JP2019062737A - Driver - Google Patents

Abstract

To provide a driver capable of restraining short circuit current at the time of short circuit detection, while ensuring short circuit tolerance.SOLUTION: A driver 100 has a control electrode for controlling the output current, and drives a power switching element 200 which controls driving of a load by applying a voltage to the control electrode. The driver 100 includes a control section 10 for controlling the voltage application timing to the control electrode. The control electrode is divided into more than one division electrodes including a first electrode 211 and a second electrode 212, and the control section 10 drives the switching element 200 in safe mode for appling a voltage to one of the first electrode 211 or the second electrode 212, in the short circuit detection period for detecting short circuit of the load. In normal driving period for driving the load normally, the switching element 200 is driven in normal mode for applying a voltage to a division electrode, not applied with voltage in at least safe mode, out of the first electrode 211 and the second electrode 212.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a drive device for driving control electrodes such as gate electrodes and base electrodes.

  The supply of the overcurrent to the switching element due to the short circuit leads to the thermal destruction of the element, so fail safe control such as stopping the switching element is performed when the short circuit is detected. In the detection of the short circuit, it is necessary to separate the overcurrent and the pulse noise due to the short circuit. In order to determine that it is an overcurrent, for example, the drive circuit described in Patent Document 1 provides a predetermined filter time. That is, when the output current flowing through the switching element is maintained at or above a predetermined threshold during a set filter time, it is determined that the current is due to a short circuit.

  Conversely, during this filter time, the state in which the output current flows in the switching element is maintained. The output current at the time of short circuit is larger than that at the time of normal driving, and power which is not originally required in the filter time is consumed, which may lead to overheating of the switching element.

  As a method of suppressing the calorific value of the switching element at the time of the short circuit, it may be considered to reduce the saturation current of the output current, to increase the heat capacity of the switching element, and to improve the response speed of the short circuit protection. Among them, as a method of suppressing the output current, there are, for example, the techniques described in Patent Document 2 and Patent Document 3.

  The semiconductor device described in Patent Document 2 improves load short circuit tolerance while realizing a low on-voltage by optimizing the size of the effective region through which the output current flows.

  Further, the load driving device described in Patent Document 3 has a gate drive circuit and a clamp circuit. At the time of a short circuit, the gate of the IGBT is controlled using a clamp voltage smaller than the gate voltage at the time of normal driving in which no short circuit occurs.

JP, 2014-23342, A JP, 2013-84922, A JP, 2012-204985, A

  However, in the semiconductor device described in Patent Document 2, since the output current is smaller than that of the conventional configuration, the current capability of the switching element is reduced. Since the current capability and the on-voltage are in a trade-off relationship, as a result of reducing the current capability, the on-voltage at the time of normal driving increases and the loss increases.

  Further, in the load drive device described in Patent Document 3, in addition to the increase in circuit scale due to the need for two power supply circuits, this technology alone ensures a sufficient load short circuit withstand capacity required in recent years. I can not do it.

  The present invention has been made in view of the above problems, and an object of the present invention is to provide a drive device capable of securing a short circuit tolerance while suppressing a loss during normal driving.

  The invention disclosed herein employs the following technical means to achieve the above object. In addition, the reference numerals in the parenthesis described in the claims and this section indicate the correspondence with specific means described in the embodiment described later as one aspect, and the technical scope of the invention is limited. It is not something to do.

  In order to achieve the above object, the present invention is a drive device for driving a power switching element (200) having a control electrode (210) that controls an output current supplied to a load by applying a voltage. And a control unit (10) for controlling the timing of voltage application to the control electrode, wherein the control electrode is divided into two or more divided electrodes including a first electrode (211) and a second electrode (212) The part drives a power switching element in a safe mode in which a voltage is applied to one of the first electrode and the second electrode in a short circuit detection period for detecting a short circuit of a load, and in a normal drive period for normal driving of a load. Driving the power switching element in a normal mode in which a voltage is applied to both the first electrode and the second electrode.

  According to this, in the safe mode which is a period for detecting a short circuit, a voltage is applied to one of the first electrode and the second electrode which form a part of the control electrode. Therefore, the area of the control electrode to which the voltage is applied is smaller than that of the entire control electrode. Therefore, the channel formation region in the switching element is smaller than when a voltage is applied to the entire control electrode, and the output current capability can be reduced. Therefore, the power consumption at the time of the short circuit in the safe mode can be suppressed, and the short circuit tolerance can be improved by suppressing the heat generation of the switching element.

  In addition, in the normal mode, the control unit applies a voltage to both the first electrode and the second electrode to drive the power switching element. According to this, in the normal mode, a channel is formed by application of a voltage to the second electrode in addition to the first electrode. Therefore, in the normal mode, a larger amount of output current can be output as compared to that in the safe mode, the conduction resistance becomes smaller, and the loss during normal driving can be reduced.

It is a block diagram which shows schematic structure of the drive device and IGBT in 1st Embodiment. FIG. 2 is a circuit diagram showing a schematic configuration of a drive circuit and a short circuit detection circuit. It is a front view which shows the electrode structure of IGBT. It is a flowchart which shows the control flow by a control part. It is a flowchart which shows the control flow by a control part. It is a flowchart which shows the control flow by a control part. It is a flowchart which shows the control flow by a control part. It is the schematic which shows the voltage-current characteristic in IGBT. It is a figure which shows the concept of heat transfer distance W. It is a conceptual diagram which shows the temperature change in case a heat source exists uniformly on one side. It is a conceptual diagram which shows the temperature change in the conditions of L> = 2W. It is a conceptual diagram which shows the temperature change in the conditions of L <2W. It is a conceptual diagram which shows the temperature change in the conditions which made D small from the state shown in FIG. It is a conceptual diagram which shows the temperature change in the conditions which made D small from the state shown in FIG. It is a conceptual diagram which shows the temperature change in the conditions which made L small from the state shown in FIG. It is a figure collectively showing simulation conditions. It is a simulation result which shows the stripe width dependence of the temperature rise of IGBT in the semiconductor substrate which has silicon as a main component. It is a simulation result which shows the stripe width dependence of the temperature rise of IGBT in the semiconductor substrate which has silicon carbide as a main component. FIG. 16 is a front view showing a region configuration of an IGBT in a first modification. FIG. 16 is a front view showing a region configuration of an IGBT in a first modification. FIG. 16 is a front view showing a region configuration of an IGBT in a first modification. It is a circuit diagram which shows schematic structure of the drive circuit in 2nd Embodiment. It is a flowchart which shows the control flow by a control part. It is a circuit diagram which shows schematic structure of the drive circuit in 3rd Embodiment. It is a flowchart which shows the control flow by a control part. FIG. 14 is a circuit diagram showing a schematic configuration of a drive circuit and an IGBT in another embodiment. FIG. 16 is a circuit diagram showing a cell configuration of an IGBT in another embodiment. FIG. 16 is a circuit diagram showing a cell configuration of an IGBT in another embodiment. It is a front view which shows the area | region structure of IGBT in other embodiment.

  Hereinafter, embodiments of the present invention will be described based on the drawings. The same reference numerals are given to parts which are the same as or equivalent to each other in the following drawings.

First Embodiment
First, with reference to FIGS. 1 to 3, a drive device according to the present embodiment and a schematic configuration of a switching element to be driven by the drive device will be described. In addition, FIG. 3 is a figure for demonstrating a concept, The scale is not exact.

  The drive device in the present embodiment is, for example, a circuit that controls on / off of a load such as a motor. As shown in FIG. 1, the supply of current to the load 300 is controlled by turning on and off a switching element 200 such as an IGBT, a MOSFET, or another transistor. The driving device 100 is a device that controls the on / off of the switching element 200 to control the supply of current to the load 300.

  In the present embodiment, an IGBT will be described as an example of the switching element 200. Therefore, switching element 200 is denoted as IGBT 200. The output current described in the claims corresponds to the collector current flowing between the collector and the emitter of the IGBT 200. Furthermore, the control electrode described in the claims corresponds to the gate electrode 210 of the IGBT 200. The first electrode as the divided gate electrode 210 is referred to as a first gate electrode 211, and the second electrode is referred to as a second gate electrode 212. That is, the divided electrodes described in the claims correspond to the first gate electrode 211 and the second gate electrode 212 in the present embodiment.

  First, the drive device 100 will be described. As shown in FIG. 1, the drive device 100 in the present embodiment includes a control unit 10, a drive circuit 20, and a short circuit detection circuit 30.

  The control unit 10 outputs a control signal to the drive circuit 20 to control application of a voltage to the gate electrode 210. The control unit 10 is configured to determine which electrode of the gate electrode 210 is to be applied with a voltage under a predetermined condition. That is, the application timing of the voltage to the first gate electrode 211 and the second gate electrode 212 is controlled. A detailed description of the control performed by the control unit 10 will be given later.

  The drive circuit 20 is a circuit for applying a voltage to the gate electrode 210 based on a control signal input from the control unit 10. The drive circuit 20 includes a first drive circuit 21 and a second drive circuit 22, and is a first gate electrode 211 which is a divided portion of the gate electrode 210 and another portion which is divided. A voltage can be applied independently to the second gate electrode 212.

  Specifically, as shown in FIG. 2, the drive circuit 20 has a first drive circuit 21 and a second drive circuit 22.

  The first drive circuit 21 includes a switch S1 and a switch S2, and a resistor R1 and a resistor R2. The first gate electrode 211 is connected to the gate potential Vg via the resistor R1 and the switch S1. The first gate electrode 211 is connected to the reference potential VSS of the drive circuit 20 via the resistor R2 and the switch S2. The switch S1 and the switch S2 correspond to the first on-side circuit and the first off-side circuit described in the claims, respectively. Although switches S1 and S2 are each represented as one switch device in FIG. 2, each of switches S1 and S2 may include a temperature control circuit or the like for relaxing the temperature characteristic of gate potential Vg, It may be a switch having a function of supplying a constant current to the gate electrode 210. The resistors R1 and R2 may be interposed between the drive circuit 20 and the gate electrode 210 of the IGBT 200, as in the resistor R3 of the second drive circuit 22 described later.

  The second drive circuit 22 includes a switch S3 and a switch S4, and a resistor R3. The second gate electrode 212 is connected to the gate potential Vg via the resistor R3 and the switch S3. Further, the second gate electrode 212 is connected to the reference potential VSS via the resistor R3 and the switch S4. The switch S3 and the switch S4 respectively correspond to a second on-side circuit and a second off-side circuit described in the claims. Each of the switches S3 and S4 may include a temperature control circuit or the like, or may be a switch having a function of supplying a constant current to the gate electrode 210. Further, the resistor R3 may be divided into the on-side circuit and the off-side circuit, like the resistors R1 and R2 of the first drive circuit 21.

  Although the reference potential VSS is often the same potential as the emitter potential of the IGBT 200, a lower potential can also be used for the purpose of preventing erroneous ON and the like, and in particular if the potential is lower than the threshold voltage of the IGBT 200. It is not limited.

  When a voltage is applied to the first gate electrode 211 to turn on the IGBT 200, the switch S2 is turned off and then the switch S1 is turned on to supply charge from the gate potential Vg side to the first gate electrode 211. On the other hand, when the first gate electrode 211 is turned off, the switch S1 is turned off and then the switch S2 is turned on to draw charges from the first gate electrode 211 to the reference potential VSS side.

  When a voltage is applied to the second gate electrode 212 to turn on the IGBT 200, the switch S4 is turned off and then the switch S3 is turned on to supply charge from the gate potential Vg side to the second gate electrode 212. On the other hand, when the second gate electrode 212 is turned off, the switch S3 is turned off and then the switch S4 is turned on to draw charges from the second gate electrode 212 to the reference potential VSS side.

  The on / off of each of the switches S1 to S4 is controlled based on a control signal output from the control unit 10. That is, the control unit 10 controls the on / off of the switches S1 to S4 to control the application of the voltage to the first gate electrode 211 and the second gate electrode 212.

  The switching speed when turning on the IGBT 200 depends on the resistance value of the resistor R1. Also, the switching speed for turning off depends on the resistance value of the resistor R2. The resistors R3 affect each other's transition time between the normal mode and the safe mode which will be described in detail later.

  The short circuit detection circuit 30 is connected to the sense emitter terminal of the IGBT 200 as shown in FIG. The sense emitter terminal is a terminal for detecting a sense current having a correlation with a collector current which is an output current. The short circuit detection circuit 30 includes a shunt resistor R4, a subtraction circuit 31, a comparison circuit 32, and a power supply that generates a reference potential Vref.

  The shunt resistor R4 is connected between the sense emitter terminal and the reference potential VSS. The sense current flows from the sense emitter terminal of the IGBT 200 to the reference potential VSS via the shunt resistor R4.

  The subtraction circuit 31 is a circuit that receives the potentials at both ends of the shunt resistor R4 and outputs the difference. That is, the subtraction circuit 31 is configured to output the voltage drop amount due to the shunt resistor R4. Since this voltage drop amount is proportional to the current value of the sense current, the sense current can be detected by detecting the voltage drop amount due to the shunt resistor R4. And, since the sense current is correlated with the collector current, it is possible to detect the current value of the collector current which is the output current as a result.

  The comparison circuit 32 is a circuit that compares the amount of voltage drop output from the subtraction circuit 31 with the reference potential Vref. The comparison circuit 32 outputs a detection signal to the control unit 10 when the voltage drop amount is equal to or higher than the reference potential Vref. The reference potential Vref is set to a voltage value corresponding to a threshold current when judging a short circuit. That is, the comparison circuit 32 is configured to output a detection signal to the control unit 10 when the collector current becomes equal to or more than the threshold current defined by the reference potential Vref.

  The reference potential Vref is variable. The control unit 10 can change the threshold current by controlling the reference potential Vref.

  Next, an IGBT 200 as a switching element will be described. As shown in FIG. 3, the IGBT 200 in the present embodiment is a generally known vertical insulated gate bipolar transistor, and is formed on, for example, a semiconductor substrate 290 containing silicon as a main component. The IGBT 200 has a gate electrode 210 as a control electrode on the surface layer of the main surface 290 a of the semiconductor substrate 290 and a pad 220 electrically connected to the gate electrode 210 for applying a voltage. Also, although not shown, it has an emitter region as a semiconductor region and a collector region. An emitter electrode is formed on main surface 290a and in contact with the emitter region, and a collector electrode on the back surface opposite to main surface 290a and in contact with the collector region. ing. The emitter electrode is connected to the reference potential VSS, and the collector electrode is connected to a power supply voltage VCC via a parallel circuit of a load 300 and a free wheeling diode 400.

  The gate electrode 210 in the present embodiment has a first gate electrode 211 and a second gate electrode 212, and is formed extending in a predetermined direction along the main surface 290a. The first gate electrodes 211 and the second gate electrodes 212 are alternately formed side by side.

  The pad 220 has a first pad 221 connected to the first gate electrode 211 via a wire, and a second pad 222 connected to the second gate electrode 212 via a wire. That is, when applying a potential to the first gate electrode 211, the control unit 10 instructs the first drive circuit 21 to apply a voltage to the first pad 221. As a result, the first drive circuit 21 applies a predetermined voltage to the first pad 221, and a potential is applied to the first gate electrode 211 accordingly. When a potential is to be applied to the second gate electrode 212, the control unit 10 instructs the second drive circuit 22 to apply a voltage to the second pad 222. As a result, the second drive circuit 22 applies a predetermined voltage to the second pad 222, and a potential is applied to the second gate electrode 212 accordingly.

  The emitter region is in contact with the gate electrode 210 via an insulating film, and carriers are moved between the emitter region and the collector region when a voltage is applied to the gate electrode 210. Then, due to the movement of carriers, a collector current which is an output current flows between the emitter electrode and the collector electrode.

  Since the IGBT 200 in the present embodiment has the first gate electrode 211 and the second gate electrode 212 as the gate electrode 210, the emitter region in contact with each of the gate electrodes 210 is for applying a voltage to the gate electrode 210. The movement path of the carrier is independently generated. That is, as shown in FIG. 3, when a voltage is applied to the first gate electrode 211, carriers move when the voltage is applied to the first region 231 which is a carrier moving path and when the voltage is applied to the second gate electrode 212. There is a second area 232 which is a route. Although FIG. 3 is a top view, different hatching is applied to each area in order to clarify the distinction of the areas.

  As described above, the first gate electrodes 211 and the second gate electrodes 212 are extended in a predetermined direction while being alternately arranged. Therefore, the first region 231 and the second region 232 are adjacent to each other, and are formed in a stripe shape along the extending direction of the gate electrode 210. In the present embodiment, as shown in FIG. 3, the first gate electrode 211 and the second gate electrode 212 are formed to be separated by the same distance from each other, and the first region 231 and the second region 232 are It is substantially equivalent except the formation position. Assuming that the width of the first region 231 in the direction perpendicular to the extending direction (the direction in which the gate electrodes 210 are aligned) is L, and the separation distance between adjacent first regions 231 in the alignment direction is D, the IGBT 200 in this embodiment Then, L = D.

  Note that it is preferable that at least one of the first region 231 and the second region 232 be formed so as to allow an area larger than the rated current defined in the switching element as an output current. In other words, when a predetermined voltage is applied to the first gate electrode 211 or the second gate electrode 212, the first gate electrode 211, the second gate electrode 212, and the emitter region flow in such a manner that a current greater than the rated current flows. It is preferable that the semiconductor substrate 290 be laid out.

  In the present embodiment, since the first region 231 and the second region 232 are equivalent to each other, even if a voltage is applied to any of the first gate electrode 211 and the second gate electrode 212, the rated current defined in the switching element The above collector current flows.

  Next, with reference to FIG. 4 to FIG. 7, the operation of the drive device 100 according to the present embodiment will be described.

  Hereinafter, an operation at the time of turn-on of the IGBT 200, an operation at the time of turning on the IGBT 200, and an operation at the time of turn-off will be separately described. In each of FIGS. 4 to 7, the detection signal is a signal output to the control unit 10 when the short circuit detection circuit 30 detects a short circuit. The short circuit detection circuit 30 outputs an H signal to the control unit 10 when a short circuit is detected and an L signal when a short circuit is not detected. The operation mode is a mode in which the control unit 10 controls the IGBT 200, a voltage is not applied to the gate electrode 210, and a stop mode in which the operation of the IGBT 200 is stopped, and a voltage is applied to a part of the gate electrodes 210. A safe mode for driving the IGBT 200 and a normal mode controlled so that the current capability is higher than that of the safe mode.

<At turn on>
First, with reference to FIG. 4, the turn-on operation of IGBT 200 in the state where no short circuit has occurred will be described.

  Before time t1 shown in FIG. 4, control unit 10 is in the stop mode, and IGBT 200 is in the off state. No voltage is applied to the first gate electrode 211 and the second gate electrode 212. Therefore, the collector current is also not flowing.

  It is assumed that control unit 10 receives a control signal to turn on IGBT 200 from the outside at time t1. Control unit 10 drives IGBT 200 in the safe mode for the purpose of confirming the presence or absence of a short circuit. Specifically, at time t1, the control unit 10 outputs a control signal to the first drive circuit 21 so as to apply a voltage to the first gate electrode 211. Thereby, a voltage is applied to the first gate electrode 211. Along with this, the current value of the collector current starts to rise. The collector current has a constant value after the occurrence of the overshoot. The control unit 10 starts counting the filter time from the start of the safe mode. This filter time corresponds to the short circuit detection period described in the claims. Note that the count of the filter time may be realized by any circuit as long as it can count a predetermined time, and either an analog circuit or a digital circuit can be adopted. Further, the filter time is not limited to the method determined by the time from time t1, but, for example, the voltage applied to the first gate electrode 211 is monitored, and the period until the voltage reaches a certain value is determined. It may be a filter time. In addition, after the voltage reaches a fixed value, a fixed time may be elapsed.

  Under the condition that the control unit 10 does not detect the detection signal from the short circuit detection circuit 30 at time t2 after a predetermined filter time has elapsed from time t1, the control unit 10 sends the second drive circuit 22 a second command. A control signal is output to apply a voltage to the two gate electrodes 212. In the present embodiment, as shown in FIG. 4, in the safe mode, the collector current does not exceed the threshold current, and thus no detection signal is input to the control unit 10. Thus, the control unit 10 shifts to the normal mode, and the second drive circuit 22 applies a voltage to the second gate electrode 212. Note that a voltage is continuously applied to the first gate electrode 211.

  At time t2, a voltage is applied to both the first gate electrode 211 and the second gate electrode 212, so that the current capability is improved in the normal mode as compared to the safe mode. As described above, the load 300 according to the present embodiment is a motor having a relatively high inductance, so that the value of the collector current hardly changes even when the mode is shifted from the safe mode to the normal mode. However, the on-resistance is reduced because the area serving as a channel in the semiconductor substrate 290 is increased. A period in which a voltage is applied to both the first gate electrode 211 and the second gate electrode 212 and the IGBT 200 normally drives the load 300 corresponds to the normal driving period described in the claims.

  Next, with reference to FIG. 5, the turn-on operation of IGBT 200 in the state where a short circuit of load 300 has occurred will be described.

  Before time t3 shown in FIG. 5, control unit 10 is in the stop mode, and IGBT 200 is in the off state. No voltage is applied to the first gate electrode 211 and the second gate electrode 212. Therefore, the collector current is also not flowing.

  It is assumed that control unit 10 receives a control signal from the outside to turn on IGBT 200 at time t3. Control unit 10 drives IGBT 200 in the safe mode for the purpose of confirming the presence or absence of a short circuit. The specific driving is as described in the turn-on operation in which no short circuit occurs, and thus the detailed description is omitted. At time t3, a voltage is applied to the first gate electrode 211, and the current value of the collector current starts to rise.

  Here, since it is assumed that the load 300 is shorted, the collector current exceeds the threshold current and rises with the saturation current as the upper limit. At time t4 when the collector current reaches the threshold current, the short circuit detection circuit 30 outputs a detection signal to the control unit 10. Under the condition that the input of the detection signal continues for a predetermined time, the control unit 10 determines that the short circuit state is present at time t5 in FIG. 5, and stops the application of the voltage to the first gate electrode 211. After time t5, both the first gate electrode 211 and the second gate electrode 212 are in the off state, and the control unit 10 is in the stop mode.

  In this example, since a short circuit is detected during the safe mode, a voltage is applied to the second gate electrode 212 and the IGBT 200 is stopped before it is fully turned on. The saturation current of the IGBT 200 is smaller in the state applied to only the first gate electrode 211 as compared with the state in which the voltage is applied to both the first gate electrode 211 and the gate electrode 212. Therefore, the power consumption of the IGBT 200 in the filter time can be reduced. Thus, the amount of heat generated by the IGBT 200 can be reduced, and thermal destruction can be prevented.

<On state>
First, with reference to FIG. 6, a state in which a short circuit does not occur while the IGBT 200 is operating in the on state will be described. It is assumed that pulse noise is superimposed on the detection signal of the short circuit detection circuit 30. Here, the on state is a state in which a voltage is applied to both the first gate electrode 211 and the second gate electrode 212, and in the present embodiment, it is a full on state.

  Before time t7 shown in FIG. 6, a voltage is applied to both the first gate electrode 211 and the second gate electrode 212, and the IGBT 200 is driven in the fully on state. Then, at time t7, pulse noise is superimposed on the detection signal. When the detection signal is input to the control unit 10, the control unit 10 shifts to the safe mode. That is, the control unit 10 stops the application of the voltage to the second gate electrode 212 and starts counting of the filter time. As described above, the filter time count may be realized by any method. Also in the transition from the normal mode to the safe mode, the rated current is secured only by the application of the voltage to the first gate electrode 211, so the value of the collector current does not change.

  Since the duration of the pulse noise is sufficiently short compared to the filter time, the pulse noise does not continue from time t7 set as the filter time to time t8. Therefore, the control unit 10 determines that the load 300 is not in a short circuit state, and shifts to the normal mode again at time t8 to apply a voltage to the second gate electrode 212. Note that a filter sufficiently shorter than the above-described filter time may be set separately during transition to the safe mode after pulse noise is superimposed on the detection signal at time t7. Thus, unnecessary switching between the safe mode and the normal mode can be prevented.

  Next, with reference to FIG. 7, an operation in the case where a short circuit occurs in the load 300 in the on state will be described.

  It is assumed that a short circuit occurs in load 300 at time t9 during which IGBT 200 is fully on and operating. Since there is no voltage drop due to load 300, power supply voltage VCC is directly applied between the collector electrode and the emitter electrode of IGBT 200. Thereby, the collector current of the IGBT 200 starts to rise.

  The collector current rises above the threshold current and above the saturation current. At time t10 when the collector current reaches the threshold current, the short circuit detection circuit 30 outputs a detection signal to the control unit 10. The control unit 10 is triggered by the input of the detection signal, and turns off the application of the voltage to the second gate electrode 212 in consideration of the possibility of a short circuit of the load 300. That is, the normal mode is shifted to the safe mode. In this example, since the occurrence of a short circuit at time t9 is assumed, the collector current continues to be equal to or higher than the threshold current, so the control unit 10 continues to receive the detection signal. The control unit 10 determines that there is a short circuit at time t11 in FIG. 7 on condition that the input of the detection signal continues for a predetermined time, and stops the application of the voltage to the first gate electrode 211. After time t11, both the first gate electrode 211 and the second gate electrode 212 are in the off state, and the control unit 10 has shifted to the stop mode.

  In this example, since a short circuit is detected in the normal mode, a voltage is applied to both the first gate electrode 211 and the second gate electrode 212 immediately after the detection. However, the application of the voltage to the second gate electrode 212 is canceled by the collector current becoming higher than the threshold current and shifting to the safe mode. The saturation current of the IGBT 200 is smaller in the state applied to only the first gate electrode 211 as compared with the state in which the voltage is applied to both the first gate electrode 211 and the gate electrode 212. Therefore, the power consumption of the IGBT 200 in the safe mode can be reduced. Thus, the amount of heat generated by the IGBT 200 can be reduced, and thermal destruction can be prevented. Then, if it is determined that the short circuit state exists after the shift to the safe mode, the IGBT 200 is turned off as it is, and the drive by the drive device 100 is safely stopped.

<At turn-off>
For turn-off of the IGBT 200, application of voltage to the first gate electrode 211 and the second gate electrode 212 is turned off from the normal mode state. As the timing of turning off the application of the voltage, the first gate electrode 211 and the second gate electrode 212 may be turned off synchronously, or may be turned off from each other.

  Next, with reference to FIGS. 8-18, the effect of the drive device 100 which concerns on this embodiment is demonstrated. A shown in FIG. 8 shows the characteristics of the normal mode in the present embodiment, and B shows the characteristics of the safe mode. Moreover, C has shown the characteristic in the voltage clamp system of patent document 3, for example. FIG. 8 is a schematic view, and the scale of each axis is not exact.

  Control unit 10 in drive device 100 has a normal mode and a safe mode as operation modes of IGBT 200. As described above, in the present embodiment, the state in which the voltage is applied only to the first gate electrode 211 is the safe mode, and the state in which the voltage is applied to both the first gate electrode 211 and the second gate electrode 212 is the normal mode is there.

  At the time of occurrence of a short circuit of the load 300, the collector-emitter voltage becomes larger than in the state where no short circuit occurs, so that the collector current as the output current is saturated as shown in FIG. Since this saturation current is smaller in the safe mode than in the normal mode, when the IGBT 200 is driven in the safe mode when a short circuit occurs, the saturation current of the IGBT 200 can be suppressed, and the collector current and the heat generation amount of the IGBT 200 can be suppressed. Then, in the normal mode, the IGBT 200 can be driven with a larger current capability and a lower on voltage than in the safe mode. That is, in the normal mode, the loss can be suppressed because of the low on-state voltage, and in the safe mode, the short circuit tolerance can be secured by the low saturation current.

  As in the conventional configuration shown in FIG. 8C, although the voltage applied to the gate electrode 210 can be clamped to a predetermined voltage smaller than that at full on, the saturation current can be suppressed, but two power supply circuits are required. As a result, the circuit scale increases. On the other hand, in the configuration as in this embodiment, it is possible to suppress the saturation current at the time of short circuit without adding a power supply circuit. In addition, when suppression of the saturation current is insufficient with the clamp method alone, suppression of the saturation current can be realized by combining, for example, the technique of Patent Document 2, but at that time the on voltage is sacrificed. Such a contradiction does not occur if the drive device 100 in this embodiment is adopted.

  By the way, generally, in a drive circuit for driving a switching element, an output stage (portions shown by switches S1 to S4 in FIG. 3) has a relatively high current capability for high speed switching for the purpose of reducing switching loss. Requires a large output stage. Furthermore, in order to reduce the loss, there are cases where the temperature characteristic adjustment circuit of the gate potential Vg or the function of making the current for charging / discharging the gate electrode 210 a constant current may be provided. On the other hand, by dividing the gate electrode 210 into a plurality of parts as in this embodiment, the gate electrode 210 that turns on the high-performance output stage having a relatively large current capability first, that is, the first gate electrode 211. Only the output stage related to. In the above description, it is described that the first gate electrode 211 and the second gate electrode 212 may be shifted and turned off at the time of turn-off, but in such a configuration, only one of the output stages on the off side is also described. It is sufficient if the output stage has a relatively large current capacity and a high function.

  In addition, as in the present embodiment, when the main surface 290 a of the semiconductor substrate 290 is divided into the first region 231 and the second region 232 and these areas have substantially the same area, the voltage to the first gate electrode 211 is The switches S1 and S2 that control the application of V.sub.2 can be half the current capability as compared to the embodiment in which the gate electrode 210 is not divided. Thereby, for example, the physique of the driver IC can be reduced.

  Further, the IGBT 200 in the present embodiment has a first region 231 and a second region 232 as regions serving as a current path of the collector current. Since the first region 231 and the second region 232 are adjacent to each other and extend in a stripe shape, for example, when the first region 231 and the second region 232 divide the main surface 290 a of the semiconductor substrate 290 into two. In comparison, the temperature rise of the IGBT 200 can be reduced. Hereinafter, this will be specifically described with reference to FIGS. 9 to 16.

  The inventor conducted a thermal simulation on an IGBT 200 in which an insulated gate bipolar transistor is formed on a semiconductor substrate 290. The simulation is a one-dimensional simulation for calculating the heat transfer in the direction in which the areas 231 and 232 are aligned.

  FIG. 9 is a diagram showing the distribution of temperature change ΔT after a predetermined time has elapsed, when a heat source is present at a certain point. The broken line in FIG. 9 indicates ΔT with respect to the position on the main surface 290a. As shown in FIG. 9, let W be the distance from the heat source to the place where the temperature change occurs. The distance W is a distance affected by temperature rise due to heat transfer from a heat source, and is hereinafter referred to as the heat transfer distance W. The heat transfer distance W depends on the thermal property value of the material constituting the semiconductor substrate 290.

  FIG. 10 shows the temperature change ΔT after the elapse of a predetermined time when the heat source emitting a predetermined amount of heat per unit time is uniformly distributed on the main surface 290 a of the semiconductor substrate 290 with respect to the position on the main surface 290 a. And a plot. The broken line shown in FIG. 10 is the same as the broken line in FIG. 9, and the superposition of the temperature change by the heat source existing on the main surface 290a is the temperature change ΔT of the entire main surface 290a. The solid line in FIG. 10 indicates the overall temperature change ΔT. As in the present embodiment, in the structure in which the main surface 290a of the semiconductor substrate 290 is not divided into the first region 231 and the second region 232, that is, in the conventional configuration, the temperature change distribution shown by the solid line in FIG.

  FIG. 11 shows the case where the IGBT 200 has the first region 231 and the second region 232 described above, a voltage is applied only to the first gate electrode 211 and the first region 231 generates heat, LL2 W. Is a diagram showing a temperature change ΔT after the elapse of a predetermined time when the above condition is satisfied. As described above, the width in the direction orthogonal to the extending direction of the first regions 231 (the direction in which the gate electrodes 210 are arranged) is L, and the separation distance between the adjacent first regions 231 in the arranging direction is D. . Further, although D = L in the present embodiment, in the following, in order to explain the function and effect of the IGBT 200 having the first region 231 and the second region 232, modes other than D = L will be described.

  As shown in FIG. 11, under the condition of L ≧ 2 W, the peak of ΔT shows the same value as the condition of FIG. That is, even in the case where the IGBT 200 has the first region 231 and the second region 232 described above, it is equivalent to the conventional one in terms of temperature rise. Since the heat transfer distance W is sufficiently smaller than the size of the semiconductor substrate 290, the condition of L ≧ 2 W is, for example, that the first region 231 and the second region 232 divide the main surface 290a of the semiconductor substrate 290 into two. It is in the state of

  FIG. 12 is a diagram showing a temperature change ΔT after a predetermined time when the D / L ratio is larger than that in FIG. 11 and L <2W is satisfied. Under this condition, since the width L of the first region 231 serving as the heat source is smaller than the condition of FIG. 11, the peak due to the superposition of the temperature change due to the heat source is smaller than the condition shown in FIG. That is, the temperature rise of the IGBT 200 can be reduced as compared to the case where the first region 231 and the second region 232 bisect the main surface 290 a of the semiconductor substrate 290. The temperature change distribution in FIG. 12 is a distribution under the condition satisfying the relation of D> W−L / 2.

  Even in a state where a voltage is applied only to the first gate electrode 211, such as in the safe mode, it is preferable that the ratio of the first region 231 to the main surface 290a be as large as possible in order to secure a sufficient collector current. . In other words, it is preferable to reduce D while satisfying L <2W. FIG. 13 is a diagram showing a temperature change ΔT after lapse of a predetermined time when D is reduced from the condition of FIG. 12 while satisfying L <2W and D> W−L / 2. Since D> W−L / 2 is satisfied, heat is not transferred from the adjacent first region 231, and the peak value of the temperature change ΔT is equivalent to the condition in FIG.

  FIG. 14 is a diagram showing a temperature change ΔT after a predetermined time elapses of the condition satisfying D ≦ W−L / 2 by further reducing D from the condition of FIG. Since D> W−L / 2, which is a condition of dimensions in FIG. 13, is not satisfied, the heat transfer from the adjacent first regions 231 reaches a certain first region 231 and causes thermal interference. For this reason, the peak value of the temperature change ΔT is larger than the condition of FIG. However, since L <2W is satisfied, the temperature rise of the IGBT 200 can be reduced as compared to the case where the first region 231 and the second region 232 bisect the main surface 290 a of the semiconductor substrate 290. Furthermore, since D / L can be made smaller than the conditions of FIG. 13, even in the state where a voltage is applied only to the first gate electrode 211, the collector current to be rated is secured even with a smaller chip area. Can. In addition, when there is a request for suppressing the temperature rise as compared with the condition of FIG. 14, L may be designed to be small. FIG. 15 is a diagram showing a temperature change ΔT under the condition in which L is reduced from the condition of FIG. As shown in FIG. 15, by reducing the width L of the first region 231, the peak value of the temperature change ΔT can be made smaller than the condition of FIG.

  The simulation results under the respective conditions regarding the dimensions of L, D and W described above will be comprehensively described with reference to FIG. FIG. 16 is a diagram in which the horizontal axis is L and the vertical axis is D, and the dimension conditions of L and D obtained by the simulation results shown in FIGS. 11 to 15 are plotted. Each point of a to e illustrated in FIG. 16 is a dimension condition of L and D from which the simulation result illustrated in FIGS. 11 to 15 is obtained.

  In region I defined by L ≧ 2 W, ΔT does not depend on L and D, and becomes equal to ΔT in the prior art that generates heat uniformly. In the region II defined by L <2W and D> W−L / 2, ΔT decreases as L decreases, but does not depend on D. In the region III defined by D ≦ W−L / 2, ΔT decreases as L decreases, but ΔT increases as D decreases. That is, the temperature rise of IGBT 200 can be reduced by setting the dimensions of L, D and W to the conditions according to region II and region III. Furthermore, by setting the dimensions of L, D and W to the conditions according to region III, a chip with a sufficient collector current smaller even in the state where a voltage is applied only to the first gate electrode 211, such as in safe mode. The temperature rise of the IGBT 200 can be suppressed while ensuring the area.

  Subsequently, the heat transfer distance W will be examined with reference to FIGS. 17 and 18.

  FIG. 17 shows the result of simulation in the case where a voltage is applied only to the first gate electrode 211 using a semiconductor substrate 290 containing silicon as its main component. The horizontal axis is a stripe width a (= L = D), and the vertical axis is a temperature change ΔT at a position where the temperature rises most, that is, the center of the first region 231. A series of plots is an elapsed time t from the voltage application to the first gate electrode 211. Also, it is plotted at 5 μs which is generally required from the occurrence of a short to the activation of the protection function. In the vertical axis, the magnitude of the temperature change ΔT when a = 1000 μm and t = 5 μs is normalized to 1. According to FIG. 17, the smaller the stripe width a, the more the temperature rise of the IGBT 200 can be suppressed, but there is almost no fluctuation of the temperature change ΔT until it falls below a = 200 μm. As described above based on FIGS. 10 to 12, in view of no change in ΔT until L falls below 2 W, the heat transfer distance W in silicon is approximately 100 μm.

FIG. 18 shows the result of simulation in the case where a voltage is applied only to the first gate electrode 211 using the semiconductor substrate 290 containing silicon carbide (SiC) as the main component. The way of taking the vertical axis and the horizontal axis conforms to FIG. According to FIG. 18, the shapes of the plots are substantially similar to the results when the main component of the semiconductor substrate 290 is silicon and the results when the main component is SiC. In other words, when the plot showing the result when the main component is SiC is expanded or reduced in the horizontal axis direction, it almost agrees with the result when the main component is silicon. The enlargement ratio or reduction ratio in the horizontal axis direction depends on the thermal property value of the material constituting the semiconductor substrate 290. Specifically, the coefficient r represented by Equation 1 is a factor of the enlargement ratio in the horizontal axis direction.

Here, c is the specific heat, ρ is the density, and k is the thermal conductivity. As shown in FIG. 17, when the main substance constituting the semiconductor substrate 290 is silicon, the heat transfer distance W is approximately 100 μm, so the specific heat, density and thermal conductivity in silicon are cSi, SiSi, kSi, respectively. Then, the heat transfer distance W in a general substance can be expressed by Equation 2. For example, when the semiconductor substrate 290 is configured mainly of SiC, the heat transfer distance W obtained by Equation 2 is approximately 158 μm.

  Further, it is preferable to satisfy the relationship of L = D as in the present embodiment. In the present embodiment, since the first gate electrode 211 and the second gate electrode 212 are formed substantially equivalent to each other, the first region 231 and the second region 232 are substantially equivalent to each other, and L = D is satisfied. In such a configuration, since the first gate electrode 211 and the second gate electrode 212 are formed substantially equivalent to each other, the manufacturing process can be simplified.

(Modification 1)
In the embodiment described above, as illustrated in FIG. 3, the first region 231 and the second region 232 are illustrated as extending in one direction while satisfying L = D. However, it is not limited to the above-mentioned form.

  The width of the first area 231 and the separation distance D of the first area 231 do not have to be constant for all the first areas 231. For example, as shown in FIG. 19, D may be set relatively large in some of the sections on the main surface 290a, and L may be set relatively small in another section. Also good.

  Further, as shown in FIG. 20, a plurality of structures in which the first region 231 and the second region 232 are repeatedly formed may be formed.

  In addition, the stripe shape formed by the first region 231 and the second region 232 does not have to point in only one direction for all the first regions 231 and the second regions 232. For example, as illustrated in FIG. 21, the direction in which the stripes face may be orthogonal to each other on a part of the main surface 290 a and another part. In such a case, it is preferable that the first regions 231, which are energized in both the normal mode and the safe mode and have a relatively large amount of heat generation, are not adjacent to each other at the boundary between the two sections. Thereby, the temperature rise in the boundary of both divisions can be suppressed.

  The formation of the first region 231 and the second region 232 in a stripe shape also includes the case where the first region 231 and the second region 232 are repeatedly arranged in the longitudinal direction of the regions 231 and 232.

Second Embodiment
The second off-side circuit in the first embodiment corresponds to the switch S4 of the second drive circuit 22. On the other hand, the second drive circuit 22 in the present embodiment has a second soft shutoff circuit 24 as the second off-side circuit 23 in addition to the switch S4, as shown in FIG. The second soft shutoff circuit 24 is connected to the second gate electrode 212 in parallel with the resistor R3. The second soft shutoff circuit 24 has a resistor R5 and a switch S5, which are connected in series between the second gate electrode 212 and the reference potential VSS. The switch S5 is turned on when the short detection circuit 30 receives a detection signal.

  The operation of the second soft shutoff circuit 24 will be described below with reference to FIG. 23, taking a case where a short circuit is detected when the IGBT 200 is on.

  As shown in FIG. 23, it is assumed that a short circuit occurs in load 300 at time t13 during which IGBT 200 is fully on and in operation. The power supply voltage VCC is directly applied between the collector electrode and the emitter electrode of the IGBT 200, and the collector current of the IGBT 200 starts to rise.

  The collector current exceeds the threshold current and rises to the upper limit of the saturation current. At time t14 when the collector current reaches the threshold current, the short circuit detection circuit 30 outputs a detection signal to the control unit 10. The control unit 10 is triggered by the input of the detection signal, and turns off the application of the voltage to the second gate electrode 212 in consideration of the possibility of a short circuit of the load 300. That is, the normal mode is shifted to the safe mode.

  By the way, at the time of transition from the normal mode to the safe mode, in the configuration without the second soft shutoff circuit 24, the second gate electrode via the resistor R3 by turning on the switch S4 substantially simultaneously with the turning off of the switch S3. Pull out the charge of 212. On the other hand, in the embodiment where the second off-side circuit 23 includes the second soft shutoff circuit 24 as in the present embodiment, as shown in FIG. 13, the control unit 10 keeps the switch S4 off. Then, instead of the switch S4, the switch S5 is turned on by the input of the detection signal. As a result, the charge accumulated in the second gate electrode 212 escapes to the reference potential VSS side via the resistor R5.

  In this example, the occurrence of a short circuit at time t13 is assumed, so the collector current is maintained at or above the threshold current. Therefore, the detection signal continues to be input to the control unit 10. The control unit 10 determines that the load 300 is in the short circuit state at time t15 on condition that the input of the detection signal continues for a predetermined time, and stops the application of the voltage to the first gate electrode 211. After time t15, both the first gate electrode 211 and the second gate electrode 212 are in the off state, and the control unit 10 has shifted to the stop mode.

  The operation and effect of the drive device 100 in the present embodiment will be described below.

  When a short circuit occurs in the normal mode and the rise of the collector current is fast, an overcurrent flows before the transition to the safe mode. Thereafter, application of a voltage to the second gate electrode 212 is turned off to shift to the safe mode, but if the off speed is too fast, the collector current sharply decreases, resulting from the induced electromotive force by the load 300. As a result, excessive surge may occur.

  In the drive device 100 according to the present embodiment, since the second drive circuit 22 includes the second soft shutoff circuit 24, the drive device 100 is applied to the second gate electrode 212 by appropriately setting the resistance value of the resistor R5. The slew rate of the voltage can be increased, and excessive surge can be suppressed.

  It should be noted that the charge extraction via the second soft shutoff circuit 24 may be performed only when the load 300 is short-circuited, and in a normal turn-off operation, the charge may be extracted via the switch S4. According to this, the slew rate of the voltage at the time of short circuit depends only on the resistance value of the resistor R5, and the resistance value of the resistor R3 can be set independently. In other words, the speed of transition between the normal mode and the safe mode can be set independently of the speed of transition from the normal mode to the safe mode at the time of short circuit. That is, surges due to large current changes can be suppressed without sacrificing the speed of transition between the normal mode and the safe mode.

Third Embodiment
The first off-side circuit in the first embodiment and the second embodiment corresponds to the switch S2 of the first drive circuit 21. On the other hand, the first drive circuit 21 in the present embodiment has a first soft shutoff circuit 26 as the first off-side circuit 25 in addition to the switch S2, as shown in FIG. The first soft shutoff circuit 26 is connected to the first gate electrode 211 in parallel with the resistor R2. The first soft shutoff circuit 26 has a resistor R6 and a switch S6, which are connected in series between the first gate electrode 211 and the reference potential VSS. The switch S6 is turned on when a short detection circuit 30 receives a detection signal.

  The operation of the first soft shutoff circuit 26 will be described below with reference to FIG. 25, taking a case where a short circuit is detected when the IGBT 200 is in the on state.

  As shown in FIG. 25, it is assumed that a short circuit occurs in load 300 at time t17 when IGBT 200 is fully on and in operation. The power supply voltage VCC is directly applied between the collector electrode and the emitter electrode of the IGBT 200, and the collector current of the IGBT 200 starts to rise.

  The collector current exceeds the threshold current to reach the saturation current. At time t18 when the collector current reaches the threshold current, the short circuit detection circuit 30 outputs a detection signal to the control unit 10. The control unit 10 is triggered by the input of the detection signal, and turns off the application of the voltage to the second gate electrode 212 in consideration of the possibility of a short circuit of the load 300. That is, the normal mode is shifted to the safe mode. The control unit 10 turns on the switch S4 substantially simultaneously with the turning off of the switch S3. Thereby, the charge of the second gate electrode 212 is drawn out.

  In this example, the occurrence of a short circuit at time t17 is assumed, so the collector current maintains a saturation region. Therefore, the detection signal continues to be input to the control unit 10. The control unit 10 determines that the load 300 is in a short circuit state at time t19 on condition that the input of the detection signal continues for a predetermined time, and stops the application of the voltage to the first gate electrode 211. That is, it shifts to the stop mode.

  By the way, at the time of transition from the safe mode to the stop mode, in the configuration without the first soft shutoff circuit 26, the first gate electrode via the resistor R2 by turning on the switch S2 substantially simultaneously with the turning off of the switch S1. Pull out the charge of 211. On the other hand, in the embodiment in which the first off-side circuit 25 includes the first soft shutoff circuit 26 as in the present embodiment, as shown in FIG. 15, the control unit 10 keeps the switch S2 off. Then, instead of the switch S2, the switch S6 is turned on by the input of the detection signal. As a result, the charge stored in the first gate electrode 211 escapes to the reference potential VSS via the resistor R6.

  Then, after time t19, both the first gate electrode 211 and the second gate electrode 212 are in the off state, and the control unit 10 is in the state of shifting to the stop mode.

  The operation and effect of the drive device 100 in the present embodiment will be described below.

  In the drive device 100 according to the present embodiment, since the first drive circuit 21 includes the first soft shutoff circuit 26, the drive device 100 is applied to the first gate electrode 211 by appropriately setting the resistance value of the resistor R6. The slew rate of the voltage can be increased, and excessive surge can be suppressed.

  It is preferable that the charge extraction via the first soft shutoff circuit 26 be performed only when the load 300 is shorted, and that the charge be extracted via the switch S2 in the normal turn-off operation. According to this, the slew rate of the voltage at the time of short circuit depends only on the resistance value of the resistor R6, and the resistance value of the resistor R2 can be set independently. In other words, it is possible to set the turn-off speed during normal driving in which a short circuit does not occur and the turn-off speed during a short circuit independently. Therefore, it is possible to suppress a surge due to a large current change at the time of a short circuit.

(Other embodiments)
The preferred embodiments of the present invention have been described above, but the present invention is not limited to the above-described embodiments and can be variously modified and implemented without departing from the spirit of the present invention.

  For example, as shown in FIG. 26, a voltage adjustment circuit 27 such as a clamp circuit may be constructed in the drive circuit 20 in each of the above-described embodiments. The voltage adjustment circuit 27 suppresses the voltage applied to the gate electrode 210 to be lower than that during normal driving when a short circuit occurs. By the safe mode by the control unit 10 and the adjustment of the gate voltage by the voltage adjustment circuit 27, it is possible to further suppress the saturation current at the time of the short circuit as compared with the above-described embodiments. Although FIG. 26 shows an example in which the voltage adjustment circuit 27 is connected only to the first gate electrode 211 to which a voltage is applied even in the safe mode, both of the first gate electrode 211 and the second gate electrode 212 are shown. It may be connected.

  Moreover, it is preferable that the area of the effective area in the first area 231 and the second area 232 be appropriately set. Here, the effective region is a portion serving as a carrier injection source when a voltage is applied to the gate electrode 210, and corresponds to an emitter region.

  When the main surface 290a of the semiconductor substrate 290 is viewed from the front, the area ratio of the effective region in the first region 231 is α1, and the area ratio of the effective region in the second region 232 is α2, and functions as an insulated gate bipolar transistor in the main surface 290a. The area ratio of the first area 231 to the area to be formed is β. That is, the area ratio of the effective area of the first area is α1 × β to the area functioning as an insulated gate bipolar transistor on the main surface 290a, and the area ratio of the effective area of the first area is α2 × (1−β ).

  In general, the saturation current monotonously increases with respect to the area of the effective area, and the on voltage monotonously decreases with respect to the area of the effective area. In the safe mode, it is necessary to flow a current equal to or higher than the rated current, and for securing short circuit withstand voltage. It is necessary to make it smaller than a predetermined current value. That is, α1 × β is determined by the requirements of the entire system including the connected load 300 and the specifications such as the standard.

  On the other hand, in the normal mode, it is preferable that the ratio of the area occupied by the effective region, α1 × β + α2 × (1−β), be as large as possible, since decreasing the on-state voltage as much as possible leads to reduction of the loss. Since the first term α1 × β is determined by the specification of the system as described above, it is required to make the second term larger. For example, the maximum value α2 = 1 is adopted for α2 not related to the short circuit withstand capacity, and α1> β is satisfied by satisfying the restrictions of the rated current and the short circuit capacity while satisfying α1 × β + α2 × (1− β) can be set as large as possible. As a result, it is possible to secure a short circuit withstand voltage while achieving a low on-state voltage.

  Further, in each of the embodiments and the modifications described above, the example in which the common gate potential Vg is applied to the first gate electrode 211 and the second gate electrode 212, though the resistors R1 and R3 are described. However, power supplies of different systems may be prepared for the first gate electrode 211 and the second gate electrode 212, respectively.

  For example, by applying a negative voltage to the gate electrode 210 to be turned off, a storage layer is formed in a region adjacent to the insulating film of the gate electrode 210, so that carriers are discharged from the emitter electrode. It can be made easy.

  In addition, for the threshold current described in each of the embodiments and modifications described above, a first threshold in the normal mode and a second threshold in the safe mode are prepared, and different values are used for the first threshold and the second threshold. It is preferable to set.

  Specifically, the first threshold in the normal mode is set larger than the second threshold in the safe mode. In the normal mode, the current capability is larger than in the safe mode, and the saturation current at the short circuit time is larger. By setting the first threshold to be larger than the second threshold, it is possible to secure a noise margin as well as detect a short circuit in the normal mode.

  Now, in each of the embodiments and the modifications described above, the sense emitter terminal to which the short circuit detection circuit 30 is connected is briefly described. Specifically, as shown in FIG. 27, the IGBT 200 has a main cell 200a for extracting an output current and a sense cell 200b having a sense emitter terminal for extracting a sense current.

  The main cell 200a has a first gate electrode 211a and a second gate electrode 212a as a gate electrode 210, and the sense cell 200b has a first gate electrode 211b and a second gate electrode 212b as the gate electrode 210. The gate electrodes 211b and 212b divided on the sense cell 200b side correspond to sense cell side divided electrodes described in the claims. The first gate electrodes 211 a and 211 b are connected to the first drive circuit 21. The second gate electrodes 212 a and 212 b are connected to the second drive circuit 22.

  When a voltage is outputted from the first drive circuit 21 or the second drive circuit 22 to the gate electrode 210, a collector current which is an output current of the main cell 200a flows and a sense current flows in the sense cell 200b. The sense current is correlated with the output current. The shunt resistor R4 of the short circuit detection circuit 30 is connected between the sense cell 200b and the reference potential VSS. The short circuit detection circuit 30 detects a short circuit by converting a sense current output from the sense cell 200b into a voltage by the shunt resistor R4.

  The sense cell 200b does not have to have two of the first gate electrode 211b and the second gate electrode 212b as the gate electrode 210. As shown in FIG. 28, the sense cell 200b may be configured of only one gate electrode 210, for example, the first gate electrode 211b. However, voltage must be applied to the gate electrode 210 of the sense cell 200b in both the normal mode and the safe mode. According to this, since the fluctuation of the sense current does not occur at the time of switching between the normal mode and the safe mode, stable detection of the sense current can be performed. In the short circuit detection circuit 30 shown in FIGS. 27 and 28, a circuit diagram on the downstream side of the subtraction circuit 31 is omitted.

  Moreover, although the example which the gate electrode 210 of IGBT200 is divided | segmented into two division electrodes was shown in each above-mentioned embodiment and modification mentioned above, a division | segmentation number may be three or more. In FIG. 29, the gate electrode 210 is divided into three, ie, a first gate electrode 211, a second gate electrode 212, and a third gate electrode (the third gate electrode is not shown), as divided electrodes. An example is shown in which a first area 231, a second area 232, and a third area 233, through which an output current flows, are formed by applying a voltage.

  In this example, the first area 231, the second area 232, and the third area 233 are repeatedly formed side by side in this order, forming a stripe. Also in this example, if a voltage is applied to only the first gate electrode 211 in the safe mode, the separation distance D described in the claims is the total value of the widths of the second region 232 and the third region 233. It corresponds to

  In such a mode, for example, a voltage can be applied to the first gate electrode 211 in the safe mode, and a voltage can be applied to the second gate electrode 212 in addition to the first gate electrode 211 in the normal mode. In addition to the first gate electrode 211, a voltage may be applied to the third gate electrode in the normal mode. Furthermore, in addition to the first gate electrode 211, a voltage can be applied to the second gate electrode 212 and the third gate electrode in the normal mode. In addition to the first gate electrode 211, a voltage may be applied to one of the second gate electrode 212 and the third gate electrode in the safe mode. The combination of the gate electrodes 210 for applying a voltage in each mode is arbitrary, and a combination according to the specifications of the rated current for driving the load 300 and the saturation current at the time of short circuit can be selected. For example, since the current flowing through the load 300 does not require a current rating equal to or higher than the command value, a method of making the area to be driven in the safe mode variable according to the command value is effective. Alternatively, since the power at the time of short circuit depends on the power supply voltage VCC, if this voltage is high, it is also effective to reduce the area of the region through which current flows in the safe mode.

  In each embodiment and modification which were mentioned above, although IGBT200 was described as an example as a switching element, a switching element is not limited to an insulated gate bipolar transistor. For example, MOSFETs, bipolar transistors, and other generally known elements can be employed. When a MOSFET is employed, the collector in the description of each of the embodiments and the modifications may be paraphrased as a drain and the emitter may be rephrased as a source. When a bipolar transistor is employed, it may be paraphrased as a gate.

  In each of the above-described embodiments and modifications, as shown in FIG. 1, the method of driving the load 300 by one switching element is described as an example, but the method of forming the upper and lower arms by a plurality of switching elements is adopted. It can also be done.

  In each embodiment and modification described above, the short detection circuit 30 for detecting a short of the load 300 is connected to the sense emitter terminal, but the sense emitter terminal, in other words, the sense cell 200b is not necessarily required. There is no need to detect through. For example, the short circuit detection circuit 30 may be constructed at the output terminal of the main cell 200a. As an example of the short circuit detection circuit 30, a method of inserting a shunt resistor or a method of inserting a current sensor can be adopted. In addition, the short circuit may be detected based on the collector-emitter voltage Vce of the IGBT 200 or the time change dVce / dt thereof. In this case, if Vce = VCC when the IGBT 200 is on, it can be determined as a short circuit. Alternatively, the potential Vge of the gate electrode 210 may be detected, and a short circuit may be detected based on the presence or absence of the mirror period.

  In the above description, although a short circuit has been described as a short circuit in load 300, the present invention is not limited to a load short circuit, but all short circuits in which current flowing in the switching element significantly increases, such as arm short circuit by diode 400 short circuit. It can be applied to the mode.

DESCRIPTION OF SYMBOLS 100 ... Drive device, 10 ... Control part, 20 ... Drive circuit, 30 short circuit detection circuit, 200 ... IGBT (switching element), 210 ... Gate electrode, 211 ... 1st gate Electrode (first electrode), 212: second gate electrode (second electrode) 300: load

The present invention provides a drive device of a power switching element capable of suppressing the temperature rise of the semiconductor substrate when a voltage is applied to only one of the divided electrodes obtained by dividing the control electrode of the power switching element. With the goal.

In order to achieve the above object, the present invention is a drive device for driving a power switching element (200) having a control electrode (210) that controls an output current supplied to a load by applying a voltage. ,
A control unit (10) for controlling the timing of voltage application to the control electrode;
The control electrode is divided into two or more split electrodes including a first electrode (211) and a second electrode (212),
The control unit has a timing of applying a voltage only to the first electrode among the two or more divided electrodes, and a timing of applying a voltage to both the first electrode and the second electrode,
The divided electrodes are extended in a predetermined direction orthogonal to the depth direction on at least a part of the main surface (290a) orthogonal to the depth direction of the semiconductor substrate (290) on which the power switching element is formed,
When the main surface is viewed from the front, the main surface is output by applying a voltage to the first region (231) through which an output current flows when the voltage is applied to the first electrode, and an output when the voltage is applied to the second electrode A second region (232) through which current flows;
The first region and the second region are formed in a stripe shape along the extension direction of the divided electrodes,
The first electrodes are arranged in a direction perpendicular to the extending direction,
Assuming that the specific heat of the material constituting the semiconductor substrate is c, the density is ρ, the thermal conductivity is k, the specific heat of silicon is cSi, the density is SiSi, and the thermal conductivity is kSi It is characterized in that it is formed to satisfy the relationship of L <2W between W and the width L of the first region.

As described above, by suppressing the temperature rise of the semiconductor substrate, the width L of the first region (231) serving as a heat source when a voltage is applied only to the first electrode satisfies the relationship of L <2W. Becomes possible.

Claims (12)

A driving device for driving a power switching element (200) having a control electrode (210) for controlling an output current supplied to a load by applying a voltage,
A control unit (10) for controlling the timing of voltage application to the control electrode;
The control electrode is divided into two or more divided electrodes including a first electrode (211) and a second electrode (212),
The control unit
Driving the power switching element in a safe mode in which a voltage is applied to one of the first electrode and the second electrode in a short circuit detection period for detecting a short circuit of the load;
Driving the power switching element in a normal mode in which a voltage is applied to both the first electrode and the second electrode in a normal driving period for normally driving the load;
The divided electrodes are extended in a predetermined direction orthogonal to the depth direction on at least a part of the main surface (290a) orthogonal to the depth direction of the semiconductor substrate (290) on which the power switching element is formed. ,
When the main surface is viewed from the front, a voltage is applied to the first electrode (231) through which an output current flows when the voltage is applied to the first electrode, and a voltage is applied to the second electrode on the main surface. And a second region (232) through which the output current flows.
The first region and the second region are formed in a stripe shape along the extension direction of the divided electrodes.
The control unit applies a voltage to the first electrode in the safe mode to drive the power switching element.
The first electrodes are arranged in a direction perpendicular to the extending direction,
Assuming that the specific heat of the material constituting the semiconductor substrate is c, the density is ρ, the thermal conductivity is k, the specific heat of silicon is cSi, the density is SiSi, and the thermal conductivity is kSi, The driving device is characterized in that a relationship of L <2W is satisfied between the distance W and the width L of the first region.

  The first electrode is formed to satisfy the relationship of D ≦ W−L / 2 between the width L of the first region, the separation distance D of the first region, and the heat transfer distance W. The drive device according to claim 1, A driving device for driving a power switching element (200) having a control electrode (210) for controlling an output current supplied to a load by applying a voltage,
A control unit (10) for controlling the timing of voltage application to the control electrode;
The control electrode is divided into two or more divided electrodes including a first electrode (211) and a second electrode (212),
The control unit
Driving the power switching element in a safe mode in which a voltage is applied to one of the first electrode and the second electrode in a short circuit detection period for detecting a short circuit of the load;
Driving the power switching element in a normal mode in which a voltage is applied to both the first electrode and the second electrode in a normal driving period for normally driving the load;
Furthermore, a first drive circuit (21) for applying a voltage to the first electrode, and a second drive circuit (22) for applying a voltage to the second electrode;
A short circuit detection circuit (30) that outputs a detection signal to the control unit when a short circuit of the power switching element is detected;
The first drive circuit includes a first on-side circuit (S1) for turning on application of a voltage to the first electrode, and a first off-side circuit for turning off application of a voltage to the first electrode. (S2, 25), and
The second drive circuit includes a second on-side circuit (S3) for turning on application of a voltage to the second electrode, and a second off-side circuit for turning off application of a voltage to the second electrode. (S4, 23), and
In the case of turning on the power switching device,
The control unit turns on only the application of a voltage to the first electrode to drive in the safe mode.
When the detection signal is not input within the short circuit detection period, the voltage application to the second electrode is turned on to shift to the normal mode,
When the detection signal is input within the short circuit detection period, if the detection signal continues during the short circuit detection period, the voltage application to the second electrode is not turned on without turning on the first electrode. A driving device controlled to turn off application of voltage. A driving device for driving a power switching element (200) having a control electrode (210) for controlling an output current supplied to a load by applying a voltage,
A control unit (10) for controlling the timing of voltage application to the control electrode;
The control electrode is divided into two or more divided electrodes including a first electrode (211) and a second electrode (212),
The control unit
Driving the power switching element in a safe mode in which a voltage is applied to one of the first electrode and the second electrode in a short circuit detection period for detecting a short circuit of the load;
Driving the power switching element in a normal mode in which a voltage is applied to both the first electrode and the second electrode in a normal driving period for normally driving the load;
Furthermore, a first drive circuit (21) for applying a voltage to the first electrode, and a second drive circuit (22) for applying a voltage to the second electrode;
A short circuit detection circuit (30) that outputs a detection signal to the control unit when a short circuit of the power switching element is detected;
The first drive circuit includes a first on-side circuit (S1) for turning on application of a voltage to the first electrode, and a first off-side circuit for turning off application of a voltage to the first electrode. (S2, 25), and
The second drive circuit includes a second on-side circuit (S3) for turning on application of a voltage to the second electrode, and a second off-side circuit for turning off application of a voltage to the second electrode. (S4, 23), and
In the case where the detection signal is input to the control unit during a normal drive period in which a current is supplied to the first electrode and the second electrode to drive the power switching element,
The control unit turns off the application of the voltage to the second electrode and shifts to the safe mode.
If the detection signal continues during the short circuit detection period, the application of the voltage to the first electrode is turned off,
If the detection signal does not continue during the short circuit detection period, control is performed to switch on application of a voltage to the second electrode again to shift to the normal mode.   The third embodiment is characterized in that the second off-side circuit has a second soft shutoff circuit (24) which makes the time change of the output current gradual when the application of the voltage to the second electrode is turned off. Or the drive device of Claim 4.   The first off-side circuit includes a first soft shut-off circuit (26) which makes the time change of the output current gradual when the application of the voltage to the first electrode is turned off. Drive device of any one of -5.   At least one of the first drive circuit and the second drive circuit has a voltage adjustment circuit (27) for adjusting a voltage applied to the first electrode and the second electrode, respectively. The drive device according to any one of claims 3 to 6. A driving device for driving a power switching element (200) having a control electrode (210) for controlling an output current supplied to a load by applying a voltage,
A control unit (10) for controlling the timing of voltage application to the control electrode;
The control electrode is divided into two or more divided electrodes including a first electrode (211) and a second electrode (212),
The control unit
Driving the power switching element in a safe mode in which a voltage is applied to one of the first electrode and the second electrode in a short circuit detection period for detecting a short circuit of the load;
Driving the power switching element in a normal mode in which a voltage is applied to both the first electrode and the second electrode in a normal driving period for normally driving the load;
And a short circuit detection circuit (30) for outputting a detection signal to the control unit when a short circuit of the power switching element is detected.
The short circuit detection circuit outputs the detection signal when the current value of the input current is equal to or more than a predetermined threshold current.
The threshold current has a first threshold in the normal mode and a second threshold in the safe mode,
The second threshold is set to a value smaller than the first threshold. The plurality of control electrodes are formed on the main surface (290a) orthogonal to the depth direction of the semiconductor substrate (290) on which the power switching element is formed,
When the main surface is viewed from the front, a voltage is applied to the first electrode (231) through which an output current flows when the voltage is applied to the first electrode, and a voltage is applied to the second electrode on the main surface. And a second region (232) through which the output current flows.
The area occupied on the main surface of at least one of the first region and the second region is set such that a current equal to or greater than a predetermined rated current flows as an output current. The driving device according to any one of to 8.   The first area and the second area are set such that an area occupied on each of the main surfaces is set such that a current equal to or more than a predetermined rated current flows as an output current. The drive described. A short circuit detection circuit (30) for outputting a detection signal to the control unit when a short circuit of the power switching element is detected;
The power switching element includes a main cell (200a) generating the output current when a voltage is applied to the control electrode, and a sense cell (200b) for monitoring the output current of the power switching element.
The drive device according to any one of claims 1 to 10, wherein the current input to the short circuit detection circuit is a sense current output from the sense cell.   The control electrode of the sense cell has sense cell side divided electrodes (211b, 212b) divided corresponding to the divided electrodes in the main cell, and the sense cell side divided electrode corresponds to the divided electrodes in the main cell The driving device according to claim 11, wherein the driving device is driven.

Images