[ATM] ATM: Thoughts about mirror cell design from astubbornRTV-head

[mbartels at bbastrodesigns.com (Mel Bartels)]

> > any distortion introduced by bending or overbending, even with a perfect
> > mirror, is inconsequential because the wavefront error is dominated by
> > seeing on those scales even in the best conditions. Bubble wrap,
> > carpeting, whiffletrees: all seem to work.

Well, yes, but you can see rather tiny distortions in the star test that
don't readily show in the at-focus image.

> I'll note that Mel Bartels claims there is noticeable distortion with even
one
> ounce of weight on a thin mirror, and if we believe this (and I do have
faith
> in Mel's skills as an observer), then even though seeing dominates the

Yes, this is with ultra-thin mirrors where the weight hangs off of the
mirror back, attached with a sticky pad.

> 1. Given the supports are at the same points, how does the force on the
back
> of the mirror differ if the supports are part of a whiffle tree or part of
a
> passive astatic system? In my mind, the forces on the whiffle tree points
of

Well, basically, not much difference if comparing well executed systems.
But there can be a difference in that the whiffle tree points all come down
to a 3 point flotation, so that each support point's contribution to the
support can vary depending on how the support bars and triangles that lie
between the support points and the 3 point flotation are constructed.  On
the astatic system, support contribution is entirely dependent on the
counterweight.

> 2. When the mirror is collimated (as is inevitable in a truss scope), the
> "hardpoints" which define the plane of the mirror move. The counterweights
> are no longer horizontal when the scope is vertical and vice versa. Does
that
> mean the counterweights need to be "tuned" each time the scope is set up?

The counterweight is hinged such that it fully supports the mirror when
mirror is face up, and so that there is no support pressure at all when the
mirror is aimed horizontally.  The trick is that the bearings much be very
good, as any friction that might hold up the counterweight will change the
support pressure contribution.  There is a very good article in an old Sky
and Telescope about a Brazilian 20 inch where the astatic support system is
well documented.  The edge is similarly supported.  The drawback is the
expense and extra weight of all that counterweighting, though with
lightweight mirrors and center of gravity desirability towards the rear of
the tube, this maybe isn't such a drawback.  Recall that the 200 inch uses
rubber bands on its levers to fine tune the support.

Mel Bartels