Why does it look like my color temperature is changing?

The exact same thing happened to me, I was shooting a hallway in a medical clinic. I freaked out at first, thinking that the white balance in my camera had gone funky, but then I realized that the color shift was not uniform; the flourescents at the far end of the hallway were shifted orange while the flourescents close to me were shifted blue and vice versa.

This only occured at certain locations at certain times; not sure why as my shutter was 1/250 the entire evening and the building has 100% flourescent lighting, so I would have expected it to be a constant problem. It wasn't.

I haven't had this problem in any other location (even under flourescents) so I figured it had something to do with the age of the building, poor electricity, faulty ballasts, etc.
 
Last edited:
The way I figure it works is this:

Fluerescent tubes run at 60Hz (the AC frequency in N.A.). I am not EXACTLY sure what happens in the balast, but the current in the tube (which is what causes the vapor inside the tube to glow) will be either on/off, or +/- sixty times a second. If you can picture 1/250 of a second is actually just under 1/4 of that 1/60th of a second. This ~ 1/4 of the AC cycle is all the camera 'sees', but it does so at 1/24 sec intervals (assuming you are shooting 24P).

So, every time the shutter 'opens', it will see a different 1/4 of the 1/60. Not only will there be varying amount of light the camera will see in that slice of time, inevitably that 1/4 will fall in the times when the tube is switching on/off or reversing polarity. Add to it the fact that 60Hz is only approximate frequency of AC, and in most buildings (and grids) it fluctuates. You just cannot shoot under 60 Hz fluoerescent tubes with any degree of predictability, unless the shutter is open long enough to read the entire AC cycle. That's why you should use 1/60 or 1/30 under fluorescent light. If you use kinoflo or other fluorescent tubes that are specifically made for film making, you are OK - those tubes run at much higher frequencies (I think it's 2kHz, but I am not sure).
 
And just this: the variable slice of the glow pulses, not only results in a variable amount of light with high shutter speeds, but also a variable color spectrum because each glow pulse has a changing spectrum due to different afterglow (persistence) of the different fosfor powders. Simular effects are seen if incandescant light is present: during the "low" period of the glow, the incandescent light (low color temp) comes into play.
XCheck, FYI, the puls frequency is 120 on a 60 Hz powerline.
 
I had a similar problem and did a search. The only difference is that everything was on auto, and I was shooting outdoors. Don't know why. Will keep looking.
 
You must log in or register to reply here.
Back
Top