Been kicked around a fair bit the last few weeks over the "raging fire" versus viscous drag, so came up today with a thought exercise to present here...only posting it to get it out of other threads, and discussion on it's own.
I've posted the two pics here that show heat flows within a crankshaft at a couple of different RPM in a fairly small 4 cylinder engine.
Now to the boundaries of the thought exercise.
Piston undercrown temperature - 250C... I chose this temperature, and also chose how it's applied...It's assumed that all of the oil that hits the piston undercrown is heated to that temperature.
Logic for it is that engines last a long time, and if the oil WAS in fact getting to those temperatures, it would be leaving deposits (and the ones I've pulled apart show some for sure)...but if the deposits became serious, then they would get thicker, and more insulating, and you would end up with a runaway thermal failure well before the engine got "old".
250 is NOACK temperature, and 100C above where RAT considers that thermal breakdown is occurring...could raise it to TEOST, but will keep it there for the time being.
Engine that these diagrammes were from didn't have squirters, the oil on the piston underside was all sprayed up there by the action of the big-end bearing.
Jaguar ran some tests with glass sides sumps, and found that the big ends produced a "sheet" of oil on either side as the oil was flung out of either side.
I've assumed (for the thought exercise) that the oil flow from the big end ends up in two different directions, either aimed 100% at the piston underside, or 100% away, and these are 50:50...obviously not that case, as I could have assumed 4 90 degree sectors, with three of them not aimed in general at the piston...
My way gives the most oil even remotely possible to be aimed squarely at the piston undercrown, and heated to 250C.
Will take density at 15C to be 890Kg/m^3, and themal capacity at 2.2KJ/KgK.
4,000 RPM case, there's 2.38cc/sec issuing from the big end, at 145C, so taking into account my envelope, 1.19cc/sec...density at 145C is therefore 800. Thats 0.95 grammes per second.
105C delta T means that the thermal capacity of that oil stream is 220W.
Remember, that's assuming that fully half of the big end oil ends up directly cooling the piston underside...the real answer is obviously a little bit (or a lot bit) smaller...and the total of 1 main and 1 big end is already 255W in the example.
And we haven't even considered the cam bearings, they'll be a lower contributor as they are running half speed.
We haven't considered piston and ring friction either, which are oft measured at 3-4 times bearing friction, but I've taken the assumption that ALL of that is conducted to the cooling system through the liner.
Try the 6.000RPM example now.
Half the oil flow is 1.86cc/sec, and at 187C, has a density of 768KG/m^3...that's 1.43 grammes per second hitting the piston underside, and can increase it's temperature by 63C to get to the 250C (arbitrary) limit.
That's 200W versus 570W+ for a main and a big end in the test engine.
Obviously my limit on oil temperature at 250C is arbitrary, and could be higher...
* bit clearly, piston underside deposits become piston crowm overheaters and piston failures would ensue...bump the temps up to 300C, and you get 323 and 355W respectively.
* but if you want to do that, you need to reduce the oil flow from my arbitrary 50% also.
* and consider also that I assumed that the full volume of that oil was heated to the undercrown temperature
I discounted piston and ring friction, even 'though these are multiples of bearing friction Ricardo Paper, see page 12.
These largely go into the coolant, and are certainly there during the warmup phase, generating heat through oil drag that heats the metal and coolant.
In this testing of an aircooled engine shielding the piston underside from the oil and crankcase vapors only added 15F (8C) to the piston temperature, while going from an SAE 30 to SAE60 oil increased piston temperature by 20F (11C)
I've posted the two pics here that show heat flows within a crankshaft at a couple of different RPM in a fairly small 4 cylinder engine.
Now to the boundaries of the thought exercise.
Piston undercrown temperature - 250C... I chose this temperature, and also chose how it's applied...It's assumed that all of the oil that hits the piston undercrown is heated to that temperature.
Logic for it is that engines last a long time, and if the oil WAS in fact getting to those temperatures, it would be leaving deposits (and the ones I've pulled apart show some for sure)...but if the deposits became serious, then they would get thicker, and more insulating, and you would end up with a runaway thermal failure well before the engine got "old".
250 is NOACK temperature, and 100C above where RAT considers that thermal breakdown is occurring...could raise it to TEOST, but will keep it there for the time being.
Engine that these diagrammes were from didn't have squirters, the oil on the piston underside was all sprayed up there by the action of the big-end bearing.
Jaguar ran some tests with glass sides sumps, and found that the big ends produced a "sheet" of oil on either side as the oil was flung out of either side.
I've assumed (for the thought exercise) that the oil flow from the big end ends up in two different directions, either aimed 100% at the piston underside, or 100% away, and these are 50:50...obviously not that case, as I could have assumed 4 90 degree sectors, with three of them not aimed in general at the piston...
My way gives the most oil even remotely possible to be aimed squarely at the piston undercrown, and heated to 250C.
Will take density at 15C to be 890Kg/m^3, and themal capacity at 2.2KJ/KgK.
4,000 RPM case, there's 2.38cc/sec issuing from the big end, at 145C, so taking into account my envelope, 1.19cc/sec...density at 145C is therefore 800. Thats 0.95 grammes per second.
105C delta T means that the thermal capacity of that oil stream is 220W.
Remember, that's assuming that fully half of the big end oil ends up directly cooling the piston underside...the real answer is obviously a little bit (or a lot bit) smaller...and the total of 1 main and 1 big end is already 255W in the example.
And we haven't even considered the cam bearings, they'll be a lower contributor as they are running half speed.
We haven't considered piston and ring friction either, which are oft measured at 3-4 times bearing friction, but I've taken the assumption that ALL of that is conducted to the cooling system through the liner.
Try the 6.000RPM example now.
Half the oil flow is 1.86cc/sec, and at 187C, has a density of 768KG/m^3...that's 1.43 grammes per second hitting the piston underside, and can increase it's temperature by 63C to get to the 250C (arbitrary) limit.
That's 200W versus 570W+ for a main and a big end in the test engine.
Obviously my limit on oil temperature at 250C is arbitrary, and could be higher...
* bit clearly, piston underside deposits become piston crowm overheaters and piston failures would ensue...bump the temps up to 300C, and you get 323 and 355W respectively.
* but if you want to do that, you need to reduce the oil flow from my arbitrary 50% also.
* and consider also that I assumed that the full volume of that oil was heated to the undercrown temperature
I discounted piston and ring friction, even 'though these are multiples of bearing friction Ricardo Paper, see page 12.
These largely go into the coolant, and are certainly there during the warmup phase, generating heat through oil drag that heats the metal and coolant.
In this testing of an aircooled engine shielding the piston underside from the oil and crankcase vapors only added 15F (8C) to the piston temperature, while going from an SAE 30 to SAE60 oil increased piston temperature by 20F (11C)