Wrong Engine...

[Mitch Cronin]

Full article TransAsia pilot shut 'wrong engine'

"Wow, pulled back the wrong side throttle." said the Captain...

[Moon The Loon]

Yup.

http://www.theglobeandmail.com/news/news-video/video-taiwan-crash-pilot-pulled-the-wrong-throttle/article25224418/

Article says Captain had failed a PPC 9 months prior to the crash. Not necessarily a factor standing-alone. Many of us have failed a ride over a 30+ year career for one reason or another. Not something to be proud of obviously but also not something that should condemn a pilot for the rest of his/her life.

Can't remember the stats on the crew - age, experience, etc. I do wonder though if culture played a part in the haste to shut down an engine, especially on an aircraft with a proven autofeather system.

[Guest]

For those not familiar with this accident the original discussion can be found at : http://theairlinewebsite.com/topic/406057-transasia-atr-72-down-in-taiwan/?hl=transasia

[DEFCON]

It would be nice to know why he failed the ride, or just how bad said failure was considering the huge number that were actively flying, but later failed to meet the grade during verbal post crash testing on engine failure procedures.

[Guest]

Apparently a double crew error.



TransAsia probe indicates crew should have aborted take-off

By: DAVID KAMINSKI-MORROW

Source:

in 5 hours

Pilots of an ill-fated TransAsia Airways ATR 72-600 should have aborted its departure from TaipeiSongshan airport after noting that a power-control system had not armed, investigators have indicated.

Flight GE235 crashed into the Keelung river shortly after take-off when the crew inadvertently reduced power from its healthy left-hand engine, instead of the right-hand engine which was exhibiting problems.

Taiwanese investigators have disclosed that the aircraft’s automatic take-off power control system (ATPCS) had not initially been armed when the turboprop began its departure roll.

The system provides automatic support – including uptrim and auto-feathering – in the event of an engine failure during take-off.

Investigators probing the fatal 4 February crash, citing flight-recorder data, state that the monitoring pilot, in the right-hand cockpit seat, declared the ATPCS was not armed about 4s into the take-off roll.

In an interview with the inquiry team a TransAsia crew training supervisor said that ATR 72-600 pilots should abort take-off if the ATPCS is not armed during the roll, and that this point was “emphasised” during pilot training.

This requirement to abort is based on there being several procedures to be carried out if the ATPCS is not armed, which are unsuitable to perform during the take-off run.

But the inquiry points out that this policy of aborting is “not clearly described” in any manuals and notices to flight crew. The situation is further complicated by a policy allowing pilots of ATR 72-500s, faced with the same issue, to continue the take-off under certain conditions.

TransAsia flight GE235’s commanding pilot acknowledged the unarmed ATPCS call-out but opted to proceed with the take-off roll.

Just after the standard 70kt speed call, the monitoring pilot exclaimed that the ATPCS had become armed, before the ATR became airborne a few seconds later.

Some 34s into the climb the ATPCS appears to have commanded an uptrim to the left-hand engine, to increase its power, and started auto-feathering the right-hand engine – these actions are the ATPCS system’s normal response to a developing problem in the right-hand powerplant.

[DEFCON]

Had the PL's even reached the angle necessary to arm the auto-fx system when the call was made? If not, other than indicating the FO's call may have been premature, there's no information in the report suggesting there ever was a problem with the aircraft, or its systems.

[Guest]

Is the arming a manual act prior to commencing the takeoff roll, in other words was it missed during their pre takeoff check list?

[DEFCON]

When setting the aircraft systems up for takeoff, the unit is turned on, which places it into a form of standby. When the thrust levers are advanced and pass through a predetermined position in the throttle quadrant, just a guess, but say it's around 65 degrees, the system will automatically arm itself.

When takeoff power is set the system will activate and feather a given prop should certain conditions be met. For example, if the torque on an engine drops below a predetermined value with both thrust levers above a certain position / angle, the system will intervene and automatically feather the offending prop with no further input on the pilot's part.

[Guest]

thanks, when you say "the unit is turned on" is this a separate manual action or automatic?

[vanishing point]

Having never flown an ATR, I can only comment on the Dash 8 which uses similar engines and (I suspect) a similar auto-fx system.

On the Dash, the auto-fx system is activated manually by a switch light on the panel. This is done by the Captain as part of the after-start checks.

The switch light stays in this position until deactivated as part of the after-take-off checks. It's not used again until the next after-start checks.

In the "activated" mode, the auto-fx will arm during take-off when a certain power-lever-angle (PLA) is met. The protocol at Jazz is that the auto-fx system must arm by the time the take-off-power torque value is met. If it does not, a rejected-take-off (RTO) is initiated. I have no idea what the PLA is to arm but I am looking at torque values (measured in percentage). The minimum power setting for take-off on a Dash 8-300 is 82%. As long as the auto-fx system arms by this value, we're going flying. Some Dash's arm at 60%; some at 70%. It's all in the rigging.

An aircraft can be dispatched with an inoperative auto-fx system as per the MEL but with certain restrictions.

[Guest]

Thanks "vanishing point" So it appears likely that the manual switch was not activated.

[vanishing point]

Perhaps but we'll have to wait for the final report.

If the auto-fx system wasn't armed or was armed but didn't activate with the #2 engine failure, the crew should still know how to feather it manually and shut it down properly.

Jazz's protocol with an engine malfunction after rotation: once max power is set and the gear is retracted, the pilot monitoring (PM) identifies the malfunction and confirms whether the prop is feathered or not. If it's NOT feathered (or if there's an engine fire), the pilot flying (PF) calls: "Engine #__ Shutdown Drill".

If the prop did feather (and there's no fire), the crew waits until "level-off height" (typically 1000' AGL unless local terrain specifies a higher height) then goes through the "Engine Shutdown Drill" and the emergency checklist.

So with the crew above, they inadvertently shutdown the good engine for whatever reason. Why? That remains to be seen.

Without reading back through the previous posts; was the #2 engine actually feathered? If it was, the auto-fx system did its job. If it wasn't, then the system did not work as designed (either wasn't armed nor failed to activate).

[DEFCON]

VP

I was wondering if you can confirm whether the switch position is guarded by the take-off configuration warning system?

[Guest]

On the Dash 100 through 400 the autofeather is not connected to the takeoff warning. It can be MEL'd on the "classics" (quite restrictive), but not on the 400. Not sure about the ATR600 though.

However, I cannot even fathom intentionally continuing a takeoff in a machine of that size with the system not functioning. It's like going flying without fire/smoke protection. They intended to fly without it, however it apparently armed just prior to getting airborne. Now we know that their decision making and leadership/followership skills are near zero.

Then the engine flames out (that's what they say in the transcript and that's what their engine warning system indicated). So one would expect that in the year 2014 that an air carrier would have some "standard calls" for such an event? Apparently not. What I'm reading in the transcript (rough translation, and also not complete, I understand) is nothing but jibberish. They might as well have been speaking Slobovian to each other. Communication skills near zero. Of course their situational awareness was also near nothing; they had no idea what their airplane was into, or what it was telling them. Icing on the cake: the PF tries twice to engage the autopilot afterwards as both engines are producing zero power, the airplane is flirting with the stick shaker and they are sinking fast through about 600'. Workload management? Well there was some increased workload with no management.

This was a perfectly flyable airplane with a simple engine flameout. It should've been a non-event. But these guys were an accident waiting to happen: they had no business in that airplane. I wouldn't have trusted the 2 of them to organize a carpooled drive to work together. With their pilots lacking CRM skills, system & procedural knowledge and plain old flying skills, the poor folks in the back were doomed from the start.

Just my opinion (frustrated one at that).

[vanishing point]

CH is correct. The auto-fx switch light is not part of the take-off warning horn system. As it is a requirement to confirm the auto-fx system "arms" prior to take-off, this would be redundant.....not that there's anything wrong with redundant systems.

As well, the PLA required to set off the warning horn is much lower than that required to arm the auto-fx system.