We added a locator to the FR5 spacer-transfer bench and got a useful supervised run with one tray. A replacement tray brought back a missed pickup. I nearly posted fixed before the swap. What should I compare so the update explains the improvement without hiding its limit?
Compare the two trays against that locator before changing the program. Did the replacement actually reach the intended stops, or just look seated from the loading side?
It looked seated, but we removed it before checking the contacts. I cannot claim it reached every stop. The two tray references are now with the toolmaker for a static comparison.
Then that's the limit to write down. Thanks for being clear about it. A known good run and an unexplained miss are useful results; you don't need to invent the seating cause to make an update.