Adjusting Focus Distance on Adapted Lens

[scott]

Hi, all--

I recently got a Voigtlander 15mm f/4.5 lens and an M39-m4/3 adapter for my G1. I'm planning on scale-focusing it in dim conditions where I need maximum depth of field (with a tripod).

According to the online DOF calculator I used, which accounts for sensor size, the lens should produce images that (for example) are in good focus from about 3.25 feet to infinity when the lens is set at f/8 and focused at 2m.

In trying out the lens, I could tell that this wasn't the case. At that setting, distant subjects looked clearly out of focus to me in pictures at full-screen monitor size, which seems roughly similar to a reasonable-size print.

So after several tests, I decided to check to see if the focus scale was accurate. In the attached picture (a 100% crop of a small printout 2m away) the left image was scale-focused at 2m (with the focus plane measured to be 2m away, of course), and it seems poorly focused. I then made the right image focusing carefully by eye, and it seems better. However, the focus scale showed that the lens was focused at a point about halfway between 2m and infinity. (There are no marks between those numbers.)

Now for the question. If the lens is focusing closer than the scale indicates, and given that infinity focus happens with the lens mechanism closest to the sensor, doesn't this mean that I need to move the lens *back* (probably by somehow shortening the adapter)? I've heard of people shimming lenses *out*, but if I understand this correctly, I have the opposite problem.

And if that's right, does anybody have any suggestions about the best way to do this? Any help would be greatly appreciated.

 

[addieleman]

What brand of adapter do you use? Not all brands are accurate. My Fotodiox adapter is about 0.1mm too long, see this thread where others also report about inaccurate adapter length and consequently inaccurate infinity focus. FWIW, I've had no issues with Novoflex adapters, but scale focussing requires an extremely accurately made adapter.

 

[scott]

0.1mm! That's going to be hard to accurately adjust.

Mine is, iirc, from rainbowimaging. I've used their adapters before, but haven't needed to scale focus with those other lenses.

I've looked at the distance the lens moves front/back between focusing at 2m by scale and the point where it actually focuses at ~2m, and it's very small. There's no way I could accurately move the threaded mount that distance, even if it will move front/back (which I don't know). I suppose I could try to sand or grind the mount face bit-by-bit until 2m=2m, but that seems like a cheeseball solution.

 

[addieleman]

Given that your lens has a deviation of a real object distance of 2m and a scale distance of 4m (halfway between 2m and ∞) that would amount to an error of about 0.06mm for the adapter length. Maybe you can unscrew the flange and replace/remove shims, of course only possible if there's a separate flange. With this adapter you will not be able to focus on infinity anyway, but this may not be a problem because the lens is only f/4.5.

 

[scott]

The adapter has two pieces, and has a gap of about 2.5mm between the back of the M39-threaded piece and the m4/3 piece. I can't see any sort of spacer between those two pieces. The M39 ring is held in by three set-screws. I'm assuming these control the orientation of the lens to the top of the camera, but I can't see how they would control front/back distance.

 

[grebeman]

Hi Scott,

You are indeed correct in your assumption as to the purpose of the three set screws on a Leica screw to :43: adapter.

My system for these lenses is to set the lens at infinity and check on a distant subject for focus accuracy, I have usually found that the lens will not fully reach infinity focus.

There are then 2 choices, 1 is to place a thin shim, usually about 3 thousands of an inch thick between the inner and outer rings of the adapter such that the inner ring is moved forwards. Strictly speaking this then makes that adapter adjusted for that particular lens. This shim can be made from the material that is used to make the cylindrical body of aluminium drinks cans. The second method is to remove the collar that encompasses the screw thread from the back of the lens, there are 3 or 4 tiny cross head screws holding it on. Note that there is usually a little dimple that aligns with the white focus spot on the lens body to enable correct positioning on reassembly. Then thin shims (I cut these from aluminium oven cooking foil, about 1 thousands of an inch thick) and insert them behind this collar. There may already be some shims there put in at manufacture. This method means that the lens has now been adjusted to suit the particular adapter and that same adapter can be used if you have other lenses shimmed in this way that need a different thickness of shimming.

Be certain that the inner screwed ring on the adapter is put back in the correct way around, the set screws bite into a v grove on that threaded ring that is not symmetrical, the action of the set screws both lock it and force it against the step in the outer ring, thus pushing it back axially against that step.

Funnily enough it has been my experience with several Voigtlander screw lenses that my copy of the 15mm, f/4.5 was the only one that didn't require shimming.


Barrie

 

[scott]

Hi, Barrie, and thanks for those details--good stuff to know about.

But this is where I think I must be misunderstanding something. Shimming the lens or the face of the adapter outward would push the lens away from the sensor. But I thought that increasing the distance from sensor to lens *decreased* the distance to the in-focus subject. [EDITed for clarity] If that was true, then moving the lens farther out would make its maximum focus farther from infinity than it was before. But now I don't know if I've got this right or not. If somebody could put me straight I'd appreciate it.

 

[scott]

OK, I was all ready to tear into the back of the lens looking for shims when, just in time, I remember my personal DIY rule-set:

1. Stop.
2. Put the expensive, delicate, and/or hard to replace thing away, well out of reach.
3. Breathe.
4. Think. Come up with something less risky.

During (4), I remembered reading that the slot holding the M39 threaded portion is asymmetrical. What if, I thought, somebody put it in backwards?

So step 5 was to remove and reverse the M39 ring. Now when I focus on a target at 2m, the focus scale says just over 1m! So the lens has gone from focusing short to focusing way beyond the infinite. But at least that means I could shim the adapter ring our rather than grinding it down or opening the lens.

Step 5(alt) was to also order a Fotodiox M39-m4/3 adapter. They claim to be able to guarantee infinity focus, so I'll give them a chance. (And return the adapter if the claim isn't true.)

In the meantime, the delicate expensive thing is going back in the case until the other adapter arrives. I hope to have this lens for years, so I can wait a few days in the interest of not destroying.

 

[grebeman]

Hi, Barrie, and thanks for those details--good stuff to know about.

But this is where I think I must be misunderstanding something. Shimming the lens or the face of the adapter outward would push the lens away from the sensor. But I thought that increasing the distance from sensor to lens *decreased* the distance to the in-focus subject. [EDITed for clarity] If that was true, then moving the lens farther out would make its maximum focus farther from infinity than it was before. But now I don't know if I've got this right or not. If somebody could put me straight I'd appreciate it.

Scott,

That's interesting because I also sat down sometime ago and wrote down my own logical argument that the lens to sensor distance needed to be less. I was in a position to machine a little off the rear face of the inner ring to achieve that objective, however when push came to shove it worked the other way, most curious.

Barrie

 

[scott]

Yeah--that's part of why I think I must be missing something.

For now, though, I'm just going to keep the tools locked up and hope that the Fotodiox adapter will be better.

 

[scott]

Well, I'm still waiting for the Fotodiox adapter, but I've had the adapter apart a couple of times. I cleaned it up a bit and put it back together the original way, and then realized that focus would be correct if I used the f/4.5 DOF mark (for 35mm) as the focus mark, since the 2m mark now fell there when I tried to focus on a test sheet. So I put a piece of yellow striping tape on that mark so I wouldn't forget. Time permitting, I'll try to get some test shots to see if I'm getting the DOF I need.

 

[grebeman]

Scott,

Since you are happy to take the adapter apart, why not make a test shim from some brown paper or thick cartridge paper and put it on the flange that the inner screwed ring is pushed against when the adapter is assembled, thus pushing the inner ring forward by the thickness of the shim and see what that does for your infinity focus problem.

Barrie

 

[scott]

Thanks, Barrie--I've been trying to find some good materials to do that with. I tried some aluminum-foil shims, but didn't quite find the right thickness. Plus they not very durable. I may go by the hobby shop and see if they have something a a bit thicker.

Do you make your shims circles (well, doughnuts...)?

 

[scott]

Here's a boring-but-useful test picture with the offset focus mark. This was at f/8 with the focus at 2m, and it looks like everything's in focus:

Subscribe to see EXIF info for this image (if available)

 

[grebeman]

Thanks, Barrie--I've been trying to find some good materials to do that with. I tried some aluminum-foil shims, but didn't quite find the right thickness. Plus they not very durable. I may go by the hobby shop and see if they have something a a bit thicker.

Do you make your shims circles (well, doughnuts...)?

Scott,

Doughnuts, yes thats a good description. I originally made some from brown parcel wrapping paper just to prove the theory. I then made some out of the aluminium body of drinks cans. I found that to be just over 3.5 thousands of an inch thick, very slightly too thick, but not a stopper. It can be cut with a strong DIY type knife, we would know them as "Stanley" knives here in the UK, named after the manufacturer. I later found some 3 thou shim from my model engineering days.

Aluminium foil (presumably the type used to wrap food in for cooking purposes) is indeed too thin for use in the adapter, ours in the UK is 1 thousands of an inch thick, however I have used it when I dismantled the lens and placed it behind the screw mounting flange at the rear of the lens, using two or three layers. In this application it is only subjected to a crushing load as you replace the flange and is perfectly usable. Since the fixing screws go through it I used segments that fitted between the screw locations rather than a complete circle, sorry "doughnut" :smile:

Barrie

 

[scott]

Thanks, Barrie--I'll see what I can find that's a bit thicker.