Hi all,

First post here, and I've got something happening here that I've never seen before, so I thought I'd see what you all had to say!

What I need to draw up & tolerance accurately before getting them made, is a series of leadscrew assemblies - they are M5x1.0 (wierd I know, but thats the old design spec) 316 stainless leadscrews of ~40mm length, and the mating PB102 phosphor bronze nuts. These need to be very precise fits, with minimal backlash & wobble.

Currently the leadscrews are screwcut oversize with a modified 1.0 pitch insert (for extra clearance), and the nuts are tapped 4H, before the whole assembly in run in mechanically to obtain a nice tight & smooth fit.

Yesterday I went to see our CNC turner to make some sample on his Star sliding head, in order to calculate accurate tolerances for fits, in order to get gauges made, and make the parts interchangable and avoid matched sets of screws/nuts. But, when screw cutting the nuts, we ran into a wierd problem that none of us had seen before.

What we were getting, in a nut 9.5 dia by 16mm long, was a good fit on the gauge at either end of the nut, but a tight fit in the middle. I estimate the middle was 5-10microns smaller than the ends. We tried cutting the thread before and after turning the OD, and also tried cutting it in the rear spindle afetr parting-off, but all the results were the same.

Now we've all seen single direction tapers on threads, which can be corrected in the G-code command, but this seems to start correctly, shrink in the middle, and then come back onto size - has anybody ever come across this problem before, or have any thoughts on it, please? Or might it even be an issue with M5x1.0 in general??

We only have about a 20micron tolerance (estimated) on the effective dia in the nut, so to loose 10 microns in the middle is causing headaches...

Any thoughts, suggestions, or advice would be gratefully recieved!!