Transasia Atr 72 Down In Taiwan

[Don Hudson]

One of the factors that may (or may not...remains to be disproved), be behind this is "confirmation bias". IIRC from the numerous stories, the captain had commented to maintenance prior to departure on the performance of the left engine. The story of "engine trouble" before takeoff, including log-book entries has been refuted by Taiwan's CAA but this may not be the end of this particular question.

Also, there were three on the flight deck, a third check-captain.

[Vsplat]

I would personally like to know who in the Taiwanese authority thought it was a good idea to post the DFDR data. Not everyone is as qualified to read it as Don is. And compelling graphs are not always truth telling. We don't yet know the why. And that is likely where the root cause ultimately will lie.

I can think of few greater threats to the protection of sensitive information - and chance for correct data interpretation before release - than this kind of behaviour.

I feel for the passengers, crew and their loved ones. This will do nothing but exacerbate an already brutally hard situation.

FWIW

Vs

[Seeker]

I would personally like to know who in the Taiwanese authority thought it was a good idea to post the DFDR data. Not everyone is as qualified to read it as Don is. And compelling graphs are not always truth telling. We don't yet know the why. And that is likely where the root cause ultimately will lie.

I can think of few greater threats to the protection of sensitive information - and chance for correct data interpretation before release - than this kind of behaviour.

I feel for the passengers, crew and their loved ones. This will do nothing but exacerbate an already brutally hard situation.

FWIW

Vs

At first I thought the same but they must be pretty confident to release this info. It is strange though that on one hand they are saying the pilots made an error while on the other hand the entire fleet is grounded to "investigate" the cause.

[Vsplat]

I agree, the decision to release reflects a high degree of confidence. Is the confidence justified at this point in events? I have to wonder if the working level investigators agree with the call. I have a hard time believing that anyone who has ever investigated an accident would be comfortable with a release like this so soon.

Wrong on too many levels for me, but then I am thousands of miles away and just a hug away from my loved ones. Not so easy for the bereaved in Taiwan. I understand the desperate need to know. I wonder if this has actually helped.

Vs

[Don Hudson]

Vsplat;

The early release of data was a surprise to me too. Normally, such release would be associated with system or component failure but in all my reading so far I have seen nothing indicating a problem with the engines or wiring of warning systems. We don't know why the Taiwanese authorities were anxious for an early release but it may be associated with the reason for grounding of the fleet which was for an inspection to ensure normal functioning of fuel and engine systems before returning the aircraft to service. As well, the interactions of government agencies may be a factor in such release.

[DEFCON]

Yesterday, the media advised the data would be released to investigators today, but the story was presented in such a way that it left the impression the authorities intended to provide the material to the public as well, which did seem a little unusual. Satisfying cultural considerations may explain the early release of the data?

[DEFCON]

The media is reporting the cockpit was occupied by three Captains, one being a Checker. If the report is accurate, the voice recorders may offer some interesting insight as to CRM's role in the decision making process that led to the actions taken.

[Boney]

CNN reported that the captain had issues with the left engine, to be confirmed. The right engine failed after takeoff. Under stress, one can see what happened next.

Lets hope that a report is issued sooner rather than later, which includes what was on the cockpit voice recorder.

My prayers go out to all of their families.

[Guest]

Is this an acknowledgement of inadequate qualifications or simply a CYA exercise?

TTransAsia says all its 71 ATR aircraft pilots face skills test following crash

PUBLISHED ON FEB 6, 2015 10:26 PM

TAIPEI (AFP) - Taiwan's TransAsia Airways said Friday all its 71 ATR pilots will have to take a flight skills test following its second deadly accident in seven months

In a statement, the airline said "71 pilots on its fleet of 10 ATR planes will be required to do a test by the Civil Aeronautics Administration and a professional unit to make sure they are all qualified on their jobs."

- See more at: http://www.straitstimes.com/news/asia/east-asia/story/transasia-says-all-its-71-atr-aircraft-pilots-face-skills-test-following-c#sthash.lZQ1fzBd.24rkIxgb.dpuf

[mo32a]

Yikes, that's a little damning. So basically they are grounding their pilots until they can prove competency.

[flyinghigh]

Don't see anything wrong with releasing accurate data. I remember the NTSB did this very quickly as well. I suspect that the data released on this crash will be just as relevant in 6 months as it is now. The bottom line is that we know that the wrong engine was shut down. No doubt in the last 24 hours there have been thousands of airline pilots that have reviewed to themselves as a reminder, how to identify a failed engine along with thinking about ways that such an error could be made such as by rushing. That is likely a safety enhancement in itself due to the information released,

[Vsplat]

You'll have to refresh my memory on the NTSB event.

I can't recall a similar sequence from them - grounding the aircraft fleet, then releasing raw DFDR data, then grounding the pilots, all within 3 days of the accident. I can recall rather aggressive media releases after the SFO accident, but nothing like this.

The concern I have with the data release in particular is what form of vetting one can really do in the short time post accident.

I agree with your concern about errors created by rushing. That phenomenon can affect the team dealing with the phalanx of stimuli post accident.

The pressure to release data is overwhelming at present. Has anyone cross checked the data to ensure the parameters are actually telling the whole story? Has anyone else ever encountered recorded data where one or more parameters was swapped, or reverse-mapped? How about engine warning systems that were cross-wired, so say an engine fire lights up the incorrect handle or shut-off? I'm not saying any of this actually happened here. It's just hard for me to square that they have verified what they should have before putting this out.

All just my opinion.

Vs

[Specs]

Anybody care to comment and educate me a bit on what is happening with PLA #1 at just prior to 02:52:43 where the trace shows the beginning of a stepped reduction?

[inchman]

One interesting thing is the interpretation of the photos as the aircraft was about to hit the bridge. We were all looking for a single feathered engine and most had picked one out. But clearly, based on the FDR data, they were both feathered. I know that we were looking at grainy photos, but even industry professionals can interpret visual information badly and draw poor conclusions in the absence of hard data and assumptions about what must have happened.

Another question often brought up often on avherald and pprune about other accidents... most recently AirAsia... is "Why didn't they call a MAYDAY?". Clearly it had no positive effect on the outcome of this accident and, in fact, may have been part of the cause.... inattention to task to make a call to someone who could not possibly have had an effect on the positive control of the aircraft. Won't know that for sure until the CVR and FDR are synced.

[Kip Powick]

Twin engine aircraft.............Engine Failure

Dead Foot = Dead Engine (which foot does not have to push rudder to keep the aircraft straight?)

CPD

Control......(get control)........Power..(Max power to operating engine)........Drag...(clean up aircraft at appropriate speeds)

Declare Emergency

At least that was what I was taught.....

[Seeker]

That's all correct Kip except for a 2 crew aircraft the pilot who identifies the engine doesn't have a dead foot - for him/her it's best to look at the gauges and decide which engine isn't doing what it's supposed to be doing and then get confirmation from the PF before slowly moving the levers.

[J.O.]

I don't mean to be glib but the confirming pilot has the dead foot.

[Kip Powick]

Perhaps I've misunderstood your comment, but it seems to me you're suggesting that the PNF/PM should move the thrust levers. In all my time flying ME aircraft I've never seen anyone but the PF handle the thrust levers. Is it done differently elsewhere?

Not in my world......the PF flying identifies the dead engine and it has to be the DEAD FOOT engine. He can ask to confirm which is the DEAD engine but the PF moves the thrust levers and calls for all other actions.

[Moon The Loon]

Perhaps I've misunderstood your comment, but it seems to me you're suggesting that the PNF/PM should move the thrust levers. In all my time flying ME aircraft I've never seen anyone but the PF handle the thrust levers. Is it done differently elsewhere?

Jeff, I've seen 6 of one, half dozen of the other techniques. My current operator has the PM handling the power lever/confirmation and subsequent shutdown. I've always liked that approach as many times the PF has his hands full just maintaining the departure track. Keep in mind complex SIDs, varying levels of automation, dark/low visibility/turbulent conditions, escape manoeuvres/special procedures. It's a lot easier to fly a memorized procedure especially if it's different from the SID and monitor the PM than fly the memorized procedure, action the failed engine while the PM is idle, and maintain the required accuracy of the flight profile.

Two pilot operation, crew operation - it all means you have to trust your partner to do his/her job. He/she watches you fly the departure accurately; he/she actions the correct confirmations and actions asking you, the PF to confirm at a convenient time.

It all comes back to training and how company's enforce their SOP's, sometimes notwithstanding the cultural difficulties involved.

My opinion.

[Moon The Loon]

Not in my world......the PF flying identifies the dead engine and it has to be the DEAD FOOT engine. He can ask to confirm which is the DEAD engine but the PF moves the thrust levers and calls for all other actions.

May not work with a bad propeller overspeed. Reliance on the engine instruments in such a case is a must. Same with an unscheduled prop feather where engine instrument gauges are more difficult to interpret. All when dealing with turboprops. I don't have enough experience on SSBB engines (radial/horizontally opposed/inline).

Bottom line in today's world. IDENTIFY... CONFIRM... ACTION/FEATHER.

Same as it was in your day!!!

Which, by the way, is rapidly becoming my day too!

[Kip Powick]

It all comes back to training and how company's enforce their SOP's, sometimes notwithstanding the cultural difficulties involved.

Very true......my experience was Mil and 'our' procedures worked well.

The A310 in WD had the PF calling for the actions as the PNF called out the ECAM instructions and before the PNF touched anything the PF confirmed what the PNF was going to do. In the "non-glass" aircraft the procedure had the PF calling for the actions with, if required, the PNF confirming items.

In this accident based on what I only saw on TV it looked like the PF applied wrong rudder and could not recover but based on what I have read, and the posted FDR...well I am still a bit confused as to what went on in the cockpit as well as what actually caused the loss of control

[J.O.]

The Airbus generic SOP states that the PF handles the thrust levers at all times, but not every operator follows their recommendation. Personally I've never understood why the PM would ever move a thrust lever other than fine tuning the takeoff thrust on an aircraft that doesn't have auto throttle

[Seeker]

I've worked at three different operations and at every one of them the PM moves the levers in a shutdown. Honestly, I didn't even think that any place would do it differently. The idea is that one pilot can devote all his attention to flying accurately while the other carefully selects the correct lever to pull, of course the PF is still tasked with confirming the lever and monitoring.

[Guest]

On the DH8 series, including the -400, one of the first things done after recognizing an engine malfunction is to disconnect the autopilot. The airplane is hand-flown through the entire shutdown drill. Since the PF is usually physically busy with that task, the PM is the one assigned to make power and condition lever movements. Only after the drill is complete and the aircraft re-trimmed does the AP get re-engaged.

I know that's not common- the autopilot on the Dash is a 2-axis design (big downfall of the -400...but that's what happens when the 'common type' camp wins). Perhaps the ATR is different? I believe the ATR500/600 has autoland so I'm sure it's autopilot is much more capable.

Moon has a valid point with the prop overspeed. It requires careful identification, and 'feel' alone won't do it.

[Boney]

2 crew operation. PF flies, NFP runs the checklist including the movement of switches and levers, with confirmation from the FP, when required, as per QRH and/or SOP.

In my earlier years, it was the PF that did this from memory, but has changed over the years. I prefer today's procedures, where memory actions are limited to just a few items.

An issue is identified by either pilot and the FP calls for the appropriate check list that the NFP read and do.

Reduces both stress and workload.