Use this URL to cite or link to this record in EThOS:
Title: Thermal management and humidity based prognostics of high-power LED packages
Author: Law, Thong Kok
ISNI:       0000 0004 7969 5908
Awarding Body: University of Glasgow
Current Institution: University of Glasgow
Date of Award: 2019
Availability of Full Text:
Access from EThOS:
Access from Institution:
While Light Emitting Diodes (LEDs) hold much potential as the future of lighting, the high junction temperatures generated during usage result in higher than expected degradation rates and premature failures ahead of the expected lifetime. This problem is especially under-addressed under conditions of high humidity, where there has been limited studies and standards to manage humidity based usage. This research provides an analysis of the factors that contribute to high junction temperatures and suggests prognostic techniques to aid in LED thermal management, specifically under humidity stress. First, this research investigates the effects of current, temperature and humidity on the electrical-optical-thermal (EOT) properties. Temperature rises within an LED because of input stressors which cause heat to build up: the input current, the operating and ambient temperature, and the relative humidity of the environment. Not only is there an accumulation of heat due to these factors that alter the thermal properties, but the electrical and optical characteristics are changed as well. By uncovering specific configurations causing the EOT performance to degrade under stress, better thermal management techniques can be employed. Second, this research proceeds to quantitatively link the EOT performance degradation to the humidity causal factor. The recent proliferation of LED usage in regions with high humidity has not corresponded with sufficient studies and standards governing LED test and usage under the humidity stressor. This has led to indeterminate use and consequentially, a lack of understanding of humidity based failures. A novel humidity based degradation model (HBDM) is successfully developed to gauge the impact of the humidity stressor by means of an index which is shown to be an effective predictor of colour degradation. This prognostication of the colour shift by the HBDM provides both academia and industry not only with an indicator of the physical degradation but also an assessment of the LED yellow-blue colour rendering stability, a critical application criterion. Using the HBDM parameters as indicators of the state of the LED, the degradation study is expanded in the development of a Distance Measure approach to isolate degraded samples exceeding a specified multivariate boundary. The HBDM and Distance Measure approach serve as powerful prognostic techniques in overall LED thermal management.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
Keywords: TK Electrical engineering. Electronics Nuclear engineering