The foucault test, in my opinion is qualitive and semi-quatitive at best, its one of these things that self proclaimed expert master opticians like to brag about being good at.
The truth about this test is you may think you have the correct readings at a particular zone ,but you could be off by .001 or .003 or more, this test yields imperical data, ive seen people use only the "couder mask" and tested at 4 zones, to me this is a vary small map of the surface profile, one should get readings every 0.5 inch across the radius, and test with and without masks so you can see the zones that are obscured between the mask openings, and its not a bad idea to use the ronchi test, this will reveal to a novice and pro alike right away, gross extreme errors like t.d.e for example.
dont get me wrong, the foucault test IS good enough for the amateur , in my opinion the foucault test is good to about 1/20 wave MAX!.
and thats if your real good,and thats good enough.
forget about those books that say to use a slit
,make about a 1mm hole with a small piece of ground glass between hole and light source, for a light try a halogen, and design it so the light/knife edge move together, the knife edge and light should be as close together as possible
,ideally use a small prism to redirect the light through the ground glass/hole at a 90 degree angle
so as to provide your eye ample clearence(so the bulb and socket are not in the eyes way),this will also allow closer spacing between hole and knife edge.
for a positioning stage, think of things like brass rod, compression and extension springs, i made one using modified drill press vices.
good luck.