RE: [ATM] Astigmatism in round robin mirror C with Foucault.

["Koehler, Steve" <Steve_Koehler@securecomputing.com>]

Dale,

t I'm going to do tonight is take the rond robin
> measurements from both diagonals compute surface
> profiles like I have been doing.  Then I will do a
> least square Zernike fit to both diameters and see
> what I come up with.

I'm curious how you will do the Zernike fit.  I don't think it makes sense to
try to fit both sets of data simultaneously to try to get a 2D approximation of
the astigmatism.  To do this would require that you rotate the mirror without
disturbing the distance to the tester.  Once you disturb the distance to the
tester, you won't be able to "see" any focus shift difference between the two
diameters, so the primary astigmatism will "disappear".

If you're going to do Zernike fits of the two diameters individually, how will
you use this to compute astigmatism?

I have a few thoughts on detection of primary astigmatism.  I think it's very
difficult to quantify with any sort of knife test, unless you can preserve the
mirror/tester distance across rotation (rotate the tester?).

However, it's is pretty easy to detect qualitatively in a null setup with the
Ronchi test.  Just center a Ronchi line and move it backward and forward through
focus.  If astigmatism is oriented at 45 degrees from vertical, the Ronchi line
will first lean one way, then lean the other way.  I have some nice simulations
that show how this looks.  Rotate the mirror in small increments until the
effect is at a maximum, and you can tell the orientation of the astigmatism.
This works very well for testing a spherical mirror at COC before parabolizing.
I'm wondering if the effect can be seen in a holographic null test, if the slit
is replaced by a slit with an embedded Ronchi grating or wire.  It seems like
this might be feasible for testing astigmatism on a finished paraboloid.  (Dale:
perhaps we should try this out.)

Finally, I was involved in some discussions on the Zambuto group regarding
detection of astigmatism with the knife test.  My position was the same as what
Scott Milligan just recently posted, i.e., that PRIMARY astigmatism is not
detectable by measuring two diameters.  Carl says that he IS able to detect
astigmatism by seeing different surface profiles on two different diameters, but
I contend that these are higher-order effects, and not primary astigmatism.
Carl's final word on this was that pure primary astigmatism probably does not
show up on real mirrors, because of they way they are figured.  I don't believe
this is the case, because any actual astigmatism can be considered a mixture of
different asymmetric modes, and the lower order ones (e.g., PA) are likely to be
the largest ones.

-- Steve Koehler
   steve_koehler@securecomputing.com

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