A new circuit for obtaining thyristor voltage zero-crossing signal

Preface

In fast reactive power compensation and harmonic filtering devices, thyristors are used as actuators to switch capacitors as TSC circuits. The actuator thyristors have different structural types of pancake, module and bidirectional thyristor according to different application occasions. The trigger circuit is different for different main circuits and different thyristor types. TSC requires triggering at the zero-crossing point of the thyristor voltage. There are two ways to determine the zero-crossing point of the thyristor voltage. One is to obtain the synchronization signal from the grid voltage, and the other is to obtain the zero-crossing signal from the anode and cathode of the thyristor.

This paper analyzes the characteristics of various existing trigger circuits, and thus introduces a new circuit for obtaining the thyristor voltage zero-crossing signal from the main circuit thyristor. This circuit supports the generation of a series of trigger circuits and achieves excellent triggering effect.

1. Introduction to the principle of thyristor switching capacitor and the requirements for fast zero-crossing triggering

The key technology of thyristor switching capacitor bank is to ensure that the current has no impact. The mechanism of thyristor switching capacitor bank is shown in Figure 1.

When the resonance number n of the circuit is 2 or 3, its value is very large. The third term of formula (2) gives the amplitude of the oscillating current when the trigger angle deviates from the optimal point; the second term of formula (2) gives the amplitude of the oscillating current when it deviates from the optimal pre-charge value. If the capacitor current ic=C*du/dt=0, then du/dt=0, that is, the thyristor must be triggered to turn on the capacitor bank at the positive or negative peak of the power supply voltage, and the capacitor is pre-charged to the peak voltage.

The functions of the trigger circuit are: current non-impact triggering; fast switching, 20ms action. This 20ms is not the time from receiving the switching command to generating the action, but the time from stopping to re-engaging the action is 20ms. When reacting quickly, in the balanced compensation circuit, unbalanced action cannot occur, that is, some phases have current and some do not.

2 Characteristics and existing problems of the trigger circuits of two types of thyristors

From the perspective of synchronization signal acquisition, there are two types of thyristor trigger circuits: one is to obtain synchronization signals from the grid voltage, and the other is to obtain synchronization signals from both ends of the thyristor.

The circuit block diagram of obtaining synchronization signal from grid voltage is shown in Figure 2:

The circuit includes a synchronous transformer, a synchronous signal processing circuit, a power drive circuit, a pulse transformer isolation circuit, etc. When a trigger command is received, a trigger pulse train is generated at the switching point, which is isolated by the pulse transformer and drives the thyristor. The synchronous signal processing circuit has a filtering processing function and can be composed of electronic circuits such as CMOS, or a single-chip microcomputer, a GAL circuit, etc. The circuit includes a phase sequence error judgment function.

The advantage of obtaining the synchronization signal from the grid voltage is that when the main circuit is not powered, the trigger command can be given, and the trigger pulse amplitude and phase of the thyristor can be measured. After the main circuit is powered, the trigger command can be given, and you can rest assured that the TSC is working correctly. For the "2+1" circuit of two thyristors + a diode, the "2+2" circuit of two thyristors + two diodes, and the "3+3" circuit of three thyristors + three diodes in the TSC circuit, the capacitor is pre-charged by the diode, and there is always a DC voltage on the capacitor. The AC and DC voltages of the thyristor remain unchanged, and the grid voltage is suitable for obtaining the synchronization signal trigger. The disadvantage is that the circuit is complicated and the cost is high for the 400V small-capacity TSC circuit.

Figure 2: Trigger circuit for obtaining synchronization signal from grid voltage

It is difficult to obtain zero-crossing signals from both ends of the thyristor. Zero-crossing triggering requires that it is cut off when the voltage is high and that it is turned on and triggered when the voltage is the lowest. It is almost impossible to find any component with this characteristic. For example, the voltage regulator tube cuts off when the voltage is low and maintains the voltage unchanged when the voltage is high. This does not meet the requirements.

At present, the typical trigger circuit for obtaining zero-crossing signals from both ends of the thyristor is MOC3083, and its block diagram is shown in Figure 3:

Figure 3: MOC3083 circuit diagram

The MOC3083 chip has a zero-crossing trigger judgment circuit inside, which is designed for a 220V grid voltage. The chip's bidirectional thyristor withstands a voltage of 800V. When the voltage at both ends of 4 and 6 is lower than 12V, if there is an input trigger current, the internal bidirectional thyristor will be turned on.

When used in a 380V power grid TSC circuit, several 3083s should be connected in series. The application of a 2-control-3 TSC circuit is shown in Figure 4:

Figure 4: 2-to-3 TSC circuit

Using two pairs of thyristor switches to control a three-phase circuit makes the circuit simpler but the control mechanism more complicated. This trigger circuit will cause many troublesome problems when giving trigger commands randomly.

When the action is fast, there is a trigger command, one pair of thyristors is turned on and the other pair of thyristors is not turned on, but the voltage increases. Due to the limited space and focus, this article does not analyze why the voltage is higher. It only sees from the measured 2-control-3 circuit that there is indeed a phenomenon and danger of voltage increase. This phenomenon is the same as the DC voltage increase of the voltage doubler rectifier circuit. Figure 5 measures the voltage waveform of two pairs of thyristors that are not working properly. There is a possibility of high-voltage breakdown of this test thyristor, so the voltage regulator is used to lower the grid voltage. When the thyristor is turned on, the voltage at both ends is zero. When it is not turned on, the thyristor has the DC voltage of the capacitor and the AC voltage of the grid. The peak-to-peak voltage when the C phase stops is 540V, and its effective value = . The peak voltage of the C phase in the figure is 810V, and the increased voltage is about a multiple of the effective value of the grid voltage:.

Figure 5 Abnormal voltage waveforms of two pairs of thyristors

*When the thyristor voltage waveform passes through zero, the MOC3083 in series is unevenly divided, causing some 3083 to conduct and some to stop. When the grid voltage increases, the previously conducting ones remain conducting, while the different ones have to withstand higher voltages, and the 3083 may break down.

* There is a certain impact when switching on for the first time. The following is the current waveform of the first switching on of a famous foreign product.

Figure 6: The first trigger shock waveform of a foreign company's product

Record the voltage across the C-phase thyristor and the current of A-phase. The current switching impact is so great that the grid voltage is deformed.

*Not suitable for rapid shock loads.

*MOC3083 may be mis-turned on at the moment of closing the switch, which may damage the thyristor.

* When the harmonic current in the filter device is large, the thyristor will not work properly and may even stop working.

*Circuit design is difficult when the grid voltage is higher than 400V.

3 A new type of circuit that collects zero-crossing signals at both ends of the thyristor, thereby generating a series of trigger circuits.

The function of the trigger is related to the circuit structure. There are D triggers and JK triggers according to the circuit structure; D triggers are divided into transmission gated type and maintenance blocking type. In addition, there are other T triggers, SR triggers, D triggers, and JK triggers that can be converted into triggers with other functions. For triggers of different structures, the trigger conditions are different, but we should remember that there are only two triggering methods, one is pulse level triggering method, and the other is pulse edge triggering method.

Pulse level trigger mode: refers to the state of the trigger changing during the high level period of the pulse signal. Note that the high level period here indicates that the state of the trigger is constantly changing during the high level 1 period of the pulse (as long as the input signal changes, the output signal changes). For example, the SR transmission gate latch is this type. Pulse edge trigger mode: There are two types, one is that the state of the trigger changes on the rising edge of the pulse edge, such as the D trigger; the other is that the state of the trigger changes on the falling edge of the pulse edge, such as the JIK trigger.

It is not easy to design a zero-crossing trigger circuit in the main circuit. According to the literature, there is a method of using a LEM module based on the Hall principle to collect zero-crossing signals. The principle block diagram of the zero-crossing trigger is shown in Figure 7, and the principle diagram of the thyristor zero-crossing voltage detection circuit is shown in Figure 8. After hard work, the author of this article abandoned the method of taking zero-crossing signals and triggering thyristors in the main circuit of MOC3083 according to the ideas of the principle block diagrams and circuit schematics of Figures 7 and 8, and developed a new type of circuit, which is characterized by collecting the zero-crossing signal of the thyristor and feeding it back to the low-voltage end of the input, and then performing signal logic processing to trigger the thyristor. The circuit block diagram is shown in Figure 9.

Figure 7 Principle block diagram of TSC zero-crossing trigger

Figure 8 Schematic diagram of thyristor zero-crossing voltage detection circuit

Figure 9: Zero-crossing acquisition control logic photoelectric drive circuit block diagram

400V grid voltage mostly uses module thyristor, and photoelectric drive thyristor can be used as shown in Figure 9. 660V grid voltage, the grid voltage is high, and pulse transformer drive is required. See Figure 10.

Figure 10: Block diagram of zero-crossing acquisition control logic pulse transformer drive circuit

Medium voltage TSC needs to use pulse magnetic ring triggering according to insulation requirements. Figure 11.

Figure 11 Medium voltage TSC uses pulse magnetic ring trigger

A new trigger circuit is adopted and a single chip microcomputer is used to make a logic time control to trigger a 2-control-3 circuit.

The switching current is relatively free of impact. Since there is no DC voltage when the capacitor is switched for the first time, it is an undesirable state and there must be a certain impact. When the ratio of the impact current to the normal stable current is ≤1.7 times, it can be considered that it does not affect the use of the thyristor and capacitor. After the switching stops, there is a peak voltage of the grid on the capacitor. Under the synthesis of the grid voltage and the DC voltage of the capacitor, the thyristor has a zero-crossing voltage. It is an ideal state to trigger the thyristor at the zero-crossing point, and there should be no impact current.

The new trigger circuit achieves a fast action of 20ms, both thyristors are in action, there is no current shock, and the thyristors have a low withstand voltage when stopped, with a maximum of 3 times the effective value voltage.

Use a dual-trace oscilloscope to test the waveform. One test lead measures the voltage across the thyristor and the other one measures the current waveform of the thyristor. In this way, it can be seen whether the thyristor is switched on at the zero crossing point, and the current impact when it is switched on can be seen. Figure 12 shows the action waveforms of the A-phase thyristor voltage and the C-phase current of the continuously switched thyristor.

Figure 13: Another picture of the A-phase thyristor voltage and C-phase current. The horizontal axis is 50ms/grid fast action

Figure 14: Phase A thyristor voltage and phase C current when starting from a long-term stop state.

The first cycle has a little impact. The peak value of the impact current is 32A, the peak value of the normal stable current is 24A, and the impact current/stable current = 1.33.

The thyristor switch is more effective when placed in a triangle, and phase control can be used to compensate for unbalanced loads.

Figure 15 Effect of placing thyristor switch inside a triangle

Figure 16. The thyristor switch is placed in a triangle and operates without impact for the first time.

4 Conclusion:

The new circuit for collecting zero-crossing signals at both ends of the thyristor meets the requirements of fast and impact-free switching of capacitors and can still operate normally under severe harmonic current conditions. It is suitable for different main circuits of TSC, different voltage levels and different thyristor forms, and has good effects. A series of trigger circuits have been produced to meet different needs.