[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]

Re: [ATM] mirror pinching



The first thing that I'd do is to do a star test followed by
Ronchi and A knife edge test on the mirror.All three tests can be
done on a star in this case, providing an excvellent view of the
subtelties of each of the tests.
Thge star test should show that the light is even across the
disk, the Ronchi should show perfectly straight lines and the KE
test shold show an even darkening of the light like a sphere on
the benbchg would show.  My bet is that you're ont going to be
seeing this but rather some figure of revolution with zones and
so forth in rings around the mirror.  There is also a
probablility that there will be some just random lumps and
depressions on the face of the mirror but those tend to be a bit
less ccommon.  Then there is also the dog-biscuit effect on the
surface of the mirror which is expressed usually in lumps on the
order of about  a quarter of an inch in size all over the surface
like measles.
Pinched mi9rrors tend to have a trefoil shape when the cell is
pinching them and a pair of lines towards the center when the
side supports are doing the pinching. Generally, if the mirror
has the ability to move a little in the sideways direction then
the side supports don't do any harm to the image.  It is when the
side sup0ports actually press on the mirrorthat the possibility
of a pinched mirror appears.  If you can slide anything between
the mirror and a third point of the cell then you don't have a
pinched mirror.  At a 10:1 ratio, you don't have any significant
problems with gravity pototato chipping the mirror in the cell.
Bob May
Due to the large volume of SPAM thaat I'm getting, add an open
square bracket to the subject line or my spam filter will reject
your post~
rmay at nethere.com
http: slash /nav.to slash bobmay
http: slash /bobmay dot astronomy.net
Replace the obvious words with the proper character.

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