Re: [ATM] Testing the limits of ATM technology

["lou krajci" <loukrajci@comcast.net>]

From: "Daniel Bakken" <dan@runway.net>

>...I am
>particularly interested in: Carbon fiber tubes for the three truss tubes,
>upper ring construction, especially carbon fiber with possibly aluminum
>honeycomb composite, or other stiff construction ideas, materials and
advice
>on a pyramidal spider (carbon fiber, Dibond?), ideas on mirror cell design,
>especially edge support...

Ideas for mirror cell:

See: http://www.atmlist.net/pipermail/atm/2004-February/000433.html  - maybe
the book mentioned is worth getting.

Concepts to think about:

A.  Minimizing stiction in mirror cell components and interface/support of
mirror.  (Edge slings can have stiction problems if the design or
implementation is bad.  Mirror cell hinges can have stiction too.)

B.  Use of three bipods as hard points to provide lateral support...and then
a whiffle tree using astatic levers to provide axial support...which implies
a way to ensure the whiffle tree only provides force in the axial direction,
not lateral.  (See below documents for ideas.)

C.  Use of compliant/flexural hinges to minimize stiction.  Various types of
compliant/flexural hinges can have different degrees of freedom.  Some have
only one axis of rotation freedom...others have two axes of rotation
freedom.

See: http://www.aura-nio.noao.edu/book/ch4/4_5_C.pdf

Especially look at fig. 5 (three bipods for lateral support) and fig. 10
(view of bipods and axial support system)...and note that they are
supporting a 1 meter piece of glass that's about 50mm thick on only 18 axial
supports. Looks promising.

Note that no edge sling is used. Note that the mirror is glued to the
support points.

But we need more details on how to make the whiffletree (axial support)
pivot assemblies...and we want these pivots to be as low in 'stiction' as
possible. Compliant, or flexing, hinges are one way to accomplish this.

See:

http://celt.ucolick.org/reports/report16_mirrorsegsupprt_mar01/report16_mirr
orsegsupprt_mar01.htm#figures

Look at CAD drawings 1 - 8 inclusive. Lots of good ideas on making/using
compliant/flexing hinges...with relatively simple materials and techniques
at relatively low cost. Also shows how you 'isolate' the axial and lateral
support forces by using "rod flexures" (see CAD 2 and especially CAD 6). The
rod flexures allow the axial support system to transmit only axial loads to
the mirror...which means the lateral support forces are provided by the
lateral support system...which is as intended in this design and the first
one I mention above.

Note that this second design proposal, for lateral mirror support, uses a
central puck/disk arrangement (instead of three bipods, as proposed  in the
first design ). Off the cuff I like the three bipods...because they also can
serve as hard points that define mirror collimation.   That way you don't
depend on the whiffletree having to provide the job of maintaining
collimation.

But the pro's use electronic actuators to provide/control the force for the
axial supports.  We won't do that.  Instead we'll use the astatic support
system, with a weight/lever that pivots on a compliant/flexing hinge.  See:

http://astrosurf.com/altaz/chrysocallevers.htm

and

http://astrosurf.com/altaz/astatic.htm

So what we have is an astatic system to 'actuate' the proper axial support
force (depending on elevation angle of the mirror) with a whiffletree to
distribute the force from a single lever/weight assembly to all the points
on the mirror's back.  And three bipods are used to provide lateral support
and 'hard point' positioning to keep collimation stable (and also provide
some axial support in concert with the rest of the whiffletree supports).
This means that this mirror cell will probably be a bit thicker/deeper than
the average ATM cell...but I think for large/thin mirrors it will perform
better.

This design avoids problems with stiction and edge support by means of
slings.

Overall I think supporting a large, thin amateur mirror is possible with
this approach. It's not too complex, doesn't use electronic actuators (it's
entirely passive/mechanical), and is relatively inexpensive.

For collimation you have to move the entire cell...or at least the three
bipods because they are the hard points that define the mirror's plane that
it's held in.

Please give me your comments and criticism of this initial design proposal.
(I've not yet built it, but I hope to do so and give it a try.)

I hope this helps.

Tom Krajci
Albuquerque, New Mexico
---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.737 / Virus Database: 491 - Release Date: 8/11/2004

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