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

127mm Mak Star test?

  • Please log in to reply
36 replies to this topic

#26 Jeff B

Jeff B

    Fly Me to the Moon

  • *****
  • Posts: 5983
  • Joined: 30 Dec 2006

Posted 24 January 2019 - 06:24 PM

Too many bars in the DPAC images.  Should really be two to three, max.  Really should use a green filter too.


  • Asbytec likes this

#27 Peter Besenbruch

Peter Besenbruch

    Fly Me to the Moon

  • *****
  • Posts: 7011
  • Joined: 21 Aug 2014
  • Loc: Oahu

Posted 24 January 2019 - 06:37 PM

Nice set up. So you wrap the entire scope and the dewshield before you take it out?

You leave the insulation on; you never remove it.



#28 Asbytec

Asbytec

    Guy in a furry hat

  • *****
  • Posts: 15666
  • Joined: 08 Aug 2007
  • Loc: Pampanga, PI

Posted 24 January 2019 - 08:11 PM

Not a direct quote, but Suiter addresses the issue in his book on star testing.  Paraphrased, he says that light waves either propogate perfectly or they do not and if they do not, then there is something modifying them. 

The problem with MCTs is that they have higher order sphrical aberration, and if too little defocus is used, this can make the pattern appear much different (because that is something modifying the pattern) but if the defocus is very small, then it is so sensitive that it is showing an error that would have little effect on the in-focus image. 

 

And here is a thought argument.  I am going to link a page that shows a bunch of MCT tests, showing both double pass gratings and defocused images.   What I would invite the reader to do is look though these and note the correlation between the inside an outside of focus images to the star test.  See how the Ronchi tests that have the straightest lines also seem to have the best secondary shadow  balance?

 

Now if Suite were wrong, we would expect to see a discrepancy between these to tests, but there are at least 60 MCTs tested here, and the ones with the best Ronchi tests also show the best pattern of star test.  How would someone explain that?  It would seem logical to me that if Suiter were correct, then these tests should agree, and this page will show that when the shadow breakouts are balanced, the Ronci tests also show a good instrument.

 

Again, this is physics.  Either the wavefront converges perfectly or it does not.  Physics demands that if it does not converge perfectly, then it will not diverge perfectly, so any deviation from the image inside and outside of focus is telling us that there is something modifying the wavefront.  It is the burden of the tester to decide whether it is a serious error, or something like some small amount of HSA that has little impact. 

 

Here is the page.  There are some excellent MCTs and SCTs shown here, and there are some horrible ones (One Meade SCT that is probably over half a wave of SA).

 

http://teleskop-spez...teleskope.html 

Yes, you are correct in all things and Suiter is not wrong. As you say, the difficulty testing a MCT is the residual higher order SA close to focus. It is the "something modifying the (final) wavefront". I am not sure how the Ronchi test would respond or whether they are accurate. They seem to show some correction error generally. I suspect they should show something like zones as each zone is either closer or further form focus depending on it's pupil radius. I am still trying to understand the Ronchi tests in the link you provided and how a MCT might test in them. 

 

If I understand the Ronchi it's a slope test, and it's the varying slope of the converging wave that causes what we see in the star test. That should show in a sensitive Ronchi test closer to focus. I am still a little puzzled why they seem to indicate general correction error at larger defocus.  I suspect because, even though it is not perfect, the wavefront still only approximates a sphere. It might be over corrected more generally as seen in the less sensitive Ronchi tests with more than a few lines. But, the perceived over correction is misleading because it is not over corrected in the pure sense of LSA alone. 

 

 

This confuses things because you are running a simulation for a design that shows a Strehl of .97.  My guess would be that this deviation from true perfection (Streh of 1) is because of the higher order spherical aberration of the design. 

 

If this is a correct interpretation on my part, this would mean that even if the scope were perfectly fabricated, it would only put 97% of the energy into the Airy Disk.     

 

And if that is a correct assumption on my part, then that means that the reason the star test is not exactly the same on each side of focus is because the wavefront is being modified by the higher order spherical aberration.

 

Also, this test is shown for 2 and 4 waves, and once again, this is not consistent with the way Suiter recommends doing the test.  If 10 waves of defocus were used and the obstruction were included in the simulation, the star test for this scope would look just about perfect  (assuming it were perfectly made).  Suiter himself says that this test is overly sensitive to HSA and that in the presence of HSA, it is important to very carefully quantify the error because most of the time, HSA will have little bearing on the in-focus image. 

I believe your assumption is correct. The high order from the meniscus cannot be fully corrected by the lower order (of opposite sign) on the primary as shown in the graphic below (curve #5). There will be some residual error deviating from, but closely approximating, a perfectly spherical wavefront. This is what Vla's illustrations show for a "perfect" all spherical MCT. The illustrations also show the RMS value (and as Roland says in his Complex Design essay) can be exceedingly good, regardless. Suiter is right, the /residual/ HSA will have little impact on the image. Which is why MCTs can be better than the perceived over correction error we see in the sketch and in the Ronchi tests in link you posted. In fact, you can have up to 0.4 wave PV HSA and still be diffraction limited. 

 

Below is an excerpt from Telescope Optics.net showing what is happening with the all spherical MCT. You can see how the final wavefront more closely approximates a reference spherical wavefront significantly improving RMS over pure HSA or LSA alone. You can also see this effect in the star test, especially as you move away from the balanced form at best focus (more like curve #3) is closer to what we are star testing. I think that curve would appear over corrected to us and to the Ronchi at some distance form best focus. 

 

https://www.telescop..._aberration.htm

 

secondary_spherical_balancing.PNG


Edited by Asbytec, 24 January 2019 - 10:43 PM.


#29 Asbytec

Asbytec

    Guy in a furry hat

  • *****
  • Posts: 15666
  • Joined: 08 Aug 2007
  • Loc: Pampanga, PI

Posted 24 January 2019 - 08:40 PM

And I see this "Complex telescopes don't test right" all the time.  The propagating wavefront is one of the most sensitive tests one can do.  It misses nothing.   You can see errors as little as 1/40th of a wave.  

 

Exactly, it misses nothing. That's why the images shown the link and in the sketch above show what they show (and why I say that is what you are supposed to see). At close to focus, you can see the effect the balanced low/high SA wavefront has on the star test, i.e., residual HSA than cannot be fully corrected. It's almost ironic that this apparently ugly, seemingly highly over corrected star test (and Ronchi) can have very good RMS. It looks as if the sketch shows some severe over correction, but it's not what is happening because pure correction (lower or higher order) is not the shape of the actual waveform across the entire radius. It's a mix of both LSA (opposite sign) from the primary and HSA from the meniscus. This allows the actual wavefront to curl back on itself to better approximate a sphere more closely than pure correction error alone. 

 

It's also interesting to me, how SCTs (also a complex design) tend to test more conventionally because the higher order term is dealt with (some some degree) on the corrector. 

 

The excerpted graph from Amateur Telescope Optics.net (above) shows why this can be true...and the beauty, IMO, of the MCT design. Quite ingenious, really. No scope can produce a perfectly spherical wavefront, not even a Newt. The trick to a good optic is how closely the actual wavefront approximates a perfectly spherical one. The MCT achieves this by adding LSA of opposite sign to HSA causing the wavefront to undulate near the perfect wavefront rather than trying to attain a perfectly spherical one like a parabola might attempt. The MCT design, with the meniscus and spherical primary, cannot achieve a perfectly spherical wavefront. But, if well executed, it can closely approximate one...and the result is seen in the star test and in Vla's simulations above.  

 

Below is a nice graphic that also illustrates what's going on. (I believe the lower order graphic at left needs to be reversed, of opposite sign to the higher order graphic. That was mentioned in the thread where John put up this beautiful graphic. Edit: I think the bottom image shows balanced LSA with defocus to the best focus location along the caustic.) This is how a MCT deals with the problem of approximating a spherical wavefront with a fast primary that, by all rights, should be heavily under corrected...and would be without the meniscus. (Note the rough semblance of the balanced final waveform to the figure on a Schmidt corrector.) You can see the final wavefront is not spherical, because it cannot be, but it's close enough. What it shows is the residual HSA we see in the star test close to focus. Some better, some worse...maybe all diffraction limited? Who knows. Check it in focus and tell me. smile.gif  (I think many in the link actually are diffraction limited.) 

 

Vla's illustration above is as "perfect" as it can be. But, it still shows different patterns either side because only a perfect spherical wavefront is exactly the same both sides. You can see the final waveform looks nothing like conventional correction error we think we see in both the star test and the Ronchi. That's why they are difficult to star test in terms of pure correction error alone because that is not what is happening. I suppose a sensitive enough Ronchi test will show this final waveform, and not pure over correction more generally. It's this close approximation to spherical that improves RMS despite the ugly star test and Ronchi. 

 

Balanced HiLow Graphic.jpg


Edited by Asbytec, 24 January 2019 - 10:40 PM.


#30 Vla

Vla

    Viking 1

  • *****
  • Posts: 554
  • Joined: 08 Mar 2014

Posted 25 January 2019 - 02:47 PM

Eddgie, confusion might be coming from Suiter using monochromatic light, and those from OSLO are for polychromatic, photopic sensitivity. That said, I don't unconditionally trust OSLO, it does some strange things. Here's slice patterns by the good old Aberrator for 0.13 wave p-v of 5th order SA (which there is not actually the 5th order spherical, rather balanced 3rd and 5th order, when minimizing the higher order residual). There are similarities, but also differences. And this is still monochromatic light.

Attached Thumbnails

  • BSA.jpg

Edited by Vla, 25 January 2019 - 02:49 PM.


#31 AlienRatDog

AlienRatDog

    Mercury-Atlas

  • *****
  • topic starter
  • Posts: 2971
  • Joined: 15 Jul 2005
  • Loc: Ann Arbor

Posted 26 January 2019 - 08:09 AM

I assume I won’t get much better by going with a C5?

#32 Wildetelescope

Wildetelescope

    Apollo

  • -----
  • Posts: 1457
  • Joined: 12 Feb 2015
  • Loc: Maryland

Posted 26 January 2019 - 08:46 AM

To my untrained eye, the stars look pretty similar to my 120ED in sharpness

Then i would call it a day an focus on observing:-).  Your scope sounds like it performing fine to me.  

 

Cheers!  

 

Jmd



#33 Asbytec

Asbytec

    Guy in a furry hat

  • *****
  • Posts: 15666
  • Joined: 08 Aug 2007
  • Loc: Pampanga, PI

Posted 26 January 2019 - 08:51 AM

I assume I won’t get much better by going with a C5?

Hard to say. Both can be pretty good well corrected scopes. Neither are premium hand crafted optics or mechanics. Still, both are potentially good scopes of nearly equal quality. Both are obstructed by the secondary and the secondary baffle. It's really a case of 6 of one, half dozen of another.

I am not sure of the hit and miss factor, but both probably fall under a similar mass production bell curve. In my opinion based on anecdotal evidence. Folks like both designs and it's rumored China optics are not that bad, anymore. I believe it.

We all take chances. I have yet to be seriously let down. I still don't know why, but there does seem to be something magical about a MCT even though there is really no reason for it. Maybe it's just the good seeing in the tropics, but I really enjoyed mine over the years.

Edited by Asbytec, 26 January 2019 - 08:52 AM.


#34 tmaestro

tmaestro

    Messenger

  • *****
  • Posts: 473
  • Joined: 22 Jul 2015
  • Loc: West of Houston, Texas

Posted 26 January 2019 - 09:45 AM

I seem to remember a story where Roland hand figured an MCT to be identical inside and outside of focus, but it didn't perceptibly change the quality of the image.

 

A quick google turned up this post with a couple further links:

https://www.cloudyni...maks/?p=4041432

 

edit: never mind, both Asbytec and Eddgie have follow up posts in that (8 year old) thread so this is being covered, and probably thoroughly.  :D


Edited by tmaestro, 26 January 2019 - 09:49 AM.


#35 Asbytec

Asbytec

    Guy in a furry hat

  • *****
  • Posts: 15666
  • Joined: 08 Aug 2007
  • Loc: Pampanga, PI

Posted 26 January 2019 - 10:37 AM

I seem to remember a story where Roland hand figured an MCT to be identical inside and outside of focus, but it didn't perceptibly change the quality of the image.

 

Here's the link to Roland's article. (Eddgie has a valid concern over the size of the obstruction test not mentioned in this thread). Other than that, it's a good essay on MCT design. 

 

http://www.csun.edu/.../startest2.html

 

Thanks for the other link, might have to peruse that one. See if I can learn anything. Thanks.

 

Edit: Oh, I see I already read it trying to understand all this stuff. Need to read it again. And the link to Roland's article is there, too. Sorry. You said you "seem to remember", so I thought you didn't have it available to review. 


Edited by Asbytec, 26 January 2019 - 10:41 AM.


#36 Edwin

Edwin

    Messenger

  • -----
  • Posts: 452
  • Joined: 16 May 2007
  • Loc: The Netherlands

Posted 28 January 2019 - 08:06 AM

 

Folks like both designs and it's rumored China optics are not that bad, anymore. I believe it.

The Russian website fidgor.narod.ru contains a lot of optical benchtests of Synta 127 maks, about 15 or so. The majority (9 out of 15) has a strehl above 0.90 and a PtV of 1/3th or better (only 4 of 15 have a PtV of 1/4 th or better). As with all Chines telescopes , the optical quality varies and you have to have some luck to get a good one.


Edited by Edwin, 28 January 2019 - 08:07 AM.


#37 Asbytec

Asbytec

    Guy in a furry hat

  • *****
  • Posts: 15666
  • Joined: 08 Aug 2007
  • Loc: Pampanga, PI

Posted 30 January 2019 - 07:46 AM

The Russian website fidgor.narod.ru contains a lot of optical benchtests of Synta 127 maks, about 15 or so. The majority (9 out of 15) has a strehl above 0.90 and a PtV of 1/3th or better (only 4 of 15 have a PtV of 1/4 th or better). As with all Chines telescopes , the optical quality varies and you have to have some luck to get a good one.

Depends on what one means by PV, I presume peak wavefront error. Due to the nature of the wavefront, it can have as much as 0.4 PV of residual balanced higher order SA as still be diffraction limited at 1/13.4 or Marechal at 1/14 waves RMS. Normally folks see 1/3 PV and cringe. But this is not primary or secondary LSA alone. 

 

https://www.telescop...k_spherical.htm




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