PPRuNe Forums - View Single Post - Air Asia Indonesia Lost Contact from Surabaya to Singapore
Old 1st Dec 2015, 18:29
  #3447 (permalink)  
alf5071h
 
Join Date: Jul 2003
Location: An Island Province
Posts: 1,257
Likes: 0
Received 1 Like on 1 Post
The report is well crafted and reflects a thorough investigation; furthermore the authors consider some speculative aspects from which we might learn. Obviously it can be read and interpreted in different ways, but there are aspects that could be applied to future activities.

I was surprised that there was not greater focus on the pitch trim which could have resulted in a nose-up moment complicating stall recovery. There may be similarities here with AF447; the condition is not a deep stall but one of a trimmed-in stall. If a separate warning is required it should be for an adverse trim condition at the stall.
Differences with FBW aircraft may be the lack of stick shake when approaching the stall, and stick push at the stall; a conventional tactile interface is more attention-getting than aural alerting during surprising and high workload situations.

Stall training in conventional aircraft use (should use) fixed trim whereby the speed of null-control-force feel represents stable, unstalled flight. Stall recovery involves a combination of pitch control force, attitude, and speed indications, forward control movement would be aided by stick push.
In unprotected FBW aircraft there is greater need to focus on attitude to reduce pitch - vice the absence of force feel. Further, if the aircraft is ‘out of trim’ then an off-centre control positon is required to reduce AOA and subsequent speed increase (pitch and yaw off centre in this instance, where the lack of yaw trim also contributed a roll control offset).
I doubt that any of these failures are trained / demonstrated in FBW aircraft or conventional aircraft.

The induced ‘upset’ is puzzling; following the inappropriate system interruption resulting in autopilot disconnect and roll attitude, the control inputs were not as expected (with hindsight). However, if crews normally use a relaxed rearward seat position during cruise with autoflight, then a sudden need for large or rapid manual control could result in unwanted inputs. Reaching for a side stick from an unusual positon would not involve the normal relationship for muscle-arm movement (learned skill), thus a roll input could have a significant pitch component. There could be similar problems with over-control in roll and with the required nose down movement (cf AF447).

The report's discussion of human factors is particularly well presented, but without factual confirmation human activity has to be speculative and carefully considered in context - surprise, knowledge base, and resultant awareness. Just because aspects are ‘easy to see’ after the event does not mean that the crew were in ‘error’, more likely as discussed in the report that the specifics of this particular rare situation and circumstances were at or beyond human ability.

Check trim, check seat position, review the assumptions made in training scenarios.
alf5071h is offline