Pages: [1]   Go Down

Author Topic: Sony A7R Potential Light Leak (pictures)  (Read 10955 times)

Graham Clark

  • Full Member
  • ***
  • Offline Offline
  • Posts: 179
    • grahamclarkphoto.com
Sony A7R Potential Light Leak (pictures)
« on: March 08, 2014, 02:12:59 am »

I'm wondering if you guys can take a look at the following image to help determine if this is the cause of a potential light leak:


Click here to download the .ARW, .TIFF and hi-res JPEG of the Sony A7R Light Leak image





Notice how when the composition is every so slightly changed it's nearly nonexistent? But the presence of any kind of leak would have a negative impact on saturation and contrast.

I'll certainly test this tomorrow, but at first glance the flare appears to be from something metal. Notice the coloration within the flare? It also has the texture of metal as opposed to something translucent.

Graham
« Last Edit: March 08, 2014, 02:58:03 am by Graham Clark »
Logged
Graham Clark  |  grahamclarkphoto.com

LKaven

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1060
Re: Sony A7R Potential Light Leak (pictures)
« Reply #1 on: March 08, 2014, 02:32:43 am »

This is consistent with some cases of light-leak through the lens barrel, noticed especially during time exposures of course.  I've seen other cases of near-IR interference from vibration-reduction sensors. 

What lens were you using here?  Can you reproduce consistently with this lens if, for example, you shine a light directly on the lens barrel during time exposures?  Does it help if you cover the lens?  Are other lenses immune?  Etc.

I would stick with the lens hypothesis here unless you can consistently show it with all lenses.  No optical viewfinder, so no chance for pollution through the eyepiece.

Graham Clark

  • Full Member
  • ***
  • Offline Offline
  • Posts: 179
    • grahamclarkphoto.com
Re: Sony A7R Potential Light Leak (pictures)
« Reply #2 on: March 08, 2014, 02:41:22 am »

This is consistent with some cases of light-leak through the lens barrel, noticed especially during time exposures of course.  I've seen other cases of near-IR interference from vibration-reduction sensors. 

What lens were you using here?  Can you reproduce consistently with this lens if, for example, you shine a light directly on the lens barrel during time exposures?  Does it help if you cover the lens?  Are other lenses immune?  Etc.

I would stick with the lens hypothesis here unless you can consistently show it with all lenses.  No optical viewfinder, so no chance for pollution through the eyepiece.

This is with the Metabones EF adapter. Having just cleaned the sensor today on the A7R I noticed that there's wear occurring on the top section of the thread mount on the A7R body. I'll take a macro shot tomorrow, but that's where the light leak may be coming from - the top connection point between the body and lens on the Metabones EF adapter.

Graham
Logged
Graham Clark  |  grahamclarkphoto.com

CptZar

  • Full Member
  • ***
  • Offline Offline
  • Posts: 157
Re: Sony A7R Potential Light Leak (pictures)
« Reply #3 on: March 08, 2014, 03:36:29 am »

Looks like lens flare to me.
« Last Edit: March 09, 2014, 04:48:49 am by CptZar »
Logged

Robert DeCandido PhD

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 271
    • http://www.BirdingBob.com
Re: Sony A7R Potential Light Leak (pictures)
« Reply #4 on: March 11, 2014, 09:06:17 pm »

Sony light leak investigated (Roger at LensRentals.com:

http://www.fredmiranda.com/forum/topic/1279812
Logged

RawheaD

  • Full Member
  • ***
  • Offline Offline
  • Posts: 145
Re: Sony A7R Potential Light Leak (pictures)
« Reply #5 on: March 12, 2014, 01:22:29 am »

Like I suggested to Graham elsewhere, I think the upper left is lens flare; the upper right I isolated definitively to leak from the mount of the camera body.

The story's been quietly making the rounds, starting with Imaging Resources, and now Lens Rentals.  It all actually started over two months ago when I first reported the problem on the SONY Alpha group on Facebook.  Since I do a lot of daytime long exposure photography, this was disturbing, so I suggested people test for it, and came up with a rather crude, quick and dirty method; set ISO to super high, lens cap on, 30s exposure mode, then shine a flashlight around the mount area.  With this method, I was able to isolate the location of the worst leak (3 O'clock position), and most importantly, that the leak was happening not between the lens & flange, but from the body itself, the gap between the silver flange that interfaces the lens, and the orange-colored outer ring.  By applying some gaffer's tape to cover that, the leak went away.

Now, having several years of doing daytime long exposure photography, I believe I've come to understand where we should and shouldn't expect light to leak.

• If you're using an SLR, you want to cap your OVF, especially when you have the sun behind you, as light will seep into the body via the mirror-up position.

• If you're using a zoom lens, or worse, a tilt-shift lens, you want to cover your lens barrel, as those moving parts have small gaps through which light can find their way inside.

• Doors to access various ports (usb, flash sync, remote, etc.) can leak light if left open. Close them, or if you are using a wired remote, cover that area.

• If you're using Lee, Cokin, Formatt-Hitech, etc., filter holder system, make sure there is no gap between the holder and the ND filter (E.g, Big Stopper), as light will find its way through there as we..

*Never* had I needed to put tape around the mount of any of my cameras to keep light out.  I've owned a K200D, K20D, K7, 5D, 5Dmk2, (also GF-1 and NEX-6, but never did LE with those).  I've also done daytime LE with film, with various Hasselblads, Rolleiflex SL66, Kiev-friggin-60, Speed Graphic, Chamonix 045N-2; never have I seen light leaking from the mount/flange area.

The only place that consistently caused leaks is the VF.  I've many shots ruined because I forgot to cap it, or forgot to fold down the waist level finder.  I've had a couple with my filter system, where, using a Lee Big Stopper with a Cokin-Z, the light seal isn't perfect.  And I experienced some light leaking through tilt shift lenses a few times, with my Hartblei Super Rotator.  But I've never seen light coming in from the mount.

And yet, my A7R is consistent.  It doesn't matter which way it's pointed.  If shot for more than 60 seconds at ISO50 in daylight, I will always get a leak showing up as the crescent shaped ghost in the upper right of the frame.

I find this very unusual.

And yet, partially because for some reason, this "ISO25,600, 30sec in broad daylight" meme sprung up, and these websites are treating that as some golden standard to mock and laugh at.  Rather disheartening.

Here's what I would consider a VERY modest daytime long exposure shot. 76 seconds at ISO50.  The second one I agree is a bit more extreme, it's 8.5 minutes at ISO100.  I'd still not consider that excessive, though.  With my Canon 5Dmk2 I could easily shoot 5-10 minutes without any leakage as long as I remember to cap the OVF.  But considering there's all sorts of things happening in that image, I'm OK with accepting that one to be extreme.  But 76s @ ISO50??  I just can't accept that.

I would however love to hear your experiences; if you've ever had to deal with light leaks in your long exposure photographs (landscape, seascape, etc.) NOT from the lens or the OVF, but specifically from the flange area, please share your experience.  Thanks!

RawheaD

  • Full Member
  • ***
  • Offline Offline
  • Posts: 145
Re: Sony A7R Potential Light Leak (pictures)
« Reply #6 on: March 12, 2014, 02:05:26 am »

Graham: I see now in the metadata that your image was 30s @ ISO50!

See, I'm so convinced.  That crescent you see on the upper right I'm 99% positive comes in from the flange area.

The top left probably has something to do with Metabones.  We're using the same setup here; A7R + Metabones 3 + 17-40F4L.  The characteristic of the flare on the upper left between yours and mine are very similar, and it's one that I've never seen with the 17-40F4L on 5Dmk2, ergo, I believe it's light reflecting inside the Metabones adapter.

But the crescent on the top right is very similar to what we've all been seeing; leak from the 3 O'clock area of the mount.

And if it's showing up at ISO50, 30s, there's no way anyone could argue that's within spec.

Graham Clark

  • Full Member
  • ***
  • Offline Offline
  • Posts: 179
    • grahamclarkphoto.com
Re: Sony A7R Potential Light Leak (pictures)
« Reply #7 on: March 12, 2014, 02:17:05 am »

Like I suggested to Graham elsewhere, I think the upper left is lens flare; the upper right I isolated definitively to leak from the mount of the camera body.

The story's been quietly making the rounds, starting with Imaging Resources, and now Lens Rentals.  It all actually started over two months ago when I first reported the problem on the SONY Alpha group on Facebook.  Since I do a lot of daytime long exposure photography, this was disturbing, so I suggested people test for it, and came up with a rather crude, quick and dirty method; set ISO to super high, lens cap on, 30s exposure mode, then shine a flashlight around the mount area.  With this method, I was able to isolate the location of the worst leak (3 O'clock position), and most importantly, that the leak was happening not between the lens & flange, but from the body itself, the gap between the silver flange that interfaces the lens, and the orange-colored outer ring.  By applying some gaffer's tape to cover that, the leak went away.

Now, having several years of doing daytime long exposure photography, I believe I've come to understand where we should and shouldn't expect light to leak.

• If you're using an SLR, you want to cap your OVF, especially when you have the sun behind you, as light will seep into the body via the mirror-up position.

• If you're using a zoom lens, or worse, a tilt-shift lens, you want to cover your lens barrel, as those moving parts have small gaps through which light can find their way inside.

• Doors to access various ports (usb, flash sync, remote, etc.) can leak light if left open. Close them, or if you are using a wired remote, cover that area.

• If you're using Lee, Cokin, Formatt-Hitech, etc., filter holder system, make sure there is no gap between the holder and the ND filter (E.g, Big Stopper), as light will find its way through there as we..

*Never* had I needed to put tape around the mount of any of my cameras to keep light out.  I've owned a K200D, K20D, K7, 5D, 5Dmk2, (also GF-1 and NEX-6, but never did LE with those).  I've also done daytime LE with film, with various Hasselblads, Rolleiflex SL66, Kiev-friggin-60, Speed Graphic, Chamonix 045N-2; never have I seen light leaking from the mount/flange area.

The only place that consistently caused leaks is the VF.  I've many shots ruined because I forgot to cap it, or forgot to fold down the waist level finder.  I've had a couple with my filter system, where, using a Lee Big Stopper with a Cokin-Z, the light seal isn't perfect.  And I experienced some light leaking through tilt shift lenses a few times, with my Hartblei Super Rotator.  But I've never seen light coming in from the mount.

And yet, my A7R is consistent.  It doesn't matter which way it's pointed.  If shot for more than 60 seconds at ISO50 in daylight, I will always get a leak showing up as the crescent shaped ghost in the upper right of the frame.

I find this very unusual.

And yet, partially because for some reason, this "ISO25,600, 30sec in broad daylight" meme sprung up, and these websites are treating that as some golden standard to mock and laugh at.  Rather disheartening.

Here's what I would consider a VERY modest daytime long exposure shot. 76 seconds at ISO50.  The second one I agree is a bit more extreme, it's 8.5 minutes at ISO100.  I'd still not consider that excessive, though.  With my Canon 5Dmk2 I could easily shoot 5-10 minutes without any leakage as long as I remember to cap the OVF.  But considering there's all sorts of things happening in that image, I'm OK with accepting that one to be extreme.  But 76s @ ISO50??  I just can't accept that.

I would however love to hear your experiences; if you've ever had to deal with light leaks in your long exposure photographs (landscape, seascape, etc.) NOT from the lens or the OVF, but specifically from the flange area, please share your experience.  Thanks!

Thanks for the thoughtful post.

It's interesting to see your experiences as you're doing quite a bit of long exposure stuff, which is where I live. 6-stop is the lowest I use, 16 is the high end of the range. I'm seeing light leak as you are at these ranges. It's amazing not having to worry about the OVF, but now knowing that there's potential light leak through the body...

In any case, I tested today the Lee Big Stopper (10-stop) vs. B+W (9-stop) and here's the results:



Warmer toned images are the B+W. Notice how cold the Big Stopper is? : (

Another thing to notice is the top right of the horizontal images, and the top left of the vertical ones. Light leak : (

Graham
Logged
Graham Clark  |  grahamclarkphoto.com

Graham Clark

  • Full Member
  • ***
  • Offline Offline
  • Posts: 179
    • grahamclarkphoto.com
Re: Sony A7R Potential Light Leak (pictures)
« Reply #9 on: March 12, 2014, 02:24:50 am »

But there is help...

http://www.sonyalpharumors.com/sony-knows-how-to-fix-the-leak-and-fuji-x-t1-has-similar-issue-too/#disqus_thread

well I guess my Sony A7R going in for repair for sensor discoloration and defective bulb feature tomorrow is convenient... add light leak to the list.

Graham
Logged
Graham Clark  |  grahamclarkphoto.com

CptZar

  • Full Member
  • ***
  • Offline Offline
  • Posts: 157
Re: Sony A7R Potential Light Leak (pictures)
« Reply #10 on: March 12, 2014, 03:41:31 am »

After looking at your pictures again, I am not convinced, that this is a light leak problem. As in the image of Graham, in both of your images there is lens flare in the upper left corners. The "light leak" you are claiming is in both images, as well as in Grahams Images symmetrical in the right upper corner.

In my humble opinion this is caused by internal reflections in Metabones adapter. It has nothing to do with a light leak.

There is a thread concerning the matter.

http://www.luminous-landscape.com/forum/index.php?topic=85249.0

« Last Edit: March 12, 2014, 03:46:56 am by CptZar »
Logged

RawheaD

  • Full Member
  • ***
  • Offline Offline
  • Posts: 145
Re: Sony A7R Potential Light Leak (pictures)
« Reply #11 on: March 12, 2014, 05:26:07 am »

After looking at your pictures again, I am not convinced, that this is a light leak problem. As in the image of Graham, in both of your images there is lens flare in the upper left corners. The "light leak" you are claiming is in both images, as well as in Grahams Images symmetrical in the right upper corner.

In my humble opinion this is caused by internal reflections in Metabones adapter. It has nothing to do with a light leak.

There is a thread concerning the matter.

http://www.luminous-landscape.com/forum/index.php?topic=85249.0




I'm sorry if I didn't provide enough information to convince you, but I've tracked it down 100%. What is seen on the upper right is NOT lens flare or reflection in Metabones adapter.

When I was out one snowy morning to take that shot, I tried everything in the book to get rid of the ghosts, originally thinking it was flare. The stuff on the upper left went away after changing angle of composition, etc., but the upper right arc would not go away.  Then I finally realized it was light leak, and was able to control it in the field by putting my gloves against the left side of the lens.

Because of that, my next guess was that the Metabones was leaking light. I went home and conducted intensive tests.  But no matter how much I covered up and taped the Metabones adapter, the leak would not go away.

Here's one of those test shots.  ISO6400, 20s, *lens cap on*.

Then, after reading about a squishy mount and a potential internal gap in the mount structure itself, I tried taping the mount area and then and only then i was able to get rid of the leak.  See second pic (sorry it's small, can't find original).  There is a very thin strip of gaffer's tape just enough to cover the gap between the silver flange and the orange outer ring.  It doesn't even reach the flange surface, meaning the tape does nothing to seal between the lens and the flange.  And yet this thin piece of tape gets rid of the leak.

RawheaD

  • Full Member
  • ***
  • Offline Offline
  • Posts: 145
Re: Sony A7R Potential Light Leak (pictures)
« Reply #12 on: March 12, 2014, 05:30:21 am »

Graham, thanks for more photos. Gad, I wish I had my A7R now so that I can do these tests myself :LOL:

Anyway, what were the exposures? 30s as your previous shot?

Like I said, the crescent regardless of composition (vertical vs. landscape) angle to sun, etc.  It's always the same shape in the same place.

CptZar

  • Full Member
  • ***
  • Offline Offline
  • Posts: 157
Re: Sony A7R Potential Light Leak (pictures)
« Reply #13 on: March 12, 2014, 05:43:17 am »

Very interesting. I looked through my files, but couldn't find one with that issue. But then again there are no long exposures, done with the Lee Big stopper. Thanks for sharing. There is a new thread on Sony Alpha Rumors confirming your statement.

Manoli

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2299
Re: Sony A7R Potential Light Leak (pictures)
« Reply #14 on: March 12, 2014, 05:23:26 pm »

In any case, I tested today the Lee Big Stopper (10-stop) vs. B+W (9-stop) and here's the results [...]
Warmer toned images are the B+W. Notice how cold the Big Stopper is? : (

What WB did you set in camera - Lee recommend 8000-9000 Kelvin with the BigStopper, is that the range you used ?
Logged

Graham Clark

  • Full Member
  • ***
  • Offline Offline
  • Posts: 179
    • grahamclarkphoto.com
Re: Sony A7R Potential Light Leak (pictures)
« Reply #15 on: March 16, 2014, 05:22:42 pm »

Graham, thanks for more photos. Gad, I wish I had my A7R now so that I can do these tests myself :LOL:

Anyway, what were the exposures? 30s as your previous shot?

Like I said, the crescent regardless of composition (vertical vs. landscape) angle to sun, etc.  It's always the same shape in the same place.

Yes, 30 seconds! : )
Logged
Graham Clark  |  grahamclarkphoto.com

Graham Clark

  • Full Member
  • ***
  • Offline Offline
  • Posts: 179
    • grahamclarkphoto.com
Re: Sony A7R Potential Light Leak (pictures)
« Reply #16 on: March 16, 2014, 05:23:24 pm »

What WB did you set in camera - Lee recommend 8000-9000 Kelvin with the BigStopper, is that the range you used ?


I didn't actually set the white balance to 8000-9000, didn't realize I had to. It's unfortunate that the lens is colder than the other high quality glass out there... : (

Graham
Logged
Graham Clark  |  grahamclarkphoto.com

Graham Clark

  • Full Member
  • ***
  • Offline Offline
  • Posts: 179
    • grahamclarkphoto.com
Re: Sony A7R Potential Light Leak (pictures)
« Reply #17 on: March 16, 2014, 05:24:54 pm »

I often shoot at maximum ISO to determine composition when using dark ND filters on both my Canon 6D and Sony A7R, and when doing so this is what I captured yesterday:



Can you determine which was shot with the Canon and which with the Sony A7R?
Logged
Graham Clark  |  grahamclarkphoto.com

Manoli

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2299
Re: Sony A7R Potential Light Leak (pictures)
« Reply #18 on: March 16, 2014, 05:56:00 pm »

I didn't actually set the white balance to 8000-9000, didn't realize I had to. It's unfortunate that the lens is colder than the other high quality glass out there... : (

Don't hold me to it - but I'm 95% certain that Lee recommend that setting to reduce the excessively cold tint that the BigStopper casts.


I often shoot at maximum ISO to determine composition when using dark ND filters on both my Canon 6D and Sony A7R, [...] Can you determine which was shot with the Canon and which with the Sony A7R?

Were you using the BigStopper again on the A7r ? Ironically, in this shot I much prefer the 'light-leak' colour. You said you sent the camera back to Sony - what did they say?  RawheaD above commented on issues with the adapter* (I assume you were using the metabones adapter on the canon L again ) - have you tried using a native FE lens to see if you still get a light leak under the same conditions & scene settings ?

Manoli

Edit:
*And possibly the mount, if you switch between the FE and the Metabones mount you'll have confirmation of whether it's the mount (still occurs, identical or part pattern) and what's not repeated will be the adapter. I re-read your posts above and you mention wear on the sony mount - what did sony say when you sent the cam in for repair ?
« Last Edit: March 16, 2014, 06:07:20 pm by Manoli »
Logged

mcbroomf

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1538
    • Mike Broomfield
Re: Sony A7R Potential Light Leak (pictures)
« Reply #19 on: March 16, 2014, 07:17:31 pm »

Did you have the eyepiece covered on the 6D?
Logged
Pages: [1]   Go Up