Hand Coding Zeiss 15/2.8

[sean_reid]

Advertisement (gone after registration)

Has anyone successfully hand coded the Zeiss 15/2.8 as a WATE? I'm testing it now and am fairly sure I have the marking in the correct position (having hand-coded a lot of lenses by now) but even with the frame lines manually set to 28/90, it doesn't trigger the WATE menu. My only thought is that the screw could be fooling it. Any one been down this road?

 

Cheers,

 

Sean

[sdai]

Sean, I have neither of these two but I'm guessing this is because the 15 Distagon doesn't have the rangefinder cam which exists in the WATE and transmit the lens' focusing distance to the camer body. So no matter what you do, it just won't work.

[marknorton]

Simon, the camera doesn't know anything about the presence or otherwise of a focussing cam. The only things sensed are the position of the frame selector lever and the lens code.

 

Here's a scan of the back of the WATE and of a hand coded lens which is recognised as a WATE when I push the frame selector lever all the way in towards the lens (this lens is actually a 50mm which is the only uncoded lens I have).

 

If there's a screw in the way, you might cover it with a sticky label...

 

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!

 

[doubice]

Simon,

 

As Mark correctly pointed out, the fact whether a lens is rangefinder coupled or not, has absolutely nothing to do with coding.

 

Cheers,

 

Jan

[sdai]

As Mark correctly pointed out, the fact whether a lens is rangefinder coupled or not, has absolutely nothing to do with coding.

Jan, I understand that part ... but I suspect that's the reason why the WATE menu cuoldn't be brought up for the 15 Distagon despite the code in place. Was there anybody successful in bringing up the WATE menu for the 15 CV? If so then it could be due to other reasons.

 

Mark's 50 is obviously a coupled lens so the camera can be fooled.

[marknorton]

Simon, try this.

 

Take an opaque white piece of paper, put a black stripe on it, 1.5mm wide. Then, holding the lens selector lever in, pass the paper over the sensor, keeping the unmarked part pressed flat against the sensor. When the mark is in the correct position (second sensor element, counting from the bottom left), the WATE menu will pop up.

 

I assume you agree a piece of paper doesn't have a focussing cam?

 

[Edit: you do of course need to have the camera set up with Lens Recognition + UV/IR and 1.102 firmware]

[marknorton]

Advertisement (gone after registration)

Simon, the reason why the CV 15 coding is difficult is because a standard LTM adapter doesn't completely cover the sensor.

[sdai]

I assume you agree a piece of paper doesn't have a focussing cam?

 

LOL ... can't beat you on that.

 

Now even a piece of paper can fool the M8, if Sean's coding is absolutely spot on then it might be something with the 15 ZM disabling the sensor or driving it nuts when mounted ... take a close look at the back mount of both the WATE and 15 ZM, are there anything even slightly different?

[doubice]

Jan, I understand that part ... Mark's 50 is obviously a coupled lens so the camera can be fooled.

 

Simon,

 

Sorry, but I don't think you do understand. The only two factors that influence the coding are the 6-bit code (or its manually applied equivalent) and the frame selector. Nothing else, and I would like to repeat nothing else will have any effect whatsoever.

 

The lens' focusing cam is only coupled mechanically to the rangefinder, there is no electronic exchange of information between the focusing cam and the camera. Thus - the absence or presence of the cam cannot influence the coding.

 

Cheers,

 

Jan

[sdai]

Sorry, but I don't think you do understand. The only two factors that influence the coding are the 6-bit code (or its manually applied equivalent) and the frame selector. Nothing else, and I would like to repeat nothing else will have any effect whatsoever.

 

The lens' focusing cam is only coupled mechanically to the rangefinder, there is no electronic exchange of information between the focusing cam and the camera. Thus - the absence or presence of the cam cannot influence the coding.

No sorry, Jan ... the fun with the forum is correcting others and being corrected by others but, I still "insist" (even when I agree with you that the cam has no direct relationship with the coding) ... as you may also agree - the camera certainly knows if the lens has a cam to engage with its rangefinder mechanically - it's like a (mechanical) switch, if it's off, the camera may want to electronically disable part of its functions.

 

Now I agree this is out of the picture because Mark proved that a piece of paper would trigger the sensor then on the contrary side it must be due to something unique (to the 15ZM) which makes the sensor not work (at least not as expected).

[doubice]

.... the camera certainly knows if the lens has a cam to engage with its rangefinder mechanically - it's like a (mechanical) switch, if it's off, the camera may want to electronically disable part of its functions.

 

Actually Simon, the only entity that knows whether the lens has a cam, is the photographer...... The image in the RF patch either moves when you turn the focusing ring, or it does not.

 

There is no mechanical or electronic switch in the RF mechanism. If you own an M8, take the lens off and push the RF roller back and forth. Do you feel a faint click of a switch? Do you see any diffrence in the EXIF data should you be foolish enough to trip the shutter while doing so?

 

Believe me - I never like to beat my own drum, but I know a little about mechanicals and electronics. Photography has been a part of my life since times when autofocus and autoexposure were just a figment of a mad designer's imagination.......

 

All the best,

 

Jan

[doubice]

Good night.

[scott kirkpatrick]

Has anyone successfully hand coded the Zeiss 15/2.8 as a WATE? I'm testing it now and am fairly sure I have the marking in the correct position (having hand-coded a lot of lenses by now) but even with the frame lines manually set to 28/90, it doesn't trigger the WATE menu. My only thought is that the screw could be fooling it. Any one been down this road?

 

Cheers,

 

Sean

 

 

Sean, you may be seeing a consequence of the same "feature" that I discovered when coding a 50/2.0 as a TE. The firmware insists on seeing both the 6bit code and the correct position of the mount lug or the frame lever, whichever is further around the mount circle (see thread from today about "TE 35 not recognized.").

 

So what framelines comesup when you install the Zeiss 15? If it is the 28/90 pair, you are home free, that's what the WATE shows. But you said you moved the frame lever over to 28/90, so I guess something different happens. If there is no frame lever lug at all, it will bring up the 35/24 combination, and you are hosed, since the toggle switches will say 35mm regardless of the frame lever position.

 

scott

[sean_reid]

Thanks for the responses. It brings up the 50/75 lines so there is a lug.

 

It's gotta be the screw, then, right? I need to get some more white and black nail polish today anyway and so I'll paint the screw area with white.

 

Mark, thank you for those pictures. The second provides a good way for me to check my positioning.

 

Added: OK, got it. It was the screw. The label (which I had used on another Zeiss lens but forgotten about, did the trick).

 

Cheers,

 

Sean

[atanabe]

I have sucessfully coded a CV 15 as a 16-18-21 TRI using the 90 LTM to M ring. So the frame (mechanical sensor is not involved). I had the most difficult time with the position of the single black mark, it was VERY picky! Once you sucessfully place the mark you will be rewarded with the 16 - 18 - 21 lense selection on the LCD.

[scott kirkpatrick]

It wasn't Sean's exact problem, but I'll paraphrase Mark Norton's answer to the question of the mechanical logic that is used in letting the firmware know about the frame lever:

 

It will know if you push but not if you pull.

 

OR:

 

mount says 35, you signal 35, can push to be a 50/75 or a 28/90 instead.

mount says 50, you signal 50, can push to be 28/90. Pull to signal 24/35 is ignored.

mount says 28/90, that's what you're gonna be.

 

I guessed it wrong in the other thread, Mark explained it, but in the negative. Mechanical logic can be just as confusing as digital.

 

scott

[sean_reid]

It wasn't Sean's exact problem, but I'll paraphrase Mark Norton's answer to the question of the mechanical logic that is used in letting the firmware know about the frame lever:

 

It will know if you push but not if you pull.

 

OR:

 

mount says 35, you signal 35, can push to be a 50/75 or a 28/90 instead.

mount says 50, you signal 50, can push to be 28/90. Pull to signal 24/35 is ignored.

mount says 28/90, that's what you're gonna be.

 

I guessed it wrong in the other thread, Mark explained it, but in the negative. Mechanical logic can be just as confusing as digital.

 

scott

 

Hi Scott,

 

I follow your post, I believe, but what do you mean by "signal"?

 

Cheers,

 

Sean

[jlm]

the 15 heliar works when coded at a WATE (no rf cam actuation, simon) and it brings up the 28mm framelines. if your zeiss brings up the 50 frame lines, isn't that the problem? unfortunately, the bayonet tab has to be longer for the 28 by about .037" compared to the tab for the 50 you can actually see the tab edge difference in Mark's photo, (compare the tab edges near the lens lock/detent at about 1:00 o'clock)

[scott kirkpatrick]

Hi Scott,

 

I follow your post, I believe, but what do you mean by "signal"?

 

Cheers,

 

Sean

 

The two toggle switches which Mark Norton's picture shows each set a bit in the EXIF. Those two bits, plus the six bit code on the lens mount, "signal" to the firmware what to do when lens detection is enabled. The surprise is that not all combinations of frame lever position and mount depth are signaled in this way. Apparently the frame lever, frames and rangefinder were designs that were not to be touched, and the electronics to detect what is going on are carefully added outside of the most critical mechanical parts. That's why the camera responds to not just the frame lever or just the mount lug, but a sort of largest deflection from rest of the two.

 

Still don't have the right words to describe, but I hope it is clear.

 

scott

[Advertisement]

Hallo,

Look here: Leica M

[rvaubel]

The two toggle switches which Mark Norton's picture shows each set a bit in the EXIF. Those two bits, plus the six bit code on the lens mount, "signal" to the firmware what to do when lens detection is enabled. The surprise is that not all combinations of frame lever position and mount depth are signaled in this way. Apparently the frame lever, frames and rangefinder were designs that were not to be touched, and the electronics to detect what is going on are carefully added outside of the most critical mechanical parts. That's why the camera responds to not just the frame lever or just the mount lug, but a sort of largest deflection from rest of the two.

 

Still don't have the right words to describe, but I hope it is clear.

 

scott

 

I hope someone is writing all of this down. The issues of getting correct lens coding on a non-Leica lens can be so complex as to be a real detriment to a non dedicated amatuer. Of course that is what Leica thinks is in their best interest but most of us think differently about that. After all this is sorted out, it would be nice if someone could provide an all-in-one coding service that involves all the coding issues including everything from the coding slots to the frameline selection issue.

Although it is kind of fun to understand how all these variables intereact, the truth of it is that it is way too much for your average bear to comprehend.

 

Rex