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

Need help understanding light cut off & coma in binoviewer

  • Please log in to reply
13 replies to this topic

#1 Volvonium

Volvonium

    Mariner 2

  • *****
  • topic starter
  • Posts: 277
  • Joined: 06 Mar 2019
  • Loc: Long Beach, CA

Posted 22 October 2019 - 12:26 PM

Hey all, I'm very new with binoviewing and I need help understanding what is going on with my recently acquired binoviewer setup.  I have some educated guesses as to what is going on, but I'm not sure what the best solution will be to address it.

 

Equipment:

Zeiss Binoviewer, 24.5mm clear aperture all around, 11.5cm light path, 1.25" nosepiece

3.75" long Siebert 1.25X OCA in 1.25" format

16" f/4.5 dobsonian reflector, focal length 1826mm

 

Issue:  

 

With the 1.25X OCA, I am able to reach focus with the binoviewer, but there seems to be excessive aperture loss and heavy coma induced.  When defocusing a star that is centered, I get a round defocused star.  When defocusing a star that is anywhere but centered, the defocused star is cut off.   

 

It kind of looks like this at the eyepiece, where the cutoff will follow the star's position.  If the star was on the left, it would look like a mirror image of my example.

binotoosmalloca.jpg

 

Considerations:

 

Telescope is well collimated, binoviewers are well collimated with identical left right image, all optical components/diopters are on axis and well centered. 

With a 2" 2X Telecentric Focal Extender, I can also reach focus and there is no odd coma induced; defocused stars show as circles regardless of position.  It is far too much magnification for my liking, on the other hand.

 

I am suspecting that the 3.75" long OCA is too small.  While I like the low magnification, i feel like the 1.25" format is too small and the light cone coming off the secondary is too large and a lot of light is getting cut off.  This is what I suspect is happening:

binotoosmalloca2.jpg

 

 

Harry Siebert's website is incredibly hard to parse/follow, so I'm unsure of what would be the best low magnification corrector for my application.   With my 2X 2" telecentric, it would seem that it's far back enough in the focuser that the light cone size is small enough by the time it hits the telecentric's lens that it will go into the BV's 1.25" nosepiece/objective without any loss.    It would seem like I would need a 2" corrector  that sits far into the drawtube that is low magnification, but also reduces the light cone size small enough for the BV's aperture.   

 

Any assistance, corrections, product recommendations, and education on this is greatly appreciated.  This is my best guess as to what is happening. 


Edited by Volvonium, 22 October 2019 - 12:27 PM.


#2 Eddgie

Eddgie

    ISS

  • *****
  • Posts: 25,605
  • Joined: 01 Feb 2006

Posted 22 October 2019 - 01:50 PM

It is almost certain that you are working at reduced aperture but the pictures are consistent with the simple fact that even if there is not aperture reduction, you would see this anyway because this is just the effect of off axis illumination cutoff and you would see that in pretty much any binoviewer used in all but the very slowest dobs.

 

Here are the number though, and let's look at the binoviewer first. If the front aperture of the binoviewer is 24.5mm, and the light path is 115mm, then you can find the fastest light cone that the binoviewer will pass by dividing the front aperture of the binovewer into the light path lenght.  In this case, that would be (115 / 24.5 =) f/4.69.  Any light cone entering larger than that will be cut down to f/6.9.

Now your OCS will make your dob effectively f/5.625, so more than enough to pass the center of the light cone, but there would be vignetting.   

 

To find the size of the fully illuminated circle, let's do the math for that. If the light cone is f/5.625, this means that for every 65.625mm that the light cone travels, the fully illuminated circle will be reduced 1mm in size.  So here, we divide 5.625 into the light path length of the binoviewer light path and we get (115 / 5.625 =) 20.44, so the 24.5mm fully illuminated circle that enters into the bioviewer is reduced to (24.5 - 20.44 =) 4.06mm.   Outside of this circle, the field is vignetted and the aperture is reduced, but this is pretty much about normal for bioviewers in dobs. 

 

So, it is not the binoviewer that is causing the issue.  You are getting full aperture over a 4mm circle with vignetting outside of that circle, and this will show up exactly the way you illustrated because off axis, you are seeing that the BV front aperture is cutting off your view of the edge of the primary mirror.

 

So far though, no aperture loss. 

 

But now we are going to look at the OCS, or should I say "you" are going to look at the OCS, because here, we would need a piece of data that I don't have and that is the aperture of the OCS. You have given us the lenght of the OCS tube, but not the size of the aperture, but let's use a made up number and you can later substitute the actual value.

Here, we are going to add the light path of the bioviewer and the light path of the OCS together and when I do that, I get (115 + 95.25 =) 210mm.

 

Here, we are going to do the exact same thing as we did with the size of the aperture, the focal ratio, and the light path except with the longer light path.. 

Now, suppose that the front aperture on your OCS is 37mm.  If you divide the total light path length by the focal ratio, you get (210 / 5.625 =) 37.33.  Now if the front aperture on the OCS is 37mm, then you would subtract the 37.33 from the size of the OCS and you get -.33, so yes, the system would be loosing aperture.  If the OCS was smaller than this, the aperture loss would be worst, and if it were longer than this you would get full aperture.  

 

To get the same size field as the binoviewer itself could show, you would need to make the OCS 4mm larger, so about 41.4mm and you are at full aperture but past this and the binoviewer itself becomes the limit as we saw earlier.

 

Now in all of these cases, the aperture is reduced outside of the fully illuminated circle because you are vignetted outside of that circle.

 

Now I could have just asked you for the diameter of your OCS, but the post was intended more as a way to help people in the future with understanding how a configuration behaves.

 

If your OCS is smaller than 37mm, you have a zero fully illuminated circle size (aperture loss) and if it is larger than that, you are working with full aperture, but the largest fully illuminated circle the bioviewer itself will show is 4mm, and past this, the inside edge of the front aperture of the BV is cutting off the off axis rays from the secondary, and that is almost always the case with binoviewers unless a telenegative type lens is used. 

 

So, assuming all the numbers are accurate, if the front aperture of your OCS is larger than 37mm, you are working at full aperture and if it is at least 41mm, then you are maxed out with a 4mm fully illuminated circle. 


Edited by Eddgie, 22 October 2019 - 02:03 PM.

  • Volvonium likes this

#3 Eddgie

Eddgie

    ISS

  • *****
  • Posts: 25,605
  • Joined: 01 Feb 2006

Posted 22 October 2019 - 02:17 PM

Andy your ray trace is showing a condition that almost always exists for a binoviewer.  The most off axis rays generally always get cut off by the binoviewer or OCS, but you don't need the most off axis rays because you can't really see the same size true field when using binoviewers due to the 1.25" limit anyway.  

The calculations above only look at the rays converging toward the exact center of the field of view and as long as a circle large enough to contain the full diameter of your subject is within that circle, it is seen at full brightness and full resolution/contrasts.

As you move to a point where you can see the defocused pattern starts to show cut-off, then at this point, the aperture is reduced along with the resolution and contrast transfer because not all of the off axis rays are able to make it to the focal plane.  

 

The amount of illumination falloff (aperture loss) can be estimated by the amount of area of the mirror that is still visible.

For example in your illustration of the secondary mirror, it appears that about 30% of your mirror is being cut offat the point in the field where you show it, so your aperture, resolution, and contrast at this point in the field would be reduced to an aperture of that size (70%) if you let your subject drift to this point in the field.  

 

In dobs when using binoviewers with wide field eyepieces, illumination falloff (aperture reduction off axis) can be reduced as much as 70%, but when using short focal lenght eyepieces for planets, it is rare to be able to get the object so far away from the center of the field that you would be using an aperture maybe smaller than maybe 70% of the full aperture.

 

(There is also a slight diffraction loss due to the arc of the front aperture as a planet or other object crosses this line, so for best best best planetary observing, the planet should be kept inside the fully illuminated circle size and if you can't see a difference, then just ignore all of this, but that does not make physics stop working.  Ostrich syndrome often prevails in situations like this. LOL. )   


  • Volvonium likes this

#4 Volvonium

Volvonium

    Mariner 2

  • *****
  • topic starter
  • Posts: 277
  • Joined: 06 Mar 2019
  • Loc: Long Beach, CA

Posted 22 October 2019 - 02:43 PM

Thank you very much Eddgie; I'll need some time to process and visualize the geometry so I can wrap my head around it, but it's absolutely helping me to better understand the optical system.

 

I'm not home to measure the aperture of the OCS, but I found a good pic of it that I can use to estimate its aperture-- it is definitely well below 37mm

 

1.25X OCA.jpg

 

By measuring pixels, the OCA's 1.25"/31.75mm body diameter equals ~174 pixels and can used as a measurement of reference...

 

... from the edge of the body to roughly where the aperture begins is roughly 24 pixels on each side, which leaves the aperture in the image at roughly 126px wide, or 72.4% of the body's diameter

 

 

72.4% of 31.75mm = 22.9mm, which seems about right.  I'd guess the OCS's aperture is 23mm +/- 2mm

 

 

The formula using ((light path binoviewer + light path OCS )/ focal ratio) is IMMENSELY helpful and I'll be able to apply that to start understanding binoviewer operation/performance and correction on other telescopes; I can start to understand why longer focal ratio refractors respond to BVs much better.  I'll have to draw this all out when I get home today to help me visually imprint this information.   


Edited by Volvonium, 22 October 2019 - 03:11 PM.


#5 Eddgie

Eddgie

    ISS

  • *****
  • Posts: 25,605
  • Joined: 01 Feb 2006

Posted 23 October 2019 - 10:13 PM

I have to apologize because it is possible that the Siebert uses something like a collimating lens pair or beam expander or something.   My answer was based on more typical doublet type lenses and with your picture, it dawns on me that the Siebert is using something different from this and this would make my previous post invalid. 

 

An example of this is found in solar HA scopes.   Since the light rays must hit the Etalon at zero degrees a lens is placed in front of the etalon, and this lens turns the converging cone into parallel rays.After it passes through the Etalon, a second lens now bends the rays back into the converging cone.

 

If this were the case, then the situation could be somewhat different and the formula above may not apply.

 

In this case, you would probably calculate it just on its distance from the focal plane. 

 

For example, suppose your focuser has to be racked out 30mm to reach focus with a regular eyepiece, but when the OCS is in place, you have to rack the focuser in 25mm.  Now, you add the 95mm light path of the OCS to this, for a total of 120m.

 

The math here is just looking at the size of the light cone as it enters into the front of the OCS and ignores the binoviewer but it is a similar concept.   If the aperture is 22mm and it is 120mm ahead of the focal plane, then you still apply the same math in that you divide 120mm by the focal ratio (4.5) and you get 26.6 so this would be the size of the aperture placed that far in front of the focal plane would have to be to allow the system to work at full aperture so again, consistent with your drawing. 

 

Now if this OCS makes your BV Parfocal with a regular eyepiece (like a Powermate, no inward focuser tube movement to reach focus), then it could in theory be working with zero light path, and this would make my math above not applicable, and I would then apologize for wasting your time.  In this case, with the front of the OCS exactly 96mm ahead of the focal plane, then the converging light cone for the rays going to the center of the field  would be 21mm wide when it hit the front of the OCS, so the OCS would be big enough, and the fully 21mm light cone would be presented to the front aperture of the binoviewer.

 

 

So, try using a regular eyepiece with the field stop at the top of the barrel in the focuser, bring it to focus, and then stick in the binoviewer. If you do not have to rack it in at all to reach focus, then the OCS is intercepting the full light cone and is not causing the aperture loss.  If though, you have to rack it in more than 4.5mm, then there would be some aperture loss and it would depend on how much further you had to rack it in. 

 

Again, I have to apologize.  I should have considered that the Siebert OCS was not a standard doublet, but is instead some kind of collimator or beam expander or something that makes eliminates its path length.  In this case, the most important measurement is how far ahead of the focal plane the front lens is. 


  • Volvonium likes this

#6 Volvonium

Volvonium

    Mariner 2

  • *****
  • topic starter
  • Posts: 277
  • Joined: 06 Mar 2019
  • Loc: Long Beach, CA

Posted 23 October 2019 - 11:02 PM

No worries eddgie, absolutely not a waste of time since it's analyzing and understanding geometries of different optical systems and how they handle the light cone.  I was under the impression The Siebert is like a telecentric barlow, akin to a powermate or ES focal extender, since it has a front and rear lens group. After i put my son to bed in a few, I'll test if the bv + OCS has parfocality with a regular eyepiece. My suspicion is that it does. (Update: no, it doesn't)

 

Sieberts verbiage on his website is a bit confusing, but he seems to recommend his 38mm front aperture OCS for fast reflectors f/2.9 and up. He recommends the type I'm using for f/5 and up iirc


Edited by Volvonium, 24 October 2019 - 12:35 AM.


#7 Volvonium

Volvonium

    Mariner 2

  • *****
  • topic starter
  • Posts: 277
  • Joined: 06 Mar 2019
  • Loc: Long Beach, CA

Posted 24 October 2019 - 12:33 AM

Ok, so i tried a few eps, focusing with the field stop of the ep at or near the top of the Focuser barrel, and then with the siebert ocs threaded onto the end of the nosepiece of the bv, with bv fully inserted into Focuser and same eyepiece fully inserted into diopter.

 

To reach focus with bv + ocs:

Bhz @ 8mm fl - 10mm in to focus

Bhz @ 24mm fl - 10.5mm in to focus

9mm plossl - 22.5mm in to focus

12.5mm edge on LER - 8mm in to focus



#8 Volvonium

Volvonium

    Mariner 2

  • *****
  • topic starter
  • Posts: 277
  • Joined: 06 Mar 2019
  • Loc: Long Beach, CA

Posted 24 October 2019 - 12:47 AM

Also i have an error in my original drawing of the defocused star-- i misrecalled it and the secondary obstruction should be  closer to the center, rather than the outer edge.  Please excuse the crude drawing, i couldn't hold my phone steady enough to take a photo. The drawing assumes a black background

Attached Thumbnails

  • 20191023_224522.jpg


#9 Eddgie

Eddgie

    ISS

  • *****
  • Posts: 25,605
  • Joined: 01 Feb 2006

Posted 24 October 2019 - 08:04 AM

Yeah, it makes sense that it would be some kind of telecentric unit with a focal reducer or even just a Barlow with a focal reducer, though the telecentric makes more sense.   I thought about it last night, and Denk and Earthwin(?) both use doublet Barlow type lenses and focal reducers to get 1.3x or 1.2x so it would not have to be anything exotic, but both of them use much larger OCS lenses (and I don't know for sure that Earthwin uses as simple doublet but it so closely resembles the Denk system that this is my guess). 

 

If the front opening is 22mm and the tube is 96mm then this would mean that it were exactly parfocal with a eyepiece that had the field stop at the top of the eyepiece holder, the front opening would be 96mm in front of the focal plane and this means that at f/4.5, the opening would fully intercept the entire on-axis light cone (96 / 4.5 = 21.33mm).

 

Now with only about 7mm inward focus, you would push the front aperture far enough forward to take off  off an additional millimeter and a half ,  so anything over that would cause aperture loss just as you are showing in your ray trace. 

 

To be fair, loosing a couple of millimeters of primary may not be the end of the world if the goal is a wider field of view, but if you are doing planets and want to let the planet drift, then you would probably want to use the a full telecentric 2x that had a bigger aperture, or kept the tube shorter.

 

Even if power switch systems were loosing aperture in low power mode (I figured it out once and it would have to be a very fast system but I forget how fast) it would not affect planets because most people would use straight through for planets, and both arms out on a power switch system requires that the focuser be racked out about 45mm, and that would move the front aperture far enough to the rear that you could intercept even very fast light cones. 

 

Again, the cut off of the mirror you see is just the vignetting of the off axis rays. 

 

And silly me.. I should have just asked you to see if you could see the entire diameter of the primary mirror when the binoviewer is at infinity focus. I am trying to have you calculate something that you should be able to just look in and see. What's wrong with me?   So, just put in the eyepiece that takes the most inward travel, focus at infinity, and measure the focuser extension.  Now in the daylight, just take out the eyepiece and look into the BV and see if you can see the whole mirror. 



#10 Eddgie

Eddgie

    ISS

  • *****
  • Posts: 25,605
  • Joined: 01 Feb 2006

Posted 24 October 2019 - 08:06 AM

And an important note on the last test.  If the mirror is not exactly centered when you do this last test, you have either collimation issues or focuser tilt, or your focuser is deflecting under the load of the binoviewers. I just had a total brain frat though and should have first suggested that you simply visually test it. Why did I make it so difficult?



#11 Volvonium

Volvonium

    Mariner 2

  • *****
  • topic starter
  • Posts: 277
  • Joined: 06 Mar 2019
  • Loc: Long Beach, CA

Posted 24 October 2019 - 11:49 AM

A good case of the old space pen vs pencil anecdote/urban legend!  It's been a helpful exercise and just checking how much of the primary mirror can be seen completely went over my head as well.  Will try out later tonight.  The simple answer would seem that I need a larger aperture aperture OCA for my focal ratio, but now there is a good foundation for understanding the why.  



#12 Eddgie

Eddgie

    ISS

  • *****
  • Posts: 25,605
  • Joined: 01 Feb 2006

Posted 24 October 2019 - 01:41 PM

A good case of the old space pen vs pencil anecdote/urban legend!  It's been a helpful exercise and just checking how much of the primary mirror can be seen completely went over my head as well.  Will try out later tonight.  The simple answer would seem that I need a larger aperture aperture OCA for my focal ratio, but now there is a good foundation for understanding the why.  

Well, the trade-off is magnification.   A Denk will have a clear aperture of 45mm, but in many scopes it might not reach low power without threading out the extension tube and if you move the extension tube out, the low power will go from 1.3x to maybe 1.4x (rarely is this much extension needed). So you might pick up some field illumination and possibly a few millimeters of aperture, but you might loose some true field. 

 

Now I still think that for Newtonians, when all of the dust settles, the power switch type systems are (in my opinion) the absolute best choice because of the low power and the fact that you now have three useful powers with one pair of eyepieces. 

 

Still, it is a big cost increase to save a few millimeters of aperture (if you are loosing aperture which we can't be positive of yet.). 


  • Volvonium likes this

#13 Volvonium

Volvonium

    Mariner 2

  • *****
  • topic starter
  • Posts: 277
  • Joined: 06 Mar 2019
  • Loc: Long Beach, CA

Posted 28 October 2019 - 02:28 PM

 

And silly me.. I should have just asked you to see if you could see the entire diameter of the primary mirror when the binoviewer is at infinity focus. I am trying to have you calculate something that you should be able to just look in and see. What's wrong with me?   So, just put in the eyepiece that takes the most inward travel, focus at infinity, and measure the focuser extension.  Now in the daylight, just take out the eyepiece and look into the BV and see if you can see the whole mirror. 

 

I had some time this weekend to do this and with just the BV and no eps or OCS in my f/4.5 scope, the BV cuts off maybe a few mm of the primary mirror, which isn't bad-- the BV's clear aperture is definitely adequate.  With the OCS installed and looking through the BV with no eyepieces, there is significant cut off.   

 

It seems that my most economical solution is to just use the BV with my slower scopes and/or use my existing 2X focal extender and just accept having a relatively high magnification factor until I get a larger aperture OCS.  Some wider FOV longer focal length eyepieces will help me eke out a wider view as well (My Baader Zooms have 50deg fov @ 24mm). 

 

Some aperture loss is going to be inevitable with faster dobs and I can live with that, since my f/4.5 isn't my only scope.  I definitely have a better understanding of the system now, so I'm very very grateful for your input and help to walk me through this.



#14 Eddgie

Eddgie

    ISS

  • *****
  • Posts: 25,605
  • Joined: 01 Feb 2006

Posted 29 October 2019 - 09:45 AM

Yes, I am not surprised, and again, I super apologize for not simply telling you to do this test, but it would appear that in my initial posts, the math was probably a decent ballpark in that I had said that with 38mm, you would be at aperture reduction and since you report considerably more than a couple of MM loss, then that aligns with my initial analysis.

 

Now the next thing is that I recommend using a Barlow for all planetary observing and even if you can reach focus with an OCS (something other than a simple Barlow type lens) I recommend that you star test to see if the added optics cause some spherical aberration.   I have used a couple of correctors that induced half a wave of SA and for planets, that imposes enough damage to easily see when compared directly to a Barlow that does not induce spherical aberration.

 

In fact, I had a dialog with Harry about this, and he said that it might be best not to use one of his GPCs for planets but that was many years ago, and maybe he has changed the design. 

 

Your post though is a great model for why I put a lot of emphasis on analyzing and testing the configuration.  People buy things like the OCA you bought, and they can work as stated, but that does not mean that they will always work great, and in your case, you can see that there was a pretty severe penalty for that configuration.  The reason I do the analysis is so that I don't buy something that will entail some very meaningful compromise.

 

Anyway, happy that you were able to make a comprehensive assessment. It is always good to know what you are working with.


Edited by Eddgie, 29 October 2019 - 09:46 AM.

  • Volvonium 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