Savitha G. Kini

Application of accelerated life testing principles to project long term lumen maintenance of LED luminaires

This work presents our effort to predict the long term reliability of LED arrays using the application of accelerated life testing principles. Assessment of long term reliability and performance of LED arrays is a testing exercise but it is also vital for successful acceptance of Solid State Lighting (SSL) systems. The objective of this work is to analyze IESNA LM-80 test data obtained from the LED manufacturers to study how failure is accelerated by stress and fit an acceleration model to the data. This acceleration model can be used to accurately project the reliability of the LED arrays under normal operating conditions. The methodology was to apply statistical analysis to LM-80 test data and obtain accelerated models for life-stress relationships and life-time distributions. The Arrhenius-Weibull, Generalised Eyring-Weibull and Inverse Power-Weibull models were obtained and were compared for their effectiveness in to predicting the reliability of LED arrays.

Prognostic algorithms for L70 life prediction of solid state lighting

The life of light-emitting diodes (LEDs) is difficult to measure by traditional testing methods as they are not likely to fail completely. The Illuminating Engineering Society of North America (IESNA) uses a standard regression approach based on short-term collected lumen data to predict the L70 lifetime of LEDs. In this paper, a model-based prognostics method is employed to determine the life of luminaires using LEDs. Unscented Kalman filter and particle filter algorithms are used for degradation model parameter estimation. An analytical approach based on three statistical models (Weibull, normal, lognormal) is employed and a best fit is determined by the Akaike information criterion. The resulting L70 is compared with L70 derived from the IESNA approach to accurately determine the best prognostic method.

Measurement of junction temperature of light-emitting diodes in a luminaire

There are vast numbers of light-emitting diode (LED) luminaires to choose from, but not all LED luminaires perform reliably. More than half of the input electrical power to a LED luminaire is wasted in the form of heat. Heat management is the one of the most critical issues faced by LED luminaire designers. The long term reliability of a LED luminaire is mainly dependent on the junction temperature of the LEDs. Hence, accurate junction temperature information during luminaire operation is critical for monitoring and assessing the health of the luminaire. In practice, it is extremely difficult to measure the junction temperature of LEDs in modern day luminaires. With the optical system and heat sink surrounding the tiny LED, measurement of junction temperature with direct methods like infrared cameras and thermocouples becomes more complex. This paper explores the possibility of monitoring and measuring the junction temperature of LEDs in a luminaire by making use of the strong correlation between the forward voltage drop at the LED junction and the temperature of that junction. Results of thermal investigations of a LED downlight are presented. The results suggest that the inherent forward voltage/ junction temperature dependency of LEDs can be used to measure and monitor LED junction temperature in a luminaire under operational conditions.

Measurement of junction-to-ambient thermal resistance of a LED lighting luminaire

Recent advancement in solid state technology has made LED lighting energy-efficient and today it is one of the rapidly-developing lighting technologies. Design of LED lighting luminaires is still repeatedly governed by misinterpretations and inappropriate use of the related solid state lighting technology. Heat is generated as byproduct of the light generation process at the LED junction; this heat needs to be dissipated in an efficient manner. The correct method of thermal management in the LED luminaire is critical for effective heat transfer from the junction to the ambient. Light output, useful life and long term reliability of an LED luminaire are strongly associated with the temperature of the LED junction. Inadequate heat dissipation influences LED life and consequently the long term reliability of the entire LED luminaire. It is therefore important to have the proper knowledge about the light-output and thermal properties of the LED luminaire. This work proposes a methodology for the measurement of the junction-to-ambient thermal resistance of a market ready LED luminaire using the linear relationship between forward voltage and junction temperature of the LEDs. Measurement of the junction-to-ambient thermal resistance of a LED luminaire is essential to assess the quality of the thermal design of the luminaire. Considering two LED luminaire with similar electrical and photometric characteristics, the LED luminaire with lower junction-to-ambient thermal resistance has better heat management system compared to the one with a higher value of the junction-to-ambient thermal resistance. The methodology used in the study and findings of this study are discussed in this paper. If two similar LED luminaires are tested using the proposed method, their thermal management systems can be assessed based on junction-to-ambient thermal resistance.

A review of existing methods for study and analysis of thermal design of LED lighting products

The lifetime of LED luminaires, made of high power LED arrays, operating under hostile conditions is determined by their thermal design, operating junction temperature and thermal transients managed by the thermal system. Thermal transient characterization is a noninvasive study of the thermal resistance characteristics or thermal design of the LED luminaire. This procedure can identify anomalies in the thermal design or irregularities in the heat conduction path from the LED junction to the ambient. Any increase of thermal resistance along the heat conduction path can cause excessive junction temperature build up and cause early failure or increased degradation of device. The scope of this work is study the existing method for thermal transient characterization and analysis using the evaluation of the structure function of the heat conduction path of LED lighting product. It can be concluded that structure function approach is a very superior procedure for thermal transients characterizing the heat conduction path and estimating junction-to-ambient thermal resistances.

LED lighting reliability from a failure perspective

Incredible long-life makes LED lighting systems a good long-term investment The higher capital cost is justified as there is long-term energy and maintenance cost savings. Operational requirements and materials used in manufacturing a of LEDs make then-failures distinctive compared to other microelectronics devices. Significant effort has gone into understanding of the failure modes and mechanisms and reliability of LED lighting. Although still very incomplete, our knowledge of the reliability issues relevant to LED lighting is increasing. This paper provides an overview of LED lighting failure modes and mechanisms that are commonly encountered. It focuses on the reliability issues of LED lighting.

FPGA based buck controller for LED white light tuning

New developmentsin solid state lighting allows us to use energy efficient LEDs in general lighting. Solid state lighting provides both energy and environmental benefits. One of the main specialties of LEDs over conventional light source is the controllability of its color temperature. This paper illustrates dynamic color temperature tuning of the multi colored LEDs to get different shades of white color using FPGA. The color tuning allows the user to control the human performance. The principle of additive color mixing algorithm is adopted for getting the desired white color from a set of multicolored LEDs. The control of intensity and color temperature is achieved by 4 channel current controlled buck converter by adjusting the PWM of each. The output current compensation for the LED is given by a type 2 PI controller. The proposed design and control algorithm assure very small color temperature variations and allows the user a full control over the luminance.

Review of methods for reliability assessment of LED luminaires using optical and thermal measurements

LED luminaires are extensively used in a variety of applications. The LEDs are driven by from fraction of an ampere to nearly one ampere. LED luminaires are required to operate at high temperatures, depending on the application. High-power LED luminaires generate too much heat. If the heat generated is not dissipated away from the LED device it will affect the useful operating life and color characteristics of the luminaire. This work examines the need and rewards of detailed thermal modeling of LED luminaires. The purpose of this work is to develop a thermal model of LED luminaire such that the model represents the actual physical system. This model will be useful in reducing the experimental analysis. The scope of this work is to develop a thermal model of LED luminaire and validate it theoretically.

Analysis of Thermal Comfort and Visual Comfort – A Soft Computing Approach

The entry of daylight into the interiors depends on the orientation of windows, size of the windows, time of the day, as well as weather conditions. The position of the blinds also adds to the above factors to allow day light inside. The illuminance inside a room varies throughout the day depending on the above factors. The introduction of daylight inside the room also aids to increase in temperature inside the room, which hinders the thermal comfort of the user inside the room. Thermal comfort can be achieved by installing Air Conditioner (AC) inside the room. But due to the excessive heat inside, the energy consumption increases to maintain the thermal comfort. This paper deals with the various measures taken to reduce the energy consumption. These values are then fed into a fuzzy logic controller to get the optimum result.

LED life prediction based on lumen depreciation and colour shift

Light emitting diodes, with advantages in energy savings, luminous efficacy and greater reliability, are becoming preferred over conventional white light sources. Currently, only light output depreciation is considered for life estimation of light emitting diode luminaires but it is recommended to include colour shift variations for applications demanding colour stability. In this paper, an extended Kalman filter is employed to determine L70 life and colour temperature degradation over life of a light emitting diode luminaire. The colour shift in terms of Duv is determined by statistical polynomial cure fitting. The variation in chromaticity coordinates over life is determined and life based on colour shift is determined by acceptable Duv limits. The results are compared to life determined by the IES-TM-21 method and the correlated colour temperature limits taken from the luminaire data sheet.