Turkish airliner crashes at Schiphol
Join Date: Aug 2005
Location: Cloudbase
Posts: 149
Likes: 0
Received 0 Likes
on
0 Posts
PJ2
sorry, but the (non aviation) engineer in me sais differently:
yes, the accident would have easily prevented by basic airmanship.
It would never have happened if the A/T and A/P systems were, by design, aware of their mutual modes, which they seem not to be. I think "mode confusion" is the term for this.
We all know that two things are safe: those you never do and those you always do. This was something that you seem to be doing extremely rarely. Not an inop radalt, but the fact that A/T disconnects as a result and A/P bleeds off speed, trying to maintain GS. As I said earlier, the systemic failure of this accident is terribly similar to the Birgenair 301 accident in 1996. In both cases, a faulty sensor was the single input to an automation system that subsequently flew the airplane into a stall. In both cases, the crew failed to notice the loss of energy and, when they finally did, were unable to recover.
Everyone will seek the cause in the realm where they can personally improve, and that's good. Pilots who are aware of their own fallability are better pilots (I do, however, sometimes wonder if some of the "idiots forgot to fly the plane" claims here don't have an underlying "couldn't have happened to me").
Engineers must, at the same time think about something else, I believe: Pilots are on the flight deck when things happen, they can adapt to the situation and react accordingly. Engineers aren't. Their decisions are, however, at the heart of the actions taken by the automation systems. How they respond to peripheral failures is decided upfront by what the design engineers could imagine. Failures they couldn't think of will often be handled poorly or not at all (that's a general observation, not specifical for aviation at all)
So what I'm saying is: yes, both accident should have turned out as non-events had the trend being spotted and identified early enough.
But: both events shouldn't have developed to anything serious in the first place, because at least theoretically, all relevant information was available to the automation systems to make a decision to either continue in a coordinated, safe mode (A/T and A/P feeding off a known good data source) or to completely hand it off to those humans who, in ambiguous situations, should know better what to do than the engineer behind his desk.
The thing I'm wondering about is whether it wasn't a full handoff to manual by the A/T system communicated to the crew by the thrust levers moving all the way back. Was that an ambiguous signal, since it was what they expected to happen when they were at an energy state that was too high to begin with?
Quite a few slices of cheese, quite a few holes lining up.
sorry, but the (non aviation) engineer in me sais differently:
yes, the accident would have easily prevented by basic airmanship.
It would never have happened if the A/T and A/P systems were, by design, aware of their mutual modes, which they seem not to be. I think "mode confusion" is the term for this.
We all know that two things are safe: those you never do and those you always do. This was something that you seem to be doing extremely rarely. Not an inop radalt, but the fact that A/T disconnects as a result and A/P bleeds off speed, trying to maintain GS. As I said earlier, the systemic failure of this accident is terribly similar to the Birgenair 301 accident in 1996. In both cases, a faulty sensor was the single input to an automation system that subsequently flew the airplane into a stall. In both cases, the crew failed to notice the loss of energy and, when they finally did, were unable to recover.
Everyone will seek the cause in the realm where they can personally improve, and that's good. Pilots who are aware of their own fallability are better pilots (I do, however, sometimes wonder if some of the "idiots forgot to fly the plane" claims here don't have an underlying "couldn't have happened to me").
Engineers must, at the same time think about something else, I believe: Pilots are on the flight deck when things happen, they can adapt to the situation and react accordingly. Engineers aren't. Their decisions are, however, at the heart of the actions taken by the automation systems. How they respond to peripheral failures is decided upfront by what the design engineers could imagine. Failures they couldn't think of will often be handled poorly or not at all (that's a general observation, not specifical for aviation at all)
So what I'm saying is: yes, both accident should have turned out as non-events had the trend being spotted and identified early enough.
But: both events shouldn't have developed to anything serious in the first place, because at least theoretically, all relevant information was available to the automation systems to make a decision to either continue in a coordinated, safe mode (A/T and A/P feeding off a known good data source) or to completely hand it off to those humans who, in ambiguous situations, should know better what to do than the engineer behind his desk.
The thing I'm wondering about is whether it wasn't a full handoff to manual by the A/T system communicated to the crew by the thrust levers moving all the way back. Was that an ambiguous signal, since it was what they expected to happen when they were at an energy state that was too high to begin with?
Quite a few slices of cheese, quite a few holes lining up.
Join Date: Jul 2008
Location: Deira
Posts: 9
Likes: 0
Received 0 Likes
on
0 Posts
Pilots are Pilots and they are humans , by nature susceptible to commit mistakes . I am pretty sure these guys knew most of the issues everybody is pointing out regarding basic skills , AP performance monitoring and so on .Maybe they would even be posting in this thread should it had happened to somebody else .
Even though it seems unthinkable at a first glance to "crash ...due RA failure ", a combination of possible events already mentioned by some previous posts , can set a scenario where many of us can commit mistakes that we might consider ourselves immunes. It is very easy to judge or stress out somebody else mistakes while we are seating in a bar with fellow pilots or posting replies for a thread....very easy .
Which of us drivers ever had that flight , with a long time friend in the cockpit , long chats and the situational awareness deeply affected .
" I was distracted I forgot ...".
The question we should make is "Why an aircraft with three pilots in the flight deck were not being properly monitored ?" . Well , time (and the investigation)will tell ......It was not the RA failure alone .
Even though it seems unthinkable at a first glance to "crash ...due RA failure ", a combination of possible events already mentioned by some previous posts , can set a scenario where many of us can commit mistakes that we might consider ourselves immunes. It is very easy to judge or stress out somebody else mistakes while we are seating in a bar with fellow pilots or posting replies for a thread....very easy .
Which of us drivers ever had that flight , with a long time friend in the cockpit , long chats and the situational awareness deeply affected .
" I was distracted I forgot ...".
The question we should make is "Why an aircraft with three pilots in the flight deck were not being properly monitored ?" . Well , time (and the investigation)will tell ......It was not the RA failure alone .
From the Boeing recommendation:
WHY do I, as a professional pilot, need to be reminded to watch my airspeed?
This is silly. It is sooo basic that I actually do not understand why Boeing would communicate this to the flying community.
It feels like telling mrs. fox niner to watch the road while driving a car.
In
addition, crews should be reminded to carefully monitor primary flight
instruments (airspeed, attitude etc.) and the FMA for autoflight modes.
addition, crews should be reminded to carefully monitor primary flight
instruments (airspeed, attitude etc.) and the FMA for autoflight modes.
This is silly. It is sooo basic that I actually do not understand why Boeing would communicate this to the flying community.
It feels like telling mrs. fox niner to watch the road while driving a car.
Join Date: Jan 2008
Location: South Bucks
Posts: 46
Likes: 0
Received 0 Likes
on
0 Posts
This is silly. It is sooo basic that I actually do not understand why Boeing would communicate this to the flying community.
Join Date: Jul 2005
Location: Ask OPS!
Posts: 1,078
Likes: 0
Received 0 Likes
on
0 Posts
Just to add my two penneth worth as an Airbus driver.
The thrust levers in the airbus are fixed, non moving. Hence there is a requirement to closely monitor ALL FMA readings during all phases of flight. A loss of auto thrust or a reduction in thrust, as seems to be the case here, would not be too difficult to spot. Also there is an SOP of 'hands on the top' of the thrust/throttle levers from interception of the final approach.
AFAIK, not being a Boeing driver, the throttle levers move with the commands of the auto thrust and, according to my colleagues on Boeings, the pilots get quite used to where they 'should' be. A prolonged reduction to idle thrust should, surely, have been noticed by either the TC or the safety pilot? One could forgive the trainee as they were, obviously, in their early stages ti warrant a Safety Pilot.
As to the 'horror' scenario of flying a manual thrust/ AP off landing? Not a horror at all, fairly normal in fact. Most airlines allow the Captain to have the latitude to disconnect the AP and/or the Autothrottle for routine line training at their or the FO's wish. Manual approaches are flown regularly at my airline by both seats.
A sad scenario and one that could, possibly, have been avoided. I for one will read the accident report with interest.
The thrust levers in the airbus are fixed, non moving. Hence there is a requirement to closely monitor ALL FMA readings during all phases of flight. A loss of auto thrust or a reduction in thrust, as seems to be the case here, would not be too difficult to spot. Also there is an SOP of 'hands on the top' of the thrust/throttle levers from interception of the final approach.
AFAIK, not being a Boeing driver, the throttle levers move with the commands of the auto thrust and, according to my colleagues on Boeings, the pilots get quite used to where they 'should' be. A prolonged reduction to idle thrust should, surely, have been noticed by either the TC or the safety pilot? One could forgive the trainee as they were, obviously, in their early stages ti warrant a Safety Pilot.
As to the 'horror' scenario of flying a manual thrust/ AP off landing? Not a horror at all, fairly normal in fact. Most airlines allow the Captain to have the latitude to disconnect the AP and/or the Autothrottle for routine line training at their or the FO's wish. Manual approaches are flown regularly at my airline by both seats.
A sad scenario and one that could, possibly, have been avoided. I for one will read the accident report with interest.
Join Date: Aug 2007
Location: Alabama
Age: 58
Posts: 366
Likes: 0
Received 0 Likes
on
0 Posts
.Formula one pilots do survive hight speed impacts because they are protected by light armored material
F1 Pilots survive high speed impact because of the crashing parts that disperse kinetic energy. Furthermore their harness is much better designed than the one used by the cockpit crew. Finally, after many pilots died due to neck/spine injuries, the HANDS device is use to avoid large "g" force on neck and head.
Cunning Artificer
Join Date: Jun 2001
Location: The spiritual home of DeHavilland
Age: 76
Posts: 3,127
Likes: 0
Received 0 Likes
on
0 Posts
It feels like telling mrs. fox niner to watch the road while driving a car
We know the how - a defective radio altimeter caused the throttles to retard at or around glideslope capture. The missing element is a human factors one -why did three professional pilots who would normally be well aware of their airspeed during an approach fail to notice the speed decay. What was the distraction?
Join Date: Sep 2003
Location: .
Posts: 557
Likes: 0
Received 0 Likes
on
0 Posts
One thing I am wondering about is in my outfit below MSA you are supposed to cover the flight controls. They were below MSA when this chain of events began. If PF was covering the controls surely he would have noticed a prolonged and unusual thrust reduction from the position of the thrust levers?? This might have then prompted a look at the FMA's? Not right, disconnect.
Also airmanship would dictate that anytime you were expecting the thrust levers to move (back or forward) cover them?
Still can't understand how 3 sets of eyes were distracted to this extent. Be interested to see the human factors involved in this. edit: (sorry post above just said this but it shows we are all wondering!!)
Also airmanship would dictate that anytime you were expecting the thrust levers to move (back or forward) cover them?
Still can't understand how 3 sets of eyes were distracted to this extent. Be interested to see the human factors involved in this. edit: (sorry post above just said this but it shows we are all wondering!!)
Join Date: Apr 2008
Location: Southampton
Posts: 10
Likes: 0
Received 0 Likes
on
0 Posts
rubik101 - The flight deck was subject to different forces to the passenger cabin - came down from a higher angle, stopped in a shorter distance - so you cannot compare the two. I also imagine that if they just had a lap belt then a sudden stop would result in a lot more injury from contact with the various panels, etc. around them.
If I were to find myself in an accident and had the choice I would go for the 5 point harness (or at least a 3 point one) every single time over a lap belt.
If I were to find myself in an accident and had the choice I would go for the 5 point harness (or at least a 3 point one) every single time over a lap belt.
Join Date: May 2006
Location: London
Age: 69
Posts: 237
Likes: 0
Received 0 Likes
on
0 Posts
From post 1053 re. Boeing's message ;
"(Extract) ........... If one LRRA provides erroneous altitude readings, typical flight deck effects, which require flight crew intervention whether or not accompanied by an LRRA fault flag, include:
- Large differences between displayed radio altitudes, including radio altitude readings of -8 feet in flight.
................."
Does that mean that the "fault" reading of a RA is "-8 ft" or merely that a reading of -8 ft whilst in flight is indicative of a RA failure ?
.
"(Extract) ........... If one LRRA provides erroneous altitude readings, typical flight deck effects, which require flight crew intervention whether or not accompanied by an LRRA fault flag, include:
- Large differences between displayed radio altitudes, including radio altitude readings of -8 feet in flight.
................."
Does that mean that the "fault" reading of a RA is "-8 ft" or merely that a reading of -8 ft whilst in flight is indicative of a RA failure ?
.
Join Date: Jul 2007
Location: Hadlow
Age: 60
Posts: 597
Likes: 0
Received 0 Likes
on
0 Posts
Why didn't the nose gear shear off like the MLG did?
Last edited by Super VC-10; 5th Mar 2009 at 08:31. Reason: typo
Join Date: Aug 2007
Location: London, New York, Paris, Moscow.
Posts: 3,632
Likes: 0
Received 0 Likes
on
0 Posts
MartinS post #1301
I would not be surprised if the crew thought that since they were using AP B then the AT system would also read from RA on that side and they had no issues due to the captain's side RA malfunction.
Join Date: Mar 2008
Location: Europe
Posts: 2
Likes: 0
Received 0 Likes
on
0 Posts
Gear down?
Quoting: "The crew, which included a captain, a first officer on a training flight and a third pilot in the cockpit, "were notified that the left radio altimeter was not working correctly" by a "landing gear must go down" warning signal, Vollenhoven said, adding that "provisional data indicates that this signal was not regarded [by the crew] to be a problem."
So, the gear was not down by the time the AT retarded to idle? Later it was said, the gear sheared off during impact. That makes me wonder if they selected the gear down without checking IAS that would have been low at that time already...
So, the gear was not down by the time the AT retarded to idle? Later it was said, the gear sheared off during impact. That makes me wonder if they selected the gear down without checking IAS that would have been low at that time already...
Join Date: Sep 2008
Location: Dark Side of West Wales
Age: 85
Posts: 161
Received 0 Likes
on
0 Posts
Human Factors
As a now retired "shrink" with many years working with the aeronautical industry I have found many of the posts on this thread very informative. Several have homed in on what is probably the real cause of this tragic crash - human failure of some sort. Who, how, what, when, etc we may never know but the cockpit voice recorder should give us some excellent clues and coupled to the data recorder hopefully something we can all learn from.
But for the Grace of God, etc.
But for the Grace of God, etc.
Per Ardua ad Astraeus
Join Date: Mar 2000
Location: UK
Posts: 18,579
Likes: 0
Received 0 Likes
on
0 Posts
Less 'heat' on Boeing, I feel - they HAD to issue the 'reminder' of how to fly once they were 'warned' by the Dutch investigation, I'm sure. They were really on a hiding to nothing there.
While I was flying the 737 at LGW with BA they ran a situation awareness course in the early 2000s to address the subtle incapacitation of active monitoring caused by increasingly sophisticated automation and instrumentation. It seems this needs to be repeated - everywhere.
Part of the 'problem' is that we are (correctly) encouraged to engage the A/P when things go wrong to allow the system to do the basic flying while we resolve the issue. This, I'm now convinced, is leading pilots into a state of nirvana where the automatics are everything and we can all relax. To some extent, the LHR BA 777 may reflect that, with the A/P being left in to follow the glide while the speed decayed towards the stall, just like AMS, and the PGF Airbus crash shows the same over-trust in the systems.
The question is, how do we address this?
While I was flying the 737 at LGW with BA they ran a situation awareness course in the early 2000s to address the subtle incapacitation of active monitoring caused by increasingly sophisticated automation and instrumentation. It seems this needs to be repeated - everywhere.
Part of the 'problem' is that we are (correctly) encouraged to engage the A/P when things go wrong to allow the system to do the basic flying while we resolve the issue. This, I'm now convinced, is leading pilots into a state of nirvana where the automatics are everything and we can all relax. To some extent, the LHR BA 777 may reflect that, with the A/P being left in to follow the glide while the speed decayed towards the stall, just like AMS, and the PGF Airbus crash shows the same over-trust in the systems.
The question is, how do we address this?
Join Date: Jan 2000
Location: U K
Posts: 469
Likes: 0
Received 0 Likes
on
0 Posts
Plugster,
Idle thrust for 40 seconds on the approach is not realy an issue. On a jet aircraft like the 73, it is common place for the crew to carry out continuous decent approaches. During a CDA it is normal to have lengthy periods of idle thrust. The question is, why did the crew not notice that when the speed had reduced to what they had selected, did the autothrust not start to come back in?
I don't believe that a hand on the thrust lever would make much difference in this case, you either notice it is time for thrust, or you don't.
Framer,
If you are handeling the aircraft you should cover the controlls, no need to have your hands welded to them.
If you fly the 73, I would be amazed if captains don't regularly ask you to remove your hand from the thrust lever in order for them to ensure the correct flap selection.
While your left hand is glued to the thrust levers, who is making all of your MCP selections while the autopilot is still engaged?
Idle thrust for 40 seconds on the approach is not realy an issue. On a jet aircraft like the 73, it is common place for the crew to carry out continuous decent approaches. During a CDA it is normal to have lengthy periods of idle thrust. The question is, why did the crew not notice that when the speed had reduced to what they had selected, did the autothrust not start to come back in?
I don't believe that a hand on the thrust lever would make much difference in this case, you either notice it is time for thrust, or you don't.
Framer,
If you are handeling the aircraft you should cover the controlls, no need to have your hands welded to them.
If you fly the 73, I would be amazed if captains don't regularly ask you to remove your hand from the thrust lever in order for them to ensure the correct flap selection.
While your left hand is glued to the thrust levers, who is making all of your MCP selections while the autopilot is still engaged?
Join Date: Jun 2004
Location: Maastricht, NL
Posts: 101
Likes: 0
Received 0 Likes
on
0 Posts
RA
In my understanding the RA didn't actually fail. It was transmitting an incorrect value (while "being" completely "sure" about it, that it's correct), thus it was not giving the failure sign, causing the INOP warning light on the GPWS panel and the other flags in.
Join Date: Dec 1998
Location: UK
Posts: 293
Likes: 0
Received 0 Likes
on
0 Posts
I think there may be good reasons why the two radalts are separate, and the No.1 is more important than the number 2. The number 1 probably is powered by a transfer bus, the number 2 by a gen bus, therefore if you lose half your electrics the aircraft can shed the no.2 whilst keeping the more important no.1 working.
Also, I believe this allows you to dispatch under the MEL with the no.2 radalt inoperative, but not the no.1. All very sensible.
I probably did know once that radalt 1 controls the a/t but I had forgotton. I think most crews would not be aware of this. However, most crews would probably notice 'Retard' on the FMA, thrust levers at idle and speed decaying by 40kts. Also, the pitch attitude must have been very unusually high close to the stall speed, would have thought you'd notice this pitch up.
The difference between this and 777 incident is that the 777 crew noticed the speed decay, and intervened, critically, BEFORE the aircraft stalled, thus saving the lives of everyone on board.
From what I understand, the aircaft would not have flared because A/P B was engaged and taking it's radalt info from the no.2. So the aircraft would have maintained the correct glideslope whilst the speed washed off.
I agree you should definitely have your hands on the thrust levers below 1000ft as your hand is hovering over the toga button in case you need to go around, or if you hit windshear, or stall, then you can react immediately.
The captain should NOT take over in the event of a stall normally (maybe different on a training flight if the trainee fails to react). The F/O is PIC/US and is fully trained and should be expected to react correctly to a stall warning. The captain taking over simply leads to a delay and a few vital seconds when everyone has to adjust to their new role at a critical flight phase. The drill isn't difficult, it's similar to windshear or GPWS -
Disconnect A/P and A/T
Manually select full power
Select pitch attitude (20 degrees nose up for GPWS, 15 for Windshear, lower the nose until the stall warner stops / pitch decreases below PLI's.
Speedbrakes retract
Wings level
Leave rest of config as it is.
Best aircraft performance is achieved whilst 'nibbling' at the stall warner (intermittant stall warn).
Were the thrust levers firewalled? I know this has gone out of fashion but probably appropriate in that situation.
Finally, we are required to be stable at 1000ft (and MUST be stable at 500')- which includes the thrust levers being at approach power. I always check this with a view to discussing if we are likely to be stable by 500' and considering a go-around. I guess they didn't......
Also, I believe this allows you to dispatch under the MEL with the no.2 radalt inoperative, but not the no.1. All very sensible.
I probably did know once that radalt 1 controls the a/t but I had forgotton. I think most crews would not be aware of this. However, most crews would probably notice 'Retard' on the FMA, thrust levers at idle and speed decaying by 40kts. Also, the pitch attitude must have been very unusually high close to the stall speed, would have thought you'd notice this pitch up.
The difference between this and 777 incident is that the 777 crew noticed the speed decay, and intervened, critically, BEFORE the aircraft stalled, thus saving the lives of everyone on board.
From what I understand, the aircaft would not have flared because A/P B was engaged and taking it's radalt info from the no.2. So the aircraft would have maintained the correct glideslope whilst the speed washed off.
I agree you should definitely have your hands on the thrust levers below 1000ft as your hand is hovering over the toga button in case you need to go around, or if you hit windshear, or stall, then you can react immediately.
The captain should NOT take over in the event of a stall normally (maybe different on a training flight if the trainee fails to react). The F/O is PIC/US and is fully trained and should be expected to react correctly to a stall warning. The captain taking over simply leads to a delay and a few vital seconds when everyone has to adjust to their new role at a critical flight phase. The drill isn't difficult, it's similar to windshear or GPWS -
Disconnect A/P and A/T
Manually select full power
Select pitch attitude (20 degrees nose up for GPWS, 15 for Windshear, lower the nose until the stall warner stops / pitch decreases below PLI's.
Speedbrakes retract
Wings level
Leave rest of config as it is.
Best aircraft performance is achieved whilst 'nibbling' at the stall warner (intermittant stall warn).
Were the thrust levers firewalled? I know this has gone out of fashion but probably appropriate in that situation.
Finally, we are required to be stable at 1000ft (and MUST be stable at 500')- which includes the thrust levers being at approach power. I always check this with a view to discussing if we are likely to be stable by 500' and considering a go-around. I guess they didn't......
Last edited by Propellerhead; 5th Mar 2009 at 09:43.
Mach 3
Join Date: Aug 1998
Location: Stratosphere
Posts: 622
Likes: 0
Received 0 Likes
on
0 Posts
Those arguing the "bad system" design case ought to consider the fact that they weren't arguing the case prior to the event. You can't claim ignorance of the architecture.
Thats the whole point!
In lieu of the brace of other system cues, the AT logic emerged as is...
(The choice in this case seems pretty stark? Decouple from the GS and CFIT, or remain coupled with the GS, and stall!)
We stand on the shoulders of those who have gone before us.
Those wondering how common events where failing to notice airspeed decays of "large" magnitudes occur might like to ask their Chief Pilot/Flight Safety Pilot the answer to the question?
Those who reckon they've never failed to spot something simple during their mandatory proficiency checks are a statistical rarity. I find the finger pointing pretty hypocritical...
What do we learn/want to achieve here on out?
More training. 3 days re-freshers? Better CAT D sims. A requirement for all the previous 2(?) weeks Tech Log entrys to be made accessible to the crew on a one page precis...
Thats the whole point!
In lieu of the brace of other system cues, the AT logic emerged as is...
(The choice in this case seems pretty stark? Decouple from the GS and CFIT, or remain coupled with the GS, and stall!)
We stand on the shoulders of those who have gone before us.
Those wondering how common events where failing to notice airspeed decays of "large" magnitudes occur might like to ask their Chief Pilot/Flight Safety Pilot the answer to the question?
Those who reckon they've never failed to spot something simple during their mandatory proficiency checks are a statistical rarity. I find the finger pointing pretty hypocritical...
What do we learn/want to achieve here on out?
More training. 3 days re-freshers? Better CAT D sims. A requirement for all the previous 2(?) weeks Tech Log entrys to be made accessible to the crew on a one page precis...