Coolers and semiconductor devices

Abstract

The present invention suppresses the occurrence of a deviated flow distribution in a cooler and the increase in pressure loss.　A cooler (10) comprises, inside a container (14): a first flow path (14e) disposed parallel with a first side wall (14a) and communicating with an introduction port (11); a second flow path (14f) disposed parallel with a second side wall (14b) and communicating with a discharge port (12); a third flow path (14g) communicating with the first flow path (14e) and the second flow path (14f); a first flow rate adjustment unit (15) disposed between the first flow path (14e) and the third flow path (14g); and a second flow rate adjustment unit (16) disposed between the second flow path (14f) and the third flow path (14g). The first flow rate adjustment unit (15) includes a first region (15a) having a first opening ratio and a second region (15b) having a second opening ratio smaller than the first opening ratio. The second flow rate adjustment unit (16) includes a third region (16a) having a third opening ratio and a fourth region (16b) having a fourth opening ratio greater than the third opening ratio.

Description

Technical Field

[0001] The present invention relates to a cooler and a semiconductor device.

Background Art

[0002] There is known a cooler integrated with a housing of a power converter, the cooler connecting a refrigerant passage and a recess whose opening is sealed by a heat generating body at a connecting portion, and a cooler in which the opening area and shape of the connecting portion are changed corresponding to the distance from the inlet of the refrigerant passage (Patent Document 1).

[0003] Also, a plurality of plate-shaped fins that form a flow path of cooling water are provided below an upper plate on which semiconductor chips are arranged, and a connecting bar having a plurality of comb teeth portions that project into the flow path is connected to the plurality of plate-shaped fins, and a plurality of openings are defined by the plurality of comb teeth portions and the plurality of plate-shaped fins according to the size based on the position of the semiconductor chips and the like. There is known a cooler (Patent Document 2).

[0004] Also, there is known a semiconductor device including a fin portion including a plurality of protrusions connected to the lower surface of a heat conductive base plate, and a cooling component connected to an inlet and an outlet of refrigerant and covering the fin portion, and a header and a water flow control plate which are water reservoir chambers are provided so that refrigerant can flow between the inlet and outlet and the fin portion (Patent Document 3).

[0005] Also, there is known a semiconductor cooler including a cooling plate on one surface of which a plurality of semiconductor modules having different heat generation amounts are arranged, and a plurality of radiation fins are erected on the other surface, and a housing portion arranged to face the cooling plate, and the channel height of a refrigerant flow path formed between the gaps of adjacent radiation fins and the wall portion of the cooling plate and the housing portion is made different depending on the facing region with the semiconductor modules having different heat generation amounts (Patent Document 4).

[0006] Furthermore, a liquid-cooled cooler is known in which a cooling container with a heat sink having heat dissipation fins as one side wall is divided into two regions by a first partition wall, a heat dissipation region in which the heat dissipation fins are exposed is formed in one region, and an inlet header region and an outlet header region are formed in the other region separated by a second partition wall, an inlet-side connecting passage and an outlet-side connecting passage are provided in the first partition wall, the heat dissipation region and the inlet header region are connected by the inlet-side connecting passage, and the heat dissipation region and the outlet header region are connected by the outlet-side connecting passage, thereby forming a cooling liquid passage (Patent Document 5).

[0007] Furthermore, a semiconductor device is known in which a base plate on which a semiconductor element is mounted on the upper surface is fitted with a plurality of cooling fins and a jacket surrounding them on the lower surface, a partition wall is provided below the plurality of cooling fins in the jacket, allowing refrigerant from the refrigerant inlet of the jacket to flow to the plurality of cooling fins and out to the refrigerant outlet of the jacket, and an inlet opening is provided at a position on the partition wall corresponding to the semiconductor element, allowing refrigerant to flow from the refrigerant inlet to the plurality of cooling fins (Patent Document 6).

[0008] Furthermore, there is an electrical device known in which a cooling jacket has multiple upstream connecting passages in the connection area between the main flow path that guides the cooling medium and the introduction passage upstream of it, and multiple downstream connecting passages in the connection area between the main flow path and the discharge passage downstream of it, and an electrical element is installed on the ceiling wall of the main flow path of the cooling jacket (Patent Document 7).

[0009] Furthermore, a cooler for semiconductor modules is known, comprising a tray-shaped cooling jacket provided with refrigerant introduction channels and refrigerant discharge channels extending parallel to each other and a cooling channel between them; a heat sink in which the flow channels are perpendicular to the refrigerant introduction channels and refrigerant discharge channels and a flow velocity adjustment plate fixed to one side extends to the boundary position with the refrigerant discharge channel; and a heat dissipation plate on which semiconductor elements are bonded to the outer surface and which closes the opening of the cooling jacket (Patent Document 8).

[0010] Furthermore, a cooler for semiconductor modules is known, comprising a water jacket having a first flow path extending from a refrigerant inlet, a second flow path arranged parallel to and spaced apart from the first flow path and extending toward a refrigerant outlet, and a third flow path connecting the first and second flow paths, and a heat sink disposed within the third flow path, wherein a flow velocity adjustment plate is provided within the second flow path of the water jacket, spaced apart from and parallel to the side surface of the heat sink (Patent Document 9). [Prior art documents] [Patent Documents]

[0011] [Patent Document 1] Japanese Patent Publication No. 2012-146759 [Patent Document 2] Japanese Patent Publication No. 2019-71330 [Patent Document 3] International Publication No. 2017 / 090106 Brochure [Patent Document 4] Japanese Patent Publication No. 2012-69892 [Patent Document 5] Japanese Patent Publication No. 2015-153799 [Patent Document 6] International Publication No. 2019 / 211889 Brochure [Patent Document 7] Japanese Patent Publication No. 2006-179771 [Patent Document 8] International Publication No. 2015 / 079643 brochure [Patent Document 9] International Publication No. 2013 / 054615 brochure [Overview of the Initiative] [Problems that the invention aims to solve]

[0012] One known technique for cooling semiconductor modules that generate heat during operation is the use of a liquid-cooled cooler. For example, a predetermined coolant such as water is circulated inside the cooler's container (also called a water jacket, etc.), and heat exchange takes place between the coolant and the semiconductor module mounted on the outside of the cooler, thereby cooling the semiconductor module.

[0013] However, in such coolers, depending on the internal configuration of the container, for example, the arrangement and shape of the flow paths on the refrigerant introduction and discharge sides, as well as the flow paths connecting them, a flow uneven distribution may occur within the cooler. This flow uneven distribution within the cooler can lead to uneven cooling efficiency in different parts of the semiconductor module, and the resulting overheating may cause a decrease in the performance or failure of the semiconductor module.

[0014] To improve such uneven flow distribution, a technique is known in which openings or plates are provided at predetermined locations within the cooler's flow path to adjust the refrigerant flow velocity. However, when employing such a technique, depending on the configuration provided to adjust the refrigerant flow velocity, it may lead to an increase in the pressure loss of the refrigerant introduced into and discharged from the cooler, potentially increasing the load on the pump that circulates the refrigerant within the cooler.

[0015] In one aspect, the present invention aims to realize a cooler that can suppress the occurrence of flow deviation and the increase in pressure loss. Furthermore, in one aspect, the present invention aims to realize a semiconductor device equipped with a cooler that can suppress the occurrence of flow deviation and the increase in pressure loss. [Means for solving the problem]

[0016] In one embodiment, the vessel has opposing first and second side walls and is equipped with a refrigerant inlet and outlet; a first flow path arranged parallel to the first side wall and communicating with the inlet within the vessel; a second flow path arranged parallel to the second side wall and communicating with the outlet within the vessel; a third flow path arranged within the vessel and communicating with the first and second flow paths; a first flow velocity adjustment section arranged between the first and third flow paths within the vessel; and a second flow velocity adjustment section arranged between the second and third flow paths within the vessel, wherein the first flow velocity adjustment section includes a first region having a first aperture ratio and a second region having a second aperture ratio smaller than the first aperture ratio; the second flow velocity adjustment section includes a third region having a third aperture ratio and a fourth region having a fourth aperture ratio larger than the third aperture ratio; and the direction of the opening of the first region toward the second flow velocity adjustment section From the end on the first flow velocity adjustment section side to the end on the second flow velocity adjustment section side The width of the opening of the second region is in the direction from the first flow velocity adjustment section to the second flow velocity adjustment section. From the end on the first flow velocity adjustment section side to the end on the second flow velocity adjustment section side Larger than the width, and in the direction from the first flow velocity adjustment section to the second flow velocity adjustment section of the opening in the third region. From the end on the first flow velocity adjustment section side to the end on the second flow velocity adjustment section side The width of the opening in the fourth region is in the direction from the first flow velocity adjustment section to the second flow velocity adjustment section. From the end on the first flow velocity adjustment section side to the end on the second flow velocity adjustment section side A cooler smaller than the width of is provided.

[0017] In one aspect, a semiconductor device is provided that includes a cooler and a semiconductor module mounted on the cooler. The cooler has a container having opposing first and second side walls and including an inlet and an outlet for a refrigerant, a first flow path disposed within the container parallel to the first side wall and communicating with the inlet, a second flow path disposed within the container parallel to the second side wall and communicating with the outlet, a third flow path disposed within the container and communicating with the first and second flow paths, a first flow rate adjustment unit disposed within the container between the first flow path and the third flow path, and a second flow rate adjustment unit disposed within the container between the second flow path and the third flow path. The first flow rate adjustment unit includes a first region having a first aperture ratio and a second region having a second aperture ratio smaller than the first aperture ratio. The second flow rate adjustment unit includes a third region having a third aperture ratio and a fourth region having a fourth aperture ratio larger than the third aperture ratio. The width of the opening of the first region in the direction from the first flow rate adjustment unit to the second flow rate adjustment unit From the end on the first flow velocity adjustment section side to the end on the second flow velocity adjustment section side is larger than the width of the opening of the second region in the direction from the first flow rate adjustment unit to the second flow rate adjustment unit From the end on the first flow velocity adjustment section side to the end on the second flow velocity adjustment section side and the width of the opening of the third region in the direction from the first flow rate adjustment unit to the second flow rate adjustment unit From the end on the first flow velocity adjustment section side to the end on the second flow velocity adjustment section side is smaller than the width of the opening of the fourth region in the direction from the first flow rate adjustment unit to the second flow rate adjustment unit From the end on the first flow velocity adjustment section side to the end on the second flow velocity adjustment section side and the semiconductor module is mounted at a position facing the third flow path of the cooler.

Advantages of the Invention

[0018] In one aspect, it becomes possible to realize a cooler that can suppress the occurrence of uneven flow distribution and an increase in pressure loss. Also, in one aspect, it becomes possible to realize a semiconductor device including a cooler that can suppress the occurrence of uneven flow distribution and an increase in pressure loss.

[0019] The above and other objects, features, and advantages of the present invention will become apparent from the following description in connection with the accompanying drawings that represent preferred embodiments of the present invention as examples.

Brief Description of the Drawings

[0020] [Figure 1] This figure illustrates an example of a semiconductor device and cooling system according to the first embodiment. [Figure 2] This figure illustrates an example of a semiconductor device according to the first embodiment. [Figure 3] This figure illustrates an example of the configuration of cooling fins provided on the heat sink plate of a cooler according to the first embodiment. [Figure 4] This figure illustrates an example of the configuration of the container of the cooler according to the first embodiment. [Figure 5] This figure illustrates an example of the configuration of the first flow rate adjustment unit and the second flow rate adjustment unit of the cooler according to the first embodiment. [Figure 6] This is a diagram (part 1) illustrating an example of the configuration of a cooler according to the first embodiment. [Figure 7] This is a diagram (part 2) illustrating an example of the configuration of a cooler according to the first embodiment. [Figure 8] This is a diagram (part 3) illustrating an example of the configuration of a cooler according to the first embodiment. [Figure 9] This is a diagram (part 1) illustrating an example of the configuration of a cooler related to a comparative example. [Figure 10] This is a diagram (part 2) illustrating an example of the configuration of a cooler related to a comparative example. [Figure 11] This is a diagram (part 3) illustrating an example of the configuration of a cooler related to a comparative example. [Figure 12] This figure shows an example of the evaluation results of the refrigerant flow rate with respect to the semiconductor element position. [Figure 13] This figure shows an example of the evaluation results for pressure loss in each type of cooler. [Figure 14] This figure shows an example of the evaluation results of semiconductor device temperature with respect to semiconductor device position. [Figure 15] This figure illustrates a first modified example of cooling fins provided on a heat sink plate of a cooler. [Figure 16]This figure illustrates a second modified example of cooling fins provided on the heat sink of a cooler. [Figure 17] This figure illustrates a third modified example of cooling fins provided on a heat sink plate of a cooler. [Figure 18] This figure illustrates a first modified example of the container of the cooler according to the second embodiment. [Figure 19] This figure illustrates a second modified example of the container of the cooler according to the second embodiment. [Figure 20] This figure illustrates a third modified example of the container of the cooler according to the second embodiment. [Figure 21] This figure illustrates a fourth modified example of the cooler container according to the second embodiment. [Figure 22] This figure illustrates a first modified example of the first flow rate adjustment unit and the second flow rate adjustment unit of the cooler according to the third embodiment. [Figure 23] This figure illustrates a second modified example of the first and second flow rate adjustment units of the cooler according to the third embodiment. [Figure 24] This figure illustrates a third modified example of the first and second flow rate adjustment units of the cooler according to the third embodiment. [Figure 25] This figure illustrates a first example of a cooler according to the fourth embodiment. [Figure 26] This figure shows the evaluation results of a thermal fluid simulation for the first example of a cooler that uses prismatic cooling fins. [Figure 27] This figure shows the evaluation results of a thermal fluid simulation for the first example of a cooler that uses cylindrical cooling fins. [Figure 28] This figure illustrates a second example of a cooler according to the fourth embodiment. [Figure 29] This figure shows the evaluation results of a thermal fluid simulation for a second example of a cooler that uses prismatic cooling fins. [Figure 30] This figure shows the evaluation results of a thermal fluid simulation for a second example of a cooler that uses cylindrical cooling fins. [Figure 31]This figure illustrates a third example of a cooler according to the fourth embodiment. [Figure 32] This figure shows the evaluation results of a thermal fluid simulation for a third example of a cooler that uses prismatic cooling fins. [Figure 33] This figure shows the evaluation results of a thermal fluid simulation for a third example of a cooler that uses cylindrical cooling fins. [Figure 34] This figure illustrates a fourth example of a cooler according to the fourth embodiment. [Figure 35] This figure shows the evaluation results of a thermal fluid simulation for a fourth example of a cooler that uses prismatic cooling fins. [Figure 36] This figure shows the evaluation results of a thermal fluid simulation for a fourth example of a cooler that uses cylindrical cooling fins. [Figure 37] This figure illustrates a fifth example of the cooler according to the fourth embodiment. [Figure 38] This figure shows the evaluation results of a thermal fluid simulation for the fifth example of a cooler that uses prismatic cooling fins. [Figure 39] This figure shows the evaluation results of a thermal fluid simulation for the fifth example of a cooler that uses cylindrical cooling fins. [Modes for carrying out the invention]

[0021] The embodiments will be described below with reference to the drawings. In the following description, "upward" refers to the direction facing upward as viewed from the page. "Upward" and "side" are merely convenient expressions to specify the relative positional relationship and do not limit the technical concept of the present invention. In the following description, "main component" refers to a case where it contains 80 vol% or more. In addition, "identical" means within a range of ±10%. In addition, "parallel" means within a range of ±10°.

[0022] [First Embodiment] Figure 1 illustrates an example of a semiconductor device and cooling system according to the first embodiment. Figure 1 schematically shows a perspective view of the main parts of an example of a semiconductor device according to the first embodiment, along with some elements of the cooling system. Figure 2 also illustrates an example of a semiconductor device according to the first embodiment. Figure 2 schematically shows a cross-sectional view of the main parts of an example of a semiconductor device according to the first embodiment. Figure 2 is a cross-sectional view taken along line II-II of Figure 1.

[0023] The semiconductor device 1 shown in Figures 1 and 2 comprises a cooler 10 and a semiconductor module 20 mounted on the cooler 10. As shown in Figure 1, the semiconductor module 20 has circuit element sections 21, 22, and 23 mounted on three different mounting areas AR1, AR2, and AR3 of the cooler 10, respectively. Each of the circuit element sections 21, 22, and 23 includes an insulating circuit board 24 and semiconductor elements 25 (also referred to as "CP1") and 26 (also referred to as "CP2") mounted on the insulating circuit board 24.

[0024] As shown in Figures 1 and 2, the insulating circuit board 24 comprises an insulating substrate 24a and conductive layers 24b and 24c provided on both sides thereof. The insulating substrate 24a can be made of alumina, alumina-based composite ceramics, aluminum nitride, silicon nitride, or the like. The conductive layers 24b and 24c can be made of metallic materials such as copper or aluminum. For example, a DCB (Direct Copper Bonding) substrate can be used for the insulating circuit board 24. Other substrates, such as an AMB (Active Metal Brazed) substrate, may also be used for the insulating circuit board 24.

[0025] For example, power semiconductor elements are used for semiconductor elements 25 and 26. Switch elements such as IGBTs (Insulated Gate Bipolar Transistors) and MOSFETs (Metal Oxide Semiconductor Field Effect Transistors) are used for semiconductor elements 25 and 26, respectively. Diode elements such as FWDs (Free Wheeling Diodes) and SBDs (Schottky Barrier Diodes) may be connected to or integrated with the switch elements used for semiconductor elements 25 and 26, respectively. As an example, reverse-conducting insulated-gate bipolar transistors, i.e., RC-IGBTs (Reverse Conducting - Insulated Gate Bipolar Transistors), are used for semiconductor elements 25 and 26.

[0026] As shown in Figures 1 and 2, semiconductor elements 25 and 26 are mounted on the conductive layer 24b side of the insulating circuit board 24, and are electrically connected to the conductive layer 24b via a bonding layer 27 such as solder, or via a wire (not shown). Although detailed illustrations are omitted here, the conductive layer 24b of the insulating circuit board 24 is provided on the insulating substrate 24a in a predetermined pattern shape so that a predetermined circuit function is realized together with the mounted semiconductor elements 25 and 26.

[0027] For example, semiconductor elements 25 and 26 are connected in series on the conductor layer 24b side of the insulating circuit board 24 and mounted on the conductor layer 24b side of the insulating circuit board 24 so as to function as an inverter circuit. For example, semiconductor element 25 is mounted to form the upper arm of the inverter circuit, and semiconductor element 26 is mounted to form the lower arm of the inverter circuit. The connection node between the series-connected semiconductor elements 25 and 26 is used as the output.

[0028] Three circuit element sections 21, 22, and 23, each having such a configuration, are connected in parallel on the conductor layer 24b side of the insulating circuit board 24. For example, the outputs of each of the circuit element sections 21, 22, and 23 correspond to the U-phase, V-phase, and W-phase outputs, respectively, and are connected to a three-phase AC motor. By performing switching control of the semiconductor elements 25 and 26 of each of the circuit element sections 21, 22, and 23, the DC current is converted to AC current, and the three-phase AC motor is driven.

[0029] The circuit element sections 21, 22, and 23 of the semiconductor module 20 are thermally connected to the cooler 10 via a bonding layer 28 on the conductor layer 24c side of the respective insulating circuit board 24, opposite to the conductor layer 24b side on which the semiconductor elements 25 and 26 are mounted.

[0030] The cooler 10, on which the semiconductor module 20 is mounted, comprises a heat sink 13 (also called a "fin base") provided with cooling fins 13a, and a container 14 (also called a "water jacket"). Circuit element sections 21, 22, and 23 of the semiconductor module 20 are thermally connected to the heat sink 13 of the cooler 10 via a bonding layer 28. The heat sink 13, provided with the cooling fins 13a, functions as a heat sink. The container 14 is connected to the heat sink 13 so as to cover the cooling fins 13a provided on the heat sink 13, and is fastened, for example, by bolts (not shown). The container 14 is connected to the heat sink 13 so as to house the cooling fins 13a of the heat sink 13 inside. The container 14 functions as a fin cover.

[0031] In the cooler 10 on which the semiconductor module 20 is mounted, a refrigerant 30 supplied from the outside flows through the internal space between the heat sink 13 and the container 14, that is, the gap between the heat sink 13 and its cooling fins 13a and the container 14. The refrigerant 30 can be water or LLC (Long Life Coolant). The cooler 10 is equipped with an inlet 11 and an outlet 12 for the refrigerant 30. The refrigerant 30 introduced from the inlet 11 flows through the refrigerant flow path (third flow path 14g) defined by the cooling fins 13a in the internal space between the heat sink 13 and the container 14 in the cooler 10, and is discharged from the outlet 12.

[0032] When the cooler 10 is in use, the inlet 11 is connected to the pump 40 by piping, and the outlet 12 is connected to the heat exchanger 50 by piping. The refrigerant 30 is introduced into the container 14 from the inlet 11 by the pump 40, circulates within the container 14, and is discharged from the outlet 12. The heat generated in the circuit element sections 21, 22, and 23 of the semiconductor module 20 is transferred to the heat sink 13 and its cooling fins 13a of the cooler 10, and heat is exchanged with the refrigerant 30 circulating within the container 14 that covers the cooling fins 13a. As a result, the circuit element sections 21, 22, and 23 are cooled. The refrigerant 30, whose temperature has risen due to the cooling of the circuit element sections 21, 22, and 23, is discharged from the outlet 12. The refrigerant 30 discharged from the outlet 12 is sent to the heat exchanger 50 for cooling. The refrigerant 30, cooled in the heat exchanger 50, is sent back to the inlet 11 by a pump 40 connected to the heat exchanger 50 by piping, and then introduced into the container 14 from the inlet 11.

[0033] In a cooling system including a semiconductor device 1 equipped with a cooler 10, a pump 40, and a heat exchanger 50, a refrigerant flow path is formed through which the refrigerant 30 flows within a closed loop including the cooler 10, pump 40, and heat exchanger 50. The refrigerant 30 is forcibly circulated within this closed loop by the pump 40. The semiconductor module 20 of the semiconductor device 1 is cooled by the forcibly circulated refrigerant 30.

[0034] Furthermore, the arrangement of the inlet 11 and outlet 12 of the cooler 10 is constrained by the routing of the piping connecting them to the pump 40 and heat exchanger 50, as well as the clearance between them and the semiconductor device 1 and the surrounding components of the cooling system including it, and therefore can be arranged in various ways. The arrangement of the inlet 11 and outlet 12 shown in Figure 1 is just one example of such various arrangements.

[0035] An example configuration of the semiconductor device 1 will be further explained with reference to Figures 3 to 8 below. First, the heat sink 13 of the cooler 10 and the cooling fins 13a provided thereon will be described with reference to Figure 3. Figure 3 illustrates an example of the configuration of cooling fins provided on the heat sink of a cooler according to the first embodiment. Figure 3(A) schematically shows a perspective view of the main part of an example of cooling fins provided on the heat sink of a cooler according to the first embodiment, and Figure 3(B) schematically shows a plan view of the main part of an example of cooling fins provided on the heat sink of a cooler according to the first embodiment. Figure 3(B) is an enlarged plan view of part Z0 of Figure 3(A).

[0036] The cooling fins 13a are provided on the heat sink 13 of the cooler 10 as pin fins, where multiple pin-shaped fins are arranged in a grid pattern, as shown in Figures 3(A) and 3(B). For example, the cooling fins 13a can be prism-shaped or roughly prism-shaped with chamfered corners. The cooling fins 13a have a rectangular or roughly rectangular planar shape (or cross-sectional shape) with a side length in the range of 1 mm to 3 mm, and a height from the mounting surface 13b of the heat sink 13 is in the range of 2 mm to 10 mm. For example, multiple cooling fins 13a are arranged in a grid pattern on the mounting surface 13b of the heat sink 13, with a side length of 3 mm and a spacing of 1.5 mm between adjacent cooling fins 13a. As an example, the cooling fins 13a shown in Figures 3(A) and 3(B) are provided on the heat sink 13 of the cooler 10 as shown in Figures 1 and 2. Note that the shape and dimensions of the cooling fins 13a shown in Figures 3(A) and 3(B) are examples only, and the optimal shape and dimensions will be selected according to the required cooling performance.

[0037] The cooling fins 13a are integrated with the heat sink 13. Metal materials such as aluminum, aluminum alloy, copper, and copper alloy are used for the heat sink 13 and the cooling fins 13a. The cooling fins 13a are manufactured in an integrated form with the heat sink 13 using die casting, brazing, or various welding techniques. Alternatively, the cooling fins 13a may be manufactured in an integrated form with the heat sink 13 using processing techniques that form convex-shaped cooling fins 13a from the material of the heat sink 13 by die casting, forging, or pressing, or by cutting or wire cutting.

[0038] Next, the container 14 of the cooler 10 will be explained with reference to Figure 4. Figure 4 illustrates an example of the configuration of a cooler container according to the first embodiment. Figure 4(A) schematically shows a perspective view of the main part of an example of a cooler container according to the first embodiment, and Figure 4(B) schematically shows a cross-sectional view of the main part of an example of a cooler container according to the first embodiment. Figure 4(B) is a cross-sectional view taken along line IV-IV of Figure 4(A).

[0039] The container 14 has an external shape that is a rectangular parallelepiped or substantially rectangular parallelepiped, as shown in Figures 4(A) and 4(B), for example. The container 14 has opposing first side walls 14a and second side walls 14b, and opposing third side walls 14c and fourth side walls 14d. The first side walls 14a, second side walls 14b, third side walls 14c and fourth side walls 14d are configured to be erected from the bottom plate 14h toward one side thereof. For example, of the opposing first side walls 14a and second side walls 14b, an inlet 11 is located on one of the first side walls 14a, and an outlet 12 is located on the other second side wall 14b.

[0040] A first flow channel 14e is arranged inside the container 14, parallel to the first side wall 14a and communicating with the inlet 11. The first flow channel 14e is a first groove that extends along the first side wall 14a at the bottom between the first side wall 14a and the second side wall 14b of the container 14.

[0041] A second flow path 14f is provided inside the container 14, which is arranged parallel to the second side wall 14b and communicates with the outlet 12. The second flow path 14f is a second groove that extends along the second side wall 14b at the bottom between the first side wall 14a and the second side wall 14b of the container 14. The second flow path 14f extends parallel to the first flow path 14e.

[0042] A third flow path 14g is further arranged inside the container 14, communicating with the first flow path 14e and the second flow path 14f. The third flow path 14g is the internal space of the container 14 above the first flow path 14e (first groove) and the second flow path 14f (second groove). As described later, a first flow velocity adjustment unit 15 is arranged at the boundary between the third flow path 14g and the first flow path 14e, and a second flow velocity adjustment unit 16 is arranged at the boundary between the third flow path 14g and the second flow path 14f. The cooling fins 13a of the heat sink 13, which is connected to cover the container 14, are housed and arranged in the third flow path 14g, which is the internal space above the first flow path 14e and the second flow path 14f (Figures 1 and 2).

[0043] The length w (which can also be said to be the length w of the first channel 14e and the second channel 14f) and width h0 of the internal space of the container 14 surrounded by the first side wall 14a, the second side wall 14b, the third side wall 14c, and the fourth side wall 14d, the width h and height t1 of the first channel 14e and the second channel 14f, and the height t2 of the third channel 14g are set appropriately based on the dimensions of the semiconductor module 20, the dimensions of the semiconductor device 1, the required cooling performance, etc.

[0044] The container 14 is made of a metal material such as aluminum, aluminum alloy, copper, or copper alloy. When such a metal material is used, the first channel 14e, second channel 14f, and third channel 14g are formed in the container 14, for example, by die casting. The inlet 11 and outlet 12 of the container 14 are formed, for example, by cutting. The container 14 is not limited to metal materials; other materials may be used as long as they have sufficient corrosion resistance and heat resistance to the refrigerant 30 circulating inside the container 14. For example, the container 14 may be made of a material containing carbon filler. Depending on the type and temperature of the refrigerant 30 circulating inside the container 14, ceramic materials or resin materials may also be used.

[0045] Next, the first flow velocity adjustment unit 15 and the second flow velocity adjustment unit 16, which are located in the container 14, will be described with reference to Figure 5. Figure 5 illustrates an example of the configuration of the first and second flow velocity adjustment units of the cooler according to the first embodiment. Figure 5 schematically shows a plan view of the main parts of an example of the first and second flow velocity adjustment units of the cooler according to the first embodiment.

[0046] As shown in Figures 4(A) and 4(B) above, a first flow velocity adjustment unit 15 and a second flow velocity adjustment unit 16 are arranged in the first flow path 14e and the second flow path 14f of the container 14, respectively, as shown in Figure 5.

[0047] The first flow velocity adjustment section 15 is formed, for example, from a plate-shaped member and is arranged parallel to and spaced apart from the bottom surface of the first flow path 14e (first groove). The first flow velocity adjustment section 15 is connected to and fixed to the first side wall 14a, for example, so as to cover the first flow path 14e of the container 14. The first flow velocity adjustment section 15 is provided with an opening for allowing the refrigerant 30 to flow from the first flow path 14e to the third flow path 14g.

[0048] The first velocity adjustment section 15 includes a first region 15a with a first slit 15aa having a first width h2 as an opening, and a second region 15b with second slits 15ba having second widths h1 and h3 as openings. For example, if the first velocity adjustment section 15 is divided into three regions in its longitudinal direction (corresponding to the direction in which the first flow path 14e extends along the first side wall 14a), the central region is designated as the first region 15a, and the remaining two outer regions are designated as the second region 15b. The first velocity adjustment section 15 shown in Figure 5 has a configuration in which the central first region 15a is sandwiched between the two outer second regions 15b. The first region 15a has a first longitudinal length of w2, and the two second regions 15b have second longitudinal lengths of w1 and w3. Furthermore, the total length in the longitudinal direction of the first flow velocity adjustment section 15 is the length w of the internal space of the container 14 shown in Figure 4 (length w of the first flow path 14e). The first length w2 of the first region 15a and the second lengths w1 and w3 of the second region 15b are each set to a length of approximately 1 / 3 of the total length w of the first flow velocity adjustment section 15.

[0049] The first width h2 of the first slit 15aa in the first region 15a, and the second widths h1 and h3 of the second slit 15ba in the second region 15b are set in the range of 1 mm to 3 mm. The first width h2 of the first slit 15aa in the first region 15a and the second widths h1 and h3 of the second slit 15ba in the second region 15b are set to be different widths from each other. For example, the second widths h1 and h3 of the second slit 15ba in the second region 15b are set to be the same width, but they may also be set to be different widths from each other. In the example in Figure 5, the first width h2 of the first slit 15aa in the first region 15a is set to be wider than the second widths h1 and h3 of the second slit 15ba in the second region 15b.

[0050] The first region 15a of the first flow velocity adjustment section 15, where the first slit 15aa is provided, has a first aperture ratio, and the second region 15b, where the second slit 15ba is provided, has a second aperture ratio smaller than the first aperture ratio of the first region 15a. Here, the first aperture ratio of the first region 15a is the ratio of the opening portion per unit area of ​​the first region 15a opened by the first slit 15aa. The second aperture ratio of the second region 15b is the ratio of the opening portion per unit area of ​​the second region 15b opened by the second slit 15ba.

[0051] The first slit 15aa and the second slit 15ba of the first flow velocity adjustment section 15 are positioned at one end of the two ends that extend in the longitudinal direction, that is, at the end that faces the first side wall 14a when the first flow velocity adjustment section 15 is positioned to cover the first flow channel 14e of the container 14. The first slit 15aa and the second slit 15ba are formed continuously, but they may be separated at the boundary between the first slit 15aa and the second slit 15ba.

[0052] Furthermore, the second flow velocity adjustment section 16 is formed, for example, from a plate-shaped member and is arranged parallel to and spaced apart from the bottom surface of the second flow path 14f (second groove). The second flow velocity adjustment section 16 is connected to and fixed to the second side wall 14b, for example, so as to cover the second flow path 14f of the container 14. The second flow velocity adjustment section 16 is provided with an opening for allowing the refrigerant 30 to flow from the third flow path 14g to the second flow path 14f.

[0053] The second velocity adjustment section 16 includes a third region 16a with a third slit 16aa having a third width h6 as an opening, and a fourth region 16b with fourth slits 16ba having fourth widths h5 and h7 as openings. For example, if the second velocity adjustment section 16 is divided into three regions in its longitudinal direction (corresponding to the direction in which the second flow path 14f extends along the second side wall 14b), the central region is designated as the third region 16a, and the remaining two outer regions are designated as the fourth region 16b. The second velocity adjustment section 16 shown in Figure 5 has a configuration in which the central third region 16a is sandwiched between the two outer fourth regions 16b. The third region 16a has a third longitudinal length of w6, and the two fourth regions 16b have fourth longitudinal lengths of w5 and w7. Furthermore, the total length of the second flow velocity adjustment section 16 in the longitudinal direction is the length w of the internal space of the container 14 shown in Figure 4 (length w of the second flow path 14f). The third length w6 of the third region 16a and the fourth lengths w5 and w7 of the fourth region 16b are each set to a length of approximately 1 / 3 of the total length w of the second flow velocity adjustment section 16.

[0054] The third width h6 of the third slit 16aa in the third region 16a, and the fourth widths h5 and h7 of the fourth slit 16ba in the fourth region 16b are set in the range of 1 mm to 3 mm. The third width h6 of the third slit 16aa in the third region 16a and the fourth widths h5 and h7 of the fourth slit 16ba in the fourth region 16b are set to be different widths from each other. For example, the fourth widths h5 and h7 of the fourth slit 16ba in the fourth region 16b are set to be the same width, but they may also be set to be different widths from each other. In the example in Figure 5, the third width h6 of the third slit 16aa in the third region 16a is set to be narrower than the fourth widths h5 and h7 of the fourth slit 16ba in the fourth region 16b.

[0055] The third region 16a of the second flow velocity adjustment section 16, where the third slit 16aa is provided, has a third aperture ratio, and the fourth region 16b, where the fourth slit 16ba is provided, has a fourth aperture ratio that is greater than the third aperture ratio of the third region 16a. Here, the third aperture ratio of the third region 16a is the ratio of the opening portion per unit area of ​​the third region 16a opened by the third slit 16aa. The fourth aperture ratio of the fourth region 16b is the ratio of the opening portion per unit area of ​​the fourth region 16b opened by the fourth slit 16ba.

[0056] The third slit 16aa and the fourth slit 16ba of the second flow velocity adjustment section 16 are positioned at one end of the two ends that extend in the longitudinal direction, that is, at the end that faces the second side wall 14b when the second flow velocity adjustment section 16 is positioned to cover the second flow channel 14f of the container 14. Although the third slit 16aa and the fourth slit 16ba are formed continuously, they may be separated at the boundary between the third slit 16aa and the fourth slit 16ba.

[0057] The first flow velocity adjustment unit 15 and the second flow velocity adjustment unit 16 are arranged in the container 14 of the cooler 10 such that their respective first region 15a and third region 16a face each other, namely the first region 15a, which has a relatively wide first slit 15aa and a relatively large opening ratio, and the third region 16a, which has a relatively narrow third slit 16aa and a relatively small opening ratio. The first flow velocity adjustment unit 15 and the second flow velocity adjustment unit 16 are arranged in the container 14 of the cooler 10 such that their respective second region 15b and fourth region 16b face each other, namely the second region 15b, which has a relatively narrow second slit 15ba and a relatively small opening ratio, and the fourth region 16b, which has a relatively wide fourth slit 16ba and a relatively large opening ratio.

[0058] The first length w2 of the first region 15a of the first flow velocity adjustment section 15, the first width h2 of the first slit 15aa of the first region 15a, the second lengths w1 and w3 of the second region 15b, the second widths h1 and h3 of the second slit 15ba of the second region 15b, and the third length w6 of the third region 16a of the second flow velocity adjustment section 16, the third width h6 of the third slit 16aa of the third region 16a, the fourth lengths w5 and w7 of the fourth region 16b, and the fourth widths h5 and h7 of the fourth slit 16ba of the fourth region 16b are set appropriately based on the dimensions of the container 14 of the cooler 10, for example, the dimensions of the first flow path 14e and the second flow path 14f, and the required cooling performance.

[0059] The first flow velocity adjustment section 15 and the second flow velocity adjustment section 16 are formed by die casting, pressing, or the like. The first flow velocity adjustment section 15 is connected to the side wall of the first flow path 14e (at least one of the first side wall 14a, the third side wall 14c, the fourth side wall 14d, and the side wall of the first flow path 14e facing the first side wall 14a) using brazing or various welding techniques so as to cover the first flow path 14e of the container 14, and is integrated with the container 14. The second flow velocity adjustment section 16 is connected to the side wall of the second flow path 14f (at least one of the second side wall 14b, the third side wall 14c, the fourth side wall 14d, and the side wall of the second flow path 14f facing the second side wall 14b) using brazing or various welding techniques so as to cover the second flow path 14f of the container 14, and is integrated with the container 14.

[0060] Furthermore, in addition to plate-shaped members, cylindrical members formed to match the groove shapes of the first flow path 14e and the second flow path 14f of the container 14 may be used for the first flow velocity adjustment section 15 and the second flow velocity adjustment section 16, respectively. For the cylindrical member used for the first flow velocity adjustment section 15, a first slit 15aa and a second slit 15ba are formed at predetermined positions on one side surface by cutting or the like. For the cylindrical member used for the second flow velocity adjustment section 16, a third slit 16aa and a fourth slit 16ba are formed at predetermined positions on one side surface by cutting or the like. By fitting these cylindrical members for the first flow velocity adjustment section 15 and the second flow velocity adjustment section 16 into the first flow path 14e and the second flow path 14f of the container 14, respectively, a container 14 in which the first flow velocity adjustment section 15 and the second flow velocity adjustment section 16 are integrated may be obtained.

[0061] Next, the cooler 10, in which the first flow velocity adjustment unit 15 and the second flow velocity adjustment unit 16 are integrated with the container 14, will be described with reference to Figures 6 to 8. Figures 6 to 8 illustrate an example of the configuration of a cooler according to the first embodiment. Figure 6 schematically shows a perspective view of the main part of an example of a cooler according to the first embodiment. Figure 7 schematically shows a plan view of the main part of an example of a cooler according to the first embodiment. Figures 8(A) and 8(B) schematically show cross-sectional views of the main part of an example of a cooler according to the first embodiment. Figure 8(A) is a cross-sectional view taken from VIIIa-VIIIa in Figure 7, and Figure 8(B) is a cross-sectional view taken from VIIIb-VIIIb in Figure 7.

[0062] A first flow velocity adjustment unit 15 and a second flow velocity adjustment unit 16, as shown in Figure 5, are arranged and connected to a container 14 (water jacket) as shown in Figures 4(A) and 4(B) above, resulting in a cooler 10 as shown in Figures 6, 7, 8(A), and 8(B). Note that the cooler 10 shown in Figures 6, 7, 8(A), and 8(B) omits the heat sink 13 (fin base) with cooling fins 13a, as shown in Figures 1 and 2 above. Also, the flow of the refrigerant 30 is schematically shown by dotted arrows in Figures 6, 7, 8(A), and 8(B).

[0063] The first flow velocity adjustment section 15 is positioned to cover the first flow channel 14e, which extends along the first side wall 14a of the container 14. The first flow velocity adjustment section 15 is positioned such that its opening, i.e., the first slit 15aa of the first region 15a and the second slit 15ba of the second region 15b, are located at the end of the first flow velocity adjustment section 15 on the side of the first side wall 14a of the container 14. It can also be said that the first slit 15aa of the first region 15a and the second slit 15ba of the second region 15b of the first flow velocity adjustment section 15 are positioned at the end of the first flow channel 14e on the side of the first side wall 14a. Of the three regions obtained by dividing the first flow channel 14e in the direction extending along the first side wall 14a, the central one corresponds to the first region 15a of the first flow velocity adjustment section 15, and the remaining two outer regions correspond to the second region 15b of the first flow velocity adjustment section 15. A first slit 15aa is provided in the first region 15a, and a second slit 15ba, which is narrower than the first slit 15aa, is provided in the second region 15b. The first region 15a has a first aperture ratio, and the second region 15b has a second aperture ratio that is smaller than the first aperture ratio of the first region 15a.

[0064] The second velocity adjustment section 16 is positioned to cover the second flow channel 14f that extends along the second side wall 14b of the container 14. The second velocity adjustment section 16 is positioned such that its opening, i.e., the third slit 16aa of the third region 16a and the fourth slit 16ba of the fourth region 16b, is located at the end of the second velocity adjustment section 16 on the side of the second side wall 14b of the container 14. It can also be said that the third slit 16aa of the third region 16a and the fourth slit 16ba of the fourth region 16b of the second velocity adjustment section 16 are positioned at the end of the second flow channel 14f on the side of the second side wall 14b. Of the three regions obtained by dividing the second flow channel 14f in the direction extending along the second side wall 14b, the central one corresponds to the third region 16a of the second velocity adjustment section 16, and the remaining two outer regions correspond to the fourth region 16b of the second velocity adjustment section 16. A third slit 16aa is provided in the third region 16a, and a fourth slit 16ba, which is wider than the third slit 16aa, is provided in the fourth region 16b. The third region 16a has a third aperture ratio, and the fourth region 16b has a fourth aperture ratio that is greater than the third aperture ratio of the third region 16a.

[0065] The first flow velocity adjustment section 15 and the second flow velocity adjustment section 16 are arranged in the container 14 such that a first region 15a, which has a relatively wide first slit 15aa and a relatively large opening ratio, and a third region 16a, which has a relatively narrow third slit 16aa and a relatively small opening ratio, face each other. The first flow velocity adjustment section 15 and the second flow velocity adjustment section 16 are arranged in the container 14 such that a second region 15b, which has a relatively narrow second slit 15ba and a relatively small opening ratio, and a fourth region 16b, which has a relatively wide fourth slit 16ba and a relatively large opening ratio, face each other.

[0066] In the examples shown in Figures 6, 7, 8(A), and 8(B), the first region 15a of the first flow velocity adjustment section 15, which has a relatively wide first slit 15aa and a relatively large opening ratio, is positioned closer to the inlet 11 of the refrigerant 30 that communicates with the first flow path 14e of the container 14 than the second region 15b, which has a relatively narrow second slit 15ba and a relatively small opening ratio. In the examples shown in Figures 6, 7, 8(A), and 8(B), the third region 16a of the second flow velocity adjustment section 16, which has a relatively narrow third slit 16aa and a relatively small opening ratio, is positioned closer to the outlet 12 of the refrigerant 30 that communicates with the second flow path 14f of the container 14 than the fourth region 16b, which has a relatively wide fourth slit 16ba and a relatively large opening ratio.

[0067] In a container 14 in which the first flow path 14e is covered by the first flow velocity adjustment section 15 and the second flow path 14f is covered by the second flow velocity adjustment section 16, a third flow path 14g is formed in the internal space above the first flow path 14e and the second flow path 14f. Specifically, the first flow velocity adjustment section 15 is positioned at the boundary between the first flow path 14e and the third flow path 14g, and the second flow velocity adjustment section 16 is positioned at the boundary between the second flow path 14f and the third flow path 14g. The first flow path 14e and the third flow path 14g are in communication through the first slit 15aa and the second slit 15ba of the first flow velocity adjustment section 15, and the second flow path 14f and the third flow path 14g are in communication through the third slit 16aa and the fourth slit 16ba of the second flow velocity adjustment section 16.

[0068] Although not shown in the diagram here, a heat sink 13 with cooling fins 13a, as shown in Figures 1, 2, 3(A), and 3(B), or a heat sink 13 with a semiconductor module 20 mounted on the opposite side of the cooling fins 13a, is arranged to cover the internal space of such a container 14. The heat sink 13 and the container 14 are fastened and connected, for example, using bolts. The cooling fins 13a of the heat sink 13 connected to the container 14 are arranged to be housed within the third flow channel 14g of the container 14, as shown in Figure 2. The cooling fins 13a are provided such that a certain clearance c1 (Figure 2) is secured between the tip of the cooling fins 13a and the bottom surface of the third flow channel 14g when the heat sink 13 is connected to the container 14.

[0069] When the cooler 10 is in use, the refrigerant 30 flows through the cooler 10 as shown by the dotted arrows in Figures 6, 7, 8(A), and 8(B). At that time, the refrigerant 30 supplied to the cooler 10 by the pump 40 (Figure 1) is introduced into the cooler 10 from the inlet 11. The refrigerant 30 introduced from the inlet 11 flows into the first flow path 14e of the container 14 which is in communication with the inlet 11, and flows from the first flow path 14e through the relatively wide first slit 15aa (Figure 8(A)) and the relatively narrow second slit 15ba (Figure 8(B)) of the first flow velocity adjustment unit 15 into the third flow path 14g. The refrigerant 30 that flows into the third flow path 14g flows from the third flow path 14g through the relatively narrow third slit 16aa (Figure 8(A)) and the relatively wide fourth slit 16ba (Figure 8(B)) of the second flow velocity adjustment unit 16 into the second flow path 14f of the container 14 which is in communication with the outlet 12. The refrigerant 30 that flows into the second flow path 14f is discharged outside the cooler 10 from the outlet 12.

[0070] The refrigerant 30 that flows from the first channel 14e into the third channel 14g flows through the refrigerant channel defined by the cooling fins 13a housed within the third channel 14g, that is, through the gaps between adjacent cooling fins 13a. While the refrigerant 30 flows through the third channel 14g, the heat transferred from the semiconductor module 20 to the heat sink 13 and its cooling fins 13a is exchanged with the refrigerant 30 flowing through the third channel 14g, thereby cooling the semiconductor module 20. The refrigerant 30, whose temperature has risen due to the heat exchange with the heat sink 13 and its cooling fins 13a, flows into the second channel 14f and is discharged outside the cooler 10 from the outlet 12. The refrigerant 30, whose temperature has been lowered by being sent to the heat exchanger 50 (Figure 1), is then introduced back into the cooler 10 from the inlet 11 by the pump 40.

[0071] The cooler 10 having the above configuration makes it possible to suppress the occurrence of uneven flow distribution and the increase in pressure loss of the refrigerant 30 circulating within the cooler 10. Furthermore, it becomes possible to realize a semiconductor device 1 equipped with a cooler 10 that can suppress such uneven flow distribution and the increase in pressure loss. This point will be explained further below.

[0072] Here, we will use the cooler and semiconductor device equipped therewith shown in Figures 9 to 11 as comparative examples. Figures 9 to 11 illustrate examples of the configuration of a cooler according to comparative examples. Figure 9 schematically shows a perspective view of the main part of an example of a cooler according to comparative examples. Figure 10 schematically shows a plan view of the main part of the first flow velocity adjustment section and the second flow velocity adjustment section of the cooler according to comparative examples. Figure 11 schematically shows a cross-sectional view of the main part of an example of a cooler according to comparative examples. Figure 11 is a cross-sectional view taken along line XI-XI in Figure 9. In addition, the flow of the refrigerant 30 is schematically shown by dotted arrows in Figures 9 and 11.

[0073] The cooler 110 shown in Figure 9 differs from the cooler 10 described in the first embodiment in that it has a configuration in which a first flow rate adjustment unit 115 and a second flow rate adjustment unit 116 are arranged as shown in Figures 9 to 11. The container 14 of the cooler 110, as well as the heat sink 13 and its cooling fins 13a that cover the container 14 (though not shown here), and the semiconductor module 20 mounted on the heat sink 13 are the same as those described in the first embodiment.

[0074] As shown in Figure 10, the first flow velocity adjustment section 115 of the comparative example cooler 110 has a configuration in which a seventh slit 115aa is provided as an opening, having a longitudinal length w4 and a constant width h4. The second flow velocity adjustment section 116 of the comparative example cooler 110 has a configuration in which an eighth slit 116aa is provided as an opening, having a longitudinal length w8 and a constant width h8. The first flow velocity adjustment section 115 and the second flow velocity adjustment section 116 are arranged to cover the first flow path 14e and the second flow path 14f of the container 14, respectively. The seventh slit 115aa of the first flow velocity adjustment section 115 is positioned at the end of the first flow velocity adjustment section 115 on the first side wall 14a side, that is, at the end of the first flow path 14e on the first side wall 14a side. The eighth slit 116aa of the second flow velocity adjustment section 116 is positioned at the end of the second side wall 14b of the second flow velocity adjustment section 116, that is, at the end of the second flow channel 14f on the second side wall 14b side.

[0075] Although not shown in the diagram here, a heat sink 13 with cooling fins 13a, or a heat sink 13 with a semiconductor module 20 mounted on the opposite side of the cooling fins 13a, is arranged to cover the internal space of such a container 14, according to the examples in Figures 1, 2, 3(A), and 3(B). The heat sink 13 and the container 14 are fastened and connected, for example, using bolts. The cooling fins 13a of the heat sink 13 connected to the container 14 are arranged to be housed within the third flow channel 14g of the container 14.

[0076] When the cooler 110 is in use, the inlet 11 of the cooler 110 is connected to the pump 40 by piping, and the outlet 12 of the cooler 110 is connected to the heat exchanger 50 by piping, as shown in Figure 1 above. The pump 40 and the heat exchanger 50 are connected by piping. The refrigerant 30 flows through the cooler 110 as shown by the dotted arrows in Figures 9 and 11. That is, the refrigerant 30 supplied to the cooler 110 by the pump 40 is introduced into the cooler 110 from the inlet 11. The refrigerant 30 introduced from the inlet 11 flows into the first flow path 14e of the container 14 which is in communication with the inlet 11, and flows from the first flow path 14e through the seventh slit 115aa of the first flow velocity adjustment unit 115 into the third flow path 14g. The refrigerant 30 that flows into the third flow path 14g flows from the third flow path 14g through the eighth slit 116aa of the second flow velocity adjustment unit 116, which has a constant width, into the second flow path 14f of the container 14 that communicates with the outlet 12. The refrigerant 30 that flows into the second flow path 14f is discharged outside the cooler 110 from the outlet 12.

[0077] The refrigerant 30 that flows from the first channel 14e into the third channel 14g flows through the refrigerant channel defined by the cooling fins 13a housed within the third channel 14g, that is, through the gaps between adjacent cooling fins 13a. While the refrigerant 30 flows through the third channel 14g, the heat transferred from the semiconductor module 20 to the heat sink 13 and its cooling fins 13a is exchanged with the refrigerant 30 flowing through the third channel 14g, thereby cooling the semiconductor module 20. The refrigerant 30, whose temperature has risen due to the heat exchange with the heat sink 13 and its cooling fins 13a, flows into the second channel 14f and is discharged outside the cooler 110 from the outlet 12. The refrigerant 30, whose temperature has been lowered by being sent to the heat exchanger 50, is then introduced back into the cooler 110 from the inlet 11 by the pump 40.

[0078] Here, the cooler 10 according to the first embodiment described above will be referred to as "Type A," and the cooler 110 according to this comparative example will be referred to as "Type B." Furthermore, a cooler using a container 14 without the first flow rate adjustment sections 15 and 115 and the second flow rate adjustment sections 16 and 116 described above will be referred to as "Type C."

[0079] The length w, width h0, width h, height t1, and height t2 of the container 14 for the Type A cooler 10, Type B cooler 110, and Type C cooler are the dimensions of the parts shown in Figure 4 above. The length w of the container 14 for the Type A cooler 10, Type B cooler 110, and Type C cooler is set to be the same as that of the container 14 for the Type A cooler 10, Type B cooler 110, and Type C cooler. The width h0 of the container 14 for the Type A cooler 10, Type B cooler 110, and Type C cooler is set to be the same as that of the container 14 for the Type A cooler 10, Type B cooler 110, and Type C cooler. The height t1 of the container 14 for the Type A cooler 10, Type B cooler 110, and Type C cooler is set to be the same as that of the container 14 for the Type C cooler 10, Type B cooler 110, and Type C cooler. The height t2 dimension of the container 14 for the type A cooler 10, the type B cooler 110, and the type C cooler is set to be the same for all of them.

[0080] The first length w2, second lengths w1 and w3, first width h2, second width h1 and h3 of the first flow velocity adjustment section 15 of the type A cooler 10 are the dimensions of the part shown in Figure 5 above. The third length w6, fourth lengths w5 and w7, third width h6, fourth width h5 and h7 of the second flow velocity adjustment section 16 of the type A cooler 10 are the dimensions of the part shown in Figure 5 above. The dimensions of the first length w2, second lengths w1 and w3 of the first flow velocity adjustment section 15 are set to lengths obtained by dividing the length w of the container 14 into approximately three equal parts. The dimension of the first width h2 of the first flow velocity adjustment section 15 is set to 2 mm as an example, and the dimensions of the second width h1 and h3 are set to 1 mm as an example. The dimensions of the third length w6, fourth lengths w5 and w7 of the second flow velocity adjustment section 16 are set to lengths obtained by dividing the length w of the container 14 into approximately three equal parts. The dimension of the third width h6 of the second flow velocity adjustment section 16 is set to 1 mm as an example, and the dimensions of the fourth widths h5 and h7 are set to 2 mm as an example.

[0081] The length w4 and width h4 of the first flow velocity adjustment section 115 of the Type B cooler 110 are the dimensions of the part shown in Figure 10 above. The length w8 and width h8 of the second flow velocity adjustment section 116 of the Type B cooler 110 are the dimensions of the part shown in Figure 10 above. The length w4 of the first flow velocity adjustment section 115 is set to the same dimension as the sum of the first length w2, second length w1, and second length w3 of the first flow velocity adjustment section 15 of the Type A cooler 10. The width h4 of the first flow velocity adjustment section 115 is set to the same dimension as the second widths h1 and h3 of the first flow velocity adjustment section 15 of the Type A cooler 10, for example, set to 1 mm. The length w8 of the second flow velocity adjustment section 116 is set to the same dimension as the sum of the third length w6, fourth length w5, and fourth length w7 of the second flow velocity adjustment section 16 of the Type A cooler 10. The width h8 of the second flow velocity adjustment section 116 is set to the same dimension as the third width h6 of the second flow velocity adjustment section 16 of the type A cooler 10, for example, to 1 mm.

[0082] The results of thermal fluid simulations and evaluations of the Type A cooler 10, Type B cooler 110, and Type C cooler, which employ the dimensions described above, are shown in Figures 12 to 14.

[0083] Figure 12 shows an example of the evaluation results of the refrigerant flow rate with respect to the semiconductor element position. Figure 13 shows an example of the evaluation results of the pressure loss in each type of cooler. Figure 14 shows an example of the evaluation results of the semiconductor element temperature with respect to the semiconductor element position.

[0084] In the thermal fluid simulation, the flow rate of the refrigerant 30 introduced from the inlet 11 of the container 14 is set to 10 L / min. In the thermal fluid simulation, heat generation is reproduced by introducing a certain amount of loss to the semiconductor module 20 as shown in Figure 1 above. Specifically, heat generation is reproduced by introducing a certain amount of loss to the semiconductor elements CP1 (semiconductor element 25) and CP2 (semiconductor element 26) in each of the three mounting areas AR1 (circuit element section 21), mounting area AR2 (circuit element section 22), and mounting area AR3 (circuit element section 23) of the semiconductor module 20 mounted on the heat sink 13 covering the container 14.

[0085] Figure 12 shows the flow velocity of the refrigerant 30 at the locations of semiconductor elements CP1 and CP2 in mounting region AR1, the flow velocity of the refrigerant 30 at the locations of semiconductor elements CP1 and CP2 in mounting region AR2, and the flow velocity of the refrigerant 30 at the locations of semiconductor elements CP1 and CP2 in mounting region AR3.

[0086] As shown in Figure 12, in a Type C cooler using a container 14 without the first flow velocity adjustment units 15 and 115 and the second flow velocity adjustment units 16 and 116 as described above, the flow velocity of the refrigerant 30 at the positions of semiconductor elements CP1 and CP2 in the central mounting region AR2 is around 0.65 m / s, and the flow velocity of the refrigerant 30 at the positions of semiconductor elements CP1 and CP2 in each of the mounting regions AR1 and AR3 at both ends is about 0.40 m / s to 0.45 m / s, resulting in a flow bias distribution. On the other hand, as can be seen from Figure 12, in the Type A cooler 10 using a container 14 equipped with a first flow velocity adjustment unit 15 and a second flow velocity adjustment unit 16, and in the Type B cooler 110 using a container 14 equipped with a first flow velocity adjustment unit 115 and a second flow velocity adjustment unit 116, the flow velocity of the refrigerant 30 at the positions of semiconductor elements CP1 and CP2 in mounting areas AR1, AR2, and AR3 is around 0.40 m / s, indicating that a more uniform flow is generated compared to the Type C cooler.

[0087] Figure 13 shows the pressure loss between the inlet 11 and outlet 12 of the container 14, that is, the decrease in the pressure of the refrigerant at the outlet 12 relative to the pressure of the refrigerant 30 at the inlet 11. As shown in Figure 13, in a Type C cooler using a container 14 without the first velocity adjustment units 15 and 115 and the second velocity adjustment units 16 and 116 as described above, the pressure loss is about 5.0 kPa, whereas in a Type B cooler 110 using a container 14 equipped with the first velocity adjustment unit 115 and the second velocity adjustment unit 116, the pressure loss increases to 9.0 kPa, an 80% increase. On the other hand, as shown in Figure 13, in a Type A cooler 10 using a container 14 equipped with the first velocity adjustment unit 15 and the second velocity adjustment unit 16, the pressure loss is about 7.0 kPa, and the increase in pressure loss from Type C is limited to 40%.

[0088] Figure 14 shows the temperatures of semiconductor elements CP1 and CP2 in mounting region AR1, in mounting region AR2, and in mounting region AR3.

[0089] As shown in Figure 14, in a Type C cooler using a container 14 without the first flow velocity adjustment units 15 and 115 and the second flow velocity adjustment units 16 and 116 as described above, the semiconductor elements CP1 and CP2 in the central mounting region AR2 (Figure 12), where the flow velocity of the refrigerant 30 is relatively high, are cooled well and their temperature becomes relatively low at around 124°C, while the temperature of the semiconductor elements CP1 and CP2 in the mounting regions AR1 and AR3 (Figure 12) at both ends, where the flow velocity of the refrigerant 30 is relatively slow, becomes relatively high at 125°C or higher. On the other hand, as can be seen from Figure 14, in the Type A cooler 10 using a container 14 equipped with a first flow velocity adjustment unit 15 and a second flow velocity adjustment unit 16, and in the Type B cooler 110 using a container 14 equipped with a first flow velocity adjustment unit 115 and a second flow velocity adjustment unit 116, the temperatures of the semiconductor elements CP1 and CP2 in all of the mounting areas AR1, AR2, and AR3 (Figure 2), where the flow velocity of the refrigerant 30 is relatively uniform, are around 124°C, indicating that the cooling is more uniform compared to the Type C cooler.

[0090] As shown in Figures 12 to 14, the Type A cooler 10 can suppress pressure loss more effectively than the Type B cooler 110 while achieving a flow distribution suppression effect and semiconductor device cooling effect equivalent to or close to that of the Type B cooler 110.

[0091] According to the Type A cooler 10, that is, the cooler 10 according to the first embodiment, it is possible to suppress the occurrence of uneven flow distribution and the increase in pressure loss of the refrigerant 30 circulating within the cooler 10. Furthermore, it becomes possible to realize a semiconductor device 1 equipped with a cooler 10 that can suppress such uneven flow distribution and the increase in pressure loss.

[0092] Generally, semiconductor modules 20 like the one described above are widely used in power conversion devices used in control systems for hybrid vehicles and electric vehicles. In these semiconductor modules 20 that constitute energy-saving control devices, power semiconductor elements that control large currents are used as semiconductor elements 25 (CP1) and semiconductor elements 26 (CP2). Ordinary power semiconductor elements are heat-generating elements that generate heat when controlling large currents, but as power conversion devices become smaller and more powerful, the amount of heat generated increases. Therefore, cooling becomes an important issue in semiconductor modules 20 that have multiple heat-generating elements.

[0093] For example, conventionally, liquid-cooled coolers have been used to cool semiconductor modules 20. In liquid-cooled coolers, measures have been taken to improve cooling efficiency, such as increasing the refrigerant flow rate or using cooling fins with a high heat transfer coefficient. However, as a result of such measures, the pressure loss of the refrigerant inside the cooler may increase, which can increase the load on the pump that circulates the refrigerant. To reduce pressure loss, it is ideal to increase cooling efficiency with a low refrigerant flow rate, which can be achieved by reducing the refrigerant flow rate and using cooling fins with a high heat transfer coefficient. However, adopting such cooling fins may lead to increased costs for the cooler and the semiconductor device using it. Furthermore, in conventional liquid-cooled coolers, a flow distribution occurs where the refrigerant flows unevenly inside the cooler due to the shape of the heat sink and refrigerant flow path, the arrangement of heat-generating elements, or the shape of the refrigerant inlet and outlet. Because such a flow distribution leads to uneven cooling performance, it has been difficult to obtain uniform and stable cooling performance with conventional coolers. As a result, the temperature of some heat-generating elements may rise, potentially leading to a decrease in their performance and lifespan, or even failure.

[0094] Conventionally, techniques for improving the uneven flow distribution within such coolers have been known, such as changing the dimensions of the openings through which the refrigerant flows by changing the position of the refrigerant inlet and the heat-generating element (for example, Patent Documents 1 and 2 above). However, such techniques can complicate the structure of the cooler and lead to increased costs. Techniques for flowing refrigerant to the heat sink through a single slit of a certain width (for example, Patent Documents 3, 4 and 5 above, or the Type B cooler 110 above), and techniques for introducing refrigerant through multiple holes or slits of the same size (for example, Patent Documents 6 and 7 above) are also known. However, with such techniques, if the shape of the heat sink and the refrigerant inlet and outlet have a large influence, the width of the slit or the diameter of the hole must be reduced to obtain a uniform flow velocity distribution, which easily leads to an increase in pressure loss. Techniques for suppressing the increase in pressure loss by providing a refrigerant flow path on the side of the heat sink are also known (for example, Patent Documents 8 and 9 above). However, with such techniques, the overall dimensions of the flow path in the cooler become large, resulting in an excessively large semiconductor device equipped with the cooler. Furthermore, if bolt holes or sealing grooves for connecting the heatsink to the cooler container are provided near the side of the heatsink, it becomes difficult to employ such technology.

[0095] In contrast, in the cooler 10 (Type A) according to the first embodiment, a first flow velocity adjustment section 15 and a second flow velocity adjustment section 16 are arranged between the parallel first flow channels 14e and second flow channels 14f in the container 14 and the third flow channel 14g communicating with them, respectively. The first flow velocity adjustment section 15 includes a first region 15a with a first aperture ratio determined by a relatively wide first slit 15aa, and a second region 15b with a second aperture ratio smaller than the first aperture ratio determined by a relatively narrow second slit 15ba. The second flow velocity adjustment section 16 includes a third region 16a with a third aperture ratio determined by a relatively narrow third slit 16aa, and a fourth region 16b with a fourth aperture ratio larger than the third aperture ratio determined by a relatively wide fourth slit 16ba. In such a cooler 10, by forming multiple types of gaps with appropriate shapes and dimensions around the first flow velocity adjustment section 15 and the second flow velocity adjustment section 16, the refrigerant 30 can flow smoothly without applying excessive pressure inside the first flow path 14e and the second flow path 14f. As a result, while keeping the size of the cooler 10 and the semiconductor device 1 equipped therewith down, it is possible to suppress an increase in pressure loss while maintaining a more uniform flow velocity distribution of the refrigerant 30.

[0096] According to the first embodiment of the cooler 10, it is possible to suppress the occurrence of uneven flow distribution of the refrigerant 30 circulating inside the cooler 10 and the increase in pressure loss, while suppressing the complexity and size of its structure and the constraints on the connection between the container 14 and the heat sink 13. Furthermore, it becomes possible to realize a semiconductor device 1 equipped with such a cooler 10.

[0097] Figure 15 illustrates a first modified example of a cooling fin provided on a heat sink of a cooler. Figure 15(A) schematically shows a perspective view of the main part of the first modified example of a cooling fin provided on a heat sink, and Figure 15(B) schematically shows a plan view of the main part of the first modified example of a cooling fin provided on a heat sink. Figure 15(B) is an enlarged plan view of part Z1 of Figure 15(A).

[0098] The mounting surface 13b of the heat sink 13, which covers the container 14 of the cooler 10 and is connected to the container 14, may be provided with cylindrical cooling fins 13a, as shown in Figures 15(A) and 15(B), in addition to the rectangular or approximately rectangular prism-shaped cooling fins 13a described above. The dimensions of the cylindrical cooling fins 13a are appropriately selected according to the required cooling performance. For example, multiple cylindrical cooling fins 13a, as shown in Figures 15(A) and 15(B), are arranged in a close-packed manner on the heat sink 13.

[0099] The cylindrical cooling fins 13a are integrated with the heat sink 13. Metal materials are used for both the heat sink 13 and the cylindrical cooling fins 13a. The cylindrical cooling fins 13a are integrated with the heat sink 13 by various welding techniques, such as die casting or brazing. Alternatively, the cylindrical cooling fins 13a integrated with the heat sink 13 may be formed by processing techniques that create convex-shaped cooling fins 13a from the heat sink 13 material by die casting, forging, or pressing, or by processing techniques that create convex-shaped cooling fins 13a from the heat sink 13 material by cutting or wire cutting.

[0100] A heat sink 13, equipped with cylindrical cooling fins 13a as shown in Figures 15(A) and 15(B), is placed on the container 14 such that the cooling fins 13a are housed in the third flow channel 14g, and is connected and fixed to the container 14. The cylindrical cooling fins 13a also transfer heat generated by the semiconductor module 20 mounted on the heat sink 13 to the cooling fins 13a, and heat exchange occurs with the refrigerant 30 flowing through the third flow channel 14g, thereby cooling the semiconductor module 20.

[0101] Figure 16 illustrates a second modified example of a cooling fin provided on a heat sink of a cooler. Figure 16(A) schematically shows a perspective view of the main part of the second modified example of a cooling fin provided on a heat sink, and Figure 16(B) schematically shows a plan view of the main part of the second modified example of a cooling fin provided on a heat sink. Figure 16(B) is an enlarged plan view of section Z2 of Figure 16(A).

[0102] The heat sink 13, which covers the container 14 of the cooler 10 and is connected to the container 14, may be provided with corrugated cooling fins 13a, or corrugated fins, as shown in Figures 16(A) and 16(B). The dimensions of the corrugated fins provided as cooling fins 13a are appropriately selected according to the required cooling performance. For example, corrugated fins as shown in Figures 16(A) and 16(B) may be arranged on the heat sink 13 as cooling fins 13a.

[0103] The corrugated fins provided as cooling fins 13a are integrated with the heat sink 13. Metal materials are used for both the heat sink 13 and the cooling fins 13a. The corrugated fins provided as cooling fins 13a are integrated with the heat sink 13 using various welding techniques, such as die casting or brazing.

[0104] A heat sink 13, equipped with corrugated fins as shown in Figures 16(A) and 16(B) as cooling fins 13a, is placed on the container 14 and connected and fixed to the container 14, such that the corrugated fins are housed in the third channel 14g. In this case, the corrugated fins are housed in the third channel 14g in such a way that the refrigerant 30 flowing through the third channel 14g from the first channel 14e toward the second channel 14f flows in a direction parallel to the mounting surface 13b of the corrugated fins of the heat sink 13 and along the direction in which the peaks or valleys of the corrugated fins extend. Even when such corrugated fins are provided as cooling fins 13a, the heat generated by the semiconductor module 20 mounted on the heat sink 13 is transferred to the corrugated fins, and heat exchange occurs with the refrigerant 30 flowing through the third channel 14g, thereby cooling the semiconductor module 20.

[0105] Figure 17 illustrates a third modified example of a cooling fin provided on a heat sink of a cooler. Figure 17(A) schematically shows a perspective view of the main part of the third modified example of a cooling fin provided on a heat sink, and Figure 17(B) schematically shows a plan view of the main part of the third modified example of a cooling fin provided on a heat sink. Figure 17(B) is an enlarged plan view of section Z3 of Figure 17(A).

[0106] The heat sink 13, which covers the container 14 of the cooler 10 and is connected to the container 14, may be provided with flat cooling fins 13a, i.e., straight fins (or blade fins), as shown in Figures 17(A) and 17(B). The dimensions of the straight fins provided as cooling fins 13a are appropriately selected according to the required cooling performance. For example, straight fins as shown in Figures 17(A) and 17(B) are arranged on the heat sink 13 as cooling fins 13a.

[0107] The straight fins provided as cooling fins 13a are integrated with the heat sink 13. Metal materials are used for the heat sink 13 and the cooling fins 13a. The straight fins provided as cooling fins 13a are integrated with the heat sink 13 by various welding techniques, such as die casting or brazing. Alternatively, the straight fins integrated with the heat sink 13 may be formed as cooling fins 13a by processing techniques that form convex-shaped straight fins from the material of the heat sink 13 by die casting, forging or pressing, or by cutting or wire cutting.

[0108] A heat sink 13, equipped with straight fins as cooling fins 13a as shown in Figures 17(A) and 17(B), is placed on the container 14 and connected and fixed to the container 14, such that the straight fins are housed in the third channel 14g. In this case, the straight fins are housed in the third channel 14g in such a way that the refrigerant 30 flowing through the third channel 14g from the first channel 14e toward the second channel 14f flows in a direction parallel to the mounting surface 13b of the straight fins of the heat sink 13 and along the direction in which the side walls of the straight fins extend. Even when such straight fins are provided as cooling fins 13a, the heat generated by the semiconductor module 20 mounted on the heat sink 13 is transferred to the straight fins, and heat exchange occurs with the refrigerant 30 flowing through the third channel 14g, thereby cooling the semiconductor module 20.

[0109] [Second Embodiment] Here, a modified example of the container 14 of the cooler 10 will be described as a second embodiment. Figure 18 illustrates a first modified example of the cooler container according to the second embodiment. Figure 18 schematically shows a perspective view of the main part of the first modified example of the cooler container.

[0110] The container 14 shown in Figure 18 has a configuration in which an inlet 11 communicating with a first flow path 14e extending along the first side wall 14a and an outlet 12 communicating with a second flow path 14f extending along the second side wall 14b is provided on the third side wall 14c connecting the first side wall 14a and the second side wall 14b. In the container 14 shown in Figure 18, a first flow velocity adjustment unit 15, for example as shown in Figure 5, is arranged to cover the first flow path 14e communicating with the inlet 11 provided on the third side wall 14c. A second flow velocity adjustment unit 16, for example as shown in Figure 5, is arranged to cover the second flow path 14f communicating with the outlet 12 provided on the third side wall 14c.

[0111] Even with a cooler 10 using a container 14 as shown in Figure 18, by arranging a first flow velocity adjustment unit 15 between the first flow path 14e and the third flow path 14g, and a second flow velocity adjustment unit 16 between the second flow path 14f and the third flow path 14g, it is possible to suppress the occurrence of uneven flow distribution of the refrigerant 30 circulating within the cooler 10 and the increase in pressure loss.

[0112] Furthermore, the opening layout of the first flow velocity adjustment unit 15 and the second flow velocity adjustment unit 16 can also be changed in accordance with the changes in the positions of the inlet 11 and outlet 12 as shown in Figure 18. For example, with respect to the first flow velocity adjustment section 15, the slit width is adjusted so that the opening ratio of the region closest to the inlet 11 among the three regions obtained by dividing the first flow path 14e in the direction extending along the first side wall 14a is greater than the opening ratios of the remaining two regions. Furthermore, with respect to the second flow velocity adjustment section 16, the slit width is adjusted so that the opening ratio of the region closest to the outlet 12 among the three regions obtained by dividing the second flow path 14f in the direction extending along the second side wall 14b is smaller than the opening ratios of the remaining two regions. As a result, the region of the first flow velocity adjustment section 15 that is closest to the inlet 11 and has a relatively large opening ratio faces the region of the second flow velocity adjustment section 16 that is closest to the outlet 12 and has a relatively small opening ratio. Furthermore, the region of the first flow velocity adjustment section 15 that is relatively far from the inlet 11 and has a relatively small opening ratio faces the region of the second flow velocity adjustment section 16 that is relatively far from the outlet 12 and has a relatively large opening ratio. The first flow velocity adjustment unit 15 and the second flow velocity adjustment unit 16, with their modified opening layouts, may be arranged in relation to the container 14 as shown in Figure 18.

[0113] Figure 19 illustrates a second modified example of the cooler container according to the second embodiment. Figure 19 schematically shows a perspective view of the main part of the second modified example of the cooler container. The container 14 shown in Figure 19 has a configuration in which an inlet 11 is provided on a fourth side wall 14d connecting the first side wall 14a and the second side wall 14b, and communicates with a first flow path 14e extending along the first side wall 14a. Furthermore, the container 14 shown in Figure 19 has a configuration in which an outlet 12 is provided on a third side wall 14c connecting the first side wall 14a and the second side wall 14b, and communicates with a second flow path 14f extending along the second side wall 14b. In the container 14 shown in Figure 19, a first flow velocity adjustment unit 15, for example as shown in Figure 5, is arranged to cover the first flow path 14e that communicates with the inlet 11 provided on the fourth side wall 14d. A second flow velocity adjustment unit 16, for example as shown in Figure 5, is arranged to cover the second flow path 14f that communicates with the outlet 12 provided on the third side wall 14c.

[0114] Even with a cooler 10 using a container 14 as shown in Figure 19, by arranging a first flow velocity adjustment unit 15 between the first flow path 14e and the third flow path 14g, and a second flow velocity adjustment unit 16 between the second flow path 14f and the third flow path 14g, it is possible to suppress the occurrence of uneven flow distribution of the refrigerant 30 circulating within the cooler 10 and the increase in pressure loss. Furthermore, in accordance with the changes in the positions of the inlet 11 and outlet 12 as shown in Figure 19, the first flow velocity adjustment unit 15 and the second flow velocity adjustment unit 16 with modified opening layouts may be arranged.

[0115] Figure 20 illustrates a third modified example of the cooler container according to the second embodiment. Figure 20 schematically shows a perspective view of the main part of the third modified example of the cooler container. The container 14 shown in Figure 20 has a configuration in which an inlet 11 communicating with a first flow path 14e extending along a first side wall 14a and an outlet 12 communicating with a second flow path 14f extending along a second side wall 14b are provided on its bottom plate 14h. In the container 14 shown in Figure 20, a first flow velocity adjustment unit 15, for example as shown in Figure 5 above, is arranged to cover the first flow path 14e communicating with the inlet 11 provided on the bottom plate 14h. A second flow velocity adjustment unit 16, for example as shown in Figure 5 above, is arranged to cover the second flow path 14f communicating with the outlet 12 provided on the bottom plate 14h.

[0116] Even with a cooler 10 using a container 14 as shown in Figure 20, by arranging a first flow velocity adjustment unit 15 between the first flow path 14e and the third flow path 14g, and a second flow velocity adjustment unit 16 between the second flow path 14f and the third flow path 14g, it is possible to suppress the occurrence of uneven flow distribution of the refrigerant 30 circulating within the cooler 10 and the increase in pressure loss. Furthermore, in accordance with the changes in the positions of the inlet 11 and outlet 12 as shown in Figure 20, the first flow velocity adjustment unit 15 and the second flow velocity adjustment unit 16 with modified opening layouts may be arranged.

[0117] Figure 21 illustrates a fourth modified example of the cooler container according to the second embodiment. Figure 21 schematically shows a perspective view of the main part of the fourth modified example of the cooler container. The container 14 shown in Figure 21 has a bottom plate 14h, in which an inlet 11 communicating with the first flow path 14e is provided at the end of the first flow path 14e extending along the first side wall 14a on the side of the fourth side wall 14d. Furthermore, the container 14 shown in Figure 21 has a bottom plate 14h, in which an outlet 12 communicating with the second flow path 14f is provided at the end of the second flow path 14f extending along the second side wall 14b on the side of the third side wall 14c. In the container 14 shown in Figure 21, a first flow velocity adjustment unit 15, for example as shown in Figure 5, is arranged to cover the first flow path 14e communicating with the inlet 11 provided on the bottom plate 14h. A second flow velocity adjustment unit 16, for example as shown in Figure 5, is arranged to cover the second flow path 14f communicating with the outlet 12 provided on the bottom plate 14h.

[0118] Even with a cooler 10 using a container 14 as shown in Figure 21, by arranging a first flow velocity adjustment unit 15 between the first flow path 14e and the third flow path 14g, and a second flow velocity adjustment unit 16 between the second flow path 14f and the third flow path 14g, it is possible to suppress the occurrence of uneven flow distribution of the refrigerant 30 circulating within the cooler 10 and the increase in pressure loss. Furthermore, in accordance with the changes in the positions of the inlet 11 and outlet 12 as shown in Figure 21, the first flow velocity adjustment unit 15 and the second flow velocity adjustment unit 16 with modified opening layouts may be arranged.

[0119] [Third Embodiment] Here, a modified version of the first flow rate adjustment unit 15 and the second flow rate adjustment unit 16 of the cooler 10 will be described as a third embodiment. Figure 22 illustrates a first modified example of the first and second flow velocity adjustment sections of the cooler according to the third embodiment. Figure 22 schematically shows a plan view of the main parts of the first modified example of the first and second flow velocity adjustment sections of the cooler.

[0120] The first velocity adjustment section 15 shown in Figure 22 has a configuration in which the first slit 15aa of the central first region 15a, one of a group of three regions divided in the longitudinal direction, is divided into multiple slits 15aa, for example into two, and the second slits 15ba of each of the two outer second regions 15b are divided into multiple slits 15ba, for example into two. The second velocity adjustment section 16 shown in Figure 22 has a configuration in which the third slit 16aa of the central third region 16a, one of a group of three regions divided in the longitudinal direction, is divided into multiple slits 16aa, for example into two, and the fourth slits 16ba of each of the two outer fourth regions 16b are divided into multiple slits 16ba, for example into two. The first velocity adjustment section 15 and the second velocity adjustment section 16, as shown in Figure 22, are arranged to cover the first flow path 14e and the second flow path 14f of the container 14, respectively. The first region 15a of the first flow velocity adjustment section 15, which has a relatively large aperture ratio, and the third region 16a of the second flow velocity adjustment section 16, which has a relatively small aperture ratio, face each other, and the second region 15b of the first flow velocity adjustment section 15, which has a relatively small aperture ratio, and the fourth region 16b of the second flow velocity adjustment section 16, which has a relatively large aperture ratio, face each other.

[0121] A cooler 10 using a first flow velocity adjustment unit 15 and a second flow velocity adjustment unit 16 as shown in Figure 22, that is, a cooler 10 in which the first flow velocity adjustment unit 15 and the second flow velocity adjustment unit 16 are arranged in the first flow path 14e and the second flow path 14f of the container 14, respectively, can also suppress the occurrence of uneven flow distribution of the refrigerant 30 circulating within the cooler 10 and the increase in pressure loss.

[0122] Furthermore, in the first flow velocity adjustment section 15, the first slit 15aa of the first region 15a may be divided into three or more parts, and the second slit 15ba of the second region 15b may be divided into three or more parts. If the opening ratio of the first region 15a is greater than the opening ratio of the second region 15b, the widths of each of the divided first slits 15aa may be the same or different from each other, and the widths of each of the divided second slits 15ba may be the same or different from each other.

[0123] Furthermore, in the second flow velocity adjustment section 16, the third slit 16aa of the third region 16a may be divided into three or more parts, and the fourth slit 16ba of the fourth region 16b may be divided into three or more parts. If the opening ratio of the third region 16a is smaller than the opening ratio of the fourth region 16b, the widths of each of the divided third slits 16aa may be the same or different from each other, and the widths of each of the divided fourth slits 16ba may be the same or different from each other.

[0124] Furthermore, the width of the first slit 15aa of the first flow velocity adjustment section 15 and the width of the fourth slit 16ba of the second flow velocity adjustment section 16 may be the same or different from each other, and the width of the second slit 15ba of the first flow velocity adjustment section 15 and the width of the third slit 16aa of the second flow velocity adjustment section 16 may be the same or different from each other.

[0125] Figure 23 illustrates a second modified example of the first and second flow velocity adjustment sections of the cooler according to the third embodiment. Figure 23 schematically shows a plan view of the main parts of the second modified example of the first and second flow velocity adjustment sections of the cooler.

[0126] The first velocity adjustment section 15 and the second velocity adjustment section 16 shown in Figure 23 are provided with holes instead of slits as openings. The first velocity adjustment section 15 shown in Figure 23 has a configuration in which, of the three regions divided in the longitudinal direction, a central first region 15a is provided with a plurality of first holes 15ab having a first diameter d1, and each of the two outer second regions 15b is provided with a plurality of second holes 15bb having a second diameter d2 smaller than the first diameter d1. The second velocity adjustment section 16 shown in Figure 23 has a configuration in which, of the three regions divided in the longitudinal direction, a central third region 16a is provided with a plurality of third holes 16ab having a third diameter d3, and each of the two outer fourth regions 16b is provided with a plurality of fourth holes 16bb having a fourth diameter d4 larger than the third diameter d3. As shown in Figure 23, the first flow velocity adjustment section 15 and the second flow velocity adjustment section 16 are arranged to cover the first flow path 14e and the second flow path 14f of the container 14, respectively. The first region 15a of the first flow velocity adjustment section 15, which has a relatively large opening ratio, and the third region 16a of the second flow velocity adjustment section 16, which has a relatively small opening ratio, face each other, and the second region 15b of the first flow velocity adjustment section 15, which has a relatively small opening ratio, and the fourth region 16b of the second flow velocity adjustment section 16, which has a relatively large opening ratio, face each other.

[0127] A cooler 10 using a first flow velocity adjustment unit 15 and a second flow velocity adjustment unit 16 as shown in Figure 23, that is, a cooler 10 in which the first flow velocity adjustment unit 15 and the second flow velocity adjustment unit 16 are arranged in the first flow path 14e and the second flow path 14f of the container 14, respectively, can also suppress the occurrence of uneven flow distribution of the refrigerant 30 circulating inside the cooler 10 and the increase in pressure loss.

[0128] Furthermore, in the first flow velocity adjustment section 15, if the aperture ratio of the first region 15a is greater than that of the second region 15b, the number of first holes 15ab in the first region 15a and the number of second holes 15bb in the second region 15b are not limited to those shown in the figure. If the aperture ratio of the first region 15a is greater than that of the second region 15b, the first diameter d1 of each of the multiple first holes 15ab may be the same or different from each other, and the second diameter d2 of each of the multiple second holes 15bb may be the same or different from each other. The multiple first holes 15ab may be arranged in multiple rows, not just one, and the multiple second holes 15bb may be arranged in multiple rows, not just one.

[0129] Furthermore, in the second flow velocity adjustment section 16, if the aperture ratio of the third region 16a is smaller than that of the fourth region 16b, the number of third holes 16ab in the third region 16a and the number of fourth holes 16bb in the fourth region 16b are not limited to those shown in the figure. If the aperture ratio of the third region 16a is smaller than that of the fourth region 16b, the third diameter d3 of each of the multiple third holes 16ab may be the same or different from each other, and the fourth diameter d4 of each of the multiple fourth holes 16bb may be the same or different from each other. The multiple third holes 16ab may be arranged in multiple rows, not just one row, and the multiple fourth holes 16bb may be arranged in multiple rows, not just one row.

[0130] Furthermore, the first diameter d1 of the first hole 15ab of the first flow velocity adjustment section 15 and the fourth diameter d4 of the fourth hole 16bb of the second flow velocity adjustment section 16 may be the same or different from each other, and the second diameter d2 of the second hole 15bb of the first flow velocity adjustment section 15 and the third diameter d3 of the third hole 16ab of the second flow velocity adjustment section 16 may be the same or different from each other.

[0131] Figure 24 illustrates a third modified example of the first and second flow velocity adjustment sections of the cooler according to the third embodiment. Figure 24 schematically shows a plan view of the main parts of the third modified example of the first and second flow velocity adjustment sections of the cooler.

[0132] The first velocity adjustment section 15 shown in Figure 24 has a configuration in which a fifth slit 15ac is provided, which narrows in width from the central part 15c in the longitudinal direction toward both ends 15d. The first velocity adjustment section 15 shown in Figure 24 can also be described as having a configuration in which a fifth slit 15ac is provided, which narrows in width from the central first region 15a toward the two outer second regions 15b of a group of regions divided into three in the longitudinal direction. The second velocity adjustment section 16 shown in Figure 24 has a configuration in which a sixth slit 16ac is provided, which widens in width from the central part 16c in the longitudinal direction toward both ends 16d. The second velocity adjustment section 16 shown in Figure 24 can also be described as having a configuration in which a sixth slit 16ac is provided, which widens in width from the central third region 16a toward the two outer fourth regions 16b of a group of regions divided into three in the longitudinal direction. As shown in Figure 24, the first flow velocity adjustment section 15 and the second flow velocity adjustment section 16 are arranged to cover the first flow path 14e and the second flow path 14f of the container 14, respectively. The first region 15a of the first flow velocity adjustment section 15, which has a relatively large opening ratio, and the third region 16a of the second flow velocity adjustment section 16, which has a relatively small opening ratio, face each other, and the second region 15b of the first flow velocity adjustment section 15, which has a relatively small opening ratio, and the fourth region 16b of the second flow velocity adjustment section 16, which has a relatively large opening ratio, face each other.

[0133] A cooler 10 using a first flow velocity adjustment unit 15 and a second flow velocity adjustment unit 16 as shown in Figure 24, that is, a cooler 10 in which the first flow velocity adjustment unit 15 and the second flow velocity adjustment unit 16 are arranged in the first flow path 14e and the second flow path 14f of the container 14, respectively, can also suppress the occurrence of uneven flow distribution of the refrigerant 30 circulating inside the cooler 10 and the increase in pressure loss.

[0134] Furthermore, the fifth slit 15ac of the first flow velocity adjustment section 15 may be divided into multiple slits at the boundary between the first region 15a and the second region 15b, or it may be divided into multiple parts within each of the first region 15a and the second region 15b, as shown in the example in Figure 22 above.

[0135] Furthermore, the sixth slit 16ac of the second flow velocity adjustment section 16 may be divided into multiple slits at the boundary between the third region 16a and the fourth region 16b, or it may be divided into multiple parts within each of the third region 16a and the fourth region 16b, as shown in the example in Figure 22 above.

[0136] Furthermore, the width of the central portion 15c of the fifth slit 15ac of the first flow velocity adjustment section 15 and the width of the end portion 16d of the sixth slit 16ac of the second flow velocity adjustment section 16 may be the same or different from each other, and the width of the end portion 15d of the fifth slit 15ac of the first flow velocity adjustment section 15 and the width of the central portion 16c of the sixth slit 16ac of the second flow velocity adjustment section 16 may be the same or different from each other.

[0137] [Fourth Embodiment] Here, the evaluation results obtained from thermal fluid simulations using various cooler configurations are described as the fourth embodiment.

[0138] <Example 1> Figure 25 illustrates a first example of a cooler according to the fourth embodiment. Figure 25(A) schematically shows a perspective view of the main part of the first example cooler and the layout of the semiconductor element mounting area. Figures 25(B) to 25(F) schematically show plan views of the main part of the flow velocity adjustment section applied to the first example cooler.

[0139] In the first example, a container 14 as shown in Figure 25(A) is used for the cooler 10. The container 14 shown in Figure 25(A) corresponds to the one shown in Figure 4 above. The container 14 shown in Figure 25(A) has an inlet 11 (IN) in the center of the first side wall 14a that communicates with the first flow path 14e, and an outlet 12 (OUT) in the center of the second side wall 14b that communicates with the second flow path 14f. The cooling fins 13a of the heat dissipation plate 13 that covers the container 14 are housed in the third flow path 14g, which is an internal space above the first flow path 14e and the second flow path 14f. In the thermal fluid simulation, the cooling fins 13a are either prismatic in shape as shown in Figures 3(A) and 3(B) above, or cylindrical as shown in Figures 15(A) and 15(B) above. Then, in the region on the heat sink 13 corresponding to the third flow path 14g (the region shown by the dotted line frame in Figure 25(A)), semiconductor elements CP1 and CP2 are arranged in each of the three mounting regions AR1, AR2, and AR3, as shown in Figure 25(A), following the example in Figure 1 and so on.

[0140] In Figure 25(A) (and Figures 25(B) to 25(F) described later), the inlet 11 side of the container 14 is represented as "IN," and the outlet 12 side is represented as "OUT." The three mounting regions AR1-AR3, and the semiconductor elements CP1 and CP2 provided in each of them, are positioned relative to the IN and OUT of the container 14 as shown in Figure 25(A).

[0141] In the thermal fluid simulation, the cooler 10 shown in Figure 25(A) is equipped with a first velocity adjustment unit 115 and a second velocity adjustment unit 116 as shown in Figure 25(B), and a first velocity adjustment unit 15 and a second velocity adjustment unit 16 as shown in Figures 25(C) to 25(F).

[0142] Here, the first flow velocity adjustment section 115 and the second flow velocity adjustment section 116 shown in Figure 25(B) are referred to as "SL1". SL1 corresponds to the first flow velocity adjustment section 115 and the second flow velocity adjustment section 116 shown in Figure 10. The first flow velocity adjustment section 115 and the second flow velocity adjustment section 116 shown in Figure 25(B) each have a slit 115e (seventh slit) and a slit 116e (eighth slit) of constant width extending in the longitudinal direction. The widths of slits 115e and 116e are set to 1 mm.

[0143] The first flow velocity adjustment section 15 and the second flow velocity adjustment section 16 shown in Figure 25(C) are referred to as "SL2". SL2 corresponds to the first flow velocity adjustment section 15 and the second flow velocity adjustment section 16 shown in Figure 5 above. The first flow velocity adjustment section 15 shown in Figure 25(C) adjusts the width of the slit 15e so that the opening ratio of the central region (first region) closest to the inlet 11 (IN), one of the three regions divided in the longitudinal direction, is greater than the opening ratio of the regions on both sides (second regions). The width of the slit 15e (first slit) in the central region closest to the inlet 11 is set to 2 mm, and the width of the slits 15e (second slit) in the regions on both sides is set to 1 mm. Furthermore, the second flow velocity adjustment section 16 shown in Figure 25(C) adjusts the width of the slit 16e so that the opening ratio of the central region (third region) closest to the outlet 12 (OUT) is smaller than the opening ratios of the regions on both sides (fourth region), out of the three regions divided in the longitudinal direction. The width of the slit 16e (third slit) in the central region closest to the outlet 12 is set to 1 mm, and the width of the slits 16e (fourth slit) in the regions on both sides is set to 2 mm.

[0144] The first velocity adjustment section 15 and the second velocity adjustment section 16 shown in Figure 25(D) are referred to as "SL3". SL3 corresponds to the first velocity adjustment section 15 and the second velocity adjustment section 16 shown in Figure 22. The first velocity adjustment section 15 shown in Figure 25(D) has a slit 15f which is obtained by dividing the slit 15e in Figure 25(C) into two parts in each of the three regions obtained by dividing the first velocity adjustment section 15 in the longitudinal direction. The second velocity adjustment section 16 shown in Figure 25(D) has a slit 16f which is obtained by dividing the slit 16e in Figure 25(C) into two parts in each of the three regions obtained by dividing the second velocity adjustment section 16 in the longitudinal direction.

[0145] The first flow velocity adjustment section 15 and the second flow velocity adjustment section 16 shown in Figure 25(E) are referred to as "SL4". SL4 corresponds to the first flow velocity adjustment section 15 and the second flow velocity adjustment section 16 shown in Figure 23. The first flow velocity adjustment section 15 shown in Figure 25(E) is configured such that the diameter of the hole 15g is adjusted so that the opening ratio of the central region (first region) closest to the inlet 11 (IN), one of the three regions divided in the longitudinal direction, is greater than the opening ratios of the regions on both sides (second regions). The diameter of the hole 15g (first hole) in the central region closest to the inlet 11 is set to 2 mm, and the diameter of the holes 15g (second holes) in the regions on both sides is set to 1 mm. Furthermore, the second flow velocity adjustment section 16 shown in Figure 25(E) adjusts the diameter of the holes 16g so that the opening ratio of the central region (third region) closest to the outlet 12 (OUT), one of the three regions divided in the longitudinal direction, is smaller than the opening ratios of the regions on both sides (fourth region). The diameter of the hole 16g (third hole) in the central region closest to the outlet 12 is set to 1 mm, and the diameter of the holes 16g (fourth hole) in the regions on both sides is set to 2 mm.

[0146] The first flow velocity adjustment section 15 and the second flow velocity adjustment section 16 shown in Figure 25(F) are referred to as "SL5". SL5 corresponds to the first flow velocity adjustment section 15 and the second flow velocity adjustment section 16 shown in Figure 24. The first flow velocity adjustment section 15 shown in Figure 25(F) is adjusted so that the opening ratio of the central region (first region) closest to the inlet 11 (IN) is greater than the opening ratios of the regions on both sides (second region), that is, the width of the slit 15h (fifth slit) narrows from the center toward both sides. The width of the slit 15h in the center is set to 2 mm, and the widths at both ends are set to 1 mm. Furthermore, the second flow velocity adjustment section 16 shown in Figure 25(F) is adjusted so that the opening ratio of the central region (third region) closest to the outlet 12 (OUT) among the three regions divided in the longitudinal direction is smaller than the opening ratios of the regions on both sides (fourth region), that is, the width of the slit 16h (sixth slit) widens from the center toward both sides. The width of the slit 16h in the center is set to 1 mm, and the widths at both ends are set to 2 mm.

[0147] In the thermal fluid simulation, SL1-SL5 shown in Figures 25(B) to 25(F) are applied to the container 14 of the cooler 10 as shown in Figure 25(A). For each case, the pressure loss between the inlet 11 and outlet 12, the refrigerant flow velocity at the positions of semiconductor elements CP1 and CP2 in the mounting region AR1-AR3, and the temperatures of semiconductor elements CP1 and CP2 are determined when a rectangular or cylindrical cooling fin 13a is applied to the heat sink 13. For comparison, the pressure loss between the inlet 11 and outlet 12, the refrigerant flow velocity at the positions of semiconductor elements CP1 and CP2 in the mounting region AR1-AR3, and the temperatures of semiconductor elements CP1 and CP2 are also determined when a rectangular or cylindrical cooling fin 13a is applied to the container 14 of the cooler 10 as shown in Figure 25(A). In the thermal fluid simulation, heat generation is reproduced by introducing a certain loss to the semiconductor elements CP1 and CP2 in the mounting region AR1-AR3. The evaluation results from the thermal fluid simulation are shown in Figures 26 and 27.

[0148] Figure 26 shows the evaluation results of a thermal fluid simulation for the first example of a cooler using prismatic cooling fins. Figure 26(A) shows an example of the evaluation results of pressure loss in the cooler. Figure 26(B) shows an example of the evaluation results of refrigerant flow velocity with respect to semiconductor element position. Figure 26(C) shows an example of the evaluation results of semiconductor element temperature with respect to semiconductor element position. In Figures 26(A) to 26(C), the flow velocity adjustment sections (first and second flow velocity adjustment sections) applied to the cooler container are represented as "SL1-SL5" (Figures 25(B)-25(F)), and the case without a flow velocity adjustment section is represented as "None".

[0149] As shown in Figure 26(A), the pressure loss of the cooler 10 increases by 90.1% when SL1 is applied, 44.3% when SL2 is applied, 54.2% when SL3 is applied, 81.6% when SL4 is applied, and 22.9% when SL5 is applied, compared to the case without a flow velocity adjustment unit (pressure loss shown by the dotted line L1 in Figure 26(A)). On the other hand, the pressure loss of the cooler 10 decreases by 24.1% when SL2 is applied, 18.9% when SL3 is applied, 4.5% when SL4 is applied, and 35.4% when SL5 is applied, compared to SL1 with a constant slit width (pressure loss shown by the dashed line L2 in Figure 26(A)). Therefore, when SL2-SL5 are applied, the increase in pressure loss compared to the case without a flow velocity adjustment unit is suppressed compared to when SL1 is applied.

[0150] As shown in Figure 26(B), without a flow velocity adjustment unit, the refrigerant flow velocity at the positions of semiconductor elements CP1 and CP2 in the central mounting region AR2 is faster than the refrigerant flow velocity at the positions of semiconductor elements CP1 and CP2 in the mounting regions AR1 and AR3 at both ends, resulting in a flow bias distribution. On the other hand, when SL1-SL5 is applied, the refrigerant flow velocity at the positions of semiconductor elements CP1 and CP2 in the mounting regions AR1-AR3 is kept relatively constant compared to the case without a flow velocity adjustment unit, resulting in a more uniform flow.

[0151] As shown in Figure 26(C), without the flow rate adjustment unit, the temperatures of semiconductor elements CP1 and CP2 in the central mounting region AR2, where the refrigerant flow rate is relatively high, are lower, while the temperatures of semiconductor elements CP1 and CP2 in the mounting regions AR1 and AR3 at both ends, where the refrigerant flow rate is relatively slow, are higher. On the other hand, when SL1-SL5 is applied, the temperatures of semiconductor elements CP1 and CP2 in mounting regions AR1-AR3 are kept relatively constant and cooled more uniformly compared to the case without the flow rate adjustment unit.

[0152] From the results in Figures 26(A) to 26(C), it can be said that in the cooler 10 of Figure 25(A) with prismatic cooling fins, when SL1-SL5 are applied, a superior flow distribution suppression effect and semiconductor device cooling effect can be obtained compared to the case without a flow velocity adjustment section. Furthermore, in the cooler 10 of Figure 25(A) with prismatic cooling fins, when SL2-SL5 are applied, it can be said that a flow distribution suppression effect and semiconductor device cooling effect equivalent to or close to that when SL1 is applied can be obtained while suppressing the increase in pressure loss compared to when SL1 is applied.

[0153] Figure 27 shows the evaluation results of a thermal fluid simulation for the first example of a cooler with cylindrical cooling fins. Figure 27(A) shows an example of the evaluation results of pressure loss in the cooler. Figure 27(B) shows an example of the evaluation results of refrigerant flow velocity with respect to semiconductor element position. Figure 27(C) shows an example of the evaluation results of semiconductor element temperature with respect to semiconductor element position. In Figures 27(A) to 27(C), the flow velocity adjustment sections (first and second flow velocity adjustment sections) applied to the cooler container are represented as "SL1-SL5" (Figures 25(B)-25(F)), and the case where no flow velocity adjustment section is applied is represented as "None".

[0154] As shown in Figure 27(A), the pressure loss of the cooler 10 increases by 86.4% when SL1 is applied, 42.4% when SL2 is applied, 52.0% when SL3 is applied, 69.6% when SL4 is applied, and 20.4% when SL5 is applied, compared to the case without a flow velocity adjustment unit (pressure loss shown by the dotted line L1 in Figure 27(A)). On the other hand, the pressure loss of the cooler 10 decreases by 23.6% when SL2 is applied, 18.5% when SL3 is applied, 9.0% when SL4 is applied, and 35.4% when SL5 is applied, compared to SL1 with a constant slit width (pressure loss shown by the dashed line L2 in Figure 27(A)). Therefore, when SL2-SL5 are applied, the increase in pressure loss compared to the case without a flow velocity adjustment unit is suppressed compared to when SL1 is applied.

[0155] As shown in Figure 27(B), without a flow velocity adjustment unit, the refrigerant flow velocity at the positions of semiconductor elements CP1 and CP2 in the central mounting region AR2 is faster than the refrigerant flow velocity at the positions of semiconductor elements CP1 and CP2 in the mounting regions AR1 and AR3 at both ends, resulting in a flow bias distribution. On the other hand, when SL1-SL5 is applied, the refrigerant flow velocity at the positions of semiconductor elements CP1 and CP2 in the mounting regions AR1-AR3 is kept relatively constant compared to the case without a flow velocity adjustment unit, resulting in a more uniform flow.

[0156] As shown in Figure 27(C), without the flow rate adjustment unit, the temperatures of semiconductor elements CP1 and CP2 in the central mounting region AR2, where the refrigerant flow rate is relatively high, are lower, while the temperatures of semiconductor elements CP1 and CP2 in the mounting regions AR1 and AR3 at both ends, where the refrigerant flow rate is relatively slow, are higher. On the other hand, when SL1-SL5 is applied, the temperatures of semiconductor elements CP1 and CP2 in mounting regions AR1-AR3 are kept relatively constant and cooled more uniformly compared to the case without the flow rate adjustment unit.

[0157] From the results in Figures 27(A) to 27(C), it can be said that in the cooler 10 of Figure 25(A) with cylindrical cooling fins, when SL1-SL5 are applied, a superior flow distribution suppression effect and semiconductor device cooling effect can be obtained compared to the case without a flow velocity adjustment section. Furthermore, in the cooler 10 of Figure 25(A) with cylindrical cooling fins, when SL2-SL5 are applied, it can be said that a flow distribution suppression effect and semiconductor device cooling effect equivalent to or close to that when SL1 is applied can be obtained while suppressing the increase in pressure loss compared to when SL1 is applied.

[0158] <Example 2> Figure 28 illustrates a second example of a cooler according to the fourth embodiment. Figure 28(A) schematically shows a perspective view of the main part of the second example cooler and the layout of the semiconductor element mounting area. Figures 28(B) to 28(F) schematically show plan views of the main part of the flow velocity adjustment section applied to the second example cooler.

[0159] In the second example, a container 14 as shown in Figure 28(A) is used for the cooler 10. The container 14 shown in Figure 28(A) corresponds to the one shown in Figure 18. The container 14 shown in Figure 28(A) has an inlet 11 (IN) communicating with the first flow path 14e and an outlet 12 (OUT) communicating with the second flow path 14f on its third side wall 14c. The cooling fins 13a of the heat sink 13 covering the container 14 are housed in the third flow path 14g, which is an internal space above the first flow path 14e and the second flow path 14f. In the thermal fluid simulation, a rectangular prism-shaped cooling fin 13a as shown in Figures 3(A) and 3(B), or a cylindrical cooling fin 13a as shown in Figures 15(A) and 15(B) is used. Then, in the region on the heat sink 13 corresponding to the third flow path 14g (the region shown by the dotted line frame in Figure 28(A)), semiconductor elements CP1 and CP2 are arranged in each of the three mounting regions AR1, AR2, and AR3, as shown in Figure 28(A), following the example in Figure 1 and so on.

[0160] In Figure 28(A) (and Figures 28(B) to 28(F) described later), the inlet 11 side of the container 14 is represented as "IN," and the outlet 12 side is represented as "OUT." The three mounting regions AR1-AR3, and the semiconductor elements CP1 and CP2 provided in each of them, are positioned relative to the IN and OUT of the container 14 as shown in Figure 28(A).

[0161] In the thermal fluid simulation, the cooler 10 shown in Figure 28(A) is equipped with a first velocity adjustment unit 115 and a second velocity adjustment unit 116 as shown in Figure 28(B), and a first velocity adjustment unit 15 and a second velocity adjustment unit 16 as shown in Figures 28(C) to 28(F).

[0162] Here, the first flow velocity adjustment section 115 and the second flow velocity adjustment section 116 shown in Figure 28(B) are referred to as "SL1". SL1 corresponds to the first flow velocity adjustment section 115 and the second flow velocity adjustment section 116 shown in Figure 10. The first flow velocity adjustment section 115 and the second flow velocity adjustment section 116 shown in Figure 28(B) each have a slit 115e (7th slit) and a slit 116e (8th slit) of constant width extending in the longitudinal direction. The widths of slits 115e and 116e are set to 1 mm.

[0163] The first and second velocity adjustment sections 15 and 16 shown in Figure 28(C) are denoted as "SL2". SL2 is a modified version of the opening layout of the first and second velocity adjustment sections 15 and 16 shown in Figure 5. The first velocity adjustment section 15 shown in Figure 28(C) has its slit width 15i adjusted so that the opening ratio of the region closest to the inlet 11 (IN) (first region) is greater than the opening ratios of the remaining two regions (second regions) among the three regions divided in the longitudinal direction. The width of the slit 15i (first slit) in the region closest to the inlet 11 is set to 2 mm, and the width of the slits 15i (second slits) in the remaining regions is set to 1 mm. Furthermore, the second flow velocity adjustment section 16 shown in Figure 28(C) adjusts the width of the slit 16i so that the opening ratio of the region closest to the outlet 12 (OUT) (third region) among the three regions divided in the longitudinal direction is smaller than the opening ratio of the remaining two regions (fourth region). The width of the slit 16i (third slit) in the region closest to the outlet 12 is set to 1 mm, and the width of the slit 16i (fourth slit) in the remaining regions is set to 2 mm.

[0164] The first velocity adjustment section 15 and the second velocity adjustment section 16 shown in Figure 28(D) are referred to as "SL3". SL3 is a modified version of the opening layout of the first velocity adjustment section 15 and the second velocity adjustment section 16 shown in Figure 22. The first velocity adjustment section 15 shown in Figure 28(D) has a slit 15j which is obtained by dividing the slit 15i in Figure 28(C) into two parts in each of the three regions obtained by dividing the first velocity adjustment section 15 in the longitudinal direction. The second velocity adjustment section 16 shown in Figure 28(D) has a slit 16j which is obtained by dividing the slit 16i in Figure 28(C) into two parts in each of the three regions obtained by dividing the second velocity adjustment section 16 in the longitudinal direction.

[0165] The first velocity adjustment section 15 and the second velocity adjustment section 16 shown in Figure 28(E) are referred to as "SL4". SL4 is a modified version of the opening layout of the first velocity adjustment section 15 and the second velocity adjustment section 16 shown in Figure 23. In the first velocity adjustment section 15 shown in Figure 28(E), the diameter of the hole 15k is adjusted so that the opening ratio of the region closest to the inlet 11 (IN) (first region) is greater than the opening ratio of the remaining two regions (second regions) among the three regions divided in the longitudinal direction. The diameter of the hole 15k (first hole) in the region closest to the inlet 11 is set to 2 mm, and the diameter of the holes 15k (second holes) in the remaining regions is set to 1 mm. Furthermore, the second flow velocity adjustment section 16 shown in Figure 28(E) adjusts the diameter of the holes 16k so that the opening ratio of the region closest to the outlet 12 (OUT) (third region) among the three regions divided in the longitudinal direction is smaller than the opening ratio of the remaining two regions (fourth region). The diameter of the hole 16k (third hole) in the region closest to the outlet 12 is set to 1 mm, and the diameter of the holes 16k (fourth hole) in the remaining regions is set to 2 mm.

[0166] The first flow velocity adjustment section 15 and the second flow velocity adjustment section 16 shown in Figure 28(F) are referred to as "SL5". SL5 is a modified version of the opening layout of the first flow velocity adjustment section 15 and the second flow velocity adjustment section 16 shown in Figure 24. The first flow velocity adjustment section 15 shown in Figure 28(F) is adjusted so that the opening ratio in the region closer to the inlet 11 (IN) (first region) is greater than the opening ratio in the region further away from the inlet 11 (second region), that is, the width of the slit 15m (fifth slit) becomes narrower as it moves away from the inlet 11. The width of one end of the slit 15m on the inlet 11 side is set to 2 mm, and the width of the other end is set to 1 mm. Furthermore, the second flow velocity adjustment section 16 shown in Figure 28(F) is adjusted so that the opening ratio in the region closer to the outlet 12 (OUT) (third region) is smaller than the opening ratio in the region further away from the outlet 12 (fourth region), that is, the width of the slit 16m (sixth slit) widens as it moves away from the outlet 12. The width of one end of the slit 16m on the outlet 12 side is set to 1 mm, and the width of the other end is set to 2 mm.

[0167] In the thermal fluid simulation, SL1-SL5 shown in Figures 28(B) to 28(F) are applied to the container 14 of the cooler 10 as shown in Figure 28(A). For each case, the pressure loss between the inlet 11 and outlet 12, the refrigerant flow velocity at the positions of semiconductor elements CP1 and CP2 in the mounting region AR1-AR3, and the temperatures of semiconductor elements CP1 and CP2 are determined when a rectangular or cylindrical cooling fin 13a is applied to the heat sink 13. For comparison, the pressure loss between the inlet 11 and outlet 12, the refrigerant flow velocity at the positions of semiconductor elements CP1 and CP2 in the mounting region AR1-AR3, and the temperatures of semiconductor elements CP1 and CP2 are also determined when a rectangular or cylindrical cooling fin 13a is applied to the container 14 of the cooler 10 as shown in Figure 28(A). In the thermal fluid simulation, heat generation is reproduced by introducing a certain loss to the semiconductor elements CP1 and CP2 in the mounting region AR1-AR3. The evaluation results from the thermal fluid simulation are shown in Figures 29 and 30.

[0168] Figure 29 shows the evaluation results of a thermal fluid simulation for a second example of a cooler using prismatic cooling fins. Figure 29(A) shows an example of the evaluation results of pressure loss in the cooler. Figure 29(B) shows an example of the evaluation results of refrigerant flow velocity with respect to semiconductor element position. Figure 29(C) shows an example of the evaluation results of semiconductor element temperature with respect to semiconductor element position. In Figures 29(A) to 29(C), the flow velocity adjustment sections (first and second flow velocity adjustment sections) applied to the cooler container are represented as "SL1-SL5" (Figures 28(B)-28(F)), and the case without a flow velocity adjustment section is represented as "None".

[0169] As shown in Figure 29(A), the pressure loss of the cooler 10 increases by 153.2% when SL1 is applied, by 96.7% when SL2 is applied, by 104.2% when SL3 is applied, by 128.2% when SL4 is applied, and by 42.5% when SL5 is applied, compared to the case without a flow velocity adjustment unit (pressure loss shown by the dotted line L1 in Figure 29(A)). On the other hand, the pressure loss of the cooler 10 decreases by 22.3% when SL2 is applied, by 19.4% when SL3 is applied, by 9.9% when SL4 is applied, and by 43.7% when SL5 is applied, compared to SL1 with a constant slit width (pressure loss shown by the dashed line L2 in Figure 29(A)). Therefore, when SL2-SL5 are applied, the increase in pressure loss compared to the case without a flow velocity adjustment unit is suppressed compared to when SL1 is applied.

[0170] As shown in Figure 29(B), without the flow velocity adjustment unit, the refrigerant flow velocity at the locations of semiconductor elements CP1 and CP2 in mounting region AR1 is faster than the refrigerant flow velocity at the locations of semiconductor elements CP1 and CP2 in mounting regions AR2 and AR3, resulting in a flow bias distribution. On the other hand, when SL1-SL5 is applied, the flow bias distribution of the refrigerant at the locations of semiconductor elements CP1 and CP2 in mounting regions AR1-AR3 is suppressed compared to the case without the flow velocity adjustment unit, resulting in a more uniform flow.

[0171] As shown in Figure 29(C), without the flow rate adjustment unit, the temperatures of semiconductor elements CP1 and CP2 in mounting region AR1, where the refrigerant flow rate is relatively high, are lower, while the temperatures of semiconductor elements CP1 and CP2 in mounting regions AR2 and AR3, where the refrigerant flow rate is relatively slow, are higher. On the other hand, when SL1-SL5 is applied, the temperatures of semiconductor elements CP1 and CP2 in mounting regions AR1-AR3 are kept relatively constant and cooled more uniformly compared to the case without the flow rate adjustment unit.

[0172] From the results in Figures 29(A) to 29(C), it can be said that in the cooler 10 of Figure 28(A) with prismatic cooling fins, when SL1-SL5 are applied, a superior flow distribution suppression effect and semiconductor device cooling effect can be obtained compared to the case without a flow velocity adjustment section. Furthermore, in the cooler 10 of Figure 28(A) with prismatic cooling fins, when SL2-SL5 are applied, it can be said that a flow distribution suppression effect and semiconductor device cooling effect equivalent to or close to that when SL1 is applied can be obtained while suppressing the increase in pressure loss compared to when SL1 is applied.

[0173] Figure 30 shows the evaluation results of a thermal fluid simulation for a second example of a cooler with cylindrical cooling fins. Figure 30(A) shows an example of the evaluation results of pressure loss in the cooler. Figure 30(B) shows an example of the evaluation results of refrigerant flow velocity with respect to semiconductor element position. Figure 30(C) shows an example of the evaluation results of semiconductor element temperature with respect to semiconductor element position. In Figures 30(A) to 30(C), the flow velocity adjustment sections (first and second flow velocity adjustment sections) applied to the cooler container are represented as "SL1-SL5" (Figures 28(B)-28(F)), and the case where no flow velocity adjustment section is applied is represented as "None".

[0174] As shown in Figure 30(A), the pressure loss of the cooler 10 increases by 176.5% when SL1 is applied, by 98.5% when SL2 is applied, by 105.4% when SL3 is applied, by 114.1% when SL4 is applied, and by 35.1% when SL5 is applied, compared to the case without a flow velocity adjustment unit (pressure loss shown by the dotted line L1 in Figure 30(A)). On the other hand, the pressure loss of the cooler 10 decreases by 28.2% when SL2 is applied, by 25.7% when SL3 is applied, by 22.6% when SL4 is applied, and by 51.1% when SL5 is applied, compared to SL1 with a constant slit width (pressure loss shown by the dashed line L2 in Figure 30(A)). Therefore, when SL2-SL5 are applied, the increase in pressure loss compared to the case without a flow velocity adjustment unit is suppressed compared to when SL1 is applied.

[0175] As shown in Figure 30(B), without the flow velocity adjustment unit, the refrigerant flow velocity at the locations of semiconductor elements CP1 and CP2 in mounting region AR3 is slower than the refrigerant flow velocity at the locations of semiconductor elements CP1 and CP2 in mounting regions AR1 and AR2, resulting in a flow bias distribution. On the other hand, when SL1-SL5 is applied, the flow bias distribution of the refrigerant at the locations of semiconductor elements CP1 and CP2 in mounting regions AR1-AR3 is suppressed compared to the case without the flow velocity adjustment unit, resulting in a more uniform flow.

[0176] As shown in Figure 30(C), without the flow rate adjustment unit, the temperatures of semiconductor elements CP1 and CP2 in the mounting region AR1-AR3 become relatively high. On the other hand, when SL1-SL5 is applied, the temperatures of semiconductor elements CP1 and CP2 in the mounting region AR1-AR3 are kept relatively constant and cooled more uniformly compared to the case without the flow rate adjustment unit.

[0177] From the results in Figures 30(A) to 30(C), it can be said that in the cooler 10 of Figure 28(A) with cylindrical cooling fins, when SL1-SL5 are applied, a superior flow distribution suppression effect and semiconductor device cooling effect can be obtained compared to the case without a flow velocity adjustment section. Furthermore, in the cooler 10 of Figure 28(A) with cylindrical cooling fins, when SL2-SL5 are applied, it can be said that a flow distribution suppression effect and semiconductor device cooling effect equivalent to or close to that when SL1 is applied can be obtained while suppressing the increase in pressure loss compared to when SL1 is applied.

[0178] <Example 3> Figure 31 illustrates a third example of a cooler according to the fourth embodiment. Figure 31(A) schematically shows a perspective view of the main part of the third example cooler and the layout of the semiconductor element mounting area. Figures 31(B) to 31(F) schematically show plan views of the main part of the flow velocity adjustment section applied to the third example cooler.

[0179] In the third example, a container 14 as shown in Figure 31(A) is used for the cooler 10. The container 14 shown in Figure 31(A) is a modified version of the one shown in Figure 19. The container 14 shown in Figure 31(A) has an inlet 11 (IN) on the third side wall 14c that communicates with the first flow path 14e, and an outlet 12 (OUT) on the fourth side wall 14d that communicates with the second flow path 14f. The cooling fins 13a of the heat dissipation plate 13 that covers the container 14 are housed in the third flow path 14g, which is an internal space above the first flow path 14e and the second flow path 14f. In the thermal fluid simulation, a rectangular prism-shaped cooling fin 13a as shown in Figures 3(A) and 3(B), or a cylindrical cooling fin 13a as shown in Figures 15(A) and 15(B) is used. Then, in the region on the heat sink 13 corresponding to the third flow path 14g (the region shown by the dotted line frame in Figure 31(A)), semiconductor elements CP1 and CP2 are arranged in each of the three mounting regions AR1, AR2, and AR3, respectively, as shown in Figure 31(A), following the example in Figure 1 and so on.

[0180] In Figure 31(A) (and Figures 31(B) to 31(F) described later), the inlet 11 side of the container 14 is represented as "IN," and the outlet 12 side is represented as "OUT." The three mounting regions AR1-AR3, and the semiconductor elements CP1 and CP2 provided in each of them, are positioned relative to the IN and OUT of the container 14 as shown in Figure 31(A).

[0181] In the thermal fluid simulation, the cooler 10 shown in Figure 31(A) is used, along with the first velocity adjustment unit 115 and the second velocity adjustment unit 116 shown in Figure 31(B), and the first velocity adjustment unit 15 and the second velocity adjustment unit 16 shown in Figures 31(C) to 31(F).

[0182] Here, the first flow velocity adjustment section 115 and the second flow velocity adjustment section 116 shown in Figure 31(B) are referred to as "SL1". SL1 corresponds to the first flow velocity adjustment section 115 and the second flow velocity adjustment section 116 shown in Figure 10. The first flow velocity adjustment section 115 and the second flow velocity adjustment section 116 shown in Figure 31(B) each have a slit 115e (seventh slit) and a slit 116e (eighth slit) of constant width extending in the longitudinal direction. The widths of slits 115e and 116e are set to 1 mm.

[0183] The first and second velocity adjustment sections 15 and 16 shown in Figure 31(C) are denoted as "SL2". SL2 is a modified version of the opening layout of the first and second velocity adjustment sections 15 and 16 shown in Figure 5. The first velocity adjustment section 15 shown in Figure 31(C) has its slit width 15n adjusted so that the opening ratio of the region closest to the inlet 11 (IN) (first region) is greater than the opening ratios of the remaining two regions (second regions) among the three regions divided in the longitudinal direction. The slit 15n (first slit) in the region closest to the inlet 11 has sections of different widths, with the wide section having a width of 3 mm and the narrow section having a width of 2 mm. The width of the slit 15n (second slit) in the remaining region is set to 1 mm. Furthermore, the second flow velocity adjustment section 16 shown in Figure 31(C) adjusts the width of the slit 16n such that the opening ratio of the region furthest from the outlet 12 (OUT) (the fourth region) among the three regions divided in the longitudinal direction is greater than the opening ratio of the remaining two regions (the third region). The slit 16n (fourth slit) in the region furthest from the outlet 12 has parts of different widths, with the wide part having a width of 3 mm and the narrow part having a width of 2 mm. The width of the slit 16n (third slit) in the remaining region is set to 1 mm.

[0184] The first flow velocity adjustment section 15 and the second flow velocity adjustment section 16 shown in Figure 31(D) are referred to as "SL3". SL3 is a modified version of the opening layout of the first flow velocity adjustment section 15 and the second flow velocity adjustment section 16 shown in Figure 22. The first flow velocity adjustment section 15 shown in Figure 31(D) is a slit 15p, in which the slit 15n in Figure 31(C) is divided into two parts in each of the three regions obtained by dividing the first flow velocity adjustment section 15 in the longitudinal direction (the region at the end closest to the inlet 11 is divided into a wide section and a narrow section). The second flow velocity adjustment section 16 shown in Figure 31(D) is a slit 16p, in which the slit 16n in Figure 31(C) is divided into two parts in each of the three regions obtained by dividing the second flow velocity adjustment section 16 in the longitudinal direction (the region at the end furthest from the outlet 12 is divided into a wide section and a narrow section).

[0185] The first velocity adjustment section 15 and the second velocity adjustment section 16 shown in Figure 31(E) are referred to as "SL4". SL4 is a modified version of the opening layout of the first velocity adjustment section 15 and the second velocity adjustment section 16 shown in Figure 23. In the first velocity adjustment section 15 shown in Figure 31(E), the diameter of the holes 15q is adjusted so that the opening ratio of the region closest to the inlet 11 (IN) (first region) is greater than the opening ratio of the remaining two regions (second regions) among the three regions divided in the longitudinal direction. The holes 15q (first holes) in the region closest to the inlet 11 have different diameters, with the large diameter set to 3 mm and the small diameter to 2 mm. The diameter of the holes 15q (second holes) in the remaining regions is set to 1 mm. Furthermore, the second flow velocity adjustment section 16 shown in Figure 31(E) adjusts the diameter of the holes 16q such that the opening ratio of the region furthest from the outlet 12 (OUT) (the fourth region) among the three regions divided in the longitudinal direction is greater than the opening ratio of the remaining two regions (the third region). The holes 16q (fourth holes) in the region furthest from the outlet 12 have different diameters, with the large diameter set to 3 mm and the small diameter to 2 mm. The diameter of the holes 16q (third holes) in the remaining regions is set to 1 mm.

[0186] The first velocity adjustment section 15 and the second velocity adjustment section 16 shown in Figure 31(F) are referred to as "SL5". SL5 is a modified version of the opening layout of the first velocity adjustment section 15 and the second velocity adjustment section 16 shown in Figure 24. In the first velocity adjustment section 15 shown in Figure 31(F), the width of the slit 15r (fifth slit) is adjusted so that the opening ratio of the region closest to the inlet 11 (IN) (first region) is greater than the opening ratio of the remaining two regions (second regions) among the three regions divided in the longitudinal direction. The width of the slit 15r in the region closest to the inlet 11 is set to 3 mm at the end on the inlet 11 side, and the width is set to narrow to 1 mm as it moves away from the inlet 11. The width of the slit 15r in the remaining regions is set to 1 mm. Furthermore, the second flow velocity adjustment section 16 shown in Figure 31(F) adjusts the width of the slit 16r (sixth slit) so that the opening ratio of the region furthest from the outlet 12 (OUT) (fourth region) among the three regions divided in the longitudinal direction is greater than the opening ratio of the remaining two regions (third region). The slit 16r of the region furthest from the outlet 12 is set to have a width of 3 mm at the end opposite to the outlet 12, and the width is set to narrow to 1 mm as it approaches the outlet 12. 16r The width is set to 1 mm.

[0187] In the thermal fluid simulation, SL1-SL5 shown in Figures 31(B) to 31(F) are applied to the container 14 of the cooler 10 as shown in Figure 31(A). For each case, the pressure loss between the inlet 11 and outlet 12, the refrigerant flow velocity at the positions of semiconductor elements CP1 and CP2 in the mounting region AR1-AR3, and the temperatures of semiconductor elements CP1 and CP2 are determined when a rectangular or cylindrical cooling fin 13a is applied to the heat sink 13. For comparison, the pressure loss between the inlet 11 and outlet 12, the refrigerant flow velocity at the positions of semiconductor elements CP1 and CP2 in the mounting region AR1-AR3, and the temperatures of semiconductor elements CP1 and CP2 are also determined when a rectangular or cylindrical cooling fin 13a is applied to the container 14 of the cooler 10 as shown in Figure 31(A). In the thermal fluid simulation, heat generation is reproduced by introducing a certain loss to the semiconductor elements CP1 and CP2 in the mounting region AR1-AR3. The evaluation results from the thermal fluid simulation are shown in Figures 32 and 33.

[0188] Figure 32 shows the evaluation results of a thermal fluid simulation for a third example of a cooler using prismatic cooling fins. Figure 32(A) shows an example of the evaluation results of pressure loss in the cooler. Figure 32(B) shows an example of the evaluation results of refrigerant flow velocity with respect to semiconductor element position. Figure 32(C) shows an example of the evaluation results of semiconductor element temperature with respect to semiconductor element position. In Figures 32(A) to 32(C), the flow velocity adjustment sections (first and second flow velocity adjustment sections) applied to the cooler container are represented as "SL1-SL5" (Figures 31(B)-31(F)), and the case without a flow velocity adjustment section is represented as "None".

[0189] As shown in Figure 32(A), the pressure loss of the cooler 10 increases by 91.2% when SL1 is applied, 52.1% when SL2 is applied, 56.1% when SL3 is applied, 72.9% when SL4 is applied, and 50.6% when SL5 is applied, compared to the case without a flow velocity adjustment unit (pressure loss shown by the dotted line L1 in Figure 32(A)). On the other hand, the pressure loss of the cooler 10 decreases by 20.4% when SL2 is applied, 18.4% when SL3 is applied, 9.6% when SL4 is applied, and 21.2% when SL5 is applied, compared to SL1 with a constant slit width (pressure loss shown by the dashed line L2 in Figure 32(A)). Therefore, when SL2-SL5 are applied, the increase in pressure loss compared to the case without a flow velocity adjustment unit is suppressed compared to when SL1 is applied.

[0190] As shown in Figure 32(B), without the flow velocity adjustment unit, the refrigerant flow velocity is slow at the locations of semiconductor elements CP1 and CP2 in mounting region AR1, and fast at the locations of semiconductor elements CP1 and CP2 in mounting region AR3, resulting in a flow bias distribution. On the other hand, when SL1-SL5 is applied, the flow bias distribution of the refrigerant at the locations of semiconductor elements CP1 and CP2 in mounting regions AR1-AR3 is suppressed compared to the case without the flow velocity adjustment unit, resulting in a more uniform flow.

[0191] As shown in Figure 32(C), without a flow rate adjustment unit, the temperatures of semiconductor elements CP1 and CP2 increase as you approach the mounting region AR1, where the refrigerant flow rate is slower. On the other hand, when SL1-SL5 is applied, the temperatures of semiconductor elements CP1 and CP2 in the mounting regions AR1-AR3 are kept relatively constant and cooled more uniformly compared to the case without a flow rate adjustment unit.

[0192] From the results in Figures 32(A) to 32(C), it can be said that in the cooler 10 of Figure 31(A) with prismatic cooling fins, when SL1-SL5 are applied, a superior flow distribution suppression effect and semiconductor device cooling effect can be obtained compared to the case without a flow velocity adjustment section. Furthermore, in the cooler 10 of Figure 31(A) with prismatic cooling fins, when SL2-SL5 are applied, it can be said that a flow distribution suppression effect and semiconductor device cooling effect equivalent to or close to that when SL1 is applied can be obtained while suppressing the increase in pressure loss compared to when SL1 is applied.

[0193] Figure 33 shows the evaluation results of a thermal fluid simulation for a third example of a cooler that uses cylindrical cooling fins. Figure 33(A) shows an example of the evaluation results of pressure loss in the cooler. Figure 33(B) shows an example of the evaluation results of refrigerant flow velocity with respect to semiconductor element position. Figure 33(C) shows an example of the evaluation results of semiconductor element temperature with respect to semiconductor element position. In Figures 33(A) to 33(C), the flow velocity adjustment sections (first and second flow velocity adjustment sections) applied to the cooler container are represented as "SL1-SL5" (Figures 31(B)-31(F)), and the case where no flow velocity adjustment section is applied is represented as "None".

[0194] As shown in Figure 33(A), the pressure loss of the cooler 10 increases by 106.8% when SL1 is applied, by 53.0% when SL2 is applied, by 56.9% when SL3 is applied, by 62.0% when SL4 is applied, and by 53.0% when SL5 is applied, compared to the case without a flow velocity adjustment unit (pressure loss shown by the dotted line L1 in Figure 33(A)). On the other hand, the pressure loss of the cooler 10 decreases by 26.0% when SL2 is applied, by 24.1% when SL3 is applied, by 21.6% when SL4 is applied, and by 26.0% when SL5 is applied, compared to SL1 with a constant slit width (pressure loss shown by the dashed line L2 in Figure 33(A)). Therefore, when SL2-SL5 are applied, the increase in pressure loss compared to the case without a flow velocity adjustment unit is suppressed compared to when SL1 is applied.

[0195] As shown in Figure 33(B), without the flow velocity adjustment unit, the refrigerant flow velocity is slow at the locations of semiconductor elements CP1 and CP2 in mounting region AR1, and fast at the locations of semiconductor elements CP1 and CP2 in mounting region AR3, resulting in a flow bias distribution. On the other hand, when SL1-SL5 is applied, the flow bias distribution of the refrigerant at the locations of semiconductor elements CP1 and CP2 in mounting regions AR1-AR3 is suppressed compared to the case without the flow velocity adjustment unit, resulting in a more uniform flow.

[0196] As shown in Figure 33(C), without a flow rate adjustment unit, the temperatures of semiconductor elements CP1 and CP2 increase as you approach the mounting region AR1, where the refrigerant flow rate is slower. On the other hand, when SL1-SL5 is applied, the temperatures of semiconductor elements CP1 and CP2 in the mounting regions AR1-AR3 are kept relatively constant and cooled more uniformly compared to the case without a flow rate adjustment unit.

[0197] From the results in Figures 33(A) to 33(C), it can be said that in the cooler 10 of Figure 31(A) with cylindrical cooling fins, when SL1-SL5 are applied, a superior flow distribution suppression effect and semiconductor device cooling effect can be obtained compared to the case without a flow velocity adjustment section. Furthermore, in the cooler 10 of Figure 31(A) with cylindrical cooling fins, when SL2-SL5 are applied, it can be said that a flow distribution suppression effect and semiconductor device cooling effect equivalent to or close to that when SL1 is applied can be obtained while suppressing the increase in pressure loss compared to when SL1 is applied.

[0198] <Example 4> Figure 34 illustrates a fourth example of the cooler according to the fourth embodiment. Figure 34(A) schematically shows a perspective view of the main part of the fourth example cooler and the layout of the semiconductor element mounting area. Figures 34(B) to 34(F) schematically show plan views of the main part of the flow velocity adjustment section applied to the fourth example cooler.

[0199] In the fourth example, a container 14 as shown in Figure 34(A) is used for the cooler 10. The container 14 shown in Figure 34(A) corresponds to the one shown in Figure 20. The container 14 shown in Figure 34(A) has an inlet 11 (IN) communicating with the center of the first flow path 14e and an outlet 12 (OUT) communicating with the center of the second flow path 14f on its bottom plate 14h. The cooling fins 13a of the heat dissipation plate 13 covering the container 14 are housed in the third flow path 14g, which is an internal space above the first flow path 14e and the second flow path 14f. In the thermal fluid simulation, a rectangular prism-shaped cooling fin 13a as shown in Figures 3(A) and 3(B), or a cylindrical cooling fin 13a as shown in Figures 15(A) and 15(B) is used. Then, in the region on the heat sink 13 corresponding to the third flow path 14g (the region shown by the dotted line frame in Figure 34(A)), semiconductor elements CP1 and CP2 are arranged in each of the three mounting regions AR1, AR2, and AR3, respectively, as shown in Figure 34(A), following the example in Figure 1 and so on.

[0200] In Figure 34(A) (and Figures 34(B) to 34(F) described later), the inlet 11 side of the container 14 is represented as "IN," and the outlet 12 side is represented as "OUT." The three mounting regions AR1-AR3, and the semiconductor elements CP1 and CP2 provided in each of them, are positioned relative to the IN and OUT of the container 14 as shown in Figure 34(A).

[0201] In the thermal fluid simulation, the cooler 10 shown in Figure 34(A) is equipped with a first velocity adjustment unit 115 and a second velocity adjustment unit 116 as shown in Figure 34(B), and a first velocity adjustment unit 15 and a second velocity adjustment unit 16 as shown in Figures 34(C) to 34(F). Figures 34(B) to 34(F) illustrate the positions of the inlet 11 (IN) and outlet 12 (OUT).

[0202] Here, the first flow velocity adjustment section 115 and the second flow velocity adjustment section 116 shown in Figure 34(B) are referred to as "SL1". SL1 corresponds to the first flow velocity adjustment section 115 and the second flow velocity adjustment section 116 shown in Figure 10. The first flow velocity adjustment section 115 and the second flow velocity adjustment section 116 shown in Figure 34(B) each have a slit 115e (7th slit) and a slit 116e (8th slit) of constant width extending in the longitudinal direction. The widths of slits 115e and 116e are set to 1 mm.

[0203] The first flow velocity adjustment section 15 and the second flow velocity adjustment section 16 shown in Figure 34(C) are denoted as "SL2". The first flow velocity adjustment section 15 shown in Figure 34(C) has a slit 15s that is the same as the slit 15e of the first flow velocity adjustment section 15 shown in Figure 25(C). The second flow velocity adjustment section 16 shown in Figure 34(C) has a slit 16s that is the same as the slit 16e of the second flow velocity adjustment section 16 shown in Figure 25(C).

[0204] The first flow velocity adjustment section 15 and the second flow velocity adjustment section 16 shown in Figure 34(D) are denoted as "SL3". The first flow velocity adjustment section 15 shown in Figure 34(D) has a slit 15t that is the same as the slit 15f of the first flow velocity adjustment section 15 shown in Figure 25(D). The second flow velocity adjustment section 16 shown in Figure 34(D) has a slit 16t that is the same as the slit 16f of the second flow velocity adjustment section 16 shown in Figure 25(D).

[0205] The first flow velocity adjustment section 15 and the second flow velocity adjustment section 16 shown in Figure 34(E) are denoted as "SL4". The first flow velocity adjustment section 15 shown in Figure 34(E) has a hole 15u that is the same as the hole 15g of the first flow velocity adjustment section 15 shown in Figure 25(E). The second flow velocity adjustment section 16 shown in Figure 34(E) has a hole 16u that is the same as the hole 16g of the second flow velocity adjustment section 16 shown in Figure 25(E).

[0206] The first flow velocity adjustment section 15 and the second flow velocity adjustment section 16 shown in Figure 34(F) are denoted as "SL5". The first flow velocity adjustment section 15 shown in Figure 34(F) has a slit 15v that is the same as the slit 15h of the first flow velocity adjustment section 15 shown in Figure 25(F). The second flow velocity adjustment section 16 shown in Figure 34(F) has a slit 16v that is the same as the slit 16h of the second flow velocity adjustment section 16 shown in Figure 25(F).

[0207] In the thermal fluid simulation, SL1-SL5 shown in Figures 34(B) to 34(F) are applied to the container 14 of the cooler 10 as shown in Figure 34(A). For each case, the pressure loss between the inlet 11 and outlet 12, the refrigerant flow velocity at the positions of semiconductor elements CP1 and CP2 in the mounting region AR1-AR3, and the temperatures of semiconductor elements CP1 and CP2 are determined when a rectangular or cylindrical cooling fin 13a is applied to the heat sink 13. For comparison, the pressure loss between the inlet 11 and outlet 12, the refrigerant flow velocity at the positions of semiconductor elements CP1 and CP2 in the mounting region AR1-AR3, and the temperatures of semiconductor elements CP1 and CP2 are also determined when a rectangular or cylindrical cooling fin 13a is applied to the container 14 of the cooler 10 as shown in Figure 34(A). In the thermal fluid simulation, heat generation is reproduced by introducing a certain loss to the semiconductor elements CP1 and CP2 in the mounting region AR1-AR3. The evaluation results from the thermal fluid simulation are shown in Figures 35 and 36.

[0208] Figure 35 shows the evaluation results of a thermal fluid simulation for a fourth example of a cooler using prismatic cooling fins. Figure 35(A) shows an example of the evaluation results of pressure loss in the cooler. Figure 35(B) shows an example of the evaluation results of refrigerant flow velocity with respect to semiconductor element position. Figure 35(C) shows an example of the evaluation results of semiconductor element temperature with respect to semiconductor element position. In Figures 35(A) to 35(C), the flow velocity adjustment sections (first and second flow velocity adjustment sections) applied to the cooler container are represented as "SL1-SL5" (Figures 34(B)-34(F)), and the case without a flow velocity adjustment section is represented as "None".

[0209] As shown in Figure 35(A), the pressure loss of the cooler 10 increases by 98.7% when SL1 is applied, 58.5% when SL2 is applied, 62.2% when SL3 is applied, 78.3% when SL4 is applied, and 38.9% when SL5 is applied, compared to the case without a flow velocity adjustment unit (pressure loss shown by the dotted line L1 in Figure 35(A)). On the other hand, the pressure loss of the cooler 10 decreases by 20.2% when SL2 is applied, 18.4% when SL3 is applied, 10.3% when SL4 is applied, and 30.1% when SL5 is applied, compared to SL1 with a constant slit width (pressure loss shown by the dashed line L2 in Figure 35(A)). Therefore, when SL2-SL5 are applied, the increase in pressure loss compared to the case without a flow velocity adjustment unit is suppressed compared to when SL1 is applied.

[0210] As shown in Figure 35(B), without the flow velocity adjustment unit, the refrigerant flow velocity at the locations of semiconductor elements CP1 and CP2 in mounting region AR2 is faster than the refrigerant flow velocity at the locations of semiconductor elements CP1 and CP2 in mounting regions AR1 and AR3, resulting in a flow bias distribution. On the other hand, when SL1-SL5 is applied, the flow bias distribution of the refrigerant at the locations of semiconductor elements CP1 and CP2 in mounting regions AR1-AR3 is suppressed compared to the case without the flow velocity adjustment unit, resulting in a more uniform flow.

[0211] As shown in Figure 35(C), without the flow rate adjustment unit, the temperatures of semiconductor elements CP1 and CP2 in mounting regions AR1 and AR3, where the refrigerant flow rate is slower than in mounting region AR2, become higher. On the other hand, when SL1-SL5 is applied, the temperatures of semiconductor elements CP1 and CP2 in mounting regions AR1-AR3 are kept relatively constant and cooled more uniformly compared to the case without the flow rate adjustment unit.

[0212] From the results in Figures 35(A) to 35(C), it can be said that in the cooler 10 of Figure 34(A) with prismatic cooling fins, when SL1-SL5 are applied, a superior flow distribution suppression effect and semiconductor device cooling effect can be obtained compared to the case without a flow velocity adjustment section. Furthermore, in the cooler 10 of Figure 34(A) with prismatic cooling fins, when SL2-SL5 are applied, it can be said that a flow distribution suppression effect and semiconductor device cooling effect equivalent to or close to that when SL1 is applied can be obtained while suppressing the increase in pressure loss compared to when SL1 is applied.

[0213] Figure 36 shows the evaluation results of a thermal fluid simulation for a fourth example of a cooler that uses cylindrical cooling fins. Figure 36(A) shows an example of the evaluation results of pressure loss in the cooler. Figure 36(B) shows an example of the evaluation results of refrigerant flow velocity with respect to semiconductor element position. Figure 36(C) shows an example of the evaluation results of semiconductor element temperature with respect to semiconductor element position. In Figures 36(A) to 36(C), the flow velocity adjustment units (first and second flow velocity adjustment units) applied to the cooler container are represented as "SL1-SL5" (Figures 34(B)-34(F)), and the case where no flow velocity adjustment unit is applied is represented as "None".

[0214] As shown in Figure 36(A), the pressure loss of the cooler 10 increases by 113.5% when SL1 is applied, by 57.9% when SL2 is applied, by 62.1% when SL3 is applied, by 68.2% when SL4 is applied, and by 36.1% when SL5 is applied, compared to the case without a flow velocity adjustment unit (pressure loss shown by the dotted line L1 in Figure 36(A)). On the other hand, the pressure loss of the cooler 10 decreases by 26.0% when SL2 is applied, by 24.1% when SL3 is applied, by 21.2% when SL4 is applied, and by 36.3% when SL5 is applied, compared to SL1 with a constant slit width (pressure loss shown by the dashed line L2 in Figure 36(A)). Therefore, when SL2-SL5 are applied, the increase in pressure loss compared to the case without a flow velocity adjustment unit is suppressed compared to when SL1 is applied.

[0215] As shown in Figure 36(B), without the flow velocity adjustment unit, the refrigerant flow velocity at the locations of semiconductor elements CP1 and CP2 in mounting region AR2 is faster than the refrigerant flow velocity at the locations of semiconductor elements CP1 and CP2 in mounting regions AR1 and AR3, resulting in a flow bias distribution. On the other hand, when SL1-SL5 is applied, the flow bias distribution of the refrigerant at the locations of semiconductor elements CP1 and CP2 in mounting regions AR1-AR3 is suppressed compared to the case without the flow velocity adjustment unit, resulting in a more uniform flow.

[0216] As shown in Figure 36(C), without the flow rate adjustment unit, the temperatures of semiconductor elements CP1 and CP2 in mounting regions AR1 and AR3, where the refrigerant flow rate is slower than in mounting region AR2, become higher. On the other hand, when SL1-SL5 is applied, the temperatures of semiconductor elements CP1 and CP2 in mounting regions AR1-AR3 are kept relatively constant and cooled more uniformly compared to the case without the flow rate adjustment unit.

[0217] From the results in Figures 36(A) to 36(C), it can be said that in the cooler 10 of Figure 34(A) with cylindrical cooling fins, when SL1-SL5 are applied, a superior flow distribution suppression effect and semiconductor device cooling effect can be obtained compared to the case without a flow velocity adjustment section. Furthermore, in the cooler 10 of Figure 34(A) with cylindrical cooling fins, when SL2-SL5 are applied, it can be said that a flow distribution suppression effect and semiconductor device cooling effect equivalent to or close to that when SL1 is applied can be obtained while suppressing the increase in pressure loss compared to when SL1 is applied.

[0218] <Example 5> Figure 37 illustrates a fifth example of the cooler according to the fourth embodiment. Figure 37(A) schematically shows a perspective view of the main part of the fifth example cooler and the layout of the semiconductor element mounting area. Figures 37(B) to 37(F) schematically show plan views of the main part of the flow velocity adjustment section applied to the fifth example cooler.

[0219] In the fifth example, a container 14 as shown in Figure 37(A) is used for the cooler 10. The container 14 shown in Figure 37(A) is a modified version of the one shown in Figure 21. The container 14 shown in Figure 37(A) has an inlet 11 (IN) on the bottom plate 14h that communicates with the end of the third side wall 14c of the first flow path 14e, and an outlet 12 (OUT) that communicates with the end of the fourth side wall 14d of the second flow path 14f. The cooling fins 13a of the heat dissipation plate 13 that covers the container 14 are housed in the third flow path 14g, which is an internal space above the first flow path 14e and the second flow path 14f. In the thermal fluid simulation, a rectangular prism-shaped cooling fin 13a as shown in Figures 3(A) and 3(B), or a cylindrical cooling fin 13a as shown in Figures 15(A) and 15(B) is used. Then, in the region on the heat sink 13 corresponding to the third flow path 14g (the region shown by the dotted line frame in Figure 37(A)), semiconductor elements CP1 and CP2 are arranged in each of the three mounting regions AR1, AR2, and AR3, respectively, as shown in Figure 37(A), following the example in Figure 1 and so on.

[0220] In Figure 37(A) (and Figures 37(B) to 37(F) described later), the inlet 11 side of the container 14 is represented as "IN," and the outlet 12 side is represented as "OUT." The three mounting regions AR1-AR3, and the semiconductor elements CP1 and CP2 provided in each of them, are positioned relative to the IN and OUT of the container 14 as shown in Figure 37(A).

[0221] In the thermal fluid simulation, the cooler 10 shown in Figure 37(A) is equipped with a first flow velocity adjustment unit 115 and a second flow velocity adjustment unit 116 as shown in Figure 37(B), and a first flow velocity adjustment unit 15 and a second flow velocity adjustment unit 16 as shown in Figures 37(C) to 37(F). Figures 37(B) to 37(F) illustrate the positions of the inlet 11 (IN) and outlet 12 (OUT).

[0222] Here, the first flow velocity adjustment section 115 and the second flow velocity adjustment section 116 shown in Figure 37(B) are referred to as "SL1". SL1 corresponds to the first flow velocity adjustment section 115 and the second flow velocity adjustment section 116 shown in Figure 10. The first flow velocity adjustment section 115 and the second flow velocity adjustment section 116 shown in Figure 37(B) each have a slit 115e (7th slit) and a slit 116e (8th slit) of constant width extending in the longitudinal direction. The widths of slits 115e and 116e are set to 1 mm.

[0223] The first and second velocity adjustment sections 15 and 16 shown in Figure 37(C) are denoted as "SL2". SL2 is a modified version of the opening layout of the first and second velocity adjustment sections 15 and 16 shown in Figure 5. The first velocity adjustment section 15 shown in Figure 37(C) has its slit width 15w adjusted so that the opening ratio of the region closest to the inlet 11 (IN) (first region) is greater than the opening ratios of the remaining two regions (second regions) among the three regions divided in the longitudinal direction. The width of the slit 15w (first slit) in the region closest to the inlet 11 is set to 2 mm, and the width of the slits 15w (second slit) in the remaining regions is set to 1 mm. Furthermore, the second flow velocity adjustment section 16 shown in Figure 37(C) adjusts the width of the slit 16w so that the opening ratio of the region furthest from the outlet 12 (OUT) (the fourth region) among the three regions divided in the longitudinal direction is greater than the opening ratio of the remaining two regions (the third region). The width of the slit 16w (fourth slit) in the region furthest from the outlet 12 is set to 2 mm, and the width of the slit 16w (third slit) in the remaining regions is set to 1 mm.

[0224] The first velocity adjustment section 15 and the second velocity adjustment section 16 shown in Figure 37(D) are referred to as "SL3". SL3 is a modified version of the opening layout of the first velocity adjustment section 15 and the second velocity adjustment section 16 shown in Figure 22. The first velocity adjustment section 15 shown in Figure 37(D) is a slit 15x, obtained by dividing the slit 15w in Figure 37(C) into two parts in each of the three regions obtained by dividing the first velocity adjustment section 15 in the longitudinal direction. Similarly, the second velocity adjustment section 16 shown in Figure 37(D) is a slit 16x, obtained by dividing the slit 16w in Figure 37(C) into two parts in each of the three regions obtained by dividing the second velocity adjustment section 16 in the longitudinal direction.

[0225] The first velocity adjustment section 15 and the second velocity adjustment section 16 shown in Figure 37(E) are referred to as "SL4". SL4 is a modified version of the opening layout of the first velocity adjustment section 15 and the second velocity adjustment section 16 shown in Figure 23. In the first velocity adjustment section 15 shown in Figure 37(E), the diameter of the hole 15y is adjusted so that the opening ratio of the region closest to the inlet 11 (IN) (first region) is greater than the opening ratio of the remaining two regions (second regions) among the three regions divided in the longitudinal direction. The diameter of the hole 15y (first hole) in the region closest to the inlet 11 is set to 2 mm, and the diameter of the holes 15y (second holes) in the remaining regions is set to 1 mm. Furthermore, the second flow velocity adjustment section 16 shown in Figure 37(E) adjusts the diameter of the hole 16y so that the opening ratio of the region furthest from the outlet 12 (OUT) (the fourth region) among the three regions divided in the longitudinal direction is greater than the opening ratio of the remaining two regions (the third region). The diameter of the hole 16y (fourth hole) in the region furthest from the outlet 12 is set to 2 mm, and the diameter of the hole 16y (third hole) in the remaining regions is set to 1 mm.

[0226] The first velocity adjustment section 15 and the second velocity adjustment section 16 shown in Figure 37(F) are referred to as "SL5". SL5 is a modified version of the opening layout of the first velocity adjustment section 15 and the second velocity adjustment section 16 shown in Figure 24. The first velocity adjustment section 15 shown in Figure 37(F) is adjusted so that the opening ratio of the region closer to the inlet 11 (IN) (first region) is greater than the opening ratio of the region further away from the inlet 11 (second region), that is, the width of the slit 15z (fifth slit) becomes narrower as it moves away from the inlet 11. The width of one end of the slit 15z on the inlet 11 side is set to 2 mm, and the width of the other end is set to 1 mm. Furthermore, the second flow velocity adjustment section 16 shown in Figure 37(F) is adjusted so that the opening ratio in the region closer to the outlet 12 (OUT) (third region) is smaller than the opening ratio in the region further away from the outlet 12 (fourth region), that is, the width of the slit 16z (sixth slit) widens as it moves away from the outlet 12. The width of one end of the slit 16z on the outlet 12 side is set to 1 mm, and the width of the other end is set to 2 mm.

[0227] In the thermal fluid simulation, SL1-SL5 shown in Figures 37(B) to 37(F) are applied to the container 14 of the cooler 10 as shown in Figure 37(A). For each case, the pressure loss between the inlet 11 and outlet 12, the refrigerant flow velocity at the positions of semiconductor elements CP1 and CP2 in the mounting region AR1-AR3, and the temperatures of semiconductor elements CP1 and CP2 are determined when a rectangular or cylindrical cooling fin 13a is applied to the heat sink 13. For comparison, the pressure loss between the inlet 11 and outlet 12, the refrigerant flow velocity at the positions of semiconductor elements CP1 and CP2 in the mounting region AR1-AR3, and the temperatures of semiconductor elements CP1 and CP2 are also determined when a rectangular or cylindrical cooling fin 13a is applied to the container 14 of the cooler 10 as shown in Figure 37(A). In the thermal fluid simulation, heat generation is reproduced by introducing a certain loss to the semiconductor elements CP1 and CP2 in the mounting region AR1-AR3. The evaluation results from the thermal fluid simulation are shown in Figures 38 and 39.

[0228] Figure 38 shows the evaluation results of a thermal fluid simulation for a fifth example of a cooler using prismatic cooling fins. Figure 38(A) shows an example of the evaluation results of pressure loss in the cooler. Figure 38(B) shows an example of the evaluation results of refrigerant flow velocity with respect to semiconductor element position. Figure 38(C) shows an example of the evaluation results of semiconductor element temperature with respect to semiconductor element position. In Figures 38(A) to 38(C), the flow velocity adjustment sections (first and second flow velocity adjustment sections) applied to the cooler container are represented as "SL1-SL5" (Figures 38(B)-38(F)), and the case without a flow velocity adjustment section is represented as "None".

[0229] As shown in Figure 38(A), the pressure loss of the cooler 10 increases by 69.8% when SL1 is applied, 50.7% when SL2 is applied, 53.1% when SL3 is applied, 61.7% when SL4 is applied, and 41.7% when SL5 is applied, compared to the case without a flow velocity adjustment unit (pressure loss shown by the dotted line L1 in Figure 38(A)). On the other hand, the pressure loss of the cooler 10 decreases by 11.2% when SL2 is applied, 9.9% when SL3 is applied, 4.8% when SL4 is applied, and 16.5% when SL5 is applied, compared to SL1 with a constant slit width (pressure loss shown by the dashed line L2 in Figure 32(A)). Therefore, when SL2-SL5 are applied, the increase in pressure loss compared to the case without a flow velocity adjustment unit is suppressed compared to when SL1 is applied.

[0230] As shown in Figure 38(B), without the flow velocity adjustment unit, the refrigerant flow velocity at the locations of semiconductor elements CP1 and CP2 in the mounting region AR1-AR3 becomes uneven, resulting in a flow bias distribution. On the other hand, when SL1-SL5 is applied, the flow bias distribution of the refrigerant at the locations of semiconductor elements CP1 and CP2 in the mounting region AR1-AR3 is suppressed compared to the case without the flow velocity adjustment unit, resulting in a more uniform flow.

[0231] As shown in Figure 38(C), without the flow rate adjustment unit, the temperatures of semiconductor elements CP1 and CP2 in the mounting region AR1-AR3 become uneven. On the other hand, when SL1-SL5 is applied, the temperatures of semiconductor elements CP1 and CP2 in the mounting region AR1-AR3 are kept relatively constant and cooled more uniformly compared to the case without the flow rate adjustment unit.

[0232] From the results in Figures 38(A) to 38(C), it can be said that in the cooler 10 of Figure 37(A) with prismatic cooling fins, when SL1-SL5 are applied, a superior flow distribution suppression effect and semiconductor device cooling effect can be obtained compared to the case without a flow velocity adjustment section. Furthermore, in the cooler 10 of Figure 37(A) with prismatic cooling fins, when SL2-SL5 are applied, it can be said that a flow distribution suppression effect and semiconductor device cooling effect equivalent to or close to that when SL1 is applied can be obtained while suppressing the increase in pressure loss compared to when SL1 is applied.

[0233] Figure 39 shows the evaluation results of a thermal fluid simulation for a fifth example of a cooler that uses cylindrical cooling fins. Figure 39(A) shows an example of the evaluation results of pressure loss in the cooler. Figure 39(B) shows an example of the evaluation results of refrigerant flow velocity with respect to semiconductor element position. Figure 39(C) shows an example of the evaluation results of semiconductor element temperature with respect to semiconductor element position. In Figures 39(A) to 39(C), the flow velocity adjustment units (first and second flow velocity adjustment units) applied to the cooler container are represented as "SL1-SL5" (Figures 37(B)-37(F)), and the case where no flow velocity adjustment unit is applied is represented as "None".

[0234] As shown in Figure 39(A), the pressure loss of the cooler 10 increases by 85.8% when SL1 is applied, 56.8% when SL2 is applied, 60.3% when SL3 is applied, 60.2% when SL4 is applied, and 47.3% when SL5 is applied, compared to the case without a flow velocity adjustment unit (pressure loss shown by the dotted line L1 in Figure 39(A)). On the other hand, the pressure loss of the cooler 10 decreases by 15.6% when SL2 is applied, 13.7% when SL3 is applied, 13.8% when SL4 is applied, and 20.7% when SL5 is applied, compared to SL1 with a constant slit width (pressure loss shown by the dashed line L2 in Figure 39(A)). Therefore, when SL2-SL5 are applied, the increase in pressure loss compared to the case without a flow velocity adjustment unit is suppressed compared to when SL1 is applied.

[0235] As shown in Figure 39(B), without the flow velocity adjustment unit, the refrigerant flow velocity at the locations of semiconductor elements CP1 and CP2 in the mounting region AR1-AR3 becomes uneven, resulting in a flow bias distribution. On the other hand, when SL1-SL5 is applied, the flow bias distribution of the refrigerant at the locations of semiconductor elements CP1 and CP2 in the mounting region AR1-AR3 is suppressed compared to the case without the flow velocity adjustment unit, resulting in a more uniform flow.

[0236] As shown in Figure 39(C), without the flow rate adjustment unit, the temperatures of semiconductor elements CP1 and CP2 in the mounting region AR1-AR3 become uneven. On the other hand, when SL1-SL5 is applied, the temperatures of semiconductor elements CP1 and CP2 in the mounting region AR1-AR3 are kept relatively constant and cooled more uniformly compared to the case without the flow rate adjustment unit.

[0237] From the results in Figures 39(A) to 39(C), it can be said that in the cooler 10 of Figure 37(A) with cylindrical cooling fins, when SL1-SL5 are applied, a superior flow distribution suppression effect and semiconductor device cooling effect can be obtained compared to the case without a flow velocity adjustment section. Furthermore, in the cooler 10 of Figure 37(A) with cylindrical cooling fins, when SL2-SL5 are applied, it can be said that a flow distribution suppression effect and semiconductor device cooling effect equivalent to or close to that when SL1 is applied can be obtained while suppressing the increase in pressure loss compared to when SL1 is applied.

[0238] The above merely illustrates the principle of the present invention. Furthermore, numerous modifications and changes are possible for those skilled in the art, and the present invention is not limited to the exact configurations and applications shown and described above. All corresponding modifications and equivalents are considered to be within the scope of the present invention as defined by the appended claims and equivalents. [Explanation of Symbols]

[0239] 1 Semiconductor device 10, 110 cooler 11 Inlet 12 Outlet 13 Heat sink 13a Cooling fins 13b Installation surface 14 Container 14a First side wall 14b 2nd side wall 14c 3rd side wall 14d 4th side wall 14e First channel 14f Second channel 14g Third channel 14h bottom plate 15, 115 1st flow rate adjustment section 15a First region 15aa First slit 15ab First hole 15ac Fifth slit 15b Second region 15ba Second slit 15bb Second hole 15c, 16c Central part 15d, 16d End part 16, 116 Second flow rate adjustment part 16a Third region 16aa Third slit 16ab Third hole 16ac Sixth slit 16b Fourth region 16ba Fourth slit 16bb Fourth hole 15e, 15f, 15h, 15i, 15j, 15m, 15n, 15p, 15r, 15s, 15t, 15v, 15w, 15x, 15z, 16e, 16f, 16h, 16i, 16j, 16m, 16n, 16p, 16r, 16s, 16t, 16v, 16w, 16x, 16z, 11s Slit 15g, 15k, 15q, 15u, 15y, 16g, 16k, 16q, 16u, 16y Hole 20 Semiconductor module 21, 22, 23 Circuit element part 24 Insulating circuit board 24a Insulating board 24b, 24c Conductor layer 25, 26, CP\(_1\), CP\(_2\) Semiconductor element 27, 28 Bonding layer 30 Refrigerant 40 Pump 50 Heat exchanger 115aa Seventh slit 116aa Eighth slit AR\(_1\), AR\(_2\), AR\(_3\) Mounting region SL\(_1\), SL\(_2\), SL\(_3\), SL\(_4\), SL\(_5\) Flow rate adjustment part

Claims

1. A container having opposing first and second side walls, and equipped with a refrigerant inlet and outlet, Within the container, there is a first flow path arranged parallel to the first side wall and communicating with the inlet, Within the container, there is a second flow path arranged parallel to the second side wall and communicating with the outlet, A third channel is provided within the container and communicates with the first channel and the second channel, A first flow velocity adjustment unit is disposed between the first flow path and the third flow path within the container, A second flow velocity adjustment unit is provided within the container, located between the second flow path and the third flow path. It has, The first flow velocity adjustment unit includes a first region having a first aperture ratio and a second region having a second aperture ratio smaller than the first aperture ratio. The second flow velocity adjustment section includes a third region having a third aperture ratio and a fourth region having a fourth aperture ratio greater than the third aperture ratio. The width of the opening of the first region from the end on the first velocity adjustment section side toward the second velocity adjustment section to the end on the second velocity adjustment section side in the direction from the first velocity adjustment section toward the second velocity adjustment section is greater than the width of the opening of the second region from the end on the first velocity adjustment section side toward the second velocity adjustment section to the end on the second velocity adjustment section side in the direction from the first velocity adjustment section toward the second velocity adjustment section. A cooler in which the width of the opening in the third region from the end on the first velocity adjustment section side to the end on the second velocity adjustment section side in the direction from the first velocity adjustment section to the second velocity adjustment section is smaller than the width of the opening in the fourth region from the end on the first velocity adjustment section side to the end on the second velocity adjustment section side in the direction from the first velocity adjustment section to the second velocity adjustment section.

2. The cooler according to claim 1, wherein the first region is arranged opposite to the third region.

3. The first region is located closer to the inlet communicating with the first flow path than the second region. The cooler according to claim 1, wherein the third region is located closer to the outlet communicating with the second flow path than the fourth region.

4. The opening in the first region is a first slit having a first width, The opening in the second region is a second slit having a second width that is narrower than the first width. The opening in the third region is a third slit having a third width, The cooler according to claim 1, wherein the opening in the fourth region is a fourth slit having a fourth width that is wider than the third width.

5. The opening in the first region is a first hole having a first diameter, The opening in the second region is a second hole having a second diameter smaller than the first diameter. The opening in the third region is a third hole having a third diameter, The cooler according to claim 1, wherein the opening in the fourth region is a fourth hole having a fourth diameter larger than the third diameter.

6. The openings in the first region and the second region are fifth slits that extend from the first region to the second region and narrow in width from the first region to the second region. The cooler according to claim 1, wherein the openings in the third region and the fourth region are sixth slits that extend from the third region to the fourth region and widen in width from the third region to the fourth region.

7. The first flow path is a first groove extending along the first side wall at the bottom between the first side wall and the second side wall of the container, The second flow channel is a second groove extending along the second side wall at the bottom, The cooler according to any one of claims 1 to 6, wherein the third flow path is an internal space of the container above the first groove and the second groove.

8. Of the three regions obtained by dividing the first flow path in the direction in which the first groove extends along the first side wall, one corresponds to the first region, and the remaining two correspond to the second region. The cooler according to claim 7, wherein of the group of regions obtained by dividing the second flow path into three in the direction in which the second groove extends along the second side wall, one corresponds to the third region and the remaining two correspond to the fourth region.

9. The opening in the first region and the opening in the second region are arranged to be located at the end of the first channel on the first side wall side, The cooler according to claim 7, wherein the opening in the third region and the opening in the fourth region are arranged to be located at the end of the second flow path on the second side wall side.

10. The cooler according to claim 7, wherein the container includes a heat sink plate that covers the third flow path and has fins disposed within the third flow path.

11. Cooler and The semiconductor module mounted on the aforementioned cooler, Equipped with, The aforementioned cooler is, A container having opposing first and second side walls, and equipped with a refrigerant inlet and outlet, Within the container, there is a first flow path arranged parallel to the first side wall and communicating with the inlet, Within the container, there is a second flow path arranged parallel to the second side wall and communicating with the outlet, A third channel is provided within the container and communicates with the first channel and the second channel, A first flow velocity adjustment unit is disposed between the first flow path and the third flow path within the container, A second flow velocity adjustment unit is provided within the container, located between the second flow path and the third flow path. It has, The first flow velocity adjustment unit includes a first region having a first aperture ratio and a second region having a second aperture ratio smaller than the first aperture ratio. The second flow velocity adjustment section includes a third region having a third aperture ratio and a fourth region having a fourth aperture ratio greater than the third aperture ratio. The width of the opening of the first region from the end on the first velocity adjustment section side toward the second velocity adjustment section to the end on the second velocity adjustment section side in the direction from the first velocity adjustment section toward the second velocity adjustment section is greater than the width of the opening of the second region from the end on the first velocity adjustment section side toward the second velocity adjustment section to the end on the second velocity adjustment section side in the direction from the first velocity adjustment section toward the second velocity adjustment section. The width of the opening in the third region from the end on the first velocity adjustment section side to the end on the second velocity adjustment section side in the direction from the first velocity adjustment section to the second velocity adjustment section is smaller than the width of the opening in the fourth region from the end on the first velocity adjustment section side to the end on the second velocity adjustment section side in the direction from the first velocity adjustment section to the second velocity adjustment section. The semiconductor module is a semiconductor device mounted in a position opposite the third flow path of the cooler.

12. The semiconductor device according to claim 11, wherein the first region is arranged opposite to the third region.

13. The first region is located closer to the inlet communicating with the first flow path than the second region. The semiconductor device according to claim 11, wherein the third region is located closer to the outlet communicating with the second flow path than the fourth region.

14. The opening in the first region is a first slit having a first width, The opening in the second region is a second slit having a second width that is narrower than the first width. The opening in the third region is a third slit having a third width, The semiconductor device according to claim 11, wherein the opening in the fourth region is a fourth slit having a fourth width wider than the third width.

15. The opening in the first region is a first hole having a first diameter, The opening in the second region is a second hole having a second diameter smaller than the first diameter. The opening in the third region is a third hole having a third diameter, The semiconductor device according to claim 11, wherein the opening of the fourth region is a fourth hole having a fourth diameter larger than the third diameter.

16. The openings in the first region and the second region are fifth slits that extend from the first region to the second region and narrow in width from the first region to the second region. The semiconductor device according to claim 11, wherein the openings in the third region and the fourth region are sixth slits that extend from the third region to the fourth region and widen in width from the third region to the fourth region.

17. The first flow path is a first groove extending along the first side wall at the bottom between the first side wall and the second side wall of the container, The second flow channel is a second groove extending along the second side wall at the bottom, The semiconductor device according to any one of claims 11 to 16, wherein the third channel is an internal space of the container above the first groove and the second groove.

18. Of the three regions obtained by dividing the first flow path in the direction in which the first groove extends along the first side wall, one corresponds to the first region, and the remaining two correspond to the second region. The semiconductor device according to claim 17, wherein, of the three regions obtained by dividing the second flow channel in the direction in which the second groove extends along the second side wall, one corresponds to the third region and the remaining two correspond to the fourth region.

19. The opening in the first region and the opening in the second region are arranged to be located at the end of the first channel on the first side wall side, The semiconductor device according to claim 17, wherein the opening in the third region and the opening in the fourth region are arranged to be located at the end of the second channel on the second side wall side.

20. The container includes a heat sink plate that covers the third channel and has fins positioned within the third channel, The semiconductor device according to claim 17, wherein the heat sink is disposed between the opposing semiconductor module and the third channel.

Images