29 replies to this topic

[dec12252]

Over the past several months I have read with great interest many CN posts on DPAC testing.  I have learned a lot by reading through (and studying) the recent threads by peleuba, Scott in NC, and others. Thanks to these informative CN threads, I found it remarkably simple to set up a testbed at relatively low cost - the most expensive component for me was the 6” flat. The other components including the mounts, Ronchi gratings, DPAC eyepieces, and LED’s in total cost less than the flat.

 

As a relative newcomer to DPAC testing I have been enjoying testing several of my small refractors and using Maciej’s excellent AOS program to analyze the results. In testing two of my refractors, a Vixen 102 fluorite and an AT72EDII, I am amazed at the power and simplicity of this test, particularly when combined with the AOS analysis program.  Maciej has been very helpful in guiding me in the use of his program to analyze Ronchigrams, and I am exceedingly grateful to him for his help with AOS. 

 

One scope I am in the process of testing is my AT72EDII refractor (shown below under test on the "chopping block").  So far I have only tested in the green (red, white, and blue are in the pipeline).  My initial analysis indicated that the lens has a slight turned down edge and a central zone but nevertheless has a Strehl ratio in the green of around 0.95.

 

I sent my inside focus Ronchigram to Maciej for a comparison and he graciously provided me with a revised analysis using a new version of AOS that he is currently developing - it is a version that includes zones and turned edges in the analysis.  His result is shown below.  The new AOS version can accurately model the turned edge in the lens and for my scope yields a green Strehl of around 0.88. A lower Strehl than my initial result (which did not account for the turned down edge) but still diffraction limited. Hopefully this new version of AOS will be available soon and I look forward to using it and continuing to test my stable of refractors. 

 

For folks who are interested in DPAC testing, don’t be afraid to jump in - you will find it to be straightforward and inexpensive and it will provide you with an excellent tool for evaluating telescope optics (I don't know why it took me so long to jump in). And lot’s of CNers are available to provide advice and valuable experience in setting up and conducting this test. I am certainly grateful for all I have learned on this topic from reading CN threads.

 

Dave

Attached Thumbnails

• 
• 

[Alrakis]

Dave, 

 

Have you thought of a mask to reduce the aperture a few millimeters to see how wide the turned edge is? 

 

Chris 

[dec12252]

Good suggestion.  I have not done that. Might give it a try.

[Scott in NC]

Very nice, Dave!  Great to see someone else jumping on the bandwagon.

[Scott in NC]

I can just barely see the central zone in the inside-focus Ronchigram that you provided, and if you hadn’t pointed it out then I probably would have missed it. Did you take at-focus and outside-focus images too?  If so, I’d be very interested in seeing them.

[dec12252]

Hi Scott,

 

Here they are.  You can see the central zone in the null image.

 

Dave

Attached Thumbnails

• 

[Scott in NC]

Thanks!  The at-focus image really makes it easy to see, but I can see it easily enough in the outside-focus image too.

[Scott in NC]

 

You can see the central zone in the null image.

 

One point that I feel I should make, just so we’re all using correct terminology (and for anyone who knows better, please correct me if you feel that I’m wrong), but I don’t think that an at-focus image can be correctly described as a “null image.” The two terms don’t mean the same thing.  Null is a state, rather than a condition of focus, but I suppose that an at-focus image could be described as a null image if the optics were perfectly nulled or optimized and aberration-free, and the at-focus image demonstrated that.

[CreatorsHand]

Hi Dave,

 

Thank you for posting this and for the encouragement. Like you, I have been following the posts on DPAC testing with great interest, and am in the final stages of putting my own DPAC testing rig together, hoping to be ready to do the first test in the next week or so. It is great being of this helpful community.

[Maciek_Cz]

Dave, Thank you for your kind words. You did a great job!
Regarding the estimation of the size of the turned edge, the AOS simulation shows that it has a width in the range of 2 - 2.5mm.

If you look closely at the simulation, you will also notice a central zone.
Below is an image of an at focus simulation.

The zone is a bit different from the one seen in Dave's photo, but the program still needs to be refined.

This middle zone does not have much impact on the results. I also marked the Strehl for the lens with and without the edge and zone.

The results may still be underestimated due to the method of estimating the Strehl coefficient.

This is a simple test that automatically finds the average focus for the entire lens, something like best focus,

and then the algorithm checks how many of the rays that have passed through the lens hit the Airy disk located at this focus.

 

I still have a lot of work to do before AOS has a user-friendly interface for defining zones and edges.

Attached Thumbnails

• 

Edited by Maciek_Cz, 06 April 2023 - 02:54 AM.

[John Huntley]

In an earlier thread on DPAC testing (not the controversial one !) I asked if an owner DPAC tests a scope or gets it DPAC tested and then, for whatever reason, decides to sell the scope, should they disclose the DPAC test results to potential buyers ?

 

I didn't really get any views on this because the thread moved on to other aspects of the process.

 

I wonder what folks feel about this ?

 

With the process being used by more and more folks, I can see that "have you had it DPAC tested ?" might become a common question from prospective purchasers.

 

Thanks.

Edited by John Huntley, 06 April 2023 - 04:13 AM.

[Alrakis]

Dave, Thank you for your kind words. You did a great job!
Regarding the estimation of the size of the turned edge, the AOS simulation shows that it has a width in the range of 2 - 2.5mm.

 

I thought that might be the case since the dpac lines looked decently straight with the hook only at the very end. It seems then that it would be worth loosing the outer 2-3mm to gain increased precision of the optical system.

 

This would be, in my mind, on of the benefits to DPAC. A way to determine if, through minor modification, you can greatly improve a telescope's optical system.

 

Chris 

[Etendue645]

Where can I get a copy of this AOS program? What an amazing and useful tool! 

 

OP, Good job on your DPAC testing setup and excellent analysis!

[Scott in NC]

In an earlier thread on DPAC testing (not the controversial one !) I asked if an owner DPAC tests a scope or gets it DPAC tested and then, for whatever reason, decides to sell the scope, should they disclose the DPAC test results to potential buyers ?

 

I didn't really get any views on this because the thread moved on to other aspects of the process.

 

I wonder what folks feel about this ?

 

With the process being used by more and more folks, I can see that "have you had it DPAC tested ?" might become a common question from prospective purchasers.

 

Thanks.

Here’s my take on it.  Buyers should always do their due diligence, and while I don’t think that it needs to be stated in an ad whether or not a scope has been DPAC tested, if a scope looks plain horrible on DPAC (or any other testing, whether it be a star test, interferometry, etc.), then an honest seller would be expected to disclose that.

 

Something like this IMO should be disclosed:

 

 

[Scott in NC]

Having said that, anyone here on CN knows that I DPAC test my scopes, and so I wouldn’t be able to get away with not disclosing my results even if I wanted to, lol.  But I wouldn’t do that.  As you guys know, I gladly share all my results, good or bad.

[dec12252]

Ditto for me.  If I sold a scope I would definitely disclose any DPAC test results I had on the scope.  I would feel honor bound to do so.

[dec12252]

Where can I get a copy of this AOS program? What an amazing and useful tool! 

 

OP, Good job on your DPAC testing setup and excellent analysis!

I suggest you email Maciek directly through CN and ask him where you can download the current version of AOS.

[DAVIDG]

 I would caution about calculating Strehl that do not include zones especially turned edges. Remember that 50% of the area of  a circle falls outside the 70% zones so a turned edge can encompass a large area of the lens or mirror. Also a turned edge has a sharp slope so it throws a lot of light outside the Airy disk. In my experience if you have  a turned edge you are already at 1/4 wave or worse no matter what the other area of the optics looks like. Also zones reduce contrast.

   As I have said before a measurement at one wavelength for an objective does not tell the whole story. Both spherical and chromatic aberration need to combined to get a true measure of the quality of an objective. For example in the spectrohelioscope I restored at Stellafane it uses a singlet lens of 16 foot focal length. If I measure the Strehl at any wavelength from violet into deep red the Strehl calculated would be diffraction limited or better for that wavelength. The problem is the focus between the blue image and the red image is 7 inches away from each other ! So the lens has a huge amount of chromatic aberration. So the total polychromatic Strehl is about 0.2 or less. The fact that the lens is used to view images of the Sun in only one wavelength at a  time and the images is refocused for each wavelength allows for diffraction limited or better quality images. 

   If a lens is designed to have no spherical aberration in the green and is shows a problem, this throws off  both the spherical and chromatic aberration of all the other wavelengths as well. So you can't assume because it has Strehl of say 0.8 ie 1/4 wave in the green that is the total Strehl for the lens. It most likely will be worse. 

  Another point is that one needs to look at the image at  the focal point  with a knife edge when doing DPAC. The Ronchi bands by themselves can make zones and surface roughness difficult to detect especially for those that are at experienced in reading them. 

  As pointed out by Scott a null is when you see no errors in the optics at the focal plane.  It will look like  a perfectly spherical mirror when tested at the radius of curvature, ie it will so nothing but a perfectly even surface.  You have to have that if your optics are going to have  a high Strehl. 

 

        - Dave 

Edited by DAVIDG, 06 April 2023 - 10:55 AM.

[Maciek_Cz]

In this case, the turned edge is only 5% of the entire surface of the lens.

[DAVIDG]

In this case, the turned edge is only 5% of the entire surface of the lens.

  The edge is most critical and having the turned edge as a slope slope that throws all that light outside the Airy disk   You also need to understand that it also affects the total spherical and chromatic aberration of the lens, unlike a mirror  So you can not  say that the total Strehl of the lens is 0.88, it lower then this.

 

             - Dave 

[Maciek_Cz]

  The edge is most critical and having the turned edge as a slope slope that throws all that light outside the Airy disk   You also need to understand that it also affects the total spherical and chromatic aberration of the lens, unlike a mirror  So you can not  say that the total Strehl of the lens is 0.88, it lower then this.

 

             - Dave 

David, I don't need to understand this because I know it perfectly well.

In the case shown in the pictures, the outer zone is very narrow, in addition, its effective curvature changes from elliptical to spherical closer to the edge.

Some of the rays coming from this area can still hit the Airy disk. Of course, I am not sure because I would have to generate an image of the rays running from the edge to the focus.

And I don't want to do that.
"Turned edges" such as in the example photo of the author of the thread AOS generates as a gradient of conic constant. As you can see, such a mathematical description of this area very well reflects the real picture from experiment. Is the Strehl obtained from the simulation real? I'm not sure because it's my first attempt. In addition, as I wrote in my first post, the result may be underestimated due to the simple criterion of hitting the Airy disk for avarage focus. See for example the astigmatism that fits in the Airy disc, but Strehl goes down dramatically.

I'm not saying you're wrong, because I might be wrong.

The software is in the experimental phase and requires many tests, and it must have a convenient interface for entering zones, and there is no such interface yet.

What would be most useful to me would be ronchigrams for telescopes that have been studied by other methods with known wavefront profiles.

This would be the best test for AOS. Without a reference point, nothing can be definitively determined.

 

Maciek

Edited by Maciek_Cz, 06 April 2023 - 02:07 PM.

[DAVIDG]

  Your program looks to be  very good at modeling the Ronchi pattern when you have a smooth figure. To have your code calculate a correct Strehl for a turned edge or any zone for that matter  it needs to correctly calculate the radius of  curvature of  that zone to determine were the light is coming to focus from that zone.  A zone maybe small in width so it only affects a small section of Ronchi band but it could have a  big impact on the results . For example a ring zone that is narrow it will  be difficult to try  to model  since it affects a  small section of  the Ronchi band to correctly determine it's radius of curvature  and hence the total affect on the Strehl calculation. 

  So my suggestion would that the user would need to enter the position of the zone,  total width and then allows the user to  enter a radius  to try to match the distortion of the band at that point to what they see in the actual image. 

 

                - Dave 

Edited by DAVIDG, 06 April 2023 - 02:36 PM.

[Maciek_Cz]

  Your program looks to be  very good at modeling the Ronchi pattern when you have a smooth figure. To have your code calculate a correct Strehl for a turned edge or any zone for that matter  it needs to correctly calculate the radius of  curvature of  that zone to determine were the light is coming to focus from that zone.  A zone maybe small in width so it only affects a small section of Ronchi band but it could have a  big impact on the results . For example a ring zone that is narrow it will  be difficult to try  to model  since it affects a  small section of  the Ronchi band to correctly determine it's radius of curvature  and hence the total affect on the Strehl calculation. 

  So my suggestion would that the user would need to enter the position of the zone,  total width and then allows the user to  enter a radius  to try to match the distortion of the band at that point to what they see in the actual image. 

 

                - Dave 

Dave, but AOS does exactly what you wrote. Approximately 120,000 rays are analyzed. The algorithm consists in finding the parameters of each of the rays of light (radius of curvature and focus). The zone definition window is constructed in this way, the columns will contain the beginning and end of the zone and the range of conic constant changes. Behind all this are numerical calculations, not just drawing pretty pictures.

[peleuba]

In an earlier thread on DPAC testing (not the controversial one !) I asked if an owner DPAC tests a scope or gets it DPAC tested and then, for whatever reason, decides to sell the scope, should they disclose the DPAC test results to potential buyers ?

 

I didn't really get any views on this because the thread moved on to other aspects of the process.

 

I wonder what folks feel about this ?

 

With the process being used by more and more folks, I can see that "have you had it DPAC tested ?" might become a common question from prospective purchasers.

 

Thanks.

 

 

Controversial thread?   <<perk>>  Did someone call?

 

DPAC testing threads are not controversial in and of themselves.  Rather, they are valuable additions to the CN community.  There is little controversary when testing in autocollimation because there are few sources of error.  The results are almost always reproducible and defendable.

 

Controversary arises when members:  (1) question the process;  (2) question the motives of the tester;  (3) wish to impart their version of reality into the threads.  Anyway, I'm of a different opinion:  The results should not be driving the controversy rather, its the unsubstantiated claims by vendors/manufacturers - in which DPAC debunks - are the drivers of controversy.     

 

In any event, your second point is a question of ethics... Basically, you are asking should a scope owner sell a scope that s/he knows is not a good performer without disclosing this to prospective buyers.  Easy question for me, I would not sell a scope - at top dollar -  knowing it tested poorly.   Others may have a different POV as there are some questionable scopes for sale.  As an example, SCT's routinely do poor in DPAC - barely ¼ wave and some worse due to the additive nature of the aberrations, but folks seem to love'em.  I have the luxury of being able to test anything that comes my way using several methodologies, not just DPAC.  There is a growing number of members who are interested in learning about optics and this is a good thing.

Edited by peleuba, 06 April 2023 - 03:17 PM.

[Scott in NC]

 

In any event, your second point is a question of ethics... Basically, you are asking should a scope owner sell a scope that s/he knows is not a good performer without disclosing this to prospective buyers.  Easy question for me, I would not sell a scope - at top dollar -  knowing it tested poorly. 

Exactly.