Aperture and image quality

A forum to ask questions, post setups, and generally discuss anything having to do with photomacrography and photomicroscopy.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

Jonas
Posts: 14
Joined: Sat Feb 25, 2017 5:36 am

Aperture and image quality

Post by Jonas »

When shooting at higher (>2x) magnification, something fishy happens with the details in my images at lower apertures: while I can see that some of the image is in focus, it is completely unusable because the in-focus area is covered in some milky low-contrast haze. I realize that stopping down too far reduces image quality, but is it also the case when stopping up?
When using 68mm extension tubes behind my tokina 100mm 2.8 and a raynox msn-202, the blur does not seem to disappear untill f/8 (my camera does not correct for the tubes, so the effective aperture is much higher). This seems too high to me and I am wondering if I am doing something wrong.
I have inserted some test images. The area between the leftmost yellow spot and the left antenna is in focus in all the images, but serverly affected by haze in the first several examples.
This problem has haunted me for a while and I really hope someone can shed some light on it.
-Jonas
Image
Image
Image
Image

rjlittlefield
Site Admin
Posts: 23564
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

The milky appearance shown here is typical of optics that have a lot of "spherical aberration", which basically means that different parts of the lens are focused at different distances. The result is that even at best focus what you're seeing is actually a combination of in-focus and out-of-focus light. It's the out-of-focus light that contributes the milky haze. Stopping down is a powerful way to reduce spherical aberration, consistent with the behavior that you're seeing.

Some amount of spherical aberration is unavoidable whenever a lens is focused differently from how it was designed.

But in your case I think the problem is particularly severe because you're using both lenses far away from their design points. The Tokina was not designed to go in front of tubes, and the Raynox was not designed to go in front of another lens that is already focused to a very close distance.

I suggest to start by removing the tubes, focus the Tokina at infinity, stick the MSN-202 in front of that, and see how the combo works. From there, you can focus the Tokina closer, and if everything is still OK at the closest focusing position of its focus ring, then you can add tubes behind that. Be sure to keep the Tokina's ring at its closest focus point, and add as few tubes as possible.

The MSN-202 is nominally a 40 mm lens (25 diopters), so using it in front of the Tokina 100 mm at infinity focus should give you 2.5X (2.5=100/40). Focusing the Tokina closer than infinity will give you some further magnification, but don't be surprised if it's less than you expect. That's because part of the Tokina's focusing mechanism probably involves making its focal length shorter as it focuses closer, which reduces the back/front ratio that was originally 100/40.

--Rik

Jonas
Posts: 14
Joined: Sat Feb 25, 2017 5:36 am

Post by Jonas »

Thank you, Rik!
This was exactly the kind of response I was looking for. I have tested mounting the Raynox on a tele-lens at 100mm without the tubes. This gives me a decent result at about f/8, which should be below the diffraction-blur threshold, but I think that it is still quite high. I have noticed that the issue appears to be exacerbated when I photograph something with a reflective surface (such as the exoskeleton of the bee) compared to more dull subjects - does this make sense?
The fact that I was using the equipment in an unintended manner is news to me and this makes me feel like I need a good source of that kind of information; can you recommend a book or online resource that would be useful in this regard? I have browsed extreme-macro.co.uk for advice on techniques, but I have not come across anything mentioning that using Raynox or tubes on macro-lenses would be a bad idea. Perhaps I have not been thorough enough...

-Jonas

rjlittlefield
Site Admin
Posts: 23564
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

Jonas wrote:I have tested mounting the Raynox on a tele-lens at 100mm without the tubes. This gives me a decent result at about f/8, which should be below the diffraction-blur threshold, but I think that it is still quite high.
I assume you mean that the rear lens was set on f/8. In that case, the MSN-202 is operating at f/3.2. That number can be calculated as the aperture diameter of the rear lens, 100mm/8=12.5mm, divided into the focal length of the front lens, 40mm/12.5mm=f/3.2. If you open the rear lens even farther, the front lens f-number would be even smaller. Noting that the MSN-202 is a simple and inexpensive design, 4 elements in 3 groups for about $80 USD, I think it would be surprising if its best performance occurred at apertures much wider than f/3.2.
I have noticed that the issue appears to be exacerbated when I photograph something with a reflective surface (such as the exoskeleton of the bee) compared to more dull subjects - does this make sense?
There are several separate but related problems that can result from the combination of reflective subjects and "hard" lighting, where the light comes from a narrow range of angles.

The underlying issue is that with a reflective subject and hard light, each point on the subject acts like a tiny mirror that reflects a narrow beam of light to just a small part of the lens aperture. Different points on the subject end up sending beams of light to different parts of the lens aperture. The exact effect depends on the position of the point in the field, the angle of the surface of the subject at that point, the size of the light beam, the aberrations of the lens for the angle and position where the beam hits it, and probably several other things that I've overlooked at this moment. Anyway, the bottom line is that you can end up with a significant loss of image quality. So yes, it does make sense that you would have more trouble with highly reflective subjects.

The best attack on this problem is to use lots of diffusion, more than you think is necessary and then add some more. Imagine yourself in the position of the subject, shrunk down to the size of the subject, looking up at the lens surrounded by "sky". You want the illumination to look like a big bright fog bank that obscures the location of the sun but provides lots of light from all around you. The usual implementation is to use a diffuser that wraps around the subject, with the actual light sources relatively far away from the diffuser so that the diffuser is illuminated evenly.

Once you get good results with very diffused illumination, then you can experiment with less diffusion to get more directional illumination for better modeling.
The fact that I was using the equipment in an unintended manner is news to me and this makes me feel like I need a good source of that kind of information; can you recommend a book or online resource that would be useful in this regard? I have browsed extreme-macro.co.uk for advice on techniques, but I have not come across anything mentioning that using Raynox or tubes on macro-lenses would be a bad idea. Perhaps I have not been thorough enough...
Sorry, I do not know any resource that concisely covers issues like we're discussing here.

Perhaps somebody else will chime in with some good suggestions.

--Rik

Jonas
Posts: 14
Joined: Sat Feb 25, 2017 5:36 am

Post by Jonas »

Thank you so much for your time, Rik. You have been most helpful. I will work on my diffusion and try not to squeeze every ounce of magnification out of my relatively inexpensive gear for a start.

Best wishes
Jonas

palea
Posts: 40
Joined: Mon Dec 17, 2018 8:09 pm

Post by palea »

rjlittlefield wrote:Sorry, I do not know any resource that concisely covers issues like we're discussing here.
I think, in general, it's your posts. Any two or three of them on a given topic typically provide a quality, concise answer. The difficulty is more in finding them, and sometimes finding the most current ones, among all the threads.
rjlittlefield wrote:Noting that the MSN-202 is a simple and inexpensive design, 4 elements in 3 groups for about $80 USD, I think it would be surprising if its best performance occurred at apertures much wider than f/3.2.
This makes me wonder what might more usefully handle the focus range of a macro without becoming overly expensive or cumbersome. The Raynoxes are fairly affordable, compact, and lightweight. They offer less diffraction than objectives and fewer support, vibration management, and lighting complexities than reversed lenses. The few objectives with longer working distances than the 202 are substantially more expensive, as are lenses like the Apo-Componons. Something like a Componon or Componon-S (often) costs more and has more elements, but is also designed to a different task and brings along more vigetting issues.

Scientific lens suppliers offer many achromat doublets but I'm hesitant on a 2/1 aspheric formula compared to the 4/3 202 or 3/2 250, partly as pricing is higher than the 505 and mounting more involved. The advantage there is perhaps more doublets are available in many focal lengths. Asymmetric doublet pairs presumably don't have much to offer over six element lenses like the Componons in image quality and they're commonly also bigger and more expensive. All the Steinheil triplets I've come across seem designed for 1:1 conjugate ratios and it's unclear to me if spherical Hastings triplets would improve over an aspheric doublet. One could buy and mount retrofocal elements before a more corrected (and more afocal?) positive group but the cost of doing so is high.

Some forum members might have the budget to commission a purpose built improvement on a Raynox but, like high NA LWD objectives, that's beyond my means. Maybe if someone did a Kickstarter.

Thoughts?

Post Reply Previous topicNext topic