.jpg)
Placing an ND filter on the lens and shooting a long exposure can give a nice moody feel. The technique is to shoot in normal daylight with ND filters and long shutter speeds.
The Sony A7R has gotten a lot of attention for light leaks that take place around the bayonet – where the chrome ring contacts the anodized orange ring.
http://community.sony.com/t5/Alpha-NEX-Cameras/A7-A7R-Light-leaks/td-p/254835
http://www.dpreview.com/forums/post/53072154
I started with black electrical tape and recently discovered a hair tie (elastics) to be an easy, non-sticky alternative. It’s easy to carry in any camera bag too. Get the Goody variety because it does not have the metal connector and “it will give you the hold you need and the comfort you crave.” They are available at drug stores or online. They are the perfect diameter to stretch around the lens and fit snugly at the joint between the lens and body. Even if you catch them with your fingers while handling the camera they snap back into the groove. Also, it’s easy enough to leave on to cut down on moisture and dust and the possibility that “normal” exposures might be effected by a light leak. The biggest bonus is that it’s NOT unattractive!
http://www.walmart.com/ip/Goody-Ouchless-No-Metal-Hair-Elastics-Little-Black-Dress-27255-36-count
Goody hair tie around lens bayonet.
Light leak present in upper right and left side. 30 seconds, f/2.8, ISO 25600. Exposure made within 3 inches of 60W incandescent bulb.
Performing the same test with the Goody hair tie in place. No light leaks.
Light Leaks with Studio Strobe (added 5/5/14)
I set up a situation that could be a real shooting scenario with the A7r and a studio strobe. The strobe was on full power 800w and the camera was held near the strobe in a manner that a photographer would stand while shooting and lighting a subject in the distance. The camera was set up as if it was a conceivable shot, not a torture test or ridiculous settings – a subject in the distance and photographer near the strobe. If my position with the camera were in such a position that when the strobe hit the bottom area of the lens on the left side there was evidence of a light leak. The exif data is 1/60second, ISO800, f2.8 (the f-stop doesn’t matter because the light leaks takes place at the mount) 35mm f/2.8 FE lens. There is an opinion that testing for light leaks on the A7(R) is not a realistic test of the camera, after all “whoever shoots at ISO25600 for 30 seconds?” Well, this test demonstrates it can happen at 1/60sec, ISO800 too. I’m not in a panic about it, after all there is the hair tie.
VIDEO of the Light Leak in the studio: https://www.youtube.com/watch?v=cbq_zXK3o0c
Light leak with camera near strobe. Strobe was placed lower left of camera. Settings: 1/60sec. ISO800, f/2.8.
Wow, that’s a torture test for any camera, my Canon 5D3 would fail that miserably with its leaky optical viewfinder! Next time you buy some Broccoli see if there’s a purple rubber band around the stems – I save those for no reason other than they’re good stout bands that are about 1/2″ wide. One could live on your lens hood until needed, it’d make for a perfect light seal.
[…] können all diejenigen, die mit den Problemen zu kämpfen haben, immernoch den simplen Trick von Ferrell McCollough verwenden, und ein Haargummi um die entsprechende Stelle am Objektiv spannen. Ja, das scheint […]
That’s why Canon provides a cap for the optical viewfinder, to prevent such leaks.
I was going to try tape on my A7r, but maybe I’ll try a wide black rubber band. 30 secs isn’t all that long an exposure either, I often shoot exposures of a few minutes during daytime with my Big Stopper. Can’t wait to see what Sony comes up with, though I won’t hold my breath.
The problem I have with the torture test is that the light gathering, barring any form of reciprocity (which sensors have their own versions of) at 30sec at 25600 is equivalent to ONE FULL HOUR at ISO 200. In other words the vast majority of photographers will never see this, unless they are deliberately duplicating the test. 3 inches away from an incandescent bulb? Who does that if they’re not looking for hidden lemon juice letters?
I agree that Sony should seal the mount, perhaps more for environmental purposes but if there aren’t enough photons to engage the sensor and register the flare until you hit 16 minutes at ISO 800, there’s not a huge issue to be made here. For most.
I completely disagree that the high-ISO test (which is something that I came up with, and thenceforth has slowly spread around the Internet) is some, highly unrealistic torture test.
First of all, the exposure time is not really an issue. Take a light tight camera & lens, put it in 12,800 or 25,600ISO and shoot it for 1s, 5s, 10s, or 30s. The longer the exposure, the more noise you’ll see across the frame coming from the heat-buildup of the sensor. However, you will NOT see the kind of crescent shape light leak you see with the A7/R. That’s because it’s light tight. The “stress test” could be ISO25,600 @ 1s, and you’ll still see the light leak as long as you have the light source up against the leaky spot on the flange. The only reason why the exposure would be long with the test is to give the tester enough time to take the light source around the flange and check the mount from all angles.
Thus, comparing the stress test to a “ONE FULL HOUR” exposure @ ISO200 is pointless.
https://dl.dropboxusercontent.com/u/131577/forums/FB%20Samples/DSC02723.jpg
Here’s the photo I’ve been pasting all over the place in response to naysayers. This is an 8 minute exposure @ ISO100, f8.0, with a total of 19 stops of ND. Before anyone slams me that that’s also too extreme,
https://dl.dropboxusercontent.com/u/131577/forums/FB%20Samples/DSC02718-Edit.jpg
Here’s a more modest 2.5 minute exposure at ISO50, f22, and just 12 stops of ND. Anybody who does daytime LE work knows that 12+ ND is not “extreme” by any means.
And you know what? The updated “Light Leaks with Studio Strobe” sample pretty much proves that I’ve been right all along, that the degree of light leaking that I saw in my images the very first time I did LE work with this camera, I might add, is nowhere near acceptable, or what should be considered “within spec”
RE: Chris Gibbs
Mmm, yeah, that’s why when you do LE with an SLR, you always put a cap on the OVF. You know, the cap that Canon (and other companies) *always* includes in their camera boxes? The OVF leaks light, there’s no way around it, because the mirror box is designed *specifically* to transmit light back and forth between the mount, sensor plane, and the OVF. In contrast a mount of a camera should always be designed to *specifically* be light tight. A very big difference.
Sony certainly did not ship a hariband with my A7R :LOL:
@ Richard Saunders
Google 5D3 light leak – its actually two-fold, the “cap” is only half the issue, the other light leak is internal.
Thank you so much for this. Talk about synchronicity! Just yesterday (March 6) I took a real world shot for my client. A shot of the famous Bund in Shanghai with the A7r. It was a long exposure in the daytime with a 3ND filter to smooth out the water and I had exactly the same problem with identical marks to the ones you’re showing. On some images just one side and on other both sides. I was very confused as it couldn’t have been flare but not it makes sense and I will follow your fix for further shots. Thanks again.
[…] cosa viene de Sony Alpha Rumors y viene de Before the coffee.com donde veréis ejemplos del […]
[…] Image courtesy of Ferrell McCollough […]
[…] Ferrell McCollough discovered that the light leak can be stifled by simply putting an elastic hairband over the area […]
[…] obvious fix is to put gaffer’s tape around the mount, but photographer Ferrell McCollough had something easier in mind. He used hair ties instead. Just pick yourself up some black hair ties at WalMart […]
[…] obvious fix is to put gaffer’s tape around the mount, but photographer Ferrell McCollough had something easier in mind. He used hair ties instead. Just pick yourself up some black hair ties at WalMart […]
[…] obvious fix is to put gaffer’s tape around the mount, but photographer Ferrell McCollough had something easier in mind. He used hair ties instead. Just pick yourself up some black hair ties at WalMart […]
[…] Sony ne s’étend pas sur le sujet, mais indique travailler à une correction (je ne voudrais pas être à la place des ingénieurs qui s’en occupent). En attendant, il vaut mieux protéger cette partie de l’appareil avec du gaffer (scotch décollable), un gros élastique plat ou un chouchou (normalement pour tenir les cheveux comme indiqué sur BTC). […]
[…] Sony does not go into great details, but says that they are working on a correction for the issue (I would not like to be one of the engineers tasked with this). In the mean time, the owners would better be using soemthing to mask this part of the camera: gaffer tape, a large elastic rubber band, or elastic hair ties (as shown on BTC). […]
I put some tape over it…. easy solved. Novoflex also told me that you could adjust the screws to tighten the adapter to the lens used. I don’t bother, ’cause the black tape strip does take care of the leak…
…of course, IF you use adapter. If not, use tape between lens and camera. The latter isn’t a good solution if you change lenses alot….
🙁
[…] if you’re an A7 or A7r user experiencing these light leaks, a photographer by the name of Ferrell McCollough has a simple but apparently effective solution for you: an elastic hair tie which you can then put […]
[…] Aceasta este solutia propusa chiar de unul dintre utilizatorii care au observat bizara problema, Ferrell McCollough (foto […]
[…] Καταρχήν να αναφέρουμε ότι οι αναφορές λένε ότι οι Sony A7 και Sony A7r εμφανίζουν το πρόβλημα "light leak" σε συγκεκριμένες συνθήκες. Για να είμαστε ακριβείς όταν στην μηχανή ρυθμιστεί το ISO στα 25.600 και η ταχύτητα του κλείστρου στα 30 δευτερόλεπτα και υπάρχει στο κάδρο μας κάποιο δυνατό φως, τότε στην τελική φωτογραφία εμφανίζονται λωρίδες φωτός. Από ότι φαίνεται η είσοδος του φωτός γίνεται από την μοντούρα της μηχανής, δηλαδή το σημείο σύνδεσης του φακού με την μηχανή. Μάλιστα η imaging-resource αναφέρει ότι το ίδιο αποτέλεσμα μπορούμε να έχουμε και σε μικρότερες τιμές του ISO και για χαμηλότερες ταχύτητες κλείστρου αν το φως που υπάρχει στο κάδρο μας είναι πολύ δυνατό. Η λύση μέχρι τώρα στο πρόβλημα, η οποία αναφέρεται από χρήστες, είναι η τοποθέτηση μαύρης ταινίας γύρω από το σημείο της σύνδεσης του φακού με το σώμα της μηχανής, ενώ την ίδια δουλειά μπορεί να κάνει και ένα λαστιχάκι για τα μαλλιά (δείτε εδώ). […]
[…] March 19, 2014. In case if you want to fix the issue right now, then a photographer by the name of Ferrell McCollough has a simple but apparently effective solution for you: an elastic hair tie which you can then put […]
[…] Goody hair tie around lens bayonet (Courtesy: Ferrell McCollough) […]
More tidy temp solution is an orange colored (that’s the color it comes in) O ring available in Amazon (comes in 25 or 100/bag). After testing various sizes, I found 2″ inside diameter x 3/16″ thick to work the best. This size allows you to stretch over the lens so that it holds tight at the gap between the lens and the mount. Just roll it up the lens when changing and leave it there. It’s cheap enough to leave one on each lens.
Sorry I meant 3/32″ thick O’ring. 3/16″ is a bit too thick.
Here’s the Amazon link:
http://www.amazon.com/gp/product/B000FMWPRE/ref=oh_details_o08_s02_i00?ie=UTF8&psc=1
[…] light leak has already been documented quite well by Imaging Resource, Ferrell McCollough, and others. We tried a couple of simple methods to determine where the light leak occurred and […]
[…] no encaixe da objetiva, vedando a área por completo. Uma outra solução, menos invasiva, é o uso de elástico para cabelo na cor preta, conforme apontado pelo fotógrafo Ferrell McCollough em seu […]
Dorem: I don’t see the o-ring as a “more tidy” solution. I’ve worked with many o-rings and thought about it for this solution but discarded the idea because their elasticity is too limited to go over a lens and hood easily.
O-rings are not designed to be “stretchy” they are designed to seal. In the cold they are even harder to work with especially 70A.
Orange is ugly, they come in black you know.
[…] light leak has already been documented quite well by Imaging Resource, Ferrell McCollough, and others. We tried a couple of simple methods to determine where the light leak occurred and […]
[…] light leak has already been documented quite well by Imaging Resource, Ferrell McCollough, and others. We tried a couple of simple methods to determine where the light leak occurred and […]
Consider that many tests show the light leak to be very overexposed at ISO25600 30sec. If you have an obvious light leak and continue to lower the ISO, just before it’s imperceptible then there is a loss of contrast. It would be subtle, maybe not even noticeable but you’d certainly give up some quality. You are probably getting some loss of contrast and you don’t even know it. A light leak is not like a light switch that gets turned on and off when you dial in ISO 25600, it’s a subtle gradation. You might be shooting in that subtle range everyday. Your images could be better and you don’t even know it.