Compressed Air Engine for a cycle

[thehitchhiker]

A friend and I are working on this idea which is fairly common, but I'm having trouble starting out. The idea is that of a compressed air engine. Where we take a normal, small capacity engine and modify it such that only compressed air, stored in tanks is sufficient to move the piston.
Now, I'm not too sure how to go about starting this. We thought we'll get a small scrap engine and see the amount of pressure we'll need to make the piston start moving. Based on what we see, we go forward. As for the calculations part, I'm completely lost.

Suggestions anyone?

 

[buffordboy23]

If you want to talk calculations, you are heading down a complicated path...

Do you know how an engine works? A 2-stroke or 4-stroke engine will make a difference as well. How many cylinders? There are a lot of variables to consider...

If you want to do calculations, I recommend a simplified approach first. Consider only a single cylinder engine. What geometry best describes the inertia of the crankshaft? What is the mass of the connecting rod/piston assembly? Your basically trying to ballpark the linear force that acts on the piston to cause a rotational force (torque) about the crankshaft axis. Another caveat is that the torque values will fluctuate over one engine cycle b/c of the the angle between the con rod and crank throw. If the force on the piston was constant, this would simplify matters too...in an internal combustion engine (ICE), this is not the case. The pressure-volume gas law (pV = nRT) and knowledge of piston-face area will allow you to compute the force, via force = pressure x area.

There is a lot to consider and I don't know your background...hopefully, I gave you some things to think about. My answer is on a basic level. Please ask questions if you need clarification.

 

[HowlerMonkey]

It would always be a "2 stroke engine" because there is no need for the compression or intake stroke.

When I say 2 stroke, it applies to cam timing and not using ports as found in a 2 stroke engine.

If you start with a conventional 4 stroke engine, first you would double the speed of the camshaft or add lobes.

You also may find any camshaft timing scheme available today for your engine of choice is not favorable to this concept.

 

[FAlonso]

If your work is application specific and you are doing this for running something, try doing load calculations for which the engine should be designed and then find the operating parameters of the engine like pressure, volume needed etc

 

[Unrest]

I think you could do a fairly simple initial analysis by assuming the cylinder is always at the supply pressure while it's filling, which would be approached if it was turning slowly:

Mechanical power = air power
torque * angular velocity = pressure * flow rate

If you know the required torque and speed, you can choose the pressure and flow rate according to other factors that suit you.

You can select the engine capacity from the flow rate and speed. flow rate = (power strokes per second) * cylinder capacity.

If you have a measurement of the pressure required to just move the piston, and the angle of the crank shaft (required pressure will depends strongly on crankshaft position), then that tells you the minimum torque it needs to provide - enough to overcome internal friction. So add the load's torque on to it. It might be easier to measure the torque directly, say by using a torque wrench or fish scales.