why additional optics, small sensor

Here you can discuss everything related to taking light micrographs and videos.
Post Reply
Message
Author
mete
Posts: 114
Joined: Thu Aug 26, 2021 8:31 am
Location: Switzerland

why additional optics, small sensor

#1 Post by mete » Wed Sep 01, 2021 8:14 am

Hello,

A few questions and/or thoughts I will be happy to hear a feedback.

I have a Zeiss Stemi stereo and I will get a Zeiss Axiolab soon. Initially I tried using Stemi with Axio ERc 5s but I was not happy with it so I moved to Fuji X-T2. I am using it without any additional optics so I have a small vignetting. For both microscopes, I do not want to use any additional optics and also preferably have minimal or no vignetting. Any known brand 1” dedicated camera is too expensive, so it seems like MFT is a good compromise. I actually bought an E-M5 mk2 to try and waiting for it. Am I missing something here ? To me:

- it sounds not reasonably to use additional optics, because why to introduce additional distortion, is it that negligible ? I am not 100% sure if it is the reason but I do not prefer using 2.5x Zeiss tube with Sony a7r2 for example comparing to direct Fuji X-T2.

- it also sounds not reasonable to use a very tiny sensor if I already have an fov quite large (eg. 1/2” sensor vs 23mm fov) and nyquist wise it seems there is no need to have anything below 4um pixel pitch (am I correct in this calculation?), so any slr like camera should be ok in this sense.

- any additional optics also have some cost and it is quite high for authentic ones. I rather would spend it somewhere else.

- only point I can think of is low light capability, but I can only compare a ~1” dedicated to an slr, where the former is more/much more expensive. I had no problem with light in Stemi but with a compound when would this be a problem ? Why/in what type of application someone needs a large sensor/dedicated/cooled camera ? Only for FL ? Only for high mag/100x work ?

Mete

Hobbyst46
Posts: 4277
Joined: Mon Aug 21, 2017 9:02 pm

Re: why additional optics, small sensor

#2 Post by Hobbyst46 » Wed Sep 01, 2021 8:56 am

Note that dedicated microscope cameras, for example from Olympus, have a relatively small sensor (<1") yet offer high sensitivity to perform under low light conditions. Fluorescence is the classical application. Cooling enhances the sensitivity.

mete
Posts: 114
Joined: Thu Aug 26, 2021 8:31 am
Location: Switzerland

Re: why additional optics, small sensor

#3 Post by mete » Wed Sep 01, 2021 10:32 am

I dont know much about FL but if it is a need I understand, similar to some astro conditions maybe. What about BF and PH ?

Hobbyst46
Posts: 4277
Joined: Mon Aug 21, 2017 9:02 pm

Re: why additional optics, small sensor

#4 Post by Hobbyst46 » Wed Sep 01, 2021 11:45 am

mete wrote:
Wed Sep 01, 2021 10:32 am
I dont know much about FL but if it is a need I understand, similar to some astro conditions maybe. What about BF and PH ?
FL demands a highly sensitive camera because the intensity of fluorescence is usually low, or very low, lower by orders of magnitude than the excitation (illumination) intensity. The brightness of BF and even phase contrast is much higher so any camera is OK.

apochronaut
Posts: 6272
Joined: Fri May 15, 2015 12:15 am

Re: why additional optics, small sensor

#5 Post by apochronaut » Wed Sep 01, 2021 12:11 pm

When people move over from telescopes to microscopes, I am seeing a lot of assumptions based on telescope experience. I would advise caution in this, since although they are both optical instruments , the design perspective for each is very different.
You mention that you don't want a photo lens because it will create added distortion. This appears to be telescope thinking but I could be wrong. In fact, a photo relay lens in a microscope is there to correct for various aberrations or distortions. Only if you can be 100% sure that your objective or objective/telecentric lens combination is creating as aberration and distortion free an image as is possible should one consider not using a photo relay lens.

mete
Posts: 114
Joined: Thu Aug 26, 2021 8:31 am
Location: Switzerland

Re: why additional optics, small sensor

#6 Post by mete » Wed Sep 01, 2021 4:52 pm

I dont have much experience with telescopes also, just happen to know the very basics. I am really interested to know the details of this if true: “a photo relay lens in a microscope is there to correct for various aberrations or distortions”. I assume if it is true, this also means eyepiece parts (things after tube lens? if I use term correct) are also correcting things. By the way I am talking only about infinity corrected designs.

Dubious
Posts: 426
Joined: Sun May 09, 2021 7:55 pm

Re: why additional optics, small sensor

#7 Post by Dubious » Wed Sep 01, 2021 7:04 pm

Yes, at least with finite systems (160mm, etc.), eyepieces usually are used to correct aberrations introduced by the objective. Since different manufacturers do it different ways, eyepieces and objectives often need to be from the same manufacturer for acceptable results. Modern Infinity systems, however, have a tube lens between which can be used to correct aberrations that were not corrected in the objective itself. A few manufacturers (is it just Nikon?) have always done more of the corrections in the objective itself. Quite different from the eyepiece on a telescope, which generally is just an eyepiece.

mete
Posts: 114
Joined: Thu Aug 26, 2021 8:31 am
Location: Switzerland

Re: why additional optics, small sensor

#8 Post by mete » Wed Sep 01, 2021 7:33 pm

If the tube lens is important in this sense, I mean if it affects the image quality, then even using the same objective, are there difference in image quality between different microscopes (of the same brand), assuming the quality of tube lens might also be different ? If so, is that the reason they say (at least Zeiss, I didnt check others) recommended objectives ? ie apochromats are not in recommended objectives list for basic models (axiolab).

Edit: I didnt mean first if actually, tube lens has to be important but I was saying anything after that is an additional complication, so I was saying why to use additional optics eg for cameras. Not related to the title of this thread but I am wondering if there is a quality difference of tube lenses between models and really there is less benefit of using a “too” good objective for a simpler microscope.

Dubious
Posts: 426
Joined: Sun May 09, 2021 7:55 pm

Re: why additional optics, small sensor

#9 Post by Dubious » Wed Sep 01, 2021 8:37 pm

That's an interesting question that someone else here can better answer, since my experience is almost entirely with 160mm microscopes, and they do not have tube lenses. But I will say that certainly, in theory, tube lenses by the same manufacturer could vary in quality between models. Still, the tube lens is not that expensive, and I doubt that a big name manufacturer would be motivated to include an inferior tube lens in some models. But there is no single standardization on tube lenses, and the needs of a large research microscope designed to accommodate a multitude of filters and prisms, different types of lighting and maybe even special objectives, are different from those of a simple student microscope, so I expect there are some differences.

Interestingly, it is possible to buy third-party tube lenses that apparently have advantages for certain applications, see ThorLabs link below. While the focal length varies by the intended group of microscope manufacturer, these lenses seem to assume that necessary corrections for objective aberrations are performed in the objective itself.

https://www.thorlabs.com/newgrouppage9. ... up_id=5834

"Microscope manufacturers design their systems with one of several standard tube lens focal lengths, including 200 mm (typical for Thorlabs, Nikon, Leica, and Mitutoyo microscopes), 180 mm (typical for Olympus microscopes), and 165 mm (typical for Zeiss microscopes). Similarly, microscope objectives are designed to provide the magnification engraved on the housing when used with a tube lens of the appropriate standard focal length. We offer infinity-corrected tube lenses in all of these focal lengths so that home-built microscope systems may make use of these industry standards."
Last edited by Dubious on Wed Sep 01, 2021 10:19 pm, edited 1 time in total.

apochronaut
Posts: 6272
Joined: Fri May 15, 2015 12:15 am

Re: why additional optics, small sensor

#10 Post by apochronaut » Wed Sep 01, 2021 9:27 pm

Using a term such as quality is a bit loaded. A tube lens or telan lens or telecentric lens are all terms that can be used to denote the lens in an infinity corrected system. It is there in every system to perform at least one function which is to convert the ray bundle from infinity to convergent with a specific desired focal length. It may also be called upon to effect other corrections or compensations depending on how the system is engineered. Since the objectives are microscope objectives not photographic objectives, there is no inherent benefit to having more of the corrections occur in the objective rather than spread out amongst the objective tube lens and or eyepiece. The ultimate goal is to get an excellent image at the eyepiece exit pupil, which is more dependent on the colour correction category of the objectives than any other optical factor, so tube lenses don't have differences in quality ; they produce the quality of image required for the system in total, no more and no less.

Scarodactyl
Posts: 2775
Joined: Sat Mar 03, 2018 9:09 pm

Re: why additional optics, small sensor

#11 Post by Scarodactyl » Wed Sep 01, 2021 10:52 pm

mete wrote:
Wed Sep 01, 2021 7:33 pm
If the tube lens is important in this sense, I mean if it affects the image quality, then even using the same objective, are there difference in image quality between different microscopes (of the same brand), assuming the quality of tube lens might also be different ? If so, is that the reason they say (at least Zeiss, I didnt check others) recommended objectives ? ie apochromats are not in recommended objectives list for basic models (axiolab).
A few systems do have different quality tiers of tube lenses for the same objectives. I've experimented with the sort of 'bog standard' middle-of-the-road olympus tube lens (the u-tlu) which is rated to a 22mm field number, and compared it against the tube lens in the much fancier superwide trinocular head with a 26.5mm field number. You do see a difference but it's really only obvious when you get into the corners of a 26.5mm field and even then it's not an insane difference, especially compared to the difference in price. That makes sense--the U TLU is basically just a doublet from what I can see, but since it's so stopped down by the objective it doesn't have much chance to add its own aberrations to the mix. The superwide head lens and similar ones from Nikon (like the Thorlabs ITL200 which they make as well) are at least triplets, a lot more optical magic goes into them but the increase in quality isn't as much as you'd expect. If you are going to be taking really high resolution photos on larger sensors that have to be perfect corner to corner this sort of thing starts to matter more, but otherwise it's not something you necessarily need to lose sleep over.

apochronaut
Posts: 6272
Joined: Fri May 15, 2015 12:15 am

Re: why additional optics, small sensor

#12 Post by apochronaut » Thu Sep 02, 2021 12:55 am

You cannot however change the tube lens without changing the component that it is engineered to be a part of unless you start to DIY a system. The U.W.F. lenses are part of systems that have design parameters required to embrace requirements of the wider field.

mete
Posts: 114
Joined: Thu Aug 26, 2021 8:31 am
Location: Switzerland

Re: why additional optics, small sensor

#13 Post by mete » Thu Sep 02, 2021 2:14 am

I actually do have Thorlabs ITL200, I didnt use it a lot yet but I have it for low mag photomacrography to use with a focus stacking rail.

If I have an apo objective, dont I need at least an apo tube lens? I understand looking it as a system but since objectives are mentioned with corrections, apo, plan etc. I intend to think they (objective and tube lens) are individually designed things so their performance should match.

Scarodactyl
Posts: 2775
Joined: Sat Mar 03, 2018 9:09 pm

Re: why additional optics, small sensor

#14 Post by Scarodactyl » Thu Sep 02, 2021 4:12 am

mete wrote:
Thu Sep 02, 2021 2:14 am
If I have an apo objective, dont I need at least an apo tube lens?
See my above linked test. The olympus superwide tube lens is apochromatically corrected, while the U-TLU is just a doublet. But when a doublet is stopped way down as a tube lens effectively is in a microscope its own aberrations matter less--not none, as you can see in the far corners, but less. If you can readily spot a difference between the center and even the edge crops with the two lenses I'd be surprised. Ultimately the objective sets the stage, and the rest of the optics matter some but not as much as you'd expect. Even if you're doing high end photography, if the corners aren't super critical it still may not matter. As an example Macro Cosmos on photomacrography and here on Instagram https://www.instagram.com/macro__cosmos/?hl=en shoots on full frame(!!) on his Olympus, and he uses a normal u tlu, not a superwide head (to be fair, he does often use a magnification changer to push the mag up too, but still).

mete
Posts: 114
Joined: Thu Aug 26, 2021 8:31 am
Location: Switzerland

Re: why additional optics, small sensor

#15 Post by mete » Thu Sep 02, 2021 8:41 am

I cant see difference or nothing easily visible at least. A part of my questions are a bit theoretical anyway, I mean they may have a very little impact for my practical use, I am not doing anything critical. However, additional optics and dedicated cameras are more expensive anyway, so what I am asking is also cost effective.

I come across with 2 articles (or 4, one of them is 3 parts), links are at the end. So my understanding is: 1) (I think) before infinity corrected designs, eyepiece was contributing to correct for the residual aberrations. 2) with infinity corrected designs, Nikon took the path of correcting everything in the objective, and Olympus also joined to this view, whereas Zeiss and Leica do most on the objective but not everything. Not everything meaning lateral color is minimized in the objective but a residual is left to be corrected in the tube lens (as far as I understand in order to minimize the complexity of the objective). Also astigmatism is also corrected in the tube lens (not sure to what extent this is done in the objective vs. tube lens).

So, my take from this:

1) for Nikon and Olympus, tube lens should have a zero/very minimal impact for correction. So for your test regarding to different tube lenses, maybe super wide tube is apochromatic only to not introduce more aberrations and maybe it is not possible to make the field of view large without making it apochromatic.

2) for Zeiss and Leica, tube lens is important for correction, so I would not be surprised if it is optimized for a certain set of objectives too. I mean, answering my previous question, a more expensive microscope might be optimized more for apo objectives, i.e. Primostar is targeted for A-Plan, Axiolab up to Neofluar, Axioscope including Apochromat (this information is from the current brochures), i.e. Primostar might have an achromatic tube lens, Axiolab semi-apo, and Axioscope apo. So having an apo objective with Axiolab might be better than Neofluar but maybe cannot use all of its benefits because tube lens do not correct them properly or introduces its own imperfections.

3) Because only tube lens is mentioned in the 3 part article, I still think additional optics for camera connection plays no role in correction. So I still think it is better not to have it if possible. However, article from Leica mentions eyepieces for field correction/astigmatism, so I dont know maybe there is also a difference between the implementations of Zeiss and Leica. If Leica leaves astigmatism to eyepiece, then what happens at camera connection.

resources:

https://www.leica-microsystems.com/scie ... icroscopy/
https://www.degruyter.com/document/doi/ ... -0002/html

apochronaut
Posts: 6272
Joined: Fri May 15, 2015 12:15 am

Re: why additional optics, small sensor

#16 Post by apochronaut » Thu Sep 02, 2021 2:16 pm

The Primostar does not use Zeiss objectives. It is incompatible with the 164.5mm reference length . It is 180mm and most likely has standard Olympus formatted Optics. The Zeisd A Plan would not likely produce a very good image and certainly one that was about 7% low in magnification.

Microscope objective engineering has come a long way. At one time, eyepiece corrections were optimized for a specific objective in a series, often the 40X. In fact, the different magnifications could be optimized by using different eyepieces.

As time moved on , manufacturers got better at bringing all the objectives in a series to the same level of corrections. In the late 60's and into the 70's, Bausch & Lomb produced their semi-objective series each of which , around 30 objectives in all was corrected to use a final common rear lens element situated above the nosepiece. From there, the parallel ray bundle was converged by a very thin tube lens in the head. It's job is 1) convergence 2) A very small amount of field flattening. The condition reaching the eyepieces is slightly overcorrected, since the eyepiece is slightly flattening and compensating.The one common back lens plus tube lens system works equally for all objectives from the earliest lowly 10X .25 flat field achromat used on student scopes to the last prototype 125X 1.4 N.A. flat field apochromat, which has such supreme resolution and is so good and ca free that it could easily be stood against modern planapos and stand tall. Using the same thin telan lens.
There is no need for manufacturers to vary tube lenses according to objective type unless there is a wider field encompassed. The technology exists to bring all objective types to an equivalent level of optimization of the corrections and doesn't really make sense, not to do that.
Scarodactyl. It makes sense that Olympus is using a 3 element tube lens in the U.W.F. head because of the extreme ray angles the farther off axis you go . An achromat would be out of it's element at that diameter but have you considered that there might also be the necessity for a field flattening effect too?

mete
Posts: 114
Joined: Thu Aug 26, 2021 8:31 am
Location: Switzerland

Re: why additional optics, small sensor

#17 Post by mete » Thu Sep 02, 2021 2:49 pm

Thanks for clarifying Primostar, I was confused by its document. I was wondering why a store told me it can only work with one type of objective, now it makes sense.

mete
Posts: 114
Joined: Thu Aug 26, 2021 8:31 am
Location: Switzerland

Re: why additional optics, small sensor

#18 Post by mete » Sat Sep 18, 2021 2:43 pm

I saw a somehow more detailed text about tube lenses (Optical Design of Microscopes by Seward, SPIE Press, 2010, chapter 13). It says typically the tube lens is a doublet pair where first pair corrects some things and other pair corrects others and it is an achromat but 3rd wavelength is also mainly corrected so it can be considered as a semi-apo. On the other hand, for example all tube lenses offered by thorlabs are apochromatic. So I still think it makes sense if there is a connection between the level/price/target of a microscope and its tube lens, ie why to use an expensive/apo tube lens if there is a very small possibility the objective used will be an expensive/apo one.

mete
Posts: 114
Joined: Thu Aug 26, 2021 8:31 am
Location: Switzerland

Re: why additional optics, small sensor

#19 Post by mete » Thu Sep 30, 2021 7:41 pm

I havent seen it officially documented before so I want to share. I saw today that a single page document comes with Axiolab 5, that basically says if you have a proper T2 adapter for the camera, you can connect it without using a lens. So I understand it as no correction needed after the tube lens.

Post Reply