Re: ATM Foucault test accuracy

["Bill T." <twentiethwave@hotmail.com>]

Ken,

Didn't mean to put down the Foucault test at all!

My own experience, based on star test validation, is that is reasonable to
make Foucault-based figure changes up through at least 1/20 wave PV
readings -- this is high confidence for any optical testing system!  I think
we agree that additional qualitative tests like the Ronchi test and the
shearing interferometer validate that the sparse Foucault data points are
truly representative of the overall surface.

Bill T.

Ken Lowther wrote,

> You don't do KE testing until you see a decent shape and surface.  So
> yes, you do take small samples, but you take them from a surface that
> you should have verified as smooth and regular before you started the
> test.  There are many of us that will yawn forever if you talk about how
> 'bad' the Foucault test is.  It works for us and we have seen the
> results at the eyepiece.  If it doesn't work for YOU, then perhaps you
> do have a problem doing the test and should evaluate your methods.