How do I stop these old clamped checks becoming our acceptance figures?

The photographs have part tags. Six rows match. Other pictures were renamed during an old folder move and I cannot link them reliably.

Start with the six, then. No need to guess the others into place.

Are those six all the same bracket revision? We have been caught by mixed retained samples.

Good question. Tags identify parts, but the spreadsheet has no drawing revision. I'll check against the work orders.

Two revisions. Four of the six belong to the one now proposed. The other two have a different gusset.

Keep the groups separate while you review them.

I ran into the mystery-correction version of this recently. Could you find what was actually changed on those four parts, or is 'corrected' all anyone wrote?

Also identify the reference used for each measurement; a tag connects a number to a part, but the photograph may still not show how that dimension was taken.

The measurement method is recorded for those four. Correction is just a tick box. We don't know what was done or whether the recorded value came before it.

To answer Thomas, they used the same inspection drawing reference. That helps with the dimensions, but doesn't fill the correction gap.

Any fixture repair history? Our inherited shim was compensating for a repaired mounting face.

There is a repair entry for one support. No dimensions attached. I'll ask maintenance what else survives before anyone copies the present packing pieces.

Can you photograph the current support arrangement cheaply while that search happens? It would give the designer something definite to work from.

Yes, maintenance has photographed it during the survey. We are treating the repair and packing as things to explain, not instructions for the new fixture.

That saved us an argument too, writing down 'we don't know why this is here' instead of giving a mystery strip the grand title of correction feature.

What will the new trial record for incoming fit-up? The old records start at the fixture, which may hide differences already present in the loose parts.

I'd include which component revisions went into each assembly as well, after finding those two gussets mixed into the old set.

Quality and the fixture designer are defining those checks. We have retained the four traceable cases as background, with their correction uncertainty stated. They won't become final acceptance figures for the new design.

Did maintenance find anything more about that repaired support? That's the loose end I'd still be curious about.