Re: [ATM] Higher order Null Lens for RC primary

[Jim Burrows <burrjaw@earthlink.net>]

At 2004-08-09 10:38 -0700, Gert Gottschalk wrote:

>I made a 10" RC primary (F2.6, SC=-1.17) and tested it
>with zone tests, only. The correction is OK but with
>the steep curves it is impossible to judge if the
>surface is free of zones.

Been there, done that - an 10"f/2.6 RC cass primary. Unfortunately figured 
with HCF - so the mirror's too rough and going to be refigured 
(sometime...when the Seattle rains begin).  The Foucault test is nearly 
impossible to do on mirrors this fast.  I used a modified Foucault test 
which I called Focal Point Foucault (FPF), involving a beam splitter and 
microscope looking at the image.  I think (but not really sure) that the 
test was barely accurate enough.

Let me push my webcam Hartmann test:

         http://home.earthlink.net/~burrjaw/public/hart2d.zip

which should work, provided the mirror isn't as rough as mine.  The 
advantage of this vs. a null test is that it will show you the quantitative 
errors relative to the best-fit conic RC parameters, rather than just 
saying there's errors of unknown magnitude relative to the fixed null test 
parameters.

It might be of interest to mention that the standard Schwarzschild conic 
section RC design (e.g., Texereau, p. 147) is just an approximation.  Born 
& Wolf, in "Principles of Optics", 7th edition, p. 214, give the exact 
formulation in terms of two differential equations.  The RMS difference, 
conic - B&W, is 12 nm for a 10" f/8 RC system.

         -- Jim Burrows
         -- mailto://burrjaw@earthlink.net
         -- http://home.earthlink.net/~burrjaw
         -- Seattle N47.4723 W122.3662 (WGS84) 

_______________________________________________
ATM mailing list http://www.atmlist.net/