JP5125530B2 - Power converter - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a power converter capable of efficiently cooling a semiconductor device that performs power conversion without causing an increase in size, cost, thermal resistance or stress concentration. <P>SOLUTION: The power converter includes: a first substrate 13 mounted on a main surface of a first radiator 17; the first semiconductor device 11 mounted on a main surface of the first substrate 13; a first electrode 15 mounted on the main surface of the first radiator 17 or the main surface of the first substrate 13; a heat spreader 19 that is disposed on an upper surface of the first semiconductor device 11 and brings a plurality of first wire bonding lines 24 that electrically connect a first upper surface electrode 23 of the first semiconductor device 11 with the first electrode 13 into contact with the first upper surface electrode 23; a protrusion 19b provided on one surface side of the heat spreader 19 and brought into contact with or positioned in proximity to the first upper surface electrode 23 through the first wire bonding lines 24; a recess 19a that is approximated to the first wire bonding lines 24; and a second radiator 18 mounted on the other surface side of the heat spreader 19. <P>COPYRIGHT: (C)2009,JPO&amp;INPIT

Description

  The present invention relates to a power conversion device, and more particularly to a power conversion device that requires a sufficient countermeasure against heat generation in order to use a large number of semiconductor elements.

Conventionally, as a power conversion device that generally performs power conversion, for example, a “semiconductor device” (see Patent Document 1) is known.
FIG. 19 is a cross-sectional explanatory view schematically showing the structure of a conventional semiconductor device. As shown in FIG. 19, the conventional semiconductor device 1 has a lower semiconductor element 2 and an upper semiconductor element 3. The lower semiconductor element 2 is mounted on the cooler 6a via an insulating substrate 5a having metal patterns on both sides connected by solder 4a, and the upper semiconductor element 3 is connected on both sides by solder 4b. It is mounted on the cooler 6b through an insulating substrate 5b having a metal pattern.

An electrode 7 a is mounted on the main surface electrode of the lower semiconductor element 2, and a cooler 6 b is placed on the electrode 7 a via an insulating material 8 a, and on the main surface electrode of the upper semiconductor element 3. An electrode 7b is mounted, and a cooler 6c is placed on the electrode 7b via an insulating material 8b.
The conventional semiconductor device 1 has the coolers 6a, 6b, and 6c on the front surface side and the back surface side of the lower semiconductor element 2 and the upper semiconductor element 3, respectively, thereby improving the cooling performance and suppressing the temperature rise. Yes. In addition, the planar size is reduced by forming a stacked structure in which the semiconductor elements 2 and 3 are stacked one above the other.
JP 2001-244407 A

However, the conventional semiconductor device 1 has the following problems.
In order to cool both surfaces of the semiconductor element 2 or 3, it is necessary to join a cooler or a heat radiator with less thermal resistance on both surfaces of the semiconductor element. In particular, when an electrode surface connected to the live potential is formed on the plane of the semiconductor element, it is very difficult to join the electrode surface, the cooler, and the heat radiating body with little thermal resistance.
An object of the present invention is to provide a power conversion device capable of efficiently cooling a semiconductor element that performs power conversion without causing an increase in thermal resistance.

  In order to achieve the above object, a power converter according to the present invention is mounted on a first radiator, a first substrate mounted on a main surface of the first radiator, and a main surface of the first substrate. A first semiconductor element, a first electrode mounted on a main surface of the first radiator or the main surface of the first substrate, and disposed on an upper surface of the first semiconductor element. A plurality of first wire bonding lines that electrically connect one upper surface electrode and the first electrode; a heat spreader that contacts the first upper surface electrode; and the first wire bonding line provided on one surface side of the heat spreader. A convex portion in contact with or close to the first upper surface electrode, a concave portion in proximity to the first wire bonding line, and a second radiator mounted on the other surface side of the heat spreader. .

According to the present invention, the first substrate is mounted on the main surface of the first radiator, the first semiconductor element is mounted on the main surface of the first substrate, and the first surface of the first radiator is A first electrode is mounted on a main surface of one substrate, and a plurality of first wire bonding lines that electrically connect the first upper surface electrode of the first semiconductor element and the first electrode on the upper surface of the first semiconductor element. And a heat spreader that is in contact with the first upper surface electrode, and a convex portion that is in contact with or close to the first upper surface electrode through the first wire bonding line on one surface side of the heat spreader, and the first wire bonding line And a second radiator is mounted on the other side of the heat spreader.
Thereby, the semiconductor element which performs power conversion can be efficiently cooled without causing an increase in size, an increase in cost, an increase in thermal resistance, and stress concentration.

The best mode for carrying out the present invention will be described below with reference to the drawings.
(First embodiment)
FIG. 1 is a cross-sectional explanatory view showing the configuration of the power conversion device according to the first embodiment of the present invention. FIG. 2 is a cross-sectional explanatory view taken along the line AA of FIG. 3 is an explanatory plan view of the back side of the heat spreader of FIG. FIG. 4 is an explanatory plan view of FIG. 2 excluding the heat spreader. 5 shows the cross-sectional structure of each part of FIG. 4, (a) is a cross-sectional explanatory view along the line BB, (b) is a cross-sectional explanatory view along the line CC, (c) is a DD line Cross-sectional explanatory drawing which follows, (d) is a cross-sectional explanatory drawing which follows the EE line.

As shown in FIGS. 1 to 5, the power conversion device 10 includes a first semiconductor element 11, a second semiconductor element 12, a first substrate 13, a second substrate 14, a first electrode 15, a second electrode 16, and a first semiconductor element 11. The radiator 17, the second radiator 18, the heat spreader 19, and the circuit board 20 are formed and formed by layering them (see FIG. 1).
The first semiconductor element 11 and the second semiconductor element 12 sandwich a heat spreader 19 that is a heat dissipating member made of a metal body for heat diffusion, and a first substrate 13 is placed outside the first semiconductor element 11. The first heat radiator 17 is located on the outside of the second semiconductor element 12, and the second heat radiator 18 is located on the outside of the second semiconductor element 12 via the second substrate 14. That is, the first semiconductor element 11, the first substrate 13, and the first radiator 17, and the second semiconductor element 12, the second substrate 14, and the second radiator 18 sandwich the heat spreader 19 so that their main surface sides face each other. (Refer to FIG. 1).

The first substrate 13 is mounted on the main surface of the first radiator 17 via the insulating layer 21, and the first semiconductor element 11 is mounted on the main surface of the first substrate 13 via the solder layer 22. (See FIG. 1). The first substrate 13 may be mounted on the main surface of the first radiator 17 directly without the insulating layer 21 or via a base plate (not shown) instead of the insulating layer 21.
Further, the first electrode 15 is mounted on the main surface of the first heat radiator 17 adjacent to the first substrate 13 via the insulating layer 21, and the first upper surface electrode 23 of the first semiconductor element 11 (see FIG. 4). And the first electrode 15 are electrically connected by a first wire bonding line 24 (see FIG. 2), and the first upper surface electrode 23 and the first wire bonding line 24 are positioned on the upper surface side of the first semiconductor element 11. The heat spreader 19 is brought into contact with a high thermal conductivity material, for example, a high thermal conductivity adhesive 25 (see FIG. 5).

The first electrode 15 may be directly mounted on the main surface of the first heat radiator 17 without the insulating layer 21 interposed therebetween, and is not on the main surface of the first heat radiator 17 but the main surface of the first substrate 13. May be implemented above.
The circuit board 20 is a gate wiring board, for example, and is connected to the control terminal 27 of the first semiconductor element 11 by a connection line 26 (see FIGS. 1 and 4).
Here, a concavo-convex portion is provided on one surface side of the heat spreader 19, the convex portion is in contact with or close to the first upper surface electrode 23 through the first wire bonding line 24, and the concave portion is the first wire bonding line 24. To be in close proximity to.

That is, generally, the connection by the first wire bonding wire 24 to the first semiconductor element 11 that performs high power conversion is performed using a metal wire (for example, aluminum wire) having a diameter of about 350 to 500 μm, and these The plurality of metal lines are drawn out in the same direction at regular intervals to some extent, and are connected to the wiring electrodes (see FIG. 4).
Therefore, a plurality of groove-shaped recesses 19a extending in the same direction as the first wire bonding lines 24 are provided apart from each other on the first upper surface electrode 23 side of the first semiconductor element 11, and a partition wall between the adjacent recesses 19a is provided. A heat spreader 19 (see FIG. 3) having a convex portion 19b is mounted. The recess 19 a of the heat spreader 19 has a shape that can be disposed close to the first wire bonding line 24.

Thereby, the convex part 19b between the adjacent concave parts 19a enters between the plurality of first wire bonding lines 24 connected to the first upper surface electrode 23 of the first semiconductor element 11, and the heat spreader 19 is moved to the first. It makes it adjoin to the upper surface electrode 23 (refer FIG. 5).
A groove-like recess 19a provided on one surface side of the heat spreader 19 corresponds to a bonding connection portion of the first wire bonding wire 24 to be connected to the first upper surface electrode 23 of the first semiconductor element 11 (FIG. 5C). Reference) is made shallower than the portion not corresponding to the bonding connection portion (see FIG. 5D).
And the 2nd heat radiator 18 is mounted in the other surface side of the heat spreader 19, ie, the opposite side to the main surface side in which the 1st heat radiator 17 is arrange | positioned (refer FIG. 1).

The second substrate 14 is mounted on the main surface of the second radiator 18 via an insulating layer 29, and the second semiconductor element 12 is mounted on the main surface of the second substrate 14 via a solder layer 30. (See FIG. 1). Note that the second substrate 14 may be mounted on the main surface of the second radiator 18 directly without the insulating layer 29 or via a base plate (not shown) instead of the insulating layer 29.
Further, the second electrode 16 is mounted on the main surface of the second radiator 18 adjacent to the second substrate 14 via the insulating layer 29, and the second upper surface electrode (not shown) of the second semiconductor element 12 and the second electrode 16 are mounted. The two electrodes 16 are electrically connected by the second wire bonding line 31. Then, the second upper surface electrode and the second wire bonding line 31 are brought into contact with the other surface side, which is the second semiconductor element 12 side, of the heat spreader 19 directly or through a high thermal conductivity material, for example, a high thermal conductivity adhesive 32. (See FIG. 1).

The second electrode 16 may be directly mounted on the main surface of the second radiator 18 without the insulating layer 29 interposed therebetween, and is not on the main surface of the second radiator 18 but on the main surface of the second substrate 14. May be implemented above.
Further, on the other surface side of the heat spreader 19, similarly to the one surface side, a plurality of groove-shaped recesses extending in the same direction as the second wire bonding line 31, and a partition wall-shaped protrusion between adjacent recesses are provided, The convex portion is in contact with or close to the second upper surface electrode through the second wire bonding line 31, and the concave portion is in proximity to the second wire bonding line 31.
The groove-shaped recess provided on the other surface side of the heat spreader 19 is the second upper surface electrode of the second semiconductor element 12 of the second wire bonding line 31, similarly to the groove-shaped recess 19 a provided on the one surface side of the heat spreader 19. The portion corresponding to the bonding connection portion to be connected to the substrate is made shallower than the portion not corresponding to the bonding connection portion.

The first electrode 15 and the second electrode 16 are electrically connected, and the first upper surface electrode 23 of the first semiconductor element 11 and the second upper surface electrode of the second semiconductor element 12 are electrically at the same potential. The configuration. Further, the first semiconductor element 11 and the second semiconductor element 12 are a combination of elements having a state where they do not perform the same electrical operation.
In this embodiment, as the power conversion device, for example, an inverter device that drives an electric motor is used, the first semiconductor element 11 is an IGBT (Insulated Gate Bipolar Transistor), and the second semiconductor element 12 is a diode. The first semiconductor element 11 has a first upper surface electrode 23 that is an emitter electrode and a control terminal 27 that is a gate electrode. The gate electrode (control terminal 27) is connected to the gate wiring board (circuit board 20) via the gate connection line (connection line 26) (see FIG. 4).

  Then, the IGBT as the first semiconductor element 11 and the diode as the second semiconductor element 12 are connected in an antiparallel arrangement in which the energization direction is electrically reversed to form the switch circuit 33 (see FIG. 1). . For example, as the three-phase inverter, the switch circuit 33 constitutes one high-side arm or low-side arm, and by forming six switch circuits 33 and electrically connecting them appropriately, A three-phase inverter can be configured.

  By having the configuration described above, the power conversion device 10 can obtain the following effects. In addition, the power converter device 10 includes the first semiconductor element 11, the first substrate 13, and the first radiator 17, and the second semiconductor element 12 and the second substrate 14 corresponding to the first constituent body. And the second structure composed of the second heat radiator 18 have a structure in which the heat spreader 19 is interposed therebetween, and the effects of the first structure have been described for the second structure as well. Applies to

  As a first effect, the first semiconductor element 11 that performs power conversion is not only from the back surface side of the first semiconductor element 11 but also from the upper surface (main surface) side of the first semiconductor element 11, that is, the first semiconductor. Heat can be radiated from both sides of the element 11. In addition, the electrical connection of the first semiconductor element 11 to the first upper surface electrode 23 can be performed by the first wire bonding line 24 in a state where it can be easily and reliably performed, and heat is radiated from the upper surface side of the first semiconductor element 11. Can be easily performed with low thermal resistance.

  That is, the metal body by the 1st wire bonding line 24 and the convex part 19b of the heat spreader 19 occupies most between the 1st upper surface electrode 23 of the 1st semiconductor element 11, and the heat spreader 19. FIG. And it becomes the shape where the high heat conductivity adhesive agent 25 fills the clearance gap which arises because the wired 1st wire bonding line 24 will be in the state which floated in the mountain shape. For this reason, the heat flow from the upper surface side of the first semiconductor element 11, that is, from the heat spreader 19 side, is transmitted to the second radiator 18 through the heat spreader 19 that is a metal body. At this time, the high thermal conductivity adhesive 25 also contributes to heat transfer.

Therefore, most of the first upper surface electrode 23 of the first semiconductor element 11 and the second radiator 18 can be thermally coupled with a low thermal resistance due to a metal body. Therefore, the heat generated in the first semiconductor element 11 is reduced from the upper surface side, that is, the heat spreader 19 side, in addition to the heat radiation from the back side of the first semiconductor element 11 according to the conventional structure, that is, the first radiator 17 side. Since heat can be dissipated by the resistor, the power converter 10 can be greatly reduced in size easily.
As a second effect, since the heat spreader 19 that is a metal body is placed immediately above the first semiconductor element 11, the heat spreader 19 serves as a shield body against noise generated from the first semiconductor element 11. Function. In particular, since it can be arranged very close to the first semiconductor element 11, a high shielding effect can be obtained.

As a third effect, even if there is a gap non-uniformity due to a variation in the height of the apparatus components or a difference in parallelism, it can be prevented from being affected by it, and the manufacturing becomes easier.
In general, a power converter is composed of a plurality of semiconductor elements. Further, it is desirable that the number of the second heat radiator 18 disposed on the upper surface side of the first semiconductor element 11 is one as the whole power conversion device from the viewpoint of reducing the number of parts and simplifying the manufacturing process. Under such circumstances, the thickness of the first semiconductor element 11 is not necessarily uniform, and the parallelism of the second radiator 18 with respect to the first substrate 13 does not always match.

In the configuration of the power conversion device 10, the high thermal conductivity adhesive 25 when the second radiator 18 or the heat spreader 19 is pressed even if there is a gap nonuniformity due to a height variation or parallelism difference between these components. By crushing, the space between the first upper surface electrode 23 of the first semiconductor element 11 and the second heat radiator 18 or the heat spreader 19 can be reliably filled.
In other words, in the case of a structure in which a plurality of structural members such as the power conversion device 10 are stacked and the thickness of each semiconductor element related part is not necessarily uniform, variation in accuracy due to dimensional tolerance of each structural member occurs. It is inevitable to do. However, by utilizing the property that the high thermal conductivity adhesive 25 in this configuration is easily deformable before the thermosetting treatment, the high thermal conductivity adhesive 25 is easily crushed, thereby easily causing variations in accuracy of each component. It can be absorbed and unaffected.

Therefore, it is possible to easily deal with the occurrence of variations in accuracy in the respective constituent members such as the first semiconductor element 11 and the heat spreader 19, and between the first upper surface electrode 23 of the first semiconductor element 11 and the second radiator 18. Can be reliably thermally coupled at all times.
Furthermore, manufacture of the power converter device 10 becomes easier. That is, since the high thermal conductivity adhesive 25 only needs to be heat-cured after application, it is a laminated structure in which the second radiator 18 or the heat spreader 19 is placed on the first semiconductor element 11 as in this configuration. However, after applying the high thermal conductivity adhesive 25 at normal temperature and then thermosetting, the structure can be easily formed.

  As a fourth effect, it is possible to avoid the occurrence of reliability concerns due to thermal stress as much as possible. In general, copper, aluminum, or the like is used for the second radiator 18 and the heat spreader 19 from the viewpoint of material cost, but these metal materials are greatly different in thermal expansion coefficient from semiconductors. On the other hand, in this configuration, thermal stress can be alleviated even when materials having greatly different thermal expansion coefficients are joined by deformation of the high thermal conductivity adhesive 25 and the first wire bonding line 24.

As a fifth effect, the temperature rise of the first semiconductor element 11 and the second semiconductor element 12 can be remarkably suppressed. That is, when the first semiconductor element 11 generates heat by energization, the heat is transmitted from the back surface of the first semiconductor element 11 (on the first radiator 17 side) to the first radiator 17 via the first substrate 13. In addition, the heat is transmitted from the upper surface of the first semiconductor element 11 (on the heat spreader 19 side) to the second semiconductor element 12 via the heat spreader 19 and further to the second radiator 18 via the second substrate 14. Thereby, the temperature rise of the first semiconductor element 11 can be remarkably suppressed, and the temperature rise of the second semiconductor element 12 can be remarkably suppressed similarly.
Here, in this configuration, since the upper surface electrode of the first semiconductor element 11 and the upper surface electrode of the second semiconductor element 12 are electrically at the same potential, the semiconductor elements 11 and 12 can be disposed sufficiently close to each other. . In other words, there is no need to provide an interval in order to obtain electrical insulation.

  That is, since the first semiconductor element 11 and the second semiconductor element 12 are the switch circuits 33 that function as the arms of the inverter device, the upper surface electrodes and the rear surface electrodes of both the semiconductor elements 11 and 12 are electrically at the same potential. Become. Furthermore, the high thermal conductivity material that joins the upper surface electrodes of the semiconductor elements 11 and 12, for example, the high thermal conductivity adhesive 25, does not need to be an insulator. In general, a bonding material having electrical conductivity has better thermal conductivity than a bonding material having insulating properties, and can be further reduced in thermal resistance.

That is, an increase in thermal resistance may be caused between the upper surface electrode of the first semiconductor element 11 and the heat spreader 19 and between the upper surface electrode of the second semiconductor element 12 and the heat spreader 19 in order to achieve sufficient electrical insulation. Absent. As a result, the heat generated from the first semiconductor element 11 and the second semiconductor element 12 can be transmitted to the other semiconductor element disposed opposite to the upper surface side with a low thermal resistance. And this other semiconductor element can also suppress a temperature rise because the back surface side is thermally coupled to a radiator (the first radiator 17 or the second radiator 18) with a low thermal resistance.
Therefore, each heat generation of the first semiconductor element 11 and the second semiconductor element 12 can be transmitted to both the upper and lower surfaces with low thermal resistance, and can be led to the radiator (the first radiator 17 or the second radiator 18). Therefore, the temperature rise of the first semiconductor element 11 and the second semiconductor element 12 can be significantly suppressed.

  As a sixth effect, a heat dissipation path can be easily and reliably provided on the upper surface side of each of the first semiconductor element 11 and the second semiconductor element 12. That is, the second semiconductor element 12 is placed so as to face the upper surface side of the first semiconductor element 11. At this time, the shape of the portion for fixing the second semiconductor element 12 to the first semiconductor element 11 is set. There is an undeniable concern that the distance between the upper surface electrodes of the semiconductor elements 11 and 12 varies depending on the accuracy. Even in this case, in this configuration, since the dimensional variation can be absorbed by the crushing of the high thermal conductivity adhesive 25, the dimensional variation can be absorbed and then cured and bonded.

  Furthermore, since the upper surface electrodes of the first semiconductor element 11 and the second semiconductor element 12 are at the same potential, there is no need to interpose an insulating material therebetween. For this reason, there is no inadequate joining due to the variation in the thickness of the insulating material, and neither the manufacturing process is complicated due to the joining of the front and back surfaces of the insulating material and the upper surface electrodes of the two semiconductor elements 11 and 12 together.

  As a seventh effect, since the power conversion device 10 has an upper and lower stacked structure in which the second semiconductor element 12 is mounted on the upper surface of the first semiconductor element 11 via the heat spreader 19, the first semiconductor element 11 and the first semiconductor element 11 2 The projected area for mounting the semiconductor element 12 can be greatly reduced as compared with the side-by-side structure. Further, since no new insulating material is interposed between the upper surface electrodes of the first semiconductor element 11 and the second semiconductor element 12 and both the upper surface electrodes are at the same potential, the two semiconductor elements 11 and 12 are sufficiently connected. Can be placed close to each other. Therefore, the increase in thickness of the power conversion device 10 can be suppressed, and the projected area of the mounting portion of the semiconductor element can be greatly suppressed, which can greatly contribute to downsizing of the entire device.

As an eighth effect, if the projected area of the power converter is reduced by the above effect, the bus bar connected from the semiconductor element to the smoothing capacitor can be significantly shortened. Therefore, since the parasitic inductance in the bus bar is reduced, the stability of the power conversion device with respect to the large current operation is remarkably increased.
As a ninth effect, the temperature increase of the first semiconductor element 11 and the second semiconductor element 12 can be further suppressed by the heat capacity of the heat spreader 19. For example, when the power conversion device 10 is mounted and used in an automobile or the like, generally, the maximum output is often limited to a short time such as when the vehicle starts. In this case, the effect of slowing the temperature rise of the semiconductor element is obtained by the heat capacity of the heat spreader 19. If the duration of the maximum output becomes limited, as a result, the temperature of the semiconductor element can be kept low.

As a tenth effect, the control terminal 27 of the first semiconductor element 11 can be easily connected to the circuit board 20. This leads to a reduction in manufacturing cost of the power conversion device 10. That is, in the state without the heat spreader 19, when the upper surface electrode of the first semiconductor element 11 and the upper surface electrode of the second semiconductor element 12 are brought close to each other for thermal coupling, as a result, the semiconductor elements 11 and 12 are mounted. The distance between the first substrate 13 and the second substrate 14 is also reduced.
In this state, when the circuit board 20 to which the control terminal 27 of the semiconductor element is connected is mounted on the first board 13 or the second board 14, the connection line 26 for the electrical connection, the wiring pattern on the circuit board 20, the resistance, etc. It is not possible to secure a sufficient space for arranging the electronic components. As a result, there is a concern that a complicated arrangement structure must be adopted, resulting in an increase in cost.

In this configuration, by having the heat spreader 19, the space between the first semiconductor element 11 and the second semiconductor element 12 can be dissipated between the two semiconductor elements 11 and 12 while maintaining a state in which heat can be radiated by a sufficiently small thermal resistance. In addition, the distance between the first substrate 13 and the second substrate 14 can be increased. Therefore, a space for mounting and arranging the connection line 26 of the control terminal 27, the circuit board 20, and the electronic component can be easily secured. Thereby, manufacturing cost can be reduced.
As an eleventh effect, the first wire bonding line 24 and the first electrode 15 and the second wire bonding line 31 and the second electrode 16 can be easily connected. This also leads to a reduction in manufacturing cost of the power conversion device 10.

That is, in the state without the heat spreader 19, as described above, the distance between the first substrate 13 and the second substrate 14 is reduced. In this state, there is a concern that the connection portion of the first wire bonding line 24 on the first electrode 15 and the connection portion of the second wire bonding line 31 on the second electrode 16 may interfere with each other. Therefore, if these two connecting portions are arranged so as to be shifted in order to prevent interference, the power conversion device 10 is increased in size.
Furthermore, in order to obtain a margin in the interval between these connection portions, even if the first electrode 15 is lower than the first substrate 13 and the second electrode 16 is lower than the second substrate 14, Problems are likely to arise.

  The first wire bonding line 24 and the second wire bonding line 31 are connected to the first electrode 15 and the second electrode 16 which are lower than the upper surface electrode after being connected to the upper surface electrode of the semiconductor element. For this reason, there is a possibility that the first wire bonding line 24 and the second wire bonding line 31 and the ends of the first semiconductor element 11 and the second semiconductor element 12 are too close to each other. Since the first wire bonding line 24 and the second wire bonding line 31 and the ends of the first semiconductor element 11 and the second semiconductor element 12 have different electrical potentials, it is necessary to ensure electrical insulation. Therefore, there is a concern that a complicated arrangement structure must be adopted, resulting in an increase in cost.

  In this configuration, by providing the heat spreader 19, the first electrode is formed from the upper surface electrode of each semiconductor element while maintaining a state where heat can be radiated between the first semiconductor element 11 and the second semiconductor element 12 by a sufficiently small thermal resistance. 15 and the second electrode 16 can be easily positioned higher. Therefore, the gap between the first wire bonding line 24 and the second wire bonding line 31 and the end portions of the first semiconductor element 11 and the second semiconductor element 12 can be easily widened. As a result, the manufacturing cost for securing electrical insulation between the first wire bonding line 24 and the second wire bonding line 31 and the end portions of the first semiconductor element 11 and the second semiconductor element 12 does not increase. Can be.

As a twelfth effect, electrical insulation inside the power converter 10 can be easily realized. As a result, the manufacturing cost can be reduced. That is, the first substrate 13 and the second substrate 14 on which each semiconductor element is mounted, the first wire bonding line 24 and the second wire bonding line 31, the first electrode 15 and the second electrode 16, and the control terminal 27. The connection lines 26 and the circuit board 20 have different electrical potentials, and it is necessary to ensure electrical insulation.
In the present configuration, since the heat spreader 19 is provided, it is possible to widen the interval between the constituent members having different electrical potentials. Therefore, electrical insulation can be achieved easily, that is, at low cost.

When the power converter 10 is, for example, a three-phase inverter, for example, the switch circuit 33 that performs power conversion has an IGBT connected to a high speed rectifier diode (FRD) in parallel. In the IGBT and FRD connected in parallel, current does not flow through both of them except for a transition state of a very short time in the switching operation. Further, when the inverter is in the power running state, the current mainly flows through the IGBT, and when the inverter is in the regenerative state, the current mainly flows through the FRD.
Therefore, for example, if the first semiconductor element 11 is IGBT and the second semiconductor element 12 is FRD, the back electrodes of the first semiconductor element 11 and the second semiconductor element 12 are at the same potential, It becomes the same potential.

  A large current flows through the first semiconductor element 11 and the second semiconductor element 12 at the same time, and no significant heat is generated. That is, since the upper surface electrodes of the first semiconductor element 11 and the second semiconductor element 12 are joined by the high thermal conductivity adhesive 25, the first semiconductor which is mainly an IGBT while the motor is driven by powering. The element 11 generates heat and is radiated from both the back surface side and the main surface side of the first semiconductor element 11. At this time, the heat generation of the second semiconductor element 12 is small and does not adversely affect the temperature suppression of the first semiconductor element 11. Further, during the regenerative drive of the electric motor, the second semiconductor element 12 which is mainly FRD generates heat and is radiated from both the back surface side and the main surface side of the second semiconductor element 12. At this time, the heat generation of the first semiconductor element 11 is small and does not adversely affect the temperature suppression of the second semiconductor element 12.

Therefore, as a thirteenth effect, the first semiconductor element 11 and the second semiconductor element 12 are stacked one above the other, but there is one element that generates a large amount of heat, and that heat can also be dissipated up and down. it can. Therefore, there is no possibility that the heat of the first semiconductor element 11 and the second semiconductor element 12 is applied and excessive heat is generated. Further, since the upper surface electrodes of the first semiconductor element 11 and the second semiconductor element 12 are at the same potential, it is not always necessary to interpose an insulating material that becomes a thermal resistance.
Therefore, the first semiconductor element 11 and the second semiconductor element 12 can be thermally coupled with a low thermal resistance. Furthermore, the thermal coupling between the first semiconductor element 11, the second semiconductor element 12, the radiator 17, and the radiator 18 is performed by using the insulating layer 21 on the back surface side of the first semiconductor element 11 and the second semiconductor element 12 as in the past. It is not difficult to perform sufficient and reliable thermal coupling of this part.

As a result, the heat generated from the first semiconductor element 11 and the second semiconductor element 12 can be flowed in both the upper and lower directions (both front and back surfaces), and further efficiently cooled.
As a fourteenth effect, the thermal resistance related to heat radiation from the upper surface electrode can be reduced. The height of the first wire bonding line 24 relative to the upper surface electrode of the first semiconductor element 11 and the height of the second wire bonding line 31 relative to the upper surface electrode of the second semiconductor element 12 are not constant. Therefore, for example, in the portion where the wire bonding line is connected to the upper surface electrode of the semiconductor element, the position of the wire bonding line is inevitably lowered. In this portion, the concave portion 19a of the heat spreader 19 can also be made shallow so that the gap between the wire bonding line and the heat spreader 19 can be kept small, so that the thermal resistance related to heat radiation from the upper surface electrode can be further reduced. In addition, the semiconductor element can be cooled.
(Second Embodiment)

6 is similar to FIG. 5 showing the cross-sectional structure of each part of the power conversion device according to the second embodiment of the present invention, (a) is a cross-sectional explanatory view along the line BB, (b) is CC Sectional explanatory drawing which follows a line, (c) is sectional explanatory drawing which follows a DD line, (d) is sectional explanatory drawing which follows an EE line.
As illustrated in FIG. 6, the power conversion device 35 includes a heat spreader 36 that covers the first semiconductor element 11 and the second semiconductor element 12. That is, the heat spreader 36 protrudes to the first semiconductor element 11 side, which is the upper surface side of the heat spreader 36, and has a tip that is located on the side of the first semiconductor element 11 and reaches the side surface of the first substrate 13. A portion 37 is provided. Similarly, on the second semiconductor element 12 side, a protruding wall portion (projecting toward the second semiconductor element 12 side and extending to the side of the second substrate 14 at the side of the second semiconductor element 12) (Not shown).

Further, as shown in FIG. 6A, on the side of the protruding wall portion 37 on the first semiconductor element 11 side where the first wire bonding line 24 is drawn, a wiring port for drawing out the first wire bonding line 24 is provided. 37a is cut away.
Other configurations and operations are the same as those of the heat spreader 19 of the first embodiment.
In the power conversion device 35, the first structure on the first semiconductor element 11 side and the second structure on the second semiconductor element 12 side corresponding to the first structure are opposed to each other with the heat spreader 36 interposed therebetween. What has been described and the description of the configuration and effects of the first structure is similarly applied to the second structure.

By having the said structure, in addition to the effect acquired by the power converter device 10 which concerns on 1st Embodiment, the power converter device 35 can also acquire the following effects.
As a first effect, in the manufacturing process of the power conversion device 35, it is easy to align the heat spreader 36 with the upper surface of the semiconductor element (the first semiconductor element 11 and the second semiconductor element 12), that is, the side surface of the heat spreader 36. .

In other words, when the respective recesses 19a (not shown) formed on the upper and lower surfaces of the heat spreader 36 are fitted into the first wire bonding line 24 and the second wire bonding line 31, the fitted part is not necessarily visually observed. The shape of the heat spreader 36 is not necessarily a shape that can be formed, but by aligning the protruding wall portion 37 of the heat spreader 36 with each end of the first semiconductor element 11 and the second semiconductor element 12, the first wire bonding is performed on one recess 19 a of the heat spreader 36. The second wire bonding line 31 can be easily fitted into the wire 24 and the other recess (not shown).
Here, it is possible to easily form the heat spreader 36 by determining the dimensions and intervals of the protruding wall portions 37 and the recessed portions 19 a based on the shapes of the first semiconductor element 11 and the second semiconductor element 12.

If an insulating material is used as the high thermal conductivity adhesive 25, the heat spreader 41, the end portions of the first semiconductor element 11 and the second semiconductor element 12, and the first semiconductor element 11 and the second semiconductor element are used. 12 can also be easily electrically insulated from the first substrate 13 and the second substrate 14 on which 12 is mounted.
As a second effect, the first semiconductor element 11 is mounted on the first semiconductor element 11 through the protruding wall portion 37 of the heat spreader 36 from the upper surface electrodes of the first semiconductor element 11 and the second semiconductor element 12. Heat can be radiated to the radiator 17.

That is, the first radiator 17 actually contributes to cooling mainly in the lower part of the first semiconductor element 11, that is, the first substrate 13 side part. With this configuration, the peripheral portion of the first semiconductor element 11, which has conventionally not greatly contributed to the cooling by the first radiator 17, is also provided with the protruding wall portion 37 around the peripheral portion. Accordingly, the first radiator 17 can be used for cooling the peripheral portion of the first semiconductor element 11. Furthermore, for the convenience of the structure of the power conversion device, even when the second radiator 18 is not used, only the first radiator 17 is used and the upper surface electrodes of the first semiconductor element 11 and the second semiconductor element 12 are used. Can also dissipate heat. The same applies to the side where the second radiator 18 is disposed.
Thereby, while being able to form the power converter mentioned above more easily, cooling performance can be improved further.
(Third embodiment)

FIG. 7 is a cross-sectional explanatory view showing the configuration of the power converter according to the third embodiment of the present invention. FIG. 8 is a cross-sectional explanatory view taken along the line AA of FIG. FIG. 9 is an explanatory side view as seen from X in FIG. FIG. 10 is a cross-sectional explanatory view taken along line BB in FIG. FIG. 11 is a cross-sectional explanatory view taken along the line CC of FIG.
As shown in FIGS. 7 to 11, the power converter 40 includes a heat spreader 41, a wiring pattern 42, and a gate wiring connection terminal 43 (see FIGS. 7, 9, and 10). The heat spreader 41 has the same configuration (see FIG. 11) as the heat spreader 36 in which the protruding wall portion 37 is formed (see FIG. 11), and also has a through electrode 41a (see FIGS. 8 to 10) penetrating the upper and lower end faces of the heat spreader 41. doing.

The control signal line wiring pattern 42 provided on the main surface of the heat spreader 41 on one side (the first semiconductor element 11 side) is electrically connected to the control terminal 27 of the first semiconductor element 11 via the gate electrode connection line 42a. The gate wiring connection terminal 43 provided in the vicinity of the end of the heat spreader 41 is electrically connected to the wiring pattern 42 through the through electrode 41a (see FIGS. 9 and 10). Note that the through electrode 41 a is formed with an insulating member around the through electrode 41 a so as to be insulated from the heat spreader 41. Further, the gate wiring connection terminal 43 and the wiring pattern 42 may be directly connected without using the through electrode 41a.
Other configurations and operations are the same as those of the power conversion device 35 (see FIG. 6) of the second embodiment.

In the power conversion device 40, the first structure on the first semiconductor element 11 side and the second structure on the second semiconductor element 12 side corresponding to the first structure are opposed to each other with the heat spreader 41 interposed therebetween. What has been described and the description of the configuration and effects of the first structure is similarly applied to the second structure.
By having the said structure, in addition to the effect acquired by the power converter device 35 which concerns on 2nd Embodiment, the power converter device 40 can also acquire the following effects.

As a first effect, a space related to the connection of the control terminal (for example, the gate electrode) 27 can be remarkably reduced. When the gate wiring substrate 20 (see FIG. 1) is disposed beside the first semiconductor element 11, the gate connection line 26 is connected to the gate wiring substrate 20, so that the size of the gate wiring substrate 20 itself, the gate wiring, The power converter has been increased in size by a space for preventing the substrate 20 and the semiconductor element from interfering with each other.
In this configuration, since the gate wiring substrate 20 can be arranged on the upper side of the first semiconductor element 11 by using the heat spreader 41, it is necessary to prepare for arranging the gate wiring substrate 20 in terms of the projected area. Most of the necessary space can be eliminated. Thereby, a power converter device can be further reduced in size.

  As a second effect, the electrical operation of the power conversion device 40 can be further stabilized. Taking an IGBT as an example of the first semiconductor element 11 having the control terminal 27, an electric on / off operation is generally performed by determining a gate potential with respect to an emitter potential. This emitter potential is the potential of the first upper surface electrode (emitter electrode) 23. In this configuration, the heat spreader 41 can be regarded as contacting a part of the first wire bonding line 24 even if a material having an insulating property is used as the high thermal conductivity adhesive 25. Therefore, the gate wiring substrate 20 can be disposed on the heat spreader 41 having the same potential as the emitter potential.

  As described above, since the heat spreader 41 functions as a shield material, the gate potential applied to a semiconductor element such as an IGBT can be reduced by preventing adverse effects on the gate potential due to noise and the like, and allowing the gate wiring to be capacitively coupled to a metal body having an emitter potential. Furthermore, it can be stabilized.

As a third effect, moisture or the like can be easily prevented from entering the upper surface of the first semiconductor element 11. Conventionally, in order to prevent intrusion of moisture or the like, a gel agent or the like is poured so as to cover the entire upper surface side of the first semiconductor element 11 and is thermally cured. For this reason, the manufacturing process increased and the increase in manufacturing cost was inevitable. In this configuration, the upper surface side of the first semiconductor element 11 is bonded to the first upper surface electrode 23, the control terminal 27, the connection line (gate electrode connection line 42 a and the wiring pattern 42) of the control terminal 27, and the circuit board portion. Can be easily covered with an agent. Therefore, entry of moisture and the like can be easily prevented without increasing the number of manufacturing steps.
(Fourth embodiment)

FIG. 12 is a cross-sectional explanatory view showing the configuration of the power converter according to the fourth embodiment of the present invention. FIG. 13 is a cross-sectional explanatory view taken along line BB in FIG. FIG. 14 is a cross-sectional explanatory view taken along the line CC of FIG.
As shown in FIGS. 12 to 14, the power conversion device 45 includes a heat spreader 46 similar to the heat spreader 41 of the power conversion device 40 according to the third embodiment (see FIG. 4). There are four semiconductor elements: a first semiconductor element 47 and a third semiconductor element 49 similar to the first semiconductor element 11, and a second semiconductor element 48 and a fourth semiconductor element 50 similar to the second semiconductor element 12. (See FIGS. 12 and 13).

  A fourth semiconductor element 50 is mounted on the first radiator 17 side in addition to the first semiconductor element 47, and a third semiconductor element 49 is mounted on the second radiator 18 side in addition to the second semiconductor element 48, respectively. Has been. In other words, the second semiconductor element 48 that does not perform the same electrical operation as the first semiconductor element 47 at the position of the second substrate 14 facing the first semiconductor element 47 is the first substrate facing the third semiconductor element 49. The fourth semiconductor element 50 that does not perform the same electrical operation as the third semiconductor element 49 is disposed at the position 13.

In addition, the power conversion device 45 has a first wiring pattern 51 similar to the wiring pattern 42 together with the heat spreader 46 on one surface side of the heat spreader 46 and a second wiring pattern 52 on the other surface side of the heat spreader 46. . The control terminal 47a of the first semiconductor element 47 is connected to the first wiring pattern 51 via the first gate electrode connection line 51a, and the control terminal 49a of the third semiconductor element 49 is connected via the second gate electrode connection line 52a. The second wiring patterns 52 are electrically connected to each other.
Further, the first wiring pattern 51 and the second wiring pattern 52 are connected to a gate wiring connection terminal (not shown) provided in the vicinity of the end of the heat spreader 46 through the through electrodes 46a that penetrate the upper and lower end faces of the heat spreader 46. Connecting.

Other configurations and operations are the same as those of the power conversion device 40 (see FIGS. 7 to 11) according to the third embodiment.
In the power conversion device 45, the first structure on the first semiconductor element 11 side and the second structure on the second semiconductor element 12 side corresponding to the first structure are opposed to each other with the heat spreader 46 interposed therebetween. What has been described and the description of the configuration and effects of the first structure is similarly applied to the second structure.
By having the said structure, in addition to the effect acquired by the power converter device 40 which concerns on 3rd Embodiment, the power converter device 45 can also acquire the following effects.

In the power conversion device 45 according to this embodiment, a plurality of IGBTs and FRDs are connected in parallel as semiconductor elements in order to increase the capacity, and the first semiconductor element 47 and the third semiconductor element 49 are connected. The IGBT, the second semiconductor element 48 and the fourth semiconductor element 50 are FRD.
As a first effect, the cooling performance can be further improved. The plurality of IGBTs 47 and 49 operate in the same electrical manner and generate heat at the same time. Therefore, if another IGBT is arranged immediately next to the IGBT, the mutual heat flow interferes, and the element temperature is further increased. The same applies to FRD. In this configuration, even when a plurality of IGBTs and FRDs are arranged in parallel, the FRDs can be arranged close to the side of the IGBT without unnecessarily increasing the element spacing. It is possible to prevent the elements from interfering with each other and further improve the cooling performance of the semiconductor element.

  As a second effect, the mounting projected area of the semiconductor element can be reduced. In general, in the case of the same current capacity, the IGBT is larger than the FRD. Therefore, if a plurality of IGBTs and FRDs are arranged on the same plane, a wasteful space is likely to be generated due to the difference in the sizes of the IGBTs and FRDs. In this configuration, an IGBT is disposed as the first semiconductor element 47, and an FRD is disposed as the second semiconductor element 48 with the heat spreader 46 interposed therebetween on the upper surface side. Then, an IGBT is mounted and arranged beside the FRD and the third semiconductor element 49, and an FRD beside the first semiconductor element 47 and the fourth semiconductor element 50.

Due to the above-described first effect, the second semiconductor element 48 and the third semiconductor element 49 do not increase the element temperature even if they are arranged close to each other, and the first semiconductor element 47 and the fourth semiconductor element 50 are similarly arranged close to each other. However, the element temperature is not increased. As a result, by arranging the semiconductor elements close to each other, an end portion of the IGBT as the third semiconductor element 49 is formed on a part of the upper surface side of the IGBT as the first semiconductor element 47 while suppressing the temperature rise of the semiconductor element. It can be close to such a degree, and can be easily arranged in high density.
As a result, it is possible to easily arrange a high-density arrangement in which the end portions of the respective semiconductor elements overlap each other when viewed from the projected area while suppressing the temperature rise of the semiconductor elements, thereby further reducing the size of the power conversion device. be able to.
(Fifth embodiment)

FIG. 15 is a cross-sectional explanatory view showing the configuration of the power converter according to the fifth embodiment of the present invention. FIG. 16 is an explanatory plan view of the main surface shape of the heat spreader of FIG.
As shown in FIGS. 15 and 16, the power converter 55 includes a heat spreader 56 in which wire bonding lines are further connected on the main surface instead of the heat spreader 46 of the power converter 45 according to the fourth embodiment. Yes. The heat spreader 56 connects the third wire bonding line 57 and the fourth wire bonding line 58 in addition to the first wire bonding line 24 and the second wire bonding line 31 on the main surface, so that the protrusions and the recesses are also formed. It has a heat spreader shape having

That is, the third wire bonding line 57 is arranged on one side (first substrate 13 side) of the main surface of the heat spreader 56 so as to be substantially parallel to the first wire bonding line 24 (see FIG. 15), and the upper surface electrode of the semiconductor element. A plurality of them are provided so as to cover a region where the layers are joined. Further, a fourth wire bonding wire 58 is provided on the other surface side (second substrate 14 side) of the main surface of the heat spreader 56 substantially in parallel with the second wire bonding wire 31 (see FIG. 15), and the upper surface of the semiconductor element. A plurality of electrodes are provided so as to cover the region S to which the electrodes are bonded (see FIG. 16).
Thereby, the 3rd wire bonding line 57 and the 4th wire bonding line 58 become the shape similar to the protrusion of the heat spreader 46 shown in 4th Embodiment. Further, the spaces between the third wire bonding lines 57 and the fourth wire bonding lines 58 have the same shape as the grooves of the heat spreader 46 shown in the fourth embodiment.

Other configurations and operations are the same as those in the fourth embodiment.
In the power conversion device 55, the first structure on the first semiconductor element 11 side and the second structure on the second semiconductor element 12 side corresponding to the first structure are opposed to each other with the heat spreader 56 interposed therebetween. What has been described and the description of the configuration and effects of the first structure is similarly applied to the second structure.
In the above description, the power conversion device 45 according to the fourth embodiment has been described as the configuration in which the third wire bonding line 57 and the fourth wire bonding line 58 are added, but the first to third embodiments are described. Similarly, the third wire bonding line 57 and the fourth wire bonding line 58 may be added to the power converter according to the embodiment.

By having the said structure, in addition to the effect acquired by the power converter device 45 which concerns on 4th Embodiment, the power converter device 55 can also acquire the following effects.
A heat spreader having recesses and protrusions on one side and the other side of the main surface can be further easily formed. Thereby, the manufacturing cost of a power converter device can further be reduced. That is, the third wire bonding line 57 and the fourth wire bonding line 58 can contribute to heat conduction through the upper surface electrode of the semiconductor element as the protrusions of the heat spreaders of the first to fourth embodiments.

Further, the heat spreader 56 portion between the third wire bonding line 57 and the fourth wire bonding line 58 is the same as the first wire bonding line 24 and the recess of each heat spreader of the first to fourth embodiments, that is, the groove part. The heat conduction through the second wire bonding line 31 can be contributed.
Therefore, the heat spreader 56 does not need to be subjected to fine uneven processing, and the same effect can be obtained by wire bonding line connection which is a conventional general technique. As a result, the manufacturing cost of the power conversion device can be further reduced.
(Sixth embodiment)

FIG. 17 is a cross-sectional explanatory view showing the configuration of the power converter according to the sixth embodiment of the present invention. As illustrated in FIG. 17, the power conversion device 60 includes two heat spreaders, a first heat spreader 61 and a second heat spreader 62, instead of the heat spreader 46 of the power conversion device 45 according to the fourth embodiment.
That is, the first heat spreader 61 has a groove-like shape that allows the first semiconductor element 47 and the fourth semiconductor element 50 to enter one side and the second semiconductor element 48 and the third semiconductor element 49 to enter the other side, respectively. The second heat spreader 62 has a fin portion 62a having juxtaposed wall-like portions protruding substantially orthogonally from one surface of the flat plate-like base portion.

The second heat spreader 62 places the flat base on the upper surface electrodes of the first semiconductor element 47, the second semiconductor element 48, the third semiconductor element 49, and the fourth semiconductor element 50, and performs first wire bonding. The wire 24 and the second wire bonding line 31 are mounted so as to open at positions corresponding to the connecting portions, and the tips of the fin portions 62 a are connected to the recesses 61 a of the first heat spreader 61. The flat base of the second heat spreader 62 is bonded to the upper surface electrodes of the first semiconductor element 47, the second semiconductor element 48, the third semiconductor element 49, and the fourth semiconductor element 50 with a high thermal conductivity adhesive or the like. Also good.
Other configurations and operations are the same as those of the power converter 45 according to the fourth embodiment.

In the above description, the first heat spreader 61 and the second heat spreader 62 are described in place of the heat spreader 46 of the power converter 45 according to the fourth embodiment, but the first to third embodiments are described. Similarly, the power converter according to the embodiment may be configured to include the first heat spreader 61 and the second heat spreader 62 instead of the heat spreader.
By having the said structure, in addition to the effect obtained by the power converter device 45 which concerns on 4th Embodiment, the power converter device 60 can also acquire the following effects.
As a first effect, it is possible to further reduce the thermal resistance related to heat radiation from the upper surface electrode side, and to mount the heat spreader (61, 62) on the upper surface electrode of the semiconductor element (47, 48, 49, 50). It becomes easier.

  This reduction in thermal resistance is due to the following reason. According to this configuration, the second heat spreader 62 is previously applied to the upper surface electrodes of the first semiconductor element 47, the second semiconductor element 48, the third semiconductor element 49, and the fourth semiconductor element 50 with a high thermal conductivity adhesive or the like. After bonding, the wire bonding lines (24, 31) can be connected. Therefore, since the second heat spreader 62 made of metal can be disposed also below the wire bonding lines (24, 31), the lower side of the first wire bonding line 24 and the second wire bonding line 31 is also highly heated. A second heat spreader 62 having a lower thermal resistance than that of the conductive adhesive can be provided. For this reason, thermal resistance can be further reduced.

As a second effect, the mounting operation becomes easy. This is due to the following reason. The positioning of the first heat spreader 61 after mounting the second heat spreader 62 on each upper surface electrode of the first semiconductor element 47, the second semiconductor element 48, the third semiconductor element 49, and the fourth semiconductor element 50 is as follows. What is necessary is just to fit the front-end | tip part of the fin part 62a of the 2nd heat spreader 62 in the groove-shaped recessed part 61a of the 1 heat spreader 61 simply. Accordingly, the heat spreader can be easily aligned and joined. Note that a high thermal conductivity adhesive may be used for this bonding.
Further, since the possibility of the heat spreader coming into contact with the first wire bonding line 24 and the second wire bonding line 31 itself can be eliminated, the first wire bonding line 24 and the second wire bonding line 31 in the mounting operation for forming this configuration can be eliminated. There is no risk of any damage to the two-wire bonding line 31.

In each of the first to sixth embodiments described above, the second semiconductor element 12 and the second semiconductor element 12 are arranged on the other surface side of the heat spreader in which the first semiconductor element 11 and the first radiator 17 are arranged on the main surface side. It demonstrated as a structure which arrange | positions the 2nd heat radiator 18. FIG. However, the arrangement is not necessarily limited to this.
FIG. 18 is a cross-sectional explanatory view showing another example of the arrangement configuration of the semiconductor element and the heat radiator. As shown in FIG. 18, the first semiconductor element 11 (47) and the fourth semiconductor element 50 are arranged on the first heat radiator 17, and the heat spreader 65 is arranged on the upper surface electrode, that is, the heat spreader. Even in the configuration in which the semiconductor element and the radiator are arranged only on one surface side of the heat spreader, instead of arranging the semiconductor element and the radiator on both sides thereof, the effects described in the above embodiments are as follows. The same can be obtained for the first semiconductor element 11 (47) and the fourth semiconductor element 50.

  As described above, the power conversion device according to the present invention includes a first radiator, a first substrate mounted on the main surface of the first radiator, and a first semiconductor mounted on the main surface of the first substrate. An element, a first electrode mounted on a main surface of the first radiator or the main surface of the first substrate, and a first upper surface electrode of the first semiconductor element disposed on an upper surface of the first semiconductor element. A plurality of first wire bonding lines that electrically connect the first electrode, a heat spreader that is in contact with the first upper surface electrode, and the first wire bonding line provided on one surface side of the heat spreader. And a convex portion that contacts or is close to the first upper surface electrode, a concave portion that is close to the first wire bonding line, and a second radiator mounted on the other surface side of the heat spreader.

In the present invention, the second substrate mounted on the main surface of the second radiator, the second semiconductor element mounted on the main surface of the second substrate, and the main surface of the second radiator or A second electrode mounted on a main surface of the second substrate; a second upper surface electrode of the second semiconductor element; and a plurality of second wire bonding lines that electrically connect the second electrode; A convex portion that is provided on the other surface side of the heat spreader and that is in contact with or close to the second upper surface electrode through the second wire bonding line, and a concave portion that is close to the second wire bonding line, Preferably, the first electrode and the second electrode are electrically connected, and the upper surface electrode of the first semiconductor element and the upper surface electrode of the second semiconductor element are electrically at the same potential.
In the present invention, it is preferable that the first semiconductor element and the second semiconductor element are a combination of elements that do not perform the same electrical operation at the same time.

Moreover, in this invention, it is preferable that the part corresponding to the bonding connection position of the said wire bonding and the said upper surface electrode of the recessed part provided in the said heat spreader is formed shallower than the part which does not correspond to a bonding connection position.
Moreover, in this invention, it is preferable that the edge part of the said heat spreader protrudes in the heat spreader upper surface side, and is located in the side of the said semiconductor element.
Also, in the present invention, the wiring pattern provided on the main surface which is one side of the heat spreader and the control terminal of each semiconductor element is electrically connected by a gate electrode connection line, and in the vicinity of the end of the heat spreader. It is preferable to electrically connect the provided gate wiring connection terminal and the wiring pattern.

  In the present invention, a first semiconductor element having a control terminal mounted on the first substrate, a third semiconductor element having a control terminal mounted on the second substrate, and one surface side of the heat spreader A first gate electrode connection line that electrically connects the first wiring pattern provided and the control terminal of the first semiconductor element; and a second wiring pattern provided on the other surface side of the heat spreader and the third semiconductor element. A second gate electrode connection line for electrically connecting the control terminal, and the first wiring pattern and the second wiring pattern provided in the heat spreader are connected to a gate wiring connection terminal provided in the vicinity of the end of the heat spreader. Do not simultaneously perform the same electrical operation as the first semiconductor element mounted on the second substrate at a position facing the through electrode and the first semiconductor element. And second semiconductor elements, mounted on the first substrate at a position opposite to the second semiconductor device, it is preferable that a fourth semiconductor device not to perform the second semiconductor element identical electrical behavior and at the same time.

In the present invention, the first wire bonding line is arranged on the one surface side of the heat spreader along the wiring direction of the first wire bonding line, and the second wire is used as the convex portion, and the second surface side of the heat spreader is the second surface of the heat spreader. It is preferable to have the 4th wire bonding which arrange | positions along the wiring direction of a wire bonding line, and is the said convex part.
In the present invention, the heat spreader includes a first heat spreader having recesses on one surface side and the other surface side, and a second heat spreader having a fin portion on a flat plate-like base, and the flat plate-like base of the second heat spreader However, it is preferable that the semiconductor device is mounted on the upper surface electrode of each semiconductor element in an opening shape corresponding to the wire bonding connection portion, and the tip of the fin portion is connected to the concave portion of the first heat spreader.

It is a section explanatory view showing the composition of the power converter concerning a 1st embodiment of this invention. FIG. 2 is a cross-sectional explanatory view taken along line AA in FIG. 1. It is a plane explanatory view on the back side of the heat spreader of FIG. FIG. 3 is an explanatory plan view excluding a heat spreader in FIG. 2. 4A and 4B are cross-sectional explanatory views taken along line BB, FIG. 4B is a cross-sectional explanatory view taken along line CC, and FIG. 4C is a cross-sectional explanatory view taken along line DD. (D) is sectional explanatory drawing which follows the EE line. FIG. 5A is a cross-sectional explanatory view taken along the line BB, and FIG. 5B is a cross-sectional view taken along the line CC, similar to FIG. 5 showing the cross-sectional structure of each part of the power conversion device according to the second embodiment of the present invention. Explanatory drawing, (c) is sectional explanatory drawing which follows the DD line, (d) is sectional explanatory drawing which follows the EE line. It is a section explanatory view showing the composition of the power converter concerning a 3rd embodiment of this invention. FIG. 8 is a cross-sectional explanatory view taken along line AA in FIG. 7. It is side surface explanatory drawing seen from X of FIG. FIG. 8 is a cross-sectional explanatory view taken along line BB in FIG. 7. FIG. 8 is an explanatory cross-sectional view taken along the line CC in FIG. 7. It is a section explanatory view showing the composition of the power converter concerning a 4th embodiment of this invention. It is sectional explanatory drawing which follows the BB line of FIG. FIG. 13 is an explanatory cross-sectional view taken along the line CC in FIG. 12. It is sectional explanatory drawing which shows the structure of the power converter device which concerns on 5th Embodiment of this invention. FIG. 16 is an explanatory plan view of the main surface shape of the heat spreader of FIG. 15. It is sectional explanatory drawing which shows the structure of the power converter device which concerns on 6th Embodiment of this invention. It is sectional explanatory drawing which shows the other example of the arrangement configuration of a semiconductor element and a heat radiator. It is sectional explanatory drawing which shows the structure of the conventional semiconductor device typically.

Explanation of symbols

10, 35, 40, 45, 55, 60 Power converter 11, 47 First semiconductor element 12, 48 Second semiconductor element 13 First substrate 14 Second substrate 15 First electrode 16 Second electrode 17 First radiator 18 Second heat radiator 19, 36, 41, 46, 56, 65 Heat spreader 19a, 61a Concave portion 19b Convex portion 20 Circuit board 21, 29 Insulating layer 22, 30 Solder layer 23 First upper surface electrode 24 First wire bonding line 25, 32 High thermal conductivity adhesive 26 Connection line 27, 47a, 49a Control terminal 31 Second wire bonding line 33 Switch circuit 37 Projecting wall part 37a Wiring port 41a, 46a Through electrode 42 Wiring pattern 42a Gate electrode connecting line 43 Gate wiring connecting terminal 49 Third semiconductor element 50 Fourth semiconductor element 51 First wiring pattern 51a First gate Electrode connecting line 52 second wiring pattern 52a second gate electrode connecting line 57 third wire bonding line 58 fourth wire bonding line 61 first heat spreader 62 second heat spreader 62a fins

Claims (9)

A first radiator;
A first substrate mounted on a main surface of the first radiator;
A first semiconductor element mounted on a main surface of the first substrate;
A first electrode mounted on the main surface of the first radiator or the main surface of the first substrate;
A plurality of first wire bonding lines disposed on an upper surface of the first semiconductor element and electrically connecting the first upper surface electrode of the first semiconductor element and the first electrode, and the first upper surface electrode are brought into contact with each other. A heat spreader,
A convex portion provided on one surface side of the heat spreader, in contact with or close to the first upper surface electrode through the first wire bonding line, and a concave portion adjacent to the first wire bonding line;
And a second radiator mounted on the other surface side of the heat spreader. A second substrate mounted on the main surface of the second radiator;
A second semiconductor element mounted on the main surface of the second substrate;
A second electrode mounted on the main surface of the second radiator or the main surface of the second substrate;
A second wire bonding line provided on the other surface side of the heat spreader, wherein a plurality of second wire bonding lines electrically connecting the second upper surface electrode of the second semiconductor element and the second electrode are brought into contact with each other. A convex portion that is in contact with or close to the second upper surface electrode, and a concave portion that is close to the second wire bonding line,
The first electrode and the second electrode are electrically connected, and the upper surface electrode of the first semiconductor element and the upper surface electrode of the second semiconductor element are electrically at the same potential. The power converter described.   The power conversion apparatus according to claim 2, wherein the first semiconductor element and the second semiconductor element are a combination of elements that do not perform the same electrical operation at the same time.   The portion corresponding to the bonding connection position of the wire bonding and the upper surface electrode of the recess provided in the heat spreader is formed shallower than the portion not corresponding to the bonding connection position. The power converter device as described in any one.   5. The power conversion device according to claim 1, wherein an end portion of the heat spreader protrudes toward an upper surface of the heat spreader and is located on a side of the semiconductor element.   A gate wiring connection terminal provided in the vicinity of the end of the heat spreader, wherein the control circuit of each semiconductor element and the wiring pattern provided on the main surface on one side of the heat spreader are electrically connected by a gate electrode connection line. The power converter according to claim 1, wherein the wiring pattern is electrically connected. A first semiconductor element having a control terminal mounted on the first substrate;
A third semiconductor element having a control terminal mounted on the second substrate;
A first wiring pattern provided on one side of the heat spreader, a first gate electrode connection line for electrically connecting the control terminals of the first semiconductor element, and a second wiring pattern provided on the other side of the heat spreader; A second gate electrode connection line for electrically connecting the control terminal of the third semiconductor element;
A through electrode connected to the gate wiring connection terminal provided in the vicinity of the end of the heat spreader, the first wiring pattern and the second wiring pattern provided in the heat spreader;
A second semiconductor element that is mounted on the second substrate at a position facing the first semiconductor element and does not simultaneously perform the same electrical operation as the first semiconductor element;
7. A fourth semiconductor element mounted on the first substrate at a position facing the second semiconductor element and not performing the same electrical operation as the second semiconductor element at the same time. 8. The power converter described. A third wire bonding disposed on one surface side of the heat spreader along a wiring direction of the first wire bonding line and serving as the convex portion;
It has 4th wire bonding which is arranged along the wiring direction of said 2nd wire bonding line on the other surface side of said heat spreader, and is said convex part. The power converter described. The heat spreader includes a first heat spreader having a concave portion on one side and the other side, and a second heat spreader having a fin portion on a flat plate-like base,
The flat base portion of the second heat spreader is mounted on the upper surface electrode of each semiconductor element in an opening shape corresponding to the wire bonding connection portion, and the tip of the fin portion is connected to the concave portion of the first heat spreader. The power conversion device according to any one of claims 1 to 7, wherein:

Images