Both trays retained. Pocket identified in the replacement. I did not record comparable run counts before the locator, so I can describe an impression, not a reduced failure rate.

Good distinction. Did anyone check that the replacement seated against the locator, rather than just looking aligned from above?

And record the sleeve in that pocket if it's retained; the tray change may not be the only changed condition.

Preserve both trays as found before adjustment. The tooling team can compare their locating features and seating under the proper isolated inspection arrangement, without turning the comparison into a new production run.

Also keep the selected program with the observation. If offsets changed between those runs, the report needs to say so rather than assign the whole difference to the new locator.

Same recorded program in the short trial. No seating check was made at the replacement. The sleeve was not retained, so that part comparison is unavailable for the miss.

Then don't infer that the replacement tray caused it merely because the miss followed the change.

You can still inspect the retained trays for a useful difference. That would inform the next controlled comparison, not recreate the missing sleeve or prove the cause of an event you did not observe closely enough.

Tooling found a locating recess difference between the trays; the replacement does not engage the new locator as the first one does. It is not an interchangeable tray for this arrangement.

Has that compatibility restriction reached the people choosing trays, not just the trial notes?

Keep the miss's cause qualified even with that finding. The missing sleeve evidence still exists as a gap, though the tray incompatibility needs action on its own.

Replacement tray held out and its reference marked incompatible with this locator. The report names the recess difference but does not claim that it explains every pickup issue.

Define the eligible tray condition and the replacement check before the next supervised run. Otherwise the first tray becomes a special case that only the original trial team can recognise.

Have the loader demonstrate that recognition from the proposed check, without someone prompting the intended answer. That can expose an unclear label before another tray reaches the bench.

That check is drafted, not tried with the loader yet. Thanks, Lucy; the tray reference is now on the hold note, so it won't return just because the outside looks familiar.

Then the locator remains a limited trial result, with a real compatibility finding and verification still to do.