Involute cutters are a compromise on shape so if the pinion was hobbed then you won't get an exact match, also may be a modified form due to the small number of teeth.

#1 cutter (out of 8 cutters in the involute cutter set, PA 20 degrees) is for 12-13 teeth, so the 10 tooth pinion cannot be formed accurately with an involute gear cutter. There probably is slight undercutting on the tooth form.

The 10 tooth pinion gear was most likely hobbed.

Andrew

Edited By mahgnia on 02/01/2018 10:06:42

Edited By mahgnia on 02/01/2018 10:14:57

I made a new quill out of free cutting steel, 40mm to house tapered bearings, for the Uniamt, and like you, noticed the issue you describe, I used a horizontal cutter holder and ground the tip to the correct profile by eye and feel, I then used a Chester milling machine to hold the new quill whilst I cut the new rack, unfortunately I'm going to have to redo one day, as the Chinese seem to have used an imperial gear in THEIR quill rack, which approximates metric, but not anywhere near the accuracy I was working to, 0.001mm, over the length of the new quill (I made a new alloy vertical head to accomodate the new quill, which has more feed depth) . Moral, never trust a cheap Chinese fine vertical feed or quill !!

i done a cnc conversation.

i had the same problemes.

my solution:

grind a flycutter insert and cut the 10teeth gear or cut with modul 1 gearcutter no1 12 teeth gear. put your stepper to 1/8 microstepping and good.

but!

you can buy a dovetail-profil.

you need only a adapter to mount the dovetail profil column instead of the original. then you get more then 25mm way in Z.

Would a ballscrew be more accurate (less backlash) than a trapezoidal spindle with a (spring loaded) anti-backlash nut?