HP/torque vs. RPM

[YellowMaranello]

Yet another idea I had at work, suppose you took a 4 cylinder car engine. Now suppose you took it up to 100,000 RPM (don't ask how, just assume, how isn't the question). Would it make more or less hp/torque at 100,000 RPM then it would at 8,000?

[454Casull]

With torque reasonably flat all the way to redline [actually, doesn't even have to be relatively straight], HP will be higher at 100K RPM than at 8K RPM.

[SaabJohan]

Lets say that we have an engine which gives 210 hp at 8000 rpm. If all values are the same, power at 100k rpm will be 2630 hp. Since all values are the same the torque will be the same. In real life the power and torque will be much lower at 100k rpm because such high revs cause the effiency to decrease since losses will be higher than at at lower rpms.

[ivymike1031]

you didn't mention that at 100,000 rpm you'll have a REALLY hard time getting charge into and exhaust out of the cylinder. I'm not sure whether you could burn gasoline at that speed either, but perhaps if you were running HCCI or something you could.

[911GT2]

Hypothetically, forgetting that engines lose energy and also forgetting that energy is consumed exponentially as speed increase (IE E=MC squared, different uses but same principle), it would obviously increase power as more revolutions are made.

But since this doesn't happen, there is a point at which power output is optimal, at about 18K revs, right where F1 engines redline. There's no point in going further, the energy loss counteracts any power gains.

And Johan, remember that very few torque curves are perfectly flat (they're called curves for a reason). So that 210 probably wouldn't be 2600, it'd be pretty well unpredictable in a hypothetical engine like we're talking about.

[SaabJohan]

As I said the the torque curve would be flat if all values were the same. And if they were that the power should be 2630 hp. To calculate this I used 2l displacement, 100% VE, 30% effiency, SAFR 14,7 and 44 MJ / kg fuel energy content.

Today F1 engines is running at 18000 rpms, but this rpm will probably increase. One of the BMW engines have actually been over 20000 rpm without taking damage, and that was something unthinkable a few years back.

The problems with high revs in F1 engines are mainly, piston acceleration and friction. In the past one big problem was caused by the valvesprings, today they use pneumatic springs.

Lets say that we want to build an engine which could go up to 100000 rpms, at first we must solve the problem with piston acceleration and speed. The engines should be a four cylinder and if we use 160 mm bore and 25 mm stroke the max acceleration will be 1478000 m/s^2, mean piston speed 83 m/s and 131 m/s in maximum speed. An F1 engine at 18000 have a mean piston speed of 25,2 m/s, max acceleration of 84400 m/s^2 and max piston speed of 40 m/s. This means that if the reciprocating parts for one cylinder should have the same weight as in the F1 engine stresses in our engine would be 17.5 times as high. Then we should have serious problems with friction and with buildning up an oilfilm on theese places.
The bearings must also be of a non contact typ, like floating bearings or magnetic bearings.
To get air into the cylinder we could use a compressor, induction wave charging and helmholtz resonator charging. The big bore should make it a littler easier to fill the cylinders. To help the exhaust out we could use pressure-wave exhaust gas scavenging.
To get the fuel to ignite and burn under short time is also a big problem, maybe homogenus-charge compression-ignition, as mentioned, could solve this partially. But variable compression is preferd in HCCI engines, and this makes it even harder to construct and build the engine.
And then we must construct valves which can be used at this engine speeds.
But the biggest question of them all is why do we want to use such high engine speeds? If we want much power from a small engine, use a turbocharged engine instead, or a gas turbine.

[ivymike1031]

SJ, the original question was whether it would make more or less HP & torque... if you assume that it makes the same torque, you've also assumed that it makes more hp, and thus "assumed away" the question completely.

I'm not sure why you used all that extraneous information in your power calculation. If you assume the same torque, then you could get power by multiplying by the speed ratio (210 * 100/8) = 2625. To carry it out an extra step, 210 hp @ 8000 rpm means 138ft*lbf torque, 138ft*lbf torque @ 100000 -> 2625hp

I won't bother picking apart the rest of what you said, because I think that you're answering the wrong question. I took the question to mean "If we assume that an engine could be built (mechanically) to run at that speed, what would be the effects on performance?" The first two things that come to my mind are:
* breathing goes in the shitter
* not enough time for combustion

If we take the original question literally (magically take a conventional 4-banger to 100,000rpm without breaking it), then the inevitable answer is that the engine would make a large negative amount of power, and a large negative amount of torque. This means that if the engine could get there mechanically, you'd have to drive it with an external power source to make it spin (in other words: it wouldn't run). I think that breathing would be the first problem you'd hit. We're all familiar with the drop-off towards the end of a conventional dyno chart. If you continue that curve for a little while, it passes zero and keeps dropping. Mechanical issues aside, the second zero-point on the hp and torque curves will be the maximum speed at which the engine can spin under its own power.

Interesting note: the HP and torque curves would start at 0hp & 0ft*lbf @ 0 rpm, cross each other on the way up (X ft*lbf & X hp @ 5252rpm if the scales are the same), and cross each other again on the way back down (0 ft*lbf and 0 hp @ Y rpm).

[911GT2]

That's great if the torque values are the same, but they never (and i repeat NEVER) are. So the curve isn't flat, but as mike said, intercept at 5252 rpm. Remember, hp=torque*rpm/5252.

[SaabJohan]

I took the question so, that we want a 4 banger to run properly at 100000 rpm. Whats the fun by rev it up by an external power source, then it's no longer an engine, just an airpump.

[911GT2]

Now that you think of it, with so many rpms, the engine's exhaust would have a similar effect to that of a turbine. It'd be cool as hell (figuratively, that is....it would actually be skin-melting hot).

[sciguyjim]

Ivymike, you said:
"If you assume the same torque, then you could get power by multiplying by the speed ratio (210 * 100/8) = 2625."

Does the relationship hold in the other direction too? For example, my stock 305 tpi is rated at 195HP at some rpm (I don't know, lets say 4000 rpm for simplicity). If I want to know what HP I'm feeling at 2000rpm all I need to do is (195*2/4)=97.5HP? Or is this too far outside the usual range of HP and rpm calculations to be a valid approximation? Thanks.

[ivymike1031]

The key to my statement was the assumption of constant torque. If you know torque and speed, you know horsepower (torque * speed = power). If you look at a dyno chart for your engine, you'll notice that the torque rises until a certain point, then begins to fall. Horsepower follows a similarly shaped curve, but falls off a bit later.

The simple answer: no, it wouldn't work for you.

On a turbocharged vehicle, where the torque trace is flat over a large rpm band, it would work.

[Chris]

Generally, the higher an engine is designed to rev, the lower the torque (big bores, as mentioned earlier, dont like torque. And a bore 7 times bigger than the stroke, Christ, it would look like a frikin nickel!!!) So this engine (if it worked), would have not much torque, ever, but would make nice chunks of horsepower. And so, you would run it through a nice great gear reduction for low speeds.

[SaabJohan]

Originally posted by Chris
Generally, the higher an engine is designed to rev, the lower the torque (big bores, as mentioned earlier, dont like torque. And a bore 7 times bigger than the stroke, Christ, it would look like a frikin nickel!!!) So this engine (if it worked), would have not much torque, ever, but would make nice chunks of horsepower. And so, you would run it through a nice great gear reduction for low speeds.

Remember that a big bore/short stroke engine gives the same torque at a given rpm as a small bore/long stroke engine if the cylinder pressure, and volume is the same in both engines. But as the rpm increase it's harder to reach high cylinder pressures.