Help wanted breaking thro the High Magnification Barrier

[MacroB]

Happy New Year to All!

Thought that the Before & After Focusing might help but it only increased the quantity of slices and made the foreground confusing.

Maxillary Palp 30X at 793 slices and 0.001 mm with alignment turned off.
Have tried less slices and thicker but it did not help.

Camera Canon 50D with Leica X16 lens system, Einstein 640, MT-24EX, poly cup diffuser and Cognisys rail and control box.

Would slabs help?? Have looked at ZS's Batch Dialogue> Current Batch & Stacking tasks but it did not mean much.

When I went to school Technology was having electricity at home. Would be grateful if someone could point me in the right dirrection.

Best Wishes,
Bob












[img][/img]

[rjlittlefield]

Bob,

To upload photos:
1. Make sure that they are no bigger than 1024x1024 pixels (resize as needed).
2. Save as JPEG with a "quality" setting that makes the file length be about 200-250 kilobytes (KB).
3. Follow carefully the directions at http://www.photomacrography.net/forum/viewtopic.php?t=7, especially the ones at Step 5 that say where you have to click to get a link inserted.

I'm looking forward to seeing the image...

--Rik

[Craig Gerard]

Bob,

A recent discussion may provide the information you seek:

[rjlittlefield]

Bob,

I think you're on the right track to try slabs. Your current image appears to have been processed entirely with DMap. That seldom works very well at high magnification, especially with a subject that has as much overlapping structure as this one does. No matter what method you use, you're going to end up with "transparent foreground" artifacts that have to be fixed with manual retouching. Both of those problems will be reduced by processing in slabs using PMax, then combining the results with DMap.

There will also be loss of contrast in some background regions, due to their being partially hidden from the lens behind foreground structures. There's no cure for that. It's an optical effect that's present in the original images.

In the image that's shown, I'm struck that nothing seems very sharp. Since you're shooting with flash, I think we can rule out vibration as a cause. So I'm left wondering about the optics.

What exactly is a "Leica X16 lens system"? Does it have an adjustable aperture, and if so, what do you have the aperture set to?

--Rik

[MacroB]

Hello Rik,

The X-16 lens system is a Leica planapochromatic zoom macroscope which I have mounted horizontally on an optical bench.

This is a PMAX image with the diaphragm set to 4, where 1 is the smallest and 5 the biggest. Probably no comparison to camera f numbers! Just occured to me does a microscope somehow work in reverse to camera regards size of aperture. Should I be using 1?

When I uploaded the jpeg it went darker than original, just checked the memory stick on main computer. Also had a look at first few images of this stack - No 1 the leading setae are OOF with No 2 probably the sharpest of all, though NOT pin sharp when viewed on large monitor.

Certainly when I was focusing and viewing on camera LCD all seemed sharp, however, I thought the disappointing sharpness poss. due to large 27" IIyama screen.

Bob

[ChrisR]

HI Bob - well, no answers so I with little knowledge, will chime in..
you almost certainly want the LARGEST HOLE at this magnification, and I think that's going to be at the "1" end!
No way to eyeball the lens??


Edit a couple of hours later:
I've just been reading of Zeiss macro lenses where an East German one used the lower mumbers for smaller apertures, but West German were the other way around.

[MacroB]

Hello Craig, thanks for the workflow for forming the slabs. I'm sure at some stage that will come in useful - though at the moment like the photo, it's all a blurr.

Hello ChrisR, Yes, I did manage to get the front lens off and check on the aperture. It was as I had always thought - the opposite to a camera lens, with 5 being the biggest.

It was worth a try, sometimes it's just a chance remark that solves the problem.

Have been looking at my software settings to see if the answer might be there. Am using Digital Photo Pro and EOS Utility.

Thanks again,
Bob

[ChrisR]

Do try it at 5 then, diffraction KILLS!

[rjlittlefield]

Bob, I may have misunderstood your words, but I'm cringing a bit at the thought of disassembling a planapochromatic lens. Part of their good correction comes from very fine control of spacing and centering between elements. If anything goes the slightest awry on reassembly, the lens won't work as well as it did when it left the factory.

Next time you need to check which way an aperture goes, just remove the camera body and look through the lens the same way the camera would. That will give you the same information about the aperture, plus let you inspect for stray reflections, all at no risk to the optics.

Now, back to your problem...

I'd like to see a couple more images:
1. Overview snapshot of your whole setup.
2. Actual-pixels crop of the sharpest portion of one source image.

Regarding the aperture, I always suggest to shoot a series of individual frames, stepping aperture from smallest to widest. Look closely at what you get, and choose whatever aperture turns out to be the sharpest. It will probably be at or near wide open, but the exercise will confirm that and will also give you some feel for what diffraction does with smaller apertures.

Once you know the sharpest aperture, then you can determine an appropriate step size -- small enough to catch each detail in good focus while still getting only 20-30% overlap between in-focus portions from step to step. At 30X with an adjustable aperture lens, I suspect that something around 0.005 mm will turn out to be reasonable, maybe even larger would be OK.

An easy way to run the test is to shoot a short stack of say 50 frames at 0.001 mm, then process the stack multiple times using every frame, every 2nd frame, every 3rd frame, and so on until you start losing detail and getting focus banding. In Zerene Stacker, there's a control for "Stack every N'th frame" at Options > Preferences > Preprocessing. You can check that box and adjust the number between runs of Stack > Align & Stack All (PMax). Or you might find it simpler and more reliable to run ZS multiple times, each time loading in just the subset of images you want to process in that run.

[Charles Krebs]

Bob,

I think I have an idea about the primary cause of your problems, but need some more info about how you are set up.

I think you are using the Leica Z-16 (not X16) zoom system, is this correct? Which main objective are you using? ( 0.5X, 0.8X 1X, 2X or 5X? I am not referring to the zoom setting, but the objective itself).

The 1X is, by far, the most commonly used. If this is what you have, then the numerical aperture (NA) of your equipment will range from 0.017 to 0.112.

You said that your image was taken at 30X. Is this actually a 30X magnification on the camera sensor? This is very important information! (How are you determining the magnification?)

If (note the big "if". I am making some suppositions here that you need to confirm or change) your magnification on sensor is 30X, and you are using the 1X main objective you will never get a critically sharp image. Nothing to do with the quality of the equipment, but completely due to the fundamental physics of light. (The Leica Z6 Apo and Z16 Apo are about as good as it gets quality-wise, but they still come under the limitations of physics, and need to be used within certain magnification parameters for best results).

To obtain good resolution at a magnification of 30X (on APS-C sized sensor) you need to have a NA of about 0.50 (somewhere in the 0.40 to 0.60 range). With a maximum NA of 0.112, the on sensor magnification should be about 5X . Anything above 6X (on sensor) will suffer quite severely from resolution loss due to diffraction.

Edited to add a few additional thoughts:

[MacroB]

Hello Rik,
Not quite as bad as it sounds. I undid the front lens which is 71/2" long and since it came in sections that I had to put together, the lens was one such piece.

What do you think about turning off all the alignment in ZS's Options?
I find it cuts out the "Bar code" at the edges and stops the dust spots developing into curly hairs. Should I perhaps leave the X & Y shift checked?

Bob

Hello Charles,

Forgive my senior moment - it is, as you stated, the Z-16! I have the
Z-16 2.0x objective. Have not seen the numerical aperture mentioned anywhere, in fact, Leica do not give out much info. I find it best to write direct to Germany.

My measurement of Magnification was not exactly scientific since I used a macrometer and photographed the jaw space, when of course I had to include a tiny overlap.

FOV as measured 0.7 mm, size of image sensor 22.3 mm ÷ 0.7 = 31.857.

So, what is going to be the optimum aperture and what is the max pin sharp image size that I can hope to achieve? I remember in one of the adverts Leica mention adding a microscope objective to the front lens of the Z-16 to achieve even greater magnification.

If I reduced the magnification and added such an objective with its smaller size glass and presumably aperture, would that give me sharpness at 30x ?

Bob

[Charles Krebs]

[rjlittlefield]

[MacroB]

Hello Rik and Charles,

Thank you both very much for all the info and your time, which is much appreciated.

Have started experimenting with the "stack selected frames" in ZS - something which I had previously not used. Will also try varying the alignment.

Before I bought the Einstein I had read about the blast moving the subject, so I did check when I first got it and all seemed ok. Must admit though that I was looking for movement on a larger scale, plus using a polystyrene cup as diffuser I thought all would be well.

I just had not thought of the blast from the studio flash moving the whole set-up or showing up as a more subtle movement! I am going to try the 30x shot using the MT-24EX and the 430EX II flash guns.

If that gives me a sharper image all well and good, if not I will see what Leica can suggest.

Thank you all once again and all the very best wishes for the New Year.
Bob

[Charles Krebs]

Bob,

I suggest doing the comparison using Einstein 640 flash, and them with the exact same set-up, using only the modeling lights (live-view. silent mode) for a comparison. That way you really have a direct comparison with same lighting angles. The flash would not cause vibration the entire set-up, but possibly with a more "gossamer" subject such as a butterfly wing (or perhaps the cup you are using for diffusion). In your case I really doubt this is the problem, but it is certainly worth checking, and it is easy to do so.
_________________
http://www.krebsmicro.com