Thanks for all the comments.

Here's the picture for Bill et. al. I believe the shrink tubing that I am using is 1/16" that I probably got from Litchfield Station, although it should be available anywhere. The length of segment (thickness of donut) is about 1/16". It must be slightly less than the height of the coupler box cavity. As Jon described, the donuts are placed so that the donut axis is vertical. In the picture you can see that they are pretty squished even when the coupler is in the centered position. They have to be placed far enough to the left to leave space for the part of the cover that extends down around the pins. Sorry that the picture is a little fuzzy around the edges. Depth of field is thin at this mag.



Regarding Andy's comment, Frank certainly made an absolutely magnificent product. But I and I think many others found them to be so cantankerous to work with that we were ready to give up.  The possibility for them to be set off center does not add to my enjoyment. Self centering helps in most circumstances and for me it's the difference between go and no go. I'll deal with the coupling on a curve issue should I encounter it.

So use this idea if you want to. It does no harm. My next quest will be to find a similar technique for the standard shank version. Others have shown various techniques for SC there but I'm hoping that this simple donut method can be adapted.

George