Pages: [1]   Go Down

Author Topic: Industrial/Gym Lighting Color Temp Question  (Read 6286 times)

kielinski

  • Newbie
  • *
  • Offline Offline
  • Posts: 10
    • Kielinski Photographers
Industrial/Gym Lighting Color Temp Question
« on: February 04, 2016, 10:13:06 am »

Ok, so my wife's business niche is shooting athletic competitions that frequently take place in industrial space gyms.  From what I can tell, light sources are typically sodium or mercury vapor lamps.  She's photographed in a few gyms where the light source has behaved oddly, though.  Color temperature is inconsistent from frame to frame - sometimes even within the frame!  You can understand what a chore this would be to deal with when processing hundreds of images.

Anyone have experience with this phenomenon?  I really want to identify what kind of lamps these things are, if that is indeed the issue.
« Last Edit: February 04, 2016, 10:17:22 am by kielinski »
Logged

bjanes

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3354
Re: Industrial/Gym Lighting Color Temp Question
« Reply #1 on: February 04, 2016, 10:36:27 am »

Ok, so my wife's business niche is shooting athletic competitions that frequently take place in industrial space gyms.  From what I can tell, light sources are typically sodium or mercury vapor lamps.  She's photographed in a few gyms where the light source has behaved oddly, though.  Color temperature is inconsistent from frame to frame - sometimes even within the frame!  You can understand what a chore this would be to deal with when processing hundreds of images.

Anyone have experience with this phenomenon?  I really want to identify what kind of lamps these things are, if that is indeed the issue.

In my experience which is about 7 years out of that game, most of those lights are Metal Halide. These have a decent spectrum but their output and color can vary according to the ballast they use. At 60 Hz there can be interaction with the shutter speed. You can use a Whibal target and take a burst of test shots at a shutter speed of 1/120 sec and above and look for variations in exposure and white balance. Electronic ballasts cycle at a much higher frequency and should pose no problem. Some gyms use fluorescent bulbs which can use magnetic or electronic ballast with the same problems.

Mercury and sodium lamps are less common and have a strong blue and yellow cast respectively. Newer installations may use LED.

Regards,

Bill
Logged

Colorado David

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1178
Re: Industrial/Gym Lighting Color Temp Question
« Reply #2 on: February 04, 2016, 10:54:20 am »

I've shot several times in Cessna citation service centers. The lights in those facilities appear to vary through the color spectrum. It is as if they are continually changing color temperature. I've never asked about this, but assumed it might be intentional for the purpose of inspecting electrical or mechanical components under different color temperatures. I could be all wrong, but it was certainly a challenge to shoot under those conditions.

kielinski

  • Newbie
  • *
  • Offline Offline
  • Posts: 10
    • Kielinski Photographers
Re: Industrial/Gym Lighting Color Temp Question
« Reply #3 on: February 04, 2016, 11:05:17 am »

Thanks bjanes, very helpful!
Since it's something she just has to deal with on the rare occasion it's not a huge deal.  Maybe 1 in 20 locations have exhibited the phenomenon. I'm attempting to attach two images captured in succession for visual reference.

TomFrerichs

  • Full Member
  • ***
  • Offline Offline
  • Posts: 108
Re: Industrial/Gym Lighting Color Temp Question
« Reply #4 on: February 04, 2016, 02:50:44 pm »

I've photographed in a rodeo arena lit with sodium lights powered by portable generators, and have observed color temp changes between frames shot in a burst.

I use Nikon, but Canon introduced an "anti-flicker" feature in the 7D Mark II that was designed to combat the problem.

https://www.learn.usa.canon.com/resources/articles/2014/eos7dmkii_antiflicker.htmlp

Tom Frerichs
Logged

dwswager

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1375
Re: Industrial/Gym Lighting Color Temp Question
« Reply #5 on: February 04, 2016, 02:54:45 pm »

Ok, so my wife's business niche is shooting athletic competitions that frequently take place in industrial space gyms.  From what I can tell, light sources are typically sodium or mercury vapor lamps.  She's photographed in a few gyms where the light source has behaved oddly, though.  Color temperature is inconsistent from frame to frame - sometimes even within the frame!  You can understand what a chore this would be to deal with when processing hundreds of images.

Anyone have experience with this phenomenon?  I really want to identify what kind of lamps these things are, if that is indeed the issue.

I shoot both in Gyms and night sports outdoors under stadium lights.  I use a ColorChecker Passport now to not only get white balance, but can create a custom conversion profile.  I formerly used an Expodisc.  The passport is not a panacea, but it allows you to get much closer to neutral than otherwise possible.  That handle color temp.

Flicker is handled in some cameras.  I know Canon has it in some and the D5 and D500 specs say they do it now too.
Logged

Ken Bennett

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1756
    • http://www.kenbennettphoto.com
Re: Industrial/Gym Lighting Color Temp Question
« Reply #6 on: February 04, 2016, 03:10:08 pm »

Sports photographers have this problem all the time. The lights cycle, different banks at different times. You can solve it by shooting at 1/60 sec, but of course that doesn't work for basketball or pretty much any other sport. High end arenas will have better lighting, but the usual big arenas are mostly terrible.
Logged
Equipment: a camera and some lenses.

dwswager

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1375
Re: Industrial/Gym Lighting Color Temp Question
« Reply #7 on: February 22, 2016, 09:56:36 pm »

Ok, so my wife's business niche is shooting athletic competitions that frequently take place in industrial space gyms.  From what I can tell, light sources are typically sodium or mercury vapor lamps.  She's photographed in a few gyms where the light source has behaved oddly, though.  Color temperature is inconsistent from frame to frame - sometimes even within the frame!  You can understand what a chore this would be to deal with when processing hundreds of images.

Anyone have experience with this phenomenon?  I really want to identify what kind of lamps these things are, if that is indeed the issue.

To my previous comment, I will add that both the output and spectrum changes when the lights cycle.  And depending on how cheaply the lighting is wired, it is sometimes that the whole damn bank will cycle simultaneously.

Interestingly enough, Auto White Balance tends to give the most consistent results rather than a constant metered white balance.   However, be prepared to correct WB in post!!
Logged

bjanes

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3354
Re: Industrial/Gym Lighting Color Temp Question
« Reply #8 on: February 23, 2016, 07:01:07 am »

Sports photographers have this problem all the time. The lights cycle, different banks at different times. You can solve it by shooting at 1/60 sec, but of course that doesn't work for basketball or pretty much any other sport. High end arenas will have better lighting, but the usual big arenas are mostly terrible.

The professional arenas often have three phase wiring. See the normalized wave form in the Wikipedia article and refer to the Three-phase Loads section of the article for a discussion of how 3 phase is applied to lighting (second paragraph). This arrangement reduces flicker, but the 60 Hz or 50 Hz flicker reduction in your camera would no longer be in phase with the lighting. In looking at the menu of my D800, I see that it does have flicker reduction for 50 and 60 Hz.

Regards,

Bill

« Last Edit: February 23, 2016, 10:58:40 am by bjanes »
Logged

dwswager

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1375
Re: Industrial/Gym Lighting Color Temp Question
« Reply #9 on: February 23, 2016, 10:05:41 am »

In looking at the menu of my D800, I see that it does have flicker reduction for 50 and 60 Hz.

Unfortunately, like my D810, I think that setting only affects movie recording and Live View.  I have not actually tried it for stills in a burst so something to check out next night game I shoot.

Nikon D810 Manual (page 371)

Logged

BernardLanguillier

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 12908
    • http://www.flickr.com/photos/bernardlanguillier/sets/
Re: Industrial/Gym Lighting Color Temp Question
« Reply #10 on: February 24, 2016, 10:48:38 pm »

Flicker is handled in some cameras.  I know Canon has it in some and the D5 and D500 specs say they do it now too.

I think the D500 has it for sure, but I am unclear about the D5. I did read somewhere it didn't have it, but haven't double checked.

For those cameras that have it, it works by delaying exposure, so it may not be usable in all cases.

Cheers,
Bernard

dwswager

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1375
Re: Industrial/Gym Lighting Color Temp Question
« Reply #11 on: February 25, 2016, 07:51:21 am »

I think the D500 has it for sure, but I am unclear about the D5. I did read somewhere it didn't have it, but haven't double checked.

For those cameras that have it, it works by delaying exposure, so it may not be usable in all cases.

Cheers,
Bernard

I've read that the D5 lacks it for Stills, but the D500 has it.  I find that hard to fathom.  But... 

At the end of the day, no matter how you try to account white balance under these lights, because the cycling changes spectrum distribution as a function of time, at faster shutters speeds over 1/50-1/60 (the cycle frequency), one is going to need to adjust WB in post processing.
Logged

Justinr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1733
    • Ink+images
Re: Industrial/Gym Lighting Color Temp Question
« Reply #12 on: February 26, 2016, 11:02:48 am »

Thanks bjanes, very helpful!
Since it's something she just has to deal with on the rare occasion it's not a huge deal.  Maybe 1 in 20 locations have exhibited the phenomenon. I'm attempting to attach two images captured in succession for visual reference.

May one ask what the WB setting was?
Logged

Justinr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1733
    • Ink+images
Re: Industrial/Gym Lighting Color Temp Question
« Reply #13 on: February 26, 2016, 11:12:31 am »

In my experience which is about 7 years out of that game, most of those lights are Metal Halide. These have a decent spectrum but their output and color can vary according to the ballast they use. At 60 Hz there can be interaction with the shutter speed. You can use a Whibal target and take a burst of test shots at a shutter speed of 1/120 sec and above and look for variations in exposure and white balance. Electronic ballasts cycle at a much higher frequency and should pose no problem. Some gyms use fluorescent bulbs which can use magnetic or electronic ballast with the same problems.

Mercury and sodium lamps are less common and have a strong blue and yellow cast respectively. Newer installations may use LED.

Regards,

Bill

Could I seek some sort of explanation on this as I must admit it's a new one on me.

Are we saying that the spectrum of a lamps output will vary over a cycle? If so what dictates the speed of the cycle and is it the full visible spectrum or just part of it? I'm assuming that the slower the cycle the more likely we are to get a variance in colour temp if using faster shutter speeds, so does the same apply to luminosity?

Cheers.
Logged

TomFrerichs

  • Full Member
  • ***
  • Offline Offline
  • Posts: 108
Re: Industrial/Gym Lighting Color Temp Question
« Reply #14 on: February 26, 2016, 12:45:04 pm »

Could I seek some sort of explanation on this as I must admit it's a new one on me.

Are we saying that the spectrum of a lamps output will vary over a cycle? If so what dictates the speed of the cycle and is it the full visible spectrum or just part of it? I'm assuming that the slower the cycle the more likely we are to get a variance in colour temp if using faster shutter speeds, so does the same apply to luminosity?

Cheers.

The speed of the cycle is dependent upon the frequency of the AC current.  In the US that frequency is 60 Hz; I think in many European countries the frequency is 50 Hz.  This means that the instantaneous voltage will change from zero to a maximum, back to zero,  to the maximum again in the opposite direction, then back to zero sixty times a second.

Incandescent bulbs, by their nature, don't react to the voltage change that quickly. They sort of "average" out the changes and are relatively constant, both as far as luminosity and color temp.

Other kinds of lights, particularly sodium and mercury based lights, do not "smooth" things over. The color temperature and light output actually change with the input voltage fairly substantially.  Fluorescent lights also have the same problem, although not as strongly as those others.

If your shutter speed is slow enough, you end up "averaging" the light through at least one or more cycles.  If, on the other hand, your shutter speed is fast you can end up with only part of the cycle, and your color temp and measured light can vary from frame to frame, depending upon when you hit your shutter.

Finally, those lights themselves, used in the US for gym, industrial, and street lighting, don't have anywhere close to a continuous spectrum at any time.  Go grab a cheap diffraction grating or prism and you'll see really strange spikes in the spectrum. These lights are bright, fairly low in heat production, fairly efficient, long lived, and cheap for their light output compared to incandescent, which explains their use.

Tom Frerichs

« Last Edit: February 26, 2016, 12:48:45 pm by TomFrerichs »
Logged

Justinr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1733
    • Ink+images
Re: Industrial/Gym Lighting Color Temp Question
« Reply #15 on: February 27, 2016, 06:39:53 am »

The speed of the cycle is dependent upon the frequency of the AC current.  In the US that frequency is 60 Hz; I think in many European countries the frequency is 50 Hz.  This means that the instantaneous voltage will change from zero to a maximum, back to zero,  to the maximum again in the opposite direction, then back to zero sixty times a second.

Incandescent bulbs, by their nature, don't react to the voltage change that quickly. They sort of "average" out the changes and are relatively constant, both as far as luminosity and color temp.

Other kinds of lights, particularly sodium and mercury based lights, do not "smooth" things over. The color temperature and light output actually change with the input voltage fairly substantially.  Fluorescent lights also have the same problem, although not as strongly as those others.

If your shutter speed is slow enough, you end up "averaging" the light through at least one or more cycles.  If, on the other hand, your shutter speed is fast you can end up with only part of the cycle, and your color temp and measured light can vary from frame to frame, depending upon when you hit your shutter.

Finally, those lights themselves, used in the US for gym, industrial, and street lighting, don't have anywhere close to a continuous spectrum at any time.  Go grab a cheap diffraction grating or prism and you'll see really strange spikes in the spectrum. These lights are bright, fairly low in heat production, fairly efficient, long lived, and cheap for their light output compared to incandescent, which explains their use.

Tom Frerichs

Many thanks Tom, that fills in another gap and would explain why I've experienced variance in colour temp on a couple of occasions. I had always considered artificial light as consistent but it now appears not. And yes, it is 50hz this side of the pond.
Logged
Pages: [1]   Go Up