Tuning the Light Spectrum To Improve Energy Efficiency
New research suggests that by tuning the light spectrum, light levels can be reduced without compromising visual performance. if this holds true, it might mean substantial energy savings in the future. By Guest Writer Sam Berman.Tuning the Light Spectrum To Improve Energy Efficiency
Because of the possibility of compromising visual performance, lighting engineers and designers generally do not recommend lowering light levels in the workplace as a means of reducing lighting energy consumption. However, researchers at the Lawrence Berkeley Laboratory (LBL) in California are proposing that by changing the spectral distribution of the light, it may be possible to decrease light levels without reducing performance. Energy savings could then accrue depending on the relative efficacy of light production of the original and spectrally modified lamps.
The rationale for this proposal is based on new findings of the LBL team about how light affects human vision. The eye is a light sensing system with a photoreceptive medium (the retina) and an aperture (the pupil). The retina is composed of two types of photoreceptors, cones and rods. The cones provide for color vision and for viewing fine detail while the rods are thought to be primarily associated with night vision. Cones and rods have different spectral responses, i.e., the cones are most receptive to green light at the wavelength of 555 nanometers while rods are most receptive to blue-green light at 507 nanometers.
Light meters and photometric devices are generally calibrated to the cone spectral sensitivity known as the photopic response. As a result, the light output of lamps (lumens) is rated only in terms of its photopic content. The rod spectral sensitivity, known as the scotopic response, has generally not been considered of relevance for interior lighting.
However, the LBL team has determined that for conditions of full-field viewing and light levels typical of building interiors, the predominant spectral determinant of the pupil aperture is scotopic rather than photopic. Although in general, increases in light level will generally cause decreases in pupil size, white light whose spectral distribution is weighted more in the blue-green will be more efficient in contracting the pupil than white light which is relatively deficient in blue-green composition.
Because most of us have some imperfections in the lens of our eye and the imperfections cause optical aberrations, visual performance is generally improved with smaller pupil size for light levels typical of building interiors. This led the LBL team to propose that a reduction in visual performance caused by reducing light levels could be compensated for by reducing pupil size via shifting the lamp spectral distribution towards a higher scotopic content. In this manner, performance would be maintained at a lower light level which could lead to substantial energy savings.
One visual performance factor that demonstrates the LBL proposition is depth of field, which decreases as pupil size increases. The LBL research has demonstrated that by replacing the original illumination by modified illumination which is scotopically richer, depth of field can be maintained but at a lower light level.
To generalize our proposition to the workplace visual environment, other types of visual tasks need to be studied. The LBL team has recently demonstrated for a sample of 24 adults between 20 and 40 years of age that for a low-contrast-recognitions task, smaller pupils result in improved performance compared to larger pupils. Furthermore, they showed that performance decrements due to contrast reductions can be compensated for by decreases in pupil size achieved by changes in the light spectrum. Presently, they are continuing to study this task, as well as other tasks, such as acuity and reading speed, to determine the trade-off between pupil size and illumination.
Should these studies establish that pupil size is a major determinant of visual performance in workplace situations, then there are very significant possibilities for saving lighting energy through changing the spectral compositon of general illumination. It is important to note, however, that the considerations discussed here are specific for achromatic tasks, e.g., reading printed material, where color discrimination is not a component of the visual task.
With this proviso in mind, consider the group of 40-watt fluorescent lamps commonly used for interior lighting that are listed in the table below. The first column shows that these whitish-color lamps are roughly equivalent in terms of their photopic light output and efficacy (photopic lumens per watt). However, because they are made with different phosphors they have different spectral distributions and their scotopic outputs listed in the second column are quite different. The third column lists the combination of photopic and scotopic lumens which the LBL group has shown determine pupil size for conditions of full-field of view. This metric is referred to as "pupil lumens". The fourth column compares the efficacy of these lamps on the basis of pupil lumens per watt.
According to the LBL proposition, the different values of pupil lumens per watt allow a comparison of these lamps on the basis of their potential for visual effectiveness. For example, an installation lit by 40-watt CW fluorescent lamps could be retrofitted with NB (5000) lamps operating at 24% lower energy use while maintaining the same pupil size. On the other hand, although replacing a 125-watt incandescent lamp with a 35-watt high pressure sodium (HPS) lamp (both producing about the same photopic light output) appears to be an energy savings strategy it might not be the case. In terms of pupil lumens per watt the two lamps are about equally efficient, but the HPS lamp would have to be three-times brighter to maintain the same pupil size.
This problem with HPS lighting has been noticed by many lighting designers and users, who have reported difficulties in seeing well, and a perception of dimness, when ambient illumination is provided by HPS lighting. We can now suspect that these effects are due to visual scotopic sensitivity despite equal or increased photopic luminance of the HPS lighting. Several other initially puzzling findings of the effects of light spectrum on visual tasks and visual clarity that have been reported over the past 20 years can be simply understood by consideration of spectral effects on pupil size and the concept of the pupil lumen.
To determine the value of the pupil luminance for a given lighting condition requires knowing both the scotopic and photopic values, and their ratio S/P. In many cases this ratio is independent of the intensity level of the illumination. Common light sources vary by a factor of 10 for S/P ratios.
Some values of S/P are listed in the table while the diagram above shows that there is a positive relationship between the S/P ratio and the color rendering index (CRI). If, over the next few years, the concept proposed by the LBL team can be shown to apply to sensitive visual tasks representative of the visual demands of the workplace environment, users will be assured of a simple and highly cost-effective strategy for improving lighting energy efficiency while maintaining desired visual performance.
Sam Berman
Sam Berman is leader of the Lighting Systems Research Group, MS 46-125, Lawrence Berkely Laboratory, Energy and Environment Divison, Berkeley, California 94720, USA.
For further information, see S.M. Berman, "Energy Efficiency Consequences of Scotopic Sensitivity", Journal of the Illuminating Engineering Society. Winter 1992.