JPS59108424A - Control circuit with impedance for auxiliary self switching on for thyristor or triac type high sensitivity semiconductor device - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

Description: TECHNICAL FIELD The present invention relates to the control of sensitive thyristors by forming a short circuit between two regions of a semiconductor and its application to the construction of bidirectional elements used as static relays.

BACKGROUND OF THE INVENTION For static relay applications, a conventional υ iris resistor, which is switched on by injecting the tip of a steep gate current of appropriate polarity, reacts with a sudden voltage change dV/
dt, and there was a shallowness where the current -valve speed di/dt was insufficient when very high control til+sensitivity was required.

In order to improve these shortcomings, on November 25, 1982, the Center National de I
a Researcher's Scientific
In the French patent application entitled "Thyristor structure with intrinsic switch-on characteristics and its application to the structure of bidirectional devices" filed by
To form an ultra-sensitive thyristor with self-switch-on characteristics at room temperature due to the effect of capacitive current due to dt, to form a short circuit between the gate and cathode that inhibits the self-switch-on effect, and to supply voltage. It has been proposed to eliminate this short-circuit condition by increasing the value of , and to control a thyristor with such an inherent switch-on characteristic by causing self-switch-on at the zero crossover point.

One drawback of thyristors with inherent switch-on characteristics is that there are some major constructional difficulties based on the basic principles of their operation.

In fact, it is very difficult to make these thyristors sufficiently sensitive for self-switching at a point of almost zero crossover of the mains voltage in a reproducible manner.

That is, as far as the self-switch-on trigger is concerned, the displacement current CdV/dt is effectively reduced, but this is due to the homogeneous distribution of this current to the surface of the central junction of the thyristor, which reduces the This is based on the fact that this current decreases hyperbolically from the point where the voltage is O, since on the other hand the capacitance of the central junction varies with 1/V for 7<n<. Taking into account the low density of this displacement current and its reduction, the Iristor has a very high gain, determined by its structure, of the NPN1 transistor on the one hand, and therefore a very low current value on the other hand. , N determined by this gain
It is necessary to form the PNt transistor and the PNP transistor so that their gains are maintained. It is extremely difficult to reproduce such a structure. On the other hand, obtaining the desired ultra-high sensitivity to voltage changes dV/dt is difficult due to the parasitic voltage changes d
It must be emphasized that obtaining sufficient immunity against V/dt is contradictory.

The present invention overcomes these drawbacks by using a control circuit to assist self-switch-on or triggered switch-on, and uses highly sensitive thyristors available on the general market to achieve the desired We propose to make this possible.

SUMMARY OF THE INVENTION The control circuit according to the invention consists of switching means for forming a gate-to-cathode short circuit in a sensitive semiconductor structure controlled by the action of an external control, assisting in self-switching. It is characterized by an impedance and by connecting the anode of the circuit to the gate control output.

In a first embodiment, the impedance is a resistor and the switch means is formed by a field effect transistor whose gate is biased by a further resistor from the anode.

In the second embodiment, the impedance is a capacitor.

In selected embodiments, the switching means comprises a phototransistor, such as a field effect transistor biased by the anode of a resistor controlled thyristor, which shorts out the gate bias 6- when activated by an optical signal. Become.

DESCRIPTION OF THE EMBODIMENTS FIG. 1 shows a semiconductor device in which the rectangular portion SG is made non-conductive by short-circuiting the gate Ca and the cathode. This device is 220V-50
An AC voltage is supplied from the power supply of H2, and the voltage is applied between the anode A and the cathode. The control means for creating a short circuit between the gate Ga and the cathode is supplied with a supply voltage between its own anode A and the cathode. The entire circuit is bipolar. The means essentially consist of an N-channel MOS + transistor, denoted by M, a zener diode N, a resistor RG and a phototransistor Ph-Tr activated by a control light signal.

This phototransistor preferably constitutes the output of the photocoupler. Diode Z is MOS l~
It plays the role of protecting the transistor, for example 10
It has a zener voltage on the order of volts. Resistance RGi, tMO
3l ~ The self-bias of the gate of the transistor from the hot point (+) of the output circuit is compensated.

When the photocoupler is activated, this gate bias is shorted and the MOS t-transistor becomes non-conductive.

On the other hand, when the F7I-to coupler is not activated, MO8
The gate of the I-transistor is biased and as soon as the voltage VGS exceeds the threshold vth of the MOS I-transistor, the M
The OS becomes an almost complete short circuit, and therefore the semiconductor device S
C is made non-conductive.

Finally, it is noted that the semiconductor device SC is always non-conductive, and its control is achieved by opening a photocoupler control relay, which is advantageous for application as a static relay.

By choosing a sufficiently large value, for example 1 MΩ, for the resistor RG, the parasitic current flowing through the phototransistor when the semiconductor device SC is in a conductive state is limited to an allowable effective value on the order of 0.2+nA. A time constant added to the delay due to the threshold value vth of the MOS transistor when the semiconductor device becomes non-conductive due to charging of the input capacitance C15s of the MOS transistor is substantially smaller than the delay when switching on the semiconductor device SC. Assuming that, the resistance value of 1 MΩ can be further increased.

According to the essential characteristics of the above circuit, for example, from 1 to IO
An additional high resistance Rea of MΩ connects the anode of the conventional high sensitivity thyristor SC to the gate. The result is
As soon as it is applied from the optical control signal, i.e. MOS t
- As soon as the transistor is made non-conducting, the thyristor S
A low level gate current is set at C. As the supply voltage increases, this auxiliary small current starts to switch on the thyristor SC once the zero crossover is reached. However, the displacement current of the thyristor SC extends the switched-on state over the entire surface of the central junction of the thyristor SC. The auxiliary gate current also increases with the supply current and tends to compensate for the hyperbolic decrease in displacement current.

The current type of MO8t-transistor has no difficulty and no appreciable voltage drop even if the current from the resistor Rca reaches a peak value of 300 μ8 for 1 MΩ at an effective 229-0 V. It is noted that this current can be adjusted without any interference.

Also, if the thyristor is switched on in a circuit with a highly inductive negative direction, i.e. the increase in current is very slow, the auxiliary current will not exceed the voltage until the thyristor has completely switched on. It should be noted that, since its size increases with time, it in any case provides a greater assistance to the switch-on and eliminates the danger of a "break-off" by the thyristor. Resistance R
The value of 'ca is the switch-on threshold Va for the thyristor.
can be calculated if given again. Actual ΔQa
is the amount of charge stored at the base of the NPN part of the thyristor from the zero crossover point of the supply voltage, a fixed amount for a given thyristor, ■ is the amplitude of the supply voltage, and ω is its pulsation. In this case, the formula ΔQa =
Va 2/2ωV・RGa can be shown.

10- In the circuit of FIG. 2, the bidirectional element is shown as consisting of two thyristor structures A+, 2, and A2, 2 connected in reverse order, each of which is shown in FIG. It is controlled by the same type of circuit as shown in .

A+-A2 and +-2 represent the respective anodes and cathodes. In the conducting state, MOS transistor M1
The phototransistors T1 and T2 shorting and M2 are themselves controlled by light emitting transistors, not shown, which are powered from a low power DC source via resistors and mechanical or electrical switches. T1 and -
The light emitting diodes for controlling [2 are each controlled at times corresponding to the beginning of the positive half-wave part and the beginning of the negative half-wave part. The control time of this diode needs to be longer than the self-switching of the thyristor. Further, there is no problem even if this control time is a long time, and for example, it may be a length corresponding to several half wavelengths of the power supply current. At the end of the control of this diode, the thyristor is already in operation and the voltage across the terminals of the circuit reaches a value of, for example, 2 V, which is too low to make the MOS conductive.

It is noted that this control device has positive safety with respect to conventional component failures. That is, even if the cause of the failure is that the photocoupler cannot conduct, safety is maintained because the MOS phototransistor is still in its natural state and self-switching-on is prevented for each half-wave.

Shorting of resistor RG, which must carry most of the voltage, is most likely to occur for high voltages in the power supply. If this happens, the voltage increase at the gate of the MOS will cause this gate to break through (introducing a small amount of leakage current), but it will also cause the MOS to leak due to excessive dissipation.
This results in a short circuit rather than the MOS becoming conductive. Only the case of two failures shows negative safety. In other words, the destruction of the phototransistor,
This is the destruction of the zener diode. In fact, since the phototransistor is protected by the Zener diode, its destruction is very unlikely. As for the Zener diode, destruction will occur only if the voltage rises to such an extent as to cause destruction of the MOS at the same time, and in the above-mentioned case, the voltage will drop.

Continuously non-conducting enrichment M
The OS transistor can be replaced by a normally conductive type element. Such elements are improved MOS
It may be a transistor, or it may be a field effect transistor having a P-chil one-thinner junction. In this way, concentrated M
Complementary control to that obtained with OS transistors becomes possible. Since a conducting state is thus obtained which compensates for the self-switching of the thyristor structure, the gate of the field-effect transistor has to be controlled in such a way that it is non-conducting and thus allows self-switching.

Of course, it is necessary to add a stage of inverter logic 13, known per se, to the input of the field effect transistor in order to maintain a logical correspondence between the control signal and the conduction state of the power element.

This variant has the inherent advantage of providing 1q of active safety independent of the control, resulting in a similar increase in overall safety.
This is because even if S has a fatal defect, it will only lead to a short circuit. On the other hand, field effect transistors are more difficult to manufacture and require more complex control.

However, this latter drawback can be alleviated by integrating the inverter and Zener diode within the same crystal, which is easy to do.

R1Ja+ and R(Ja2 are self-switch-on assisting resistors and play the same role as RGa in FIG. 1.

In the case of the bidirectional element shown in Figure 3, two
It consists of two thyristors Th+ and Th2, capacitors C1 and C2 connecting their anodes A1 and A2 and gates G1 and G2, respectively, and the gates (3+ and G2
are connected to cathodes 1 and 2 via switches 11 and I14-2.

Considering one of the two (Neuristors) in this circuit, taking Th+ as an example, when switch 1 opens, as soon as the voltage is established at AI, the current!=C+ dv/
dt is injected into the gate Th+. Switch-on occurs after each zero crossover by increasing the value of the supply voltage as soon as this current is sufficient. 11
When closed, the current C+ d from the cathode junction
Since v/dth<, the voltage II at the gate becomes approximately equal to the cathode voltage and the cyrisk can no longer be switched on.

At the time of first switch-on, the discharge current is C+dv/dt
It is noted that its value is low since it does not exceed , making it pointless to connect a resistor in series with the capacitor to limit this current.

During the non-conducting period, in the case of dv/dt parasitics, current flows through the internal resistance of 11, so that if the magnitude of this internal resistance is equal to If it is much lower than the C of the thyristor.
Protection against untimely switching on from I is completely ensured. For example, dv/dt is 1000V/μS,
If C+ = 02 = 100PF, the short circuit switch needs to flow a current of 100 mA, and the voltage between the terminals needs to be maintained at a maximum of 0.2V. This allows the MOS
The l-transistor will have a static resistance of 2Ω in the conducting state. Furthermore, the switch must fill this resistance value very quickly. This is because the thyristor must be activated before it is switched on, and if the thyristor is triggered by a current supplied by a capacitor, switching on takes place relatively quickly.

This device works accurately when using highly sensitive thyristors. If a conventional thyristor is used, the auxiliary capacitor must have a high value, so that the leakage current flowing in the semiconductor device and the load in the non-operational state is too high.

The triac circuit illustrated in FIG. 4 operates in the same manner.

It is worth noting that shorter switch-on times are obtained for circuits with switch-on auxiliary capacitors compared to circuits with auxiliary resistors (because the current injection into the gate of the thyristor is steeper). be. As long as the resistance in the closed state of the switch is well defined and of a very low value (eg electronic contacts), these circuits are useful and provide good protection against parasitic operations. When?
In the case of a circuit that is assisted by a resistor, the current type M
Operated by a switch consisting of an OS l-transistor,
It is very useful in some applications.

If a compromise is to be made between the characteristics of these two types of circuits, a parallel resistor-capacitor circuit can be used as an impedance for auxiliary switch-on.

It goes without saying that various different modifications may be made to the circuitry shown and described above without departing from the spirit of the invention.

[Brief explanation of drawings]

FIG. 1 shows a schematic diagram of a control circuit for a simple highly sensitive semiconductor structure according to a first embodiment of the invention, and FIG. 2 shows an embodiment of the circuit when the structure to be controlled is bidirectional. 3 and 4 show two types of semiconductor control circuits for two thyrisk devices and a triac, respectively, according to a second embodiment of the invention. Description of symbols of main parts SC: Semiconductor device A: Anode K: Cathode Ga, G: Gate M: MOS transistor Z... Zener diode Th... Thyristor agent Patent attorney Motohiko Fujimura 18-

Claims (8)

[Claims] (1) A control circuit for a semiconductor device comprising switching means for forming a short circuit between the gate and cathode of the semiconductor device under external control, further comprising a switch means for forming a short circuit between the gate and the cathode of the semiconductor device, A control circuit characterized in that it includes an impedance for connecting and assisting in self-switching. (2) The impedance is a resistance, and the switch means is connected to a field effect transistor whose gate is biased from the anode through another resistor, and when activated by an optical signal, the switch means is connected to the field effect transistor whose gate is biased. 2. The control circuit according to claim 1, comprising a phototransistor that short-circuits a bias. (3) A control circuit according to claim 2, characterized in that the resistor for assisting the self-switch-on has a value on the order of 1M ohms to several MJ-ohms. (4) The control circuit according to claim 1, wherein the impedance is a capacitor. (5) The control circuit according to claim 2, wherein the optical signal is set for a time longer than the self-switch-on time of the semiconductor device at the time of zero crossover of the supply voltage. (6) The control device according to claim 5, wherein the field effect transistor has a size capable of interrupting a switch-on current flowing within the semiconductor device. (7) The control device according to claim 2, wherein the field effect transistor is of an MQS concentration type. (8) The control device according to claim 2, wherein the field effect transistor is an MO8 deficient type or a P-channel junction type, and an inverter logic circuit is coupled to its input.