[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]

[ATM] Exact Conic Fit.

Sixtests finds the mirror surface as precisely as possible, interpolating 
test readings and integrating differential equations which model the test 
geometry.  Best-fit conic sections are found by first fitting 5 symmetric 
Zernike polynomials and then fitting a conic section to those 5 
coefficients - but such a second-order fit is just an approximation.

Steve Koehler recently pointed out that it may require many Zernike 
polynomials to do an adequate job of representing a "zony" mirror.  For 
example, the 5th symmetric Zernike polynomial only has 3 high/low zones 
from center to edge.  An alternative is to do an exact conic fit to a "high 
frequency" surface output (Sixtests outputs a file with surface heights 
every mm of radius).  This is of value when figuring a mirror with a conic 
target, such as a Cassegrain secondary, or primary and secondary surfaces 
of a Ritchey-Chrétien system.  This download has an executable and a sample 
surface file (my RC cass primary with lotsa zones due to HCF ripple):

         http://home.earthlink.net/~burrjaw/public/sixsrf.zip 
(2005-01-21, Win32, 240k)

James - this is an improved version of the program I sent you sometime ago.

         -- Jim Burrows
         -- mailto://burrjaw@earthlink.net
         -- http://home.earthlink.net/~burrjaw
         -- Seattle N47.4723 W122.3662 (WGS84)  

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