alexanderspencer
Member
Morning folks,
I recently had a Canon 0.95 converted from its original mount to Leica M.
I was running some other tests on my Leica M6 TTL and noticed that this is the only lens in my kit that is focusing outside the lens' distance range. Seems like a horizontal misalignment occurs.
The lens is supposed to focus from 1m -- infinity (as I'm sure you know). If I want to focus to infinity, focus is achieved just beforehand as seen on the distance scale (shown in image 1), tested by focusing on the moon.
The distance range/scale ring also turns to before the 1m marking (shown in image 2), by the same margin on the scale as when focus is achieved before infinity. Should there be a hard stop at 1m on the 0.95?
What effect will this have to shooting? I'm unlikely to shoot at infinity, however, if I needed to, could I just focus to where I hit infinity and ignore the error on the distance scale?
Many thanks in advance!
I recently had a Canon 0.95 converted from its original mount to Leica M.
I was running some other tests on my Leica M6 TTL and noticed that this is the only lens in my kit that is focusing outside the lens' distance range. Seems like a horizontal misalignment occurs.
The lens is supposed to focus from 1m -- infinity (as I'm sure you know). If I want to focus to infinity, focus is achieved just beforehand as seen on the distance scale (shown in image 1), tested by focusing on the moon.
The distance range/scale ring also turns to before the 1m marking (shown in image 2), by the same margin on the scale as when focus is achieved before infinity. Should there be a hard stop at 1m on the 0.95?
What effect will this have to shooting? I'm unlikely to shoot at infinity, however, if I needed to, could I just focus to where I hit infinity and ignore the error on the distance scale?
Many thanks in advance!
Attachments
Shoot a roll of film, and test it wide-open at min focus as measured by the RF and again at Infinity as measured by the RF.
It sounds like the shim used for the lens is very slightly too thin. This means that actual focus, the RF, and distance scale will not agree. When doing a conversion like this- it is most important to get the actual focus and the RF to agree.
If your camera agrees perfectly with other lenses- then the converted Canon has the issue. If the actual focus and RF disagree, you need to send the camera and the lens to the person that did the conversion.
Hard stop at 1.0m: no, it goes past 1m.
It sounds like the shim used for the lens is very slightly too thin. This means that actual focus, the RF, and distance scale will not agree. When doing a conversion like this- it is most important to get the actual focus and the RF to agree.
If your camera agrees perfectly with other lenses- then the converted Canon has the issue. If the actual focus and RF disagree, you need to send the camera and the lens to the person that did the conversion.
Hard stop at 1.0m: no, it goes past 1m.
raid
Dad Photographer
I would send the lens back to the person who did the lens mount adapting.
I once had a strange focus issue with a Canon 50/1.2 ltm. The focus area changed from left side to right side in each image. DAG took the lens apart and he told me that some glass elements had moved and caused the focus problem.
I once had a strange focus issue with a Canon 50/1.2 ltm. The focus area changed from left side to right side in each image. DAG took the lens apart and he told me that some glass elements had moved and caused the focus problem.
alexanderspencer
Member
Thanks for your responses so far!
Seems that in all likelihood that this will be going back to the serviceman.
I would also like to point out that the rear element protrudes ever so slightly from the base of the M mount when focused to infinity. Is this normal?
Best
Seems that in all likelihood that this will be going back to the serviceman.
I would also like to point out that the rear element protrudes ever so slightly from the base of the M mount when focused to infinity. Is this normal?
Best
raid
Dad Photographer
There are several past threads in such conversions. Going through them may reveal to you some differences seen in your lens.
retinax
Well-known
If the shim is indeed too thin, the scale will be off but actual focus and rf focus will agree. This is the case with all 50mm lenses because the rf feeler feels the actual distance of the lens, i.e. rangefinder cam pitch equals focusing pitch of a (nominal) 50mm. An ltm 50 will also focus correctly according to rf when it's not fully screwed on (it just won't reach infinity).
Nonetheless I'd expect proper alignment of the distance scale, had I shelled out for a conversion.
Mackinaw
Think Different
Mine does. Conversion done by Ken Ruth before he retired. Focus is perfect on my M3 and MP.
Jim B.
alexanderspencer
Member
Sorry, I used the term 'hard stop' to refer to the start of the lens' closest distance. It starts before 1.0m, as shown in image 2. Is this normal? I mention this because the margin between where the focus ring starts and 1.0m on the distance scale is the same margin as shown between where infinity focus is achieved and where the infinity marker actually is on the ring (see image 1).
Last edited:
alexanderspencer
Member
Does this mean that the lens can focus correctly but that the alignment with the distance scale is just a bit off? See above comment.
Many thanks all.
retinax
Well-known
Well, if it's just a spacer issue, then yes
. You need to test accurate focus. But I'd ask the person who converted it to fix it. Even if it's just that, an accurate distance/DOF scale is great to have.
My Canon 50/0.95 has a close-focus less than 1m, it does not stop at the 1M mark. It looks like the same position as yours.
The spacing required to get the Distance scale to infinity is about 0.03mm, at least that is where I would start.
I would like to see pictures taken with the lens on your camera, as mentioned before. If the actual focus vs RF is accurate, then changing the shim and Cam of the lens must both be done for distance scale, RF, and focus to agree.
The spacing required to get the Distance scale to infinity is about 0.03mm, at least that is where I would start.
I would like to see pictures taken with the lens on your camera, as mentioned before. If the actual focus vs RF is accurate, then changing the shim and Cam of the lens must both be done for distance scale, RF, and focus to agree.
Phil_F_NM
Camera hacker
One other possibility is that the helicoid has been reassembled incorrectly, just off by one thread. This would cause the hard stop to hit "early" before infinity, if it is bottoming out at the end of the threads. If the 50/.95 is constructed the same as the 50/1.2 and 50/1.4 (all Canon RF lenses) then the problem is not with a shim to space the optical unit within the focusing unit. You can fit the focusing unit of the lens on the camera and "focus" the 1.2 and 1.4 lenses without the actual glass in the focusing unit.
Looking at your image of the lens, the focusing index mark and the aperture index mark line up (as they should since the optical part of the lens should be keyed to fit only in the focusing stage one way) and the issue is the focusing barrel. Could be just the indicator ring being turned a little bit one way, or it could be more in depth, like the incorrect reassembly I mentioned above.
Either way, return it, IMO.
Phil Forrest
Looking at your image of the lens, the focusing index mark and the aperture index mark line up (as they should since the optical part of the lens should be keyed to fit only in the focusing stage one way) and the issue is the focusing barrel. Could be just the indicator ring being turned a little bit one way, or it could be more in depth, like the incorrect reassembly I mentioned above.
Either way, return it, IMO.
Phil Forrest
peterm1
Veteran
If the repairman disassembled the helicals to service them I would agree with the poster who says this is a possible source of the problem. But the helical may not have been touched given that the conversion you had done does not require this. One thing I am not sure of in your description is whether (at say infinity) the lens distance scale does not reach infinity but the lens is actually in focus at infinity or if you are saying the lens does not reach infinity and the focus is off at infinity. This is an important distinction and the repair man needs to know which it is.
Subject to the above, I would say it is also possible that when undertaking the conversion the repairman got the lens flange distance slightly wrong and it does not correspond to the flange distance it had before conversion. It is also possible that the lens came from the factory wrongly shimmed (as unlikely as this sounds) or that for some reason the shim was swapped out at some stage and a slightly incorrect one installed. I only say this because I grappled with a Canon 50mm f1.2 which had exactly this issue and as it had a factory correct brass shim in it I can only surmise that it was incorrectly installed and never picked up during QA checks. It took my camera tech three tries to get it just right by thinning the shim (mine was forward focusing) but now it is spot on for the rangefinder.
The best way I have found to diagnose exactly the nature and extent of the forward or back focus issue is to set up a target - any small object that you can focus on accurately, at a known distance (when I did it I used 3 metres as this distance is also marked on the lens's distance scale and that helps) and also place a measuring scale (e.g. a tape measure) before and behind the target.
Then using a mirrorless camera and an adapter that is known to be correct, take a few images of the target and then also check to see what the distance scale on the lens says when the target is correctly in focus. You can reverse the process too, by scale focusing (set the distance scale on the lens to 3 metres - if that is where the target is) then see where the lens is ACTUALLY focusing. It should correspond to the distance in front or behind found in the previous test described above.
If you have access to a digital Leica M this can be double checked by then mounting the lens on it and using the rangefinder to see where the lens is actually focusing compared to where the rangefinder "thinks" it is focusing.
You can of course do the same with a film M but it is more difficult for obvious reasons as you lack the instant access to feedback on results. But starting with using a mirrorless camera to do the initial test helps a lot as it gives a pretty reliable indication of the nature and extent of the problem.
You could of course just send it back to the technician but it helps if you can give specific and detailed info about how far forward or backwards the focus is.
Subject to the above, I would say it is also possible that when undertaking the conversion the repairman got the lens flange distance slightly wrong and it does not correspond to the flange distance it had before conversion. It is also possible that the lens came from the factory wrongly shimmed (as unlikely as this sounds) or that for some reason the shim was swapped out at some stage and a slightly incorrect one installed. I only say this because I grappled with a Canon 50mm f1.2 which had exactly this issue and as it had a factory correct brass shim in it I can only surmise that it was incorrectly installed and never picked up during QA checks. It took my camera tech three tries to get it just right by thinning the shim (mine was forward focusing) but now it is spot on for the rangefinder.
The best way I have found to diagnose exactly the nature and extent of the forward or back focus issue is to set up a target - any small object that you can focus on accurately, at a known distance (when I did it I used 3 metres as this distance is also marked on the lens's distance scale and that helps) and also place a measuring scale (e.g. a tape measure) before and behind the target.
Then using a mirrorless camera and an adapter that is known to be correct, take a few images of the target and then also check to see what the distance scale on the lens says when the target is correctly in focus. You can reverse the process too, by scale focusing (set the distance scale on the lens to 3 metres - if that is where the target is) then see where the lens is ACTUALLY focusing. It should correspond to the distance in front or behind found in the previous test described above.
If you have access to a digital Leica M this can be double checked by then mounting the lens on it and using the rangefinder to see where the lens is actually focusing compared to where the rangefinder "thinks" it is focusing.
You can of course do the same with a film M but it is more difficult for obvious reasons as you lack the instant access to feedback on results. But starting with using a mirrorless camera to do the initial test helps a lot as it gives a pretty reliable indication of the nature and extent of the problem.
You could of course just send it back to the technician but it helps if you can give specific and detailed info about how far forward or backwards the focus is.
I read the OP to mean infinity focus as seen through the viewfinder was short of the infinity mark on the distance scale, that the lens would drive the RF beyond infinity. I've seen misassembled helicals- typically the error is larger than this.
The conversion process requires that the original mount be removed and that a new M-Mount be put on the camera. I would use a shim between the base of the lens and the new mount to get the RF cam of the lens to make the distance scale and the RF cam agree, then change the main shim for actual focus to agree.
The conversion process requires that the original mount be removed and that a new M-Mount be put on the camera. I would use a shim between the base of the lens and the new mount to get the RF cam of the lens to make the distance scale and the RF cam agree, then change the main shim for actual focus to agree.
alexanderspencer
Member
That's right. When seen through the viewfinder, infinity is passed but falls short of the infinity mark on the distance scale. It's quite minor (a hair's width or two hairs), but the difference is there and therefore noticeable to my eye.
I've been in touch with the serviceman who is going to take a look at the lens again. However, he did mention that he checked the converted lens for accurate collimation and that it was collimated correctly. I suppose he'll have a look and review.
It would be best of you send him the camera and the lens. I've adjusted a couple hundred lenses for people in the past- there are differences between digital RF cameras, film RF cameras, and even monochrome-digital RF cameras. The Canon 50/0.95 was made for the Canon 7. Assumptions for film and digital are different. The problem with adapting lenses to a different mount- even more complex.
peterm1
Veteran
That's how I would do it as well and for me, on reflection, this failure to shim correctly is probably the main suspect. Fortunately it is not at all hard to fix by any competent tech.
Phil_F_NM
Camera hacker
The OP's issue isn't a shim issue if the construction of the 50mm f/.95 is the same as the f/1.2. The shims have nothing to do with the way the lens mechanically focuses, they only provide proper registration distance for the one-piece optical unit. The focusing cam is not a part of the optical unit, it is built in to the focusing mount. This is for the f/1.2 and f/1.4 versions of this lens.
Taking apart the focusing unit down to individual helicals would be good practice if there was any cutting or filing of the original mount necessary for the modification. Just a single tiny brass filing would jam up the fine threads in the lens helicoid quite easily.
Ensuring the lens focuses to infinity is one thing but ensuring it does so when hitting the hard infinity stop AND matching the indicator mark is how it should be done.
Phil Forrest
Taking apart the focusing unit down to individual helicals would be good practice if there was any cutting or filing of the original mount necessary for the modification. Just a single tiny brass filing would jam up the fine threads in the lens helicoid quite easily.
Ensuring the lens focuses to infinity is one thing but ensuring it does so when hitting the hard infinity stop AND matching the indicator mark is how it should be done.
Phil Forrest
This lens has a new Mount put on it. That mount replaces the original, and probably uses some sort of a shim. If it does not, adding a thin shim between the mount and base of the lens will correct the infinity mark not agreeing with the RF. After that is done, then the actual focus must be brought into agreement with the RF. That means either building up/ or filing down the RF cam of the lens, moving the rear group, or moving the optical block with respect to the focus mount. I'm not taking my Canon 50/0.95 apart to find out. Easiest to build up the RF cam.
The Canon 50mm F1.2 barrel is held in the focus mount using a retaining ring. Stand-off distance uses a shim.
https://johanniels.com/canon-rf-50mm-1-2-cleaning-leica/
On converting the Canon 50/0.95: I left mine in Canon 7 mount. Bought the lens for $200 back in the day when it was proclaimed to be worse than using a Coke bottle for photography. Focus was perfect on my Canon 7, picked up the fitted case for it for $35. With the 50/1.1 Nokton and other super-speed choices for well under $1000, hard to justify converting it. I found a 50/1.2 with perfect glass on a Canon V-T Deluxe for $400 last year, was Christmas 2019 present to myself. 1/2 F-Stop difference, and much the same look as the 50/0.95. Using the latter for film, Canon 7 works well for me. I've converted 4 SLR lenses to M-Mount, could probably do the 50/0.95 myself. Then I would have nothing to put into the fitted case.
Total side note: The Nikon Nikkor-T 10.5cm F4 lens: the hard stop for infinity and actual focus for infinity is at the edge of the infinity mark, not the center. That drove me nuts. Looks like the OP's lens, except it was made that way.
The Canon 50mm F1.2 barrel is held in the focus mount using a retaining ring. Stand-off distance uses a shim.
https://johanniels.com/canon-rf-50mm-1-2-cleaning-leica/
On converting the Canon 50/0.95: I left mine in Canon 7 mount. Bought the lens for $200 back in the day when it was proclaimed to be worse than using a Coke bottle for photography. Focus was perfect on my Canon 7, picked up the fitted case for it for $35. With the 50/1.1 Nokton and other super-speed choices for well under $1000, hard to justify converting it. I found a 50/1.2 with perfect glass on a Canon V-T Deluxe for $400 last year, was Christmas 2019 present to myself. 1/2 F-Stop difference, and much the same look as the 50/0.95. Using the latter for film, Canon 7 works well for me. I've converted 4 SLR lenses to M-Mount, could probably do the 50/0.95 myself. Then I would have nothing to put into the fitted case.
Total side note: The Nikon Nikkor-T 10.5cm F4 lens: the hard stop for infinity and actual focus for infinity is at the edge of the infinity mark, not the center. That drove me nuts. Looks like the OP's lens, except it was made that way.
alexanderspencer
Member
Absolutely!
Just for some added context... when I spoke with my tech guy yesterday he mentioned that all was optically sharp when checking for collimation on his digital Leica M. I did specify that I am using an M6TTL. Might this have made a difference?
I'm tempted to get a second opinion. Any suggestions for UK techs who do this sort of thing? Cameraworks has an insane turnaround of 6months+.