Alignment Rotation Question

That would probably work, it seems fine. I was trained to be sure my alignment used actual hit data for each X, Y, and Z depending on which I am rotating to. So in this case your Z data is all you have (assuming Y workplane, and tab is on right side of part). I would also take a third hit in X relative to the Z of that midpoint created by those first 2 hits so that I have X data.

I made a very crude drawing in paint. Assuming slot is on the right of the part. Viewing in Y workplane.

^^^^ Like that.

However, if the slot was somehow machined off-center to your datum A axis, such as in a lathe with live tooling, or in a mill with an inept operator... That will be difficult to detect on a small part with a CMM, and also give you a less accurate rotation line.

Adding a second alignment to refine things would also help. (repeat the datum alignment again) 

When I was first trained here on the CMM I was taught to repeat my first alignment in all my programs to refine the data as you said above. It's very helpful.

Sorry William, can you please explain a bit more? I'm honestly lost.

Here is a picture of my part. Sorry, I meant to include this picture in my original post. So far, my plan is to physically clock to the slot as best as I can on the CMM grid plate.

Obviously, I will level to the OD of the part. After that, I'm not sure how to go about this..!

Usually with something like this, I try to measure a couple features and construct intersection points, like this:

Then, I would construct a midpoint of the 2 intersection points.

However, in this case, I don't think I can construct those 2 intersection points...

I am not entirely sure what I am looking at, but based on your second picture. Are you able to make a line at at each side of the slot at the same depth your origin circle is measured, and do a pierce/intersect point where that line pierces or intersects the circle?

I would think that you could measure a line on each side of the slot and then construct a mid line. then i would rotate to the mid line. If you rotate from the center of -A- you will not get accurate results as to where the actual location of the slot would be.

William, that's exactly what I was thinking. That's what I was referring to with this picture. I tried it, and the pierce point comes out on the other side of the part Even if I try reversing the line. Maybe it would work regardless of where it is piercing, but it's a bit concerning. Any other ideas? I wish it was possible to intersect a plane and a cylinder OD to create an intersection line...

William Johnson, another poster, just recommended lines on each side of the slot, a midline, and rotating to that midline. I could do that, but I am concerned about accuracy. This slot is pretty short.

William, yes, that is the default approach. However, I'm concerned that this slot is a little too short to use as a sole rotation feature. Any slight measurement inaccuracy, for example, and the rotation of my coordinate system will be off. I think my program would be slightly more repeatable if I somehow connected the slot with the OD. Do you agree? I might be wrong on this, but this is the general approach I have taken over the years.

That is what your dimension requires you to do. If you use anything else you will be getting false information. Do you have any vision systems? If you do that would be a easier way to check it. Or at lease verify what the CMM is giving you..

You need to make sure that the lines that you use for the midline are aligned to prior to measuring the lines. This way you ensure that the points taken are always in the same location on every part based on the actual location that they are at.

so I would measure those lines 2x aligning on the 1st ones and using the 2nd ones for midline construction and rotation.

yes it is a short line but it is all you have to use. You could also build a hard gage to check it.