Differences between color temperature and color rendering index
The frequent use of the terms Color Temperature and Color Rendering Index (CRI) confuses users. What do these concepts really mean?
The color temperature of a light source is determined by its warmth or coldness and is expressed in degrees Kelvin (K). The term comes from the theory of physics. When an object, called a "black body-emitter", is heated, its color changes from black to red, then to yellow, white and finally to blue. At the bottom of this scale, the object is considered "warmer" in color, while at the top, its color is considered "colder." In the warmer range of the scale, the candle will have a color temperature of approximately 1800 K, while the sky in the northern hemisphere will reach 28,000 K. In practice, we usually consider the colors of artificial light sources in the range of approximately 2000 to 10,000 K.
Interestingly, two different types of lamps can have the same color temperature, but reproduce colors differently. For example, General Electric's SP and SPX fluorescent lamps have approximately the same color temperature as incandescent lamps, but the former have much less energy in the red region of the spectrum. This makes reds appear brighter under incandescent lighting than with fluorescent light sources. In turn, CRI is defined as a measure of the degree to which the color of an object illuminated by a light source deviates from its color when illuminated by a reference light source of comparable color temperature. The term originated around the 1960s and 1970s, when a system was developed that mathematically compares how much a light source changes position on the spectral scale of eight specific pastel colors compared to the same colors illuminated by a reference color source of the same color temperature, as defined by the International Lighting Commission (CIE). The average differences are then subtracted from 100 to give the color rendering index. The six complementary colors are sometimes used for special needs, but they are not used to calculate the CRI. By definition, if there is no difference in how objects' colors look, the light source is assigned a CRI of 100. Thus, small differences will have a CRI closer to 100, while larger differences will result in a lower CRI value. When comparing color temperatures in the 2000 K to 5000 K range, the reference light source is the "black body emitter" and daylight for color temperatures above this range. It is noteworthy that the color rendering index for both incandescent lamps and the sky of the Northern Hemisphere is considered equal to 100, while none of them is really flawless. Incandescent light bulbs are very weak in blue tones (try, for example, distinguishing a dark blue sock from a black sock in a room lit by incandescent light bulbs). In turn, the northern sky at 7500 K is weak when illuminated with red tones. However, CRI, despite its limitations and weaknesses, is still applicable and useful for determining the "quality" of a color. CRI was originally designed to compare continuous light sources with CRIs above 90, because below 90 you can have two light sources with the same CRI but very different color rendering. Technically, CRI can only be compared for light sources that have the same color temperature. Generally, however, light sources with high CRIs (80-100) tend to make people and things look better than light sources with lower CRIs.
CRI and LEDs
Research is currently under way and it is found that white light emitted by mixing red, green and blue LEDs is preferable to light emitted by halogen and incandescent lamps, even though the latter have higher CRIs. In fact, the International Commission on Illumination White LED Light Source Color Rendering Technical Report states: "The Technical Committee concluded that the Color Rendering Index developed by the panel is generally not applicable to predict the color rendering performance of a set of light sources if the set includes white LEDs." ...
This recommendation stems from the study of a variety of academic analyzes that looked at both phosphorus-coated white LEDs and red-green-blue (RGB) LED clusters. Reviewers evaluated the appearance of illuminated scenes using lamps with different CRIs and found that, in general, there was no exact relationship between classifications and calculated CRIs. In many cases, RGB LEDs had color rendering indices in the region of 20, but they performed well at rendering colors. One possible explanation for this is that they tend to increase the perceived saturation of most colors without shifting the hue.
The US Department of Energy recommends the following: “Long-term research and development is underway to create an updated metric system for assessing color quality that is applicable to all light sources. In the meantime, the color rendering index can be considered one of the information parameters when evaluating LED products and systems based on them. It should not be used to select a specific lighting product without prior personal assessments and product testing at the intended location.
1. Determine the visual tasks to be performed when illuminated by a particular light source. Where color fidelity is critical (for example, in a space where colors or fabrics compare in both daylight and electric light), the existing metric CRI values can be useful and useful for evaluating LED products.
2. Color rendering index can only be compared for light sources of equal color temperature. This applies to all light sources, not just LEDs. In addition, differences in CRI values less than five units are not significant. This means that light sources with a color rendering index of 80 and 84 are practically the same.
3. If color appearance is more important than color fidelity, do not rule out white LEDs just because of their relatively low CRI. Some LED solutions with CRIs as low as 25 still emit a visually pleasing white light.
4. In cases where fidelity of color reproduction or the appearance of colors are important factors, evaluate LED systems personally, and if possible, at the site of intended use.
Conclusion
Then why use CRI when it has so many disadvantages? It is currently the only internationally recognized color rendering rating system that provides consumers with some guidance. However, it is worth noting that the US State Institute of Standards and Technology (NIST) is working in this area, developing a Color Quality Scale to solve some of the problems of the existing CRI color rendering system, but this scale has not yet been universally accepted.
Dick Erdmann, Process Engineer, GE
