Re: Are Two Engine 757 & 767 Jets Dangerous?

Date:         27 Dec 96 13:32:18 
From:         bauwens@acs.ucalgary.ca (Luc Bauwens)
Organization: The University of Calgary
References:   1 2 3
Next article
View raw article
  or MIME structure

In article <airliners.1996.2811@ohare.Chicago.COM>,
Karl Swartz <kls@ohare.Chicago.COM> wrote:
>
>The big question is what happens when both engines on one wing fail.
>The certification requirements require that you be able to take off
>and climb if *one* engine fails after reaching V1, but what about
>a double failure?  I assume you have to handle this in cruise since
>the pre-ETOPS ICAO rules for long overwater flights mandated the
>ability to limp to safety after the loss of two engines, but losing
>two at cruise, with relatively low thrust levels is a very different
>matter from losing two during takeoff or climb, at high thrust levels.
>
>The El Al crash in Amsterdam, as well as the 1991 China Air crash near
>Taipei (both 747-200Fs) suggest that losing two during climb may not
>be recoverable -- I'm not sure if any of the wing damage in El Al was
>significant with regard to control of the plane, and I know even less
>of the details of the China Air crash.

Finally, this discussion going in the right direction :-)...?

Seems to me that it would be useful to divide this discussion in two parts:
take off and enroute.  And then try to compound the odds.  (Also, since
all these probabilities are pretty small, we can just add them.)

Take off:

If we assume that indeed, only one engine can be lost, then the fatal
scenarios include

- independent loss of two engines.  The probability of which, all other
things equal, would be double with four engines compared with two.  But
for two-engine ETOPS, the odds of loosing an engine are less, so, the risk
with fours engines is more than double.  Also, during take off with four
engines, the engines have two be used at a fraction of their maximum thrust
that is 50 % more than in a two-engined one.  Surely this has a dramatic
effect on the probability of failure.  (This is just a way of saying that
it's much safer at take off to have two big engines with lots of margin
than four wimpy ones.  For god's sake, look at these 340 :-)..)  Anyway,
the probability of loosing one engine in an ETOPS twin is probably about
1/10 of a four-engine.  Altogether, the independent (not common mode)
failure of two engines during take off might well be around twenty times
less, or better, in the ETOPS twin?

- Then there is the El Al scenario under discussion: uncontained failure
of one engine affecting a second one.  Clearly, the odds of a second engine
failing as a result of a first one having failed are significantly bigger
than the odds of loosing one engine, especially during take off?   Possibly
by a factor of 20 or even more, much more?   (And this is made worse by
the larger probability of uncontained failure of one engine in the
four-engine plane, discussed in the paragraph above.)

- Then there is the common mode.  Since the four-engined plane is more
complex, the odds of a failure that affects all engines will be larger
than in a twin? Not twice, but still, anywhere between 20 and 50% perhaps?

- There is the risk of the crew shutting off the wrong engine.  But if
only one engine can be lost, this affects equally both sides in the
comparison.

- Beside engine loss scenarios, the extra margin during take off increases
safety in any twin because of the increased flexibility in dealing with
other emergencies.

So, the one thing left in favor of the four engines is that there will be
scenarios in which the airplane will be light enough to take off on two
engines.  Otherwise, the ETOPS twin will be distinctly safer during take
off/climb.

But then there is cruise.  During which the four-engined plane will
probably be able to survive even the loss of three engines (at least for
a while :-)).  But the probability of loosing two engines in an ETOPS twin
due to two independent failures still will be quite small.

So, it seems to me that the much larger probability of a catastrophic loss
of two engines during take off in a four engine plane should overwhelm
the pretty small odds of loosing the two in cruise in the ETOPS twin?
(Especially keeping in mind that by far, most engine losses occur during
take off, at high thrust and high turbine temperatures, and not in cruise.)

In any event, I am not aware of any occurence of independent failure of
two engines on an ETOPS twin ever?   While, besides the El Al and the
China Air under discussion, I vaguely recall of at least a couple of other
cases where the loss of one engine affected the other one on the same side
in four-engine planes (Bae146)?

Of course, all of this ought to be quantified based upon any figures that
might be available...   Guesses being just guesses.

Bottom line: I would personally rather cross the Atlantic on a 767, 777,
330 or A310 than on a 340 with these (very obviously) wimpy engines, or
on a 747  (especially aging ones operated by near bankrupt airlines and
whose fuel tanks blow up).

And I would bet that airlines who get 340s when the 330 would do the job
are purely motivated by marketing considerations.  It's pretty clear that
the public still has the perception that four engines are more reassuring
than two.  Which, I submit, is probably quite wrong.

(Why do we have two lungs, two kidneys etc.  and not four :-)...?)

Luc

--
Luc Bauwens                        bauwens@acs.ucalgary.ca
The University of Calgary,         Department of Mechanical Engineering
Phone: (403) 220 5792	           Fax: (403) 282 8406
http://www.ucalgary.ca/~bauwens/