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

help with experimental optimization of achromat

  • Please log in to reply
15 replies to this topic

#1 Bob4BVM

Bob4BVM

    Surveyor 1

  • *****
  • topic starter
  • Posts: 1988
  • Joined: 23 Mar 2015
  • Loc: W. Oregon

Posted 26 June 2019 - 11:30 PM

I have an old Jaegers 6"F5 achro that I built a scope for, but I have always wanted to go back and take the time to optimize the objective.  The scope itself is all square at both ends and collimates well so I think I have the bases covered in a sound OTA.

 

What I do not know is how well the crown & flint are aligned rotationally and also whether their air-spacing is correct value.

 

Not being an optics guy by any stretch of the imagination, I would appreciate hearing from anyone who knows what to look for when adjusting air gap & rotation alignment. I do not have Jaegers specs on gap and if there were any marks on the lens edges, they are long gone. 

Also can the lens be tested out of the OTA or do I need to reassemble the whole thing in the OTA at every iteration ?  If so what would that test setup look like ?

 

Thanks

Bob

 


  • kksmith likes this

#2 MKV

MKV

    Cosmos

  • *****
  • Posts: 8203
  • Joined: 20 Jan 2011
  • Loc: Florida, USA

Posted 27 June 2019 - 12:42 AM

Also can the lens be tested out of the OTA or do I need to reassemble the whole thing in the OTA at every iteration ?  If so what would that test setup look like ?

Bob, unless you have a 7-inch optical flat, you may consider testing the lens using an oil flat as illustrated below. It's best to test it in a cell. One of the lenses will have to be rotated after each test run and retested. 

 

Just curious, what makes you think Jaegers spacers are incorrect? Is this the original Jaegers lens, or was it pre-owned and possibly tampered with?

 

Mladen

 

testing refractor lens.jpg


  • ngc7319_20 likes this

#3 ngc7319_20

ngc7319_20

    Apollo

  • *****
  • Posts: 1017
  • Joined: 25 Oct 2015
  • Loc: MD

Posted 27 June 2019 - 06:34 AM

I would suggest to first do a so-called "star test" on a bright star with a green filter, and decide if there is a problem or not. Since it seems to be on an alt-az mount, perhaps Polaris is the best target.  Let the scope thoroughly acclimate, and wait for some good seeing.  Use something like 300x.   And study what the in-focus star image looks like.  Then look at the patterns inside and outside focus.  Maybe make some drawings.  You can compare to output of a simulation program like Aberrator and get some qualitative and quantitative estimates of the errors.  If it looks similar to a "perfect" lens, then I'd leave it alone. 

 

If you see astigmatism, you might be able to improve it by rotating the elements.  But only if both are somewhat astigmatic.  If all the astigmatism is in one element, then there is not much you can do.

 

If there is coma aberration, then it is likely that the air space is wedged and not uniform.  In other words, the spacing is not the same all the way around the edge of the glasses.  If there are three spacers, you can increase or decrease one of them slightly.  The direction of the coma will help you decide which one.

 

You can also use a Cheshire collimation tool, and study the reflections from the lens surfaces.  This can tell you if one or both elements are tilted in the cell.  Or if the cell is tilted on the scope. Or (again) if the air space is wedged. 

 

An incorrect average spacing will cause spherical aberration.  Slower F-ratio lenses I've worked were relatively insensitive to changes in the element separation.  But I don't have any experience with F/5.

 

If you do a star test with no filter, and see that the star images are red on one side, and blue on the other -- this indicates one or both of the elements are wedged.  In effect, a thin prism has been added to the lens.  Here again rotation can help, if the elements are equally wedged.  But if all the wedge is in one element, rotation won't help.  Sometimes wedge can be improved by decentering the glasses -- but only if they are loose in the cell to start with.  Otherwise you would have to grind-off the edge of one of the glasses to get more space in the cell, and it is a big project.

 

Aberrator software is here:

http://aberrator.astronomy.net/


Edited by ngc7319_20, 27 June 2019 - 06:36 AM.


#4 ed_turco

ed_turco

    Vanguard

  • *****
  • Posts: 2437
  • Joined: 29 Aug 2009
  • Loc: Lincoln, RI

Posted 27 June 2019 - 11:11 AM

I would suggest to first do a so-called "star test" on a bright star with a green filter, and decide if there is a problem or not. Since it seems to be on an alt-az mount, perhaps Polaris is the best target.  Let the scope thoroughly acclimate, and wait for some good seeing.  Use something like 300x.   And study what the in-focus star image looks like.  Then look at the patterns inside and outside focus.  Maybe make some drawings.  You can compare to output of a simulation program like Aberrator and get some qualitative and quantitative estimates of the errors.  If it looks similar to a "perfect" lens, then I'd leave it alone. 

 

If you see astigmatism, you might be able to improve it by rotating the elements.  But only if both are somewhat astigmatic.  If all the astigmatism is in one element, then there is not much you can do.

 

If there is coma aberration, then it is likely that the air space is wedged and not uniform.  In other words, the spacing is not the same all the way around the edge of the glasses.  If there are three spacers, you can increase or decrease one of them slightly.  The direction of the coma will help you decide which one.

 

You can also use a Cheshire collimation tool, and study the reflections from the lens surfaces.  This can tell you if one or both elements are tilted in the cell.  Or if the cell is tilted on the scope. Or (again) if the air space is wedged. 

 

An incorrect average spacing will cause spherical aberration.  Slower F-ratio lenses I've worked were relatively insensitive to changes in the element separation.  But I don't have any experience with F/5.

 

If you do a star test with no filter, and see that the star images are red on one side, and blue on the other -- this indicates one or both of the elements are wedged.  In effect, a thin prism has been added to the lens.  Here again rotation can help, if the elements are equally wedged.  But if all the wedge is in one element, rotation won't help.  Sometimes wedge can be improved by decentering the glasses -- but only if they are loose in the cell to start with.  Otherwise you would have to grind-off the edge of one of the glasses to get more space in the cell, and it is a big project.

 

Aberrator software is here:

http://aberrator.astronomy.net/

What about the expression that begins "Fools rush in . .  ?''  I wouldn't dare try this project and I do have a modicum of experience in optical matters.


  • Oregon-raybender likes this

#5 Bob4BVM

Bob4BVM

    Surveyor 1

  • *****
  • topic starter
  • Posts: 1988
  • Joined: 23 Mar 2015
  • Loc: W. Oregon

Posted 27 June 2019 - 12:58 PM

Bob, unless you have a 7-inch optical flat, you may consider testing the lens using an oil flat as illustrated below. It's best to test it in a cell. One of the lenses will have to be rotated after each test run and retested. 

 

Just curious, what makes you think Jaegers spacers are incorrect? Is this the original Jaegers lens, or was it pre-owned and possibly tampered with?

 

Mladen

 

attachicon.gif testing refractor lens.jpg

Great info on the oil flat, thank you ! Seems that will be easiest to try many iterations since I would not have to reassemble the scope every time. Now I just have to build the setup and learn how to interpret what I see.

Yes, it is  original Jaegers lens, but I bought it bare, no cell, just crown & flint, so the spacers are my own 'guess', about .003" paper IIRC.

CS

Bob



#6 Bob4BVM

Bob4BVM

    Surveyor 1

  • *****
  • topic starter
  • Posts: 1988
  • Joined: 23 Mar 2015
  • Loc: W. Oregon

Posted 27 June 2019 - 01:03 PM

I would suggest to first do a so-called "star test" on a bright star with a green filter, and decide if there is a problem or not. Since it seems to be on an alt-az mount, perhaps Polaris is the best target.  Let the scope thoroughly acclimate, and wait for some good seeing.  Use something like 300x.   And study what the in-focus star image looks like.  Then look at the patterns inside and outside focus.  Maybe make some drawings.  You can compare to output of a simulation program like Aberrator and get some qualitative and quantitative estimates of the errors.  If it looks similar to a "perfect" lens, then I'd leave it alone. 

 

If you see astigmatism, you might be able to improve it by rotating the elements.  But only if both are somewhat astigmatic.  If all the astigmatism is in one element, then there is not much you can do.

 

If there is coma aberration, then it is likely that the air space is wedged and not uniform.  In other words, the spacing is not the same all the way around the edge of the glasses.  If there are three spacers, you can increase or decrease one of them slightly.  The direction of the coma will help you decide which one.

 

You can also use a Cheshire collimation tool, and study the reflections from the lens surfaces.  This can tell you if one or both elements are tilted in the cell.  Or if the cell is tilted on the scope. Or (again) if the air space is wedged. 

 

An incorrect average spacing will cause spherical aberration.  Slower F-ratio lenses I've worked were relatively insensitive to changes in the element separation.  But I don't have any experience with F/5.

 

If you do a star test with no filter, and see that the star images are red on one side, and blue on the other -- this indicates one or both of the elements are wedged.  In effect, a thin prism has been added to the lens.  Here again rotation can help, if the elements are equally wedged.  But if all the wedge is in one element, rotation won't help.  Sometimes wedge can be improved by decentering the glasses -- but only if they are loose in the cell to start with.  Otherwise you would have to grind-off the edge of one of the glasses to get more space in the cell, and it is a big project.

 

Aberrator software is here:

http://aberrator.astronomy.net/

 

Thx for the link and for all the defect descriptions, that is a lot of what I was looking for, "what causes what" …

For decentering, if it comes to that, I think I would just make some space in the cell ID before attacking the glass. shocked.gif 

CS

Bob



#7 Bob4BVM

Bob4BVM

    Surveyor 1

  • *****
  • topic starter
  • Posts: 1988
  • Joined: 23 Mar 2015
  • Loc: W. Oregon

Posted 27 June 2019 - 01:11 PM

What about the expression that begins "Fools rush in . .  ?''  I wouldn't dare try this project and I do have a modicum of experience in optical matters.

 

"...  ...where angels fear to tread ! "

 

Yeah Ed, I won't be messing with the glass, much easier to enlarge the cell ID if needed.

 

That said, some of us did some objective lens-edge glass removal in making our widefield DIY Nikon binocs with no apparent ill effects, but that was on a little 3X optical system...



#8 MKV

MKV

    Cosmos

  • *****
  • Posts: 8203
  • Joined: 20 Jan 2011
  • Loc: Florida, USA

Posted 28 June 2019 - 01:56 AM

Great info on the oil flat, thank you ! Seems that will be easiest to try many iterations since I would not have to reassemble the scope every time. Now I just have to build the setup and learn how to interpret what I see. Yes, it is  original Jaegers lens, but I bought it bare, no cell, just crown & flint, so the spacers are my own 'guess', about .003" paper IIRC.

Bob, the setup is easy. Here's a setup I used recently on a 153 mm f/7.8 Wollensak doublet for a quick look. 

 

oil flat DPAC rig .jpg

 

the lens

 

oil flat_1.jpg

 

and the Ronchi test

 

oil flat DPAC ronchigram.jpg

 

The surface features are just dust and particle contamination of the oil surface. These can be easily spooned off.

 

The lens was not properly aligned relative to the oil flat because it was on a fixed platform such as I originally drew it. Basically, it shows a fairly decent wavefront (almost straight bars). The kinks at the edges are probably artifacts of the screen frequency (diffraction effects). This can be verified with a screen of lower frequency (about 70-72 lpi vs 133).

 

Some curvature of the bars is most likely due to a slight misalignment. I know for a fact that the lens was reversed in the cell intentionally and possibly not reassembled correctly 

put together as it should be.

 

For critical assessment you'll need a leveling platform, such as this one used to test a C8. 

 

20150306_065507.jpg

 

I've updated my earlier drawing to ease in visualization. 

 

testing refractor lens_leveling.jpg

 

This allows rotation of the rear element relative to the front for best optical definition. The rear element should be gently lifted, rotated a small amount and retested. Make sure all the spacers are of the same thickness, otherwise on-axis coma may result. 

 

Familiarize yourself with various error observed in a DPAC Ronchi test.There are plenty of illustrations online for that. An excellent course is Daniel Malacara Optical Sop Testing.]

 

It's also important to use vibration-dampening devices such as already mentioned (dogfood/catfood, lentil, rice, sand bangs, etc.), and the thickest motor oil you can find for your flat.

 

Mladen

 

______

 

PS Do you happen to know the glass substrate used by Jaegers. My quick look at a 6-inch f/5 BK7/F2 doublet suggests an air gap of about half of what you're using, or about 0.0015 inches. 


  • eros312, davidc135, Bob4BVM and 1 other like this

#9 Bob4BVM

Bob4BVM

    Surveyor 1

  • *****
  • topic starter
  • Posts: 1988
  • Joined: 23 Mar 2015
  • Loc: W. Oregon

Posted 28 June 2019 - 10:11 PM

Bob, the setup is easy. Here's a setup I used recently on a 153 mm f/7.8 Wollensak doublet for a quick look. 

 

attachicon.gif oil flat DPAC rig .jpg

 

the lens

 

attachicon.gif oil flat_1.jpg

 

and the Ronchi test

 

attachicon.gif oil flat DPAC ronchigram.jpg

 

The surface features are just dust and particle contamination of the oil surface. These can be easily spooned off.

 

The lens was not properly aligned relative to the oil flat because it was on a fixed platform such as I originally drew it. Basically, it shows a fairly decent wavefront (almost straight bars). The kinks at the edges are probably artifacts of the screen frequency (diffraction effects). This can be verified with a screen of lower frequency (about 70-72 lpi vs 133).

 

Some curvature of the bars is most likely due to a slight misalignment. I know for a fact that the lens was reversed in the cell intentionally and possibly not reassembled correctly 

put together as it should be.

 

For critical assessment you'll need a leveling platform, such as this one used to test a C8. 

 

attachicon.gif 20150306_065507.jpg

 

I've updated my earlier drawing to ease in visualization. 

 

attachicon.gif testing refractor lens_leveling.jpg

 

This allows rotation of the rear element relative to the front for best optical definition. The rear element should be gently lifted, rotated a small amount and retested. Make sure all the spacers are of the same thickness, otherwise on-axis coma may result. 

 

Familiarize yourself with various error observed in a DPAC Ronchi test.There are plenty of illustrations online for that. An excellent course is Daniel Malacara Optical Sop Testing.]

 

It's also important to use vibration-dampening devices such as already mentioned (dogfood/catfood, lentil, rice, sand bangs, etc.), and the thickest motor oil you can find for your flat.

 

Mladen

 

______

 

PS Do you happen to know the glass substrate used by Jaegers. My quick look at a 6-inch f/5 BK7/F2 doublet suggests an air gap of about half of what you're using, or about 0.0015 inches. 

 

 thanks for this  Mladen,

Most excellent,  this it really helps me visualize the process !

I have more or less stayed away from making or testing optics all my life, just because I tend to do everything in a deep-dive fashion, and I know I already have way too many hobbies/irons in the fire.  But this looks simple enough, I think I can do this without too mych trouble now that you've made it so clear.

 

I do not know the Jaeger glass types on this F5, but I will try different shims while I am set up for testing for rotation. What effect tells me I have the right  shim spacing ?

 

CS

Bob



#10 MKV

MKV

    Cosmos

  • *****
  • Posts: 8203
  • Joined: 20 Jan 2011
  • Loc: Florida, USA

Posted 29 June 2019 - 02:35 AM

Hi Bob,

I'm glad you liked it. Not everything with optics is complicated. If you're going to wonder about your optics, you'll have to do some testing unless you can hire someone to do ti for you. It's your choice.

 

Incorrect shims would result in more spherical aberration. This would manifest itself as curved Ronchi lines .Again, you'll have to do find what Ronchi patterns mean what. There are many examples to be found online. I already gave you one source. 

 

Basically, your test will require a Ronchi screen (100 lines/inch is a good general-use frequency), and a good high-power eyepiece. The eyepiece would be used in place of a Ronchi screen to test for image quality. The Ronchi screen would tell you if you have astigmatism, spherical aberration etc. The eyepiece will show you

what a star image would look like,etc.There are many causes of astigmatism, coma, spherical aberration, chromatic aberration, etc. Shims may be one of them.

 

For chromatic aberration, use a white LED as a light source. For correction of the lens as a whole use a green LED.

 

If you know how to use the knife edge (as in a Foucault test) it can give show you the figure of the wavefront  very clearly. You can achieve that by using only one band of the Ronchi test with somewhat less clarity. Ideally the wavefront should present as a flat disk, or straight Ronchi bands (with 3 to 5 bands max) showing. You may be surprised how many optics fail this simple test.This one is clearly unacceptable.

 

SAM_4661.JPG

 

I tested a 5-inch f/9.4 doublet and found out that the lenses were inverted in the cell.

 

SAM_4288_LR(2).JPG

 

After turning them around, the objecitve turned out to test excellent. (When judging the results, don't forget that DPAC shows the errors doubled)

 

SAM_4332_CR (2).JPG

 

Another objective showed clear coma on-axis due to a uneven spacers

 

SAM_4343_LR.JPG

 

Mladen


Edited by MKV, 29 June 2019 - 02:38 AM.


#11 KerryR

KerryR

    Soyuz

  • *****
  • Posts: 3867
  • Joined: 05 Dec 2007
  • Loc: West Michigan

Posted 29 June 2019 - 09:11 AM

I nearly eliminated over-correction in my 6" f8 by increasing element spacer width. For testing, I used (real) stars with an added 33% obstruction, a green band-pass filter, and relied on shadow breakout at 10 waves intra and extra focally. The process was: remove the existing spacers and start with a single layer of spacers made from Post-It notes, because they're sticky but easily removed, and re-test. Add spacing, re-test. I kept this up 'til I identified when the spacing was too much, and the objective shifted towards under-correction, then began working backwards. Once I had a good breakout, I made new more accurately cut spacers, again from a stack of the proper number of Post-It notes, and replaced the messy stacks used for the initial testing. The results have been very favorable.

At some point, I'll do the oil-flat thing. I didn't know about that process back when I did the above...


Edited by KerryR, 29 June 2019 - 10:05 AM.


#12 BGRE

BGRE

    Mercury-Atlas

  • *****
  • Posts: 2517
  • Joined: 21 Mar 2016
  • Loc: New Zealand

Posted 29 June 2019 - 10:03 AM

If the SA etc are measured for several spacer thicknesses and plotted against spacer thickness then the ideal value for the spacer thickness can be easily deduced.

#13 KerryR

KerryR

    Soyuz

  • *****
  • Posts: 3867
  • Joined: 05 Dec 2007
  • Loc: West Michigan

Posted 29 June 2019 - 10:56 AM

If the SA etc are measured for several spacer thicknesses and plotted against spacer thickness then the ideal value for the spacer thickness can be easily deduced.

Wouldn't this require fairly accurate objective measurement of the degree of SA change? What'd be a good way for a kitchen-counter atm to measure this? 
 



#14 MKV

MKV

    Cosmos

  • *****
  • Posts: 8203
  • Joined: 20 Jan 2011
  • Loc: Florida, USA

Posted 29 June 2019 - 01:40 PM

Wouldn't this require fairly accurate objective measurement of the degree of SA change? What'd be a good way for a kitchen-counter atm to measure this? 

Double-pass autocollimation test.



#15 davidc135

davidc135

    Viking 1

  • -----
  • Posts: 720
  • Joined: 28 May 2014
  • Loc: Wales, UK

Posted 29 June 2019 - 02:48 PM

In the 'Amateur Observational Handbook' Sidgwick says that .22mm is the max allowable focus variation due to SA in a F/5 system. But errors are doubled by light traversing the lens twice in a DPAC test and are further doubled if the test light source stays still. So, .88mm travel when using a knife edge should correspond to a 1/4 wave error. (I've taken max allowable to mean a 1/4 wave but he doesn't say so?) So, not ultra precision.

This figure of .22mm increases as the square of the F ratio so a scope of F/10 would be 4 times easier to test.

 

David


Edited by davidc135, 29 June 2019 - 02:51 PM.


#16 BGRE

BGRE

    Mercury-Atlas

  • *****
  • Posts: 2517
  • Joined: 21 Mar 2016
  • Loc: New Zealand

Posted 29 June 2019 - 06:19 PM

Using a knife edge to measure the axial focus for several zones in a DPAC setup is one method.
One could also use the Roddier test (essentially a quantitative version of the star test that uses intrafocal and extrafocal images to measure the wavefront curvature using the intensity transport equation).
Suitable Roddier test software is readily available. Otherwise an interferometer could be used.

Edited by BGRE, 29 June 2019 - 06:20 PM.



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