Diesel aircraft engines and are the light jets pushing out the twins?

In rec.aviation.owning Dean Wilkinson wrote:

True, but the probability of losing all of the engines at the same time
is greater with only two engines as opposed to four.

Matt

Not necessarily...

There has never been a historical case of a twin engine jetliner
losing both engines at once due to unrelated failures. All twin
engine failures have been due to a common cause; fuel starvation being
the prime reason.

Here are some examples of related engine failures:

A four engine 747 had all four engines flame out at the same time when
it flew into the ash cloud of Mt. Redoubt in Alaska, and only managed
to restart three of them after losing over 10,000 feet of altitude.

A four engine Airbus A340 made a dead-stick landing at Lajes in the
Azores after running of fuel due to a combination fuel leak and fuel
system management problem.

A 767 (twin) made an emergency landing in Canada on a drag strip after
losing both engines due to a miscalculation during fueling.

The probability of an ETOPS plane losing both engines in a single
flight due to unrelated failures is extremely remote. That doesn't
mean it can never happen, but it is less likely than winning the
lottery.

Not quite; the probability of all engines failing decreases with the
number of engines if all engines have the same probability of failing.

Whether or not this will actually happen is highly dependent on the
probability of the individual engine failing.

Since airline engines tend to be well maintained, and hence the probablity
of failure low, one could reasonably say the chances of multiple engine
failures (no common cause) is quite remote.

However, a friend of mine that spent lots of time in B-52s relates the
tale of the time they lost 3 engines (no common cause) during flight
and a fourth engine on final getting "that big piece of crap" back on
the ground.

--
Jim Pennino

Remove -spam-sux to reply.

Jim,

Please don't reply in such a way as to make it appear that I said something
that I did not. It was Matt Whiting that made the comment about losing all
engines on a twin being more likely than losing all four. I didn't say
that, and I know that is not true.

Dean
wrote in message
...
In rec.aviation.owning Dean Wilkinson wrote:

True, but the probability of losing all of the engines at the same
time
is greater with only two engines as opposed to four.

Matt

Not necessarily...

There has never been a historical case of a twin engine jetliner
losing both engines at once due to unrelated failures. All twin
engine failures have been due to a common cause; fuel starvation being
the prime reason.

Here are some examples of related engine failures:

A four engine 747 had all four engines flame out at the same time when
it flew into the ash cloud of Mt. Redoubt in Alaska, and only managed
to restart three of them after losing over 10,000 feet of altitude.

A four engine Airbus A340 made a dead-stick landing at Lajes in the
Azores after running of fuel due to a combination fuel leak and fuel
system management problem.

A 767 (twin) made an emergency landing in Canada on a drag strip after
losing both engines due to a miscalculation during fueling.

The probability of an ETOPS plane losing both engines in a single
flight due to unrelated failures is extremely remote. That doesn't
mean it can never happen, but it is less likely than winning the
lottery.

Not quite; the probability of all engines failing decreases with the
number of engines if all engines have the same probability of failing.

Whether or not this will actually happen is highly dependent on the
probability of the individual engine failing.

Since airline engines tend to be well maintained, and hence the probablity
of failure low, one could reasonably say the chances of multiple engine
failures (no common cause) is quite remote.

However, a friend of mine that spent lots of time in B-52s relates the
tale of the time they lost 3 engines (no common cause) during flight
and a fourth engine on final getting "that big piece of crap" back on
the ground.

--
Jim Pennino

Remove -spam-sux to reply.

In rec.aviation.owning Dean Wilkinson wrote:
Jim,

Please don't reply in such a way as to make it appear that I said something
that I did not. It was Matt Whiting that made the comment about losing all
engines on a twin being more likely than losing all four. I didn't say
that, and I know that is not true.

Ummm, I posted a followup to the entire post as received without editing
any previous content.

The depth of the '' characters at the beginning of the lines should show
who said what and my response was to the latest post, i.e. that of Matt
Whiting.

--
Jim Pennino

Remove -spam-sux to reply.

wrote:
Dean Wilkinson wrote:

The probability of an ETOPS plane losing both engines in a single
flight due to unrelated failures is extremely remote. That doesn't
mean it can never happen, but it is less likely than winning the
lottery.

Not quite; the probability of all engines failing decreases with the
number of engines if all engines have the same probability of failing.

That looks fair enough at first sight, but, as you go on to say, it
is "highly dependent on the probability of the individual engine
failing".

The whole point of ETOPS is that the *requirements* for the engines
are rather stricter than those for airliners with three or more
engines, since once you've got a single failure the other fan had
jolly well better keep turning.

With three or more engines, a second failure during diversion is
much less likely to be catastrophic.

Of course, if you take two pairs of ETOPS engines, fit them to a
four-motor aeroplane, and maintain them to ETOPS standards then
the probability of losing all of them from unrelated failures is
exceedingly small - *way* down in the noise of multiple failures
with a *common* cause.

In rec.aviation.owning Robert Briggs wrote:
wrote:
Dean Wilkinson wrote:

The probability of an ETOPS plane losing both engines in a single
flight due to unrelated failures is extremely remote. That doesn't
mean it can never happen, but it is less likely than winning the
lottery.

Not quite; the probability of all engines failing decreases with the
number of engines if all engines have the same probability of failing.

That looks fair enough at first sight, but, as you go on to say, it
is "highly dependent on the probability of the individual engine
failing".

The whole point of ETOPS is that the *requirements* for the engines
are rather stricter than those for airliners with three or more
engines, since once you've got a single failure the other fan had
jolly well better keep turning.

With three or more engines, a second failure during diversion is
much less likely to be catastrophic.

Of course, if you take two pairs of ETOPS engines, fit them to a
four-motor aeroplane, and maintain them to ETOPS standards then
the probability of losing all of them from unrelated failures is
exceedingly small - *way* down in the noise of multiple failures
with a *common* cause.

After posting it occured to me that the above was an incorrect statement.

It should be, the probability of all engines failing decreases with the
number of engines.

While the probability of all engines failing will increase with the
probabilities of individual enginge failure, that number will always
be less than any individual probability.

This is a consequence of the laws of probability and nothing else.

In the real world, we attempt to keep those probability numbers low
so that such an occurance becomes highly unlikely.

The probability of getting 3 jackpots in a row on a Vegas slot machine
is a number greater than zero, but does not form a valid basis for a
retirement plan, for example.

--
Jim Pennino

Remove -spam-sux to reply.

wrote:
Robert Briggs wrote:

The whole point of ETOPS is that the *requirements* for the
engines are rather stricter than those for airliners with
three or more engines, since once you've got a single failure
the other fan had jolly well better keep turning.

While the probability of all engines failing will increase with
the probabilities of individual engine failure, that number will
always be less than any individual probability.

This is a consequence of the laws of probability and nothing else.

There is more than simple "the laws of probability" at work here.

In pre-ETOPS days, three or more engines were required for, say,
transatlantic airliners because the risk of multiple engine
failures was deemed to be too high.

ETOPS certification requires engines which are *demonstrated* to
be more reliable than those on 707s, early 747s, and the like.

It is by no means impossible for four independent engine failures
on an older aeroplane to be more likely than two on an ETOPS kite.

Of course, now that manufacturers are building ETOPS-certified
engines there's not a great deal of point in deliberately making
*less reliable* versions for airliners with three or more engines,
so it is rather likely that *newer* 747s are less prone to losing
all four engines independently than twins are to losing both.

Dean Wilkinson wrote:

Jim,

Please don't reply in such a way as to make it appear that I said something
that I did not. It was Matt Whiting that made the comment about losing all
engines on a twin being more likely than losing all four. I didn't say
that, and I know that is not true.

How do you know this is not true? If I recall correctly, the
probability of independent events occuring simultaneously is equal to
the product of the probabilities of each event occurring. If we rule
out common cause failures such as fuel exhaustion and look at only
random failures, the the probability of all engines failing
simultaneously is the product of the probability of failure of each
engine separately. Assuming that each engine has the same probability
of failure, means that with two engines the probability of both failing
is P^2 whereas with four engines the probability of all failing is P^4.
Since 0=P=1, P^4 will be less than P^2.

As someone else said, the probability of having AN engine fail on any
given flight is higher with more engines, but I believe the probability
of ALL engines failing on a given flight is less with more engines.


Matt

In rec.aviation.owning Robert Briggs wrote:
wrote:
Robert Briggs wrote:

The whole point of ETOPS is that the *requirements* for the
engines are rather stricter than those for airliners with
three or more engines, since once you've got a single failure
the other fan had jolly well better keep turning.

While the probability of all engines failing will increase with
the probabilities of individual engine failure, that number will
always be less than any individual probability.

This is a consequence of the laws of probability and nothing else.

There is more than simple "the laws of probability" at work here.

In pre-ETOPS days, three or more engines were required for, say,
transatlantic airliners because the risk of multiple engine
failures was deemed to be too high.

ETOPS certification requires engines which are *demonstrated* to
be more reliable than those on 707s, early 747s, and the like.

It is by no means impossible for four independent engine failures
on an older aeroplane to be more likely than two on an ETOPS kite.

Of course, now that manufacturers are building ETOPS-certified
engines there's not a great deal of point in deliberately making
*less reliable* versions for airliners with three or more engines,
so it is rather likely that *newer* 747s are less prone to losing
all four engines independently than twins are to losing both.

Mathematics doesn't care about certifications or maintenance
programs. Programs are based on the mathematics.

If the probability of an engine failing is 0.1, the probabilities for
all engines failing a

1 0.1
2 0.01
3 0.001
4 0.0001

If the probability of an engine failing is 0.7, i.e. really ratty engines,
the probabilities for all engines failing a

1 0.7
2 0.49
3 0.34
4 0.24

Things like ETOPS exist because the numbers say getting the probabilities
low is a good thing.


--
Jim Pennino

Remove -spam-sux to reply.

If the probability of an engine failing is 0.1, the probabilities for
all engines failing a

1 0.1
2 0.01
3 0.001
4 0.0001


5: not enough information to say. You do not specify that the probability of
an engine failing is independent of the other engines failing. Fly through a
volcano cloud and I bet you'll find that if one engine fails, the others will
too... quickly. Volcanos don't care about mathematics.


Things like ETOPS exist because the numbers say getting the probabilities
low is a good thing.


True enough. For independent failurs, get the "one engine" number down low
enough and the "two engine" number will also go down low enough. I'm sure they
also address dependent systems in ETOPS land too, and want a dependent failure
probability to be "low enough".

Jose


--
(for Email, make the obvious changes in my address)

Hi NG,

If the probability of an engine failing is 0.1, the probabilities for
all engines failing a

1 0.1
2 0.01
3 0.001
4 0.0001


5: not enough information to say. You do not specify that the
probability of an engine failing is independent of the other engines
failing. Fly through a volcano cloud and I bet you'll find that if
one engine fails, the others will too... quickly. Volcanos don't
care about mathematics.

:-)

somewhere along the discussion it was already stated, that there are
engine failures due to common causes. Of course, the probability of all
engines failing due to fuel starvation or a vulcano doesn't change
whether you have 2, 4, or 8 engines.

For independent failurs, get the "one engine" number
down low enough and the "two engine" number will also go down low
enough.

I guess with ETOPS engines, this "two engines" number is already
significantly lower than the probability number for a common cause
making all engines fail on a plane. So the total number for "all
engines fail" is not significantly greater on ETOPS 2-engine planes
than on 4-engine planes.

regards,
Friedrich

--
for personal email please remove "entfernen" from my adress