JPH07108098B2 - Power semiconductor module - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a power semiconductor module used in a power conversion circuit such as an inverter circuit or a chopper circuit.

[Conventional technology]

In recent years, the main elements used in power converters have changed from thyristors, which have only an ON function, to self-extinguishing semiconductor elements such as GTOs, bipolar transistors, IGBTs, Bi MOS transistors that have both ON functions. Have been replaced. Since this self-extinguishing type semiconductor element has a higher operating speed than a thyristor, in the case of a power conversion device using this as a main element, not only the power conversion efficiency can be increased and the component configuration can be simplified, but also the operation sequence can be improved. Can be simplified.

FIG. 4 is a circuit diagram showing the basic operation principle of a single-phase bridge voltage type PWM inverter which is an example of a typical power converter. In the figure, the switches A, B, C, D are each configured by connecting a flywheel diode in antiparallel to a baopolar transistor, which is a self-arc-extinguishing type semiconductor device described above. The series circuit and the series circuit of the switches C and D are connected in parallel, the constant voltage power source V is connected between both terminals of the parallel circuit, and the connection points of the switches A and B and the connection points of the switches C and D are connected. An inductance L is connected between them. The switches A and D and the switches B and C are paired to perform on / off operation. That is, when the transistor of switch A turns on and off, the transistor of switch D also turns on and off, and when the transistor of switch B turns on and off, switch C
Each transistor is controlled so that the transistor is turned on / off.

FIG. 5 is a timing chart showing the operation of the inverter, of which FIG. 5 (1) shows each of the switches A to
Waveform of the control signal that controls the on / off operation of D (modulation wave
a, signal waves b1, b2), FIG. 5 (2) shows the waveform of the output voltage E _out when the signal wave is at a high voltage, and FIG. 5 (3) is an output voltage when the signal wave is at a low voltage. The waveform of E _out is shown. Fifth
A section in which the level of the modulated wave a shown in FIG. 1A is lower than the levels of the signal waves b1 and b2 (for example, shown in FIGS. 5B and 5C).
In the section of t1 and t2), an ON signal is given to the bases of the transistors of the switches A and D, and the level of the modulation wave a changes to the signal waves b1 and b.
Sections higher than level 2 (eg, Figure 5 (2),
In the section of t3 and t4 shown in (3), an ON signal is given to the bases of the transistors of the switches B and C. When the amplitude is large as in the case of the signal wave b1, FIG. 5 (2)
The output power of the inverter becomes large as shown by the dotted line in FIG. 3 and when the amplitude is small as in the case of the signal wave b2, the output power of the inverter becomes small as shown by the dotted line in FIG. The frequency of the inverter output corresponds to the frequencies of the signal waves b1 and b2.

By the way, in the inverter circuit shown in FIG. 4, for example, from the state in which the transistors of the switches A and D are turned on and the load current shown by the arrow is flowing, the transistors of the switches A and D are turned off after a certain rest period. on the other hand,
It is assumed that an ON signal is given to the transistors of the switches B and C. At this time, the current flowing in the inductance L in the direction of the arrow decreases, so that a counter electromotive force is generated in the inductance L and a return current flows as shown by the arrow. This return current flows in the switches B and C through a flywheel diode connected in antiparallel to the transistors, so that the transistors in the switches B and C are applied with a voltage in the opposite direction to that in normal operation. Therefore,
During this period, the normal ON operation of the transistors of the switches B and C cannot be waited for, but nevertheless, the ON signal is input to these transistors (this condition will be referred to as " Called "blank mode"). When the period in which the ON signal is given to the transistors of the switches B and C is very short, as in the period t5 shown in Fig. 5 (2), the above-mentioned return current is generated even if the transistors of the switches B and C are turned on. The load current indicated by the arrow in FIG. 4 does not flow, and only the “idle hit mode” occurs. Even when the period in which the ON signal is applied to the transistors of the switches B and C is not short, the transistors of the switches B and C are in the steady state through the “idle driving mode”.

On the other hand, when the ON signal is applied again to the transistors of the switches A and D, these transistors shift to the ON operation while holding the voltage (power supply voltage) held at the OFF time. At this time, if the period in which the ON signal is applied to the transistors of the switches B and C is very short, a large current is passed through the flywheel diode connected in antiparallel to the transistor of the switch A and the transistor of switch B. Flows. Similarly, a large current flows through a flywheel diode connected in antiparallel to the transistor of switch C and the transistor of switch D. Because the flywheel diode of the switches B and C had a return current flowing at least immediately before this stage, the flywheel diode must be removed until the charge accumulated inside these flywheel diodes is removed. This is because the diode is considered to be in a short circuit state. At the end of the period when these flywheel diodes are short-circuited, the voltage held by the transistors of switches A and D until then is held by the transistors of switches B and C, This switching of the voltage sharing generally occurs in a very short time.

In the above operation, since a large current flows in the transistors of the switches A and D while the power supply voltage is held, it is expected that these transistors are likely to be destroyed. The safe operating area of the transistor in this mode is indicated by the index "short-circuit SOA".

However, in practice, it is empirically known that the operation in the “blank driving mode” is performed and the transistors on the switches B and C are much more likely to be damaged. It is also empirically known that at this time, in driving the bases of the transistors on the switches B and C, the breakdown is more likely to occur when the forward base current is large.

Note that when the transistors of switches B and C are turned on and the load current shown by the arrow is flowing, and the ON signal is given to the transistors of switches A and D, the "blank driving mode" is applied to the transistors of switches A and D. Resulting in a large current switch
It flows to the B and C transistors.

Experience such as increasing the transistor margin and adding a circuit to prevent the base current from flowing more than necessary as a measure to improve the breakdown tolerance of the transistor corresponding to the above-mentioned "blank driving mode" There was a specific response.

Although the analysis of the above-mentioned “blank-firing mode” operation has not been sufficiently performed so far, it is expected that the following mechanism is working.

In other words, in the "blank driving mode" operation, even if the collector current does not flow in the transistor, a state in which a slight amount of electric charge exists inside the transistor appears due to the forward base current provided (at least, the transistor is typically In a saturated state, which can be considered to be turned on, a given forward base current flows from the base to the collector). At the end of the period when the flywheel diode connected in anti-parallel to this transistor exhibits a short-circuit state, as described above, a high voltage equivalent to the power supply voltage is applied to such a transistor in an internal state in an extremely short time. Will be applied. In this way, the situation in which a voltage is suddenly applied to a transistor that holds electric charges inside is generally not very good for stable operation of the transistor, according to the reference <H. Nishiumi et al, “High Volt
age High Power Transistor Modules for 440V AC Line
Voltage Inverter Application "Conference record of
It is also known by IPEC Tokyo '83〉.

Considering such a mechanism of "blank mode",
As shown in FIG. 6, the structure of the base drive circuit that does not cause the "blank driving mode" also improves the transistor breakdown resistance.
It has been proposed as one measure. In FIG. 6, two NPN transistors Q1 and Q2 are connected in Darlington, and a flywheel diode D is connected in antiparallel between the collector and emitter of the NPN transistor Q1.
A current detection coil 1 for detecting a reverse current (reflux current) is inserted in the collector side electric circuit of the NPN transistors Q1 and Q2, and the base amplifier AMP is connected while the detection output of the current detection coil 1 indicates the reverse current. Deactivate and NPN transistor
It is configured so that no forward base current flows to Q2.

[Problems to be Solved by the Invention]

However, as described above, there is a problem in that the cost is increased by taking measures such as increasing the margin of the transistor or adding a circuit for preventing an unnecessary base current from flowing to the base driving circuit. .

Further, in the method using the current detection coil 1, a current detection coil 1 having a size capable of flowing a main current is required, and it takes time and effort to connect such a large coil for each transistor, and the cost also increases. There are problems such as.

The present invention has been made to solve such problems, and an object thereof is to obtain a power semiconductor module having a simple structure and capable of improving the breakdown resistance of a self-arc-extinguishing type semiconductor element.

[Means for Solving the Problem] A power semiconductor module according to claim 1 of the present invention is a self-extinguishing semiconductor element and a self-extinguishing element connected in antiparallel with the self-extinguishing semiconductor element. In a power semiconductor module including a flywheel diode that causes a return current to flow in a direction opposite to the conduction direction of the semiconductor element, the flywheel diode includes a shunt composed of a series circuit of another diode and a shunt resistor. Parallel to the self-arc-extinguishing type semiconductor element and connected in anti-parallel to the self-extinguishing semiconductor element, terminals are provided at both ends of the shunt resistor, and the presence or absence of the return current is detected from the polarity of the voltage between the terminals. And

A power semiconductor module according to a second aspect of the present invention is the power semiconductor module according to the first aspect, wherein the flywheel diode and the other diode are formed on a common semiconductor substrate. It is characterized by

[Action]

In the invention of claim 1, when the return current is flowing, the voltage across the terminals of the shunt resistor has a reverse polarity,
By controlling so that the ON signal is not applied to the self-arc-extinguishing semiconductor element during this polarity, the self-arc-extinguishing semiconductor element is prevented from being broken due to the return current.

In the invention described in claim 2, since the flywheel diode and the other diode are formed on a common semiconductor substrate, it is possible to easily improve the precision of the electrical characteristics and the precision of the dimensions of these diodes. You can

〔Example〕

FIG. 1 is a circuit diagram showing an embodiment of a power semiconductor module according to the present invention. This power semiconductor module is, for example, a circuit used as a switch of the above-described inverter circuit shown in FIG. 4, and flows a reflux current in antiparallel to the NPN transistor Q3 which is a self-extinguishing semiconductor element. Is connected to a flywheel diode D1 and a series circuit consisting of a second diode D2 and a shunt resistor R _S, which have substantially the same basic electric characteristics as the above flywheel diode D1 and are small in size, and fly. Connect in parallel to the wheel diode D1 and connect the shunt resistor R _S and diode D2 to the shunt resistor R _S.
The detection terminal T for detecting the polarity of the inter-terminal voltage is extracted.

FIG. 2 is a sectional view showing an example in which the flywheel diode D1 and the second diode D2 of the power semiconductor module are formed on one chip. In this way, by forming the two diodes D1 and D2 on the same chip, the diodes D1 and D2 having the same electrical characteristics and different dimensions are formed.
Can be easily obtained. As for the shunt resistance R _S connected in series with the second diode D2, the diode D
It can be easily formed on the same chip as 1, D2 by a conventionally known technique. Second, 2 is a back surface electrode common to the diodes D1 and D2.

In this power semiconductor module, when the return current flows through the flywheel diode D1, the current corresponding to the area ratio of the flywheel diode D1 and the second diode D2 is branched from the return current to cause the diode D2 and the shunt resistance. It is diverted to shunt a series circuit of R _S, the voltage between the terminals of the shunt resistor R _S is reverse polarity. Therefore, based on the detection output of this reverse polarity detected from the detection terminal T, by controlling so that the forward direction base current is not applied to the base of the NPN transistor Q3 during this period, it is possible to prevent the occurrence of the "blank mode". , NPN caused by reflux current
The breakdown of the transistor Q3, that is, the self-arc-extinguishing type semiconductor element is prevented.

The second diode D2 described above functions as a sensor for detecting the return current, and it is preferable that the reverse recovery time is shorter because the influence on the NPN transistor Q3, which is the main active element, can be reduced. Such control of the reverse recovery time can be performed regardless of whether the diodes D1 and D2 are formed on the same chip or formed on separate chips.
This can be easily performed by using a general lifetime control technique.

In the power semiconductor module, the amplitude of the voltage applied to the diode D2 is usually high, exceeding several hundred volts, and it is practically impossible to extract the control signal from the voltage generated in the diode D2. However,
In the power semiconductor module of this embodiment, since the voltage across the shunt resistor R _S is detected, the voltage of the constant voltage power supply V is set to several hundreds by properly setting the resistance value of the shunt resistor R _S according to the application. It is possible to detect the reflux current even in the application exceeding V. Moreover, in the power semiconductor module in which the flywheel diode D1 and the diode D2 are formed on a common semiconductor substrate as shown in FIG. 2, the electrical characteristics of these diodes are well matched. Since the dimensional accuracy of these diodes is also high, there is an advantage that the accuracy of detecting the return current is high.

Further, in recent years, with the higher functionality of the power semiconductor element, a structure in which a drive circuit is incorporated into the module has been attempted. However, the power semiconductor module of this embodiment is mounted on the same chip as described above. Since the diodes D1 and D2 and the shunt resistor R _S are easily formed, it is possible to easily realize a configuration in which the drive circuit is also formed on the same chip.

Although the above embodiment has described the case where one NPN transistor Q3 is used as the self-arc-extinguishing type semiconductor device, the same operation can be performed by using the Darlington transistor.

Also, when using a cathode short GTO or collector short IGBT as a self-extinguishing semiconductor element,
When an ON signal is applied to these self-arc-extinguishing semiconductor devices while the main voltage is applied in the direction opposite to the normal operation, the main current flows in the direction opposite to the normal time as in the case of the transistor. Since it has a function, the same operation as that in the above-described case can be performed by applying the same configuration as this embodiment to these self-arc-extinguishing type semiconductor elements.

Furthermore, as shown in FIG. 3, in the case of a Bi-MOS device in which a MOS transistor Q5 is cascode-connected in the preceding stage of the bipolar transistor Q4 like a Darlington transistor, the preceding-stage MOS transistor Q5 has its built-in diode. It is known that this function causes a current to flow in the opposite direction regardless of the gate signal, and that a bipolar transistor is essentially parasitic on this MOS transistor Q5. On the other hand, a resistor R is placed between the base and the emitter of the bipolar transistor Q4 in the latter stage of this BiMOS element.
When using this Bi-MOS device as a self-turn-off type semiconductor device of a power semiconductor module because it is connected to the The current flows in the direction of, and in other elements other than the Bi-MOS element described above, if an OFF signal is given during the operation in the "blank mode", the flow of charges into the inside of the element is suppressed. Will be different. That is, the Bi • MOS device cannot positively suppress the inflow of charges into the device. But this Bi
-Even if it is a MOS element, it is possible to send an OFF signal to the bipolar transistor Q4 in the subsequent stage during the detection period of the return current, and at least prevent the operation of the bipolar transistor Q4 in the subsequent stage, which is the main transistor. Is possible, and in this case as well, it is possible to suppress a large current operation that would lead to destruction.

〔The invention's effect〕

According to the first aspect of the present invention, the period in which the voltage across the terminals of the shunt resistor has the opposite polarity is detected as the period during which the return current is flowing, and the ON signal is not applied to the self-turn-off type semiconductor element during this period. Since the control is performed as described above, it is possible to prevent the self-arc-extinguishing type semiconductor element from being broken due to the return current with a simple configuration.

In the invention described in claim 2, since the flywheel diode and the other diode are formed on a common semiconductor substrate, it is possible to easily improve the precision of the electrical characteristics and the precision of the dimensions of these diodes. You can Therefore, the accuracy of detecting the return current can be easily increased.

[Brief description of drawings]

FIG. 1 is a circuit diagram showing an embodiment of a power semiconductor module according to the present invention, and FIG. 2 shows a structure in which a flywheel diode and a second diode in the power semiconductor module are formed on the same chip. A sectional view, FIG. 3 is a circuit diagram showing the configuration of a Bi-MOS element applicable as a self-extinguishing type semiconductor element of the power semiconductor module, and FIG. 4 shows an operation principle of a typical power conversion device. FIG. 5 is a timing chart showing the operation of the power converter, and FIG. 6 is a circuit diagram showing a proposed circuit configuration for improving the breakdown tolerance of the self-extinguishing semiconductor element in the power converter. Is. In the figure, Q3 is an NPN transistor (self-extinguishing semiconductor device), D1 is a flywheel diode, D2 is a second diode, R _S is a shunt resistor, and T is a detection terminal. In the drawings, the same reference numerals indicate the same or corresponding parts.

Claims (2)

[Claims] 1. A self-arc-extinguishing semiconductor element, and a flywheel diode connected to the self-arc-extinguishing semiconductor element in anti-parallel and flowing a return current in a direction opposite to the conduction direction of the self-arc-extinguishing semiconductor element. In a power semiconductor module including a shunt, another shunt consisting of a series circuit of a diode and a shunt resistor is connected in parallel to the flywheel diode and in antiparallel to the self-extinguishing semiconductor element, and the shunt is connected. A power semiconductor module, wherein terminals are provided at both ends of the resistor, and the presence or absence of the return current is detected from the polarity of the voltage between the terminals. 2. The power semiconductor module according to claim 1, wherein the flywheel diode and the other diode are formed on a common semiconductor substrate.