JP2010098204A - Cooling structure of light source - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a cooling structure of a light source, which can improve a heat transfer rate and can also improve the flexibility and versatility in a passage design compatibly in a cooling jacket. <P>SOLUTION: In the cooling structure of the light source 2 for making a metal heat slinger 3 adhere to the back surface of an insulating ceramic substrate 1 on the surface of which the light source 2 used as a heating element is mounted, for forming the cooling jacket 5 between the heat slinger 3 and a refrigerant storage container 4 to be joined to the heat slinger 3, and for cooling the light source 2 by pouring the refrigerant to the cooling jacket 5, a refrigerant passage is formed in the rugged shape on the back surface facing the inside of the cooling jacket 5 of the heat slinger 3, and a rectangular parallelepiped (or cylindrical) fins 6 are intensively erected right near the back, on which the light source 2 of the heat slinger 3 is mounted, as the rugged shape. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a cooling structure that cools a light source such as an LED with a coolant such as cooling water flowing through a cooling jacket.

  In recent years, the power consumption of semiconductors has increased at an accelerating rate, and the amount of heat generated by semiconductors has increased accordingly, and the demand for higher performance cooling systems has increased. The electric power consumed by the high-density integrated circuit is increasing, and the heat generation density is rapidly increasing simultaneously with the heat generation amount. Therefore, in the conventional forced absorption by the Peltier element, the forced air cooling structure by the heat sink and the fan, etc., the cooling capacity starts to be limited, and a liquid cooling system having a higher cooling capacity has been adopted.

  Also in the LED, which is an optical semiconductor, along with the improvement in brightness, the way of increasing the power from the conventional display application to the illumination application is being followed. The biggest problem with LED elements is that even if the luminous efficiency is improved, most of the input power becomes heat and the brightness is lowered by the heat generated by itself.

  In particular, in a high-power LED, a package / heat dissipation structure that can receive heat of several watts per chip is required, and metal core substrates and ceramic substrates excellent in thermal conductivity have been put into practical use. In particular, ceramic substrates are attracting attention for two main reasons: improved thermal conductivity due to advances in material technology, and the fact that an insulating layer is not required unlike metal core substrates because it is a highly insulating substrate. Yes. A liquid cooling system having a high cooling performance for such a high power LED is considered to be an effective heat dissipation means.

  For example, in Patent Document 1, when a semiconductor component having a high heat generation density is cooled by a liquid cooling system, in order to keep the thermal resistance from a heat generation source to external air as low as possible, a substrate for mounting an electric circuit and a heat dissipation There has been proposed a cooling structure that integrates a ceramic substrate and a tube wall through which a refrigerant passes.

  Further, Patent Document 2 proposes a cooling structure in which a liquid cooling jacket is formed by providing a box so as to cover the outside of a conventional heat sink and flowing a refrigerant between fins as a simpler implementation method. Yes.

  Further, in Patent Document 3, attention is paid to an aluminum nitride substrate having a high thermal conductivity, and a circuit pattern and a heat sink are bonded to each other by brazing to provide an integrated double-sided copper-clad substrate.

  On the other hand, since the conventional heat exchanger technology can be applied to the liquid cooling jacket, many proposals have been made regarding the flow path structure of the cooling liquid. For example, Patent Document 4 proposes a technique for improving the heat transfer coefficient by causing a three-dimensional turbulent flow using a rib structure, and Patent Document 5 uses a block structure arranged radially. A technique for achieving a uniform cooling effect by controlling the refrigerant to flow uniformly has been proposed.

  Further, Patent Document 6 proposes a technique for reducing the thickness while maintaining a necessary strength by using a concavo-convex structure for forming a flow path as a reinforcing portion of the wall surface, and Patent Document 7 discloses a technique for reducing the upper and lower sides of the heating element. Techniques have been proposed for improving the cooling performance by flowing the refrigerant twice.

Further, Patent Document 8 proposes a technique for forming a flow path by laser processing on a highly insulating ceramic substrate itself.
JP 2002-329938 A JP 2003-086744 A JP 2005-011922 A JP 2002-250572 A JP 2003-051689 A JP 2007-010277 A JP 2008-041806 A JP 2008-140877 A

  In order to improve the equivalent heat transfer coefficient from solid to fluid (= amount of heat to move / envelope cross-sectional area), it is important to optimize the flow path structure of the refrigerant inside the liquid cooling jacket. The flow path proposed in is a simple rectangular parallelepiped or a structure in which they are connected, and the technique described in Patent Document 2 merely uses the shape of an existing heat sink as it is.

  In addition, the technique proposed in Patent Document 3 is intended to improve the overall thermal conductivity by devising the shape of the heat sink and bonding the heat sink to an aluminum nitride substrate having a high thermal conductivity. However, aluminum nitride and copper have poor consistency in thermal expansion coefficient, and there is a possibility that peeling and cracking may occur due to stress coupled with the fluidity of wax.

  Further, in the configurations proposed in Patent Documents 4 to 7, although the effects of improving the thermal conductivity and improving the strength of the jacket can be obtained by devising the flow path structure, each of them is designed for exclusive use. There is a problem that flexibility is low and versatility is poor. In addition, if a complicated flow channel structure is pursued in order to improve the heat transfer coefficient, the pressure loss increases, and it is not possible to secure the required flow rate. Become.

  In addition, with the technique proposed in Patent Document 8, the cost of laser processing of a ceramic substrate that is harder than that of metal is high, and it is not easy to construct a complicated flow path pattern, and a striped groove structure is practical. There is a place.

  As described above, there has conventionally been a problem that it is difficult to improve both the heat transfer coefficient of the liquid cooling jacket and the flexibility and versatility of the flow path design.

  In order to obtain a highly versatile liquid cooling jacket with excellent thermal conductivity and flexibility that can meet the required specifications while suppressing manufacturing costs, the arrangement of the heat source and the arrangement of the refrigerant inlet and outlet in the jacket, It is important to be able to freely and inexpensively design a flow path structure with a low pressure loss in accordance with the capacity of the circulating device to be used.

  The present invention has been made in view of the above circumstances, and the object of the process is a light source capable of achieving both improvement in heat transfer coefficient in a liquid cooling jacket and improvement in flexibility and versatility of flow path design. It is to provide a cooling structure.

  In order to achieve the above object, the invention according to claim 1 is characterized in that a metal heat sink is bonded to the back surface of an insulating ceramic substrate on which a light source serving as a heating element is mounted, and is bonded to the heat sink. In the cooling structure of the light source that cools the light source by forming a cooling jacket with the refrigerant storage container, and flowing the refrigerant through the cooling jacket, the refrigerant is formed in an uneven shape on the back surface of the heat radiating plate facing the cooling jacket. A flow path is formed, and a rectangular parallelepiped or a cylindrical fin is erected in the vicinity of the back of the heat sink where the light source is mounted as the uneven shape.

  According to a second aspect of the present invention, in the first aspect of the invention, the insulating ceramic substrate is mounted on a surface including a layer containing a fine ceramic material having a thermal conductivity of 50 [W / m · K] or more. It has a multilayer structure including an electric circuit layer of the light source.

  According to a third aspect of the present invention, in the first or second aspect of the present invention, the heat radiating plate has a thermal conductivity of 150 [W / m · K] or more and a heat comparable to that of the insulating ceramic substrate. It is composed of a single metal or an alloy having an expansion coefficient.

  The invention according to claim 4 is the invention according to any one of claims 1 to 3, wherein the light source is an LED.

  According to the present invention, in the cooling jacket, the part forming the refrigerant flow path can be made independent of other components (refrigerant storage container, lid, refrigerant inlet / outlet, etc.). Therefore, in the case where the specifications of the light source that is a heating element and the refrigerant circulation device are different, it is only necessary to change the design of the independent refrigerant flow path portion, and the remaining components can be used in common as they are. In addition, since the coolant flow path is formed in a concavo-convex shape on a metal heat sink that is easy to process and has high thermal conductivity, there is no risk of peeling or cracking due to stress, reducing processing costs and design freedom. The improvement of both can be achieved.

  Further, in the refrigerant flow path structure according to the present invention, the ratio of the fin volume to the cooling jacket inner volume is less than 15%, and it is not a structure in which fins are extremely fine like a microchannel. Therefore, it is possible to cool the light source by circulating the refrigerant efficiently.

  Furthermore, in the vicinity of the light source that greatly contributes to the heat transfer performance, the area where heat exchange with the refrigerant can be performed can be increased up to about four times compared to a flat surface without unevenness, and heat can be efficiently generated. Can communicate.

  Note that, as disclosed in Patent Document 5, the structure is basically thermally advantageous because the refrigerant blows out from directly under the light source, but the refrigerant blows out under the light source due to external restrictions. Even when it is difficult to provide the opening, the cooling structure according to the present invention can freely control the flow rate of the refrigerant in the cooling jacket by the arrangement of the fins, so that the cooling performance can be improved.

  Embodiments of the present invention will be described below with reference to the accompanying drawings.

  The present invention makes use of the characteristics of a ceramic substrate having both high insulation and high thermal conductivity, and the flow path necessary for improving the heat transfer coefficient to the refrigerant by configuring the refrigerant flow path with a metal material that is easy to process. It is characterized by the freedom of road design.

  FIG. 1 is an exploded perspective view showing a basic configuration of a cooling structure according to the present invention. As shown in the drawing, a light source 2 such as an LED as a heating element is mounted at the center of the surface of a rectangular insulating ceramic substrate 1. Yes. A metal rectangular heat sink 3 is bonded to the back surface of the insulating ceramic substrate 1 opposite to the side where the light source 2 is provided, and a shallow rectangular box-shaped refrigerant storage container 4 is attached to the heat sink 3. A cooling jacket 5 is formed between the cooling containment vessel 4 and the heat radiating plate 3. Here, a plurality of rectangular parallelepiped fins 6 projecting radially from the center are provided on the rear surface of the heat radiating plate 3 (the surface facing the cooling jacket 5). The refrigerant storage container 4 is provided with a pipe-like refrigerant inlet 7 and a refrigerant outlet 8, and one end of each of the refrigerant inlet 7 and the refrigerant outlet 8 is opened to the center portion and the corner portion of the cooling jacket 5. ing.

  Thus, the refrigerant flowing into the cooling jacket 5 from the refrigerant inlet 7 flows radially outward from the center through the refrigerant flow path defined by the plurality of fins 6 projecting radially from the heat radiating plate 3. In the process, the heat emitted from the light source 2 is taken and the light source 2 is cooled, and then discharged from the refrigerant outlet 8 to the outside of the cooling jacket 5. That is, the heat emitted from the light source 2 mounted on the insulating ceramic substrate 1 diffuses inside the insulating ceramic substrate 1 by heat conduction. At this time, the insulating ceramic substrate 1 has a temperature gradient. In general, it is known that the distribution has a normal distribution with the light source 2 as the center. Accordingly, among the fins 6 erected on the back surface of the heat radiating plate 3, the one near the light source 2 affects the heat transfer performance, and the one far from the light source 2 is not enough for heat transfer. Does not contribute.

  The pressure loss of the refrigerant in the entire cooling jacket 5 is determined by the ratio of the fins 6 standing inside to the internal volume of the cooling jacket 5 and the fluid resistance due to the size of the fins 6. Therefore, as the number of fins 6 is increased or the size of the fins 6 is reduced, the contact surface area of the fins 6 with the refrigerant increases and the equivalent heat transfer coefficient is improved. However, the pressure loss increases in inverse proportion. End up.

  Therefore, in the present invention, the surface area in contact with the refrigerant and the flow velocity of the refrigerant are locally increased by concentrating the fins in the vicinity of the back of the light source mounted on the surface of the insulating ceramic substrate to increase the heat transfer performance. And reducing the pressure loss of the refrigerant in the entire cooling jacket by eliminating the fins on the outer periphery of the cooling jacket that do not contribute much to the cooling performance.

  By the way, in the present embodiment, the insulating ceramic substrate 1 is 50 [W / m · K] or more as typified by aluminum nitride, boron nitride, silicon nitride, aluminum oxide, silicon carbide / diamond-like carbon, etc. Preferably, it has a multilayer structure of two or more layers including a layer containing a fine ceramic material having a thermal conductivity of 150 [W / m · K] or more and an electric circuit layer for the heating element 2 mounted on the surface. ing. As for the electric circuit layer, one formed by vapor deposition or sputtering on a ceramic substrate, one obtained by bonding a thin plate by brazing, one having a thermal via for heat sinking on a thin film flexible substrate, etc. The thermal resistance satisfying the target specification can be freely selected within a range where it can be realized.

  Further, the heat radiating plate 3 having a plurality of fins 6 projecting from the back surface is made of a material having excellent thermal conductivity such as copper, aluminum or an alloy thereof (150 [W / m · K] or more). The material is preferably made of a material having thermal conductivity, and has the same thermal expansion coefficient as that of the insulating ceramic substrate 1 to be bonded, and it is an essential condition that no peeling or cracking occurs due to stress. The heat radiating plate 3 is molded in advance by a processing method such as press molding, extrusion, casting, forging, die casting, or spatula drawing, and is bonded to the insulating ceramic substrate 1 by brazing or a ceramic adhesive. Alternatively, the fins 6 are formed by additionally processing a metal plate previously bonded to the insulating ceramic substrate 1 by etching or vapor deposition.

  Next, examples of the present invention will be described. The channel structure is not limited to the one shown below.

<Flow channel model>
The fluid analysis was performed on the four examples and the two comparative examples for the channel model of the cooling jacket structure. The features of each example and comparative example will be described below.

<Common items>
Regarding the material of each layer structure, the insulating ceramic substrate is aluminum nitride having a thermal conductivity of 170 [W / m · K], the uneven heat sink has a 0.5 mm thick aluminum plate, the refrigerant storage container for storing the refrigerant, A3003 series aluminum was selected as the material of the refrigerant inlet and the refrigerant outlet. Also, press molding was adopted as a method for processing the heat sink made of aluminum, and brazing using a brazing containing an active metal was selected as a method for bonding the heat sink to the insulating ceramic substrate.

  The size of the insulating ceramic substrate was 50 mm × 50 mm, and the depth (inner dimension) of the refrigerant storage container was 3 mm. The height of the fins erected on the back surface of the heat radiating plate is also 3 mm, and the refrigerant reaches the ceiling of the cooling jacket so that the refrigerant is sufficiently filled in the refrigerant flow path between the fins. Since the fin width (corresponding to the diameter in the case of a cylindrical fin) is 1 mm, the ratio of fin height to width (= fin height / fin width) is 3. This is because if the ratio of the height and width of the fin exceeds 4 to 5, heat is not sufficiently transferred to the tip of the fin, and heat cannot be efficiently exchanged with the refrigerant throughout the fin.

  The electric circuit layers formed on the surface of the insulating ceramic substrate were the same, and the entire surface pattern without grooves was used for simplicity in analysis. As a heating element, an aggregate of a plurality of LED chips was assumed, and a ceramic heater with 10 W × 10 mm × 1.5 mm, 20 W output was used. The heating elements were arranged in the center for Examples 1 to 3 and Comparative Example 1 and in four for Example 4 and Comparative Example 2 (5 mm × 5 mm), and four heating elements were arranged at equal intervals in the vertical and horizontal directions.

  Pure water was used as the refrigerant, and the volume flow rate was set to 0.4 liter / min as a typical value of a small pump used in a water-cooled personal computer or the like.

  In this embodiment, as shown in FIG. 2, the refrigerant inlet 7 is provided at the center, the refrigerant outlet 8 is provided at the end portion, and the rectangular parallelepiped fins 6 are erected in the cooling jacket 5 so as to spread radially from the center. Yes. In such a cooling structure, the refrigerant flows into the cooling jacket 5 from the central refrigerant inlet 7, flows radially outward through the radial refrigerant flow path defined by the fins 6, and is cooled from the refrigerant outlet 8. It is discharged out of the jacket 5.

  The structure of the fin 6 is roughly divided into three. That is, first, a cross-shaped fin 6a that divides the entire cooling jacket 5 into four, and secondly, a total of four fins 6b installed in a 45 ° direction with respect to the cross-shaped fin 6a, one for each region. Third, it is composed of a total of eight fins 6c installed in a direction of 22.5 ° with respect to the two fins 6a and 6b, two in each region. The first cross-shaped fin 6 a does not completely divide the entire cooling jacket 5, and a gap is provided in the periphery of the cooling jacket 5. Thereby, the refrigerant flowing in from the center can reach the refrigerant outlet 8 through the gap in the peripheral portion.

  The first cross-shaped fin 6 a is installed so as to straddle the refrigerant inlet 7 and acts to reduce the cross-sectional area of the refrigerant inlet 7. Therefore, the flow rate of the refrigerant can be increased at the refrigerant inlet 7, which can contribute to the improvement of the equivalent heat transfer coefficient. Further, the distance between the fins 6 (6a to 6c) increases from the central part to the peripheral part of the cooling jacket 5, and the standing density of the fins 6 (6a to 6c) decreases as the peripheral part increases. Loss is reduced.

  In this embodiment, as shown in FIG. 3, the refrigerant inlet 7 is provided at the center and the refrigerant outlet 8 is provided at the end, and the cylindrical fins 9 are provided in the cooling jacket 5 so as to spread radially from the center. Standing up. In such a cooling structure, the refrigerant flows into the cooling jacket 5 from the central refrigerant inlet 7, passes through the refrigerant flow path formed between the columnar fins 9, and passes from the refrigerant outlet 8 at the end to outside the cooling jacket 5. Is discharged.

  Each of the spiral structures formed by the cylindrical fins 9 is composed of a total of six trajectories that are point-symmetric about a heating element (not shown). The reason why the cylindrical fins 9 have a spiral structure instead of a linear radial shape is that the equivalent heat transfer coefficient is improved by increasing the length of the substantial refrigerant flow path.

  As the cooling jacket 5 moves from the central part to the peripheral part, the distance between the spiral trajectories of the fins 9 increases, and the standing density of the fins 9 increases toward the central part, contributing to the improvement of the equivalent heat transfer coefficient. The higher the density, the lower the density of the fins 9 and the smaller the pressure loss of the refrigerant.

  In this embodiment, as shown in FIG. 4, the refrigerant inlet 7 is provided at the center, the refrigerant outlet 8 is provided at the end, and the cylindrical fins 9 are provided in the cooling jacket 5 so as to spread radially from the center. Standing up. In such a cooling structure, the refrigerant flows into the cooling jacket 5 from the central refrigerant inlet 7, passes through the refrigerant flow path between the fins 9, and is discharged out of the cooling jacket 5 from the refrigerant outlet 8 at the end. .

  The linear radial shapes of the fins 9 are arranged concentrically around a heating element (not shown), and the number of fins 9 arranged in a single circle is a multiple of six. That is, this radiation shape can be divided into six points symmetrically, which is a unit structure. The external shape of the standing shape is a circle, and a line in which the fins 9 are aligned along the diameter direction is a dividing boundary line. Although the refrigerant flowing in from the central part in the cooling jacket 5 is blocked by the fins 9 and dispersed in the flow, most of the refrigerant has a regular shape based on a hexagon along the boundary of the six divisions. Continue to draw and reach the outer periphery. Therefore, the flow of the refrigerant can be controlled by providing regularity as compared with the structure in which the fins 9 are arranged in a radial manner, and the pressure loss of the refrigerant can be reduced.

<Comparative Example 1>
This comparative example corresponds to the reference experiment of Examples 1 to 3, and as shown in FIG. 5, the refrigerant inlet 7 is provided at the center, the refrigerant outlet 8 is provided at the end, and the fins are provided in the cooling jacket 5. It is set as the structure which does not provide. In such a cooling structure, the refrigerant flows into the cooling jacket 5 from the central refrigerant inlet 7, passes through the cooling jacket 5, and is discharged out of the cooling jacket 5 from the refrigerant outlet 8 at the end.

  In this comparative example, since the fins are not erected, it is expected that the equivalent heat transfer coefficient will remain low although the pressure loss of the refrigerant is the lowest as compared with Examples 1-3.

  In this embodiment, as shown in FIG. 6, the heating element 2 is divided into four parts (5 W each), and the insulating ceramic substrate 1 of the insulating ceramic substrate 1 is equidistant from the end face of the insulating ceramic substrate 1 at a position equal to 1/4. Thirty-seven columnar fins 9 are arranged in a lattice pattern on the back surface and concentrated on the position just behind the heating element 2 on the back surface. In such a cooling structure, the refrigerant flows into the cooling jacket 5 from the refrigerant inlet 7 disposed at one corner portion on the diagonal line, passes through the refrigerant flow path, and is disposed at the other corner portion on the diagonal line. 8 is discharged out of the cooling jacket 5.

  In the present cooling structure, the refrigerant flow is dispersed only in the portion where the fins 9 are erected, and heat exchange is performed on the side surfaces of the fins 9. On the other hand, the pressure loss of the refrigerant is zero in the portion where the fins 9 are not erected. In this way, by clearly separating the heat exchange part and the other parts, the variation in the flow rate of the refrigerant at the back surface position of each heating element 2 is suppressed, and the equivalent heat transfer coefficient is made uniform, and at the same time, the cooling jacket 5 as a whole. The aim is to reduce the pressure loss of the refrigerant.

<Comparative example 2>
This corresponds to the reference experiment of Example 4, and is an example in which columnar fins 9 are uniformly arranged on the back surface without changing the surface of the insulating ceramic substrate 1 as shown in FIG. The refrigerant enters the cooling jacket 5 from the refrigerant inlet 7, passes through the refrigerant flow path between the fins 9, and is discharged from the refrigerant outlet 8.

  In this comparative example, since the fins 9 are erected over the entire cooling jacket 5, the equivalent heat transfer coefficient is improved as compared with the fourth embodiment, but the pressure loss of the refrigerant is also expected to increase. It is.

[Fluid analysis results and discussion]
FIG. 8 shows the temperature distribution on the ceramic substrate surface in Examples 1 to 3 and Comparative Example 1. Although there is an inclination along the flow of the refrigerant, it can be seen that heat is diffused substantially concentrically around the heating element. In this embodiment, since the size of the heating element is somewhat larger than the ceramic substrate, the shape is closer to a quadrangle than the perfect circle, but when the size of the heating element is small, the shape becomes more perfect circle. The shape is close and the temperature gradient is expected to be close to that of the normal distribution.

Below, the result verified about each Example and a comparative example is demonstrated.
(1) When there is one heating element in the center:
Table 1 shows a list of results.

実施例１〜３までの全体を見ると、比較例１と比べて発熱体の温度が低減されており、等価熱伝達率が向上しているのが分かる。圧力損失については比較例１を基準として、フィンを追加したことによる影響が少ない方が望ましい。従って、実施例２が本解析においては最良の形状であったと言える。又、図９〜図１１に実施例１〜３における流速ベクトルの分布を示す。流路に沿って流速ベクトルが分布していることと、実施例１〜３の何れにおいても、中央部の流速ベクトルの大きさ（長さではなく色）が同程度であることが分かる。

When the whole to Examples 1-3 is seen, it turns out that the temperature of a heat generating body is reduced compared with the comparative example 1, and the equivalent heat transfer rate is improving. With respect to the pressure loss, it is desirable that the influence of adding the fin is small with Comparative Example 1 as a reference. Therefore, it can be said that Example 2 was the best shape in this analysis. 9 to 11 show the distribution of flow velocity vectors in the first to third embodiments. It can be seen that the flow velocity vector is distributed along the flow path and that the flow velocity vector in the central portion has the same size (not length but color) in any of the first to third embodiments.

<Comparison of Example 1 and Example 2>
Since the equivalent heat transfer coefficient is the same, it can be considered that the surface area related to heat transfer in the vicinity of the heating element is approximately the same in Example 1 and Example 2. On the other hand, when attention is paid to the fin volume ratio and the pressure loss, it can be seen that both of the first embodiment is about twice as large as the second embodiment and are in a substantially proportional relationship. Therefore, it was verified that the pressure loss is less dependent on the fin shape, and that the fin volume ratio and the arrangement requirement have a greater influence on the pressure loss.

<Comparison between Example 2 and Example 3>
Since the equivalent heat transfer coefficient is the same as in the previous section, it can be considered that the surface area related to heat transfer in the vicinity of the heating element is the same in Example 2 and Example 3.

In Example 2, the reduction in pressure loss is dealt with by lowering the standing density by thinning out the fins, whereas in Example 3, the arrangement density is maintained while maintaining the standing density as it is. This is achieved by controlling the refrigerant flow path and lowering the substantial fluid resistance. In this study, it was verified that the method of lowering the standing density is superior.
<Comparison between Example 1 and Example 3>

  The magnitude relationship between the fin volume ratio and the pressure loss is reversed from the comparison between the first embodiment and the second embodiment. As described in the previous section, in Example 3, the flow density is controlled by giving the arrangement structure regularity while maintaining the standing density, and the pressure loss is reduced by lowering the substantial fluid resistance. Because of that. Accordingly, it was verified that the latter contributed to a substantial reduction in pressure loss in terms of fin volume ratio and arrangement requirements.

As described above, the second embodiment has a low heat transfer performance (effective heat transfer surface area near the center) and reduced pressure loss (temperature distribution of the substrate caused by heat generation, etc. It was verified that it was the best shape at two points (excluding peripheral fins that did not contribute).
(2) When four heating elements are dispersed:
Here, it verifies about the case where arrangement | positioning of a heat generating body changes. In Example 4 and Comparative Example 2, the list of results is shown in Table 2, and the distribution of temperature and flow velocity vector is shown in FIGS.

  When Example 4 and Comparative Example 2 are compared, the difference between the maximum and minimum temperatures of the heating elements is 0.3 ° C., but the temperature difference between each heating element (maximum temperature-minimum temperature) is small. I understand that The operating temperature of the heating element is an important factor that affects the characteristics and life of the element, and the fact that the temperature difference can be reduced means that there is an effect of suppressing the bias in characteristics and life.

  With regard to the refrigerant flow path structure, it can be said that effective cooling can be achieved by creating fins in the vicinity of the heating element to create density of the flow velocity and concentrating the flow velocity vector in the vicinity of the heating element. At the same time, it can be said that pressure loss can be greatly reduced by removing fins that are far from the heating element and do not contribute much to heat transfer.

It is a disassembled perspective view which shows the basic composition of the cooling structure which concerns on this invention. It is a perspective view which shows the cooling structure which concerns on Example 1 of this invention. It is a perspective view which shows the cooling structure which concerns on Example 2 of this invention. It is a perspective view which shows the cooling structure which concerns on Example 3 of this invention. It is a perspective view which shows the cooling structure which concerns on the comparative example 1 of this invention. It is a perspective view which shows the cooling structure which concerns on Example 4 of this invention. It is a perspective view which shows the cooling structure which concerns on Example 2 of this invention. It is a figure which shows the temperature distribution of the ceramic substrate surface in Examples 1-3 and Comparative Example 1 of this invention. It is a figure which shows the flow velocity vector distribution in Example 1 of this invention. It is a figure which shows the flow-velocity vector distribution in Example 2 of this invention. It is a figure which shows the flow-velocity vector distribution in Example 3 of this invention. (A) is a temperature distribution in Example 4 of this invention, (b) is a figure which shows flow-velocity vector distribution. (A) is a temperature distribution in the comparative example 2 of this invention, (b) is a figure which shows flow velocity vector distribution.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Insulating ceramic substrate 2 Light source 3 Heat sink 4 Refrigerant storage container 5 Cooling jacket 6 Rectangular fin 7 Refrigerant inlet 8 Refrigerant outlet 9 Columnar fin

Claims (4)

A metal heat sink is bonded to the back surface of the insulating ceramic substrate on which a light source serving as a heating element is mounted on the surface, and a cooling jacket is formed between the heat sink and the refrigerant storage container joined thereto, In the cooling structure of the light source that cools the light source by flowing a coolant through the cooling jacket,
A refrigerant flow path is formed in a concavo-convex shape on the back surface of the heat radiating plate facing the cooling jacket, and a rectangular parallelepiped or a cylindrical fin as the concavo-convex shape is concentrated near the back of the heat radiating plate where the light source is mounted. A cooling structure of a light source characterized by that.   The insulating ceramic substrate has a multilayer structure including a layer containing a fine ceramic material having a thermal conductivity of 50 [W / m · K] or more and an electric circuit layer of the light source mounted on the surface. The light source cooling structure according to claim 1.   The heat radiating plate is made of a single metal or alloy having a thermal conductivity of 150 [W / m · K] or more and a thermal expansion coefficient comparable to that of the insulating ceramic substrate. Item 3. The light source cooling structure according to Item 1 or 2.   The light source cooling structure according to claim 1, wherein the light source is an LED.

Images