Welcome to Photo.net: A Community of Photographers

Community > Forums > Nikon > Nikon SLRs > Strange problem with D300:...

Strange problem with D300: Fluorescent Lighting

Jerd Jetsrisuparb , Nov 05, 2008; 04:30 a.m.

I ve encounter a strange problem with my D300 after I had the firmware updated. I am guessing it has got something to do with the white balance, and higher ISO setting.

Tonight I was bored out of my mind and took a picture at my wall at high ISO, it looked as if I put bi color filter on my lens. Some part of the photo were really yellow. On ISO 200 the color was accurate. This happen whens I shoot under fluorescent light.

I have tried to do this with 2 lenses, without using any filter. Nikon 50 f/1.8, and Nikon 70-300 VR, they produce same kind of color so I am pretty sure it is not from the lens. I attached the pictures I took. D300 users with updated firmware, please try to take picture under fluorescent with high ISO. I hope it's not my D300.


Attachment: fileA9uPdV.jpg

Responses

Jerd Jetsrisuparb , Nov 05, 2008; 04:34 a.m.

Response to Strange problem with D300

Sorry I have to attached the other picture seperately

Large photo attachment:
(ISO 5000 -- 1424 x 2144 photo)

David Haas , Nov 05, 2008; 06:23 a.m.

Response to Strange problem with D300

One of the "fixes" in the new firmware was the auto-white balance. You don't mention if you used auto or self set.

Try one on self set wb - set for fluorescent - yellow.

Also - I'd send this on to Nikon Tech Support.

Dave

Jerd Jetsrisuparb , Nov 05, 2008; 07:04 a.m.

Response to Strange problem with D300

Sorry, I didn't provide all the information. The pictures above were taken in Auto WB. Also, I have shot in fluorescent WB and PRE the result were similar, the color was off in high ISO, and accurate in low ISO. I am also sending this to Nikon tech support.

Rainer T , Nov 05, 2008; 07:07 a.m.

Response to Strange problem with D300

This is a typical and well known problem when shooting images with the main light source being fluorescent tubes.

The "problem" with flourescent light is, it hasn't got a fixed color temperatur. Its color temperatur oscillates with the double frequency of the mains power. Therefore, even if you set your cameras WB to fluorescent light, you set it to an averaged value. Such pictures will appeanice (like your first image), when the exposure time is long enough to cover at least one full cycle of the oscillation. (Your first image was taken with 1/30 sec). Now, your second image was taken with 1/1250sec. This is only a smaller part of one cycle of the oscillation. You might be lucky to take the capture while the colortemperature was near the average value the camera uses, but you might also take the image at a time where the actual color temperature is quite different from that. This leads to an image like your second one. The last effect to be observed is an image where the color varies within the image. It depends on the X-sync speed of the camera if you can produce this effect or not. Typically cameras with X-sync speeds at 1/125 to 1/250 can produce those images with exposure times at 1/250 to 1/500.

Jay Poel , Nov 05, 2008; 10:32 a.m.

Response to Strange problem with D300

Rainer explained it better than I would have. I had similar problems shooting in a school gymnasium - keep the shutter speed up and not some uch of a problem. But if I took a burst of shots 4 or 5, you could actually see the colour temperature changing in the pictures.

John Schroeder , Nov 05, 2008; 11:47 a.m.

Fluorescent lights flicker at 60 cycles a second. You captured the flicker.

Rainer T , Nov 05, 2008; 12:37 p.m.

-- "Fluorescent lights flicker at 60 cycles a second."

John, the light emited is the product of two sine functions (voltage and current) ... both functions have a frequency of 50htz in some countries, and 60htz in the US. The product (and that the light emited) of these two functions is a function with a frequency of 100htz or 120htz (the double of the root frequency). By the way ... standard bulbs do also flicker at this frequency (but the amplitude isn't that big as it is with tubes).

But the main point is, fluorescent light varies the color temperature. The original poster has NOT captured a flicker.

Joseph Wisniewski , Nov 05, 2008; 12:59 p.m.

120 times/sec in the US.

AC power oscillates 60 times per second, each oscillation is one cycle of a sine wave, with both a positive peak and a negative peak. So two peaks, 60 cycles/sec = 120 peaks/sec.

Jerd Jetsrisuparb , Nov 05, 2008; 03:11 p.m.

Thanks for the inputs

Back to top

Notify me of Responses