Re: [ATM] Friction and sticking in multipoint mounts

[James P Crombie <jpcrombie@pei.eastlink.ca>]

I would agree with the compliant hinge usage.  My cell uses machined 
aluminum triangles held in place by stripper bolts with several thou 
cleance, and rubber washers between the cell and the triangles.  There 
is also a light spring on the back of the bolt but they really dont do 
much except keep it from rattling.  I think as long as the back of the 
mirror is relatively flat and the cell is uniform and flat then there 
will be very little movement in the cell itself.  The rubber washers 
will adjust for any flexture in the cell and the use of a sling should 
compensate for any lateral loads at low angles. 
http://www.jamescrombie.com/mirrorcell/mrcellpt2.jpg

Nils Olof Carlin wrote:

>Art Bianconi wrote:
>
>  
>
>>I've worked a fair bit with FEA software and while I have limited
>>exposure to PLOP, I will assume that it's a good program and
>>yields accurate information.
>>    
>>
>
>Some have compared results to simulations with commercial FEA programs (very
>expensive, I take it) and found they agree within reason - this is no proof
>of course, and the model assumed is of course a simplified description of
>reality.
>  
>
>>However, the problem I have is the assumptions so many appear to
>>make about the viability of mechanically hinged support systems.
>>For us to build 27 point supports with so many mechanical
>>fulcrums and expecting the moments and C of F's to be uniform is
>>naive.
>>    
>>
>
>They won't be exactly uniform, but Plop offers facilities to try the
>sensitivity to errors - a feature I am not well acquainted with. But it
>would be more useful to calculate the actual sensitivity to small
>differences in forces - with some care, forces wouldn't vary more than
>occasional percents between adlacent points or by adjacent groups of
>points - provided you don't make the support points too wide, "fuzzy" and
>poorly defined.
>  
>
>>I've buddies that worked as machinists in the prototype shop for
>>Bell Labs where exotic designs are (were?) common place. Those
>>guys made more money that many engineers because of their ability
>>to deliver on super accurate, highly sensitive designs for
>>scientific research. And, they had access to a machine shop that
>>the guys at NASA would envy and equally large budgets.
>>
>>When I showed them a typical 27 point mount and tell them what is
>>expected, they roll their eyes. They hear ". . . to 1/12th wave
>>accuracy"  and they look at me as if I am crazy!
>>    
>>
>
>We don't know the reason for their looking at you, rolling their eyes <G> -
>it may or may not be based on a good sensitivity analysis.
>
>  
>
>>There HAS to be a better way!
>>    
>>
>
>I am sure you didn't miss the opportunity to ask those buddies how to do it
>better. Come on, what did they suggest???
>
>But there are certainly questions you should ask yourself about a mirror
>cell - such as having made it to balance the forces pretty accurately, how
>critical is it to ensure that the mirror is held accurately centered on the
>cell - and do you really check? And what about edge support - have you made
>sure it does not introduce deforming forces far in excess of those by a
>somewhat unbalanced back support?
>
>Nils Olof
>
>
>
>
>
>
>
>
>_______________________________________________
>ATM mailing list http://www.atmlist.net/
>
>  
>

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