Re: [ATM] Basic mirror support question...

[Mark Holm <mdholm@telerama.com>]

Ah, yes.  That is exactly the rub.  In order to support a mirror well,
you can't tolerate it just flexing down to touch an almost flat surface.
  The thing that newcomers usually fail to realize is just how precisely
the surface of a mirror must be maintained, and how difficult a
proposition it really is.  For really good performance, you would like
your mirror surface to not deviate from theoretical perfection by more
than about 10 nanometers RMS or roughly 25 nanometers peak to valley.
This is way smaller than normal machining tolerances and, is small
enough that even materials as stiff as glass, aluminum and steel flex,
under their own weight, by this much over small spans.

Note that 0.001 inch, a reasonably good machine shop tolerance for all
but highly precise work, equals 25400 nanometers.  Thus, our mirror
surface needs to be roughly 1000 times better than ordinary machine shop
tolerances.  The tolerance on flexure is just as tight.  This factor of
1000 is where all the trouble originates.

Jerry Reddell wrote:

> 
>>From a practical standpoint, there is enough flexibility / compressibility
> in real materials  that essentially all points will be touching.



-- 
Mark Holm
mdholm@telerama.com

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