Re: [ATM] A challenging 28" project

[Mike Lockwood <melockwo@uiuc.edu>]

Hi,

Richard wrote:
> It's not the power that matters, it's the numerical aperture, 
> although I admit they do tend to go hand in hand. The 'standard', 
> if there can be such a thing, on these spherical wave 
> interferometers tends to be NA 0.25 which often goes with a power 
> of 10x. Works for me and is easily capable of a beam of f/3 or 
> better, ie mirror under test of up to about f/1.5. To get a wider 
> beam for Mikes' f/1.1 I would have though you would want to go to a
>  smaller NA, so a NA 0.1 5x may well be the right direction rather 
> than a NA 0.65 40x.

Yes, an F/1.1 tests with an F/2.2 cone (because it is tested at the 
ROC).  NA goes up (not down, Richard) as focal ratio decreases and the 
cone expands, so I would need greater than 0.25.

NA = n sin(theta), where n is the index of refraction (1 for air) and 
n is the half angle of the cone of light going out of the lens if it 
is fully illuminated from the back.  So, for an F/2, the half angle is 
26.57 degrees, so the NA required is sin (26.57)*1 = 0.4472.  At first 
I used an objective that I think was 10X with an NA of 0.5.

I found objectionable fringe artifacts with the microscope objectives 
I had which were broadband AR coated - there were some reflections 
that interfered with each other or the transmitted wavefront.  Since 
I'm only dealing with laser light I found that Edmunds has a 
reasonably priced, mounted (in a microscope objective-type housing 
with the same threads) laser singlet that eliminates that problem for 
the HeNe wavelength:
http://www.edmundoptics.com/onlinecatalog/displayproduct.cfm?productID=1922

I used the shortest focal length.  I actually bought two of these - 
one is used in the tester, the other illuminates the 5u pinhole that 
is the light source for my collimator.

The mounted singlets were hard to find online - you have to look under 
the spatial filter heading in their online catalog.  I first saw it in 
quite an old printed catalog that I chanced upon and started paging 
through.

> The critical element is the small concave mirror which is the 
> spherical reference surface. Ideally, it would be optically 
> perfect. In the real world it needs to be better than 1/20th wave 
> on the wavefront. Surplus Shed sell (or used to sell) a Pyrex 2" 
> f/2 uncoated mirror for about $3. Mike has measured some of these, 
> I think they weren't bad, better than 1/4 wave. If you can figure 
> one as a reference element you can save a fortune. The rest is a 
> small HeNe laser (Ebay ~$50) and a beam splitter of some kind like 
> a pellicle or cube beam splitter (Surplus Shed ~$25). The former 
> has less problems in use than the latter, but is extremely 
> delicate.

Actually I will have to simplify my test setup in order to test these 
small, fast mirrors.  Much easier than testing a 28" F/1.1 hyperbola!

	Mike Lockwood

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