Elsevier

Microelectronics Journal

Volume 108, February 2021, 104969
Microelectronics Journal

Orthogonal study and analysis of variance on a thermal management system for high-power LED package

https://doi.org/10.1016/j.mejo.2020.104969Get rights and content

Abstract

Determination of influences of various factors is important for improvement and optimization of a thermal management system for microelectronics. In this study, a series of orthogonal experiments and analysis of variance were performed on an LED cooling system which integrates thermoelectric cooler with microchannel heat sink using nanofluid as coolant. The significance of four factors, including nanofluid concentration, ambient temperature, nanofluid temperature, and power of the thermoelectric cooler, were evaluated. Substrate temperature was measured and temperature distribution of the LED was estimated. Results indicate that nanofluid demonstrated better cooling capacity compared with water. Results of difference analysis and analysis of variance indicate that all the four factors showed significant effects on the LED temperature, while the dominant factor was nanofluid concentration. The optimum nanofluid concentration was obtained. The variation of the nanofluid effect on the cooling performance with ambient temperature and nanofluid temperature were observed. The influence of dispersion stability of the nanofluid on the cooling performance was discussed.

Introduction

Thermal management is important for improving for the efficient an lifetime of high-power LEDs [1] since poor thermal management will result in reduced reliability and shorten operation life, sustainability and reliability of the LED [2]. Even high-performance LEDs can suffer from limited lifetime without carefully optimized thermal management and driving conditions [3].

As yet, thermal management of LED is still a vital issue. There are various methods for the thermal management of LED. The technology of air-cooled heat sinks has become the most and viable solution for cooling electronic devices [4]. However, air cooling methods have been found to be in-sufficient for high-power electronic devices. Neither air nor passive liquid cooling is sufficient to maintain LED junction temperature in practical application [2].

One of the most effective methods for cooling electronic devices is based on microchannel cooler [5] whose performance can be further enhanced using nanofluid coolant [6]. To improve the heat dissipation of LED, a cooling system with thermoelectric cooler (TEC), was investigated [7]. The results show that the junction temperature was decreased by 17 ​°C compared to the condition without thermoelectric cooler. The cooling system with a thermoelectric cooler was suggested as the most efficient cooling system for the high-power multichip LEDs [8]. As reported in the previous study of the author [9], a novel LED cooling system integrating microchannel heat sink (MHS) with TEC achieved good performance. The main factors affecting LED cooling performance of the water-cooled system include the power of the thermoelectric cooler (PTEC), the inlet temperature of coolant (Ti), and the ambient temperature of LED (Ta). Nanofluid was introduced to attain improved cooling performance to the system [10]. However, the effect of nanofluid, among other factors such as the TEC power and the ambient temperature, on the performance of the cooling system still worth to be revealed.

In this paper, the nanofluid concentration (ϕ) was taken as another factor because it greatly affects the properties of the nanofluid. The dominant factors in the LED cooling system which integrates TEC and MHS were revealed by Taguchi method and analysis of variance (ANOVA). Taguchi method is a suitable way to investigate the effects of various factors by minimum trials as compared with a full factorial design [11]. Analysis of variance can be used to find out the dominant factors from experimental results [12]. The effects of the significant factors on the substrate temperature of LED were investigated. Dispersion stability of the nanofluid was investigated to evaluate its influence on the cooling performance of the system. Moreover, the temperature distribution of the LED was estimated by infrared thermography [13], which is a method to obtain temperature information in a fast and nondestructive way without any contacting sensor and without interruption of the system [14].

Section snippets

Experimental devices and processes

Fig. 1 shows the diagram of the cooling system and the test platform. The experimental set-up is based on that described in the authors’ previous work [9], and nanofluid is introduced in this study. Briefly, an LED lamp with a rated power of 20 ​W and a current of 0.6 A was cooled by integrated thermoelectric cooler and MHS, using water and nanofluid of different concentration as coolant in the MHS. A thermostatic water tank was used to maintain the temperature of the nanofluid (25 ​°C, 35 ​°C,

Start-up process of the system with different coolant

The start-up process of the cooling system by water and nanofluid is shown in Fig. 3, with PTEC, Ti, and Ta, set at 20 ​W, 25 ​°C and 40 ​°C, respectively. The Ts increased after the LED was turned on when water was used for cooling but decreased when the nanofluid was used as coolant. The start-up process gradually became stable typically after 400s when the Ts reached a constant value. The steady Ts decreased by about 7 ​°C when nanofluid was used instead of water. The result indicates that

Conclusion

Performance of nanofluid coolant combined with microchannel heat sink and thermoelectric cooler for LED cooling was investigated. Four factors, including nanofluid concentration, inlet temperature of nanofluid, operating power of the thermoelectric cooler, and ambient temperature were studied through orthogonal experiments and ANOVA to reveal their influence on the performance of the cooling system. The conclusions were obtained:

  • (1)

    The effects of ϕ, Ti, Ta on the Ts were highly significant, while

Author statement

Jiarong Ye: Investigation, Writing – original draft. Xiaohui Lin: Investigation. Songping Mo: Conceptualization, Methodology, Writing – review & editing, Supervision. Lisi Jia: Writing – review & editing. Ying Chen: Writing – review & editing. Zhengdong Cheng: Writing- Reviewing and Editing.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This work was supported by the Guangzhou Municipal Science and Technology Project (201704030107), and the National Natural Science Foundation of China (grant number 51976040). Ying Chen acknowledges the support from the Guangdong Special Support Program (2017 ​T ​× ​04 ​N371).

Cited by (10)

  • Ionic-liquid-enhanced solvent extraction mechanism: A novel concept

    2022, Journal of Environmental Chemical Engineering
  • Experimental performance of a LED thermal management system with suspended microencapsulated phase change material

    2022, Applied Thermal Engineering
    Citation Excerpt :

    However, the cooling capacity of an MHS is greatly affected by its coolant [12]. Hence, the role of coolant is vital in the performance of the thermal management system [13]. Compared with conventional active cooling, thermal management systems with phase change material (PCM) have various advantages, such as rapid absorption of heat [14] and uniform temperature distribution [15].

  • Study of the core-shell structure MoSi<inf>2</inf>@ Al<inf>2</inf>O<inf>3</inf> powder prepared by the sol-gel method in a low-vacuum atmosphere

    2022, Surface and Coatings Technology
    Citation Excerpt :

    Therefore, analysis of variance was used to determine the effect of a single factor from orthogonal experimental data (Table 4). The parameters of variance assessment contain the square deviations (SSj) for each factor, the degree of freedom (df), the mean square deviation (MS), the factor contributions (F and P-values) [25]. According to the P-value obtained by the analysis of viscosity results, the primary and secondary sequences of shelling performance were temperature, pH, and the molar ratio of deionized water to aluminum isopropoxide.

View all citing articles on Scopus
View full text