I don't know why this is happening—I don't think it's an artifact of the changes made in #585, but I suppose that could be the case. I think this behavior has been observed before, so I figure we should document this here.
Since this occurs really inconsistently (not sure if this means anything, but I wasn't able to reproduce this after adding a single debug console.log() statement to the top of PlateViewer.modifyWell()), my guess is that this is some sort of race condition—perhaps between the calls to PlateViewer.updateUnknownsAndDuplicates() and to that.grid.updateRow() in PlateViewer.patchWell() (formerly in PlateViewer.modifyWell()).
If this proves to be a significant annoyance, I can spend some time trying to track this down. But given how hard this is to reproduce, I don't think it's that serious of a problem.