HELP - Reflections (maybe?) appearing in pics during night shots (examples within)

[Morbo]

bad news guys - i was out tonight - and i took off the UV filter - so I was shooting with just the lens and nothing else - and the reflections were still there.

I will post pics soon.

[Morbo]

they are basically the same spots/photos as before - seeing as how it's a good way to try it out...but, even without the UV filter, it still happened.

these were with a Tamron 28-75 on a K100Dsuper

[Wheatfield]

If it makes you feel any better, I've managed to get that effect with my 77mm lens. It is more likely to happen with more complex lens designs that have more elements (zooms). I did a shoot some years ago with my 77 on the LX. The customer wanted a romantic, candle lit set up, and so, rather than do the smart thing, I set up a bunch of candles and started shooting. I believe I had one of the Pentax Ghostless filters on the lens. The results were pretty enough, but there were probably a dozen reflections of the candles in odd places around the images.
I suppose it is possible you have managed to invoke some sensor ghosting, but I've never seen it to know what it is, and don't really know what it looks like.

[Venturi]

Originally Posted by Morbo

they are basically the same spots/photos as before - seeing as how it's a good way to try it out...but, even without the UV filter, it still happened.

these were with a Tamron 28-75 on a K100Dsuper

Trying to learn something here. Are you saying you are able to repeat the results with different lenses and/or different focal lengths of the same zoom lens on the same camera body?

Looking closely at the shots you posted it seems the anomaly is symmetrically opposed from the subject light source in all the photos.

Not sure if that's the right terminology but if you were to imagine the full circular image being projected back and then draw a line bisecting the center perpendicular to a line connecting the center of the real light and the flare you would see that the lengths of the lines from the bisecting point to each "light" is equal in length.

What this tells my addled brain, and I could be way off here, is that the reflection is happening at the camera/sensor end of things. If it were internal lens elements, or a filter, causing the issue you would not necessarily see such a symmetrical reflection. And this would seem to bear out the "sensor ghosting" theory - the subject light is bouncing off the sensor back at the rear lens element and back to the sensor again.

(I haven't talked like that in years - geometry is hard, m'kay)

If it is sensor ghosting and happens often is there anything correctable by the "end user"? Micro dust on the sensor or residue perhaps?

[Canada_Rockies]

Originally Posted by Venturi

Trying to learn something here. Are you saying you are able to repeat the results with different lenses and/or different focal lengths of the same zoom lens on the same camera body?

Looking closely at the shots you posted it seems the anomaly is symmetrically opposed from the subject light source in all the photos.

Not sure if that's the right terminology but if you were to imagine the full circular image being projected back and then draw a line bisecting the center perpendicular to a line connecting the center of the real light and the flare you would see that the lengths of the lines from the bisecting point to each "light" is equal in length.

What this tells my addled brain, and I could be way off here, is that the reflection is happening at the camera/sensor end of things. If it were internal lens elements, or a filter, causing the issue you would not necessarily see such a symmetrical reflection. And this would seem to bear out the "sensor ghosting" theory - the subject light is bouncing off the sensor back at the rear lens element and back to the sensor again.

(I haven't talked like that in years - geometry is hard, m'kay)

If it is sensor ghosting and happens often is there anything correctable by the "end user"? Micro dust on the sensor or residue perhaps?

I agree that it appears to be sensor ghosting - the clue is the green colour of the ghosts. I think (correct me if wrong) that the IR filter causes the colour as the light passes through a "no red shall pass" filter three times before it is recorded as a ghost - the initial pass to the sensor, the bounce back, and another pass through the sensor. I find that ghosts from the lens elements are much closer to the actual colour of the light that is reflected. An example is one image I took with my DA 12-24 here, where the ghost in the trees is purple, the colour of the SMC coatings. Also, the ghost is not relected at the exact angle of the sun reversed against the center line of the lens. Elk River, BC-5 on Flickr - Photo Sharing!

[Morbo]

ok, so, how do I fix it? :\