90MM Apo-Summicron focus problem

[Quorrell]

Advertisement (gone after registration)

New M8 arrived this morning. Updated firmware to latest. Took a few pics of boring subjects with my Tri-Elmar 28-35-50 (not coded) all fine. Took some with my coded 90mm F2 APO noticed a marked focusing problem. I have used this lens with my M7's and its spot on. Has any body else found a problem with this lens and the M8? The picture enclosed was taken with flash on a tripod. You can clearly see the that the background object is sharper whilst the ink bottle which was the object focused on is blurred (aperture f4). No post processing of the image apart from converting to jpeg and cropping. No difference either with or with out lens detection

 

Cheers

 

Phill

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!

[Advertisement]

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

[pascal_meheut]

It is hard to focus precisely with such a lens. The 1.25 magnifier helps.

Also, if your M8 rangefinder is not spot on, you may very well not detect it with the TE at f/4 although it is obvious with the 90mm.

[Quorrell]

Cheers Pascal for that, I agree that it's hard to focus but I did use the 1.25 magnifier. The camera to subject distance was 125cm and the distance between the front of the bottle and the rear object was 35mm. I tried it with my Tri-Elmar and the results were perfect.

[pascal_meheut]

Try to check your rangefinder on 3 books on a shelf at 1m for instance. Put the left one 2cm further than the center one and the right one 2cm closer.

Focus on the edge of the center one as it is easier to be very precise then.

 

And check on the image where the focus is. Also, can you check the 90mm at infinity ? i.e. the moon for instance ? Or close infinity by shooting a TV antenna or something like that and bracketing the focus whilst on a tripod ?

[billh]

Yes, I have exactly the same back focus problem with my APO90 f2 and the M8. It also exists with some other lenses, especially the 50 f1.4 ASPH and APO135 f3.4. With my APO75 f2 the focus is exact. I do not have these focusing problems with film Ms and these lenses, and it is the camera, not my eyes, and it exists always, in repeated tests, with and without the 1.25x. It is enormously frustrating. Why do some lenses focus perfectly, while others do not? And why the difference on the M8 vs film cameras? It makes it impossible to know what to attempt fix, and the idea of sending the camera into never-never land seeking the possibility of a fix is not enticing. Eventually someone will figure out what is happening and hopefully post a fix for it.

 

Bil

[Englander]

Yes, I have had this problem but it only occured AFTER my lens was coded by Leica. I had some test shots with the M8 and the 90 AA from when I was evaluating the M8's accuracy but before I sent in the 90 for coding. When it was returned, the focusing was off. For the last several weeks It has been back to Leica for adjustment.

 

I use what has been called a complicated system for evaluating focus with the M: the camera is mounted on a macro focusing rail, it is focused as accurately as possible with a magnifier on a fixed resolution chart, and then the camera is moved back and forth in 5 or 10 mm increments without changing the original focus to evaluate how much the focus is off at 3 m. If I find the focus is off, I repeat the test for confirmation. A series of receding books can serve the same purpose but since I have the chart and the macro rail, I prefer to use them as they provide a repeatible test that allows subsequent comparisions.

 

I mentioned on this forum sometime ago that I noticed my 35/1.4 ASPH came back from coding with inaccurate focusing. I returned it and now it is perfectly accurate at 1.4, suffers typical focus shift on stop-down but never exceeds depth of field when evaluated at 100%. Of all the lenses I had coded--24/2.8, 35/1.4 ASPH, 50/2, 50/1.4 ASPH, and 90 AA--both the 35 and the 90 showed a change in focus that was almost identical, the others remained accurate. I do not know what procedure is used to change the mounts but something happens that can affect the focus. I do not know if lenses that are purchased new with coding are subject to the same variables.

 

If your focusing test is accurate, your lens should be focused wide-open at the place you focus. If it is not, you should look to DAG or Leica to adjust it.

 

Joe

[billh]

Advertisement (gone after registration)

Yes, I have had this problem but it only occured AFTER my lens was coded by Leica. I had some test shots with the M8 and the 90 AA from when I was evaluating the M8's accuracy but before I sent in the 90 for coding. When it was returned, the focusing was off. For the last several weeks It has been back to Leica for adjustment.

 

I use what has been called a complicated system for evaluating focus with the M: the camera is mounted on a macro focusing rail, it is focused as accurately as possible with a magnifier on a fixed resolution chart, and then the camera is moved back and forth in 5 or 10 mm increments without changing the original focus to evaluate how much the focus is off at 3 m. If I find the focus is off, I repeat the test for confirmation. A series of receding books can serve the same purpose but since I have the chart and the macro rail, I prefer to use them as they provide a repeatible test that allows subsequent comparisions.

 

I mentioned on this forum sometime ago that I noticed my 35/1.4 ASPH came back from coding with inaccurate focusing. I returned it and now it is perfectly accurate at 1.4, suffers typical focus shift on stop-down but never exceeds depth of field when evaluated at 100%. Of all the lenses I had coded--24/2.8, 35/1.4 ASPH, 50/2, 50/1.4 ASPH, and 90 AA--both the 35 and the 90 showed a change in focus that was almost identical, the others remained accurate. I do not know what procedure is used to change the mounts but something happens that can affect the focus. I do not know if lenses that are purchased new with coding are subject to the same variables.

 

If your focusing test is accurate, your lens should be focused wide-open at the place you focus. If it is not, you should look to DAG or Leica to adjust it.

 

Joe

 

Joe,

 

This seems to confirm what is the only plausible explanation to my mind - that somehow the focus of the actual lenses in question is the problem on the M8, not the camera.

 

Does anyone know what they do to adjust the focus in a lens (verses making an adjustment in a rangefinder on the camera)?

[Quorrell]

Thanks for these suggestions I've just done a quick test as suggested by Pascal with three books each book is actually 3cm thick I stacked them together then carefully focused on the edge of the middle. I took this shot at f2 as you can see the middle book and the top book are quite soft whilst the book at the back is fairly sharp. If you look closely you can see a double image on the edge of the middle book. I need to get out and test it on some real subjects but it is frustrating. The lens is only a couple of months old and coded and like I've said it's fine on my M7's see second pic although not a fair comparison it shows no sign of the problem.

 

Phill

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!

[Englander]

Bill,

They insert shims. I hypothesize that the shims were left out when my lenses were reassembled with the new coded mount.

 

The focusing mechanism and the lens unit are assembled separately. Lens units are matched to focusing mechanisms. A certain amount of manufacturing variation is allowed and that is minimized using shims to bring the match even closer. The use of shims in precision equipment is quite common. As Mark Norton pointed out, there are shims used to accurately position sensor in the M8, too.

 

Joe

[billh]

Bill,

They insert shims. I hypothesize that the shims were left out when my lenses were reassembled with the new coded mount.

 

The focusing mechanism and the lens unit are assembled separately. Lens units are matched to focusing mechanisms. A certain amount of manufacturing variation is allowed and that is minimized using shims to bring the match even closer. The use of shims in precision equipment is quite common. As Mark Norton pointed out, there are shims used to accurately position sensor in the M8, too.

 

Joe

 

Joe, where do they put shims in a lens? And do you know what they use for shims? I wonder if we can try this at home to correct a lens(es)?

[Englander]

Bill,

They are circular and I believe they were brass, but I do not know for certain what is being used now. They position the lens unit relative to the focusing unit. Any mismatch between the focusing unit--which moves the rangefinder cam and can show perfect focus at infinity, etc--and the lens unit is most apparent at close focus where depth of field is at its minimum.

 

Unlike previous Visoflex lenses and older 50 Summicrons, I think none of the recent lenses are meant to be disassembled by owners. I believe you are dealing with levels of precision that require a colliminator.

 

Joe