Beyond CCT: The spectral index system as a tool for the objective, quantitative characterization of lamps

Highlights

• A simple system for a physically meaningful, quantitative characterization of lamp spectra.

• Spectral indices are straightforward to compute from the standard spectra currently obtained at any lab.

• A natural link between lighting engineering and astrophysics, relevant for the study of artificial light at night.

• A system potentially useful for industrial certification, legal regulation and biophysical studies.

Abstract

Correlated color temperature (CCT) is a semi-quantitative system that roughly describes the spectra of lamps. This parameter gives the temperature (measured in kelvins) of the black body that would show the hue more similar to that of the light emitted by the lamp. Modern lamps for indoor and outdoor lighting display many spectral energy distributions, most of them extremely different to those of black bodies, what makes CCT to be far from a perfect descriptor from the physical point of view. The spectral index system presented in this work provides an accurate, objective, quantitative procedure to characterize the spectral properties of lamps, with just a few numbers. The system is an adaptation to lighting technology of the classical procedures of multi-band astronomical photometry with wide and intermediate-band filters. We describe the basic concepts and we apply the system to a representative set of lamps of many kinds. The results lead to interesting, sometimes surprising conclusions. The spectral index system is extremely easy to implement from the spectral data that are routinely measured at laboratories. Thus, including this kind of computations in the standard protocols for the certification of lamps will be really straightforward, and will enrich the technical description of lighting devices.

Introduction

Observational astronomy progressed for centuries having black bodies as its almost only matter of study: the stars. And for millennia, the only detector system in astronomy was the human eye, unaided or aided by optical devices, what defined the sensitivity curve of human sight as the only spectral band effectively available for the study of the universe.

The end of the xix^th century brought the photographic revolution and, with it, a different sensitivity curve that covered a slightly different spectral region, biased towards bluer wavelengths. Even though photographic emulsions are less sensitive than the eye, this new technology allowed the study of much fainter celestial objects, thanks to the possibility to accumulate light during very long exposure times.

Approximately at the same time, spectroscopic techniques led to the discovery of non-thermal emitters in astrophysical contexts: emission nebulae whose light is made up mainly from narrow lines of ionized atoms such as hydrogen, oxygen or sulfur.

More and more non-thermal astrophysical sources have been discovered since then. Also, technological progress opened the whole electromagnetic spectrum to astrophysics, and many different bands have been defined, even inside the optical window (that roughly covers from the near-UV to the near-IR). One of the most used photometric systems in observational astronomy is the so called Johnson-Cousins, based on a set of five filters (shown in Fig. 1, and that will be commented later in Section 2.1 and Table 2), but many others exist.

Interestingly enough, there exists a strong analogy between the evolution of observational astronomy and that of lighting engineering, that we have outlined in Table 1. Also this field began with black bodies as the only working matter (combustion of solids or gas, and incandescent lamps), and only one sensitivity curve was considered at the beginning: The photopic (day-time) sensitivity curve of the human eye. Somewhat later, the scotopic (darkness-adapted) sensitivity curve was added, and it was found to be much more sensitive, and biased towards the blue. Later on, new light sources have appeared, that are not thermal emitters, such as discharge lamps and light-emitting diodes (LEDs).

Huge advancements have happened in recent times, in the research of photo-sensitive pigments in humans and in other species both animal and vegetal, what implied characterizing many spectral sensitivity curves that complement the traditional ones. In this context of non-thermal emitters and multiplicity of spectral bands, analog to the evolution experienced in astronomy, it arises the need to re-think those concepts used in lighting engineering that are based upon the properties of human vision and of thermal light sources.

Correlated Color Temperature (CCT) is a way to link human perception of the hue of lamps to the thermodynamic temperature of black bodies. The current official definition can be found at [5]. For non-black-bodies, CCT lacks rigorous physical meaning and it provides just a perceptual indication of the hue of the light. Significant differences in the perceived hue are admitted, even for sources having the same CCT. In a multi-band (even non-human-band) and non-thermal context, CCT loses most of its meaning, even in spite the efforts to bring this parameter to the limit of maximum numerical accuracy as in [4]. We have to ask ourselves whether better methods do exist, to characterize the spectral properties of lamps. From the qualitative and unsatisfactory, one-number CCT descriptor, to the heavy power of giving the whole spectrum in high resolution as suggested by Lucas et al. [13], there has to be some middle point that allows us to work with just a few numbers, with univocal and clear physical meanings, that should even make it possible to perform meaningful calculations (something completely out of place with CCT).

We explore the promising prospects that arise from the adaptation of some of the techniques developed in multi-band astronomical photometry, to lighting engineering. In particular we will work on the so-called color index system, that we propose to translate into a format suited to the description of lamps under the name of spectral index system: Converging solutions for two fields of study that have followed parallel trajectories during the last two centuries.