PPRuNe Forums - View Single Post - Use of autopilot for instrument rating tests
Old 20th Mar 2011, 09:55
  #3 (permalink)  
Piltdown Man
 
Join Date: Jun 2002
Location: Wor Yerm
Age: 68
Posts: 4
Likes: 0
Received 0 Likes on 0 Posts
I can fully accept that manually flying an approach might be a required by a national authority to renew an instrument rating. I also think most European authorities have the same requirement but, and it's a big but, it will be done with the use of a flight director. But to my knowledge, there are few if any (modern) aircraft where an autopilot can be used without a flight director. And on my own aircraft, there is one piece of automation that can not be discarded, the Flight Path Vector.

For my own company (all jet), if the FD was not available the failure rate would certainly not be 100% but it would be greater than it is at the moment - without practice. So therefore, quite reasonably, we would insist on regular recurrent training for manual flying. But to train to pass a recurrent flight test is a complete waste of everybody's time and money.

As for a company requiring permanent automation - that is complete lunacy and such an operator needs a complete change of personnel at the top and the Flight Ops. inspector sacking.

But moving on -
Eventually pilot's pure flying skills would return to pre-automation days and perhaps that would be a good thing and reduce the number of accidents caused by loss of control in IMC?
Yes and No. To get to the level of manual skills required, virtually all flying would have to be manual, without A/P or F/D. But then we would rapidly increase the number of accidents in IMC until we got to the accident level we suffered before automation. The F/D and A/P were developed to make flying in IMC safer. The problems attributed to "lack of automation awareness" are where the design of the automation in display, performance and documentation terms is lacking. For example, my own aircraft has LNAV and VNAV functions. Using both together is a bit like using Douglas Adams' Improbability Drive. I'm convinced Honeywell have incorporated a random number generator in the VNAV. Basically, the thing is complete rubbish and should never, ever be relied upon, especially when IMC.

What may be lacking is proper monitoring of the automation. Pilots regularly fail to detect incorrect modes and excursions from the desired flight path early enough and therefore allow unintended outcomes to occur. It is in this where we have to spend the training money to improve standards.
Piltdown Man is offline