i've got .064" tip grooves that were machined before the tips were straightened for true position. of course i'm having troubles with the parts running on the cmm. the plane circle path of the slots making most of the trouble...
i've tried to tie the groove circle plane target locations (x,y,z) and rotations (i,j,k) to previously probed features. PLN_512 (used for the Y location) is the back side of the tip (right side in the pic) and CYL_181 (used for the X and Z) is the adjacent cylinder (just to the left of PLN_512).
it doesnt seem to be working as well as i'd hoped. the probe still occasionally contacts the side of the slots when probing the circle planes for the slot both when approaching the part and when moving from one touch point to the next indicating the Y axis math is not being done and/or the circle path is not canted with the same IJK as PLN_512.
Auto-plane relies on your spacer value, even though you update the centroid and linear location it could allow for touches outside the desired location. (Current instance, hit in the fillet radius or just catch the outer edge of your diameter, maybe both if the features aren't actually concentric.)
You can have the points shift polar radial distance and linear distance easily with your pre-existing features, this would eliminate concern of touching an incorrect surface. You get to control how it would adjust.