Re: [ATM] Simple Foucault tester for astigmatism

[Mike Lockwood <melockwo@uiuc.edu>]

Hi,

Marc J. Sibilia wrote:
> Continuing Ken Hunter's idea, you could use a V groove instead of
> an L and use a conventional Foucault tester.  Put the LED at the
> vertex of the V and using the left angled edge or the right angled
> edge by cutting it in or out.  You have to get used to the idea of
> the knife cutting in from two different sides

I was going to build something like this (drew some plans), but I
haven't got around to it yet.  A raising an lowering of the angled
edges to adjust the illumination of the zones would probably work,
rather than a horizontal insertion of the KE as is conventionally done.

I think a lot can be gleaned with a "conventional" KE tester by simply
removing the mask and looking at the shadows on the mirror.  The
mirror should be rotated at several different angles, and shadows
examined for asymmetry at all angles.  If there is an astigmatism
problem, I believe this will reveal at least hints of it.  (At least I
hope it will.)  Doing this examination will surely improve one's skill
at observing shadows and becoming attunded to their peculiarities.

As others have mentioned, the most sensitive test for astigmatism
using a conventional Foucault tester will be done when the mirror is
close to spherical.  A careful examination at this state (rotating the
mirror on the test stand, without a mask) should be done, and the
shadows should look the same no matter how the mirror is rotated.  If
the shadows change consistently with rotation, something is wrong.

Test stand astigmatism may vary, however, since the edges of the
mirror may not be even, and the edge pressures will almost certainly
change with rotation (as has been found by those doing
interferometry).  This will wreak havoc when looking for shadow
asymmetricity on thin mirrors, but it shouldn't be an issue with
smaller mirrors.

Personally, after making nearly 30 mirrors, I haven't had any problems
(that I know of) problems with astigmatism.  I check the anneal of the
mirrors that I work on, and take care to support them as evenly as
possible.  I grind the back flat if I am using abrasives on the
mirror, and I try to avoid refiguring mirrors with uneven backs.  Each
time I work on the mirror, if it is on the bottom, it is placed on the
work stand in a different orientation.  I also work around the mirror
in even numbers of turns.  Those are my main precautions, and they
have served me well.

J Killea wrote:
> I have gone ahead and star tested this newly oblate, TUE, sphere,
> and I do see some elliptical shape to the diffraction pattern,
> although much less than before.  **Also, the ellipse seems
> non-existent when centered in the field, but as the star slides up
> and off the diagonal that's when the ellipse really shows up
> stronger.**   So, I will rotate the mirror 120-degrees in the cell
> and test again tonight.

Star testing is the ultimate test for astigmatism.  The caveat here is
that for faster mirrors, ***if the collimation is off at all, 
astigmatism will be seen and can easily be mistaken for error on the 
glass!***  Careful collimation will solve this, and a tracking 
telescope will keep the star in the center of the field, where the 
image must be evaluated.

Additionally, cooling fans and heated diagonal mirrors can induce
astigmatism in the primary and secondary, respectively.  Poorly made
secondaries will cause astigmatism.  Rotating the primary in its cell
will tell where the problem lies, but if the primary is unevenly
cooled, the astigmatism will at first seem to rotate with the primary,
and then the astigmatism will move back to the previous orientation.
(One fan blowing horizontally across the center of the face of the
mirror can cause astigmatism, because it doesn't cool the face evenly.)

	Mike Lockwood

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