Re: [ATM] 'Commercial' annealed, rather than 'fine' annealed

[Mark Cowan <toolontop@yahoo.com>]

OOps, I guess I left that out... :(

Here's my take on the testing, based on examining
dozens of pieces of Pyrex and large quantities of
plate as well.

A precision annealed piece, under crossed polarizers
(you only need a small viewing polarizer), will show
no "cross" or any visible differences under careful
examination.  There may be tiny stress loci that
generate tiny (mm sized) "maltese crosses" or a kind
of a bowtie effect.  Unless these are at the surface
they don't matter.  Sometimes an impurity is visible
in the glass (piece of sand), this will always
generate a stress locus.  

One time I had a finished mirror with flawless surface
that showed a diffraction spot under Foucault testing.
 I marked the spot, couldn't find anything on the
glass.  Under the polarizer there was a tiny localized
stress bowtie visible, no inclusion or anything - just
happened to come out to the surface - but man it was
visible easily by the effect on the Foucault image at
that point.  Manufacturer (Newport) had no explanation
for it better than mine, which was...right, an
invisible flaw.  Overall the glass showed no strain,
as expected.  Remember, Pyrex is only theoretically a
homogeneous glass.

Fine annealing - may show barely visible maltese cross
or variation.  May have visible veining or striations
(internal structure with stress) that can sometimes be
spotted without polarizers.  Makes a fine mirror
anyway.

Commercial annealing - can't say, as I haven't
knowingly tested any.  Some glass I've seen of
uncertain vintages has shown more strain than I'd be
comfortable with.

But the bottom line is, if it shows no visible strain
effects under polarizers (ie, looks like ice or
whatever, nothing moves with the polarizer angle) then
it's good enough.  I think the test is sensitive
enough to weed out any bad blanks even without
numbers.

To extract numbers I believe you need to record the
behavior (% rotation) of the light and dark fields
against the angle of the polarizer.

To get a tolerance out of it - the only way I could
see to do it would be to make identical mirrors on
good and bad annealed glass and then test them at
different temps.  Theoretically I believe you could
calculate the surface deformation under assumptions of
strain relief and arrive at some numbers that way. 
But I'm not drawn to try to do it.

THEN FOR FUN, STICK A PIECE OF OPTICAL CALCITE IN YOUR
POLARIZER TESTER!

Cheers,
Mark Cowan
Salem, OR USA

Where we will have a Venus transit in 2012...any links
to predicitions on that one?

--- Michael Lindner <mikell@optonline.net> wrote:
> On Saturday 24 April 2004 05:49 pm, Mark Cowan
> wrote:
> > This is why inspection is vital!  You could get
> large
> > sheets of Polaroid from the Edmund's surplus
> outlet...
> 
> The only thing missing from this equation is how to
> test. Yes, a "cross" is 
> bad, but how bad? How do you use this test to
> quantify the strain levels 
> enough to say "this won't do for a 1/10 wave mirror,
> but will do for a 1/4 
> wave mirror"?
> 
> -- 
> Michael Lindner
> http://www.starastronomy.org ***
> http://home.att.net/~mikel
> http://www.atmsite.org *** http://www.atmlist.net
> 



	
		
__________________________________
Do you Yahoo!?
Yahoo! Photos: High-quality 4x6 digital prints for 25¢
http://photos.yahoo.com/ph/print_splash
_______________________________________________
ATM mailing list http://www.atmlist.net/