[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: CF light temps

"allan Mlodick" <allangm at hotmail_com> wrote:

>I agree with what you have to say, but I find the Kelvin rating very 
>misleading. A 9235 K All Glass Aquarium lamp looks as though it has a strong 
>red and blue component. Even with the strong red component it is still 
>listed at 9325 K, not far from 10,000 K. This is why I asked for opinions 
>from people, like yourself, who have actually seen the lamp under planted 
>tank conditions. My guess is the All Glass 9235 K lamp has a very low CRI, 
>but when I saw it at the LFS it appeared fairly balanced (a little pink). 
>Unfortunately, all manufacturers do not provide spectral density plots for 
>their lamps, if accurate they would be helpful in selecting lights.
>Is there a better way of determining the appearance of a lamp than to go by 
>the K temp and CRI?
>For now, I am in agreement with you and will most likely purchase the 6700 K 
>lamps at AHSupply, although the Pink/White Realux does sound interesting.
>Thanks for all your help.

I have a recollection of having written this before, but here it goes: the
concept of color temperature only applies to spectra that have a minimum
resemblance to a black-body spectrum (or Planck curve). Natural light and
all artifical light designed to look like natural light, have such a 
spectrum (at least when examined with very poor spectral resolution). The 
red-blue-enhanced aquarium bulbs, on the contrary, have a spectrum that 
looks like the opposite of a black body, if such thing can be defined. 
So the color temperature rating of these bulbs is totally meaningless. 
The rating manufacturers provide is probably the result of a formal, 
standardized industrial measurement procedure that works fine for 
black-body-like light sources, but gets fooled when applied to these 
weird spectra. And the CRI of these bulbs is not just very low, it is 
meaningless as well, by the same basic reason.

Spectral curves can serve just as a rough guide in determining the appearance 
of a lamp when used to light an aquarium. _If_ the bulb is of "full spectrum" 
type (meaning the spectrum is more or less black-body-like) OR is high
CRI type (>80-90, meaning it was designed to look like natural light) then
the color temp rating makes sense and can be used as a guide. Otherwise, 
I'm afraid the only reliable indicator is to see an actual lamp over an 
actual aquarium. Color perception by the human visual system is a 
differential process, meaning we cannot see "absolute" colors outside of 
a context. So such things as the ambient where the aquarium is, the colors 
of its surroundings, the light level of its surroundings, the color of the
light at its surroundings, all affect our final perception of color. And 
to complicate a little further, it is variable from person to person too. 
Have you ever noticed how public aquariums in museums, exhibit halls and 
such, are always kept in dimly lit surroundings ? This is in part to achieve 
better control on the looks, if not for other reasons.

- Ivo Busko
  Baltimore, MD