460 Ford Forum banner
1 - 7 of 7 Posts

· Registered
137 Posts
Discussion Starter · #1 ·
During one of the Saturday Night chat sessions some time ago, a couple of us were discussing which elements of an engine give the most friction.

I finally found some details on it all. According to Charles Fayette Taylor's "The internal combustion engine in theory and practice" ISBN 0262700263

Bearings and Valve gear - 25%
Pistons and Rings - 75%

For the same piston speed - the longer the stroke:
the lower the bearings and valve gear friction;
the higher the piston and ring friction.

For the same piston speed - the shorter the stroke:
the higher the bearings and valve gear friction;
the lower the piston and ring friction.

In terms of pistons and rings:
1. The average force of friction on the compression and exhaust stroke is nearly the same.
2. The average force of friction on the power stroke is about twice that of the intake stroke.
3. Forces tend to be high just after top and bottom centres, probably because these points where the rings have metallic contact with the cylinder walls.
4. The force is not zero at top and bottom centers. This is probably due to deflection of the engine parts such that piston velocity does not reach zero exactly at the top and bottom center positions of the crank.

The best way to cut friction, follow the example of aircraft engines:
1. Light reciprocating parts, which minimize inertia loads.
2. Large piston to cylinder clearances, which limit the extent of the piston oil film.
3. Short pistons, with the non-thrust surfaces cut away.

So there you have it. :D

In my mind this #2 means, the maximum allowable clearance that doesn't allow oil to sneak past the ring and blow by is at an appreciable level.

Propane Steve

· Premium Member
1,203 Posts
Great info!!

Along those lines, has anyone ever seen information about exactly how much power is being absorbed by friction alone? I'm sure it's an eye opener...

Would it be RPM or power dependent, or combo of both? I'd say combo, as RPM definitely has an influence and as power goes up, it would be forcing pistons against cylinders more and creating more friction. I think one of the BIG lessons learned (at least for me) from the EMC is that managing friction in an engine can pay big dividends. It's one of the few things you can do on an engine that has very few compromises (besides money). It will improve performanc everywhere in the RPM range which doesn't happen often in building engines...

· Registered
137 Posts
Discussion Starter · #3 ·
Friction increases with piston speed

Surprisingly, it is not such a huge amount. The example given in the book is for a 4.00 bore, 3.5 stroke, 5 main bearings at 2.7in with 0.003 clearance, 8 rod bearings at 2.5in with 0.003 clearance - pretty close to a 351 or cheby 350 if you ask me. The test RPM is 4000 and the total hp loss on the rods and mains is only 1.32hp, or 0.74psi in the cylinders. This example is without load on the bearings. Put 3000lbf on the rod bearings and 3500lbf and power loss jumps to a whopping 3.86hp or 2.17psi.

Piston and ring friction for the same motor having pistons with a 3 inch skirt and 0.010 skirt clearance is at 8.6hp.

So total losses without load are 8.6 + 1.32 =9.92HP and with load are 8.6 + 3.86=12.46HP.

Mind you, this is only at 4000 rpm and in this motor the piston speed is about 2670ft/min - I'll calculate later what RPM that is in a stock 460.

Propane Steve

· Registered
792 Posts
Somethings not quite there...

The percentages and directional assumptions look right - - but the calculated numbers don't...its a rather simplified look at a very complicated design problem.

Try turning a shortblock over to 4000 RPM and I think you'll find it takes a bit more than 12 horsepower. Ever look at a spintron?

Ring friction will also up exponentially with both RPM and power output as a result of cylinder pressure - - - especially at the top ring, but to a lesser extent at the second ring as well.

Skirt friction is pretty clearly tied to side load - - but any given side load is distributed across the available skirt area to arrive at a p.s.i. unit loading value. This value must be used to design/develop the appropriate oil film thickness for durability. Moving a given piston too far beyond its designed skirt clearance will result in HIGHER unit load and more friction rather than less as the oil film fails. A thin oil film is actually stronger than a thick one is.

Inertial losses are not truly tied to friction and have to be measured and addressed seperately. There are interrelationships such as piston rockover at TDC and BDC - - but friction is by definition a result of component contact (considering oil as a component).

· Registered
137 Posts
Discussion Starter · #5 ·
back to basics

The numbers presented in the example above relate to pure mechanical friction, not pumping friction. The numbers were calculated with the heads off the engine. Rip the heads off your engine and motor it up to 4000rpm, I think you will find that if it has oil in it and reasonable tolerances, it wont use a lot of HP to get there. I also bet it will take a lot more to start to turn it than to keep it turning.

I will provide the pumping losses as a seperate figure. Have to dig the book out again.

· Registered
137 Posts
Discussion Starter · #7 ·
Pumping losses

For every 100lbs of IMEP (Indicated Mean Effective Pressure) piston friction losses will increase - above and beyond the 'normal' mechanical friction - by approximately 3lbs. The majority of this is caused by 'butting' of the rings against the cylinder walls during compression and by pure compression by itself.

Approximately 80% of the total piston/ring friction, even during pumping, is from the rings only.

For those mathematically inclined:
The estimated sum of mechanical and pumping friction mep (mean effective pressure) can be expressed as follows:

fmep = fmep0 + x(pe-pi) + y(imep-100)

fmep = Friction Mean Effective Pressure
fmep0 = Friction Mean Effective Pressure during motoring
x = exhaust mep - inlet mep as a function of Z (mach index)
pe = pressure of exhaust
pi = pressure of inlet
y = value of motoring test without heads - value of motoring test with heads and a constant air pressure applied to both intake and exhaust
imep = indicated mean effective pressure

I hope I'm helping somebody here! :p
1 - 7 of 7 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.