RE: ATM An $18.00 Null Test

["Wagner,Rick [CIS-ISB]" <Rick.Wagner@EC.GC.CA>]

This is the Ross null test - lens is between the (whole) tester and the
mirror.  You want a high quality (since light traverses it twice)
plano-convex lens with a focal length in the 10-20 inch range depending on
mirror size and f/ratio.  The best thing to do is to ray-trace the lens
mirror combination for minimum spherical aberration.  There is also software
available either from Mel Bartel's page for the original pascal version by
Doug George.  I am working on a windows version which currently works but
which does not yet have the requisite copyright notices etc.  The test is
described in detail in TM 39 (I think) and TM 44 (also I think) described
the software.
Rick Wagner
Ottawa Centre  RASC
45*27'16.6"N  76*15'30.1"W


> ----------
> From:         Bertapelle, Ken[SMTP:Bertapelle.Ken@tci.com]
> Reply To:     Bertapelle, Ken
> Sent:         Tuesday, 27 July 1999 11:41
> To:   Aplanatic@aol.com; atm@shore.net
> Subject:      RE: ATM An $18.00 Null Test
> 
> 
> Dave, et. al.,
> 
> I have a question about null tests that's been nagging at me for
> a while.  I have a slitless focault tester that works as follows:
> An intense light source is delivered via fiber optic cable to a
> little gadget that I'll refer to as the "viewing node".  The node
> has a small opaque window in it that points toward the mirror and
> allows the light to shine on the mirror.  Also mounted on the node
> and right next to the light window is a sliver of razor that acts 
> as the knife edge for the focault test.  The whole node, not 
> counting mounting hardware, is about the size of a dime, and maybe
> 3/8" thick.  (BTW, neither the design nor the construction are my
> own - a friend has loaned me the tester so I can make a copy.)
> 
> I've found that this focault tester works very well because it 
> delivers a *lot* of light, and the shadows are much easier to 
> see than those with the first tester I made patterned after the
> design in Sam Brown's book.
> 
> However, the light source and knife edge move together on the 
> viewing node, and they are only separated by about 1/8".  Thus, in
> any null test the light would have to go through the lens twice -
> once on its way to the mirror, and once on return.  My question is
> whether a null test can be made to work with such a tester.  If
> so, what kind of lens does one use?  
> 
> Thanks in advance.
> 
> - KenB
>