[ATM] Fw: Sixtests and P-V Error

[burrjaw at earthlink.net (Jim Burrows)]

At 2004-02-27 11:58 +0530, Ulhas Deshpande wrote:

> >  For example, I sent my mirror off for coating when I got a Strehl of 0.959
> > and an RMS error of 9.9 nm.
> >  Does this translate to an RMS P-V error of 9.9/550= 1/55.5 waves?
> >  Similarly the Peak to Valley diff on the surface error curve is 40 nm i.e
> >  40/550= 1/13.75 waves

Ulhas is right except for the terminology.  Sixtests outputs the surface 
RMS; you have subtract the min deviation from the max deviation on the 
error curve to get the surface PV - there's no straight-forward 
relationship between RMS and PV.  PV is not as good a mirror evaluator as 
RMS, but it does give a vague indication of how much more figuring effort 
lies ahead.

To get the wavefront RMS and PV (in nanometers), simply multiply those 
numbers by 2.  Finally, to get those numbers in waves, divide by your 
choice of light wavelength (standard is 550 nm (you get more impressive 
numbers using deep red, say 700 nm <g>)).

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