A Technique for Improving the Precision of the Direct Measurement of Junction Temperature in Power Light-Emitting Diodes
Demetrio Iero,
Massimo Merenda,
Riccardo Carotenuto,
Giovanni Pangallo,
Sandro Rao,
Gheorghe Brezeanu,
Francesco G. Della Corte
Affiliations
Demetrio Iero
Department of Information Engineering, Infrastructure and Sustainable Energy (DIIES), Mediterranea University of Reggio Calabria, 89124 Reggio Calabria, Italy
Massimo Merenda
Department of Information Engineering, Infrastructure and Sustainable Energy (DIIES), Mediterranea University of Reggio Calabria, 89124 Reggio Calabria, Italy
Riccardo Carotenuto
Department of Information Engineering, Infrastructure and Sustainable Energy (DIIES), Mediterranea University of Reggio Calabria, 89124 Reggio Calabria, Italy
Giovanni Pangallo
Department of Information Engineering, Infrastructure and Sustainable Energy (DIIES), Mediterranea University of Reggio Calabria, 89124 Reggio Calabria, Italy
Sandro Rao
Department of Information Engineering, Infrastructure and Sustainable Energy (DIIES), Mediterranea University of Reggio Calabria, 89124 Reggio Calabria, Italy
Gheorghe Brezeanu
Faculty of Electronics Telecommunications and Information Technology, University Politehnica Bucharest, 061071 Bucharest, Romania
Francesco G. Della Corte
HWA srl-Spin Off dell’Università Mediterranea di Reggio Calabria, Via Reggio Campi II tr. 135, 89126 Reggio Calabria, Italy
Extending the lifetime of power light-emitting diodes (LEDs) is achievable if proper control methods are implemented to reduce the side effects of an excessive junction temperature, TJ. The accuracy of state-of-the-art LED junction temperature monitoring techniques is negatively affected by several factors, such as the use of external sensors, calibration procedures, devices aging, and technological diversity among samples with the same part number. Here, a novel method is proposed, indeed based on the well-known technique consisting in tracking the LED forward voltage drop when a fixed forward current is imposed but exploiting the voltage variation with respect to room temperature. This method, which limits the effects of sample heterogeneity, is applied to a set of ten commercial devices. The method led to an effective reduction of the measurement error, which was below 1 °C.
