(Piston) Engine Failure Rate: 1/3200 hrs, yikes!

[Bell206]

If there is such a point, that performance level would probably be the bread-and-butter of turbine aircraft, rendering all of this purely academic.

But even before turbines when they needed greater engine performance they added rows of cylinders to the air-cooled engines vs build a V24 liquid cooled engines. Regardless, liquid cooled aircraft engines did have their niche markets until technology eclipsed them. However, in the context of putting a liquid cooled engine on your average Piper or Cessna the numbers never did add up nor I doubt they ever will. Besides, even turbine engines are technically a air-cooled engine as well..

 

[Mikey52]

And the liquid cooled fighter weren't expected to last more than 25 missions. Neither were the air cooled ones. Some did, some didn't.

 

[Dan Thomas]

y 2 cents. Aircraft engines have relatively large bore and stroke compared to an automotive engine. All this weight tossing back and forth on top of relatively loose clearances make for a ton of vibration and pretty enormous forces on the bearings, case etc. Yes it's counterbalanced, but it's still a lot of mass accelerating and decelerating. It's also slinging around a giant propeller that not everyone gets dynamically balanced.

Compare this with a solid block modern engine where they run at much higher RPM to produce a given HP, but with a much shorter stroke. Even a Porsche 911 engine (the old air-cooled versions) have 74mm stroke vs 111mm for a Lycoming O360. They are each 180hp, but the Porsche (911SC) produces that at 5,500 rpm vs half that for the Lycoming.

There is a very good reason for big cylinders: Horsepower at low RPM to allow direct-drive while using fewer than a dozen cylinders or so.

Horsepower is defined as torque (in foot-pounds) times RPM divided by 5252. In more complex terms, it's 33,000 foot-pounds per minute. A hypothetical younger, healthy horse could pull hard enough and long enough to lift, via ropes and pulleys, 330 pounds 100 feet up in one minute, or 3300 pounds 10 feet up in a minute. Whatever. On an engine, it's measured using a Prony brake, which has a drum attached to the engine's crankshaft, and a brake assembly over the drum that has a lever that pushes on a scale. If that lever's scale end is one foot from the crank centerline, and the scale reads, say, 100 pounds when the engine is at full throttle and the brake is tightened to keep the engine right at redline RPM, we have 100 foot-pounds of torque being delivered. If that engine's redline is 2500, we get 100 x 2500 = 250,000, divided by 5252, which gives us 47.6 HP.

Now, we can get more HP by running that engine faster, and that's what auto engines do. They have transmissions with gear reductions anyway, so the gearing just reduces the RPM further to what the wheels need. But a heavy propeller has massive inertia that wants to keep that prop moving at a steady rate, while the engine's rotation surges every time a cylinder fires, and lags on every compression stroke, and the propeller gearing experiences forces far beyond what the net HP would indicate. At certain resonant frequencies it will simply explode if not properly engineered.

So a direct-drive affair is SO much simpler and SO much lighter, that the engine is designed for it. That propeller has to have its tips kept pretty much subsonic to avoid an enormous increase in drag, so low RPM is necessary. Yup, you could use a really short propeller and spin it faster, but the loss in efficiency is terrific doing that. Many people have been there and done just that. Some small racing airplanes using Continental O-200s have been hopped up to spin at 4400 RPM with a small prop. The object is a high cruise speed, not good takeoff or climb, which both suffer dangerously in this sort of use. What we have in GA are the result of a whole lot of compromises.

Lots of cubic inches in a few cylinders to make the torque necessary and the low RPM necessary. Contrary to popular belief, aircraft engines are NOT undersquare, with stroke longer than bore diameter. That went away a VERY long time ago.