Re: [ATM] Hyperbola and TDE - Tenacious

[Jim Burrows <burrjaw@earthlink.net>]

At 2005-01-21 17:51 -0800, J Killea wrote:

>However, it's hard for me to imagine that a mirror with even a little TDE 
>could really have such nice specs.  So I would like to get rid of the TDE 
>once and for all, as well as take out the central deep area.  (Sixtests 
>data file is below.)
>
>I am tempted to try Tex's suggested TOT 1/4 W  stroke ( p.92, Fig. 42 ) to 
>take down the high area and am wondering what anyone's thoughts on that 
>are - too much risk for more TDE?
>
>Below is the Sixtests file after one little round of a center-deepening 
>parabolizing stroke that seems to have taken me in the wrong direction a 
>bit  (deepened the center and the TDE slightly,  and also reduced Strehl 
>and raised RMS error).

Thanks for chucking in the Sixtests input file ("One page of test data is 
worth a thousand Ronchi pictures" <g>).  As you found out, you don't want 
to use parabolizing or W strokes on a mirror with b = -1.24.  Sixtests says 
the RMS relative to that hyperbola is 7.2 nm, while the RMS relative to the 
best-fit parabola is 15.1, so most of that RMS is due to the overcorrection 
[ sqrt(15.1² - 7.2²) = 13.3 ].

Theoretically, any length COC stroke will reduce the correction.  So short 
spells of COC and keep on testing - you might edge in on a 7.2 nm RMS 
relative to a parabola, .974 Strehl, good enough for anybody.

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

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