SEMICONDUCTOR COMPONENT AND POWER CONVERTER ARRANGEMENT

Abstract

A semiconductor component comprising: a power semiconductor element (1); a collector substrate (11) electrically connected to the power semiconductor element (1); a collector main terminal (2) electrically connected to the collector substrate (11); an emitter main terminal (3) connected to an emitter electrode surface (12) of the power semiconductor element (1) via a plurality of first bonding wires (15); an emitter reference terminal (4) connected to the emitter electrode surface (12) of the power semiconductor element (1) via a second bonding wire (16); a first voltage measuring circuit (5) for measuring a first voltage, which is a difference between a potential at the collector main terminal (2) and a potential at the emitter main terminal (3);a second voltage measuring circuit (6) for measuring a second voltage, which is a difference between a potential at the emitter reference terminal (4) and a potential at the emitter main terminal (3); and a deterioration region detection unit (9) for referring to correspondence information defining a deteriorated region of a plurality of bonding regions (20) on the emitter electrode surface (12) to which the first bonding wires (15) are connected, for a combination of a temporal change in the first voltage and a temporal change in the second voltage, and for identifying the deteriorated region corresponding to a combination of a temporal change in the first voltage measured by the first voltage measuring circuit (5) and a temporal change in the second voltage measured by the second voltage measuring circuit (6).

Description

TECHNICAL FIELD

[0001] The present invention relates to a semiconductor device and a power converter device. STATE OF THE ART

[0002] Power modules for use in power converter assemblies and the like have a plurality of bond wires connected to a power semiconductor element.

[0003] For example, Patent Document 1 describes a method for detecting breakage of a plurality of bonding wires. The device in Patent Document 1 measures the electrical characteristics of four first wires by applying a weak current to the four first wires and detects breakage of the four first wires by comparing the measured electrical characteristics with a reference value.

[0004] Patent Document 2 relates to a motor driving device having a function for detecting a failure in a power element.

[0005] Patent Document 3 relates to a semiconductor device with deterioration detection function.

[0006] Patent Document 4 describes a method for monitoring the contact of wire bonds or other electrically conductive connections with power semiconductor devices, in particular in IGBT and MOSFET power semiconductor switches.

[0007] In Non-Patent Literature 1, the peak voltage at the parasitic inductance between the Kelvin and the power emitter in the turn-off transient is proposed as a real-time aging indicator. STATE OF THE ART Patent Document 1: Japanese Patent Application Laid-Open No. JP 2013-206997A Patent document 2: DE 10 2018 001 826 A1 Patent document 3: WO 2018 / 211 735 A1 Patent document 4: DE 103 16 357 A1

[0008] Non-patent literature 1: ZHENG, R. [et al.]: Online Aging Parameter Extraction with Induced Voltage V eEbetween Kelvin and Power Emitter in Turn-off Progress for IGBT Modules. In: 2018 IEEE Energy Conversion Congress and Exposition (ECCE), Portland, OR, USA, 2018, pp. 362-366, - ISSN 2329-3748 SUMMARY OF THE INVENTION Problems to be solved by the invention

[0009] Unfortunately, the technique described in Patent Document 1 cannot detect which of the bonding regions of the bonding wires to a power semiconductor element has deteriorated. Therefore, an object of the present invention is to provide a semiconductor device and a power converter assembly capable of detecting which of the bonding regions of a plurality of bonding wires to a power semiconductor element has deteriorated. Means to solve the problems

[0010] A semiconductor device according to the present invention includes: a power semiconductor element; a collector substrate electrically connected to the power semiconductor element; a collector main terminal electrically connected to the collector substrate; an emitter main terminal connected to an emitter electrode surface of the power semiconductor element through a plurality of first bonding wires; an emitter reference terminal connected to the emitter electrode surface of the power semiconductor element through a second bonding wire; a first voltage measuring circuit for measuring a first voltage, which is a difference between a potential at the collector main terminal and a potential at the emitter main terminal; a second voltage measuring circuit for measuring a second voltage, which is a difference between a potential at the emitter reference terminal and a potential at the emitter main terminal.and a deterioration region detection unit for referring to correspondence information defining a deteriorated region of a plurality of bonding regions to the emitter electrode surface to which the first bonding wires are connected, for a combination of temporal change of the first voltage and temporal change of the second voltage, and for identifying the deteriorated region corresponding to a combination of temporal change of the first voltage measured by the first voltage measuring circuit and temporal change of the second voltage measured by the second voltage measuring circuit.; Effect of the invention

[0011] According to the present invention, for a combination of temporal change of the first voltage and temporal change of the second voltage, it can be detected which of the bonding regions of a plurality of bonding wires to the power semiconductor element has deteriorated by referring to correspondence information defining a deteriorated region of a plurality of bonding regions to the emitter electrode surface to which the first bonding wires are connected. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a diagram showing an overall configuration of a semiconductor device according to a first embodiment; Fig. 2 is a diagram showing a cross section of a power module included in the semiconductor device; Fig. 3 is a perspective view showing a connection state between a power semiconductor element 1 and peripheral elements; Fig. 4 is a picture showing the direction in which the main current flows and part of the size of the main current; Fig. 5 is a diagram for explaining the distance between a deteriorated bonding portion and a connecting portion of a second bonding wire 16; Fig. 6 is a diagram illustrating an example of the correspondence relationship between a first deteriorated region and a combination of temporal change ΔVce of a first voltage Vce and temporal change ΔVee of a second voltage Vee; Fig. 7 is a diagram showing an example of a look-up table for identifying the first deteriorated area; Fig. 8 is a diagram showing a configuration of a semiconductor device according to a second embodiment; Fig. 9 is a diagram showing the correspondence relationship between a second deteriorated region and a combination of temporal change of the first voltage Vce and temporal change of the second voltage Vee in a case where the first deteriorated region is a bonding region 20A; Fig. 10 is a diagram showing a look-up table for identifying the second deteriorated region in a case where the first deteriorated region is a bonding region 20A; Fig. 11 is a diagram showing a configuration of a semiconductor device according to a third embodiment; Fig. 12 is a diagram illustrating the configuration of a semiconductor device according to a fourth embodiment; Fig. 13 is a diagram showing the relationship between collector current Ic and first voltage Vce in an IGBT; Fig. 14 is a diagram for explaining the correction of the first voltage Vce at a time (n-1) before the current time n; Fig. 15 is a diagram for explaining the correction of the first voltage Vce at the current time n; Fig. 16 is a diagram showing the relationship between the collector current Ic and the second voltage Vee in an IGBT; Fig. 17 is a diagram for explaining the correction of the second voltage Vee at time (n-1); Fig. 18 is a diagram for explaining the correction of the second voltage Vee at time n; Fig. 19 is a diagram showing the relationship between the collector current Ic and the second voltage Vee at three temperatures Tk in the power semiconductor element 1; Fig. 20 is a diagram showing a configuration of a semiconductor device according to a fifth embodiment; Fig. 21 is a flowchart illustrating a specific method for estimating the temperature of the power semiconductor element 1 and identifying a damaged area according to the fifth embodiment; Fig. 22 is a graph illustrating the time change of the standard voltage Vee0; Fig. 23 is a diagram showing a configuration of a semiconductor device according to a sixth embodiment; Fig. 24 is a flowchart showing a specific method for temperature estimation of the power semiconductor element 1 and for identifying a deteriorated portion according to the sixth embodiment; Fig. 25 is a block diagram showing a configuration of a power converter system according to a seventh embodiment; Fig. 26 is a diagram illustrating a specific configuration of a deterioration monitoring unit 205 according to the first to sixth embodiments; Fig. 27 is a diagram illustrating an example in which a processing circuit 1001 is implemented by software; Fig. 28 is a diagram illustrating another example of the correspondence relationship between the first deteriorated region and a combination of the time change ΔVce of the first voltage Vce and the time change ΔVce of the second voltage Vee, and Fig. Figure 29 is a diagram showing another example of the lookup table for identifying the first deteriorated area. DESCRIPTION OF THE EMBODIMENTS

[0012] Embodiments of the present invention will be described below with reference to the drawings. Embodiment 1

[0013] Fig. 1 is a diagram showing an overall configuration of a semiconductor device according to a first embodiment.

[0014] The semiconductor device includes a power semiconductor element 1, a collector main terminal 2, an emitter main terminal 3, an emitter reference terminal 4, and a degradation monitoring unit 205.

[0015] The deterioration monitoring unit 205 includes a first voltage measuring circuit 5, a second voltage measuring circuit 6, a look-up table storage unit 19, a deterioration region detection unit 9, a first output terminal 7, a second output terminal 8, and a third output terminal 10.

[0016] The power semiconductor element 1 is designed, for example, as an insulated-gate bipolar transistor (IGBT). The power semiconductor element 1 is connected to the main collector terminal 2, the main emitter terminal 3, and the emitter reference terminal 4.

[0017] The first voltage measuring circuit 5 is connected between the main collector terminal 2 and the main emitter terminal 3. The first voltage measuring circuit 5 outputs information representing a first voltage Vce, which is a difference between a potential at the main collector terminal 2 and a potential at the main emitter terminal 3, to the first output terminal 7.

[0018] The second voltage measuring circuit 6 is connected between the emitter reference terminal 4 and the emitter main terminal 3. The second voltage measuring circuit 6 outputs information representing a second voltage Vee, which is a difference between a potential at the emitter reference terminal 4 and a potential at the emitter main terminal 3, to the second output terminal 8.

[0019] The lookup table storage unit 19 stores a lookup table. The deterioration area detection unit 9 refers to the correspondence information in the lookup table to identify a deteriorated area corresponding to a combination of the temporal change ΔVce of the first voltage measured by the first voltage measuring circuit 5 and the temporal change ΔVee of the second voltage measured by the second voltage measuring circuit 6. The deterioration area detection unit 9 outputs information representing the identified deteriorated area to the third output terminal 10.

[0020] How the deterioration area detection unit 9 identifies a deteriorated area is described below based on the operation of the power semiconductor element 1.

[0021] Fig. 2 is a diagram showing a cross section of a power module included in the semiconductor device.

[0022] The power module includes a power semiconductor element 1, a collector substrate 11, solder 13, a collector main terminal 2, an emitter main terminal 3, an emitter reference terminal 4, an emitter-side main current bonding wire (hereinafter first bonding wire) 15, an emitter reference bonding wire (hereinafter second bonding wire) 16, and a collector-side main current bonding wire (hereinafter third bonding wire) 14.

[0023] The power semiconductor element 1 is bonded to the collector substrate 11 via solder 13. In the power semiconductor element 1, a surface facing the collector substrate 11 functions as the collector, and a surface opposite the surface functions as the emitter. The surface acting as the emitter is an emitter electrode surface 12. A layer of a highly conductive metal, such as aluminum or copper, is formed on one surface of the emitter electrode surface 12.

[0024] The collector main terminal 2, the emitter main terminal 3, and the emitter reference terminal 4 are connected to the power semiconductor element 1 by a plurality of bonding wires 14, 15, and 16. The collector main terminal 2 is connected to the collector substrate 11 via the third bonding wire 14. The emitter main terminal 3 is connected to the emitter electrode surface 12 via the first bonding wire 15. The emitter reference terminal 4 is connected to the emitter electrode surface 12 via the second bonding wire 16.

[0025] The bonding areas are firmly connected to the counter metal by a process that presses the bonding areas of these bond wires 14, 15 and 16 onto the counter metal using ultrasound.

[0026] The lifetime of the power module is generally determined by the lifetime of the bonding areas of these bond wires 14, 15, and 16. Specifically, the first phenomenon that occurs is the deterioration of the bonding areas of the bond wires 14, 15, and 16 with the counterpart metal, and the bonding areas become separated. The lifetime of the power semiconductor element 1 itself, the lifetime dependent on the material of the bond wires 14, 15, and 16, and the lifetime dependent on the metal material constituting the collector main terminal 2, the emitter main terminal 3, and the emitter reference terminal 4 are far longer than the lifetime of the connection areas of the bond wires 14, 15, and 16 with the counterpart metal.

[0027] The lifetime of the solder 13 connecting the power semiconductor element 1 and the collector substrate 11 is shorter than the lifetime depending on the materials described above, but longer than the lifetime of the connection areas of the bond wires 14, 15, and 16 with the counterpart metal. The reason for this is as follows.

[0028] Fig. Figure 2 shows a path 17 of the main current flowing from the main collector terminal 2 to the main emitter terminal 3. The main current passes through the following areas in this order: main collector terminal 2, third bond wire 14, collector substrate 11, solder 13, power semiconductor element 1, first bond wire 15, and main emitter terminal 3.

[0029] In the main current path 17, the power semiconductor element 1 is the part that generates heat. Unlike the other elements, the power semiconductor element 1 has a resistance value that changes with the gate control for turning the main current on and off. Since the resistance of the power semiconductor element 1 is low, the power semiconductor element 1 has the following characteristics. When the power semiconductor element 1 interrupts the current, a high voltage is applied to the power semiconductor element 1, but no current flows through the power semiconductor element 1. When current flows through the power semiconductor element 1, a large current flows through the power semiconductor element 1, but there is almost no voltage across the power semiconductor element 1.

[0030] That is, the power semiconductor element 1 consumes almost no power both at the time of power off and at the time of power on. However, since both current and voltage exist at the time of switching, the amount of heating energy of the power semiconductor element 1 at the time of switching is large. This heating only occurs at the time of switching, and the heating is less when the switching frequency is low. However, the power semiconductor element 1 generally switches at a high frequency of more than a few kHz and repeats the instantaneous heating at an extremely high frequency. Fig. 2 shows the heater 18.

[0031] The power semiconductor element 1 repeatedly heats up at a high frequency, repeatedly subjecting the components connected to the power semiconductor element 1 to thermal stress. As a result, the connected components are damaged. The collector substrate 11 is connected to the collector side of the power semiconductor element 1 via the solder 13, and the first bonding wire 15 is connected to a surface of the emitter electrode surface 12, which is the emitter side of the power semiconductor element 1. This exerts thermal stress on the solder 13 and the first bonding wire 15.

[0032] Fig. 2 shows a region 119 where deterioration may occur due to the heating 18 of the power semiconductor element 1 in the solder 13. A region 20 where deterioration may occur due to the heating 18 is also shown at the connection region of the first bond wire 15 to the emitter electrode surface 12.

[0033] The solder 13 is connected to the entire surface on the collector side of the power semiconductor element 1. On the other hand, the area of the bonding region of the first bonding wire 15 to the emitter electrode surface 12 is equal to the cross-sectional area of the first bonding wire 15 or at most several times larger than the cross-sectional area of the first bonding wire 15. In other words, the current density in the bonding region of the first bonding wire 15 is greater than that in the solder 13. For this reason, and due to the electromigration effect caused by the high current density, the bonding region 20 of the first bonding wire 15 reaches the end of its service life first.

[0034] Therefore, to estimate the lifetime of the semiconductor device, one must estimate the lifetime of the bonding region 20 of the first bonding wire 15 on the emitter electrode surface 12. In the actual semiconductor device, a plurality of first bonding wires 15 are bonded to the power semiconductor element 1. For example, in some cases, more than 10 first bonding wires 15 are bonded to the power semiconductor element 1.

[0035] Fig. Figure 3 is a perspective view showing a connection state between power semiconductor element 1 and peripheral elements. Here, four first bonding wires 15A to 15D are connected to power semiconductor element 1 in bonding regions 20A to 20D.

[0036] If some of a plurality of bonding regions 20A to 20D are damaged, the performance of the semiconductor device deteriorates, but the semiconductor device can still operate. The criterion for determining that the semiconductor device has reached the end of its life when some of the bonding regions 20A to 20D deteriorate depends on the intended use of the semiconductor device. It may be immediately determined that the semiconductor device has reached the end of its life when some of the bonding regions 20A to 20D deteriorate, or it may be determined that the semiconductor device has reached the end of its life when the deterioration progresses to a predetermined number or more of the bonding regions 20A to 20D.

[0037] As in Fig. 3, when a plurality of bonding wires 15A to 15D are arranged, the lifetime of the semiconductor device is to be estimated to determine how long the power module can be used until the final deterioration state by determining the degree of a current deterioration state of the power module with respect to the final deterioration state, which is regarded as the end of the lifetime of the semiconductor device.

[0038] If the accuracy of determining the deterioration state of the power module is high, the power module can be replaced at a convenient time. If the accuracy of determining the deterioration state of the power module is low, the semiconductor device may suddenly reach the end of its service life at an unexpected time, resulting in disruption to the user's operation of the power converter assembly.

[0039] To accurately determine the deterioration state of the power module, the deterioration of the bonding regions 20A to 20D of the first bonding wires 15 must be accurately detected. However, the conventional technique has limitations in the accuracy of deterioration detection. This is because it cannot detect which part of the bonding regions 20A to 20D of the first bonding wires 15A to 15D is damaged. The conventional technique can detect the deterioration of a plurality of bonding wires as a whole, but cannot detect the deterioration of each individual bonding wire.

[0040] When a plurality of first bonding wires 15 are attached, the order of their deterioration is not always fixed. The order of deterioration of a plurality of first bonding wires 15 varies with the usage state, that is, with the driving method of the power semiconductor element 1.

[0041] For example, in a first driving method, the bonding region 20C or the bonding region 20D may be worn first, and after a short time, the bonding region 20D or the bonding region 20C may be worn. In this case, half of the remaining first bonding wires 15 are damaged within a short time after the first deterioration. On the other hand, in a second driving method, the bonding region 20A or the bonding region 20B may be damaged first, and only after a long time may the bonding region 20B or the bonding region 20A be damaged.

[0042] In such a case, the lifetime of the semiconductor device cannot be accurately estimated simply by determining that one of the four first bonding wires 15A to 15D is damaged. Even if it is determined that the lifetime of the semiconductor device ends when half of the first bonding wires are damaged, there is a difference in the duration until the end of the lifetime between the time when the bonding region 20C is worn first and the time when the bonding region 20A is worn first.

[0043] Conventional technology requires a method such as checking the resistance value by applying current to each individual bond wire in some way. This means that deterioration of each individual bond wire cannot be detected without using expensive equipment and a complicated testing procedure.

[0044] The present embodiment can detect which of the bonding regions of a plurality of first bonding wires 15 to the emitter electrode surface 12 of the power semiconductor element 1 has deteriorated without requiring a special device. The principle will now be described in detail.

[0045] The first voltage Vce is the difference between a potential at the collector main terminal 2 and a potential at the emitter main terminal 3. The first voltage Vce is the voltage between the collector main terminal 2 and the emitter main terminal 3, which are originally installed in the power module. Thus, the first voltage Vce can be measured without requiring any special configuration to expand the power module. The first voltage Vce is measured while the power semiconductor element 1 is energized. The power semiconductor element 1 is connected in series with the bonding region 20 of the first bonding wire 15 between the collector main terminal 2 and the emitter main terminal 3. Thus, the magnitude of the first voltage Vce is equal to or greater than the sum of the magnitude of the voltage generated by the resistance of the power semiconductor element 1 and the magnitude of the voltage in the bonding region 20 of the first bonding wire 15.

[0046] The second voltage Vee is the difference between a potential at the emitter reference terminal 4 and a potential at the emitter main terminal 3. The second voltage Vee is the voltage between the emitter main terminal and the emitter reference terminal 4, which are originally installed in the power module. The gate terminal 21 is connected to the gate of the power semiconductor element 1 via a gate bonding wire 22. The emitter reference terminal 4 is used to provide a reference potential that serves as a reference for the voltage applied to the gate terminal 21. This reference potential must be equal to the potential at the emitter electrode surface 12. Therefore, the emitter reference terminal 4 is connected to the emitter electrode surface 12 via a second bonding wire 16.

[0047] The main emitter terminal 3 cannot be used to provide this reference potential. This is because the main current supplying the power semiconductor element 1 causes a voltage drop due to the resistance of the path from the emitter electrode surface 12 to the main emitter terminal 3, and the potential at the main emitter terminal 3 does not accurately reflect the potential at the emitter electrode surface 12. Since the main current does not flow through the emitter reference terminal 4 and the second bonding wire 16, there is almost no difference between the potential at the emitter reference terminal 4 and the potential at the emitter electrode surface 12. The emitter reference terminal 4 can therefore provide a correct reference potential.

[0048] Furthermore, electromigration does not occur because no high-density current flows through the second bonding wire 16. The second bonding wire 16 is typically connected near the edge, where the temperature of the power semiconductor element 1 tends to be low. Therefore, deterioration of the second bonding wire 16 is less likely. Regarding the second bonding wires 16, multiple second bonding wires 16 can be provided to ensure redundancy.

[0049] Since the path of the second voltage Vee does not include the power semiconductor element 1 itself, the resistance of the power semiconductor element 1 has no influence on the magnitude of the second voltage Vee. The second voltage Vee is considered to be the potential difference between the emitter electrode surface 12 and the emitter main terminal 3, but strictly speaking, they are not the same. This will be described later.

[0050] The principle of generating the first voltage Vce will now be described. As described above, the magnitude of the first voltage Vce is equal to or greater than the sum of the magnitude of the voltage generated by the resistance of the power semiconductor element 1 and the magnitude of the voltage in the bonding region 20 of the first bonding wire 15. However, the first voltage Vce is influenced not only by the resistance of the first bonding wire 15, but also by the resistance due to the current flowing through the interior of the emitter electrode surface 12.

[0051] Fig. 4 is a diagram showing the direction in which the main current flows and part of the magnitude of the main current. Since the main current flows from the collector to the emitter, the main current flows from the collector substrate 11 into the power semiconductor element 1, then flows to the emitter electrode surface 12 of the power semiconductor element 1, and continues to the emitter main terminal 3 through the first bonding wires 15A to 15D. At this time, the current flowing through the first bonding wires 15A to 15D is not uniform. For example, part of the current flowing immediately below the bonding region 20A follows a path to the emitter main terminal 3 through the first bonding wire 15A.

[0052] The remaining part of the current flowing immediately below the bonding region 20A flows through the emitter electrode surface 12 to the bonding region 20C and then follows a path to the emitter main terminal 3 through the first bonding wire 15C. On the other hand, the current flowing immediately below the bonding region 20C mainly follows a path to the emitter main terminal 3 through the first bonding wire 15C. This is because the other paths are far away and the resistance values of the other paths are large.

[0053] As a result, for example, even if the bonding region 20A deteriorates and the energization of the first bonding wire 15A stops, the influence is relatively small. Even if no current can flow through the first bonding wire 15A, the current flowing to the bonding region 20A can flow from the bonding region 20C to the emitter main terminal 3 through the first bonding wire 15C with relatively little loss. In comparison, the influence is relatively large when the bonding region 20C deteriorates and the energization of the first bonding wire 15C stops. This is because the current flowing from the bonding region 20C directly to the emitter main terminal 3 through the first bonding wire 15C must flow from another bonding region to the emitter main terminal 3 through a path with a larger resistance.

[0054] In other words, the increase in the first voltage Vce caused when the bonding region 20A of the first bonding wire 15A deteriorates is relatively small. On the other hand, the increase in the first voltage Vce caused when the bonding region 20C of the first bonding wire 15C deteriorates is relatively large. For example, if the bonding region 20A deteriorates first, the first voltage Vce increases by 20%, while if the bonding region 20C deteriorates first, the first voltage Vce increases by 30%.

[0055] However, the deteriorated bonding region of the first bonding wires 15A to 15D cannot be identified solely by the temporal change of the first voltage Vce. This is because, when a large number of first bonding wires 15A to 15D are arranged, other first bonding wires may be in the same condition. For example, if the bonding region 20B deteriorates first, the first voltage Vce may increase by 20%, and if the bonding region 20D deteriorates first, the first voltage Vce may also increase by 30%. Therefore, the damaged bonding region cannot be identified solely by the first voltage Vce.

[0056] The second voltage Vee is now described.

[0057] It is often assumed that the second voltage Vee reflects the resistance values of the first bonding wires 15A to 15D because it is the difference between the potential at the emitter reference terminal 4 and the potential at the emitter main terminal 3, but there is more. The surface of the emitter electrode surface 12 of the power semiconductor element 1 has a resistance component even though it has a metal layer. Since, as described above, current flows through the interior of the emitter electrode surface 12, the second voltage Vee also reflects the potential difference due to the resistance component. From another perspective, the reason why the potential at the emitter reference terminal 4 and the potential at the emitter main terminal 3 are substantially the same is that the emitter reference terminal 4 and the emitter main terminal 3 are coupled by the first bonding wire 15.In other words, without the first bonding wire 15, the voltage between the emitter reference terminal 4 and the emitter main terminal 3 may increase. Therefore, the influence of the deterioration of the first bonding wire 15 on the second voltage Vee depends greatly on how close the bonding region of the second bonding wire 16 is to the deteriorated region of the bonding regions 20A to 25D of the first bonding wire.

[0058] Fig. 5 is a diagram for explaining the distance between a damaged bonding area and a connecting area of the second bonding wire 16. In Fig. 5, the distance difference between the damaged bonding area and the connection area of the second bonding wire 16 is shown by arrows.

[0059] The bonding region 20B of the first bonding wire 15B is close to the connection region of the second bonding wire 16 to the emitter electrode surface 12. The bonding region 20A of the first bonding wire 15A is far from the connection region of the second bonding wire 16 to the emitter electrode surface 12. When the bonding region 20B deteriorates first, the second voltage Vee increases significantly, while the second voltage Vee does not increase significantly when the bonding region 20A deteriorates first. In other words, the increase value of the second voltage Vee varies depending on the deteriorated bonding region of the first bonding wire 15. For example, when the bonding region 20B deteriorates first, the second voltage Vee increases by 40%, while the second voltage Vee increases by only 10% when the bonding region 20A deteriorates first.

[0060] However, if there are a large number of first bonding wires 15, there may be other first bonding wires 15 under the same condition. In the case of Fig. 4, for example, if the bonding area 20D deteriorates first, the second voltage Vee increases by 40%, and if the bonding area 20C deteriorates first, the second voltage Vee also increases by 10%. The deteriorated bonding area cannot be identified solely by the second voltage Vee.

[0061] The present embodiment differs from the conventional art in that the deterioration monitoring unit 205 in the present embodiment considers the characteristics of the power semiconductor element 1 as described above, then measures the first voltage Vce and the second voltage Vee, and identifies a deteriorated portion according to their temporal changes.

[0062] Fig. 6 is a diagram showing an example of the correspondence relationship between the first deteriorated region and a combination of the temporal change ΔVce of the first voltage Vce and the temporal change ΔVce of the second voltage Vee.

[0063] When the first deteriorated area is the bonding area 20A, the first voltage Vce increases by 20% ± ΔV2 and the second voltage Vee increases by 10% ± ΔV1. When the first deteriorated area is the bonding area 20B, the first voltage Vce increases by 20% ± ΔV2 and the second voltage Vee increases by 40% ± ΔV4. When the first deteriorated area is the bonding area 20C, the first voltage Vce increases by 30% ± ΔV3 and the second voltage Vee increases by 10% ± ΔV1. When the first deteriorated area is the bonding area 20D, the first voltage Vce increases by 30% ± ΔV3 and the second voltage Vee increases by 40% ± ΔV4. For example, ΔV1 is 1%, ΔV2 is 2%, ΔV3 is 3% and ΔV4 is 4%.

[0064] Fig. Figure 7 is a diagram showing an example of a lookup table for identifying the first deteriorated area. This lookup table can be derived from the correspondence relationship in Fig. 6 can be derived.

[0065] The lookup table contains correspondence information defining a deteriorated area of a plurality of bonding areas on the emitter electrode surface 12 to which a plurality of first bonding wires 15 are connected for a combination of temporal change ΔVce of the first voltage and temporal change ΔVee of the second voltage.

[0066] As in Fig. As shown in Figure 7, the first deteriorated region is defined as the bonding region 20A when the first voltage Vce increases by 20% ± ΔV2 and the second voltage Vee increases by 10% ± ΔV1. The first deteriorated region is defined as the bonding region 20B when the first voltage Vce increases by 20% ± ΔV2 and the second voltage Vee increases by 40% ± ΔV4. The first deteriorated region is defined as the bonding region 20C when the first voltage Vce increases by 30% ± ΔV3 and the second voltage Vee increases by 10% ± ΔV1. It is defined that the first deteriorated area is the bonding area 20D when the first voltage Vce increases by 30% ± ΔV3 and the second voltage Vee increases by 40% ± ΔV4.

[0067] As indicated in the following equation (1), the deterioration region detection unit 9 calculates the temporal change ΔVce between the first voltage Vce(n) at the current time n and the first voltage Vce(n-1) at the previous time (n-1). As indicated in the following equation (2), the deterioration region detection unit 9 calculates the temporal change ΔVee between the second voltage Vee(n) at the current time n and the second voltage Vee(n-1) at the previous time (n-1). ΔVce=Vce(n)−Vce(n−1) ΔVee=Vee(n)−Vee(n−1)

[0068] Alternatively, as indicated in the following equation (1A), the deterioration region detection unit 9 may calculate the temporal change ΔVce between the first voltage Vce(n) at the current time n and the first voltage Vce(n-1) at the previous time (n-1). As indicated in the following equation (2A), the deterioration region detection unit 9 may calculate the temporal change ΔVee between the second voltage Vee(n) at the current time n and the second voltage Vee(n-1) at the previous time (n-1). ΔVce={Vce(n)−Vce(n−1)} / Vce(n−1) ΔVee={Vee(n)−Vee(n−1)} / Vee(n−1)

[0069] The deterioration region detection unit 9 refers to the lookup table and identifies the deteriorated region corresponding to the combination of the temporal change ΔVce of the first voltage Vce and the temporal change ΔVce of the second voltage Vee. If the combination of the temporal change ΔVce of the first voltage Vce and the temporal change ΔVce of the second voltage Vee is not included in the lookup table, the deterioration region detection unit 9 determines that a plurality of bonding regions of the emitter electrode surface 12 to which a plurality of first bonding wires 15 are connected do not have a deteriorated region.

[0070] As described above, the semiconductor device according to the first embodiment can detect which of the bonding regions of a plurality of first bonding wires 15 to the power semiconductor element 1 has deteriorated by measuring the voltage between two existing terminals and referring to the prestored lookup table, without adding any special structure. This allows the lifetime of the semiconductor element to be accurately estimated. Embodiment 2

[0071] With a large number of first bond wires, the order in which the bond regions of the first bond wires deteriorate may vary depending on the operating conditions and the last state of the power module, i.e., the drive method. In some cases, the next deteriorated region can be accurately estimated based on the history of deteriorated bond regions. A second embodiment uses a lookup table corresponding to the history of deteriorated bond regions.

[0072] Fig. 8 is a diagram showing a configuration of a semiconductor device according to the second embodiment. A semiconductor device according to the second embodiment includes a deterioration region history recording unit 23 in addition to the components of the semiconductor device according to the first embodiment.

[0073] The deterioration area history recording unit 23 records the history of the deteriorated bonding areas of the first bonding wires. If none of several bonding areas deteriorates, the deterioration area detection unit 9 refers to a Fig. 7 and identifies the deteriorated area. When one of a plurality of bond areas deteriorates, the deterioration area detection unit 9 refers to a lookup table corresponding to a deteriorated area and identifies the deteriorated area.

[0074] Fig. 9 is a diagram showing the correspondence relationship between a second deteriorated region and a combination of the temporal change ΔVee of the first voltage Vce and the temporal change of the second voltage Vee in a case where the first deteriorated region is the bonding region 20A.

[0075] When the second deteriorated area is the bonding area 20B, the first voltage Vce increases by 40% ± ΔV4 and the second voltage Vee increases by 80% ± ΔV8. When the second deteriorated area is the bonding area 20C, the first voltage Vce increases by 60% ± ΔV6 and the second voltage Vee increases by 20% ± ΔV2. When the second deteriorated area is the bonding area 20D, the first voltage Vce increases by 60% ± ΔV6 and the second voltage Vee increases by 80% ± ΔV8. For example, ΔV2 is 2%, ΔV4 is 4%, ΔV6 is 6%, and ΔV8 is 8%.

[0076] Fig. Fig. 10 is a diagram showing a look-up table for identifying the second deteriorated region in the case where the first deteriorated region is the bonding region 20A. This look-up table can be obtained from the correspondence relationship in Fig. 9 can be derived.

[0077] The second deteriorated region is defined as the bonding region 20B when the first voltage Vce increases by 40% ± ΔV4 and the second voltage Vee increases by 80% ± ΔV8. The second deteriorated region is defined as the bonding region 20C when the first voltage Vce increases by 60% ± ΔV6 and the second voltage Vee increases by 20% ± ΔV2. The second deteriorated region is defined as the bonding region 20D when the first voltage Vce increases by 60% ± ΔV6 and the second voltage Vee increases by 80% ± ΔV8.

[0078] When the first deteriorated area is the bonding area 20A, the deterioration area detection unit 9 refers to the Fig. 10 and identifies the second deteriorated area corresponding to the measured ΔVce and ΔVee. Similarly, if the first deteriorated areas are the bond areas 20B, 20C, and 20D, the deterioration area recognition unit 9 refers to the lookup tables corresponding to areas 20B, 20C, and 20D and identifies the second deteriorated area corresponding to the measured ΔVce and ΔVee.

[0079] The deterioration area recognition unit 9 can identify the i-th deteriorated area by referring to the i-th deteriorated area identification lookup table corresponding to the first to (i-1)-th deterioration areas.

[0080] As described above, the present embodiment can identify the next deteriorated area in accordance with the history of the deteriorated areas. Embodiment 3

[0081] The semiconductor device according to a third embodiment estimates the lifetime of the semiconductor device and suggests changing the driving method based on the history of the deteriorated bonding regions and the driving method used.

[0082] Fig. 11 is a diagram showing a configuration of a semiconductor device according to the third embodiment. The semiconductor device according to the third embodiment includes a lifetime estimation unit 25 in addition to the components of the semiconductor device according to the second embodiment.

[0083] The history of deteriorated regions output by the deteriorated region history recording unit 23 is sent to the lifetime estimation unit 25. The lifetime estimation unit 25 estimates the lifetime of the semiconductor device using the current driving method based on the history of the deteriorated bonding regions and the driving method of the power semiconductor element 1.

[0084] The control method is represented by a combination of switching states of the power semiconductor element 1. In particular, the control method includes the collector voltage, the collector current, the gate voltage, the gate current, and the switching frequency applied to the power semiconductor element 1.

[0085] As described below, the lifetime of the semiconductor device may vary with the driving method.

[0086] In driving method A, the power efficiency is high, but the electromagnetic noise is also large. When driving method A is used, one of the bonding regions 20C or 20D deteriorates first. Then, if deterioration occurs in one of the bonding regions 20C or 20D, deterioration occurs in the other bonding region 20C or 20D after a short time, and the semiconductor device reaches the end of its life. In driving method B, the power efficiency is somewhat low, but the electromagnetic noise is small. When driving method B is used, one of the bonding regions 20A or 20B deteriorates first. Then, if deterioration occurs in one of the bonding regions 20A or 20B, deterioration occurs in the other bonding region 20A or 20B after a long time, and the semiconductor device reaches the end of its life.

[0087] For example, if "Drive Method A" is used continuously from the beginning and deterioration of the bonding region 20C of the first bonding wire 15C is detected after 10,000 hours, the lifetime estimation unit 25 estimates that, for example, if drive method A is used, half of the bonding regions of the first bonding wire 15 will deteriorate after another 100 hours, and the semiconductor device will then reach the end of its lifetime. Alternatively, if "Drive Method B" is used continuously from the beginning and deterioration of the bonding region 20A of the first bonding wire 15D is detected after 10,000 hours, the lifetime estimation unit 25 estimates that, for example, if drive method B is used, half of the bonding regions of the first bonding wire 15 will deteriorate after another 1,000 hours, and the semiconductor device will then reach the end of its lifetime.

[0088] In this way, the history of degradation areas of a plurality of bond regions is sometimes related to the subsequent degradation rate, and this can be used to improve the accuracy in predicting the lifetime of the semiconductor device.

[0089] The lifetime estimation unit 25 gives the user a hint to recommend switching to a control method that extends the lifetime.

[0090] For example, if the driving method A is used and deterioration of the bonding region 20C or 20D is first detected, the lifetime estimation unit 25 recommends switching to a driving method B in which the time until the end of life is reached is longer.

[0091] The lifetime estimation unit 25 can further perform automatic switching to a driving method that extends the lifetime, together with a control circuit not shown. For example, if driving method A is used and deterioration of the bonding region 20C or 20D is first detected, the lifetime estimation unit 25 instructs the control circuit to switch to driving method B, and the control circuit controls the switching to driving method B.

[0092] As described above, the semiconductor device according to the third embodiment can estimate the lifetime of the semiconductor device and propose a driving method that extends the lifetime based on the history of the deteriorated bonding regions and the driving method of the power semiconductor device. Embodiment 4

[0093] Fig. 12 is a diagram showing a configuration of a semiconductor device according to a fourth embodiment. The semiconductor device according to the fourth embodiment includes a temperature sensor 27 in addition to the components of the semiconductor device according to the first embodiment.

[0094] The temperature sensor 27 measures the temperature Tk of the power semiconductor element 1. The temperature sensor 27 may be a separate element using a thermistor or a thermocouple, or it may be a temperature measuring diode integrated into the power semiconductor element 1.

[0095] Fig. Figure 13 is a diagram showing the relationship between the collector current Ic and the first voltage Vce in the IGBT.

[0096] The first voltage Vce changes with the collector current Ic and the temperature Tk of the power semiconductor element 1.

[0097] Even if the power semiconductor element 1 is energized with the same collector current Ic, the resistance value of the power semiconductor element 1 changes with the temperature Tk of the power semiconductor element 1. As a result, for example, the first voltage Vce generated at 25 °C differs from the value of the first voltage Vce generated at 125 °C.

[0098] Therefore, the value of the first voltage Vce changes with the temperature Tk of the power semiconductor element 1, even if the degree of deterioration of the first bonding wire 15 does not change at all. If this is not taken into account, it may be mistakenly assumed that the first bonding wire 15 is deteriorating, even though it is not deteriorating. As shown in Fig. As shown in Figure 13, an IGBT typically has a collector current Ic_nd at which the first voltage Vce does not change even when the temperature Tk of the power semiconductor element 1 changes within a certain temperature range. When the temperature Tk of the power semiconductor element 1 is between 25°C and 125°C, the first voltage Vce measured with the collector current Ic_nd is a fixed value Vce_nd, and therefore, a change in the temperature Tk of the power semiconductor element 1 can be avoided. However, when the temperature Tk of the power semiconductor element 1 is outside the temperature range (25°C to 125°C), the first voltage Vce measured with the collector current Ic_nd is sometimes not Vce_nd.

[0099] The lookup table described in the first embodiment in Fig. Figure 10 defines the deteriorated area of the bonding areas of the first bonding wires to the power semiconductor element 1, for the change ΔVce of the first voltage and the change ΔVee of the second voltage, where the temperature Tk of the power semiconductor element 1 is the standard temperature (25 °C). Therefore, if the temperature Tk of the power semiconductor element 1 deviates from the standard temperature, the change ΔVce of the first voltage and the change ΔVee of the second voltage must be corrected using the lookup table.

[0100] In the present embodiment, the deterioration region detection unit 9 corrects the temporal change ΔVce of the first voltage Vce depending on the temperature Tk of the power semiconductor element 1.

[0101] Fig. 14 is a diagram for explaining the correction of the first voltage Vce at a time (n-1) before the current time n.

[0102] If the temperature Tk of the power semiconductor element 1 is 125 °C and the first voltage Vce is equal to Vce (n-1, 125 °C) at time (n-1), as in Fig. 14, the deterioration region detection unit 9 determines the first voltage Vce (n-1, 25 °C) when the temperature Tk of the power semiconductor element 1 is 25 °C, under the condition that the collector current Ic is equal.

[0103] Fig. Figure 15 is a diagram explaining the correction of the first voltage Vce at the current time n.

[0104] If the temperature Tk of the power semiconductor element 1 is 75 °C and the first voltage Vce at time n, as in Fig. 15, Vce (n, 75 °C), the deterioration region detection unit 9 determines the first voltage Vce (n, 25 °C) when the temperature Tk of the power semiconductor element 1 is 25 °C, under the condition that the collector current Ic is equal.

[0105] The temporal change ΔVce of the first voltage without correction with the temperature Tk of the power semiconductor element 1 is given as follows. ΔVce=Vce(n, 75 °C)−Vce(n−125 °C)

[0106] In contrast, the deterioration region detection unit 9 according to the present embodiment determines the temporal change ΔVce (25 °C) of the first voltage from Vce (n, 25 °C) and Vce (n-1, 25 °C). ΔVce(25 °C)=Vce(n, 25 °C)−Vce(n−1, 25 °C)

[0107] Fig. Figure 16 is a diagram illustrating the relationship between collector current Ic and second voltage Vee in an IGBT.

[0108] The second voltage Vee changes with the collector current Ic and the temperature Tk of the power semiconductor element 1.

[0109] Even if the power semiconductor element 1 is supplied with the same collector current Ic, the resistance value of the metal layer on the surface of the emitter electrode surface 12 of the power semiconductor element 1 changes with the temperature Tk. As a result, for example, the second voltage Vee generated at 25 °C differs from the second voltage Vee generated at 125 °C.

[0110] In the present embodiment, the deterioration region detection unit 9 corrects the temporal change ΔVee of the second voltage Vee depending on the temperature Tk of the power semiconductor element 1.

[0111] Fig. 17 is a diagram for explaining the correction of the second voltage Vee at time (n-1).

[0112] If the temperature Tk of the power semiconductor element 1 is 125 °C and the second voltage Vee is equal to Vee (n-1, 125 °C) at time (n-1), as in Fig. 17, the deterioration region detection unit 9 determines the second voltage Vee (n-1, 25 °C) when the temperature Tk of the power semiconductor element 1 is 25 °C, under the condition that the collector current Ic is equal. Fig. Figure 18 is a diagram explaining the correction of the second voltage Vee at time n.

[0113] If the temperature Tk of the power semiconductor element 1 is 75 °C and the second voltage Vee at time n is Vee (n, 75 °C), as in Fig. 18, the deterioration region detection unit 9 determines the second voltage Vee (n, 25 °C) when the temperature Tk of the power semiconductor element 1 is 25 °C, under the condition that the collector current Ic is equal.

[0114] The temporal change ΔVee of the second voltage without correction with the temperature Tk of the power semiconductor element 1 is given as follows. ΔVee=Vee(n, 75 °C)−Vee(n−1, 125 °C)

[0115] In contrast, the deterioration region detecting unit 9 according to the present embodiment determines the temporal change ΔVee (25 °C) of the second voltage from Vee (n, 25 °C) and Vee (n-1, 25 °C). ΔVee(25 °C)=Vee(n, 25 °C)−Vee(n−1, 25 °C)

[0116] The deterioration area detection unit 9 uses the lookup table in Fig. 9 and identifies the degraded area corresponding to the combination of the corrected change ΔVce (25 °C) of the first voltage in equation (4) and the corrected change ΔVee (25 °C) of the second voltage in equation (6).

[0117] As described above, according to the present embodiment, even when the temperature Tk of the power semiconductor element 1 deviates from the standard temperature assumed by the look-up table, the deteriorated region can be determined using the look-up table for the standard temperature by correcting the change ΔVce of the first voltage and the change ΔVee of the second voltage based on the temperature Tk of the power semiconductor element 1. Embodiment 5

[0118] In a fifth embodiment, the change ΔVce of the first voltage and the change ΔVee of the second voltage due to the temperature Tk of the power semiconductor element 1 are corrected in the same manner as in the fourth embodiment, but the temperature Tk of the power semiconductor element 1 is estimated without using the temperature sensor 27.

[0119] Fig. 19 is a graph showing the relationship between the collector current Ic and the second voltage Vee at three temperatures Tk in the power semiconductor element 1.

[0120] As in Fig. As shown in Figure 19, the second voltage Vee exhibits some variation depending on the temperature Tk of the power semiconductor element 1 at a constant collector current Ic, as long as no deterioration occurs in the first bonding wires 15. This is because the second voltage Vee elements are configured with only metal such as the emitter electrode surface 12 and the bonding wires 15.

[0121] When the collector current Ic is constant, the second voltage Vee changes at a predetermined rate in accordance with the change in temperature Tk of the power semiconductor element 1.

[0122] For example, assume that the second voltage at 25°C of the power semiconductor element 1 is Vee (25°C) when the collector current Ic is equal to Ic0. When the collector current Ic is equal to Ic0, the second voltage Vee (75°C) at 75°C of the power semiconductor element 1 is 1.2×Vee (25°C), and the second voltage Vee (125°C) at 125°C of the power semiconductor element 1 is 1.4×Vee (25°C).

[0123] When the collector current Ic is equal to and when the second voltage Vee at the standard temperature T0 of the power semiconductor element 1 is the standard voltage Vee0, the second voltage Vee measured when the temperature of the power semiconductor element 1 is Tk has the following relationship. Vee=K1(Tk−T0)+Vee0

[0124] The resistance Ree (= Vee / Ic) (hereinafter second resistance) of a current path between emitter reference terminal 4 and emitter main terminal 3, which is represented by the value obtained by dividing the second voltage Vee by the collector current Ic, does not depend on the collector current Ic, and the second resistance Ree changes at a predetermined rate in accordance with the change in temperature Tk of the power semiconductor element 1.

[0125] For example, the second resistance at 25°C of power semiconductor element 1 is Ree (25°C). The second resistance Ree (75°C) at 75°C of power semiconductor element 1 is 1.2×Ree (25°C), and the second voltage Vee (125°C) at 125°C of power semiconductor element 1 is 1.4×Ree (25°C).

[0126] If the second resistance Ree at the standard temperature T0 of the power semiconductor element 1 is the standard resistance Ree0, the second resistance Ree measured when the temperature of the power semiconductor element 1 is Tk has the following relationship. Ree=K2(Tk−T0)+Ree0

[0127] The above condition does not apply if the deterioration occurs in the bonding area 20 of the first bonding wire 15.

[0128] Typically, the impact of deterioration of the bonding region 20 of the first bonding wire 15 on the second voltage Vee occurs immediately. The deterioration of the bonding region 20 of the first bonding wire 15 is caused by separation (lifting) of the bonding region. Even if the partial separation of the bonding region 20 gradually progresses and the bonding region gradually shrinks, the second voltage Vee hardly changes until complete separation. This is because the bonding region 20 is extremely thin and the resistance value of the bonding region 20 is initially extremely small.

[0129] To explain, a simple example: Suppose the cross-sectional area of the bonding region 20 is the same as the cross-sectional area of the first bonding wire 15, the thickness of the bonding region 20 (the thickness of a crack starting from the edge) is 1 nm, and the crack starting from the edge of the bonding region 20 reduces the cross-sectional area of the bonding region 20 to 1 / 1000. The increase in resistance corresponds to the increase in resistance when the length of the first bonding wire 15 is 1000 times larger than 1 nm, i.e., 1 µm. The increase in resistance when the area of the bonding region 20 decreases to one millionth corresponds to the increase in resistance when the length of the first bonding wire 15 increases by 1 mm.

[0130] The separation of the bonding region 20 occurs completely, and at the moment the first bonding wire 15 completely separates from the emitter electrode surface 12, the resistance of the bonding region 20 becomes instantaneous infinite, and the second voltage Vee fluctuates significantly. That is, the change in the second voltage Vee occurs instantaneously. Assuming that the temperature immediately before the instantaneous increase of the second voltage Vee is the same as the temperature immediately after the instantaneous increase, the standard voltage Vee0 or the standard resistance Ree0 can be corrected if the fluctuations of the second voltage Vee are always recorded.

[0131] Fig. 20 is a diagram showing a configuration of a semiconductor device according to the fifth embodiment. The semiconductor device according to the fifth embodiment includes, in addition to the components of the semiconductor device according to the first embodiment, a voltage history recording unit 31 and a temperature estimation unit 32.

[0132] The voltage history recording unit 31 stores the history of the second voltage Vee.

[0133] When the collector current Ic flowing through the power semiconductor element 1 is constant, the temperature estimation unit 32 estimates the temperature Tk of the power semiconductor element based on the second voltage Vee, the standard temperature T0, and the standard voltage Vee0, that is, the second voltage Vee at the standard temperature T0. When the time rate of change dVee of the second voltage is equal to or greater than a reference value TH1, the temperature estimation unit 32 updates the standard voltage Vee0 in accordance with the time rate of change dVee of the second voltage.

[0134] Fig. 21 is a flowchart showing a specific method of temperature estimation of the power semiconductor element 1 and identification of a deteriorated portion according to the fifth embodiment.

[0135] In step S101, the temperature estimation unit 32 sets the standard voltage Vee0, that is, the second voltage Vee for the standard temperature T0 of the power semiconductor element 1, to an initial value V0. The temperature estimation unit 32 sets the collector current Ic to a specific value Ic0.

[0136] In step S102, the temperature estimation unit 32 detects the first voltage Vce measured by the first voltage measuring circuit 5 and the second voltage Vee measured by the second voltage measuring circuit 6.

[0137] In step S103, the temperature estimation unit 32 calculates the temporal change rate dVee of the second voltage. Specifically, when the most recent second voltage is Vee(n) and the second voltage recorded at a previous time in the voltage history recording unit 31 is Vee(n-1), the temperature estimation unit 32 calculates the temporal change rate dVee of the second voltage Vee according to the following equation. dVee={Vee(n)−Vee(n−1)} / Vee(n−1)

[0138] In step S104, if the time rate of change dVee of the second voltage is equal to or greater than the reference value TH1, the process proceeds to step S105, and if the time rate of change dVee of the second voltage is smaller than the reference value TH1, the process proceeds to step S106.

[0139] In step S105, the temperature estimation unit 32 updates the standard voltage Vee0. Specifically, the temperature estimation unit 32 sets a value resulting from multiplying the current standard voltage Vee0 by (1+dVee) as the new standard voltage Vee0.

[0140] In step S106, the temperature estimation unit 32 estimates, as indicated in the following equation, the temperature Tk of the power semiconductor element 1 based on the standard temperature T0, the standard voltage Vee0, and the second voltage Vee. Tk=(Vee−Vee0) / K1+T0

[0141] In step S107, the deterioration region detecting unit 9 corrects the change ΔVce of the first voltage and the change ΔVee of the second voltage based on the temperature Tk of the power semiconductor element 1 in the same manner as in the fourth embodiment.

[0142] In step S108, the deterioration area detection unit 9 refers to the look-up table and identifies the deteriorated area using the corrected change ΔVce of the first voltage and the corrected change ΔVee of the second voltage.

[0143] Fig. Figure 22 is a graph showing the time change of the standard voltage Vee0.

[0144] If none of the bonding areas of a plurality of first bonding wires 15 deteriorates, the standard voltage Vee0 is set to the initial value V0. If a first of the bonding areas of the bonding wires 15 deteriorates, step S105 is executed for the first time, and the standard voltage Vee0 is set to V1. If a second bonding area of the bonding wires 15 deteriorates, step S105 is executed for the second time, and the standard voltage Vee0 is set to V2. If a third bonding area of the bonding wires 15 deteriorates, step S105 is executed a third time, and the standard voltage Vee0 is set to V3.

[0145] In this way, the standard voltage Vee0 is updated in accordance with the deterioration of the first bonding wire 15, which can always correctly correct the temperature influence.

[0146] To avoid missing a momentary increase in the second voltage Vee, the second voltage Vee is sampled at short intervals. The voltage history recording unit 31 can store the second voltage Vee at a previous time.

[0147] As described above, the semiconductor device according to the fifth embodiment can estimate the temperature Tk of the power semiconductor element 1 based on the measured value of the second voltage Vee without installing a temperature sensor. Embodiment 6

[0148] Fig. 23 is a diagram showing a configuration of a semiconductor device according to a sixth embodiment.

[0149] The semiconductor device according to the sixth embodiment includes, in addition to the components of the semiconductor device according to the first embodiment, a resistance history recording unit 131 and a temperature estimation unit 132.

[0150] The resistance history recording unit 131 stores the history of the second resistance Ree.

[0151] The temperature estimation unit 132 estimates the temperature Tk of the power semiconductor element based on the second resistance Ree, the standard temperature T0, and the standard resistance Ree0, that is, the second resistance Ree at the standard temperature T0. When the time rate of change dRee of the second resistance is equal to or greater than a reference value TH2, the temperature estimation unit 132 updates the standard resistance Ree0 according to the time rate of change dRee of the second resistance.

[0152] Fig. 24 is a flowchart showing a specific method of temperature estimation of the power semiconductor element 1 and identification of a deteriorated portion according to the sixth embodiment.

[0153] In step S201, the temperature estimation unit 132 sets the standard resistance Ree0, ie, the resistance Ree for the standard temperature T0 of the power semiconductor element 1, to an initial value R0.

[0154] In step S202, the temperature estimation unit 132 detects the first voltage Vce measured by the first voltage measuring circuit 5 and the second voltage Vee measured by the second voltage measuring circuit 6.

[0155] In step S203, the temperature estimation unit 132 calculates the second resistance Ree(Vee / Ic) by dividing the second voltage Vee by the collector current Ic.

[0156] In step S204, the temperature estimation unit 132 calculates the time change rate dRee of the second resistance Ree. That is, when the last second resistance is Ree(n) and the second resistance at a previous time is Ree(n-1), the temperature estimation unit 132 calculates the time change rate dRee of the second resistance Ree according to the following equation. dRee={Ree(n)−Ree(n−1)} / Ree(n−1)

[0157] In step S205, when the time rate of change dRee of the second resistance Ree is equal to or greater than the reference value TH2, the process proceeds to step S206, and when the time rate of change dRee of the second resistance Ree is smaller than the reference value TH2, the process proceeds to step S207.

[0158] In step S206, the temperature estimation unit 132 updates the standard resistance Ree0. Specifically, the temperature estimation unit 132 sets a value resulting from multiplying the current standard resistance Ree0 by (1+dRee) as the new standard resistance Ree0.

[0159] In step S207, the temperature estimation unit 132 estimates the temperature Tk of the power semiconductor element 1 based on the standard temperature T0, the standard resistance Ree0, and the second resistance Ree. Tk=(Ree−Ree0) / K2+T0

[0160] In step S208, the deterioration region detecting unit 9 corrects the change ΔVce of the first voltage and the change ΔVee of the second voltage based on the temperature Tk of the power semiconductor element 1 in the same manner as in the fourth embodiment.

[0161] In step S209, the deterioration area detection unit 9 refers to the look-up table and identifies the deteriorated area using the corrected change ΔVce of the first voltage and the corrected change ΔVee of the second voltage.

[0162] As described above, the semiconductor device according to the sixth embodiment can estimate the temperature Tk of the power semiconductor element 1 based on the measured value of the second voltage Vee and the second resistance Ree obtained from the collector current Ic without installing a temperature sensor. The semiconductor device according to the fifth embodiment requires that the collector current Ic be constant to estimate the temperature Tk of the power semiconductor element 1. In comparison, the semiconductor device according to the sixth embodiment can estimate the temperature Tk of the power semiconductor element 1 even if the collector current Ic is not constant. Embodiment 7

[0163] In the present embodiment, the semiconductor device according to the foregoing first to sixth embodiments is applied to a power converter assembly. Although the present embodiment is not limited to a specific power converter assembly, a three-phase inverter to which the first to sixth embodiments are applied will be described below.

[0164] Fig. 25 is a block diagram showing a configuration of a power conversion system according to a seventh embodiment.

[0165] The power converter system includes a power supply 100, a power converter assembly 200, and a load 300.

[0166] The power supply 100 is a DC power supply and supplies DC power to the power converter assembly 200. The power supply 100 can be configured with a variety of power supplies. For example, the power supply 100 can be configured with a DC power system, a solar cell, or a storage battery. The power supply 100 can be configured with a rectifier circuit or an AC / DC converter connected to an AC system. The power supply 100 can be configured with a DC / DC converter that converts the DC power output from a DC power system into a predetermined power.

[0167] The power converter assembly 200 is a three-phase inverter connected between the power supply 100 and the load 300. The power converter assembly 200 converts the DC power supplied by the power supply 100 into AC power and supplies the AC power to the load 300. The power converter assembly 200 includes a main converter circuit 201 for converting DC power to AC power and outputting the AC power, and a control circuit 203 for outputting a control signal to the main converter circuit 201 for controlling the main converter circuit 201.

[0168] Load 300 is a three-phase motor driven by the AC voltage supplied by power converter assembly 200. Load 300 is not limited to specific applications and is, for example, a motor incorporated into a variety of electrical devices. Load 300 is a motor incorporated, for example, into a hybrid vehicle, an electric vehicle, a rail vehicle, an elevator, or an air conditioning system.

[0169] The details of the power converter assembly 200 are described below.

[0170] The main converter circuit 201 includes a semiconductor module 202. The semiconductor module 202 contains switching elements and freewheeling diodes (not shown). The switching elements are switched to convert the DC voltage supplied by the power supply 100 into AC voltage, which in turn is supplied to the load 300. As shown in Fig. As shown in Figure 25, the degradation monitoring unit 205 in the first to sixth embodiments may be disposed within the semiconductor module 202. Alternatively, the degradation monitoring unit 205 may also be disposed outside the semiconductor module 202.

[0171] There are a variety of specific circuit configurations of the main converter circuit 201. The main converter circuit 201 according to the present embodiment is a two-stage three-phase full-bridge circuit. This circuit can be configured with six switching elements and six freewheeling diodes connected in anti-parallel to the respective switching elements. Each switching element in the main converter circuit 201 corresponds to the power semiconductor element 1 according to the above first to sixth embodiments. The semiconductor module 202 corresponds to the power module according to the above first to sixth embodiments. Two switching elements of the six switching elements are connected in series and form upper and lower arms. The upper and lower arms each form a phase (U-phase, V-phase, W-phase) of the full-bridge circuit. The output terminals of the upper and lower arms, i.e.three output terminals of the main converter circuit 201 are connected to the load 300.

[0172] The main converter circuit 201 includes a driver circuit (not shown) for driving each switching element. However, the driver circuit may be included in the semiconductor module 202, or the driver circuit may be provided separately from the semiconductor module 202. The driver circuit generates a drive signal for driving a switching element of the main converter circuit 201 and supplies the drive signal to the control electrode of the switching element of the main converter circuit 201. Specifically, the driver circuit outputs a drive signal for turning on a switching element and a drive signal for turning off a switching element to the control electrode of each switching element in accordance with a control signal from the control circuit 203. When the switching element is to remain on, the drive signal is a voltage signal (ON signal) equal to or higher than a threshold voltage of the switching element.If the switching element is to remain off, the drive signal is a voltage signal (OFF signal) equal to or lower than the threshold voltage of the switching element. The control circuit 203 controls the switching elements of the main converter circuit 201 so that a desired power is supplied to the load 300. Specifically, the time (ON time) during which each switching element of the main converter circuit 201 is to be in the ON state is calculated based on the power to be supplied to the load 300. For example, the control circuit 203 can control the main converter circuit 201 using pulse width modulation (PWM), which modulates the ON time of the switching elements depending on a voltage to be output.The control circuit 203 outputs a control command (control signal) to a drive circuit in the main converter circuit 201, so that, at the respective time, an ON signal is output to a switching element to be turned on and an OFF signal is output to a switching element to be turned off. The drive circuit outputs an ON signal or an OFF signal as a drive signal to the control electrode of each switching element in accordance with this control signal.

[0173] The power module of the power converter assembly according to the present embodiment includes a deterioration monitoring unit 205 in the first to sixth embodiments. This unit can detect which of the bonding regions of a plurality of first bonding wires 15 to the power semiconductor element 1 is deteriorating and estimate the lifetime of the power semiconductor element 1 and the semiconductor module 202. The deterioration monitoring unit 205 outputs the information about the deterioration or lifetime to the control circuit 203 and, if necessary, modifies a drive signal output by the drive device.

[0174] In the present embodiment, an example is described in which the first to sixth embodiments are applied to a two-level three-phase inverter. However, the first to sixth embodiments can be applied to a variety of power converter assemblies. Although the present embodiment is a two-level power converter assembly, it can be a three-level or multi-level power converter assembly. The first to sixth embodiments can be applied to a single-phase inverter when powering a single-phase load. When powering a DC load or a similar load, the first to sixth embodiments can be applied to a DC / DC converter or an AC / DC converter. The load 300 connected to the power converter assembly 200 is not limited to a motor.The power converter assembly 200 can be used, for example, as a power supply device for an electrical discharge machine, a laser beam machine, an induction heating oven, or a wireless charging system. The power converter assembly 200 can be used as a power conditioner for a photovoltaic system or an energy storage system.

[0175] Fig. 26 is a diagram showing a specific configuration of the deterioration monitoring unit 205 according to the first to sixth embodiments.

[0176] For example, the functions of the deterioration area detection unit 9 and the look-up table storage unit 19 are Fig. 1 is realized by a processing circuit 1001.

[0177] The processing circuit 1001 can be implemented by dedicated hardware. The processing circuit corresponds to a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or a combination thereof.

[0178] Fig. 27 is a diagram showing an example in which the processing circuit 1001 is implemented by software.

[0179] As in Fig. As shown in Figure 27, the processing circuit 1001 can implement the function of each unit through a processor 1002 that reads and executes a program stored in a memory 1003. The processor 1002 corresponds to a central processing unit (CPU), a processing device, a computing device, a microprocessor, a microcomputer, or a digital signal processor (DSP).

[0180] The functions of the deterioration area detection unit 9 and the lookup table storage unit 19 are realized by software, firmware, or a combination of software and firmware. The software or firmware is described as a program and stored in the memory 1003. The memory 1003 corresponds, for example, to a volatile or non-volatile semiconductor memory such as a random access memory (RAM), a read-only memory (ROM), a flash memory, an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a magnetic disk, a flexible disk, an optical disk, a compact disc, a mini-disc, or a digital versatile disc (DVD).

[0181] Some of the functions of the deterioration area detection unit 9 and the lookup table storage unit 19 may be implemented by dedicated hardware, and some of them may be implemented by software or firmware.

[0182] In this way, the processing circuit 1001 can implement the functions described above through hardware, software, firmware, or a combination thereof. Modifications

[0183] (1) In the preceding embodiments, an IGBT is used as the power semiconductor element 1. However, the type of power semiconductor element is not limited to an IGBT; a metal oxide semiconductor field-effect transistor (MOSFET), for example, may also be used. In this case, "collector" should be read as "drain" and "emitter" as "source," so that the embodiments can be applied equally. Similarly, "Vce" should be read as "Vds" and "Vee" as "Vss."

[0184] (2) Fig. 28 is a diagram showing another example of the correspondence relationship between the first deteriorated region and a combination of the time change ΔVce of the first voltage Vce and the time change ΔVce of the second voltage Vee.

[0185] When the first deteriorated area is the bonding area 20A, the first voltage Vce increases by 20% or more and the second voltage Vee increases by 10% or more. When the first deteriorated area is the bonding area 20B, the first voltage Vce increases by 20% or more and the second voltage Vee increases by 40% or more. When the first deteriorated area is the bonding area 20C, the first voltage Vce increases by 30% or more and the second voltage Vee increases by 10% or more. When the first deteriorated area is the bonding area 20D, the first voltage Vce increases by 30% or more and the second voltage Vee increases by 40% or more.

[0186] Fig. Figure 29 is a diagram showing another example of the lookup table for identifying the first deteriorated area. This lookup table can be obtained from the correspondence relationship in Fig. 28 can be derived.

[0187] It is defined that the first deteriorated region is the bonding region 20A when the first voltage Vce increases by 20% or more and less than 30% and the second voltage Vee increases by 10% or more and less than 40%. It is defined that the first deteriorated region is the bonding regions 20A and 20C when the first voltage Vce increases by 30% or more and the second voltage Vee increases by 10% or more and less than 40%. It is defined that the first deteriorated region is the bonding regions 20A and 20B when the first voltage Vce increases by 20% or more and less than 30% and the second voltage Vee increases by 40% or more. It is defined that the first deteriorated area is the bonding areas 20A, 20B, 20C and 20D when the first voltage Vce increases by 30% or more and the second voltage Vee increases by 40% or more.

[0188] The deterioration area detection unit 9 refers to the lookup table in Fig. 29 and identifies the deteriorated region according to the combination of the temporal change ΔVce of the first voltage Vce and the temporal change ΔVce of the second voltage Vee. If the combination of the temporal change ΔVce of the first voltage Vce and the temporal change ΔVce of the second voltage Vee is not included in the lookup table, the deterioration region detection unit 9 determines that a plurality of bonding regions of the emitter electrode surface 12 to which a plurality of first bonding wires 15 are connected do not have a deteriorated region.

[0189] The embodiments described herein should be understood as illustrative only. List of reference symbols 1 power semiconductor element 2 Collector main connection 3 Emitter main connection 4 Emitter reference terminal 5 first voltage measuring circuit 6 second voltage measuring circuit 7 first output connection 8 second output connection 9 Deterioration area detection unit 10 third output connection 11 Collector substrate 12 Emitter electrode surface 13 lots 14 third bond wire 15, 15A - 15D first bond wire 16 second bond wire 17 Main current flow path 18 Heating 19 Lookup table storage unit 20A-20D bonding area 21 Gate connection 22 Gate bond wire 23 Deterioration area history recording unit 25 Lifetime estimation unit 27 Temperature sensor 31 Voltage history recording unit 32, 132 Temperature estimation unit 100 Power supply 119 Deterioration area 131 Resistance History Recording Unit 200 power converter arrangement 201 Main converter circuit 202 semiconductor module 203 Control circuit 205 Deterioration Monitoring Unit 300 load 1001 processing circuit 1002 processor 1003 memory

Claims

[1] Semiconductor device comprising: a power semiconductor element (1); a collector substrate (11) electrically connected to the power semiconductor element (1); a collector main terminal (2) electrically connected to the collector substrate (11); an emitter main terminal (3) connected to an emitter electrode surface (12) of the power semiconductor element (1) via a plurality of first bonding wires (15); an emitter reference terminal (4) connected to the emitter electrode surface (12) of the power semiconductor element (1) via a second bonding wire (16); a first voltage measuring circuit (5) for measuring a first voltage which is a difference between a potential at the collector main terminal (2) and a potential at the emitter main terminal (3); a second voltage measuring circuit (6) for measuring a second voltage which is a difference between a potential at the emitter reference terminal (4) and a potential at the emitter main terminal (3); and a deterioration area detection unit (9) for referring to correspondence information defining a deteriorated area of a plurality of bonding areas (20) on the emitter electrode surface (12) to which the first bonding wires (15) are connected, for a combination of temporal change of the first voltage and temporal change of the second voltage, and for identifying the deteriorated area corresponding to a combination of temporal change of the first voltage measured by the first voltage measuring circuit (5) and temporal change of the second voltage measured by the second voltage measuring circuit (6). [2] A semiconductor device according to claim 1, further comprising a recording unit (23) for recording the history of the identified deteriorated region. [3] A semiconductor device according to claim 2, wherein then, if none of the bonding areas in the history is deteriorated, the deterioration area recognition unit (9) refers to the first correspondence information for no deterioration as the correspondence information and identifies the deteriorated area, and then, when one of the bonding regions in the history is deteriorated, the deterioration region recognition unit (9) refers to the second correspondence information corresponding to the deteriorated region in the history as the correspondence information and identifies the deteriorated region. [4] A semiconductor device according to claim 2 or 3, further comprising a lifetime estimation unit (25) for estimating the lifetime of the semiconductor device based on the history of the deteriorated region and a driving method of the power semiconductor element (1). [5] The semiconductor device according to claim 4, wherein the lifetime estimation unit (25) switches a driving method of the power semiconductor element (1) based on the estimated lifetime. [6] The semiconductor device according to claim 4, wherein the lifetime estimation unit (25) gives a user a hint to recommend switching a driving method of the power semiconductor element (1) based on the estimated lifetime. [7] Semiconductor component according to one of claims 1 to 6, further comprising a temperature sensor (27) for measuring the temperature of the power semiconductor element (1). [8] A semiconductor device according to any one of claims 1 to 6, further comprising a temperature estimation unit (32) for estimating the temperature of the power semiconductor element (1) based on the second voltage, a standard temperature and a standard voltage which is the second voltage at the standard temperature when the collector current flowing through the power semiconductor element (1) is constant. [9] The semiconductor device according to claim 8, wherein the temperature estimation unit (32) updates the standard voltage in accordance with the rate of temporal change of the second voltage when a rate of temporal change of the second voltage is equal to or greater than a reference value. [10] The semiconductor device according to claim 9, wherein the temperature estimation unit (32) sets, when the rate of temporal change of the second voltage is dVee, a value obtained by multiplying the current standard voltage by (1+dVee) as a new standard voltage. [11] A semiconductor device according to any one of claims 1 to 6, further comprising a temperature estimation unit (132) for estimating the temperature of the power semiconductor element (1) based on a second resistance obtained by dividing the second voltage by the collector current flowing through the power semiconductor element (1), a standard temperature, and a standard resistance which is the second resistance at the standard temperature. [12] The semiconductor device according to claim 11, wherein the temperature estimation unit (132) updates the standard resistance in accordance with the rate of temporal change of the second resistance when a rate of temporal change of the second resistance is equal to or greater than a reference value. [13] The semiconductor device according to claim 12, wherein the temperature estimation unit (132) sets, when the time rate of change of the second resistance is dRee, a value obtained by multiplying the current standard resistance by (1+dRee) as a new standard resistance. [14] A semiconductor device according to any one of claims 7 to 13, wherein the correspondence information defines a deteriorated region of the bonding regions for a combination of temporal change of the first voltage and temporal change of the second voltage at a standard temperature, and the deterioration region detection unit (9) corrects the temporal change of the measured first voltage and the temporal change of the measured second voltage depending on the temperature of the power semiconductor element (1) and refers to the correspondence information to identify the deteriorated region corresponding to a combination of the corrected temporal change of the first voltage and the corrected temporal change of the second voltage. [15] Power converter assembly comprising: the semiconductor component according to one of claims 1 to 14; a main converter circuit (201) for converting the input power and outputting the converted power; and a control circuit (203) for outputting a control signal for controlling the main converter circuit (201) to the main converter circuit (201).

Images