Thanks @TorqueMaster. I ended up using my mill to open up the U. Was a simple operation. Plunk it in the vise, indicate it and mill the channel to 14.25mm. Actually it was a lot easier than re-drilling out the holes which I did for the earlier version. The good news is everything fit on the read head, and on the lathe. I have RTV'd the end cap on the glass scale, and will mount the X DRO scale on my lathe tomorrow. Then I have to do the Z scale, but I expect that one will be a standard install.It would have been nice to have it come out right to begin with, but seeing as I had already sunk in a lot of time making multiple prints, I just wanted it done. It's interesting, no, its more of an irritation, that one cannot constrain the slicer to maintain a dimension. I tried using variable slicing and it really didn't work. Actually I was very disappointed with the variable height slicing, as the slicer insisted on using ridiculously thin layers which didn't print that well, rather than simply adjusting a few of the layers to be thicker or thinner to hit the dimension. That part of the slicer needs more work, in my opinion.
Technically, no. But I did measure a piece that I printed. The piece should have been 286mm long, but measured to be 285mm. However, that piece was printed on a diagonal, so hard to say what each axis error should be. Do I simply print a cube for this test? I'm still pretty green at 3d printing, so I have a lot to learn yet.