Real stats on engine failures?

(Snowbird) wrote
In a SE airplane w/ an engine powered vacuum pump and w/out a backup
electric pump or backup electric gyros, engine failure will become
gyro failure in a minute or so.

Not true unless the engine seizes. If it keeps windmilling, the gyros
will keep spinning.

Michael

(Snowbird) wrote in message . com...
(Captain Wubba) wrote in message . com...
Andrew Rowley wrote in message . ..
Why do you exclude fuel exhaustion, fuel contamination etc? Don't they
happen if you're IFR?

If you're IFR or at night it doesn't really matter WHY it stops.

Because I can control these problems. If I do a proper preflight, the
probability of fuel contamination is very, very low. If I do the
proper fuel calculations and check the fuel levels and carry proper
reserves, I'm not going to run out of gas.

"Cap",

Just curious.

When you fill the tanks after each cross country flight, do you
calculate the fuel you actually had remaining, and compare it
to your calculated fuel reserve?

Not after every one. But after some percentage...probably around 1/4
of the time. I've caught one FBO that didn't give me all the fuel I
asked for this way. They were not trying to cheat me, but it was a
miscomminication with their lineboy. Wasn't a serious problem...I
always carry at least a 2 hour reserve (60 gallon tanks on a Beech
Musketeer that drinks 9 GPH), so I got in with one and a half hour
reserve instead of 2.5. Hard to notice how 10 gallons looks in a tank.

If so, have they ever disagreed?

Yep. In the case above, and when we were having carb problems. Part of
the reason we started suspsecting carb problems.



I really don't want to go there again either -- this topic has been
thrashed out previously and anyone who cares could go Google for it,
but there are a number of factors which make fuel calculations for
a GA aircraft somewhat more uncertain than most pilots would apparently
like to believe.

These uncertainties include:
*aircraft parked on a slope while refueling
*aircraft tachometer not calibrated and no fuel flow meter
*leakage of fuel in flight
*OAT colder than expected or charted and pilot doesn't compensate
*and so forth

Indeed. But I keep two hour reserves on cross country flights in my
Musketeer. No reason not to...it's almost always just me and my wife
(and soon our little one, so no reason not to carry plenty of
gas...when you have 6-hour tanks and a 3-hour bladder, you might as
well put the extra tank space to use


We've had our "ulp" moment where we landed safely and fueled,
and while we had legal reserves we in fact had considerably
less fuel than our proper calculations and preflight checking
led us to expect (for one or more of the above reasons), and
it would have bitten us on the butt if we'd had to exercise
"Plan B".

I don't disagree at all with the philosophy that one should
exercise control to minimize whatever risks one can.

I just feel that it's a mistake to conclude that no pilot
who runs out of fuel in flight did so, or that no pilot
who does so will ever run out of fuel in flight.

I didn't mean to imply that. But that's part of the risk management. I
eat lunch at the airport almost every day, sitting in the GA lot
watching the planes. And you would be stunned by the number of pilots
I see who don't do *any* preflight. We've had two fuel-exhaustion
crashes at my airport over the last decade or so. One unfortunately
killed two innocent people on the ground as well as the pilot. And
both were directly caused by *astounding* stupidity on the part of the
pilots. From reviewing the NTSB database, it appears the majority of
fuel-exhaustion accidents are not the result of a simple
miscalculation. They tend to be a chain of bad decisions (as do most
pilot-error accidents). An example is the one I mentioned above. A
pilot rented a 152, flew it out to Indiana for the day (1.5 hours),
flew a buddy around (45 minutes), then tried to fly back (2 hours)
fully aware that the tanks held 4 hours of fuel. He didn't want to pay
the higher prices at the Indiana airport. He wasn't night current, but
flew back at night. He ran out of gas, then tried to land *into*
traffic on a highway at night. Killed himself, and two women in the
minivan he flew into. Stunning stupidity all the way around.

I'm sure there are many fuel-exhaustion accidents that happen despite
good efforts by the pilot to be diligent. But I think they are much
less common than those that happen because the pilot abrogated his
responsibility to manage all of the risks he could.


Regards,
Sydney


Cheers,

Cap

(Snowbird) wrote
One observation from the recent ASF/FAA vacuum failure study
was that pilots who lost only their AI (electric HSI did not
fail) did not lose control of the airplane, while a significant
number of pilots (same aircraft) lost control when they lost
both. There was no correlation to time in type or total time.

This result suggests to me that it might be a mistake to
extrapolate from "lose AI no problem" to "lose gyros no
problem".

That's somewhat valid. I don't worry about this situation for two
reasons: First, I have dual vacuum pumps, so loss of both gyros
simultaneously is very, very unlikely. The same is true of a wet pump
setup. I've already made my feelings about flying IMC with a single
dry pump and no backups well known, but in case anyone missed it -
it's stooopid. Second, my GPS has an HSI mode.

The reason I say it's somewhat valid is this - my (admittedly somewhat
limited) experience as an instrument instructor is that most people
miss having the DG a lot more than they miss having the AI. While
there is a case to be made that a jet can't be flown without an AI (no
jet crew that lost all attitude indicators in IMC has ever survived)
light piston airplanes most certainly can be.

There are currently no "real stats" which prove or disprove
the contention that this ugliness is entirely due to improper
training.

No, but that's the way to bet. It's certainly how my insurance
company is betting - I'm now required to take a full IPC with engine
cuts every year in make and model, regardless of recency of
experience, if I want to keep my relatively low rates.

Michael

Mike

As they say, "Train for the worst. Hope for the best".

Big John

On Wed, 26 Nov 2003 02:38:51 GMT, "Mike Rapoport"
wrote:

I have also had a gyro fail (in a Turbo Lance that had only one AI) in IMC
flight along with an partial electrical failure (lost the alternator) and
managed to get to my destination after shooting a localizer approach to
pretty much minimiums with a Garmin 12XL that I had to program the approach
waypoints into while flying partial panel AND it was in freezing rain. No
****, this really happened. Every emergency I have ever had was on that one
flight which happened to be my first serious IFR flight after getting the IR
(accross the Sierra From Minden to San Jose in a major blizzard)

That experience doesn't convince me that there are not plenty of senarios
where it wouldn't have had a happy ending.

Mike
MU-2


"Michael" wrote in message
. com...
"Mike Rapoport" wrote
True but I would assume that they thought that they had given the
subject
adequate consideration. It is arogant to believe that everyone else is
a
fool and you are not. My fovorite ezample are those pilots who are
confident that they could handle an IMC gyro failure when the record
shows
that many (most?) cannot.

Yeah, I've heard that song before. Even believed it. Then I had my
AI tumble. At night. In IMC. On the climbout. While being
rerouted. In spite of what everyone told me, it was a complete
non-event. Used the copilot side AI for a while, but quickly decided
it was too much hassle, and flying partial panel was easier. Since I
still had the copilot side AI, I was legal to continue the flight -
and I did. Shot the NDB at my destination, but the weather was crap
and the runway lights were inop, so I couldn't get in. Wound up
shooting the ILS to near mins in the rain at my alternate. No big
deal. Gyro failure is not a big deal if you train properly. I could
even argue that without the backup AI, I would have been safer that
night because I would have had to turn back and land.

On the other hand, an engine failure in a single engine airplane under
the same conditions would have been very, very ugly.

Michael

750 hrs here (OK, 250 of them in gliders). One impending failure due
to broken shaft seal on the crankcase. The engine did not show oil
pressure any more, but I made it back to the airport with the engine
still running at near idle. It happened on the climbout, about 3000
AGL.

Tobias

Rich

Some other data to put in the pot.

The Air Force paid some one (Rand Corporation or some other think
tank) to do a study on accidents vs flying time.

It basically came out that there were two spikes, one around 500 hours
and the other around 1000 hours. The 500 hour accidents were
attributed to cocky over confidence. Not sure right now what the 1000
spike was but it was caused by something we could train around or
change procedures, etc. to reduce as I recall.

Big John



On 25 Nov 2003 07:57:26 -0800, (Rich Stowell)
wrote:

Sorry I can't point you to the "harder" data you're looking for, but
here's perhaps a little perspective on the issue:

According to one NTSB Study, pilots with fewer than either 500 hours
total time, or 100 hours in type, are more likely to encounter an
inadvertent stall/spin than to have a genuine engine failure (i.e.: a
random-event engine failure, not one attributed to such pilot errors
as fuel mismanagement).


In my case, over 6,400 hours with 5,600+ hours of instruction given
(mostly doing spin, emergency maneuver, aerobatic, and tailwheel
training -- the type of flying that might be considered harder on an
engine than more routine types of flying), I've had several
non-critical engine anomalies that were successfully dealt with,
including:

Prop stoppages during spins due to a couple of students hanging on so
tight to the throttle that it choked off the engine -- we call that
"fright idle";

Clogged fuel injectors during take-off that only revealed themselves
at full throttle;

Primer controls that were not truly "in and locked" which has lead to
prop stoppages during idle power landings.


In addition, two legitimate engine failures as follows:

The first, a fuel injector failure as we entered the traffic pattern
(after practicing off field landings, no less!) -- landed without
further incident;

The second, carb ice in a Champ during a flight review choked off the
engine during a touch and go -- touched down on the taxiway abeam the
departure end of the runway, hit a parked Porshe, bent the airplane,
walked away without so much as a scratch.

Rich
http://www.richstowell.com



(Captain Wubba) wrote in message . com...
Indeed. Interesting. But I'd still like to see some hard data. This is
the kind of problem I run into...most of your pilot friends report
that they have had a failure, but the majority of mine report none.
And none of the 2000+ hour CFI types I asked (I asked 4 of them) have
ever experienced an engine failure. My dad was a pilot with well over
12,000 hours and never had one. Another relative had fewer than 500
hours total in his flying carrer and lost one on his first solo XC.

I asked another A&P I ran into at the airport tonight, and he said he
thought it should be at least 40,000 hours per in-flight engine
failure, but really wasn't sure. Since a big part of flying is risk
management, it would be very helpful to *really* know the risks
involved. If the odds of losing an engine are 1 in 50,000 hours, then
night/hard-IFR single-engine flying becomes a great deal more
appealing than if it is 1 in 10,000 hours.

I'll try to go over the NTSB data more thoroughly, I think a
reasonable extrapolation would be that at least 1 in 4 in-flight
engine failures (probably more) would end up in the NTSB database.
But the cursory review I made earlier made me think the numbers were
much less negative than I had considered before. And the opinions of
these A&Ps are very interesting, because while failure might not
require a total overhaul, it will require *something* to be done by a
mechanic...and if these guys are seeing 30-40 engines make it to TBO
for every one needing repair due to an in-flight failure, that might
well support the 40,000 to 50,000 hour hypothesis.

Cheers,

Cap


(Michael) wrote in message om...
(Captain Wubba) wrote
Howdy. I was discussing with a friend of mine my concerns about flying
single-engine planes at night or in hard IFR, due to the possibility
of engine failure. My buddy is a CFI/CFII/ATP as well as an A&P, about
3500 hours, and been around airplanes for a long time, so I tend to
give credence to his experiences. He asked me how often I thought a
piston engine had an in-flight engine failure. I guestimated once
every 10,000 hours or so. He said that was *dramatically*
over-estimating the failure rate. He said that in his experience it is
at least 40,000 to 50,000 hours per in-flight engine failure.

The only vaguely official number that I've ever seen came from a UK
accident report for a US-built twin. The UK investigators queried the
FAA on engine failure rates for the relevant engine, and the only
answer they got was that piston engines have failure rates on the
order of 1 in 1000 to 1 in 10000 hours. This is consistent with my
experience. I've had one non-fuel-related engine failure (partial,
but engine could only produce 20-30% power) in 1600+ hrs. Most people
I know with over 1500 GA hours have had an engine failure.

50,000 hours is not realistic. Excluding a few airline pilots (who
have ALL had engine failures) all my pilot friends together don't have
50,000 hours, and quite a few of them have had engine failures.

I've heard the maintenance shop thing before, but you need to realize
that most engine failures do not result in a major overhaul. Stuck
valves and cracked jugs mean that only a single jug is replaced;
failure of the carb or fuel injection system (my problem) affects only
that component. And oil loss will often seize an engine and make it
not worth overhauling.

There are no real stats on engine failures because engine
manufacturers and the FAA don't want those stats to exist. The FAA
could create those stats simply by requiring pilots to report engine
failures for other than fuel exhaustion/contamination reasons, but
will not.

The truth is, FAA certification requirements have frozen aircraft
piston engines in the past, and now they're less reliable than
automotive engines (not to mention ridiculously expensive).

Michael

(Michael) wrote in message . com...
(Rich Stowell) wrote
Sorry I can't point you to the "harder" data you're looking for, but
here's perhaps a little perspective on the issue:

According to one NTSB Study, pilots with fewer than either 500 hours
total time, or 100 hours in type, are more likely to encounter an
inadvertent stall/spin than to have a genuine engine failure (i.e.: a
random-event engine failure, not one attributed to such pilot errors
as fuel mismanagement).

Really? If that were true, then there would be hard data.


Yes, really -- see "A Study of Light Plane Stall Avoidance and
Suppression." By D.R. Ellis, Report No. FAA-RD-77-25, 1977, p. 6. As
for the "hard data" behind this finding, that's for you to follow up
on since this is your research project


What the NTSB study REALLY says is that these low time pilots are more
likely to encounter an inadvertent stall/spin LEADING TO AN ACCIDENT
than to have a genuine engine failure LEADING TO AN ACCIDENT.


True, but that's stating the obvious since NTSB only gets involved in,
and thus only reports on, those encounters that have led to actual
accidents.


This is
because an engine failure rarely leads to an accident (at least if the
ones known to me are any indication) but an inadvertent stall/spin
usually leads to an accident.


Define "rarely." From an industrial accident prevention standpoint,
the theoretical ratio 1:30:300 is often applied wherein for every 331
hazardous encounters of a similar type, only one will progress as far
as an actual accident (significant damage and/or injury). The rest
fall under "incidents" and "hazards."

In other words, typically 1 out of 331 encounters of a similar type
results in an accident, whether it's precipitated by an engine failure
or an inadvertent stall/spin. In the case of NTSB data, one could
extrapolate to get a feel for the order of magnitude of problems
pilots deal with in a particular category by multiplying the number of
accidents by 331.

While it is true that one accident classification may be more
prevalent than another (e.g.: more stall/spin fatalities than ground
loop fatalities), the ratio of accidents-to-total encounters may very
well be equal. In that case, 1 out of 331 would be the same for engine
failures leading to accidents as for stall/spins leading to accidents,
or any other accident type. I guess one could argue that 1 accident
out of every 331 hazardous encounters is "rare" regardless of the
cause. In that context, one could then argue that compared to the
total number of stall/spin encounters, stall/spin accidents are
equally as "rare" as engine failure accidents.


For that matter, most engine failure fatalities in light singles are
not the result of collision with terrain (which is usually survivable)
but of failure to maintain flying speed (which usually isn't). That's
basically a stall/spin anyway.


Two things: First, approximately 19 percent of stall/spin accidents
are preceded by an engine failure. But the primary accident cause is
still listed as "stall/spin." See "General Aviation Pilot Stall
Awareness Training Study," by William C. Hoffman and Walter M.
Hollister, Report No. FAA-RD-77-26, 1976, p. 6.

Second, the contention that "failure to maintain flying speed" is
"basically a stall/spin anyway" is pure myth. Spins are the result of
two ingredients that must coexist: yaw and stall. And neither yaw nor
stall is a function of airspeed. Up to the point where the wings
decide to bend or break, stalls and spins can and do occur at any
airspeed, and in any attitude.

For example, stall while at 1-g and Vso and give it some yaw = spin;
stall at 1.95 times Vso with +3.8-g's (that's the same as saying "Va
and the design limit in the Normal Category") and give it some yaw =
spin; give any airplane the right amount of g's at a given airspeed
and give it some yaw = spin.

In my experience, and based on the research I've read, I'd postulate
that the majority of stall/spin accidents occur with the airplane
operating somewhere between 1.07 to 1.20 times Vso and 1.15 to 1.41-g.
In other words, with pilots pulling into an uncoordinated, accelerated
stall while turning at bank angles between 30 and 45 degrees.

Rich
http://www.richstowell.com

(Captain Wubba) writes:

So what is it? If the engine-failure rate is one failure for every
50,000 flight hours, I'll feel much less reticent about night/IFR
single-engine flying than if it is one in 10,000 hours. Anybody have
any facts or hard data, or have any idea where I might be able to
track some down?

Don't forget that you're safest with a single-cylinder engine. If
you have a six-cylinder, you're *six* times as likely to have a
failure.

....or at least that's what I've learned from some of the geniuses
who talk about twins vs. singles.

--kyler

"Kyler Laird" wrote in message
...
Don't forget that you're safest with a single-cylinder engine. If
you have a six-cylinder, you're *six* times as likely to have a
failure.

If the only thing that could go wrong with an engine was some sort of
failure of the cylinder, then that would actually be a pretty close
approximation of the truth. And in fact, if you have a six-cylinder engine,
you ARE (about) six times as likely to have a failure *of a cylinder* as you
would with a single-cylinder engine.

In the single vs. twin analysis, you have nearly double the chance of an
engine failure as with a single, all else being equal. If X (a number
between 0 and 1) is the chance of an engine failure for a single engine,
it's not that you have 2 * X chance of an engine failure for two engines.
You actually have 1 - ((1-X) * (1-X)) chance of an engine failure. But when
X is small (as it is in this case), the square of 1-X is pretty close to 1 -
(2 * X).

If all that could fail on an engine was a cylinder, or component related to
a cylinder, then a six-cylinder engine would be 1 - ((1-X) ^ 6) likely to
fail, where X is the chance of failure for a single-cylinder engine. But
just as 1 - ((1-X) ^ 2) is very close to 2 * X for small X, so too 1 -
((1-X) ^ 6) *is* actually very close to 6 * X for small X.

Now, with that essay out of the way, the real reason that six cylinder
engines aren't six times as likely to fail is that a number of failure modes
have nothing to do with the cylinder. They involve one or more other parts,
parts which exist in the same number regardless of the number of cylinders.

Note also that just as having two engines provides a benefit to offset the
very real increased opportunity for failure, having four, six, or more
cylinders provides a benefit to offset the very real increased opportunity
for *cylinder failure*. That is, with a six cylinder engine, if something
that IS specific to a cylinder fails, often the result is simply reduced
power, not a complete power failure.

...or at least that's what I've learned from some of the geniuses
who talk about twins vs. singles.

Sounds like you've got some good geniuses advising you. Stick with them.

Pete

"Mike Rapoport" writes:

Knowing *NOTHING* about turbocharged engines, I was wondering. Would a loss
of the turbocharger still allow the engine to produce the same power as a
non-turbo engine of the same size at the same altitude? From the

It depends on the type of turbo failure. The typical failure is the bearing
in which case the engine will not make much, if any power.

It apparently depends on the system too. For my plane, it's recommended
to shut off the turbos for takeoff below 1000'. I've also flown quite a
few hours with blown turbo bearings, so I think it's safe to say it does
not cause a huge decrease in power.

--kyler