RE: [ATM] Mirror cell design

["mbyorick" <mbyorick@starband.net>]

Jeff Anderson-Lee wrote:
>How thick is your mirror and what kind of glass/ceramic is it?  How
many 
>point in your cell?
>Rather than say "it works", please elucidate "it works under these 
>conditions".

The mirror to which I previously referred is ~1.6" thick Pyrex, so it's
not a particularly 'thin' mirror by present standards.  At any rate, my
'observation', more formally stated, is the following: 
For a mirror of ~10/1 diameter/thickness ratio (made from a
similar-to-Pyrex material, which I believe would include most types of
glass/other ceramics), the process of gluing the mirror, via a
sufficiently compliant adhesive (e.g., RTV silicone), to a flotation
cell (even a half-baked homemade one), results in star tests indicating
negligible or non-existent observable astigmatism attributable to the
mirror attachment method. 
 
The above is not set forth in a comparative sense ('clips vs. glue'),
because it was not feasible (for me) to determine exactly how tightly
each of the original mirror clips was honked down (i.e., possibly
'over-tightened') (but Richard S. might be glad to know that the clips
were tight enough for the mirror to retain collimation when subjected to
at least a 2 G force induced in the process of 'GEM-flipping' the OTA)
[Geez, Richard, your sense of humor is nearly as warped as mine].
Perhaps(?) if each of the clips were carefully torqued down into place,
at some optimum (but unknown) force, the mirror would not suffer
astigmatically as a result.  But then again, how would the magnitude of
this hypothetically optimum force be determined, much less measured in
place?  

OK, Jeff, here are the conditions under which the above 'observation'
was made:

EQUIPMENT:
The cell to which the mirror is RTV-ed is a PLOP-compliant (bad pun)
[one that 'conforms' to PLOP-generated data] 18-point cell made from
aluminum stock.  Well, I not sure that anyone actually knows what
'PLOP-conforming' really means when each support 'point' of the mirror
cell is actually a relatively large, compliant 'blob' of RTV.  At any
rate, the mirror cell glue 'points' are 10-24 TEE nuts, the mean
diameter of the RTV blobs is ~1", and the thickness of each blob is ~1.5
mm.  I used a clear 100% RTV silicone (GE Silicone II). 

[Note: I believe that RTV blob thickness is important, more from the
standpoint of preventing unnecessary joint flexure (caused by excessive
thickness), than from the standpoint of being sufficiently compliant to
avoid inducing stress on the mirror because of thermal mismatch.  Hey,
the big boys have found the stuff to be sufficiently compliant for their
multi-meter-sized mirrors.  More specifically, Dow-Corning 840 RTV
(available from McMaster-Carr for about 8 bucks a tube) is apparently
the adhesive of choice for this application.  Interestingly, this stuff
is alcohol-based, so it has to smell a lot better than traditional
acetic-acid-based RTVs.]

PROCEDURE/RESULTS:
I was initially able to simulate, indoors, the astigmatism I observed
with real star images, using an artificial star made from a bright LED
and a VERY small hole I finally managed to poke in a piece of aluminum
foil.  When I removed the mirror clips from the original (Meade) cell,
the astig. disappeared, and after RTV-ing the mirror into the new cell,
the original astig. did not re-appear in 'real' star tests.

Hopefully the above, although not particularly quantitative, is
sufficiently qualitative so as to convey some possibly useful info to
the ATM troops.

Cheers,
Mike Byorick


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