We're all familiar with pumping losses as they apply to an engine's intake and exhaust systems. For instance, we know that we get lower fuel flow if we run WOT and lower RPM vs part throttle and higher RPM (for a given power output). However, I wonder how much thought has been given to air pumping losses within the crankcase.
I did a top overhaul on my IO-360 a couple of years ago. Since I had virtually no prior experience with Lycoming engine internals, I had expected to see into the sump once I removed the cylinders (like it would be on a car or motorcycle engine). Instead, I was struck by how compact and relatively closed the crankcase volume for each cylinder pair appeared to be. There are small air/oil passages beneath each main bearing and an additional small slot into the sump beneath the crank volume for cylinders 3 and 4. I don't have any measurements but my memory (and some pictures I've found on the web) indicate the passages may be small enough to present a significant impediment to air flow.
Over each revolution of an IO-360, the crankcase volume for each pair of pistons varies by 180 ci (cubic inches) between TDC and BDC. In principle, that means that the air passages must vent 180 ci out then 180 ci in on each revolution of the crank (per pair). By comparison, that's four times the air flow ingested through the intake ports. How much power that requires depends on the pressure it takes to pump the air. This is where it gets complicated. There are so many variables (orifice size, air density, crankcase volume, etc) that I haven't been able to get a good handle on the math yet.
One can get an idea of the potential losses by examining a simplified scenario. Let's say my concerns turn out to be valid and the average crankcase delta-P (crankcase pressure difference between piston down-stroke and up-stroke) is 10 psi over an entire revolution at 2700 rpm:
Energy loss per rev per cylinder = 10 psi * 90 ci = 900 inch-pounds = 75 foot-pounds/rev/cyl.
Energy loss per rev for the engine = 75 foot-pounds/rev/cyl * 4 cyl = 300 foot-pounds/rev.
Pumping power loss for the engine = 300 foot-pounds/rev * 2700 rev/min = 810,000 foot-pounds/min = 24.5 hp.
I have no idea how close that scenario is to reality. My Google searches have yielded no indication that anyone else has even considered this loss let alone measured or calculated it. If the average delta-P is even 1/10 of my scenario, it would still be of interest to those trying to wring the most efficiency from their engines. That's especially true for those running higher revs since the power loss is proportional to engine speed cubed.
I'll keep trying to figure out how to estimate the losses from the physics. There are so many variables that I think one would need to take measurements on a running engine to really know what's going on. That wouldn't be too difficult. An oscilloscope and a couple of miniature fast-response pressure probes should do the trick. A test stand or dyno would be best; though it could conceivably be done on an installed engine. Maybe I'll get bored and try it someday.
In my searches, I stumbled across one link that supports my hypothesis. I haven't seen the data myself but apparently aircraft engines become more efficient with altitude. I.e., at a given setting of RPM/MAP/FF, an engine will produce more power at higher altitudes than it does at sea level. Since pumping losses are proportional to density, they will be less than half the sea level value when flying at 25000 feet and could explain this effect.
If it turns out my hypothesis is correct, then the next question is what to do about it. One could enlarge the various air passages but I'm not sure one could make enough difference without overly weakening the crankcase. If the losses are proven someday, then maybe the engine manufacturers could be convinced to produce new crankcase castings with better internal air flow.
Sorry for my long-windedness but I wanted to get my thoughts recorded. Perhaps one of the many knowledgeable people on this forum knows the answer. For all I know it was hashed out and understood long ago.
Kev
I did a top overhaul on my IO-360 a couple of years ago. Since I had virtually no prior experience with Lycoming engine internals, I had expected to see into the sump once I removed the cylinders (like it would be on a car or motorcycle engine). Instead, I was struck by how compact and relatively closed the crankcase volume for each cylinder pair appeared to be. There are small air/oil passages beneath each main bearing and an additional small slot into the sump beneath the crank volume for cylinders 3 and 4. I don't have any measurements but my memory (and some pictures I've found on the web) indicate the passages may be small enough to present a significant impediment to air flow.
Over each revolution of an IO-360, the crankcase volume for each pair of pistons varies by 180 ci (cubic inches) between TDC and BDC. In principle, that means that the air passages must vent 180 ci out then 180 ci in on each revolution of the crank (per pair). By comparison, that's four times the air flow ingested through the intake ports. How much power that requires depends on the pressure it takes to pump the air. This is where it gets complicated. There are so many variables (orifice size, air density, crankcase volume, etc) that I haven't been able to get a good handle on the math yet.
One can get an idea of the potential losses by examining a simplified scenario. Let's say my concerns turn out to be valid and the average crankcase delta-P (crankcase pressure difference between piston down-stroke and up-stroke) is 10 psi over an entire revolution at 2700 rpm:
Energy loss per rev per cylinder = 10 psi * 90 ci = 900 inch-pounds = 75 foot-pounds/rev/cyl.
Energy loss per rev for the engine = 75 foot-pounds/rev/cyl * 4 cyl = 300 foot-pounds/rev.
Pumping power loss for the engine = 300 foot-pounds/rev * 2700 rev/min = 810,000 foot-pounds/min = 24.5 hp.
I have no idea how close that scenario is to reality. My Google searches have yielded no indication that anyone else has even considered this loss let alone measured or calculated it. If the average delta-P is even 1/10 of my scenario, it would still be of interest to those trying to wring the most efficiency from their engines. That's especially true for those running higher revs since the power loss is proportional to engine speed cubed.
I'll keep trying to figure out how to estimate the losses from the physics. There are so many variables that I think one would need to take measurements on a running engine to really know what's going on. That wouldn't be too difficult. An oscilloscope and a couple of miniature fast-response pressure probes should do the trick. A test stand or dyno would be best; though it could conceivably be done on an installed engine. Maybe I'll get bored and try it someday.
In my searches, I stumbled across one link that supports my hypothesis. I haven't seen the data myself but apparently aircraft engines become more efficient with altitude. I.e., at a given setting of RPM/MAP/FF, an engine will produce more power at higher altitudes than it does at sea level. Since pumping losses are proportional to density, they will be less than half the sea level value when flying at 25000 feet and could explain this effect.
If it turns out my hypothesis is correct, then the next question is what to do about it. One could enlarge the various air passages but I'm not sure one could make enough difference without overly weakening the crankcase. If the losses are proven someday, then maybe the engine manufacturers could be convinced to produce new crankcase castings with better internal air flow.
Sorry for my long-windedness but I wanted to get my thoughts recorded. Perhaps one of the many knowledgeable people on this forum knows the answer. For all I know it was hashed out and understood long ago.
Kev
