Sigma 180 2.8 (Tube Lens) and Mitutoyo 10x - bad image quality

[Patison]

Hello!
I am facing a problem when using a Sigma 180mm lens as a tube lens for a Mitutoyo 10x microscope objective, there are strange artifacts appearing in the resulting images, leading to poor image quality.
In my opinion, the Sigma 180 2.8 has a high image resolution and there should be no problems using it as a tube lens?

The only thing I find that could be a problem is I'm not using it set to infinity.
Because i am stacking using the AF motor.

Examples:
here is 100% crop - mitutoyo 10x


50% crop - same picture mitutoyo 10x


100% crop with sharp light




Any idea ?

[lothman]

Any idea ?

may be it is the Mitutoyo lens, because the other corners look OK. Just rotate the Mitutoyo and see whether this blurry area also rotates.

[rjlittlefield]

The first image reminds me a lot of a problem that I had with a defective Mitutoyo 10X. See viewtopic.php?t=20594 for discussion.

The cross-shaped artifacts typically result from astigmatism combined with focus stacking. Astigmatism means that the lens does not focus tangentially and radially in the same depth plane. In individual source frames, a sharp point feature on the subject ends up appearing as either a streak pointing toward the center of the image, or a streak at right angles to that, or a fuzzy blob, depending on focus. The feature is never sharply focused in all directions at the same time. The focus stacking process then combines the two streaks to produce a cross-shaped pattern.

This sort of problem occurs fairly often in used Mitutoyo objectives, probably as a result of being dropped or knocked over at some time. This causes one or more lens elements to shift sideways by a few microns inside the lens, often leaving no mark on the outside.

The randomly oriented streaky artifacts with sharp light often occur even with good lenses. They result from bright specular reflections coming from surfaces that are sloped at different angles. In that same image, on the right side, I see a lot of cross-sharped artifacts that again make me suspect problems with the Mitutoyo. But for diagnosis, you need to use highly diffused light to avoid misleading artifacts.

--Rik

[Patison]

The first image reminds me a lot of a problem that I had with a defective Mitutoyo 10X. See viewtopic.php?t=20594 for discussion.

The cross-shaped artifacts typically result from astigmatism combined with focus stacking. Astigmatism means that the lens does not focus tangentially and radially in the same depth plane. In individual source frames, a sharp point feature on the subject ends up appearing as either a streak pointing toward the center of the image, or a streak at right angles to that, or a fuzzy blob, depending on focus. The feature is never sharply focused in all directions at the same time. The focus stacking process then combines the two streaks to produce a cross-shaped pattern.

This sort of problem occurs fairly often in used Mitutoyo objectives, probably as a result of being dropped or knocked over at some time. This causes one or more lens elements to shift sideways by a few microns inside the lens, often leaving no mark on the outside.

The randomly oriented streaky artifacts with sharp light often occur even with good lenses. They result from bright specular reflections coming from surfaces that are sloped at different angles. In that same image, on the right side, I see a lot of cross-sharped artifacts that again make me suspect problems with the Mitutoyo. But for diagnosis, you need to use highly diffused light to avoid misleading artifacts.

--Rik

here is today test with mitutoyo 10x

full resolution stacked image : soft light
mitu x10 by Radoslaw, on Flickr

[Adalbert]

that looks good :-)

[Patison]

i play a little more

here is two photos:
one is on the closest focusing range (macro)

secound is on the middle focusing range i think difference is a lot...
https://gfycat.com/dopeynaturaldalmatian

[rjlittlefield]

For testing purposes, I recommend using a flat subject that fills the frame.

A moth or butterfly wing works OK, but has the problem that the scales do actually change from place to place, so for example there can be problems comparing one corner with another.

Laser scan copies glued to flat glass can be excellent, for example as I used for testing at viewtopic.php?t=23898 .

--Rik

[rjlittlefield]

i play a little more

here is two photos:
one is on the closest focusing range (macro)

secound is on the middle focusing range i think difference is a lot...
https://gfycat.com/dopeynaturaldalmatian

The animation was flipping several times per second so I could not see anything at first. Then I figured out how to pause the animation and use the arrow buttons to switch images.

I agree, big differences in image quality.

It is possible that things go bad because of focusing the rear lens away from infinity. This would be the worst I have ever seen, by far.

If the image quality is good at infinity focus, then the Mitutoyo is OK.

--Rik

[Patison]

For testing purposes, I recommend using a flat subject that fills the frame.

A moth or butterfly wing works OK, but has the problem that the scales do actually change from place to place, so for example there can be problems comparing one corner with another.

Laser scan copies glued to flat glass can be excellent, for example as I used for testing at viewtopic.php?t=23898 .

--Rik

i will test.. but in the mean time please check gifs

https://gfycat.com/medicalrigidamericanindianhorse

[Scarodactyl]

You probably should not use a 10x objective at anything but infinity focus. That kind of tomfoolery can be OK with a 5x where everything's a bit looser but once you're firmly in the micro range there's less room to mess around.

[rjlittlefield]

I have used AF motor focusing with 10X and even a 50X objective, as documented at viewtopic.php?t=14569 .

I have not seen image degradation nearly as bad as what Patison is showing, though there are obvious limitations especially at high magnification.

But if his combo looks good at long focus on the rear lens, as it seems to, then I don't see anything to blame other than close focus.

For me the takeaway will be that different rear lenses have wide variation in behavior at closer-than-infinity focus with objectives. That's not terribly surprising, since one of the side effects of using an objective is to place the aperture stop at a position that will be far different from what the rear lens is designed for.

--Rik

[Patison]

from the middle range to infinity stack (300 pictures)

[url=https://flic.kr/p/2ojkcNY]

[FotoChris]

Hello!

I know this is a slightly older thread but I might have an explanation as to what is happening here;

First of all I think both your Sigma lens and Mitutoyo objective are fine.

However like any lens with internal focussing, the Sigma 180mm 2.8 shortens it focal length when you set the focus to a close distance. In fact most macro lenses will have a much shorter focal length at 1:1 compared to infinity, in case of the 180mm Sigma it's close to 110mm at 1:1.
Since the 10x Mitutoyo requires a 200mm tube lens to achieve 10x magnification (with its focal length of 20mm) this will result in a magnification of around 5,5x - which may still deliver a sharp image in the center of the photo but the image quality will noticeably degrade towards the edges and corners resulting in many optical artefacts.

And because you use the autofocus to focus stack, every time the focus is changed it will also change the focal length a bit - as well as the position of the internal lens elements. That will result in a change of magnification between images, even if it's hardly noticeably between two images it will add up over the whole stack. If you were to shoot a stack from infinity to the minimum focus distance that'll range from 8,75x - 5,5x. Your stacking software will probably be able to compensate for some of these changes between images but the further away from the center you go the more noticeably these distortions will be because that's where the software won't be able to account for changes in field curvature, coma, astigmatism,...

Photonstophotos has a decent database concerning lenses and how their focal lengths change during focussing, here's the Sigma 180mm 2.8:
https://www.photonstophotos.net//Genera ... xis,FocalI
When you click on "Focus" you can change the focussing and see how the elements and focal length change - and you'll see a lot more useful information like changes in aperture etc.

[Lou Jost]

That's a great explanation! I use this technique a lot, and it never occurred to me that the objective's good image circle must shrink as the lens focuses closer. Thanks for that insight.