Increase FOV

[Veysel]

Hello,

I am making automated microscope system. I am scanning 15mmx15mm tissue with 40x Objective, 0.35x Adaptor and 1x tube lens, and also my camera is 12MP CMOS area scan camera

- My first Objective is Olympus Plan 40x 0.65NA and FN Value 20. When imaging the tissue, my field of view is 0.84 mm x 0.63 mm on the camera

- My second Objective is 40x UPlanApo 0.95NA and FN Value 26.5. When imaging the tissue, my field of view is 0.7 mm x 0.525 mm on the camera

- My third Objective is Olympus 40x UPLFLN 0.75NA and FN Value 26.5. When imaging the tissue, my field of view is 0.64 mm x 0.48 mm on the camera

So how field of view increase or decrease ? what parameter important for big field of view ? For example Leica GT450 say that, it is field of view is 1mm. How can I see 1mm field of view when using 40x Objective ? As i know, FN value important but Olympus 40x UPLFLN 0.75NA objective's FOV very small, how is it working ?

[PeteM]

You might want to post a picture of your setup. The FOV numbers you list for various Olympus 40x objectives aren't making sense to me.

There are many things, besides the objective field number, that can affect field of view. As an example, various Olympus heads will constrain the achievable field number from 18mm (older BH scopes) to 20mm (BH2 and low end CX) to 22mm (typical BX) to 26.5mm (Superwide head). Try to image with a C-mount adapter (and likely your .35x adapter), and that's another constraint.

There's also the issue of plan field of view. Your least expensive 20mm FN objective could offer a still-wider field of view, but with image sharpness seriously deteriorating at the periphery.

[apochronaut]

The fov on the camera sensor is determined by the setup of the camera system not the objective unless the camera sensor is larger than the objective f.o.v.., then the objective image circle would determine the f.o.v. on the sensor. An objective with a larger image circle should show as such then.
As Pete says, your set up needs to be examined because something is not as it should be. It sounds like there may be vignetting going on somewhere ?

[Veysel]

First of all thank you for your answers.

Adaptor and Tube Lens purchased from Olympus. We use C-mount 0.35x adaptor Here is the setup and images taken with different objectives

Our Camera's sensor info is 1/1.7“(7.40x5.55)

Yes 40x 0.65 objective showing wide area but quality is lower than Olympus 40x 0.75NA.

There is also tube lens information ;

Focal length : 180mm
Projection area: 26.5mm,
Focal Plane Position: 151.3mm from the reference plane

[apochronaut]

So, no vignetting and the frame is completely filled so the only anomaly I see here is that the 40X Olympus UPLFLN objective is magnifying more than the 40X Olympus Plan objective. The non-Olympus UplanApo, might have a requirement for a longer tube length, possibly 200mm, thus it shows as producing a slightly less magnified image than the Olympus UplanApo, which makes perfect sense. Is that a Chinese objective? Can you post a picture of it?
The two images taken with the Olympus 40X .65 Plan objective are consistent, just laterally displaced from each other.

Why the two Olympus 40X objectives are giving different magnifications in the system is a bit of a mystery unless you have moved the photo relay lens from one to the next. Do they also magnify differently through the eyepieces ? Are they parfocal through the eyepieces? Are all objectives infinity corrected?

[Scarodactyl]

I wouldn't expect the olympus objectives to be so far off spec. Even the generic uplanapo seems like it shouldn't be that far off, I am pretty sure they are meant as a drop in replacement for an Olympus. Did you make sure the tube lens is focused at infinity?

[Veysel]

Yes @apochronaut i agree with you 40X Olympus UPLFLN objective is magnifying more than the 40X Olympus Plan objective. Yes 40x UPlanApo 0.95NA objective is purchased from Chinese and it is not Olympus. I also viewed with microscope(Olympus CX23) and I see that 40x 0.65 is not magnify 40x or others is not magnify 40x. As you see from the photos, all objectives are infinity corrected

@ Scarodactyl, "Did you make sure the tube lens is focused at infinity?" Yes it is infinity focused, length can be adjusted

@apochronaut " Are they parfocal through the eyepieces?" If I am not wrong, I checked with these 3 objectives. First I focused with 40x 0.65 olympus and then checked other 2 objective images are tolerable focused (but 3-5 micrometer focus is required)


(Dark background images taken from microscope with mobile phone)

[Veysel]

Also objectives are added

[apochronaut]

I do know that Chinese objectives resembling yours were produced for a 200mm tube. They were called NIS by the factory and the ones I saw were called Planapo, not UplanApo. One would assume that a UplanApo would be the 180mm Olympus format but Olympus might not allow that much leeway in their production sharing agreement with their cooperating factory(ies) in China. They might not be able to do anything about an NIS producer using UplanApo to make a Nikon format objective. They do make them as both 45 and 60mm parfocal in China.
That might account for the mag. difference between the two 26.5mm f.n. objectives but it is only about 5%.

the 40X .65 is a mystery. If it is from a CX23, is it possible that a CX23 uses a different tube length and unique objectives?

[Veysel]

As I understand I should play with tube lens specification right ? Please see the attached photos 40x Chinese Not Olympus objective show biggest area but in microscope and camera system 40x 0.65 Olympus objective show biggest area

Yes 40x 0.65 is from CX23 Microscope. When I look at the CX23 Microscope specification, it says eyepiece has 20 FN value

[apochronaut]

A microscope's optical tube is the determining factor of the f.n. and the factory eyepiece field stop is the ultimate control on that. Objectives designed for the system have an image circle that exceeds that of the eyepiece f.n. and if plan or flat field, that planar or flat field circle at least should project to equal the f.n. of the eyepiece but is often greater. Two objectives with different diameter image circles will each function at an f.n. of 20 if the image circle of each is at least an f.n. of 20 at the eyepiece field stop.

When using a camera frame as the field stop, it functions in the system exactly in the same fashion as the eyepiece field stop if set for parfocality, just in a different shape but there is often one optical change. There can be an extra lens, a photo relay lens or system of lenses but it acts universally on each objective's image. If all of the objectives are the same magnification and the same reference focal length, then the camera frame will receive the same object field diameter from each objective irregardless of the size of the objective's image circle.

In your system, something is causing the individual objectives to magnify differently and the image circle or f.n. of each objective is of no concern. With infinity corrected objectives, it is the reference focal length that usually alters magnification because the infinity space is usually quite short. If you have extended the infinity space considerably beyond the specification that each objective was designed for, that may be what is going on but you say that each objective even magnifies differently in a CX 23?
Can you test them in another Olympus infinity model? Is that 40X .65 plan objective from a CX 23? How did you determine that the objectives do not magnify 40X in a CX 23?

The visual diameter of an objective's lenses either from the front or back does not indicate too much about the field size. It does generally : for instance a high N.A. planapo with a big image circle will likely present with a wider back lens and smaller diameter front lens than a similar magnification planachro with a modest N.A. and image circle but because there are so many possible permutations of lens design in order to achieve similar performance, two objectives with similar specifications worked out by different optical designers can have widely variable diameters of the lens components. I have examples right here of 40X .75 PlanF objectives made in the same factory ahout 3 years apart and they have quite different lens diameters, despite performing similarly.

[Veysel]

Hi, sorry for the late reply. Yes you're right tube lens is important factor. "How did you determine that the objectives do not magnify 40X in a CX 23?" As I understand from the images taken with my phone camera, same objects has different pixel length.

Actually I tested with 20x objectives that field of view is about same. Please see attached photos. (Images was laterally displaced from each other when changing objectives)

Can you recommend specific lens features that I can achieve bigger FOV ? "something is causing the individual objectives to magnify differently and the image circle or f.n. of each objective is of no concern" Actually my setup consist of 0.35x adaptor, 1x tube lens(attached above) and objective and also CMOS camera. I think tube lens affect this situation.

"40X .65 plan objective from a CX 23" Yes it is from CX23. I have one microscope

Thank you for your time

[apochronaut]

You see. You just increased the f.o.v. by using 20X objectives. Lower magnification increases the f.o.v. and the focal length of each objective affects the magnification. From the images using the 20X objectives, it appears that they have the same f.l. , so the 40X examples from the same sources would too. The difference between those higher N.A. and colour corrected 40X objectives is unusual but relativelly small, around 5% or so. That is peculiar but not that great a difference. It is somewhat normal for objective magnification to be rounded off. There are lots of examples of objectives being 5% off of their marked magnification for several reasons.

The difference with the 40X .65 objectives is large. I asked whether the 40X .65 objectives were from a CX 23 because I wonder if there is something different about the CX 23 optics that makes them unique to the CX 23. The CX 23 appears to only be made in China and is all over the Made in China website, sold as an Olympus microscope by many brokers or even factories. It is not uncommon for a microscope company to market an entry level microscope that has incompatabilities with their main production. Both Nikon and Zeiss have a similar arrangement and formerly others, notably Bausch & Lomb. It is economically valuable for a company to cherry pick a design from another economically depressed country for their student or entry level stuff that does not conflict with their sales goals for more upscale designs. Zeiss doesn't exactly want customers upgrading a PrimoStar by putting in a higher end Zeiss objectives, so you can't. They want people to buy a whole other Zeiss microscope.
It is also not unusual for objectives to be marked with their effective magnification, not their actual magnification. There are extreme examples of this. It is possible that the CX 23 system has objectives that have lower magnification than that marked and uses a small multiplication factor in the telan lens, so with another tube lens they don't achieve 40X, resulting in lower magnification and a greater f.o.v. I would look into whether the CX 23 objectives are unique to that model. I don't know for sure but it seems like a good possibility to explain their wider field.

The only way to increase the f.o.v. captured by your sensor is by reducing the magnification, either in the objective or the tube lens. Objectives these days are pretty incremental. There used to be 30 or even 35X made and some optics companies will still custom make just about anything at great cost.
You could also try using objectives from other infinity systems with different focal lengths, also at great cost, or more economically ; change the focal length and or magnification of your tube lens.

[Veysel]

Thank you for your time and your detailed answers.

So I will try with 0.5x tube lens and i will investigate the how 0.5x tube lens affect magnification and field of view

Thank you again

[Scarodactyl]

It'll be interesting to see if you can push it that far. As is if my calculations are right you are pretty close to using the full rated image circle of the better objectives but they can often be pushed a bit further.

[apochronaut]

If you can find the patent # for the Olympus 40X 0.75 objective, you can likely find what it's design f.n. is, which will possibly be larger than it's application f.n. I looked briefly at the 40X ones I knew of from Japan with a 180mm reference focal length and .75 N.A. but none were as high as 26.5, yet others with higher N.A.s were up to 30mm.
I wouldn't count on finding a patent for the Chinese objective unless it is made by Motic.
Sometimes, the f.n. is higher than that marked.

Historically as an example , Arthur Shoemaker designed a great number of AO's 34mm parfocal infinity objectives and almost all have f.n. s of 24 or 25mm, yet A.O. did not exceed a 20mm f.n. with any 34mm parfocal system.

[Veysel]

Hi again,

I have played some configuration within our optic system. I just put tube lens reversed and see that magnification decreased and field of view increased but image was not good as before. (I just wondered that how this will affect the situation)

Field of view increased about 0.1 mm after tube lens reversed

I also tried with 0.5x lens, magnification increased and field of view decreased(please see the images), maybe I am wrong somewhere

I did not play with exposure time , so some frames can be dark or bright

What I am trying to do is that increase field of view without changing the quality(~%5) for example as you said, lens attributes is important factor. I know there is a limit to see bigger field of view, I just want to reach to that limit with high quality and acceptable magnification.

Setup is not well structured for quick test(for now) please don't mind this

[Veysel]

Please see other images

[apochronaut]

You have two limitations. The size of the sensor and the magnification of the objective. Increasing the sensor size will capture more of the field that the objective sees, or decreasing the magnification will project a larger field on the existing sensor. Each will achieve the same goal of capturing a wider field to the sensor.

The magnification can be altered in the tube lens but becareful not to introduce spherical aberration. In some situations, placing a helicoid or variable extension tube governing the tube lens position into the photo relay system in order to trim the framing and focus more precisely. They are available in various extensions and thus compacted depth. Many/most are M42. Search helicoid adapter or variable extension tube on ebay.

[Veysel]

Thank you for your time and help

I will also look at the helicoid adapter or variable extension tube at some later. I will go ahead same setup for now

Have a good day

[Scarodactyl]

https://www.ebay.com/itm/186045045314
This lens has proven to be an excellent shorter focal length tube lens, and this listing is from a trusted community member. You'd need to do some adapting to make it work and the question of whether your objectives can cover that wider area is an open one, but it could give you a wider fov.

[Veysel]

Thank you @Scarodactyl I will also consider your shared product. So just to clarify something, How much bigger area can I see with best optical system at 40x ?

For example https://www.leicabiosystems.com/digital ... io-gt-450/ (please navigate to COMPARE section)in this site there are some Leica scanner devices and it's attributes. Some of the models has 0.5mm f.o.v but GT450 has 1mm f.o.v

To scan a 15mm x 15mm area I have to take about 700 overlapped photo. If I see about 1mm x 0.8mm with my camera, then I need about 300 overlapped photos (like 20X)

[Scarodactyl]

Almost nobody officially specs objectives for ultrawide coverage at high magnifications, though some of them nevertheless deliver a good image outside their specced area (others show serious image degradation even within their spec). Your options would be to try a shorter focal length tube lens and see if it happens to work or get a 20x objective to use instead--a 20x/0.75 like the inexpensive Nikon 20x/0.75 planapo would give a better image than your 40x/0.65.

[apochronaut]

Years ago I played around with using a teaching bridge and s 200mm camera lens as a photo system. The Nikkor Q still has pretty good relative quality, especially in the corners and it is cheap. It outperforms many more modern zoom telephoto lenses, especially in the corners. RAF camera made me up a 2" male dovetail to 52mm threaded adapter to attach the lens filter thread to the bridge but I suppose you could mount it directly on the stand too. You focus at infinity, wide open. With a 16x9 aspect ratio using I think it was an AO/Reichert 40X .70 planfluor, I was getting more than a 28mm f.n. but there was a little chromatic falloff after around 25mm. I never did much more with that system but the potential to collect more of an objective's image circle is distinctly possible based on that little experiment . The question is whether the aberrations outside the corrected f.n. can be dealt with in a telan( tube) lens. In this case, the Nikkor Q lens was acting as both a telan lens and a photo relay lens and it's capacity to provide the needed corrections beyond 25mm wasn't there because it was designed to self correct , not as a corrective adjunct lens . Perhaps another corrective 1X lens in the system could be used. Essentially , I reduced the magnification of the objective dramatically to the sensor, probably to about 25X.
If you can find an objective's patent, you can find the f.n. and image circle usually but the value of collecting some of that extra peripheral area of the circle is questionable in light of the possibility of using a lower magnification, higher N.A. objective. It provides adequate known corrections over the field in question, albeit at possibly 1/2 the magnification. In the case above, I could have used then, and do use now, a 25X .65 planapo in a standard fully corrected trino. photo tube, which covers the same ground as the 40X at a 28mm f.n. and probably provides as good an image. The Reichert 40X planfluor is marked as a planfluor apo in another incarnation so it is no resolution slouch. Colour correction is additive to N.A. in terms of achieving resolution. The best evidence of this are the Lomo 45mm parfocal planapo objectives which have conventional N.A.s. The 40X .65 or 60X .85 versions literally would wipe most equivalent N.A. plan achromat objectives off the slate, so I see no reason why a 20X .75 planapo wouldn't do most of the good work that a 40X .75 planfluor does, just at a wider field and lower mag. There are 3 20-25X apochromats I am very familiar with. Nikon 20X .75 Planapo, Bausch & Lomb 25X .65 Flat Field Apochromat and Reichert Austria 25X .65 Planapo. They are all about equivalent in performance and could sub. in for a 40X planfluor.