Originally Posted by
Malcolm McLeod
To be rigorous, it would be wise to spec that an edge should be compared to the other and then swap end-for-end and compared again. Certain shapes, symmetric about the center, could ‘pass’ this test otherwise (think about a gentle s-curve :: all 3 could mate).
Admittedly this would be virtually impossible to create on a jointer, but after the art-work-as-end-stop thread, I just had to toss this grenade over the fence.
Hah! Good point, so swapping end for end or moving them around would reveal a matching s-curve. Anything is possible especially as they become flat and minor inconsistencies reveal themselves.
For what it is worth I setup a jointer with a pair of high precision spirit levels after checking each table separately for flatness. I have a 40” long box square (cast iron tube that has been scraped flat/true on all sides) to compare to the tables.
Check the tables in an x pattern then length wise and note any errors. Set the outfeed parallel to the head across its width. After which check them for parallel both along the lengths and across their widths. Set their heights, then cut the wood.
Bumbling forward into the unknown.