Single cylinder pneumatic "race" engine

[taiden]

Hi all,

Part of my BSMET program is to build a steam engine from plans we are given. This is to get us used to working with machine tools before we take on larger projects.

At the end of the semester, we "race" our engines and see who gets the highest RPM from the standard design. The standard design is single cylinder, double acting, 1.125" bore, 0.468" stroke, 4.000" rod. The record speed was in the 4000 RPM range. Destruction is not uncommon. The test is done under no load.

I'd like to design and build my own steam engine to test alongside all the others. To keep things 'fair' I have decided to keep the single cylinder and 1.125" bore. From there, it's pretty open ended.

Does anyone have experience building single cylinder "race" engines? I'm looking for wisdom regarding bore/stroke ratio and rod/stroke ratio. I'm also looking for ideas for adjustable timing systems that will hold up to the high RPM use. But most importantly, I'm just looking for guidance in finding designs that are proven to work, so I can learn from them.

 

[barnesrickw]

Lots of people on here have done many air powered engines. Chuck F. had some cool valve ideas that look like they would increase torque. The only advice I have is short stroke means high RPM.

 

[vederstein]

Think about friction reduction:

If you can get away with it, make it single acting so you don't need a crosshead. This would eliminate reciprocating mass and friction. Also use a bash valve (plastic or rubber) so you don't have valve gear friction and mass. Make it uni-flow so you don't have an exhaust valve.

Use roller, not plain bearings, if the bearings can handle the reciprocating load. Use a small flywheel with low mass moment intertia. You only need a big enough flywheel to return the piston to the bash valve.

I agree with Barnesrickw, use a short stroke with a short connecting rod. Make your reciprocating components as light as feasible.

Since you're going for speed and I assume you have an unlimited supply of air, I recommend erring on the side of sloppiness vs. good piston sealing. Of course this can easily be overdone.

I hope this helps...

...Ved.

 

[Tin Falcon]

IMHO this is a poor test of engineering and machining skills. But I am not setting up the rules.
Hook up a baker fan and that would be a real test of the engines then see whose engine runs the fastest. Suggest to the instructor you use a baker fan. for the test.

Ved pretty much covered what needs to be done. I will add the tighter the tolerance you can keep on the machining the better. Not a tight fit but tight tolerances. Also balance the flywheel and the crank . At high rpm it will be vibration that kills the engine.
Tin

 

[Jyman]

Couple questions,

1. I know you said steam engine, but is steam or air going to be used to run the engine?

2. What pressure is the supply at?

 

[barnesrickw]

keeping inertia low is a good idea. If you can use more than one cylinder, twin can reduce the need for a flywheel. A heavy flywheel takes a lot of energy to get spooled up. I'm thinking opposing twin with 90 degree offset crank. Or, Scottish yoke. I think there is an engine called "Double Scotch" out there that may be worth a look.

 

[taiden]

IMHO this is a poor test of engineering and machining skills. But I am not setting up the rules.
Hook up a baker fan and that would be a real test of the engines then see whose engine runs the fastest. Suggest to the instructor you use a baker fan. for the test.

The class itself is meant as an intro to machining, and the test is steeped in tradition at the school. A baker fan would be best, no doubt about that. As far as a test of engineering goes, I think anything is a valid test of engineering if there are real design constraints and a specific goal. IE: 1.125" piston, single cylinder.... maximum sustained RPM under no load. That's one reason why I'm taking it upon myself to design something from scratch.

At this point I'm considering a double acting single cylinder wobbler with a dead piston in a horizontally opposed configuration to match the driven piston with equal and opposite impulse. I'm thinking it could have replaceable or adjustable valve plates to adjust the timing.

That said, my original plan was to do a traditional fixed cylinder single acting like Ved suggested. If I can figure out a good valvetrain for that it might be the ticket.

Also, while we refer to it as "the steam engine project" it is in fact powered by compressed air. I don't know what the working pressure is, I'll try to figure that out. But I'll probably build a small boiler just for fun.

 

[taiden]

Here's a quick animation of the standard steam engine that I will be competing against. [ame]http://www.youtube.com/watch?v=Pi-OcbBU6Dw[/ame]

 

[lohring]

Flash steam tethered boat engines exceed 10,000 rpm under load. It's a complete package with the pumps, burners, and boiler as important as the engine. Search for flash steam in this site or look at this thread for more detail.

Lohring Miller

 

[Tin Falcon]

Taden Thanks for sharing the engine animation. any chance you can share the plans here?
I see a lot of room for flex and unwanted oscillation at higher rpms.
IMHO if you shorten the con rod from say 4" to 3" make your two legged table a 4 legged table and work in a second bearing point between the eccentric and the crank you will have a lot more solid engine.
Also increase the base thickness and mass .
Oh yes lock tight on fasteners. this can help engines from falling apart on a test run at high rpm DAMHIKT. 8)
Tin

 

[Tin Falcon]

The class itself is meant as an intro to machining, and the test is steeped in tradition at the school.

I figured as much.

A baker fan would be best, no doubt about that.

Knowing about the existence of a baker fan and the principles at work is an important step.

I think anything is a valid test of engineering if there are real design constraints and a specific goal.

IMHO it is also important to quantify and verify results.

A bit difficult to repeat this sort of test. I guess the important thing here is knowing there are better testing methods than how fast will it go before it breaks into pieces.

Baker fans and dynamos are easier on the engine . even a prony bake will not have the engine running to the point of failure. Although more strain on the parts.
But I expect most engines survive it is just fasteners fall out and the engine can no longer run.
Tin

 

[taiden]

I will first say that all of the following documents are property of the University of Maine's Mechanical Engineering Technology Department... now that we've got that out of the way...

Please note that these drawings SHOULD be complete but there may be errors. I drew them, and I wouldn't be surprised if there are changes made as the engine is built.

Enjoy

https://dl.dropboxusercontent.com/u/23865811/STATIC/UMO-BSMET-Steam-Engine-2013.pdf

 

[Tin Falcon]

Thanks. I can not promise that this engine will not be built buy a few here. I like it . But I think you can trust people here will not use it as there own design or otherwise abuse or distribute the plans. It is a little big for my personal taste so if I built one it would likely be smaller.

A couple of things I noticed :
1) the lack of a cross head to support and guide the valve rod.

2) i would suggest you add bearings bushings to the ends the main bearing block. and then add an oil cup to the middle. that way there will be less friction as the shaft will be supported on both ends of the bearing blocck but not in contact the whole length.
Tin

 

[taiden]

Oh I don't think anyone will mind if people use the plans to build their own. That's not a problem at all. Just needed to give the "standard disclaimer." I'd love to see people use these plans for their own designs.

 

[Tin Falcon]

yep understood



Also if you have the time and transportation you may want to join a few of the folks here at the NEMES show in a few weeks http://www.neme-s.org/show2014.pdf there are people that drive down from maine.
Tin

 

[Heffalump]

Those drawings look very professional Luke, what software did you use?

 

[deverett]

... A couple of things I noticed :
1) the lack of a cross head to support and guide the valve rod.

2) i would suggest you add bearings bushings to the ends the main bearing block. and then add an oil cup to the middle. that way there will be less friction as the shaft will be supported on both ends of the bearing blocck but not in contact the whole length.
Tin

There seems to be a lot of overhang on the crankpin. Would it be possible to make a shorter bearing at the flywheel end and put a support bearing between valve eccentric and crank?

Also, if you use roller or ball bearings instead of bushings, you will reduce friction considerably, so your top speed should improve.

Very nicely drawn, by the way.

Dave
The Emerald Isle

 

[taiden]

Those drawings look very professional Luke, what software did you use?

Those drawings were done in Solid Edge. I also have all of the solid part files available.