Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

Unmasked Foucault testing and related topics

  • Please log in to reply
66 replies to this topic

#26 mark cowan

mark cowan

    Vendor (Veritas Optics)

  • *****
  • Vendors
  • topic starter
  • Posts: 8680
  • Joined: 03 Jun 2005
  • Loc: salem, OR

Posted 01 June 2016 - 07:17 PM

My previous life.  :lol:


  • brucesdad13 likes this

#27 mark cowan

mark cowan

    Vendor (Veritas Optics)

  • *****
  • Vendors
  • topic starter
  • Posts: 8680
  • Joined: 03 Jun 2005
  • Loc: salem, OR

Posted 01 June 2016 - 07:22 PM

Mark,

 

   When you do your write-up with the fresh data, please include comments about methods you may use to combine or correlate the data from three runs on different axes.  That is something I have never found a satisfying way to do. The best I have been able to do is treat each as a totally separate run and simply visually compare the profiles given in SixTests or FigureXP.

 

   I really like this test and have used it on a meager number of mirrors since 2001. I am a big fan of the ease of use and repeatability of the testing. I also look forward to more of your commentary on the best-practice methods and results you have found.

 

   Your automated analysis would be a welcome relief to the image processing / manual analysis methods I use. It only takes me about 10 minutes to run 15+ zones on a mirror but processing the image data in PixInsight takes at least 45 minutes before I can load up the final data into SixTests.

 

 

I don't correlate those, I just superimpose them.  They are after all different profiles of the same mirror.

 

But what I've found striking so far with the 16" (a lot of tedium involved manually!) is the two axes I tested on agree almost completely digitally, closer than the manual tests do.  I'm learning a lot about how to implement it by working through the process in detail manually and I'll be explaining that as I go along.

 

I'm trying to catch up on earlier comments now and will be supplying the data and results later on.



#28 mark cowan

mark cowan

    Vendor (Veritas Optics)

  • *****
  • Vendors
  • topic starter
  • Posts: 8680
  • Joined: 03 Jun 2005
  • Loc: salem, OR

Posted 01 June 2016 - 07:52 PM

Nice work, Mark. A couple of things -- just my observation, not critcism. I'm wondering how long does it take to measure 7 zones (I suppose several times for a good average), then data processing. It seems that this would be a very tedious process during the final stage of figuring -- when it counts the most -- as figuring spells get shorter, cooling periods longer and longer.

 

...

 

Please, don't get me wrong, your method is fascinating, and I love your presentations, but it seems terribly time-consuming and tedious to me, as  I fail to see where the advantage is -- especially if it's applied where in counts the most -- in the final phases of figuring a mirror, when figuring spells last a couple of minutes at most. Hopefully you can see my point and not take this comment the wrong way. Apparently, the method seems to work for you, and that's all that counts in the final analysis.

 

 

 

It is tedious by hand, manual Foucault is a lot faster which is why I use it and have for many years. :lol:

 

This test looks like it reaches 1/50th wave precision in terms of the absolute measurement in it's fairly raw form, and with software averaging of the image "null" radii pixels that should reduce down to the limits of the test itself, which is at least 1/100th wave PVW for everything that it's sensitive to.   I will be demonstrating that 1/50th wave potential as the thread goes along most likely.  This test does not measure astigmatism or other errors of figure of revolution, and the bugaboo is always going to be controlling them with other tests, all of which are less sensitive IME.

 

But this is the implementation step for a fully automated unmasked Foucault tester, a device I started building quite a while ago.  It will interface with the robotic figuring system I've got under final construction.  For that kind of precision control quantitative results are necessary, and they confer the ability to converge the figuring directly on to the desired surface profile as it proceeds, with no need for corrective actions here and there that lead to zonal defects, roughness, and all the usual sins of pinpoint figuring.

 

And as I've said before, I don't need the full surface map or even the surface profile of the axis under test, I just need the deviation from the desired slope along the axis of the mirror - that is, the longitudinal position at each "zone" - in order to drive the figuring.  And that's what this produces, after inverting the "zones" and known longitudinals to produce the estimated longitudinals at specific zones - a chart that is the same as the raw numbers read from Foucault (which is almost all I ever use to drive figuring) only with a lot more resolution.

 

Regarding speed, well, even if doing averaging of multiple exposures, it should be able to acquire and process detailed profiles of 20-50 points along a given profile (and produce the necessary analysis as it goes) in 10-20 seconds.  Mostly that speed depends on how long it takes the stage to settle down after each move, as the camera isn't connected to the moving stage but only to the base of the machine.  Manually I can run a 7-10 zone Foucault test in maybe 10 minutes, so that's a whole lot faster regardless.

 

It takes longer to set the mirror on the test stand and line it up, by comparison.  OTOH setting up a Foucault test is dead simple in terms of alignment and the two numbers you need for analysis (ROC and mirror optical diameter).

 

Here's an old picture of the partial build; the crude linear screw mechanism will be replaced with proper M6 leadscrews:

 

leon_partial_stage_small.jpg

 

The absolute accuracy and speed is what it's all about, software takes care of all of the tedium, put the mirror on the stand, line it up, and ZIP it's done...


Edited by mark cowan, 01 June 2016 - 11:01 PM.

  • Chriske likes this

#29 MKV

MKV

    Fly Me to the Moon

  • *****
  • Posts: 7107
  • Joined: 20 Jan 2011
  • Loc: Florida, USA

Posted 01 June 2016 - 10:22 PM

Mark, although you don't mention it,  it should go without saying that any claim to accuracy must be validated by another test, independently, such as an interferometric test. If the results of both agree very closely it's a done deal. 

 

Mladen


  • DAVIDG and brucesdad13 like this

#30 brucesdad13

brucesdad13

    Apollo

  • *****
  • Posts: 1034
  • Joined: 25 Jan 2015
  • Loc: Rutland, MA

Posted 01 June 2016 - 10:42 PM

How do you have time for all of this while making mirrors? lol ;-) Sounds like a fun project. I'm hoping to at least make a modest optical bench in part of the basement. Mechanized would be very cool... post some pics of the other half of the setup... unmasked Foucault in.... polishing pattern, speed, other parameters out?


Edited by brucesdad13, 01 June 2016 - 10:43 PM.


#31 mark cowan

mark cowan

    Vendor (Veritas Optics)

  • *****
  • Vendors
  • topic starter
  • Posts: 8680
  • Joined: 03 Jun 2005
  • Loc: salem, OR

Posted 01 June 2016 - 10:43 PM

Yes, I just edited the post to clarify that, must have felt you walking on my grave there.  :lol:

 

Interferometry in the usual form is incapable of validating levels of 1/50th wave PVW.  The fiber test that Bruce posted awhile back, however difficult it is to implement, reaches subnanometer surface error resolution, easily that good.

 

Of commonly available tests in general, none will suffice.  But the geometrical argument alone will serve to demonstrate the precision, and I'll get to that eventually.

 

Ordinary interferometry should validate it partially at least.  One could arrange some mirror swapping about to get a few different tests to compare results between. 


  • brucesdad13 likes this

#32 mark cowan

mark cowan

    Vendor (Veritas Optics)

  • *****
  • Vendors
  • topic starter
  • Posts: 8680
  • Joined: 03 Jun 2005
  • Loc: salem, OR

Posted 01 June 2016 - 10:45 PM

How do you have time for all of this while making mirrors? lol ;-) Sounds like a fun project. I'm hoping to at least make a modest optical bench in part of the basement. Mechanized would be very cool... post some pics of the other half of the setup... unmasked Foucault in.... polishing pattern, speed, other parameters out?

 

It's part of the job of making the mirrors better, and I've been working on it for quite a while.    

 

You won't really get to see the other half of the setup, unfortunately - at least not in detail or in action - as that's all trade secrets. :shrug:


Edited by mark cowan, 01 June 2016 - 10:46 PM.

  • brucesdad13 likes this

#33 brucesdad13

brucesdad13

    Apollo

  • *****
  • Posts: 1034
  • Joined: 25 Jan 2015
  • Loc: Rutland, MA

Posted 01 June 2016 - 10:46 PM

What's the mirror test stand look like? :) Builtin fans? How does it handle various sized mirrors?



#34 brucesdad13

brucesdad13

    Apollo

  • *****
  • Posts: 1034
  • Joined: 25 Jan 2015
  • Loc: Rutland, MA

Posted 01 June 2016 - 10:49 PM

Trade secrets :sad: 

 

I've been reading this type of thing for several months... 'scoping out UofA website etc all very interested. Nothing beyond the realm of daydreams for me anytime soon

 

Non-sequential optimization technique for a computer controlled ...
www.loft.optics.arizona.edu/documents/.../2009_daewook_oe_nonsequential.pdf
by DW Kim - ‎2009 - ‎Cited by 39  - ‎Related articles 1College of Optical Sciences, University of Arizona, 1630 E. University Blvd, ... R. A. Jones, “Computer-controlled polishing of telescope mirror segments,” Opt.



#35 mark cowan

mark cowan

    Vendor (Veritas Optics)

  • *****
  • Vendors
  • topic starter
  • Posts: 8680
  • Joined: 03 Jun 2005
  • Loc: salem, OR

Posted 01 June 2016 - 10:50 PM

Fans?  We don't need no stinkin' fans.  It has pegs to adjust to mirror size.  Here is pic:

 

test stand 2.jpg

 

 


  • brucesdad13 likes this

#36 brucesdad13

brucesdad13

    Apollo

  • *****
  • Posts: 1034
  • Joined: 25 Jan 2015
  • Loc: Rutland, MA

Posted 01 June 2016 - 10:50 PM

Here's another http://www.loft.opti...arge_optics.pdf



#37 mark cowan

mark cowan

    Vendor (Veritas Optics)

  • *****
  • Vendors
  • topic starter
  • Posts: 8680
  • Joined: 03 Jun 2005
  • Loc: salem, OR

Posted 01 June 2016 - 10:52 PM

Trade secrets :sad:

 

I've been reading this type of thing for several months... 'scoping out UofA website etc all very interested. Nothing beyond the realm of daydreams for me anytime soon

 

Non-sequential optimization technique for a computer controlled ...
 

 

Lots of way to confuse a cat.  All you really need is one good deterministic polishing control method and the means to deploy it accurately over the workpiece.  Uhm, plus a test method.  Oh, and an interface that knows what to do with both.

 

OK, that's really quite a lot of stuff.  Heck, it's a big project, but it's intended to make big fast mirrors accurately and in reasonable time frames.


Edited by mark cowan, 01 June 2016 - 10:54 PM.


#38 brucesdad13

brucesdad13

    Apollo

  • *****
  • Posts: 1034
  • Joined: 25 Jan 2015
  • Loc: Rutland, MA

Posted 01 June 2016 - 10:57 PM

http://fp.optics.ari...lisher_1998.ppt



#39 MKV

MKV

    Fly Me to the Moon

  • *****
  • Posts: 7107
  • Joined: 20 Jan 2011
  • Loc: Florida, USA

Posted 02 June 2016 - 01:02 AM

Interferometry in the usual form is incapable of validating levels of 1/50th wave PVW.  The fiber test that Bruce posted awhile back, however difficult it is to implement, reaches subnanometer surface error resolution, easily that good.

 

Again, I am not judging here, but it makes one wonder why the professional community abandoned the Foucault test as the final arbiter years ago when you say the Foucault can outdo general-use interferometry hands down, in no time at all, and practically on any mirror of any focal ratio?

 

But I will gladly read as you address these issues.

 

Mladen


  • brucesdad13 likes this

#40 Mark Harry

Mark Harry

    Cosmos

  • *****
  • Posts: 7633
  • Joined: 05 Sep 2005
  • Loc: Northeast USA

Posted 02 June 2016 - 06:06 AM

 I should tell you Mladen, there was just such an instance years ago that a KE setup of Dave Kelley was used at work that could easily surpass sensitivity and overall accuracy results of what was at the time a Zygo IF setup that was worth  a quarter-M-$$$ at the time. The optics under test were all long focus concave spherical surfaces, but the glaring superiority was obvious. Calculations showed 1/50-1/100 PVW errors that were obtainable and verifiable.
   They haven't abandoned it, but use it where it's most advantageous. (IE, concave surfaces with no nulling optics necessary- such as your Newt mirrors) Same with IF, where it's most useful and expedient.
M.


  • kfrederick and MKV like this

#41 coinboy1

coinboy1

    Apollo

  • *****
  • Posts: 1123
  • Joined: 03 May 2011
  • Loc: Tulsa, OK

Posted 02 June 2016 - 08:36 AM

Awesome work Mark! This is really great and looks like a very accurate test method!



#42 MKV

MKV

    Fly Me to the Moon

  • *****
  • Posts: 7107
  • Joined: 20 Jan 2011
  • Loc: Florida, USA

Posted 02 June 2016 - 10:09 AM

 I should tell you Mladen, there was just such an instance years ago that a KE setup of Dave Kelley was used at work that could easily surpass sensitivity and overall accuracy results of what was at the time a Zygo IF setup that was worth  a quarter-M-$$$ at the time. The optics under test were all long focus concave spherical surfaces, but the glaring superiority was obvious. Calculations showed 1/50-1/100 PVW errors that were obtainable and verifiable.
   They haven't abandoned it, but use it where it's most advantageous. (IE, concave surfaces with no nulling optics necessary- such as your Newt mirrors) Same with IF, where it's most useful and expedient.
M.

 

Thank you, Mark. Indeed, that is amazing. I asked Mark Cowan for the same reason I ask others about similar things -- why aren't their methods and configurations used (or mentioned) in professional literature if they're that good. The answer, so far, seems to hinge on practicality, usefulness, expediency, etc. Just so no one thinks I'm picking on the Foucault method, I was dissecting the Bath IF with similar questions, and it all came down to several (very sensible, I might ad) reasons, but the major ones are: it cannot be certified, it's not a general-use IF. But it doesn't mean it's not accurate. Admittedly, when you read professional literature, hardly anyone talks about the Foucault test. Maybe, Mark Cowan's method will bring it back. It certainly seems to tackle the expediency and, with it, greater usefulness quite nicely. So, is it safe to say, then, that the Foucault is the most sensitive and accurate optical test (as is indirectly suggested, since not other test comes close), but has generally fallen in disuse in research and industrial practice for  reasons of expediency or practicality? 

 

Mladen


  • brucesdad13 likes this

#43 kfrederick

kfrederick

    Soyuz

  • *****
  • Posts: 3996
  • Joined: 01 Feb 2008

Posted 03 June 2016 - 05:19 AM

MKV   I keep reading your posts What are you trying to convey?    I do not understand ?  Everyone knows what the others are saying /Like me I say on the offaxis/ Some who sell mirrors say how they have better ways / ATM Like to show there telescopes /What are you saying in your posts?  This should be taken as a question not  a jab .The question if a knife edge only// can make a perfect mirror// has been answered many times // every day// by the best mirror makers in the world . 


Edited by kfrederick, 03 June 2016 - 05:21 AM.


#44 MKV

MKV

    Fly Me to the Moon

  • *****
  • Posts: 7107
  • Joined: 20 Jan 2011
  • Loc: Florida, USA

Posted 03 June 2016 - 01:28 PM

MKV   I keep reading your posts What are you trying to convey?    I do not understand ?  Everyone knows what the others are saying /Like me I say on the offaxis/ Some who sell mirrors say how they have better ways / ATM Like to show there telescopes /What are you saying in your posts?  This should be taken as a question not  a jab .The question if a knife edge only// can make a perfect mirror// has been answered many times // every day// by the best mirror makers in the world . 

 

Do you read professional optical literature, Kevin?



#45 brucesdad13

brucesdad13

    Apollo

  • *****
  • Posts: 1034
  • Joined: 25 Jan 2015
  • Loc: Rutland, MA

Posted 03 June 2016 - 02:29 PM

:grouphug:



#46 mark cowan

mark cowan

    Vendor (Veritas Optics)

  • *****
  • Vendors
  • topic starter
  • Posts: 8680
  • Joined: 03 Jun 2005
  • Loc: salem, OR

Posted 26 September 2016 - 01:12 PM

Alexis (Sixela) has posted on Astroforum (and in PM to me) about Jeroen Vleggaar and use of this method (translation here) and then started a thread here that led to Jac Brosens post (with pics) of a very nice tester, and his website is here (Dutch).

 

I would be cautious about taking vertical measurements due to test stand astigmatism (not a factor for horizontal use, as demonstrated by FEM modeling of mirror sag) but work I'm pursuing on measuring along angles other than horizontal gives some promise of using this method and a couple of runs at different orientations to get full coverage on any given mirror (at least for the balance of left-and-right which is what Foucault actually measures).  Also I've found that a slitless tester setup is superior for minimizing diffraction artifacts, not sure if that got into the thread earlier.

 

More to come in due time...


Edited by mark cowan, 26 September 2016 - 01:13 PM.

  • brucesdad13 likes this

#47 brucesdad13

brucesdad13

    Apollo

  • *****
  • Posts: 1034
  • Joined: 25 Jan 2015
  • Loc: Rutland, MA

Posted 26 September 2016 - 01:26 PM

It's all Dutch to me. ;) At least it's somewhat readable without knowing the language since I once knew a little German. Did you use Google to translate the PDF?

 

Alexis (Sixela) has posted on Astroforum (and in PM to me) about Jeroen Vleggaar and use of this method (translation here) and then started a thread here that led to Jac Brosens post (with pics) of a very nice tester, and his website is here (Dutch).

 

I would be cautious about taking vertical measurements due to test stand astigmatism (not a factor for horizontal use, as demonstrated by FEM modeling of mirror sag) but work I'm pursuing on measuring along angles other than horizontal gives some promise of using this method and a couple of runs at different orientations to get full coverage on any given mirror (at least for the balance of left-and-right which is what Foucault actually measures).  Also I've found that a slitless tester setup is superior for minimizing diffraction artifacts, not sure if that got into the thread earlier.

 

More to come in due time...



#48 mark cowan

mark cowan

    Vendor (Veritas Optics)

  • *****
  • Vendors
  • topic starter
  • Posts: 8680
  • Joined: 03 Jun 2005
  • Loc: salem, OR

Posted 26 September 2016 - 02:14 PM

Oops, I used google to translate pieces of it (too large otherwise) and then I pasted it into Word with the pics. If you want to read it, I put it out here temporarily:

 

http://www.raddobs.c...op Foucault.doc


  • brucesdad13 likes this

#49 sixela

sixela

    Hubble

  • *****
  • Posts: 15216
  • Joined: 23 Dec 2004
  • Loc: Boechout, Belgium

Posted 26 September 2016 - 02:31 PM

I can translate whatever isn't clear, BTW.
  • brucesdad13 likes this

#50 mark cowan

mark cowan

    Vendor (Veritas Optics)

  • *****
  • Vendors
  • topic starter
  • Posts: 8680
  • Joined: 03 Jun 2005
  • Loc: salem, OR

Posted 26 September 2016 - 02:41 PM

Thanks!


  • brucesdad13 likes this


CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics