Jump to content

Risked life and limb to get this shot and the $7000 piece of crap let me down again

Guest NEIL-D-WILLIAMS

By Guest NEIL-D-WILLIAMS
in Leica M (Type 240 / 262)

Recommended Posts

  • Replies 231
  • Created
  • Last Reply
DigitalHeMan

DigitalHeMan

Posted
Posted
Wait so you're saying that this method of testing is definitive?! Stopping a lens down with an ND filter and a lens cap is "SURE" to block all incoming light?! Oh wait, you're shining the light into only the lens mount area. Hmm...unless it's a laser and you can guarantee no reflected light, how can you prove light is not entering through the lens front?

 

The lens cap was never designed to block 100% of incoming light.

 

 

 

Not trying to be negative, but I really don't see how this "test" proves that light is leaking through the mount.

 

 

 

I think the issues are:

 

1) is this really a light leak at the mount?

 

2) if 1 is true, and I have yet to be convinced, then is it a flaw or was the camera simply not designed for this type of photography (enter examples of DLSRS (yes, I know it's a different design from a rangefinder, that's not really the point -- intended use for these cameras is probably the same)

 

3) if a flaw, is it only with some Ms or all?

 

 

 

But until you can prove 1, the whole thread is pointless.

 

I'm amazed at how quick people are lead to believe it "must" be a mount leak, because they've "tested" it.

 

 

Doesn't the fact that the 'problem' can be fixed by putting a scrunchie around the mount kind of prove that that is where the light leak is coming from? :)

Link to post
Share on other sites
kdriceman

kdriceman

Posted
Posted

Advertisement (gone after registration)

Wait so you're saying that this method of testing is definitive?! Stopping a lens down with an ND filter and a lens cap is "SURE" to block all incoming light?! Oh wait, you're shining the light into only the lens mount area. Hmm...unless it's a laser and you can guarantee no reflected light, how can you prove light is not entering through the lens front?

 

The lens cap was never designed to block 100% of incoming light.

 

 

 

Not trying to be negative, but I really don't see how this "test" proves that light is leaking through the mount.

 

 

 

I think the issues are:

 

1) is this really a light leak at the mount?

 

2) if 1 is true, and I have yet to be convinced, then is it a flaw or was the camera simply not designed for this type of photography (enter examples of DLSRS (yes, I know it's a different design from a rangefinder, that's not really the point -- intended use for these cameras is probably the same)

 

3) if a flaw, is it only with some Ms or all?

 

 

 

But until you can prove 1, the whole thread is pointless.

 

I'm amazed at how quick people are lead to believe it "must" be a mount leak, because they've "tested" it.

 

 

I feel like I'm communicating with my wife.

Link to post
Share on other sites
Surge

Surge

Posted
Posted

I tried the test you described, and my camera is fine. No light leaks from anywhere. And while your test can lead to false positives, it won't lead to false negatives. Look that up if unsure of the meaning.

 

What this says is that there is a problem with SOME units (assuming you believe it IS leaking from the mount).

 

So what you're saying is that not all the M cameras and/or the lens mounts are perfect. Why is that such a revelation of Biblical proportions?

 

I wouldn't call it a "$7000 piece of crap" because your example is less than perfect.

 

Return it for repair. End of story.

 

The way I conducted the test is:

- 28mm M lens at f/22, hood and cap on (no ND filter, why introduce a 3rd-party variable?)

- shutter at 8 seconds / 200 ISO

- camera in pitch black room

- exposure 1 was with no light, "reference image"

- exposures 2 and 3 was with a bright LED penlight shining directly at the mount, at approx. 12 o'clock (when looking from above) and 6 o'clock

 

Low and behold, exposures 2 and 3 looked identical to the reference image. i.e., Black!

 

So go and get yours fixed, if you're sure it's not light leaking from the ND filter. Or buy a 2c hair band and forget it.

 

End of story! I think some of you would rather argue and whine then just move on with your lives and enjoy the hobby.

Link to post
Share on other sites
250swb

250swb

Posted
Posted
I tried the test you described, and my camera is fine. No light leaks from anywhere. And while your test can lead to false positives, it won't lead to false negatives. Look that up if unsure of the meaning.

 

What this says is that there is a problem with SOME units (assuming you believe it IS leaking from the mount).

 

So what you're saying is that not all the M cameras and/or the lens mounts are perfect. Why is that such a revelation of Biblical proportions?

 

I wouldn't call it a "$7000 piece of crap" because your example is less than perfect.

 

Return it for repair. End of story.

 

The way I conducted the test is:

- 28mm M lens at f/22, hood and cap on (no ND filter, why introduce a 3rd-party variable?)

- shutter at 8 seconds / 200 ISO

- camera in pitch black room

- exposure 1 was with no light, "reference image"

- exposures 2 and 3 was with a bright LED penlight shining directly at the mount, at approx. 12 o'clock (when looking from above) and 6 o'clock

 

Low and behold, exposures 2 and 3 looked identical to the reference image. i.e., Black!

 

So go and get yours fixed, if you're sure it's not light leaking from the ND filter. Or buy a 2c hair band and forget it.

 

End of story! I think some of you would rather argue and whine then just move on with your lives and enjoy the hobby.

 

 

 

What is the point of an 8 second exposure? You've already been told it is a 30 second exposure + that can replicate the problem. That the OP had the problem at only 8 seconds is possibly caused by hanging a great big heavy Noctilux off the front of the camera that opens the gap up, not a light weight 28mm lens. Good grief, will you ever read what's been said already?

 

Steve

Link to post
Share on other sites
AlanG

AlanG

Posted
Posted
Wait so you're saying that this method of testing is definitive?! Stopping a lens down with an ND filter and a lens cap is "SURE" to block all incoming light?! Oh wait, you're shining the light into only the lens mount area. Hmm...unless it's a laser and you can guarantee no reflected light, how can you prove light is not entering through the lens front?

The lens cap was never designed to block 100% of incoming light.

 

Oh, I also forgot to mention that I painted both the front and back elements with lithographer's black opaque... just to be sure.

Link to post
Share on other sites
kdriceman

kdriceman

Posted
Posted
I tried the test you described, and my camera is fine. No light leaks from anywhere. And while your test can lead to false positives, it won't lead to false negatives. Look that up if unsure of the meaning.

 

What this says is that there is a problem with SOME units (assuming you believe it IS leaking from the mount).

 

So what you're saying is that not all the M cameras and/or the lens mounts are perfect. Why is that such a revelation of Biblical proportions?

 

I wouldn't call it a "$7000 piece of crap" because your example is less than perfect.

 

Return it for repair. End of story.

 

The way I conducted the test is:

- 28mm M lens at f/22, hood and cap on (no ND filter, why introduce a 3rd-party variable?)

- shutter at 8 seconds / 200 ISO

- camera in pitch black room

- exposure 1 was with no light, "reference image"

- exposures 2 and 3 was with a bright LED penlight shining directly at the mount, at approx. 12 o'clock (when looking from above) and 6 o'clock

 

Low and behold, exposures 2 and 3 looked identical to the reference image. i.e., Black!

 

So go and get yours fixed, if you're sure it's not light leaking from the ND filter. Or buy a 2c hair band and forget it.

 

End of story! I think some of you would rather argue and whine then just move on with your lives and enjoy the hobby.

 

yawn.

Link to post
Share on other sites
MarkP

MarkP

Posted
Posted

Hypothesis:

light leak through lens mount flange where lens meets camera on digital M-mount cameras. Of significance during prolonged daylight exposure

 

Equipment & Methods:

M9 ISO 160, Monochrom ISO 320, M240 ISO 200

2.0/28 Summicron ASPH, 1.4/50 Summilux ASPH, 4.0/90 Macro-Elmar ASPH

 

Lenses all set to smallest aperture, no filters, and sealed to prevent stray light entering through front lens element. Rangefinder windows obscured on cameras.

 

LedLenser M7 torch directed to:

lens mount flange at 12 oclock, 2-3 o'clock, below the rangefinder window.

rangefinder windows

Bulb at 8 sec and 45 sec

With and without daughter's black scrunchie at base of lens covering lens mount flange.

 

Results:

Light leak artefact to lower right corner of image at 45 sec exposure clearly demonstrated with quite marked sensor blooming artefact on all cameras with all lenses. Artefact more apparent with light exposure to 2-3 o'clock than 12 o'clock on lens mount.

 

Artefact was not visualised at 8 sec exposure with any lens on any camera at both exposures but interestingly, although the image appeared black with no artefact the camera itself identified light reaching the lower right hand corner of the image. The blue underexposure warning on the M240 lit up all of the screen blue except the bottom right hand corner, and some signal on the histogram in the 0-8 range, suggesting some light did leak through even at 8 seconds (but not enough to be clinically relevant).

 

Intense light exposure to any of the rangefinder windows did not replicate the artefact.

 

External obstruction of lens flange to light source with a black cloth-covered hair elastic completely prevented the artefact.

 

Discussion:

A light leak resulting in a consistent and reproducible artefact was clearly demonstrated in controlled conditions affecting three Leica cameras and three lenses (all relatively new and with no damage) with more prolonged exposure to intense light at the lens mount/camera mount interface. Even at 8 seconds the camera's underexposure warning identified light exposure to the bottom right corner of the image although as this was not visually apparent. A slight increase in exposure of up to about 8 (on a 0-255 scale) would probably not be identified on an otherwise properly exposed file.

 

The light leak artefact was completely prevented by the use of a black cloth-covered scrunchie)

 

There is no light leak through rangefinder windows (see previously published experiments. This exact artefact pattern is also reproducible by other users in both real world (long exposure in sunlight) and experimentally.

 

Possible site of light leak is via the camera lens mount screw holes where there is only about 1mm or so solid flat metal on each side of screw hole. Screw holes in the lens mount do not overlap those on the camera body lens mount, but the metal on each side of the screw holes on lens mount is thicker than that on the camera body.

 

Interestingly the lens mount flange on Zeiss lenses is not of equal depth for the full width of the lens mount with a small step down around the outer part of the mount which may explain why my Zeiss lenses demonstrate worse artefacts at similar exposure times (unpublished data).

 

This is highly reproducible across cameras and lenses in this experiment, my previously published experiment, and with numerous other users.

This may not occur on all cameras and all lenses as tolerances may vary slightly with individual cameras and lenses, and with use.

 

Conclusion:

Stupid lens mount design for long exposures in direct bright light :p

Suggestions - use a black scrunchie and shield the lens mount from direct sunlight during long exposures such as with a x10 ND filter.

 

quod erat demonstrandum

 

 

Nevertheless Mr Surge Engineer (no offence to other engineer forum members) , read carefully what we have written and go and do your experiment more carefully I enjoy my hobby but when a days shooting was ruined by this artefact I wanted to know what it was and how to fix it.

Link to post
Share on other sites
kdriceman

kdriceman

Posted
Posted
Hypothesis:

light leak through lens mount flange where lens meets camera on digital M-mount cameras. Of significance during prolonged daylight exposure

 

Equipment & Methods:

M9 ISO 160, Monochrom ISO 320, M240 ISO 200

2.0/28 Summicron ASPH, 1.4/50 Summilux ASPH, 4.0/90 Macro-Elmar ASPH

 

Lenses all set to smallest aperture, no filters, and sealed to prevent stray light entering through front lens element. Rangefinder windows obscured on cameras.

 

LedLenser M7 torch directed to:

lens mount flange at 12 oclock, 2-3 o'clock, below the rangefinder window.

rangefinder windows

Bulb at 8 sec and 45 sec

With and without daughter's black scrunchie at base of lens covering lens mount flange.

 

Results:

Light leak artefact to lower right corner of image at 45 sec exposure clearly demonstrated with quite marked sensor blooming artefact on all cameras with all lenses. Artefact more apparent with light exposure to 2-3 o'clock than 12 o'clock on lens mount.

 

Artefact was not visualised at 8 sec exposure with any lens on any camera at both exposures but interestingly, although the image appeared black with no artefact the camera itself identified light reaching the lower right hand corner of the image. The blue underexposure warning on the M240 lit up all of the screen blue except the bottom right hand corner, and some signal on the histogram in the 0-8 range, suggesting some light did leak through even at 8 seconds (but not enough to be clinically relevant).

 

Intense light exposure to any of the rangefinder windows did not replicate the artefact.

 

External obstruction of lens flange to light source with a black cloth-covered hair elastic completely prevented the artefact.

 

Discussion:

A light leak resulting in a consistent and reproducible artefact was clearly demonstrated in controlled conditions affecting three Leica cameras and three lenses (all relatively new and with no damage) with more prolonged exposure to intense light at the lens mount/camera mount interface. Even at 8 seconds the camera's underexposure warning identified light exposure to the bottom right corner of the image although as this was not visually apparent. A slight increase in exposure of up to about 8 (on a 0-255 scale) would probably not be identified on an otherwise properly exposed file.

 

The light leak artefact was completely prevented by the use of a black cloth-covered scrunchie)

 

There is no light leak through rangefinder windows (see previously published experiments. This exact artefact pattern is also reproducible by other users in both real world (long exposure in sunlight) and experimentally.

 

Possible site of light leak is via the camera lens mount screw holes where there is only about 1mm or so solid flat metal on each side of screw hole. Screw holes in the lens mount do not overlap those on the camera body lens mount, but the metal on each side of the screw holes on lens mount is thicker than that on the camera body.

 

Interestingly the lens mount flange on Zeiss lenses is not of equal depth for the full width of the lens mount with a small step down around the outer part of the mount which may explain why my Zeiss lenses demonstrate worse artefacts at similar exposure times (unpublished data).

 

This is highly reproducible across cameras and lenses in this experiment, my previously published experiment, and with numerous other users.

This may not occur on all cameras and all lenses as tolerances may vary slightly with individual cameras and lenses, and with use.

 

Conclusion:

Stupid lens mount design for long exposures in direct bright light :p

Suggestions - use a black scrunchie and shield the lens mount from direct sunlight during long exposures such as with a x10 ND filter.

 

quod erat demonstrandum

 

 

Nevertheless Mr Surge Engineer (no offence to other engineer forum members) , read carefully what we have written and go and do your experiment more carefully I enjoy my hobby but when a days shooting was ruined by this artefact I wanted to know what it was and how to fix it.

 

Bravo Mark. You're more patient than I. And I suspect if you increase exposure by 2-3 stops on the 8 second exposure you will see evidence of 'The Artifact' in the lower right corner of the image as would Mr. Surge if he actually performed the experiment.

Link to post
Share on other sites
jdsheepdog

jdsheepdog

Posted
Posted

Two responses that sum up the case for filter leaks and reflected light:

 

I've made up my mind, don't confuse me with facts.

You don't think I will allow mere proof to sway my opinion.

 

Since my daughters are grown and far away, and my wife wears her (very attractive) hair short, I am in a quandary as to how to get a "Scrunchie." I may have to wait until Leica adopts my previous suggestion.

 

An "a la carte" Scrunchie to match the camera leather, now that sounds like serious money. :)

Link to post
Share on other sites
AlanG

AlanG

Posted
Posted

FWIW I did some tests with the Nex 6. I used a lens cap and wrapped black tape around it so all light through the lens would be blocked. (I ran out of black paint.)

 

I have a very bright LED bike light and could get a reproducible light leak when holding the light at the left side of the lens mount using 6400 ISO and a 30 second exposure. When holding the light at the top or bottom of the mount, the frame was black. Perhaps a light held on the right side would also cause a problem but the grip kept me from placing the light directly perpendicular to the lens mount so I didn't try it. Besides I only cared about finding a single place on the mount that leaked.

 

At ISO 400 and 30 seconds of exposure, lit the same way, the frame was also black. I also tested my Canon 5DII the same way at 6400 ISO and could not get any example of a leak. (I did not use a weather-proofed lens that has a rubber seal.) Of course one could spend a lot of time testing various cameras and lenses. I suggest you do so at the highest ISO for at least 30 seconds to see if there is any leakage, before trying to pin it down any further.

 

Presumably a bright enough light aimed at the right spot for a long enough time at a high enough ISO are factors that need to be considered to see if there may be a "real world" problem with a specific camera/lens combo.

Link to post
Share on other sites

Archived

This topic is now archived and is closed to further replies.

Go to topic listing

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...