Re: [ATM] Ross null test?

["Mark Suchting" <masuch@cia.com.au>]

Ric,

In summary your response to my point about the difficulty and importance of
making very good surfaces for a double pass lens is  that `1/10 wave lens
wavefront accuracy to make a 1/10 wave mirror ' is ok.    Most people
producing a null lens would like to see as little error in their set up as
possible, so that the Null is truly a good wavefront mirror, and you aren't
just polishing a bad lens wavefront into your mirror.   Your suggestion to
use the mirror itself as a test plate means the Ross lens must have an
unnecessary large diameter ( maybe three times bigger) used aperture, making
the job of fabricating the lens to this aperture even harder. Any old piece
of plate glass itself will have an  unknown refractive index too.

Making a good surfaces with freedom of zones to 1/20 wave is the minimum I'd
consider needed over the diameter for a good test wavefront, and not easy
for any lens maker to produce.  Same applies to any Null test reference , a
reflecting one must be even better, ( although Dall is more forgiving ) .
I'm not trying to put anyone off it , but I don't think trivialising the
importance and care that has to be taken for good surfaces, is doing the ATM
community any service.

About the only time I pipe up these days is to throw a bit of caution into
the ring and it must  get very boring . You on the other hand strike my as a
zealout, the wavefront figures to `1/500' wave in the same sentence as
`plate glass' are dead give away  ;-)

Mark Suchting


----- Original Message ----- >

 Well,as you know the longer the focal lenght of the
> lens the "better" the null..yes?
> So from my experience of using Ross lenses from 4 inch
> to 13 inch in size and focal lenght from 15 inches to
> 91 inches,depending on the mirror being tested you can
> with long focal length Ross elements get some really
> outstanding correction on the surface to the point of
> being unbelieveable - such as 1/500 wave on a 8 inch
> f/6 mirror using a 91 inch focal lenght Ross!
>
> If the atm uses a plate glass tool on said 8 inch f/6
> mirror and makes that into the Ross element with a 96
> inch r.c. then even if the surface of the convex side
> is not exactly spherical there may be some deviation
> from an exact 1/500 wave surface but it would be a
> good null none the less.
>
> It's all in the focal lenght of the Ross element as
> your calculations well show.Optical glass is nice but
> not needed.We test cass secondaries through the back
> using pyrex - not ideal - but it can work and th
> Foucault can be used to see surface roughness.
>
> I'm not saying the convex surface can be highly
> irregular in profile,spherical is the ideal but errors
> seen from a Ross lens that will show 1/500 wave
> correction can be disregarded if the goal is only 1/10
> wave.As you know there are many other variables that
> affect the test results so the longer the focal lenght
> the better.If  stuck with a 3inch Ross  it had better
> be a good one - with that I agree but the atm
> community hasn't explored the limits of the  Ross null
> test.Drop the idea of using just optical glass,include
> the idea of very long focal lenght and the Ross is
> "forgiving".After all who wonts a mirror rated to
> 1/500 wave....In the software provided by George there
> is an error "box" that shows the degree of shift for
> 1/4 wave tolerance - this box gets larger as the focal
> lenght of the Ross element gets longer.
>
> Of course in picking a plate glass tool one needs to
> test it for extreme straie in the material.
>
> Views??
>
> Ric
> Toronto
>
> P.S.
> Do you still wont any of the 4 inch Zerodur squares?
>
>
>
>
>
>
> _______________________________
> Do you Yahoo!?
> Express yourself with Y! Messenger! Free. Download now.
> http://messenger.yahoo.com
>

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