Re: ATM Autocollmination Flat Tolerance

[Mark Suchting <masuch@dap.CSIRO.AU>]

On Mon, 29 Jul 1996, ric rokosz wrote:

>    e = the desired correction you wont on the mirror (1/10 wave is 0.1 
> ,1/20 wave is 0.05 etc )

Just to clarify this bit which seems to be a bit misleading. This 
variable is where you plug in the maximum allowable abberation that will 
occur on your optics if you figure to a `null' in your test. Its the same 
as in the artificial star test. Allowing curvature on your flat is akin 
to bringing your artificial star up close so that  your incoming light 
is no longer parallel, and you are figuring your mirror to give a perfect 
image of a close object. This is a somewhat different thing than `desired 
correction'. 

For anyone thinking of making an autocollimation flat, freedom from
asphericity and zoning is all that need concern you. A 10" flat with a 30
fringe concavity, actually has a focal length of 150 meters...we are
talking of a long focal length telescope mirror here ;) I think this
formula dating back to 1938 is nothing more than an intellectual
curiosity, because the use of a high quality steel straight edge while
fine grinding will get you an order of magnitude of flateness better than
this..( simply by feel: if the straightedge spins on its middle you are
convex, if it drags on both edge you are cave ). Provided you do a rough
check on flatness in the Ritchey-Common test before you start figuring,
the absolute radius of the `flat' will be of no concern if you are within
the envelope. All you will do is figure for a clean knife cut off ( or
straight ronchi lines oriented horizontal). If your return sphere is F10
or more at the Centre of Curvature you will have great sensitivity. 

Most ATM's with a reasonable even polishing tecnique would be interested 
to know that in polishing their mirrors they are probably holding the 
radius of curvature constant to a few waves of sagitta anyhow..so if you 
have successfully polished a nice sphere, you allready have the skills to 
make an autocollimation flat. Perhaps the word `flat' has connotations 
which prevent more amateurs from producing a very useful piece of optics.

Apologies to all for labouring a point so, but I have seen this thread 
pop up before, and the word really needs to be drummed in.

regards,

mark