I’m clocking 0.04 mm TIR at 35k RPM on our lab micromotor with a 1.6 mm test pin, and it’s showing up as ragged zirconia finish lines and uneven die contact. After ultrasonic collet cleaning, fresh O-rings, and correct bur torque I’m only down to 0.02 mm — what maintenance step or collet spec has put you under 0.01 mm and stabilized device fabrication accuracy?
And replacing the collet and nut as a matched pair and ‘blue the taper’ before install dropped my runout from about 0.02 to 0.006 mm; I lightly lap the spindle taper, then clock the collet to the spindle’s low spot and mark it. The Nakanishi/NSK 1.6 mm h6 precision collet (≤5 µm TIR) plus a class ZZ 1.600 gauge pin for verification has been the most stable combo for clean zirconia margins.
What finally got me under 0.01 mm was controlling bur “stick‑out”: with the 1.6 mm pin I seat it until it just kisses bottom, back off about 0.2 mm, then torque — if it’s fully bottomed, the collet’s draw at 35k nudges it and shows up as runout on zirconia margins. Small caveat: I always clock TIR after a 2–3 min warm‑up so the bearings are at temp. @OP do you see your 0.02 mm change cold vs warm?
At 35k, redoing front bearing preload took me from ‘0.02’ to 0.007. Clock spindle without collet?
Quick check: warm the motor 3–5 minutes and clock it with a short solid‑carbide test blank (ISO 1797 Type 3) instead of a long steel pin, then ‘match‑mark’ the collet to the spindle at the lowest reading and stick with an AA‑grade collet spec (≤3 µm) — any oil on the taper will spike it faster than coffee jitters. @chloe_b97’s stick‑out tip pairs well with that. What collet brand are you on?