Incandescent lamps, simple artifacts with radiation spectrum very similar to a black-body emitter, are traditional standards in photometry. Nowadays LEDs are broadly used in lighting, with great variety of spectra, and it is convenient to use standards for photometry with spectral distribution similar to that of the measured artifact. Research and development of such standards occur in several National Metrology Institutes. In Brazil, Inmetro is working on a practical solution for providing a LED based standard to be used for luminous flux measurements in the field of general lighting. This paper shows the measurements made for the developing of a prototype, that in sequence will be characterized in photometric quantities.
IntroductionFor a long time, laboratory standards for photometry have been incandescent lamps calibrated at the correlated color temperature (CCT) of 2 856 K, which corresponds to the so called illuminant A [1,2]. The spectral power distribution of light emitted by incandescent lamps is very similar to that of a blackbody radiator, which was mathematically described by Planck. Following Planck's law, the measurement of the spectrum of an incandescent lamp allows for the calculation of its CCT and, conversely, the knowledge of the CCT allows the determination of its spectrum.Since the spectral response of instruments do not match perfectly to the V(λ) function, the spectrum of the standard to be used should be similar to the spectrum of the lamp to be measured, or of the existing luminous source where a calibrated instrument will be used. Thus, the development of LED based standards is crucial.Spectral power distributions of LED sources used in general lighting (phosphor-based white LED) have great similarity, with a rather sharp peak emission in the blue region accompanied by a broad peak in the yellow, what permits the use of a small number of different spectra to cover the entire range of non-special LED light sources [3]. This boosted the application of LED to general lighting.LED standards generally require control of the junction temperature because of its strong influence on the luminous emission [4]. Alternatively, this can be achieved by waiting the stabilization of temperature and emission, within the bounds specified by the LED manufacturer. The first solution has greater complexity and costs, and the second leads to long waiting periods (sometimes, in excess of one hour), to obtain the stabilization, which implies in waste of time and reduced number of utilizations of the standard between calibrations, and depends upon the ambient temperature.
