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

Understanding, Using, and Choosing Refractors (a guide)

beginner equipment optics refractor
  • Please log in to reply
42 replies to this topic

#1 jay.i

jay.i

    Mercury-Atlas

  • *****
  • topic starter
  • Posts: 2,764
  • Joined: 11 Jun 2017
  • Loc: Minneapolis, MN

Posted 12 November 2018 - 06:05 PM

Hello!

 

Over the past few months, I have been slowly working on this amateur guide on understanding refracting optics, how to use them properly, and choosing the right refractor for you. In this guide you may find some errors or things you disagree with. There are a lot of ways to describe some of these concepts and you may disagree with the way I've explained them. If you do, please let me know and I'd be happy to update the guide! I am a bit of a perfectionist, and feel like I will never post this for fear of being wrong, but even worse, I might mislead people. That said, I figure the best thing to do is post this. I have gotten some great feedback from people already and feel like it's probably high time I release this to the general public, perfectionism be darned! Note that this guide is not meant to be an exhaustive resource, but a primer to get people on the right track and explain some concepts they may have heard but not understood.

 

Please, if you have any comments or critique, don't hesitate to let me know. I hope that some people can learn from this and better understand this sometimes confusing world of optics. I know that I still have plenty to learn as well! If you learned anything from this, please do let me know. Feel free to share this with members of your local astronomical community too!

 

Guide (v1.02) as of Nov 12, 2018https://www.dropbox....actors.pdf?dl=0

 

PDF scanned with VirusTotalhttps://www.virustot...8bf65/detection

 

If you'd like the guide in another format, such as a rudimentary webpage with each slide as an image, let me know and I can try to put that together. If you have any other questions or concerns, as always, don't hesitate to reach out!

 

Thanks for looking, and I wish you clear and steady skies!

 

-j


Edited by jay.i, 12 November 2018 - 11:10 PM.

  • Paul Hyndman, rustynpp, punk35 and 31 others like this

#2 Erik Bakker

Erik Bakker

    Cosmos

  • *****
  • Moderators
  • Posts: 8,395
  • Joined: 10 Aug 2006
  • Loc: Netherlands, Europe

Posted 12 November 2018 - 06:22 PM

Thank you for putting together this very informative and well written guide Jay bow.gif



#3 punk35

punk35

    Apollo

  • *****
  • Posts: 1,294
  • Joined: 26 Jan 2005
  • Loc: Adrian Michigan

Posted 12 November 2018 - 06:31 PM

Excellent write up. Thanks for taking the time to do that and for sharing it with us.



#4 Steve Allison

Steve Allison

    Apollo

  • *****
  • Posts: 1,242
  • Joined: 25 Nov 2016
  • Loc: Olympia, Wash. 98502

Posted 12 November 2018 - 07:01 PM

I like it. Great job!

 

Steve



#5 Tyson M

Tyson M

    Skylab

  • *****
  • Posts: 4,157
  • Joined: 22 Jan 2015
  • Loc: Alberta, Canada

Posted 12 November 2018 - 07:43 PM

Well written and put together Jay! 

 

I think this deserves a pin to this forum for newbies wanting to jump into the madness that is the refractor forum. 


  • rerun likes this

#6 vtornado

vtornado

    Mercury-Atlas

  • *****
  • Posts: 2,505
  • Joined: 22 Jan 2016
  • Loc: Northern Illinois

Posted 12 November 2018 - 08:29 PM

Hello a very enjoyable read, I learned a lot.

 

I am not an optical technician but I thought that Spherical Abberation was due to the front element, being a section of a sphere.  Just as a spherical mirror has spherical abberation.  An just as a spherica mirror, the slower the  f ratio of the lens, the smaller the effects of SA.  The optical designer can account for the spherical aberration of the first element by the second having the opposite abberation.  (see I'm not an optical engineer).

 

One way to get around this is to have the elements be Ashpherical, but these lenses are very difficult to shape. 

 

Am I kind of correct?


Edited by vtornado, 12 November 2018 - 08:32 PM.


#7 RaulTheRat

RaulTheRat

    Viking 1

  • -----
  • Posts: 675
  • Joined: 09 Oct 2018

Posted 12 November 2018 - 08:35 PM

Nice guide, thanks. I think you have the use of the term depth of field the wrong way round though, I have always understood it in the sense that more DoF means more in focus at once (and never seen anyone use the term in the opposite sense).

#8 jay.i

jay.i

    Mercury-Atlas

  • *****
  • topic starter
  • Posts: 2,764
  • Joined: 11 Jun 2017
  • Loc: Minneapolis, MN

Posted 12 November 2018 - 09:02 PM

Hello a very enjoyable read, I learned a lot.

 

I am not an optical technician but I thought that Spherical Abberation was due to the front element, being a section of a sphere.  Just as a spherical mirror has spherical abberation.  An just as a spherica mirror, the slower the  f ratio of the lens, the smaller the effects of SA.  The optical designer can account for the spherical aberration of the first element by the second having the opposite abberation.  (see I'm not an optical engineer).

 

One way to get around this is to have the elements be Ashpherical, but these lenses are very difficult to shape. 

 

Am I kind of correct?

You might be! I haven't heard of it in that context before. Primary mirrors for Newtonian reflectors still have spherical aberration, but I'm not sure that the secondary mirror accounts for any of the SA in the primary. This would translate to my understanding of how refracting lenses cause/correct SA, but it's possible I'm just not understanding it correctly. I think the more basic description of SA in the presentation is more fitting for a document of its caliber (in terms of how deep it goes and who it's intended for) but I certainly would like to learn more about your claim about SA being caused by the front element of a lens system. There are objectives with all spherical elements (like the AP130GTX and new Stowaway), so surely SA doesn't require aspherical elements for correction.



#9 jay.i

jay.i

    Mercury-Atlas

  • *****
  • topic starter
  • Posts: 2,764
  • Joined: 11 Jun 2017
  • Loc: Minneapolis, MN

Posted 12 November 2018 - 09:05 PM

Nice guide, thanks. I think you have the use of the term depth of field the wrong way round though, I have always understood it in the sense that more DoF means more in focus at once (and never seen anyone use the term in the opposite sense).

I was struggling with how to phrase it. I thought more depth of field would mean the field seems more deep, there is more blurring, more subject separation, and less depth of field would mean the field seems less deep, with less blurring and more stuff is in focus. I didn't want to say the focal plane is larger or smaller because technically the focal plane is just the plane at which the image comes to focus, but depth of field affects how much of it is in focus. At least that's how I understand it. If other people mention this wording (less vs more depth) I'll go change it, since I was on the fence about it already.



#10 Moondust

Moondust

    Viking 1

  • -----
  • Posts: 618
  • Joined: 14 Nov 2006
  • Loc: The Show Me State

Posted 12 November 2018 - 09:30 PM

Nice guide, thanks. I think you have the use of the term depth of field the wrong way round though, I have always understood it in the sense that more DoF means more in focus at once (and never seen anyone use the term in the opposite sense).

This is the correct use of the term. 



#11 jay.i

jay.i

    Mercury-Atlas

  • *****
  • topic starter
  • Posts: 2,764
  • Joined: 11 Jun 2017
  • Loc: Minneapolis, MN

Posted 12 November 2018 - 09:32 PM

This is the correct use of the term. 

Whose use, mine or Raul's?



#12 DVexile

DVexile

    Vostok 1

  • *****
  • Posts: 125
  • Joined: 17 Mar 2016

Posted 12 November 2018 - 09:52 PM

Whose use, mine or Raul's?

Raul’s

 

https://en.m.wikiped.../Depth_of_field



#13 DVexile

DVexile

    Vostok 1

  • *****
  • Posts: 125
  • Joined: 17 Mar 2016

Posted 12 November 2018 - 10:15 PM

This is good stuff, nice job!

 

Corrections:

 

Already mentioned depth of field you have the sense backwards.

 

Focal plane has nothing to do with F-ratio.  Depth of field most certainly does, but not the focal plane which by definition is an infinitely thin surface.

 

The section on spherical aberration is entirely wrong.  Has nothing to due with smoothness of the optics.  Nothing to do with perfect spheres either, in fact in most cases caused by the fact that we naturally grind spherical surfaces but in fact to avoid SA we need aspherical surfaces which are harder to grind (the paraboloid Newtonian mirror being a classic example).

 

Similar error on last bullet in spherochromaticism section - nothing to do with smoothness of the optical elements. 

 

On field curvature the statement it only has to due with focal length is only true for the Newtonian reflector.  For a refractor it is dependent on the optical design and can be made nearly arbitrarily better or worse in trade off with optimizing other aberrations. But in general all else being equal short faster refractors tend to have more FC than longer slower refractors. 

 

Will need to to read more later, but good job putting this together so far.  With some edits this will be an amazing reference for folks. 


  • IMB likes this

#14 jay.i

jay.i

    Mercury-Atlas

  • *****
  • topic starter
  • Posts: 2,764
  • Joined: 11 Jun 2017
  • Loc: Minneapolis, MN

Posted 12 November 2018 - 10:17 PM

Yep, shoulda probably Googled it. I'll update it here shortly.



#15 Moondust

Moondust

    Viking 1

  • -----
  • Posts: 618
  • Joined: 14 Nov 2006
  • Loc: The Show Me State

Posted 12 November 2018 - 10:26 PM

Whose use, mine or Raul's?

Rauls. I was a professional photographer before resently retiring.



#16 jay.i

jay.i

    Mercury-Atlas

  • *****
  • topic starter
  • Posts: 2,764
  • Joined: 11 Jun 2017
  • Loc: Minneapolis, MN

Posted 12 November 2018 - 10:29 PM

[...]

 

The section on spherical aberration is entirely wrong.  Has nothing to due with smoothness of the optics.  Nothing to do with perfect spheres either, in fact in most cases caused by the fact that we naturally grind spherical surfaces but in fact to avoid SA we need aspherical surfaces which are harder to grind (the paraboloid Newtonian mirror being a classic example).

 

Similar error on last bullet in spherochromaticism section - nothing to do with smoothness of the optical elements. 

 

On field curvature the statement it only has to due with focal length is only true for the Newtonian reflector.  For a refractor it is dependent on the optical design and can be made nearly arbitrarily better or worse in trade off with optimizing other aberrations. But in general all else being equal short faster refractors tend to have more FC than longer slower refractors. 

 

[...]

Thanks for the feedback; let me just clarify a few things so I can understand better.

 

When you say the section on SA is entirely wrong... how would you describe it? Perhaps you can speak to the difference between a smooth polish and a "smooth figure". I have heard it blurred together too much to really know what the difference is.

 

How can a wavefront be smooth if the optics are not? Or can the optics be smooth but the figure rough? What exactly is the difference? Is the figure an actual design to be polished into the optical surface? This would make sense I suppose but it's hard to understand exactly how these things combine to make the whole. Anyway, if I can wrap my head around this then I can get rid of the misleading information in the deck before I can't edit the main post anymore.

 

As for field curvature, I checked here just now: https://www.telescop...t/curvature.htm

 

It talks about field curvature for a lens being affected by not only the focal length but also the refractive index. I have seen people say that some triplets show less field curvature than maybe a doublet of equal focal length, but I haven't seen many if any explain how it's done, other than "it's in the design." This is something else I'd love to know more about. I should change the wording in the presentation.

 

Edit: Update v1.01 is live as of 9:34pm CST with depth of field and field curvature corrections.

 

Edit2: Update v1.02 is live as of 10:10pm CST with more depth of field corrections. lol


Edited by jay.i, 12 November 2018 - 11:11 PM.


#17 DVexile

DVexile

    Vostok 1

  • *****
  • Posts: 125
  • Joined: 17 Mar 2016

Posted 12 November 2018 - 11:13 PM

I’m probably the wrong person to best describe the various figuring terms, but I think some of the confusion may come from the whole reflector vs. refractor thing.  And then there is ambiguity in “smooth” as well!

 

For a Newtonian reflector if you have a paraboloid figure to the mirror (figure meaning the shape of the cross section of the mirror) you will have no SA.  When you grind a mirror it will form a spherical surface naturally with no effort - if you think about it if you rub two blanks together they can *only* form a spherical shape between them.

 

Once you have a spherical surface of the right focal length it must be adjusted to form a paraboloid.  This means removing material from the outer edge and requires different polishing strokes that preferentially remove material from the outer edge. This is very challenging to get right.  You might remove too much from say 50-75% of the way to the edge and not enough from 75% out to the edge.  This will result in SA because two different “zones” (donut shaped regions centered on the mirror) now focus light at a different focal length.

 

I think “smooth figure” usually means that the surface properly follows the paraboloid shape but often the result is somewhat wavy.  A key point though is that a sphere is not the correct shape to begin with though!  If you make a perfectly smooth spherical mirror you will have SA.

 

Now the above is again all for reflectors which is probably a distraction in a document for refractors!  So I’d drop the whole idea of smooth and figuring as that really doesn’t apply to refractors at all.

 

As far as I know all amateur refractors use spherical lenses. Aspheres are common in camera lenses these days because we can economically make small aspheric elements but telescope objectives are far too large to do that economically. 

 

The way SA is dealt with in a refractor that only has spherical surfaces is by combining multiple spherical surfaces of different radii. In a doublet there are four spherical surfaces (two on each element).  The focal length actually doesn’t constrain any one of those surfaces!  It to a first order only constrains the relationship between the front most surface and the rear most surface.  So you can in theory choose *any* radius of curvature for the front surface of a doublet and your chosen focal length will the tell you what the rear surface curvature must be.  Completely unconstrained are the inner surfaces of the doublet.  You can change those however you want and not effect the focal length of the doublet.  And then there is the spacing of the two elements. And then the thickness of the two elements.  And the index of refraction for the two different glasses...

 

The key thing to understand is that in a refractor for a given focal length there are multiple spherical shapes of the lens elements that will give the same focal length.  Depending on which shape you choose you will get more or less of a particular aberration - SA being just one.  

 

This shows how the shape of a single element lens optimizes for SA: http://hyperphysics....eoopt/aber.html

 

With a Newtonian reflector there is only one “correct” shape for a given focal length and F-ratio.  That is a paraboloid and it will have no SA and no chromatic aberration. It will have coma and there is nothing you can do about it with the mirror shape.  For a Newtonian once you set the focal length and F-ratio there are absolutely no choices left for the designer.

 

As described for a doublet there are a host of choices that allow the designer to trade one aberration against another. There is no “correct” design since one can decide they want to optimize on axis SA at the expense of off axis coma or they can choose the opposite depending on the application or desired use. 

 

I’m not sure that was so illuminating... let me try this summary:

 

In a Newtonian SA is the result of a manufacturing *error* - the mirror is the wrong shape and perhaps this could be called a non-smooth figure.

 

In a refractor SA is usually the result of a design *choice* - the designer must balance multiple aberrations against each other in the design and if they over optimize SA they may end up with worse performance due to a different aberration (coma, astigmatism, field curvature).  

 

Obviously if someone screws up manufacturing a refractor that could also introduce even more SA than should be there!  But that isn’t the typical source of SA in a refractor, rather it is the result of design choices. 


  • IMB and jay.i like this

#18 jay.i

jay.i

    Mercury-Atlas

  • *****
  • topic starter
  • Posts: 2,764
  • Joined: 11 Jun 2017
  • Loc: Minneapolis, MN

Posted 12 November 2018 - 11:31 PM

Give me some time to do some more reading, but that all makes sense I suppose. I feel like I have seen a lot of people talk about smooth figures with refracting lenses; I know I have. If you were going to describe why spherical aberration happens, in a way that isn't too complicated given the scope of the guide, how would you do it? It's corrected for differently in reflecting and refracting objectives, that makes sense now, but why does it happen to both mirrors and lenses?

 

Thanks very much for that write-up!!



#19 Kiwi-bloke

Kiwi-bloke

    Explorer 1

  • -----
  • Posts: 78
  • Joined: 12 Mar 2017
  • Loc: New Zealand

Posted 13 November 2018 - 03:24 AM

Thanks mate, that article was very informative, and slightly disapointed me that my WO ZS103 was a mass produced chinese made scope that many make. Though it produces fantastic images and i enjoy it immensly and as you said, that's what counts.

 

I actually thought William made the scopes in house, so that was news to me.

 

You live and learn.

 

Cheers.



#20 DVexile

DVexile

    Vostok 1

  • *****
  • Posts: 125
  • Joined: 17 Mar 2016

Posted 13 November 2018 - 06:15 AM

If you were going to describe why spherical aberration happens, in a way that isn't too complicated given the scope of the guide, how would you do it? It's corrected for differently in reflecting and refracting objectives, that makes sense now, but why does it happen to both mirrors and lenses?

The reason it happens is actually right in the name!  It is an aberration that occurs when we use spherical surfaces instead of paraboloid surfaces to make a lens or mirror.  If you want to focus a wavefront to a point the correct geometric shape to do that (with a lens or a mirror) is a paraboloid.  Unfortunately paraboloids are hard to grind and when we grind optics the natural shape we get are spherical surfaces.  A sphere is a reasonable approximation of a paraboloid for large radii of curvature (i.e. slow optics) and a worse and worse approximation for small radii of curvature (i.e. fast optics).

 

For a Newtonian reflector there is only one surface to work with at all in the first place - the primary mirror surface.  If we want to make a fast mirror the only option we really have to combat SA is to actually make the surface a paraboloid.  The process of converting the natural spherically shaped mirror into a paraboloid is called "figuring" in telescope maker parlance.  In more complex reflectors (various cats) there are more optical elements in play and so other kinds of optimization can be done which means they can use a spherical primary mirror instead which is easier to manufacture.

 

Again for a Newtonian reflector if we make the perfect paraboloid we will still end up with coma and we can use a second optical system (a coma corrector near the eyepiece) to try to improve that.  For a short focal length we'd still have field curvature and we could again use a second optical system (a field flattener near the eyepiece) to try to improve that.

 

For a refractor it turns out to be impractical to make lenses with paraboloidal surfaces.  Instead two (or more) lenses made with spherical surfaces are used.  By altering the radii of curvatures used on all the surfaces and their spacing (and their indices of refraction) the designer can focus a wavefront nearly to a point despite not having used paraboloidal surfaces.  The thing is all of the other aberrations (coma, astigmatism, field curvature, distortion) also depend on those very same parameters so once you are designing a doublet you are now trading off all these aberrations with each other as you choose the radii of curvature and spacings of the surfaces.  But again you are not using a paraboloidal surfaces in a typical amateur refractor doublet.

 

They key thing to understand though is that "spherical aberration" isn't because we failed to make a spherical surface, rather it is *because* we used a spherical surface when we should have used a paraboloid!  In a refractor we use multiple spherical surfaces to approximate the effect of a paraboloidal surface.  In a reflector we actually try to make a paraboloid and when we fail to do a good job of that we get light focused at different focal lengths just like if we had used a spherical surface.  So in the reflector case we call the result of improperly figuring the mirror to a paraboloid shape "spherical aberration" because the result is similar to what we'd get if we had just used a spherical surface.

 

Clear as mud?


Edited by DVexile, 13 November 2018 - 06:17 AM.

  • IMB and jay.i like this

#21 CounterWeight

CounterWeight

    Star walker

  • *****
  • Posts: 10,928
  • Joined: 05 Oct 2008
  • Loc: PDX, OR.

Posted 13 November 2018 - 01:05 PM

Great write up and enjoyable to read - nice job!.

 

As there is a good amount of technical data in there i would recommend to add that 'Strehl'  in general and where available, and the byproducts of it's measurement are 'manufacturer specific', and not an actual standard used between manufacturers.  I am not criticizing Strehl as a measurement, just I feel it critical to understand it within the real world brackets. 

 

Hard to communicate well about star testing as to me that is something that experience reveals? Glad to see mention of 'the book'. That said, the mistake I've encountered most often with new refractor users is being under mounted or not allowing enough acclimation time, cannot achieve focus, choosing objects low to horizon, testing under jet stream, .... , maybe include something about first light and what to try, things to know about sky conditions including LP, how to go about it to a new users benefit.?



#22 peleuba

peleuba

    Vanguard

  • *****
  • Posts: 2,059
  • Joined: 01 Dec 2004
  • Loc: North of Baltimore, MD

Posted 13 November 2018 - 02:30 PM

Great write up and enjoyable to read - nice job!.

 

As there is a good amount of technical data in there i would recommend to add that 'Strehl'  in general and where available, and the byproducts of it's measurement are 'manufacturer specific', and not an actual standard used between manufacturers.  I am not criticizing Strehl as a measurement, just I feel it critical to understand it within the real world brackets. 

 

Strehl is a mathematical expression derived from an interfemotric measurement of an optical wavefront/surface.  The standard is this:  a Strehl of 1.00 (unobtanium perfect lens) has  82% of the light contained inside the Airy disk.   A Strehl of 0.95 has 95% of the perfect 82% of light concentrated in the Airy disk.  

 

There is some variability in the use of Strehl between manufacturers so its best to ask how that number was derived.  Some will state the design spec Strehl in their marketing materials.  Others actually measure each optic and incorporate Strehl measurement to help plan subsequent figuring sessions. Even others will measure Strehl at the end of the process as a sort of pass/fail metric.  Its important to be an informed consumer as different manufactures adhere to different standards of quality.  If your favorite manufacturer imports the lenses they really need to test a statistically significant sample in order to judge the goodness of a particular run of telescopes. 

 

Optical testing is often a wobbly stack with many variables.   Some interferometers rely on auxiliary optics that need to be extremely accurate so as not to add any error to the test results.  These instruments, like those from ZYGO, usually have certified optics traceable to a standards bureau and use coherent lasers as the light source.  Others (common path ie. Bath) use off the shelf parts but can be extremely accurate, too.  In either case, the data reduction software is needed to compute the PtV, RMS and Strehl.  This software can be used to subtract out the inherent test error in the setup.

 

In any event, even when manufacturers differ, if Strehl is derived in an honest, consistent and skilled manner with sound methodology the numbers will be surprisingly close.  


Edited by peleuba, 13 November 2018 - 02:53 PM.

  • Paul G and IMB like this

#23 peleuba

peleuba

    Vanguard

  • *****
  • Posts: 2,059
  • Joined: 01 Dec 2004
  • Loc: North of Baltimore, MD

Posted 13 November 2018 - 02:51 PM

In a Newtonian SA is the result of a manufacturing *error* - the mirror is the wrong shape and perhaps this could be called a non-smooth figure.

 

In a refractor SA is usually the result of a design *choice* - the designer must balance multiple aberrations against each other in the design and if they over optimize SA they may end up with worse performance due to a different aberration (coma, astigmatism, field curvature).  

 

 

Of everything you wrote, the above is the most important take-away.

 

Correcting for SA in a Newtonian mirror is orders of magnitude easier then correcting it in a lens.  SA in a mirror is an error in figuring.  SA in a lens has a color component that comes from the differing wavelengths of light each focusing at slightly different spots.  Its also inherent in a refractor design that is left all spherical.  However, if the focal ratio is long enough, the SA may be below the threshold of being obtrusive.  For short focal length refractors, assuming proper glass matching, the best manner for controlling SA is to have an air spaced design and aspherize it.  


Edited by peleuba, 13 November 2018 - 02:51 PM.

  • Paul G, DVexile and IMB like this

#24 CounterWeight

CounterWeight

    Star walker

  • *****
  • Posts: 10,928
  • Joined: 05 Oct 2008
  • Loc: PDX, OR.

Posted 13 November 2018 - 06:23 PM

Hi Paul, 

 

My point was that it is not a standard, a method is not a standard. I think it is great / fantastic that some vendors will do a bench test and provide their Strehl ratio number - so I am never intending to 'put down' the method. Methodology is usually a part of any standard, but in and of itself is not a standard.  Confusing the two I think just increases the misunderstanding. If someone has never done metrology then I suppose the confusion might seem unimportant.  A good paper here

 

I hope at some future date an agreed on mesh sample density per area  and numerical methods (along with all the other things) can be agreed on by all and done the same way so that Strehl ratio number would be 'portable' across manufacturers, and independent testing companies.



#25 RadioAstronomer

RadioAstronomer

    Mariner 2

  • *****
  • Posts: 280
  • Joined: 13 Jun 2018
  • Loc: Tucson, AZ

Posted 14 November 2018 - 01:16 AM

This is great. Thank you for posting it!


Edited by RadioAstronomer, 14 November 2018 - 01:17 AM.



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





Also tagged with one or more of these keywords: beginner, equipment, optics, refractor



Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics