Our brass sleeve samples were loaded by hand and show different exposed lengths. The sheet contains no separate stock dimensions or repeated-seat readings. Automatic loading is not authorised. What observations would distinguish blank variation from seating before either becomes an assumed FR5 correction?
Name the face used for the exposure measurement first. If two people measured from convenient but different edges, you could spend a lot of time explaining variation the sheet created for you.
The setter confirms the same chuck face was used. Individual blank identities were not recorded, so the old readings cannot be paired reliably with separately measured samples now.
Make a new identified manual comparison. Measure each blank's length and end condition, then record repeated seating against the intended stop using the same exposure reference. Keep within-blank variation separate from differences between blanks
I'd have the person doing that work review the sheet before starting. We've managed to collect excellent numbers that nobody could connect to the same piece later, simply because the identity box was on a separate page.
Will the repeated-seat comparison preserve the same loading and measurement method, Dinesh? Otherwise operator or gauge changes could enter the seating result unnoticed.
Yes, Imran. Setter and inspection agreed one method, with blank identity on every row. The first new comparison finds different overall lengths but consistent repeat seating for each of the identified samples.
Do those lengths explain the exposed-length differences against the defined seating reference, and do they remain within the required stock range? Repetition alone does not answer either question
They explain the differences in this sample group, and all measured lengths are within the agreed stock range. Ends and contact surfaces were recorded as sound. We did not recreate the old unpaired readings.
Then report the new group, not a diagnosis of every old entry. A clean explanation for identified samples doesn't give the earlier anonymous measurements identities after the fact.
Who receives the permitted stock range for the loading proposal? It needs to reach the person assessing seating and grip, not remain an interesting inspection spreadsheet nobody uses.
The integrator and toolmaker have the range and identified repeat-seat evidence. Their automatic loading and confirmation assessment is still pending. The manual study is useful input, not permission to transfer its repeatability claim to the robot.
Thank you for separating those results. Keeping the original sheet with its limitations will also stop someone treating the new measurements as corrections to data you could not pair.