7
\$\begingroup\$

My brother took a number of pictures in the alps recently, and a few of these showed defects similar but not identical to the one below. The temperature was -14 C which doesn't seem unreasonably cold. Any ideas what this might be? Full res here.

Edit: Here are the color versions, and it's still somewhat unclear if it's Newton's rings or some other effect.

Low res version

\$\endgroup\$
3
  • 2
    \$\begingroup\$ Can you explain the equipment used to capture this photo? Canon T3i at f/14, 1/200th, ISO 100 it looks like from the EXIF. What lens was used? \$\endgroup\$
    – dpollitt
    Commented Jan 17, 2015 at 20:35
  • \$\begingroup\$ I believe it was the 18-55 mm kit lens, but am checking now. \$\endgroup\$
    – ARM
    Commented Jan 17, 2015 at 22:00
  • 3
    \$\begingroup\$ While -14 °C may not be "unreasonably" cold for you, it's well outside the official operating range for the 600D / T3i (0 to 40 °C). As far as I can tell Canon don't give temperature ranges for their lenses, but it's reasonable to assume they'd be similar to the cameras; even the 1D X has a lower limit of 0 °C. \$\endgroup\$
    – Philip Kendall
    Commented Jan 17, 2015 at 22:41

2 Answers 2

3
\$\begingroup\$

I could be very well wrong, but those look a lot like Newton's rings, which can be caused by two layers of dissimilar materials that are normally in close contact expanding or contracting differently, making one of them buckle slightly. And by slightly, I mean that the spacing between them varies by fractions of a wavelength of light. (Another, similar possibility is differential stress polarization effects, again with a thermal basis. It's really hard to tell in B&W, though.) The "origin" of the pattern probably indicates a relative "hot spot". If the camera is shot cold, or allowed a little while to warm up completely, you probably wouldn't see a problem, but if it's been on just long enough to partially warm up, there might be enough difference in temperature across the sandwich that is the sensor, its lenses and filters, to cause a visible pattern.

\$\endgroup\$
3
  • \$\begingroup\$ We did consider interference effects, but it seems to me that the boundaries between the regions are too sharp, although I guess this could possibly result from the image being taken in B&W. I didn't see a diagram for this particular lens online, but if it's similar to the EF-M one (cdn.audiencemedia.com/var/photoreview/storage/images/media/…) the front lens pair is a likely candidate for something like this. \$\endgroup\$
    – ARM
    Commented Jan 17, 2015 at 22:29
  • 3
    \$\begingroup\$ I'd suspect the sensor sandwich rather than the lens; Bayer demosaicing can find "detail" where there is none. \$\endgroup\$
    – user35658
    Commented Jan 17, 2015 at 22:44
  • 2
    \$\begingroup\$ Also apparently the original was taken in color, I'll post a copy as soon as I have one. If is an interference effect it should hopefully be much more obvious in the original. \$\endgroup\$
    – ARM
    Commented Jan 17, 2015 at 23:47
-1
\$\begingroup\$

This looks like the sun was just outside the picture. While it is therefore not imaged directly, the secondary reflections from the various lens elements do end up in the picture.

This is exactly why you see people shading their lens when they point it near the sun. Try to arrange the picture so that the sun is not just outside the image. That makes it easier to use your hand or something to shade the lens but without it ending up in the picture.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.