Re: Engines CAN jettison (Was Rear engined aircraft. (727 DC9 MD80) )

From:         kls@ohare.Chicago.COM (Karl Swartz)
Organization: Chicago Software Works, Menlo Park, California
Date:         15 Dec 94 05:17:37 
References:   1 2
Followups:    1
Next article
View raw article
  or MIME structure

>The pilots can't purposely select to jettison an engine but they *CAN*
>be jettisoned:

To me, "jettison" connotes a more decisive act than you suggest.  At
the risk of starting a debate over razor-fine semantic differences, I
checked my handy (American Heritage) dictionary, which says, in part:

    1.  To cast off or overboard.
    2.  To discard (unwanted or burdensome articles).

While "cast off" suggests an action and not a mindless response, I
can't say that either of those definitions really supports my position.
Nevertheless, I think you're stretching the meaning of "jettison" quite
a bit.  Wings can be "jettisoned" too, if you bend 'em hard enough.

>(1) They are designed in a ditching situation to shear off and flip over
>    the top of the wing rather than dig into the water & pull the wing off.

This is also for landings on runways without landing gear, but I think
the concern is more to keep the aircraft from flipping than it is for
tearing off the wings.

>(2) In case of a nasty bad engine fire that is out of control the engine
>    can jettison itself (which happened once on a B747 flight). I don't
>    know how this is triggered (or suppressed) but I know of at least one
>    false jettison (reported in the press as an engine "falling off").

On the 747, and many other aircraft, there are bolts or pins in the
pylon (called fuse pins on the 747, at least) which are the weakest
part of the system.  Thus, if the stresses exceed the design limits
of the pylon, due to dragging on water (or ground) as in your first
example, or due to a severe imbalance, these parts break, allowing
the engine to depart the airframe before some more important, like
a wing or fuselage (on a 727, for example), breaks.  Unless a fire
somehow causes such high stress, the engine is staying firmly on the
airframe -- there are no explosive bolts or whatnot, and I seriously
doubt an airframe could be certified with such devices.

>(3) The DC-10 crash in Chicago in the late '70s (that resulted in the
>    grounding of all DC-10's) was because an overstressed connection
>    caused a premature jettison of an engine that flipped over the wing
>    because it was under thrust & was designed to come off that way.

The engine separation in the AA 191 crash was not the result of an
overstress condition, at least not beyond normal operating limits.
However, due to improper maintenance procedures, a flange in the
pylon had cracked, and takeoff thrust *was* sufficient to overstress
the already-damaged pylon.  To describe this as jettisoning the engine
surely overstresses the meaning of the word.

>    The engine jettison caused hydraulic disruptions that deployed a
>    spoiler which the pilot didn't recognize & retract. Coupled with
>    not reducing climb rate, both contributed to the crash.

No.  See earlier discussions of this crash in this newsgroup.  The
failure caused an uncommanded *retraction* of leading edge slats on
the left wing, leading to an asymmetric lift condition which was not
reported to the pilots due to inadequate redundancy.  The pilot did
indeed reduce climb rate, along with airspeed, in accordance with
the airline's FAA-approved emergency procedures.  However, the slat
retraction on the left wing increased the stall speed of that wing
beyond the aircraft's airspeed, leading to a stall of the inboard
left wing followed by an unrecoverable roll to the left.

The pilots did exactly what they were supposed to do, given the
information they had.  In simulators afterwards, every set of pilots
crashed the aircraft.  When given a functioning asymmetric slat
warning system, pilots successfully flew the simulators to a safe
landing every time.

Karl Swartz	|INet
1-415/854-3409	|UUCP	uunet!decwrl!ditka!kls
		|Snail	2144 Sand Hill Rd., Menlo Park CA 94025, USA
 Send sci.aeronautics.airliners submissions to