JP2016086460A - Semiconductor AC switch - Google Patents

Abstract

PROBLEM TO BE SOLVED: To block an AC power supply at a current zero point by softly turning on the AC power supply using a semiconductor switch.SOLUTION: Four MOSFETs made of a reverse conduction semiconductor device are used and a capacitor is used as an AC switch by being connected to a DC terminal, then a gate control device thereof detects the phase of a voltage to turn on S1, S3 at 270 degrees of the voltage phase at the turning-on time and then further turns on S2, S4 with a delay of 180 degrees. Though S1, S2, S3 are turned off with the delay of 270 degrees of the voltage phase when blocked, S4 is turned off with the delay of 180 degrees.SELECTED DRAWING: Figure 1

Description

    The present invention relates to an AC power supply on / off switch using a semiconductor device.

  Conventionally, a switch for turning on and off an AC power source has been conventionally turned on and off by driving a metal contact with an electromagnetic coil. In recent years, it has become possible to turn on and off the current using a semiconductor switch, but the on-loss is still large, and it does not reach the low resistance of metal contacts, the leakage current at the time of interruption, the problem of cost, etc. Not popular. In recent years, SiC (silicon carbide) MOSFET semiconductor devices have been able to achieve a high current, a large current, and a low on-resistance.

Although MOSFET semiconductor devices have reverse conducting diodes called parasitic diodes or body diodes, they exist in parallel, but their on-voltage is not low. Therefore, as a feature of the MOSFET semiconductor switch, a parallel operation is possible with a low on-resistance in both forward and reverse directions by a gate-on signal and a positive characteristic with respect to temperature, so that the on-resistance can be further reduced by parallel connection. In order to make this switch an AC switch, a bridge configuration is known as shown in FIG. In this case, if all the switches are turned on, the on-resistance is equivalent to one switch.

Further, for example, FIG. 6 shows a thyristor AC switch in order to suppress an increase in the regenerative voltage generated when the AC current is cut off, but a snubber circuit including a resistor and a capacitor is required. The snubber circuit is a function that consumes the energy of the current at the time of interruption, preventing overvoltage breakdown and noise generation of the semiconductor switch. The thyristor AC switch shown in FIG. 6 has a problem that leakage current due to charging / discharging of the capacitor in the series snubber circuit of the resistor and the capacitor continues to flow even when cut off.

This has already been patent-registered as a “current forward / reverse bidirectional switch for regenerating snubber energy” in a configuration called a magnetic energy regenerative circuit (hereinafter referred to as Magnetic Energy Recovery Switch, hereinafter referred to as MERS).

JP 2000-358359 A

When all four MOSFETs of the MERS switch are turned on, they have low on-resistance in reverse current and are 2 series and 2 parallel, so it was not possible with the conventional AC switch configuration of semiconductors such as thyristors. Although a low on-resistance can be realized, the sequence based on detailed studies such as the timing of turning on and off the AC voltage and the effect thereof have not yet been disclosed.

The problem to be solved is to clarify the sequence when the power is turned on and the sequence when the power is shut off when MERS is used as a switch for the AC power supply. It must be operated so that a voltage shock is not applied to the load as much as possible at the time of turning on, and no interruption surge voltage is generated at the time of interruption. Because, when MERS is turned on, all of the gates are turned on at the same time, the capacitor charge is short-circuited, and when all the gates are turned off simultaneously, the current is instantaneously cut depending on the timing, and the energy of the total current is accumulated in the capacitor. There is a risk of high voltage exceeding the withstand voltage of the switch.

  The present invention detects the phase of the power supply voltage and operates the voltage waveform in order to operate the four switches of the MERS circuit individually on or off in accordance with the voltage phase to become a more ideal AC switch. Perform sequence control by meeting

The control method and control circuit of the MERS AC switch according to the present invention turns on the power at a voltage phase with little voltage shock with respect to the load. Absorb. Further, when conducting, an on signal is given to the four switches to turn them all on, and the resistance of the AC switch becomes the on resistance of one switch. It is characterized by no leakage current when the switch is off.
In particular, in the case of a low voltage and large current power source, in the case of a conventional thyristor or triac, the on-voltage is about 1 V or more and does not change even when connected in parallel. Low loss can be achieved.

These are the explanatory views showing the method of carrying out the present invention. Is a simulation circuit of the present invention. Example of gate signal sequence control Is the simulation result at the time of input. Iacin is a power supply current, Iout is a load current, Vc is a capacitor voltage, Vin1 is a power supply voltage, Vout is a load voltage, I (S1) to I (S4) are MOSFET currents S1 to S4, and Vg1 to Vg4 are MOSFETs, respectively. The gate command is 1 for ON and 0 for OFF. Is a simulation result at the time of interruption. Is a circuit example of a conventional thyristor AC switch.

  The power supply is assumed to be AC 50 Hz 100 V and the load is 10 ohm and 30 mH in series.

  FIG. 1 is a diagram showing a circuit diagram and an implementation method of an embodiment of the device of the present invention. As shown in FIG. 1, there are four MOSFETs S1 to S4, each of which is turned on / off by a gate control device. Turned off. Turns on when the gate voltage is high and turns off when the gate voltage is low. The control device 4 detects the voltage phase of the power supply, receives the switch-on command and the shut-off command for the entire switch, and performs gate control of each MOSFET. Here, a circuit for detecting the phase of the power supply voltage is incorporated.

  When a power-on command is received, the MOSFETs of S1 and S3 are first turned on at a voltage phase of 270 degrees, that is, at a phase where the AC voltage is minimized. As a result, the voltage of the capacitor C charged up to the peak of the power supply voltage is added in series and combined with the negative maximum of the power supply voltage to start energizing the load at a voltage of zero. This avoids applying a pulsed voltage to the load. After that, the phase is delayed by 180 degrees, and at 50 Hz, 10 ms later, S2 and S4 are turned on to turn on all the MOSFETs, and this state is maintained until a cutoff command is received.

  When the cutoff command comes, the control device detects the phase of the voltage and turns off S1, S2, and S3 at 270 degrees. Although the route of current flowing from S1 to S2 is blocked, the entire current continues to flow as it is through the routes of S3 and S4, which are another current route. Thereafter, the current crosses zero and tries to reverse, but since S3 is a current-carrying state of the reverse conducting diode, the forward current is blocked and the snubber is interrupted by natural commutation at zero current. Thereafter, S4 is also turned off with a delay of 180 degrees (10 mS) from the turn-off of S1, S2, and S3. This is possible because there is a 10 ms delay for automatically crossing zero current. Since the load current is a lagging current, the phase is lagging and is always cut off by the reverse conducting diode of S2. The reactivation voltage at the time of interruption is interrupted at the current zero point, and this reactivation voltage is absorbed by the two snubber capacitors.

In this embodiment, an alternating current of 50 Hz is assumed. Even at 60 Hz, the timing of the voltage phase of 270 degrees and the delay time between 180 degrees are the same if operated in accordance with the timing. Description of the electronics and control system is omitted. Obviously, S2 and S4 can be exchanged.

  The case where the circuit of FIG. 1 is applied is analyzed by a circuit simulation program.

FIG. 2 shows a circuit diagram of the PSIM simulation code. The load of Example 1 is a series of 30 mH and 10Ω.
The analysis result is shown in FIG. 4, and the input command comes, and the gate signals Vg1 and Vg3 are made high at the voltage phase of 270 degrees, and Vg2 and Vg4 are made high 180 degrees (10 mS) later. The load voltage Vout, Iac1 is shown in FIG. 4. Since the voltage of the capacitor that charges the peak voltage starts at the maximum negative voltage, a difference voltage of about zero voltage is applied to the load, and the load current is It can be seen that the flow starts mainly from 0 degree of the voltage phase. The capacitor discharges slowly over 90 degrees during this period, and it can be seen that there is no pulse-like change in voltage and current in both the switch and the load.

  FIG. 5 shows a case where a shut-off command comes and shuts off in the sequence of the present invention, but the gate signals S1, S2, and S3 are simultaneously turned off at the voltage phase of 270 degrees to shut off the currents S1 and S2 of the MOSFET The load current continues to flow in the forward direction in S4 and in S4. Thereafter, the current is automatically cut off near the current zero point by the diode of S3. Thus, it can be seen that S4 is naturally cut off by the diode without being involved in the cut-off operation because the gate is turned off after the current is cut off.

  In the case of three-phase alternating current, first, two phases are turned on, but when the line voltage is seen, the ON is simultaneously started at the same minimum voltage value (270 degrees). The last phase should be turned ON after 90 degrees. As a result, a DC component does not occur in the current after being turned on, so that it is possible to prevent the excitation rush phenomenon to the transformer.

  A semi-conductor AC switch can be realized with a minimum of on-resistance when it is composed of four reverse conducting MOSFETs, but it is more ideal for the load by turning on and off each switch in accordance with the voltage phase. A sequence that can realize a typical voltage waveform is presented. In this case, the current waveform rises from zero to a sine wave when the power is turned on, and is cut off at the zero current point when cut off. In the future, computer-controlled semiconductor AC switches will be frequently used, but this sequence is used at that time.

Since this switch can be operated in accordance with the AC voltage phase, it is possible to avoid an inrush current (saturation phenomenon due to magnetic bias) to the transformer when the three-phase AC power supply is turned on. That is, the two phases are turned on at the interphase voltage of 270 degrees, and the remaining phases may be turned on after a delay of 90 degrees. However, this control is impossible with conventional mechanical contacts.

The industrial advantage of adopting this switch is that it can be avoided before the influence of the accident current by one cycle of high-speed cutoff after the cutoff command, and there are many cost advantages because it can reduce the tolerance of the equipment such as avoiding excitation inrush. The AC switch, control method, and control apparatus of the present invention are AC switch systems that use semiconductor devices that are indispensable for advanced power distribution systems that will be ICT controlled in the future.

1 Reverse conducting semiconductor switch MOSFET
2 Snubber capacitor 3 AC power supply 4 Control device 5 Load

Claims (4)

When a current bidirectional switch using four MOSFETs is used as an AC switch to input current, the gate signal is applied to the MOSFETs of S1 and S3 at the time when the AC voltage is minimum (the time when the negative voltage becomes maximum), that is, at an electrical angle of 270 degrees. An AC switch that turns on and conducts, and then sends a gate-on signal to S2 and S4 with a delay of 10 mS in the case of 180 degrees and 50 Hz, thereby bringing all the switches into conduction. When a current bidirectional switch using four MOSFETs is used as an AC switch for interrupting current, a gate-off signal is applied to S1, S2, and S3 when the voltage is minimum (the time when the voltage becomes negative), that is, at an electrical angle of 270 degrees. , And the current of S1 and S3 is concentrated in the path of S2 and S4, and the current is cut off by the reverse conducting diode of S2 at the current zero point, and then delayed by 10 mS at 180 degrees and 50 Hz. AC switch that shuts off by sending a gate-off signal to S4. Three sets of AC switches according to claims 1 and 2 are used for three-phase AC, but in order to prevent excitation inrush current to the transformer, the A and B phase C and A phase C switches are connected to the line voltage between CA. The AC switch and its control device according to claim 1 and 2 for controlling a closing phase in which a B-phase switch is turned on at the same time at a phase of 270 degrees and then delayed by 90 degrees (5 ms delay in the case of 50 Hz).   Three sets of AC switches according to claims 1 and 2 are used for three-phase AC. When a cutoff command is issued, the A, B and C phase A and C phase switches are connected to the line voltage between CAs. 4. The AC switch according to claim 1 and 2, wherein the AC phase is controlled at the same time with a phase of 270 degrees, and then the closing phase for controlling the B-phase switch is controlled with a delay of 180 degrees thereafter (10 ms delay in the case of 50 Hz). To control device.

Images