RE: ATM A fairly good mirror, I think

[Aplanatic@aol.com]

Dear Glass Pushers:

As we all should know by now, the P-V rating of a mirror is not the best measure of mirror quality.  Using this old and tired metric leads to these not-very-useful discussions of whether a mirror is 1/8 wave or 1/20 wave, etc.  For instance, I can cook up an example of a "1/4 wave P-V" mirror that performs better in an RMS sense than a "1/10 wave P-V" mirror, etc.

It should be clear that the Strehl ratio is a much better way to charaterize the performance of a mirror.  So, the interesting question to ask is:  Can an experienced tester reliably distiquish (in the presence of all errors, including systematic ones) the difference between a mirror having a Strehl of 0.8 and one having a Strehl of 0.9?  If so, then I consider the test adequately good to figure a mirror of very good or excellent quality.

The accuracy of the test, of course, is dependent upon the diameter and f/ratio of the mirror.  So, under some reasonable error assumptions, at what diameter and f/ratio does Foucault stop meeting the above accuracy criteria?

Mike Peck, have any comments?

Dave Rowe