Re: ATM f/ what ...my name is Rumak!

[Aplanatic@aol.com]

Hi Anthony:

> One concern I have is flexure of the miniscus.  To deal with this the
>  curves should be shallow and the element thick.  This spelled Rumak to me.
>  I played around with various self generated designs, optimising in OSLO LT,
>  but at no time did I even come close to the performance of the design
>  prescribed in "Telescope Optics."   However, if anyone wants to propose a
>  Mak prescription which beats the performance of the "Telescope Optics"
>  Rumak, I'll change my plans.  Besides, I'm thinking at least 8 months just
>  for the corrector.

The R and vV Rumak designs that I have worked with on TRACE show excellent 
performance.  As you recall this is an f/15 design which I consider to be 
just a little too slow for my tastes even in a visual instrument.  I 
re-optimized the design for f/10 and it still works well, showing that all 
on-axis rays are within about a five micron circle at the focal plane over 
420 nm to 700 nm.  At f/10 and 550 nm, the Airy disk is about 13 microns in 
diameter, so the inherent normalized transverse aberration in this design is 
less than 0.5, which I consider excellent.

The big problem with this optical system is the fabrication of the corrector 
plate, vis. the accuracy of the radii of curvature of the two surfaces and 
the CT.  There was a discussion of these tolerance issues in a previous 
thread a few months ago.  At that time I showed that the radii needed to be 
controlled to about 0.1% to keep the transverse aberration less than 1 for 
the on-axis spots.  However, if the second surface is fabricated after 
knowing the RoC of the first surface, and if its RoC is adjusted based upon 
that knowledge, then the fabrication problem is somewhat easier.  

That leaves a measurement problem.  How does one measure the RoC of the 
surfaces to the accuracy required?  Well, if the second surface is measured 
through the first surface by the Foucault method, then the *absolute* 
measurement accuracy can be as poor as 0.5% as long as *difference* in RoC 
can be measured to about 0.1%.  This should not be that difficult.

If you like I would be very pleased to forward the designs to you for the 
Rumaks that I have and also participate in a discussion of the tolerance 
issues.

>  The only draw back, the about 1/2 inch of excess glass I must remove from
>  the blank is more than what is needed to dispose of the surface defects
>  near the edges.  I guess only amateurs worry about wasting CT.

Why don't you send the blank to Newport to have it pre-generated?

Dave Rowe
Torrance, CA
Medium Format Astrophotography:
http://members.aol.com/aplanatic/photos/astro.html