My Elbow Engine

[two dogs]

Well, I've decided to make an Elbow Engine. The plans can be had here:
http://www.john-tom.com/MyPlans/Steam Engines/ElbowEngine.pdf

The plans make it look easy, and the pieces look simple to make; only 10 parts.
I've read several posts here about people having frustrations with theirs. So simple parts that need to be made with a lot of precision to get it to work.
I've compiled some ideas from others here and added a few of my own to come up with something that looks very
promising, in fact as you can see from the included short video, it runs without a flywheel!
Over the next few posts, I'll outline what I did to get a running engine and why.
First and probably foremost, I made the pistons from brass, thinking that brass would be easier to bend to 90 deg.
They MUST be exactly 90 deg. It took a bit of tinkering to get that, so I can't imagine what it would be like with
drill rod. Second, I drew up the pistons up to make a bending fixture and the dimensions they gave didn't work, so I made some changes and cut and bent one. It bottomed out. I'm not sure what people did here because that wasn't one of the stated problems. I kept changing the piston design til I got a design to work; to bend repeatably and to be the correct length.
Well, that's enough for now, just a teaser. More tomorrow

Mark


View attachment Elbow.wmv

 

[rake60]

Very Well Done Mark! Thm:

The Elbow Engine is one of the most difficult engines to build.
When I finished mine and it finally ran my first thoughts were to
smash it with the 20 pound hammer to recover some of the frustration
it had caused me. LOL

I hope you don't mind, I've taken the liberty of hosting your video
so it can be embedded here.

[ame]http://www.youtube.com/watch?v=kBkVcwK9ZVM[/ame]

Great Work!
Rick

 

[b.lindsey]

Very nice Mark. I will be looking forward to hearing more about your re-design and bending of the pistons. That seems to be the major problem that most have with these engines...and from the video you seem to have gotten beyond that quite well.

Bill

 

[two dogs]

:big:
Thanks guys!
Rick, thanks for embedding the video, I'm still learning how to manipulate pics and things.
I too think the pistons are the biggest hurdle on this engine. I decided to try and use brass to make it easier to bend. The brass stock I used was .2495 in dia within about .0002 or .0003" thru the whole 3' piece and seemed to be very cylindrical as well. I didn't turn the dia, just polished it up with some Scotchbrite.
I drew one up to the straight dimensions and then "bent" it on my cad to get an idea for the bending fixture. I must have done something wrong because it seemed like the center section wasn't wide enough; it looked like the larger diameters would interfere when it was bent. So I designed it wider (.563 instead of .437) cut it, bent it and installed it in the cylinders.

I must backtrack here and mention that the rest of the engine except for the pistons and flywheel were made and assembled.
Seemed to turn OK but the ends bottomed out. Next I changed the 1.25 dim to 1.1
Made 3 and bent them

Oh yeah, bending fixture

Checked with my precision square and got them as good as I could.
Next gripped them (gently!) in my mill and indicated them. The first was .002 out of square. I tried to get it better but only got worse. The matl wanted to "unbend" much easier than to bend, so it took 15 min to get it back to .002"
The other two were .001 so I left them. Not too bad using a square :big:
Don't try to run the indicator back and forth, unless you have it perfectly level you'll run off the high spot. move the indicator up and down at each end of the piston and compare readings.

Installed them in the engine and it would move but they still bottomed out! Took them out and milled .1 off each end and that did the trick. I'll post a drawing soon with the mods I've made
I had figured that I would get it to work fairly smoothly before I even tried to put the air to it so I spent a little time swapping piston to cylinder bore. Got it pretty close then put a little toothpaste on each cylinder to act as lapping compound. Worked it around by hand for a few minutes then pulled it apart and cleaned very well.
By now if you turned both cylinders, it would rotate pretty easily, but turning one would not turn the other. I thought, "wonder what some air would do?" Hooked it up and by turning both cylinders it took off! Ran a bit and stopped. Put some oil in the air inlet, upped the pressure and it started running. Let it run for a few minutes and it seemed to loosen up. Turning one cylinder still won't turn the other, but that doesn't seem to be a problem. It almost always starts just by applying air.
Just went and checked, it will run down to 15 PSI.

Anyhow that's it for now, will post again soon and also get a drawing package together

Mark

 

[two dogs]

Got the drawing of the piston done
I should explain the bend fixture; I had put in holes for pins to locate one end and the bend itself, but decided it wasn't necessary. A pc of material with a .250 wide groove .200" dp or so will work great

edit: removed drawing; download improved one below
mc

 

[kvom]

Seeing your pics, I'm thinking an edge finder could be used to check the angle as well as a DI. Perhaps easier?

This thread was worth a karma point ;D

 

[kustomkb]

Nice work!

Thanks for sharing, this is one I definitely would like to build.

 

[b.lindsey]

Mark, how many times did you have to anneal the brass during the bending process? Also, do you plan on adding a flywheel even though it runs without one?
Bill

 

[two dogs]

Bill, I didn't anneal it at all, it bent quite easily. I will make a flywheel for it but it has some air leaks that I want to fix first

Mark

 

[mnbylcr2]

Excellent job two dogs. I have noticed that a few members who have built running engines have used valve port blocks on both cylinders. Do you think this would also help future builds? - clearly yours is working with just the one but I`m thinking of this build and I`m unsure whether to incorperate one at the base of both cylinders. If one is not needed on the vertical support then what is the need for the valve port block on the base cylinder?

Excellent job.

 

[two dogs]

Kvom, thanks for the karma point! :big:
I'm sure you could use an edgefinder instead of a DTI, i think it would be faster with a DTI tho.

mnbylcr2, the valve porting is built into the vertical piece. Not sure why it has one on the base but not on the vertical.

Mark

 

[two dogs]

Two more drawings
Added an "as bent" view for the piston.
Drew up the base plate and dimensioned it better (I think) and added counterbores for o'rings to seal the upright and valve port block as I'm having some leaks there

Mark

View attachment piston.pdf

View attachment base plt.pdf

 

[kvom]

The reason I suggested an edge finder is that it would not be necessary for the piston to be exactly horizontal.

 

[two dogs]

The reason I suggested an edge finder is that it would not be necessary for the piston to be exactly horizontal.

If that works better for people, then that's the correct way ;D

 

[two dogs]

The rest of the parts can be made as on the plans (with a couple of possible exceptions, mentioned later) as long as care is taken in certain areas. The upright (flywheel post on the plans) needs to have the valve side surface and side that mounts to the base exactly perpendicular, likewise the valve port block needs to have the two sides perfectly parallel. On the base drawing, I made the valve port block mounting holes a little oversize to give a little adjustment. This was because I needed the adjustment on mine.
The cylinders need some care too. After turning them from aluminum I used a rotary table with a 5C collet attachment, but a 3 or 4 jaw chuck would work too, just indicate it to get it centered. Last in order of preference would be to trig out the xy positions and move to each location. Whichever method, I would do each hole complete before moving on to the next. I spot drilled, drilled .156 dia, drilled with a 2 flute 3/16 end mill, drilled again with a .242 drill then reamed .2515", again doing each hole complete. I then did the center hole on both by the same method.
I then made the center bushing from brass to .005" longer than the cylinder and a nice running fit to the center hole. The bushing has a head slightly larger than the socket head cap screw I used x .100 thk. After doing this, I lapped the bushing and head together on 600 grit emery cloth to get the bushing closer to the cylinders length. Do a little then bolt the cylinder (and bushing) to the mating part. Do this til you get a nice fit, still able to rotate with only maybe .001" of up and down movement.
The only changes I would make, as mentioned earlier, would be to drill the center hole in the piston bushing to .166 and the mounting holes in the upright and valve port block to 8-32 threads instead of the 10-24 as on the original drawings. The way I did it(the 10-24 holes were already done) there is only a .03 wall on the bushing. My next one I'll make this change.
That's about it, I'll leave the flywheel up to you. I'll probably try a couple of ideas, one for this one and one for my next one; I'm going to try and bling it up some.

Any questions, please post and good luck!
Mark

 

[two dogs]

Does anyone see any problem with my drawing up a complete set of plans, incorporating the changes I've made and posting them here? I, of course won't present it as my own design, I'd just like to redimension and clarify a few things

Mark

 

[two dogs]

An update and a little more learning on my part here.
I had pulled it apart to add the counterbores on the baseplate for the orings. I had labeled the pistons and cylinders so I could get everything back the same way. Good enough? No.
I put it back together and it just wouldn't run at anything less than 30 PSI. It got tight at 1 or 2 places in the rotation. Tried moving the valve port block a little and nothing. Remember, I had built in a little adjustment into the locating screw holes. After about 10 minutes frustration I rotated the valve port block 180 degrees on the baseplate. WOW! What a difference.
Moral to this story: as you're "tuning" it label things so you can repeat what you've done. These engines are very "twitchy"
I'm going to experiment when I get done with this one. Maybe I'll try putting a little taper in part of the pistons so only maybe .200 or so is really contacting the cylinder bore. Maybe even try oringing the pistons so they are more forgiving.
I also have a leak between the lower cylinder and the vavle port block, only on one side, probably owing to only having .03 surface per side on the bushing to keep the cylinder perpendicular. This one, I'll mill the face of the valve port block to seal it better, but next time I'll go with the smaller thru hole.

I'll keep posting as I think of things, going to try to get a flywheel done here in a couple of days

Mark

 

[1Kenny]

When I made my pistons, I set them on a 3/8" thick plate glass after getting them 90 degrees wiih a square. You can feel them rock on the glass if they are out of alignment and tweek them into true. Used an oiled news paper as a gasket on the lower cylinder base, that worked well. If I try to turn it with the flywheel it will bind, but when it is running on air it will run slow with 9 psig.

Kenny

 

[bearcar1]

Mark, it would be interesting to see if you could in fact get this engine to run with some type of seal on the piston/bores. From what I have seen and read these oddballs have a nasty habit of spitting oil all over the place when in motion. Perhaps even incorporating a spring hold down of some configuration to assist in sealing the base plate to the bottom cylinder. All purely speculative at best but interesting to see if it is feasible.

BC1
Jim

 

[Blogwitch]

If you can access all the old posts from when the site started, making elbow engines was a fad then, and everything you are discussing now, was all done at that time.

With access to that material, and all the trials that went on to get them to run and work better, all your questions would be answered.

Just to recap on the requirements to get this engine running on low pressure

1 Everything has to be perfectly square and in line, no if's and but's, if it ain't square, it won't run on low pressure. Just rough handling the pistons can be enough to knock them out enough to stop the engine running. Non hardened silver steel would definitely be my recommendation for pistons.

2 Use a fairly thick oil for lubrication, WD40 or 3 in 1 is OK for door hinges, but not for this engine. An auto engine oil is a good place to start.

3 The engine requires gaps for it to run, so don't even think about tightening it up to stop oil spatter, if you can get them to within a couple of thou, it should run.

4 The alignment of both cylinders along the centre line is critical, any deviation side to side will stop the engine running. You should be able to spin the engine over very freely in both directions by using the flywheel, until that setting is reached, don't expect the engine to run on low pressure. You will most probably be able to force it over with higher pressures, but that will do the engine no good at all, in fact it might wear it out to a stage it wouldn't run after fine tuning.

It took very little time for me to make my engine, but it took a lot longer to get it running correctly. But now it is set, even a couple of years later, I can wack a few drops of oil onto it and it will start running straight away at any pressure above 5 psi. It does required lubricating every 10 to 15mins run time, as it throws any oil in there out, as you will discover.

[ame]http://www.youtube.com/watch?v=5ZRKNlY7enM[/ame]


Blogs