Drifting focus

[Derbyshire Man]

Advertisement (gone after registration)

Anyone had drifting focus? My camera is a couple of months old but keeps drifting into front focus. Was perfect when delivered and adjusted twice since. Front focussing again.

Is this a thing or am I just blessed?

[LeicaR10]

Derbyshire Man,   Good question.  But would you let us know what type of camera/lens you are using?    r/ Mark

[Derbyshire Man]

No problem, 50mm summilux latest version, 35mm summilux ASPH FLE mk1 and CV 28mm 2.0

All exhibit the same initial perfection and gradual shift to front focus.

I should maybe note that when i first go the camera, lens engagement was stiff and the RF mechanism didn't always engage with the lens resulting in only partial movement in the RF as the lens was focussed. That settled however within a few days.

[LeicaR10]

Derbyshire Man,  IMO, all three lenses should not be drfiting to front focus.  I think your RF needs to be serviced.  What model of M camera are you using?  The M10 RF was totally redesigned by Leica.  The M10 onwards the RF is far more reliiable and less prone to needing calibration.   If using earlier M cameras, digital or film, the RF mechanism  would need calibration on occasion.  Just my experience.  r/ Mark

Edited February 10, 2024 by LeicaR10

[DadDadDaddyo]

I agree with Leica R10's assessment. 

Though not an M10 with its newer rangefinder mechanism, I had experience with an issue something like this on my first Monochrom, which was an M246. In fact, I came to suspect that this issue was the reason the prior owner had put it up for private sale. 

The cam in the body that the back edge of the focusing sleeve on the lens engages with had developed a kind of "sticktion". If, while focusing, the lens pushed the cam backward into the body (which is normal) the cam developed a tendency to stick there, in other words, its freedom to pivot was attenuated to the extent that the energy stored in the return spring was unable to overcome the resistance; it just kind of got stuck there. You could move it with your finger, but the return spring couldn't do it on its own (as it should), though sometimes, then, it would overcome the resistance, resulting in a focus shift if the lens rim was not in contact with the cam.

The result of this, when focusing, was that as you brought the rangefinder images together, if you went a little past that and then tried to return, the lens rim would move forward but the cam would sometimes be left behind. Then the cam might decide later to slip forward, as if it was following the lens that had left it behind. 

Because I was judging focus by the rangefinder image, this sometimes resulted in a focus shift after I had finished focusing, especially if I'd gone "back and forth" a couple of times through the ideal focus spot. 

This may be, in part, why folks often advise practicing with the rangefinder to learn to turn the lens just to the point of convergence and not to go back and forth too much if you can help it. It takes practice, but it can be learned, and it's faster too.

Tolerances within the rangefinder are teenie, teenie, tiny. I mean, really teenie! Mechanical slop would be very bad to have here, but so would even a tiny bit of excess tightness in the mechanism; we're talking watch movement tolerances here, I suspect. 

Much cheaper and easier to put brains on a chip and build an electromechanical device designed to operate for a million or so actuation cycles. Make it plastic for consumers and more bulletproof for pros and charge accordingly. There! Annoying photography problem solved!

I remember reading in the 1970s that camera companies had figured out that the average length of time a typical  camera owner owned and used a camera was around three years, after which they'd want something new, even though the camera was working properly. Folks would just get bored. I recall thinking at the time, Man, they're going to start designing to that expected service life!

I had sometimes used the expression, "built like a camera, not like a toaster oven," but then realized I'd have to put conditions on that statement. 

Most things are built to last only a particular amount of time. Few are the examples of priority being placed on getting it really right for the long haul. 

As my kid sister the Classicist likes to say, "Sic transit gloria mundi..."

[lct]

18 hours ago, Derbyshire Man said:

No problem, 50mm summilux latest version, 35mm summilux ASPH FLE mk1 and CV 28mm 2.0

All exhibit the same initial perfection and gradual shift to front focus. [...]

In RF or LV mode? If in RF mode, the RF would be probably the culprit. If in LV mode, it could be your eye sight i'm afraid.

[Derbyshire Man]

Advertisement (gone after registration)

Yep, all focussed with RF. I've noticed that when I have the FR adjusted to get the focal point of lenses towards the back of the zone of focus but leaving just a little behind (rather than everything a millimetre beyond being out of focus and a couple of feet in front in focus at 1.4) I lose the ability to get infinity focus at all.

The oddest thing isn't just this but the way the focus adjustment drifts forwards across the days until it's adjusted back and then repeats the cycle.

[lct]

I would show your camera to a Leica rep if it is under warranty.

[Bikie John]

There are some threads about focussing problems scattered around (though not specifically with it drifting over time). There seems to be a strong body of opinion that it is best to always approach focus from the same direction, which in practice would probably be from infinity, and then stop as soon as it is achieved. Try not to overshoot and then adjust to compensate. This may well be to do with the stickiness mentioned by @DadDadDaddyo above.

John

[Derbyshire Man]

Thanks all.

After some experimentation it's evident that I can the RF adjusted so that stars (infinity) are a single point and then on all of the lenses the point of focus is 95% in front of the RF patch focus. Or I can have 35% behind and 75% in front but then I can no longer get stars to be a single point.

Is this a feature or a problem? 🤣

[lct]

Not sure what you mean by 35% behind and 75% in front but focusing should be accurate in both the RF and the photos.

[Derbyshire Man]

6 minutes ago, lct said:

Not sure what you mean by 35% behind and 75% in front but focusing should be accurate in both the RF and the photos.

When the RF is perfectly convergent, the field of focus over the point in the rangefinder (a mark on an oblique tape measure) shows the focal field acceptable 25% behind the mark and for about 75% in front. In other words the thing you want in focus is comfortably in focus. In this situation infinity (stars) show as double (only just).

In the situation where infinity is perfect (stars are indeed pin points) then in that situation, the point focussed on when the RF image is fully overlapping is right at the back of the zone of acceptable focus, so focussed but somewhat hazy with a large area of the ruler in front of that in perfect focus.

[lct]

I'm not good at using rulers but the subject matter should in focus in both OVF and EVF at the same aperture and subject distance normally.

[Derbyshire Man]

7 hours ago, lct said:

I'm not good at using rulers but the subject matter should in focus in both OVF and EVF at the same aperture and subject distance normally.

I suppose my summary point is that once you do that at close and mid distances infinity is not focussed.

[lct]

Last time i tried was on my M240 so i have no experience with the M11 sorry. Assuming nothing changed in the M11's RF, i would follow Adam Bonn's advice here:

 

[Jewl]

I had with my M11 a smiliar problem that I needed to hand in the M11 + main lens (35mm lux FLE) as the rangefinder + lens needed to be adjusted (front focus - not possible to focus on infinity / the moon). 

So far, knock on wood, it has not changed to the worse again! Meaning, for the same body/lens setup, one adjustment was sufficient. 

Edited February 12, 2024 by Jewl

[jaapv]

It is also a good thing to consider less likely, but banal causes. M M240 started misfocusing and it took me quite a while to discover that a piece of fluff had worked itself into the roller wheel.