And ask the loader how the bracket gets into it. A guide may look redundant once seated while still doing useful work during loading.

The drawing calls every pad support, with no distinction between guides and locating contacts; I have asked the designer to mark their intended roles.

That's not enough definition to approve a stronger clamp. Does the inspection datum scheme appear on their drawing at all?

Bruno, the role drawing matters, but I'd still preserve the condition of the actual bench samples. We could otherwise blame the design for one damaged face.

Inspection is checking the samples; I am keeping the seating observations separate from its dimensional results so one does not substitute for the other.

Agreed. I was objecting to approving the clamp, not claiming we'd proved conflicting contacts.

For Clara Carter's comparison, keep the loading direction recorded too; don't make the operator improvise a new approach on every repeat.

The same bracket repeats consistently, but another sits on a different combination of pads; inspection found both within their checked drawing requirements, and the designer is reviewing those contacts.

That points the next review towards how permitted part variation meets the nest. It doesn't mean every within-tolerance bracket has been demonstrated. Keep the sample identities with the contact observations.

Has the designer proposed a contact change, or are they still asking for a harder clamp?

They are separating the guide function from final location in a revised proposal; thanks Yasmin, I nearly asked them to delete a guide the loader actually needs.

Glad it stayed in the discussion. I'd have the loader review the revised approach before the bench comparison, then let the comparison decide whether seating is actually improved.