Re: [ATM] What shape is this mirror?

[Mark Holm <mdholm@telerama.com>]

In order to interpret Foucault numbers, we need to know if you are using
a fixed or moving source tester.  This means, does you light source stay
in one place while you test, or does it move toward and away from the
mirror with the knife edge?

Either way, Sixtests shows an oblate ellipse, with the edge down
(compared to the best fit oblate ellipse).  Compared to a sphere, or to
a parabola you are high in the middle and high toward the edge while low
at about 67%.  Probably, changing from 1/3 center over center to 1/3 W
stroke would start moving you toward spherical, and possibly past,
toward parabola - hyperbola land.  A W stroke will also help some with
zones, such as your central hole, but do as Ken suggested and make sure
the lap stays well pressed, because if the hole is cause by poor lap
contact, the W stroke will just make a bigger hole.

Sixtests is my favorite Foucault reduction program.  It was written by
Jim Burrows.  Jim is quite good with math, and Sixtests is probably
about as sophisticated mathematically as is useful for Foucault data.
Jim also has a good handle on the limitations, both of the math, and the
optics, so Sixtests doesn't try to do things that are not possible.  You
can get Sixtests at
http://home.earthlink.net/~burrjaw/atm/atm_math.lwp/atm_math.htm  (You
want the Windows version.)

Since you are using inch measurement, and Sixtests uses millimeters
only, you may find my program Frontsix useful.  Frontsix is a front end
program for Sixtests.  It accepts multiple Foucault measurement sets (as
you get close to the parabola, multiple measurements are a good idea)
and also does the unit conversion for you.  It has a single click
interface to send your data to Sixtests.  (Well, it is single click
after you once let it know where to find Sixtests on your computer.)

Frontzix is at http://www.atmsite.org/contrib/Holm/frontsix/frontsix.html

A program that puts fuctionality similar to Frontsix and Sixtests in one
package is Foucault.exe by Andreas Reifke  Find it at
http://www.foucault-test-analysis.de/
Andreas borrowed Jim Burrows' math from Sixtests, so the results are
essentially identical.

Mark Holm
mdholm@telerama.com

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