boojum
Ignoble Miscreant
I think the biggest problem in discussing Jupiters, specifically J8's in my case, is that we are not talking about one lens but a family of lenses from different places. And then there is the QC problem. The old Soviet joke about, "We pretend to work, they pretend to pay us." comes into play here. Unmotivated work forces cannot be always depended upon to turn out high quality products. The workers did not always feel valued. But we all know there are some diamonds in the coal heap. I believe cautious shopping can raise the chances of getting a good lens. And when you get a good J8 you've gotten yourself a really fine lens with superb color and resolution. It is as if the old Nazi goal of putting a German head on a Russian body bore fruit. Fortunately not as The Third Reich envisioned it.
I post these not as examples of my prowess, little as it is, but of the prowess of the lens.
I post these not as examples of my prowess, little as it is, but of the prowess of the lens.
The articles on adjusting the Jupiter-3 shim and focal length are mine, as well as converting a Contax mount Sonnar to Leica mount.
The type of glass used is a major factor in a lens. Refractive index and dispersion are critical, and are properties of specific formulations of glass.
The KMZ v1 Jupiter-3 is similar in performance to a wartime Sonnar 5cm F1.5. Including my ZK 5cm F1.5, I currently have nine KMZ version 1 Jupiter=3 lenses and five KMZ v2 lenses. Eight v1 with perfect glass and Four v2 with perfect glass. The v1 has better flatness of field and edge-to-edge sharpness. The wartime Sonnars and v1 Jupiter-3 perform better than the Nikkor 5cm F1.5 in terms of edge-to-edge sharpness. In terms of edge-to-edge sharpness and flatness of field, the v1 KMZ Jupiter-3 performs better than the much newer Jupiter-3+.
Nikon was known for having tight tolerances, and also for optimizing a lens for close-up/wide-open performance. The deviation in focal length of the Nikon lenses is low.
dexdog
Veteran
Wow! I did not think it possible but I actually tied Sonnar Brian for J-3s, except I have 4 ZKs and 5 KMZ Jupiters.
You have me beat on ZK's! Maybe Sonnars as well.
I have a total of Twenty-Two Jupiter-3's ready-to-use, that is including the two J-3+. About 6 more without mounts in good enough condition to convert to Contax mount using the spares from the LTM Sonnar conversions. Twelve 5cm F1.5 Sonnars, Ten in LTM. Because I converted Six of those to LTM.
Back when I was Adjusting J-3's for Leica mount, I always tested with each and every lens on film or on digital. I've shot with a couple hundred J-3's. The ones with German Glass, ZK and v1 KMZ: absolutely the best, and very close to the best wartime Sonnar 5cm F1.5. The J-3 LTM focus mount is much sturdier than the Zeiss LTM focus mount. SO: Given the price of wartime Sonnars, the KMZ J-3 v1 is a bargain and just as good.
Remember that focus shift occurs with changing Aperture and with deep color filters. I have J-3's shimmed to work with Orange filters on the M Monochrom, which requires a slightly thicker shim. Focus shifts towards infinity. The same lens on color needs to be stopped down to F2. Rather than labeling them- I just leave the filter on them.
dexdog
Veteran
You have lot more Sonnars than I do, except I have 7 original f1.5 wartime Sonnars in LTM. Not even close in other categories
You need to do a thread shooting with the Seven 5cm F1.5 Sonnars. What I've seen on the ones taken apart: "slight differences" in construction, little things like some with a hidden set screw holding the namering in place, others skip that, some with set screws for the rear fixture in the barrel- others, skip it. Coatings- some seem to have been rushed, not as durable. Things must have been rushed during the war.
flavio81
Well-known
This is great information Brian, many thanks, and thanks for your PDFs, i do appreciate them.
Sacrilege!!!! lol
Thankyou. I plan on doing more articles on the Sonnars and Jupiters and have collected a lot of photos of taking them apart- "when I retire", which is not long off.
Agree that it can be sacrilege... But the conversions are easily reversed. I keep the Contax Mount with original shims for the most rare of the Sonnars, keep the SN written inside the mount. "Back in the day" when the Sonnars were new, some were converted to Leica Mount by the better repair shops. I've handled several of them, not converted using an adapter- but custom brass focus mounts made for them. Leica had the Xenon 5cm F1.5, Zeiss had the Sonnar. Leica was much smaller and less expensive than the Contax.
Sonnar_1936_2 by fiftyonepointsix, on FlickrMy uncoated/ beautiful bloom 1936 5cm F1.5 Sonnar, housed in the J-3 LTM focus mount. Wide-Open on the M Monochrom.
BUT... using the Sonnars in a really good J-3 mount, on the digital Leica is "so Smooth". I have Four Amedeo adapters, use them with S-Mount Nikkors and post-war Sonnars. The J-3 mount with the heavier pre-war Sonnar has some weight to it, and is much smoother/more accurate than the wartime LTM Mounts. I have one wartime LTM mount that had so much wobble that I made a sleeve inside the mount for the RF cam to stay straight as you focus. I also use the Amedeo adapter for the First Batch 5cm F1.5 Sonnar. Dexdog had the very first 5cm F1.5 Sonnar, donated it back to Zeiss. Mine is the 68th made, a batch of 100 pre-production lenses.
nzeeman
Well-known
They do. You're just not looking hard enough.
I got tired of hearing unsubstantiated claims from both sides of the fence, so I did a controlled test, putting three different lenses (Jupiter 8, Industar 22 and Industar 26-M) on a Leica IIIf, positioned exactly 1m from film plane to the name plate of a FED 2, and used the rangefinder to focus on the name plate instead of using the distance scale on the lens (this is important - I'll come back to this). All of the lenses were untampered with by me and had come from three very different sources. A Summicron was used as a control to show that the Leica was properly calibrated and could focus accurately. Then I wet-printed the negatives to remove any possible negative scanning issues and scanned the prints. Here's the results:
Soviet Lens Tests, Redux. on Flickr
For every FSU lens, the rangefinder has back-focused to the point that the Kiev positioned behind the FED is more in focus than the FED is.
Soviet lenses - assuming no modification either before or after purchase, back-focus on a Leica by a consistent degree. It is predictable, measureable, and provable both in testing like this and through simply checking the lens' distance reading on close focus shots - due to the different standards that Soviet lenses were based on, focusing on something 1m away with a Leica will result in the lens showing something closer to 1.1m or 1.2m on the distance scale. The same happens in reverse; a Leica lens on a FED (that hasn't been adjusted) will front-focus for the same reason. Try it for yourself - it's an easy test to do.
On further distances and stopped down, it's less obvious. Depth of field can cover it, and most people aren't that picky anyway. But the difference does exist.
so rangefinder is showing wrong? if you put object on 1m from film plane and put lens on 1m it will be ok? then i dont see a problem with russian lenses - its a problem with rf being not compatible...
nzeeman
Well-known
Attachments
The LTM Leica is 51.6mm standard. The M-Mount rangefinder is calibrated for this nominal focal length.
M-Mount lenses use an indexed cam to translate the focal length to the cam follower. The lock on the M-Mount camera sets the position of the lens in the mount precisely to allow the use of an indexed cam. LTM lenses may come up slightly off-center, which would preclude the use of an indexed cam. The Nikon wide-angle lenses, and some others, in LTM use an indexed cam- but require absolute precision in the starting point of the threads.
Leica and Zeiss built their lenses to different standards, the FSU lenses follow the Zeiss standard. Nikon lenses follow the Leica standard.
mapgraphs
Established
Brian, thanks for posting the data sheets!
I'm taking the 28.8mm as the flange registration distance. If so, it helps explain something I've encountered where I was speculating that my FSU lenses generally seem to have a longer registration distance than my Leica lenses. Dante Stella quotes 27.8mm as a Barnak registration distance (27.95mm for the M) (http://dantestella.com/technical/flange.html). Using a J12 on a Barnak, M-240 and a mirrorless body with adapter for instance, the lens doesn't focus to infinity at the hard stop, it's just short. My guess was that there was a difference of about 1mm in registration. Some of my other FSU lenses seem to do the same. So, slightly off, maybe more or less with manufacturing tolerance variation...
I'm taking the 28.8mm as the flange registration distance. If so, it helps explain something I've encountered where I was speculating that my FSU lenses generally seem to have a longer registration distance than my Leica lenses. Dante Stella quotes 27.8mm as a Barnak registration distance (27.95mm for the M) (http://dantestella.com/technical/flange.html). Using a J12 on a Barnak, M-240 and a mirrorless body with adapter for instance, the lens doesn't focus to infinity at the hard stop, it's just short. My guess was that there was a difference of about 1mm in registration. Some of my other FSU lenses seem to do the same. So, slightly off, maybe more or less with manufacturing tolerance variation...
I've bought "new old Stock" Jupiters that needed to be adjusted up to 0.5mm thicker shims. They were way off. I had one Jupiter-12 that required an extra 0.3mm shim for the Leica.
I like those lenses: they are typically in better condition than ones that were perfectly adjusted from the factory. No one could use them.
I like those lenses: they are typically in better condition than ones that were perfectly adjusted from the factory. No one could use them.
Coldkennels
Barnack-toting Brit.
That's exactly the point I'm making. Use the Soviet lens on a Soviet body (or on a modern mirrorless with through-the-lens focusing) and they're fine. Good, even.
The problem is when you take them and use them on a rangefinder that they aren't compatible with (i.e. a non-Soviet body, assuming the lens hasn't been modified).
You can, of course, use an external/accessory rangefinder and go about things that way. Good luck getting precise focus at 1m with a Jupiter 3 wide open, though.
We had a long discussion on RFF back in 2009 on the subject of using an Indexed Cam to translate the focus from a 52.4mm Zeiss Contax lens to Leica M-Mount calibrated for 51.6mm. Amedeo was part of that discussion, I posted pictures of the Summicron 50/2 with the Indexed Cam used on my Leica lens.
Amedeo brought out his Contax to Leica M-Mount adapter for internal mount lenses after this, they are absolutely perfect across range from 0.7m to infinity.
This image shows the indexed cam that Amedeo uses. This is my adapter, and I've used it with several Sonnars. I also have the Nikon S-Mount to Leica version of this adapter, which does not have the indexed cam.
The "small cut" shown on the index is about what I add to a Jupiter-3 and Jupiter-8 to get the focus optimized for close-up and wide-open for the Leica. Focus shift as you stop down shifts focus towards infinity. This works well 90% of the time, shooting infinity by F2.8.
I should start marking the lenses that I calibrated for the Leica. I do this for the Contax lenses and Jupiters that I calibrate for Nikon RF's. A little "N" on the mount.
Someday, when I'm gone, someone will fall into a lens that has been modified to shoot on a Leica, by me, and think it came from the factory like this.
I spent 8 hours on this one. Focus good from infinity to close-up, wide-open.
On the Jupiter-8 with rotating barrel: you could dedicate an LTM to M-Mount adapter to it, and cut an indexed cam like that shown in the Amedeo adapter. I use a custom indexed cam in an external mount Contax to Leica M adapter for my pre-war 8.5cm F2 Sonnar, focus is perfect across range.
The M-Mount allows for use of an Indexed Cam, and looser tolerance in actual focal length of each lens. Deviations could be corrected mechanically. My Summicron 50/2 Type 2 was marked 51.9mm internally, and had a slight index to the cam.
I currently own Sixteen Jupiter-3 lenses in LTM, from 1949 ZK through to two Jupiter-3+ lenses. The fourteen originals- I adjusted all of them to Leica Perfection. All of them needed adjustment. I've worked on a lot of Jupiter-3 lenses, and a lot of Jupiter-8 lenses. I've worked on lots of Leica Mount thread mount 50mm FSU, Japanese and German lenses. None use an indexed cam, all have a 1:1 movement of the optics with the RF cam. FSU lenses are made to the Zeiss standard. If you get one that is perfect on your Leica, it's most likely because it was adjusted for it. I've done that for about 250 Jupiter 50mm lenses.