Canon 28mm 2.8 LTM focusing issue

[jmp]

Advertisement (gone after registration)

Hi!

I have an issue with a recently purchased Canon 28mm 2.8 LTM lens.

Trying the lens (but not yet tested with a film) on both my M2 (CLA’ed) and my IIIf with comparisons with my other lenses (Summaron 35mm 2.8 and Summitar 50 2) show discrepancies on the distance indication ring.

I mean, focusing on an object 5 meters away gives an indicated distance of 3 meters
Infinite is wrong, I focus on infinite around 6 meters, and if I continue to turn to the infinite lock, then focus is wrong.

My Summaron & Summitar behave as expected.

The seller said he used the lens for 3 years and never had a problem and advice first to test with a film.
I can do a roll but I’m afraid I will just loose valuable film…

Lens is clean but has been opened.

Questions:
- do you know or have you experienced this kind of discrepancies on this Canon 38mm (aka is it a known problem)?
- if not, what would be wrong or misaligned?

Thanks!

[spydrxx]

Yes I recently experienced the issue with a Canon 28/3.5 I purchased. It was gorgeous on a mirrorless digital, but the distance shown on the focus ring was way off, and I returned it assuming it would also be off on a film body as all of my other 28mm lenses seemed to indicate (almost) the infinity mark at objects 100-200 yards away. I purchased another 28/3.5, which exhibited similar behavior, but to a lesser extent. So I put it on a film body which had a back I could open up, and using my film plane focus device confirmed that at both close focus and infinity, the scale on the lens was fine. I thought perhaps my screwmount adapter ring for my digital body was out of spec, but it worked fine with my other lenses, so, although I'm perplexed, I decided to quit worrying about it and just enjoy shooting with the Canon 28/3.5.

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[wizard]

vor einer Stunde schrieb jmp:

I focus on infinite around 6 meters, and if I continue to turn to the infinite lock, then focus is wrong.

That seems to be off a little bit too much, but you never know. The easiest way to find out would be to check the lens with the adapter ring on a digital M camera. If the infinity focus is correct when the lens is set to its infinity position, then all should be in good order on film cameras, too.

[jmp]

to be precise:

I have same problem on:

1) my III f (screw mount, no adapters)

2) my M2, using original Leica adapter

How can I be sure I am correctly focused if, if set on infinity lock, the focus is off?

 

[spydrxx]

I found that focus was a tad off with the infinity lock, but was spot on just adjacent to the lock. Screwmout and M body viewfinders said it was in focus at that point, and when tested on both a Nicca screwmount, which has a back which lifts up like an M3, and on an adapter on my M2, with a groundglass and 10x magnifier it was indeed perfectly focused. 

[jankap]

I have this lens too. It never became a CLA, so it is more or less in the factory condition. I bought it new in 1968. Could you check the following?

At infinity on the backside there is a brass ring. I measure (primitively), that it protrudes 3mm, seen from the side.

At the MFD of 1 meter the brass ring, seen from the side, is not visible any more.

Edited January 10, 2023 by jankap

[Jerry Attrik]

Advertisement (gone after registration)

Perhaps a clue?  JMP in his post #1 states"...The lens is clean but has been opened..." 
The weird staggering of the actual focus and the indicated focus might suggest a
misalignment of the focus barrel and the helix in which it travels. This can, and, does happen
with multistart helix units.  A crude check--- remove lens from camera hold the
mount steady and rotate the focus lever to infin. Look at the back of the lens and
see where the focus barrel; the brass threaded cylinder between the lens barrel (black)
and the stainless steel mount flange,is in relation to the lens mounting flange. 
It will stick out by about 3 mm. now wind the focus down to minimum; the flat edge of the
brass barrel should be exactly flush with the top edge of the mount thread.
This is how my 28mm F1:2.8 Canon lens behaves.  I have no trouble at all.

A.T.B.    D. Loxl

(Thinks..bother) Pre-empted by Jankap as I was writing D.L.

Edited January 10, 2023 by Jerry Attrik

[jankap]

+1

[Jerry Attrik]

My first Leica, 3f in 1960 courtesy of my father.
Second a Leica 3f in 1977.
Still use them both.

D. Lox

[spydrxx]

6 hours ago, jankap said:

I have this lens too. It never became a CLA, so it is more or less in the factory condition. I bought it new in 1968. Could you check the following?

At infinity on the backside there is a brass ring. I measure (primitively), that it protrudes 3mm, seen from the side.

At the MFD of 1 meter the brass ring, seen from the side, is not visible any more.

Similar on mine. The protrusion at infinity measures 3.52mm.

[jmp]

Crude measurements (I don't have a precision caliper on hand)

at 1m: protrude approx 1 mm
at infinity lock: protrude approx 4 mm

 

[jankap]

Perhaps a visit at OBI or Hagebau?

But your lens seems to be wrong compared with the lenses of Jerry and myself. You wrote, that the lens was opened, perhaps for a CLA. As Jerry wrote, other than a metric or Withworth screw the screw to adjust the distance of a lens carries a multithread. Of course 2 or 3 threads only, that is to win height with less rotation. That means, that if one - after a CLA - has to put the lens together, the right thread should be taken. Perhaps this is the reason of the problem.

Personally I wonder, why people let clean their lenses. I do nothing, perhaps a bellow. With the lens in question, putting a fingerprint on the glasses would cost effort. Also for my digital cameras, I use a small bellow at most.

Edited January 11, 2023 by jankap

[jerzy]

I am afraid that you would need special measuring equipment to measure protruction, caliper is not precise enough. It is roghly 0,5mm protruction difference between infinity and 6m (measured on Summitar). Assuming that the threads are cut properly (and with Canon it is to be expected) the problem might be in inproper assembly.

As usual for all lenses with focal length different than 50mm distance which optics moves between 1m (example) and infinity is different than the deflection of rangefinder arm, it is equal only for standard focal length. With the exception of early Elmars 35mm (that I know) which have sloped cam  nowadays solution with 2 helicoids is in use - one is for optics, the second for rangefinder arm. Both are mechanicaly coupled, but having different pitch provide that both focus and distance shown on the distance scale are correct.
In addition to possible reason as indicated by jankap above the incorrect assembly of both helicoids could be an issue here.

[jankap]

Thank you, you are right; there must be two separate threads. The movements - lens, scale and rangefinder output -  are too different. I have never disassembled a lens, let alone a rangefinder lens.

I wonder now, how a technician could assemble the lens wrong. I think, I should have look in my LTM lens collection, which one I can miss.😜

Edited January 11, 2023 by jankap

[pegelli]

Do you know somebody who has an M240 or later digital M on which you would be allowed to test the lens?

Those bodies are ideal to check rangefinder vs. live view and find discrepancies.

Edited January 11, 2023 by pegelli

[jerzy]

vor 5 Stunden schrieb jankap:

I wonder now, how a technician could assemble the lens wrong

humans make errors.... I do not know mechanical built of Cannon 28mm but here is example of  Hektor 28mm. Helicoid 1 moves optics, 2 rangefinder. Both are coupled with 2 small blocks. If you do not make markings it is easy mount them being rotated by 180 degree. As mentioned, Cannon may be even more sensitive and allow more possibilities to to it wrong

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

[roydonian]

I've just tried my 28mm f/2.8 Canon, and it is in good agreement with the rangefinder results from three Leitz-era lenses. All four showed that my test object was just over 7ft away.

[jankap]

Would you have expected something else?

[jmp]

Thanks gents! I think this is now pretty clear the lens do have a problem

[Advertisement]

Hallo,

Look here: Interesting Leica gear

[jankap]

On 1/11/2023 at 4:40 PM, pegelli said:

Do you know somebody who has an M240 or later digital M on which you would be allowed to test the lens?

Those bodies are ideal to check rangefinder vs. live view and find discrepancies.

You could try the proposal of pegelli. I think, that the optics are OK, but that the rangefinder mechanism is off. If so, the repair is simple.

Perhaps the seller of your lens used it on non rangefinder camera.