PPRuNe Forums - View Single Post - AMR 587 Airbus Crash (merged)
View Single Post
Old 22nd May 2003, 23:44
  #237 (permalink)  
cwatters
 
Join Date: Dec 2001
Location: England
Posts: 1,389
Likes: 0
Received 0 Likes on 0 Posts
> .......we go back around in circles in this topic!

Not quite my intention. I too have been back through all 16 pages.

I wanted to know if anyone had seen any research on how well (or how badly) people are able to use variable sensitivity controls. Can people adapt or do they react badly when something unexpected happens? I spent some time looking for papers on the web but couldn't find anything.

I suspect that when someone is "surprised" they subconsiously revert to using controls in a way that represents the sum or average of their lifetime experience, rather than how they are behaving at that particular moment.

Jshg wrote..

>I recall a cross-wind approach some years ago when
>the PF reverted to his previous turboprop type and
>applied rudder in the last 500'.

Wino also made the point..

>You will see that Sten used no more than 1/3rd or so
>of the available rudder travel to keep the plane
>rolling down the runway. When the wake encounter
>happened, those same inputs are now FULL TRAVEL inputs.
>There is no way in 40 seconds or so he could adjust to
>the increased sensitivity,

I suggest that even if he had adjusted to the increased sensitivity he may have been powerless to prevent himself "reverting" when suddenly surprised by the wake encounter.

A common design issue with all systems that contain feedback is the need to maintain stability and avoid conditions that cause the system to oscillate. Anything that adds a time delay in the loop usually makes this harder to achieve. This paragraph I found on reaction times might be relevant..

http://www.visualexpert.com/Resources/reactiontime.html

>Psychological Refractory Period
>
>Following a response, people exhibit a "psychological
>refractory period." During this period, new responses
>are made more slowly than if there had been no previous
>behavior. For example, suppose a driver suddenly steers
>lefts and then right. The steer-right response will
>occur more slowly because it immediate followed the
>steer-left.

Which seems to imply that when you are trying to damp out oscillation your reaction times are reduced. "Heavy" forces would also add a time delay.

Whatever the true position, it appears the pilot in 587 felt the plane was oscillating and applied several rudder reversals to try and stop this.

(and yes I have read the posts that say he shouldn't have used rudder anyway).
cwatters is offline