[ATM] Foucault test errors due to source and knife edge separation

[bobmay at nethere.com (Bob May)]

That kind of depends upon the direction that the difference is.  If you are
using a slitless tester, the light source can be very close to the viewing
edge and that will make things very fine in the vertical direction (assuming
the "normal" vertical KE that most build in a tester) with no error in the
horizontal direction.
The error that happens is that you will end up making a mirror that is
slightly short of a parabola as you will be testing the mirror with one
focus of the eliptical shape being the light source and the second focus
being the KE where the mirror shape is seen.  Generally the error is so
small that the typical error bars of the measurements will be larger than
the error from the setup of the test.
If you don't like the slitless tester, you can put a second edge over the
light source of the slitless tester and make a slit at that point.
Bob May
http://nav.to/bobmay
bobmay@nethere.com
NEW! http://bobmay.astronomy.net