Re: URGENT: Information please! Aviation Safety Reporting System

From:         rdd@cactus.org (Robert Dorsett)
Organization: Capital Area Central Texas UNIX Society, Austin, Tx
Date:         11 Aug 93 13:31:04 PDT
References:   1 2
Next article
View raw article
  or MIME structure

In article <airliners.1993.561@ohare.Chicago.COM> msb@sq.sq.com (Mark Brader) wrote:

> I don't know about the Aviation Safety Reporting System, but I thought I
> did know about that crash.  Wasn't it caused by the icing interfering with
> the *instrumentation*, so that the crew thought the engines were at a
> higher power level than they really were, and the plane simply stalled?
> The way I remember it, noise regulations prohibited the technique of just
> applying full throttle for takeoff.

The NTSB found two pages of problems, ranging from Air Florida's training
of their flight crews to the "company culture" in-cockpit to personality
conflicts (moody, inexperienced captain, submissive F/O) to improper anti-
ice procedures to bad de-icing to sloppy ATC conduct.

Two notes:

1.  That they took off with less than full power.  This was caused by the 
Pt2 engine inlet probe being frozen over: since EPR is the primary power 
instrument, related by Pt7 (exhaust) divided by Pt2, Pt7 went up while Pt2 
stayed ambient--hence they thought they were at a lot more power than they 
actually were.  This isn't really a "cause," though, since the crew (most 
crews) should have noticed that their N2, N1, and fuel flow were all way too 
low.  But would also not have even had a problem if they had turned their 
engine anti-on before takeoff like they were supposed to: this diverts 
high-pressure hot air to the cowling, thus preventing any ice build-up.

2.  After lift-off (a long roll), they were very close to stall.  What
happened at this point is open to speculation, but the NTSB was concerned
that leading-edge contamination might contribute to a severe "pitch-up"
characteristic in 737's.  If this were the case, then they would have been
low on airspeed, fighting a pitch-up, which could have pushed them right
over the stall limit.  Nance went into much more detail on this in his 
book, citing a number of pilots and airlines (mainly in Europe) which have
reported similar all-but-uncontrollable pitch-up characteristics in icy
weather.  Both parties cite simulator studies, but, well, GIGO.  This stuff
is hard to model.

NTSB cites the low, dirty, low-power situation: the crew had a stickshaker 
warning almost immediately after takeoff; the crew did succeed in lowering 
the nose a bit, to pick up airspeed, which caused the plane to descend.  
Inexplicably, they didn't advance power simultaneously, so the airplane 
collided with the ground.  NTSB concludes that if they had done so, the 
flight path would have been recoverable.

Now, the "inexplicably" is an open question, since IF both pilots were 
applying 400 pounds of down-pressure to overcome that pitch-up moment, they
wouldn't have a limb handy to advance the thrust levers.  This is probably
an argument in favor of alpha-floor protection (have engines automagically 
advance to TOGA after a critical AOA is exceeded).  

Leading edge ice contamination is mentioned in the report and conclusions
(Findings, Probable Cause, and Recommendations), but the pitch-up moment
isn't listed as a factor that we *know* the crew had to contend with.  

Like most crashes, it's a loooong string of screw-ups, starting from the 
moment that everyone got up that morning.  But certainly one to learn from:
some of the more appalling mistakes single-handedly sold cockpit resource 
management (CRM) programs to many airlines, and generally triggered a grass-
roots re-evaluation of operational philosophies and rescue procedures.




---
Robert Dorsett
rdd@cactus.org
...cs.utexas.edu!cactus.org!rdd