LED Lighting System Lifetime Accelerated Testing And Evaluation

LED (Light Emitting Diodes) is widely used in many different areas with goodadvantage of lower energy consumption, long lifetime, green which benefits topeople’s life. Using of LED lighting will give an annual global energy bill saving of€300billon and a reduction of1,000MT of CO2emission. Hence, LED lighting isrecognized as the second revolution in the history of lighting.As any other products, the reliability of LED lighting system needs to beevaluated before they can be applied at the end of customers. To ensure the system’sperformance meets the design target and customers’ needs. Currently, the customersexpect the lifetime of LED lighting system can be up to50,000hours and the warrantywith3~5years. Compare to traditional lighting (eg, incandescent lamp or fluorescentlamp) only with few thousands hours lifetime, it is a very scientific challenges. Thetest method for conventional lighting is very simple since the short lifetime and singlefailure mode. Generally, the samples are used to ageing in the lab with shorterperiod.There are some lifetime related standards published among the LED industry,such as LM80from IES and ERP from EU. However, the problem is these standardsonly focus on the Lumen maintenance, and the test must be operated with6000hoursat least. Obviously, it is a big limitation to release the LED products to marketquickly.This thesis will start the study from the LED lighting system, and aim at the everymain failure mode to perform the accelerated life test base on the failure mechanisms.Because of the LED lighting system consists different functional subsystems whichworking in close collaboration. And the subsystems include the optics, electronics,LED PCB, and thermal design. It is difficult to know the lifetime of system with oneaccelerated testing due to the subsystems or components have different failure modes.It’s a challenge for system reliability that accelerated testing condition for onesubsystem if often too strict for another subsystem. Alternatively, even the acceleratedrate possible for one subsystem may be too low to be of any use for yet anothersubsystem. According to reliability series mode, integrate all the different failuredistributions to system failure distribution which gained from the accelerated life test.Finally, the project team can know the design level, reliability metrics and predicate the warranty failure as soon as possible by this system lifetime evaluation method.