TransAsia in the water?
Join Date: Feb 2010
Location: Norden
Posts: 138
Likes: 0
Received 0 Likes
on
0 Posts
It. is all in the data.Indeed.Just read it..
Nothing wrong with No2.
Uptrim No1 for no reason.
Bleed off No1 correct,but TQ NOT increasing! Objective TQ is a calculation only.
2,15 Second after incomplete uptrim of No1 autofeather on the good engine No2.
Nothing wrong with No2.
Uptrim No1 for no reason.
Bleed off No1 correct,but TQ NOT increasing! Objective TQ is a calculation only.
2,15 Second after incomplete uptrim of No1 autofeather on the good engine No2.
Join Date: Jul 2013
Location: equatorial side of the Polar Jet
Posts: 193
Likes: 0
Received 0 Likes
on
0 Posts
non standard off the book malfuction
Firstly.does anyone have a record of the time on ATR72-600 this combined crew had.Had they flown any other ATR series before?Have they flown any other turboprop before before? How much multi crew experience did they have before or were they predominantly military single pilot operation cultivation? This might lead to explain how they behaved if at all in appropriate.
I reckon non standard (non QRH referable) malfunctions, such as this MIGHT have been, at such a low altitude on a critical departure phase wuld likely lead to instincts taking over as there would be litle time to diagnose, formulate and decide and execute appropriately.I posted earlier an unwanted auto feather on an ATR72 on decent.It would be interesting to discover on which engine that took place and what the status of ATPCS, EEC and other related indications and mechanisms such as the Propeller Valve Module were commanding and indicating.The constant PL auto regulated power Management is quite unique to ATR as a turboprop and I wonder what anomalies were discovered during flight testing prior to certification. It might be that ATR the company knows something about this but is keeping tight lipped for legitimate reasons? The truth can only be collborated through CVR analysis...and the aircraft's techlog review..I suspect Transasia Engineers also might have something to contribute to this mystery.
I reckon non standard (non QRH referable) malfunctions, such as this MIGHT have been, at such a low altitude on a critical departure phase wuld likely lead to instincts taking over as there would be litle time to diagnose, formulate and decide and execute appropriately.I posted earlier an unwanted auto feather on an ATR72 on decent.It would be interesting to discover on which engine that took place and what the status of ATPCS, EEC and other related indications and mechanisms such as the Propeller Valve Module were commanding and indicating.The constant PL auto regulated power Management is quite unique to ATR as a turboprop and I wonder what anomalies were discovered during flight testing prior to certification. It might be that ATR the company knows something about this but is keeping tight lipped for legitimate reasons? The truth can only be collborated through CVR analysis...and the aircraft's techlog review..I suspect Transasia Engineers also might have something to contribute to this mystery.
Last edited by Trackdiamond; 11th Feb 2015 at 09:52.
Join Date: Mar 2008
Location: Milano
Age: 38
Posts: 309
Likes: 0
Received 0 Likes
on
0 Posts
In my opinion the first event is the engine 2 who went idle by himself without a change of the PL, this generates the ATPCS intervention whith UPTrim and Autofeather after 2.15 sec.
Uptrim worked correctly but someone reduce the PLA to keep 90%...
Uptrim worked correctly but someone reduce the PLA to keep 90%...
Join Date: Apr 2003
Location: Germany
Age: 76
Posts: 1,561
Likes: 0
Received 0 Likes
on
0 Posts
I guess we will need the full report to understand who did what when, and in response to what oral commands. It's very difficult to understand what happened from the little we know.
At a guess, having two captains instead of the usual crew, a captain and an FO, may have led to some non-standard CRM after whatever triggered this accident. This is something that only an analysis of the CVR information can tell about, I think.
It seems from what we have read so far that the crew must have first either lost power on the right engine for some unknown reason, or else have been presented with a false indication of loss of power on the right engine. It seems that they then shut down either the (good) left engine or perhaps both engines, in response to whatever happened first.
Whatever happened to cause that, the crew seems to have found themselves with no power, at low altitude and low speed, in a place where there were not many places to make a reasonably safe off-airport landing.
Those videos seem to show an aircraft that is stalled and out of control, since it's first sinking, and then dropping its nose and left wing as it passes out of sight over the bridge. Too, the images of the wreckage in the river show it as inverted, which must mean t a fairly high rate of roll in the last few seconds of flight to end up that way.
At a guess, having two captains instead of the usual crew, a captain and an FO, may have led to some non-standard CRM after whatever triggered this accident. This is something that only an analysis of the CVR information can tell about, I think.
It seems from what we have read so far that the crew must have first either lost power on the right engine for some unknown reason, or else have been presented with a false indication of loss of power on the right engine. It seems that they then shut down either the (good) left engine or perhaps both engines, in response to whatever happened first.
Whatever happened to cause that, the crew seems to have found themselves with no power, at low altitude and low speed, in a place where there were not many places to make a reasonably safe off-airport landing.
Those videos seem to show an aircraft that is stalled and out of control, since it's first sinking, and then dropping its nose and left wing as it passes out of sight over the bridge. Too, the images of the wreckage in the river show it as inverted, which must mean t a fairly high rate of roll in the last few seconds of flight to end up that way.
Last edited by chuks; 11th Feb 2015 at 09:20.
Join Date: Apr 2010
Location: U.K.
Posts: 33
Likes: 0
Received 0 Likes
on
0 Posts
It. is all in the data.Indeed.Just read it..
Nothing wrong with
No2.
Uptrim No1 for no reason.
Bleed off No1 correct,but TQ NOT
increasing! Objective TQ is a calculation only.
2,15 Second after
incomplete uptrim of No1 autofeather on the good engine No2.
Everything, airframe wise, worked as it should, Flameout detected so #2 engine reduced to idle without any PLA movement and autofeathered, #1 Engine Objective TQ increased to 100% and the engine responded, the uptrim of #1 was due to #2 reducing.
#1 engine followed exactly what was commanded, the crew reduced #1 PLA and the engine followed, why they did this is the mystery, the data shows no evidence of an engine stall or surge.
During the decrease of #1 engine PLA they also increased the #2 PLA, which at this point did nothing as the engine was at idle, my thought is they either had not noticed the #2 engine producing no power or had firewalled the power lever in an attempt to get some thrust back.
Someone also mentioned the spikes on the TQ data, I have seen this before on other aircraft and believe it's due to the signal being zero on that channel, if you notice the Oil px it does exactly the same when the #1 engine is shut down.
I find it hard to find a reason why they would shut down #1 engine if they knew there was an issue with the #2, even if the #1 was stalling, surging, making noises or anything else, so close to an airport, over urban terrain.
The only exception would be fire from #1, however the gradual reduction in PLA does not show standard engine fire procedures.
They either didn't know #2 was the trouble or if they did then they moved all the wrong engine levers.
Join Date: Dec 2014
Location: Bangor, Northern Ireland
Posts: 10
Likes: 0
Received 0 Likes
on
0 Posts
10 of TransAsia's pilots suspended after failing oral test
"... Taiwan's regulator revealed 10 of TransAsia's 49 ATR pilots have failed oral proficiency tests on handling an aircraft during engine failure"
See halfway down TransAsia Offers £307,000 To Crash Families
See halfway down TransAsia Offers £307,000 To Crash Families
Join Date: May 2009
Location: UK
Posts: 2
Likes: 0
Received 0 Likes
on
0 Posts
A bit more detail provided here...
Taiwan orders all airlines review safety after bad test results | Reuters
Taiwan orders all airlines review safety after bad test results | Reuters
Taiwan's aviation regulator said 10 of TransAsia's 49 ATR pilots had failed oral proficiency tests on handling the aircraft during engine failure. A further 19 pilots did not take the test, due to sickness or because they were not in Taiwan, the Civil Aviation Authority (CAA) said. The 29 pilots who failed or did not take the test have been suspended, the CAA said.
[snip]
"Some of us have stayed up all night to prepare for the tests. The result will affect our career developments significantly," said the pilot, declining to be identified due to the sensitivity of the issue.
[snip]
"Some of us have stayed up all night to prepare for the tests. The result will affect our career developments significantly," said the pilot, declining to be identified due to the sensitivity of the issue.
Join Date: Jul 2013
Location: South East England
Age: 70
Posts: 39
Likes: 0
Received 0 Likes
on
0 Posts
FlightSpanner:
Indeed, and although this faulty "thought" caused the string of events to start, I wonder if they would have done any better if there was a real Flameout? I would expect a real flameout to be shown by a *really* sudden, low ITT reading, as it would be being sprayed by unlit fuel/air mixture, which would cool it down very fast (unless the flameout was due to fuel starvation, obviously, in which case it would just be cold air, but still much cooler than burning gasses).
Can anyone who's seen one comment on what a real flameout looks like on the gauges?
...
To me it suggests they thought that No.1 had failed, and they did a manual "uptrim" of what they thought was the good engine, without checking the instruments to confirm. What indication is there that the propeller is feathered, as it was when this firewalling of PLA2 took place?
That was me! So the system shows that sawtooth pattern to indicate: "This sensor is working, but is reading zero"? Sounds like a good idea, just surprised nobody has mentioned that - maybe a little-known characteristic?
It's all in the data, except for why #2 Engine detected Flame Out, from the data it does indeed look like there was nothing wrong with #2, however the EEC thought there was a Flameout.
Can anyone who's seen one comment on what a real flameout looks like on the gauges?
...
During the decrease of #1 engine PLA they also increased the #2 PLA, which at this point did nothing as the engine was at idle, my thought is they either had not noticed the #2 engine producing no power or had firewalled the power lever in an attempt to get some thrust back.
Someone also mentioned the spikes on the TQ data, I have seen this before on other aircraft and believe it's due to the signal being zero on that channel, if you notice the Oil px it does exactly the same when the #1 engine is shut down.
Join Date: Jan 2015
Location: Canada
Posts: 11
Likes: 0
Received 0 Likes
on
0 Posts
"Some of us have stayed up all night to prepare for the tests. The result will affect our career developments significantly," said the pilot, declining to be identified due to the sensitivity of the issue."
What?? What does that say about their proficiencies if they stayed up all night?? If they were flying and were faced with this kind of emergency I would hope they would know what to do, and not study all night to address the problem! I don't think the aircraft would be aloft long enough in an emergency for them to "study".
What?? What does that say about their proficiencies if they stayed up all night?? If they were flying and were faced with this kind of emergency I would hope they would know what to do, and not study all night to address the problem! I don't think the aircraft would be aloft long enough in an emergency for them to "study".
Join Date: Feb 2005
Location: UK
Age: 85
Posts: 697
Likes: 0
Received 0 Likes
on
0 Posts
no more room for pilots on modern airplanes.
At any given moment there are over 5,000 aircraft in the air, 24 hours a day and 7 days a week.
Because of the complexity most of these will be controlled by computers which can take decisions quicker than any human and store masses of information.
Once in a while, maybe once every month or so, i.e. once every million flights, the aircraft presents problems that are outside of the ability of the computer.
In these situations that aircraft will crash unless a 'human' takes over.
In 90% of the cases the pilot, due to training and experience, will rectify the situation and resolve the problem, often with little loss of life or aircraft.
Occasionally the pilot is not up to the situation either because the situation is beyond him/her or due to lack of personal ability. This results in tragedy.
The human pilot is always the last chance when everything else goes wrong. The answer lies in better pilot training and company procedures not removing the human totally.
Last edited by funfly; 11th Feb 2015 at 11:38. Reason: grammar
Join Date: Apr 2010
Location: U.K.
Posts: 33
Likes: 0
Received 0 Likes
on
0 Posts
HDRW
Unfortunately I cannot answer if this is by design or not, I've seen this when testing installations, once valid data was received the plot was normal, again I am no FDR analysis expert, just what I have seen during installation testing.
EnchevEG - can you backup claims that the FDR plot is fake? Must admit it did for a nano second cross my mind too, mainly because it was released so quick and also (as many others have said) it just messes with my head to think they shut the wrong motor down, all the indications/warnings were there for a #2 engine problem.
In the cockpit were six eyes, all able to see #2 Engine Flameout CAS and #2 engine TQ, ITT, NP & Fuel flow indicators.
It beggars belief!!
That was me! So the system shows that sawtooth pattern to indicate: "This
sensor is working, but is reading zero"? Sounds like a good idea, just surprised
nobody has mentioned that - maybe a little-known characteristic?
sensor is working, but is reading zero"? Sounds like a good idea, just surprised
nobody has mentioned that - maybe a little-known characteristic?
EnchevEG - can you backup claims that the FDR plot is fake? Must admit it did for a nano second cross my mind too, mainly because it was released so quick and also (as many others have said) it just messes with my head to think they shut the wrong motor down, all the indications/warnings were there for a #2 engine problem.
In the cockpit were six eyes, all able to see #2 Engine Flameout CAS and #2 engine TQ, ITT, NP & Fuel flow indicators.
It beggars belief!!
Psychophysiological entity
. . . and the fuel booster pump question I've asked before.
Leaving them off has caused similar symptoms on earlier marks. At least twice with one company.
So, automatic on the 600 or still a vital something 6 eyes could miss?
No one?
Leaving them off has caused similar symptoms on earlier marks. At least twice with one company.
So, automatic on the 600 or still a vital something 6 eyes could miss?
No one?
Join Date: Mar 2009
Location: Malaysia
Posts: 112
Likes: 0
Received 0 Likes
on
0 Posts
Closing the stable door after the horse has bolted?!
Ten TransAsia pilots fail tests on how to handle plane if engine fails - Asia - World - The Independent
Can anyone please explain the airline management rationale for testing all it's pilots on type after the event?
Ten TransAsia pilots fail tests on how to handle plane if engine fails - Asia - World - The Independent
Can anyone please explain the airline management rationale for testing all it's pilots on type after the event?
Join Date: Dec 2013
Location: Dallas
Posts: 108
Likes: 0
Received 0 Likes
on
0 Posts
The discussion about whether the #2 flameout warning and automatic systems intervention was legitimate is quite beside the point. That sort of failure may occur, although highly unlikely.
More significantly, after the #2 autofeather the aircraft continued to climb quite nicely. For a few moments, that is.
It appears that proper cockpit procedures would have made this a non-event.
More significantly, after the #2 autofeather the aircraft continued to climb quite nicely. For a few moments, that is.
It appears that proper cockpit procedures would have made this a non-event.
Join Date: Apr 2003
Location: Germany
Age: 76
Posts: 1,561
Likes: 0
Received 0 Likes
on
0 Posts
A Cockpit Voice Recorder is really there to aid in crash investigation. If the authorities want to, then I suppose they could release the CVR recording or its transcript to the general public ahead of the full report, which usually takes at least a year to prepare, but that might not be a good idea.
For one thing, in this case I suppose the crew were speaking Tawainese, so that most of us would need a translation to understand what was said. Too, we would usually need to combine the translated CVR transcript with the FDR data to see both what was said, but also what was done.
For instance, hypothetically, perhaps the CVR shows the PF commanding "Shut down Engine Number Two," but then the FDR shows the left engine, the wrong one, being shut down by the PNF. In a situation such as that one you would need more than just the CVR to understand what happened.
For one thing, in this case I suppose the crew were speaking Tawainese, so that most of us would need a translation to understand what was said. Too, we would usually need to combine the translated CVR transcript with the FDR data to see both what was said, but also what was done.
For instance, hypothetically, perhaps the CVR shows the PF commanding "Shut down Engine Number Two," but then the FDR shows the left engine, the wrong one, being shut down by the PNF. In a situation such as that one you would need more than just the CVR to understand what happened.
Join Date: Nov 2011
Location: Wales
Age: 73
Posts: 39
Likes: 0
Received 0 Likes
on
0 Posts
Yes it is incompetent training.
Training in these areas, especially EFATO has been dumbed down over the years. The main blame can be laid at the door of the regulators by not ensuring technical knowledge competence after the scrapping of Performance A exams. This is a complex subject not understood by a great number of pilots. Whilst I was training new entry pilots I was amazed by the lack of knowledge. The previous poster was suprised that the pilots that were tested stayed up all night revising. Well I am sorry to shater your world, but the majority of aircrew do study for exams and sim checks.
There has been a rising number of accidents caused by lack of knowledge and this will continue to rise.
Training in these areas, especially EFATO has been dumbed down over the years. The main blame can be laid at the door of the regulators by not ensuring technical knowledge competence after the scrapping of Performance A exams. This is a complex subject not understood by a great number of pilots. Whilst I was training new entry pilots I was amazed by the lack of knowledge. The previous poster was suprised that the pilots that were tested stayed up all night revising. Well I am sorry to shater your world, but the majority of aircrew do study for exams and sim checks.
There has been a rising number of accidents caused by lack of knowledge and this will continue to rise.
Join Date: Jul 2004
Location: Found in Toronto
Posts: 615
Likes: 0
Received 0 Likes
on
0 Posts
I agree. Incompetent training and/or incompetent recurrent training and checking.
It would only take a day or two to examine the FDR and CVR to have a good understanding of what happened here.
The fact that the Taiwan's Civil Aeronautics Administration has ordered the pilots to undergo proficiency tests tells me they probably know exactly what happened and why.
It would only take a day or two to examine the FDR and CVR to have a good understanding of what happened here.
The fact that the Taiwan's Civil Aeronautics Administration has ordered the pilots to undergo proficiency tests tells me they probably know exactly what happened and why.