Re: A brief commentary

From:         glenn@ims.uni-stuttgart.de (Glenn Carroll)
Organization: IMS, University of Stuttgart
Date:         16 Jul 96 13:59:48 
References:   1 2 3 4 5 6
Followups:    1 2
Next article
View raw article
  or MIME structure

In article <airliners.1996.1262@ohare.Chicago.COM>, "Rudi Vavra"
<flying@ozemail.com.au> writes:
|
|> The pilot slowed down to V2 BECAUSE he was trained to do so in that
|> scenario. With the loss of one engine (remember, the crew did not
|> know that they lost the whole engine including the pylon, the
|> assumption was that the engine lost power), the SOP was to slow down
|> to V2. This airspeed probably corresponds closely with the "best angle
|> of climb" airspeed, and this procedure was implemented so that all
|> ground obstacles are cleared safely before the crew "clean up" the
|> airplane and decide on further action.

Alright, I now know well enough why the captain decelerated to V2, but none
of the answers have done much to enlighten me about the substance of the
question.  V2 is the speed one is supposed to fly with one engine out.  In
this case, with the outboard slats inappropriately retracted, V2 was not
sufficient to prevent a stall.  The aircraft could have been flown and landed
at a higher speed, so:  why is V2 so slow?  Put another way:  what are the
factors that go into choosing V2 for a particular aircraft?  I presume that
higher speed implies more power from the remaining outboard engine which
implies more of a yaw problem.  Following this line of speculation, V2 would
not be the same for the loss of the center engine in a tri-jet such as the
DC-10, since there wouldn't be a yaw problem.  Similar considerations would
apply to the inboard/outboard engines on 747s and A340s.  Is this true?

One of the replies implied that it's decided by each airline individually--AA
revised their manuals following the crash.  That doesn't seem right;  surely V2
isa property of the aircraft, and should not be left to each airline to
determine.


|> Aaah, yes, but again, you are analysing this with the benefit of
|> hindsight. Spatial disorientation is a very powerful and deceiving
|> sensation. The only way to combat it is to trust your instruments.
|> When your instruments don't agree, it's very hard to decide which
|> instruments to trust.

I don't know what spatial disorientation you're talking about, as no one
else has mentioned it on this thread.  Care to clarify what you mean here?

Of course the nut of the problem for Birgenair (pending the NTSB report ;-)
is deciding which intstruments are right, and to what extent.  My point
was that in this case the decision criteria seemed straightforward:  one
instrument, the ASI, said "too fast", and another, the stick-shaker, said,
"too slow".  This impasse can be resolved through an instrument all pilots
carry with them:  "their ass strapped to the hardware", as someone else
put it.  At a given throttle setting and nose attitude, a B757 is not
about to suddenly accelerate past its airframe speed limit.  That eliminates
"too fast" as a threat.  "Too slow" can be taken care of by a moderate
throttle setting/pitch, and then one can start worrying about which
instrument(s) are wrong, and what one should do next.

|> We are all human, and one of the human traits is to stick with your
|> convictions. Once the crew make a decision and elect to trust one set
|> of instruments, they will go with that decision even if all
|> subsequent events try to tell them that their decision was wrong. It
|> takes a clear analytical mind to evaluate these signs and break the
|> chain.

Pilots are human, sure enough, and they perform their task better than any known
alternative.  Nonetheless, they are also supposed to be selected for their
ability to make the right decision in such cases, and then trained to do so.
Fatigue is a factor one cannot weed out through selection or training.  Panic
and/or poor decision making skills, however, are just the sort of thing that can
largely be eliminated.


8<==============================================
Glenn Carroll					glenn@ims.uni-stuttgart.de
Institut fuer Maschinelle Sprachverarbeitung
Azenbergstr. 12
70174 Stuttgart					(49)711-121-1387  office
Germany						(49)711-121-1366  fax