Low Temperature Differential Stirling Engine
(using found computer parts)
First James Senft wrote a book, [An Introduction to Low Temperature Differential Stirling Engines ISBN-13: 978-0965245517] detailing a smartly engineered engine using two pistons (type gamma). then I found a link via the Build Magazine® website to Viacheslav Slavinsky's engine using computer disk platters and I thought I can do that , and possibly make a few changes?. His solution was inspirational, but I wanted something that you could see the displacer moving. An Australian Steam model site also provided some good constructional info: A low temperature Stirling Engine by Penn Clower
If you want to construct a similar engine, here are some notes that may help you.
There are only a few main parts; no major tools needed to make it. One exception is a pillar drill or similar just to ensure that holes are drilled square to the surface. The most used hand tools are a hacksaw and a file. I have used adhesives where possible instead of screws.
What do you need?
The answer is, remarkably little: a few disk drives (hard disks and a CD-ROM), a 70mm diameter piece of foam, a teflon/ptfe desoldering tool nozzle, a CD, a short length of copper pipe, some silicon sealant, some adhesive for metal and a few machine nuts and screws.
First find your disk drives, you are likely to need at least two or three of them. I had access to a stack of old drives that were being trashed. Get yourself a T9 Torx drive, it fits "most" of the screws holding the drives together. Take off the electronics and then take the lid off. There are usually one or two screws that are hidden under the big sticky identification label Once the lid is off. You get to see the bits you want. The platter(s), the bearing in the arm assembly and if lucky the aluminium hub. The arm bearing will be used for the "crankshaft" and the top of the hub as a mounting for the CD/DVD flywheel.
I had to dismantle a couple of old 10Gb disks (see left) to get enough parts, there was only one platter in each. I did find some SCSI, drives that had up to six platters in them! The Barracuda drives shown had a hub that was just the right size to fit the hole in the CD/DVD "flywheel". Save the little screws, you can often reuse some of them.
| Geek info: I found that there were 2 platter thicknesses, 0.8mm and 1.25mm. They all have a standard diameter of 95mm** and a central hole 25mm. When there are multiple platters they are separated by aluminium rings; these are 31-33 mm diameter. I used the thicker platters, for the top and bottom, but I'm sure the thinner ones would work just as well. **I found a SCSI disk drive with 6 stacked platters which were 84mm diameter, I reckoned that these were too small to use. Why use a ptfe/teflon nozzle? See PTFE-The Most Slippery Substance in the World for details. |
The actual construction work starts with the displacer shaft. The shaft was from a CD/DVD drive. Drill a hole in one end. Flatten part of the end to make the drilling of the hole easier. This hole will be for the connection to one of the cranks, so ensure the the hole has no sharp or rough edges. The shaft will run in a gland that is made from a ptfe/teflon nozzle from a desoldering pump. My shaft was 3 mm. I drilled a 2.9mm hole in the nozzle . The fit on the shaft need to be smooth, but not loose. it is designed to be airtight, which providing the fit is good can be achieved, Putting the shaft in the drill and running it whilst sliding the nozzle back and forth, helped achieve this sliding fit.
The shaft is then fitted into a piece of wood dowel. The displacer is made from packing /building foam. I used some expanded polypropylene, but other closed cell foam is good. Make sure it is flat. Mine is nominally 10mm thick. It must not touch the walls, so 70mm maximum size is about right. Cut a hole in the foam for the dowel (using a cutter made from an old telescopic aerial). Glue in the shaft/dowel assembly. When set cut the foam to size. In that way, you will have a displacer that is concentric to the shaft.
Next prepare the platters. I planned six holes around the periphery for the screws that will hold it all together. I bought some 3mm x 30mm long countersink nylon screws (from RS components). They solve the double problem of keeping the platters and cylinder together, whilst maintaining thermal insulation. I planned to countersink the top holes, ie the nuts would be at the bottom, nominally out of sight. Tape two platters together for alignment and if possible use a third sacrificial platter at the back of the stack to ensure a better finish. There is only 6½mm of platter either side of the cylinder, so the drilling/countersinking of the holes needs to be accurate.
The platters have one obvious drawback, the 25mm centre hole that needs to be filled. The bottom one was done by simply gluing a disk cut from a spare platter. To stop the adhesive going where it shouldn't, I used masking tape. When filing the shape, create a temporary sandwich using a spacer ring either side of the piece of platter as a filing guide. Make another one using the same technique for the top
For the top one, I included one of the spacing rings between the platter and the disk. This extra height allows sufficient space for the back of the nozzle to fit above the displacer. Drill a ?? mm hole in the centre and insert the nozzle with a nut selected from the scraps bin to lock the nozzle in place. The nut will cut its thread into the nozzle. Glue the assembly to a spacer ring, but do not glue to the platter yet.
The cylinder is made from a piece of 82mm drain pipe, a size usually used on larger commercial buildings. Cut an 18mm high ring. I had already decided that I wanted to see into the chamber, so I cut three 10mm holes as viewing ports, nominally at 4, 6 and 8 o'clock. From a clear plastic juice bottle, I cut a strip that fits inside. Use adhesive around the holes to keep the two pieces airtight. Use the ahesive sparingly to avoid it squeezing out into the "viewing" ports. See picture for my fixing technique. The cylinder must have a seal with the platters. The one on the top is permanent, but the bottom seal is done by placing a thin bead of silicon sealant around the bottom edge, allow it to dry/harden before proceeding.
Invert the cylinder assembly. Place the spacer ring/gland top assembly on the underside. Fit the displacer and shaft through the nozzle. Rotate the displacer to ensure it does not foul the cylinder. Use some pieces of masking tape to hold the components in the correct place. Apply adhesive to the back of the spacer ring/gland assembly and adjust to achieve the same concentricity as before.
Next I considered the central support that will hold the bearing. This will have the CD/DVD flywheel attached on one side. From this bearing are connections to both the displacer and the piston.
One of the other disk drives I salvaged, was build onto a very simple aluminium base. This provided the material from which I cut the upstand support that will house the "crankshaft" bearing. Under the long sides there are strengthening ribs. The face and edge were square, minimising any metal bashing needed to get the upstand to be at right angles to the platter. First I drilled out the hole for the bearing. I had to open this up carefully with a file, because the bearing was slightly over 11mm diameter. Using a hacksaw I cut the basic template and then used files to refine the shape
I put in small machine screw to hold the bearing in place to allow fine adjustment of its position.
The tricky bit was drilling the holes in the top platter and upstand that will hold the two together. These need to be countersunk on the underside of the platter to have the minimum of interference with the displacer. I started with a small hole, (1.8mm) tapping size for the screw I used (2mm). I used two screws.
The "power" cylinder fits on the other side of the upstand. Cut the copper pipe and remove any sharp edges or burrs. I made the piston by casting it in the tube from epoxy, I coated the inside of the tube in graphite dust in an attempt to reduce the friction. The trick is to get a good cylinder to piston fit, without the whole thing forming into one big lump! I aimed for a piston about 10mm long. When it was set, I hollowed out the inside, to reduce the reciprocating mass. Finally I drilled a hole for a shaft (from the CD drive). The shaft is reduced in length to ?? mm. before gluing it in place.
I used ordinary metal nuts on the nylon screws, do not overtighten them.
