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

Star Test--How Important Are the Inner Rings?

This topic has been archived. This means that you cannot reply to this topic.
31 replies to this topic

#1 Kent10

Kent10

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 5,051
  • Joined: 08 May 2012

Posted 24 June 2016 - 12:34 PM

I have read Suiter's book and use it for reference along with many sources online.  In star testing my telescopes, I have been mostly looking at the outer ring inside and outside focus.  I don't expect any scope to test perfectly but I am trying to figure out the amount of Spherical Aberration and comparing what I see in my star test to images in the Suiter book and online. 

 

I can see some difference in the intensity of the outer ring inside and outside focus that is similar to the picture of 1/8 wave in the Suiter book.  But the inner rings in the Suiter book are quite different.  When the outer ring is bright the inner rings have less intensity and the area is dark.  Yet I don't think my inner rings are quite as dark.  Hard to tell but they don't look as dark as in the book.  Perhaps there are other aberrations in the mix?

 

I realize the pictures in the Suiter aren't real star tests but just drawings.  How accurate are they to a real star test?  I wish there was a compilation of star test images with someone evaluating and comparing them.  I haven't found anything like this, at least, not all in one place.

 

I have yet to make a 33% obstruction to test my scopes.  Suiter recommends this.  But even with the obstruction, it is going to be very subjective and I will have to guess if the results are 1/10 wave, 1/8 wave etc.

 

No one said it was easy and I am sure it take lots of experience but I want to learn :).

 

Thanks, Kent



#2 J A VOLK

J A VOLK

    Apollo

  • *****
  • Posts: 1,195
  • Joined: 24 May 2006

Posted 24 June 2016 - 12:45 PM

It is a little easier to consider the secondary shadow "breakout". Check the size and intensity of the shadow at the same distance inside and outside of focus - Mel Bartels has a good discussion and sketches on this (google 'Bartels star testing'). A larger obstruction helps. To me this is a little easier to assess than intensity of the rings.

#3 Kent10

Kent10

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 5,051
  • Joined: 08 May 2012

Posted 24 June 2016 - 12:48 PM

It is a little easier to consider the secondary shadow "breakout". Check the size and intensity of the shadow at the same distance inside and outside of focus - Mel Bartels has a good discussion and sketches on this (google 'Bartels star testing'). A larger obstruction helps. To me this is a little easier to assess than intensity of the rings.

Thanks.  I do need to get around to making that 33% obstruction.  It should not be difficult and it will teach me a lot.



#4 Eddgie

Eddgie

    ISS

  • *****
  • Posts: 26,634
  • Joined: 01 Feb 2006

Posted 24 June 2016 - 10:38 PM

It is hard to tell much about spherical aberration by using the defocused rings unless you obstruct the scope.

The 33% obstruction method eliminates a lot of guesswork and the brightness sensitivity of the human eye, and with refracting objectives, color mixing different ways on different sides of focus can make the rings do very unpredictable things.

 

The obstruction test is about geometry.  If the shadow is 1/3rd the diameter at exactly 10 waves of defocus, the scope has perfect SA. 

 

Why then do you need to do both sides?  Well, if the shadow is not exactly 1/3rd of the pattern on one side, then it is the difference in size between the 10 waves of defocus inside and the 10 waves of defocus outside that let you quantify the error.

 

If there is a perfect SA correction though, the secondary shadow would be exactly the same size 10 waves inside and 10 waves outside.

 

And don't cheat on the 10 waves.  Do the math and use the exact measurement or as close as you can get.  This is because the 10 waves will keep errors like higher order spherical aberration (often present in small, fast refractors) from causing a mis-reading.

 

Ignore how the rings appear.  For the SA test, the only thing that matters is how big the secondary shadow is as a percentage of the image.  A 33% obstruction will make a 33% sized shadow if the SA is perfect. 

 

Simple geometry. 



#5 Uwe Pilz

Uwe Pilz

    Gemini

  • *****
  • Moderators
  • Posts: 3,120
  • Joined: 16 May 2008

Posted 25 June 2016 - 12:18 AM

I recommend to simulate the patterns with the aberrator software and compare it with what you see. Works fine for me. It is important to know how far outside or inside the focus you are (in lambdas). Suiter explains how to determine that. If you are in a doubt how to handel that, I may give you an explanation.



#6 Eddgie

Eddgie

    ISS

  • *****
  • Posts: 26,634
  • Joined: 01 Feb 2006

Posted 25 June 2016 - 07:29 AM

I recommend to simulate the patterns with the aberrator software and compare it with what you see. Works fine for me. It is important to know how far outside or inside the focus you are (in lambdas). Suiter explains how to determine that. If you are in a doubt how to handel that, I may give you an explanation.

 

 

I 100% agree.    I use Aberrator 3.0 to make a model of the pattern as best as I can, and this tool allows one to plot the MTF curve so that the tester can actually see how much contrast transfer would be affected by the error.

 

Now today, I would tell people that the Roddier test would perhaps be an easier way for them because a lot of people simply won't take the time to read Suiter's book, which is too bad because it is an amazing book about telescopes and things that they can and can't do. It is one of the best books I have ever owned.

 

100% Agree though that modeling in Aberrator 3.0 (freeware) is a huge plus to help quantify the error.



#7 Kent10

Kent10

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 5,051
  • Joined: 08 May 2012

Posted 25 June 2016 - 08:35 AM

OK.  Thanks.  I have had the 33% obstruction on my mind for some time now so I just have to make one and try it out.

 

I have seen the Aberrator 3.0 software but have never tried it.  I looked at it recently and it is so old that I wondered if it would work on my Windows 10 computer.  Does anyone know if I will have problems?  I don't want to risk my computer though I do image so if I get major problems I can restore.



#8 Kent10

Kent10

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 5,051
  • Joined: 08 May 2012

Posted 25 June 2016 - 08:38 AM

I just found the zip of Aberrator 3.  The link worked on the home page.  From the downloads page it was a dead link.  I finally have it on my computer.  I am excited to try it out.  Thanks!



#9 Kent10

Kent10

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 5,051
  • Joined: 08 May 2012

Posted 25 June 2016 - 08:49 AM

One thing I have never understood is what are the different orders of error.  Low order, high order, 3rd order, 5th order etc.  I'll have to review my Suiter or is it easy for someone to explain the differences to me.  Thanks. 



#10 Uwe Pilz

Uwe Pilz

    Gemini

  • *****
  • Moderators
  • Posts: 3,120
  • Joined: 16 May 2008

Posted 25 June 2016 - 12:14 PM

> orders of error.

 

This are the so called Seidel aberrations. The degree ist the power of the deviation in the Seidel model. The higher the Seidel number the more "curly" is the wavefront deformation

The lower ones can be easy explained:

- 1th  Order are defocus and tilt

- 3th order are coma, astigmatism, distorsion and spherical aberration

- 5th order: there exist 6 which move between spherical aberration (higher degree) to distorsion (higher degree).

 

Some more explanation: Seidel explained the wave front deformation with two parameters: One (usually r) increases with aperture and the other (usually y)  increases with field. The Seidel degree gives the sum of exponents of both parameters. For the first degree there is only possible

- r (to the power 1)  (defocus, aberration increases with aperture)

- y (tilt, aberration increases with field)

 

There exist only the uneven degrees. For the third degree we could have r³ (spherical aberration), r²y (coma), ry² (astigmatism) and y³ (distorsion). You see that spherical aberration appear more with aperture, and distorsion is only visible in the field.

 

For the fifth degree (or even more for higher degrees) we have multiple aberrations, not all of them have common names. For 5th we have

 

r^5  r^4y  r²y³  r³y²  ry^4  y^5



#11 Kent10

Kent10

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 5,051
  • Joined: 08 May 2012

Posted 25 June 2016 - 06:36 PM

Thanks Uwe.  That helps a lot and I am going to keep reading some more. 



#12 Kent10

Kent10

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 5,051
  • Joined: 08 May 2012

Posted 26 June 2016 - 10:11 AM

It started out cloudy last night so I took the opportunity to make a 33% obstruction.  That would be 60mm diameter for my 180mm Tec.  I attached it and had a look.  I focused, pulled the eyepiece out and then compared the image by focusing again and then putting the eyepiece back in.  I tried several different eyepiece distances.  The size of the obstruction did look slightly different on each side of focus and I think it was inside focus there was a brighter ring around the obstruction.  So now my task is to compare to the Suiter images and Aberrator with SA and 33% obstruction. 

 

I am looking at the 1/8 wave SA in Suiter and at first even at 1/8 the difference in size of the obstruction outside/inside focus appears quite large.  But the images are misleading.  I think in comparing you should use the same # of rings inside/outside focus.  Correct?  It is difficult to see because of the different colors of the rings.  Some are now gray. 

 

Now to compare in Aberrator. 

 

In Aberrator I use the Intra/Extra example.  I put in 180mm for aperture, 33% obstruction, focal ratio of F7.  I leave everything else the same and then enter 0.125 3rd LSA and use 5 waves of focus.  I assume 0.125 is 1/8 wave.  Again, inside/outside focus, the obstruction size is misleading because if you look at the black area only the difference is large.  So I count 3 waves or rings and use that.  Correct?  I measured the size of each of these 3 dark rings with my ruler and they are almost the same even though they don’t look the same because of the different colors.  I see the thin bright ring on one side of focus around the 3rd black ring.  I suppose that looks similar to what I am seeing in my Tec.  Then there is a brighter thicker ring too but I don’t recall if I saw that.

 

In my images at the Tec I don’t remember seeing any diffraction rings within the dark obstruction like I see in Suiter and Aberrator.  Is that due to seeing?  Or maybe I don’t recall correctly.

 

This is fun.  Thanks for the Aberrator recommendation which I have always wanted to try and thanks for helping me with my evaluations.  It is still difficult to measure what I am seeing in my scope.



#13 Uwe Pilz

Uwe Pilz

    Gemini

  • *****
  • Moderators
  • Posts: 3,120
  • Joined: 16 May 2008

Posted 26 June 2016 - 11:42 AM

> It is still difficult to measure what I am seeing in my scope.

 

Yes, it needs some time. But I am sure you will master it.



#14 Kent10

Kent10

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 5,051
  • Joined: 08 May 2012

Posted 26 June 2016 - 01:15 PM

> It is still difficult to measure what I am seeing in my scope.

 

Yes, it needs some time. But I am sure you will master it.

Thanks for the encouragement, Uwe.  Little by little I am learning.  Ideally I need to star test many scopes.  That would help immensely.



#15 Mike Spooner

Mike Spooner

    Vendor (mirrors)

  • *****
  • Vendors
  • Posts: 774
  • Joined: 06 Aug 2010

Posted 27 June 2016 - 12:52 PM

One thing I have never understood is what are the different orders of error.  Low order, high order, 3rd order, 5th order etc.  I'll have to review my Suiter or is it easy for someone to explain the differences to me.  Thanks. 

Using Aberrator you can check different mixes of errors. Trying to recall from years past but I think I remember getting slightly different results using the different versions of the simulator. After star testing hundreds of optics I don't bother trying to quantify results with any real precision. Inputting various amounts of 3rd and 5th order spherical (over/under mixes) reveals how subtle compound errors cause misinterpretation. Aberrator is limited trying to model effects of a hole or hill or different radial zones all of which I've seen in real optics. I think Aberrator is a fantastic resource but just as with Suiter's book there's a limit to how far we can push the star test with authority for all cases. That's why I tend to prefer watching the close to focus performance. If an optic is perfect, it's pretty easy but mixed aberrations with seeing conditions can be frustrating. Polychromatic errors, secondary edge diffraction interactions with zones, hills or holes and thermal effects in the glass all can add to difficulty of assessing good/great/superb subjective evaluation. 

I've read a bit about the Roddier test but have never used it but glad Eddgie mentioned it as it may provide a more objective or clinical result. The old sci.astro.amateur forum had lots of discussion in addition to the many discussions here and other sites. I have only a few thousand hours of star testing/ bench testing experience and I'm still learning (sometimes surprising) things here and in the field. Sometimes I do focus the scopes and actually observe and have found that sharp images aren't always equated with only perfect star tests.  :)

 

Mike Spooner 



#16 Eddgie

Eddgie

    ISS

  • *****
  • Posts: 26,634
  • Joined: 01 Feb 2006

Posted 27 June 2016 - 08:11 PM

Use exactly 10 waves.   Less than this and the inner ring will not clear the secondary shadow.

 

Also, higher order spherical aberration that would be to small t have any impact on the in-focus image can also show if the defocus is not around 10 waves.

 

Much more than 10 waves and it is not sensitive enough. 

 

Use 10 waves.  !0 waves is exactly 2.156mm at f/7.   If you can't get exactly 10 waves, round up.but get as close to this figure as you can.



#17 Kent10

Kent10

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 5,051
  • Joined: 08 May 2012

Posted 27 June 2016 - 08:18 PM

OK thanks Ed.  I'll give that a go at some point.  It might be a while now since we will be going on a holiday soon. 

 

I have also been wondering how much cooling down a lens affects SA.  I know mirrors are affected so do you know if a lens is affected much.  I do cool my scopes as the sun is going down but with a 180mm triplet it might never keep up with our falling temperatures. 

 

Mike, thanks for sharing your experience.



#18 Eddgie

Eddgie

    ISS

  • *****
  • Posts: 26,634
  • Joined: 01 Feb 2006

Posted 27 June 2016 - 08:24 PM

And be aware of this as welll.  At f/7 and 180mm, unless one element has been aspherized, there is going to be some amount of higher order spherical aberration, and this will indeed influence the outcome and can make the rings different in brightness than what they would be without HSA.

 

This could sllighly influence the ring breakout, but unlikely more than a tiny bit.

 

Not sure at all why you are questioning the quality of this scope.  TEC builds high quality instruments. The point I would make with my post is that just looking at the out of focus rings tells you little, but if you are going to star test, while the test is easy to do, it is very specific in methods and one should have a reasonable undertanding of the different errors and how they present themselves.



#19 Kent10

Kent10

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 5,051
  • Joined: 08 May 2012

Posted 27 June 2016 - 08:44 PM

Ed, I don't question the quality of the Tec.  I love my Tec and believe I will never sell it.  I know from reading Roland's article, not to expect a perfect star test.  The reason I am doing the star test is to learn.  I love science and optics and although I don't build or make my own I am interested in learning all about it.  Of course, like others, I want my Tec and all my scopes to test to be perfect but I understand that is unlikely and so I enjoy understanding any "flaws" there might be but at the same time doubting I would ever notice the difference at the eyepiece.  Even if my Tec has 1/8 wave and other slight aberrations I do trust it is top notch and the views prove it.  On the other hand, I don't have a lot of experience to know if the views could possibly be better in, for example, an Astro Physics 180mm.  But I do have confidence in Tec and my scope to believe it is amongst the best :)

 

Star testing has helped me find the out of collimation problems in my 2 Televue scopes, NP-101 and TV-140.  So star testing does have practical value for me as well.



#20 Uwe Pilz

Uwe Pilz

    Gemini

  • *****
  • Moderators
  • Posts: 3,120
  • Joined: 16 May 2008

Posted 27 June 2016 - 11:59 PM

> Ed, I don't question the quality of the Tec.

 

It is worth performing the star test to see that the instrument is nearly perfect. Seeing means believing. :winky:



#21 Derek Wong

Derek Wong

    Apollo

  • *****
  • Posts: 1,266
  • Joined: 27 Apr 2008

Posted 28 June 2016 - 12:33 AM

Hi Kent:

 

Star testing is very interesting to study, along with the optical theory.  However, as you have seen, putting an exact number on the wavefront error is fraught with errors especially when you mix in color errors, which are reduced but not taken away by most filters.

 

Here is a specific example that has come up before in fast refractors from the following page:

 

http://www.telescope..._aberration.htm

 

"As the numbers show, fully minimizing pure higher order spherical aberration by balancing it with primary spherical, results in 4.6 times smaller P-V, and six times smaller RMS wavefront error."

 

That means a fast scope with high order spherical aberration and no primary spherical aberration is WORSE than one with added primary spherical if it is in the right direction.  Doing a component analysis by breaking the star test pattern up in high order spherical and low order spherical is not possible for 99.9999% of us.  Something like a Roddier test will have a much better chance at interpreting patterns correctly.

 

Derek



#22 Kent10

Kent10

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 5,051
  • Joined: 08 May 2012

Posted 28 June 2016 - 08:57 AM

Thanks Derek.  Yes, I can see it is not so easy.  I am impressed by those who can look at a star test image and can estimate the Strehl.  For example, Markus Ludes here with the refractor list on the left, http://aberrator.ast....net/scopetest/  .  I know he has seen hundreds if not thousands of tests.



#23 turnerjs085

turnerjs085

    Apollo

  • *****
  • Posts: 1,252
  • Joined: 14 Sep 2012

Posted 28 June 2016 - 09:10 AM

If you are interested in testing your scope further, you can make a knife edge or Ronchi "eyepiece" to use on stars in addition to the classic star test. A bit of dremel butchery of an extension tube will probably work to make a knife holder. The good bit about using an extension tube is that it has filter threads for using color filters to isolate different wavelengths. It is much easier to explore spherochromatism and and optical smoothness with a KE test than it is with a star test.

It is too hard to write up a huge post about making the tester using my phone, but if you do some reading on KE and Ronchi testers you should get the idea for how to make one. Basically you just need the knife apparatus, your light source is a star and the micrometer stage is the telescope focuser. Both of these are null tests when used with a star. Not necessarily easy to back out a reliable P/V wavefront or strehl, but very easy to see both the location and magnitude of defects.

Jeremy

#24 Kent10

Kent10

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 5,051
  • Joined: 08 May 2012

Posted 28 June 2016 - 09:25 AM

Thanks Jeremy.  I do have a Ronchi eyepiece.  I bought this one a few months ago.  http://www.gerdneuma...i-eyepiece.html

 

I have used it several times and I notice on the Tec that the lines bow out slightly when focused inwards.  It is slight but obvious.  If I use 3-5 lines, I can see any lines in the middle look straight but the ones near the left or right side have slight bowing.  What is strange is that I believe I should see the same amount of bowing going inwards when focused outwards but sometimes I only see a little and at other times it looks as though there is none while I always see the bowing when focused inwards.  This is why I am wondering if there is something going on with seeing or the cooling down of the lens.

 

In my Televue scopes, the Ronchi lines look straight but these are smaller aperture scopes (101mm and 140mm) so either they have less SA or they are not as affected from cooling.  Maybe I need to take my scopes out and test without cooling to see if I see the lines as curved even more and also in the Televue.  That might answer my question regarding whether cooling affect the SA.



#25 Derek Wong

Derek Wong

    Apollo

  • *****
  • Posts: 1,266
  • Joined: 27 Apr 2008

Posted 28 June 2016 - 09:35 AM

Thanks Derek.  Yes, I can see it is not so easy.  I am impressed by those who can look at a star test image and can estimate the Strehl.  For example, Markus Ludes here with the refractor list on the left, http://aberrator.ast....net/scopetest/  .  I know he has seen hundreds if not thousands of tests.

 

Hi Kent:

 

Markus (and other people like Roland or TMB) have tested thousands of lenses and have the interferometer reports next to the star test.  Even he gets into friendly arguments/discussions with Mr. Rohr about star testing.  Markus also says that a great many people who give estimates are way off, plus he readily admits that the interferometer may give different results than his estimates.  I have seen many people look through a lens and give an exact figure, and I have a lot of objections to that.  One can make a good estimate especially in a slower scope, but exact quantification is fraught with errors for the reasons that Mike Spooner detailed in his post.

 

Derek




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