To: dobsonstathis{at}planet-interkom.de, atm{at}shore.net
From: mdholm{at}telerama.com
Reply-To: mdholm{at}telerama.com


Stathis,

I think the most likely explanation is stress.  Not too many things rotate
the plane of polarization, stress is one.  If you rotate one of the
filters, the light and dark patterns should swap places: light becomes dark
and vice versa.

Make sure the pattern moves with the mirror blank, and not with either filter.

It is possible for most of the stress in a thick glass part to be near the
surface.  Grinding down both surfaces and polishing might remove the
stress. Still, it is a bad sign.  It indicates poor annealing.  There is no
guarantee that the stress will be only near the surface.

Try a molded plastic part.  Most will show some stress, especially near the
mold gate, where the plastic entered the mold.

Also, take the filters outside to an automobile.  The side and rear windows
should show a strong pattern of light and dark spots very roughly 20 mm
diameter.  The pattern is there because the tempered glass used in side and
rear windows is purposely made with a large amount of internal stress.  It
is the opposite of annealing.  I can often see it in car windows with only
the natural polarization that arises from atmospheric phenomena and low
angle reflection.  With polarizing sunglasses, it is easy to see.

Oh,

Did you oil the ground surface?  I'm not sure what effect sending the light
through a ground surface would have.  Oiling the ground surface should
minimize any effect of the ground surface.  Vegetable oil, or mineral oil
should work.

I do not know if the ground surface itself might induce enough stress to
cause the patern you see.  Certainly grinding does cause local stress, but
it should be much too small a pattern to see witout a microscope.

Is the glass ground all over or were only high areas ground leaving other
areas smooth?  If so, the local stress in the ground areas would rotate
polarized light.  The rotation would be of varying amounts on a small
scale. Averaged out it would look colorless.

Perhaps others with more experience in stress testing glass blanks can
contribute here.

Mark Holm
mdholm{at}telerama.com

--- BBBS/NT v4.01 Flag-5