Tube Lens tests - medium format and full frame

Have questions about the equipment used for macro- or micro- photography? Post those questions in this forum.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

nathanm
Posts: 222
Joined: Thu Jun 02, 2016 8:13 pm
Location: Bellevue, WA
Contact:

Tube Lens tests - medium format and full frame

Post by nathanm »

My colleagues and I have just finished a round of tube lens testing. We used Mitutoyo 5X objective for this.

One aim was to get a tube lens that would cover the PhaseOne (53.7 x 40.4 mm sensor). The other was to see what was best for full frame (35mm) sensors.

We did a ton of testing on an optical bench which allowed us to test both forward and reverse, and different tube lens placement.

We found that in general quality in the corners and lack of vignetting improved the closer the objective was to the tube lens. That was not always true, but often we found that putting the objective nearly touching the tube lens was best.

The lenses we tried were primarily close up lenses. Since Raynox 150, 250 are widely used we tried a lot of alternatives. We tried some large format lenses (Apo Gerenon) and camera lenses, but none of those came out on top.

Quality was judged by stacking enough frames that the corners and center were both in focus, and then looking at the stacked result. That means that field flatness was not selected for - which is appropriate for lenses used for stacking.

A Thorlabs test target was used, which makes it easy to determine magnification.

For PhaseOne there were several winners.

4.8X Raynox DCR5320AB set (72mm diameter), forward, objective close (but not very close).

4.9X Hoyer & Speyer 160mm focal length doublet, forward, objective almost touching

5.6X Leica Elpro 2 close up lens, reverse position, objective close

6X Century Precision Optics +2 diopter, 72mm diameter, reverse objective almost touching

For Canon Full Frame

4.1X Leica Elpro VI B, forward, objective almost touching (tiny amount of vignetting)

4.3X Leica E60 (1:2 converter), forward ,objective close

4.8X Raynox DCR5320AB forward, objective close

5.6X Leica Elpro 2 close up lens, reverse position, objective close

Note that I did not try tube lenses significantly longer than the nominal 200mm intended for Mitutoyo.

Note that the magnification was determined from the best combination. We started with tube lens focused at infinity, but sometimes the best result was slightly off of that.

In these tests the Raynox 150 and 250 (which I had previously used) were not as good as any of these. Obviously there are many people on the forum who use them successfully - but they had more CA in our tests than the lenses above.
nathanm

Lou Jost
Posts: 5942
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

Excellent information at last about best tube lenses. Thanks!

Could you tell us which camera lenses you tested and how they fared?

nathanm
Posts: 222
Joined: Thu Jun 02, 2016 8:13 pm
Location: Bellevue, WA
Contact:

Post by nathanm »

Mamiya and Schneider lenses for Phase One - several of them (120mm, 210mm, 300mm)

240 mm Apo Gerenon
nathanm

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

Nice -- thanks.

A couple of comments/questions...

1. Google search on "DCR5320AB" returns nothing useful (nothing at all at the time I did the search). Raynox makes a DCR-5320, which is a +2 and +3 set. Is that what you mean, with both components used? If so, in what order and does order matter?

2. I was struck by the bit about "4.9X Hoyer & Speyer 160mm focal length doublet". At infinity focus, a 160 mm tube lens would give 4X with a Mitutoyo 5X objective. I see your comment that "We started with tube lens focused at infinity, but sometimes the best result was slightly off of that." So I'm thinking that for this lens, "slightly off" must have been with about 22% extra extension. Do I have that right, or was the lens actually not near 160 mm to start with?

3. Did you actually succeed in getting corner-to-corner sharpness on the PhaseOne sensor (53.7 x 40.4 mm), at the magnifications noted above?

4. Did you try anything at higher magnifications? My guess is that the Mitty 10X is the one that most people would be interested in first.

--Rik
Last edited by rjlittlefield on Fri Oct 28, 2016 8:55 pm, edited 1 time in total.

nathanm
Posts: 222
Joined: Thu Jun 02, 2016 8:13 pm
Location: Bellevue, WA
Contact:

Post by nathanm »

I should add that none of the camera lenses did well in the tests. Mostly the problem was vignetting.

The close up lenses are all very fast - for example the Raynox DCR5320AB has a focal length in this test of 192 mm. It is 72mm in diameter so that is about an f/2.7 to f/2.8 lens.

They are also pretty thin (although DCR5320AB is one of the thickest).

Vignetting is determined by a complicated mix of the size of the entrance and exit pupils and their locations. We had strong vignetting for the camera lenses and the large format lens (Apo Generon) used as tube lenses.

Because my main goal was covering the PhaseOne sensor this is pretty important.
nathanm

nathanm
Posts: 222
Joined: Thu Jun 02, 2016 8:13 pm
Location: Bellevue, WA
Contact:

Post by nathanm »

Yes, the DCR-5320 is a set with two components which the pamphlet with it calls A and B. This is both together in the order they recommend. I did not try other orientations.

You are correct that we played with the focus until we got the best result. The S&H lens is marked 160mm focal length, but we are using it a bit longer.

That one by the way was the best.

Yes we get corners reasonably sharp and bright. The center still wins but for my uses this is quite good. I am traveling at the moment but I will post an example when I can.
nathanm

nathanm
Posts: 222
Joined: Thu Jun 02, 2016 8:13 pm
Location: Bellevue, WA
Contact:

Post by nathanm »

We will confirm with 10X soon
nathanm

nathanm
Posts: 222
Joined: Thu Jun 02, 2016 8:13 pm
Location: Bellevue, WA
Contact:

Post by nathanm »

Here is a photo of the 5X Mitutoyo mounted on the Raynox DCR-5320 pair.

They take 72mm filter mount, so we made an adapter that threads to 72mm, it is concave just enough to put the end of the objective about 1 mm in front of the front element of the DCR 5320 pair.

Image

We made this after the optical bench tests.

The DCR 5320 is threaded into a 72-77 mm step down adapter, with the 77 mm end threaded into Thorlabs SM3 tube.
nathanm

dmillard
Posts: 636
Joined: Tue Oct 24, 2006 7:37 pm
Location: Austin, Texas

Post by dmillard »

nathanm wrote:
The close up lenses are all very fast - for example the Raynox DCR5320AB has a focal length in this test of 192 mm. It is 72mm in diameter so that is about an f/2.7 to f/2.8 lens.
Thanks for doing this huge amount of work, and sharing your results!

Could you please confirm your measured focal length on this converter? Raynox describes it as having a focal length of 170mm (see here), which would provide a magnification at infinity of 4.25X with a Mitutoyo 5X objective.

David
Last edited by dmillard on Sat Oct 29, 2016 6:23 am, edited 2 times in total.

Lou Jost
Posts: 5942
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

Nathan, did some of the camera lenses and large-format lenses do better than the close-up lenses in center sharpness?

Any enlarger lenses in the test?

How important is that non-standard objective-to-DCR-5320 distance? Does the nonstandard adapter make a big difference? Is the difference primarily in field coverage or is there also a difference in optical quality near the center? Do you have any extra adapters you'd like to sell?

Again, thanks for this test!

nathanm
Posts: 222
Joined: Thu Jun 02, 2016 8:13 pm
Location: Bellevue, WA
Contact:

Post by nathanm »

We didn't measure the focal length, we used the spacing of lines on the Thorlabs test target to determine the magnification directly.

So in the case of the DCR 5320 we are getting 4.8X so the effective focal length must be 192mm

I did not try enlarger lenses. My plan was to try them next if the close up lenses didn't work. There aren't that many enlarger lenses in the 200 mm focal length range which is roughly what we are looking for in a tube lens.

I was surprised that I didn't find better low mag choices for full frame. The Phase One at 4.8x has a field of view that should be about 3.2X on FF. So the objective can do the job. But the lowest mag combination we found for FF was much higher.
nathanm

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

nathanm wrote:I was surprised that I didn't find better low mag choices for full frame. The Phase One at 4.8x has a field of view that should be about 3.2X on FF. So the objective can do the job. But the lowest mag combination we found for FF was much higher.
I'm unclear on your strategy.

To get 3.2X on FF, with the tube lens at infinity focus, would require focal length 3.2/5*200 = 128 mm.

I think I recall seeing some reports of good results on full frame with a 135 mm tube lens, which would be pretty consistent with that.

What tube lenses did you test around that focal length?

--Rik

nathanm
Posts: 222
Joined: Thu Jun 02, 2016 8:13 pm
Location: Bellevue, WA
Contact:

Post by nathanm »

About to board a flight to Australia, will post the list of lenses we tested after I get there
nathanm

Chris S.
Site Admin
Posts: 4042
Joined: Sun Apr 05, 2009 9:55 pm
Location: Ohio, USA

Re: Tube Lens tests - medium format and full frame

Post by Chris S. »

Nathan, a few thoughts.
nathanm wrote:Note that I did not try tube lenses significantly longer than the nominal 200mm intended for Mitutoyo.
I respectfully ask, “Why not?”

My thinking is as follows: The diameter of the high-quality image circle of Mitutoyo lenses is roughly 24mm, depending on one’s standards; some use more of the image circle than this, while others do not. In my work, I use 24x16 mm (Nikon DX-APS-C) sensors and a 200mm converging lens with Mitutoyo objectives, because the just-under 29mm diagonal of these sensors is a close match to the best portion of the Mitutoyo image circle. But I can understand the rational of folks who prefer a 36x24 sensor, with its 43mm diagonal, for these optics; they lose quality at the edges, but gain field of view. This said, stretching this approach to a 53.7mm x 40.4mm sensor of your PhaseOne body, with its 67mm diagonal, strikes me as excessive. You are likely to be capturing a large portion of the objective’s low-quality image circle.

If I needed to enlarge the high-quality image circle of an infinity microscope objective, my intuition suggests using a longer focal length lens. In your situation, a 500 converging lens—if it doesn’t vignette—would seem ideal. This would of course need to be tested. So in my mind, your tests should include some much longer focal length converging lenses.
nathanm wrote:In these tests the Raynox 150 and 250 (which I had previously used) were not as good as any of these. Obviously there are many people on the forum who use them successfully - but they had more CA in our tests than the lenses above.
Kudos for including chromatic aberration in your evaluation; my sense has long been that CA has been undervalued in converging lens tests.
nathanm wrote:Quality was judged by stacking enough frames that the corners and center were both in focus, and then looking at the stacked result. That means that field flatness was not selected for - which is appropriate for lenses used for stacking.

A Thorlabs test target was used, which makes it easy to determine magnification.
Great that you comparied stacks, rather than single shots! Unlike quite a few other comparisons, you have, in this step, avoided not just field flatness, but potential focus errors. IMO, all such lens tests should be done on stacks.

This said, would you mind telling us which Thorlabs target you used? Our interpretation of your tests should include this information as a base.

And in a much more important point, could you possibly make your test images available to forum members for personal evaluation? Granted, this takes a lot of time. But the value is enormous, and the applicability of your tests to other users devolves upon it. People perceive visual information in very different ways; my eyes/brain have differed from those of esteemed forum colleagues more times than I can count. I haven’t much disagreed with their methods, but have differed in my visual interpretation of their data. So I think that tests like yours must, to be useful, make data available for widespread interpretation.

Cheers, and very best regards,

--Chris

nathanm
Posts: 222
Joined: Thu Jun 02, 2016 8:13 pm
Location: Bellevue, WA
Contact:

Post by nathanm »

I just arrived in Australia, and I don't have everything with me.

Here is the list of lenses that we tested, with their approximate focal length

Leica Elpro1, 398mm
Leica Elpro2, 203mm
Leica Elpro3, 602mm
Leica Elpro4, 1333mm
Raynox DCR150, 208mm
Raynox DCR250, 125mm
Marumi DHG +5, 200mm
Leica E60, 216mm (ESTIMATED)
S+H 160mm, 160mm
Century +7, 142mm
Century +2 small C167073, 500mm
Century +2 large C41191, 500mm
Opteka 10x, 100mm
Raynox DCR5320 A+B, 170mm
Raynox DCR5320 A, 486mm
Raynox DCR5320 B, 311mm
Zeiss/Hasselblad 0.5, 500mm
Zeiss/Hasselblad 1, 1000mm
Zeiss/Hasselblad 2, 2000mm
110 mm Mamiya, 110mm
120 mm Mamiya, 120mm
300mm Mamiya, 300mm
210mm Mamiya, 210mm
Leica ElproVIIa, 602mm
Leica ElProVIb, 203mm
Leica Elpro VIIb, 1333mm
Leica Elpro VIb, 203mm
Leica Elpro VIa, 398mm
L14728 BINOCULAR OBJECTIVE 45MM DIA BY 160MM FL

I previously tested the Apo Gerenons
nathanm

Post Reply Previous topicNext topic