DESIGN AND RELIABILITY ASSESSMENT OF HIGH POWER LED AND LED-BASED SOLID STATE LIGHTING
MetadataShow full item record
Lumen depreciation and color quality change of high power LED-based solid state light (SSL) are caused by the combination of various degradation mechanisms. The analytical/experimental models on the system as well as component-level are proposed to analyze the complex reliability issues of the LED-based solid SSL. On the system-level front, a systematic approach to define optimum design domains of LED-based SSL for a given light output requirement is developed first by taking cost, energy consumption and reliability into consideration. Three required data sets (lumen/LED, luminaire efficacy, and L70 lifetime) to define design domains are expressed as contour maps in terms of two most critical operating parameters: the forward current and the junction temperature (If and Tj). Then, the available domain of design solutions is defined as a common area that satisfies all the requirements of a luminaire. Secondly, a physic of failure (PoF) based hierarchical model is proposed to estimate the lifetime of the LED-based SSL. The model is implemented successfully for an LED-based SSL cooled by a synthetic jet, where the lifetime of a prototypical luminaire is predicted from LED lifetime data using the degradation analyses of the synthetic jet and the power electronics. On the component-level front, a mathematical model and an experimental procedure are developed to analyze the degradation mechanisms of high power LEDs. In the approach, the change in the spectral power distribution (SPD) caused by the LED degradation is decomposed into the contributions of individual degradation mechanisms so that the effect of each degradation mechanism on the final LED degradation is quantified. It is accomplished by precise deconvolution of the SPD into the leaked blue light and the phosphor converted light. The model is implemented using the SPDs of a warm white LED with conformally-coated phosphor, obtained before and after 9,000 hours of operation. The analysis quantifies the effect of each degradation mechanism on the final values of lumen, CCT and CRI.