(BH-2) Any solutions for widefield (26.5mm) photography?

[mneium]

Per the title, is there any solution that exists for imaging the full field produced by widefield-compatible objectives? (achromats and apochromats)
It seems that the photo tube of the superwide head produces the same image diameter as the non-superwide head.
I understand that you could remove the head entirely and go for "direct projection", but this seems to be non-ideal because eyepieces must be used for final image correction. Are there any jigs for attaching cameras (potentially smartphones) directly to BH-2 eyepieces?

[apochronaut]

The idea being that you want to image as closely as possible what your eyes are seeing through the eyepieces? That would be determined by whichever projection eyepiece you are using in your phototube and depending on the eyepiece, it may not have the ability to maintain it's corrections outside of a narrow range of eyepiece to sensor distance. So, changing to a lower magnification eyepiece might be the only option.
With the systems I use for photos, probably due to the fact that the photo eyepieces have neutral corrections , I can adjust the coverage of the sensor by changing the eyepiece to sensor distance. I lose parfocality but that isn't of much consequence because I always trim the focus on the sensor anyway, with a focus magnifier.

With certain objectives ; a 60X for instance, the image you would be seeing through 26.5mm eyepieces could be approximated by taking the photo with a 40X objective, acceptable as long as the N.A. of the 40X is high enough. I wouldn't want to push a 40X .65 to represent what I was seeing with a 60X .95 for instance. The quality of the sensor would affect that as well.

[mneium]

...

The NFK photo eyepieces don't come in a super-wide flavor (as far as I am aware) and cover 20mm max. So, I think I have to either do direct projection (bad coz no eyepiece corrections) or focus a (small - smartphone sized, I suppose) sensor on one of the normal SWHK 26.5mm eyepieces.

[Scarodactyl]

Afocal imagining through a superwide eyepiece might work. You'd have to do some hacky modifications to the trinoc port to make it accept a 30mm eyepiece (or whatever diameter?), but once you have the eyepiece in focus in the trinoc port you can image through it with a pancake lens on a dslr (as long as the eyepoint is high enough anyway). For a normal eyepiece you'd use a 40mm pancake for aps-c, but for an ultrawide you'd likely need something shorter. Suspending the camera with lens above the eyepiece will also require some hacky stuff.
The other potential issue with this is that I am not sure the corrections in the viewing eyepiece are the same as the photo eyepiece--this might be wrong, but the way they present the photo eyepieces in the catalogue makes it seem like their corrections might be better.
You'd want to test it first by photographing with the lens through the binoc to see how well it works (And of course you'd also only want to do this if you like messing around with hacky stuff).

[apochronaut]

Eyepieces are made to do the job and usually no better. Otherwise, there is wasted production cost, the microscope has to be more expensive, losing ground to an economical competitor. Achromats are corrected for two wavelengths, apochromats, typically for 3. The human eye has only so much accuity and colour perception, so even with apochromats there is some small ca that is hard to perceive for most people. Not so with a camera film or sensor, where it can be quite evident. Some microscopes have photo eyepieces that are more colour corrected than the visual versions. They may also be ground to closer tolerances; 1/4 𝛌 rather than the more common 1/2 𝛌, giving a sharper image.

[mneium]

Eyepieces are made to do the job and usually no better. Otherwise, there is wasted production cost, the microscope has to be more expensive, losing ground to an economical competitor. Achromats are corrected for two wavelengths, apochromats, typically for 3. The human eye has only so much accuity and colour perception, so even with apochromats there is some small ca that is hard to perceive for most people. Not so with a camera film or sensor, where it can be quite evident. Some microscopes have photo eyepieces that are more colour corrected than the visual versions. They may also be ground to closer tolerances; 1/4 𝛌 rather than the more common 1/2 𝛌, giving a sharper image.

The super wide eyepieces were (are) pretty expensive, and the optics are visibly different from the non-SW ones (the coating is a different color.)



They could be worse then the photo pieces in ways I can't detect, but between definitely losing 50% of the view field and possibly losing some color correction, it seems like keeping the view field is a better trade. I'm going to be taking videos mainly, not photographs.

I am honestly surprised Olympus didn't provide some official means of doing this. It seems like they thought of everything else when designing the BH-2.

Examining the inside of the phototube dovetail assembly, it seems like it would be impossible to bore it out far enough for it to accept the 30mm eyepiece. Maybe the best approach will be to just remove the dovetail assembly entirely and fashion a new spacer using something like PVC.

[apochronaut]

What is the magnification of your photo eyepiece?

[mneium]

What is the magnification of your photo eyepiece?

Don't actually have one atm. BH-2 has 1.67x (rare - $750+,) 2.5x, 3.3x, 5x, 6.7x - http://www.alanwood.net/olympus/photo-eyepieces.html
I don't think even the 1.67x has a 26.5+ field number like the SWHK eyepieces

[apochronaut]

The field # of a photo eyepiece is irrelevant to a large degree. It is the magnification that determines the real field or f.o.v. You cannot increase the field of view captured to your sensor by maintaining a certain magnification and increasing the field # of the eyepiece because the sensor is of a fixed size. The way to capture a wider real field to it, is to view a wider field in the photo tube. The way to do that is to use a photo eyepiece with a lower magnification.



So you are not using a photo eyepiece? You are going direct to sensor?

[mneium]

The way to capture a wider real field to it, is to view a wider field in the photo tube. The way to do that is to use a photo eyepiece with a lower magnification.

All of the NFK eyepieces including the 1.67x are 23mm pieces. Unless I'm misunderstanding, it should be impossible for them to capture 26.5mm FoV. Yeah, I'm looking to use a non-photo eyepiece to project onto the sensor as it seems like the only option that can capture the full field.

[apochronaut]

Your 26.5mm f.o.v. is the apparent field, not the real field. Your eye can expand it's vision to view a wider field. With a 60X objective and a 10X eyepiece with a 26.5 mm f.o.v., the real field, the field you are seeing is about 440 microns. The real field , if the eyepieces were 20mm f.o.v. would be about 335 microns. If you move the magnification up or down by changing the objective the real field changes by a factor directly related to the magnification. If you used a 30X objective the real field would double to 880 microns but the apparent field would stay the same, 26.5mm. The apparent field of view is limited by the field stop in the eyepiece but it is expandable based on the eyepiece, as long as the head and objective image circle are large enough.

The photo sensor cannot expand it's visual field and take in a wider field , like your eyes can. It is fixed.

In the photo tube the apparent field of view is also determined by the field stop, determined not only by the field stop in the eyepiece but also by the inflexibility of the sensor frame. As long as the field stop in the eyepiece projects a larger circle than the sensor frame, the sensor is filled. In order to increase the real field of view and therefore capture what the 26.5mm eyepieces are seeing visually, you have to reduce the magnification in the photo tube and capture the same real visual field ( ex. 440 microns with a 60X objective) on the sensor.

Having a 30mm photo eyepiece will not change the inflexibility of the sensor frame.

[mneium]

Having a 30mm eyepiece will not change the sensor size. The only value of having more than 23.2mm in the photo tube is if the photo eyepiece is incapable of filling the sensor frame, a condition that I have never encountered.

I'm not sure I understand what you're saying. With a 26.5mm photo tube there is more information leaving the microscope. From there, there is some particular sensor size and crop that will fit this information well...

[apochronaut]

Yes. a larger sensor would theoretically capture more information and therefore be likely a better tool for the job but the apparent field size of the visual image would still , not be directly transducible to the sensor. In order to capture the same real field on the sensor, you would have to adjust your photo tube magnification so that the same real field ( 440 microns for instance) was captured on the sensor. You get the ultra widefield imaging of the 26.5mm eyepieces on the sensor. You cannot expand the sensor to see more, like you can with your eyes, you have to reduce the magnification of the photo system.
edit note
I absentmindedly used mm instead of microns in the above post, in my hypothetical image diameters. ..changed to microns.

[mneium]

Yes. a larger sensor would theoretically capture more information and therefore be likely a better tool for the job but the apparent field size of the visual image would still , not be directly transducible to the sensor. In order to capture the same real field on the sensor, you would have to adjust your photo tube magnification so that the same real field ( 440 microns for instance) was captured on the sensor. You get the ultra widefield imaging of the 26.5mm eyepieces on the sensor. You cannot expand the sensor to see more, like you can with your eyes, you have to reduce the magnification of the photo system.
edit note
I absentmindedly used mm instead of microns in the above post, in my hypothetical image diameters. ..changed to microns.

The cameras I'll probably end up using (smartphone cameras) let you play with the sensor a lot. On my ageing Samsung S7, setting the 6.86 x 5.55mm sensor to 1:1 ratio (1440x1440 - plenty adequate resolution even for plan apo objectives at this pixel size, I think) almost perfectly crops it to match the eyepiece's output, for example. I don't necessarily know that I'll be using the S7, but matching the sensor to the eyepiece should not be an issue... just a matter of finding the right sensor and crop.

[apochronaut]

So you see the point. The camera can be adjusted to match the real field . You aren't going to get great quality of imaging with a smartphone camera but apparently you now see that the f.o.v. of the visual eyepieces is irrelevant to the camera set up. You adjust the field capture at the camera. You have to be careful with cropping though. It inherently reduces the resolution. My original input was based on a review of photo eyepiece choices. Cropping in a smartphone is a fair way from that.

[mneium]

You aren't going to get great quality of imaging with a smartphone camera

It should be 1000~3000% more resolution than is needed, according to https://www.youtube.com/watch?v=2JhWaMFBm4I - small-size sensors (or rather, pixels) are apparently way better for microscopy.

you now see that the f.o.v. of the visual eyepieces is irrelevant to the camera set up.

I still don't get what you mean, obviously if I use a 26.5mm SWHK eyepiece I get 2x as much information being emitted from the microscope compared to a 20mm one - then it's just a matter of capturing it with the right sized sensor and crop. This conversation is confusing. Anyway, thanks I think.

[75RR]

Have a look at this link if you haven't already: http://krebsmicro.com/relayDSLR/relayoptics1.html

[Scarodactyl]

No offense Apo but I think you've lost sight of what he's after. The idea is to capture as much of the good FoV as the objective can deliver on this BH2. The nfk photo eyepieces crop those edges, regardless of their stated mag. So (if corrections are to be maintained) afocal is the only way to go. Afocal involves (effectively) a reducing lens going on the camera so you just have to use a shorter focal length to match a bigger FoV in the eyepiece to your sensor size.

Smartphones are also doing afocal imaging, but vs a properly set up good quality camera they'll get poorer results because of cheaper lenses and smaller, cheaper sensors (regardless of stated mp rating). You tend to get distortions at the edges and poorer overall image--certainly not an unusable image, and often superior to an older or poorly installed camera (hence why so many labs use this technique--for basic documentation it's much nicer than tangling with an old microscope camera that needs a pci bus or whatever). I've helped a friemd set up a dslr on their lab microscope (a leitz so it was done afocally) where they had previously used a smartphone, and difference in photo quality was pretty stark.
Small sensor sizes are not better, for the record. They are convenient for some purposes (I believe including active cooling) but mostly they're cheaper.

[mneium]

Have a look at this link if you haven't already: http://krebsmicro.com/relayDSLR/relayoptics1.html

Thanks for the link. It seems to confirm that smaller pixel size (ie. smaller sensor) is indeed better, at least for meeting the resolution requirements of low-power plan apo optics? I could be missing something.
From what I'm seeing, it appears that 35mm cameras actually should be rather bad for microscopy (particularly low-power microscopy with high-NA objectives)

The idea is to capture as much of the good FoV as the objective can deliver on this BH2. The nfk photo eyepieces crop those edges, regardless of their stated mag. So (if corrections are to be maintained) afocal is the only way to go. Afocal involves (effectively) a reducing lens going on the camera so you just have to use a shorter focal length to match a bigger FoV in the eyepiece to your sensor size.

Smartphones are also doing afocal imaging, but vs a properly set up good quality camera they'll get poorer results because of cheaper lenses and smaller, cheaper sensors (regardless of stated mp rating).

You've got what I'm going for, but I think I can perhaps get good quality with a smartphone. Lenses need not necessarily get in the way - I just have to choose the right sized sensor for direct projection from the eyepiece, and there is huge variety with smartphones.



I tore down an old (dead) phone and found that the lens of the sensor can be removed, revealing the bare sensor. Since this phone is broken, and I don't want to butcher my current phone's camera module, I cannot immediately test this - but I will be ordering a few replacement rear camera modules for my working phone (they cost just $3 on aliexpress?!) and testing this as soon as they arrive.

[apochronaut]

No offense Apo but I think you've lost sight of what he's after. The idea is to capture as much of the good FoV as the objective can deliver on this BH2. The nfk photo eyepieces crop those edges, regardless of their stated mag. So (if corrections are to be maintained) afocal is the only way to go. Afocal involves (effectively) a reducing lens going on the camera so you just have to use a shorter focal length to match a bigger FoV in the eyepiece to your sensor size.

Smartphones are also doing afocal imaging, but vs a properly set up good quality camera they'll get poorer results because of cheaper lenses and smaller, cheaper sensors (regardless of stated mp rating). You tend to get distortions at the edges and poorer overall image--certainly not an unusable image, and often superior to an older or poorly installed camera (hence why so many labs use this technique--for basic documentation it's much nicer than tangling with an old microscope camera that needs a pci bus or whatever). I've helped a friemd set up a dslr on their lab microscope (a leitz so it was done afocally) where they had previously used a smartphone, and difference in photo quality was pretty stark.
Small sensor sizes are not better, for the record. They are convenient for some purposes (I believe including active cooling) but mostly they're cheaper.

Probably, because it isn't apparent. Of what value is it to set up afocal with a smartphone in order to capture the periphery of a 26.5mm f.o.v., when that extra 3.25mm of apparent field circle will contain a considerable amount of distortion not to mention that the resolution of the image will be poor. Since there is a photo tube, surely a reduction in the photo tube will at least yield acceptable corrections to the edge even if it doesn't collect the entirety of the intended field, plus the opportunity to capture the original resolution fairly accurately. Is anyone out there making good afocal pictures with a phone? Can you post them, please.

[mneium]

Since I'm going to be using all SWHK eyepieces for this, I am not sure if it actually will be afocal.
They allow focal adjustment inside the eyepiece from +2 to -8 (not sure what units, but that's how it's labeled on the eyepiece.) And I'll be going for parfocality between two of them, so it's seemingly +4 to -16 (mm?) of room to work with

Even if it ends up being afocal, I can't say it would be a huge deal.

Also, using a smartphone offers a lot of advantages. Many of the new ones can do 960FPS slowmo, 4K 60FPS (which isn't a feature even on $2,000+ SLR cameras,) etc. To top it off, for static photography, smartphones natively do a thing that many photomicrographists do: take 10-20 images and stack them for contrast. Google Camera can take up to 48 images and stack them with one button press, all computed within the smartphone in a few seconds.

The signal-to-noise ratio is also favorable, even in high-MP smartphone sensors, because of pixel binning

[Hobbyst46]

Since I'm going to be using all SWHK eyepieces for this, I am not sure if it actually will be afocal.
They allow focal adjustment inside the eyepiece from +2 to -8 (not sure what units, but that's how it's labeled on the eyepiece.) And I'll be going for parfocality between two of them, so it's seemingly +4 to -16 (mm?) of room to work with

Even if it ends up being afocal, I can't say it would be a huge deal.

Also, using a smartphone offers a lot of advantages. Many of the new ones can do 960FPS slowmo, 4K 60FPS (which isn't a feature even on $2,000+ SLR cameras,) etc. To top it off, for static photography, smartphones natively do a thing that many photomicrographists do: take 10-20 images and stack them for contrast. Google Camera can take up to 48 images and stack them with one button press, all computed within the smartphone in a few seconds.

The signal-to-noise ratio is also favorable, even in high-MP smartphone sensors, because of pixel binning

Having tried using a smartphone (Samsung Galaxy S5; afocally!), I felt it is very inconvenient, regardless of the phone camera optics, megabytes, speed etc:
1. Proper alignment of the smartphone with the optical axis of the microscope is a time-consuming trial and error job - even with smartphone adapters;
2. The smartphone camera software is not well adept to continuous photography;
3. Zooming the phone camera each time to catch the FOV without vignetting was annoying;
So, IMHO, although these difficulties are surmountable, a permanent dedicated camera is a better way.
Admittedly, the phone was used on a 18mm FOV.

[mneium]

1. Proper alignment of the smartphone with the optical axis of the microscope is a time-consuming trial and error job - even with smartphone adapters;
2. The smartphone camera software is not well adept to continuous photography;
3. Zooming the phone camera each time to catch the FOV without vignetting was annoying;
So, IMHO, although these difficulties are surmountable, a permanent dedicated camera is a better way.
Admittedly, the phone was used on a 18mm FOV.

1) No doubt, this will be the most difficult part. I've already accepted I'll have to make a spacer with PVC and it will require a lot of adjustment. Once it's made it'll be no issue though.
2) There are lots of different camera software solutions. Not many people know there is a large suite of third-party camera apps for almost every phone (made by users). You can also control an Android smartphone from your desktop using various solutions: https://www.howtogeek.com/430466/how-to ... indows-pc/
3) Definitely would be annoying, but I'm just going to select a smartphone sensor that matches the output of the eyepiece so I don't have to deal with that.

[Scarodactyl]

Of what value is it to set up afocal with a smartphone in order to capture the periphery of a 26.5mm f.o.v.

Ah, I get what you're saying now. Yeah, that in particular is not a good idea.

Since I'm going to be using all SWHK eyepieces for this, I am not sure if it actually will be afocal...Even if it ends up being afocal, I can't say it would be a huge deal.

Just to clarify, that's not what afocal means. Afocal means that you're projecting the image to infinity(~ish) with your viewing eyepiece and then focusing it onto the sensor with another lens.

Also, using a smartphone offers a lot of advantages. Many of the new ones can do 960FPS slowmo, 4K 60FPS (which isn't a feature even on $2,000+ SLR cameras,) etc. To top it off, for static photography, smartphones natively do a thing that many photomicrographists do: take 10-20 images and stack them for contrast. Google Camera can take up to 48 images and stack them with one button press, all computed within the smartphone in a few seconds.

There are lots of cool processing tricks that can compensate for small sensors and small lenses, but I doubt they'd be that well-suited for serious photomicrography.
I'm not sure about stacking for contrast per se--typically stacks in photomacrography are done to compensate for the narrow DoF of high-resolution optics. It sounds like focus bracketing, which is a very handy tool but probably not ideal for imaging through an eyepiece. You generally want your camera's lens to remain focused on infinity and adjust focus by moving the objective closer or further from the subject--adjusting the focus of the lens over the eyepiece is probably going to get you inferior results since you're pushing the objective out of spec that way (effectively using it at an out-of-spec working distance). Objectives can tolerate some variation on that of course (or else you couldn't adjust the oculars for your eyes) but I'm not sure how much. It'd also be a real challenge to get the software to do autofocusing and everything else with a lens other than what's in the phone, so you'd be stuck with phone optics for that.
Similarly, the stacking algorithm is definitely going to be making some simplifying assumptions to get such quick stacks on a phone processor, which are likely not optimized for photomicrography. It'd be interesting to see someone try it though.
I don't do much with video so I can't really comment on slomo or 4k.

The signal-to-noise ratio is also favorable, even in high-MP smartphone sensors, because of pixel binning

Pixel binning is just downsampling less-sensitive small pixels to act like more-sensitive large pixels. It's not something inherent to the sensor at all, it's just a (very basic) algorithm you can do to an image from any camera.

Thanks for the link. It seems to confirm that smaller pixel size (ie. smaller sensor) is indeed better, at least for meeting the resolution requirements of low-power plan apo optics? I could be missing something.
From what I'm seeing, it appears that 35mm cameras actually should be rather bad for microscopy (particularly low-power microscopy with high-NA objectives)

Sensor size/pixel pitch requirements are kind of irrelevant when you're using relay optics. You can capture more pixels of image by making your pixels smaller, or by making the image bigger to cover a larger sensor with the same density.
But there are some inherent issues with smaller sensors and smaller lenses. There's a reason these are being used in phones and not in professional rigs, impressive as their capabilities have gotten over the years.

Don't get me wrong, this would be an extremely interesting project, but I think it would end up being a ton of time, effort and expense for an ultimately not-that-optimal result.

[apochronaut]

With infinity optics you can use a teaching bridge as a photo tube and by pass the telan lens. With an adaptor for the dovetail on the bridge( same as the head) that has a thread on the other side to fit the filter thread of a prime lens , you can use a prime lens set at infinity, mounted vertically on the teaching bridge. A 200 mm will display about a 28mm f.o.v. on an APS-C . A 135 to 150mm lens will put the entire image circle of the objective onto the sensor. The determining factors in image quality are how good the native corrections of the objective are outside of it's intended image field and how flat and well corrected the photo lens is. I use a Nikkor Q, which is a really old cheap lens but fairly renowned for edge to edge sharpness and freedom from coma. I tried a Minolta apo , which wasn't as good. I'm sure there are some cracker lenses out there to use but they may be pricey. A Nikkor Q 200mm is about 50.00

There is probably a way of adapting 160mm Nikon CF objectives to this method and most decent modern infinity corrected objectives should give at least a 24mm field, without distortion. Obviously, objectives that need further correction in either the telan lens or in the eyepieces are not suitable for this method of full field photomicrography.

[mneium]

With infinity optics you can use a teaching bridge as a photo tube and by pass the telan lens. With an adaptor for the dovetail on the bridge( same as the head) that has a thread on the other side to fit the filter thread of a prime lens , you can use a prime lens set at infinity, mounted vertically on the teaching bridge. A 200 mm will display about a 28mm f.o.v. on an APS-C . A 135 to 150mm lens will put the entire image circle of the objective onto the sensor. The determining factors in image quality are how good the native corrections of the objective are outside of it's intended image field and how flat and well corrected the photo lens is. I use a Nikkor Q, which is a really old cheap lens but fairly renowned for edge to edge sharpness and freedom from coma. I tried a Minolta apo , which wasn't as good. I'm sure there are some cracker lenses out there to use but they may be pricey. A Nikkor Q 200mm is about 50.00

There is probably a way of adapting 160mm Nikon CF objectives to this method and most decent modern infinity corrected objectives should give at least a 24mm field, without distortion. Obviously, objectives that need further correction in either the telan lens or in the eyepieces are not suitable for this method of full field photomicrography.

Did not know you can do this. I'd definitely give it a try if not for infinity corrected systems being so darn expensive. What is the entire field of view of modern infinity-corrected plan apos? It didn't even occur to me that it might be higher than 26.5mm (which is the standard for widefield even today, it seems.)

[apochronaut]

I'm not sure if you could adapt the principle to a fixed tube system. It might need some complicated optics and conversions. B & L did that in the Balplan, where they had a series of modified fixed tube objectives converted to infinity in a telescope lens system and then converted back to convergent in a telan lens in the head. Makers have increased the objective diameter too, widening the area of view that requires less correction, so infinity systems are more likely to provide a wider field. The teaching bridge in the system I used is about 16" long, resting on the back of the microscopes arm. Good support for the camera and relatively heavy lens system.

The quality of this sort of system depends entirely on the quality of the camera lens ; how flat and well corrected it is. However, I did this as an experiment and have taken it no farther and although the f.o.v. is very wide, the edge corrections aren't the best. That's probably due to an optic mismatch. Probably, there are some expensive apochromat lenses out there that would trump the Nikkor Q, however for a cheap lens, the Nikkor Q is hard to beat.
As far as cost? The entire system, camera, lens, bridge and microscope cost under 600.00. The Reichert based D.I.N. infinity optical systems have been around for 50 years. They do a 24mm f.o.v. at 10X but I think they obviously aren't limited to that. The planachros are always available at low cost. They are as good as any in their class. Infinity microscopes have been available inexpensively for years, it's just that some people can't leave their brand snobbishness behind.

Here are a couple of pictures I did with the system about 5 years ago. You can see the ca that creeps in as you go off axis. The first is the Minolta apo , the second is a Nikkor Q . The Minolta apo was a fairly low end one, made for a specific compact camera system. The diameter of the lenses in the Minolta are about 2/3 that of the old Nikkor. Both lenses were adapted to work on a Sony a5000, mirrorless. I'm pretty sure that a superior camera lens would give a flatter more well corrected image.
The subject is potato starch in DF. The f.o.v. across the frame edge to edge is about the equivalent of 28mm with the Q. I think I cropped the Minolta lens image a bit because the peripheral distortion was poor. The objective was a 40X .70 planfluor.

Here is a specific review covering a number of lenses in mostly the 180-200mm class for center sharpness....mostly Nikon.
https://www.kenrockwell.com/nikon/200mm ... arison.htm
Similar review for corner performance.
https://www.kenrockwell.com/nikon/200mm-comparison.htm

[c-krebs]

This thread has bounced around a bit, but I think I know what you were originally asking. The BH series S Plans and S Plan Apos were "rated" to provide a 26.5mm image circle ("intermediate image"). The Super Widefield head and viewing eyepieces would "see" that full intermediate image, but the Olympus photo eyepieces were too narrow to accept an image that size. The largest photo eyepiece field number of the NFK series was 21.6mm for the NFK 2.5x. So, in essence, you are immediately cropping into the intermediate image produced by the objective when they are used.

My understanding is that you want to record as much as the intermediate image as possible. Since the Olympus finite LB series objectives use the eyepieces for chromatic correction, I can only think of one approach for this. You would want to use an Olympus Super Widefield corrective viewing eyepiece. How this would be set up depends on what you have, and what camera you will use.

I don't have the widefield head for my BH2's but on my TR-30 trinocular tubes it is a simple matter to remove the plate with the hole at the top for the photoeyepiece. This could be replaced with an appropriate plate to accept the larger diameter SWHK eyepiece. (This assumes that the head is indeed passing the full size intermediate image into the trinocular tube... I don't know if it does). With the larger eyepiece now in place you have two choices. If you have a camera with no lens (i.e. direct access to sensor) you can "raise" the eyepiece in the tube, which causes it to project a real image that can be placed on a sensor.

If you want to use a camera with attached lens you would use the afocal method, placing the lens up close to the eyepiece. Most cell phones these days have a lens focal length of around 4.5mm with a 1/2.5" sensor. The ~4.5mm focal length (with that size sensor) is too short to properly fit the image onto the sensor. You would record a field size of 40mm which is much larger than the intermediate image, so you get a circular image with the black edges. For a 1/2.5" sensor the ideal focal length would be about 6.8mm. With this focal length you would record 26.5mm diagonally on the sensor.