A Light Leak in my D300?

In summer 2009,  some 18 months after switching to the Nikon D300 body, I noted a strange artifact on some of my photos. Here's a sample shot that highlights the problem:

18 - 200 mm super zoom @ ISO 200, 27 mm, f/11, 1/320"

Initially, I didn't even notice the defect which was most conspicuous when it fell onto a clear sky or other monotonous area. I just thought "strange!" — and patched it away in Aperture. My primary suspicion was that this may be a reflection caused by the skylight filter that I used to protect the front lens of my 18 - 200 mm VR super zoom lens: sun light (the artifact seemed to be linked to pictures taken in bright sunshine) can be reflected from the filter in front of the sensor, then reflected back on the backside of the UV or Skylight filter, producing an extra reflex / artifact — this is a known phenomenon. However, if this was the problem, why then would the artifact always appear in the same location??

My next thought was that it might be some object / dust particle towards the back end of the zoom lens, which might explain why the artifact looked like a diffraction pattern (with dust on the sensor I would expect well-defined, dark spots). So, I was hoping that better lenses would cure the problem. Wrong.

In spring 2010 I switched from the super zoom to professional (full frame) quality 70 - 200 mm f/2.8 and 16 - 35 mm f/4 lenses; I did not use an extra UV filter, just to make sure this does not cause extra headaches. The lenses produced very good, high quality, clear images — but sure enough, the problem was still there! Sometimes it took while to realize the issue was present, like here:

70 - 200 mm f/2.8 @ ISO 200, 105 mm, f/10, 1/400

But once I realized that it was at a given location in the raw image, it became hard to ignore it:

One some background it wasn't easy to make it disappear — sometimes I needed to use Photoshop, or at least very careful patching in Aperture. Here's a close-up look at the artifact in the above shot:

Now I was more worried about my camera body. But — if it wasn't the lens(es), could it still be some strange kind of dust particle on the sensor filter? I took a series of blank, defocused shots of a uniform surface, at various lighting conditions — nothing at all, not the tiniest sign of sensor dust anywhere, let alone such (not so) funny circles!

So, I continued investigating and noted that the artifact only showed up when I had direct sunlight on the camera body — and particularly when the sun was almost or directly in the picture. My hypothesis then was that only a tiny light leak at some distance from the sensor plane would explain the circle-like diffraction pattern on photos, and why it would only appear when direct sunlight fell onto the front part of the camera body. In October 2010, I sent the body to Nikon for repair, along with pictures highlighting the issue. The warranty had expired, but the repair was done at no cost; the repair sheet spelled out some warning about not letting dust enter the body — but I'm sure this was not the issue! I took some shots with the sun in the picture — and I'm glad to report that the problem is gone and has since not reappeared:

24 - 120 mm f/4 @ ISO 200, 36 mm, f/10, 1/400"

24 - 120 mm f/4 @ ISO 200, 36 mm, f/14, 1/800"