The Paradoxes of Color Temperature
Posted in Physics, Science, Seattle, Sustainability on February 22nd, 2009 by leodirac – Be the first to comment
Last week I went to the Indoor Sun Shoppe in Fremont and got a couple new CF bulbs for the house. I love their selection — they have everything from tiny 7W candelabra bulbs to these massive 150W bulbs that look like death-rays. A giant 105W bulb (pictured) is now trying to make my monstera deliciosa's home in the living room a little more like tropical mexico and less like winter-in-seattle.
In addition to a huge range of powers, they also clearly show you the color temperature of each bulb. Some of my friends have avoided CF bulbs because of their harsh color. But not all CF bulbs cast a vampirish hue on everything. In fact if you know what to look for, you can tell how cool or warm the color will be by reading the box. But not always. Depends on the brand.
The key is to look for a color temperature number like 5000 K or 2700 K. The higher the number, the more cool or blue the light will be. The lower numbers will be warmer or more yellow. Bulbs that are described as "full spectrum" typically do so because their color temperature matches that of regular sunlight — 5000 K or 6000K, but indoors these lights look pretty blue. A typical incandescent bulb will be more like 3000 K. Here is a good page showing what color temperature numbers typically mean.
Indoor Sun has CF bulbs at 2700 and 4000. They're not quite as efficient, but they're still a lot cleaner than incandescent, and if it pushes you away from "I won't use them because they're ugly" then that little efficiency drop is well worth it.
A little science
The irony of color temperatures is in our vocabulary for describing them. What we call a "cooler" light with more blue in it actually corresponds to a hotter temperature. When we describe a light as 5000 K we mean this is the spectrum of light that would be emitted by something heated to 5000 degrees Kelvin, or about 8500 Farenheit. (Technically, it's a black box radiation spectrum, but most hot objects radiate pretty darned close to a theoretical black body.) Just as bluer flames represent hotter combustion, so with color temperature. But we still call lights "warm" when they've got plenty of yellow and red in them and not so much blue.
Putting these numbers in context gives us a little physical grounding for lighting. With a basic incandescent bulb, we really are heating a tiny filament up to about 3000 Kelvin, just to see it glow. Incandescent bulbs are ancient, incredibly simple, and really inefficient. The color temperature of sunlight is about 6000 K, because that's just how hot the surface of the sun is. Thinking about how the sun is this amazingly hot nuclear fire that powers practically everything on the planet, it might be surprising that we can achieve about the same temperature in a piece of wire protected by nothing more than a couple inches of glass globe. The discrepency there is because the atom smashing fun doesn't happen at 6000 K on the surface — the real power is in the middle of the sun where things are well over 10,000,000 Kelvin. And even heating your bit of wire that hot would start a nuclear fire without the incredible pressure caused by gravity pushing things together. So in case you were worried, there really is no danger of making a hydrogen bomb out of a lightbulb, just because you can get it as hot as the surface of the sun.
[Oh and props to Six Apart for updating the typepad editor and supporting Chrome. Thanks!]
