I thought I would give my two cents on how I align cylindrical parts in a rotary table, in hopes that those in the niche market can get something from it, and perhaps share their ideas so we can all have new or different ideas.
I see that there is some talk about how alignments work, and definitely some good examples, but there's something that every operator or programmer, hell even every person, should keep in mind:
There are no perfect examples that will work every time in every scenario. There is always going to be some work one's end to do whatever it is they're trying to do. You can get a perfect CAD model, it is unlikely to get a perfect part, let alone an entirely perfect environment and absolutely repeatable-to-the-umpteenth-decimal measurements.
It's science.
However, there are ways to compensate.
First and foremost, most important, can't stress it enough, there aren't enough bold letters and asterisks to signify the relevance of this:
Alignment is really ******* important
Now that the initial yelling is over, let's move on....
-------------------------
Essentially, aligning a part using the rotary table is
terrible in that it is certainly different from, say, laying a flat part on the table or using fixtures in your alignment.
There's not a whole lot grabbing onto the part.
So there is some tilt, one way or the next. Okay cool that's fine, I'll just take a cylinder on the OD and level to it. Now the tilt is settled. Right?
Sure, sure.
Actually, negative. That's all well and good, but say you can't take a perfect cylinder because of some wierd way your part is cut. That's easy, just take a couple circles and construct a cylinder to level to. Done?
Not quite.
Not only is your part aligned like this, but when you rotate with the rotary, the rotation is a bit funky. You can get things straightened pretty good, say a notch upright or whatever, but rotating any degrees will move that tilt by however much. Let me explain.
If you put in the part at 0° and it is tilted slightly downward, then rotate 90°, the part will now be tilted slightly sideways. That's a whole new part as far as the CMM is concerned.
If I rotate, I usually create a cylinder at each degree of rotation useful to me (which is usually increments of 30° or 45°) and, when measuring a feature at that rotation, align to the corresponding cylinder.
It's typically useful to use the same naming convention for similar features at different rotations, e.g. OD_45, OD_225, Notch_XX_XXX_225, Notch_XX_XXX_45, etc.
-------------
I haven't even gotten into the nitty gritty of things, but please, share your ideas or thought process and help me (and you, collectively
us) gain a better understanding of rotary/cylindrical alignments!
OK,thanks, i understand what you mean.
i do save my registry settings as it took me ages to get them how i wanted & i don't want to risk losing the layout.
is there any other solution to this?