Re: Peruvian 757 crash -- possible cause reported

Date:         23 Nov 96 03:36:24 
From:         Stefano Pagiola <>
Organization: World Bank
References:   1 2
Followups:    1
Next article
View raw article
  or MIME structure

Adam Dobrzycki wrote:
> "David G. Davidson" <> writes:
> => As I recall, the incident involving the
> => Piedmont 737 that went off the end of the runway and ended up against
> => railroad tracks was very similar to the LH A320 in Warsaw.
> Pardon my ignorance, but I have never heard of this incident. When/where was
> that? If this happened before the A320 crash in Warsaw, it would be amazing
> that the logic preventing this type of accidents wasn't built into Airbus'
> software...

The Piedmont incident involved 737-222 N752N (msn19073) on 25 Oct 1896,
at Charlotte.  The problem was similar to that later experienced by the
LH A320 at WAW: the safeguards built in to prevent inflight spoiler
activation weren't met (on the 737, they involved either the main wheels
spinning through at least a quarter revolution or some amount of pressure
on the oleos) because the plane was hydroplaning on a wet runway.  In this
case, it wasn't a computer, it was just a simple link between microswitches
on the gear and the spoiler activation mechanism.  But the principle was
the same.

Now why, you ask, were measures to prevent this kind of accident not built
into the A320 software?  The problem is that the cause of these accidents
is itself a safeguard against another kind of accident.  You don't want
spoilers (or thrust reversers) activating in flight.  Several DC-8s were
lost, for example, because of inadvertent spoiler activation in flight.
So you build in measures to ensure this won't happen until you're on the
runway.  How do you know?  As these accidents illustrate, there's no
foolproof way.  You try to find a middle ground between alternatives.
And sometimes you get it wrong.  Or you just get unlucky.  I'm not sure
what specific changes were made in the 737 and A320 (and perhaps other
models) in response to these accidents, but I suspect that unless someone
came up with something really clever, the result is probably a system
that's a little less vulnerable to runway over-runs and a little more
vulnerable to inadvertent in-flight spoiler deployment.  Hopefully the
combination of circumstances which will expose that higher vulnerability
won't happen... too often.

A final note: As I recall, there were no injuries in the Piedmont 737.
At WAW, on the other hand, a ditch and a wall in the runway over-run area
resulted in the aircraft breaking in two and catching fire, with 2
fatalities.  That again is the luck of the draw.

All opinions expressed are my own