Re: hydraulic problems with DC-10's??

From:         Robert Dorsett <rdd@rascal.ics.utexas.edu>
Date:         10 Dec 92 00:52:08 PST
References:   1 2 3 4
Next article
View raw article
  or MIME structure


Before starting: the left pylon assembly weight, btw, was 13,477 lbs, from 
the accident report.  A whole bunch of figures had been floating around...


In <airliners.1992.106@ohare.Chicago.COM> Michael Weiss writes:
 
>In any case, my point is that there would have been a severe weight unbalance
>between the wings, and I have doubts that it could have been countered by the
>ailerons.  The whole reason that there was a negative roll moment was that the
>left wing STALLED, not that it lost lift directly from the retracting slats.
>I'm still not convinced that even WITH the slats extended it could have been
>prevented.

Allow me to throw in my $0.02 worth.  

1.  A slat increases the maximum effective lift coefficient for a wing section. 
How this is done is irrelevant to this discussion: the result is that the 
lift coefficient goes up.  Slats permit the wing to produce a greater lift at 
slower airspeeds, i.e., they drive the stall speed down.  When used with 
trailing edge flaps, they offer even better lift characteristics, and improve 
handling characteristics.

2.  If we take away the slats, then the maximum lift coefficient goes down.  
By definition.  This means the stall speed goes up for the resulting 
configuration.  By definition.

3.  The accepted procedure was to climb at V2 until 800' AGL, then to lower 
the nose and accelerate. 

For a normal, undamaged aircraft, at 379,000 lbs, V2 was 153 knots.  

In the damaged aircraft, the minimum controllable airspeed, with a 4 degree 
left bank, into the missing engine, was 159 knots.

Therefore, if the crew were to fly a standard engine-out profile, at 153
knots, they would have been beneath the minimum controllable airspeed for
the damaged aircraft (159 knots).

During the investigation, the NTSB asked 13 qualified pilots to fly various
takeoff profiles.  70 takeoff simulations were flown.  All crashed the
airplane when flying the crash profile.  Several pilots, when left to their
own devices, and with extensive knowledge of the events, managed to control 
the airplane, nonetheless, by recognizing the initial roll and applying full
opposite aileron and significant rudder, and lowering the nose to gain air-
speed.  All pilots who received appropriate feedback, via a functioning 
stickshaker, and who increased their airspeed to stay above the stickshaker 
value--168 knots--saved the airplane.



I really fail to see what the problem is, here.  The engine fell off after
V1.  This didn't affect the aerodynamic characteristics of the wing itself: 
it became a control problem.  It also killed the electrical system driving 
the captain's stick-shaker, and killed a hydraulic system.  The latter 
caused the slats to retract within 20 seconds of failure.

The slat retraction DID affect the wing: it then became both a control and 
aerodynamic problem.  Exercising established control practices in an 
unknown aerodynamic regime crashed the airplane (I'd love to know whether
this went into Airbus's "pilot error" database :-)).  Had the slats remained
down, the airplane would have survived the engine failure, even with the 
failure of the stall warning system.  Other airframe manufacturers have 
manual locking mechanisms for their slat jackscrews.  McDonnell Douglas 
relied on hydraulic pressure to hold it all together.

Incidentally, this problem wasn't corrected: the SUX DC-10 also experienced 
extension of its slats after it lost all its hydraulics.

I'd suggest you obtain a copy of the accident report (NTSB-AAR-79-17), and
look it over, closely.  It has more than enough data for back-of-the-
envelope calculations.  Nothing in it suggests that weight or moments 
following engine separation played a significant role.


Lastly, I'd note that there was SIGNIFICANT public and industry concern 
about the DC-10's safety after this crash: the FAA's extraordinary grounding
of the airplane, inappropriate though it may have been, is testimony to 
that.  All of the manufacturers had something to contribute, and a great 
deal of manpower was invested in finding the cause.  There was REAL concern
that the airplane wasn't airworthy, even by FAA's standards.  Every analysis 
or comment I've ever seen on this crash has concentrated on the slat retrac-
tion being the proximal cause for the crash.  I've never seen the weight 
issue raised.  If you have "hard" evidence that it HAS been, some references 
would be useful, since it's not a well-publicised theory.  If you're
basing your comments on classroom experience, as I believe you indicated,
it might be a worthwhile learning exercise to raise it in class, or print
this discussion and privately  discuss it with your professor: but be sure 
to let us know the outcome.

 





--
Robert Dorsett
Internet: rdd@rascal.ics.utexas.edu
UUCP: ...cs.utexas.edu!rascal.ics.utexas.edu!rdd