Archived

This topic is now archived and is closed to further replies.

Cactus521

Recent crash in Netherlands, focusses on altimeter

Recommended Posts

Help AVSIM continue to serve you!
Please donate today!

That's interesting. I certainly hope the investigators get to the bottom of this accident, as it is quite a disturbing scenario to have three crew on the flight deck and come down in such a curious way.It seems odd to imagine that the crew would not be closely monitoring speed, altitude and rate of descent when in low visibility. Kind of makes me wonder if the P1 and monitoring P2 were distracted by having a third person on the flight deck and possibly placed too much trust in the autopilot whilst chatting. I understand the third person on the FD was a trainee, so it seems reasonable to assume they'd be briefing the trainee about stuff, and if that proved a distraction from monitoring the instruments, then they wouldn't be the first crew to have an accident because of that.But that's speculation of course, and I guess you'd have to hear the cockpit voice recording to determine the possibility of such a scenario.Al

Share this post


Link to post

Briefly, this is what I gathered from the initial report:The radio altimeters went from 1950 to -8 due to some fault, the A/T followed this and went into retard mode as if the aircraft was in flare. The aircraft stalled and the crew couldn't recover from it. It seems that the aircraft was in autoland mode.I didn't know that the A/T would follow the radio altimeter. BTW, it has also been said that this same fault was seen recently on this aircraft.EDIT: The first thing that came to my mind was the lack of monitoring. How could the crew have missed the discrepancy?

Share this post


Link to post

""According to conversation recorded between the plane's captain, first officer and an extra first officer on the flight, the pilots noticed the faulty altimeter but didn't consider it a problem and didn't react, Van Vollenhoven said.""Being Dutch, I analysed the official announcent by Mr. van Vollenhove, but IMO it's a bit ambigous whether the pilots actually noticed the faulty RadioAlt meter or not. That is: when listening to his statement.What he said was (free translation):"... that the crew noticed the faulty left RadioAlt meter, namely because they got a Gear warning ...."and"... that the crew didn't consider this signalling to be a problem ...."What puzzles me is the second statement, where he says "... that the crew didn't consider this SIGNALLING to be a problem ..."Instead of "....the crew didn't consider the faulty RadioAlt meter to be a problem..."Which is a big difference IMO.Usually (because of possible legal consequences), such announcements are very well and accurately formulated.So I wonder whether the crew actually did notice the faulty Altmeter or not.Time will tell ....But whatever happened exactly, I find it very hard to understand why the crew didn't notice the decrease in airspeed (way below normal), and high pitch angle (because the aircraft was trying to follow glideslope with too low airspeed), untill it was too late.I don't want to pass judgement on the crew at all, but at least you can conclude it's another example of aircraft crew relying on 'automation' too much c.q. the crew 'procedures' to check if the 'automation' is making sence, isn't strict enough.In short, another classic example of where the intuitive human brain and human observation could have corrected faulty 'automation'.Which was programmed by other humans, by the way. (I'm referring to the coupling of AutoThrottle to the left RadioAlt meter). Rob

Share this post


Link to post

Excellent post Rob. I agree with what you say on the crew not noticing the anomalies. I wonder if that altimeter fault was on the MEL list since it was also reported before.I hope we, the general public, will get more information. This initial report has raised more questions than it has answered.

Share this post


Link to post
I didn't know that the A/T would follow the radio altimeter. BTW, it has also been said that this same fault was seen recently on this aircraft.
It seems to make sense (theoretically) that the A/T would respond to the radio altimeter. The FMC database, though it has airport elevation, wouldn't be as accurate as a real-time reading from the altimeter, so the autoland would use the radio altimeter to coordinate the flare.
The radio altimeters went from 1950 to -8 due to some fault...
Damn birds...On a more serious note, maybe the radio altimeter was briefly confused by something which it wasn't programmed to deal with well; I don't imagine a 738's altimeter reads -8 as a matter of course.

Share this post


Link to post
Being Dutch, I analysed the official announcent
Rob,Perhaps you can answer a question for us. Was this aircraft ever cleared for landing?I can only find a clearance down to 4,000 feet, and from then on, no communication between the tower and this aircraft until after it has crashed very close to the runway.Cheers,

Share this post


Link to post
I don't imagine a 738's altimeter reads -8 as a matter of course.
I would have to dig again into Boeing manuals but if I recall the negative altitude is a valid reading of the RA during autoland when aircraft reaches the phase of FLARE and then ROLLOUT which is annunciated on the PFD. Obviously it should never happen the way it happened here when aircraft was still hundreds of feet above ground.

Share this post


Link to post

I assume the "signalling" might have meant that the pilots looking at their Baro alt knew that the gear warning wasn't a problem per se, but the question is did they troubleshoot what caused the alarm?Interesting in that we often hear that the bus FBW throttle is inferior because of its lack of movement, while if this idea of A/T retard (if not flare) is correct, the throttle levers must have moved to idle without crew recognizing it. scott s..

Share this post


Link to post
the throttle levers must have moved to idle without crew recognizing it.
If (as alleged) their heads were turned back towards the trainee sitting on the jumpseat whom they were instructing on use of the checklist then it is easy to understand how they missed everything. Not only they missed the throttle movement but they missed all the annunciations on the PFD. It is not the first time that a minor technical issue leads to major crash becasue attention of the whole crew is diverted to something else.

Share this post


Link to post

I wouldn't wish to second guess the accident investigators, but I've just had a look at the B737 manual for stuff relating to the radio altimeter on approach and some related stuff. I thought it might be of interest to people reading this thread:If you have a dual AP approach set up on the MCP, the automatic flare commences when the radio altimeter reads 50 feet and the autothrottle then retards at 27 feet. I can't find any mention of a negative value on the RA (at least not in the section I read). But based on those values, it doesn't need to get near a negative value to commence flaring and retarding the throttle.The autothrottle disengages 2 seconds after touchdown, but whether touchdown is read as a parameter from the RA, or the weight on the wheels, is not made clear in the manual, although one suspects it is from the RA based on what has occurred in this accident and the fact that the process begins when airborne.If you hit TOGA when in a dual autopilot mode, the FD commands a 15 degree nose pitch up, in addition to the thrust increase, which is obviously likely to precipitate a stall if the speed is very low and the nose goes up before the engines spool up properly and succeed in getting some speed on the clock. Which could explain the tail striking first in the case of this accident, if the crew did hit the TOGA switch. (see the note below)The B737 manual warns that at low airspeeds, such as when windshear might be encountered on an ILS approach - a situation presumably not dissimilar to that which the crew of the Turkish 737 found themselves in, with minimal airspeed on approach (albeit for different reasons) - both the FD and the autopilot will still try and hold the airplane on the glideslope without regard to either angle of attack or stick shaker limitations, and the manual further warns that the airspeed could decrease into a stall if the pilot does not intervene by either hitting the TOGA switch, or disconnecting the autopilot and manually flying the aircraft.It goes on to state that: 'the flight crew must, if necessary to avoid ground contact, be prepared to disconnect the autothrottle, advance the thrust levers to the forward stop, disconnect the autopilot, and manually fly the airplane.' Which of course means that if you hit TOGA when there's 'nothing on the clock but the maker's name', it is likely to make matters worse in comparison to disconnecting everything, ramming the thrust levers to the stops and manually keeping the nose down.This - other than the factual stuff quoted from the manual - is fairly speculative with regard to what occurred in this incident, and I think without hearing the comments on the flight deck it's not really going to be possible to say what happened. Naturally, it is easy to be wise about such things when sitting reading a manual drinking a cup of coffee. Quite different of course when you have seconds to make a choice.Al

Share this post


Link to post
if the crew did hit the TOGA switch.
They never did, based on press reports.

Share this post


Link to post
Which of course means that if you hit TOGA when there's 'nothing on the clock but the maker's name', it is likely to make matters worse in comparison to disconnecting everything, ramming the thrust levers to the stops and manually keeping the nose down.
Aside from the press reports, I find it a bit dubious that the crew would have thought to hit the TOGA switch. This is nothing but speculation on my part (disclaimer: I am not a pilot, have never taken flight lessons, and have never read airplane manuals, but have read a fair bit of theory), but I'm led to believe that the training the crew would have received would have ground into them the procedure for near-stall and stall recovery, and that they would have known that hitting the TOGA switch, even if they didn't specifically remember the section in the manual you quoted, would not produce a desirable reaction (pitching the nose up rather than down).

Share this post


Link to post
Rob,Perhaps you can answer a question for us. Was this aircraft ever cleared for landing?I can only find a clearance down to 4,000 feet, and from then on, no communication between the tower and this aircraft until after it has crashed very close to the runway.Cheers,
Hi Kevin,Although that wasn't explicitely said in that 22 min. statement, I can only assume it was.Because of the statement "everything appeared to be normal".If it hadn't be cleared for landing, I'm sure that would have been reported now.But as said: this statement probably generated more questions than that it answered.Like:- SHOULD the crew have noticed the fatal decrease in airspeed according to 'procedures' ? (I.o.w.: do procedures require them to monitor airspeed in an automated landing? And how often ? )- And if so, WHY didn't they ? Or, if they did notice it, why didn't they react earlier ? - According to the black boxes, this Altmeter problem happened already twice in the previous 10 flights. But the statement doesn't clarify whether this was actually noticed by the other crews or maintenance personel, and if so, what their reaction has been or if/what was done about it.In short, this statement was just a preliminary result of the investigation commitee 's findings, and one can doubt whether it was wise to come with such a statement when all those questions are still unanswered, other than the actual technical problem that lead to the crash.Because it already has unfortunately lead to different "opinions" in Holland and Turkey about the accident. Like, in Turkey, the crew have been presented in the press as "heroes", because they managed to "crash" the aircraft into such a (relatively) good location (muddy clay), thereby saving a lot of lives.The least you can conclude from the committee's statement, that there was nothing deliberate about the location of the crash, and that they were actually trying to save the aircraft and avoid the crash at all.IMO it was wise though, since such a disaster could happen at any time again if pilots aren't aware what the consequences of a faulty left Altmeter can be (even IF they are aware of the Altmeter being faulty). Hence the warning to Boeing (about the A/T-coupling) made by the committee.So it will take a lot of time longer to establish ALL the facts leading/contributing to this crash.And even when all the "facts" are known, there will probably still remain discussions on how to weigh/interprete them; as in many air crash investigations.Because it undoubtedly (and IMO unfortunately) usually leads to the question "who's to blame" (because of legal/financial consequences).Like, even todate, there are still a different opinions on "who's to blame" in the "Tenerife" crash (where a landed, taxiing 747 collided with a departing 747).I've seen 3 documentories on that (US, Dutch, Spanish made), and they all come to a different interpretation/weighing of the "facts".Rob

Share this post


Link to post
- SHOULD the crew have noticed the fatal decrease in airspeed according to 'procedures' ? (I.o.w.: do procedures require them to monitor airspeed in an automated landing? And how often ? )
You bet they SHOULD. It is part of basic airmanship skills that at least one pilot (if there are 2) constantly performs instrument scan. Every pilot learns it during his/her IFR training and then airline crews get additional training called Crew Resource Management that teaches them how to divide responsibilities between two pilots. There is simply no excuse for BOTH pilots to have their attention diverted away for so long from monitoring such basic performance numbers like speed, altitude and engine power. It reminds me of two other famous crashes with many fatalities when crews forgot to monitor their aircraft - the really famous Everglades crash of L-1011 and the DC8 crash in Portland that ran out of fuel while crew wrestled with the nose gear not being fully extended.

Share this post


Link to post

From the preliminary investigation it appears that the A/P was engaged in CMD B, which would take its alt ref from the right (correct) RALT, and "autoland" was not a factor. The pilots were flying a coupled ILS approach. The A/T takes its alt ref from the left RALT, which was providing faulty output. It wasn't stated, but it seems as though the left RALT did not flag a "fail". It looks like the A/T will command a retard when the RALT is 27 ft regardless of A/P function. scott s..

Share this post


Link to post
Like:- SHOULD the crew have noticed the fatal decrease in airspeed according to 'procedures' ? (I.o.w.: do procedures require them to monitor airspeed in an automated landing? And how often ? )- And if so, WHY didn't they ? Or, if they did notice it, why didn't they react earlier ? - According to the black boxes, this Altmeter problem happened already twice in the previous 10 flights. But the statement doesn't clarify whether this was actually noticed by the other crews or maintenance personel, and if so, what their reaction has been or if/what was done about it.
Rob,As always, all excellent insights. Thanks for the info. I've not heard audio of this guy being cleared final or for landing, and I've searched for it (I just assume I can't find it, since it would be highly unusual and the Tower at Schiphol is highly trained and respected.)I'm amazed at the quality of the investigations conducted by the NTSB, which has a 4-man advisory team in on the investigation. They are incredibly thorough, and will leave no stone unturned. My initial feeling, just looking at the pictures, was that it was most likley wind shear, probably caused by the preceding aircraft's wake turbulence. However, knowing that there was a known faulty equipment of such import, one naturally turns their eye to the indifference factor.Cultural sensitivities will also be in play as this unfolds, I suspect. The Turks are investing a lot of face in the "hero" storyline (whether that be for cultural or financial reasons, probably both).

Share this post


Link to post
From the preliminary investigation it appears that the A/P was engaged in CMD B, which would take its alt ref from the right (correct) RALT, and "autoland" was not a factor. The pilots were flying a coupled ILS approach. The A/T takes its alt ref from the left RALT, which was providing faulty output. It wasn't stated, but it seems as though the left RALT did not flag a "fail". It looks like the A/T will command a retard when the RALT is 27 ft regardless of A/P function.
This is really some solid info. Thanks Scott.
Cultural sensitivities will also be in play as this unfolds, I suspect. The Turks are investing a lot of face in the "hero" storyline
You bet. When a 737 crashed years ago in eastern Turkey during snowy weather (captain flew "illegal" approach, ignored warnings and hit the mountain) and when the final NTSB report came out blaming the crew Turks did not accept it and called pilot a "martyr". By the same token Egyptians never accepted similar reports blaming their pilots for Flash Air and EgyptAir crashes and instead blamed unspecified equipment failures. Certain cultures have very hard time accepting any blame. I think one Muslim cleric even hinted that a Muslim pilot can't make a mistake.

Share this post


Link to post
This is really some solid info. Thanks Scott.You bet. When a 737 crashed years ago in eastern Turkey during snowy weather (captain flew "illegal" approach, ignored warnings and hit the mountain) and when the final NTSB report came out blaming the crew Turks did not accept it and called pilot a "martyr". By the same token Egyptians never accepted similar reports blaming their pilots for Flash Air and EgyptAir crashes and instead blamed unspecified equipment failures. Certain cultures have very hard time accepting any blame. I think one Muslim cleric even hinted that a Muslim pilot can't make a mistake.
Please, let us take care that this thread doesn't end up in a debate about "countries" and "cultures".Because (but now I'm getting very philysofical), what are "facts" ??To give you one example (and I should blame myself for that ):Fact is (and you have to take my word on that) I don't speak the Turkish language. Like most of you probably don't as well.So if I state "Like, in Turkey, the crew have been presented in the press as "heroes", and I call that a "fact", I'm not correct.Because I assume that what the Dutch press reports about Turkish press on this, is actually correct. But is it ??Our main problem, even in this day-and-age, is that if we read/hear/see things in the media presented as a "fact", we are inclined to assume it IS a "fact".But if history teaches us one thing, presented "facts" in the media aren't always "facts" at all (everyone can fill in his own examples !) Heck, even the "fact" that the left Altmeter was faulty is only a "fact" to me, because I cannot imagine that this Investigation committee could be wrong (or even worse: would lie) about that.Just to put things into perspective :-)Rob

Share this post


Link to post
Fact is (and you have to take my word on that) I don't speak the Turkish language.
Neither do I. And I have never been to Australia and frankly I am not sure if I can believe media/books that this continent exists. I never touched or surveyed it to verify its true geographic span. And I wasn't on the Moon when Armstrong is alleged to make his first step - so can't really trust it either. Just putting things in perspective ;)

Share this post


Link to post

Very interesting posts so far. However, a few things must be clarified.First of all, the media in Turkey is going to be extremely harsh on Turkish Airlines. That has already started. The reason these pilots are presented as heros is because there is a major highway and buildings close to the crash site. Whether the crew crashed there intentionally or not is not known at this time. Personnally, I give little credit to this but you never know. My opinion is that that it was of pure luck. Unfortunately all of the 3 pilots died, they would have helped with the investigation.I would also like to point out that it is totally useless and incorrect to refer to "cultural issues" for the simple reason that the majority of the members here probably know absolutely nothing about Turkey or the Turkish way of dealing with these types of tragedies. Please do not compare Turkey to other countries either. Here the population is extremely critical towards Turkish Airlines when such incidents occur. Moreover, I wouldn't be surprised if many Turks prefered to fly with other companies for a while. Oh, and for the record, in Turkey, pilots who lose their lives are called martyrs for the simple fact that an unwritten code says state employees killed in the line of duty die as a martyr. As Rob said, the best way to debate on this is by sticking to the facts. A poster also mentioned that he couldn't find the ATC recording that corresponds to the landing clearance. You can find it here, http://www.liveatc.net It's right there on the top of the home page.What intrigues me most and the most of you judging by your posts is the lack of reaction from the crew. I certainly hope this will be clarified in future reports. It is hard to believe that the 3 of them just sat there while the aircraft just stalled itself...

Share this post


Link to post
A poster also mentioned that he couldn't find the ATC recording that corresponds to the landing clearance. You can find it here, http://www.liveatc.net It's right there on the top of the home page.
Yes, I have listened to this recording. I do not hear a landing clearance. Perhaps it is in there, but I cannot discern it. There are some communications at the end which are so garbled as to make them incomprehensible to me. Does anyone know if a transcript of the public tower communications has been released in Holland?

Share this post


Link to post
Yes, I have listened to this recording. I do not hear a landing clearance. Perhaps it is in there, but I cannot discern it. There are some communications at the end which are so garbled as to make them incomprehensible to me. Does anyone know if a transcript of the public tower communications has been released in Holland?
Yeah, you're right, sorry. I forgot about it, this one is actually approach clearance and hand off to tower. Oddly enough the tower recordings aren't released. Maybe they were enclosed for investigation.I don't think the communications with tower are available or surely someone would have noticed them by now. If there aren't any legal issues, hopefully they should come out in the future.

Share this post


Link to post
Yeah, you're right, sorry. I forgot about it, this one is actually approach clearance and hand off to tower. Oddly enough the tower recordings aren't released. Maybe they were enclosed for investigation.I don't think the communications with tower are available or surely someone would have noticed them by now. If there aren't any legal issues, hopefully they should come out in the future.
As said, the main reason why the investigation committee felt they had to disclose what they did at this moment, is because it can happen at any time again under simular circumstances.For the rest: it must be frustrating for everybody involved, but I think many questions will remain unanswered for a long time untill everything leading/related to this crash is cleared up. If even possible.Hence, I think much additional info (even if allready known), will remain "under the hood" for the time being.Regards, Rob

Share this post


Link to post