green tick after an undocumented trip to the press

Final inspection, according to the inspector. But the failed reading before the press isn't on these cards. Only 'upright low'.

Can the inspector still find the final readings, even if the earlier ones are gone?

Rosa, was that a software import problem or did somebody deliberately keep only the last reading?

Cards are originals. Inspector found final results in a separate book. They link by batch, not individual bracket. No magical missing column, unfortunately.

Luca, the export collapsed them. Nobody had asked it to preserve the distinction, so it looked tidier than the job actually was. We kept the original readings and correction note together once we found them.

Could the old operator show which feature they corrected on a retained bracket? I would ask that before trying to infer the correction from the final measurements alone.

I'd try that too, though I wouldn't ask them to recreate the pressing from memory just to explain it

Press work is missing from the estimate, Oliver. Ada, there are retained brackets. Operator is available to point out the feature, not demonstrate a correction.

A marked photo would help the software and planning people as well. 'Upright low' could be a height or an angle to someone who has never held the bracket.

Operator pointed to the upright angle. They used the press after the bracket had cooled. Final book measures that angle from the mounting face, but the fixture check just says against stop.

Was the same drawing revision used for the final book and the new proposal?

Revision F in both. Found the reference on the book's cover. Fixture drawing is older and doesn't identify which bracket revision it was checked against.

That leaves a fixture comparison and a missing process cost. I'd keep both visible. Otherwise everyone solves the drawing question and forgets the press again.

I would still use the final book to understand what they accepted. It cannot tell you the amount of correction, but throwing all the history away would lose a useful starting point.

Agreed on keeping it, Ada, though mixed batches worry me if anyone starts calculating an average correction from those numbers

No correction average. Keeping the final acceptance history and operator's feature photo. New trial needs paired measurements. Planning has put the press work back as an unresolved cost.

Thanks for coming back with the costing bit. Usually that's where the thread goes quiet.