I saw a brief mention in another thread about using a 3-cylinder engine as an air compressor. My mind was already headed along that route when I came across the reference so I guess I'm not the first one to come up with the idea, but I think I'd like to run my version of the idea by you guys to see if my plan is sound.
My current plan is to find an engine from a Geo Metro. Those engines were either 1.0L 3-cylinder or 1.3L 4-Cylinder engines depending on the model (LS or LSI). I'd strip all the electronics and exhaust off the engine and seal up each exhaust port individually to isolate the cylinders from each other. I'd remove the spark plugs and thread pipe into the spark plug ports. The pipe from each plug port would thread into a Y of some sort that would have 2 check valves in opposition to each other. The check valve that would allow air to flow back into the cylinder would then be routed back to the intake and the other check valve would be routed to some sort of manifold that would gather the compressed air from all three cylinders and route it to the holding tank. That arrangement of check valves would convert the exhaust stroke into an additional compression stroke doubling the engines efficiency as a compressor.
The engine flywheel would be removed and replaced with a pulley to allow the engine to be driven from an electric motor.
The questions I have would be how large of an electric motor would be required? I've got a couple of motors laying around from swimming pool pumps that are rated 1HP @ either 1800 RPM or 3600 RPM. Can't remember which at the moment. I'm thinking of using at least a 2:1 reduction between the electric motor and the engine to raise the torque at the engine's crankshaft.
The next question would be how many cubic feet per minute could I expect from this arrangement assuming a 1.0L displacement running at 900 RPM? If I'm looking at the info on Wikipedia correctly then the displacement is based on the total air moved through an engine during one complete 4-stroke cycle which would comprise 2 complete rotations of the crankshaft (At least for a 4-cylinder engine). If that is correct then my check valve manifold would actually double the engine displacement making it 1L per rotation. 1L = .035cf so at 900 RPM I would be getting close to 30cfm. Does that sound right?
I've done a compression test on my Geo Tracker's engine and get between 150 and 170 psi on each cylinder. Would it be reasonable to expect a compressor like this to produce a working pressure in that same range?
Would that mean I'd end up with a compressor that could produce close to 30CFM at 150 PSI or would the compression ratio of the engine need to be taken into account in calculating the cfm?
My current plan is to find an engine from a Geo Metro. Those engines were either 1.0L 3-cylinder or 1.3L 4-Cylinder engines depending on the model (LS or LSI). I'd strip all the electronics and exhaust off the engine and seal up each exhaust port individually to isolate the cylinders from each other. I'd remove the spark plugs and thread pipe into the spark plug ports. The pipe from each plug port would thread into a Y of some sort that would have 2 check valves in opposition to each other. The check valve that would allow air to flow back into the cylinder would then be routed back to the intake and the other check valve would be routed to some sort of manifold that would gather the compressed air from all three cylinders and route it to the holding tank. That arrangement of check valves would convert the exhaust stroke into an additional compression stroke doubling the engines efficiency as a compressor.
The engine flywheel would be removed and replaced with a pulley to allow the engine to be driven from an electric motor.
The questions I have would be how large of an electric motor would be required? I've got a couple of motors laying around from swimming pool pumps that are rated 1HP @ either 1800 RPM or 3600 RPM. Can't remember which at the moment. I'm thinking of using at least a 2:1 reduction between the electric motor and the engine to raise the torque at the engine's crankshaft.
The next question would be how many cubic feet per minute could I expect from this arrangement assuming a 1.0L displacement running at 900 RPM? If I'm looking at the info on Wikipedia correctly then the displacement is based on the total air moved through an engine during one complete 4-stroke cycle which would comprise 2 complete rotations of the crankshaft (At least for a 4-cylinder engine). If that is correct then my check valve manifold would actually double the engine displacement making it 1L per rotation. 1L = .035cf so at 900 RPM I would be getting close to 30cfm. Does that sound right?
I've done a compression test on my Geo Tracker's engine and get between 150 and 170 psi on each cylinder. Would it be reasonable to expect a compressor like this to produce a working pressure in that same range?
Would that mean I'd end up with a compressor that could produce close to 30CFM at 150 PSI or would the compression ratio of the engine need to be taken into account in calculating the cfm?
