Re: [ATM] Slit and source arrangment part 3.

["Bob May" <rmay@nethere.com>]

OK, try this.
Convert the error into a horizontal error and a vertical error
and look at each of them seperately.  Since the Foucault test
only looks at the equator of the mirror, you can do this.  With
the slitless tester where the source and KE are one above each
other, the horizontal error will be 0 as the two focii are at a 0
distance relative to each other.  While this distance between the
two focii are different because the source is generally below the
COC point of the mirror and the KE reading the results is above
the COC point, this error shows up as an error from pole to pole
of the mirror's surface.  You don't see that error when you're
looking at the shadows at the equator.
Bob May

rmay at nethere.com
http: slash /nav.to slash bobmay
http: slash /bobmay dot astronomy.net

----- Original Message -----
From: Jerry <wa4guu@verizon.net>
To: 'Bob May' <rmay@nethere.com>
Sent: Tuesday, April 29, 2008 12:23 PM
Subject: RE: [ATM] Slit and source arrangment part 3.


I think I see your point. I don't think you are seeing mine, but
I have not
done a good job of explaining it. I have thought about a better
way of
saying it.

Using the slitless tester with knife vertical:
I'm thinking of the surface under test as part of, as Nils Olof
described it
the solid ellipsoid with an equator (a circle) that coincides
with the
diameter of the mirror under test. A spherical mirror surface
centered
anywhere on the equator would test as a null across that
diameter. I think
that is the point you are getting at. I see that.

The center of that circle (center of curvature of the mirror) is
not at the
foci, it is midway between them. When testing the parabola the
knife
movements to null various zones is a measure of a difference of
length from
that zone of the mirror to the foci, not to the mirror's center
of
curvature.

That seems to me to be similar to the situation of lateral offset
of the
vertical knife from a vertical slit. The micrometer readings of
the zones
are not precisely measures of the change in ROC. That is my
point.

Jerry


-----Original Message-----
From: Bob May

Jerry, you're missing the point here.  When doing the Foucault
test, there will indeed be astig. showing with the slitless
tester BUT there are two things that migitate the problem.
First is that the false astig. is showing in the vertical
direction and the little secret here is that you have that part
of the mirror masked off with the zonal mask.  A full Couder mask
will show a very little bit with a fast mirror as the holes in
the mask are biased to one direction and the other as you go
through the zones.  This will show up on the calculations as a
zig-zag of the readings on a perfect mirror across the perfect
zero error line.
The second is that you can't move that astig. around as yo test
for other angles across the mirror.  The error just sticks to the
vertical axis and thus can be ignored as it is an artifact of the
test procedure rather than reality of the errors in the mirror.
If you're doing the test with the Everest pin method of finding
the zones, you may see the astig. in the vertical part of the
mirror but the results are valid as you're not checking the
vertical problems but rather the horizontal ones.
I'll also note for those who are barely following this
discussion, that the error is, on a typical mirror, within the
tolerance of what you can see on the mirror to begin with and it
really takes an extremely good tester eye and a fast mirror to be
able to detect this problem.
Bob May

rmay at nethere.com
http: slash /nav.to slash bobmay
http: slash /bobmay dot astronomy.net

----- Original Message -----
From: Jerry <wa4guu@verizon.net>
To: 'Bob May' <rmay@nethere.com>
Sent: Sunday, April 27, 2008 8:19 PM
Subject: RE: [ATM] Slit and source arrangment part 3.



Bob,

You seem to be saying the slitless implementation to be 100%
accurate
because the offset of source and image is perpendicular to the
knife, and
the laterally off set vertical knife and slit arrangement
inaccurate because
the offset is in line with the diameter of the test.

Let's assume perfect alignment of the tester.
With the slitless tester the optical axis passes midway between
the source
and image on the knife. With perfect alignment of the tester
movement the
source will remain at the same distance from the mirror's optical
axis and
therefore so will the image. The angle of reflection off the
mirror changes
as the distance of the knife changes. As the knife is moved from
the null of
the mirror's center to the null of the mirror's edge the angle of
reflection
becomes smaller. Doesn't that throw a kink into the micrometer
reading just
as a laterally offset knife would? Trivial of course, just as I
think the
lateral offset is. Are you sure which of those kinks is larger?

Maybe I am mistaken about what you think of the magnitude of the
errors.
Are you saying that with a separation of say 1/2" or less that
you can see
that a spherical mirror looks to be an over-corrected ellipse? Or
are you
saying that you get different test readings using a slitless
tester compared
to a laterally offset knife?  How much difference in measurement
do you get?
Jerry

-----Original Message-----
From: Bob May

Actually, you are wrong here.
A slitless tester will not show errors in the direction of the
line between the source and reading edge.  Typically this is in
the vertical direction.  It will show accurately errors in the
horizontal direction tho.  Putting the source and the KE
horizontally won't show any problems in the vertical direction
but will show all of the error in the horizontal direction and
those, assuming a vertical alignment of the KE, will definitely
be in error.
Only the alignment where the distance between the source and the
KE from the mirror will show a proper null for a figure of
revolution properly.
Bob May





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



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