I Know Nothing About Car Systems

[Occasionally6]

The two types of starter-generator that I can think off the top of my head are those used by GM in the mild hybrid Vue et al and the PHT Silverado.

The former is belt driven with some belt tension magic to maintain belt tension when either driven by or driving the engine. (The tension side of the belt shifts depending on which it is doing.)

The PHT has the starter-generator attached to the ring gear on the back of the engine and is direct drive, concentric with the crankshaft.

[jeff88]

Is it possible to have an engine with enough pistons to not need a flywheel? I know it's not really a feasible proposition, but I just want to know if it is theoretically possible? If so, how many would it need? If the crank turns 360°, how many times do you need a piston to "power" the crank to eliminate the flywheel? Would you need 360 (one every degree) or could you get away with less?

These are the random thoughts I come up with...

[some_other_dave]

I don't think it's a matter of the number of cylinders. The pistons, after all, only go up and down. You need the flywheel to store rotational energy, not up and down energy. It also serves as a very convenient place to get the power out of the engine, and even to mount the starter in most cases.

Note that the early aircraft rotary engines (think the Sopwith Camel and similar) did not have any flywheel. Instead, they bolted the end of the crankshaft directly to the airframe, and bolted the propeller directly to the engine. The whole engine spun around, while the crankshaft was held still. It effectively used most of the mass of the engine as a flywheel.

-soD

[Occasionally6]

If we idealize the crankshaft and pistons as having no inertia or friction, the energy requirement reduces to that needed to compress the charge and perform the pumping work. One working cylinder will do that

Since the power stroke (4-stroke ICE) occurs over 180 degrees of crankshaft rotation, out of 720, and assuming even firing, 4 cylinders is sufficient to ensure that at least one cylinder has fired and is releasing energy at any crankshaft position.

A real engine differs in that there is friction and that the pistons and crankshaft do have inertia.

With very low (or idealized to none) crankshaft + flywheel inertia the engine speed will fluctuate throughout the crankshaft rotation as the torque from each cylinder varies with cylinder pressure and crank throw.

[bestclimb]

I think 5.

the flywheel stores rotational energy from the power stroke to drive the engine through intake, compression and exhaust.

A power stroke lasts 180 degrees (actually less as torque comes to 0 on the top and bottom of the stroke) in a zero inertia engine, or if the load (work being done and internal friction) exceeded inertia the engine could stop. torque reaches zero at the bottom of the stroke, and rings and bearings and pumping losses continue.

A 4 cylinder 4 stroke engine must go through 720 degrees for each cylinder to fire. so in 2 revolutions each jug fires once. with 0 overlap which means one fire is going out, just before the next starts.

with 5 cylinders there is some 36 degrees of overlap as one cylinder is about 18 degrees from ending the next one is already 18 degrees into it's power stroke.

with 5 cylinders there is 900 degrees of power stroke by the time all of them fire. As 360 does not go into 900 nicely there is some overlap.

It would not be a smooth running engine as the power pulses would not be smoothed by the flywheel.

as for starting, one could rotate the engine to just after TDC then fire a spark off.

[Occasionally6]

I get what you are saying but 4 should be enough. The pressure in the cylinder at the bottom of the expansion stroke is always higher than that at the top of the compression stroke in the next cylinder to fire. (There are more moles of gas and they're hotter.) At WOT it's that simple and the engine will run.

To simplify it further, picture a 2 cylinder engine without valves, each 180 degrees apart on the crankshaft, with each cylinder containing the same mass of air as the other. In one cylinder the air is hot and the other it's cold. Where do the cylinders end up?

Throttled would require subtracting the pressure difference between that in the cylinder that is on the inlet stroke and that in the cylinder on the exhaust stroke, from that in the cylinder that has just fired. As long as the sum of those is still greater than the pressure in the cylinder that is on compression - also at lower pressure than with WOT - and it will be, the engine will run.

Even allowing for the exhaust valve to be opened before BDC and ignition advance in the next cylinder in the firing order, that will be the case.

(jeff88, you ask good questions.)

[pgfpro]

Jeff I always enjoy your questions.

[bestclimb]

I get what you are saying, but with rotational inertia being very low there is not much to insure that the pressurized fuel air charge pushes the piston down the correct side of the crank. a 4 pot might work, a 5 pot would work.

[Occasionally6]

It might be interesting to investigate what combination of ignition and exhaust valve timing defines the limits before there would be a possibility of the firing cylinder trying to turn the engine backwards.

If the ignition and exhaust timing were never such that the pressure in the firing cylinder exceeded that in the exhaust cylinder until the direction was decided, it would be OK with 4.

As a rule of thumb, ignition timing is typically set so that the maximum pressure in the cylinder is achieved about 10 degrees past TDC, ignition advance from 30 to 0 degrees BTDC. Not sure if that's enough info. to allow some thought about it.

I'll have to think about when the crankshaft speed is greatest. If it's not at a minimum at TDC/BDC then the engine should keep turning. The crank doesn't stop with the pistons and will have a peak speed as many times as there are cylinders but I'm not sure what the angular disposition is.

[jeff88]

Originally Posted by Occasionally6

jeff88, you ask good questions.

Originally Posted by pgfpro

Jeff I always enjoy your questions.

Thanks for the support guys. I once had a professor that told me to get A+ answers you have to ask A+ questions. Although my ideas may be outside of the box, I try to ask intuitive questions.

I wish I had access to resources to just 'play' with stuff. Like a lab at school that was specifically designed for you to create, adjust, record and test your ideas. I could "build" an engine and work out these hypothesis to see what works. Oh, if only.

Could pressure equalization, intake/exhaust timing and spark advance/retarding make it possible to eliminate the flywheel or is something else involved?