Note: I would expect that by now the CVR and the FDR have been played together in a 738 sim for comparison, that the CVR transcript along with the FDR data has been distributed to all 'interested' (read as official) members of the investigating team), that recordings/transcripts of all tower communications have been completed and also distributed to those same members.

I have noted several quotes that KLM has reported some 15 or so similar failures of the radalt on 737 type aircraft. Was these on 738 or mixed versions? Secondly, has any other large nexgen 737 carrier reported similar findings....especially in the US...such as AMR or Southwest....or in Europe such as Ryanair?

[Graybeard]

The radalt is a pretty good corrosion detector when it comes to that type of failure. Antenna to fuselage corrosion provides a signal leakage path direct from transmit to receive antenna; hence, the -8 feet indication.

The antennas should be dropped at the C check and the area cleaned. The belly collects a lot of crap, including that corrosive blue water.

-8 feet can be a valid altitude, as the 757 antennas are way forward, and will indicate about that when on ground. 0 feet, of course, is gear extended, at typical touchdown pitch.

GB

[snowfalcon2]

What is the schedule for the next update
----------------------------------------------------------------
Have the Dutch announced any future briefings/meetings/hearings in regards to this accident...or has it gone 'behind closed doors' for the time being?

As always, ICAO Annex 13 regulations stipulate that the investigating agency shall issue a preliminary report within 30 days after the accident. The preliminary report is "The communication used for the prompt dissemination of data obtained during the early stages of the investigation."

[Hiflyer1757]

Tks SF

Appreciate that info on ICAO regs and the 30 days...I think a lot of aviation industry professionals not officially members of the investigation would like to see a cvr transcript even if it would probably be redacted somewhat at this stage. The what, when, who, how, and why questions are pretty much pointing to the cvr as a possible guide to some answers at this point.