Rangefinder lens disassembled: how to calibrate focus?

[OAPOli]

@mariozelaschi I'm sure you can get it back to working order; consider that it was adjusted during the initial assembly and that you can do the same.

The focus scale is for indication only, only the physical infinity stop is important. And the rangefinder is adjusted after the lens has been collimated.

Fix the focus bracket to the lens at the position shown in your picture. Place a piece of glass (35mm tall) on the film rails, preferably with two pieces of scotch tape on the rails to offset the glass slightly. Draw a mark on the glass with a felt pen (lens side), open the lens on B, f/2.8. Hold the glass in place with a rubber band or whatever.

Take a digital camera with a long lens focused to infinity (go outside and find a distant building), place it in front of the Olympus lens. Two tripods work. Rack the focus on the Olympus and inspect the felt pen mark using live-view on the camera. When the mark is the sharpest, the Olympus is at infinity. Then adjust the position of the focus bracket until infinity is at the stop.

Then adjust the RF on a distant building to synchronise it.

 

[mariozelaschi]

@mariozelaschi I'm sure you can get it back to working order; consider that it was adjusted during the initial assembly and that you can do the same.

The focus scale is for indication only, only the physical infinity stop is important. And the rangefinder is adjusted after the lens has been collimated.

Fix the focus bracket to the lens at the position shown in your picture. Place a piece of glass (35mm tall) on the film rails, preferably with two pieces of scotch tape on the rails to offset the glass slightly. Draw a mark on the glass with a felt pen (lens side), open the lens on B, f/2.8. Hold the glass in place with a rubber band or whatever.

Take a digital camera with a long lens focused to infinity (go outside and find a distant building), place it in front of the Olympus lens. Two tripods work. Rack the focus on the Olympus and inspect the felt pen mark using live-view on the camera. When the mark is the sharpest, the Olympus is at infinity. Then adjust the position of the focus bracket until infinity is at the stop.

Then adjust the RF on a distant building to synchronise it.

Thanks! Yesterday I've tried it using another slr with split screen and an 85 f1.8, using some tape. I can try with a piece of frosted glass and the tape to offset.
Just a doubt: is it important the distance between the two cameras? I put them almost in touch with a lens hood to filter out external light.
This is the result I have from my tests (I used a piece of tape, so I'll redo it with glass and more precision)

 

[flavio81]

Thanks! Yesterday I've tried it using another slr with split screen and an 85 f1.8, using some tape. I can try with a piece of frosted glass and the tape to offset.
Just a doubt: is it important the distance between the two cameras? I put them almost in touch with a lens hood to filter out external light.
This is the result I have from my tests (I used a piece of tape, so I'll redo it with glass and more precision)

That "rangefinder" on the focusing screen needs to be looked at in a straigtly orthogonal way. From the looks of the picture, your eye (or camera) is not positioned exactly orthogonal to the focusing screen. This will give you a focus error.

 

[mariozelaschi]

That "rangefinder" on the focusing screen needs to be looked at in a straigtly orthogonal way. From the looks of the picture, your eye (or camera) is not positioned exactly orthogonal to the focusing screen. This will give you a focus error.

To be sure to be orthogonal, then I will have to draw a circle with a cross in the exact middle so I will know that is centered, right? Is a 50mm a good lens to use to do so, or is it better to choose a longer focal length? Also, I am using a analog slr, would be a digital one better?

 

[OAPOli]

@mariozelaschi try to use a clear piece if glass, not frosted. A split-prism rangefinder on a film SLR can get dim in these conditions but it can work. The mark should be perpendicular to the split line for max accuracy.

A digital camera (SLR/mirrorless) would be better because live-view has high magnification. The distance between the lenses isn't critical but getting them close and coaxial is better. Ideally use a lens that's 5x the focal length of the tested lens.

 

[mariozelaschi]

Ok. I will do some tests these days and post results and photos. It might also be useful for other users: there is very little documentation on this procedure, especially from the visual point of view, which for some people is more important than words to visualize the procedure to be done.

 

[OAPOli]

Also, a diffused light source behind the glass + mark is nice. Placing the setup next to a sunlit window works well.

 

[mariozelaschi]

Small update, I did a more precise test and the result is much more convincing to me.
I'm still not sure it's perfect, though: I used matte tape attached to the inner guides with a symbol drawn in marker and aligned the lines using the focus of my SLR, but I'm afraid there might be a slight discrepancy from what the film would be..
I think I'll take a small piece of exposed but blank film (I have a recent roll of film developed in half due to a problem with a machine, I think it's perfect) on which I might draw some lines and use this system:

Camera repair pages

 

[mariozelaschi]

Hello everyone,
I’ve moved forward with a more serious setup, but I’m encountering some unusual results that I’d like to share.

First of all, the setup:​

Here’s the setup using my Konica T4 + 85mm f/1.8, though I later switched to an EOS 5D Mark II + 28-135mm (set at 135mm, manual focus to infinity) to make focusing easier and to capture some photos for comparison. Both cameras are aligned and parallel to the floor.

To adjust the focus in the rangefinder camera, I initially used frosted tape attached to the inner rails but later switched to a piece of blank film with pen-drawn lines.
The film was held in place using a roll of tape and a rubber band. (In the photo, the roll of tape isn’t aligned properly, but I placed it correctly after mounting the camera on the tripod.)

Now, the weird results:​

Let me clarify a couple of points:

• There was no noticeable difference between the frosted tape and the film method; both provided the same results. However, the film made it easier to determine if the image was in focus when looking at the picture.
• Similarly, there was no difference in the focus appearance between the two cameras (analog + 85mm and digital + 135mm).

These are positive signs, as they suggest the system is consistent.

The issue arose when I purchased another secondhand 35 RC, which the seller claimed was in flawless working condition.
I removed the ISO ring to expose the lens mount (without unmounting the lens this time!! ) and confirmed that the factory seals on the screws were intact (so the camera's focus is still the factory one).

When I tested the infinity focus on this untouched camera, it turned out to be slightly out of focus!

Here’s the comparison:

• On the left: 2 images from my old camera (focus adjusted using the tape and film methods).
• On the right: the image from the untouched, factory-sealed camera using film.

The focus is noticeably sharper on the 2 images in the left (adjusted lens by me) compared to the most right image (untouched factory setup)!

Upon closer inspection of the modified camera, I noticed that the original factory screw position was approximately 1mm (or slightly less) above the position I set for optimal focus.
Moving the lens back to match the factory screw marks resulted in less sharpness than my adjusted position.

What I was expecting:​

• The untouched camera would have a perfectly sharp infinity focus.
• The screw marks on the modified camera would align with the position I found to achieve optimal focus.

​

What are the two possibilities:​

1. Both cameras are slightly out of focus from the factory.
2. My adjustment method is not completely accurate, as the difference between my focus point and the factory setting is about 1mm (or less).

I strongly believe that number 2 is the most reasonable one...

What might I be missing, and how can I refine my method for better accuracy?

Thanks to all!

 

[OAPOli]

@mariozelaschi When there is film in the camera with the back closed, the film floats between the inner and outer rails, and generally sits *slightly* behind the film rails. It's preferable to adjust the camera with an autocollimator and film in the camera but that's not available to most people.

This is why I mentioned in posts #16 and #26 to offset your reference plane. A piece of scotch tape is 0.05mm and should work well. Try it with the factory-adjusted camera and you will likely find that the lens is collimated to that plane. It will focus past infinity if the reference plane is at the film rails.

 

[koraks]

What might I be missing, and how can I refine my method for better accuracy?

I'd start by verifying the infinity point you set with your method using the method I described earlier. If you get good focus of an actual far-away image, this should be a clue that your method works.

 

[mariozelaschi]

@mariozelaschi When there is film in the camera with the back closed, the film floats between the inner and outer rails, and generally sits *slightly* behind the film rails. It's preferable to adjust the camera with an autocollimator and film in the camera but that's not available to most people.

This is why I mentioned in posts #16 and #26 to offset your reference plane. A piece of scotch tape is 0.05mm and should work well. Try it with the factory-adjusted camera and you will likely find that the lens is collimated to that plane. It will focus past infinity if the reference plane is at the film rails.

I tried to do that right now: I cut a piece of plexiglas so that it would lay on the outer rails, and placed the film correctly in the inner rail. I used the same tape roll method to keep the plexi plate in place on the outer rails. This means I had the exact situation of the real film, and here are the results.

First pic original factory focus, second pic adjusted focus:

@koraks Having changed the focus calibration of the lens, the rangefinder should also be recalibrated, as it follows the focus of the lens with a pin attached to the metal mount of the lens itself.

 

[mshchem]

When you're finished you'll have the sharpest 35 RC on the planet

 

[OAPOli]

I tried to do that right now: I cut a piece of plexiglas so that it would lay on the outer rails, and placed the film correctly in the inner rail. I used the same tape roll method to keep the plexi plate in place on the outer rails. This means I had the exact situation of the real film, and here are the results.

First pic original factory focus, second pic adjusted focus:

This is not what I suggested. Normally film would be tensioned between the cassette and take-up spool; developed film is a little bit stiffer than undevelopped film.

Try with a physical offset and a rigid surface for the mark. I've done it on the same camera model and it matched the results achieved with an autocollimator.

 

[baachitraka]

If you have any focusing screen say Mamiya RB then it can remove all the errors if you place the screen on the film gate.

*Take this suggestion as pinch of salt.

 

[Tel]

The difference between your adjusted camera and the “factory” setting may well be explained by 55 years of use and abuse. I’d hardly expect a camera from the 1970s to have kept its factory settings unless no one ever used it and it was never subjected to temperature or humidity changes. To me, it’s not surprising that your recently collimated camera exceeds the sharpness of the other one.

And Koraks is right: put a roll of film in it and see how it performs in the real world. That is, ultimately, the test that matters.

When you're finished you'll have the sharpest 35 RC on the planet

+1

 

[koraks]

@koraks Having changed the focus calibration of the lens, the rangefinder should also be recalibrated, as it follows the focus of the lens with a pin attached to the metal mount of the lens itself.

Sure, but ignoring the rangefinder for a bit: if you calibrate the focus on the lens to infinity using the camera-as-a-collimator approach, the lens set at the same focus point should then correctly focus an actual far-away image on the film plane. So even without addressing any issues with the rangefinder, you can verify this way if both approaches to set infinity focus correspond to each other. Once they do, you can then start to fix the rangefinder if it happens to be out of calibration as well.

And Koraks is right: put a roll of film in it and see how it performs in the real world. That is, ultimately, the test that matters.

Testing with film is great of course, but kind of slow. Simply aiming the camera as it is now (on a tripod, back open, makeshift ground glass/tape on the film gate) at a far-away scene and observing the projected image with a loupe should also give a good impression of proper infinity focus.

 

[mariozelaschi]

Sure, but ignoring the rangefinder for a bit: if you calibrate the focus on the lens to infinity using the camera-as-a-collimator approach, the lens set at the same focus point should then correctly focus an actual far-away image on the film plane. So even without addressing any issues with the rangefinder, you can verify this way if both approaches to set infinity focus correspond to each other. Once they do, you can then start to fix the rangefinder if it happens to be out of calibration as well.



Testing with film is great of course, but kind of slow. Simply aiming the camera as it is now (on a tripod, back open, makeshift ground glass/tape on the film gate) at a far-away scene and observing the projected image with a loupe should also give a good impression of proper infinity focus.

This may be a bit funny, but I don't have a magnifying glass. Which one can I buy? Or, can I use a digital camera with a macro lens?

 

[koraks]

Or, can I use a digital camera with a macro lens?

Yeah, that would work.
I use some lens elements I robbed from a broken rangefinder camera.

 

[brbo]

The difference between your adjusted camera and the “factory” setting may well be explained by 55 years of use and abuse. I’d hardly expect a camera from the 1970s to have kept its factory settings unless no one ever used it and it was never subjected to temperature or humidity changes. To me, it’s not surprising that your recently collimated camera exceeds the sharpness of the other one.

Exactly. I wouldn't assume an unopened camera to automatically be a good reference. It might be, but also might be not.

 

[MattKing]

A 50mm f/1.8 or so lens for a 70s - 80s era SLR makes a good magnifying glass

 

[mariozelaschi]

Hi everyone,
First of all, a huge thank you for all the answers, suggestions, advice, and support. This community is incredible!

I've been thinking (and overthinking) about my situation, and it feels so strange that two cameras are out of focus in the exact same way and to the same degree. One of them was adjusted by me, while the other still has the factory seals on the screws—so I'm confident nothing shifted there. Sure, rangefinders can definitely go out of focus, but the lenses themselves? That's much rarer, especially when we're talking about two cameras with identical focus issues.

Then, like a lightbulb moment, it hit me: why not test this with one of my other cameras that I know works perfectly?
So, I grabbed my trusty Konica TC with the wonderful Hexanon 50mm f/1.7 (sharp as a razor), set it to infinity focus at maximum aperture, and guess what? It's also out of focus.

This clearly means the problem isn’t with the lenses but with my homemade collimator. I’m doing something wrong, but I can’t quite figure out what.

The setup I’ve built mimics the closed cover perfectly. The film is pushed against the inner rails and slides smoothly without any issues.

Does anyone have any ideas about what might be going wrong? I know, I know—tomorrow I’ll take this outside (I live in Berlin, Germany, and it’s -1°C right now, but hey, science demands sacrifices!) and do a live focus test by pointing at distant objects. I have a 50mm and 28mm lenses that will be perfect for a loupe.

But in the meantime, I’d really appreciate any insights into what I might be overlooking here.

Thanks again for your help!

The out of focus Hexanon 50mm (that, instead, is razor sharp and perfectly in focus). PS: it appears out of focus also using another analog SLR as the collimator camera.

Here's my contraption with the marked film on the emulsion side.

 

[OAPOli]

This clearly means the problem isn’t with the lenses but with my homemade collimator. I’m doing something wrong, but I can’t quite figure out what.

The setup I’ve built mimics the closed cover perfectly. The film is pushed against the inner rails and slides smoothly without any issues.

Does anyone have any ideas about what might be going wrong? I know, I know—tomorrow I’ll take this outside (I live in Berlin, Germany, and it’s -1°C right now, but hey, science demands sacrifices!) and do a live focus test by pointing at distant objects. I have a 50mm and 28mm lenses that will be perfect for a loupe.

But in the meantime, I’d really appreciate any insights into what I might be overlooking here.

Normally film would be tensioned between the cassette and take-up spool; developed film is a little bit stiffer than undevelopped film.

Try with a physical offset and a rigid surface for the mark. I've done it on the same camera model and it matched the results achieved with an autocollimator.

It's preferable to adjust the camera with an autocollimator and film in the camera but that's not available to most people.

This is why I mentioned in posts #16 and #26 to offset your reference plane. A piece of scotch tape is 0.05mm and should work well. Try it with the factory-adjusted camera and you will likely find that the lens is collimated to that plane. It will focus past infinity if the reference plane is at the film rails.

 

[koraks]

@OAPOli I highly doubt the 0.05mm offset is relevant. I've recalibrated focus on old cameras plenty of times without this added distance. It always ended up perfectly fine.

 

[baachitraka]

I would put a Mamiya RB focusing screen without film or anything in between and just focus anything at infinity or perhaps anything at distance >100 m.

That is all.

But if you want to fine tune, you can put a 50mm lens on the focusing screen to obtain magnification. Other methods work too...