PPRuNe Forums - View Single Post - Ethiopian airliner down in Africa
View Single Post
Old 30th Mar 2019, 00:05
  #2747 (permalink)  
fdr
 
Join Date: Jun 2001
Location: 3rd Rock, #29B
Posts: 2,956
Received 861 Likes on 257 Posts
Elephants

The details on the trigger event have blocked out the major issue in these events, that is, why the crew were unsuccessful in coping with the problems that they encountered?

Note: MCAS is not a stall prevention system, it is a system to achieve compliance with a static stability criteria for a part of the envelope.
The deficiency in the natural stability could lead to excessive g loading or inadvertent stalls etc as a consequence as attitude maintenance tasks need greater monitoring than with a compliant system. The rules themselves are not necessarily as relevant as they were in the days of early attitude displays, and with shorter body aircraft. The flight crew in a modern aircraft will notice the response on their coffee cup as they pull a bit harder on the prong, which is not simulated well in current 6-DOF FFS.

Two crews had issues that presented in such a way that what is obvious post fact was not at the time. Why? That is the question. While Bill Boeing needs to sort out his system that appears to have been based on various assumptions of response that did not occur as expected, the fundamental case remains that all conceivable failure modes of the total aircraft as a system are unlikely to be run down prior to experiencing the same in the air, and the crew response needs to be able to recover from unknown symptoms in a time critical period.

Lion Air is poignant where it appears the pilot during the fault finding attempted to apply human factor training concepts, and in doing so control was finally lost. Every time we push on a balloon it reshapes the balloon... actions have consequences. The fundamental tenets of priorities have not changed, nor does human factor training expect to do so, but on the day, the intervention that was holding ground and keeping the flight path managed was lost with the transfer of control. The transfer was undertaken to give the pilot more cognitive capacity to deal with the issue, however the preventive action was not maintained by the FO...

Control reversion with true FBW aircraft, alternate modes on Boeing FBW aircraft are quite different to Airbus, however as Boeing always uses a speed trim input from the trim switches, the reversion from full FBW to none on the Boeing is not a significant change. On the Airbus, the degradation of modes results in the introduced requirement to use manual trim, which is also displayed on the ECAM, however history shows that the crews in a high work load and stress environment do not always respond with manual trim inputs, resulting in loss of control.

AOA gauges are great, and indexers of any type are good tools in achieving stable approaches. HUD's are also great, but neither of them is an absolute necessity to the safe flight of the aircraft. Given adequate funding, they are the first equipment on the options list to ask for. For these cases, the symptom of the problem was the out of trim case, and the inability to simply re-trim and cure the out of trim case. The trim system on any aircraft is not a Möbius loop, it will end at some point, and the continued demand to re-trim was inconsistent with normal operation of the trim on any aircraft.

Somewhere in the design architecture decision making, the necessity of a correct crew response to a fault was either not considered, or was considered and assumed that the crew would readily identify the problem symptom as a trim runaway, and intervene using the method that has been incorporated for more than 50 years. That a crew would not be able to recognise and respond appropriately shows that human factors are alive and well in the design and certification process. It is the human that may make an incorrect assumption on a design or certification matter who also intervenes and ensures that other matters that would not be caught purely by compliance with preexisting rules are mitigated. To err is human, but removing humans from decision making is a high risk game.
fdr is offline