Mitutoyo 50x Sharpness Problems?

[rjlittlefield]

Rik - the "diffraction cutoff" that you mention - does that mean that anything beyond f/16 will appear equally soft?

There is some symmetry in this situation: I don't know what you mean by "anything beyond f/16 will appear equally soft". Communication is tough!

But the answer to your question is surely "no, that's not what I mean".

Let me try again in more detail.

When speaking of a lens, "diffraction cutoff" refers to the finest level of detail that can be resolved at all. On an MTF chart, it is the level where contrast drops to zero. For an f/16 lens, that number computes to be about 114 line pairs per mm on sensor.

When speaking of a sensor, the finest level of detail that the sensor can possibly be trusted to capture is one line pair for every two pixels. This is the "Nyquist limit", and for your sensor it turns out to be 119 line pairs per mm.

119 is about equal to 114, hence my comment that your sensor has resolution corresponding to diffraction cutoff at f/16.

Lenses that are significantly wider than effective f/16 will produce images with higher contrast for fine detail, but the sensor will not capture more than 119 line pairs per mm regardless of how sharp the lens is.

Lenses that are significantly narrower than effective f/16 will produce images that are more blurred in proportion to the f-number. Your 50x/0.42, running at effective f/60, will produce an image that is very blurred on sensor, and your sensor will capture all the detail that is present in it. Your 20x/.40, running at effective f/25, will produce an image that is much less blurred, but is still blurred enough that your sensor will capture all of that image detail also.

Is the solution to get a camera with a larger sensor or more MP?

No. The sensor that you have is fine.

More MP on the same size sensor would simply prepare the sensor to capture finer detail that is not in the optical image in the first place, because of diffraction.

More MP on a larger sensor might help, IF the objective could cover the larger sensor, which it probably can't. That limitation is because microscope objectives are designed to cover a sensor no larger than APS-C at rated magnification. Some of them do actually cover more, but that's an area I don't want to get into right now. The simple and reliable approach is to just pair a microscope objective with an APS-C sensor, like you're doing.

You also mention that you use a finite lens and I noticed that it asks for a 210mm tube lens.

No, it asks for a 210mm tube, no lens. Finite objectives are designed to be used on empty tubes, no tube lens.

And large-NA objectives are very picky about the length of that empty tube. The standard graph of tolerance for tube length shows that an NA 0.80 dry objective can only tolerate about 2 mm deviation from nominal. Sticking this objective on your Tamron would give an atrociously blurred image, if you could get it to focus at all.

--Rik

[Chris S.]

If I were to get a 50x/.55 instead, would there be a noticeable difference in sharpers from my current 50x/.42?

Some time ago, I posted an image very similar to your 50x/0.42 SL image, but taken with a Mitutoyo 50x/0.55 objective. You and I may even have used the same species of subject? Mine was the sunset moth (Urania rhipheus, aka Chrysiridia rhipheus). See the fifth image down in Mitutoyo 50x with chips in glass, tested.

Comparing the two images, I think the one from the 50/0.55 objective looks quite a bit better than the one from your 50/0.42. This is expected, but you may find the amount of difference a useful guidepost. (For whatever it’s worth, the image I posted came from the lesser of two 50x/0.55 lenses, the one with chips in its front face.)

I find the 50x/0.55 Mitutoyo a very useful lens, with its 13mm of working distance. Yes, higher NA objectives will deliver more resolution, but at the cost of working distance.

You mentioned the Mitutoyo HR 50x/0.75. I tested one, and found it exquisite with flat subjects. But when I tried to shoot a bug with it—specifically, to look into the pits on the surface of that bug, and see what was at the bottom of these pits—I could not do it. This objective’s working distance of 5.2mm—combined with its broad, flat, front—forced me to use shallow, glancing light. Light from this small set of angles left the bottoms of the pits in complete shadow. Almost as bad, this light accentuated the surface texture of the bug way too much (it looked bad and didn’t convey reality). So even if one can illuminate around an objective with short working distance, the effect of this light may not be desireable for a given subject.

Another member of this forum was able to photograph these pits using an objective with a more narrow, pointy end. His image also more attractively and accurately conveyed the surface texture of the bug. IIRC, his objective had less working distance, but the narrowness of the objective’s end gave him more angles to light from. I’ve come to think that for photomacrography, working distance should not be expressed only as a number of millimeters, but with a second number that conveys the maximum angle available to illuminate the subject.

--Chris S.

[Zack]

Rik - All your answers clear things up significantly for me, so thank you! Communication is certainly tough when I don't know what I am talking about It definitely sounds like my options are to stick with the APS-C sensor, stay with infinity-corrected objectives, and maybe just be content with the 20x and crop if needed. I am going to try another lens with much shorter WD, but I am not confident that I will be able to sneak sufficient light in there for photography.

Chris - Yes, I am also photographing the Madagascar Sunset Moth, and the image you linked is significantly sharper than mine (and probably even better when not viewed in 1024 pixels online!), so maybe that is convincing evidence in favor of the Mitutoyo 50x/.55. I also really appreciate your comments on the Mitutoyo HR 50x/.75, which I would predominantly use for butterfly scales that are essentially flat (maybe 50 images stacked front to back at 2 microns), so perhaps the textural lighting you mention would not be as much of an issue for me there? That being said, I believe the HR is quite a bit more expensive, and I really have not been able to even find one for sale from a reputable dealer, despite scouring the internet.

[Chris S.]

. . . the Mitutoyo HR 50x/.75, which I would predominantly use for butterfly scales that are essentially flat (maybe 50 images stacked front to back at 2 microns), so perhaps the textural lighting you mention would not be as much of an issue for me there? That being said, I believe the HR is quite a bit more expensive, and I really have not been able to even find one for sale from a reputable dealer, despite scouring the internet.

The Mitutoyo HR 50/0.75 would be outstanding for butterfly scales. Sadly, the objective almost never appears on the used market--I've been watching. New, the cost is almost $6,000 USD. If this works for you, Edmund Optics lists it as "in stock" and is very reputable.

--Chris S.

[Scarodactyl]

Yeah, it's just this one https://www.ebay.com/itm/Mitutoyo-BD-Pl ... 3214247454
On the one hand, too risky. On the other, if you get the shell off it might be marginally better than the brightfield version.

[Zack]

Thanks for the link, Scarodactyl. In reply to both you and Chris, while those lenses may be a dream, the prices are certainly prohibitive. On a school teacher's salary, I'm not sure I can even afford to look at those kind of prices

I did order a returnable Olympus 50x/.8 from Ebay that I will try next week and post the results of, though like I said, I am not confident I can sneak any light into the 1mm WD that it offers...

[Scarodactyl]

My olympus 50x/0.8 is one of my favorite objectives. I think you may enjoy it.

[rjlittlefield]

I did order a returnable Olympus 50x/.8 from Ebay

At eBay, I see a number of different "Olympus 50x/.8". Can you point to the one that you got?

--Rik

[Lou Jost]

I am not confident I can sneak any light into the 1mm WD that it offers...

For some subjects, epi lighting solves that problem completely. It is not attractive on all subjects, but on some subjects it works well (eg electronics wafers).

[rjlittlefield]

For some subjects, epi lighting solves that problem completely.

For the benefit of Zack...

I think that Lou is talking about illumination through the lens. In microscopy that would specifically be called "episcopic brightfield" because a focused planar mirror looks bright, as opposed to "episcopic darkfield" which refers to illumination from the lens side of the subject but around the outside of the lens so that a focused planar mirror looks dark. The unqualified term "epi" can refer to either. In fact one of my favorite objectives is labeled by Nikon as "CF Plan 50X NA 0.55 inf/0 EPI ELWD" because it is designed for use with opaque subjects, hence illuminated from the lens side with no cover slip, even though it has no special provision for illumination through the lens and is normally used with illumination around the lens.

Personally I prefer to use "through the lens" for clarity. The term "epi" can be helpful for brevity, but it works well only within a community that has agreed to understand that shorthand in a certain restricted way.

--Rik

[Zack]

Scarodactyl - Excellent to hear! I hope I am as enamored as you are with it

Lou - Being as inexperienced as I am, I am not sure exactly what episode lighting is... although I only photograph insects, I am still open to learning any lighting techniques that would improve my photography

Rik - I ordered the Olympus UMPlanFl 50x/.80 BD. I understand that it is not apochromatic and I will likely have to do a significant amount of touching up in post-processing (or just live with the color chromatic aberrations). I also believe that there is something else needed to accommodate for a BD objective, like blocking the outer channel from light with black felt or similar...? I think my excitement to do microscopy is overpowering my lack of knowledge on these subjects.

Also, Rik, thank you for the clarification on the "epi" term. I just did a little research of my own and most of what I found for "episcopic brightfield" is related to reflected light microscopy. Currently, I have two flash units with attached diffusers, and I simply point them in the direction of the subject and fire away. Is there a better way to achieve similar lighting? When I think of "reflecting" lighting, I think of having a flash pointed away from the subject instead and bouncing that light onto the subject, thereby further softening/diffusing it. But maybe I am also confusing terms here. As it now stands, I have to use my two flashes at around 1/4 power each to illuminate my subject. That's a fair amount of flash power and burns through batteries, but if that is the solution, then that is what I will continue to do. I know that an alternative is increasing my ISO, but I believe (correct me if I am wrong) that ISO noise will stack in Zerene and so even a little bump in ISO - to 200 or 320, maybe - will result in a considerable amount of noise in the finished stack?

[WojTek]

Hi Zack,
I like Oly 50x/0.80 too
e.g.
viewtopic.php?f=27&t=44106&p=277058&hil ... us#p277058
Best, ADi

[Zack]

Adi - your photo of the Swallowtail is stunning - that's exactly the sort of subject I would like to photograph, though I could never hope to get results as good as yours! Would you mind telling me more about your lighting and how you achieved the shot? If you have a photo of your setup, I would love that too to help me visualize everything. I believe I still have a long way to go before conquering my lighting/short WD battle.

[rjlittlefield]

Rik - I ordered the Olympus UMPlanFl 50x/.80 BD. I understand that it is not apochromatic and I will likely have to do a significant amount of touching up in post-processing (or just live with the color chromatic aberrations). I also believe that there is something else needed to accommodate for a BD objective, like blocking the outer channel from light with black felt or similar...?

Thanks for the added info. You will definitely have to block the outer channel, or your images will suffer from a horrible amount of veiling glare. You will also have to remove the outer barrel, or the angles available for shining light on the subject from around the lens will be very limited.

I just did a little research of my own and most of what I found for "episcopic brightfield" is related to reflected light microscopy.

In microscopy, most applications use transmitted illumination, where light comes through a transparent subject. It is like shooting glass subjects against a bright background. For opaque subjects obviously that won't work, so light has to be reflected off the "front" or the "top" of the subject, whatever surface is facing the lens.

Within the microscopy community, the technical terms are "diascopic" for light that is transmitted through the subject, versus "episcopic" for light that is reflected off the subject.

Outside the microscopy community, I prefer to use "transmitted through the subject" or "reflected off the subject", because surely "diascopic" and "episcopic" will cause more confusion than insight.

Anyway, "episcopic" systems are further broken down as "brightfield" or "darkfield", depending on what a focused planar mirror would look like. When illumination is provided through the lens, the mirror will reflect it back into the lens, so this is "episcopic brightfield". When illumination is provided around the lens, the mirror will reflect it back around the lens, so this is "episcopic darkfield".

Some butterfly scales are strongly mirrorlike and have dramatically different appearances depending on whether illumination is provided through the lens versus around the lens. At viewtopic.php?f=27&t=42795&p=269228 , those things that look like black "holes" between blue ridges are not holes at all! Instead, they are shiny solid membranes that are reflecting the black interior of the camera where no light is coming from. If illumination were provided through the lens, then those very same scales would have bright spots where the black is now. At the end of that thread, there is a link to a pdf HERE that shows the true structure of the scales. Figure 6 shows what they look like when light comes through the lens.

Currently, I have two flash units with attached diffusers, and I simply point them in the direction of the subject and fire away. Is there a better way to achieve similar lighting? When I think of "reflecting" lighting, I think of having a flash pointed away from the subject instead and bouncing that light onto the subject, thereby further softening/diffusing it. But maybe I am also confusing terms here.

The final image does not depend on whether light reaches the subject by being bounced off some other surface or transmitted through some sort of diffuser. However, it very much does depend on the range of angles from which the light arrives. In photography at larger scale, light from a wide range of angles is often provided by bouncing it off the environment, which is the sort of "reflected" that you're talking about. For small subjects, light from a wide range of angles is usually provided by sending the illumination through some sort of diffuser. Commonly the diffuser wraps around the subject so that light comes to the subject from a very wide range of angles, much wider than can be achieved by putting a diffuser on the flash head. At high magnification (large NA), that wide range of angles is needed to avoid an assortment of artifacts that can range from reduced resolution to false colors to subject features "squirming around" laterally as focus is changed.

I know that an alternative is increasing my ISO, but I believe (correct me if I am wrong) that ISO noise will stack in Zerene and so even a little bump in ISO - to 200 or 320, maybe - will result in a considerable amount of noise in the finished stack?

Noise does tend to accumulate in the stacking process, so lots of light and base ISO will give the cleanest result.

--Rik

[Zack]

Thanks for all the answers, Rik. Your point about the tiny dimples in the Ulysses wings is one of the reasons why I got into macro in the first place - those infinitesimal wonders of the world that we never notice. Incredible. I think I have a much better understanding of the types of lighting and light transmission as well, although I will continue to read and learn. In terms of the BD objective, do you have any specific advice on how to safely remove the outer barrel? I obviously do not want to damage the objective through negligence.