On Tue, 29 Jun 2004 07:32:04 -0000, Dylan Smith
wrote:

Perhaps auto engines aren't as feeble as people like to make out?

Maybe it's time for me to post the article written by a Detroit auto
engineer about exactly what kind of torture auto engines go through
during their development and testing. He was in charge of GM's
"Premium Engine Program" and wrote this article for "Contact!"
magazine about 8 years ago.

Begin article:

"PERFORMANCE

The engine in production form for 1999 develops 215 HP at 5600 RPM and
230 foot pounds of torque at 4400 rpm. As a routine part of an engine
development program we tested the engine at full power, maximum RPM.
We ran it at 6000 RPM, pulling 215 HP at wide open throttle, for 265
hours. That's a continuous 265 hours of wide open throttle, far worse
than autobahn driving, because even on the German Autobahn, you
wouldn't be at 6000 RPM. THAT IS A STANDARD DURABILITY TEST.
(emphasis mine) We run many engines through this test as a matter of
course.

Specific development focus is on the crank, pistons, rods, block
structure, timing drive wear; we get a lot of full load cycles in a
hurry. It isn't necessarily designed to replicate customer driving
but to get development answers. Wear and fatigue are accelerated.
The test is particularly applicable in proving out dampers and their
effectiveness. If the damper is not properly tuned to the engine the
crankshaft will inevitably break in that time period. (note, this is
evidence you should not discard the stock damper when using the auto
engine for aircraft power)

A number of other engine tests are utilized. We use a variety of
specific tests to accelerate engine wear and to look at fatigue
failures. The cyclic endurance test is now called PTED (power train
endurance). It closely approximates cyclic durability. The engine is
cycled from its torque peak to its horsepower peak, at wide open
throttle, then down to idle, then accelerates up to shift points, then
back down to the torque peak and then horsepower peak. This test is
run for 400 hours. Once again, it's a wide open throttle test for 400
hours. The RPM for this engine, ranged between 4400 and 6000 RPM,
back and forth in about a 5 minute cycle. The dyno computer will
occasionally bring the engine down to idle, up to 6500 RPM shift
points, and then back to the 4400 - 6000 RPM 5 minute cycle.

Thermal cycle tests are run to define engine capability under cold
weather condition. We run the engine at full throttle at 4000 RPM,
bring it down to idle, stop it, switch the coolant valves to drain the
hot coolant, pump the chilled coolant from the chiller until the metal
temperature stabilizes at 0 degrees F. Frost forms on the outside of
the block, as the cold coolant rushes into the engine. When it
stabilizes at 0 F, we motor the engine, start it, come to full
throttle at 4400 RPM, the valves switch and the coolant temperature
starts to climb. It climbs back up to 260 degrees F. It takes 10 -11
minutes to complete one cycle. The engine must pass 600 cycles
without any sign of failure. We typically run 1200 cycles and a probe
test will run 1600 cycles. That's a (sic) excellent gasket killer
test. Head gaskets are the first to fail because of the rapid
expansion and contraction.

A powertrain endurance test simulates in-vehicle operation. The
Ypsilanti plant uses it for testing transmission. We, of course, use
it to look at engine performance. The equipment consists of an
engine/transmission combination, which sits on a dyno with large steel
inertia wheels. The inertia wheels are being driven by the
transmission output shaft, just like in a car. They cycle is brutal;
the engine is at idle in gear. The engine accelerates wide open to
6200 RPM, upshift occurs, 6200 RPM is reached, upshift occurs to 3rd,
6200 RPM is reached, upshift occurs to 4th, the wheels turn up to 135
MPH depending on the application. The second half of the cycle calls
for a closed throttle down to 70 MPH, then wide open throttle with a
downshift to 2nd, the engine goes back up to top speed, coasts down so
that the transmission selects down to a lower range. The engine is in
an overrun condition all the way down to idle; i.e., the engine is
being used for braking. That's one cycle. One transmission life
cycle is typically 12K - 13K cycles of the above test. We will run an
engine through 4 or 5 transmissions. This is a very harsh schedule
for the engine, particularly because of the overrun braking.
Cylinders and rings suffer the most on this test.

We run some idle tests to verify low speed operation. The engine is
run at idle for about 2000 hours to make sure of adequate oil flow at
idle.

We use all those engine tests in addition to fleet tests and extensive
vehicle road testing. The customer can be assured that the PV6 engine
is a thoroughly tested advanced design that matches or exceeds
competing offerings."

End of article

This is merely typical of auto engine development, all the
manufacturers run some variation of this kind of torture. BMW and
Mercedes probably beat up their engines even worse.

As Peter Duniho pointed out, auto engines have been in use in the
homebuilt community for many years now. One guy managed to run his
Ford 3.8L engine to 2,000 hours before tearing it down for a rebuild.
He found no significant wear in any of the engine components, but he
replaced the Gates toothed belt for the PSRU anyway. The engine tear
down and rebuild, including the belt, cost him around $1,000 if I'm
remembering correctly.

Corky Scott