Analysis of the heat dissipation capacity of commonly used heat dissipation methods in power electronic equipment
2026-04-06 07:38:26··#1
1 Introduction With the continuous development of electronic assembly technology, the size of electronic devices tends to be miniaturized and the system tends to be more complex. High heat density has become an irresistible development trend. In order to meet the demand for high heat density, traditional heat dissipation methods such as fans and heat sinks are constantly being innovated, and new and efficient heat dissipation methods are emerging one after another. In the face of many heat dissipation methods, distinguishing the heat dissipation capacity of various heat dissipation methods and choosing an economical and reliable heat dissipation method has become a major concern for designers. This paper focuses on two commonly used heat dissipation methods, air cooling and water cooling, and summarizes the research results of these two heat dissipation methods in domestic and foreign literature, and summarizes the heat dissipation capacity of these two heat dissipation methods, providing a reference for thermal designers to choose an economical and reasonable heat dissipation method. 2 Analysis of the heat transfer capacity of various heat transfer methods The approximate range of heat transfer coefficients of various heat transfer methods is shown in the attached table [1]. For air, the heat transfer coefficient of natural air cooling is very low, with a maximum of 10w/(m2k). If the temperature difference between the heat sink surface and the air is 50℃, the maximum amount of heat carried away by the air per square centimeter of heat dissipation area is 0.05w. The heat transfer mode with the strongest heat transfer capacity is the heat transfer process with phase change. The heat transfer coefficient of water in the phase change process is on the order of 103 to 104. The reason why the heat transfer capacity of the heat pipe is so large is that the heat transfer process of its evaporation section and condensation section is phase change heat transfer. Reference [2] gives the reference basis for selecting the heat dissipation mode according to the heat dissipation volume and thermal resistance, as shown in Figure 1. For example, for the heat dissipation mode with a thermal resistance requirement of 0.01℃/w, if the volume is limited to 1000 in3 (1in3=16.4cm3), the air cooling mode can be selected, but a high-efficiency air cooling radiator must be equipped; while if the volume is limited to 10 in3, only the water cooling mode can be selected. 3 Air cooling The air cooling mode has low cost and high reliability, but due to its small heat dissipation capacity, it is only suitable for situations where the heat dissipation power is small and the heat dissipation space is large. The current research hotspot of air-cooled heat sinks is to integrate heat pipes with heat sink fins, and use the high heat transfer capacity of heat pipes to transfer heat evenly to the surface of fins, improve the uniformity of fin surface temperature, and thus improve its heat dissipation efficiency. Forced air convection cooling is a commonly used heat dissipation method for power electronic components. Its common structure is a heat sink plus a fan. Although this structure is easy to implement and has a low cost, its heat dissipation capacity is limited. Taking the cooling of Intel Pentium 4 CPU (2.2 GHz) as an example, we can illustrate the heat dissipation range of the common air-cooled structure. The CPU generates about 55 W of heat, the surface allowable temperature is 70 °C, the chip size is 12 × 12.5 × 1.5 mm, and the heat diffusion copper plate size is 31 × 31 mm [3]. The limited installation space for the heat sink plus fan is 80 × 60 × 50 mm. Manish Saini conducted an experimental study on the maximum heat dissipation of the common air-cooled structure under this condition [4]. Icepak simulations show that the thermal resistance of a 31×31mm heat-diffusing copper plate is equal to that of a 16×16mm copper plate under uniform heating. The experiment used a 16×16mm uniformly heated copper plate as the heat source and a standard heat dissipation structure. Results show that when the CPU surface temperature is 70℃ and the ambient air temperature is 35℃, within a heat dissipation space of 80×60×50mm, the maximum heat dissipation is 89.4W when the fan is in a top-blowing configuration and 78.2W when it is in a side-blowing configuration. Analysis of this experimental data indicates that the maximum heat flux density of the heat source is 34.9w/cm² when the fan is in a top-blowing configuration and 30.5w/cm² when it is in a side-blowing configuration. To adapt air-cooling systems to the new requirements of high heat density heat dissipation, thermal designers are modifying the packaging technology and form of electronic components and designing new air-cooling devices to broaden the applicability of air-cooling systems. In terms of changing the packaging form, flip chip manufacturing using substrate-on-top flip packaging technology and ball grid array (BGA) using printed circuit boards as substrates have improved the heat dissipation performance of the packaging module. In the design of new air-cooling devices, ralph l.webb, shinnobu yamauchi[5] et al. designed a heat dissipation device called air-cooled thermosyphon for computer CPUs, the structure of which is shown in Figure 2. The device consists of a thermosiphon and a heat sink. The shell material of the thermosiphon is aluminum, the working fluid is R134a, and the air-facing area of the heat sink is 75×90mm (16mm wide). The experimental method is still to uniformly heat a 16×16mm copper plate. The experimental results show that when the CPU surface temperature is within the permissible range, the maximum heating density of the copper plate is 39w/cm2, that is, the device can remove 100w of heat from the CPU. This is the air-cooling device with the largest heat dissipation capacity reported so far. The drawback of this device is that it can only be installed vertically because there is no core inside the thermosiphon; the liquid can only return to the heated surface by gravity. Based on the above analysis, if the uniform heat flow from the heat source at the bottom of the radiator is used as the standard for the heat dissipation capacity of the air-cooled device, when limited by the heat dissipation space, the heat dissipation limit of the air-cooled device is approximately 40 W/cm². If not limited by the heat dissipation space, increasing the fan airflow and increasing the radiator area will result in a higher heat dissipation capacity for the air-cooled system. Designers can choose a suitable air-cooled device based on the heat dissipation density and the size of the heat dissipation space. 4. Water Cooling and Other Heat Dissipation Systems Although air-cooling technology is constantly improving, air cooling itself is limited by its heat dissipation capacity. With the continuous increase in heat flux density, the application of water-cooled devices with greater heat dissipation capacity will become increasingly prevalent. According to the attached table, the heat transfer coefficient of forced convection gas is approximately 20-100 W/(m2℃), while the heat transfer coefficient of forced convection water is as high as 15000 W/(m2℃), which is more than 100 times that of forced convection gas. The heat transfer coefficient of boiling water is even higher, reaching 25000 W/(m2℃). The maximum heat dissipation capacity of water cooling devices has not yet been studied. The heat dissipation capacity of water cooling systems is explained below through the heat dissipation performance of several cooling devices. A common way to cool printed circuit boards or hybrid circuit substrates is to connect them to cold plates that are cooled by air or liquid [6]. The cold plate adopts a hollow structure, usually with a honeycomb or swirling structure inside. The working fluid is usually water. The cooling water carries away heat through forced convection cooling. The flow of water in the pipeline can be divided into three flow states according to the size of the Reynolds number (re): laminar flow, transitional flow, and turbulent flow. The literature [7] gives empirical formulas for the Nusset number under different flow states, which can be used as the basis for calculating the heat dissipation of the cold plate. For cold plates, users are most concerned with the two parameters of thermal resistance and flow resistance. During the design process, designers aim to obtain the relationship between the thermal resistance and flow resistance of the cold plate; that is, under a certain thermal resistance requirement, the lower the flow resistance, the better. This relationship is obtained partly through the designer's experience and partly through theoretical analysis; currently, theoretical research in this area is still insufficient. Boiling heat transfer has a greater heat transfer capacity than forced convection cooling. Currently, various cooling devices have been designed to cool high heat flux density chips through the boiling heat transfer of liquids. Heffington et al. designed a device that uses vibration-induced droplet atmozation (vida) to cool the heated surface. The device structure is shown in Figure 3. The device is a closed cavity with heat sinks around its perimeter, with a diameter of 50 mm and a thickness of 20 mm. A piezoelectric actuator and a small amount of liquid (water or FC-72) are installed at the bottom of the cavity. The actuator vibrates to generate atomized droplets, which splash onto the heated surface, forming a continuous liquid film. Simultaneously, the liquid film vaporizes and carries away heat. The vapor inside the cavity is cooled by a radiator on the outer surface of the cavity, forming a liquid that returns to the bottom of the cavity under gravity. Experiments by Heffington et al. showed that if air cooling is used on the outer surface of the cavity, the maximum heat dissipation capacity of the device is 100 W/cm² when the heated surface temperature is 100°C; if water cooling is used, the heat dissipation capacity can reach 200 W/cm². In the mid-1980s, American scholars Tucherman and Pease reported a microchannel structure as shown in Figure 4. This structure is made of a material with high thermal conductivity (e.g., silicon). The channel width (wc) and channel wall thickness (ws) are both 50 μm, and the channel aspect ratio (b/wc) is approximately 10. The heat q applied to the bottom surface (w×l) is conducted through the microchannel wall into the channel and then carried away by the forced convection fluid. Due to the small size of the microchannel, the heat transfer behavior within the channel is completely different from that of large-scale channels. Their experiments showed that when the water flow rate is 10 cm³/s and the water temperature rise is 71°C, the cooling heat flux reaches as high as 790 W/cm². It is currently the water-cooled device with the largest heat dissipation capacity. The emergence of microchannels has met the cooling needs of the ever-increasing heat flux density of microelectronic chips. Undoubtedly, it also has broad application prospects as an efficient and compact heat exchanger or cooling device in other fields. Based on the above introduction and analysis of water-cooled devices, it can be seen that its heat dissipation capacity is 1 to 2 orders of magnitude higher than that of air-cooled devices. Moreover, its heat dissipation capacity has not been fully explored. With the continuous increase of heat flux density, its application will become more and more widespread. 5 Conclusion This paper summarizes the research results of air-cooling and water-cooling, two common heat dissipation methods, in domestic and foreign literature, and summarizes the heat dissipation capacity and applicable scope of these two heat dissipation methods, providing a reference for thermal designers to choose an economical and reasonable heat dissipation method. (1) When limited by heat dissipation space, the heat dissipation limit of the air-cooled system is about 40w/cm2. If not limited by heat dissipation space, increasing the fan air volume and increasing the heat sink area will make the heat dissipation capacity of the air-cooled system even higher. (2) The heat dissipation capacity of water cooling system is 1 to 2 orders of magnitude higher than that of air cooling system. Its heat dissipation potential has not been fully explored. At present, the cooling method of forced convection of water in microchannel is the method with the largest heat dissipation capacity in water cooling system, which can reach 790w/cm2. References [1] Yang Shiming. Heat Transfer [M]. Beijing: Higher Education Press, 1997. [2] Sukhvinder Kang, Cooling Technologies for Power Electronics, Report in Xi'an Jiaotong University, 2009. [3] Intel Pentium 4 Data Sheet, Table 37, p93, January 2002. [4] Manish Saini, Ralph L. Webb. Heat rejection limits of air cooled plane fin heat sinks for computer cooling. The eighth intersociety conference on thermal and thermo-mechanical phenomena in electronic system. 2002. [5] Ralph L. Webb, Shinobu Yamauchi, et al. Remote heat sink concept for high power heat rejection. IEEE Transactions on Components and Packaging Technologies, Volume: 25, Issue: 4, Dec 2002. [6] Hu Zhiyong. Development trend of cooling technology for electronic devices today [J]. Electromechanical Engineering, 1999(2). [7] Jerry Sergent, Al Krum. Thermal Management Handbook. 1998