Uneven exposure bands, top exposed properly rest underexposed

[Adam]

Advertisement (gone after registration)

Hi,

 

I have been shooting in low light using narrow aperture on an M 240 (Safari) plus 75  Summicron and noticed on two images exposure bands, where top of image seems exposed properly the rest underexposed. 

 

In all the other cases the pics were fine, even when shot  a second later with the same or similar settings (between f11 and f16, exposure 12-15 seconds).  I have been using this setup for a while and never had such problems. I also cannot replicate the results. BTW my card is Sandisk extreme pro 64, which has been reliable for over a year. I was also using an EVF.

 

Does anyone know what this could be?

 

I searched the forum, but I was not sure how to call the phenomenon. In this case a picture is a thousand words, here are the two examples. I would really appreciate any help!

 

Thanks.

 

Adam

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!

[edwardkaraa]

Looks like a shutter problem to me.

[Adam]

Looks like a shutter problem to me.

That was one of the things I was suspecting. Would it be so intermittent?

[edwardkaraa]

That was one of the things I was suspecting. Would it be so intermittent?

It goes through many stages before it fails, assuming it's the shutter.

[Adam]

It goes through many stages before it fails, assuming it's the shutter.

thanks!

[edwardkaraa]

It's starting to get worrying. Too many shutter problems recently. It seems that the M240 shutter isn't durable. Steve in the other thread had three shutter failures on the same body.

[Adam]

Advertisement (gone after registration)

It's starting to get worrying. Too many shutter problems recently. It seems that the M240 shutter isn't durable. Steve in the other thread had three shutter failures on the same body.

I just read the Seve's  thread. I hope it is not some systemic problem with the M-P/M 240, but rather, at worst, a bad batch of shutters.

 

I never had shutter problems before. My camera is a year old, around 16K actuations. So this is not as bad as Steve's case. 

But I did have my sensor replaced (not sure why, I sent it to New Jersey for cleaning, and they replaced the whole board. They never told me why.). If this is the shutter, this will be my second major repair within a year, and that is starting to get me worried.

[jdlaing]

What metering method were you using?

[Adam]

What metering method were you using?

Classic, center-weighed.

 

Because of the EVF Live view was on.

 

I use classic center weighed, or spot most of the time. But rarely EVF/Live view combo. Do you think it could be connected with the metering method? 

[ralphh]

Seems unlikely to me. Regardless of method, the camera starts the exposure from a closed shutter, it shouldn't affect whether or not you get two or three different exposures in one frame.

 

I don't suppose there was a sudden, short flash of light like lighting or similar just as you pressed the button that you failed to mention....?

 

In all seriousness tho, is the noise the same across the frame? Reason for asking is that if the top half has been boosted (or recorded to bright, if you like) by some weird card write error or firmware bug it'll have more noise. Look for a tone in the top sky and beach that are close to the same and see if the noise matches.

 

If it does match, they actually got the same amount of light even through they shouldn't have, and you probably had a shutter issue. If they don't match its electronic and could be your camera or card did a fart. I would probably replace the card unless it was uber expensive, then wouldn't worry unless it starts to happen again regularly.

 

If it is noiseless then push the exposure up a bunch and see what it looks like.

[jdlaing]

Classic, center-weighed.

 

Because of the EVF Live view was on.

 

I use classic center weighed, or spot most of the time. But rarely EVF/Live view combo. Do you think it could be connected with the metering method?

 

Possibly but I would have thought more so if classic than advanced as classic spot would meter off whatever part of the dark scene close to the center of the frame.

[Adam]

Seems unlikely to me. Regardless of method, the camera starts the exposure from a closed shutter, it shouldn't affect whether or not you get two or three different exposures in one frame.

 

I don't suppose there was a sudden, short flash of light like lighting or similar just as you pressed the button that you failed to mention....?

 

In all seriousness tho, is the noise the same across the frame? Reason for asking is that if the top half has been boosted (or recorded to bright, if you like) by some weird card write error or firmware bug it'll have more noise. Look for a tone in the top sky and beach that are close to the same and see if the noise matches.

 

If it does match, they actually got the same amount of light even through they shouldn't have, and you probably had a shutter issue. If they don't match its electronic and could be your camera or card did a fart. I would probably replace the card unless it was uber expensive, then wouldn't worry unless it starts to happen again regularly.

 

If it is noiseless then push the exposure up a bunch and see what it looks like.

 

Thanks!

 

I did consider card error, especially since on that day my card reported as full although it still had 33%. But there is no noise in the upper frame as far as I can see.

 

I also checked what happens when you increase exposure. I am attaching two images. The first image is the messed up one but exposure pushed 2.55 in Lightroom. The second image is an example of a shot just seconds later.

 

Note the sky at the top in the second image below has the same noise as in the initial messed up image (obviously the sky is blown in the image with exposure compensation).

 

The weird part is that the messed up image has longer exposure 1/12 sec than the non messed up one, which is brighter (1/15 sec).

 

But there were no bright flashes, UFO's, etc.  Just very windy. But I doubt that would mess up exposure...

 

I never had the problem before, nor did I replicate it today. Mind you these were two images from a set of 140, so maybe whatever is wrong is manifesting itself slowly.

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!

[Adam]

Possibly but I would have thought more so if classic than advanced as classic spot would meter off whatever part of the dark scene close to the center of the frame.

Thanks.

 

I changed metering to spot today, but the camera did not mess up.

 

For me it looks like the top of the image was exposed correctly and the rest underexposed.

[edwardkaraa]

Every thing is possible but I don't remember having seen such issue caused by a card or the sensor. M shutters are normally rated at 150k clicks, so 16k is still in the low range. LV probably puts a lot more pressure on the shutter though.

[Adam]

Every thing is possible but I don't remember having seen such issue caused by a card or the sensor. M shutters are normally rated at 150k clicks, so 16k is still in the low range. LV probably puts a lot more pressure on the shutter though.

 

I agree. My sensor was replaced last fall, not sure why. But it never misbehaved aside from the occasional freeze.

 

I rarely use LV - mostly for tripod shots and when using an EVF. We could say 1/3  of all shots were with LV, so  another 5K max. I never use advanced metering, which means roughly overall  20K, which is still in the low range. The shutter should not be misbehaving with such usage. 

[ralphh]

I have seen all kinds of effects, but the chances are slim - I don't really doubt it's a shutter issue, but worth making a few simple checks, esp as you hard other card issues that day

 

Another test for you. If it is a card / firmware fart it'll be pixel based, as where even at f16 the edge of the shutter blade will be soft.

 

Look at it at 100%. If exposure change has a razor sharp, one pixel to the next edge it's a data issue. If it's slightly feathered it's a shutter issue.

[Steve Ricoh]

Hi Adam,

I'm sorry to hear about this.

 

I suppose I'm somewhat preconditioned to lean towards shutter issues. That aside, the shutter runs vertically (I believe) so the banding in the image would be consistent with a shutter having variable speed.

 

It could be electronic, but i doubt the SD card is at fault. I would save the image on the current card as evidence (you may need it later) and change over to a different card.

 

With the shutter being the prime suspect, I would hit it with continuous operation at different shutter speeds to see whether normal operation occurs on every shot. Since I don't use LV, I have no idea whether continuous is possible in this mode, if it is I would 'hit' it here too.

 

You probably know this already, but just in case it's worth knowing that the camera keeps an error log within its internal memory, so on return (if it comes to this) Leica will be able to interrogate the log and diagnose the fault, if it is a camera issue.

 

You mention the SD card has been reliable for over a year, so I'm guessing the camera is of a similar age. Anyway 16-20k clicks should be nothing for a robust M240.

[sblutter]

LV may be the culprit.  Does it happen when that's off?

 

Shouldn't happen regardless - but may point to a solution.  Perhaps can be fixed in firmware

[Guest]

Adam,

 

I am bit confused .... nothing new with that state .... but in your original post exposures were 12 to 15 seconds ... in the latter post 1/12 to 1/15.

 

Are these two different times of day?

 

Consider that classic CW metering with a 75 is pretty much a large area as 1/3 of the image covers a good portion of the transition from dark cloud bank 

and distant hills to the much brighter sky.

 

If the proportion of sky to lower area increases with CW metering then the chance of underexposure is dramatically increased. So spot metering 

would allow for a better exposure.

 

The banding to me looks normal sensor related due to underexposure ... 

 

If the exposure times were on the order of 12 seconds the relative open to close time compared to the total time open would be

so small that I doubt it would show as dramatically as it appears here even if it were stepping. Hard to imagine that from 1/12 to 1/15

of a second would require 2.55 stops in post. So at these shutter speeds looks like the values are wrong.

 

The dynamic range of your scene is huge ... without a graduated filter or multiple exposures rendering it without underexposed foreground or

overexposed sky would be difficult.

 

Sorry for the run-on ... it is confusing at best.

 

Bob

[Advertisement]

Hallo,

Look here: Leica M

[ralphh]

Bob, I think you need to look at those images again. See how the sky is #1 is three distinct tones. It shouldn't be. Underexposure does not cause that.