Is Chroma Shift the Achilles' heel of LEDs

The Illuminating Engineering Society (IES) first published the LM-80, "Measuring Lumen Maintenance of LED Light Sources" in 2008, and updated it in 2015 ("IES Approved Method: Measuring Luminous Flux and Color Maintenance of LED Packages, Arrays and Modules (IES Approved Method: Measuring Luminous Flux and Color Maintenance of LED Packages, Arrays and Modules)"). These specifications provide functions for chromatic shift measurements in addition to lumen maintenance. All versions of the LM-80 standard require the LED package to operate for at least 6,000 hours, with lumen and colorimetric measurements recorded at least every 1,000 hours.

Another concern for the lighting industry is the way in which long-term lumen maintenance and chromaticity shift predictions are made based on the limited data set obtained in the LM-80 test. In response, IES published TM-21, Technical Memorandum: Projecting Long Term Lumen Maintenance of LED Light Sources, in 2011, and updated it in 2019 to include radiation and photons Prediction of flux projection. Also released in 2019 is TM-35, Technical Memorandum: Projecting Long-term Chromaticity Coordinate Shift for LED Packages, Arrays and Modules of LED Packages, Arrays, and Modules.

Data collected using the LM-80-08 and LM-80-15 methods formed the basis for the March 2020 DoE report by Lynn Davis and Monica Hansen. This report characterizes the lumen and chromaticity maintenance of LED packages as an updated version of the report launched in 2015, which only focused on lumen maintenance. The new 2020 edition evaluates four different LED package types used in general lighting applications: ceramic-based, polymer-based, chip-on-board (COB) and wafer-level. A total of 223 LM-80 datasets, representative products from leading LED package manufacturers, were used to predict lumen maintenance and chromaticity shift according to the protocols presented in TM-21-11 and TM-35-19, respectively.

Unsurprisingly, the results of the 2020 study show improvements in LED package lifetime across all package types since the study began in 2015 (except for wafer-level packages not included in the earlier study). In the 2015 study, about one-third of LED packages were expected to last more than 60,000 hours (L70); in the 2020 study, 96% of those using L70 packages were expected to last more than 60,000 hours. The report also noted that polymer-based encapsulation has achieved the greatest gains due to improvements in polymer materials.

However, the industry's main point of interest, the area of projected chromaticity shift, was not covered by any standard until the publication of TM-35-19. This study used the Chrominance Shift Mode (CSM) defined in the table below to characterize the chromatic shift observed according to the LM-80-15 protocol test.

Three of the four LED package types included in the 2020 study showed a chromaticity shift during LM-80 testing. With the exception of COB packaging, the pattern of chrominance shift appears to be related to the package type, so no conclusions can be drawn, and the study recommends further research into the mechanisms used for COB package chrominance shift.