Re: Dominican Republic 757 crash (RISKS 17.82)

From:         rdd@netcom.com (Robert Dorsett)
Organization: Netcom Online Communications Services (408-241-9760 login: guest)
Date:         07 Mar 96 02:04:42 
References:   1
Followups:    1 2
Next article
View raw article
  or MIME structure

In article <CMM.0.90.1.825716031.risko@chiron.csl.sri.com> Peter G. Neumann writes:
>Investigators have concluded their analysis of the 6 Feb 1996 Boeing 757
>flight that ended in the ocean, killing all 189 people aboard.  The disaster
>was apparently due to a faulty velocity indicator that misled pilots,
>leading them to believe that their speed was adequate when they were flying
>at 7000 feet.

I don't think the investigators have concluded anything--they may have
completed their first sweep of the DFDR and CVR recovered earlier this week,
but the accident investigation's just starting.

More precisely, the comments today suggested that an erroneous speed
indication is an issue under investigation.  There can be a significant
difference between that and saying the indicator is bad.

For the "indicator" to be bad, both the pilots' airspeed indicators would
need to have failed.  And the standby speed indicator.  A highly improbable
event.

The captain and first officer each receive airspeed data from independent
air data computers.  Each air data computer, in turn, receives data from two
independent pitot/static systems, distributed on each side of the aircraft.

In the case of an unreliable data source, warning flags appear on their
instruments.  There are also flags indicating electrical failure and other
mechanical faults.

The standby system receives its feed from the right auxiliary pitot source:
it is "raw" data, not having been massaged by an ADC.

I think that we can conclude that simultaneous instrumentation failure
is unlikely.  For all THREE instruments to fail, clearly, the single source
of failure is the loss of the respective pitot input sources or a loss of the
system itself.

Assuming this is correct, it hails back to the loss of a Northwest Airlines
727-251, near Thiells, New York, on December 1, 1974.  In that one, the
airplane was on a ferry flight, no passengers.  The crew forgot to turn
on probe anti-ice, and the sensors iced over.  The result was an erroneously
high airspeed indication, which resulted in an effort by the crew to pull
back on the stick to slow down.  The airplane was destroyed.

An airline pilot friend recently commented that one of his instructors once
said that you need 5000' of altitude to recover from 10 degrees nose-down
of pitch in a transport, no matter what the reason.  In the Northwest case,
the airplane descended some 28,000 feet in 83 seconds, giving an average
rate of descent of 20,000 feet per minute.

So one obvious question is, could the crew have forgotten to turn on the
pitot anti-ice?  No: Boeing apparently learned from this, and, on the 757,
probe anti-ice is automatically turned on when an engine is is on (this is
a broad statement based upon my knowledge of the Eastern build; details like
this tend to vary according to the airlines purchasing the airplane).

Question #2, could there have been pitot failure without the crew knowing
about it?  Again, unlikely: there are warning annunciators on the overhead
panel.

Question #3, could there have been a maintenance fault?

Question #4, they were taking off into lousy weather.  Maybe they hit
sudden and severe icing conditions within the storm?  Who knows.

The point to all this is that the Dominican Republic's statement raises
far more questions than it really answers.  As with most crashes, there
is not likely to be a single source of failure.





--
Robert Dorsett                         Moderator, sci.aeronautics.simulation
rdd@netcom.com                         aero-simulation@wilbur.pr.erau.edu
                                       ftp://wilbur.pr.erau.edu/pub/av