Color Spectra and Light Sources
|
Photographic color temperature is different from the color temperature used in physics and colorimetry. Color temperature as used in physics and colorimetry refers to the temperature at which an ideal black body would emit light of the same color (i.e., having the same chromaticity) as that of the light being measured. The important issue here is the ratio of blue-to-red (B/R) since higher B/R ratios result in higher color temperatures. When measuring the color temperature of a light source, most measurement methods assume that light source has a continuous spectral power distribution. Measurement of light sources that are not continuous (those that contain sharp peaks like fluorescent) is complicated by the nature of spike spectra. The figure below shows spectral power distributions for sunlight on the left, and on the right is the spectra of a typical fluorescent lamp.
Notice how the spectra of sunlight is continuous, albeit not uniform, over the range of 300 to 700 nm (nano meters). In contrast, the fluorescent lamp's spectra is also non-uniform over the same frequency range but exhibits large spikes of energy at about 425 nm, 525 nm, and 550 nm and then rapidly drops off past 550 nm. This is characteristic of non-continuous sources. Other non-continuous sources are metal-halide lamps and the narrow band fluorescent lamp which are shown in the following figure.
Color film is traditionally made up of 3 layers. One layer is sensitive to blue light, one to green, and one to red. Combining RGB in different ratios creates the different colors of the spectrum. However, different types of light sources may appear to the human eye as being the same. Yet, comparing the RGB characteristics of each, these sources are quite different from one another. Consequently, using the same film, different light will cause the reproduced color to change. Color film is formulated such that its sensitivity to RGB will provide correct color reproduction under specific light sources. For example, daylight film is designed to reproduce correct colors under photographic daylight. Tungsten film is divided into two types--A and B. Type A is formulated of use under photoflood lamps whereas Type B is formulated for use using studio lamps. To determine the proper color balance filtration needed to adjust a film to prevailing light source is by use of a color meter. Color meters generally have two sensors--one for red and one for blue. Again, the ratio of blue/red will determine the correct light balance (LB) filter for specific film-light combinations. However, the only way to correctly and accurately measure incident light for color balance is by use of a three-color sensor that is responsive to red, blue and green. Two-color meters will measure only two colors. These meters are perfectly acceptable for daylight but will not give accurate indications for fluorescent light sources. The reason for this is that fluorescent sources require both LB and color correction (CC) filters. Blue/red figures determine the LB filter while the G/R ratio determines the CC filter. Without making three color readings, a source with the same photographic color temperature as another source may cause dramatically different results on film. Light sources of the same photographic color temperature do not necessarily have the same characteristics. The figure below shows the sensitivity of daylight color film to red, blue, and green.
The color mix between RGB for daylight at noon, morning, evening, and overcast is shown below with the RGB mix for a typical tungsten lamp.
Some experience with different films and light sources will aid in establishing a relationship between light, film, and workable filtration. However, there is little substitute for accurate measurement of light spectra. This is particularly important for non-continuous light sources where a high quality color temperature meter is essential in making quantitative evaluation of light vs. film. Other types and sources of will give different color temperature values. The table below lists some measured temperatures for different sources of light:
Several conclusions can be inferred from this table. First, notice that the 12V halogen lamp and the grow flood lamp are virtually the same temperature. This is confirmed by similar light balance filtration. Yet, although very close in temperature, and using the same light balance filtration, their color correction filtration requirements are quite different. For example, the 12V lamp requires a 05 Magenta filter while the flood lamp requires an 08 Green filter. What does this mean? Possibly this lamp is getting close to its end of life point. Nevertheless, a reading with just two colors, red and blue, will not give a true indication of color correction. The LB or light balance filter is used to correct the B/R ratio. Alternatively, the color correction or CC filter is used to correct the G/R ratio. If the green value is insufficient in relation to the red quantity, a green filter (high green transmittance) value would be indicated. If the green value was too high, then a magenta filter (low green transmittance) would be indicated. As indicated previously, identical color temperatures do not necessarily indicate identical light-source characteristics. What about flash photography? A properly functioning electronic flash will be very close to 5500°K of standard daylight. However, when shooting in lighting conditions that are different from standard daylight, under fluorescent or incandescent light sources for example, different strategies are indicated. Using a leaf shutter, one can synchronize the flash at any shutter speed. Changing the shutter speed will allow more or less of the ambient light to reach the film in relation to the electronic flash. Consequently, the exposure time can be used to control the amount of non-daylight that contributes to exposure, thus adjusting the overall color image of the photograph. Moonlight The reflection factor of the moon is twice as much for the red as for the violet; thus it has a lower color temperature at about 4,125K. The low intensity moonlight appears bluer than it actually is as in accordance with Purkinje's phenomoenon. But moonlight is extremely variable. At full moon, it is 150,000 times weaker than sunlight. Accordingly, exposures must be doubled when the moon is at 45 degrees elevation yet multiplied by seven when it is seven days before and after a full moon. |
All Photographs Copyright © by Gary Gaugler - All Rights Reserved - Use By Permission Only |