Lenses for use at 4-5X on an APS-sized sensor

[naturephoto1]

Hi Rik,

You had mentioned using the Nikon CFI BE Plan Achromat 4X NA 0.10 with a tube lens with a teleconverter. I have never used a tube lens so I have experience. Would the 4X Nikon lens need to be used with another lens and possibly with a teleconverter? Or, could you possibly use the Nikon 4X lens with a bellows and/or extension tubes and a 1.4X or a 2X teleconverter (as the tube lens) and still get good results?

Thanks,

Rich

[nucleobyte]

Hey Rik,

I must be missing something obvious, but I was surprised not to to see the Nikon 4X/0.10 160/- finite conjugate from Edmunds in this line-up. I've been getting very nice results from it from 4X up to 8X on a full sized sensor (although the corners do fall off a bit at 4X.)

Greg

[rjlittlefield]

You had mentioned using the Nikon CFI BE Plan Achromat 4X NA 0.10 with a tube lens with a teleconverter. I have never used a tube lens so I have experience. Would the 4X Nikon lens need to be used with another lens and possibly with a teleconverter? Or, could you possibly use the Nikon 4X lens with a bellows and/or extension tubes and a 1.4X or a 2X teleconverter (as the tube lens) and still get good results?

If you haven't read it already, I recommend FAQ: How can I hook a microscope objective to my camera?.

Short answer to your question is that "infinity" objectives like this one are aberration-corrected to produce their best images when focused so their output is "at infinity", each point on the subject represented by a bundle of parallel rays leaving the lens. To form a real image from that, you need a "tube lens" that again is focused "at infinity", accepting as input bundles of parallel rays and refocusing each bundle to a single point on the sensor. If you remove the tube lens and make the objective form a real image on the sensor all by itself, then the lens develops spherical aberration, just like an ordinary landscape lens will do if you push it too far toward macro. The resulting image is soft, with reduced contrast for details.

Adding a teleconverter to an existing tube lens serves to make the image bigger, or equivalently, the system acts as if the tube lens has a longer focal length. The teleconverter will also make the image bigger if used behind a bellows or extension tubes, but that doesn't help with an infinity-corrected objective because the damage has already been done by introducing spherical aberration because of the change in focusing relationships.

--Rik

[rjlittlefield]

I must be missing something obvious, but I was surprised not to to see the Nikon 4X/0.10 160/- finite conjugate from Edmunds in this line-up. I've been getting very nice results from it from 4X up to 8X on a full sized sensor (although the corners do fall off a bit at 4X.)

The study started off as a search for a good, cheap, 4X objective that could be easily obtained by anybody in the world. That's why the two no-name Chinese objectives are in there. (One is pretty good, the other not.) The AmScope got added as a potentially good, cheap, 4X objective that could be easily obtained by anybody in the US. People outside the US regularly complain about Edmund being expensive and/or difficult to deal with, so I left out their finite Nikon for this round of testing.

It's interesting & very promising that you've been getting good results from Edmund's finite Nikon even at 4X on a full frame sensor. We've always thought that the Edmund's finite Nikons were either identical or very closely related to the discontinued Nikon CF E objectives like the one I tested. But as shown, at least that one sample of the CF E is not good in the corners even on APS, while your Edmund's lens must be quite good if it comes close to covering full frame too. I'll revisit the Edmund's offering at some future time. It would be nice to have a finite 4X with many of the same characteristics as the infinite CFI BE.

--Rik

[rovebeetle]

Excellent thread and very helpful, since I am in search of a suitable lens for the 1:1-5:1 range right now. Shooting Nikon, the MP-E is no choice.
So far I am torn between the EL-Nikkor 50/2.8 or the Rodenstock Apo-Rodagon N 50/2.8.
I already use an EL-Nikkor 75/4 but somehow I am not really content. Maybe I am just spoilt by the quality of the Mitutoyo objectives which I use above that ratio.

Cheers

[ChrisR]

That's perhaps one of the more underwhelming lenses you could have chosen, Harry.
Any of the 6 element (yours is 4) enlarging lenses from the major makers will beat it.
You're up against a few things though
some boundaries :
1) diffraction
2) the lenses were mostly specified for use down to about 2x, so about half the range you're looking for is "off the end". I have a number of candidates but haven't compared them all at both ends of the scale.

What I have found, is that they vary quite a lot, it depends on the lighting and the subject itself. Different parts of the same image can look better or worse. I have a couple of particularly poor examples too, but not enough to be statistically very meaningful.

The apos are certainly less prone to odd/undesirable colour effects like flares and aberrations, but how much you'll notice them, if at all, or be bothered by a few minutes in post processing, is hard to quantify.
The often lauded El Nikkor 50mm f/2.8 is of course a good'un. Often to be found in Germany selling at low prices, not for export. Well worth a try if they'll post to Austria!

[rovebeetle]

Chris, thanks for your input. German sellers always ship to Austria as well, so no problem here. One of them I even had dealings with before.
I am pretty sure that I will be using it mostly in the 2:1 - 3:1 range, therefore no extensive extension needed. That should keep diffraction quite under control. It is just a small gap that has to be closed, for 1:1 and larger specimens I have my Mikro- and PC-Nikkors.

The El-Nikkors are cheap indeed, the Rodenstock costing about twice as much (still quite affordable, though), but since I do not have to pay for it ... .

Cheers

[Craig Gerard]

Harry,

A Schneider-Kreuznach Apo-Componon HM 40/2.8 is another lens to consider. They come in various barrels (some require reverse mounting). The lens likes to be used in the 5 to 7x range.

http://www.photomacrography.net/forum/v ... php?t=6330


Craig

[mlackey]

This is an infinity objective that has to be used in conjunction with a tube lens. It's designed to deliver 4X on 200 mm, but it can be pushed cleanly down to 2.7X on a 135 mm, or even down to 2X on a 100 with some darkening and softening of the corners. At 2.7X and above, the image is sharp and clean corner to corner -- very nice!

I know you have not tried this, but based on your experience with the 4x, would you think the 10x on a 70-200 (at 200) would yield similar results?

Note that unlike the Nikon CFI objectives that we're most familiar with, which are CFI60's, the BE objectives use a standard DIN thread and parfocal distance. This means that BE objectives need a different adapter from the Nikon CFI 10X objective seen earlier in this topic.

I hope you don't mind me asking, but... what is standard DIN thread and parfocal distance? Or can you point me in the right direction?

Edit: standard DIN thread is 20.1 mm and 36 TPI?

Edit: found the answer to the parfocal length question:
http://www.microscopyu.com/articles/for ... focal.html

[rjlittlefield]

I know you have not tried this, but based on your experience with the 4x, would you think the 10x on a 70-200 (at 200) would yield similar results?

That's my expectation, but I'll report back when the CFI BE 10X arrives.

Actually the BE 4X was sort of an afterthought that I bundled into an order for the BE 10X , but it turned out that OpticsPlanet is back-ordered on the 10X. Their exact quote was "They are very popular and hard to keep around. We do expect to receive the merchandise back in stock within 5-10 business days." I have no idea what the 10X will even look like. Often 10X plans are very short on working distance.

If a 10X finite will meet your needs, then Edmund's Nikon Finite Conjugate Achromat 10X NA 0.25 is a known good lens. See http://www.photomacrography.net/forum/v ... php?t=9393 for details about that.

I hope you don't mind me asking, but... what is standard DIN thread and parfocal distance?

Edit: standard DIN thread is 20.1 mm and 36 TPI?

Edit: found the answer to the parfocal length question:
http://www.microscopyu.com/articles/for ... focal.html

Both threads are the same: RMS (Royal Microscopical Society), as defined HERE. JIS uses a 36mm parfocal distance (from shoulder of mounting threads to plane of focused subject) while DIN uses 45mm and the Nikon CFI60 uses 60 mm. DIN finite objectives are designed for 160 mm tube length, while JIS uses 170 mm. "Tube length" means the optical path length from shoulder of mounting threads to end of eyepiece tube. The objective is actually designed to form its image 10 mm down from the end of the tube, so the distance from the shoulder of the mounting threads to the camera sensor is 150 mm for DIN and 160 for JIS. The difference in tube length is not significant at 4X (other than changing the magnification by a little bit), but the reduction in parfocal distance with JIS means that you often can't mix-and-match within the focus range of a microscope stage, and when you stick the objectives on extension in front of a camera, JIS gives about 9 mm less working distance than you'd think from just looking at the objective.

--Rik

[mlackey]

If a 10X finite will meet your needs, then Edmund's Nikon Finite Conjugate Achromat 10X NA 0.25 is a known good lens.l

Rik,

Thank you very much.

After reading the "how to attach" FAQ I started accumulating the small pieces needed to attach a finite objective. Being new to this, the choices (both new and used) seemed overwhelming. I do not have sufficient knowledge to know what I'm really looking for. I asked about the BE infinite because I was considering change of plan. You by chance mentioned the Nikon 10x, which is exactly what I need to get back on track.

Thanks again.

[pierre]

Rik,

Thank you very much for this test !
Instructive as always

[Tim M]

Hi Rik,

I tried a Nikon CFI 10x Plan like yours on a Canon 55-250 f/4.0-5.6 lens and was seeing significant dark corners down at 100mm and even 135mm (but it sure rocks up at 200-250mm!). It's bad enough at 100mm that the field is effectively reduced by over 30%. I think I had the lens wide open. Would you expect that result, or am I possibly doing something incorrect? Would a faster lens help?

Thanks, and for extra credit, where would your reversed 18-55 Canon kit lens rank in your 4x-5x lens collection? I just tried my first images with that set-up and found it to be right handy Kept me from switching back and forth between 4x CF N Plan, el Nik 50/2.8 and 80/5.6

Thanks again,
Tim

[ChrisR]

Tim that's because it's a Zoom I'm sure. The entrance pupil moves too far back. They usually "work" at the longer end and not the shorter.
The 10x works well on a Micro Nikkor 105mm. As a cheaper option there are many 100mmm enlarger lenses going through ebay, which wouldn't be too hard to connect up, with the right step rings etc. Ordinary Rodagons and Componons aren't a lot of money. I have a couple so they're on my list to try - though that's getting longer.

[Tim M]

As a cheaper option there are many 100mmm enlarger lenses

Thanks Chris,
Silly question but do I reverse those enlarger lenses, as I do when using the 50mm & 80mm el Niks directly for macro? (I assume not). That's gonna be a strange looking setup on the camera! And I assume to mount the enlarger lens as close to the camera as possible?

Thanks,
Tim