Accuracy of Machinist Levels

[Larry$]

About particle board tops: The stuff sold at the big box stores is less dense/stable than the industrial board (48#s / cu. Ft.) But both are pretty unstable with changes in moisture. They are also quite compressible. Less than ideal for a machine bench. You can slow the changes by trying to seal the board. You have to treat both sides the same and especially the edges. If you don't treat both sides the same the panel will curl (even worse!) A laminated maple top would be better. Seal it with several coats of urethane varnish, all around. Best alternative: STEEL!

 

[Optic Eyes]

That's an interesting approach to taking the twist out of a bed.
At first blush, I like it.

Warner and other Hand screw machines would machine quite big parts repeatedly year after year many times 3 shifts a day, production would never consider changing jobs on a machine to distribute wear evenly, until Repairman told them to use a different area of the machine, when they wore out we bought another one.
The point is you can overreact on engine lathes, many Tool room lathes like Harding's are on three point suspension and only need simple leveling, Colchester's from England are as well and was the Navy's choice for some time.

 

[Optic Eyes]

That's an interesting approach to taking the twist out of a bed.
At first blush, I like it.

The problem with levels is you are adjusting a surface using qualitive analysis, if you used quantitive analysis by comparing the surface to a datum you would know how high or low each spot was. Optics does this much faster than using levels giving direct readings of elevations.