Semiconductor module, power conversion device and manufacturing method for the semiconductor module

Abstract

Semiconductor module (900, 900A) comprising: a semiconductor device (300) comprising: a semiconductor element (155, 156, 157, 158), a pair of conductors (410 to 413) arranged so that the semiconductor element (155, 156, 157, 158) is located therebetween, facing each other in a thickness direction and each connected to the semiconductor element (155, 156, 157, 158), a first heat dissipation member (800) disposed on a surface of a side opposite to the semiconductor element (155, 156, 157, 158) of one conductor (410 to 413) of the conductor pair (410 to 413) via an insulation member (451, 453) and including a first connecting portion (650) extending outwardly from an outer peripheral side surface of the one conductor (410 to 413), a second heat dissipation member disposed on a surface of a side opposite to the semiconductor element (155, 156, 157, 158) of the further conductor (410 to 413) of the conductor pair (410 to 413) via an insulating member (451, 453) and including a second connecting portion (810) extending outwardly from an outer peripheral side surface of the further conductor (410 to 413), and a resin (850) for sealing the outer peripheral side surfaces of the conductor pair (410 to 413); and a flow path forming body (600, 600A, 600B) connected to the first connecting portion (650) of the first heat dissipation element (800) and the second connecting portion of the second heat dissipation element of the semiconductor device (155, 156, 157, 158), wherein a first elastically deformed portion (801 to 803) which is elastically deformed is provided such that a distance in a thickness direction between an outer peripheral end of the first connecting portion of the first heat dissipation member (800) and an outer peripheral end of the second connecting portion (810) of the second heat dissipation member is smaller than a distance in a thickness direction between an intermediate portion (804) of the first connecting portion (650) of the first heat dissipation member (800) and an intermediate portion of the second connecting portion (810) of the second heat dissipation member, and the resin (850) is filled between the first connecting portion (650) of the first heat dissipation element (800) and the second connecting portion (810) of the second heat dissipation element.

Description

Technical field

[0001] The present invention relates to a semiconductor module, a power conversion device and a manufacturing method for the semiconductor module. Technical background

[0002] A semiconductor module including a power semiconductor element performing a switching operation has high conversion efficiency and is widely used in consumer applications, in-vehicle applications, railway applications, substation equipment, and the like. Since the power semiconductor element generates heat when turned on, the semiconductor module requires high heat dissipation. Particularly for in-vehicle applications, a highly efficient cooling system using a liquid refrigerant such as water is used to cool the semiconductor module to reduce its size and weight.

[0003] An example of the structure and manufacturing process of such a semiconductor module is illustrated below.

[0004] A pair of upper and lower cases, between which the power module having a power semiconductor element is sandwiched and which have a U-shaped cross section and whose peripheral wall portions are bent substantially vertically, are arranged such that the end surfaces of the peripheral wall portions of each case face each other. The pair of upper and lower cases are pressurized from the outside, and the peripheral wall portions of the cases are deformed so that the distance between them becomes smaller, so that the end surfaces of the pair of upper and lower cases come into contact with each other. In this state, the contact portions on the end surfaces are connected by welding or the like. In such a process, the power module is fixed in the case (see, for example, PTL 1). Citation listPatent literature

[0005] PTL 1: JP 2016- 39 224 APTL 2: DE 11 2018 005 627 T5

[0006] PTL 2 describes a power semiconductor device comprising: a circuit portion including a conductor for transmitting a current and a power semiconductor element; a first base portion and a second base portion facing each other, the circuit portion being disposed between the first base portion and the second base portion;and a transfer molding member in contact with the conductor and the power semiconductor element and inserted into a space between the first base portion and the second base portion, wherein the first base portion includes a first flat portion connected to a peripheral edge of the first base portion and a first bent portion plastically deformed by connecting the first flat portion and another portion of the first base portion, and the transfer molding member is integrally formed in a state in contact with the first flat portion, and wherein the transfer molding member covers one end of the first flat portion.; Summary of the inventionTechnical problem

[0007] The semiconductor module of PTL 1 requires a process of crimping the power module to the package and then welding the bonding surface of the package, resulting in low productivity. Furthermore, PTL 1 does not describe a semiconductor module capable of cooling with a refrigerant. Solution to the problem

[0008] A semiconductor module according to one aspect of the invention includes: a semiconductor device including a semiconductor element, a pair of conductors arranged with the semiconductor element therebetween, facing each other in a thickness direction, and each connected to the semiconductor element, a first heat dissipation member disposed on a surface of a side opposite to the semiconductor element of one conductor of the conductor pair via an insulation member, and including a first connection portion extending outward from an outer peripheral side surface of the one conductor, a second heat dissipation member disposed on a surface of a side opposite to the semiconductor element of the other conductor of the conductor pair via an insulation member, and including a second connection portion extending outward from an outer peripheral side surface of the other conductor,and a resin for sealing the outer peripheral side surfaces of the conductor pair; and a flow path forming body connected to the first connecting portion of the first heat dissipation member and the second connecting portion of the second heat dissipation member of the semiconductor device. A first elastically deformed portion, which is elastically deformed, is provided such thatthat a distance in a thickness direction between an outer peripheral end of the first connecting portion of the first heat dissipation member and an outer peripheral end of the second connecting portion of the second heat dissipation member is smaller than a distance in a thickness direction between an intermediate portion of the first connecting portion of the first heat dissipation member and an intermediate portion of the second connecting portion of the second heat dissipation member. The resin is filled between the first connecting portion of the first heat dissipation member and the second connecting portion of the second heat dissipation member. Advantageous effects of the invention

[0009] According to the invention, the productivity of a semiconductor module having a flow path forming body can be improved. Brief description of the drawings [ Fig. 1] Fig. 1 is an external perspective view of an embodiment of a semiconductor device constituting a semiconductor module of the invention. [ Fig. 2] Fig. 2(a) is a cross-sectional view taken along line II-II of the semiconductor device shown in Fig. 1 is illustrated, and Fig. 2(b) is an enlarged view of a connecting portion 810 shown in Fig. 2(a). [ Fig. 3] Fig. 3 is a circuit diagram showing an example of a circuit of the semiconductor device shown in Fig. 1 is illustrated. [ Fig. 4] Fig. 4(a) to Fig. 4(c) are cross-sectional views in each process for explaining a manufacturing method of the semiconductor device shown in Fig. 1 is illustrated. [ Fig. 5] Fig. 5(a) to Fig. 5(b) are cross-sectional views in each process for explaining the manufacturing method of the semiconductor device according to Fig. 4(a) to Fig. 4(c). [ Fig. 6] Fig. 6(a) to Fig. 6(c) are perspective views of the processes, each Fig. 4(a) to Fig. 4(c). [ Fig. 7] Fig. 7(a) and Fig. 7(b) are perspective views of the processes, each Fig. 5(a) and Fig. 5(b). [ Fig. 8] Fig. 8(a) is a cross-sectional view of a process of installing a pre-sealing semiconductor device configuration in a mold and performing resin molding, Fig. 8(b) is an enlarged view of an area VIIIb of Fig. 8(a) and Fig. 8(c) is a side view illustrating the shape of the connecting portion 810 of a fin base 800 before the semiconductor device configuration is resin molded prior to sealing. [ Fig. 9] Fig. 9(a) is a diagram for explaining the action of the resin in a state where the resin is molded into the shape shown in Fig. 8(a), and Fig. 9(b) is an enlarged view of an area XIb of Fig. 9(a). [ Fig. 10] Fig. 10 is a cross-sectional view illustrating an example of a first embodiment of a semiconductor module according to the invention. [ Fig. 11] Fig. 11 is a cross-sectional view illustrating another example of the first embodiment of the semiconductor module according to the invention. [ Fig. 12] Fig. 12 is a circuit diagram of a power conversion device using a semiconductor module according to the invention. [ Fig. 13] Fig. 13 is an external perspective view showing the example of the power conversion device used in Fig. 12 is illustrated. [ Fig. 14] Fig. 14 is a cross-sectional view taken along the line XIV-XIV of the power conversion device shown in Fig. 13 is illustrated. [ Fig. 15] Fig. 15(a) is a perspective view of the power conversion device shown in Fig. 14, seen from above and Fig. 15(b) is a perspective view of the power conversion device shown in Fig. 14, seen from below. [ Fig. 16] Fig. 16 is a cross-sectional view taken along the line XVI-XVI of Fig. 15(a) was taken. [ Fig. 17] Fig. 17 is a cross-sectional view illustrating a second embodiment of the semiconductor module according to the invention. [ Fig. 18] Fig. 18(a) is a cross-sectional view of a third embodiment of the semiconductor module according to the invention and Fig. 18(b) is an enlarged cross-sectional view of a process of forming the connecting portion of the slat base shown in Fig. 18(a). [ Fig. 19] Fig. 19 is a schematic view for explaining deformation of the connecting portion of the fin base by a mold in a state where the semiconductor device configuration is installed in the mold before sealing. Description of the embodiments - First embodiment -

[0010] Embodiments of the invention will be described below with reference to the drawings. The following description and the following drawings are illustrative for describing the invention and are omitted and simplified as appropriate to clarify the description.

[0011] The position, size, shape, area, and the like of each component illustrated in the drawings may not necessarily represent the actual position, size, shape, area, and the like in order to facilitate understanding of the invention.

[0012] Fig. 1 is an external perspective view of an embodiment of a semiconductor device constituting the semiconductor module of the invention.

[0013] In the following description, the X-direction, the Y-direction and the Z-direction apply as illustrated in the drawings.

[0014] Fig. 1 is an external perspective view of an embodiment of the semiconductor device according to the invention.

[0015] The semiconductor device 300 includes a device body 301, which is a resin package in which internal electronic components are sealed with a resin 850, a fin base 800, a plurality of power terminals for inputting / outputting a large current, and a plurality of signal terminals for inputting / outputting signals. The device body 301 has a substantially regular parallelepiped shape, in other words, a substantially rectangular shape when a main surface 302 having the largest area is viewed from the vertical direction in a plan view. The plurality of power terminals and the plurality of signal terminals protrude from one side 301a in the longitudinal direction (the X direction) of the device body 301 and the other side 301b facing the one side.The fin base 800, which has a large number of fins 800a, is provided on both the main surface 302 of the device body 301 and a back surface 303, which is the end surface of the main surface 302. At the outer peripheral edge of each fin base 800, the connecting portion 810 is formed with a flow path forming body 600 (see FIG. Fig. 10), which forms a space for arranging the refrigerant.

[0016] Power terminals such as a positive electrode side terminal 315B and a negative electrode side terminal 319B protrude from the other side 301b of the device body 301. An AC side terminal 320B protrudes as a power terminal from one side 301a of the device body 301.

[0017] Signal terminals such as a lower arm gate signal terminal 325L, a mirror emitter signal terminal 325M, a Kelvin emitter signal terminal 325K, and a collector detect signal terminal 325C protrude from the further side 301b of the device body 301. Signal terminals such as an upper arm gate signal terminal 325U, a temperature detect signal terminal 325S, the mirror emitter signal terminal 325M, the Kelvin emitter signal terminal 325K, and the collector detect signal terminal 325C protrude from one side 301a of the device body 301. When these signal terminals are described comprehensively, one signal terminal 325 will be used.

[0018] As in Fig. 1, the positive electrode side terminal 315B and the negative electrode side terminal 319B, which are power terminals, and the AC side terminal 320B are provided on the other side 301b and one side 301a of the device body 301 facing each other.

[0019] The plurality of power terminals and the plurality of signal terminals protrude in the longitudinal direction (the +X direction and the -X direction), and the tip is vertically bent and extends in the height direction (the +Z direction). By routing the plurality of signal terminals in the same +Z direction, it becomes easy to connect the control circuit and the drive circuit. Furthermore, since the control terminal is divided into two sides, one side 301a and the other side 301b of the device body 301, and protrudes, the creepage distance and spatial distance between the terminals are ensured.

[0020] The positive electrode side terminal 315B and the negative electrode side terminal 319B are arranged adjacent to each other in the Y direction on the other side 301b of the device body 301. Further, the positive electrode side terminal 315B and the negative electrode side terminal 319B are arranged such that the side surfaces, which are small surfaces deflected in an L-shape, face each other, so that the input / output currents are brought close to each other to reduce the inductance. In addition, since the positive electrode side terminal 315B and the negative electrode side terminal 319B, which are DC terminals, are connected to the capacitor module 500 (see Fig. 12) connected to a battery, the terminals protrude from the same side of the further side 301b in such a way that the effect of simplifying the inverter layout is obtained. The AC-side terminal 320B protrudes from the end surface opposite to the surface on which the DC-side terminal protrudes. After connecting to a current sensor 180 (see Fig. 12), the AC side terminal 320B protrudes from the power conversion device and is connected to motor generators 192 and 194 (see Fig. 12). Therefore, there is an effect that the inverter layout can be simplified by protruding in a direction different from the DC terminal connected to the capacitor module 500.

[0021] Fig. 2(a) is a cross-sectional view taken along line II-II of the semiconductor device shown in Fig. 1 is illustrated, Fig. 2(b) is an enlarged view of the connecting portion 810 shown in Fig. 2(a), and Fig. 3 is a circuit diagram showing an example of the circuit of the semiconductor device shown in Fig. 1 is illustrated.

[0022] The semiconductor device 300 includes an upper-arm circuit having a switching function including an active element 155 and a diode 156, and a lower-arm circuit having a switching function including an active element 157 and a diode 158. The active elements 155 and 157 and the diodes 156 and 158 are referred to as semiconductor elements. This semiconductor element is not particularly limited as long as it has a switching function, but transistors such as IGBT (Insulated Gate Bipolar Transistor) and MOSFET (Metal Oxide Semiconductor Field Effect Transistor) are used as the active elements 155 and 157. As the diodes 156 and 158, an SBD (Schottky Barrier Diode), an FRD (Fast Reverse Recovery Diode), and the like are used. Si is often used as a material for forming the semiconductor element, but SiC, GaN, GaO or the like may also be used.

[0023] As in Fig. 3, the positive electrode side terminal 315B is connected to a third conductor 412. The collector electrode of the active element 155 and the cathode electrode of the diode 156, which constitute the switching element of the upper arm circuit, are electrically connected by the third conductor 412. The emitter electrode of the active element 155 and the anode electrode of the diode 156 are electrically connected by the second conductor 411.

[0024] The negative electrode side terminal 319B is electrically connected to a fourth conductor 413. The emitter electrode of the active element 157 and the anode electrode of the diode 158, which constitutes the switching element of the lower arm circuit, are electrically connected by the fourth conductor 413. The collector electrode of the active element 157 and the cathode electrode of the diode 158 are electrically connected by a first conductor 410. The first conductor 410 and the second conductor 411 are electrically connected via an intermediate electrode portion 414. The AC side terminal 320B is electrically connected to the first conductor 410. The Kelvin emitter signal terminal 325K is connected to the emitter electrode of both the upper arm circuit and the lower arm circuit.The collector detect signal terminal 325C of the upper arm circuit is electrically connected to the third conductor 412, and the collector detect signal terminal 325C of the lower arm circuit is electrically connected to the first conductor 410.

[0025] The active elements 155 and 157 may be configured to include multiple active elements 155 and 157, respectively.

[0026] As in Fig. 2(a), the collector electrode of the active element 155 and the anode electrode of the diode 156 are connected to the third conductor 412 by means of a metal connecting element 51 such as solder or sintered metal. The emitter electrode of the active element 155 and the cathode electrode of the diode 156 are connected to the second conductor 411 by means of the metal connecting element 51 such as solder or sintered metal. The collector electrode of the active element 157 and the diode 158 (the anode electrode (which in Fig. 2) are connected to the first conductor 410 by means of the metal connecting element 51 such as solder or sintered metal. The emitter electrode of the active element 157 (which is shown in Fig. 2 is not illustrated) and the cathode electrode of the diode 158 are connected to the fourth conductor 413 by the metal connecting element 51 such as solder or sintered metal. The first conductor 410 is connected to the intermediate electrode portion 414 (see also Fig. 6(a)), which is integrally formed with the second conductor 411, is connected by the metal connecting member 51. As a result, the first conductor 410 and the second conductor 411 are electrically connected.

[0027] The entire bottom surface of the active elements 155 and 157 is a collector electrode, the entire bottom surface of the diodes 156 and 158 is an anode electrode, and the active area of ​​the top surface is a cathode electrode.

[0028] The first to fourth conductors 410 to 413 are formed of copper or aluminum, but other materials may be used as long as they are materials having high electrical conductivity. A collector-side wiring plate 423 is arranged on the lower surface (in the +Z direction) of the first conductor 410 and the third conductor 412. The collector-side wiring plate 423 is connected to the first conductor 410 and the third conductor 412 by a metal connecting member 51 such as solder or sintered metal. The collector-side wiring plate 423 is configured by forming a wiring 452 made of copper or aluminum on the front and back surfaces of an insulating plate 451 made of ceramic or the like. The first conductor 410 and the third conductor 412 are connected to the wiring 452 by the metal connecting member 51.The conductors and wirings to be metallically connected may be plated or provided with fine irregularities to increase the bonding strength. The electrodes of the active elements 155 and 157 are each connected to the wiring formed on the collector-side wiring board 423 by a wire 840 and are connected to the signal terminal 325 exposed to the outside of the resin 850 by a wire 841. The wires 840 and 841 may be formed of continuous wires depending on the connection layout. The connection between the electrodes of the active elements 155 and 157 and the wiring will be described later.

[0029] An emitter-side wiring board 422 is arranged on a surface of the top side (in the -Z direction) of the second conductor 411 and the fourth conductor 413. The emitter-side wiring board 422 is connected to the second conductor 411 and the fourth conductor 413 by a metal connecting member 51 such as solder or sintered metal. The emitter-side wiring board 422 is configured by forming a wiring 454 made of copper or aluminum on the front and back surfaces of an insulation board 453 made of ceramic or the like. The second conductor 411 and the fourth conductor 413 are connected to the wiring 454 formed on the emitter-side wiring board 422 by the metal connecting member 51.

[0030] In Fig. 2(a), the fin bases 800 are connected to the bottom surface of the collector-side wiring board 423 and the top surface of the emitter-side wiring board 422, respectively. The collector-side wiring board 423 or the emitter-side wiring board 422 and the fin base 800 are connected by a metal connecting member 51 such as solder or sintered metal.

[0031] The upper and lower fin bases 800 are sealed with the resin 850. The resin 850 is formed, for example, by molding such as transfer molding.

[0032] The fin base 800, which has a large number of fins 800a, is a heat dissipation member and has the connecting portion 810 extending outward from an outer peripheral end 422a of the emitter-side wiring board 422 or an outer peripheral end 423a of the collector-side wiring board 423. The connecting portion 810 has low rigidity and is formed to a thickness substantially equal to or thinner than the thickness (the length in the Z direction) from the bottom surface of the fin base 800 to the base of the fin 800a. (Slat base 800)

[0033] As in Fig. As illustrated in Fig. 2(b), the connecting portion 810 has a planar intermediate portion 804 approximately parallel to the XY plane. Further, the connecting portion 810 has first to third elastically deformed portions 801 to 803. The first elastically deformed portion 801 is formed on the outer peripheral end 810a side of the connecting portion 810. The second elastically deformed portion 802 is formed in a region corresponding to the outer peripheral end 422a of the emitter-side wiring board 422 or the outer peripheral end 423a of the collector-side wiring board 423, which is the root side of the connecting portion 810. The third elastically deformed portion 803 is formed on the side opposite to the first elastically deformed portion 801 side in the intermediate portion 804.The distance in the thickness direction (the Z direction) between the outer peripheral ends 810a of the upper and lower fin bases 800 is smaller than the distance in the thickness direction (the Z direction) between the intermediate portions 804 of the upper and lower fin bases 800.

[0034] The resin 850 covering the outer peripheral side surfaces of the first to fourth conductors 410 to 413 is filled between the connecting portions 810 of the upper and lower fin bases 800. The first to third elastically deformed portions 801 to 803 of the connecting portion 810 are formed during molding when a semiconductor device configuration 304 is formed before sealing (see Fig. 5(a)) in a form 852 (see Fig. 8(a)) is installed and a resin material 850S (see Fig. 8(a)) is fed into the mold 852. This has the effect of reducing the variation in the distance between the intermediate portions 804 of the connecting portions 810 in the upper and lower fin bases 800. This will be described later.

[0035] The shape of the connecting section 810, which is located to the left and right of the slat base 800 in Fig. 2(b) is the mold after it has been sealed with resin. The connecting portion 810 of the fin base 800 before resin sealing has the shape of a thin flat plate extending in the X direction. Then, as described later, when it has been loaded into the mold for casting, as shown in Fig. 8(a) and Fig. 8(b), the connecting portion 810 of the pair of upper and lower slat bases 800 is pressed and elastically deformed between the upper and lower molds 852a and 852b and obtains the shape shown in Fig. 2(b), in the molding process for filling resin. (Manufacturing process)

[0036] Fig. 4(a) to Fig. 4(c) are cross-sectional views in each step for explaining the manufacturing method of the semiconductor device shown in Fig. 1 is illustrated, and Fig. 5(a) to Fig. 5(b) are cross-sectional views in each process for explaining the manufacturing method of the semiconductor device based on Fig. 4(a) to Fig. 4(c) follows. Fig. 6(a) to Fig. 6(c) are perspective views of the processes, each Fig. 4(a) to Fig. 4(c), and Fig. 7(a) to Fig. 7(b) illustrate perspective views of the processes, each Fig. 5(a) to Fig. 5(b).

[0037] With reference to Fig. 4(a) to Fig. 4(c), Fig. 5(a) to Fig. 5(b), Fig. 6(a) to Fig. 6(c) and Fig. 7(a) to Fig. 7(b) shows the manufacturing process of the semiconductor device 300 shown in Fig. 1 is illustrated.

[0038] As in Fig. 4(a) and Fig. As illustrated in Figure 6(a), the collector electrode of the active element 155 and the cathode electrode of the diode 156 are connected to the third conductor 412 through the metal interconnection member 51. Similarly, the collector electrode of the active element 157 and the cathode electrode of the diode 158 are connected to the first conductor 410 through the metal interconnection member 51.

[0039] Furthermore, the emitter electrode of the active element 155 and the anode electrode of the diode 156 are connected to the second conductor 411 through the metal connecting element 51. Similarly, the emitter electrode of the active element 157 and the anode electrode of the diode 158 are connected to the fourth conductor 413 through the metal connecting element 51.

[0040] In Fig. 4(a) to Fig. 4(c), Fig. 5(a) and Fig. 5(b), the AC side terminal 320B is integrally formed in the first conductor 410 to increase productivity, however, the AC side terminal 320B may be formed separately from the first conductor 410.

[0041] Then, as in Fig. 4(b) and Fig. 6(b), the collector-side wiring board 423 is connected to the bottom surfaces of the first conductor 410 and the third conductor 412 by the metal connecting member 51, and the electrodes of the active elements 155 and 157 are electrically connected to the wiring 452 of the collector-side wiring board 423 by the wire 840, respectively. Furthermore, all the wiring 452 and all the signal terminals shown in Fig. 1 are connected by the wire 841.

[0042] The following processes are in Fig. 4(c) and Fig. 6(c). As illustrated in these drawings, the wiring 454 on the lower surface (the Z-direction side) of the emitter-side wiring board 422 is connected to the upper surfaces of the second conductor 411 and the fourth conductor 413 through the metal connecting member 51.

[0043] In this embodiment, the first conductor 410 and the third conductor 412, which are collector-side conductors, and the collector-side wiring board 423 are separated from each other. Although the thickness of the wiring 452 of the collector-side wiring board 423 is thin, the thickness of the first conductor 410 and the third conductor 412 is thick, so that heat can be dissipated in the plane direction. By reducing the thickness of the wiring 452 of the collector-side wiring board 423, the collector-side wiring board 423 can be made cheaper, and since the thickness of the wiring 452 is thin, the wiring pattern can be miniaturized, and the area of ​​the collector-side wiring board 423 is reduced, and miniaturization becomes possible.

[0044] The same applies to the emitter side, and the second conductor 411 and the fourth conductor 413, which are the emitter-side conductors, and the emitter-side wiring board 422 are separated from each other, whereby heat can be dissipated by the second conductor 411 and the fourth conductor 413 in the plane direction and the emitter-side wiring board 422 can be designed and miniaturized at low cost.

[0045] The 800 lamella bases are on the front and back of the intermediate product produced in the process of Fig. 4(c). That is, as in Fig. 5(a) and Fig. As illustrated in Fig. 7(a), the fin base 800 is connected to the bottom surface of the collector-side wiring board 423 and the top surface of the emitter-side wiring board 422, respectively, through the metal connecting member 51. The fin base 800 is made of, for example, aluminum. When the wiring 452 of the collector-side wiring board 423 and the wiring 454 of the emitter-side wiring board 422 are formed of copper, the fin base 800 is deformed due to the difference in thermal expansion between aluminum and copper. However, in this embodiment, the fin base 800 is connected through the metal connecting member 51 to the collector-side wiring board 423 connected to the first conductor 410 and the third conductor 412, and the emitter-side wiring board 422 connected to the second conductor 411 and the fourth conductor 413, respectively.Therefore, it is possible to reduce warpage when joining the fin base 800. Therefore, the joining process of the fin base 800 can be a low-pressure or no-pressure joining process instead of a pressure joining process. As a result, the cost of the production equipment can be reduced.

[0046] The bonding surface of the fin base 800 may be nickel-plated. Furthermore, the collector-side wiring board 423 and the emitter-side wiring board 422 may be connected to the fin base 800 in advance by a metal connecting member 51 or the like.

[0047] Before the semiconductor device 300 is sealed with the resin 850 as shown in Fig. 5(a) and Fig. 7(a), it is referred to as a semiconductor device configuration 304 before sealing.

[0048] The semiconductor device configuration 304 before sealing, which in the process of Fig. 5(a) and Fig. 7(a) is resin sealed. That is, the semiconductor device configuration 304 before sealing, which is provided between the pair of upper and lower fin bases 800, is sealed with the resin 850 as shown in Fig. 5(b) and Fig. 7(b). Sealing with resin 850 is performed by transfer molding. Prior to resin encapsulation, the semiconductor device configuration 304 may be coated with a thin layer of resin prior to sealing. (Shed)

[0049] Fig. 8(a) is a cross-sectional view of the process of installing the semiconductor device configuration before sealing it in a mold and performing resin molding, Fig. Figure 8(b) is an enlarged view of an area VIIIb of Fig. 8(a) and Fig. 8(c) is a side view illustrating the shape of the connecting portion 810 of the fin base 800 before the semiconductor device configuration is resin molded prior to sealing.

[0050] With reference to Fig. 8(c), the connecting portion 810 of the fin base 800 before resin molding has the shape of a thin flat plate parallel to the X-direction. Next, with reference to Fig. 8(b), when the semiconductor device configuration is loaded into the mold for resin molding before sealing, the pair of upper and lower connecting portions 810 are pressed and deformed in the Z direction at stepped portions 855 of the upper and lower molds 852a and 852b, as shown in Fig. 8(b). The details are described below.

[0051] As in Fig. 8(a), the semiconductor device configuration 304 before sealing, which is shown in Fig. 5(a), is installed in the cavity of the mold 852, which is composed of a lower mold 852a and an upper mold 852b. As described with reference to Fig. 8(c), each fin base 800 is formed with a low-rigidity connecting portion 810. When the semiconductor device configuration 304 is installed in the upper and lower molds 852a and 852b before sealing, the outer peripheral end 810a of the connecting portion 810 abuts against a first surface 857 of the stepped portion 855b of the upper mold 852b, as shown in Fig. 8(b). Accordingly, the outer peripheral end 810a, although this is shown in Fig. 8(b), to the first surface 857 of the stepped portion 855a of the lower mold 852a. The reason for this is illustrated below.

[0052] Fig. Fig. 19 is a schematic view for explaining deformation of the connecting portion of the fin base by a mold in a state where the semiconductor device configuration is installed in the mold before sealing. For description, reference is also made to Fig. 19 is referred to.

[0053] The reference number 810-1, which is indicated by a solid line in Fig. 19 indicates the shape of the connecting portion 810, which is parallel to the X direction, before deformation before resin molding.

[0054] The lower mold 852a and the upper mold 852b are formed with a stepped portion 855a and a stepped portion 855b, respectively. The structure of the stepped portion 855a of the lower mold 852a and the stepped portion 855b of the upper mold 852b are the same, and the stepped portion 855a and the stepped portion 855b are representatively described below as the stepped portion 855. Further, the stepped portion 855 has the first surface 857 and a second surface 858 extending in the X direction and facing the Z direction.

[0055] For the length in the X direction, the length between the outer peripheral ends 810a of the connecting portion 810, ie, the dimension X 810 , which in Fig. 19, greater than the length between the vertical side surfaces 856 of the stepped portion 855, ie the dimension X 856 , which in Fig. 19. The distance in the thickness direction (the Z direction) between the first surface 857 of the stepped portion 855b of the upper mold 852b and the first surface 857 of the stepped portion 855a of the lower mold 852a, ie, the dimension Z 857 , which in Fig. 19 is set to be smaller than the distance between the connecting portions 810 of the upper and lower fin bases 800 of the semiconductor device configuration 304 before sealing in the thickness direction (the Z direction), that is, the dimension Z 810 , which in Fig. 19. Further, the distance between a second surface 858 of the stepped portion 855b of the upper mold 852b and the second surface 858 of the stepped portion 855a of the lower mold 852a in the thickness direction (the Z direction), ie, the dimension Z 858 , which in Fig. 19, is set to be greater than the distance Z 810 between the connecting portions 810 of the upper and lower fin bases 800 of the semiconductor device configuration before sealing 304.

[0056] The first surface 857 and the second surface 858 of the stepped portions 855a and 855b of the mold 852 are formed in a flat surface substantially parallel to the XY surface.

[0057] As described above, the dimensions of the connecting portion 810 of the fin base 800 and the first surface 857 and the second surface 858 of the stepped portions 855a and 855b of the mold are adjusted as described above. Therefore, when the semiconductor device configuration 304 is installed in the cavity of the mold 852 before sealing and the mold 852 is closed, the portions near the outer peripheral ends 810a of the connecting portions 810 of the upper and lower fin bases 800 respectively correspond to corners where the first surface 857 of the stepped portion 855b or the stepped portion 855a and a vertical side surface 856 intersect, as shown in Fig. 8(a) and Fig. 8(b), and a connecting portion 810-1 before deformation, which is shown by a solid line in Fig. 19 is bent like a connecting portion 810-2 after deformation indicated by a two-dot chain line 810-2.

[0058] Fig. 9(a) is a diagram for explaining the action of the resin in a state where the resin is molded into the shape shown in Fig. 8(a), and Fig. 9(b) is an enlarged view of an area XIb of Fig. 9(a).

[0059] The resin material 850S is supplied in the state shown in Fig. 8(a) and Fig. 8(b) is fed into the mold 852. The resin material 850S flows into the cavity of the mold 852 and is filled between the connecting portions 810 of the upper and lower fin bases 800 to seal the outer peripheral side surfaces of the first to fourth conductors 410 to 413. As described above, in this state, the connecting portions 810 of the upper and lower fin bases 800 are pressed against the first surface 857 of the stepped portions 855b and 855a, respectively. Therefore, the resin material 850S fed between the upper and lower fin bases 800 is suppressed from leaking at the contact portion between the connecting portion 810 of the upper and lower fin bases 800 and the first surface 857 of the stepped portion 855b or the stepped portion 855a, and no loss to the second surface 858 side of the step 855b or the step 855a occurs.

[0060] When the pre-sealing semiconductor device configuration 304, to which the pair of upper and lower fin bases 800 are connected, is strongly clamped by the mold 852, excessive stress is generated in the active elements 155 and 157, etc. However, since the fin base 800 is provided with the connecting portion 810 and the fin base 800 is configured to bend with a small load at the low-rigidity connecting portion 810, the stress acting on the active elements 155 and 157, etc., can be relaxed.

[0061] Furthermore, as in Fig. 8(a), a spring mechanism 864 is provided in the mold 852. The spring mechanism 864 has a peeling prevention function, which acts on the active elements 155 and 157, etc., via the first to fourth conductors 410 to 413 and the collector-side / emitter-side wiring boards 422 and 423. Peeling is the following phenomenon. Namely, a hydrostatic pressure Ps, which expands the space between the fin bases 800, acts through the resin material 850S arranged around the semiconductor device configuration 304 before sealing, which is installed in the cavity of the mold 852 (see Fig. 9(a) and Fig. 9(b)), is filled on the upper and lower fin bases 800. Therefore, a peeling force acts on the active elements 155 and 157, etc., via the first to fourth conductors 410 to 413 and the collector-side / emitter-side wiring boards 423 and 422. By making the pressing force on the semiconductor device configuration 304 before sealing by the spring mechanism 864 larger than the pressing force on the semiconductor device configuration 304 before sealing generated by the mold clamping forces of the upper and lower molds 852a and 852b, the peeling force acting on the active elements 155 and 157, etc., can be suppressed.

[0062] The active elements 155 and 157, etc., are strong against the pressing force but weak against the peeling force, causing breakage or failure. By making the pressing force on the semiconductor device configuration 304 before sealing by the spring mechanism 864 greater than the peeling force generated by the pressure of the resin material 850S, it is possible to prevent the active elements 155 and 157, etc., from being destroyed or damaged during resin molding.

[0063] As in Fig. 9(a), the resin material 850S having flowability before curing flows into the semiconductor device configuration 304 before sealing installed in the cavity of the mold 852, and thus the pressure applied to the resin material 850S is applied as the hydrostatic pressure Ps to the mold 852 and the semiconductor device configuration 304 before sealing.

[0064] As in Fig. As illustrated in Fig. 9(b), the hydrostatic pressure Ps generated by the resin material 850S deforms the connecting portion 810 of the fin base 800 and presses the second surface 858 of the stepped portions 855a and 855b of the upper and lower molds 852a and 852b. At this time, the first elastically deformed portion 801, the second elastically deformed portion 802, the third elastically deformed portion 803, and the intermediate planar portion 804 are formed in the connecting portion 810.

[0065] As in Fig. As illustrated in Fig. 9(b), the connecting portion 810 of the fin base 800 is deformed by the first elastically deformed portion 801 such that the outer peripheral end 810a is located above the first surface 857 of the stepped portions 855a and 855b. The connecting portion 810 of the fin base 800 is also deformed by the second elastically deformed portion 802 and the third elastically deformed portion 803, and the intermediate portion 804 is deformed to be flat according to the surface of the second surface 858 of the mold 852. The connecting portion 810 of the fin base 800 is deformed by the second elastically deformed portion 802 on the root side such that it extends obliquely outward and toward the third elastically deformed portion 803.

[0066] When the injection pressure of the resin material 850S is 5 MPa, if the connecting portion 810 is made of an aluminum material of 0.6 mm or less, the connecting portion 810 having the first to third elastically deformed portions 801 to 803 and the intermediate portion 804 can be formed.

[0067] Normally, due to component tolerances and variations during assembly, a variation of about 0.1 mm occurs in a height position in the thickness direction of the connecting portion 810 of the upper and lower fin bases 800 of the semiconductor device configuration before sealing 304, in other words, the Z direction (which may be simply referred to as "height position" hereinafter), in one semiconductor device configuration 304 before sealing itself. Furthermore, a variation of about 0.2 mm occurs in the plurality of semiconductor device configurations 304 before sealing.

[0068] On the other hand, in this embodiment, the intermediate portions 804 of the connecting portions 810 of the upper and lower fin bases 800 are pressed against the flat second surface 858 of the upper and lower molds 852a and 852b by the resin materials 850S during molding. The connecting portion 810 is formed by elastic deformation to maintain this state. That is, even if the height positions of the intermediate portions 804 of the connecting portions 810 of the upper and lower fin bases 800 of the semiconductor device configuration 304 vary before sealing, the height positions of the intermediate portions 804 of all the connecting portions 810 of the semiconductor device configuration 304 before sealing can be adjusted to the position of the second surface 858 of the upper and lower molds 852a and 852b.Therefore, the variation in the height position of the intermediate portions 804 of the connecting portions 810 of the upper and lower fin bases 800 of the semiconductor device 300 can be extremely reduced. In the present inventor's study, the variation in the height position of the intermediate portions 804 of the connecting portions 810 of the upper and lower fin bases 800 could be made approximately 0.01 mm even among the plurality of semiconductor devices 300.

[0069] In the description of the elastic deformation of the connecting portion 810 of the lamella base 800, this was described with respect to the X-direction. However, as can be seen from Fig. 10, the connecting portion 810 of the lamella base 800 itself elastically deforms in the Y direction similar to the X direction.

[0070] Although not illustrated, the assembly of the first to fourth conductors 410 to 413, the power terminals, and the signal terminals is performed in a state where the first to fourth conductors 410 to 413, the power terminals, and the signal terminals are connected by a tensile bar until resin molding. After resin molding, the tensile bar is cut off, and the power terminal and the signal terminal are processed into a predetermined shape, thereby manufacturing the semiconductor device 300 formed in Fig. 1 is illustrated.

[0071] Fig. 10 is a cross-sectional view illustrating an example of the first embodiment of the semiconductor module according to the invention. A semiconductor module 900 includes the semiconductor device 300 and the flow path forming body 600.

[0072] As described above, in the semiconductor device 300, the fin bases 800 are arranged above and below the first to fourth conductors 410 to 413, and the first to fourth conductors 410 to 413 are sealed by the resin 850 filled between the connecting portions 810 of the fin base 800. The connecting portion 810 has the intermediate portion 804 exposed on the top and bottom surfaces (in the Z direction) of the resin 850.

[0073] The flow path forming body 600 has an upper housing 601a and a lower housing 601b. The upper housing 601a is connected to the upper (Z-direction) fin base 800, and the lower housing 601b is connected to the lower (Z-direction) fin base 800. The connection structure between the upper housing 601a and the fin base 800 and the connection structure between the lower housing 601b and the fin base 800 are the same. Hereinafter, the upper housing 601a and the lower housing 601b are represented by the housing 601, and the connection structure between the housing 601 and the fin base 800 will be described.

[0074] The casing 601 includes a base portion 602 having a rectangular frame shape in plan view, and a cover portion 603 integrally formed with the base portion 602. The base portion 602 is formed in a planar shape substantially parallel to the XY plane and is connected to the intermediate portion 804 of the connecting portion 810 of the fin base 800. The cover portion 603 rises from the base portion 602 to a height at which a gap is formed between the base portion 602 and the tip of the fin 800a of the fin base 800. The gap between the cover portion 603 and the fin 800a of the fin base 800 forms a cooling flow path Cw through which a refrigerant such as water flows.

[0075] The cooling flow path Cw between the upper housing 601a and the lower housing 601b is provided with a refrigerant inlet 13 (see Fig. 16) and a refrigerant outlet 14 (see Fig. 16) communicating with each other in a region (not illustrated). The casing 601a and the lower casing 601b are assembled to form the flow path forming body 600.

[0076] The base portion 602 of the housing 601 and the intermediate portion 804 of the connecting portion 810 of the fin base 800 are connected by a connecting portion 650. The connecting portion 650 is formed over the entire circumference in the resin 850 that seals the outer peripheral side surfaces of the first to fourth conductors 410 to 413.

[0077] Bonding or welding using resin can be used to join the base portion 602 of the case 601 and the connecting portion 810 of the fin base 800, but welding, which has excellent durability, is preferred. Laser welding can be used as the joining by welding. Generally, with laser welding, the risk of welding defects increases when a gap of 0.1 mm or more is created between the members to be joined. As described above, in this embodiment, the variation in the height position of the intermediate portions 804 of the connecting portions 810 of the upper and lower fin bases 800 could be made to about 0.01 mm even among the plurality of semiconductor devices 300.Therefore, the reliability and workability of the joining between the flow path forming body 600 and the fin base 800 can be improved, and thus the productivity can be improved.

[0078] Fig. 11 is a cross-sectional view illustrating another example of the first embodiment of the semiconductor module according to the invention.

[0079] A 900A semiconductor module that Fig. 11 includes two semiconductor devices 300 and a flow path forming body 600A. The flow path forming body 600A includes an upper case 601c and a lower case 601d. The upper case 601c is connected to the upper (Z-direction) fin base 800, and the lower case 601d is connected to the lower (Z-direction) fin base 800. The connection structure between the upper case 601c and the fin base 800 and the connection structure between the lower case 601d and the fin base 800 are the same. Hereinafter, the upper case 601c and the lower case 601d are represented by the case 601, and the connection structure between the case 601 and the fin base 800 will be described.

[0080] The housing 601 has a frame-like shape, is arranged between the two semiconductor devices 300, and is connected to both semiconductor devices 300. That is, the housing 601 is connected to the connecting portion 810 of the fin base 800 of one semiconductor device 300 at the connecting portion 650 and is connected to the connecting portion 810 of the fin base 800 of the other semiconductor device 300 at the connecting portion 650.

[0081] The upper housing 601c and the lower housing 601d are connected to the refrigerant inlet 13 (see Fig. 16) and the refrigerant outlet 14 (see Fig. 16) communicating with each other in a region (which is not illustrated), and the upper case 601c and the lower case 601d are assembled to form the flow path forming body 600A.

[0082] In addition, in the embodiment described in Fig. As illustrated in FIG. 11, in this embodiment, the fluctuations in the height position of the intermediate portions 804 of the connecting portions 810 of the upper and lower fin bases 800 of the semiconductor device 300 can be made to be about 0.01 mm even among the plurality of semiconductor devices 300. The upper case 601c and the lower case 601d are connected to the intermediate portion 804 of the connecting portion 810 of the fin base 800, which has such a small fluctuation. Therefore, the flow path forming body 600A and the fin base 800 can be well and effectively connected in terms of strength and reliability.

[0083] In Fig. 11, the semiconductor module 900A is illustrated as a structure in which two semiconductor devices 300 are connected by a flow path formation body 600A. However, the number of semiconductor devices 300 may be three or more, and the adjacent semiconductor devices 300 may be connected by the flow path formation body 600A.

[0084] According to the first embodiment, the following effects are obtained.

[0085] (1) The semiconductor modules 900 and 900A include a semiconductor device 300 including a first fin base (a heat dissipation member) 800 having a first connecting portion 810, a second fin base (a heat dissipation member) 800 having a second connecting portion 810, and a resin 850 for sealing the outer peripheral side surfaces of the first to fourth conductors 410 to 413, and flow path forming bodies 600 and 600A connected to the first connecting portion 810 of the first fin base 800 and the second connecting portion 810 of the second fin base 800.The first elastically deformed portion 801 is provided such that the gap in the thickness direction between the outer peripheral end 810a of the first connecting portion 810 of the first fin base 800 and the outer peripheral end 810a of the second connecting portion 810 of the second fin base 800 is smaller than the gap in the thickness direction between the intermediate portion 804 of the first connecting portion 810 of the first fin base 800 and the intermediate portion 804 of the second connecting portion 810 of the second fin base 800. The resin 850 is filled between the first connecting portion 810 of the first fin base 800 and the second connecting portion 810 of the second fin base 800. According to this configuration, the semiconductor modules 900 and 900A can be assembled only by connecting the flow path forming bodies 600 and 600A and the first and second fin bases 800 of the semiconductor device 300.Therefore, the procedure of aligning the end faces of the peripheral side portions of the upper and lower cases is not required, and the productivity of the semiconductor modules 900 and 900A can be improved.

[0086] Since the gap in the thickness direction between the outer peripheral ends 810a of the first and second fin bases 800 is elastically deformed to be smaller than the gap in the thickness direction between the intermediate portions 804, it is possible to prevent the resin material 850S from leaking outward from the first and second connecting portions 810 during molding.

[0087] (2) In the method for manufacturing the semiconductor module, the outer peripheral end 810a of each of the first connecting portion 810 of the first fin base 800 and the second connecting portion 810 of the second fin base 800 is brought into contact with the step (adjacent portion) 855 of the mold 852, the resin material 850S is filled between the first connecting portion 810 of the first fin base 800 and the second connecting portion 810 of the fin base 800, and the first connecting portion 810 of the first fin base 800 and the second connecting portion 810 of the second fin base 800 are each elastically deformed such that the gap in the thickness direction between the intermediate portion 804 of the first connecting portion 810 of the first fin base 800 and the intermediate portion 804 of the second connecting portion 810 of the fin base 800 is larger than the gap in the thickness direction between the steps 855.

[0088] According to this method, the first and second connecting portions 810 of the first and second fin bases 800 are elastically deformed by the pressure during the injection of the resin material 850S. Therefore, it is not necessary to separately perform the procedure of elastically deforming the first and second connecting portions 810 of the first and second fin bases 800, and productivity is improved.

[0089] In addition, the resin material 850S fed between the upper and lower fin bases 800 is suppressed from leaking at the adjacent portion between the connecting portion 810 of the upper and lower fin bases 800 and the stepped portion 855b or the stepped portion 855a, and no loss to the second surface 858 side of the step 855b or the step 855a occurs.

[0090] Furthermore, the connecting portion 810 of the upper and lower fin bases 800 is elastically deformed to be held at a position pressed against a bottom surface 858 of the stepped portion 855 of the mold 852. That is, even if the gap between the connecting portions 810 of the upper and lower fin bases 800 of the semiconductor device configuration 304 varies before sealing, the Z-height positions of the intermediate portions 804 of all the connecting portions 810 of the semiconductor device configuration 304 before sealing can be adjusted to the position of the bottom surfaces 858 of the upper and lower molds 852a and 852b. Therefore, the fluctuation of the gap between the intermediate portions 804 of the connecting portions 810 of the upper and lower fin bases 800 can be extremely reduced.As a result, the reliability and workability of joining the upper and lower fin bases 800 and the flow path forming bodies 600 and 600A can be improved, and thus the productivity can be improved.

[0091] Fig. 12 is a circuit diagram of a power conversion device using the semiconductor module according to the invention.

[0092] A power conversion device 200 includes inverter circuit units 140 and 142, an inverter circuit unit 43 for auxiliary equipment, and a capacitor module 500. The inverter circuit units 140 and 142 include a plurality of power semiconductor devices 300 and form a three-phase bridge circuit by connecting them. Further, when a current capacity is large, the semiconductor devices 300 are connected in parallel, and these parallel connections are made for each phase of the three-phase inverter circuit so that the current capacity can be increased. Furthermore, the increase in current capacity can be accommodated by connecting the active elements 155 and 157 and the diodes 156 and 158, which are power semiconductor elements incorporated in the semiconductor device 300, in parallel.

[0093] The inverter circuit unit 140 and the inverter circuit unit 142 have the same basic circuit configuration, and the control method and operation are substantially the same. Since the outline of the circuit operation of the inverter circuit unit 140 and the like are well known, a detailed description will be omitted here.

[0094] As described above, the upper-arm circuit includes an upper-arm active element 155 and an upper-arm diode 156 as a switching power semiconductor element. The lower-arm circuit includes a lower-arm active element 157 and a lower-arm diode 158 as a switching power semiconductor element. The active elements 155 and 157 receive a drive signal output from one or the other of the two drive circuits constituting a drive circuit 174, perform a switching operation, and convert DC power supplied from a battery 136 into three-phase AC power.

[0095] As described above, the upper arm active element 155 and the lower arm active element 157 include a collector electrode, an emitter electrode, and a gate electrode. The upper arm diode 156 and the lower arm diode 158 include two electrodes, a cathode electrode and an anode electrode. As shown in Fig. As illustrated in Figure 3, the cathode electrodes of the diodes 156 and 158 are electrically connected to the collector electrodes of the IGBTs 155 and 157, respectively, and the anode electrodes are electrically connected to the emitter electrodes of the active elements 155 and 157, respectively. As a result, the current flows from the emitter electrode of the upper-arm active element 155 and the lower-arm active element 157 to the collector electrode in the forward direction.

[0096] Furthermore, a MOSFET (metal oxide semiconductor field-effect transistor) can be used as the active element. In this case, the diode 156 for the upper arm and the diode 158 for the lower arm become unnecessary.

[0097] The positive electrode side terminal 315B and the negative electrode side terminal 319B of each upper / lower arm series circuit are connected to the DC terminals for connecting the capacitors of the capacitor module 500, respectively. AC power is generated in each connection portion between the upper arm circuit and the lower arm circuit, and the connection portion between the upper arm circuit and the lower arm circuit of each upper / lower arm series circuit is connected to the AC side terminal 320B of each semiconductor device 300. The AC side terminal 320B of each semiconductor device 300 of each phase is connected to the AC output terminal of the power conversion device 200, and the generated AC power is supplied to the stator winding of a motor generator 192 or 194.

[0098] A control circuit 172 generates a timing signal for controlling a switching timing of the upper-arm active element 155 and the lower-arm active element 157 based on input information from a control device or a sensor (e.g., the current sensor 180) on the vehicle side. The drive circuit 174 generates a drive signal for switching the upper-arm active element 155 and the lower-arm active element 157 based on the timing signal output from the control circuit 172.

[0099] It should be noted that 181, 182 and 188 are connectors.

[0100] The upper / lower arm series circuit includes a temperature sensor (not illustrated), and the temperature information from the upper / lower arm series circuit is input to a microcomputer. Furthermore, voltage information about the positive DC electrode side from the upper / lower arm series circuit is input to the microcomputer. The microcomputer performs overtemperature detection and overvoltage detection based on the information. When overtemperature or overvoltage is detected, the microcomputer stops the switching operation of all upper arm active elements 155 and lower arm active elements 157, protecting the upper / lower arm series circuit from overtemperature or overvoltage.

[0101] Fig. 13 is an external perspective view showing an example of the power conversion device used in Fig. 12 is illustrated, and Fig. 14 is a cross-sectional view taken along the line XIV-XIV of the power conversion device shown in Fig. 13 is illustrated. Fig. 15(a) is a perspective view of the power conversion device shown in Fig. 14 is illustrated, seen from above, Fig. 15(b) is a perspective view of the power conversion device shown in Fig. 14, seen from below and Fig. 16 is a cross-sectional view taken along the line XVI-XVI of Fig. 15(a) was taken.

[0102] The power conversion device 200 is composed of a lower case 11 and an upper case 10, and includes a case 12 formed in a substantially regular parallelepiped shape. A semiconductor module 900B, the capacitor module 500, and the like, which are arranged in Fig. 15 are housed in the housing 12. The semiconductor module 900B includes a flow path forming body 600B. A refrigerant inflow line 13 and a refrigerant outflow line 14 connected to the cooling flow path Cw (see Fig. 10) of the flow path forming body 600B, protrude from a side surface of the housing 12. As shown in Fig. As illustrated in Fig. 14, the lower case 11 has an opening on the top, and the upper case 10 is attached to the lower case 11 by closing the opening of the lower case 11. The upper case 10 and the lower case 11 are formed of an aluminum alloy or the like, and are sealed and fixed to the outside. The upper case 10 and the lower case 11 may be integrated. Since the case 12 has a simple regular parallelepiped shape, it can be easily attached to a vehicle or the like, and it can be easily manufactured.

[0103] As in Fig. 13, a connector 17 is attached to a side surface of the housing 12 in the longitudinal direction, and an AC terminal 18 is connected to the connector 17. Further, a connector 21 is provided on the surface from which the refrigerant inflow line 13 and the refrigerant outflow line 14 are led out.

[0104] As in Fig. 14, the semiconductor module 900B is housed in the housing 12. The control circuit 172 and the drive circuit 174 are arranged on the top side of the semiconductor module 900B, and a capacitor module 500 is housed on the bottom side of the semiconductor module 900. As shown in Fig. 15(a) and Fig. 15(b), the semiconductor module 900B has a 6-in-1 structure including three semiconductor devices 300 having a 2-in-1 structure. That is, one of the inverter circuit units 140 and 142 shown in Fig. 12 is included. In addition, Fig. 15(b) to illustrate the arrangement of the semiconductor device 300, the fin base 800 through the flow path forming body 600.

[0105] The AC side terminal 320B of the semiconductor device 300 penetrates the current sensor 180 and is connected to a bus bar 361. Further, the positive electrode side terminal 315B and the negative electrode side terminal 319B, which are DC terminals of the semiconductor device 300, are connected to positive and negative electrode terminals 362A and 362B of the capacitor module 500, respectively.

[0106] In the semiconductor device 300, which is in Fig. As illustrated in Fig. 14, the AC-side terminal 320B is not bent and is straight. Furthermore, the positive electrode side terminal 315B and the negative electrode side terminal 319B have a short shape that is cut off on the root side.

[0107] The power conversion device 200 is manufactured such that the capacitor module 500 is housed in the lower case 11, the semiconductor module 900B, which has been manufactured in advance, is housed in the capacitor module 500, and the control circuit 172 and the drive circuit 174 are housed in the semiconductor module 900B. When the semiconductor module 900B is housed, the AC side terminal 320B of each semiconductor device 300 is connected to the bus bar 361, and the positive electrode side terminal 315B and the negative electrode side terminal 319B are connected to the positive and negative electrode terminals 362A and 362B of the capacitor module 500, respectively.When the control circuit 172 and the drive circuit 174 are incorporated, the signal terminal of each semiconductor device 300 and the connection terminals (not illustrated) of the control circuit 172 and the drive circuit 174 are connected. The power conversion device 200 shown in FIG. Fig. 13 can be obtained by accommodating the semiconductor module 900B, the capacitor module 500, the control circuit 172 and the drive circuit 174 in the lower case 11 and then sealing them with the upper case 10.

[0108] How is Fig. 15(a), Fig. 15(b) and Fig. As illustrated in FIG. 16, the semiconductor module 900B has an elongated regular parallelepiped shape. The flow path forming body 600B of the semiconductor module 900 is formed of iron, an aluminum alloy, or the like.

[0109] As in Fig. 16, the flow path forming body 600B has a structure in which the flow path forming body 600 shown in Fig. 10, and the flow path forming body 600A shown in Fig. 11, and includes an upper flow path cover 610, a flow path case 620, and a lower flow path cover 630. The flow path case 620 is provided with a frame 621 for connecting adjacent semiconductor devices 300 to each other. The upper flow path cover 610 and the flow path case 620 are assembled by a fastener (not illustrated). The flow path case 620 and the lower flow path cover 630 are connected via an O-ring 631 to form a waterproof structure.

[0110] As in Fig. 11, the frame 621 is connected by a connecting portion 650 of the connecting portion 810 of the fin base 800 of each adjacent semiconductor device 300 and connects the semiconductor devices 300 on both sides.

[0111] The lower flow path cover 630 is provided with the refrigerant inflow line 13 and the refrigerant outflow line 14. The flow path case 620 is formed with a throughflow path 612 penetrating the frame 621 in the thickness (the Z direction). The refrigerant flowing in from the refrigerant inflow line 13 flows through the flow path provided between the bottom side (Z direction) of each semiconductor device 300 and the lower flow path cover 630 to cool each semiconductor device 300 from the bottom. Further, the refrigerant flowing in from the refrigerant inflow line 13 flows through the flow path provided between the top side (Z direction) of each semiconductor device 300 and the upper flow path cover 610 via the throughflow path 612 to cool each semiconductor device 300 from the top.The refrigerant that cools each semiconductor device 300 from the top and bottom flows out from the refrigerant outflow line 14. The cooling flow path Cw for cooling each semiconductor device 300 is formed in the lower flow path cover 630, the flow path case 620, and the upper flow path cover 610.

[0112] In this way, the power conversion device 200 having a 6-in-1 structure is formed using three semiconductor devices 300 having a 2-in-1 structure. - Second embodiment -

[0113] Fig. 17 is a cross-sectional view illustrating a second embodiment of the semiconductor module according to the invention.

[0114] In the second embodiment, a thick portion 811 is provided in the intermediate portion 804 of the connecting portion 810 of the fin base 800 of the semiconductor device 300.

[0115] The thick portion 811 is provided so as to protrude to the side opposite to the flow path forming body 600A side of the connecting portion 810. Since the thick portion 811 is provided in the intermediate portion 804 between the first elastically deformed portion 801 and the third elastically deformed portion 803, the connecting portion 810 can be elastically deformed by the first to third elastically deformed portions 801 to 803 without inhibiting the elastic deformation of the connecting portion 810. When the connecting portion 810 of the fin base 800 and the flow path forming body 600A are joined by laser welding, the laser can penetrate the connecting portion 810 of the fin base 800 and the flow path forming body 600A in the thickness direction due to the fluctuation of the laser output.By providing the thick portion 811 in the connecting portion 810 of the fin base 800, it is possible to suppress such penetration of the laser and ensure the bonding strength.

[0116] Other configurations in the second embodiment are similar to those in the first embodiment.

[0117] Therefore, the second embodiment has the same effect as the first embodiment. - Third embodiment -

[0118] Fig. 18(a) is a cross-sectional view of a third embodiment of the semiconductor module according to the invention and Fig. 18(b) is an enlarged cross-sectional view of a process of forming the connecting portion of the slat base shown in Fig. 18(a).

[0119] In the third embodiment, the semiconductor device 300 has a structure in which a recess 812 having a triangular cross section is provided in the intermediate portion 804 of the connecting portion 810 of the fin base 800. The recess 812 is provided in a ring shape over the entire circumference at substantially the center of the width (the length in the X direction) of the intermediate portion 804 of the connecting portion 810, and the connecting portion 810 is divided into two planar portions extending in the X direction and the -X direction with the recess 812 as a boundary.

[0120] The mold 852 is provided with a projection 860 having a triangular cross-section for forming the recess 812 in a ring shape over the entire circumference.

[0121] When the resin material 850S is fed into the mold 852, the resin material 850S is filled between the connecting portions 810 of the upper and lower fin bases 800, and the connecting portion 810 of each fin base 800 is pressed toward the second surface 858 of the stepped portion 855b by the resin pressure. Since the mold 852 is formed with the protrusion 860, the connecting portion 810 is formed with elastically deformed portions 805, 806, and 807 on the upper surface and the root portions on both sides of the upper surface, which are the corners of the protrusion 860, respectively, and this state is maintained.

[0122] As in Fig.As illustrated in Fig. 18(a), the upper and lower casings 601 of the flow path forming body 600A are connected to the connecting portion 810 of the fin base 800 of each semiconductor device 300 through two connecting portions 650. The two connecting portions 650 are sequentially provided in the X-direction flat portion region and the -X-direction flat portion region of the recess 812 of the connecting portion 810.

[0123] In this way, the connecting portion 810 of a fin base 800 is connected to each housing 601 of the flow path forming body 600 at two locations to ensure the reliability of the connection strength and improve the waterproof performance.

[0124] The other configurations in the third embodiment are the same as those in the first embodiment.

[0125] Therefore, the third embodiment has the same effect as the first embodiment.

[0126] In the embodiment described above, the semiconductor modules 900 and 900A are illustrated as a 2-in-1 structure in which one pair of upper arms and one lower circuit are provided, or a 6-in-1 structure in which three pairs are provided. However, the semiconductor modules 900 and 900A may also have a 3-in-1 structure or a 4-in-1 structure.

[0127] The 3-in-1 structure has, for example, a structure in which three circuits of an upper branch are encapsulated, or a structure in which three circuits of a lower

[0128] A semiconductor module having a 6-in-1 structure can be formed by combining an upper-arm package in which three upper-arm circuits are packaged and a lower-arm package in which three lower-arm circuits are packaged.

[0129] In the above-described embodiment, the structure in which the first to fourth conductors 410 to 413 and the wirings 452 and 454 of the wiring boards 422 and 423 are connected, and the wirings 452 and 454 of the wiring boards 422 and 423 and the fin base 800 are connected by the metal connecting member 51 was illustrated. However, other joining methods such as conductive adhesive, welding, and fusion bonding by ion beam irradiation may be used instead of joining with the metal connecting member 51. List of reference symbols 155, 157 Active element (semiconductor element) 156, 158 Diode (semiconductor element) 200 power conversion device 300 semiconductor devices 304 Pre-sealing semiconductor device configuration 410 to 413 First to fourth leaders 422 Emitter-side wiring board 423 Collector-side wiring board 451, 453 insulation plate 600, 600A, 600B flow path forming bodies 601 Housing (Frame) 601a, 601c Upper case (frame) 601b, 601d Lower case (frame) 602 Base section (frame) 603 Cover section 621 frames 650 connecting section 800 fin base (heat dissipation element) 800a slat 801 to 803 First to third elastically deformed sections 804 Intermediate section (flat section) 810 connecting section 810a outer circumference end 811 Thick Section 812 recess 850 resin 850S resin material 852 Shape 855, 855a, 855b Stepped section (adjacent section) 860 lead 900, 900A semiconductor module Cw cooling flow path Ps Hydrostatic pressure

Claims

[1] Semiconductor module (900, 900A) comprising: a semiconductor device (300) comprising: a semiconductor element (155, 156, 157, 158), a pair of conductors (410 to 413) arranged so that the semiconductor element (155, 156, 157, 158) is located therebetween, facing each other in a thickness direction and each connected to the semiconductor element (155, 156, 157, 158), a first heat dissipation member (800) disposed on a surface of a side opposite to the semiconductor element (155, 156, 157, 158) of one conductor (410 to 413) of the conductor pair (410 to 413) via an insulation member (451, 453) and including a first connecting portion (650) extending outwardly from an outer peripheral side surface of the one conductor (410 to 413), a second heat dissipation member disposed on a surface of a side opposite to the semiconductor element (155, 156, 157, 158) of the further conductor (410 to 413) of the conductor pair (410 to 413) via an insulating member (451, 453) and including a second connecting portion (810) extending outwardly from an outer peripheral side surface of the further conductor (410 to 413), and a resin (850) for sealing the outer peripheral side surfaces of the conductor pair (410 to 413); and a flow path forming body (600, 600A, 600B) connected to the first connecting portion (650) of the first heat dissipation element (800) and the second connecting portion of the second heat dissipation element of the semiconductor device (155, 156, 157, 158), wherein a first elastically deformed portion (801 to 803) which is elastically deformed is provided such that a distance in a thickness direction between an outer peripheral end of the first connecting portion of the first heat dissipation member (800) and an outer peripheral end of the second connecting portion (810) of the second heat dissipation member is smaller than a distance in a thickness direction between an intermediate portion (804) of the first connecting portion (650) of the first heat dissipation member (800) and an intermediate portion of the second connecting portion (810) of the second heat dissipation member, and the resin (850) is filled between the first connecting portion (650) of the first heat dissipation element (800) and the second connecting portion (810) of the second heat dissipation element. [2] The semiconductor module (900, 900A) according to claim 1, wherein the first heat dissipation member (800) and the second heat dissipation member each include fins (800a) for heat dissipation. [3] Semiconductor module (900, 900A) according to claim 1, wherein both the first connecting portion (650) of the first heat dissipation member (800) and the second connecting portion (810) of the second heat dissipation member include a flat portion between a root portion facing the outer peripheral side surface of the conductor and the outer peripheral end, wherein the root portion is a corner of a projection (860) of a mold (852) into which the resin (850) is pressed. [4] The semiconductor module (900, 900A) according to claim 3, wherein the root portion of each of the first connecting portion (605) of the first heat dissipation member (800) and the second connecting portion (810) of the second heat dissipation member includes a second elastically deformed portion elastically deformed such that a distance in a thickness direction toward the flat portion side becomes larger. [5] The semiconductor module (900, 900A) according to claim 3, wherein the flow path forming body (600, 600A, 600B) is connected to each of the first connecting portion (605) and the second connecting portion (810) in the planar portion of both the first connecting portion (605) of the first heat dissipation member (800) and the second connecting portion (810) of the second heat dissipation member. [6] The semiconductor module (900, 900A) according to claim 5, wherein the flow path forming body (600, 600A, 600B) is connected to the first connecting portion (650) of the first heat dissipation member (800) and the second connecting portion (810) of the second heat dissipation member by welding. [7] The semiconductor module (900, 900A) according to claim 1, wherein the flow path forming body (600, 600A, 600B) includes a cover portion (603) provided to cover a surface on a side opposite to the conductor (410 to 413) of each of the first heat dissipation member (800) and the second heat dissipation member, and forms a flow path for a refrigerant between the conductor (410 to 413) of each of the first heat dissipation member (800) and the second heat dissipation member and the opposite surface. [8] A semiconductor module (900, 900A) according to claim 1, comprising: a plurality of semiconductor devices (300), wherein the flow path forming body (600, 600A, 600B) includes a frame (621) that connects the first connecting portions (650) of the first heat dissipation member (810) of the adjacent semiconductor devices (300) to each other and the second connecting portions (810) of the second heat dissipation member of the semiconductor devices (300) to each other. [9] A power conversion device (200) which accommodates a plurality of semiconductor modules (900, 900A) according to any one of claims 1 to 8 in the flow path forming body (600, 600A, 600B). [10] A manufacturing method of the semiconductor module (900, 900A) according to any one of claims 1 to 8, comprising: Bringing the outer peripheral end of each of the first connecting portion (650) of the first heat dissipation member (800) and the second connecting portion (810) of the second heat dissipation member into contact with an adjacent portion of a mold; Filling the resin (850) between the first connecting portion (650) of the first heat dissipation element (800) and the second connecting portion (810) of the second heat dissipation element and elastically deforming both the first connecting portion (650) of the first heat dissipation member (800) and the second connecting portion (810) of the second heat dissipation member such that a distance in a thickness direction between an intermediate portion (804) of the first connecting portion (650) of the first heat dissipation member (800) and an intermediate portion of the second connecting portion (810) of the second heat dissipation member is greater than a distance in a thickness direction between the adjacent portions.

Images