That’s an interesting video on reducing the self-feed speed and is well worth looking at. The only issue with the WM180 is that it does not have a reverse tumbler for the lead screw. The introduction of an additional gear reduction pair rotates the lead screw the wrong way, needs to be thought about. I tried to work out an additional reduction that preserved the correct rotation, but it was very complicated, so I didn’t pursue it any further.
The real issue here is not so much the actual speed of the self-feed, but the surface finish that it produces. The slowest self-feed with a tool with, say, a tip radius of 0.04mm will produce a surface finish that is visually good and sufficient for most non running surfaces.
For a running surface in a bearing, this may not be good enough. The asperities in the surface may be great enough to break through the oil film thickness, leading to premature wear and probably eventual failure. There is the option of increasing the radius of the cutting edge to reduce the height of the peaks, or as we probably all have done, polish the surface with a strip of abrasive cloth.
In one of the books I have on Stirling Engines, the author has included drawing for internal and external laps for fine finishing the machined surfaces. Another option to consider.
The ELS idea is the ultimate solution provided that it produces the surface finish that you need. Lots to think about. Just consider what you are really after.
Neil