[ATM] mirror testing: environment and equipment requirements?

[Dale Eason]

For testing with a Bath interferometer:

Vibration can be a problem.  It can be mitigated by using a fast shutter speed.  If you are the only person in the workshop and there are no machines running then it would be much easier and not be an issue.  Vladimir test in an apartment building.

Air currents are tamed by building a simple test tunnel from things like painters drop cloth sheeting.  Also take about 5 to 10 igrams and analyze the results (not the igrams).  OpenFringe makes that easy to do.

Roc and mirror diameter measurments need to be accurate to 1/16 of an inch.  Most people can measure to that precision.  OpenFringe will compute the error margins for you. It depends upon the mirror parameters being measured.  For example an 8 inch F7 measured with a 1/16 inch error will produce only a 1/100 wave difference.

Setup and alignment precision is just like Foucault.  No precision needed other than getting the return beam to fall at the right place.  That can be done visually just like you do with Foucault.

Parts are so inexpensive that it would not hurt most people to try it and if it does not work for them they have not lost alot.  However I bet that there would be very few cases where it would not work.

Dale Eason


--- On Tue, 8/5/08, Gert Verhoog <gjv at feestpaleis.net> wrote:

> From: Gert Verhoog <gjv at feestpaleis.net>
> Subject: [ATM] mirror testing: environment and equipment requirements?
> To: "ATM List" <atm at atmlist.net>
> Date: Tuesday, August 5, 2008, 11:31 PM
> One thing that's not clear to me regarding foucault and
> interferometry  
> testing are requirements related to the test environment
> and  
> fabrication and alignment of test equipment.
> 
> I was under the impression that a vibration-free
> environment is much  
> more important for interferometry than it is for foucault
> testing.  
> Would I be able to get meaningful igrams on a wooden floor?
> Or with  
> air currents caused by the heat of sunlight on a timber
> wall/curtain?  
> I'm afraid those are the constraints I have to live
> with in my current  
> "workshop".
> 
> Another thing that I remember reading, please correct me if
> I'm wrong,  
> is that igram analysis is extremely picky about focal/roc
> distance. If  
> you get the distance wrong just a little, the results will
> be entirely  
> different. Could anyone shed some light on this? Is that
> different  
> with foucault testing?
> 
> It seems logical to me that you need to have a decent test
> setup  
> (sturdy floor, accurate roc distance measurements, good
> alignment) in  
> order to get the much increased precision that
> interferometry could  
> provide. But is it also a minimal requirement? I would
> rather use  
> foucault to test a mirror up to 1/10th wave in sub-optimal
> conditions  
> than discover my workshop is not suitable for
> interferometry at all...
> 
> cheers,
> gert
> _______________________________________________
> ATM mailing list http://www.atmlist.net/