Used Leica Apo-Summicron-M 90 - Bokeh balls not round wide open. Is this normal?

[Xelamoon]

Advertisement (gone after registration)

I just received a new to me lens from MPB, and wanted to ask if that is normal or could be caused by maybe misaligned diaphragm? Maybe the lens opens too far?

Attached is a quick example shot of LEDs taken with my SL2-S (if that matters). Only downsizes a bit so that it passes the size limit. 

Third image is from my Tele Elmarit M to rule out the LEDs as a culprit.

Input of other lens owners is welcome.

 

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

[jankap]

One can observe:

Picture 1 and 3: Not all bokeh-balls are round.

Picture 2: The balls have an onion structure.

Picture 1 and 2: The balls don´t have a round edge.

The Leica APO-Summicron-M 90 has an element with one aspherical surface.

The Tele-Elmarit is of a classical design.

 

[Xelamoon]

Thanks for your observation, but they aren't all that helpful and only stating the obvious.

My question was not regarding the ORB, and the change of shape due to optical vignetting. 

It was specifically about the jagged balls of picture two.

[jankap]

The onion structure (picture 2) is clear?

The jagged edge can only be the diaphragm. The number of the edges gives the number of blades.

Wide open could be the problem and also the form of the blades. Have you compared the blades of both lenses? Wide open and at f/4?

[farnz]

I had a 90/2 APO-Summicron-M asph for about 10 years and wide open the bokeh balls were perfectly spherical.

It looks like the diaphragm on your lens is not opening fully and is obscuring the round barrel, which is what makes the bokeh balls spherical when the lens is wide open.  It might just be a piece of gunk that's got in behind the diaphragm and is stopping it from fully opening.

If it was me I'd be inclined to send the lens back to MPB if they didn't declare the problem.

Pete.

[Xelamoon]

vor 12 Stunden schrieb jankap:

The onion structure (picture 2) is clear?

The jagged edge can only be the diaphragm. The number of the edges gives the number of blades.

Wide open could be the problem and also the form of the blades. Have you compared the blades of both lenses? Wide open and at f/4?

Don’t get hung up on the comparison lens it doesn’t really matter. The shape of the blades should be of no consequence wide open as Leica lenses create a perfect circle wide open (as do almost all other lenses), except if the Apo Summicron is an outlier which I doubt. So it seems this a faulty lens

vor 7 Stunden schrieb farnz:

I had a 90/2 APO-Summicron-M asph for about 10 years and wide open the bokeh balls were perfectly spherical.

It looks like the diaphragm on your lens is not opening fully and is obscuring the round barrel, which is what makes the bokeh balls spherical when the lens is wide open.  It might just be a piece of gunk that's got in behind the diaphragm and is stopping it from fully opening.

If it was me I'd be inclined to send the lens back to MPB if they didn't declare the problem.

Pete.

Thanks for the info, that’s what I thought. I will consult the local Leica repair center on Monday and see what how much a service would be, as the price of the lens was very good.

Also I think the lens might be opening too wide because those jagged edges could be the shape that aperture is mounted to.

 

[adan]

Advertisement (gone after registration)

19 minutes ago, Xelamoon said:

The shape of the blades should be of no consequence wide open as Leica lenses create a perfect circle wide open (as do almost all other lenses), except if the Apo Summicron is an outlier which I doubt.

Not necessarily.

Leica over the years (and versions of the APO-Summicron can date back into the last century - 1998):

1. has NOT made all its (hand-assembled) lenses of a type identical. Check your focus ring, for example, to see the small number that indicates the EXACT focal length of your personal 90 APO. Which may be 16 (91.6mm) or 00 (90mm) or 10 (90.1mm) or 95 (89.5mm) or 09 (90.9mm) and so on.

The design specs for the 90mm APO-Summicron, as listed in the Leica data sheet, are for a true focal of 90.9mm (marked 09) - but many lenses will not match that.

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!

2. If the precise focal length is different (after assembly), obviously the wide-open diameter that equates precisely to "f/2.0" MUST be slightly different, to produce the correct exposure. 91.6 (f) divided by 2, equals an opening of 45.8mm; 89.5 (f) divided by 2 = 44.75 mm opening. 

3. Leica will build its machined metal lens barrels to a single size en masse (generally slightly oversized to cover all possibilities up to 45.8mm), and then "fit and adjust" the aperture mechanism/blades on a given lens to the exact glass optics being installed. For correct exposures - but not filling the barrel opening completely

I have just checked all eight of my Leica-built M lenses (various vintages 1970 to 2014) - and exactly one (1) of them does NOT show the aperture blades creeping into the edges of the lens opening at full aperture. Happens to be a 1990s 50mm f/1.4 pre-ASPH, with a marked focal length of 51.6mm.

Some are polygonal, others more round - that will depend on the exact number and curvature of the aperture blades.

Just one of the "joys" of getting lenses that are hand-crafted and hand-fitted, rather than robot-ground and assembled.

 

Now, it certainly could be out of adjustment also or instead. But don't be surprised if it gets returned as "within specifications" - for the exact pieces of glass it contains.

[M11 for me]

@adan Thank you for your most instructive explanation. You give me more to think about.

And did I understand you correctly: All lenses that are missing such a small number besides the "m" letter are not manually assembeled?

Edited 9 hours ago by M11 for me

[evikne]

29 minutes ago, adan said:

Not necessarily.

Leica over the years (and versions of the APO-Summicron can date back into the last century - 1998):

1. has NOT made all its (hand-assembled) lenses of a type identical. Check your focus ring, for example, to see the small number that indicates the EXACT focal length of your personal 90 APO. Which may be 16 (91.6mm) or 00 (90mm) or 10 (90.1mm) or 95 (89.5mm) or 09 (90.9mm) and so on.

The design specs for the 90mm APO-Summicron, as listed in the Leica data sheet, are for a true focal of 90.9mm (marked 09) - but many lenses will not match that.

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!

2. If the precise focal length is different (after assembly), obviously the wide-open diameter that equates precisely to "f/2.0" MUST be slightly different, to produce the correct exposure. 91.6 (f) divided by 2, equals an opening of 45.8mm; 89.5 (f) divided by 2 = 44.75 mm opening. 

3. Leica will build its machined metal lens barrels to a single size en masse (generally slightly oversized to cover all possibilities up to 45.8mm), and then "fit and adjust" the aperture mechanism/blades on a given lens to the exact glass optics being installed. For correct exposures - but not filling the barrel opening completely

I have just checked all eight of my Leica-built M lenses (various vintages 1970 to 2014) - and exactly one (1) of them does NOT show the aperture blades creeping into the edges of the lens opening at full aperture. Happens to be a 1990s 50mm f/1.4 pre-ASPH, with a marked focal length of 51.6mm.

Some are polygonal, others more round - that will depend on the exact number and curvature of the aperture blades.

Just one of the "joys" of getting lenses that are hand-crafted and hand-fitted, rather than robot-ground and assembled.

 

Now, it certainly could be out of adjustment also or instead. But don't be surprised if it gets returned as "within specifications" - for the exact pieces of glass it contains.

Thank you! Then I've learned something new today.

[Xelamoon]

vor 2 Stunden schrieb adan:

Not necessarily.

Leica over the years (and versions of the APO-Summicron can date back into the last century - 1998):

1. has NOT made all its (hand-assembled) lenses of a type identical. Check your focus ring, for example, to see the small number that indicates the EXACT focal length of your personal 90 APO. Which may be 16 (91.6mm) or 00 (90mm) or 10 (90.1mm) or 95 (89.5mm) or 09 (90.9mm) and so on.

The design specs for the 90mm APO-Summicron, as listed in the Leica data sheet, are for a true focal of 90.9mm (marked 09) - but many lenses will not match that.

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!

2. If the precise focal length is different (after assembly), obviously the wide-open diameter that equates precisely to "f/2.0" MUST be slightly different, to produce the correct exposure. 91.6 (f) divided by 2, equals an opening of 45.8mm; 89.5 (f) divided by 2 = 44.75 mm opening. 

3. Leica will build its machined metal lens barrels to a single size en masse (generally slightly oversized to cover all possibilities up to 45.8mm), and then "fit and adjust" the aperture mechanism/blades on a given lens to the exact glass optics being installed. For correct exposures - but not filling the barrel opening completely

I have just checked all eight of my Leica-built M lenses (various vintages 1970 to 2014) - and exactly one (1) of them does NOT show the aperture blades creeping into the edges of the lens opening at full aperture. Happens to be a 1990s 50mm f/1.4 pre-ASPH, with a marked focal length of 51.6mm.

Some are polygonal, others more round - that will depend on the exact number and curvature of the aperture blades.

Just one of the "joys" of getting lenses that are hand-crafted and hand-fitted, rather than robot-ground and assembled.

 

Now, it certainly could be out of adjustment also or instead. But don't be surprised if it gets returned as "within specifications" - for the exact pieces of glass it contains.

Thank you very much for the detailed explanation. Mine says 07, so 90.7mm which should be really close to the 90.9mm, right?

Also isn't 10 (91.0mm) and not 90.1mm? And the 95 is interesting, because I was wondering what the high number on my Elmarit-C meant, as it could of course not be 99.5mm.

 

[adan]

4 hours ago, M11 for me said:

And did I understand you correctly: All lenses that are missing such a small number besides the "m" letter are not manually assembeled?

Not at all! So far as I know, Leica hand-assembles all the M lenses - part of the cachet (and price).

The main reason for those numbers is being sure the lens focus is correctly calibrated for "proxy" rangefinder focusing, with longer focal lengths (50mm through 135mm) that have narrow DoF and higher magnification - should there be a future need for disassembling/reassembling the lens for repairs or cleaning.

Wideangle M lenses with larger inherent DoF and lower magnification don't ever need them. Nor do later R-system lenses, which are focused "through the lens" where the photographer's eye corrects for any variation in exact focus/focal-length needed.

Leica eventually realized it was not critical in that case, and the engraving was a useless expense. It was (apparently) also not used on the "discount" Summarit f/2.4-2.5 50/75/90 lenses.

3 hours ago, Xelamoon said:

Thank you very much for the detailed explanation. Mine says 07, so 90.7mm which should be really close to the 90.9mm, right?

Also isn't 10 (91.0mm) and not 90.1mm? And the 95 is interesting, because I was wondering what the high number on my Elmarit-C meant, as it could of course not be 99.5mm.

Don't forget that while the optical design spec is for 90.9mm - the metal lens barrel opening has to be able to accomodate lenses up to 91.6mm. Almost 1mm larger than your 90.7 actual uses, so the aperture blades/polygon will be visible at f/2.0.

You are correct that the mark "10" means 91.0mm focal length. I just lost my place, juggling a reference book and my keyboard in the pre-dawn hours. 😰

[NigelG]

Have you looked to see if it’s possible to turn the aperture ring a tiny bit past f2.0 and fully retract the blades?
I’m pretty sure I had a 90 (maybe one of the non-APO Summicrons I used to have) where the blades were not fully retracted at f2.0.

I used to put a tiny bit of pressure on the aperture ring with the side of a finger to hold the blades fully open if I knew there would be bokeh balls otherwise they “crept” into the shape like yours.

[Xelamoon]

vor 48 Minuten schrieb adan:

Don't forget that while the optical design spec is for 90.9mm - the metal lens barrel opening has to be able to accomodate lenses up to 91.6mm. Almost 1mm larger than your 90.7 actual uses, so the aperture blades/polygon will be visible at f/2.0.

You are correct that the mark "10" means 91.0mm focal length. I just lost my place, juggling a reference book and my keyboard in the pre-dawn hours. 😰

Thank you for taking the time and your detailed explanation! Makes sense, and also interesting that this isn't more widely discussed. 

vor 33 Minuten schrieb NigelG:

Have you looked to see if it’s possible to turn the aperture ring a tiny bit past f2.0 and fully retract the blades?
I’m pretty sure I had a 90 (maybe one of the non-APO Summicrons I used to have) where the blades were not fully retracted at f2.0.

I used to put a tiny bit of pressure on the aperture ring with the side of a finger to hold the blades fully open if I knew there would be bokeh balls otherwise they “crept” into the shape like yours.

It actually gets worse if I overturn the aperture ring, and looks better if I move it just a hair towards 2.8. Which led me to believe that the retracted blade mount is visible.