[ATM] UltraThin support investigation

[David Davis <vorblesnak@peak.org>]

bob wrote ..

 > Seems to me (here we go again) that the best "edge support" solution is
 > none.  Get your lateral support by gluing the mirror to the cell with
 > silicone.  Note that one of the objections to this, having the support
 > displaced to the rear of the COG, is lessened by the fact that this 
mirror
 > is slumped.  In fact, with a thin and fast enough slumped mirror, the 
outer
 > support points will be ahead of the center support points!

Richard Schwartz commented on a support idea he had,  but I have not had 
the time to follow it up.  I think he too suggested attaching the mirror 
to the cell but only at the middle,  maybe just one spot.

You make a good observation here in that the outer supports are ahead of 
the central supports.  In a cell made for a flat back mirror this would 
apply more pressure to the outer the outer ring of support, unless the 
mirror had enough weight to overcome the pivot friction of the cell 
components.  I don't think this mirror does.

I tried using a ring at the edge, a bicycle rim of the right diameter, 
both attached with silicone rubber and as a simple edge support with 
points for the glass to rest on.  I found distortion in both cases.  The 
cell I sent to Dale Eason provides support points in the outer two rings 
of a PLOP design.  This worked as it did not show a detectable 
distortion.  Dale's interferometer is much more sensitive and may show 
some print through of the support points, or distortion.

The 15.5 is coming back and I will see if I can work the central hump 
out and smooth the curve.  That glass did not respond as well to the 
soft point idea and I am thinking I will try Richard's idea if I can 
find the details on it.

Dale was going to narrow the distance between the pegs on his test stand 
and try again.  It will be interesting to see if there is a noticeable 
change in the mirror.

David Davis
Toledo, Oregon 97391
_______________________________________________
ATM mailing list http://www.atmlist.net/