Re: [ATM] Best locations for 4-Pt. Wiffletree edge support PLS.

["Mel Bartels" <mbartels@bbastrodesigns.com>]

> Was surprised that my thread question lead to so much discussion and that
> edge control for thin amateur mirrors  is still a "frontier" and hasn't
> yet
> been the subject of  3D modeling or analysis, like back support has.  Too
> bad that the latest solutions adopted by high-budget observatory solutions
> using very complex active systems  don't translate well for us atms.



Well, one of my main points is that it HAS been analyzed, at the pro level.
See Wilson's Reflecting Telescope Optics II, section 3.4.3, Lateral Supports
for Mirrors, and the conclusions on subsequent pages. Some snippets:

--- ideal radial support is equal distribution around the rim, in the plane
of center of gravity, each pulling or pushing at slight angle inward at
cosine of angle (24 pts for 8m blank)

--- slings though common are not very good - significant astigmatism remains

What we're missing is translation of the pro analysis into practical
engineering for amateur sized optics. 

The pros' ideal edge support is a 360 deg ring of support points that pushes
up on the mirror bottom and pulls up on the mirror top, each support point
pushing/pulling per cosine of angle, with the ring of supports set at the
mirror's center of gravity.  The 90 degree dual support is an intriguing
substitute as astigmatism is cancelled (other errors remain).

However, there's a universe of difference between the pros engineering and
the amateur engineering.  Consider the sling: the pros would have a sling at
the correct position, backed up by specific analysis of that mirror.  The
amateurs throw in stretchy seatbelts and all sorts of interesting but
totally unanalyzed slings with various constructions and angles of slings.
No wonder the results are all over the map.

The whiffle tree can easily if not accidentally be built with even
distribution around the rim but setting the forces for cosine of angle is
more difficult.  Counter-weighting or springs are in order, but the analysis
comes with a caution that forces need to be controlled to 1% (?).

You have two ways to go about this: start with the dual 90 deg supports,
have them analyzed using a decent FEA, elaborating from there.  Or be ready
to try several cells using feedback from star tests until satisfied.  The
latter though doesn't advance ATM in general because only a specific
solution to a specific problem was solved, and it's impossible to generalize
to other problems (other mirrors) since the constraints were never really
established.  But it's a way to get going - I mean you can build 2 or 3
mirror mounts in the time spent theorizing.

Mel Bartels


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