Subject: [interferometry] Digest Number 2044
From: interferometry@yahoogroups.com
Date: 8/17/2012, 4:15 AM
To: interferometry@yahoogroups.com

interferometry
Yahoo! Groups
interferometry Group

11 New Messages

Digest #2044
1a
Can a Bath effective test a 1 meter f:3.66 paraboloid? by "Steve Follett" steve_follett
1b
Re: Can a Bath effective test a 1 meter f:3.66 paraboloid? by "atmpob" atmpob
1c
Re: Can a Bath effective test a 1 meter f:3.66 paraboloid? by "Arjan (udjat)" atm_6_6_6
1d
Re: Can a Bath effective test a 1 meter f:3.66 paraboloid? by "Stephen Koehler" steve_koehler
1e
Re: Can a Bath effective test a 1 meter f:3.66 paraboloid? by "Michael Peck" mikepeck5440
1f
Re: Can a Bath effective test a 1 meter f:3.66 paraboloid? by "Michael Peck" mikepeck5440
1g
Re: Can a Bath effective test a 1 meter f:3.66 paraboloid? by "Pete" peterpekurar
1h
Re: Can a Bath effective test a 1 meter f:3.66 paraboloid? by "atmpob" atmpob
1i
Re: Can a Bath effective test a 1 meter f:3.66 paraboloid? by "pjifl@bigpond.com.au" pjifl
1j
Re: Can a Bath effective test a 1 meter f:3.66 paraboloid? by "vladimir galogaza" vladimirgalogaza
2a
Just to clarify by "m_scherman" m_scherman

Messages

1a

Can a Bath effective test a 1 meter f:3.66 paraboloid?

Thu Aug 16, 2012 2:15 pm (PDT) . Posted by:

"Steve Follett" steve_follett

I'm a member of a group who are constructing a 40 inch
telescope for Robert Ferguson Observatory in Sonoma County, CA.

The mirror we are working on is

40 inch F:3.66 paraboloid
2.25 inch thick (at the edge)

Can I use a Bath interferometer to reliably test a mirror
this big and this fast?

I posted this question a couple years ago, and it wasn't clear
then that it would work well. Some posts suggested that a PSI
Bath might work. In the meantime we went ahead and made the
mirror and tested it as best we could.

We are now at the point that I would like to switch to
interferometry if we can.

We have tested up to this point using the King wire test and
Foucault testing. We were able to tell we had astigmatism by
testing on several axes and comparing the results.

We used a test that Richard Schwartz described in a post to
the Amateur Telescope Making listserv to measure the
astigmatism. The test used a mask with holes in a 'X' pattern,
similar to a Hartmann mask. The mask was set up so that the
pattern fell on the 45 degree axes of the mirror.

We then measured the ROC of each axis. Any difference between
the two axes would indicate astigmatism and allow us calculate
the size of the error (on those two axes). We then rotate the
mirror 22.5 degrees and tested again until we had measurements
on 16 axes (i.e. every 22.5 degrees).

We had measurable astigmatism initially of about 5 waves. We
corrected this using local polishing to the point that we
couldn't reliably detect it anymore.

We have as a result a mirror with some remaining astigmatism
(detectable a star test) and probably a very bumpy figure from
the local polishing. The problem is that we are seeing things
in a star test that I don't understand. The image does come to
a focus but outside of focus two lobes appear to collapse back
toward the center of the pattern.

Here's a link to video of the star test:
https://dl.dropbox.com/u/74884057/Project%2040%20StarTest%2020120723.mp4

I would really like to have a surface map to understand what's
going on and improve the mirror.

BTW, I want to try making a PSI Bath and I need a source for a
polarizing beam splitter and a suitable wave plate at minimal
cost.

Thanks,
Steve Follett

1b

Re: Can a Bath effective test a 1 meter f:3.66 paraboloid?

Thu Aug 16, 2012 2:34 pm (PDT) . Posted by:

"atmpob" atmpob

If you can expand the laser enough to cover the mirror it can be done using a Bath and OpenFringe. Since it is so cheap to build a Bath interferometer it can not hurt to try it with the standard configuration.

You will need a good DSLR camera to capture the igrams because of the fine narrow fringes you will need to get. Most any DSLR 6 years old or newer will do.

Dale Eason

--- In interferometry@yahoogroups.com, "Steve Follett" <steve_follett@...> wrote:
>
> I'm a member of a group who are constructing a 40 inch
> telescope for Robert Ferguson Observatory in Sonoma County, CA.
>
> The mirror we are working on is
>
> 40 inch F:3.66 paraboloid
> 2.25 inch thick (at the edge)
>
> Can I use a Bath interferometer to reliably test a mirror
> this big and this fast?
>

1c

Re: Can a Bath effective test a 1 meter f:3.66 paraboloid?

Thu Aug 16, 2012 2:44 pm (PDT) . Posted by:

"Arjan (udjat)" atm_6_6_6

You need a high numerical aperture measurement beam to cover the whole
mirror. This means either a very short focus lens in the Bath or widen
the beam from the laser.
I did the latter for an F/3 and F/2.4 mirror respectively, although not
as big as yours. For beam expansion I took off the laser lens, and
replaced with a longer focal length F/1 projector objective. See here:
http://atm.udjat.nl/tester/bathtester.html

Arjan

Steve Follett wrote:
>
> I'm a member of a group who are constructing a 40 inch
> telescope for Robert Ferguson Observatory in Sonoma County, CA.
>
> The mirror we are working on is
>
> 40 inch F:3.66 paraboloid
> 2.25 inch thick (at the edge)
>
> Can I use a Bath interferometer to reliably test a mirror
> this big and this fast?
>
> I posted this question a couple years ago, and it wasn't clear
> then that it would work well. Some posts suggested that a PSI
> Bath might work. In the meantime we went ahead and made the
> mirror and tested it as best we could.
>
> We are now at the point that I would like to switch to
> interferometry if we can.
>
> We have tested up to this point using the King wire test and
> Foucault testing. We were able to tell we had astigmatism by
> testing on several axes and comparing the results.
>
> We used a test that Richard Schwartz described in a post to
> the Amateur Telescope Making listserv to measure the
> astigmatism. The test used a mask with holes in a 'X' pattern,
> similar to a Hartmann mask. The mask was set up so that the
> pattern fell on the 45 degree axes of the mirror.
>
> We then measured the ROC of each axis. Any difference between
> the two axes would indicate astigmatism and allow us calculate
> the size of the error (on those two axes). We then rotate the
> mirror 22.5 degrees and tested again until we had measurements
> on 16 axes (i.e. every 22.5 degrees).
>
> We had measurable astigmatism initially of about 5 waves. We
> corrected this using local polishing to the point that we
> couldn't reliably detect it anymore.
>
> We have as a result a mirror with some remaining astigmatism
> (detectable a star test) and probably a very bumpy figure from
> the local polishing. The problem is that we are seeing things
> in a star test that I don't understand. The image does come to
> a focus but outside of focus two lobes appear to collapse back
> toward the center of the pattern.
>
> Here's a link to video of the star test:
> https://dl.dropbox.com/u/74884057/Project%2040%20StarTest%2020120723.mp4
>
> I would really like to have a surface map to understand what's
> going on and improve the mirror.
>
> BTW, I want to try making a PSI Bath and I need a source for a
> polarizing beam splitter and a suitable wave plate at minimal
> cost.
>
> Thanks,
> Steve Follett
>
>

1d

Re: Can a Bath effective test a 1 meter f:3.66 paraboloid?

Thu Aug 16, 2012 3:18 pm (PDT) . Posted by:

"Stephen Koehler" steve_koehler

Steve,

Besides needing a wide enough beam to cover the short focal ratio mirror,
you also need to consider the waves of SA caused by testing a paraboloid at
focus. This value is related to the mirror diameter and is inversely
related to the focal ratio. If you have too many waves of SA, you will
have a very difficult time creating interferograms, because you can't avoid
having very fine fringes.

To give you an idea of where your 40" f/3.66 falls in the range of SA
values, you will have a Zernike SA of -12. This is doable, with a standard
Bath and the Vortex method in OpenFringe, but it will be a bit
challenging. An SA of less than about -10 becomes increasingly difficult,
with -16 about as far as you can go.

You may have seen this before, but I did a quick feasibility study for a
20" f/2.6, which has a Zernike SA of -16.7. I demonstrate with a simulated
interferogram that under ideal conditions it can be analyzed. This may
give you some idea of what you will be up against. Here's the report:

http://tech.groups.yahoo.com/group/interferometry/files/Steve%20Koehler/Gert%20Mirror/rc_20/gert_20.html

I would say, go for it. It you can get good, clean igrams, you should be
able to analyze them.

I'm very curious to see what gives rise to the very strange star test
patterns you showed in the video. You didn't follow Richard Schwartz's
advice on tight clamps, did you. :-)
--
Steve Koehler

1e

Re: Can a Bath effective test a 1 meter f:3.66 paraboloid?

Thu Aug 16, 2012 3:53 pm (PDT) . Posted by:

"Michael Peck" mikepeck5440

On 8/16/2012 4:15 PM, Steve Follett wrote:
> I'm a member of a group who are constructing a 40 inch
> telescope for Robert Ferguson Observatory in Sonoma County, CA.
>
> The mirror we are working on is
>
> 40 inch F:3.66 paraboloid
> 2.25 inch thick (at the edge)
>
> Can I use a Bath interferometer to reliably test a mirror
> this big and this fast?
>
> I posted this question a couple years ago, and it wasn't clear
> then that it would work well. Some posts suggested that a PSI
> Bath might work. In the meantime we went ahead and made the
> mirror and tested it as best we could.
>
> We are now at the point that I would like to switch to
> interferometry if we can.
>
>

The suggestion that a PSI Bath might work was based on the fact that you
could test at zero tilt. If I did my arithmetic correctly you will have
about 24 fringes in a bullseye pattern at 0 tilt and defocus, which
should keep you within the Nyquist limit everywhere with a reasonable
camera.

You might be able to do the same with static interferometry and vortex.
You probably won't get good results for the center of the mirror, but
you really don't need to care about that since it will be obstructed anyway.

I'm not sure why someone who would spend 2 years figuring a mirror would
be reluctant to take on the extra complexity and couple hundred dollars
expense to build a PSI Bath, but that's your call.

Mike P.

--
Michael Peck
http://wildlife-pix.com
mlpeck54 -at- earthlink.net

1f

Re: Can a Bath effective test a 1 meter f:3.66 paraboloid?

Thu Aug 16, 2012 4:04 pm (PDT) . Posted by:

"Michael Peck" mikepeck5440

On 8/16/2012 5:53 PM, Michael Peck wrote:

Actually, to be fair I should add that you need a pretty quiet testing
environment to use PSI effectively, so if vibrations or air turbulence
are significant problems you might have no choice but to use static
interferometry.

Mike P.

--
Michael Peck
http://wildlife-pix.com
mlpeck54 -at- earthlink.net

1g

Re: Can a Bath effective test a 1 meter f:3.66 paraboloid?

Thu Aug 16, 2012 5:33 pm (PDT) . Posted by:

"Pete" peterpekurar

"I'm not sure why someone who would spend 2 years figuring a mirror would
be reluctant to take on the extra complexity and couple hundred dollars
expense to build a PSI Bath, but that's your call"

I'll tell you why... He wants to be sure the technique yields an excellent mirror. I asked the same questions regarding my 24" f/3.3 and the responses at the time were not very favorable (ie. near the limits of the standard interferometer)

Sent from my iPad

On 2012-08-16, at 6:53 PM, Michael Peck <mpeck1@ix.netcom.com> wrote:

> I'm not sure why someone who would spend 2 years figuring a mirror would
> be reluctant to take on the extra complexity and couple hundred dollars
> expense to build a PSI Bath, but that's your call.

1h

Re: Can a Bath effective test a 1 meter f:3.66 paraboloid?

Thu Aug 16, 2012 8:37 pm (PDT) . Posted by:

"atmpob" atmpob

So the alternative is to do nothing? Which of course is unlikely to lead to an excellent mirror. In your case while it may be near the limit it is doable and so is better than not doing so. It has a higher probability of leading to a better mirror than not doing so.

I find when I put in his values to OpenFringe and then have it make a simulated igram that the analysis is very good as long as you can capture the fringes with enough resolution.

It may be hard to get complete coverage of that mirror but it is worth a try. Even if one can't get coverage out to the edge I bet one can figure out the major surface flaws of the mirror. Then once those are fixed then other tests like ronchi can diagnose the edge easily.

Heck they probably spent more on grit and polishing supplies than a PSI Bath would cost. I bet they spent more just on pitch than a Bath would cost.

Dale Eason

--- In interferometry@yahoogroups.com, Pete <peter@...> wrote:
>
> "I'm not sure why someone who would spend 2 years figuring a mirror would
> be reluctant to take on the extra complexity and couple hundred dollars
> expense to build a PSI Bath, but that's your call"
>
> I'll tell you why... He wants to be sure the technique yields an excellent mirror. I asked the same questions regarding my 24" f/3.3 and the responses at the time were not very favorable (ie. near the limits of the standard interferometer)
>

1i

Re: Can a Bath effective test a 1 meter f:3.66 paraboloid?

Fri Aug 17, 2012 3:37 am (PDT) . Posted by:

"pjifl@bigpond.com.au" pjifl

It sounds like one of the main things you want to investigate is
Astigmatism - or at least some non symmetric defect.

Even if the SA when testing a large Paraboloid at COC makes it difficult, as
long as you can get fringes, I think the Bath will still be by far your best
bet for these asymmetrical aberrations. Even if the SA test was not so
useful the results should still be very useful.

Maybe I am wrong - what do others think ??

Peter Smith.

1j

Re: Can a Bath effective test a 1 meter f:3.66 paraboloid?

Fri Aug 17, 2012 4:09 am (PDT) . Posted by:

"vladimir galogaza" vladimirgalogaza

Steve,

>I want to try making a PSI Bath and I need a source for a
>polarizing beam splitter and a suitable wave plate at minimal
>cost.

Polarizing beam splitter cubes:
Roithner Lasertechnik : www.roithner-laser.com
office@roithner-laser.com
PBS-10-ZF1, Pol. Beamsplitter Cube, 10x10x10mm, λ/4, coating 450-650nm, $ 120.75

1/4 wave plate:
Until you find something else, and for quick start
I can send you a piece of 1/4 wave film for 560 nm (free). Give me the address .
(That will require use of the green laser.)

But before you start with PSI (or in parallel),
I suggest to buy from SurplusShed: http://www.surplusshed.com
one unpolarized beamsplitter cube http://www.surplusshed.com/pages/item/l10563d.html
for immediate use in non-PSI Bath ($18)
one biconvex lens: item PL1098 ( $ 5)
One flat mirror: item L3493 ( $ 7.5) ( or extract one from scrapped copiers).
One laser pointer 1-5 mW
And you can start gathering experience in interferometry in a couple of days.

Regards
Vladimir.

2a

Just to clarify

Thu Aug 16, 2012 11:44 pm (PDT) . Posted by:

"m_scherman" m_scherman

Hi Dale
Just to be sure does this all sound right? ... I am using a standard right angle Bath interferometer (laser pointed at mirror, camera at beam splitter). I am getting some very nice contrasty ingrams. I take 10 or so images of the mirror at each orientation (0 degrees, 90, etc). After I do the 0 degree position, I rotate the mirror (as I am facing it in the stand, optical surface towards me) clockwise (CW) 90 degrees for the 90 degree position. Again 90 degree more clockwise for the 180 position and 90 more CW for the 270 position.

Then I analyze each individual ingram photo in OpenFringe, be sure unwrap errors are (much) less than 1000, and and do FFT analysis on each image and save the resuts for each ingram as an FFT Wavefront file. I usually save the cropped ingram image for posterity too as a .png file. During analysis, sometimes (but not always) I get a dialog box saying "the wavefront appears to be inverted, do I want to invert it?" and I always click Yes (is that ok?). After FFT analysis, I usually hit the FFT filter button at 0.06 to clean it up a bit too.

When done I average the FFT Wavefront files to generate an average FFT wavefront file for that orientation. This should average out periodic errors such as air turbulence. Then, I take the 4 average FFT Wavefront files (0, 90, 180, and 270) and average them together as done before. This serves to remove test stand induced astigmatism. Therefore, this last average FFT Wavefront file is the final report of the mirror's surface quality (figure).

Questions: Would you counter rotate each of (say, the 90 degree) FFT wavefront files by 90 degrees first before saving each FFT wavefront file (in Openfringe I would rotate them 90 degrees CW by selecting the "Transforms" tab and the "rotate the Fringe Files 90 CW" option and NOT the "Rotate Fringe CW 90"), or would you do the analysis on all the 90 degree files first without any Transformation, average them, and the counter rotate the average FFT Wavefront file (for that group? Again I have the right angle Bath, so I would assume so.

Thank you very much! I have a nice guide I am preparing with some great build photos too.
Mike

--- In interferometry@yahoogroups.com, "atmpob" <atmpob@...> wrote:
>
> You counter rotate the wavefront file from the FFT analysis. There are no fringe files associated with FFT analysis and wavefront files.
>
> There is no new help file. It has a few updates to include the new features of versions 12_3 but it is pretty much the same as for 10.
>
> The right angle Bath can be used two ways. The way with the laser pointing at the mirror will flip the image. If instead you have the camera pointing at the mirror (my preference) then the image will not be flipped.
>
> Dale Eason
>
> --- In interferometry@yahoogroups.com, "m_scherman" <m_scherman@> wrote:
> >
> > Thank you Dale, that clears it up alot! I am using a right angle Bath. Is is best to counter rotate the igram image in Photoshop first or in OpenFringe? In Openfringe would I counter rotate the fringes or fringe files in Openfringe? I will delve into the new help file tonight too. I hopefully will image the 16 inchg tonight.
> > regards
>

We are making changes based on your feedback, Thank you !
The Yahoo! Groups Product Blog
Yahoo! Groups
CHANGE SETTINGS
>
TERMS OF USE
>
UNSUBSCRIBE
>