M8 sensor plane position?

[microview]

Advertisement (gone after registration)

Can anyone confirm how far back from the front face of the M8 the sensor plane lies, within the 36mm thickness of the body? I suspect my 50mm Summilux ASPH is – at close distances – focusing around 25mm+ in front of what the rangefinder suggests. (Using a calibrated strip with the camera angled on a tripod at 45 degrees to the strip shows this to be the case.) I would like to set the focus scale on the lens exactly as per a tape measure from camera body to subject, and see if that confirms a problem (or gives one a bypass for best focus at nearest distances by actual measurement when setting up the camera).

Thanks in advance for your replies.

[Advertisement]

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

[luigi bertolotti]

The answer is easy: 27,8 mm: this is the "historical" flange-to-film distance of Leica M (1 mm less than the old Leica 39x1 screwmount) that of course is the flange-to-sensor distance in Leica M8. The distance in the focus scale refers to focus plane (film-sensor) to subject, so your evaluation of an "error" of 25 mm in front focus from CAMERA BODY to subject has a sense.

[SJP]

Indeed, to confirm, the scale is from the sensor plane to the object, simplifying a bit for a simple lens 1/f=1/O + 1/B and the barrel indicates the distance O+B.

[hoppyman]

Microview, be very careful with your setup for measuring this yourself. It is possible to introduce errors due to the displacement of the RF above the lens etc etc. Many people who try focus adjustments themselves later report problems, for example with other lenses. However you may not be intending to do that yourself? Personally I don't recommend that you try it yourself.

 

25mm is a large error at close range, especially for that lens design. I guess you know about focus shift as aperture is changed? This lens and the APO 75 are two that have no issue there in my experience. Even so, I thought that mine were perfect in practical use and Leica checked them for me when I had my M8 upgraded. "Your lenses are good. We can make them better!"

 

Any adjustment error can exist in camera, lens or operator or a combination of those . The focus magnifier helps a lot as well. Of course the sensor plane of focus tolerance is essentially zero, unlike film.

 

I watched the factory experts measure and record the focus accuracy on a set of the new White M8 bodies. Their computer display gives predicted and actual points which they move into coincidence. They must be ensuring that the body is within tolerance but I don't know how the camera uses that information. Perhaps it is used by the firmware?? That is a guess. I imagine that would be commercial in confidence anyway? The jig they use is right across a large room. They have the angled targets at 8 or 10 points, starting at .7m then .9 etc out to much longer. Their measuring apparatus is of course very sophisticated and not by peering through the finder with naked eye naturally.

 

If you establish that you think you do have a problem, Leica customer service can of course adjust body and lens together for you. I think that lens mounts of slightly different dimensions are used, or perhaps shims which would make sense. In any event naturally their adjustments are of course far more sophisticated than anything you can achieve for yourself and adjusting the camera body is fraught, in my view. You want your camera to be exactly in the centre of the tolerance range in a perfect world and your lens/lenses adjusted to work perfectly to that.One person you can send English enquiries to is:

Andrea.Frankl@leica-camera.com

Good luck and let us know how you go .

Can anyone confirm how far back from the front face of the M8 the sensor plane lies, within the 36mm thickness of the body? I suspect my 50mm Summilux ASPH is – at close distances – focusing around 25mm+ in front of what the rangefinder suggests. (Using a calibrated strip with the camera angled on a tripod at 45 degrees to the strip shows this to be the case.) I would like to set the focus scale on the lens exactly as per a tape measure from camera body to subject, and see if that confirms a problem (or gives one a bypass for best focus at nearest distances by actual measurement when setting up the camera).

Thanks in advance for your replies.

[mjh]

I would like to set the focus scale on the lens exactly as per a tape measure from camera body to subject

The focus distance is the distance between the subject and the image plane (i.e. the sensor position), not the distance from subject to camera body. It has always been defined that way, and as already mentioned, the difference is 27.8 mm.

[microview]

The focus distance is the distance between the subject and the image plane (i.e. the sensor position), not the distance from subject to camera body. It has always been defined that way, and as already mentioned, the difference is 27.8 mm.

 

Perhaps I didn't make my intentions clear, but I'm not planning on just measuring to the front of the camera body – that's why I wanted to know about that 27.8mm in my OP.

 

Anyway, thanks for the info. And for all your responses. (Hoppyman, I would not dream of trying to adjust either lens or rangefinder setup myself! But as you suggest, possibly the lens is outside spec. And I have the 24mm Elmarit as a cross reference.)

 

I thought this dahlia jpeg (from about 800mm) ought to be sharper....

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!

[hoppyman]

Advertisement (gone after registration)

Microview, a 24 would be a rotten benchmark for focus accuracy! DoF obviously much larger and your 'Lux at 1.4 in close. Better to try your lens on another camera under identical conditions and vice versa .

Perhaps I didn't make my intentions clear, but I'm not planning on just measuring to the front of the camera body – that's why I wanted to know about that 27.8mm in my OP.

 

Anyway, thanks for the info. And for all your responses. (Hoppyman, I would not dream of trying to adjust either lens or rangefinder setup myself! But as you suggest, possibly the lens is outside spec. And I have the 24mm Elmarit as a cross reference.)

[microview]

Microview, a 24 would be a rotten benchmark for focus accuracy! DoF obviously much larger and your 'Lux at 1.4 in close. Better to try your lens on another camera under identical conditions and vice versa .

 

Thanks Geoff.

[lars_bergquist]

Obviously you cannot measure sensor position without any reference 'null' position, and that is the front flange plane of the bayonet. By the way, the complete specification, as far as I know, is 27.8mm +/- 0.01mm. Can you do this kind of accuracy?

 

I have had focusing problems with two lenses. One was a Zeiss Distagon 18mm, so Carl Zeiss had to exchange it for a correctly calibrated one. The other was a mint 90mm Elmarit, last model -- a mystery, really. That went back to Solms for calibration. This lens was calibrated expressly for my M8, which was NOT meddled with. It is now astonishingly sharp.

 

The conclusion is that quality control is a human endeavour, and thus fallible. Sometimes it is only too human. But QC is also a very sophisticated science, and some adjustments are best left to people who have the training and the instrumentation to do it right.

 

Older Leica cameras had a focal plane mark on the top plate, but that was intended only for focus measurements with a tape measure, much too rough for a critical assessment of focus.

 

The old man from the Age of the Slit Screwdriver (now living in the Age of the IKEA Key)

[marknorton]

Another way to look at it is that a correctly calibrated M lens will bring the image of an object at infinity to focus 27.8mm behind the bayonet flange. The question is then: where do you mount the sensor to obtain best image sharpness? Where is the sweet spot? The sensor has a small, but finite, thickness compared to film and is mounted at three points using shims to an accuracy of +/- 0.005mm - the thinnest shim I found in my camera was 0.01mm. By changing the shims, Leica can mount the sensor at the required distance behind the bayonet flange.

 

Have Leica, over the lifetime of the M8, changed the position of the sensor - shifted it forwards or back by a fraction of a mm compared to the first cameras off the line?

 

I cannot be certain but there's much less discussion here about front and back focussing than there used to be and my own cameras came back from upgrading (which would have required the sensor to be removed and, presumably, realigned) working much better with much less focus inconsistency across the lenses I have.

 

Something's changed and the truth is, I think, that Leica have learned a lot about building the M8 over the nearly 3 years it's been in production.

[rob_x2004]

Out of interest, what has the sensor distance got to do with rangefinder coupling and alignment?

Additionally, just because one lens works and the other doesnt, doesnt completely rule out that both lenses are fine, or otherwise. It can still be the camera.

Mark, youve had enough of them apart. Ever chatted with someone in the know?

[hoppyman]

Mark, what you say makes perfect sense and of course I have looked in awe (and a few cringes:)) at your dissassembled camera series.

To expand on that line of thought, it would be fascinating to know what process follows the final checking of the camera body on a benchmark lens with Leica's very sophisticated rig and expert operator. You might have read that a group of forum members observed some of this procedure during a recent factory visit. Unless it was just a check that they were in tolerance at the specific positions. Disasssembly would not have been normal at that point I would think (unless a fault was found naturally).

My camera was upgraded and then the technician took my 50 & 75 for check against the reassembled camera figures, presumably from that process.

Another way to look at it is that a correctly calibrated M lens will bring the image of an object at infinity to focus 27.8mm behind the bayonet flange. The question is then: where do you mount the sensor to obtain best image sharpness? Where is the sweet spot? The sensor has a small, but finite, thickness compared to film and is mounted at three points using shims to an accuracy of +/- 0.005mm - the thinnest shim I found in my camera was 0.01mm. By changing the shims, Leica can mount the sensor at the required distance behind the bayonet flange.

 

Have Leica, over the lifetime of the M8, changed the position of the sensor - shifted it forwards or back by a fraction of a mm compared to the first cameras off the line?

 

I cannot be certain but there's much less discussion here about front and back focussing than there used to be and my own cameras came back from upgrading (which would have required the sensor to be removed and, presumably, realigned) working much better with much less focus inconsistency across the lenses I have.

 

Something's changed and the truth is, I think, that Leica have learned a lot about building the M8 over the nearly 3 years it's been in production.

[jaapv]

Microview, be very careful with your setup for measuring this yourself. It is possible to introduce errors due to the displacement of the RF above the lens etc etc. Many people who try focus adjustments themselves later report problems, for example with other lenses. However you may not be intending to do that yourself? Personally I don't recommend that you try it yourself.

 

25mm is a large error at close range, especially for that lens design. I guess you know about focus shift as aperture is changed? This lens and the APO 75 are two that have no issue there in my experience. Even so, I thought that mine were perfect in practical use and Leica checked them for me when I had my M8 upgraded. "Your lenses are good. We can make them better!"

 

Any adjustment error can exist in camera, lens or operator or a combination of those . The focus magnifier helps a lot as well. Of course the sensor plane of focus tolerance is essentially zero, unlike film.

 

I watched the factory experts measure and record the focus accuracy on a set of the new White M8 bodies. Their computer display gives predicted and actual points which they move into coincidence. They must be ensuring that the body is within tolerance but I don't know how the camera uses that information. Perhaps it is used by the firmware?? That is a guess. I imagine that would be commercial in confidence anyway? The jig they use is right across a large room. They have the angled targets at 8 or 10 points, starting at .7m then .9 etc out to much longer. Their measuring apparatus is of course very sophisticated and not by peering through the finder with naked eye naturally.

 

Geoff, you are 100% spot on. I happened to be talking on this subject yesterday with Will van Manen, respected camera repairman. He advised strongly against the hex-key method of adjusting - of which, I must admit, I have been guilty myself in the past. I transpires that any change there has a magnifying effect -usually negative- on the close-distance focussing. Both the close focus and infinity focus must be adjusted together against a reference lens, or, as you say, on the factory rig.