JP5702986B2 - Resin-sealed semiconductor device - Google Patents

Description

The present invention relates to a resin-sealed semiconductor device having a heat sink.

  Various resin-encapsulated semiconductor devices having a heat sink for dissipating heat generated in a semiconductor chip have been conventionally proposed (see, for example, Patent Document 1).

  FIG. 12 is a view for explaining the resin-encapsulated semiconductor device 900 described in Patent Document 1. In FIG. A resin-encapsulated semiconductor device 900 described in Patent Document 1 (hereinafter referred to as a conventional resin-encapsulated semiconductor device 900) includes a heat sink 910 made of a metal plate and a heat sink 910 as shown in FIG. The semiconductor chip 920 disposed on the upper surface side, the ceramic substrate 930 disposed between the heat sink 910 and the semiconductor chip 920 and having high thermal conductivity and insulation, and the upper surface side of the ceramic substrate 930 are formed. An external connection lead 940 electrically connected to the lower surface side electrode 922 of the semiconductor chip 920 via the wiring pattern 932 and a wiring pattern 924 formed on the upper surface side of the semiconductor chip 920 via a connector 960 made of a metal plate. Electrically connected external connection lead 950, heat sink 910, semiconductor chip 920, ceramic substrate 930, and a mold resin 970 for fixing together the external connection leads 940 and 950 and connector 960.

  According to the conventional resin-encapsulated semiconductor device 900, the ceramic substrate 930 having high thermal conductivity and insulation is disposed between the heat sink 910 and the semiconductor chip 920, and the lower surface side electrode of the semiconductor chip 920 is ceramic. The wiring pattern 932 formed on the upper surface side of the substrate 930 is connected to the external connection lead 940, and the upper surface side electrode of the semiconductor chip 920 is connected to the external connection lead 950 via a connector 960 made of a metal plate. Therefore, it is possible to realize a resin encapsulated semiconductor device for high current that has high heat dissipation and is in demand for various uses such as in-vehicle use.

JP 2009-238804 A

  However, in recent years, in order to be able to cope with higher performance and higher functionality of electronic devices and the like, even higher performance and higher functionality are required in resin-encapsulated semiconductor devices. In particular, when a resin-encapsulated semiconductor device is used for in-vehicle use, a resin-encapsulated semiconductor device capable of flowing a larger current is required.

  In this type of resin-encapsulated semiconductor device, it is also important to take measures to prevent damage to the semiconductor chip when some force is applied to the semiconductor chip. In particular, when a large amount of current is passed through the semiconductor chip, the amount of heat generation also increases. Therefore, a force such as “twist” is applied to the semiconductor chip due to thermal contraction due to a difference in thermal expansion coefficient between the semiconductor chip and the connector. There is a risk of joining. Further, when the resin-encapsulated semiconductor device is used for in-vehicle use, a force such as vibration or impact may be applied to the semiconductor chip. For this reason, it is also important to take measures against the force applied to these semiconductors. In the present specification, the force applied to the semiconductor chip by thermal contraction, the force applied to the semiconductor chip by vibration / impact, etc. are collectively referred to as “stress”.

  An object of the present invention is to provide a resin-encapsulated semiconductor device capable of flowing a larger current than a conventional resin-encapsulated semiconductor device, and to relieve stress applied to a semiconductor chip. It is an object to provide an elastic connector.

  [1] The resin-encapsulated semiconductor device of the present invention relieves stress applied to the semiconductor chip, a heat sink having conductivity and thermal conductivity, a semiconductor chip disposed on the upper surface side of the heat sink, and the semiconductor chip. An elastic connector disposed between the heat sink and the semiconductor chip, an external connection lead electrically connected to the upper surface side electrode of the semiconductor chip, and a lower surface of the heat sink And a mold resin that seals the heat sink, the semiconductor chip, the elastic connector, and the external connection lead in a state where the side and the tip end side of the external connection lead are respectively exposed. The heat sink is electrically connected to the lower surface side electrode of the semiconductor chip through the elastic connector.

  According to the resin-encapsulated semiconductor device of the present invention, the lower surface side electrode of the semiconductor chip is connected to the heat sink via the elastic connector. As described above, in the resin-encapsulated semiconductor device of the present invention, the heat sink having a large area on the surface orthogonal to the current flow direction also serves as the external connection lead of the lower surface side electrode of the semiconductor chip. This resin-encapsulated semiconductor device can pass a larger current than the above-described conventional resin-encapsulated semiconductor device.

  Further, according to the resin-encapsulated semiconductor device of the present invention, the semiconductor chip is connected to the heat sink through the elastic connector in a wide area, so that the heat generated in the semiconductor chip is efficiently transferred from the elastic connector to the heat sink. The heat can be transmitted well, and the heat generated in the semiconductor chip can be efficiently radiated from the heat sink to the outside.

  Further, according to the resin-encapsulated semiconductor device of the present invention, since the elastic connector having the stress relaxation structure is disposed between the heat sink and the semiconductor chip, the stress applied to the semiconductor chip can be relaxed. Further, there is an effect that damage to the semiconductor chip due to stress applied to the semiconductor chip can be prevented.

  [2] In the resin-encapsulated semiconductor device of the present invention, the stress relaxation structure is formed at a predetermined position on the one end side between the one end and the other end of the flat metal plate. A first folded portion, a second folded portion is set at a predetermined position on the other end side, a space between the one end portion and the first folded portion is a first flat plate portion, When the second flat plate portion is between the other end and the second folded portion, and the third flat plate portion is between the first folded portion and the second folded portion, the first flat plate The first flat plate portion is folded by rotating the first flat plate portion in the first rotation direction so that the portion has a predetermined interval with respect to the third flat plate portion and is parallel, and The second flat plate portion has a predetermined interval and is parallel to the third flat plate portion. It is preferably formed by folding the from the first rotational direction in the second flat plate portion and the second folded portion so as to rotate in a second rotational direction opposite.

  When the elastic connector has the stress relaxation structure formed as described above, the elastic connector becomes excellent in elasticity, and thereby, stress applied to the semiconductor chip can be relaxed.

  [3] In the resin-encapsulated semiconductor device of the present invention, the stress relaxation structure is formed at a predetermined position on the one end side between the one end and the other end of the flat metal plate. A first folded portion, a second folded portion is set at a predetermined position on the other end side, a space between the one end portion and the first folded portion is a first flat plate portion, When the second flat plate portion is formed between the other end portion and the second folded portion, and the third flat plate portion is formed between the first folded portion and the second folded portion, the one end The first flat plate portion is folded by rotating in the first rotation direction at the first folding portion so that the portion contacts or approaches the third flat plate portion, and the other end portion is the first flat portion. 3 The second flat plate portion is the second folded portion so as to come into contact with or close to the three flat plate portions. Serial or may be formed by folding so as to rotate in a second rotational direction opposite to the first rotational direction.

  Even when the elastic connector has the stress relaxation structure formed as described above, the elastic connector is excellent in elasticity, and thereby the stress applied to the semiconductor chip can be reduced.

  As such a stress relaxation structure, the first flat plate portion side will be described as an example. Only a part of the first flat plate portion has a flat plate portion parallel to the third flat plate portion, and An example can be illustrated in which the portion from the flat plate portion to the tip portion of the first flat plate portion (one end portion of the flat metal plate) is an inclined portion toward the third flat plate portion. Further, the first flat plate portion may be formed so as to be curved. Similarly, the stress relaxation structure can be formed on the second flat plate portion side.

  [4] In the resin-encapsulated semiconductor device of the present invention, the first folded portion and the second folded portion are formed by folding the first flat plate portion with the first folded portion and the second flat plate portion. Positions of the first folded portion and the second folded portion so that the one end portion and the other end portion are separated from each other when the second folded portion is folded. Is preferably set.

  By setting the first folded portion and the second folded portion to such positions and folding the first flat plate portion and the second flat plate portion, the elastic connector can be connected to one end portion of the flat metal plate and the other. Since it will be in the state which does not overlap with the edge part of this, it will be excellent in elasticity.

  [5] In the resin-encapsulated semiconductor device of the present invention, the stress relaxation structure is arranged at a predetermined position on the one end side between the one end and the other end of the flat metal plate. A first folded portion, a second folded portion is set at a predetermined position on the other end side, a space between the one end portion and the first folded portion is a first flat plate portion, When the second flat plate portion is between the other end and the second folded portion, and the third flat plate portion is between the first folded portion and the second folded portion, the third flat plate portion is used. The third flat plate portion is folded back so as to rotate in the first rotation direction at the first folding portion so that the portion forms an acute angle with the first flat plate portion, and the second flat plate portion is The second flat plate portion is made to be parallel to the first flat plate portion and the second folded portion The may be one which is formed by folding so as to rotate in a second rotational direction opposite to the first rotational direction.

  By forming the stress relaxation structure in this way, the elastic connector has a Z-shape. Also by adopting such a shape, the elastic connector is excellent in elasticity, so that the stress applied to the semiconductor chip can be relaxed.

  [6] In the resin-encapsulated semiconductor device of the present invention, when the length between the one end and the other end is L, the first folded portion is the one terminal. It is preferable that the second folded portion is set at a position L / 3 or substantially L / 3 from the other terminal and the other terminal.

  By setting the first folded portion and the second folded portion at such a position and folding the first flat plate portion and the second flat plate portion, the elastic connector can be made in a shape having excellent stability, And it can be excellent in elasticity.

  [7] In the resin-encapsulated semiconductor device of the present invention, the stress relaxation structure sets one folded portion at a predetermined position between one end and the other end of the flat metal plate. When the first flat plate portion is between the one end portion and the folded portion, and the second flat plate portion is between the other end portion and the folded portion, the second flat plate portion is the first flat portion. The second flat plate portion is arranged in the first rotation direction or the second rotation direction opposite to the first rotation direction at the folded portion so as to be parallel to the flat plate portion and parallel to the flat plate portion. The stress relieving structure is formed by turning back so as to rotate, and one turn-up portion is set at a predetermined position between one end and the other end of the flat metal plate, A space between the one end and the folded portion is a first flat plate portion, and the other end is The second flat plate portion is folded back so that the second flat plate portion has a predetermined interval with respect to the first flat plate portion and is parallel to the second flat plate portion. It may be formed by folding back so as to rotate in the first rotation direction or the second rotation direction opposite to the first rotation direction.

  By forming the stress relaxation structure in this way, the elastic connector becomes a shape obtained by rotating the U shape by 90 degrees. Since the elastic connector having such a shape is also excellent in elasticity, the stress applied to the semiconductor chip can be relaxed.

  [8] In the resin-encapsulated semiconductor device of the present invention, when the length between the one end portion and the other end portion is L, the folded portion is the one terminal or the other end. It is preferable to set at a position of L / 2 or almost L / 2 from the terminal.

  By setting the folded portion to such a position and folding the second flat plate portion, the elastic connector can have a shape with excellent stability and excellent elasticity.

  [9] In the resin-encapsulated semiconductor device of the present invention, the folded portion is preferably folded so that a rounded portion is formed at least on the outer peripheral surface side of the folded portion.

  As described above, when the elastic connector is soldered to the semiconductor chip and the heat sink by being folded so that the rounded portion is formed at least on the outer circumferential surface side of the folded portion, the solder is rounded on the outer circumferential surface side. By entering the portion, a good solder fillet is formed between the semiconductor chip and the heat sink, and the bonding state between the elastic connector, the semiconductor chip and the heat sink becomes strong.

  [10] In the resin-encapsulated semiconductor device of the present invention, it is preferable that the elastic connector has a chamfered portion formed on an edge of a surface in contact with the semiconductor chip and the heat sink.

  As described above, the chamfered portion is formed at the edge of the surface of the elastic connector that is in contact with the semiconductor chip and the heat sink, so that when the elastic connector is soldered to the semiconductor chip and the heat sink, soldering is performed. As a result, the solder fillet is formed between the semiconductor chip and the heat sink, and the bonding state between the elastic connector, the semiconductor chip and the heat sink becomes strong. .

  [11] In the resin-encapsulated semiconductor device of the present invention, the semiconductor chip is a plurality of semiconductor chips arranged to be connected in parallel, and the elastic connector is the plurality of semiconductor chips. It is preferable that it is disposed between the heat sink and the plurality of semiconductor chips so as to straddle.

  When a plurality of semiconductor chips are connected in parallel as described above, it is not necessary to provide an elastic connector for each semiconductor chip by providing one elastic connector so as to straddle the plurality of semiconductor chips. For this reason, the number of parts (in this case, the number of elastic connectors) can be reduced, and the advantage that the number of assembly steps can be reduced is also obtained. In addition, by using a plurality of semiconductor chips such as diodes connected in parallel, when a current of the same magnitude is passed, each semiconductor chip can be downsized compared to the case of using one semiconductor chip, When individual semiconductor chips are downsized, it is advantageous in that chip cracks are less likely to occur.

  [12] In the resin-encapsulated semiconductor device of the present invention, it is preferable that the elastic connector has the stress relaxation structure corresponding to each semiconductor chip of the plurality of semiconductor chips.

  By setting it as such a structure, the stress added to each semiconductor chip can be relieved for every semiconductor chip.

  [13] In the resin-encapsulated semiconductor device of the present invention, between the stress relaxation structures corresponding to each of the semiconductor chips, a stress interference suppression unit that suppresses the interference of the stress between the semiconductor chips. Is preferably provided.

  When the elastic connector is arranged so as to straddle a plurality of semiconductor chips, when stress is applied to each semiconductor chip, the stress interferes with each other, and there is a possibility that `` twist '' or the like occurs in the semiconductor chip. By providing the stress interference suppression unit that suppresses stress interference, it is possible to suppress the stress generated between the semiconductor chips from interfering with each other.

  [14] The elastic connector according to the present invention includes a heat sink having conductivity and thermal conductivity, a plurality of semiconductor chips arranged to be connected in parallel to the upper surface side of the heat sink, and the plurality of semiconductor chips. An elastic connector disposed between the heat sink and the semiconductor chip so as to straddle, an external connection lead electrically connected to an upper surface side electrode of the plurality of semiconductor chips, the heat sink, and the semiconductor chip The elastic connector used in a resin-encapsulated semiconductor device including a mold resin that seals the elastic connector and the external connection lead, and relieves stress applied to each semiconductor chip of the plurality of semiconductor chips. And a stress relaxation structure is provided corresponding to each of the semiconductor chips of the plurality of semiconductor chips. .

  According to the elastic connector of the present invention, the stress applied to the semiconductor chip can be relaxed. The elastic connector of the present invention is disposed between the heat sink and the semiconductor chip so as to straddle a plurality of semiconductor chips, and a stress relaxation structure is provided corresponding to each semiconductor chip. Therefore, in the case of using a plurality of semiconductor chips connected in parallel, it is not necessary to provide an elastic connector for each semiconductor chip, the number of parts can be reduced, and the number of assembly steps can be reduced, And the stress added to each semiconductor chip can be relieved for every semiconductor chip.

  [15] In the elastic connector according to the present invention, between the stress relaxation structures corresponding to the respective semiconductor chips, a stress interference suppressing portion that suppresses the interference of the stress between the respective semiconductor chips is provided. It is preferable.

  By adopting such a configuration, it is possible to suppress the stress generated between the semiconductor chips from interfering with each other.

1 is an external perspective view of a resin-encapsulated semiconductor device 10 according to Embodiment 1. FIG. 1 is a perspective view showing a configuration of a frame unit 1000 in a resin-encapsulated semiconductor device 10 according to Embodiment 1. FIG. FIG. 2 is a cross-sectional view of the resin-encapsulated semiconductor device 10 shown in FIG. It is a figure shown in order to demonstrate the elastic connector 1200,1300. It is an enlarged view at the time of seeing the part of the broken-line frame A in FIG. 2 from the side surface. It is a figure shown in order to demonstrate the 1st modification of the elastic connector 1200. FIG. It is a figure shown in order to demonstrate the 2nd modification of the elastic connector 1200. FIG. It is a figure shown in order to demonstrate the 3rd modification of the elastic connector 1200. FIG. It is a figure shown in order to demonstrate the resin sealing type semiconductor device 20 which concerns on Embodiment 2. FIG. It is a figure shown in order to demonstrate the elastic connector 1800. FIG. It is a figure shown in order to demonstrate the modification of the elastic connector 1800. FIG. It is a figure shown in order to demonstrate the resin sealing type semiconductor device 900 described in patent document 1. FIG.

  Hereinafter, embodiments of the present invention will be described.

[Embodiment 1]
FIG. 1 is an external perspective view of a resin-encapsulated semiconductor device 10 according to the first embodiment. FIG. 2 is a perspective view showing the configuration of the frame portion 1000 in the resin-encapsulated semiconductor device 10 according to the first embodiment. FIG. 3 is a cross-sectional view of the resin-encapsulated semiconductor device 10 shown in FIG. 3A is a cross-sectional view taken along the line aa in FIG. 1, and FIG. 3B is a cross-sectional view taken along the line bb in FIG.

  As shown in FIGS. 1 to 3, the resin-encapsulated semiconductor device 10 according to the first embodiment includes a frame unit 1000, a first semiconductor chip (referred to as a diode) 310 attached to the frame unit 1000, and a first unit. 2 semiconductor chip (referred to as a diode) 320 and a mold resin 400 for sealing the frame portion 1000, the first semiconductor chip 310 and the second semiconductor chip 320. Note that nut portions 411 to 414 for screwing are provided on the upper surface side of the mold resin 400.

  The frame unit 1000 includes a heat sink 1100 made of a metal plate having excellent conductivity and thermal conductivity, a first elastic connector 1200 made of a metal plate having excellent conductivity and thermal conductivity, and the same conductivity and thermal conductivity. The second elastic connector 1300 made of a metal plate excellent in resistance, a pair of terminal fittings 1410 and 1420 made of a metal member excellent in conductivity, and a pair of terminal fittings 1510 and 1520 made of a metal member also excellent in conductivity. And have. The heat sink 1100 is in a state where the lower surface side is exposed from the mold resin 400 in a state where the mold resin 400 is formed.

  The first elastic connector 1200 is disposed between the heat sink 1100 and the first semiconductor chip 310 disposed on the upper surface side of the heat sink 1100, and the lower surface side electrode (the anode side electrode and the first semiconductor chip 310). It is electrically connected to 311. As a result, the anode side electrode 311 of the first semiconductor chip 310 and the heat sink 1100 are electrically connected, and the heat sink 1100 also serves as an external connection lead for the anode side electrode in the first semiconductor chip 310. .

  The second elastic connector 1300 is disposed between the heat sink 1100 and the second semiconductor chip 320 disposed on the upper surface side of the heat sink 1100, and the lower surface side electrode (anode side) of the second semiconductor chip 320. It is electrically connected to 322. As a result, the anode side electrode 321 of the second semiconductor chip 320 and the heat sink 1100 are electrically connected, and the heat sink 1100 also serves as an external connection lead of the anode side electrode in the second semiconductor chip 320. .

  Incidentally, in the resin-encapsulated semiconductor device 10 according to the first embodiment, the first semiconductor chip 310 and the second semiconductor chip 320 are illustrated as being used in parallel connection. By using two semiconductor chips connected in parallel, each semiconductor chip can be reduced in size compared to the case where one semiconductor chip is used when the same current flows. Thus, if each semiconductor chip is miniaturized, it is advantageous in that chip cracks are less likely to occur.

  In the case where the first elastic connector 1200 and the second elastic connector 1300 are collectively described, they may be referred to as elastic connectors 1200 and 1300. Similarly, when the first semiconductor chip 310 and the second semiconductor chip 320 are collectively described, they may be referred to as semiconductor chips 310 and 320.

  The elastic connectors 1200 and 1300 have a stress relaxation structure for relaxing stress applied to the semiconductor chips 310 and 320. This stress relaxation structure is formed by folding a flat metal plate, and details thereof will be described later.

  The terminal fitting 1410 of the pair of terminal fittings 1410 and 1420 has a function as an attachment fitting when the resin-encapsulated semiconductor device 10 is attached to a substrate or the like, and the upper surface side electrode of the first semiconductor chip 310. It is electrically connected to a 312 (referred to as a cathode side electrode) (see FIG. 3A). Therefore, the terminal fitting 1410 also has a function as an external connection lead of the cathode side electrode in the first semiconductor chip 310. The terminal fitting 1420 of the pair of terminal fittings 1410 and 1420 is a terminal fitting that is not electrically connected to the first semiconductor chip 310, and is used when the resin-encapsulated semiconductor device 10 is attached to a substrate or the like. It functions as a mounting bracket.

  In a state where the mold resin 400 is formed, the pair of terminal fittings 1410 and 1420 is in a state where the portion from the midway part to the tip part is exposed from the mold resin 400. Screw through holes 1411 and 1421 are provided on the tip side.

  The terminal fitting 1510 of the pair of terminal fittings 1510 and 1520 has a function as an attachment fitting when the resin-encapsulated semiconductor device 10 is attached to a substrate or the like, and the upper surface side electrode of the second semiconductor chip 320. (Cathode side electrode) 322 is electrically connected. For this reason, the terminal fitting 1510 also has a function as an external connection lead of the cathode side electrode in the second semiconductor chip 320. The terminal fitting 1520 of the pair of terminal fittings 1510 and 1520 is a terminal fitting that is not electrically connected to the second semiconductor chip 320, and is used when the resin-encapsulated semiconductor device 10 is attached to a substrate or the like. It functions as a mounting bracket.

  Similarly to the terminal fittings 1410 and 1420, these terminal fittings 1510 and 1520 are exposed from the mold resin 400 between the middle part and the tip part when the mold resin 400 is formed. Screw holes 1511 and 1521 are provided on the distal end side of the terminal fittings 1510 and 1520.

  The pair of terminal fittings 1410 and 1420 and the pair of terminal fittings 1510 and 1520 are in a state of extending in opposite directions until the molding resin 400 is formed by the resin sealing process. . Then, after the mold resin 400 is formed by sealing the resin, the tip portions of the pair of terminal fittings 1510 and 1520 are bent and the tip portions of the pair of terminal fittings 1410 and 1420 are opposed to each other. Are bent so as to face each other (see FIGS. 1 to 3).

  In this manner, the screw holes 1411 of the terminal fitting 1410 are bent by the nut portions of the mold resin 400 by bending the tip ends of the pair of terminal fittings 1410 and 1420 and the pair of terminal fittings 1510 and 1520 to face each other. The screw hole 1421 of the terminal fitting 1420 faces the nut portion 412 of the mold resin 400, and the screw penetration hole 1511 of the terminal fitting 1510 faces the nut portion 413 of the mold resin 400, and the terminal fitting 1520. The screw through hole 1521 faces the nut portion 414 of the mold resin 400.

  FIG. 4 is a view for explaining the elastic connectors 1200 and 1300. In this case, since the elastic connectors 1200 and 1300 have the same shape, only the first elastic connector 1200 will be described here. 4A is a perspective view showing a flat metal plate 1210 (hereinafter referred to as a flat metal plate 1210) used for the elastic connector 1200, and FIG. 4B is a perspective view of the elastic connector 1200. FIG. FIG. 4C is a view of FIG. 4B viewed from the direction of the arrow y. The elastic connector 1200 shown in FIGS. 4B and 4C has a stress relaxation structure formed by folding a flat metal plate 1210.

  The stress relaxation structure of the elastic connector 1200 can be formed by the following procedure. First, a flat metal plate 1210 as shown in FIG. As shown in FIG. 4A, the flat metal plate 1210 is at least one of the upper surface side and the lower surface side of the flat metal plate 1210 (the upper surface side in FIG. 4A). A chamfered portion C is formed at the edge.

  In such a flat metal plate 1210, a first folded portion P1 (hereinafter referred to as a first folded portion P1) is formed at a predetermined position on one end t1 side between the one end t1 and the other end t2 of the flat metal plate 1210. Then, the first folded portion P1) is set, and the second folded portion P2 (hereinafter referred to as the second folded portion P2) is set at a predetermined position on the other end t2. Here, a portion between one end t1 and the first folded portion P1 is a first flat plate portion 1211, and a portion between the other end t2 and the second folded portion P2 is a second flat plate portion 1212. A portion between the portion P1 and the second folded portion P2 will be described as a third flat plate portion 1213.

  Then, the first flat plate portion 1211 is arranged in the first rotation direction (first rotation direction P1) so that the first flat plate portion 1211 is parallel to the third flat plate portion 1213 and is parallel to the third flat plate portion 1213. In this case, it is turned counterclockwise.) And the second flat plate portion 1212 has a predetermined distance d with respect to the third flat plate portion 1213 and is parallel to the third flat plate portion 1213. The part 1212 is turned back so as to rotate in a second rotation direction (in this case, clockwise) opposite to the first rotation direction at the second turn-back part P2 (FIGS. 4B and 4C). reference.). When the flat metal plate 1210 is folded at the first folded portion P1 and the second folded portion P2, a rounded portion is formed on at least the outer peripheral surface side of the first folded portion P1 and the second folded portion P2. Wrap like so.

  When the first flat plate portion 1211 is folded at the first folded portion P1 and the second flat plate portion 1212 is folded at the second folded portion P2 (the states shown in FIGS. 4B and 4C). It is preferable to set the position of the first folded portion P1 and the position of the second folded portion P2 so that one end t1 and the other end t2 of the flat metal plate 1210 are separated from each other. In the example shown in FIG. 4, the length from one end t1 to the other end t2 is L, the length from one end t1 to the first folded portion P1 is L1, and the other end When the length from the portion t2 to the second folded portion P2 is L2, L1 and L2 are set to be approximately L / 5.

By setting the first folded portion P1 and the second folded portion P2 to such positions, one end t1 and the other end t2 of the flat metal plate 1210 are opposed to each other in a separated state. Therefore, the length of the metal plate for forming the elastic connector (the length from one end t1 to the other end t2) can be obtained. ) Can be shortened, and the weight can be reduced.
By folding the flat metal plate 1210 in such a procedure, an elastic connector 1200 as shown in FIGS. 4B and 4C can be created.

  In the resin-encapsulated semiconductor device according to the first embodiment, the rotation direction when the flat metal plate 1210 is folded is the first rotation direction counterclockwise and the second rotation direction clockwise. Although the case has been illustrated, the reverse, that is, the first rotation direction may be clockwise and the second rotation direction may be counterclockwise. As described above, when the first rotation direction is the clockwise direction and the second rotation direction is the counterclockwise direction, the chamfered portion C is formed at the edge on the lower surface side of the flat metal plate 1210. The chamfered portion C may be provided on both the upper surface side and the lower surface side of the flat metal plate 1210.

  The first elastic connector 1200 has been described above, but the second elastic connector 1300 can also be created in the same procedure and has the same stress relaxation structure.

  The elastic connectors 1200 and 1300 having such a stress relaxation structure are disposed between the semiconductor chips 310 and 320 and the heat sink 1100, and the lower surface side electrodes (anode side electrodes) of the semiconductor chips 310 and 320 and the heat sink 1100 are arranged. It is fixed by soldering to the upper surface side. In forming the mold resin 400 (see FIG. 1) by sealing the resin, a space (in FIG. 4C) formed between the elastic connectors 1200 and 1300 and the heat sink 1100 is formed. The space (shown in gray) is filled with resin.

  According to the resin-encapsulated semiconductor device 10 according to the first embodiment configured as described above, the semiconductor chips 310 and 320 are connected to the heat sink 1100 via the elastic connectors 1200 and 1300. A larger current can be passed through each of the semiconductor chips 310 and 320 than in the resin-encapsulated semiconductor device.

  That is, in the resin-encapsulated semiconductor device 10 according to the first embodiment, the heat sink 1100 having a large area on the surface orthogonal to the current flow direction is the external connection lead of the lower surface side electrodes (anode side electrodes) of the semiconductor chips 310 and 320. It also plays a role. In the resin-encapsulated semiconductor device 10 according to the first embodiment, as can be seen from FIG. 3, the length of the current path between the heat sink 1100 and the lower surface side electrodes (anode side electrodes) of the semiconductor chips 310 and 320. Therefore, the resistance component can be reduced. From these points, the resin-encapsulated semiconductor device 10 according to the first embodiment can pass a larger current than the conventional resin-encapsulated semiconductor device (see FIG. 12).

  Further, since the semiconductor chips 310 and 320 are connected to the heat sink 1100 through the elastic connectors 1200 and 1300 in a wide area, the heat generated in the semiconductor chips 310 and 320 is transferred from the elastic connectors 1200 and 1300 to the heat sink 1100. Can be transmitted efficiently. Thereby, the heat generated in the semiconductor chips 310 and 320 can be efficiently radiated from the heat sink 1100 to the outside.

  Further, since the elastic connectors 1200 and 1300 have a stress relaxation structure as shown in FIGS. 4B and 4C, the stress applied to the semiconductor chips 310 and 320 can be relaxed, and the semiconductor Damage to the semiconductor chips 310 and 320 due to stress applied to the chips 310 and 320 can be prevented in advance.

  Further, when the flat metal plate 1210 is folded at the first folded portion P1 and the second folded portion P2, a rounded portion is formed at least on the outer peripheral surface side of the first folded portion P1 and the second folded portion P2. It is folded back. Further, the flat metal plate 1210 has a chamfered portion C formed on the edge portion on the upper surface side of the flat metal plate 1210 (see FIG. 4A), and the flat metal plate 1210 is used. The elastic connectors 1200 and 130 produced as shown in FIG. 4 have a chamfered portion C at the edge of the surface in contact with the semiconductor chips 310 and 320 and the heat sink 1100 (FIG. 4B and FIG. 4). 4 (c).) For this reason, when performing soldering, favorable soldering is attained. This will be described with reference to FIG.

  FIG. 5 is an enlarged view of the portion of the broken line frame A in FIG. 2 viewed from the side. In FIG. 5, the symbol F indicates a solder fillet generated by soldering, and the symbol R indicates a rounded portion formed on the outer peripheral surface side of the first folded portion P1 and the second folded portion P2.

  As shown in FIG. 5, the first elastic connector 1200 has rounded portions R formed on the outer peripheral surface sides of the first folded portion P1 and the second folded portion P2, and the semiconductor chips 310 and 320 and the heat sink 1100. A chamfered portion C is formed at the edge of the surface in contact with the surface. Therefore, the solder enters the rounded portion R and the chamfered portion C on the outer peripheral surface side of the first folded portion P1 and the second folded portion P2, and a good solder fillet F is formed between the semiconductor chip 310 and the heat sink 1100. The Thereby, the bonding state between the semiconductor chip 310 and the heat sink 1100 becomes strong.

  In FIG. 5, the soldering state between the first elastic connector 1200, the first semiconductor chip 310, and the heat sink 1100 has been described. However, the second elastic connector 1300, the second semiconductor chip 310, and the heat sink 1100 The soldering state is also good.

  In the resin-encapsulated semiconductor device 10 according to the first embodiment, the elastic connectors 1200 and 1300 have a stress relaxation structure as shown in FIG. 4, but are not limited to this, and the elastic connectors As 1200 and 1300, those having various stress relaxation structures can be used. Below, the modification of an elastic connector is demonstrated.

[First Modification of Elastic Connector]
FIG. 6 is a view for explaining a first modification of the elastic connector 1200. Here, a modified example of the first elastic connector 1200 will be described. As a first modification of the elastic connector 1200, two modifications (see FIGS. 6A and 6B) will be described here. The elastic connectors in these two modifications are denoted by reference numerals 1202 and 1204 and will be described as the elastic connectors 1202 and 1204. The elastic connector 1202 shown in FIG. 6 (a) and the elastic connector 1204 shown in FIG. 6 (b) are different in how the first flat plate portion 1211 and the second flat plate portion 1212 are folded in the elastic connector 1200 shown in FIG. Since only the case of FIG. 4 is different, the same parts as those of the elastic connector 1200 shown in FIG.

  The stress relaxation structure of the elastic connector 1200 shown in FIG. 4 includes the first flat plate portion 1211 and the second flat plate portion so that the first flat plate portion 1211 and the second flat plate portion 1212 are parallel to the third flat plate portion 1213, respectively. The first flat plate portion 1211 and the second flat plate portion 1212 are not necessarily parallel to the third flat plate portion 1213. For example, FIG. 6A and FIG. It may be configured as shown in b).

  That is, the elastic connector 1202 shown in FIG. 6A and the elastic connector 1204 shown in FIG. 6B have the first flat plate portion so that one end t1 is in contact with or close to the third flat plate portion 1213. The second flat plate portion 1212 is moved in the second direction so that the other end portion t2 is in contact with or close to the third flat plate portion 1213 while being turned back by rotating the 1211 in the first turning direction at the first turn-back portion P1. The folded portion P2 is folded so as to be rotated in the second rotational direction. Although not shown in FIG. 6, in practice, the elastic connector 1202 and the elastic connector 1204 are arranged on the edge of the surface in contact with the semiconductor chip 310 and the heat sink 1100 as shown in FIG. A chamfered portion C as shown in FIG.

  In the stress relaxation structure of the elastic connector 1202, as shown in FIG. 6A, only a part of the first flat plate portion 1211 is parallel to the third flat plate portion 1213 on the first flat plate portion 1211 side. And the inclined portion 1211b from the flat plate portion 1211a to the tip of the first flat plate portion 1211 (one end t1 of the flat metal plate 1210) toward the third flat plate portion 1213. On the second flat plate portion 1212 side, the flat plate portion 1212a is such that only a part of the second flat plate portion 1212 is parallel to the third flat plate portion 1213, and the flat plate portion. A portion from 1212a to the tip of the first flat plate portion 1212 (the other end t2 of the flat metal plate 1210) is formed as an inclined portion 1212b toward the third flat plate portion 1213. It has become a thing.

  Further, as shown in FIG. 6B, the elastic connector 1204 has the first flat plate portion 1211 on the first flat plate portion 1211 side so that the tip portion (one end t1 of the flat metal plate) is the first plate portion. 3 The flat plate portion 1213 is formed so that the entire first flat plate portion 1211 is curved so as to be in contact with or close to the flat plate portion 1213, and also on the second flat plate portion 1212 side, the tip end portion of the first flat plate portion 1212 The other end portion t2) of the second flat plate portion 1213 is formed so as to bend so that the entire second flat plate portion 1212 is curved.

  In the elastic connector 1202 having the stress relaxation structure as shown in FIG. 6A, the same effect as that of the elastic connector 1200 having the stress relaxation structure as shown in FIG. 4 can be obtained. Further, the elastic connector 1204 having the stress relaxation structure as shown in FIG. 6B also provides the same effect as the elastic connector 1200 having the stress relaxation structure as shown in FIG. In the above, the modification of the first elastic connector 1200 has been described. However, the second elastic connector 1300 can be similarly modified.

[Second Modification of Elastic Connector]
FIG. 7 is a view for explaining a second modification of the elastic connector 1200. Here, a modified example of the first elastic connector 1200 will be described. The elastic connector as the second modified example is denoted by reference numeral 1600 and will be described as the elastic connector 1600. FIG. 7A is a perspective view showing a flat metal plate 1610 used for the elastic connector 1600, and FIG. 7B is a perspective view of the elastic connector 1600. FIG.

  The elastic connector 1600 can be formed by the following procedure. First, a flat metal plate 1610 as shown in FIG. In FIG. 7A, the flat metal plate 1610 is shown with no chamfered portion. However, in the flat metal plate 1610, the edge on the upper surface side of the flat metal plate 1610 is shown. It is preferable that a chamfered portion is formed in each of the edge portion and the edge portion on the lower surface side.

  In such a flat metal plate 1610, the first folded portion P1 is set at a predetermined position on the one end t1 side between the one end t1 and the other end t2 of the flat metal plate 1610. At the same time, the second folded portion P2 is set at a predetermined position on the other end t2 side. Here, a portion between one end t1 and the first folded portion P1 is a first flat plate portion 1611, and a portion between the other end t2 and the second folded portion P2 is a second flat plate portion 1612. A portion between the portion P1 and the second folded portion P2 will be described as a third flat plate portion 1613.

  Then, the third flat plate portion 1613 is arranged at the first turn-back portion P1 so that the third flat plate portion 1613 forms an acute angle with respect to the first flat plate portion 1611. And the second flat plate portion 1612 is moved in the second folded portion P2 so that the second flat plate portion 1612 is parallel to the first flat plate portion 1611. It is folded back so as to rotate in a second rotation direction (in this case, counterclockwise) opposite to the rotation direction. Also in this case, when the flat metal plate 1210 is folded at the first folded portion P1 and the second folded portion P2, a round portion is provided on at least the outer peripheral surface side of the first folded portion P1 and the second folded portion P2. It is folded to form.

  By folding back the flat metal plate 1610 in such a procedure, the elastic connector 1600 can be formed. The elastic connector 1600 has a Z-shaped stress relaxation structure as shown in FIG. It will have. 7B shows the elastic connector 1600 in which the chamfered portion is not formed, the edge portion on the side in contact with the semiconductor chip 310 and the heat sink 1100 also in the elastic connector 1600. It is preferable that a chamfered portion is formed.

  Here, when the length between the one end t1 and the other end t2 of the flat metal plate 1610 is L, the first folded portion P1 is L / 3 from the one end t1 or substantially the same. The second folded portion P2 is preferably set at a position L / 3 or substantially L / 3 from the other end t2. Specifically, when the length from one end t1 of the flat metal plate 1610 to the first folded portion P1 is L1, and the length from the other end t2 to the first folded portion P2 is L2. It is preferable to set the positions of the first folded portion P1 and the second folded portion P2 so that L1 = L2 = L / 3, but the positions may be slightly shifted. By setting the first folded portion and the second folded portion at such positions, the elastic connector 1600 can have a shape with excellent stability and excellent elasticity.

The elastic connector 1600 having such a stress relaxation structure is slightly longer than the elastic connector 1200 shown in FIG. 4 in terms of the length of the current path, but in terms of elasticity, the elastic connector 1600 shown in FIG. 1200 is not inferior.
The modification of the first elastic connector 1200 has been described above, but the second elastic connector 1300 can be similarly modified.

  In the second modification, the rotation direction when the flat metal plate 1610 is folded is exemplified by the case where the first rotation direction is the clockwise direction and the second rotation direction is the counterclockwise direction. The reverse, that is, the first rotation direction may be counterclockwise and the second rotation direction may be clockwise.

[Third Modification of Elastic Connector]
FIG. 8 is a view for explaining a third modification of the elastic connector 1200. Here, the first elastic connector 1200 will be described. It should be noted that the elastic connector as the third modification is denoted by reference numeral 1700 and will be described as the elastic connector 1700. FIG. 8A is a perspective view showing a flat metal plate 1710 used for the elastic connector 1700, and FIG. 8B is a perspective view of the elastic connector 1700.

  The elastic connector 1700 can be formed by the following procedure. First, a flat metal plate 1710 as shown in FIG. In FIG. 8A, the flat metal plate 1710 is shown with no chamfered portion, but the flat metal plate 1710 also has a flat metal plate 1210 (FIG. 4A). It is preferable that a chamfered portion similar to that in () is formed.

  In such a flat metal plate 1710, the folded portion P1 is set at a predetermined position on one end side between the one end t1 and the other end t2 of the flat metal plate 1710. Here, a description will be given assuming that a portion between one end t1 and the folded portion P1 is a first flat plate portion 1711 and a portion between the other end t2 and the folded portion P2 is a second flat plate portion 1712.

  Then, the second flat plate portion 1712 is folded back at the first rotation direction (in this case) so that the second flat plate portion 1712 has a predetermined distance d with respect to the first flat plate portion 1711 and is parallel to the first flat plate portion 1711. Fold it so that it rotates clockwise. In this case as well, when the flat metal plate 1710 is folded at the folded portion P1, it is folded so that a rounded portion is formed at least on the outer peripheral surface side of the folded portion P1.

  By folding the flat metal plate 1710 in such a procedure, an elastic connector 1700 can be formed. The elastic connector 1700 rotates the U-shape as shown in FIG. 8B by 90 degrees. It has a stress relaxation structure with a different shape. Although not shown in FIG. 8B, the elastic connector 1700 has a chamfered portion similar to that in FIG. 4B at the edge of the surface in contact with the semiconductor chip 310 and the heat sink 1100. Is preferably formed.

  Here, when the length between one end t1 and the other end t2 of the flat metal plate 1710 is L, the folded portion P1 is L from one end t1 or the other end t2. It is preferable to set at a position of / 2 or approximately L / 2. Specifically, if the length from one end t1 of the flat metal plate 1710 to the first folded portion P1 is L1, and the length from the other end t2 of the metal plate to the folded portion P1 is L2, It is preferable to set the position of the folded portion P1 so that L1 = L2 = L / 2, but the position may be slightly shifted. By setting the folded portion P1 to such a position, the elastic connector 1700 can have a shape with excellent stability and excellent elasticity.

The elastic connector 1700 having such a stress relaxation structure has the same effect as the elastic connector 1200 having the stress relaxation structure as shown in FIG.
The modification of the first elastic connector 1200 has been described above, but the second elastic connector 1300 can be similarly modified.

  In the third modification, the rotation direction when the flat metal plate 1710 is folded is exemplified by the case where the first rotation direction is the clockwise direction, but the opposite, that is, the first rotation direction is the same. It may be counterclockwise. Thus, when the first rotation direction is the counterclockwise direction, when the chamfered portion C is formed on the edge of the flat metal plate 1210, the lower surface side edge of the flat metal plate 1210 is formed. A chamfered portion C is formed. The chamfered portion C may be provided on both the upper surface side and the lower surface of the flat metal plate 1210.

[Embodiment 2]
FIG. 9 is a view for explaining the resin-encapsulated semiconductor device 20 according to the second embodiment. FIG. 9 is a view of the heat sink 1100 from the upper side, that is, the direction of the semiconductor chips 310 and 320. Further, the elastic connector used in the resin-encapsulated semiconductor device 20 according to the second embodiment is denoted by reference numeral 1800 and described as the elastic connector 1800.

  The resin-encapsulated semiconductor device 20 according to the second embodiment is different from the resin-encapsulated semiconductor device 10 according to the first embodiment described above in that one elastic connector 1800 extends over the semiconductor chips 310 and 320. The other points are the same as those of the resin-encapsulated semiconductor device 10 according to the first embodiment, except that it is disposed between 1100 and the semiconductor chips 310 and 320. For this reason, in FIG. 9, only the positional relationship between the heat sink 1100, the elastic connector 1800, and the semiconductor chips 310 and 320 is shown.

  As shown in FIG. 9, the elastic connector 1800 is disposed between the heat sink 1100 (indicated by a broken line in the drawing) and the semiconductor chips 310 and 320 so as to straddle the two semiconductor chips 310 and 320. The elastic connector 1800 has a stress relaxation structure corresponding to each of the semiconductor chips 310 and 320. For easy understanding, the stress relaxation structure corresponding to the first semiconductor chip 310 will be described as “stress relaxation structure 1800A”, and the stress relaxation structure corresponding to the second semiconductor chip 320 will be described as “stress relaxation structure 1800B”. .

  FIG. 10 is a view for explaining the elastic connector 1800. FIG. 10A is a plan view showing a flat metal plate 1810 used for the elastic connector 1800, and FIG. 10B is a perspective view of the elastic connector 1800. FIG.

  As shown in FIG. 10, the elastic connector 1800 has a stress relaxation structure 1800A corresponding to the semiconductor chip 310 and a stress relaxation structure 1800B corresponding to the semiconductor chip 320. As the stress relaxation structures 1800A and 1800B, any of the stress relaxation structures of the elastic connectors 1200, 1202, 1204, 1600, and 1700 shown in FIGS. 4, 6, 7, and 8 can be used. However, here, it is assumed that the stress relaxation structure of the elastic connector 1200 shown in FIG. 4 is used. 10, the same parts as those in FIG. 4 are denoted by the same reference numerals. The stress relaxation structures 1800A and 1800B of the elastic connector 1800 can be formed by a procedure substantially similar to the procedure described in FIG.

  The elastic connector 1800 includes a connecting plate portion 1820 that connects the stress relaxation structure 1800A and the stress relaxation structure 1800B between the stress relaxation structure 1800A and the stress relaxation structure 1800B. The connecting plate portion 1820 is integrated with the stress relaxation structures 1800A and 1800B.

  In such a connecting plate portion 1820, a stress interference suppressing portion 1830 is formed to suppress stresses from interfering with each other between the semiconductor chips 310 and 320. The stress interference suppression portion 1830 is formed by providing notches 1831 and 1832 at positions opposite to one edge 1821 and the other edge 1822 of the connecting plate portion 1820.

  By forming such a stress interference suppressing portion 1830 on the elastic connector 1800, stress applied to one semiconductor chip (for example, the first semiconductor chip 310) is applied to the other semiconductor chip (for example, the second semiconductor chip 320). It is possible to suppress transmission. The stress interference suppression unit 1830 is not limited to the configuration shown in FIG. 10, and various configurations are conceivable.

  FIG. 11 is a view for explaining a modification of the elastic connector 1800. FIG. 11A is a diagram illustrating a first modification of the elastic connector 1800, and FIG. 11B is a diagram illustrating a second modification of the elastic connector 1800. The elastic connector as a first modification of the elastic connector 1800 is denoted by reference numeral 1802 and will be described as the elastic connector 1802. Further, an elastic connector 1804 as a second modified example of the elastic connector 1800 is denoted by reference numeral 1804 and described as an elastic connector 1804. The elastic connector 1802 and the elastic connector 1804 are different from the one shown in FIG.

The stress interference suppressing portion 1830 of the elastic connector 1802 has a recess 1833 formed on a line orthogonal to one edge 1821 and the other edge 1822 as shown in FIG. .
The downward protrusion length of the recess 1833 (the length from the lower surface of the connecting plate portion 1820 to the tip of the outer peripheral surface of the recess 1833) h1 is the predetermined interval d described in FIG. The length is equal to or less than h2 corresponding to the added thickness. By setting the downward protruding length h1 of the concave portion 1833 in this manner, when the elastic connector 1800 is disposed between the heat sink 1100 and the semiconductor chips 310 and 320, the front end portion of the outer peripheral surface of the concave portion 1833 The elastic connector 1800 can be disposed in a stable state without contacting the heat sink 1100.

  Further, as shown in FIG. 11B, the stress interference suppressing portion 1830 of the elastic connector 1804 is a combination of the notches 1831 and 1832 shown in FIG. 10 and the recess 1833 shown in FIG. It's a thing. By configuring the stress interference suppression unit 1830 as shown in FIG. 11B, the stress interference suppression effect can be further enhanced.

  Note that the present invention is not limited to the above-described embodiment, and various modifications can be made without departing from the gist of the present invention. For example, the following modifications (1) to (4) are possible.

  (1) In the embodiment described above, the case where there are two semiconductor chips has been described. However, the number of semiconductor chips may be one, or may be three or more.

  (2) The stress relaxation structure of the elastic connector is not limited to the structure described in each of the above embodiments and modifications, and various structures can be employed.

  (3) In each of the above embodiments, the semiconductor chip is a diode, but the present invention is not limited to this.

  (4) In each of the above embodiments, the chamfered portion C has been described as having the chamfered portion C formed on the edge of at least one of the upper surface side and the lower surface side of the flat metal plate. Is not necessary.

  DESCRIPTION OF SYMBOLS 10,20 ... Resin sealing type | mold semiconductor device, 310 ... 1st semiconductor chip, 320 ... 2nd semiconductor chip, 400 ... Resin mold, 1000 ... Frame part, 1100 ... Heat sink 1200 ... first elastic connector, 1202, 1204 ... first modification of the elastic connector, 1210, 1610, 1710, 1810 ... flat metal plate, 1211, 1611, 1711 ... first 1 flat plate portion, 1212, 1612, 1712 ... second flat plate portion, 1213, 1613 ... third flat plate portion, 1300 ... second elastic connector, 1410, 1420, 1510, 1520 ... pair of Terminal fitting, 1600, second modified example of elastic connector 1200, 1700, third modified example of elastic connector 1200, 1800, elastic connector, 1802,. First modified example of connector 1800, 1804 ... first modified example of elastic connector 1800, 1820 ... connecting plate portion, 1830 ... stress interference suppressing portion, 1831, 1832 ... notched portion, 1833: concave portion, C: chamfered portion, P1: first folded portion, P2: second folded portion, R: rounded portion, t1: one end portion, t2 ... The other end

Claims (5)

A heat sink having electrical conductivity and thermal conductivity;
A semiconductor chip disposed on the upper surface side of the heat sink;
Entirely made flat metal plate, and said have a stress relaxation structure for alleviating the stress applied to the semiconductor chip, and an elastic connection element that is disposed between the heat sink and the semiconductor chip ,
An external connection lead electrically connected to the upper surface side electrode of the semiconductor chip;
A resin-sealed mold comprising: a mold resin that seals the heat sink, the semiconductor chip, the elastic connector, and the external connection lead in a state where the lower surface side of the heat sink and the tip end side of the external connection lead are exposed. A semiconductor device,
The heat sink is electrically connected to the lower surface side electrode of the semiconductor chip via the elastic connector,
The stress relaxation structure is
A first folded portion is set at a predetermined position on the one end side between one end and the other end of the flat metal plate, and a second is set at a predetermined position on the other end side. Is set as a first flat plate portion between the one end portion and the first folded portion, and a second flat plate portion is provided between the other end portion and the second folded portion. When the third flat plate portion is between the first folded portion and the second folded portion,
The first flat plate portion is turned back by rotating the first flat plate portion in the first rotation direction, and the second flat plate portion is turned by the second turn back portion opposite to the first rotation direction. The first flat plate portion and the second flat plate are formed so as to be folded back so as to be rotated in the second rotation direction and the one end portion and the other end portion are separated from each other. The portion is in contact with the heat sink side, and the third flat plate portion is disposed in contact with the semiconductor chip side ,
The semiconductor chips are a plurality of semiconductor chips arranged to be connected in parallel,
The elastic connector is
Arranged between the heat sink and the plurality of semiconductor chips so as to straddle the plurality of semiconductor chips,
The stress relaxation structure corresponding to each of the semiconductor chips of the plurality of semiconductor chips,
A connecting plate portion that exists between the stress relaxation structures corresponding to each of the semiconductor chips and that connects the stress relaxation structures corresponding to the respective semiconductor chips;
The connecting plate portion has one edge and the other edge parallel to the one edge,
The connection plate portion is provided with a stress interference suppression portion that suppresses interference of the stress between the semiconductor chips,
The resin-encapsulated semiconductor device, wherein the stress interference suppression portion is formed by providing at least a concave portion formed on a line orthogonal to the one edge portion and the other edge portion . The resin-encapsulated semiconductor device according to claim 1,
The stress relaxation structure is
The first flat plate portion is rotated in the first rotation direction by the first folded portion so that the first flat plate portion has a predetermined interval with respect to the third flat plate portion and is parallel to the third flat plate portion. And the second flat plate portion has a predetermined interval with respect to the third flat plate portion and is parallel to the second flat plate portion so that the second flat plate portion is parallel to the first rotation direction. A grease-sealed semiconductor device, wherein the grease-sealed semiconductor device is formed by being folded back so as to be rotated in a second rotation direction opposite to the first direction. The resin-encapsulated semiconductor device according to claim 1,
The stress relaxation structure is
The first flat plate portion is turned back in the first rotation direction so that the one end portion is in contact with or close to the third flat plate portion, and the other end is turned. The second flat plate portion is turned back by rotating the second flat plate portion in the second rotation direction opposite to the first rotation direction so that the portion contacts or approaches the third flat plate portion. A resin-encapsulated semiconductor device, characterized by being formed. The resin-encapsulated semiconductor device according to any one of claims 1 to 3,
The resin-encapsulated semiconductor device, wherein the folded portion is folded so that at least the outer peripheral surface side of the folded portion is rounded. The resin-encapsulated semiconductor device according to any one of claims 1 to 4,
2. The resin-encapsulated semiconductor device according to claim 1, wherein a chamfered portion is formed on an edge portion of a surface of the elastic connector that is in contact with the semiconductor chip and the heat sink.

Images