Lumen and chromaticity maintenance lifetime prediction for LED lamps using a spectral power distribution method

In this study, a spectral power distribution (SPD) based method is developed to analyze and predict reliability characteristics of Light Emitting Diode (LED) lamps, such as lumen and chromaticity maintenance lifetimes. Usually, an entire SPD of a LED lamp can be regarded as the superposition of an asymmetric blue with a phosphor converted spectrum, each of which can be fitted by an Asymmetric Gaussian (AGaussian) model. During the ageing process, the decomposed SPD model parameters extracted from the test data of a LED lamp are fitted further to an exponential degradation model. Then at any given time, the lumen maintenance φ and color coordinates u' and v' can be calculated by predicting the SPD of a LED lamp. In order to demonstrate the application of the proposed method, a batch of ten LED PAR20 lamps equipped with four phosphor converted white LEDs (pc-white LEDs) were aged at 25 °C for 4000 hours. And a statistical analysis was performed to calculate the median values (50 percentile) of φ and du'v' with both upper boundary (5 percentile) and lower boundary (95 percentile). The results show that most of the measured φ and du'v' can be enveloped within the lower boundary curves predicted by the proposed model. In addition, there is a good prediction result of the φ curves by using the proposed method and a published statistical method. At last, the comparison between lumen and chromaticity maintenance lifetime predictions shows that the luminous flux depreciation failure may be the primary concern than the color shift in the selected test samples.