From:         "Larry Stone, United Airlines, 415-634-4725" <STONE_L@Eisner.DECUS.Org>
Organization: DECUServe
Date:         30 Jun 95 03:47:13 
References:   1 2
Next article
View raw article
  or MIME structure

In article <airliners.1995.827@ohare.Chicago.COM>, (David T. Medin) writes:

> I think the main exception when compared to Airbus is that Boeing
> allows pilot override of the computers, either through fighting the
> control inputs or switching flight control modes, and gives simulated
> tactile feedback of conventional control feel. Airbus does neither.
> That is the issue...

As a (light-plane) pilot, I strongly feel the need for both the feedback and
to be the "final authority". There are many situations where the "long-run"
optimal strategy is not the "short-run" optimal strategy and it may be for
reasons that the computer can't know about (a good example is engine failure
training in a single - you're taught early that best glide speed is the
airspeed to fly [it may not be due to wind, weight, and other factors but let's
not worry about that] - that's all well and good but if I'm descneding at best
glide and notice some power lines in my way, you'd better believe I'll
sacrifice some glide range to avoid those wires).

But my major concern with the Airbus approach is a multiple emergency scenario
that is beyond anything that might have been envisioned. I'm sure we all
remember the Aloha "convertible". This was a situation that involved multiple
emergencies - if I remember correctly, besides the obvious explosive
decompression/structural failure, there was lost comm. (their headsets "exited"
the aircraft although they later got a handhold mike out and switched to the
overhead speaker), they lost an engine from FOD, the nose gear never indicated
down and locked, and they had buffetting at full flaps on final that forced a
less than normal flaps landing. Let's face it - the moment that incient
started, the crew became test pilots flying an unproven and never before flown
type. What they knew about flying a 737-200 was only a good starting point for
flying what they had. They had to learn to fly this plane from scratch and
fortunately, as the flap incident showed, did learn and "threw away" the book
when experience showed the book no longer was right. Could a computer,
programmed for a 737-200, not this new type of airplane, have done as well.
What if the new configuration (with much added drag) required the controls to
be positioned "outside the envelope"?

The multiple emeregency aspects of this is interesting too. Due to the added
drag and the failed engine, they could not climb or even maintain level flight.
Therefore, when they didn't get a green on the nose gear, even after doing what
part of the emergency gear extension procedure they had time for, they did the
only thing could do - essentially said "screw it". Imagine a computer trying to
say "you're not in the landing configuration, I won't let you land." (Note: I
am not saying Airbus planes would do this, I'm just trying to point out the
hazards of emergency procedures that conflict with each other or with the
airplane's current capabilities. Somewhere, a judgment call has to be made to
resolve conflicting instruction.)

I once was looking at landing speeds cards for a 747-200. Besides the usual
speeds for various "legal" landing weights, there was a shaded section marked
"for use when an overweight landing is deemed the best course of action." In
other words, even though the "book" says "Thou shall not land overweight",
here's the information in case you have no other choice.

Larry Stone                             | United Airlines
VAX and HP-UX Systems Administrator     | Maintenance Operations Center                | San Francisco, CA  415-634-4725
All opinions are mine, not United's.