Apo-Ronar shims--one, two, or none?

[Drew Wiley]

Pere - why are you deferring to very superficial web banks of lens data when Rodenstock has already published resolution graphs that factor in tangential as well as axial performance figures as well as how this behaves at different spheres of the image circle? What good is a mere axis reading in view camera work? I have seen Sinar components specially configured for high resolution diagnostic work. But a lot goes into that. All the tolerances have to be tightened to the point it would make them unrealistic for ordinary photography. Then you'd have to precisely collimate the film and lens nodal plane using a series of front surface mirrors etc (I won't go into detail), and finally you'd need some kind of very dimensionally stable beam mount (I know how to make em). Then, of course, you'd either have a very precise vacuum holder or a means to measure the aerial image at multiple positions. And having looked at some of those Perez etc results, I too am convinced that they're often worth LESS than nothing because they're so misleading. The characteristics of Apo Ronars are pretty well understood. For your own objective of just determining shim usage, adapting a DLSR to the rear position in a carefully leveled Norma would probably be sufficient in a practical sense, then comparing the result at different distances from near to infinity. But quantifying that sort of thing is a different story. And it's an utter myth that all view lenses are optimized at f/22. It all depends. A lot of us like to stop down 4x5 work to f/22 or f/32 not only for sake of depth of field, but also to minimize the effect of the film plane inevitably being a little off in the holder. With graphics lenses, the question includes the ability to precisely apochromatically align dots in all three or four color separations clear out to the corners of the recommended field of usage. A copy camera routinely used flat vacuum targets for sake of precision at that end. In that particular industry, f/22 was a convention related to their own kind of application. Ordinary photographic applications are rarely that nitpicky. And as per actual optical quality, as an example, my own Apo Nikkor barrel lenses are superbly sharp from f/11 down - more sharp and better apo corrected than any of the high-reputation regular view camera lenses I own. But all of this is overkill talk. Sometimes convenience is more important, as well as the specific "character"of image rendering you are seeking. I think you're making a mountain out of a mole hill. Just go out and enjoy your lens.

[Pere Casals]

Pere - why are you deferring to very superficial web banks of lens data when Rodenstock has already published resolution graphs....

You know, there was a sample to sample variation in the production run. Manufacturer graphs show average MTF at low cycles because in that case they can provide consistent information, but peak performance at extintion were never guaranted. They had some product segmentation, the same SCH product could be sold under 4 brands: Linhof, Sinar, Schneider and Calumet, prices were different in each segment, all lenses completely capable but perhaps not the same QC requirements for the Technika or the Calumet stamps, we can guess that from discounts .

All the tolerances have to be tightened to the point it would make them unrealistic for ordinary photography. Then you'd have to precisely collimate the film and lens nodal plane using a series of front surface mirrors etc (I won't go into detail)....

Drew, even me I can calculate the accurate MTF graph and the extintion cycles with high precision, the "collimation" (aligment in this case) it's very easy to do, you don't require any sort of sofisticated equipment to make very accurate measurements, you only need the knowledge and refined procedures.

And having looked at some of those Perez etc results, I too am convinced that they're often worth LESS than nothing because they're so misleading.

I simply don't agree, it is a test with limitations and some flaws, just read the disclaimer at the bottom of the document, but if understanding the limitations then that test is really a great source of information, see this example:

This is a Symmar 150 convertible:
f/11 42 48 38
f/16 42 48 42
f/22 48 64 42

This is the 265 conversion of the same lens:
f/16 48 48 23
f/22 48 48 33
f/32 42 42 38

This is the 265 conversion with a #15 yellow filter:
f/16 48 48 21
f/22 48 48 32
f/32 30 38 33

This shows a pretty nice consistence in the measurements, it also shows what happens with the conversion in the corners, and it tells that a filter (removing possible chromatic aberration) does not improve the result, pointing that the flaws of the conversion come from the spheric aberration. And the conversion is not bad at all ! Probably the values in blue are inconsistent, because the the filter addition has only an effect at f/32, but if you look at the "general picture" this great information.

Very interesting, isn't it?

It is possible that some Pérez measurements were flawed and the lens was better, but never worse than what he says, because to say a number he had to see a certain Group/Element. It has to be pointed that Mr Perez made his career in a a prestigious instrumentation company (Tektronix, IIRC), we may guess he knows what accuracy and methodology are.

The characteristics of Apo Ronars are pretty well understood.

Yes, a narrow angle around 48 degrees only, 4 elements only, first-class for all formats whose long side is not larger than half the focal length. Light, compact, affordable and superior in sharpness and similar coverage angle to tele lenses, but requires more bellows than teles....

A lot of us like to stop down 4x5 work to f/22 or f/32

I shooting a lot f/32 we don't need expensive lenses, diffraction limit is 50, so in practice around 45.

Ordinary photographic applications are rarely that nitpicky.

True... in 3D scenes with near subjects the circle of confusion grows easily in the DOF. A distant shot may exploit better the lens performance, another thing is needing that performance overkill or not...

Apo Nikkor barrel lenses are superbly sharp from f/11 down

It is easy making a good lens with with a low angle of coverage (37º to 46º). If the corrections have cover 70º or 100º then we require more elements and a refined design.

[Bob Salomon]

Pere - why are you deferring to very superficial web banks of lens data when Rodenstock has already published resolution graphs that factor in tangential as well as axial performance figures as well as how this behaves at different spheres of the image circle? What good is a mere axis reading in view camera work? I have seen Sinar components specially configured for high resolution diagnostic work. But a lot goes into that. All the tolerances have to be tightened to the point it would make them unrealistic for ordinary photography. Then you'd have to precisely collimate the film and lens nodal plane using a series of front surface mirrors etc (I won't go into detail), and finally you'd need some kind of very dimensionally stable beam mount (I know how to make em). Then, of course, you'd either have a very precise vacuum holder or a means to measure the aerial image at multiple positions. And having looked at some of those Perez etc results, I too am convinced that they're often worth LESS than nothing because they're so misleading. The characteristics of Apo Ronars are pretty well understood. For your own objective of just determining shim usage, adapting a DLSR to the rear position in a carefully leveled Norma would probably be sufficient in a practical sense, then comparing the result at different distances from near to infinity. But quantifying that sort of thing is a different story. And it's an utter myth that all view lenses are optimized at f/22. It all depends. A lot of us like to stop down 4x5 work to f/22 or f/32 not only for sake of depth of field, but also to minimize the effect of the film plane inevitably being a little off in the holder. With graphics lenses, the question includes the ability to precisely apochromatically align dots in all three or four color separations clear out to the corners of the recommended field of usage. A copy camera routinely used flat vacuum targets for sake of precision at that end. In that particular industry, f/22 was a convention related to their own kind of application. Ordinary photographic applications are rarely that nitpicky. And as per actual optical quality, as an example, my own Apo Nikkor barrel lenses are superbly sharp from f/11 down - more sharp and better apo corrected than any of the high-reputation regular view camera lenses I own. But all of this is overkill talk. Sometimes convenience is more important, as well as the specific "character"of image rendering you are seeking. I think you're making a mountain out of a mole hill. Just go out and enjoy your lens.

Right on!!!

[Pere Casals]

!!!

[Drew Wiley]

Pere, ordinarily if an Artar or Apo-Ronar was FACTORY fitted in a conventional shutter like a Copal or Comupur 3 it was intended for general-purpose shooting and will probably be excellent at infinity as is. If it is in barrel only and meant for something like a Compur electronic shutter, then you should just ask Rodenstock or Schneider how to proceed. In all such cases angle of coverage figures can be misleading. No, you won't get as big an angle of coverage as with a plasmat of the same focal length. But the published specs tend to be very conservative because they had repro usage in mind; and besides, in such applications the image circle often became significantly bigger due to close usage, possibly right up to 1:1 magnification. And although a lot of process lenses are on the used market rather affordably at the moment, that certainly doesn't mean they were cheaper than plasmats when new. And as far as Mr Perez goes, why should I trust anyone who does not describe his methodology up front? I'm sure not gonna guess at what he might or might not have done. What I do know, is that, having used certain lenses he lists and knowing their performance quite well, some of his results just don't make sense. Others have noted the same thing. And I'm a bit skeptical about the notion of there being significant lens to lens variation with modern expensive lenses from the "big four" manufacturers. That might have been true at one time, but there are no reasons to suspect it now. The symmetrical elements of Dagors had to be hand tuned; but no dagors have been made for for large format for quite awhile except for a rare Schneider intended for ULF use. Who knows what vintage of lenses Perez tested tended to be. Therefore I place that kind of information in the same category as Ken Rockwell - take it with a grain of salt.

[Bob Salomon]

Most of the lenses, if not all, tested by Perez were not new!

[CreationBear]

Pere, ordinarily if an Artar or Apo-Ronar was FACTORY fitted in a conventional shutter like a Copal or Comupur 3 it was intended for general-purpose shooting and will probably be excellent at infinity as is.

OP, here....[waves hands over head]--Drew actually raises a question I had earlier: is there any way to tell if an in-shutter lens came from the factory that way (versus having been retrofitted into a modern shutter by some owner down the line?) Coincidently, one of the larger lens vendors currently has an Apo-Ronar 360/9 that's described being factory installed in a Copal 3--whence this confidence?

And, finally, if for some reason I decided that the Apo-Ronar I've bought needs a different set of shims, what resources are available to rectify the issue? Would Rodenstock still have shims on-hand, or would I be looking at a custom-machining job?

[Bob Salomon]

OP, here....[waves hands over head]--Drew actually raises a question I had earlier: is there any way to tell if an in-shutter lens came from the factory that way (versus having been retrofitted into a modern shutter by some owner down the line?) Coincidently, one of the larger lens vendors currently has an Apo-Ronar 360/9 that's described being factory installed in a Copal 3--whence this confidence?

And, finally, if for some reason I decided that the Apo-Ronar I've bought needs a different set of shims, what resources are available to rectify the issue? Would Rodenstock still have shims on-hand, or would I be looking at a custom-machining job?

Rodenstock has shims but they come in different thicknesses. If you really want the lens to be adjusted to factory specs then you would need to send it to them.

Contact Dieter Wenzel here:

http://www.rodenstock-photo.com/service/?lang=en

[Pere Casals]

Pere, ordinarily if an Artar or Apo-Ronar was FACTORY fitted in a conventional shutter like a Copal or Comupur 3 it was intended for general-purpose shooting and will probably be excellent at infinity as is.

Let me explain how IMHO a bare amateur may tuneup perfectly an Apo-Ronar like if it was "FACTORY fitted" for distant subjects:


https://www.ebay.es/itm/WF30X-30mm-B...kAAOSwtM9cdq~W

$25, free shipping... x30


You place that in the camera back, you may use a lens board for it substituting the GG+Graflock of the monorail, with the APO Ronar in the front: ready for action. Then print a target (http://www.takinami.com/yoshihiko/ph..._test/USAF.pdf) and place it far, some 8m to 15m is excellent for a low quality print of the target in the link, depending on the focal and what the printer/paper was able (that pdf has group 7 !!!).

Then unscreew the Ronar's front cell in half turns and focus again, repeat while the thing improves. You are done. Now screew the lens by counting the tours, your shim is the tours multiplied by the thread pitch. You can also repeat for the mid/corner to balance the tuneup.

If our pocket is in shape ($200) we may buy a better ocular, not necessary, but nice:



Or used/good/cheap ($35) microscope/telescope oculars, a x50 would be great for MF and 35mm lenses.


____________________________________________


With that simple gear we may also rate easily any LF lens, without exposing film or making dslr shots: No ALIGNMENT problems, you focus again for each spot in the circle, you navigate in the circle with rise/shift and focus again. Of course you multiply the lp/mm reading (tables here: https://en.wikipedia.org/wiki/1951_U...ion_test_chart) by the magnification, so the single thing you have to calculate accurately is the magification, which is pretty easy, just placing again the GG and using a cheap digital caliper on it....


This kind of measurement is said "in the air", it's the perfect way to compare lenses because we remove the film flatness miss, film resolving power influence, alignments, etc, etc, etc..


This is a simple way in what a bare amateur may make very precise optical ratings at extintion.

Those ratings do not include the filed curvature, wich is irrelevant, but if wanting to measure the field curvature an amateur also has simple ways to do it: You substitute the GG by a clear class, it can be anti-reflective True-Vue, then you align the camera (max aperture) until all corners are perfectly sharp (we need a flat wall) seen with the ocular, then the center may not be in perfect focus, by measuring the camera (tripod) displacement for a perfect center focus we measured the flatness miss, or by measuring the additional bellows draw we have the reading for the film plane so we can calculate the impact in the field.

Substituting the GG by a flat clear glass is also a way to explore the circle with the ocular, instead removing the graflock+GG and replacing it with lensboard+ocular. The naked ocular works better than the GG+loupe, because the GG grain degradates the image, so with GG we cannot see all performance that the film will see.

________

Measurements in the air are perfect for rating glass, but making practical tests by shooting film (like Pérez-Croell) also has sense because we see the practical effect on film after some degradation factors.

________


Of course modern LF lenses also have a remarkable sample to sample variation !!!! which is lower in the last models but it's there. Another thing is when such variability has a practical impact. At f/32 no impact...

If usually shooting at f/22 and beyond we won't see the sample to sample variation because diffraction masks it, at f/32 even we won't see a difference of a 1950s lens vs a 2000s one, coating apart.

[Bob Salomon]

Suit yourself, you can do it right or your way.