The curious case for the 40mm Summicron... With images

[wlaidlaw]

Advertisement (gone after registration)

I have to admit that now I have a 35 ASPH Summicron as a far lighter alternative to my chrome/brass 35 ASPH Summilux, my 40 Summicron-C sits all the time on my film CL. Even that rarely gets used due to its erratic metering, even though it has been fettled twice over recent years. I suspect it is the fault of the MR-9 PX625 battery replacer. The metering is far more accurate if you remember to press the battery check light button two or three times before using the camera for the first time of the day. I think this "wakes" up the battery regulation circuit in the MR-9. I find the same phenomenon with the MR-9 625 replacer in my MR-4 Leicameter. 

Wilson

[wlaidlaw]

7 hours ago, Michael Geschlecht said:

Hello Wison,

Do you have a photo of an M5 with an "offset RF cam roller"?

Best Regards,

Michael

I have never had an M5 so I am going on what Malcolm Taylor told me when I sent him my 40-C for service and tweaking, when I asked him about focus accuracy of the 40C on non-CL Leicas. I think there is also a Dante Stella article on this.  I am not even sure what Malcolm means by "an offset RF cam" but I am sure he would be correct, as one of the Leica repair gurus. He is currently repairing my 250FF reporter and servicing my motor drive IIIa.

Wilson

[Advertisement]

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

[lct]

10 hours ago, james.liam said:

I remember reading something by E. Puts about this prototype and he thought it was a chien of a lens.  

Marco Cavina's review of the Elmarit-C 40/2.8 looks more documented:

http://www.wetzlar-historica-italia.it/elmarit4028.html

[Michael Geschlecht]

8 hours ago, wlaidlaw said:

I have never had an M5 so I am going on what Malcolm Taylor told me when I sent him my 40-C for service and tweaking, when I asked him about focus accuracy of the 40C on non-CL Leicas. I think there is also a Dante Stella article on this.  I am not even sure what Malcolm means by "an offset RF cam" but I am sure he would be correct, as one of the Leica repair gurus. He is currently repairing my 250FF reporter and servicing my motor drive IIIa.

Wilson

Hello Wilson,

What the "offset cam" sometimes refers to is that some of the earlier M3's & some other earlier M cameras, had focusing rollers that were somewhat off center. That is to say: Not precisely at "12:00 high" when looking at the lens opening in the body. They might have arrived from the factory & focused accurately, with the roller a little to the left or to the right of top dead center.

Precise alignment of the roller, left or right of center, was not important because there was enough of the appropriate cam surface of the lens touching the roller on the body, even if the roller was a little off center.

While the geometry of the rangefinder cams of the (Film) CL lenses required that the roller be more exactly centered.

Best Regards,

Michael

Edited April 23, 2019 by Michael Geschlecht

[Michael Geschlecht]

8 hours ago, wlaidlaw said:

I have to admit that now I have a 35 ASPH Summicron as a far lighter alternative to my chrome/brass 35 ASPH Summilux, my 40 Summicron-C sits all the time on my film CL. Even that rarely gets used due to its erratic metering, even though it has been fettled twice over recent years. I suspect it is the fault of the MR-9 PX625 battery replacer. The metering is far more accurate if you remember to press the battery check light button two or three times before using the camera for the first time of the day. I think this "wakes" up the battery regulation circuit in the MR-9. I find the same phenomenon with the MR-9 625 replacer in my MR-4 Leicameter. 

Wilson

Hello Wison,

This might also be due to the sometimes "erratic" performance of alkaline batteries when used as replacements for mercury batteries.

I use silver oxide alternatives (357/303/SR44) in both an MR meter & a Minolta 101 and both work just as fine & just as smoothly as they did with their mercury predecessors (PX13/PX625).

Best Regards,

Michael

Edited April 23, 2019 by Michael Geschlecht

[a.noctilux]

18 hours ago, Michael Geschlecht said:

Hello Wison,

Do you have a photo of an M5 with an "offset RF cam roller"?

Best Regards,

Michael

Like this ?

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!

a bit to the right if we admit that the "wax seal" is at "12h"

my other black M5 has almost the roller at the same place (a very small bit less to right)

I examined other Ms, some have that offset more or less,

but in real life use, those offsets give little "impact" in sharpness and stay in the dof "2" limits (largely) with M-Rokkor 2/40

I've just tried that out with M5 + M-Rokkor 40 which has that steep coupling ramp, compared to plain Summicron-M 35mm at same points of focus.

[wlaidlaw]

Advertisement (gone after registration)

3 hours ago, Michael Geschlecht said:

Hello Wison,

This might also be due to the sometimes "erratic" performance of alkaline batteries when used as replacements for mercury batteries.

I use silver oxide alternatives (357/303/SR44) in both an MR meter & a Minolta 101 and both work just as fine & just as smoothly as they did with their mercury predecessors (PX13/PX625).

Best Regards,

Michael

Michael, 

I always use Silver Oxide SR43 cells in my MR-9 adapters, due to their flatter voltage/discharge curve. The only place I don't use those is in my R4 and M7, where I use lithium-managanese CR 1/3 DN. I am sure the fault lies with the MR-9 adapters as I have had a couple of them die on me. First they become more difficult to wake then die totally. I believe that some cameras that used the PX 625 mercury cell, can use the SR44W cells, which are the same size as the mercury cell, as they have an internal voltage regulation circuit. Sadly I understand the CL is not one of them. Whereas I can get my MR-4 meters recalibrated properly (in contrast to a well known meter man in California, who made a total mess of doing one of mine) by Kamera Technik Langer in Karlsruhe for €60 each. I bought one of my MR-4 meters from him and it is far more accurate than either the one re-calibrated in the USA or the one using the PX625 replacer. He said he can do the CL but some seem to work and some don't. 

Wilson

[lct]

1 hour ago, a.noctilux said:

but in real life use, those offsets give little "impact" in sharpness and stay in the dof "2" limits (largely) with M-Rokkor 2/40

The Rokkor for Minolta CLE (not CL) has not the same focus cam as the Summicron. 

Edited April 23, 2019 by lct

[Michael Geschlecht]

1 hour ago, a.noctilux said:

Like this ?

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!

a bit to the right if we admit that the "wax seal" is at "12h"

my other black M5 has almost the roller at the same place (a very small bit less to right)

I examined other Ms, some have that offset more or less,

but in real life use, those offsets give little "impact" in sharpness and stay in the dof "2" limits (largely) with M-Rokkor 2/40

I've just tried that out with M5 + M-Rokkor 40 which has that steep coupling ramp, compared to plain Summicron-M 35mm at same points of focus.

Hello A. Noctilux,

Curious. Because the reason that many earlier "M" cameras did not necessarily have their rollers directly at 12:00 was that: Within certain parameters : It didn't matter where the roller was placed. Curious because by the time that M5's were being built the technical ability to align the roller was better & the need for its alignment for use with the film CL lenses was known.

Best Regards,

Michael

Edited April 23, 2019 by Michael Geschlecht

[wlaidlaw]

Michael, 

I think there must be more to the M5/40C problem than a tiny displacement to one side of the roller. Could it be something to do with the swing down light meter, requiring different levers on the RF linkage, in order to clear the meter arm. 

Wilson

[dau]

1 hour ago, wlaidlaw said:

Michael, 

I think there must be more to the M5/40C problem than a tiny displacement to one side of the roller. Could it be something to do with the swing down light meter, requiring different levers on the RF linkage, in order to clear the meter arm. 

Wilson

I just had a look at some lenses. The "real" M lenses all use a lever which moves in parallel to the optical axis of the lens to press on the roller. The rotational position of the roller with respect to the mount is irrelevant. The -C lenses have just a ramp cut into the cylinder which turns with the lens when focussing. The rotational position of the roller must therefore have an influence on the distance measured by the RF in the body.

[adan]

Wilson, are you considering the following notable difference in how C and M lenses work and are constructed?

As we know, the Leica RF geometry is constructed around the movement of a 50mm lens when focused. For any focal length other than 50mm, a lens has to have a "differential movement" to move the rangefinder as though a 50mm lens was mounted, to match the hard-wired RF geometry (move the roller x-amount to move the RF image y-amount).

In M lenses, this is done by adding an additional internal focus helix, and cylindrical cam. Turn the lens, and the lens moves in and out to focus, and in addition, the inner cylinder moves in and out a different amount on its own thread, to simulate the movement of a 50mm lens. The back surface of the cylinder lens cam is flat, therefore it does not matter if the camera's roller cam is offset one side or the other.

For weight and simplicity (cost) reasons, the C 40mm and 90mm lenses do not have a separate threaded focus cylinder/cam with a flat contact surface. The back of the lens tube is the cam, machined with a slope to move the camera roller in and out with rotation and provide the differential movement needed for a 40 or 90 vs. a 50. But because of the slope, offsetting the camera's roller to one side or the other means it no longer touches the lens at the correct place. And that will de-calibrate the focus.

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!

The C lenses were never intended to be used on an M body. There are no framelines for 40mm, and there were plenty of certified "M" 90mm f/4 lenses already available for M-mount. The shallower cam-slope and DoF of a 40, and/or chance coincidence of the roller-cam positions, may allow it to work - the 90, with less DoF, is more susceptible to focus problems.

However, the CL body will work fine with M lenses and their more sophisticated focus-cam engineering.

All that was understood 45 years ago when the C lenses were in production - today that knowledge may have been lost in the fog of time.

(Edit - this of course is exactly what dau just said - we posted together. The "lever" he mentions only occurs on some larger lenses in some models (e.g. 90 f/2 APO), to save weight - the cylindrical lens focus cam is short and buried in the lens barrel, while the "lever" or follower-bar, much smaller and lighter, extends the cam movement to the back of the lens)

Edited April 23, 2019 by adan

[dau]

As Andy said; here some quick and nasty shots:

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!

[lct]

Been using both M-Rokkor for CLE and Summicron-C 40/2 for many years on film and digital Ms and i've never got focusing issues despite their different focus cam i must say. Same for M-Rokkor and Elmar-C 90/4 BTW. FWIW.

[jdlaing]

Same here.

[jmahto]

Same here

 

[wlaidlaw]

Has anyone on this thread got an M5 plus any other M. Would it be possible to see if due to the exigencies of the light metering arm, the M5's roller moves further sideways when depressed than regular M's roller. If this is the case, then that would be the explanation for people saying that the Summicron C with its sloped RF cam (which some Russian lenses also have) would be less accurate on the M5 than other M cameras.

When I got my Summicron-C 40 back from its tweak and service by Malcolm Taylor, I tested its focus accuracy on my M240, reading zoomed live view with peaking against rangefinder convergence. It seemed to focus just fine. I bought a new Noctilux just after the Summicron-C came back from service and used the same focus checking method with the Noctilux. It immediately showed that unlike the Summicron, the Noctilux was far from focusing accurately. I had to arrange a pick up and repair by Leica, as there were no other new replacements available and it was going to be two months until the next batch was built. 

Wilson