By Mark E. Duffy, PhD, Manager—Global Industry Standards, GE Lighting
With the transformation of our lighting industry well underway, consumers are beginning to learn what lighting professionals have known for many years: light sources have color, and it can be important in application.
Globally, as the industry transitions to more efficient lighting for general service, the market is shifting to light source technologies that can be spectrally tailored, such as fluorescent and light-emitting diodes (LEDs). Tailoring a spectrum is much more than designing the light source correlated color temperature (CCT) or chromaticity coordinates; it involves detailing every aspect of color perception or characteristics to meet specific application needs.
For example, retail stores may use more saturated colors to illuminate merchandise to attract shoppers. Restaurant owners may seek softer, warmer spectrums to blend with candlelight for a romantic ambience. Eldercare facilities, facing the age-related challenges of its patients, may choose a time-controlled combination of cooler (bluish) and warmer (orange-reddish) light source spectrums to enhance circadian entrainment objectives. Horticultural facilities may use light sources with specialized spectrums to promote plant growth. Industries relying on color discrimination may benefit from spectrums featuring larger color gamut areas.
The list of applications is nearly endless, and the value of and need for spectral variety is undeniable. The new technologies enable light source manufacturers to design the spectral characteristics and better serve the specific needs of each application.
To realize this potential for spectral tailoring, lighting manufacturers need spectral freedom, which requires restraint from enacting regulatory restrictions on light sources based on their spectrum, due to the wide variation of intended applications requiring different spectral qualities.
Spectral freedom involves a paradigm shift to the evaluation of light-source efficiency in terms of radiated power in the visible range rather than lumens per watt. This article explains the concept of a new approach to light-source efficiency, reflects on the responsibilities associated with spectral freedom, and urges regulatory restraint.
Shifting toward Measuring all Visible Light
Since the CIE (International Commission on Illumination) standardized the photopic luminous efficiency function V(λ) in 1924, approximating the response of the human eye to normal light levels, the lumen has been used as a measure of luminous flux. For this discussion, it is important to note that the luminous flux does not measure how much light is emitted by the light source. The use of V(λ) to weight the spectral power distribution (SPD) integration results in a quantity (lumens) that measures the amount of green and yellow light and, to a much lesser extent, red and blue light that is emitted from the light source.
The ratio of lumens to input power (rate of energy consumption) of a light source is called the efficacy, measured in lumens per watt (lm/W). It is a popular misconception that the efficacy of a light source determines its efficiency.
A paradigm shift toward measuring all of the visible light (blue, green, yellow, and red) produced by a light source is needed for spectral freedom. The SPD, obtained to determine the luminous flux above, is integrated from 380 nanometers (nm) to 760 nm without any weighting function to calculate the radiant power (W) in the visible region. In other words, the radiant power measures the rate of visible light energy produced by a light source.
This is actually much simpler than the luminous flux to calculate. When the radiant power is divided by the electrical input power consumed by the light source, the resulting quotient is a measure of the actual radiant efficiency for conversion of electrical energy into light. Use of the radiant efficiency for visible light production by a light source is color neutral, applying equally to traditional and new technologies. As the relatively newer LED technology improves efficient visible light production, the radiant efficiency serves as a spectrally unbiased measure of efficiency.
With Freedom Comes Responsibility
With freedom comes responsibility, even for spectral freedom. As noted earlier, spectral or color quality is application-specific. Spectral quality requires a set of ratings for evaluating light sources according to the wide variety of potential applications.
I like to use an analogy with the wine industry: Wine has tremendous variety expressed in many different ways—white or red; sweet or dry; pairs well with seafood, poultry, meat, or dessert. Reading the label on a wine bottle provides insight into the vast creativity of producers to market their wines using descriptive language involving taste, smell, and sight. They use rating systems that are subjective scores assigned by wine critics and used to inform the consumer in the selection process.
Now I have whetted your appetite! Similarly, responsible spectral design and selection must use a variety of objective ratings tailored to inform specific application needs. Every existing or proposed color quality metric uses the SPD as input for the calculation. Each metric differs by the use of one or more weighting functions for the SPD as well as in the complexity of the calculation to reduce many integration quantities into a single valued metric.
These weighting functions are sometimes called action spectra to describe the effectiveness of light at each wavelength to produce the desired effect. Luminous flux remains a valuable metric that uses V(λ) as the action spectrum. Horticulture applications may use chlorophyll absorption action spectra to promote plant growth. Circadian applications rely on the melatonin suppression action spectrum of intrinsically photosensitive retinal ganglion cells (ipRGC). The CIE color rendering index (CIE Ra) uses the average of eight test-color (action spectra) samples.
Color saturation and discrimination metrics use gamut areas derived from the chromaticity appearance of multiple color samples, each with a specific reflection spectrum. Finally, color preference indexes that combine effects of enhanced saturation, minimal hue distortion, and color appearance are also proposed.
The effort by industry experts to transform metrics along with the technology is vital to responsible use of spectral freedom. A system that uses action spectrum ratings as a measure of the spectral value for each application could be considered. Such a rating system would need to be developed that provides a simple means for users and designers to select light sources suitable for each application. Each light source could receive ratings according to suitable applications in order to avoid misapplication.
Spectral freedom can be hampered by well-intended regulations aimed at optimizing one or more of the spectral metrics. Although regulations have the clear intent of improving light source efficiency and color quality—both noble goals—they carry unintended consequences.
Existing regulatory requirements based on light source efficacy (i.e., lumens per watt) have the unintended consequence of restricting spectral freedom by rewarding greenish spectra and penalizing broader, more natural, saturated, or preferred spectra. The use of luminous efficacy instead of radiant efficiency of visible light in regulations has a subtle, indirect influence on spectral design away from viewer preference in general service applications.
As creative experts in the lighting industry develop more efficient light sources, the next round of energy regulations has the opportunity to provide spectral freedom by changing the paradigm from luminous efficacy to radiant efficiency.
Potential regulatory restrictions based on color metrics will directly interfere with spectral freedom. These restrictions include color rendering (Ra), the red color rendering index (R9), gamut area, and action spectra metrics for specific general service or specialty applications.
Each of these metrics is intended to provide specific information about a spectral characteristic for proper use in a relevant application. They are under considerable revision as the part of the lighting industry’s technology transformation.
The academic community is rapidly building the base of knowledge for human vision and lighting-related science. Innovative metrics that both refine existing spectral metrics and address new application areas will be developed. Regulatory restraint requires regional authorities worldwide to refrain from attempting to improve the color quality of lighting products by setting regulations that oversimplify a very complex science.
The challenge of shifting the paradigm for the electric lighting industry, now well over a century old, is obviously great, but now is the time for spectral freedom!
This piece was originally published in the February 2016 issue of ei, the magazine of the electroindustry.
 Commission Internationale de l’Eclairage proceedings, 1924. Cambridge University Press, Cambridge, 1926
 Illuminating Engineering Society, IES TM-18-08 Light and Human Health: An Overview of the Impact of Light on Visual, Circadian, Neuroendocrine and Neurobehavioral Response, 2008
 Commission Internationale de l’Eclairage, Method of Measuring and Specifying Colour Rendering Properties of Light Sources. CIE Publication 13.3, Vienna: CIE 13.3, 1995
 Vick K. and Allen G., “Quantifying Consumer Lighting Preference,” Proceedings of the 14th International Symposium on the Science and Technology of Light Sources, Landmark Paper LP93, Como, Italy, 2014