Jump to content


Photo

Zambuto Criteria #4 on Strehl

  • Please log in to reply
318 replies to this topic

#51 Rusty35

Rusty35

    Lift Off

  • -----
  • Posts: 11
  • Joined: 30 Jul 2013

Posted 13 October 2013 - 12:27 PM

Im confused,
I dont know anything about mirror making, or testing.
Was planing on ordering a 14"from Zambuto, but after hearing what MKV is saying it sounds like maybe I would be better of looking elsewhere.
Would Discovery be a better choice, they use interferometer testing?

#52 Alan French

Alan French

    Night Owl

  • *****
  • Posts: 4387
  • Joined: 28 Jan 2005
  • Loc: Upstate NY

Posted 13 October 2013 - 12:55 PM

Im confused,
I dont know anything about mirror making, or testing.
Was planing on ordering a 14"from Zambuto, but after hearing what MKV is saying it sounds like maybe I would be better of looking elsewhere.
Would Discovery be a better choice, they use interferometer testing?


Now, now - don't equate disagreements and discussions over testing methodology with production issues. Carl has an excellent reputation for consistently producing fine mirrors.

There are a number of excellent mirror makers around today. It is a good time for people who appreciate fine optics.

Clear skies, Alan

#53 JimMo

JimMo

    I'd Rather Do It Myself

  • *****
  • Posts: 5112
  • Joined: 08 Jan 2007
  • Loc: Under the SE Michigan lightdome

Posted 13 October 2013 - 12:55 PM

You'd pass on a Zambuto mirror because of a post on the internet? :ohmy:

#54 Pinbout

Pinbout

    Cosmos

  • *****
  • Posts: 7842
  • Joined: 22 Feb 2010
  • Loc: nj

Posted 13 October 2013 - 01:02 PM

It is a good time for people who appreciate fine optics.




you don't need IF to know this needs work... :grin:

Posted Image

#55 Ed Jones

Ed Jones

    Mercury-Atlas

  • *****
  • Posts: 2695
  • Joined: 06 Apr 2004
  • Loc: Sin-sin-atti

Posted 13 October 2013 - 01:06 PM

No Carl make great mirrors and has a good reputation. Are his other Criteria clear? #1 and 2 are straightforward but the rest are rather subjective "exceptionally smooth, functionally negligible levels, levels undetectable at the eyepiece, no evidence of any deviation outside the norm." Sounds good.

OTOH anyone producing a mirror using IF free of surface roughness, turned edge and good 2D curve conformance should expect similar performance.

#56 Arjan

Arjan

    Mariner 2

  • -----
  • Posts: 214
  • Joined: 21 Jan 2009
  • Loc: Netherlands

Posted 13 October 2013 - 01:23 PM

So, a 2mm roughness in 200mm mirror would spread (too tiny to notice) energy in a halo about 100 times the Airy disc diameter.

Did you mean to say 2 nm, not mm?


No, I think 2mm as in (average) lateral dimension. The height indeed should be very small.
Look up the theory here: Amateur Telescope Optics under 4.5 "Fabrication errors".

I should have looked here before asking anyway...

#57 MKV

MKV

    Mercury-Atlas

  • *****
  • Posts: 2576
  • Joined: 20 Jan 2011

Posted 13 October 2013 - 02:02 PM

Im confused,
I dont know anything about mirror making, or testing.
Was planing on ordering a 14"from Zambuto, but after hearing what MKV is saying it sounds like maybe I would be better of looking elsewhere.
Would Discovery be a better choice, they use interferometer testing?

I neither said nor implied any such thing. My objection was and is to Zambuto's characterization of interferometry and not to his workmanship. Zambuto has a top-notch reputation for excellence in mirror making.

Mladen

#58 MKV

MKV

    Mercury-Atlas

  • *****
  • Posts: 2576
  • Joined: 20 Jan 2011

Posted 13 October 2013 - 02:08 PM

you don't need IF to know this needs work...

You don't nee the Foucualt test either. :)

#59 Mike Lockwood

Mike Lockwood

    Vendor, Lockwood Custom Optics

  • *****
  • Vendors
  • Posts: 473
  • Joined: 01 Oct 2007
  • Loc: Usually in my optical shop

Posted 13 October 2013 - 02:18 PM

But I am still looking for answers, so please answer these three simple questions

I am tired of having arguments twisted around into questions that are not relevant to the thread topic, and then being asked (or another poster like Mark being asked) to answer them, like it's a quiz or piece of homework. I am under no obgligation to answer them, but since they were posted after my previous post, I will elaborate a bit and comment.

(1) do you think Vla's statement was "wrong and groundless"?

If he was looking at the comparison images of the two mirrors as shown in the web page listed by the OP, and as far as I can tell he was, then yes I think he is wrong and there are no grounds for saying one image was better quality than the other. Period.

(2) do you know how wide is a virtual "slit" in a slitless tester (and therefore how sensitive is the test), and how do you determine that width?

No I don't. Please measure it and report back to Mark and I.

I thank Mark for his excellent comparison of slit/slitless testing, which I recall had been posted before. Perhaps you missed it, but I remember it.

(3) do you think it's okay to throw interferometry under the bus?

I fail to see where this question comes from. A correction of facts about slitless testing does not constitute an attack on interferometry.

Quite frankly I've NEVER thrown interferometry under the bus, though I have been highly unkind to poor testing practices. I use interferometry, and will use it in the future, in far more ways than you know. I even use it for your much-mentioned Ritchey-Common testing, since it quantifies errors.

With proper sampling techniques and phase shifting, most of the roughness in the example should be visible after analysis, though Foucault makes it easily visible with a fraction of the effort/time, and thus for a fraction of the cost.

#60 Pinbout

Pinbout

    Cosmos

  • *****
  • Posts: 7842
  • Joined: 22 Feb 2010
  • Loc: nj

Posted 13 October 2013 - 02:26 PM

you don't need IF to know this needs work...

You don't nee the Foucualt test either. :)


but it is a autocollimation image which make these gross errors even grosser-er. :lol:

#61 mark cowan

mark cowan

    Vendor (Veritas Optics)

  • *****
  • Vendors
  • Posts: 4098
  • Joined: 03 Jun 2005
  • Loc: salem, OR

Posted 13 October 2013 - 02:34 PM

Also theoretically: why would a surface roughness of RMS 100nm be worse than, let's say, spherical abberation of the same amount?
The latter will direct more light from Airy disk to the rings surrounding this. The roughness would probably just enlarge the Airy disk itself, but I expect not further than that ring?


I think it's related to something I mentioned earlier, the larger scale structure of small scale errors - primary ripple, for example, has a repeating pattern, rather than being random across the surface, so the slope errors, though individually small, can reinforce. SA of the same overall amount of random surface roughness would be more obviously damaging for the same reason. Random error will scatter randomly, increasing veiling glare, but periodic error scatters preferentially, meaning it's at least more detectable.

Best,
Mark

#62 mark cowan

mark cowan

    Vendor (Veritas Optics)

  • *****
  • Vendors
  • Posts: 4098
  • Joined: 03 Jun 2005
  • Loc: salem, OR

Posted 13 October 2013 - 05:06 PM

and the slitless version had no slit at all on the source...

Mark, the slitless test forms a virtual" slit. The width of that slit is determined by the lateral position of the k-e.

regards,
Mladen


Why are you telling me this? Figure out exactly what the effective slit width is for any Foucault apparatus working at maximum resolution and then we can talk some more. Either build one and try it (which is what I did) or work it out on paper, I don't much care which. :shrug:

Best,
Mark


#63 richard7

richard7

    Not Quite

  • *****
  • Moderators
  • Posts: 5892
  • Joined: 02 Nov 2007
  • Loc: Sacramento

Posted 13 October 2013 - 06:25 PM

It's getting closer.
:locked:

#64 jhayes_tucson

jhayes_tucson

    Apollo

  • *****
  • Posts: 1036
  • Joined: 26 Aug 2012
  • Loc: Bend, OR

Posted 13 October 2013 - 11:00 PM

Mark,
My point is that on the Igram the surface roughness is quantified and in other methods it is not. The KE shows roughness but what is good and what isn't? Carl seems to be throwing IF under the bus so to speak but yet he uses IF to test his flats. Surface roughness on a flat is just as damaging as in the primary. Sure, good mirrors can be made without IF but don't bash IF for the sake of self promotion.


I agree with what Ed has to say here and I want to add a thought. It is interesting that there has been little mentioned about a key difference between interferometry and the other tests. Interferometry provides a direct, quantitative measurement of surface shape with respect to a reference surface. With PSI, that measurement is done over a high resolution, uniform grid over the entire surface. PSI measurements are very precise and with proper calibration, can provide a very high level of accuracy. The KE test is a slope test that provides data that must be integrated to get to surface information--in only one direction. The KE test is indeed qualitatively extremely sensitive to small slope errors but it is very poor at producing direct surface data. While it is possible to analyze the effects of slope errors over different spatial frequencies, the KE test does not give sufficient data for good quantitative analysis. There is no doubt that a smooth mirror will generally be better than a rough mirror, but it's the amplitude as shown in a power spectral density plot that counts. PSI data is probably the best way to get real data for that kind of analysis. Think of it this way: Strehl performance is most affected by low spatial frequency figure errors; while high frequency, low amplitude slope errors generally drives scatter. Clearly, scatter can reduce contrast but not enough to be noticeable for visual instruments with sub-quarter wave amplitudes (in the wavefront,) which are easily seen with a KE test.

When I see marketing statements about perfectly smooth mirrors, I am reminded of the first time I visited a Swiss binocular maker back in the 80's. They actually didn't know much (if anything) about optical testing so their approach was to "just make it perfect". That meant that they had (mostly) very high quality products but they spent so much time (an hence money) producing each unit that they eventually were driven out of business by competitors who understood what "good enough" meant.

The folks at Zambuto may make some very good mirrors but without real data, it's impossible to say how good they really are. They are certainly good at marketing and have a lot of happy customers, which may be all that counts.

John

#65 Starman1

Starman1

    Vendor (EyepiecesEtc.com)

  • *****
  • Vendors
  • Posts: 22897
  • Joined: 23 Jun 2003
  • Loc: Los Angeles

Posted 14 October 2013 - 12:41 AM

tests:
http://www.astro-for...Your-Mirror&...
http://www.astro-for...wton&p=35325...
Obviously not a significant quantity of Carl's mirrors, but indicative.

#66 Alan A.

Alan A.

    Mariner 2

  • -----
  • Posts: 263
  • Joined: 08 Feb 2007
  • Loc: San Francisco

Posted 14 October 2013 - 03:40 AM

I wonder how they test for smoothness on small scales in large mirrors like the 8 meter mirrors at Steward if you can't do it with an interferometer?



I know someone who was able to answer this question for me, and just got their response. To evaluate micro roughness on a surface - they use a laser fed "point source microscope" that gives a 2D readout in Angstroms RMS. For a special application mirror requiring a superpolished surface - they can make and measure it down to 1 ANGSTROM RMS OR LESS!! I find this to be a staggering achievement.




Would it be correct to say that a 10-20 Angstroms RMS mirror by IF would be basically a functionally perfect mirror for visual observing if it's Strehl were also high( eg >0.96) ?

#67 kfrederick

kfrederick

    Gemini

  • *****
  • Posts: 3015
  • Joined: 01 Feb 2008

Posted 14 October 2013 - 06:59 AM

The knife edge reading taken and the mirror rotated and a new set taken and they match! Is how it is done .More accurate than a single set of readings for seeing Astig and other things . I think in many ways a better test as no other optic to be used .

#68 Ed Jones

Ed Jones

    Mercury-Atlas

  • *****
  • Posts: 2695
  • Joined: 06 Apr 2004
  • Loc: Sin-sin-atti

Posted 14 October 2013 - 07:12 AM

No, if you move the mirror you lose the frame of reference. Rotating a mirror is more likely to detect astigmatism however but not quantify it.

#69 wh48gs

wh48gs

    Surveyor 1

  • *****
  • Posts: 1616
  • Joined: 02 Mar 2007

Posted 14 October 2013 - 07:15 AM

Well, Zambuto mirrors are smoother than the others shown, but local irregularities are there, and am sure on better or larger photos more microrughness would show in the null shot as well. Figure is less than perfect. But those two mirrors are certainly good enough to be top performers. And so are those shown as "defective" on Zambuto's site. So what is, really, the difference?

Vla

#70 wh48gs

wh48gs

    Surveyor 1

  • *****
  • Posts: 1616
  • Joined: 02 Mar 2007

Posted 14 October 2013 - 07:27 AM

The first mirror interferogram given in Carl Zambuto's site has an rms of .008 wave, and the wavelength used is 632.8 nm, so it looks like the surface roughness is 50.6 Angstroms RMS - correct?

Would it be correct to say that a 10-20 Angstroms RMS mirror by IF would be basically a functionally perfect mirror for visual observing if it's Strehl were also high( eg >0.96) ?



What is the source for those figures? It is hard to tell from the photo on Zambuto's site, because it is all blurred, but for this level of RMS error the average linear extent of the ripple structure shouldn't be larger than a couple of mm. Which could as well be. If so, this terrible roughness degrades the Strehl by 0.996 ratio, or so. Going down to 10-20 Angstroms wouldn't make no difference whatsoever.

Zambuto says the magnitude of microrriple cannot be measured - which is not correct - but at the same time feels free to imply that it significantly degrades mirror performance. A bit presumptuous?

Vla

#71 careysub

careysub

    Surveyor 1

  • *****
  • Posts: 1902
  • Joined: 18 Feb 2011
  • Loc: Rancho Cucamonga, CA

Posted 14 October 2013 - 08:18 AM

Im confused,
I dont know anything about mirror making, or testing.
Was planing on ordering a 14"from Zambuto, but after hearing what MKV is saying it sounds like maybe I would be better of looking elsewhere.
Would Discovery be a better choice, they use interferometer testing?


There have been a few thousand Zambuto mirrors sold thus far over many years.

There have been many, many opportunities for people to have their Zambuto mirrors tested, and given the price, the cost and/or effort of doing so is incrementally insignificant.

See if you can find a reference to a single Z-mirror testing badly. I have never heard of one.

#72 pstarr  Happy Birthday!

pstarr

    Mercury-Atlas

  • *****
  • Posts: 2949
  • Joined: 17 Sep 2004
  • Loc: NE Ohio

Posted 14 October 2013 - 08:47 AM

It's very simple. Those of you who think Zambuto is exaggerating or lying about his mirrors, or if his test methods are antiquated or unreliable, buy one from somewhere else. We're beating a dead horse here. People who enjoy his mirrors most likely don't care about the fact that he didn't use a interferometer. All they know is they are among the best mirrors they have ever owned. Call him, I'm sure he will take the time to discuss in detail every claim he has ever made. He is that kind of guy.

#73 marcosbaun

marcosbaun

    Sputnik

  • -----
  • Posts: 32
  • Joined: 01 Jan 2012

Posted 14 October 2013 - 09:28 AM

Here in Brazil there is a saying : Speak good or bad but talk about me. Zambutto company should be happy for such propaganda. Ok, they deserve.The question is: To what level you need to polish a mirror so that it provides images of the highest quality? To what extent is advantage we pay much more dearly for a mirror that level if another mirror too and just well polished, can provide images as good and almost indistinguishable in quality compared to the super polished, often for a price much more advantageous? Well, you are the one who decides.

Marcos

#74 MKV

MKV

    Mercury-Atlas

  • *****
  • Posts: 2576
  • Joined: 20 Jan 2011

Posted 14 October 2013 - 09:28 AM

This topic was never about Zambuto's mirrors, but mostly about his claim that (1) a certified 0.996 Strehl 1/15 wave P-V mirror is a piece of junk you wouldn't want in your telescope and (2) that the primary ripple is highly injurious to the image quality.

Some very seasoned individuals (Ed Jones, Vla, John, etc), with various degrees and years of educational, theoretical, professional and practical background and experience, beg to disagree.

regards,
Mladen

#75 pstarr  Happy Birthday!

pstarr

    Mercury-Atlas

  • *****
  • Posts: 2949
  • Joined: 17 Sep 2004
  • Loc: NE Ohio

Posted 14 October 2013 - 10:01 AM

I don't see anywhere where the mirror in question is called a pc of junk by Zambuto. To answer the second question, I would not want a mirror with that amount of primary ripple in my scope if I the had the choice of getting one that had a smooth surface. In actual use, I have found that the Zambuto mirror I owned showed pitch black sky around the moon without any haze or glare that I have seen with other mirrors. Same with planets. I saw a difference. Have you ever compared one of his mirrors with a mass produced mirror in actual use?.






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics