Re: [ATM] 6-points cell voor 20 inch mirror?

[Mark Holm <mdholm@telerama.com>]

Arjan te Marvelde wrote:
> Mark,
> 
> I don't think this statement scales very well. Some cell designs are less
> susceptible to construction errors than others.
This is a testable hypothesis.  Plop can probably be made to do the
testing, but it may be tedious to set up.  One thing to keep in mind is
the extreme influence of mirror thickness to diameter ratio.  Based on
what I have seen, thinner mirrors are probably much more susceptible to
cell construction errors than thicker ones.  Furthermore, the
appropriate thickness to diameter ratio to use for comparisons is not
constant with increasing diameter.

 I believe that especially
> the 3 and 6 point cells can take a smaller error margin (i.e. factor 2
> rather than 4).
Possible, but be sure to take the thickness to diameter thing into
proper account.

Also the pad size will influence this factor.  Either Tom Krajci or Jeff
Anderson Lee did some Plop work simulating increasing pad size.  He was not
looking specifically at construction tolerance effects though.
Increasing pad size had a modest beneficial effect on overall deformation.

I started looking at the effect of pad size on construction tolerances
in 6-point cells.  I didn't really finish the job, so it isn't in
publishable form.  What I was seeing was that increasing pad size did
not seem to decrease sensitivity to construction error.  Now I know Plop
is not a great tool for simulating pads.  You have to simulate them as
an array of points and they haven't any give.  The next step, not a
small one, would be to model it up in a real FEA program.  I'm not in a
position to do it.

Mark Holm
mdholm@telerama.com

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