Determining power curve of steam/air engine.

[Brian Rupnow]

It occured to me this afternoon while messing about with the valve timing on my new twin horizontal engine, that it would be nice to have some way to meaure the difference it makes in a small engines power output when the valve timing is adjusted. I can think of a couple or 3 ways of doing this. Probably the low tech method would be with a tachometer. Assuming that the air pressure and the inherent friction of the system were unchanged, "better" valve timing would give higher RPM, while "worse" valve timing would slow things down. Somewhat "higher tech" would be to drive a small generator and measure the change in generated amperage/voltage that result from timing changes. One of the rather neat ways of measuring would be with the "floating ping pong ball" method. This is a system I devised for measuring air flow through die passages of an injection mold. Any time that you have a stream of air coming out of a nozzle, the velocity of the cone of air exiting from a nozzle is greatest around the perimeter of the air stream. If the nozzle is pointed straight up, you can float a ping pong ball on the air stream, and it won't fall out because of the higher pressure/velocity around the outside of the air stream.--And the more efficient the air passage is designed, the higher the ping pong ball will float, because the more air will flow thru the passage.--I won a hundred dollar bet one time with a die casting engineer who thought his die design was superior to mine. We ran air at equal pressure thu his design and thru mine, with a vertical nozzle attached to the exhaust. My ping pong bal floated about 4" higher than his---unfortunately, he owned the company I was working for at the time, but thats another story----If you hooked a small steam engine to an air pump (aquarium bubbler size) you could measure a change in power output by seeing how much higher or lower the ping pong ball floated----

 

[wareagle]

Brian, your description sounds like flow testing the engine for efficiency rather than for power output. The two are related, but there are a couple of other things to consider regarding power output on these little engines. Torque output really comes into play when determining the engine's output. RPM alone really doesn't give the story.

Torque is what really does the work. Think of a big truck's diesel engine compared to an indy car's engine. The diesel turns about 2000 RPMs for max power and the indy car's turn 13,000 or more. The indy car engine has a higher horsepower rating, but I bet you it wouldn't pull a trailer down the highway. The same is true of any engine, including our little models.

For flow testing purposes, the ping pong ball test would be the best indicator of flow. As far as testing the actual power output, some sort of small dynometer would be needed to get an accurate output measurement. It could be done with electrical or hydraulic means.

Hope that makes sense!!

 

[mklotz]

A Prony (not pony) brake

http://en.wikipedia.org/wiki/De_Prony_brake

is the traditional method for measuring small engine delivered (brake) horsepower. For our small engines, a simple wooden clamp and a kitchen (or jeweler's) scale can be used. Some form of tachometer to measure engine speed is needed.

Once delivered HP is known, you can use the STEAMHP program from my page to calculate the theoretical horsepower and, from the two values, calculate an approximate efficiency for the engine.

[Aside: The early developers of the steam engine were interested in power, of course, but were just as interested in efficiency since the latter translated directly into how much fuel their customers needed to buy and fuel was expensive then as now.]

Inferring HP from generators or air pumps driven by the engine bothers me since it assumes a linear relationship between engine power and driven device output that may not really exist. Of course, it all depends on how thorough you want to be about the test.

 

[BobWarfield]

The engine dynos I've seen for car engines used water pumps.

What are the linearity issues that concern you Marv, and why won't the De Prony brake have similar issues? It sounds fiddly to me to get the friction right on that arm. There must be linearity issues there too.

 

[Bluechip]

Hi Folks

A Sellers [ Sellars? ] Brake is easier to make and set up. It's what I used years ago.

You can use a dynamo/load for comparative testing, it won't give an accurate HP figure, but will indicate whether any change is for the better or worse. By the way, a car wiper motor is about 60-65% efficient as a generator.

You need to use a wire wound resistor as the load, filament lamps vary enormously with temp. Gives some misleading figures. How do I know? Been there.

Dave.

 

[Tin Falcon]

one of the older more common ways of engine testing was the Baker Fan invented by Abner D. Baker, Sorry the site that has the best info on it seems to be down.
But it was used to test and run in steam traction engines and puts less stress and stain on the engine than a prony brake.IIRC The load put on the engine is proportional to the square of the rpm.
I have seen these built in many sizes and should not be difficult to build
Tin

 

[Maryak]

Brian,

If you want a very very rough idea of what the output may be try this. ??? ??? ???

(2xPLAN/33000)*0.7

P = Cylinder pressure (psig)
L = Length of stroke (ft)
A = Area of Piston (sq ins) Mean of topside and underside which is smaller due to piston rod
N= rpm
2 because it's double acting
33000 ftlbs/min = 1 HP
0.7 allows for generally accepted friction and thermal losses in a recip engine.

Best Regards
Bob ;D ;D