[rs4race]

I received a new ball screw in the mail today. I was wondering what kind of tolerance or runout I should expect for the machined end of the ball screw.

My current screw seems pretty bad. Its a 1605 single nut, it has a couple bends and seems that the machined end isn't eccentric with the rest of the screw. The backlash was ok at .003, but thats double my other axis even with repacking the nut.

This new ball screw and nut is a 2005. It has machined ends with 15mm diameter for the bearings. I clamped the ballnut in a vise and set up a dial indicator on the bearing section of the machined end. Rotating the screw I measure +- .001 in runout with the ball nut placed near the machined end. This is a lot better than my previous screw, but I'm not sure how this compares to a screw from one of the typical kits you would find.

I attempted to measure backlash in the same configuration but the screw wants to turn. Trying to restrict its turning will show less than half thou backlash, but I think it will be a more accurate measurement mounted on the mill.

Does anyone have any information on what acceptable tolerances are? I looked into this ball screw because of slow max speeds, and relatively high backlash. I think the backlash is going to be good and I would guess the machined end is fine, but if it is actually too much I want to get it fixed before I install it.

[AVRnj]

To me, the .001 runout seems great. I would think it would be hard to detect that at all in use. A half a thou of backlash is great, if you can keep the AC bearing backlash below a half a thou as well, you will be under one thou of backlash, which is IMO a great number and tough to consistently beat. My understanding is getting below a thou of backlash, consistently, is very difficult on these hobby machines.

[Fastest1]

.003 is high backlash?