In the absence of any discernible datum on the print and no knowledge of the function, use the longest/largest or most tightly toleranced feature as the primary, then the next most for the secondary.
Normally I work with iterative alignments, so I think I'm gonna need to break this down even more.
I can use the two main faces on the inside of the pin as the origin and align. We want to check that those two faces are good to the notch on the shaft closer to the bottom... so 3 planes to each other.
the way I would go about it is thus.... if you can get your probe down in the slot. get the plane on the top of the part and the two planes on the inside of your slot plus the circle around the top plane. Create two lines at the intersection of your top plane and your slot planes. Ensure both lines are pointing in the same direction. create two points using the pierce function with your lines and circle. Create a line between your two points then do an alignment using plane-line-line. That's all IF you can get your probe in the slot. If not then I'm not sure. Been trying to figure out how to align to cylinders myself.
We measure cylinders, albeit large ones. If a cylinder is mounted on the CMM table oriented with the 'butt' toward the operator and the 'front' toward the back of the machine, from memory I believe we align in the following way -
Measure cylinder and level to it. If the cylinder is measured in the direction of the operator to the back of the machine, level to Y-. I don't believe any rotation is necessary in such a case. Origin Z and X to the cylinder. Make a point on the butt end to serve as the Y origin.
This works for us, although we are simply checking these cylinders for conformance of diameter.
Hope that helps. We pretty much have to figure things out for ourselves, having had very minimal training.
