Extending the magnification range of Printing Nikkors

[ray_parkhurst]

Do you know of any of the PN 105 A that are already configured for DSLR use . . . ?

If a Printing Nikkor 105mm is not already configured for DSLR use, making it so is not difficult. Start with Rafcamera's eBay part 142382208066, add a 52mm-thread extension tube, then a reversing ring for your camera mount.

Back in 2012, before the Rafcamera adapter was available, Ray Parkhurst helped a few of us get custom adapters made by his machinist. These adapters were (and still are) very nice, and are the equivalent of the Rafcamera adapter part plus a length of 52mm reversing ring, though as a single piece. If Ray has any of these left, or if his machinist cares to make more, I'd highly recommend them. But with the Rafcamera part readily available off-the-shelf (and given the good quality of Rafcamera items), going with the Rafcamera part seems like an easy and very workable solution.

--Chris S.

I have a couple of these adapters still around, but I had them made for the 95PN in order to mount it in 1:2 configuration. For the 105PN, or the 95PN in 2:1 configuration, I highly recommend Raf's M45-M42 adapter:

http://www.ebay.com/itm/M45x0-75-female ... Sw42JZFfRU

There is a trick to make this work...the front trim ring, which has m43 filter threads, is removable, and reveals an M45 thread that allows the Raf adapter to attach.

Remember, the 105PN is optically symmetrical, so it doesn't matter which way you mount it. I've been using mine with the Raf adapter and it works great.

[austrokiwi1]

Still thinking hard about this. Lou you are using a MFT camera. With such a small sensor you are "looking" through the best part of the glass both with and without the teleconvertor. My teleconvertor isn't ideal for this so read this with a caution: When I try the technique out with my ff frame camera I am seeing IQ deterioration near the edges of the frame. When I switch to my NMFT camera I see no such deterioration. I suspect that The technique may well be best suited for small sensors

[Lou Jost]

That's a good point, austrokiwi. But in relative terms, what I've shown is that the TC image is better than the PN +extension in the central 17mm image circle. That holds true in FF as well. Though a FF camera's bigger pixels may not show as strong a difference as my 50-80 Mp 17mm sensor.

What TC did you use? If you get deterioration on the edges of FF with the TC, might you also get the same or worse deterioration when using the P105 with just extension at the same m?

I have sometimes used a shift adapter which lets me test the image circle out to almost the corners of FF. I can look at that in the future when I have more time.

Where did you put the TC? That matters a lot.

[ray_parkhurst]

I've usually thought of the Vivitar 2x MFTC as one of the best 2x TCs out there. It is certainly better than the 2x Nikon TC's I have, which start to show a little softening on APS-C. I thought it was good enough to do the "new" test method for lens evaluation I was proposing, see here:

http://www.photomacrography.net/forum/v ... highlight=

However, I started down the path and found the corner results were all about the same for several lenses with widely different expected qualities. Looking closer, I found the Vivitar TC was unfortunately slightly softening the corner, enough that great lenses were looking mediocre. I expected almost no softening vs the center given the test method views the performance well within even the APS-C coverage, and verified this with a non-TC test.

Needless to say, I won't be using the TC method for lens tests, and have lowered my expectations for TC quality in general.

[austrokiwi1]

The TC was a Minolta MD 2X it was placed between the camera and the bellows. Not sure what you mean with the extension. I set the bellows extension so the PN 105 was at 1X. I tried it with and with out the TC (with the TC I was getting 2X as I should have). If I set the extesion on the bellows so the PN was at @X I know I would get shockingly poor IQ( I tried it in the past) much worse than what I was seeing to day with the TC.

[Lou Jost]

Ray, in my experience the TC method does work very well to distinguish the quality of lenses, especially long telephoto ones. I have used that method informally all my life. But I have looked mostly at center resolution. For that purpose I do think it is a great way to quickly check quality of a lens.

But a milder TC is better. Nikon now makes 1.2, 1.4, and 1.7 TCs and they are exquisite, far better than older TCs. There have been big recent advances in TC design.

Austrokiwi, yes, that's where the TC needs to go. It sounds like your experience proves my point, right? When you use a 2x TC to get m=2, you find that using the TC produces corner degradation, but using just added extension (instead of the TC) to get m=2 gives you much worse IQ ("shockingly poor") than using the TC. Is that correct?

I'd strongly recommend using a really good modern 1.4x TC.

[austrokiwi1]

yes... but it also suggests to me the technique will not work with FF sensors... and might still show issues with APSc ( Nikkon and Sony perhaps less so with Canon)

[ray_parkhurst]

Ray, in my experience the TC method does work very well to distinguish the quality of lenses, especially long telephoto ones. I have used that method informally all my life. But I have looked mostly at center resolution. For that purpose I do think it is a great way to quickly check quality of a lens.

But a milder TC is better. Nikon now makes 1.2, 1.4, and 1.7 TCs and they are exquisite, far better than older TCs. There have been big recent advances in TC design. ...

Lou...certainly the center is no issue. In the center I see all the benefits of using the TC to improve the final image resolution that I've shown in past posts. It is indeed the corners that are the problem. The issue is subtle, and could be overlooked if I wasn't being extremely picky, but ANY degradation (or lack of improvement) vs the center becomes highly visible when you're doing testing.

Unfortunately, the logic of the test falls apart with anything other than a 2x TC, so unless I can find a 2x TC that is essentially perfect out to the corners of APS-C, I am dead in the water with this method...Ray

[rjlittlefield]

Perhaps it would help to formulate the issues in a little different way...

Suppose you have the following equipment:

1. MFT sensor (17.3 mm wide)
2. Printing Nikkor lens
3. Bellows
4. 2X teleconverter

and your goal is to:

6. image an 8.65 mm subject width.

Question: which of the following approaches will give you the best image?

A. Set the bellows so the Printing Nikkor runs at 1:1, and crop to the center half of the sensor. [This gives an ideal optical image, but the captured image likely will be limited by sensor resolution.]

B. Extend the bellows so the Printing Nikkor runs at 2:1, and use the whole sensor. [The optical image will be degraded by aberrations from running the PN away from its design point, but sensor limits will have less impact.]

C. Set the bellows so the Printing Nikkor runs at 1:1, add the 2X TC, and use the whole sensor. [The optical image will be degraded by aberrations from the TC, but again less impact from sensor limits.]

The essence of Lou's argument is that with a high quality TC, option C will give the best image.

But everything is connected to everything. If the sensor is extremely high quality, then conceivably option A could be best, and with a low quality TC, option B might be best.

Changing the sensor size and/or the size of the subject area to be captured will change the imaging tradeoffs, but the conceptual framework stays the same.

If you want to image other sizes of subject areas, or you have a TC of different strength, then there are more options that involve different crops and/or setting the bellows at other lengths so as to run the Printing Nikkor at other magnifications different from 1:1. Again, I think that Lou's point would be that with a high quality TC, you'll get a better result from using the TC than from any other approach.

--Rik

[ray_parkhurst]

I think Lou is specifying use of a 1.4x TC rather than 2x in order to find the sweet spot for the method.

[Lou Jost]

Rik's caveat is important; my conclusion that the TC is better than pure extension is only valid at current and near-future pixel densities.

Note however that I used a 50Mp 17mm-wide image, and still the TC version was far better than the cropped-and-enlarged 1:1 version. It is going to be a long time before sensors are so dense that cropping can compete with optical enlargement in this situation.

Edit to respond to Ray- I used the 1.4x because it only reduces the aperture by one stop. More than that and I am into diffraction territory for this lens in the hi-res MFT mode. So for my purposes a 2x converter is not a viable option, and I would not try to push the PN105 that far. I'd search for better methods to get m=2, using coupled lenses or the Myutron or Rayfact or Schneider lenses designed for this m, rather than the PN105.

So I would add an option D to Rik's list:
Buy a more appropriate lens for the job.

[ray_parkhurst]

...

Edit to respond to Ray- I used the 1.4x because it only reduces the aperture by one stop. More than that and I am into diffraction territory for this lens in the hi-res MFT mode. So for my purposes a 2x converter is not a viable option, and I would not try to push the PN105 that far. I'd search for better methods to get m=2, using coupled lenses or the Myutron or Rayfact or Schneider lenses designed for this m, rather than the PN105.

So I would add an option D to Rik's list:
Buy a more appropriate lens for the job.

That's what I meant by sweet spot. 1.4x is enough to be a useful additional magnification where 2x is too much from diffraction perspective.

[Lou Jost]

Yes, exactly.

Based on the resolution curves of the PN105 on coinimaging.com, I think a very slight bit of extension beyond 1:1 would also give good results when magnified by the teleconverter. The graphs show that center resolution increases slightly beyond 1:1 while the corner resolution drops suddenly. On an MFT sensor the corners will hold up a little longer than in the coinimaging.com tests as extension is added, so maybe I could get to something like m=1.2 by extension and then add the 1.4x converter to reach m=1.7. Probably not worth going beyond that.

[RobertOToole]

Do you know of any of the PN 105 A that are already configured for DSLR use . . . ?

If a Printing Nikkor 105mm is not already configured for DSLR use, making it so is not difficult. Start with Rafcamera's eBay part 142382208066, add a 52mm-thread extension tube, then a reversing ring for your camera mount.

Back in 2012, before the Rafcamera adapter was available, Ray Parkhurst helped a few of us get custom adapters made by his machinist. These adapters were (and still are) very nice, and are the equivalent of the Rafcamera adapter part plus a length of 52mm reversing ring, though as a single piece. If Ray has any of these left, or if his machinist cares to make more, I'd highly recommend them. But with the Rafcamera part readily available off-the-shelf (and given the good quality of Rafcamera items), going with the Rafcamera part seems like an easy and very workable solution.

--Chris S.

I have a couple of these adapters still around, but I had them made for the 95PN in order to mount it in 1:2 configuration. For the 105PN, or the 95PN in 2:1 configuration, I highly recommend Raf's M45-M42 adapter:

http://www.ebay.com/itm/M45x0-75-female ... Sw42JZFfRU

There is a trick to make this work...the front trim ring, which has m43 filter threads, is removable, and reveals an M45 thread that allows the Raf adapter to attach.

Remember, the 105PN is optically symmetrical, so it doesn't matter which way you mount it. I've been using mine with the Raf adapter and it works great.

Ray didn't you mention once that the Printing nikkor shares threads with the Rodenstock 45mm modular mount and extension tubes? I am just curious since I have a few of those laying around so I might keep them in case I ever buy a PN lens.

Thanks

Robert

[ray_parkhurst]

Ray didn't you mention once that the Printing nikkor shares threads with the Rodenstock 45mm modular mount and extension tubes? I am just curious since I have a few of those laying around so I might keep them in case I ever buy a PN lens.

Thanks

Robert

Indeed it does, but the 105PN can be mounted only in the reverse direction, with the front trim ring removed. This is not a problem since the lens is symmetrical. The 95PN can mount either direction.

edited to add: I am curious if anyone owns or has even seen a 75PN?