Good info, thanks! I'll copy that for future reference.
My "math is hard" comment mostly comes from the backlash Mattel received a handful of years ago when they released a talking Barbie, and that was actually one of her preprogrammed phrases! Cue female STEM student shrieking...
No, I actually got it right on. I got the bolt center measurement by threading the bolts in until they bottomed, tightening them square in their bores. I used the corners of their hex heads to scribe lines across their heads to find and strike a center in each head. I used the caliper to measure between centers and verify each to each. That's how I got my equilateral side measurement.
I took this, divided one side in half to create a right triangle, and used the good old Pythagorean theorem to get a center line. The epicenter where each crosses each is 2/3c, so that gave me my bolt circle radius. I subtracted the register hub diameter from the calculated bolt circle diameter and divided by 2 to get the difference movement at the cross slide.
I used a trepanning tool to indicate off the register hub of the back plate, then moved out that cross slide distance and inscribed a mark there. I then punched a mark at an arbitrary point on that circle, used the dividers to mark my other two points from that point, and punched those as well. But problem- the distance between points? A and B and points A and C were the same, the distance between points B and C was different from the other two! Slightly longer.
Oddly enough, what hamstrung me was not taking into consideration the fact that the tip of the tool was 0.03125". This drew the tip of my center punch out away from my actual proposed bolt circle. Had I remembered this, my marks would have been right on. As it was, I went back to my calculations and used the difference in calculated bolt circle vs register hub diameter to measure out from the hub and discover my mistake.
Okay, so now! Got my new marks made and punched, they are ACTUALLY equidistant at my actual bolt distance, time to drill. Went to the drill press, chucked up a center drill, and ... wait. Why does it look like it's wobbling? Put a dial gauge on it and discovered my chuck is THIRTY-SEVEN THOUSANDTHS OUT OF TRUE. Holy crow! Well, what do you want from a 1970s Craftsman press that was made for drilling wood... yeah, that weird one with the 1/8" belt and the 850 RPM "slow" speed. Use what ya got, except the chuck is pinned in place instead of being either threaded or Morse tapered... and the point at which it's out of true is right AT the roll pin, so it was drilled off square from the factory. Balls, there's no fixing that without a lot of surgery.
Fortunately, a neighbor has a Shop Fox benchtop mill he is generous enough to allow me to use. It may not be a Bridgeport, but it should be precise and square *enough* to let me drill my three holes with reasonable precision so I don't have to try to make my own Adjust-tru chuck.
So that's where I'm at. Plate is done, holes are marked, all I lack is drilling them so I can finally mount that chuck. Frustrating! So close.
