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

Calling OSLO experts - tilted plates in optical path

  • Please log in to reply
20 replies to this topic

#1 TG

TG

    Mercury-Atlas

  • *****
  • topic starter
  • Posts: 2,826
  • Joined: 02 Nov 2006
  • Loc: Latitude 47

Posted 26 October 2021 - 12:14 PM

I've been considering a plate beam splitter for testing purposes but wasn't sure whether it would cause aberrations. So, I modeled two plates in OSLO: one a full aperture plate, tilted a small amount as an optical window would be, and a second one tilted 45° close to focus of a f/6 paraboloid. The full aperture plate causes no discernible aberration, even as thick as 10mm and tilted a significant amount but, to my utter surprise, the one close to focus caused 1 wave of astigmatism, even when it was only 1mm thick. I had been planning to try a clear UV/IR filter as a poor man's plate beam splitter but looks like this idea is no good.

 

13rEbhvl.png

 

Not being an OSLO or optical expert, I would appreciate help checking my model (attached).

 

Does the above result make sense?

 

 

TG

Attached Files



#2 BGRE

BGRE

    Soyuz

  • *****
  • Posts: 3,601
  • Joined: 21 Mar 2016
  • Loc: New Zealand

Posted 26 October 2021 - 01:20 PM

You could compare your results with those predicted by the formulae in:
https://wp.optics.ar...cal_Testing.pdf

Tilted plates in uncollimated light produce axial coma, astigmatism, as well as axial CA, transverse CA, and SA.

The relevant formulae start on page 42.

Edited by BGRE, 26 October 2021 - 03:46 PM.

  • TG and RichA like this

#3 DAVIDG

DAVIDG

    Voyager 1

  • *****
  • Posts: 10,550
  • Joined: 02 Dec 2004
  • Loc: Hockessin, De

Posted 26 October 2021 - 02:15 PM

 You didn't have your coordinate system setup correctly. Try this version. You still have some astigmatism and lateral color that one would expect.

 

                                 - Dave 

Attached File  Tilted Plates v2.len   652bytes   14 downloads


  • MKV likes this

#4 TG

TG

    Mercury-Atlas

  • *****
  • topic starter
  • Posts: 2,826
  • Joined: 02 Nov 2006
  • Loc: Latitude 47

Posted 26 October 2021 - 07:09 PM

Dave, did you save the file before uploading? I see no difference from my file in terms of setup and the same P-V and RMS values.

 

TG



#5 freestar8n

freestar8n

    Vendor - MetaGuide

  • *****
  • Vendors
  • Posts: 11,000
  • Joined: 12 Oct 2007
  • Loc: Melbourne, Australia

Posted 26 October 2021 - 07:52 PM

If intended for use near focus, consider a pellicle.

Frank

#6 TG

TG

    Mercury-Atlas

  • *****
  • topic starter
  • Posts: 2,826
  • Joined: 02 Nov 2006
  • Loc: Latitude 47

Posted 26 October 2021 - 08:49 PM

If intended for use near focus, consider a pellicle.

Frank

I intend to use it as a beam splitter for star testing in DPAC mode, possibly in conjunction with Roddier software or just placing a Ronchi grating at focus. I didn't know about pellicle beam splitters - thanks for the tip - but they seem extremely fragile and at best 1λ surface quality. A cube beam splitter seems a better choice here but I was wondering if a tilted UV/IR filter would work as well.

 

TG



#7 BGRE

BGRE

    Soyuz

  • *****
  • Posts: 3,601
  • Joined: 21 Mar 2016
  • Loc: New Zealand

Posted 26 October 2021 - 08:54 PM

https://www.slidesha...ight-a-new-type

Shows one method of compensating the transmitted beam aberrations produced by a plate beamsplitter when used in non collimated light.

#8 BGRE

BGRE

    Soyuz

  • *****
  • Posts: 3,601
  • Joined: 21 Mar 2016
  • Loc: New Zealand

Posted 26 October 2021 - 09:10 PM

Another option when using the Roddier test is to mathematically separate rotationally dependent errors such as axial coma and astigmatism in the system under test from those produced by the beamsplitter.
The wavefront aberrations are measured at various clockings of the test system with respect to the beamsplitter.
This can be done to remove the inherent field astigmatism when testing with a Bath interferometer when using DFTFringe.
The same procedure could be applied to results obtained with the Roddier test.
  • TG likes this

#9 TG

TG

    Mercury-Atlas

  • *****
  • topic starter
  • Posts: 2,826
  • Joined: 02 Nov 2006
  • Loc: Latitude 47

Posted 26 October 2021 - 09:16 PM

Thanks. BGRE. That might indeed be possible with the Roddier software but it would obviously not be helpful if the optic itself has astigmatism or coma (say, due to misalignment).

 

TG



#10 BGRE

BGRE

    Soyuz

  • *****
  • Posts: 3,601
  • Joined: 21 Mar 2016
  • Loc: New Zealand

Posted 26 October 2021 - 09:44 PM

Actually the astigmatism due to the optics and that due to the beamsplitter can be separated allowing the astigmatism in the optics to be measured/found.
This is what happens when measuring large fast primaries with a Bath when the field coma due to the lateral separation of the incident and return test beam foci can be significant.
The amount of astigmatism inherent in the optics under test and that due to the beamsplitter etc are both measured by the rotation/derotation process.
Its can also be useful for backing out test stand induced astigmatism.

Dale describes the process in some detail e.g.
https://groups.io/g/...20,2,0,85419066
  • MKV likes this

#11 MKV

MKV

    Cosmos

  • *****
  • Posts: 9,912
  • Joined: 20 Jan 2011
  • Loc: Southeast, USA

Posted 26 October 2021 - 11:32 PM

If intended for use near focus, consider a pellicle.

Expensive, easily damaged, cannot be cleaned, and not top optical quality.



#12 DAVIDG

DAVIDG

    Voyager 1

  • *****
  • Posts: 10,550
  • Joined: 02 Dec 2004
  • Loc: Hockessin, De

Posted 27 October 2021 - 09:18 AM

Dave, did you save the file before uploading? I see no difference from my file in terms of setup and the same P-V and RMS values.

 

TG

 Yes it is  a different file. On your file I turned on the focal plane so it was drawn and it was not in the right place. That shows that the your tilts and bends were not setup correctly. So I fixed that. We got that the same result by accident but it does show the problem of placing a filter of  the thickness that was specified and on large tilt angle  close to the focal plane. The angle of the converging light cone is high so it is easily affected by the filter placed at an angle  but when placed over the front of the scope the light is parallel  so filter has much less affect at distorting the wave front when place on an angle 

 

                        - Dave 



#13 TG

TG

    Mercury-Atlas

  • *****
  • topic starter
  • Posts: 2,826
  • Joined: 02 Nov 2006
  • Loc: Latitude 47

Posted 27 October 2021 - 11:10 AM

 Yes it is  a different file. On your file I turned on the focal plane so it was drawn and it was not in the right place. That shows that the your tilts and bends were not setup correctly. So I fixed that. We got that the same result by accident but it does show the problem of placing a filter of  the thickness that was specified and on large tilt angle  close to the focal plane. The angle of the converging light cone is high so it is easily affected by the filter placed at an angle  but when placed over the front of the scope the light is parallel  so filter has much less affect at distorting the wave front when place on an angle 

 

                        - Dave 

Not sure what you mean: I'm looking at the focal planes in both your and my versions and they're identical (see below). But the only difference I can see is that in the coordinate data for the second surface of the (close to focus) plate, you have Pickup type set to "Plus". What does this mean? It seems to make no difference in the final numbers...

h7MvNABl.png

xkIWNGll.png



#14 TG

TG

    Mercury-Atlas

  • *****
  • topic starter
  • Posts: 2,826
  • Joined: 02 Nov 2006
  • Loc: Latitude 47

Posted 27 October 2021 - 11:12 AM

In any case, looks like there is an unacceptable amount of aberration for a plate to be used as a beam splitter in a visual star test using DPAC. Software might be able to separate out the induced aberrations, but visually, the "star" would look a mess.

 

TG.



#15 MKV

MKV

    Cosmos

  • *****
  • Posts: 9,912
  • Joined: 20 Jan 2011
  • Loc: Southeast, USA

Posted 27 October 2021 - 12:41 PM

In any case, looks like there is an unacceptable amount of aberration for a plate to be used as a beam splitter in a visual star test using DPAC. Software might be able to separate out the induced aberrations, but visually, the "star" would look a mess.

You should analyze the effect of a star diagonal with a right angle prism rather than plate beam splitter. Normally, plate beam splitters are used with collimated (parallel) light beams and wave-retardation plates.

 

If you wish to use a beam splitter rather than a star diagonal, then use a cube.

 

But I wish to command you for using OSLO.  As you probably realize by now, such programs will help you learn and understand more optics and unlock more mysteries of optical science this way then by sharing "anecdotal evidence." 



#16 TG

TG

    Mercury-Atlas

  • *****
  • topic starter
  • Posts: 2,826
  • Joined: 02 Nov 2006
  • Loc: Latitude 47

Posted 27 October 2021 - 04:14 PM

Yes, MKV, I'm doing a cube beam splitter + DPAC analysis right now. Will post in a separate thread. This is what I have actually built and tested briefly with a f/5.4 Tele Vue refractor but what OSLO is telling me is surprisingly different from what I saw with a 200lpi Ronchi grating.

 

TG



#17 MKV

MKV

    Cosmos

  • *****
  • Posts: 9,912
  • Joined: 20 Jan 2011
  • Loc: Southeast, USA

Posted 27 October 2021 - 04:20 PM

Yes, MKV, I'm doing a cube beam splitter + DPAC analysis right now. Will post in a separate thread. This is what I have actually built and tested briefly with a f/5.4 Tele Vue refractor but what OSLO is telling me is surprisingly different from what I saw with a 200lpi Ronchi grating.

What is OSLO telling you? OSLO doesn't lie if done correctly, and optical tests don't lie if done correctly either. And what did you see with the Ronchi grating? Are you sure it's actually 200 lpi and not 100 lpi? Different vendors had them classified differently -- as number of single shade lines or as cycles or line pairs per unit length. Big difference.


  • Oregon-raybender likes this

#18 MKV

MKV

    Cosmos

  • *****
  • Posts: 9,912
  • Joined: 20 Jan 2011
  • Loc: Southeast, USA

Posted 27 October 2021 - 04:28 PM

Not sure what you mean: I'm looking at the focal planes in both your and my versions and they're identical (see below). But the only difference I can see is that in the coordinate data for the second surface of the (close to focus) plate, you have Pickup type set to "Plus". What does this mean? It seems to make no difference in the final numbers...

Depends how many degrees off-axis the upper cone is. It's unlikely (unless decidedly compensated) that the off-axis bundle will lie on the same pane as the paraxial one..

 

OSLO has a user's manual online that explains everything. Look up how they specify tilted components, and what each entry means.



#19 TG

TG

    Mercury-Atlas

  • *****
  • topic starter
  • Posts: 2,826
  • Joined: 02 Nov 2006
  • Loc: Latitude 47

Posted 27 October 2021 - 04:57 PM

What is OSLO telling you? OSLO doesn't lie if done correctly, and optical tests don't lie if done correctly either. And what did you see with the Ronchi grating? Are you sure it's actually 200 lpi and not 100 lpi? Different vendors had them classified differently -- as number of single shade lines or as cycles or line pairs per unit length. Big difference.

See my new thread for details but the cube should induce enough aberration at f/5.4 that it should have been easily visible. I got my grating from Ronchiscreens.com and they had 80,133,200 and 250 available, so it looks like they were line-pair specifications?

 

TG



#20 TG

TG

    Mercury-Atlas

  • *****
  • topic starter
  • Posts: 2,826
  • Joined: 02 Nov 2006
  • Loc: Latitude 47

Posted 27 October 2021 - 04:58 PM

 

Depends how many degrees off-axis the upper cone is. It's unlikely (unless decidedly compensated) that the off-axis bundle will lie on the same pane as the paraxial one..

 

OSLO has a user's manual online that explains everything. Look up how they specify tilted components, and what each entry means.

 

I read the relevant section but it's wall of text and without any diagrams whatsoever, very hard to understand.



#21 Gleb1964

Gleb1964

    Mariner 2

  • -----
  • Posts: 243
  • Joined: 13 Jul 2016
  • Loc: Sweden

Posted 28 October 2021 - 04:05 AM

The full aperture plate causes no discernible aberration, even as thick as 10mm and tilted a significant amount but, to my utter surprise, the one close to focus caused 1 wave of astigmatism, even when it was only 1mm thick.

 

Just want to mention to prevent you derive wrong conclusion from that.

The amount of aberrations induced by parallel glass plate inserted in converging beam does not depend on the plate position along axis. If glass plate closer to focus or to aperture that changes only a beam's footprint on the plate. Aberrations do not change.   


Edited by Gleb1964, 28 October 2021 - 04:06 AM.

  • Mike I. Jones 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