New power piston and displacer for 3D printed Stirling engine

To simplify the building of the 3D printed PE 2 Stirling engine I’ve designed and tested two modifications. The parts were printed on a Stratasys printer. Neither of these modifications is required, nor will they alter the performance. If you have built or are building the original design there is no need to change. If you haven’t built those parts you may find the modified parts save a little time and effort. The STL files for the parts are available at thingiverse.

the new power piston

The power piston design uses a redesigned power piston insert with an O-ring that both seals and secures the power piston insert to the aluminum sleeve. This makes the machining of the aluminum part of the power piston a simple and relatively quick operation. The 0.875 inch diameter x 0.065 inch wall aluminum tube only has to be turned to the proper diameter for an easy slip fit in the brass power cylinder, and then cut to a 1.00 inch length.

The AS568A-115 O-ring (11/16 ID, 7/8 OD, 3/32 width) secures the piston insert in the aluminum sleeve for normal operation, yet it can be pushed off if necessary for connecting rod removal. Most of the internal gas pressure on the power piston reacts directly against the piston insert so there is very little force on the aluminum sleeve. This design is quite secure on an engine like this with gas pressures of less than 1 psi and cycle speeds below 500 rpm. For an engine using very high rpm or a gas pressure of more than a few psi, you would want to use a more secure method of holding the sleeve stationary with respect to the piston insert.

This modification uses the same power piston connecting rod so there are no changes to any of the other 3D printed parts.

The new displacer

3D printed Stirling engine displacer parts

The new displacer uses two 3D printed parts instead of one part as in the previous design. The benefit of the new design is that it eliminates the four foam disks that were required on the previous displacer. Two holes must be drilled and one must be tapped for a set screw. After that the two pieces must be glued together with an air-tight seal.

The displacer needs to be stiff enough not to flex due to the pressure changes in the engine, yet have low mass so that it doesn’t add a lot of reciprocating mass to the engine. The new displacer including the shaft weighs 21 grams, one gram heavier than the old displacer.

The design uses small diameter columns to transfer the tension and compression forces between the two displacer faces when operating. It’s important that each of these columns be glued to the thin disk that seals the displacer. It’s also very important that the glue joint that seals the thin disk to the main part of the displacer be absolutely air tight. Leaks will drastically reduce the engine performance. I know this from firsthand experience.

Before you glue the two pieces together, you need to perform the drilling and tapping operations.

First, you need to drill out he displacer shaft hole to .063 inch diameter. The hole should be kept perpendicular to the bottom (or top) of the displacer.

Second, you need to drill the set screw hole for tapping with 4-40 threads. The set screw hole is oriented 45 degrees to the displacer shaft. This is most easily done by securing the main displacer piece at a 45 degree angle and drilling from what will be the inside of the displacer when it is finished. Holding the displacer at an angle lets you start the drill on a surface that is perpendicular to the hole. The photo shows my setup.

After tap drilling the hole you can tap it from the same setup. Your tap may not be long enough to cut full threads out other side. If that is the case you should be able to open up the clearance hole on the other side far enough to get to the full threads. You want to be sure to test the displacer shaft and set screw before you glue the cover on and lose access to the inside.

I am not concerned about very small leaks along the set screw threads or the displacer shaft. These should be extremely small and will help equalize pressure if you travel through different altitudes.

glue problems

To glue the two pieces together I used plastic model airplane glue. You want to put a generous drop of glue on the top of each column and then run a generous bead of glue around the edge where the disk seals the displacer. After applying the glue I pressed on the disk to seal the displacer, turned it upside down on a flat surface with wax paper, and weighted the displacer to get what I hoped would be a good glue joint.

I left the glue to dry overnight. The next day I assembled the engine and tested it. The engine would barely run. I decided that a leak was probably the only thing that would make it work so poorly. I ran another bead of glue around the seam and tested it again. This time it worked better for a while then again barely worked.

To make sure where the leak was I removed the set screw and put a small brass tube with a hose in the hole, sealed it to the displacer with plasticine clay, and then blew through the hose with the displacer partly under water. It was still leaking at the seam. I put a little more glue in the problem area. Again the engine worked for a while and then barely ran.

the acetone treatment

I decided the glue was not going to fix the problem. I used a metal pick to scratch most of the glue out of the part of the joint that was visible. After that I used a small amount of acetone with a cotton swab to go around the joint several times. The acetone will slightly dissolve the ABS material and then as the acetone evaporates the ABS solidifies again, providing a sealing/gluing action. This cotton swab and acetone method can also be used to help smooth surfaces on 3D printing parts. If you use this method be sure to only use the acetone in a well-ventilated area with no ignition sources (flames, sparks etc). Only a small amount of acetone is required.

After the acetone treatment the new displacer finally performed identically to the original displacer. I’ve run the engine with the hot plate at 170 degF to make sure it would not have high-temperature problems. I’m telling the full story of my experience so that anyone wanting to use this new displacer design is fully aware of the potential problems. The particular glue I used (Testors cement for plastic models) does not appear to make a good bond to the ABS plastic.

The original displacer design using foam may be more trouble to make, but you will not have any leak problems. I’m concerned that a person building this engine the first time with a leaky displacer would not have a good way to judge whether the displacer was working properly or not. I think if you are careful gluing the seam you should not have a problem. Do use plenty of glue and don’t use a glue that might dry too quickly before you have assembled the pieces. I think I would test epoxy (or another adhesive) to verify it bonds well with the ABS and us it next time.