Semiconductor device

Abstract

A semiconductor device according to the present invention comprises at least one terminal. The terminal includes a cylindrical holder having conductivity, and a metal pin inserted into the holder. The semiconductor device further comprises a terminal support that supports the holder, and a sealing resin that covers the terminal support and a portion of the holder. The sealing resin has a resin main surface facing one side in the thickness direction. The holder has a first surface positioned at an edge thereof on one side in the thickness direction, and a first outer side surface extending in the thickness direction. The first surface is at a position different from the resin main surface in the thickness direction. The first outer side surface is in contact with the sealing resin. The metal pin protrudes further to one side in the thickness direction than the resin main surface.

Description

Semiconductor Devices

[0001] The present disclosure relates to semiconductor devices.

[0002] Conventionally, semiconductor devices including power switching elements such as MOSFETs (Metal Oxide Semiconductor Field Effect Transistors) and IGBTs (Insulated Gate Bipolar Transistors) have been known. Such semiconductor devices are installed in a wide variety of electronic devices, from industrial equipment to home appliances, information terminals, and automotive equipment. Patent Document 1 (Patent Document 1) discloses a conventional semiconductor device (power module). The semiconductor device described in Patent Document 1 includes a semiconductor element and a support substrate (ceramic substrate). The semiconductor element is, for example, an IGBT made of silicon (Si). The support substrate supports the semiconductor element. The support substrate includes an insulating base material and conductor layers stacked on both sides of the base material. The base material is, for example, made of ceramic. Each conductor layer is, for example, made of copper (Cu), and a semiconductor element is bonded to one of the conductor layers.

[0003] Japanese Patent Application Laid-Open No. 2021-190505

[0004] In recent years, there has been a demand for electronic devices to be more energy-efficient, perform better, and be more compact. To achieve this, it is necessary to improve the performance and reduce the size of the power modules installed in electronic devices.

[0005] An object of the present disclosure is to provide a semiconductor device that is improved over conventional semiconductor devices. In particular, in view of the above-mentioned circumstances, an object of the present disclosure is to provide a semiconductor device that is suitable for achieving improved performance and miniaturization.

[0006] A semiconductor device provided by a first aspect of the present disclosure comprises a conductive cylindrical holder, at least one terminal including a metal pin inserted into the holder, a terminal support supporting the holder, and a sealing resin covering a portion of the holder and the terminal support, wherein the sealing resin has a resin main surface facing one side in a thickness direction, the holder has a first surface located at an end on one side in the thickness direction and a first outer surface extending in the thickness direction, the first surface being at a different position from the resin main surface in the thickness direction, the first outer surface being in contact with the sealing resin, and the metal pin protruding to one side in the thickness direction beyond the resin main surface.

[0007] A semiconductor device provided by a second aspect of the present disclosure comprises a support substrate having a main surface facing one side in a thickness direction, at least one terminal arranged on the main surface and including a conductive holder and a metal pin inserted into the holder, and a sealing resin having a resin main surface facing one side in the thickness direction and covering at least a portion of the support substrate, wherein at least one of the at least one terminal has all of the holder exposed from the sealing resin, and the metal pin protrudes to one side in the thickness direction beyond the resin main surface.

[0008] According to the above configuration, it is possible to provide a structure that is preferable for improving performance and miniaturizing the semiconductor device.

[0009] Other features and advantages of the present disclosure will become more apparent from the following detailed description taken in conjunction with the accompanying drawings.

[0010] FIG. 1 is a perspective view showing a semiconductor device according to a first embodiment of the present disclosure. FIG. 2 is a perspective view showing a main portion of the semiconductor device according to the first embodiment of the present disclosure. FIG. 3 is a perspective view showing a main portion of the semiconductor device according to the first embodiment of the present disclosure. FIG. 4 is a plan view showing a semiconductor device according to the first embodiment of the present disclosure. FIG. 5 is a plan view showing a main portion of the semiconductor device according to the first embodiment of the present disclosure. FIG. 6 is a side view showing a main portion of the semiconductor device according to the first embodiment of the present disclosure. FIG. 7 is an enlarged plan view showing a main portion of the semiconductor device according to the first embodiment of the present disclosure. FIG. 8 is a plan view showing a main portion of the semiconductor device according to the first embodiment of the present disclosure. FIG. 9 is a plan view showing a main portion of the semiconductor device according to the first embodiment of the present disclosure. FIG. 10 is a side view showing a semiconductor device according to the first embodiment of the present disclosure. FIG. 11 is a bottom view showing a semiconductor device according to the first embodiment of the present disclosure. FIG. 12 is a cross-sectional view taken along line XII-XII in FIG. 5. FIG. 13 is a cross-sectional view taken along line XIII-XIII in FIG. 5. FIG. 14 is an enlarged cross-sectional view showing a main portion of the semiconductor device according to the first embodiment of the present disclosure. FIG. 15 is an enlarged cross-sectional view showing a main portion of the semiconductor device according to the first embodiment of the present disclosure. FIG. 16 is a partially enlarged view of a portion of FIG. 13 . FIG. 17 is a partially enlarged view of a portion of FIG. 4 . FIG. 18 is a cross-sectional view taken along line XVIII-XVIII in FIG. 5 . FIG. 19 is a cross-sectional view taken along line XIX-XIX in FIG. 5 . FIG. 20 is a cross-sectional view taken along line XX-XX in FIG. 5 . FIG. 21 is a cross-sectional view taken along line XXI-XXI in FIG. 5 . FIG. 22 is a cross-sectional view taken along line XXII-XXII in FIG. 5 . FIG. 23 is an enlarged cross-sectional view similar to FIG. 16 , showing a semiconductor device according to a first modified example of the first embodiment. FIG. 24 is an enlarged cross-sectional view similar to FIG. 16 , showing a semiconductor device according to a second modified example of the first embodiment. FIG. 25 is an enlarged cross-sectional view similar to FIG. 16 , showing a semiconductor device according to a third modified example of the first embodiment. FIG. 26 is an enlarged cross-sectional view similar to FIG. 16 , showing a semiconductor device according to a fourth modified example of the first embodiment. FIG. 27 is an enlarged cross-sectional view similar to FIG. 16 , showing a semiconductor device according to a fifth modified example of the first embodiment. Fig. 28 is a perspective view showing a semiconductor device according to a second embodiment of the present disclosure. Fig. 29 is an enlarged cross-sectional view similar to Fig. 16 showing a semiconductor device according to a second embodiment of the present disclosure.FIG. 30 is a perspective view showing a semiconductor device according to a first embodiment based on a second aspect of the present disclosure. FIG. 31 is a plan view showing a semiconductor device according to a first embodiment based on a second aspect of the present disclosure. FIG. 32 is a view showing the sealing resin in the plan view of FIG. 31 with imaginary lines. FIG. 33 is a view showing the sealing resin and the second conductive member in the plan view of FIG. 32 with the sealing resin and the second conductive member omitted. FIG. 34 is a view showing the plan view of FIG. 33 with the first conductive member omitted. FIG. 35 is a bottom view showing a semiconductor device according to a first embodiment based on a second aspect of the present disclosure. FIG. 36 is a cross-sectional view taken along line XXXVI-XXXVI of FIG. 32. FIG. 37 is a partially enlarged cross-sectional view of a portion (near the first element) of FIG. 36. FIG. 38 is a partially enlarged cross-sectional view of a portion (near the second element) of FIG. 36. FIG. 39 is a cross-sectional view taken along line XXXIX-XXXIX of FIG. 32. FIG. 40 is a cross-sectional view taken along line XL-XL of FIG. 32. 41 is a cross-sectional view taken along line XLI-XLI in FIG. 32 . FIG. 42 is a cross-sectional view taken along line XLII-XLII in FIG. 32 . FIG. 43 is a cross-sectional view taken along line XLIII-XLIII in FIG. 32 . FIG. 44 is a partially enlarged cross-sectional view showing a part of FIG. 40 . FIG. 45 is a cross-sectional view showing a step of a method for manufacturing a semiconductor device according to the first embodiment based on the second aspect of the present disclosure. FIG. 46 is a plan view showing a semiconductor device according to a first modified example of the first embodiment of the second aspect. FIG. 47 is a cross-sectional view taken along line XLVII-XLVII in FIG. 46 . FIG. 48 is a cross-sectional view taken along line XLVIII-XLVIII in FIG. 46 . FIG. 49 is a cross-sectional view similar to FIG. 40 , showing a semiconductor device according to a second modified example of the first embodiment of the second aspect. FIG. 50 is a cross-sectional view similar to FIG. 47 , showing a semiconductor device according to a third modified example of the first embodiment of the second aspect. Fig. 51 is a cross-sectional view similar to Fig. 40, showing a semiconductor device according to a fourth modified example of the first embodiment of the second side. Fig. 52 is a plan view showing a semiconductor device according to a fifth modified example of the first embodiment of the second side. Fig. 53 is a cross-sectional view taken along line LIII-LIII in Fig. 52.

[0011] Preferred embodiments of the present disclosure will be described in detail below with reference to the drawings. First, a semiconductor device according to a first aspect of the present disclosure will be described with reference to FIGS. 1 to 29. Then, a semiconductor device according to a second aspect of the present disclosure will be described with reference to FIGS. 30 to 53. Note that the reference symbols used in FIGS. 1 to 29 (first aspect) and those used in FIGS. 30 to 53 (second aspect) are independent of each other. Therefore, for example, the same reference symbol may indicate different components on the first side and the second side, or different reference symbols may indicate the same (or similar) components on the first side and the second side.

[0012] The terms "first," "second," "third," etc. in this disclosure are used merely for identification purposes and are not intended to necessarily assign any order to their objects.

[0013] In this disclosure, unless otherwise specified, "a certain object A is formed on a certain object B" and "a certain object A is formed on a certain object B" include "a certain object A is formed directly on a certain object B" and "a certain object A is formed on a certain object B with another object interposed between the certain object A and the certain object B." Similarly, "a certain object A is disposed on a certain object B" and "a certain object A is disposed on a certain object B" include "a certain object A is disposed directly on a certain object B" and "a certain object A is disposed on a certain object B with another object interposed between the certain object A and the certain object B" unless otherwise specified. Similarly, "a certain object A is located on a certain object B" includes "a certain object A is located on a certain object B with a certain object A in contact with the certain object B" and "a certain object A is located on a certain object B with another object interposed between the certain object A and the certain object B." Furthermore, unless otherwise specified, the phrase "an object A overlaps an object B when viewed in a certain direction" includes "an object A overlaps the entire object B" and "an object A overlaps a part of an object B." Furthermore, in the present disclosure, "a surface A faces in (one side or the other side of) direction B" is not limited to the case where the angle of surface A with respect to direction B is 90°, but also includes the case where surface A is tilted with respect to direction B.

[0014] 1 to 22 show a semiconductor device according to a first embodiment based on a first aspect of the present disclosure. The semiconductor device A1 of this embodiment includes a plurality of first semiconductor elements 10A, a plurality of second semiconductor elements 10B, a support substrate 3, a first terminal 41, a second terminal 42, a plurality of third terminals 43, a fourth terminal 44, a plurality of control terminals 45, a control terminal support 48, a first conductive member 5, a second conductive member 6, and a sealing resin 8.

[0015] FIG. 1 is a perspective view showing the semiconductor device A1. FIGS. 2 and 3 are perspective views of a main portion of the semiconductor device A1. FIG. 4 is a plan view showing the semiconductor device A1. FIG. 5 is a plan view of a main portion of the semiconductor device A1. FIG. 6 is a side view of a main portion of the semiconductor device A1. FIG. 7 is an enlarged plan view of a main portion of the semiconductor device A1. FIGS. 8 and 9 are plan views of a main portion of the semiconductor device A1. FIG. 10 is a side view of the semiconductor device A1. FIG. 11 is a bottom view of the semiconductor device A1. FIG. 12 is a cross-sectional view taken along line XII-XII in FIG. 5. FIG. 13 is a cross-sectional view taken along line XIII-XIII in FIG. 5. FIGS. 14 and 15 are enlarged cross-sectional views of a main portion of the semiconductor device A1. FIG. 16 is a partially enlarged view of a portion of FIG. 13. FIG. 17 is a partially enlarged view of a portion of FIG. 4. FIG. 18 is a cross-sectional view taken along line XVIII-XVIII in FIG. 5. Fig. 19 is a cross-sectional view taken along line XIX-XIX in Fig. 5. Fig. 20 is a cross-sectional view taken along line XX-XX in Fig. 5. Fig. 21 is a cross-sectional view taken along line XXI-XXI in Fig. 5. Fig. 22 is a cross-sectional view taken along line XXII-XXII in Fig. 5.

[0016] For ease of explanation, the three mutually orthogonal directions are referred to as the x-direction, y-direction, and z-direction. The z-direction is an example of the thickness direction, and the x-direction is an example of the first direction. One side of the x-direction is referred to as the x1 side of the x-direction, and the other side of the x-direction is referred to as the x2 side of the x-direction. One side of the y-direction is referred to as the y1 side of the y-direction, and the other side of the y-direction is referred to as the y2 side of the y-direction. One side of the z-direction is referred to as the z1 side of the z-direction, and the other side of the z-direction is referred to as the z2 side of the z-direction.

[0017] The multiple first semiconductor elements 10A and the multiple second semiconductor elements 10B are electronic components that are central to the functionality of the semiconductor device A1. The constituent material of each of the first semiconductor elements 10A and each of the second semiconductor elements 10B is a semiconductor material primarily composed of, for example, silicon carbide (SiC). This semiconductor material is not limited to SiC and may be silicon (Si), gallium nitride (GaN), diamond (C), or the like. Each of the first semiconductor elements 10A and each of the second semiconductor elements 10B is, for example, a power semiconductor chip with switching function, such as a metal oxide semiconductor field effect transistor (MOSFET). In this embodiment, the first semiconductor elements 10A and the second semiconductor elements 10B are MOSFETs, but are not limited thereto and may be other transistors such as insulated gate bipolar transistors (IGBTs). Each of the first semiconductor elements 10A and each of the second semiconductor elements 10B is the same element. Each of the first semiconductor elements 10A and each of the second semiconductor elements 10B is, for example, an n-channel MOSFET, but may also be a p-channel MOSFET.

[0018] 14 and 15 , the first semiconductor element 10A and the second semiconductor element 10B each have an element main surface 101 and an element back surface 102. In each of the first semiconductor elements 10A and the second semiconductor elements 10B, the element main surface 101 and the element back surface 102 are spaced apart in the z direction. The element main surface 101 faces the z1 side in the z direction, and the element back surface 102 faces the z2 side in the z direction.

[0019] In this embodiment, the semiconductor device A1 includes four first semiconductor elements 10A and four second semiconductor elements 10B. However, the number of first semiconductor elements 10A and the number of second semiconductor elements 10B are not limited to this configuration and may be changed as appropriate depending on the performance required of the semiconductor device A1. In the example shown in FIGS. 8 and 9, four first semiconductor elements 10A and four second semiconductor elements 10B are arranged. The number of first semiconductor elements 10A and the number of second semiconductor elements 10B may be two, three, or five or more. The number of first semiconductor elements 10A and the number of second semiconductor elements 10B may be equal to or different from each other. The number of first semiconductor elements 10A and the number of second semiconductor elements 10B is determined by the current capacity handled by the semiconductor device A1.

[0020] The semiconductor device A1 is configured, for example, as a half-bridge switching circuit. In this case, a plurality of first semiconductor elements 10A form an upper arm circuit of the semiconductor device A1, and a plurality of second semiconductor elements 10B form a lower arm circuit. In the upper arm circuit, the plurality of first semiconductor elements 10A are connected in parallel to each other. In the lower arm circuit, the first semiconductor elements 10A are connected in parallel to each other. In the lower arm circuit, the plurality of second semiconductor elements 10B are connected in parallel to each other. Each first semiconductor element 10A and each second semiconductor element 10B are connected in series to form a bridge layer.

[0021] As shown in FIGS. 8 , 9 , and 21 , each of the multiple first semiconductor elements 10A is mounted on a first conductive portion 32A of a support substrate 3 (described later). In the example shown in FIGS. 8 and 9 , the multiple first semiconductor elements 10A are aligned, for example, in the y direction and spaced apart from one another. Each first semiconductor element 10A is conductively bonded to the first conductive portion 32A via a conductive bonding material 19. When each first semiconductor element 10A is bonded to the first conductive portion 32A, the element back surface 102 faces the first conductive portion 32A. Note that, unlike the present embodiment, the multiple first semiconductor elements 10A may be mounted on a metal member other than a part of the DBC substrate or the like. In this case, the metal member corresponds to the first conductive portion in the present disclosure. This metal member may be supported by, for example, the DBC substrate or the like.

[0022] As shown in FIGS. 8 , 9 , and 20 , each of the second semiconductor elements 10B is mounted on a second conductive portion 32B of a support substrate 3 (described later). In the example shown in FIGS. 8 and 9 , the second semiconductor elements 10B are aligned, for example, in the y direction and spaced apart from one another. Each second semiconductor element 10B is conductively bonded to the second conductive portion 32B via a conductive bonding material 19. When each second semiconductor element 10B is bonded to the second conductive portion 32B, the element back surface 102 faces the second conductive portion 32B. As can be seen from FIG. 9 , the first semiconductor elements 10A and the second semiconductor elements 10B overlap when viewed in the x direction, but they do not necessarily need to overlap. Unlike the present embodiment, the second semiconductor elements 10B may be mounted on a metal member other than a part of a DBC substrate or the like. In this case, the metal member corresponds to the second conductive portion in the present disclosure. This metal member may be supported by, for example, a DBC substrate or the like.

[0023] The plurality of first semiconductor elements 10A and the plurality of second semiconductor elements 10B each have a first principal surface electrode 11, a second principal surface electrode 12, a third principal surface electrode 13, and a back surface electrode 15. The configurations of the first principal surface electrode 11, the second principal surface electrode 12, the third principal surface electrode 13, and the back surface electrode 15 described below are common to each of the first semiconductor elements 10A and each of the second semiconductor elements 10B. The first principal surface electrode 11, the second principal surface electrode 12, and the third principal surface electrode 13 are provided on the element main surface 101. The first principal surface electrode 11, the second principal surface electrode 12, and the third principal surface electrode 13 are insulated by an insulating film (not shown). The back surface electrode 15 is provided on the element back surface 102.

[0024] The first principal surface electrode 11 is, for example, a gate electrode, to which a drive signal (for example, a gate voltage) for driving the first semiconductor element 10A (second semiconductor element 10B) is input. In the first semiconductor element 10A (second semiconductor element 10B), the second principal surface electrode 12 is, for example, a source electrode, through which a source current flows. The second principal surface electrode 12 of this embodiment has a gate finger 121. The gate finger 121 is, for example, made of a linear insulator extending in the x direction, and divides the second principal surface electrode 12 into two parts in the y direction. The third principal surface electrode 13 is, for example, a source sense electrode, through which a source current flows. The back surface electrode 15 is, for example, a drain electrode, through which a drain current flows. The back surface electrode 15 covers the entire area (or substantially the entire area) of the element back surface 102. The back surface electrode 15 is, for example, formed by Ag (silver) plating.

[0025] When a drive signal (gate voltage) is input to the first principal surface electrode 11 (gate electrode), each first semiconductor element 10A (each second semiconductor element 10B) switches between a conductive state and a non-conductive state in response to the drive signal. In the conductive state, current flows from the back surface electrode 15 (drain electrode) to the second principal surface electrode 12 (source electrode), while in the non-conductive state, current does not flow. In other words, each first semiconductor element 10A (each second semiconductor element 10B) performs a switching operation. The semiconductor device A1 converts a DC voltage input between the fourth terminal 44 and the first and second terminals 41 and 42, for example, into an AC voltage through the switching functions of the multiple first semiconductor elements 10A and multiple second semiconductor elements 10B, and outputs the AC voltage from the third terminal 43. Each of the multiple first semiconductor elements 10A corresponds to a first switching element in this disclosure. Each of the multiple second semiconductor elements 10B corresponds to a second switching element in this disclosure.

[0026] 5, 8, 9, etc., the semiconductor device A1 includes a thermistor 17. The thermistor 17 is used as a temperature detection sensor. Note that the semiconductor device A1 may include, for example, a temperature-sensitive diode in addition to the thermistor 17, or may not include the thermistor 17.

[0027] The support substrate 3 supports a plurality of first semiconductor elements 10A and a plurality of second semiconductor elements 10B. The specific configuration of the support substrate 3 is not limited in any way, and may be, for example, a DBC (Direct Bonded Copper) substrate or an AMB (Active Metal Brazing) substrate. The support substrate 3 includes an insulating layer 31, a support conductor 32, and a back surface metal layer 33. The support conductor 32 includes a first conductive portion 32A and a second conductive portion 32B. The dimension of the support substrate 3 in the z direction is, for example, not less than 0.4 mm and not more than 3.0 mm.

[0028] The insulating layer 31 is made of, for example, ceramics with excellent thermal conductivity. Examples of such ceramics include silicon nitride (SiN). The insulating layer 31 is not limited to ceramics and may be an insulating resin sheet or the like. The insulating layer 31 has, for example, a rectangular shape in a plan view. The dimension of the insulating layer 31 in the z direction is, for example, 0.05 mm or more and 1.0 mm or less.

[0029] The first conductive portion 32A supports a plurality of first semiconductor elements 10A, and the second conductive portion 32B supports a plurality of second semiconductor elements 10B. The first conductive portion 32A and the second conductive portion 32B are formed on the upper surface (the surface facing the z1 side in the z direction) of the insulating layer 31. The constituent material of the first conductive portion 32A and the second conductive portion 32B includes, for example, Cu (copper). The constituent material may include, for example, Al (aluminum) other than Cu (copper). The first conductive portion 32A and the second conductive portion 32B are spaced apart in the x direction. The first conductive portion 32A is located on the x1 side in the x direction relative to the second conductive portion 32B. The first conductive portion 32A and the second conductive portion 32B are each, for example, rectangular in plan view. The first conductive portion 32A and the second conductive portion 32B, together with the first conductive member 5 and the second conductive member 6, constitute a path of the main circuit current switched by the plurality of first semiconductor elements 10A and the plurality of second semiconductor elements 10B.

[0030] The first conductive portion 32A has a first main surface 301A. The first main surface 301A is a flat surface facing the z1 side in the z direction. A plurality of first semiconductor elements 10A are bonded to the first main surface 301A of the first conductive portion 32A via conductive bonding material 19. The second conductive portion 32B has a second main surface 301B. The second main surface 301B is a flat surface facing the z1 side in the z direction. A plurality of second semiconductor elements 10B are bonded to the second main surface 301B of the second conductive portion 32B via conductive bonding material 19. The material of the conductive bonding material 19 is not particularly limited and may be, for example, solder, a metal paste material, or a sintered metal. The dimension of the first conductive portion 32A and the second conductive portion 32B in the z direction is, for example, 0.1 mm or more and 1.5 mm or less.

[0031] The back surface metal layer 33 is formed on the lower surface (surface facing the z2 side in the z direction) of the insulating layer 31. The constituent material of the back surface metal layer 33 is the same as the constituent material of the support conductors 32. The back surface metal layer 33 has a back surface 302. The back surface 302 is a flat surface facing the z2 side in the z direction. In the example shown in FIG. 11 , the back surface 302 is exposed from the sealing resin 8, for example. A heat dissipation member (for example, a heat sink) (not shown) can be attached to the back surface 302. The back surface 302 may not be exposed from the sealing resin 8, but may be covered by the sealing resin 8. The back surface metal layer 33 overlaps both the first conductive portion 32A and the second conductive portion 32B in a planar view.

[0032] The first terminal 41, the second terminal 42, the plurality of third terminals 43, and the fourth terminal 44 are each made of a plate-shaped metal plate. This metal plate contains, for example, Cu (copper) or a Cu (copper) alloy. In the examples shown in Figures 1 to 5, 8, 9, and 11, the semiconductor device A1 includes one each of the first terminal 41, the second terminal 42, and the fourth terminal 44, and two third terminals 43, but the number of each terminal is not limited in any way.

[0033] A DC voltage to be converted into power is input to the first terminal 41, the second terminal 42, and the fourth terminal 44. The fourth terminal 44 is a positive electrode (P terminal), and the first terminal 41 and the second terminal 42 are each a negative electrode (N terminal). An AC voltage converted into power by the first semiconductor element 10A and the second semiconductor element 10B is output from the plurality of third terminals 43. The first terminal 41, the second terminal 42, the plurality of third terminals 43, and the fourth terminal 44 each include a portion covered with the sealing resin 8 and a portion exposed from the sealing resin 8.

[0034] As shown in FIG. 13 , the fourth terminal 44 is conductively bonded to the first conductive portion 32A. The conductive bonding method is not limited, and methods such as ultrasonic bonding, laser bonding, welding, or methods using solder, metal paste, sintered silver, etc. may be appropriately adopted. As shown in FIGS. 8 and 9 , the fourth terminal 44 is located on the x1 side in the x direction with respect to the multiple first semiconductor elements 10A and the first conductive portions 32A. The fourth terminal 44 is conductively connected to the first conductive portion 32A and, via the first conductive portion 32A, to the back surface electrode 15 (drain electrode) of each first semiconductor element 10A.

[0035] The first terminal 41 and the second terminal 42 are electrically connected to the second conductive member 6. In this embodiment, the first terminal 41 and the second conductive member 6 are integrally formed. The term "integrally formed" refers to a configuration in which the first terminal 41 and the second conductive member 6 are formed, for example, by cutting and bending a single metal plate material, without using any bonding material or the like to join them together. Furthermore, in this embodiment, the second terminal 42 and the second conductive member 6 are integrally formed. Note that the first terminal 41 and the second terminal 42 may be configured to be electrically connected to the second conductive member 6, and may have a bonding portion to join them, unlike this embodiment. As shown in FIGS. 5 and 8 , the first terminal 41 and the second terminal 42 are each located on the x1 side in the x direction with respect to the multiple first semiconductor elements 10A and the first conductive portion 32A. The first terminal 41 and the second terminal 42 are each electrically connected to the second conductive member 6 and, via the second conductive member 6, to the second principal surface electrode 12 (source electrode) of each second semiconductor element 10B.

[0036] 1 to 5 and 11 , the first terminal 41, the second terminal 42, and the fourth terminal 44 each protrude from the sealing resin 8 toward the x1 side in the x direction in the semiconductor device A1. The first terminal 41, the second terminal 42, and the fourth terminal 44 are spaced apart from one another. The first terminal 41 and the second terminal 42 are located on opposite sides of the fourth terminal 44 in the y direction. The first terminal 41 is located on the y1 side of the fourth terminal 44 in the y direction, and the second terminal 42 is located on the y2 side of the fourth terminal 44 in the y direction. The first terminal 41, the second terminal 42, and the fourth terminal 44 overlap one another when viewed in the y direction.

[0037] As can be seen from FIGS. 8 , 9 , and 12 , the two third terminals 43 are each conductively bonded to the second conductive portion 32B. The conductive bonding method is not limited to any particular method, and methods such as ultrasonic bonding, laser bonding, and welding, or methods using solder, metal paste, silver sintered body, etc., may be appropriately employed. As shown in FIG. 8 and other figures, the two third terminals 43 are each located on the x2 side in the x direction relative to the multiple second semiconductor elements 10B and the second conductive portions 32B. Each third terminal 43 is conductively connected to the second conductive portion 32B and, via the second conductive portion 32B, to the back electrode 15 (drain electrode) of each second semiconductor element 10B. The number of third terminals 43 is not limited to two and may be, for example, one or three or more. For example, if there is one third terminal 43, it is preferably connected to the center portion of the second conductive portion 32B in the y direction.

[0038] Each of the plurality of control terminals 45 is a pin-shaped terminal for controlling the driving of each first semiconductor element 10A and each second semiconductor element 10B. Each of the plurality of control terminals 45 is, for example, a press-fit terminal. The plurality of control terminals 45 includes a plurality of first control terminals 46A to 46E and a plurality of second control terminals 47A to 47D. The plurality of first control terminals 46A to 46E are used for controlling each first semiconductor element 10A, etc. The plurality of second control terminals 47A to 47D are used for controlling each second semiconductor element 10B, etc.

[0039] The multiple first control terminals 46A-46E are arranged at intervals in the y direction. As shown in Figures 8, 13, and 22, each of the first control terminals 46A-46E is supported by the first conductive portion 32A via a control terminal support 48 (a first support portion 48A described below). As shown in Figures 5 and 8, each of the first control terminals 46A-46E is located in the x direction between the multiple first semiconductor elements 10A and the first terminal 41, the second terminal 42, and the fourth terminal 44.

[0040] The first control terminal 46A is a terminal (gate terminal) for inputting a drive signal for the plurality of first semiconductor elements 10A. A drive signal for driving the plurality of first semiconductor elements 10A is input to the first control terminal 46A (for example, a gate voltage is applied).

[0041] The first control terminal 46B is a terminal (source sense terminal) for detecting source signals of the multiple first semiconductor elements 10A. The first control terminal 46B detects a voltage (a voltage corresponding to a source current) applied to each second principal surface electrode 12 (source electrode) of the multiple first semiconductor elements 10A.

[0042] The first control terminal 46C and the first control terminal 46D are terminals that are electrically connected to the thermistor 17.

[0043] The first control terminal 46E is a terminal (drain sense terminal) for detecting drain signals of the multiple first semiconductor elements 10A. The first control terminal 46E detects a voltage (a voltage corresponding to a drain current) applied to each back electrode 15 (drain electrode) of the multiple first semiconductor elements 10A.

[0044] The second control terminals 47A to 47D are spaced apart in the y direction. As shown in FIGS. 8 and 13 , each of the second control terminals 47A to 47D is supported by the second conductive portion 32B via a control terminal support 48 (a second support portion 48B described below). As shown in FIGS. 5 and 8 , each of the second control terminals 47A to 47D is located between the second semiconductor elements 10B and two third terminals 43 in the x direction.

[0045] The second control terminal 47A is a terminal (gate terminal) for inputting drive signals for the multiple second semiconductor elements 10B. A drive signal for driving the multiple second semiconductor elements 10B is input to the second control terminal 47A (for example, a gate voltage is applied). The second control terminal 47B is a terminal (source sense terminal) for detecting source signals for the multiple second semiconductor elements 10B. The second control terminal 47B detects a voltage (a voltage corresponding to a source current) applied to each second main surface electrode 12 (source electrode) of the multiple second semiconductor elements 10B. The second control terminal 47C and the second control terminal 47D are terminals that are electrically connected to the thermistor 17.

[0046] Each of the plurality of control terminals 45 (the plurality of first control terminals 46A to 46E and the plurality of second control terminals 47A to 47D) includes a holder 451 and a metal pin 452.

[0047] The holder 451 is made of a conductive material. As shown in Figures 14 and 15, the holder 451 is bonded to the control terminal support body 48 (a first metal layer 482 described below) via a conductive bonding material 459. As shown in Figure 16, the holder 451 includes a cylindrical portion 453, a first flange portion 454, and a second flange portion 455.

[0048] The cylindrical portion 453 extends in the z direction and has, for example, a cylindrical shape. The cylindrical portion 453 has a first outer surface 453a and a first inner surface 453b. The first outer surface 453a is a surface that faces radially outward of the cylindrical portion 453 when viewed in the z direction and extends in the z direction. The first inner surface 453b faces the opposite side from the first outer surface 453a and is a surface that faces radially inward of the cylindrical portion 453 when viewed in the z direction and extends in the z direction.

[0049] The first flange 454 is connected to the end of the cylindrical portion 453 on the z1 side in the z direction. The first flange 454 has a first surface 454a and a second surface 454b. The first surface 454a faces the z1 side in the z direction. The first surface 454a is located at the end of the holder 451 on the z1 side in the z direction. The first surface 454a has an annular shape (a circular ring or a rectangular ring) when viewed in the z direction. The second surface 454b is located on the z2 side in the z direction of the first surface 454a and faces the z2 side in the z direction.

[0050] The second flange 455 is connected to the end of the cylindrical portion 453 on the z2 side in the z direction. In the present embodiment, the second flange 455 is joined to the control terminal support body 48 (a first metal layer 482 described below) via a conductive bonding material 459.

[0051] A metal pin 452 is inserted through at least the first flange 454 and the cylindrical portion 453 of the holder 451. A portion of the holder 451 is covered with sealing resin 8. At least the first outer surface 453a (cylindrical portion 453) is in contact with the sealing resin 8. In the example shown in FIG. 16 , the first outer surface 453a of the cylindrical portion 453 and the second surface 454b of the first flange 454 are in contact with the sealing resin 8.

[0052] The metal pin 452 is a rod-shaped member extending in the z direction. The metal pin 452 is supported by being press-fitted into the holder 451. The metal pin 452 is electrically connected to the control terminal support 48 (a first metal layer 482 described below) at least via the holder 451. In the example shown in FIGS. 14 to 16 , the metal pin 452 is not inserted all the way to the lower end (the end on the z2 side in the z direction) of the holder 451, and the lower end of the metal pin 452 is separated from the conductive bonding material 459. In this case, the metal pin 452 is electrically connected to the control terminal support 48 (the first metal layer 482) via the holder 451. Unlike the example shown in the drawings, if the lower end (the end on the z2 side in the z direction) of the metal pin 452 is in contact with the conductive bonding material 459 within the insertion hole of the holder 451, the metal pin 452 is electrically connected to the control terminal support 48 via the conductive bonding material 459. The metal pin 452 protrudes from the upper surface of the sealing resin 8 (a resin main surface 81 described later) toward the z1 side in the z direction.

[0053] The control terminal support body 48 supports the plurality of control terminals 45. The control terminal support body 48 is interposed between the first main surface 301A and the second main surface 301B and the plurality of control terminals 45 in the z direction.

[0054] The control terminal support 48 includes a first support portion 48A and a second support portion 48B. The first support portion 48A is disposed on the first conductive portion 32A and supports a plurality of first control terminals 46A to 46E among the plurality of control terminals 45. As shown in FIG. 14 , the first support portion 48A is bonded to the first conductive portion 32A via a bonding material 49. The bonding material 49 may be conductive or insulating, and is, for example, solder. The second support portion 48B is disposed on the second conductive portion 32B and supports a plurality of second control terminals 47A to 47D among the plurality of control terminals 45. As shown in FIG. 15 , the second support portion 48B is bonded to the second conductive portion 32B via the bonding material 49.

[0055] The control terminal support 48 (each of the first support portion 48A and the second support portion 48B) is made of, for example, a direct bonded copper (DBC) substrate and includes an insulating layer 481, a first metal layer 482, and a second metal layer 483 stacked on top of each other.

[0056] The insulating layer 481 is made of, for example, ceramics and has, for example, a rectangular shape in plan view.

[0057] As shown in FIGS. 14 and 15 , the first metal layer 482 is formed on the upper surface of the insulating layer 481. Each control terminal 45 is provided upright on the first metal layer 482. The first metal layer 482 includes, for example, Cu (copper) or a Cu (copper) alloy. As shown in FIG. 8 , the first metal layer 482 includes a first portion 482A, a second portion 482B, a third portion 482C, a fourth portion 482D, a fifth portion 482E, and a sixth portion 482F. The first portion 482A, the second portion 482B, the third portion 482C, the fourth portion 482D, the fifth portion 482E, and the sixth portion 482F are spaced apart and insulated from one another.

[0058] A plurality of wires 71 are bonded to the first portion 482A, and the first portion 482A is electrically connected to the first principal surface electrodes 11 (gate electrodes) of the first semiconductor elements 10A (second semiconductor elements 10B) via the wires 71. A plurality of wires 73 are connected between the first portion 482A and the sixth portion 482F. As a result, the sixth portion 482F is electrically connected to the first principal surface electrodes 11 (gate electrodes) of the first semiconductor elements 10A (second semiconductor elements 10B) via the wires 73 and 71. As shown in FIG. 8 , the first control terminal 46A is bonded to the sixth portion 482F of the first support portion 48A, and the second control terminal 47A is bonded to the sixth portion 482F of the second support portion 48B.

[0059] A plurality of wires 72 are joined to the second portion 482B, and the second portion 482B is electrically connected to the third principal surface electrode 13 (source sense electrode) of each of the first semiconductor elements 10A (each of the second semiconductor elements 10B) via each of the wires 72. As shown in Fig. 8 , the first control terminal 46B is joined to the second portion 482B of the first support portion 48A, and the second control terminal 47B is joined to the second portion 482B of the second support portion 48B.

[0060] The thermistor 17 is joined to the third portion 482C and the fourth portion 482D. As shown in Fig. 8 , the first control terminals 46C and 46D are joined to the third portion 482C and the fourth portion 482D of the first support portion 48A, and the second control terminals 47C and 47D are joined to the third portion 482C and the fourth portion 482D of the second support portion 48B.

[0061] A wire 74 is joined to the fifth portion 482E of the first support portion 48A, and the fifth portion 482E is electrically connected to the first conductive portion 32A via the wire 74. As shown in FIG. 8 , a first control terminal 46E is joined to the fifth portion 482E of the first support portion 48A. The fifth portion 482E of the second support portion 48B is not electrically connected to other components. Each of the wires 71 to 74 is, for example, a bonding wire. The material of each of the wires 71 to 74 includes, for example, any of Au (gold), Al (aluminum), and Cu (copper).

[0062] 14, 15, etc., the second metal layer 483 is formed on the lower surface of the insulating layer 481. As shown in Fig. 14, the second metal layer 483 of the first support portion 48A is joined to the first conductive portion 32A via a bonding material 49. As shown in Fig. 15, the second metal layer 483 of the second support portion 48B is joined to the second conductive portion 32B via a bonding material 49.

[0063] The first conductive member 5 and the second conductive member 6, together with the first conductive portion 32A and the second conductive portion 32B, form a path for a main circuit current switched by the plurality of first semiconductor elements 10A and the plurality of second semiconductor elements 10B. The first conductive member 5 and the second conductive member 6 are spaced apart from the first main surface 301A and the second main surface 301B on the z1 side in the z direction and overlap the first main surface 301A and the second main surface 301B in a plan view. In this embodiment, the first conductive member 5 and the second conductive member 6 are each made of a metal plate material. The metal includes, for example, Cu (copper) or a Cu (copper) alloy. Specifically, the first conductive member 5 and the second conductive member 6 are made of an appropriately bent metal plate material.

[0064] The first conductive member 5 is connected to the second principal surface electrode 12 (source electrode) of each first semiconductor element 10A and the second conductive portion 32B, thereby electrically connecting the second principal surface electrode 12 of each first semiconductor element 10A and the second conductive portion 32B. The first conductive member 5 forms a path for a main circuit current switched by the multiple first semiconductor elements 10A. As shown in FIGS. 7 and 8 , the first conductive member 5 includes a main portion 51, multiple first bonding portions 52, and multiple second bonding portions 53.

[0065] The main portion 51 is located between the plurality of first semiconductor elements 10A and the second conductive portions 32B in the x direction and is a strip-shaped portion extending in the y direction in a plan view. The main portion 51 overlaps both the first conductive portions 32A and the second conductive portions 32B in a plan view and is spaced apart in the z direction from the first main surface 301A and the second main surface 301B on the z1 side in the z direction. As shown in FIG. 18 and other figures, the main portion 51 is located on the z2 side in the z direction with respect to a third path portion 66 and a fourth path portion 67 of the second conductive member 6 (described later) and is closer to the first main surface 301A and the second main surface 301B than the third path portion 66 and the fourth path portion 67.

[0066] In this embodiment, the main portion 51 is disposed parallel to the first main surface 301A and the second main surface 301B.

[0067] As shown in FIG. 8 and other figures, the main portion 51 extends continuously in the y direction to correspond to the region in which the multiple first semiconductor elements 10A are arranged. In this embodiment, as shown in FIGS. 7 , 8 , 13 , and other figures, multiple first openings 514 are formed in the main portion 51. Each of the multiple first openings 514 is, for example, a through hole that penetrates in the z direction (the plate thickness direction of the main portion 51). The multiple first openings 514 are aligned at intervals in the y direction. The multiple first openings 514 are provided corresponding to each of the multiple first semiconductor elements 10A. In this embodiment, four first openings 514 are provided in the main portion 51, and these first openings 514 and the multiple (four) first semiconductor elements 10A are positioned at the same position in the y direction.

[0068] 8, 13, etc., in the present embodiment, each first opening 514 overlaps with a gap between the first conductive portion 32A and the second conductive portion 32B in a plan view. The multiple first openings 514 are formed to facilitate the flow of the resin material between the upper side (the z1 side in the z direction) and the lower side (the z2 side in the z direction) near the main portion 51 (first conductive member 5) when injecting the flowable resin material to form the sealing resin 8.

[0069] As shown in FIG. 8 and other figures, the multiple first bonding portions 52 and the multiple second bonding portions 53 are each connected to the main portion 51 and are arranged to correspond to the multiple first semiconductor elements 10A. Specifically, each first bonding portion 52 is located on the x1 side of the main portion 51 in the x direction. Each second bonding portion 53 is located on the x2 side of the main portion 51 in the x direction. As shown in FIG. 14 , each first bonding portion 52 is bonded to the corresponding second principal surface electrode 12 of one of the first semiconductor elements 10A via a conductive bonding material 59. Each second bonding portion 53 is bonded to the second conductive portion 32B via the conductive bonding material 59. The material of the conductive bonding material 59 is not particularly limited and may be, for example, solder, a metal paste material, or a sintered metal. In this embodiment, the first bonding portion 52 has two portions spaced apart in the y direction. These two portions are joined to the second principal surface electrode 12 on both sides in the y direction, with the gate finger 121 of the second principal surface electrode 12 of the first semiconductor element 10A sandwiched therebetween.

[0070] The second conductive member 6 electrically connects the second main surface electrode 12 (source electrode) of each second semiconductor element 10B to the first terminal 41 and the second terminal 42. The second conductive member 6 is integrally formed with the first terminal 41 and the second terminal 42. The second conductive member 6 constitutes a path for a main circuit current switched by the plurality of second semiconductor elements 10B. As shown in FIGS. 5 to 7 , 12 , 13 , and 18 to 22 , the second conductive member 6 includes a plurality of third joint portions 61, a first path portion 64, a second path portion 65, a plurality of third path portions 66, and a fourth path portion 67. In the illustrated example, the second conductive member 6 also includes a first step portion 602 and a second step portion 603.

[0071] The multiple third bonding portions 61 are portions that are individually bonded to the multiple second semiconductor elements 10B. Each third bonding portion 61 and the second main surface electrode 12 of each second semiconductor element 10B are bonded via a conductive bonding material 69. The material of the conductive bonding material 69 is not particularly limited and may be, for example, solder, a metal paste material, or a sintered metal. In this embodiment, the third bonding portion 61 has two flat portions 611 and two first inclined portions 612.

[0072] The two flat portions 611 are aligned in the y direction. The two flat portions 611 are spaced apart from each other in the y direction. The shape of the flat portions 611 is not limited in any way, and in the illustrated example, they are rectangular. The two flat portions are joined to the second main surface electrode 12 on both sides in the y direction, sandwiching the gate finger 121 of the second main surface electrode 12 of the second semiconductor element 10B therebetween.

[0073] The two first inclined portions 612 are connected to the outside of the two flat portions 611 in the y direction. That is, the first inclined portion 612 located on the y1 side in the y direction is connected to the y1 side in the y direction of the flat portion 611 located on the y1 side in the y direction. Furthermore, the first inclined portion 612 located on the y2 side in the y direction is connected to the y2 side in the y direction of the flat portion 611 located on the y2 side in the y direction. The first inclined portion 612 is inclined so that the farther it is from the flat portion 611 in the y direction, the closer it is to the z1 side in the z direction.

[0074] The first path portion 64 is interposed between the plurality of third joint portions 61 and the first terminal 41. In the illustrated example, the first path portion 64 is connected to the first terminal 41 via a first step portion 602. The first path portion 64 overlaps the first conductive portion 32A in a plan view. The first path portion 64 has a shape that extends in the x-direction as a whole.

[0075] The first path portion 64 includes a first strip portion 641 and a first extending portion 643. The first strip portion 641 is located on the x2 side in the x direction with respect to the first terminal 41 and is substantially parallel to the first main surface 301A. The first strip portion 641 has a shape that extends in the x direction as a whole. In the illustrated example, the first strip portion 641 has a recess 649. The recess 649 is a portion of the first strip portion 641 that is recessed toward the y1 side in the y direction. In FIGS. 5 and 7 , the first conductive portion 32A appears through the recess 649.

[0076] The first extending portion 643 extends from the side end of the first strip portion 641 on the y1 side in the y direction to the z2 side in the z direction. The first extending portion 643 is spaced apart from the first conductive portion 32A. In the example shown, the first extending portion 643 is shaped along the z direction and has an elongated rectangular shape with the x direction as the longitudinal direction. Note that the first path portion 64 may not have the first extending portion 643.

[0077] The second path portion 65 is interposed between the plurality of third joint portions 61 and the second terminal 42. In the illustrated example, the second path portion 65 is connected to the second terminal 42 via the second step portion 603. The second path portion 65 overlaps the first conductive portion 32A in plan view. The second path portion 65 has a shape that extends in the x-direction as a whole.

[0078] The second path portion 65 includes a second band-shaped portion 651 and a second extending portion 653. The second band-shaped portion 651 is located on the x2 side in the x direction with respect to the second terminal 42 and is substantially parallel to the first main surface 301A. The second band-shaped portion 651 has a shape that extends in the x direction as a whole. In the illustrated example, the second band-shaped portion 651 has a recess 659. The recess 659 is a portion of the second band-shaped portion 651 that is recessed toward the y2 side in the y direction. In FIGS. 5 and 7 , the first conductive portion 32A appears through the recess 659.

[0079] The second extending portion 653 extends from the side end of the second strip portion 651 on the y2 side in the y direction to the z2 side in the z direction. The second extending portion 653 is spaced apart from the first conductive portion 32A. Like the first extending portion 643, the second extending portion 653 has a shape that extends along the z direction and is an elongated rectangle with the x direction as the longitudinal direction. Note that the second path portion 65 may not have the second extending portion 653.

[0080] The multiple third path portions 66 are individually connected to the multiple third joint portions 61. Each third path portion 66 has a shape extending in the x direction and is arranged spaced apart from one another in the y direction. There is no limitation on the number of multiple third path portions 66, and in the example shown, five third path portions 66 are arranged. Each third path portion 66 is arranged so as to be located between the multiple second semiconductor elements 10B in the y direction or to be located outward in the y direction from the multiple second semiconductor elements 10B.

[0081] Recesses 669 are formed in the two third path portions 66 located on both outer sides in the y direction. The recesses 669 are recessed from the inside toward the outside in the y direction. In the illustrated example, one recess 669 is formed in each of the two third path portions 66. In FIGS. 5 and 7 , the second conductive portion 32B appears through these recesses 669.

[0082] In the present embodiment, one third joint 61 is disposed between two third path portions 66 adjacent to each other in the y direction. In one third joint 61, the first inclined portion 612 located on the y1 side in the y direction is connected to the third path portion 66 located on the y1 side in the y direction, of the two third path portions 66 adjacent to each other in the y direction. In one third joint 61, the first inclined portion 612 located on the y2 side in the y direction is connected to the third path portion 66 located on the y2 side in the y direction, of the two third path portions 66 adjacent to each other in the y direction.

[0083] The fourth path portion 67 is connected to the ends of the plurality of third path portions 66 on the x1 side in the x direction. The fourth path portion 67 has a shape that extends elongatedly in the y direction. The fourth path portion 67 is connected to the ends of the first band portion 641 of the first path portion 64 and the second band portion 651 of the second path portion 65 on the x2 side in the x direction. In the example shown, the first path portion 64 is connected to the end of the fourth path portion 67 on the y1 side in the y direction. Furthermore, the second path portion 65 is connected to the end of the fourth path portion 67 on the y2 side in the y direction.

[0084] The sealing resin 8 covers the plurality of first semiconductor elements 10A, the plurality of second semiconductor elements 10B, the support substrate 3 (excluding the back surface 302), portions of the first terminal 41, the second terminal 42, the plurality of third terminals 43, and the fourth terminal 44, portions of the plurality of control terminals 45, the control terminal support 48, the first conductive member 5, the second conductive member 6, and the plurality of wires 71 to 74. The sealing resin 8 is made of, for example, black epoxy resin. The sealing resin 8 is formed, for example, by molding. The sealing resin 8 has, for example, a dimension in the x direction of approximately 35 mm to 60 mm, a dimension in the y direction of approximately 35 mm to 50 mm, and a dimension in the z direction of approximately 4 mm to 15 mm. These dimensions are the sizes of the largest portions along each direction. The sealing resin 8 has a resin main surface 81, a resin back surface 82, and a plurality of resin side surfaces 831 to 834.

[0085] As shown in FIGS. 10 , 12 , and 20 , the resin main surface 81 and the resin back surface 82 are spaced apart in the z direction. The resin main surface 81 faces the z1 side in the z direction, and the resin back surface 82 faces the z2 side in the z direction. Metal pins 452 of the control terminals 45 (the first control terminals 46A to 46E and the second control terminals 47A to 47D) protrude from the resin main surface 81. As shown in FIG. 11 , the resin back surface 82 has a frame shape surrounding the back surface 302 (the lower surface of the back surface metal layer 33) of the support substrate 3 in a plan view. The back surface 302 of the support substrate 3 is exposed from the resin back surface 82 and is, for example, flush with the resin back surface 82. The resin side surfaces 831 to 834 are each connected to both the resin main surface 81 and the resin back surface 82 and are sandwiched between them in the z direction. As shown in Figure 4 and other figures, the resin side surface 831 and the resin side surface 832 are spaced apart in the x direction. The resin side surface 831 faces the x2 side in the x direction, and the resin side surface 832 faces the x1 side in the x direction. Two third terminals 43 protrude from the resin side surface 831, and the first terminal 41, the second terminal 42, and the fourth terminal 44 protrude from the resin side surface 832. As shown in Figure 4 and other figures, the resin side surface 833 and the resin side surface 834 are spaced apart in the y direction. The resin side surface 833 faces the y2 side in the y direction, and the resin side surface 834 faces the y1 side in the y direction.

[0086] 1 , 4 , 13 , 22 , etc., a plurality of first recesses 810 are formed in the resin main surface 81. Each of the plurality of first recesses 810 is recessed toward the z2 side in the z direction from the resin main surface 81. The plurality of first recesses 810 are provided corresponding to each of the plurality of control terminals 45.

[0087] 16 and 17 , the first recess 810 overlaps the entire cylindrical portion 453 of the holder 451 in a plan view. In the illustrated example, the first recess 810 has a recess inner side surface 811 and a recess bottom surface 812. The recess inner side surface 811 is connected to the resin main surface 81 and extends to the z2 side in the z direction. In the illustrated example, the cross section of the recess inner side surface 811 perpendicular to the z direction is circular. The recess bottom surface 812 is connected to the end of the recess inner side surface 811 on the z2 side in the z direction and is a flat surface facing the z1 side in the z direction.

[0088] The recess bottom surface 812 surrounds the first surface 454a of the holder 451 (first flange portion 454) in a plan view. The first surface 454a and the recess bottom surface 812 are flush with each other. The first recess 810 is a trace of molding the sealing resin 8 while pressing the upper end (first flange portion 454) of the holder 451 with a pin or the like having a shape corresponding to the first recess 810. The fact that the first recess 810 is a trace of molding also applies to the first recesses 810 in the various modified examples described below. The first flange portion 454 is located on the z2 side in the z direction with respect to the resin main surface 81. As shown in FIG. 16 , the entire first outer surface 453a of the tubular portion 453 and the entire second surface 454b of the first flange portion 454 are in contact with the sealing resin 8. 16 and 17 , the first recess 810 overlaps the entire first flange 454 when viewed in the z direction. As a result, the diameter (maximum value of the inner diameter dimension) of the first recess 810 is larger than the outer diameter dimension of the first flange 454.

[0089] The first surface 454a, which is flush with the recess bottom surface 812, is located at a different position in the z direction from the resin main surface 81. Specifically, the first surface 454a is located on the z2 side of the resin main surface 81 in the z direction. In the present embodiment, a first dimension L1, which is the distance in the z direction between the resin main surface 81 and the first surface 454a, is smaller than a second dimension L2, which is the length of the holder 451 in the z direction. Preferably, the ratio of the distance in the z direction between the resin main surface 81 and the first surface 454a (first dimension L1) to the length in the z direction of the holder 451 (second dimension L2) is ⅓ or more.

[0090] In the examples shown in FIGS. 16 and 17 , the recess inner surface 811 is cylindrical, but a draft gradient may be provided during molding. When the recess inner surface 811 is provided with a draft gradient, the recess inner surface 811 is formed into a conical shape that slopes so that the inner diameter decreases toward the z2 side in the z direction. The draft gradient angle of the recess inner surface 811 is appropriately set, for example, within a range of 0 to 30°. Furthermore, when the recess inner surface 811 slopes conically, if the slope angle is relatively large, the inner diameter of the lower end (the end on the z2 side in the z direction) of the recess inner surface 811 may be smaller than the outer diameter of the first flange 454. In such a case, the recess bottom surface 812 described above is not formed. The lower end of the recess inner surface 811 contacts the first surface 454a and forms the recess edge.

[0091] As shown in FIG. 4 , a plurality of recesses 832 a are formed on the resin side surface 832. Each recess 832 a is a portion recessed in the x-direction in plan view. The plurality of recesses 832 a include one formed between the first terminal 41 and the fourth terminal 44 and one formed between the second terminal 42 and the fourth terminal 44 in plan view. The plurality of recesses 832 a are provided to increase the creepage distance along the resin side surface 832 between the first terminal 41 and the fourth terminal 44 and the creepage distance along the resin side surface 832 between the second terminal 42 and the fourth terminal 44.

[0092] As shown in FIGS. 1 , 12 , and 13 , the sealing resin 8 has multiple protrusions 851. Each of the multiple protrusions 851 protrudes from the resin main surface 81 toward the z1 side in the z direction. The multiple protrusions 851 are arranged near the four corners of the sealing resin 8 in a plan view. A protrusion end surface 851a is formed at the tip (the end on the z1 side in the z direction) of each of the protrusions 851. Each of the protrusion end surfaces 851a of the multiple protrusions 851 is parallel (or approximately parallel) to the resin main surface 81 and is on the same plane (x-y plane). Each of the protrusions 851 has, for example, a bottomed, hollow truncated cone shape. The multiple protrusions 851 are used as spacers when the semiconductor device A1 is mounted on a control circuit board or the like of a device that uses power generated by the semiconductor device A1. Each of the multiple protrusions 851 has a recess 851b and an inner wall surface 851c formed in the recess 851b. The shape of each protrusion 851 may be columnar, and is preferably cylindrical. The shape of the recess 851b is preferably cylindrical, and the inner wall surface 851c is preferably a single perfect circle in plan view.

[0093] The semiconductor device A1 may be mechanically fixed to a control circuit board or the like by a method such as screwing. In this case, a female screw thread may be formed on the inner wall surface 851c of the recessed portion 851b of the plurality of protruding portions 851. An insert nut may be embedded in the recessed portion 851b of the plurality of protruding portions 851.

[0094] Next, the operation of this embodiment will be described.

[0095] The holder 451 constituting each control terminal 45 has a first surface 454a and a first outer surface 453a. The first surface 454a is located at the end of the holder 451 on the z1 side in the z direction. The first surface 454a is located at a position different from the resin main surface 81 in the z direction. The first outer surface 453a extends in the z direction and contacts the sealing resin 8. The metal pin 452 constituting each control terminal 45 protrudes toward the z1 side in the z direction beyond the resin main surface 81. With this configuration, the multiple control terminals 45 are arranged in an area surrounded by the resin main surface 81 (sealing resin 8) in a planar view. This semiconductor device A1 can be made smaller in planar view. Furthermore, the first surface 454a is located at a position different from the resin main surface 81 in the z direction. With this configuration, the creepage distance along the surface of the sealing resin 8 (such as the resin main surface 81) can be increased between adjacent control terminals 45. Therefore, the semiconductor device A1 is suitable for increasing the withstand voltage of the adjacent control terminals 45 while achieving miniaturization in plan view.

[0096] The holder 451 includes a cylindrical portion 453 extending in the z direction and a first flange 454 connected to the end of the cylindrical portion 453 on the z1 side in the z direction. The first flange 454 has a first surface 454a facing the z1 side in the z direction. The sealing resin 8 has a first recess 810. The first recess 810 is recessed from the resin main surface 81 toward the z1 side in the z direction. The first flange 454 is located on the z2 side in the z direction relative to the resin main surface 81. When the sealing resin 8 has the above-described first recess 810, the first surface 454a (first flange 454) can be appropriately positioned at a different position from the resin main surface 81 in the z direction.

[0097] Furthermore, in a plan view (as viewed in the z direction), the first recess 810 overlaps the entire cylindrical portion 453. This allows the metal pin 452 to be inserted into the holder 451 (cylindrical portion 453) while the lower end of the metal pin 452 is inserted into the first recess 810, improving workability during the press-fitting process.

[0098] The first recess 810 has a recess inner side surface 811 and a recess bottom surface 812. The recess bottom surface 812 faces the z1 side in the z direction and surrounds the first surface 454a when viewed in the z direction. Furthermore, the entire first surface 454a is exposed from the sealing resin 8. This configuration provides excellent visibility of the first surface 454a (first flange portion 454) surrounded by the recess bottom surface 812 in a plan view. This further improves the ease of press-fitting the metal pin 452 into the holder 451. Furthermore, the configuration in which the recess bottom surface 812 of the first recess 810 surrounds the first surface 454a (first flange portion 454) in a plan view allows the creepage distance along the surface of the sealing resin 8 to be increased between adjacent control terminals 45. This is preferable for increasing the withstand voltage of the adjacent control terminals 45.

[0099] The distance in the z direction between the resin main surface 81 and the first surface 454a (first dimension L1) is smaller than the length in the z direction of the holder 451 (second dimension L2). The ratio of the distance in the z direction between the resin main surface 81 and the first surface 454a (first dimension L1) to the length in the z direction of the holder 451 (second dimension L2) is, for example, 50% or more. With this configuration, it is possible to increase the creepage distance along the surface of the sealing resin 8 between adjacent control terminals 45 while avoiding an increase in the dimension of the sealing resin 8 in the z direction.

[0100] First Modification of First Embodiment (First Side): FIG. 23 shows a semiconductor device according to a first modification of the first embodiment. FIG. 23 is an enlarged cross-sectional view of a main portion of a semiconductor device A11 according to this modification, and is a cross-sectional view similar to FIG. 16. In FIGS. 23 to 29, elements that are the same as or similar to those in the semiconductor device A1 of the above embodiment are given the same reference numerals as in the above embodiment, and descriptions thereof will be omitted as appropriate. Furthermore, the configurations of the various parts in the modifications and embodiments of FIGS. 23 to 29 can be combined with each other as appropriate within the scope of not causing technical contradictions.

[0101] In the semiconductor device A11 of this modification, the configuration of the first recess 810 differs from that of the semiconductor device A1 of the above embodiment. In the semiconductor device A11, the first recess 810 has a recess edge 813 and a cylindrical inner surface 814. The cylindrical inner surface 814 is cylindrical and extends from the resin main surface 81 toward the z2 side in the z direction. The recess edge 813 is located at the lower end (the end on the z2 side in the z direction) of the cylindrical inner surface 814. The recess edge 813 contacts the first surface 454a. In this modification, the recess edge 813 contacts the radially intermediate position of the first surface 454a. A radially outer portion of the first surface 454a is covered with the sealing resin 8, and the remaining radially inner portion is exposed from the sealing resin 8. The outer peripheral edge of the first flange 454 surrounds the first recess 810 in a plan view. As a result, the diameter (maximum value of the inner diameter dimension) of the first recess 810 is smaller than the outer diameter dimension of the first flange portion 454. Note that, although the cylindrical inner surface 814 is cylindrical in the example shown in Fig. 23, a draft may be provided on the cylindrical inner surface 814. When a draft is provided on the cylindrical inner surface 814, the cylindrical inner surface 814 is formed into a conical shape that slopes so that the inner diameter dimension decreases toward the z2 side in the z direction.

[0102] In the semiconductor device A11 of this modification, the holder 451 constituting each control terminal 45 has a first surface 454a and a first outer surface 453a. The first surface 454a is located at the end of the holder 451 on the z1 side in the z direction. The first surface 454a is located at a position different from the resin main surface 81 in the z direction. The first outer surface 453a extends in the z direction and contacts the sealing resin 8. The metal pin 452 constituting each control terminal 45 protrudes toward the z1 side in the z direction beyond the resin main surface 81. With this configuration, the multiple control terminals 45 are arranged in an area surrounded by the resin main surface 81 (sealing resin 8) in a planar view. This semiconductor device A11 can be made smaller in planar view. Furthermore, the first surface 454a is located at a position different from the resin main surface 81 in the z direction. With this configuration, the creepage distance along the surface of the sealing resin 8 (such as the resin main surface 81) can be increased between adjacent control terminals 45. Therefore, the semiconductor device A11 is suitable for achieving a small size in plan view while increasing the withstand voltage of the adjacent control terminals 45. In addition, within the scope of the same configuration as the semiconductor device A1 of the above embodiment, the same effects as those of the above embodiment are achieved.

[0103] Second Modification of First Embodiment (First Side): Figure 24 shows a semiconductor device according to a second modification of the first embodiment. Figure 24 is an enlarged cross-sectional view of a main part of a semiconductor device A12 according to this modification, and is a cross-sectional view similar to Figure 16. In the semiconductor device A12 according to this modification, the configuration of the first recess 810 differs from that of the semiconductor device A1 according to the above embodiment.

[0104] The first recess 810 has a recess edge 813, a cylindrical inner surface 814, and a tapered inner surface 815. The cylindrical inner surface 814 is cylindrical and extends from the resin main surface 81 toward the z2 side in the z direction. The tapered inner surface 815 connects to the lower end (the end on the z2 side in the z direction) of the cylindrical inner surface 814. The recess edge 813 is located at the lower end (the end on the z2 side in the z direction) of the tapered inner surface 815. The tapered inner surface 815 is inclined so that the inner diameter dimension increases toward the z1 side in the z direction. The recess edge 813 contacts the first surface 454a. In this modification, the recess edge 813 contacts the radially middle position of the first surface 454a. A radially outer portion of the first surface 454a is covered with the sealing resin 8, and the remaining radially inner portion is exposed from the sealing resin 8. The outer peripheral edge of the first flange 454 surrounds the first recess 810 in plan view.

[0105] In the semiconductor device A12 of this modified example, the holder 451 constituting each control terminal 45 has a first surface 454a and a first outer surface 453a. The first surface 454a is located at the end of the holder 451 on the z1 side in the z direction. The first surface 454a is located at a position different from the resin main surface 81 in the z direction. The first outer surface 453a extends in the z direction and contacts the sealing resin 8. The metal pin 452 constituting each control terminal 45 protrudes toward the z1 side in the z direction beyond the resin main surface 81. With this configuration, the multiple control terminals 45 are arranged in an area surrounded by the resin main surface 81 (sealing resin 8) in a planar view. This semiconductor device A12 can be made smaller in planar view. Furthermore, the first surface 454a is located at a position different from the resin main surface 81 in the z direction. With this configuration, the creepage distance along the surface of the sealing resin 8 (such as the resin main surface 81) can be increased between adjacent control terminals 45. Therefore, the semiconductor device A12 is suitable for increasing the withstand voltage of the adjacent control terminals 45 while achieving miniaturization in plan view.

[0106] A recess edge 813 located on the z2 side of the first recess 810 in the z direction contacts the first surface 454a of the first flange 454. The first recess 810 also has a tapered inner surface 815 connected to the recess edge 813, and the tapered inner surface 815 has an inner diameter that increases toward the z1 side in the z direction. With this configuration, when the metal pin 452 is press-fitted into the holder 451, the metal pin 452 that has entered the first recess 810 can be guided by the tapered inner surface 815 toward the holder 451 (the cylindrical portion 453). This improves the workability of press-fitting the metal pin 452 into the holder 451. Additionally, within the same configuration as the semiconductor device A1 of the above embodiment, the same effects as those of the above embodiment can be achieved.

[0107] Third Modification of First Embodiment (First Side): Figure 25 shows a semiconductor device according to a third modification of the first embodiment. Figure 25 is an enlarged cross-sectional view of a main part of a semiconductor device A13 of this modification, and is a cross-sectional view similar to Figure 16. In the semiconductor device A13 of this modification, the configuration of the first recess 810 differs from that of the semiconductor device A1 of the above embodiment.

[0108] The first recess 810 has a recess edge 813, a cylindrical inner side surface 814, and a tapered inner side surface 815. In this modification, the cross-sectional shapes of the cylindrical inner side surface 814 and the tapered inner side surface 815 are the same as those of the semiconductor device A12 shown in FIG. 24 . Meanwhile, in this modification, the recess edge 813 contacts the radially inner end of the first surface 454a. As a result, the entire first surface 454a (or substantially the entire first surface 454a) is covered with the sealing resin 8. The outer peripheral edge of the first flange 454 surrounds the first recess 810 in a plan view.

[0109] In the semiconductor device A13 of this modified example, the holder 451 constituting each control terminal 45 has a first surface 454a and a first outer surface 453a. The first surface 454a is located at the end of the holder 451 on the z1 side in the z direction. The first surface 454a is located at a position different from the resin main surface 81 in the z direction. The first outer surface 453a extends in the z direction and contacts the sealing resin 8. The metal pin 452 constituting each control terminal 45 protrudes toward the z1 side in the z direction beyond the resin main surface 81. With this configuration, the multiple control terminals 45 are arranged in an area surrounded by the resin main surface 81 (sealing resin 8) in a planar view. This semiconductor device A13 can be made smaller in planar view. Furthermore, the first surface 454a is located at a position different from the resin main surface 81 in the z direction. With this configuration, the creepage distance along the surface of the sealing resin 8 (such as the resin main surface 81) can be increased between adjacent control terminals 45. Therefore, the semiconductor device A13 is suitable for increasing the withstand voltage of the adjacent control terminals 45 while achieving miniaturization in plan view.

[0110] A recess edge 813 located on the z2 side of the first recess 810 in the z direction contacts the first surface 454a of the first flange 454. The first recess 810 has a tapered inner surface 815 connected to the recess edge 813, and the inner diameter of the tapered inner surface 815 increases toward the z1 side in the z direction. With this configuration, when the metal pin 452 is press-fitted into the holder 451, the metal pin 452 inserted into the first recess 810 can be guided by the tapered inner surface 815 toward the holder 451 (the cylindrical portion 453). In this modification, the recess edge 813 contacts the radially inner end of the first surface 454a. This ensures that the metal pin 452 inserted into the first recess 810 can be reliably directed toward the holder 451 (the cylindrical portion 453) when the metal pin 452 is press-fitted into the holder 451. This further improves the workability when press-fitting the metal pin 452 into the holder 451. In addition, within the scope of the same configuration as the semiconductor device A1 of the above embodiment, the same effects as those of the above embodiment are achieved.

[0111] Fourth Modification of First Embodiment (First Side): Figure 26 shows a semiconductor device according to a fourth modification of the first embodiment. Figure 26 is an enlarged cross-sectional view of a main part of a semiconductor device A14 according to this modification, and is a cross-sectional view similar to Figure 16. In the semiconductor device A14 according to this modification, the configuration of the first recess 810 differs from that of the semiconductor device A1 according to the above embodiment.

[0112] The first recess 810 has a recess inner side surface 811 and a recess bottom surface 812. In the example shown in FIG. 26 , the recess inner side surface 811 has a conical shape that tapers so that the inner diameter decreases toward the z2 side in the z direction. The recess bottom surface 812 is connected to the end of the recess inner side surface 811 on the z2 side in the z direction and is a flat surface facing the z1 side in the z direction. The recess bottom surface 812 surrounds the first surface 454a of the holder 451 (first flange portion 454) in a planar view. In this modification, the recess bottom surface 812 is located closer to the z2 side in the z direction than the first surface 454a. Therefore, the first surface 454a and the recess bottom surface 812 are not flush with each other but are positioned at different positions in the z direction. In the example shown in FIG. 26 , the outer periphery of the first flange portion 454 is exposed from the sealing resin 8.

[0113] In the semiconductor device A14 of this modified example, the holder 451 constituting each control terminal 45 has a first surface 454a and a first outer surface 453a. The first surface 454a is located at the end of the holder 451 on the z1 side in the z direction. The first surface 454a is located at a position different from the resin main surface 81 in the z direction. The first outer surface 453a extends in the z direction and contacts the sealing resin 8. The metal pin 452 constituting each control terminal 45 protrudes toward the z1 side in the z direction beyond the resin main surface 81. With this configuration, the multiple control terminals 45 are arranged in an area surrounded by the resin main surface 81 (sealing resin 8) in a planar view. This semiconductor device A14 can be made smaller in planar view. Furthermore, the first surface 454a is located at a position different from the resin main surface 81 in the z direction. With this configuration, the creepage distance along the surface of the sealing resin 8 (such as the resin main surface 81) can be increased between adjacent control terminals 45. Therefore, the semiconductor device A14 is suitable for achieving a small size in plan view while increasing the withstand voltage of the adjacent control terminals 45. In addition, within the scope of the same configuration as the semiconductor device A1 of the above embodiment, the same effects as those of the above embodiment are achieved.

[0114] Fifth Modification of First Embodiment (First Side): Fig. 27 shows a semiconductor device according to a fifth modification of the first embodiment. Fig. 27 is an enlarged cross-sectional view of a main part of a semiconductor device A15 of this modification, and is a cross-sectional view similar to Fig. 16. The semiconductor device A15 of this modification further includes a first resin filling portion 89.

[0115] In this modification, the first resin filling portion 89 is filled into the first recess 810 so as to fill the first recess 810. The first resin filling portion 89 is made of, for example, epoxy resin like the sealing resin 8, but may be made of a material different from that of the sealing resin 8. This modification makes it possible to prevent foreign matter (including moisture) from entering the first recess 810 exposed from the sealing resin 8. The semiconductor device A15 configured as described above is preferable in terms of durability and reliability. In addition, the semiconductor device A15 exhibits the same effects as the semiconductor device A1 of the above embodiment.

[0116] Second Embodiment (First Side): Figures 28 and 29 show a semiconductor device according to a second embodiment of the present disclosure. Figure 28 is a perspective view showing a semiconductor device A2 according to this embodiment. Figure 29 is an enlarged cross-sectional view of a main part of the semiconductor device A2, similar to Figure 16. In the semiconductor device A2 according to this embodiment, the sealing resin 8 does not have the above-mentioned first recess 810. On the other hand, the semiconductor device A2 has a plurality of first protrusions 852.

[0117] Each of the multiple first protrusions 852 protrudes from the resin main surface 81 toward the z1 side in the z direction. The multiple protrusions 851 are provided corresponding to each of the multiple control terminals 45 and overlap the multiple control terminals 45 in a plan view. Each metal pin 452 of the multiple control terminals 45 protrudes from the first protrusion 852. The first protrusion 852 is cylindrical. The first protrusion 852 covers a portion of the holder 451 for each control terminal 45. As shown in FIG. 29 , the first outer surface 453 a of the cylindrical portion 453 of the holder 451 and the second surface 454 b of the first flange 454 are in contact with the sealing resin 8. Specifically, a portion of the first outer surface 453 a and the entire second surface 454 b are in contact with the first protrusion 852. Meanwhile, the first surface 454 a of the first flange 454 is exposed from the sealing resin 8.

[0118] The first protrusion 852 has a protrusion top surface 852a. In a plan view, the protrusion top surface 852a surrounds the first surface 454a of the holder 451 (first flange 454). The first surface 454a and the protrusion top surface 852a are flush with each other. The first surface 454a, which is flush with the protrusion top surface 852a, is located at a different position in the z direction from the resin main surface 81. Specifically, the first surface 454a is located on the z1 side in the z direction with respect to the resin main surface 81.

[0119] Next, the operation of this embodiment will be described.

[0120] In the semiconductor device A2 of this modified example, the holder 451 constituting each control terminal 45 has a first surface 454a and a first outer surface 453a. The first surface 454a is located at the end of the holder 451 on the z1 side in the z direction. The first surface 454a is located at a position different from the resin main surface 81 in the z direction. The first outer surface 453a extends in the z direction and contacts the sealing resin 8. The metal pin 452 constituting each control terminal 45 protrudes toward the z1 side in the z direction beyond the resin main surface 81. With this configuration, the multiple control terminals 45 are arranged in an area surrounded by the resin main surface 81 (sealing resin 8) in a planar view. This semiconductor device A2 can be made smaller in planar view. Furthermore, the first surface 454a is located at a position different from the resin main surface 81 in the z direction. With this configuration, the creepage distance along the surface of the sealing resin 8 (such as the resin main surface 81) can be increased between adjacent control terminals 45. Therefore, the semiconductor device A2 is suitable for increasing the withstand voltage of the adjacent control terminals 45 while achieving miniaturization in plan view.

[0121] The sealing resin 8 has a first protrusion 852. The first protrusion 852 protrudes from the resin main surface 81 toward the z1 side in the z direction. The first flange 454 is located on the z1 side in the z direction with respect to the resin main surface 81. When the sealing resin 8 has the above-described first protrusion 852, the first surface 454a (first flange 454) can be appropriately positioned at a position different from the resin main surface 81 in the z direction.

[0122] The semiconductor device according to the first aspect of the present disclosure is not limited to the above-described embodiment, and the specific configuration of each part of the semiconductor device can be freely designed and modified in various ways.

[0123] The first aspect of the present disclosure includes the embodiments described in Supplementary Notes 1 to 16 below.

[0124] Supplementary Note 1. A semiconductor device comprising: a conductive cylindrical holder, at least one terminal including a metal pin inserted into the holder; a terminal support body supporting the holder; and a sealing resin covering a portion of the holder and the terminal support body, wherein the sealing resin has a resin main surface facing one side in a thickness direction, the holder has a first surface located at an end on one side in the thickness direction and a first outer side surface extending in the thickness direction, the first surface is at a different position in the thickness direction from the resin main surface, the first outer side surface is in contact with the sealing resin, and the metal pin protrudes to one side in the thickness direction beyond the resin main surface. Supplementary Note 2. The semiconductor device according to Appendix 1, wherein the holder includes a cylindrical portion extending in the thickness direction and a first flange portion connected to one end of the cylindrical portion in the thickness direction, the first flange portion having the first surface facing one side in the thickness direction and a second surface located on the other side in the thickness direction of the first surface and facing the other side in the thickness direction, the cylindrical portion having the first outer surface, and the first outer surface and the second surface all being in contact with the sealing resin. Appendix 3. The semiconductor device according to Appendix 2, wherein the sealing resin has a first recess recessed from the resin main surface to the other side in the thickness direction, the first flange portion being located on the other side in the thickness direction relative to the resin main surface, and the first recess overlapping the entire cylindrical portion when viewed in the thickness direction. Appendix 4. The semiconductor device according to Appendix 3, wherein at least a portion of the first surface is exposed from the sealing resin. Appendix 5. Appendix 4: The semiconductor device according to Appendix 4, wherein the entire first surface is exposed from the sealing resin, and the first recess has a recess inner side surface connected to the resin main surface and a recess bottom surface connected to an end of the recess inner side surface on the other side in the thickness direction and facing one side in the thickness direction, and the recess bottom surface surrounds the first surface when viewed in the thickness direction. Appendix 6: The semiconductor device according to Appendix 3, wherein the first recess is located on the other side in the thickness direction and has a recess edge in contact with the first surface. Appendix 7: The semiconductor device according to Appendix 6, wherein the first recess has a tapered inner side surface connected to the recess edge, and the tapered inner side surface is inclined so that an inner diameter dimension increases toward one side in the thickness direction.Appendix 8. The semiconductor device according to Appendix 6, wherein an outer peripheral edge of the first flange surrounds the first recess when viewed in the thickness direction. Appendix 9. The semiconductor device according to any of Appendixes 3 to 8, wherein a first dimension, which is the distance in the thickness direction between the resin main surface and the first surface, is smaller than a second dimension, which is the length of the holder in the thickness direction. Appendix 10. The semiconductor device according to Appendix 9, wherein a ratio of the first dimension to the second dimension is 1 / 3 or more. Appendix 11. The semiconductor device according to Appendix 3, further comprising a first resin filling portion filled in the first recess. Appendix 12. The semiconductor device according to Appendix 2, wherein the sealing resin includes a first protruding portion protruding from the resin main surface to one side in the thickness direction, and wherein a portion of the first outer side surface and the entire second surface are in contact with the first protruding portion. Appendix 13. The semiconductor device according to Appendix 12, wherein the first protrusion has a protrusion top surface facing one side in the thickness direction, the protrusion top surface surrounds the first surface when viewed in the thickness direction, and the first surface and the protrusion top surface are flush with each other. Appendix 14. The semiconductor device according to Appendix 1 or 2, further comprising a support conductor supporting the terminal support body, and at least one semiconductor element electrically connected to the at least one terminal, the at least one semiconductor element being supported by the support conductor. Appendix 15. The semiconductor device according to Appendix 14, wherein the at least one terminal is a control terminal for controlling the at least one semiconductor element. Appendix 16. The semiconductor device described in Appendix 15, wherein the support conductor includes a first conductive portion and a second conductive portion spaced apart in a first direction perpendicular to the thickness direction; the at least one semiconductor element includes a first switching element joined to the first conductive portion and a second switching element joined to the second conductive portion; the control terminal includes a first control terminal for controlling the first switching element and a second control terminal for controlling the second switching element; and the terminal support includes a first support portion supporting the first control terminal and a second support portion supporting the second control terminal.

[0125] 30 to 44, a semiconductor device according to a first embodiment of the present disclosure will be described. The semiconductor device B1 of this embodiment includes a support substrate 11, a plurality of power terminals 13, a plurality of semiconductor elements 21, a thermistor 22, a first conductive member 31, a second conductive member 32, a plurality of wires, a plurality of control terminals 45, a control terminal support 48, and a sealing resin 50. The plurality of power terminals 13 include a first power terminal 14, two second power terminals 15, and two third power terminals 16. The plurality of wires include a plurality of first wires 41, a plurality of second wires 42, a plurality of third wires 43, and a fourth wire 44.

[0126] FIG. 30 is a perspective view showing the semiconductor device B1. FIG. 31 is a plan view showing the semiconductor device B1. FIG. 32 is a plan view showing the semiconductor device B1, with the sealing resin 50 indicated by imaginary lines. FIG. 33 is a plan view showing the semiconductor device B1, with the sealing resin 50 and the second conductive member 32 omitted from the plan view of FIG. 32. FIG. 34 is a plan view showing the first conductive member 31 omitted from the plan view of FIG. 33. FIG. 35 is a bottom view showing the semiconductor device B1. FIG. 36 is a cross-sectional view taken along line XXXVI-XXXVI in FIG. 32. FIGS. 37 and 38 are partially enlarged cross-sectional views of a portion of FIG. 36. FIG. 39 is a cross-sectional view taken along line XXXIX-XXXIX in FIG. 32. FIG. 40 is a cross-sectional view taken along line XL-XL in FIG. 32. FIG. 41 is a cross-sectional view taken along line XLI-XL in FIG. 32. Fig. 42 is a cross-sectional view taken along line XLII-XLII in Fig. 32. Fig. 43 is a cross-sectional view taken along line XLIII-XLIII in Fig. 32. Fig. 44 is a partially enlarged cross-sectional view of a part of Fig. 40.

[0127] In the following description, reference will be made to a thickness direction z, a first direction x, and a second direction y, which are perpendicular to one another. The thickness direction z corresponds to the thickness direction of the semiconductor device B1. Furthermore, "plan view" refers to a view in the thickness direction z. The first direction x is perpendicular to the thickness direction z. The second direction y is perpendicular to both the thickness direction z and the first direction x.

[0128] The semiconductor device B1 converts a DC power supply voltage applied to the first power terminal 14 and two second power terminals 15 into AC power using a plurality of semiconductor elements 21. The converted AC power is input from two third power terminals 16 to a power supply target such as a motor.

[0129] As shown in Figures 34, 36 to 39, 41, and 42, the support substrate 11 supports a plurality of semiconductor elements 21 in the thickness direction z. The support substrate 11 is formed, for example, from a DBC (Direct Bonded Copper) substrate. As shown in Figures 33 to 43, the support substrate 11 includes an insulating layer 111, support conductors 112, and a back surface metal layer 113. As shown in Figures 35 to 43, the support substrate 11 is covered with a sealing resin 50 except for a portion of the back surface metal layer 113.

[0130] 36 to 43, the insulating layer 111 includes a portion interposed between the support conductor 112 and the back surface metal layer 113 in the thickness direction z. The insulating layer 111 is made of a material with relatively high thermal conductivity. The insulating layer 111 is made of ceramics containing aluminum nitride (AlN), for example. The insulating layer 111 may be made of an insulating resin sheet instead of ceramics.

[0131] As shown in FIGS. 33 , 34 , and 36 to 43 , the support conductor 112 is located above (on the z1 side of) the insulating layer 111 in the thickness direction z. The composition of the support conductor 112 includes copper (Cu). As shown in FIGS. 42 and 43 , the support conductor 112 is surrounded by the periphery of the insulating layer 111 in a plan view. As shown in FIGS. 36 to 43 , the support conductor 112 has a main surface 1120. The main surface 1120 is a plane facing the z1 side in the thickness direction z. As shown in FIGS. 33 , 34 , and 36 to 43 , the support conductor 112 includes a first conductive portion 1121 and a second conductive portion 1122. The first conductive portion 1121 and the second conductive portion 1122 are each rectangular in a plan view. The first conductive portion 1121 and the second conductive portion 1122 are spaced apart from each other in the first direction x. The first conductive portion 1121 is located on the x1 side in the first direction x with respect to the second conductive portion 1122. Each of the plurality of semiconductor elements 21 is bonded to either the first conductive portion 1121 or the second conductive portion 1122.

[0132] As shown in Figures 36 to 43, the back surface metal layer 113 is located below (on the z2 side of) the insulating layer 111 in the thickness direction z. As shown in Figure 35, the back surface metal layer 113 is exposed from the sealing resin 50. A heat dissipation member (e.g., a heat sink) (not shown) can be attached to the lower surface (the surface facing the z2 side) of the back surface metal layer 113. The composition of the back surface metal layer 113 includes copper. The back surface metal layer 113 has a rectangular shape in a plan view. The back surface metal layer 113 is surrounded by the periphery of the insulating layer 111 in a plan view.

[0133] As shown in FIGS. 34 and 36 to 39 , each of the multiple semiconductor elements 21 is mounted on either the first conductive portion 1121 or the second conductive portion 1122. Each semiconductor element 21 is, for example, a metal-oxide-semiconductor field-effect transistor (MOSFET). Alternatively, each semiconductor element 21 may be a switching element such as an insulated gate bipolar transistor (IGBT) or a diode. In the description of the semiconductor device B1, the semiconductor element 21 is an n-channel MOSFET with a vertical structure. The semiconductor element 21 includes a compound semiconductor substrate. The compound semiconductor substrate contains silicon carbide (SiC) or silicon (Si).

[0134] As shown in FIGS. 34 and 36 to 39 , in the semiconductor device B1, the multiple semiconductor elements 21 include multiple first elements 21A and multiple second elements 21B. The structure of each of the multiple second elements 21B is the same as the structure of each of the multiple first elements 21A. The multiple first elements 21A are mounted on a first conductive portion 1121. The multiple first elements 21A are arranged along the second direction y. The multiple second elements 21B are mounted on a second conductive portion 1122. The multiple second elements 21B are arranged along the second direction y. Each of the multiple first elements 21A corresponds to a first switching element in the present disclosure. Each of the multiple second elements 21B corresponds to a second switching element in the present disclosure.

[0135] As shown in FIGS. 34, 37, and 38, the plurality of semiconductor elements 21 have a first electrode 211, a second electrode 212, a third electrode 213, and two fourth electrodes 214.

[0136] 37 and 38 , the first electrode 211 faces either the first conductive part 1121 or the second conductive part 1122. A current corresponding to the power before being converted by the semiconductor element 21 flows through the first electrode 211. In other words, the first electrode 211 corresponds to the drain electrode of the semiconductor element 21.

[0137] 34 , 37 , and 38 , the second electrode 212 is located on the opposite side of the first electrode 211 in the thickness direction z. A current corresponding to the power converted by the semiconductor element 21 flows through the second electrode 212. In other words, the second electrode 212 corresponds to the source electrode of the semiconductor element 21.

[0138] 34 , the third electrode 213 is located on the same side as the second electrode 212 in the thickness direction z. A gate voltage for driving the semiconductor element 21 is applied to the third electrode 213. That is, the third electrode 213 corresponds to the gate electrode of the semiconductor element 21. As shown in FIG. 34 , the area of ​​the third electrode 213 is smaller than the area of ​​the second electrode 212 in a plan view.

[0139] 34 , 37 , and 38 , the two fourth electrodes 214 are located on the same side as the second electrode 212 in the thickness direction z and adjacent to the third electrode 213 in the first direction x. In the illustrated example, the two fourth electrodes 214 are arranged on both sides of the third electrode 213 in the second direction y, with the third electrode 213 sandwiched therebetween. The potential of each fourth electrode 214 is equal to the potential of the second electrode 212. The fourth electrode 214 corresponds to a source sense electrode. Unlike the illustrated example, each semiconductor element 21 may include only one of the two fourth electrodes 214, or may include neither of the two fourth electrodes 214.

[0140] As shown in FIGS. 37 and 38 , the conductive bonding layer 23 is interposed between one of the first conductive portions 1121 and 1122 and the first electrode 211 of one of the plurality of semiconductor elements 21. The conductive bonding layer 23 is, for example, solder. Alternatively, the conductive bonding layer 23 may include a sintered body of metal particles. The first electrodes 211 of the plurality of first elements 21A are conductively bonded to the first conductive portion 1121 via the conductive bonding layer 23. As a result, the first electrodes 211 of the plurality of first elements 21A are electrically connected to the first conductive portion 1121. The first electrodes 211 of the plurality of second elements 21B are conductively bonded to the second conductive portion 1122 via the conductive bonding layer 23. As a result, the first electrodes 211 of the plurality of second elements 21B are electrically connected to the second conductive portion 1122. Unlike the present embodiment, the first elements 21A and the second elements 21B may be mounted on a metal member that is different from a part of the DBC substrate or the like. In this case, the metal member corresponds to the first conductive portion and the second conductive portion in the present disclosure. This metal member may be supported by the DBC substrate or the like, for example.

[0141] The plurality of power terminals 13 are electrically connected to the plurality of semiconductor elements 21, respectively. A current corresponding to the power before being converted by the plurality of semiconductor elements 21 or a current corresponding to the power after being converted by the plurality of semiconductor elements 21 flows through the plurality of power terminals 13. The plurality of power terminals 13 include a first power terminal 14, two second power terminals 15, and two third power terminals 16.

[0142] As shown in FIGS. 33 and 39 , the first power terminal 14 is joined to the first conductive portion 1121. This joining is not limited in any way and may be done using a conductive joining material (e.g., solder) (not shown), laser welding, or crimping. The first power terminal 14 is electrically connected to the first electrodes 211 of the multiple first elements 21A via the first conductive portion 1121. The first power terminal 14 is a P terminal (positive electrode) to which a DC power supply voltage to be converted into power is applied. As shown in FIG. 33 , the first power terminal 14 is located on the opposite side of the second conductive portion 1122 in the first direction x, with the first conductive portion 1121 sandwiched therebetween. The first power terminal 14 extends from the first conductive portion 1121 toward the x1 side in the first direction x and protrudes from the sealing resin 50 toward the x1 side in the first direction x. 32 , the first power terminal 14 includes a portion covered with the sealing resin 50 and a portion exposed from the sealing resin 50. In the first power terminal 14, the portion covered with the sealing resin 50 is joined to the first conductive part 1121. In addition, in the first power terminal 14, the portion exposed from the sealing resin 50 is used as the aforementioned P terminal of the semiconductor device B1.

[0143] A second conductive member 32 is joined to the two second power terminals 15. The two second power terminals 15 are electrically connected to the second electrodes 212 of the multiple second elements 21B via the second conductive member 32. The two second power terminals 15 are N terminals (negative electrodes) to which a DC power supply voltage to be converted is applied. The two second power terminals 15 are spaced apart from each other in the second direction y. The first power terminal 14 is located between the two second power terminals 15. As shown in FIG. 33 , the two second power terminals 15 are located on the same side as the first power terminal 14 with respect to the first conductive portion 1121 and the second conductive portion 1122 in the first direction x. The two second power terminals 15 are spaced apart from the first conductive portion 1121 and the second conductive portion 1122. Each of the two second power terminals 15 extends in the first direction x and protrudes from the sealing resin 50 toward the x1 side in the first direction x. As shown in Fig. 32 , each of the two second power terminals 15 includes a portion covered with the sealing resin 50 and a portion exposed from the sealing resin 50. In each second power terminal 15, a second conductive member 32 is joined to the portion covered with the sealing resin 50. In addition, in each second power terminal 15, the portion exposed from the sealing resin 50 is used as the aforementioned N terminal of the semiconductor device B1.

[0144] As shown in FIGS. 33 and 36 , the two third power terminals 16 are each joined to the second conductive portion 1122. This joining is not limited to any particular method and may be performed using a conductive joining material (e.g., solder) (not shown), laser welding, or crimping. Each of the two third power terminals 16 is electrically connected to the first electrodes 211 of the plurality of second elements 21B via the second conductive portion 1122. Each of the two third power terminals 16 is also electrically connected to the second electrodes 212 of the plurality of first elements 21A via the second conductive portion 1122 and the first conductive member 31. AC power converted by the plurality of semiconductor elements 21 (the plurality of first elements 21A and the plurality of second elements 21B) is output from the two third power terminals 16. In other words, each of the two third power terminals 16 is an output terminal for the AC power. The two third power terminals 16 are spaced apart from each other in the second direction y. As shown in FIG. 33 , the two third power terminals 16 are located on the opposite side of the first conductive portion 1121 in the first direction x, with the second conductive portion 1122 sandwiched therebetween. Each of the two third power terminals 16 extends from the second conductive portion 1122 toward the x2 side in the first direction x and protrudes from the sealing resin 50 toward the x2 side in the first direction x. As shown in FIG. 32 , each of the two third power terminals 16 includes a portion covered with the sealing resin 50 and a portion exposed from the sealing resin 50. In each third power terminal 16, the portion covered with the sealing resin 50 is joined to the second conductive portion 1122. In addition, in each third power terminal 16, the portion exposed from the sealing resin 50 is used as the aforementioned output terminal of the semiconductor device B1.

[0145] In this embodiment, the semiconductor device B1 includes four first elements 21A and four second elements 21B. However, the number of first elements 21A and the number of second elements 21B are not limited to this configuration and may be changed as appropriate depending on the performance required of the semiconductor device B1. In the example shown in FIG. 34 , four first elements 21A and four second elements 21B are arranged. The number of first elements 21A and the number of second elements 21B may be two, three, or five or more. The number of first elements 21A and the number of second elements 21B may be equal to or different from each other. The number of first elements 21A and the number of second elements 21B is determined by the current capacity handled by the semiconductor device B1.

[0146] The semiconductor device B1 is configured, for example, as a half-bridge switching circuit. In this case, the multiple first elements 21A configure an upper arm circuit of the semiconductor device B1, and the multiple second elements 21B configure a lower arm circuit. In the upper arm circuit, the multiple first elements 21A are connected in parallel with each other, and in the lower arm circuit, the multiple second elements 21B are connected in parallel with each other. Each first element 21A and each second element 21B are connected in series to configure a bridge layer.

[0147] Each of the plurality of control terminals 45 is a pin-shaped terminal for controlling the driving of each of the first elements 21A and each of the second elements 21B. Each of the plurality of control terminals 45 is, for example, a press-fit terminal. The plurality of control terminals 45 includes a plurality of first control terminals 46A to 46C and a plurality of second control terminals 47A to 47D. The plurality of first control terminals 46A to 46C are used for controlling each of the first elements 21A, etc. The plurality of second control terminals 47A to 47D are used for controlling each of the second elements 21B, etc.

[0148] The multiple first control terminals 46A to 46C are arranged at intervals in the second direction y. As shown in Figures 34, 39, and 40, each of the first control terminals 46A to 46C is supported by the first conductive portion 1121 via a control terminal support body 48 (a first support portion 48A described below). As shown in Figures 33 and 34, each of the first control terminals 46A to 46C is located in the first direction x between the multiple first elements 21A and the first power terminal 14 and two second power terminals 15.

[0149] The first control terminal 46A is a terminal (gate terminal) for inputting a drive signal for the plurality of first elements 21A. A drive signal for driving the plurality of first elements 21A is input to the first control terminal 46A (for example, a gate voltage is applied).

[0150] The first control terminal 46B is a terminal (source sense terminal) for detecting source signals of the multiple first elements 21 A. The first control terminal 46B detects a voltage (a voltage corresponding to a source current) applied to each second electrode 212 (source electrode) of the multiple first elements 21 A.

[0151] The first control terminal 46C is a terminal (drain sense terminal) for detecting the drain voltages of the multiple first elements 21 A. The first control terminal 46C detects the voltage (voltage corresponding to the drain current) applied to each first electrode 211 (drain electrode) of the multiple first elements 21 A.

[0152] The second control terminals 47A to 47D are spaced apart in the second direction y. As shown in Figures 34, 39, 43, etc., each of the second control terminals 47A to 47D is supported by the second conductive portion 1122 via a control terminal support 48 (a second support portion 48B described below). As shown in Figures 33 and 34, each of the second control terminals 47A to 47D is located between the second elements 21B and the two third power terminals 16 in the first direction x.

[0153] The second control terminal 47A is a terminal (gate terminal) for inputting drive signals for the multiple second elements 21B. A drive signal for driving the multiple second elements 21B is input to the second control terminal 47A (for example, a gate voltage is applied). The second control terminal 47B is a terminal (source sense terminal) for detecting source signals for the multiple second elements 21B. The second control terminal 47B detects a voltage (a voltage corresponding to a source current) applied to each second electrode 212 (source electrode) of the multiple second elements 21B. The second control terminal 47C and the second control terminal 47D are not electrically connected to any of the multiple second elements 21B. The second control terminal 47C and the second control terminal 47D are terminals electrically connected to the thermistor 22.

[0154] Each of the plurality of control terminals 45 (the plurality of first control terminals 46A to 46C and the plurality of second control terminals 47A to 47D) includes a holder 451 and a metal pin 452.

[0155] The holder 451 is made of a conductive material. The holder 451 is disposed on the main surface 1120 of the support conductor 112 (support substrate 11). In this embodiment, the holder 451 is bonded to the control terminal support 48 (a first metal layer 482 described below) via a conductive bonding layer 459, as shown in FIG. 44 . As shown in FIG. 44 , the holder 451 includes a cylindrical portion 453, a first flange portion 454, and a second flange portion 455.

[0156] The tubular portion 453 extends in the thickness direction z and has, for example, a cylindrical shape.

[0157] The first flange 454 is connected to the end of the cylindrical portion 453 on the z1 side in the thickness direction z. The first flange 454 has a first surface 454a. The first surface 454a faces the z1 side in the thickness direction z. The first surface 454a is located at the end of the holder 451 on the z1 side in the thickness direction z. The first surface 454a is annular (circular in the illustrated example) when viewed in the thickness direction z.

[0158] The second flange 455 is connected to the end portion on the z2 side in the thickness direction z of the cylindrical portion 453. In the present embodiment, the second flange 455 is joined to the control terminal support body 48 (a first metal layer 482 described below) via a conductive bonding layer 459.

[0159] A metal pin 452 is inserted through a first flange 454 and a part of the cylindrical portion 453 of the holder 451. The entire holder 451 is exposed from the sealing resin 50.

[0160] The metal pin 452 is a rod-shaped member extending in the thickness direction z. The metal pin 452 is supported by the holder 451 by being press-fitted into the holder 451. The metal pin 452 is electrically connected to the control terminal support body 48 (a first metal layer 482 described below) via the holder 451 and the conductive bonding layer 459. The metal pin 452 protrudes toward the z1 side in the thickness direction z from the upper surface of the sealing resin 50 (a resin main surface 51 described below).

[0161] The control terminal support body 48 supports the plurality of control terminals 45. The control terminal support body 48 is interposed between the main surface 1120 of the first conductive portion 1121 and the main surface 1120 of the second conductive portion 1122 and the plurality of control terminals 45 in the thickness direction z.

[0162] The control terminal support 48 includes a first support portion 48A and a second support portion 48B. The first support portion 48A is disposed on the first conductive portion 1121 and supports a plurality of first control terminals 46A to 46C among the plurality of control terminals 45. As shown in FIG. 44 , the first support portion 48A is bonded to the first conductive portion 1121 via a bonding layer 49. The bonding layer 49 may be conductive or insulating, and may be made of solder, for example. The second support portion 48B is disposed on the second conductive portion 1122 and supports a plurality of second control terminals 47A to 47D among the plurality of control terminals 45. Like the first support portion 48A, the second support portion 48B is bonded to the second conductive portion 1122 via a bonding layer (not shown).

[0163] The control terminal support 48 (each of the first support portion 48A and the second support portion 48B) is made of, for example, a direct bonded copper (DBC) substrate. The control terminal support 48 has an insulating layer 481, a first metal layer 482, and a second metal layer 483 stacked on top of each other.

[0164] The insulating layer 481 is made of, for example, ceramics and has, for example, a rectangular shape in plan view.

[0165] As shown in FIG. 44 and other figures, the first metal layer 482 is formed on the upper surface of the insulating layer 481. Each control terminal 45 is provided upright on the first metal layer 482. The first metal layer 482 includes, for example, Cu (copper) or a Cu (copper) alloy. As shown in FIG. 34 and other figures, the first metal layer 482 includes a first portion 482A, a second portion 482B, a third portion 482C, a fourth portion 482D, and a fifth portion 482E. The first portion 482A, the second portion 482B, the third portion 482C, the fourth portion 482D, and the fifth portion 482E are spaced apart and insulated from one another.

[0166] The fourth portion 482D has a plurality of first wires 41 joined thereto and is electrically connected to the third electrodes 213 (gate electrodes) of the first elements 21A (second elements 21B) via the respective first wires 41. A plurality of third wires 43 are connected between the fourth portion 482D and the first portion 482A. As a result, the first portion 482A is electrically connected to the third electrodes 213 (gate electrodes) of the first elements 21A (second elements 21B) via the respective third wires 43 and first wires 41. As shown in FIG. 34 , a first control terminal 46A is joined to the first portion 482A of the first support portion 48A, and a second control terminal 47A is joined to the first portion 482A of the second support portion 48B.

[0167] The second portion 482B has a plurality of second wires 42 joined thereto, and is electrically connected to the fourth electrode 214 (source sense electrode) of each first element 21A (each second element 21B) via each second wire 42. As shown in Fig. 34 , a first control terminal 46B is joined to the second portion 482B of the first support 48A, and a second control terminal 47B is joined to the second portion 482B of the second support 48B.

[0168] The second control terminal 47C is joined to the third portion 482C. As shown in FIG. 34 , the second control terminal 47C is joined to the third portion 482C of the second support portion 48B. The first control terminal 46C and the second control terminal 47D are joined to the fifth portion 482E. The first control terminal 46C is joined to the fifth portion 482E of the first support portion 48A. A fourth wire 44 is joined to the fifth portion 482E of the first support portion 48A, and electrical conduction is established with the first electrode 211 (drain electrode) of each first element 21A via the fourth wire 44. The second control terminal 47D is joined to the fifth portion 482E of the second support portion 48B.

[0169] The thermistor 22 is conductively joined across the third portion 482C and the fifth portion 482E of the second support portion 48B. The thermistor 22 is, for example, an NTC (Negative Temperature Coefficient) thermistor. An NTC thermistor has the characteristic that its resistance decreases gradually with increasing temperature. The thermistor 22 is used as a temperature detection sensor for the semiconductor device B1.

[0170] Each of the plurality of first wires 41, the plurality of second wires 42, the plurality of third wires 43, and the fourth wire 44 is, for example, a bonding wire. The constituent material of each of the first wires 41, the plurality of second wires 42, the plurality of third wires 43, and the fourth wire 44 is not particularly limited and may include, for example, Au (gold), Al (aluminum), or Cu (copper). Note that the plurality of first wires 41, the plurality of second wires 42, the plurality of third wires 43, and the fourth wire 44 are omitted in Figures 32, 36 to 40, and 43.

[0171] As shown in Fig. 44, the second metal layer 483 is formed on the lower surface (surface facing the z2 side in the thickness direction z) of the insulating layer 481. As shown in Fig. 44, the second metal layer 483 of the first support portion 48A is joined to the first conductive portion 1121 via a bonding layer 49. The second metal layer 483 of the second support portion 48B is joined to the second conductive portion 1122 via a bonding layer (not shown), similar to the second metal layer 483 of the first support portion 48A.

[0172] As shown in FIGS. 33 and 36 , the first conductive member 31 is conductively bonded to the second electrodes 212 of the plurality of first elements 21A and the second conductive portion 1122. This electrically connects the second electrodes 212 of the plurality of first elements 21A to the second conductive portion 1122. The composition of the first conductive member 31 is not particularly limited and may include, for example, copper. The first conductive member 31 is a metal clip. As shown in FIGS. 33 and 36 , the first conductive member 31 has a main body 311, a plurality of first joint portions 312, and a plurality of second joint portions 313.

[0173] The main body portion 311 forms a main portion of the first conductive member 31. As shown in FIG. 33 , the main body portion 311 extends in the second direction y. As shown in FIGS. 33 and 36 , the main body portion 311 straddles the first conductive portion 1121 and the second conductive portion 1122. As shown in FIG. 33 , a plurality of through holes 310 are formed in the main body portion 311. Each of the plurality of through holes 310 penetrates the main body portion 311 in the thickness direction z. In a plan view, the plurality of through holes 310 overlap between the first conductive portion 1121 and the second conductive portion 1122. This allows the sealing resin 50 to flow smoothly downward in the thickness direction z of the main body portion 311 (toward the z2 side in the thickness direction z) during the formation of the sealing resin 50.

[0174] As shown in FIGS. 33 and 36 , the multiple first joints 312 are individually joined to the second electrodes 212 of the multiple first elements 21A. Each of the multiple first joints 312 faces one of the second electrodes 212 of the multiple first elements 21A. In a plan view, each first joint 312 extends from the main body 311 toward the x1 side in the first direction x. In the illustrated example, the multiple first joints 312 are bifurcated from the main body 311, but they do not necessarily need to be bifurcated. The base end of each first joint 312 (the end connected to the main body 311) is bent downward in the thickness direction z (toward the z2 side in the thickness direction z). Therefore, the tip of each first joint 312 (the end opposite to the end connected to the main body 311) is located downward in the thickness direction z (toward the z2 side in the thickness direction z) relative to the main body 311 in the thickness direction z.

[0175] 33 and 36 , the multiple second joints 313 are joined to the second conductive portion 1122. Each of the multiple second joints 313 faces the second conductive portion 1122. In a plan view, each second joint 313 extends from the main body portion 311 toward the x1 side in the first direction x. The base end of each second joint 313 (the end connected to the main body portion 311) is bent downward in the thickness direction z (toward the z2 side in the thickness direction z). Therefore, the tip end of each second joint 313 (the end opposite to the end connected to the main body portion 311) is located downward in the thickness direction z (toward the z2 side in the thickness direction z) relative to the main body portion 311 in the thickness direction z.

[0176] 37 , the semiconductor device B1 further includes a first conductive bonding layer 33. The first conductive bonding layer 33 is interposed between the second electrodes 212 of the plurality of first elements 21A and the plurality of first bonding portions 312. The first conductive bonding layer 33 conductively bonds the second electrodes 212 of the plurality of first elements 21A to the plurality of first bonding portions 312. The first conductive bonding layer 33 is, for example, solder. Alternatively, the first conductive bonding layer 33 may include a sintered body of metal particles.

[0177] 36 , the semiconductor device B1 further includes a second conductive bonding layer 34. The second conductive bonding layer 34 is interposed between the second conductive portion 1122 and the second bonding portion 313. The second conductive bonding layer 34 conductively bonds the second conductive portion 1122 and the second bonding portion 313. The second conductive bonding layer 34 is, for example, solder. Alternatively, the second conductive bonding layer 34 may include a sintered body of metal particles.

[0178] As shown in FIG. 32 , the second conductive member 32 is conductively joined to the second electrodes 212 of the plurality of second elements 21B and the two second power terminals 15. This electrically connects the second electrodes 212 of the plurality of second elements 21B to the two second power terminals 15. The composition of the second conductive member 32 is not particularly limited and may include, for example, copper. The second conductive member 32 is a metal clip. As shown in FIGS. 32 , 36 , and 39 to 42 , the second conductive member 32 has a pair of main body portions 321, a plurality of third joint portions 322, a pair of fourth joint portions 324, a plurality of intermediate portions 326, a plurality of cross beam portions 327, and a pair of hanging portions 328.

[0179] As shown in Fig. 32 , the pair of main bodies 321 are positioned apart from each other in the second direction y. The pair of main bodies 321 extend in the first direction x. As shown in Figs. 36 and 40 , the pair of main bodies 321 are arranged parallel to the upper surfaces of the first conductive part 1121 and the second conductive part 1122. The pair of main bodies 321 are positioned further away from the first conductive part 1121 and the second conductive part 1122 than the main body part 311 of the first conductive member 31.

[0180] 32 , 41 , and 42 , the intermediate portions 326 are spaced apart from one another in the second direction y and are located between the pair of main body portions 321 in the second direction y. The intermediate portions 326 extend in the first direction x.

[0181] As shown in FIGS. 32 and 42 , the third joints 322 are individually joined to the second electrodes 212 of the second elements 21B. Each of the third joints 322 faces one of the second electrodes 212 of the second elements 21B. In a plan view, the third joints 322 extend in the second direction y from the intermediate portions 326. The base end of each third joint 322 (the end connected to the intermediate portion 326) is bent downward in the thickness direction z (toward the z2 side in the thickness direction z). Therefore, the tip of each third joint 322 (the end opposite to the end connected to the intermediate portion 326) is located downward in the thickness direction z (toward the z2 side in the thickness direction z) from the intermediate portion 326 in the thickness direction z.

[0182] 32 and 36 , the pair of fourth joint portions 324 are individually joined to the two second power terminals 15. Each of the pair of fourth joint portions 324 faces a corresponding one of the two second power terminals 15.

[0183] As shown in Fig. 32 , the multiple cross beam portions 327 are arranged along the second direction y. In a plan view, the multiple cross beam portions 327 include regions that individually overlap the multiple first joint portions 312 of the first conductive member 31. As shown in Fig. 32 and Fig. 41 , of the multiple cross beam portions 327, the cross beam portion 327 located at the center in the second direction y is connected on both sides in the second direction y to the multiple intermediate portions 326. Of the multiple cross beam portions 327, the remaining two cross beam portions 327 are connected on both sides in the second direction y to one of the pair of main body portions 321 and one of the multiple intermediate portions 326.

[0184] 32 and 41 , the pair of hanging portions 328 are individually connected to the pair of main body portions 321. As shown in FIG. 41 , each of the pair of hanging portions 328 extends downward in the thickness direction z (toward the z2 side in the thickness direction z) from the corresponding one of the pair of main body portions 321. Each of the pair of hanging portions 328 is connected to the outer edge of the corresponding one of the pair of main body portions 321 in the second direction y. In the illustrated example, the lower ends (the edges on the z2 side in the thickness direction z) of the pair of hanging portions 328 overlap the first conductive portion 1121 when viewed along the second direction y.

[0185] 38 , the semiconductor device B1 further includes a third conductive bonding layer 35. The third conductive bonding layer 35 is interposed between the second electrodes 212 of the plurality of second elements 21B and the plurality of third bonding portions 322. The third conductive bonding layer 35 conductively bonds the second electrodes 212 of the plurality of second elements 21B to the plurality of third bonding portions 322. The third conductive bonding layer 35 is, for example, solder. Alternatively, the third conductive bonding layer 35 may include a sintered body of metal particles.

[0186] 36 , the semiconductor device B1 further includes a fourth conductive bonding layer 36. The fourth conductive bonding layer 36 is interposed between the two second power terminals 15 and the pair of fourth joints 324. The fourth conductive bonding layer 36 conductively bonds the two second power terminals 15 and the pair of fourth joints 324. The fourth conductive bonding layer 36 is, for example, solder. Alternatively, the fourth conductive bonding layer 36 may include a sintered body of metal particles.

[0187] As shown in FIGS. 30 to 43 , the sealing resin 50 covers the semiconductor elements 21, the first conductive members 31, the second conductive members 32, the first wires 41, the second wires 42, and the third wires 43. Furthermore, the sealing resin 50 covers a portion of each of the support substrate 11, the power terminals 13, and the control terminal support 48. The sealing resin 50 has electrical insulation properties. The sealing resin 50 includes, for example, a black epoxy resin. The sealing resin 50 is formed, for example, by molding. As shown in FIGS. 30 to 32 and 35 to 43 , the sealing resin 50 has a resin main surface 51, a resin back surface 52, resin side surfaces 531 to 534, a plurality of first recesses 511, and a pair of recesses 531 a.

[0188] As shown in FIGS. 36 and 39 to 43, the resin main surface 51 faces the same direction in the thickness direction z as the upper surface (main surface 1120) of the first conductive portion 1121 and the upper surface (main surface 1120) of the second conductive portion 1122. Metal pins 452 of the control terminals 45 (the first control terminals 46A to 46C and the second control terminals 47A to 47D) protrude from the resin main surface 51. As shown in FIGS. 36 and 39 to 43, the resin back surface 52 faces the opposite side of the resin main surface 51 in the thickness direction z. As shown in FIG. 35, the resin back surface 52 has a frame shape in a plan view that surrounds the lower surface (the surface facing the z2 side in the thickness direction z) of the back surface metal layer 113 of the support substrate 3. The back surface metal layer 113 of the support substrate 11 is exposed from the resin back surface 52. The lower surface of the rear surface metal layer 113 (the surface facing the z2 side in the thickness direction z) is flush with the resin rear surface 52, for example.

[0189] As shown in Figures 31, 32, 36, and 39, the resin side surface 531 and the resin side surface 532 are spaced apart from each other in the first direction x. The resin side surface 531 and the resin side surface 532 face opposite each other in the first direction x and extend in the second direction y. The resin side surface 531 and the resin side surface 532 are connected to the resin main surface 51. The resin side surface 531 faces the x1 side in the first direction x, and the resin side surface 532 faces the x2 side in the first direction x. A first power terminal 14 and two second power terminals 15 each protrude from the resin side surface 531. Two third power terminals 16 each protrude from the resin side surface 532.

[0190] 31 , 32 , and 40 to 43 , the resin side surface 533 and the resin side surface 534 are spaced apart from each other in the second direction y. The resin side surface 533 and the resin side surface 534 face opposite each other in the second direction y and extend in the first direction x. The resin side surface 533 and the resin side surface 534 are connected to the resin main surface 51 and the resin back surface 52. The resin side surface 533 faces the y1 side in the second direction y, and the resin side surface 534 faces the y2 side in the second direction y.

[0191] 30 , 39 , 40 , 43 , and 44 , each of the multiple first recesses 511 is recessed from the resin main surface 51 toward the z2 side in the thickness direction z. In the present embodiment, the multiple first recesses 511 are individually provided corresponding to the multiple control terminals 45. The multiple control terminals 45 are individually arranged corresponding to the multiple first recesses 511.

[0192] 31 , 39 , 40 , 43 , and 44 , the first recess 511 overlaps with all of the holders 451 of the corresponding control terminals 45 in a plan view. In the present embodiment, as shown in Fig. 44 , the first recess 511 has a first recess inner surface 512 and a chamfered portion 515. The first recess inner surface 512 extends in the thickness direction z and has a conical shape that slopes downward so that the inner diameter decreases toward the z2 side in the thickness direction z.

[0193] 44 , the first recess inner surface 512 has a first edge 513 and a second edge 514. The first edge 513 is located at the end of the first recess inner surface 512 on the z2 side in the thickness direction z, and is in contact with the control terminal support body 48 (first metal layer 482). The second edge 514 is located at the end of the first recess inner surface 512 on the z1 side in the thickness direction z. The second edge 514 surrounds the first edge 513 in a plan view.

[0194] The chamfered portion 515 is connected to the resin main surface 51 and is interposed between the resin main surface 51 and the first recess inner side surface 512. The specific shape of the chamfered portion 515 is not particularly limited, and examples thereof include a curved R-chamfered shape or a C-chamfered shape. In the illustrated example, the chamfered portion 515 has an R-chamfered shape.

[0195] The first recess 511 is a trace of the sealing resin 50 formed by molding while pressing the control terminal support 48 with, for example, a pin having a shape corresponding to the first recess 511. FIG. 45 is a cross-sectional view similar to FIG. 44 , showing a manufacturing process of the semiconductor device B1. As shown in FIG. 45 , for example, a cylindrical pin 911 is provided in a mold 91 for molding. A holder 451 is placed in the inner space of the cylindrical pin 911, and a fluid resin material is injected into a cavity 919 of the mold 91 while pressing the lower end (the end on the z2 side in the thickness direction z) of the cylindrical pin 911 against the control terminal support 48 (first metal layer 482). As can be seen from FIGS. 44 and 45 , the first recess inner surface 512 of the first recess 511 has a draft angle corresponding to the outer peripheral surface of the cylindrical pin 911. Also, as shown in FIG. 45 , the mold 91 has a rounded corner 915 at the base of the cylindrical pin 911. The chamfered portion 515 of the first recess 511 has a shape corresponding to the rounded corner portion 915 of the mold 91. After the sealing resin 50 is formed by molding using the mold 91, the holder 451 arranged in the inner space of the cylindrical pin 911 is entirely exposed from the sealing resin 50.

[0196] Depending on the positioning of the holder 451 for the control terminal 45, the lower end of the cylindrical pin 911 shown in FIG. 45 may be pressed across the first metal layer 482 and the insulating layer 481 during molding. On the insulating layer 481, there is a step in the thickness direction z at the boundary between the portion where the first metal layer 482 is formed and the portion where the first metal layer 482 is not formed. Therefore, when the lower end of the cylindrical pin 911 is pressed across the first metal layer 482 and the insulating layer 481, a gap may be formed between the lower end of the cylindrical pin 911 and the control terminal support body 48 (insulating layer 481). To address this issue, a resist layer may be formed on the portion of the upper surface of the insulating layer 481 where the first metal layer 482 is not formed, thereby eliminating the step between the portion where the first metal layer 482 is formed and the portion where the first metal layer 482 is not formed on the insulating layer 481. Alternatively, the lower end of the cylindrical pin 911 may be made of a cushioning material. In this case, when the cylindrical pin 911 is pressed against the control terminal support 48, the step is absorbed by the cushioning material, and no gap is created between the lower end of the cylindrical pin 911 and the control terminal support 48 (insulating layer 481).

[0197] The method for forming the first recess 511 is not limited to the method described above with reference to Fig. 45. For example, the sealing resin 50 may be formed by molding while pressing the control terminal support body 48 with a columnar solid pin corresponding to the first recess 511. In this case, the holder 451 is not disposed on the control terminal support body 48 during molding. After molding, the holder 451 is disposed in the first recess 511 on the control terminal support body 48.

[0198] In the illustrated example, the first surface 454a of the holder 451 (first flange portion 454) is located on the z2 side in the thickness direction z with respect to the resin main surface 51. As a result, the holder 451 is entirely housed in the first recess 511.

[0199] 31 , the pair of recesses 531a are recessed from the resin side surface 531 toward the x2 side in the first direction x. The pair of recesses 531a extend from the resin main surface 51 to the resin back surface 52 in the thickness direction z. The pair of recesses 531a are located on both sides of the first power terminal 14 in the second direction y.

[0200] Next, the operation of this embodiment will be described.

[0201] The holders 451 constituting each control terminal 45 are disposed on the main surface 1120 of the support conductor 112 (support substrate 11). The metal pins 452 constituting each control terminal 45 protrude further toward the z1 side in the thickness direction z than the resin main surface 51. With this configuration, the multiple control terminals 45 are disposed in an area surrounded by the resin main surface 51 (sealing resin 50) in a plan view. Such a semiconductor device B1 can be made smaller in size in a plan view.

[0202] All of the holders 451 of the control terminals 45 are exposed from the sealing resin 50. This configuration can prevent the sealing resin 50 from flowing into the holders 451 into which the metal pins 452 are inserted. Therefore, in the semiconductor device B1, electrical continuity between the holders 451 and the metal pins 452 can be properly maintained, and the control terminals 45 configured to include the holders 451 and the metal pins 452 can function properly.

[0203] The control terminals 45 are arranged in first recesses 511 in the sealing resin 50. In this embodiment, the sealing resin 50 has a plurality of first recesses 511, and the control terminals 45 are individually arranged corresponding to the plurality of first recesses 511. The first recesses 511 (first recess inner surface 512) have a first edge 513 that contacts the control terminal support body 48 (first metal layer 482). This configuration allows the creepage distance between adjacent control terminals 45 along the surface of the sealing resin 50 (the resin main surface 51, the first recess inner surface 512 of the first recess 511, etc.) to be increased. Therefore, the semiconductor device B1 is suitable for increasing the withstand voltage of adjacent control terminals 45 while achieving miniaturization in a planar view.

[0204] In a plan view, the first recess 511 overlaps the entire holder 451 of the corresponding control terminal 45. With this configuration, the holder 451 surrounded by the first recess 511 is highly visible in a plan view. This further improves the workability when press-fitting the metal pin 452 into the holder 451.

[0205] Modifications of the semiconductor device according to the second aspect of the present disclosure will be described below. The configurations of the components in each modification can be combined with each other as long as no technical contradiction occurs.

[0206] 46 to 48 show a semiconductor device according to a first modified example of the first embodiment of the second side. FIG. 46 is a plan view showing a semiconductor device B11 of this modified example. FIG. 47 is a cross-sectional view taken along line XLVII-XLVII in FIG. 46. FIG. 48 is a cross-sectional view taken along line XLVIII-XLVIII in FIG. 46. In the drawings from FIG. 46 onwards, elements that are the same as or similar to those in the semiconductor device B1 of the above embodiment are given the same reference numerals as in the above embodiment, and descriptions thereof will be omitted where appropriate.

[0207] In the semiconductor device B11 of this modification, the configuration of the first recesses 511 in the sealing resin 50 differs from that of the semiconductor device B1 of the above embodiment. As shown in FIGS. 46 to 48 , in the semiconductor device B11, the sealing resin 50 has two first recesses 511. One of the two first recesses 511 corresponds to the multiple control terminals 45 (first control terminals 46A to 46C), and the multiple control terminals 45 (first control terminals 46A to 46C) are arranged in this one first recess 511. In a plan view, one first recess 511 overlaps all of the holders 451 for each of the multiple control terminals 45 (first control terminals 46A to 46C). The other of the two first recesses 511 corresponds to the plurality of control terminals 45 (second control terminals 47A to 47D), and the plurality of control terminals 45 (second control terminals 47A to 47D) are arranged in the other first recess 511. The other first recess 511 overlaps all of the holders 451 of the plurality of control terminals 45 (second control terminals 47A to 47D) in a plan view.

[0208] In the semiconductor device B11 of this modified example, the holders 451 constituting each control terminal 45 are disposed on the main surface 1120 of the support conductor 112 (support substrate 11). The metal pins 452 constituting each control terminal 45 protrude further toward the z1 side in the thickness direction z than the resin main surface 51. With this configuration, the multiple control terminals 45 are disposed in an area surrounded by the resin main surface 51 (sealing resin 50) in a plan view. This semiconductor device B11 can be made smaller in size in a plan view.

[0209] All of the holders 451 of the control terminals 45 are exposed from the sealing resin 50. This configuration can prevent the sealing resin 50 from flowing into the holders 451 into which the metal pins 452 are inserted. Therefore, in the semiconductor device B11, electrical continuity between the holders 451 and the metal pins 452 can be properly maintained, and the control terminals 45 configured to include the holders 451 and the metal pins 452 can function properly.

[0210] The multiple control terminals 45 are arranged in first recesses 511 of the sealing resin 50. In the semiconductor device B11, the sealing resin 50 has two first recesses 511. The multiple control terminals 45 (first control terminals 46A to 46C) are arranged in one first recess 511, and the multiple control terminals 45 (second control terminals 47A to 47D) are arranged in the other first recess 511. With this configuration in which the multiple control terminals 45 are arranged together in one first recess 511, the sealing resin 50 can be formed relatively easily by molding.

[0211] Fig. 49 shows a semiconductor device according to a second modification of the first embodiment of the second side. Fig. 49 is a cross-sectional view showing a semiconductor device B12 of this modification, and is a cross-sectional view similar to Fig. 40. The semiconductor device B12 of this modification further includes a first resin part 55, and differs from the semiconductor device B1 of the above embodiment in that it includes the first resin part 55.

[0212] The first resin portion 55 fills at least a portion of the first recess 511 and contacts at least a portion of the holder 451. In the semiconductor device B12, the first resin portion 55 fills each first recess 511 so as to fill the first recess 511. The first resin portion 55 covers the entire holder 451 placed in the first recess 511. The constituent material of the first resin portion 55 is not particularly limited. The first resin portion 55 may be the same material as the sealing resin 50 or a different material from the sealing resin 50. In the semiconductor device B12, for example, the constituent material of the first resin portion 55 is different from the constituent material of the sealing resin 50. In the semiconductor device B12, for example, the elastic modulus of the first resin portion 55 is smaller than the elastic modulus of the sealing resin 50. In this case, when the elastic modulus of the first resin portion 55 is smaller than the elastic modulus of the sealing resin 50, the constituent material of the first resin portion 55 is not particularly limited, and examples thereof include silicone resin and silicone gel.

[0213] In the semiconductor device B12 of this modified example, the holders 451 constituting each control terminal 45 are disposed on the main surface 1120 of the support conductor 112 (support substrate 11). The metal pins 452 constituting each control terminal 45 protrude toward the z1 side in the thickness direction z beyond the resin main surface 51. With this configuration, the multiple control terminals 45 are disposed in an area surrounded by the resin main surface 51 (sealing resin 50) in a plan view. Such a semiconductor device B12 can be made smaller in size in a plan view.

[0214] All of the holders 451 of the control terminals 45 are exposed from the sealing resin 50. This configuration can prevent the sealing resin 50 from flowing into the holders 451 into which the metal pins 452 are inserted. Therefore, in the semiconductor device B12, electrical continuity between the holders 451 and the metal pins 452 can be properly maintained, and the control terminals 45 configured to include the holders 451 and the metal pins 452 can function properly.

[0215] In the semiconductor device B12, each first recess 511 is filled with a first resin portion 55. The first resin portion 55 covers the holder 451 disposed in each first recess 511. The elastic modulus of the first resin portion 55 is smaller than that of the sealing resin 50. This configuration reduces stress around the holder 451 covered by the first resin portion 55. Furthermore, in the semiconductor device B12, the inclusion of the first resin portion 55 prevents foreign matter (including moisture) from entering the first recess 511 exposed from the sealing resin 50. The semiconductor device B12 configured as described above is preferable for improving durability and reliability. Additionally, the semiconductor device B12 achieves the same effects as the above embodiment within the same configuration as the semiconductor device B1 of the above embodiment.

[0216] Fig. 50 shows a semiconductor device according to a third modified example of the second aspect of the first embodiment. Fig. 50 is a cross-sectional view showing a semiconductor device B13 of this modified example, and is a cross-sectional view similar to Fig. 47. The semiconductor device B13 of this modified example further includes a first resin part 55, and differs from the semiconductor device B11 of the above modified example in that it includes the first resin part 55.

[0217] The first resin portion 55 fills at least a portion of the first recess 511 and contacts at least a portion of the holder 451. In the semiconductor device B13, the first resin portion 55 fills a portion of the first recess 511. The first resin portion 55 covers a portion of each of the multiple holders 451 arranged in the first recess 511. The constituent material of the first resin portion 55 is not particularly limited. The first resin portion 55 may be the same material as the sealing resin 50, or may be a different material from the sealing resin 50. In the semiconductor device B13, for example, the constituent material of the first resin portion 55 is different from the constituent material of the sealing resin 50. In the semiconductor device B13, for example, the elastic modulus of the first resin portion 55 is greater than the elastic modulus of the sealing resin 50. In this case where the elastic modulus of the first resin portion 55 is greater than the elastic modulus of the sealing resin 50, the constituent material of the first resin portion 55 is not particularly limited, and examples thereof include an epoxy potting material.

[0218] In the semiconductor device B13 of this modified example, the holders 451 constituting each control terminal 45 are disposed on the main surface 1120 of the support conductor 112 (support substrate 11). The metal pins 452 constituting each control terminal 45 protrude toward the z1 side in the thickness direction z beyond the resin main surface 51. With this configuration, the multiple control terminals 45 are disposed in an area surrounded by the resin main surface 51 (sealing resin 50) in a plan view. This semiconductor device B13 can be made smaller in size in a plan view.

[0219] All of the holders 451 of the control terminals 45 are exposed from the sealing resin 50. This configuration can prevent the sealing resin 50 from flowing into the holders 451 into which the metal pins 452 are inserted. Therefore, in the semiconductor device B13, electrical continuity between the holders 451 and the metal pins 452 can be properly maintained, and the control terminals 45 configured to include the holders 451 and the metal pins 452 can function properly.

[0220] In the semiconductor device B13, each first recess 511 is filled with a first resin portion 55. The first resin portion 55 covers at least a portion of the holder 451 disposed in each first recess 511. The elastic modulus of the first resin portion 55 is greater than the elastic modulus of the sealing resin 50. This configuration improves the earthquake resistance of the holder 451 covered by the first resin portion 55. The semiconductor device B13 configured as described above is preferable in terms of improving performance. In addition, the semiconductor device B13 exhibits the same effects as the semiconductor device B11 of the above-described modified example.

[0221] FIG. 51 shows a semiconductor device according to a fourth modification of the first embodiment of the second side. FIG. 51 is a cross-sectional view similar to FIG. 40 showing a semiconductor device B14 of this modification. The semiconductor device B14 of this modification further includes a first resin portion 55. Furthermore, in the semiconductor device B14, the dimension of the sealing resin 50 in the thickness direction z is smaller than that of the semiconductor device B1 of the above embodiment. Accordingly, the dimension of each first recess 511 in the thickness direction z is also smaller than that of the semiconductor device B1. The holder 451 of each control terminal 45 protrudes toward the z1 side in the thickness direction z beyond the resin main surface 51 of the sealing resin 50. The first surface 454a of the holder 451 (first flange portion 454) is located on the z1 side in the thickness direction z relative to the resin main surface 51. As a result, a portion of the holder 451 is accommodated in the first recess 511.

[0222] The first resin portion 55 fills at least a portion of the first recess 511 and contacts at least a portion of the holder 451. In the semiconductor device B14, the first resin portion 55 fills each of the first recesses 511 so as to fill the recess. The constituent material of the first resin portion 55 is not particularly limited. The first resin portion 55 may be the same material as the sealing resin 50, or may be a different material from the sealing resin 50. In the semiconductor device B14, for example, the constituent material of the first resin portion 55 is different from the constituent material of the sealing resin 50. In the semiconductor device B14, for example, the elastic modulus of the first resin portion 55 is smaller than the elastic modulus of the sealing resin 50. In this way, when the elastic modulus of the first resin portion 55 is smaller than the elastic modulus of the sealing resin 50, the constituent material of the first resin portion 55 is not particularly limited, and examples thereof include silicone resin and silicone gel.

[0223] In the semiconductor device B14, the first resin part 55 has a portion located on the z1 side in the thickness direction z of the resin main surface 51. The portion of the first resin part 55 located on the z1 side in the thickness direction z of the resin main surface 51 is a portion that protrudes along the outer peripheral surface of the holder 451 (cylindrical part 453) toward the z1 side in the thickness direction z due to, for example, the surface tension of the first resin part 55. In the illustrated example, the first surface 454a of the holder 451 (first flange part 454) is exposed from the first resin part 55.

[0224] In the semiconductor device B14 of this modified example, the holders 451 constituting each control terminal 45 are disposed on the main surface 1120 of the support conductor 112 (support substrate 11). The metal pins 452 constituting each control terminal 45 protrude further toward the z1 side in the thickness direction z than the resin main surface 51. With this configuration, the multiple control terminals 45 are disposed in an area surrounded by the resin main surface 51 (sealing resin 50) in a plan view. This semiconductor device B14 can be made smaller in size in a plan view.

[0225] All of the holders 451 of the control terminals 45 are exposed from the sealing resin 50. This configuration can prevent the sealing resin 50 from flowing into the holders 451 into which the metal pins 452 are inserted. Therefore, in the semiconductor device B14, electrical continuity between the holders 451 and the metal pins 452 can be properly maintained, and the control terminals 45 configured to include the holders 451 and the metal pins 452 can function properly.

[0226] In the semiconductor device B14, each first recess 511 is filled with a first resin portion 55. The first resin portion 55 covers the holder 451 disposed in each first recess 511. The elastic modulus of the first resin portion 55 is smaller than the elastic modulus of the sealing resin 50. This configuration reduces stress around the holder 451 covered by the first resin portion 55. Furthermore, in the semiconductor device B14, the first surface 454a of each holder 451 is exposed from the first resin portion 55. This allows the metal pin 452 to be press-fitted into the holder 451 after the first recess 511 is filled with the first resin portion 55. This stabilizes the operation of press-fitting the metal pin 452 into the holder 451. Additionally, the semiconductor device B14 achieves the same effects as the above embodiment within the same configuration as the semiconductor device B1 of the above embodiment.

[0227] 52 and 53 show a semiconductor device according to a fifth modified example of the first embodiment of the second side. Fig. 52 is a plan view showing a semiconductor device B15 of this modified example. Fig. 53 is a cross-sectional view taken along line LIII-LIII in Fig. 52. In the semiconductor device B15 of this modified example, the sealing resin 50 has a plurality of second recesses 517. The semiconductor device B15 differs from the semiconductor device B1 of the above embodiment in that the sealing resin 50 has the second recesses 517.

[0228] The second recess 517 is recessed from the resin main surface 51 toward the z2 side in the thickness direction z. In the semiconductor device B15, the sealing resin 50 has a plurality of second recesses 517. Each of the plurality of second recesses 517 is provided to correspond to one of the plurality of first recesses 511. As shown in FIG. 52 , the second recess 517 surrounds the corresponding first recess 511 in a plan view. In the illustrated example, the second recess 517 has an annular shape in a plan view.

[0229] 53 , the second recess 517 has a second recess bottom surface 518. The second recess bottom surface 518 is located at the end of the second recess 517 on the z2 side in the thickness direction z. The second recess bottom surface 518 is spaced apart from the control terminal support body 48 on the z1 side in the thickness direction z.

[0230] In the semiconductor device B15 of this modified example, the holders 451 constituting each control terminal 45 are disposed on the main surface 1120 of the support conductor 112 (support substrate 11). The metal pins 452 constituting each control terminal 45 protrude further toward the z1 side in the thickness direction z than the resin main surface 51. With this configuration, the multiple control terminals 45 are disposed in an area surrounded by the resin main surface 51 (sealing resin 50) in a plan view. Such a semiconductor device B15 can be made smaller in size in a plan view.

[0231] All of the holders 451 of the control terminals 45 are exposed from the sealing resin 50. This configuration can prevent the sealing resin 50 from flowing into the holders 451 into which the metal pins 452 are inserted. Therefore, in the semiconductor device B15, electrical continuity between the holders 451 and the metal pins 452 can be properly maintained, and the control terminals 45 configured to include the holders 451 and the metal pins 452 can function properly.

[0232] In the semiconductor device B15, the sealing resin 50 has a second recess 517. The second recess 517 surrounds the first recess 511 in a plan view. The end of the second recess 517 on the z2 side in the thickness direction z (the bottom surface 518 of the second recess 517) is spaced from the control terminal support 48 on the z1 side in the thickness direction z. This configuration increases the creepage distance along the surface of the sealing resin 50 (the resin main surface 51, the inner side surface 512 of the first recess 511, the second recess 517, etc.) between the control terminal 45 surrounded by the second recess 517 in a plan view and the adjacent control terminal 45. The semiconductor device B15 can be miniaturized in a plan view while further increasing the withstand voltage of the adjacent control terminal 45. Additionally, the semiconductor device B15 achieves the same effects as the above-described embodiment within the same configuration as the semiconductor device B1 of the above-described embodiment.

[0233] The semiconductor device according to the present disclosure is not limited to the above-described embodiment, and the specific configuration of each part of the semiconductor device according to the present disclosure can be freely modified in various ways.

[0234] In the above embodiment and each modified example, the case where the entire holder 451 for each of the plurality of control terminals 45 is exposed from the sealing resin 50 has been described, but the present disclosure is not limited to this. For example, the holder 451 for some of the plurality of control terminals 45 may be covered by the sealing resin 50.

[0235] A second aspect of the present disclosure includes the configurations described in Supplementary Notes 1B to 17B below. Supplementary Note 1B. A semiconductor device comprising: a support substrate having a main surface facing one side in a thickness direction; at least one terminal disposed on the main surface and including a conductive holder and a metal pin inserted into the holder; and a sealing resin having a resin main surface facing one side in the thickness direction and covering at least a portion of the support substrate, wherein at least one of the at least one terminal is exposed from the sealing resin with the entire holder, and the metal pin protrudes beyond the resin main surface to one side in the thickness direction. Supplementary Note 2B. The semiconductor device according to Supplementary Note 1B, further comprising a terminal support interposed between the support substrate and the at least one terminal in the thickness direction, wherein the holder is supported by the terminal support. Supplementary Note 3B. The semiconductor device according to Supplementary Note 2B, wherein the sealing resin covers a portion of the terminal support. Supplementary Note 4B. The semiconductor device according to Appendix 3B, comprising a plurality of the terminals, wherein the sealing resin has at least one first recess recessed from the resin main surface toward the other side in the thickness direction, wherein the at least one first recess has a first edge contacting the terminal support, and wherein the plurality of terminals are arranged in the at least one first recess. Appendix 5B. The semiconductor device according to Appendix 4B, comprising a plurality of the first recesses, wherein the plurality of terminals are individually arranged corresponding to the plurality of first recesses, and wherein each of the plurality of first recesses overlaps with all of the holders for the corresponding terminal when viewed in the thickness direction. Appendix 6B. The semiconductor device according to Appendix 4B, wherein the first recess overlaps with all of the holders for the corresponding terminal when viewed in the thickness direction. Appendix 7B. The semiconductor device according to Appendix 5B or 6B, further comprising a first resin portion filling at least a portion of the first recess, wherein the first resin portion contacts at least a portion of the holder. Appendix 8B. The semiconductor device according to Appendix 7B, wherein a constituent material of the first resin portion is different from a constituent material of the sealing resin, and an elastic modulus of the first resin portion is smaller than an elastic modulus of the sealing resin.Appendix 9B. The semiconductor device according to Appendix 7B, wherein the constituent material of the first resin portion is different from the constituent material of the sealing resin, and wherein the elastic modulus of the first resin portion is greater than the elastic modulus of the sealing resin. Appendix 10B. The semiconductor device according to any of Appendixes 4B to 9B, wherein the sealing resin has a second recess recessed from the resin main surface toward the other side in the thickness direction, and wherein the second recess surrounds the first recess as viewed in the thickness direction. Appendix 11B. The semiconductor device according to Appendix 10B, wherein the second recess has a second recess bottom surface located at an end on the other side in the thickness direction, and wherein the second recess bottom surface is spaced apart from the terminal support to one side in the thickness direction. Appendix 12B. The semiconductor device according to any one of Supplements 4B to 11B (or any one of Supplements 4B to 6B), wherein the at least one first recess includes a first recess inner side surface, the first recess inner side surface having the first edge located at an end on the other side in the thickness direction and a second edge located at an end on one side in the thickness direction, and the second edge surrounding the first edge when viewed in the thickness direction. Supplementary Note 13B. The semiconductor device according to Supplementary Note 12B, wherein the at least one first recess has a chamfered portion interposed between the resin main surface and the first recess inner side surface. Supplementary Note 14B. The semiconductor device according to any one of Supplements 1B to 13B (or any one of Supplements 1B to 6B), wherein the holder has a first surface located at an end on one side in the thickness direction, and the first surface is disposed on the other side in the thickness direction relative to the resin main surface. Supplementary Note 15B. The semiconductor device according to any one of Supplementary Notes 2B to 13B (or any one of Supplementary Notes 2B to 6B), further comprising at least one semiconductor element disposed on the main surface and electrically connected to the at least one terminal.Supplementary Note 16B: The semiconductor device according to Supplementary Note 15B, wherein the at least one terminal is a control terminal for controlling the at least one semiconductor element.Appendix 17B. The semiconductor device according to Appendix 16B, wherein the support substrate includes a first conductive portion and a second conductive portion spaced apart in a first direction orthogonal to the thickness direction, the at least one semiconductor element includes a first switching element bonded to the first conductive portion and a second switching element bonded to the second conductive portion, the control terminal includes a first control terminal for controlling the first switching element and a second control terminal for controlling the second switching element, and the terminal support includes a first support portion supporting the first control terminal and a second support portion supporting the second control terminal.

[0236] (Explanation of symbols related to first side surface) A1, A11, A12, A13, A14, A15, A2: semiconductor device 10A: first semiconductor element 10B: second semiconductor element 101: element main surface 102: element back surface 11: first main surface electrode 12: second main surface electrode 121: gate finger 13: third main surface electrode 15: back surface electrode 17: thermistor 19: conductive bonding material 3: supporting substrate 301: supporting surface 302: bottom surface 31: insulating layer 32: supporting conductor 32A: first conductive portion 32B: second conductive portion 321: first bonding layer 33: back surface metal layer 41: first terminal 42: second terminal 43: third terminal 44: fourth terminal 45: control terminal 451: holder 452: metal pin 453: Cylindrical portion 453a: First outer surface 453b: First inner surface 454: First flange portion 454a: First surface 454b: Second surface 455: Second flange portion 459: Conductive bonding material 46A, 46B, 46C, 46D, 46E: First control terminal 47A, 47B, 47C, 47D: Second control terminal 48: Control terminal support (terminal support) 48A: First support portion 48B: Second support portion 481: Insulating layer 482: First metal layer 482A: First portion 482B: Second portion 482C: Third portion 482D: Fourth portion 482E: Fifth portion 482F: Sixth portion 483: Second metal layer 49: Bonding material 5: First conductive member 51: Main portion 51 514: First opening 52: First bonding portion 53: Second bonding portion 59: Conductive bonding material 6: Second conductive member 602: First step portion 603: Second step portion 61: Third bonding portion 611: Flat portion 612: First inclined portion 64: First path portion 641: First band-shaped portion 643: First extending portion 649: Recess 65: Second path portion 651: Second band-shaped portion 653: Second extending portion 659: Recess 66: Third path portion 669: Recess 67: Fourth path portion 69: Conductive bonding material 71, 72, 73, 74: Wire 8: Sealing resin 81: Resin main surface 810: First recess 811: Recess inner side surface 812: Recess bottom surface 813: Recess edge 814: Cylindrical inner surface 815: Tapered inner surface 82: Resin back surface 831,832: Resin side surface 832a: Recess 833, 834: Resin side surface 851: Protrusion 851a: Protrusion end surface 851b: Recess 851c: Inner wall surface 852: First protrusion 852a: Protrusion top surface 89: First resin filling portion L1: First dimension L2: Second dimension (Explanation of symbols related to second side surface) B1, B11, B12, B13, B14, B15: Semiconductor device 11: Support substrate 111: Insulating layer 112: Support conductor 1120: Main surface 1121: First conductive portion 1122: Second conductive portion 113: Back metal layer 13: Power terminal 14: First power terminal 15: Second power terminal 16: Third power terminal 21: Semiconductor element 21A: First element (first switching element) 21B: Second element (second switching element) 211: First electrode 212: Second electrode 213: Third electrode 214: Fourth electrode 22: Thermistor 23: Conductive bonding layer 31: First conductive member 310: Through hole 311: Main body portion 312: First bonding portion 313: Second bonding portion 32: Second conductive member 321: Main body portion 322: Third bonding portion 324: Fourth bonding portion 326: Intermediate portion 327: Horizontal beam portion 328: Hanging portion 33: First conductive bonding layer 34: Second conductive bonding layer 35: Third conductive bonding layer 36: Fourth conductive bonding layer 41: First wire 42: Second wire 43: Third wire 44: Fourth wire 45: Control terminal (terminal) 451: Holder 452: Metal pin 453: Cylindrical portion 454: First flange portion 454a: First surface 455: Second flange portion 459: Conductive bonding layer 46A, 46B, 46C: First control terminal 47A, 47B, 47C,47D: Second control terminal 48: Control terminal support (terminal support) 48A: First support portion 48B: Second support portion 481: Insulating layer 482: First metal layer 482A: First portion 482B: Second portion 482C: Third portion 482D: Fourth portion 482E: Fifth portion 483: Second metal layer 49: Bonding layer 50: Sealing resin 51: Main resin surface 511: First recess 512: Inner side surface of first recess 513: First edge 514: Second edge 515: Chamfered portion 517: Second recess 518: Bottom surface of second recess 52: Back surface of resin 531, 532, 533, 534: Side surface of resin 531a: Recess 55: First resin portion 91: Mold 911: Cylindrical pin 915: rounded corners 919: cavity space

Claims

1. a conductive cylindrical holder and at least one terminal including a metal pin inserted into the holder; A terminal support body that supports the holder; a sealing resin that covers a part of the holder and the terminal support, the sealing resin has a resin main surface facing one side in a thickness direction, The holder has a first surface located at one end in the thickness direction and a first outer surface extending in the thickness direction, the first surface is at a position different from the resin main surface in the thickness direction, the first outer surface is in contact with the sealing resin, The metal pin protrudes beyond the resin main surface to one side in the thickness direction.

2. the holder includes a cylindrical portion extending in the thickness direction and a first flange portion connected to one end of the cylindrical portion in the thickness direction, the first flange portion has the first surface facing one side in the thickness direction, and a second surface located on the other side in the thickness direction relative to the first surface and facing the other side in the thickness direction, The cylindrical portion has the first outer surface, The semiconductor device according to claim 1 , wherein the first outer surface and the second surface are entirely in contact with the sealing resin.

3. the sealing resin has a first recess recessed from the resin main surface to the other side in the thickness direction, the first flange portion is located on the other side in the thickness direction with respect to the resin main surface, The semiconductor device according to claim 2 , wherein the first recess overlaps the entire cylindrical portion when viewed in the thickness direction.

4. The semiconductor device according to claim 3 , wherein at least a portion of the first surface is exposed from the sealing resin.

5. the entire first surface is exposed from the sealing resin, the first recess has a recess inner side surface connected to the resin main surface, and a recess bottom surface connected to an end of the recess inner side surface on the other side in the thickness direction and facing one side in the thickness direction, The semiconductor device according to claim 4 , wherein the bottom surface of said recess surrounds said first surface when viewed in the thickness direction.

6. The semiconductor device according to claim 3 , wherein the first recess is located on the other side in the thickness direction and has a recess edge in contact with the first surface.

7. The first recess has a tapered inner surface that is connected to an edge of the recess, 7. The semiconductor device according to claim 6, wherein said tapered inner side surface is inclined so that an inner diameter dimension thereof increases toward one side in said thickness direction.

8. The semiconductor device according to claim 6 , wherein an outer periphery of said first flange portion surrounds said first recess portion when viewed in said thickness direction.

9. 9. The semiconductor device according to claim 3, wherein a first dimension, which is a distance in the thickness direction between the resin main surface and the first surface, is smaller than a second dimension, which is a length in the thickness direction of the holder.

10. The semiconductor device according to claim 9 , wherein a ratio of the first dimension to the second dimension is equal to or greater than 1 / 3.

11. 9. The semiconductor device according to claim 3, further comprising a first resin filling portion filled in said first recess.

12. Further comprising at least one semiconductor element electrically connected to the at least one terminal, The semiconductor device according to claim 1 , wherein the at least one semiconductor element is supported by a supporting conductor.

13. The semiconductor device according to claim 12 , wherein the at least one terminal is a control terminal for controlling the at least one semiconductor element.

14. the supporting conductor includes a first conductive portion and a second conductive portion spaced apart from each other in a first direction perpendicular to the thickness direction, the at least one semiconductor element includes a first switching element joined to the first conductive portion and a second switching element joined to the second conductive portion; 14. The semiconductor device according to claim 13, wherein the control terminal includes a first control terminal for controlling the first switching element, and a second control terminal for controlling the second switching element.

15. A conductive ring-shaped holder portion, and at least one terminal including a metal pin inserted into the holder portion; a sealing resin that covers a portion of the holder portion, the sealing resin has a resin main surface facing one side in a thickness direction and a first recess recessed from the resin main surface to the other side in the thickness direction, the holder portion has a first surface located at one end in the thickness direction, the first surface is at a position different from the resin main surface in the thickness direction, The holder portion includes a ring-shaped portion extending in the thickness direction, the ring-shaped portion has the first surface facing one side in the thickness direction, The metal pin protrudes beyond the resin main surface to one side in the thickness direction.

16. A semiconductor device as described in claim 15, wherein at least a portion of the first surface is exposed from the sealing resin.

17. The entire first surface is exposed from the sealing resin, the first recess has a recess inner side surface connected to the resin main surface, and a recess bottom surface connected to an end of the recess inner side surface on the other side in the thickness direction and facing one side in the thickness direction, The semiconductor device according to claim 16 , wherein the bottom surface of the recess surrounds the first surface when viewed in the thickness direction.

18. A semiconductor device as described in claim 15, wherein the first recess is located on the other side of the thickness direction and has a recess edge that contacts the first surface.

19. The first recess has a tapered inner surface that connects to an edge of the recess, 20. The semiconductor device according to claim 18, wherein the tapered inner side surface is inclined so that an inner diameter dimension thereof increases toward one side in the thickness direction.