MicrobeHunter.com Microscopy Forum

I have to admit I'm out of practice with photomicrography, but I was trying out a 40x/0.85 inf pan fluor lens on the Amscope T720 using a diatom strew. It's set up to use my Canon 1100d (Rebel T3) for imaging. Everything is ok except that my images seem to be coming out with a violet/purple colour cast. The T720 uses LED lighting and I thought it might be to do with that. Neither 'Auto' or the other built-in white balance options gave good results. So I thought I'd try a custom colour balance which I did via Liveview. But the output still came out with a violet tinge. It must be something obvious! But not sure what. I tried with a 10x pan fluor but about the same. Any ideas, suggestions?
Here is a crop from the 40x:
Or is it just my eyes??

Thanks for any input

Louise

I don't see a violet tinge, just typical axial CA. Could it be that your monitor has gotten off calibration?

What projection method do you use? Afocal or direct projection?

Scarodactyl wrote: ↑

Sat Sep 17, 2022 4:47 pm

I don't see a violet tinge, just typical axial CA. Could it be that your monitor has gotten off calibration?

Hi

Um what is 'axial CA' ? It's actually not an all-over tinge - just in certain places like the flat area in the centre of the central diatom.
It seems to be independent of the monitor. Here is a crop of the crop with increased saturation enhancing violet and green
Louise

Wes wrote: ↑

Sat Sep 17, 2022 4:52 pm

What projection method do you use? Afocal or direct projection?

It's an infinity microscope so direct projection after the tube lens. I'm not sure whether there are any additional optics after the tube lens. The view through the oculars doesn't show any tinge.

Louise

Thats interesting, I thought direct projection largely eliminates CA, not sure about axial CA tho. You can eliminate the magenta and greens with photo editing software but I suspect you want to eliminate it at the source.

Wes wrote: ↑

Sat Sep 17, 2022 5:24 pm

Thats interesting, I thought direct projection largely eliminates CA, not sure about axial CA tho. You can eliminate the magenta and greens with photo editing software but I suspect you want to eliminate it at the source.

Yeah, I wasn't expecting CA with a hq plan fluor but the fact that I don't see any via the oculars made me think it was the camera. Alternatively, maybe the Amscope tube lens isn't so good? Hmm... Or is there an extra lens in the trinocular optical pathway? I actually removed the original trinocular tube because it wasn't any good for connecting a dslr. Maybe I'll try and connect the camera via an ocular port to see if I can get a better image.

Louise

LouiseScot wrote: ↑

Sat Sep 17, 2022 4:56 pm

Hi

Um what is 'axial CA' ?

It's what you're seeing. There are two kinds of chromatic aberration, color separation horizontally (lateral ca) and color separation axially (ie towards or away from your eye). It's typically this purple and green color vs the more classic red blue lateral ca.
Fluorites are often better corrected for lateral than axial ca, and they can vary a lot in quality. It's also possible that the eyepieces and/or tube lens on your scope are compensating to some degree. That would take experimentation to figure out.

Wes wrote: ↑

Sat Sep 17, 2022 5:24 pm

Thats interesting, I thought direct projection largely eliminates CA, not sure about axial CA tho. You can eliminate the magenta and greens with photo editing software but I suspect you want to eliminate it at the source.

On the systems it is appropriate on direct projection mitigates the possibility of extra added CA from relay optics but it doesn't eliminate CA already present in the objective's image. Axial CA is notoriously harder to remove witb software than lateral though that eill vary with the subject.

Scarodactyl wrote: ↑

Sat Sep 17, 2022 6:27 pm

LouiseScot wrote: ↑

Sat Sep 17, 2022 4:56 pm

Scarodactyl wrote: ↑

Sat Sep 17, 2022 4:47 pm

I don't see a violet tinge, just typical axial CA. Could it be that your monitor has gotten off calibration?

Hi

Um what is 'axial CA' ?

It's what you're seeing. There are two kinds of chromatic aberration, color separation horizontally (lateral ca) and color separation axially (ie towards or away from your eye). It's typically this purple and green color vs the more classic red blue lateral ca.
Fluorites are often better corrected for lateral than axial ca.

Ok but I don't see it when observing via the oculars so it must be something in the camera pathway? I'll do some tests/experiments tomorrow to try and narrow it down.
Thanks
Louise

I believe it is the LED light.
Occasionally I observed my diatom slides under halogen light, and there were no purple fringes.
Under my current LED, there are fringes, especially under oblique light. I will try and post an example.

But, diatoms are not the best demonstrators of CA. The diatom frustule does all funny things with light - scattering, refraction, dispersion etc. When in doubt, I would suggest to inspect the simplest target - a stage micrometer. There, CA - if present - is obvious.

Here is an example.
The same diatom was imaged with a plain (non phase contrast) 40X0.75 Neofluar objective.
Afocal photography, my previous 6500K LED light, filtered through a KR12 amber filter (my present light is 4000K, high CR LED).
One image is brightfield, the other oblique.
Both images were considerably cropped. Also done - slight tweaking of the brightness and contrast, simultaneously on both photos.

Hobbyst46 wrote: ↑

Sat Sep 17, 2022 7:50 pm

I believe it is the LED light.
Occasionally I observed my diatom slides under halogen light, and there were no purple fringes.
Under my current LED, there are fringes, especially under oblique light. I will try and post an example.

But, diatoms are not the best demonstrators of CA. The diatom frustule does all funny things with light - scattering, refraction, dispersion etc. When in doubt, I would suggest to inspect the simplest target - a stage micrometer. There, CA - if present - is obvious.

Hiya
The LED illumination certainly could be a factor. But I still wonder why it only shows up on camera images and not when viewing via the oculars. The T720 has Kohler so the illumination should be even. Having said that, I'll have to checkout my trinocular adapter to make sure the camera sensor is aligned. I'll also have a look via the trinocular with an eyepiece.
Thanks
Louise
ps hope you are well

LouiseScot wrote: ↑

Sat Sep 17, 2022 9:15 pm

Hiya
The LED illumination certainly could be a factor. But I still ....

My camera too registers more CA than I do. Try fiddling with white balance a bit.

Hobbyst46 wrote: ↑

Sat Sep 17, 2022 8:12 pm

Here is an example.
The same diatom was imaged with a plain (non phase contrast) 40X0.75 Neofluar objective.
Afocal photography, my previous 6500K LED light, filtered through a KR12 amber filter (my present light is 4000K, high CR LED).
One image is brightfield, the other oblique.
Both images were considerably cropped. Also done - slight tweaking of the brightness and contrast, simultaneously on both photos.

Yep, that's pretty much the same as I'm getting on my images. I know what you mean about diatom-induced aberrations! I wasn't looking for it - I was just checking out the 40x/0.85 plan fluor I got from China (at relatively great expense!). So was having a critical look at the image quality and resolution. I did set up scale bars yesterday and didn't notice any CA on the micrometer slide image. It's on another computer but will post an image of it tomorrow.

Thanks

Louise

imkap wrote: ↑

Sat Sep 17, 2022 9:21 pm

LouiseScot wrote: ↑

Sat Sep 17, 2022 9:15 pm

Hiya
The LED illumination certainly could be a factor. But I still ....

My camera too registers more CA than I do. Try fiddling with white balance a bit.

Hiya

Um, see my original post!

Louise

It's interesting that the violet color stood oit to you but not the equal and opposite green also present in tbe imagr. There might still be a screen issue of some sort that's making that color more apparent.

Scarodactyl wrote: ↑

Sat Sep 17, 2022 9:55 pm

It's interesting that the violet color stood oit to you but not the equal and opposite green also present in tbe imagr. There might still be a screen issue of some sort that's making that color more apparent.

Ah, there was more green too, originally, but that lessened after I did the custom white balance. The white balance was why I posted in the first place because I thought the issues were down to the Canon 1100d. Further experimentation and tests required!

Louise

Is your camera parfocal with the eyepieces?

MicroBob wrote: ↑

Sun Sep 18, 2022 1:10 am

Is your camera parfocal with the eyepieces?

Almost! Though I fine focus the camera image via the laptop screen anyway I'll be having a closer look at things today.

Louise

In case it is far from parfocal you use the objective with the wrong distance to the object. In some cases not much happens, in other cases the image suffers a lot.

MicroBob wrote: ↑

Sun Sep 18, 2022 11:28 am

In case it is far from parfocal you use the objective with the wrong distance to the object. In some cases not much happens, in other cases the image suffers a lot.

Hi

I did some fine tuning earlier to make sure the camera sensor is as parfocal with the oculars as possible. It was very close anyway.
I've done some other checks and found that I could see the purple aberration in some places by using a 25x EP which naturally gave a bigger image of the diatom I'd centred on (40 x 25 = 1000). I get the same effect with the EP positioned above the trinocular path. So, it's either intrinsic to the microscope/objective (rather than anything to do with the camera) or it's an artifact of the particular slide/diatom. I'll try some other slides and double check the micrometer image!

Thanks

Louise

A quick update! (where does all the time go??) Anyway, I decided to try the 40x plan fluor on my DIY rail system. However, that was somewhat in pieces as I'd been working on an epi-illuminator for it . The rail setup generally still needs some, erm, enhancements etc. but is just about useable. It's essentially monocular but with either EP or camera attached. It's fitted with a Fujifilm X-T30 APS-C mirrorless camera, and which is 22-26 MPixels with 3.74 um pixels.
A first try seems to show a little less CA than on the AmScope T720 with the Canon:


Maybe less green with the Fuji/rail?

I can't easily tell how much is due to the differences between cameras and/or microscopes. I can fit a 40x CFI60 Plan Fluor and/or a 20x CFI60 plan Apo to the rail nosepiece so will compare them when I get a chance.

Louise

That does look significantly better!

Scarodactyl wrote: ↑

Fri Oct 14, 2022 4:59 pm

That does look significantly better!

I confess I did to some peak alignment in levels though I did the same with T720 image anyway. I'll post again, when I can, with some comparisons using the Nikons.
Louise

Here are a couple of images I took using the Nikon CFI60 40x/0.75 Plan Fluor on the DIY rail with the Fuji X-T30. I have some issues with the rail in that it's hypersensitive to movement especially since I moved it to a different location in my living room. Needs some fettling and engineering. I might try moving it again - it might just be a floorboards issue! Anyway, I think the images have come out pretty much the same as the Chinese 40x/0.75 Plan Fluor. Some purple fringing but little green on my DIY rail system with LED lighting. My conclusion is that the Chinese 40x plan fluor is ok but the issues I thought I had on the AmScope T720 / Canon DSLR could possibly be intrinsic to the T720 stand optics - I'm not sure how much the tube lens might contribute.

The above pics are heavily cropped from the original 6240 x 3512 frames so pushing the limits. I'm expecting a Chinese UPlanFLN 100x/1.30 in the post soon - better for high resolution sharpness!

Louise

Should have bought the Indian planF. You would have saved a fair amount of cash, too.

apochronaut wrote: ↑

Mon Oct 17, 2022 2:05 pm

Should have bought the Indian planF. You would have saved a fair amount of cash, too.

I'm happy with the Chinese one. The Nikon is good too. I wouldn't have bought the Chinese one if I could have fitted the Nikon on the AmScope.

Louise

I set up a halogen bulb and took some images with the Nikon CFI60 40x just to compare:
Using tungsten halogen illumination definitely changes things a bit! It was a 12v 30W bulb which I actually ran at just over 6V so gave a very warm image. I could have run it at 12V but didn't want things to get too hot!
Anyway, I did some peak aligning and the above was the result.

Louise

I was thinking as I read through the series of posts that your LED and not the DSLR is probably a major culprit here. They often have been for me...

ldflan wrote: ↑

Tue Oct 18, 2022 6:58 pm

I was thinking as I read through the series of posts that your LED and not the DSLR is probably a major culprit here. They often have been for me...

Hobbyst46 suggested that in #10 and #11. I'm not completely convinced and I don't exactly understand why it would be. But I'm not ruling it out! I'll try the halogen in the AmScope 720 when I have time later in the week. I see some current Nikon scopes use halogens, others use LED. I wonder if it's the particular LED that can make a difference?

Thanks

Louise