Re: [ATM] Flexed mirrors

[David Weinshenker <daze39@earthlink.net>]

Bill Kelley wrote:

> Why flexing for the ATM ? :
>   1. A spherical surface is the easiest to make.
>   2. A spherical surface is the easiest to test.
>   3. Extreme smoothness can be achieved because the tool and mirror fit in 
> all positions of the polishing stroke. The larger
>       and faster the mirror the worse the fit, and consequent small scale 
> surface defects. The resultant light scatter reduces
>       resolution and explains why many refractors provide better resolution 
> than reflectors.

I wonder if this explains why the small mass-produced Maksutov/Cassegrain
scopes work as well as they seem to?

>   4. Among the many other advantages your flex mirror can be detuned to an 
> appropriate ellipsoid  for near range imaging.

Very interesting... or similarly to suit e.g. Dall-Kirkham systems with
interchangeable secondaries of different curvatures for different final
magnifications (and readjust the primary to "fully parabolized" for use
with a flat secondary in the Newtonian manner).

Certainly providing a direct "field adjustment" for spherical "correction"
might be as useful as having the means to fine-tune the collimation...
some observers, at least, might value the ability to tune out a visible
aberration under actual use conditions... grab the big knob sticking out
of the back of the mirror cell and adjust it until all of the light goes
through focus at the same time.

I guess the real question is how exactly a bent, figured-spherical
plate can be made to match the exact desired curve. There might be
something in a design resembling the "conical" style mirrors that
are advertised by R.F.Royce, which are supported on a single peg
at the center and have a thinned, unsupported edge. If something
like this were figured spherical on the unstressed glass, and then
flexed by using the center post to pull the edge back against a
support ring, then the exact thickness profile from center to edge
might be used as a degree of freedom to establish the exact pattern
pf bending.

-dave w
_______________________________________________
ATM mailing list http://www.atmlist.net/