Hand-loaded spacer blanks show different exposed lengths in our proposed FR10 lathe-loading study. The sheet records only stick-out, so I can't tell blank variation from seating variation. Automatic loading is not authorised. What should the manual measurement study record before this becomes a robot problem?
Use identified blanks with overall length and relevant seating features measured, then record repeated manual loading of the same blank under the approved setup. Keep the datum and measurement method consistent, alongside seating observations and any burr or contamination condition.
The original sheet used different blanks, not repeat loads of one. I've asked the setter and quality to build the comparison around identified samples, retaining the old readings as limited evidence.
Compare between-part geometry and within-part repeatability separately. A length difference may account for some projection variation without explaining all of it, and a seating observation needs its own evidence rather than being inferred from the remaining arithmetic.