ATM Central Mirror Support

[Aplanatic@aol.com]

Howdy,

So many questions, so little time.

Support a primary mirror from its center only, as on a baffle tube for a Cassegrain.  Grind a conical shape into the back of the mirror so that the edge of the mirror is thinner than the center.  Under gravity, what is the RMS error in the mirror's surface given a cross-sectional profile for the mirror?  Assume that 2nd order deflection is removed by re-focusing.

What is the best shape for the back of the mirror to minimize the RMS error?  How large and thin (light-weight) can we make the mirror with this support?  Is there a better shape for the back of the mirror than conical?

When the mirror is tipped toward the horizon, how does the shape of the mirror change?  Is this support method better or worse than a multi-point cell and a sling?

Is there a finite element analysis program available on the web that can handle this geometry?  When gravity is normal to the mirror, the problem reduces to one dimension.  It should be a fairly simple analysis for this case.  Any FEA experts out there?

Thanks,

Dave Rowe