From: "Vladimir Galogaza" 
To: "ATM shore" 
Reply-To: "Vladimir Galogaza" 


Scott,

Your calculation as it seems to me is assuming that entire 1/8 wavelength
error comes from mirrors  sphericity. So you calculated its ROC from its
sagita
and then back the sagita for smaller part of the same mirror.

Usually 1/8, or whatever,  is statistical error of the mirrors surface.
Sort of average. By using smaller part of the mirror by no means warrants
that it is better than whole. It can be even worse. This is so even if 1/8
is PV, that is nonstatistical error, if both P and V happened to be
"illuminated" in both cases.

Results of your calculations were for the perfect sphere and therefore the
outcome is correct. But this is seldom in practice that error is only of
one particular and known type.

Is there an error in this kind of reasoning?

Vladimir.

--- BBBS/NT v4.00 MP