Re: [ATM] Use of bearings in (static) mirror cells

["Olivier Biot" <olivier.biot@versateladsl.be>]

From: Don Clement

>> ... There is a variety of different kinds of MICROSCOPIC
>> MOVEMENT of the mirror in the cell; a bearing offers no resistance to
> these
>> microscopic movements, allowing the mirror to settle, with no strain, ...
>
> Flexures are the ideal bearing for a mirror cell. Flexures work not only 
> at
> the microscopic level but also at the molecular level. The movement of a
> flexure is intermolecular within the solid. Flexures can have zero 
> stiction
> almost immeasurable friction. Unlike other bearing types, flexures are
> almost immune to environmental conditions and contamination. Flexures need
> no lubrication. An RTV blob is one type of flexure. But there are many 
> other
> geometric forms for flexures and material types. If the RTV blob alone 
> does
> not offer sufficient rigidity then perhaps additional flexures in
> conjunction with a thin layer of RTV would suit the requirements of 
> rigidity
> in the needed direction.

Hi Don,

That was what I was initially thinking. Maybe we can influence the geometry 
of this RTV (AKA Silicone) "blob", and shape it like an old circus toy named 
"diabolo", or even a cooling tower of a power plant. I thought that, if I am 
able to narrow the RTV blob in the middle, then it would flex much easier. 
Maybe this complexity is not needed, but...

One reason why I feel somehow reluctant about bearings in static support 
systems, is that although much of the friction is removed, there's still 
*some* friction in the bearing, which would require some wind gust or some 
light (or heavier) tapping in order to overcome the friction. Another reason 
why I prefer avoiding bearings is their inherent cost and the significant 
machining effort you need when developing your cell.

It looks like there are 2 "camps", a larger group in favour of bearings, and 
another favouring RTV. It would be interesting to know whether machine shop 
experts favour the bearing approach and the people without machine shop 
experience (like me) maybe prefer techniques they can master...

What I am trying to do, is building a mirror cell consisting of the 
"collimation frame" supporting the "mirror cell frame" supporting the mirror 
(18 points). So I'm *not* using any support points of the cell for 
collimation. This allows much easier collimation, and to my eyes a simpler 
(albeit slightly heavier) cell design.

Best regards,

Olivier 


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