Realization of Integrated High Power LED Street Lamp Radiator
At present, high-power LED light sources are divided into two types. One is array distributed high-power LED light source, which distributes several LEDs in the array, as shown in Figure 1. The other is an integrated high-power LED light source, which integrates and packages several LEDs together, as shown in Figure 2. The two types of LED lamps are different in light distribution curve, occupied space and heat dissipation due to different LED chip layout. Relatively speaking, the lamps made of integrated high-power LED light source have lighter quality and less materials in terms of packaging materials. Compared with array distributed high-power LED light source, they can also meet the requirements of street lamp lighting, which is the development trend of street lamps in the future. However, because the heat dissipation is more difficult than the array type, the service life is shortened, which has become a key problem hindering the development of integrated high-power LED light source.Figure 1 array distributed high power LED light sourceFigure 2 integrated high power LED light sourceThis paper mainly uses ANSYS finite element software to optimize the structure of LED street lamp radiator with integrated high-power heat source. The service temperature of high-power LED lamps is required to be below 75. Therefore, the purpose of this optimization is to reduce the quality of the radiator while minimizing the junction temperature of the LED chip and less than 75.1 heat transfer theory and thermal analysis1.1 basic theory of heat transferThere are three main methods of heat transfer: heat conduction, heat convection and heat radiation. In the heat dissipation system of LED street lamp, there are three heat transfer modes, but mainly heat conduction and heat convection. The strength of thermal conductivity depends on the product material, which has been studied in many articles, and the research shows that the key to solve the LED heat dissipation problem is not to find the material with high thermal conductivity, but to change the heat dissipation structure or heat dissipation mode of LED. Therefore, this paper mainly considers the difference of heat dissipation effect caused by different heat spreader structures.The basic calculation formula of convective heat transfer is Newton's cooling formula. If the temperature difference is recorded as â³ T and it is agreed that it is always positive, then Newton's cooling formula is:Where h surface heat transfer coefficient, unit: w / (M2 K).A heat exchange area, M2.It can be seen from the convective heat transfer rate equation (1) that to increase the convective heat transfer can be achieved by increasing the temperature difference, increasing the surface heat transfer coefficient and increasing the heat transfer area. For the LED street lamp with natural convection heat transfer, the method of increasing the temperature difference and surface heat transfer coefficient is not convenient. Therefore, this paper mainly increases the heat exchange surface area.The use of fins is an effective method to increase the heat exchange surface. It can make the heat flow conduct along the height direction of the rib and dissipate heat to the surrounding environment by convection or convection plus radiation. The larger the heat dissipation area, the better the heat dissipation effect, but it is not a simple proportional relationship.1.2 radiator model establishmentThe preliminary design of this paper adopts flat fin radiator, as shown in Figure 3. Its structural parameters include fin thickness, height, length and substrate length, width and thickness. These six parameters are analyzed by ANSYS software for the structural design of the radiator.Figure 3 primary radiator model.The outer surface of the radiator in contact with the air is set as natural convection, the convection coefficient is 7.5W / (M2 Â· K), and the ambient temperature is set as 40 â, so as to ensure that the working temperature of LED street lamps is below 75 â. Due to the sealing effect of the lampshade, other surfaces of the model are defined as thermal insulation. The volume of the light source is 60 mm & tides; 60mm &TImes; 8mmã The power of LED street lamp is 50W, of which 15% is converted into light energy and 85% into heat energy. Therefore, the heat generation rate load of (1.47 & times; 106) W M - 3 is applied to the chip entity. The radiator is made of ZL104 aluminum alloy, with thermal conductivity of 147w / M and density of 2650 kg / m3. Under the condition of conventional pressure and surface roughness, the contact thermal resistance between aluminum and aluminum is 4.55 & ties; 10-4m2Â· K /W ã1.3 optimization designOrthogonal experimental design method has the advantages of less experimental times, uniform distribution of data points, and the corresponding range analysis method can be used to analyze the test results.In this paper, in order to reduce the operation scale of simulation and analyze the influence of the structural size change of radiator on its temperature field, orthogonal experiments are designed to carry out multiple thermal analysis of the parametric model. The six radiator structural parameters affecting the final temperature field distribution are taken as factors, and each factor takes 5 levels (see Table 1). Taking the radiator quality and the maximum chip temperature as the test indexes, the orthogonal table L25 (56) is selected.Comprehensively considering the size of LED wick, the design structure of the whole lamp body and the requirements for the quality and volume of radiator, the number of fins a is (5 - 17), the fin height B is (20 - 60) mm, the fin thickness C (1 - 3.8) mm, the substrate thickness d (1 - 3) mm, and the substrate length E and width f are (150 - 250) mm. The specific values of five levels are shown in Table 1 below.Table 1 Parameters of orthogonal test1.4 analysis of test resultsThe experimental results and analysis are shown in Table 2.Table 2 test result data.It can be seen from table 2 that the number of fins has the greatest impact on the chip junction temperature, followed by fin height, followed by substrate length, substrate thickness, fin thickness and substrate width. A > b > e > d > C > F.Fin thickness has the greatest impact on the quality of radiator, followed by fin height, followed by the number of fins, substrate length, substrate width and substrate thickness. That is, C > b > a > e > F > D.According to the analysis results, the influence diagram of different levels of various factors on the temperature target is drawn, as shown in Figure 4.According to the quality formula, when other parameters remain unchanged, the parameter value is directly proportional to the quality result. The greater the value, the greater the quality, so the curve is no longer drawn.Fig. 4 Effects of six factors at different levels on the maximum temperature of the chipFrom the range analysis results, we can know that different factors have different effects on the two goals, and the same factor has different effects on the two goals. Therefore, the selection of different factor values should be based on the principle of keeping the maximum temperature of the chip to the minimum as the main goal and the minimum quality of the radiator as the secondary goal. For example, the influence of fin thickness on the maximum temperature of the chip ranks sixth, but it has the greatest influence on the quality. Therefore, a smaller fin thickness can be selected to reduce the quality without increasing the temperature as much as possible.Among the 25 experiments, it can be known that the effect is the best at the 25th time, that is, a5b 5C 4d3e 2F1. At this time, the temperature is 59. 61 â, and the mass of the radiator is 1. 61 kg. The results are shown in Figure 5. The optimized result is a5b 5c1d 5e5f 1. It is verified that the temperature can be reduced to 58.09 â and the mass of radiator can be reduced to 0.98 kg. The results are shown in Figure 6.It can be seen that the purpose of double objective optimization design is achieved through orthogonal analysis.Fig. 5 steady state temperature field of a 5B 5c4d 3E 2F 1 heat dissipation structure.Fig. 6 steady state temperature field of a 5B 5c1d 5E 5F 1 heat dissipation structure.2 conclusion and ProspectIn this paper, the integrated high-power light source LED street lamp radiator is studied by combining orthogonal test method and simulation experiment. With a few simulation experiments, the test data that can basically reflect the overall situation are obtained, and the influence of different parameters on LED heat dissipation and quality is studied, so as to obtain a set of optimized parameter combinations. This optimization method is also applicable to other fin forms, and is of great significance to the popularization and application of high-power centralized heat source LED lamps.