canner wrote: Well in the time since I last wrote this lathe has been used to turn a thread, thank you so much for the help. I was struggling to hit a consistent od and id of parts and started looking at the cross slide. There is an extra 5 thou. of slop in the x axis then several years ago when I first set up the machine and so I will be pulling the cross slide of to see what is happening. I really don't think this is in the ball screw but we'll see.
The thrust bearings for the screw are the other likely place to look, but perhaps it is just a general "lathe thing"? Boring bars do tend to deflect on manual lathes, and there is no reason to expect a CNC lathe to be different.
If the same size final cut gives a different finished size on one component than another, though, it might be a sign of something slipping. Backlash or slop would be expected to be consistent, you can make consistent parts with a sloppy leadscrew on a manual lathe, after all.
Given that the machine is running with stepper motors I would be considering the possibility that the motors might be losing steps, perhaps due to over-aggressive acceleration numbers. This better fits the randomness you are hinting at. One way to check might be to home the machine and MDI a move to a particular diameter then measure a convenient distance between parts on the cross slide. Then run some parts, repeat the MDI move (in the same direction, ie manual lathe practice) and see if the distance measures the same. If it doesn't then I would very much suspect loss of steps or a slipping pulley rather than a backlash problem.
but the best I could tell there was a solid signal coming out that was equal on A, B, and Z no mater if the spindle was rotating or not. So I broke out a smaller encoder that I had originally purchased when I was planning on fixing the mpg's, sired it into the 7i76 and ran a test g33.