I've been through this with a misconstructed meade 8 (crooked rear cell epoxied on, resulting in a crooked baffle tube), and a large Celestron 11 where all the original markings had been misaligned, in addition, the more recent factory addition of placement of the corrector plate over the baffle tube was gone.
The marking on the secondary mirror was, in part, made after the factory rotated the corrector and secondary assembly, to center the secondary mirror over the baffle tube. If the baffle tube was properly centered and parallel to the tube walls, little adjustment or intentional off-centering of the secondary was needed. If the baffle tube is crooked, the factory selected, apparently, a secondary glued off center to the puck. This way, when the corrector was rotated the secondary moved eccentrically, allowing them to line it up with the baffle tube and, if the primary was glued properly (yours wasn't, yikes), lining it up with the optical axis of the primary.
My Meade had the baffle tube 8 mm or so off center at the corrector end, and the secondary mirror was glued two to three centimeters off center from the aluminum puck. A mess to be sure. By shifting the corrector plate all the way over to one side, getting a slight realignment of the baffle tube, and rotating the secondary over to the same side, the secondary is now aligned over the baffle tube. I started out with horrible coma even when collimated and now have a pretty good performer that will improve once the terribly blistered and discolored secondary is recoated. Silvered secondaries, another failed experiment by the manufacturer.
So the short answer is, don't despair, in reshimming the primary you did the hardest part of the job!
If you, as I would anticipate, carefully centered the secondary on the secondary puck, and equally carefully centered your threaded hole triangle, the secondary should not move eccentrically when the corrector is rotated, and you should be able to simply center the corrector perforation over the baffe tube to complete the opto-mechanical alignment.
When it comes to your particular situation, don't think about aligning things radially, think about aligning the baffle tube axis with the center of the secondary. You may get comments or see comments that the spherical secondary has no optical axis, but the fact that the colliimation pivot is very distant from the center of curvature means you can treat it as if it does have one.
However you do it, centering the secondary mirror itself over the baffle tube should be your goal.
If this means the corrector plate appears uncentered in the cell, that isn't a problem since the goal is not centering it in the tube, but rather, centering it over the baffle axis. The Celestrons Edge HDs actually have set screws to more easily align the corrector 'off center'. also allowing it to be cleaned and restored without losing alignment.
On mine, I completed this step with a holographic bullseye projection, but there are other ways to do it. The damaged secondary surface ironically helped see the pattern.
On the 11, the secondary seems to have been glued concentrically, so I simply centered the corrector perforation over the baffle tube axis with the holographic projection. Once I get a 3D printer I'll make a slip on mask to allow me to physically center the secondary itself, but performance at this point is extremely good. When I first got the 11 it was virtually unusable with horrible soft images. The mirror said itself gives an excellent Ronchi test. It's performance now is either optimal, or close to it.
Edited by markb, 03 March 2021 - 03:38 PM.