Re: ATM Foucault - Interferometer Test Comparison

["James W. Burrows" <burrjaw@earthlink.net>]

At 18:24 2001-08-14 -0700, Thomas Bennett wrote:

>    residual wavefront variations evaluated at fringe centers (in waves)
>   # of points     peak        valley        P-V        RMS        Strehl 
> Ratio
>     525             0.149        -0.132       0.281       0.066       0.844
>
>residual wavefront variations over uniform grid (in waves)
>    # of points    peak         valley       P-V         RMS   Strehl ratio
>     2783          0.162          -0.137      0.299        0.064   0.849
>
>So looks like what we
>have here is a comparison of a Foucault test and interferometer test on the
>same mirror. In case anyone is interested, the Foucault test results were:
>
>Optical diameter: 16
>Radius of curvature: 144
>  2 
> h(x)              3.5800     4.8300     5.8200     6.6700     7.4200  8.1000
>  8 D12               0.2160     0.2610     0.3300     0.4170     0.5000 
> 0.5380
>Maximum wavefront error = 1 / 3.1 wave at zone 1

1)  Their Strehl ratios are slightly different from the Majahan 
approximation in terms of RMS, but no quibble about differences in the 
third decimal place, since only 2 digits in wft RMS:

                         surface         M approx.
         wft RMS RMS, nm         Strehl ratio
         .066            18.2            0.842
         .064            17.6            0.851

2)  Running the Foucault readings into Sixtests, I get only marginal 
agreement:  surface RMS = 23.8 +- 7.2 nm (assuming +-.002" readings), 
Strehl ratio = 0.743.

The interferometer results are barely within the 1-sigma RMS confidence 
interval around the Foucault results, so the difference could be Foucault 
reading errors.  Another way out is the possibility that the Foucault 
readings were along a "bad" diameter and the rest of the mirror is better.

         -- Jim Burrows
         --              mailto:burrjaw@earthlink.net
         -- Seattle      N47.47233, W122.36620 (WGS84)