Re: [ATM] Another Astigmatism question, but at the sphere stage

[Guy Brandenburg <gfbrandenburg@yahoo.com>]

FWIW, about a dozen or so rather experienced ATMers
(including myself as 'very experienced' might be a
joke, I donno) tested 3 mirrors over the past 2 or 3
years in something that was called the Mirror Round
Robin (or something like that). Why do I mention this?
Because one of those mirrors was astigmatic, and IIRC,
NONE OF US CAUGHT THE ASTIGMATISM.
Bottom line, as far as I can tell, is that Ronchi and
Foucault/Couder/zonal knife-edge tests are absolutely
unable to catch astigmatism. 
If your technique with rotating everything in a
systematically random manner is good, that's your best
defense against astigmatism.
A different ROC by 0.5 mm in a ronchi or knife-edge
testing means, I think, virgually nothing: you cannot
place the mirror back onto the stand with that sort of
accuracy. 
The only accurate tests that I know of for astigmatism
are elaborate interferometic instruments or else the
star test (in some form).
That's my $0.025.
Guy

--- Michael Peck <mpeck1@ix.netcom.com> wrote:

> At 19:47 8/27/05, J Killea wrote:
> >
> >HOWEVER, I just did three sets of Foucault
> measurements along each 
> >of 2 diameters and Sixtests (set to "Target b=0"
> with variable ROC 
> >calculation) came back with the ROCs of each
> diameter differing by 
> >about 0.5mm.  That's more delta than what I had
> when I thought my 
> >parabola was done the first time.
> >
> >My question is whether it's too early to judge
> astigmatism, since 
> >I'm only at the sphere stage, and whether Sixtests
> with Foucault 
> >data is even a valid way to judge astigmatism.  I
> might just stick 
> >it back in the 'scope
> 
> Where is everybody? I can't believe nobody has
> offered a helpful 
> response after over 12 hours. Anyway, those radius
> of curvature 
> estimates returned by sixtests and figurexp have no
> particular 
> physical significance and certainly shouldn't be
> used to judge 
> astigmatism. If those were my programs I wouldn't
> even report those 
> numbers since people are easily misled by them.
> 
> I believe the tests of choice for detecting
> astigmatism in a 
> spherical mirror are ronchi, or perhaps an
> artificial star created 
> with a ball bearing as described in Tex.
> 
> >
> >Is it time to go back to AlOx #15 or 20?  (Please
> say it ain't so!)
> 
> No.
> 
> Mike Peck
> 
> 

Guy  Brandenburg
Washington, DC
My home page:
http://home.earthlink.net/~gfbranden/GFB_Home_Page.html
_______________________________________________
ATM mailing list http://www.atmlist.net/