V-twin compressor pump...

[vile_fly]

I won't be using air cooling, but I will be using the fan/flywheel in addition to 1/4" copper pipes flattened out slightly to increase contact area and placed in a plastic water container. The hot end will use 5/32" brake lines....lots of them. I might use copper if my fuel supply is too feeble to melt it. (sterno camping fuel) The original cylinder heads are junk. Too much dead air space. I am using premade flat 4" x 4" steel plates about 1/8" thick. The head gaskets will be cut from gasket paper and then painted with aluminum spray paint just before assembly. I will probably have to use a copper gasket on the hot side. So I have drilled and tapped holes in the plates for 1/8" pipe thread-to-1/4" tubing compression fittings to attach to the copper pipe. Those fittings are hard to find, even though the inside diameter is exactly the same as the tubing.

As to the piston rings, I HAVE to let them do their job breaking in and polishing the rough cylinder wall. Yes, that crosshatch pattern may be smooth to the finger, but I tell you it is still rough. Trust your eyes. Ever see the inside of a lawnmower cylinder after it has run for some time? I has a MIRROR finish. Breaking in the engine will be done by turning it with an electric motor and testing the temperature differential capability of this design....safely. When the mirror finish is there, only then will I switch to teflon and rubber seals. Besides, 80mm graphite bar stock is really expensive. I am considering using castable alumina ceramic for making the hot piston, but it is $96 per pint at http://www.graphitestore.com
and I am not rich enough to risk it on an untested design. If I can't use it as a stirling engine, then I will use it as a stirling cooler. At least that way, I won't be wasting much money. If my research is correct, I should be able to reach a maximum of -100 C on the hot end. Cool.

Oh, by the way someone else is working on this, too. Has disassembly pictures to save me the trouble of showing everyone.
http://fieldlines.com/board/index.php?topic=143774.0

[Ian S C]

As pistons go, a well fitting cast iron piston can be made very light, it does'nt have the stresses of an IC engine, or a compressor, and a number of shallow grooves on the skirt form a labyrinth seal. The bores can be lapped, using a cylindrical lap made of wood, do this then make the pistons to fit. On the hot piston put a light weight stainless steel cap, Constructed similar to a displacer, it only needs to be about one diameter high. I'm running a similar type of thing but parallel cylinders with a Ross Yoke, its only 35mm bore and 20mm stroke, but planning something similar, but 50mm+ bore. The current one will be pressurised once its run in.
I think you said you were a certified mechanic NO, but after you'v been building these things a while some one else may suggest that you need to be certified!
I'm not a qualified mechanic, and my ideas ar'nt always right, but some times they maybe a help. Ian S C

[vile_fly]

Well, you have a point. The thermal conductivity of stainless steel is 16 W/mK, alumina ceramic is 24, and graphite is 25-470, depending on the grade. I hate working with stainless, but I think I can come up with something.
The frictional coefficent and wear characteristics of cast iron cylinders are not too good against a cast iron piston under heat, but stainless steel is a better match. A lot of old japanese imported cars started out using steel piston rings instead of iron rings, and had longer lasting wear characteristics. That's why they lasted so long. I probably should use cast iron pistons if possible, its just so expensive to buy in bar stock form, I might as well be buying titanium.
I did buy some .002" stainless foil for a hot cap on a different design engine. I couldn't destroy it in the open air with a mapp gas torch, just make it glow. Looks promising if I trap it between two flanges on the end of a cylinder.
Still having trouble finding metric (80mm) teflon seals retail online.
Oh, and this v-twin compressor block will accept stock volkswagen air-cooled cylinders. Just have to use thinner studs or enlarge the holes so you can use a thicker stud. Not a perfect match on the bolt pattern, but close enough.

Definition of a stirling engine: thermodynamic nightmare.

[Ian S C]

For displacers and hot caps, I use thin walled stainless tube with an end TIG welded in, thickness around .007", I use a little thicker for displacer cylinders, I'v got a tame TIGger, who has a sheet metal shop near by, he does'nt really like the very thin stuff. Ian S C

[vile_fly]

Oh, by the way, if I were to change out and modify the bearing surfaces, I'd use shielded bearings because the sealed bearings drag too much. Shielded bearings spin much easier, and are still permanently lubed. Bearing #6200Z is 30mm o.d. and 10mm bore, and seems to almost fit inside the connecting rods, being 9mm wide.

[Ian S C]

For smaller engines its a good idea to flush out the grease with petrol, or similar, and replace it with light oil.
with sheilded bearings its possible to pop the shield out of one side that will be protected when installed, makes cleaning out the grease easy. If you get a sealed bearing, you can take the seals out and use it that way, if you'r like me, you use what you can get. Quite often bearings from old machinery are perfectly OK, even though the rest of the machine is clapped out. Thought I had some bearings, old skate board, but the bearings had .020" to .040" slop, even then I used two of them in a prototype engine, and they work sort of. Ian S C

[jimlarsen]

I have tried using skate bearings too. I found roller blades at a thrift store. My problem was that the bearings had so much dirt in them that they were almost too worn for my purpose. I was able to salvage 1 or 2 from the 8 bearings in the skates.

[Ian S C]

Hi Jim, my bearing supplier has 8mm bore bearings that he sells for replacement for skate boards etc., $ NZ5 each, at one time one of the chain stores had sets in the sports dept $nz 16 for 16 bearings, I did'nt get round to buying any, maybe they'll get more one day, they'd have to be pretty bad to be a waist of cash. One thing about bearings in NZ, metric is about half the price of imperial, so all new bearings I get are metric now. Ian SC

[vile_fly]

The bearings I am fond of are electric motor shielded bearings. From time to time, I get hold of a timing belt tensioner bearing. They are quite sturdy. Axle bearings are always loose enough to be of use, being oversized and easy to turn.
Am still machining the crankshaft. Got the front snout done, working on the counterweight and crankpin. Still debating on going oiless. I can make use of my graphite stash and make some graphite piston wristpins to avoid oil on the cylinder walls.
My worst habit is to keep changing the design as I go along. Gotta stop doing that or I will never get done.

[Ian S C]

Where I work from time to time we use cheap angle grinders, and wear them out fairly quickly, but the bearing at the comutator end is usually OK, the ones I get from there are 8mm bore, sealed race. The boss gets the grinders, then when they give up, I get the job of seeing if I can fix then to get a few more hours use, by the time i"v done that we could have bought a new grinder, but I only work there!
I use a number of bearings from old VHS vidio recorders, old printers and scanners, Old photo copiers can also be a good place to find gears bearings etc. Ian S C

[vile_fly]

Update:
Got the connecting rod bearings I will use. #6804zz is a deep groove double shielded ball bearing (20mm I.D. x 30mm O.D. x 7mm wide) that spins quite easily. As the picture shows, it is a perfect fit without me having to resize the connecting rod or reduce the crankpin size and strength too much. I won't be removing the shields, by the way. Just will drill the rods for tiny roll pins to keep the sucker from walking around under load.
The other pictures show how slowly I work on a project. Still have to finish the front snout with a keyway. I have to get some v-blocks and get some practice in cutting half-moon keyways before I can finish it.
The crank disc didn't fit on my lathe, so I used my shop's brake lathe to clean and true it up. It is marked for a 30mm stroke and 40mm stroke in case I have too much dead air space.
Not done with the cold cylinder head (the square thing) machining, by the way. Needs more holes for copper pipe fittings.

[onecycleDan1990]

Here's an example of what I think would be a very good place to start from. It is a prototype Vee twin similar in size to the compressor in question. I would recommend building something much like this and optimizing various dimensions.

Here is a link to the full article in pdf format: http://journals.tubitak.gov.tr/engineer ... -98073.pdf

A team headed by Halit Karabulut, working for Gazi University built a moderate size V twin Alpha engine. It is very simple mechanically, but as built put out less power than originally expected and required very tight tolerances for the pistons and bores. The tight tolerances were used in order to use no oil in the cylinder bores, as the engine was to be pressurized.
If the engine is not to be heavily pressurized than a larger clearance, oil control grooves, and oil lubrication will give less friction and be far easier to machine.
It used very long Heylandt crowns on both hot and cold pistons. these extended into their respective hot and cold caps.
Instead of complex fin or tube heat exchangers, the transfer pipe was connected to the bottom of the cylinders, entering BELOW the heylandt crowns. Air was thus forced through the annular gap between the piston crowns and cylinder extensions, making for a simple heat exchanger with quite a bit of surface area. This system was used by the old Rider pumping engines.

The engine was never optimized after initial testing and the mistake was made to pressurize the working space but not the crankcase.
Because of this the output was only 65 watts at 2.5 bar air.

An extreme example of what a high temperature high compression unpressurized alpha vee twin can do can be seen herehttp://www.youtube.com/watch?v=5pdqDQwehlk (sorry, but there is an ad in front of it) but do not expect anywhere near these results your first, or even fifth time. The output claimed in the youtube video probably twice or more its actual value, but it is still very impressive for its size.

[vile_fly]

Yup, I am familiar with this paper. I think the biggest mistake they made was the excessive stroke length. I tend to look at a stirling engine as a thermal bucket brigade. Keep the size of the bucket small so that it can work faster with less heat and less regenerator capacity...at least that's how I see it. Also, I think the pistons should be as wide as one can get away with (within reason) to exploit the pressure differentials with maximum efficiency. The higher surface area is a nice thermal bonus, too. I am still trying to keep this engine fairly modular in design so that anyone can use it as a guide to making a power producing stirling engine. I just hope I don't completely fall on my face. The large hollow crown design may be beyond my machining talents at this time, but I will not dismiss its potential.
That engine in the video is a nice Andy Ross design. I was outbid on the issue of the magazine that was dated 1982, so I still have no specifications on it. The regenerator looks interesting, though. Perhaps it is a labyrinth design, or something like it.
Oh, I intend to pressurize this one to about 150psi or so...including crankcase.

[Ian S C]

You could lose some weight on the con rod, by drilling a row of quite large holes in the web, proberbly fit six in the space, and if your having a dry sump, chop off the oil flinger. You could make it a bit narrower too. Its proberbly just me, I'm always looking to see where can remove surplus metal , Some times a bit too much. Ian S C

[vile_fly]

Narrowing the connecting rods is a decent idea. The webbing of the I-beam is so thin, I doubt it will help much with holes in it.

I redesigned the hot head. 55 holes for 3/16 brake line. Bubble-flared brake line will be trapped between 2 plates with 55 holes, possible using copper washers behind the flare and a 45 degree countersunk set of holes on the other plate to seal against the flare tip. Here is what I got done so far. I used 10mm triangle graph paper for the template.