To: 
From: MLThiebaux 
Reply-To: MLThiebaux 


>...
>An error of lambda/3.5 is greater than that stated in the 'Rayleigh
>criterion', which seems to be an excessive error to me, especially for an
>on-axis image.  With your chosen example of a perfect 16" f4.5 objective,
>would any owner of such a mirror knowingly accept an error that large in a
>diagonal?   Perhaps for specific purposes, but not for general observing...
>...

Sorry if I suggested that one should/might accept these errors.  I was JUST
going through a dreamed-up example to show quantitatively that a largish
error in a diagonal is equivalent to a smallish error in the primary (where
we feel on more familiar ground.)  The underlying formulas were based on
elementary ray tracing with enough detail given to customize the arithmetic
to one's own interests.

The equivalent error is directly proportional to the diagonal error and
also scales as (DL/Fd)^2 where D=primary diameter, L=diagonal-to-image
plane distance, F=primary focal length, and d=diagonal diameter.

Martial Thiebaux
Rawdon Hills, Nova Scotia

--- BBBS/NT v4.00 MP