ATM Distant-source testing

[Aplanatic@aol.com]

Howdy,

Davey asks:

> P.S.-Does anyone remember the trick of null testing a mirror (or a lens)
> with the optics trained on a very far distant and very small electric light
> source? Might be pretty neat for null figuring, if the tube is already to
> take the optics, and one could run the combo out into the country and set up
> on some kind of cradle, or better yet, live in a place 
> where there is a very
> distant light source viewable from one's open garage door...

There is really no trick to it.  If the light source is small enough and far enough away then the test can be treated as a nearly-null test.  The trouble is:  for large, fast systems the light source has to be a long way away.  I calculated and posted the distance needed a few years ago. 

When the light source is not far enough away to get a decent null, another approach is the offset Dall null test.  Place the light source as far way as possible and then use a nulling lens in front of the knife edge.  This nulling configuration can be designed/analyzed in a ray tracing program, and the tolerances are achievable.  I have used this approach with an inexpensive lens to test several optical systems to high accuracy.  The cheap lens must be accurately measured to do so.

A third approach is to dispense with the nulling lens entirely in the above.  Calculate the longitudinal aberration one would expect from the system if it were perfect and then figure to that specification.  Or, calculate the effective residual conic constant one would expect from that system, and use this to evaluate the performance, and for figuring.

Dave Rowe