JPS6114675B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a zero voltage switching circuit using a photo-triggerable bidirectional thyristor.

Conventionally, as disclosed in, for example, Japanese Patent Application Laid-Open No. 100991/1982, the above-mentioned Bidirectional photothyristors have been proposed in which each thyristor element can be fired, but in this type of thyristor, a control (gate) electrode cannot be connected because a voltage trigger signal is not applied. is common, and it is impossible to achieve the zero voltage switching function described below.

Incidentally, recently, optically coupled solid-state relays using photocouplers have been commercialized and are attracting attention. This relay uses a photocoupler on the input side and a control circuit with a zero voltage switching function on the output side, and combines them into a unit. Compared to conventional electromagnetic relays, etc., (1) it is a non-contact relay, so it is resistant to mechanical vibration and has a long life; (2) the input and output are optically coupled using a photocoupler, so the input and output (3) Zero voltage switching function (switching only when the AC power supply is close to zero voltage) ), so there is very little so-called radio frequency interference, and it also has the advantage of reducing inrush current protection circuits such as lamp loads and capacitive loads, so it can be used in places where noise is a problem or where switching is frequently done. Ideal for places with a bad atmosphere or where high reliability is required.

The circuit configuration of the optically coupled solid-state relay differs depending on what is used as the photocoupler's light receiving element, but here we will briefly explain the operation of the circuit configuration when a photothyristor is used. FIG. 1 shows an example of a conventional optically coupled solid state relay using a photocoupler using a photothyristor as a light receiving element.

When an input signal is applied to the light emitting diode LD via the current limiting resistor R1, a forward current flows through the light emitting diode LD, causing it to emit light. Photothyristor receives this light
TY is fired, this current becomes the gate current of the bidirectional thyristor AS of the main switch, the bidirectional thyristor AS becomes conductive, and current flows from the AC power supply 10 to the load 12. Here the photothyristor TY
The transistor TR added between the gate and cathode of the photothyristor TY performs zero voltage switching, that is, allows the photothyristor TY to fire only when the voltage of the AC power supply 10 is close to zero voltage.

A base current determined by resistors R2 and R3 and the AC power supply voltage flows through the transistor TR. Therefore, when the AC power supply voltage is high, the transistor
TR becomes conductive, the gate and cathode of photothyristor TY are short-circuited, and even if an input signal is added and the photothyristor is irradiated with light, photothyristor TY does not fire. Note that R _g is a gate parallel resistance, and C is a hysteresis capacitor.

This method uses a bidirectional thyristor AS in the main switch.
This method uses a photocoupler for the gate circuit, but it has the disadvantage of a large forward loss due to the forward voltage drop of two diodes in the diode bridge DB. Furthermore, since the number of parts constituting the circuit is large, there are many problems in terms of cost and spatial size when modularizing these parts.

FIG. 2 shows another conventional example having a zero voltage switching function. The circuit for this example is
This is advantageous in that it does not require the use of the diode bridge in Fig.
Since the photothyristors TY1 and TY2 are designed to irradiate light, there will inevitably be differences in the response sensitivity and amount of incident light between each photothyristor TY1 and TY2, and it is necessary to increase the amount of emitted light and optical coupling efficiency so that these differences can be virtually ignored. This has the disadvantage of not being possible. That is,
In the circuit shown in Figure 2, two
The two photothyristors TY1 and TY2 generate an AC current of 10
and a load 12, and each time the instantaneous value of the AC voltage becomes substantially zero (including exactly zero and its vicinity), the bias generation circuit 14 connects the gate parallel resistance R between the gate and cathode of each thyristor.
A bias signal is generated to make control transistors TR1 and TR2 connected together with _g1 and R _g2 non-conductive. Therefore, thyristor TY1, TY
2 is in a state where it can be ignited by the trigger light from the light emitting diode only when the AC voltage is essentially zero, but in this case, each thyristor does not necessarily emit light in the same way due to variations in characteristics and differences in the amount of incident light. It does not necessarily mean that it will be fired. In other words, a situation occurs in which thyristor TY1 is light-ignited in one half cycle of AC, but thyristor TY2 is not light-ignited in the next half cycle. In order to eliminate this kind of inconvenience, it is necessary to increase the amount of light emitted by the light emitting diode LD, increase the optical coupling efficiency between the light emitting and receiving elements,
Countermeasures such as making the photosensitive characteristics of the photothyristors uniform are possible, but none of them are practical.

As mentioned above, although the circuit of Fig. 1 suffers from the disadvantages of using a diode bridge and the large number of components, it does not have the non-uniformity of firing as in the circuit of Fig. 2 (1 (Two light emitting diodes illuminate one photothyristor.)
Although the circuit of FIG. 2 has the advantage of overcoming the disadvantages and shortcomings mentioned above in the circuit of FIG. have

An object of the present invention is to utilize a bidirectional photothyristor with variable switching sensitivity that can effectively perform a zero-voltage switching function, and to utilize the advantages of the conventional zero-voltage switching circuit described above while eliminating its disadvantages. The object of the present invention is to provide a novel zero-voltage switching circuit.

One feature of the present invention is that in an element that can fire two thyristor regions arranged in parallel by applying the same optical signal to the light-receiving surface of a semiconductor substrate, two thyristor regions arranged in antiparallel Two p-base regions corresponding to each of the thyristors are exposed on the main surface of the light-receiving side, two external terminals are provided there, and an impedance is connected between the main terminal and the light ignition sensitivity, electrical The point is that the characteristics can be adjusted appropriately.

Another feature of the present invention is that by using the above-mentioned element in the part that connects the input and output of the optically coupled solid state switch, the number of circuit components is small, the forward loss is small, and the light source The object of the present invention is to make it possible to realize an optically coupled solid-state switch circuit that requires less power and has a zero-voltage switching function.

Preferred embodiments of the present invention will be described in detail below with reference to the drawings.

3 and 4 illustrate a planar bidirectional photothyristor according to an embodiment of the present invention, in which 20 indicates a pair of main surfaces 31 and 32 located opposite each other, and It is a silicon semiconductor substrate having connecting side end surfaces 33. This semiconductor body 10 has two four-layer thyristor regions 34, 35 each containing a thyristor element.
It is equipped with

One thyristor region 34 has p-type emitter layers P ₂₁ , n
It has a cross-sectional structure in which a type base layer N ₃₁ , a p-type base layer P ₁₁ , and an n-type emitter layer N ₁ are arranged in a laminated manner, and the other thyristor region 35 is directed from the main surface 31 side to the 32 side. Then, the p-type emitter layer P ₁₂ ,
It has a cross-sectional structure in which an n-type base layer N ₃₂ , a p-type base layer P ₂₂ , and an n-type emitter layer N ₂ are arranged in a stacked manner. In the thyristor region 34, due to the above layer arrangement,
PN junctions J ₁ , J ₃₁ and J ₄₁ are formed as shown in the figure, and in the thyristor region 35, pn junctions J ₂ , J ₄₂ and J ₃₂ are formed as shown in the figure by the above layer arrangement. The pn junctions J ₂ , J ₃₁ , J ₄₁ , J ₃₂ , and J ₄₂ all terminate at the main surface 31.

The two thyristor regions 34, 35 are each surrounded by an n-type isolation region _N0 and are electrically isolated from each other. The layers N ₁ , P ₁ /, N ₃₁ , P ₂₁ included in the thyristor region 34 form a first thyristor element, and the main electrode layer 22a is deposited on the n-type layer N ₁ on the cathode surface thereof. A main electrode layer 23a is also deposited on the p-type layer _P21 on the opposite anode surface. The layers P ₁₂ , N ₃₂ , P ₂₂ , and N ₂ included in the thyristor region 35 cause the second
A thyristor element is formed, a main electrode layer 22b forming a pair with the main electrode layer 22a is deposited on the p-type layer P ₁₂ on the anode surface, and a main electrode layer 22b on the n-type layer N ₂ on the opposite cathode surface is deposited. A main electrode layer 23b, which is paired with layer 23a, is applied. Main electrode layer 2
2a, 22b are connected together to one main terminal T1, and the main electrode layers 23a, 23b are also connected to the other main terminal T2.
are connected together.

On the one main surface 31 side of the base body 20, at the boundary between the two thyristor regions 34 and 35, the photosensitive regions of the respective thyristor elements, for example, the junctions J ₃₁ and J ₄₂ and their neighboring regions are exposed, and the boundary By irradiating the portion with trigger light from the light source 26, both thyristor elements can be ignited almost equally. A gallium arsenide light emitting diode is suitable as the light source 26.

Further, on one main surface 31 side of the base body 20, p-type layers are provided at positions away from the planned light irradiation area.
A portion of P ₁₁ and a portion of p-type layer P ₂₂ are exposed;
Auxiliary electrode layers 24 and 2 are provided on each exposed portion, respectively.
5 is coated. Each auxiliary electrode layer 24, 25 is connected to a corresponding auxiliary terminal T3, T4, respectively. An impedance device 3 such as a resistor or a transistor is provided between the auxiliary terminal T3 and the main terminal T1 and between the auxiliary terminal T4 and the main terminal 2, respectively.
6 and 37 are connected, and are used to adjust the switching sensitivity of each thyristor element by appropriately setting its impedance value.

In addition, in FIG. 4, 27 indicates one main surface 3.
1 is an oxide film that covers the parts not in contact with the main electrode layers 22a, 22b and the auxiliary electrode layers 24, 25, and 28 is an oxide film that covers the main electrode layer 23a, 28 including the surface of the isolation region _N0 on the other main surface 32. 23b
This is an oxide film that covers the parts that are not in contact with the surface.

Such planar bidirectional photothyristors can be made by selective diffusion techniques. For example, n
The exposed portions of the n-type layers N ₃₁ and N ₃₂ and the main surface 3 of the isolation region N ₀ are formed by both sides of the silicon substrate.
1 and 32 are covered with an oxide film as a mask, p-type impurities are diffused, and then an n-type layer is formed.
Both principal surfaces 3 are placed so as to expose the formation positions of N ₁ and N ₂ .
A predetermined bonding structure is obtained by covering 1 and 32 with an oxide film as a mask and diffusing n-type impurities, and then forming a main electrode layer and an auxiliary electrode layer by a method such as vapor deposition or plating. Figures 3 and 4
The device shown in the figure is obtained.

Next, the operation of the apparatus shown in FIGS. 3 and 4 will be explained. First, a voltage is applied between the main terminals T1 and T2 such that one main terminal T1 has a more negative potential than the other main terminal T2, and the pn junctions J ₁ and J ₄₁ are forward biased and the pn junction J ₃₁ is reverse biased. Explain the case where In this state, one main surface 3 is
When light energy hν (h: Planck's constant, ν: frequency of light) is irradiated to the area to be irradiated with light,
Electron-hole pairs are generated near the exposed portion of the p-n junction J ₃₁ , and the electrons are collected in the n-type layer N ₃₁ and the holes are collected in the p-type layer P ₁₁ , respectively. Photocurrent due to these electron-hole pairs flows through the electrode 24 and the impedance device 36 to the main terminal _T1 . At this time, a potential difference generated across the device 36 occurs in the direction of lowering the potential of the p-type layer _P11 and increasing the potential of the n-type layer _N1 .
This potential difference is the _fault voltage (built-in
voltage), at this point n
Injection of electrons from the p-type layer N ₁ to the p-type layer P ₁₁ is started. Electrons injected into the p-type layer P ₁₁ pass through the p-n junction J ₃₁ and are diffused into the n-type layer N ₃₁ and the like. The potential of the n-type layer N _{31 increases due to the electrons collected by this electron diffusion and the electron-hole pairs generated by the above-mentioned light irradiation, which are collected in the n-type layer N 31 .} ,pn
When the voltage applied to junction J ₄₁ exceeds the barrier voltage, p
Holes are injected from the type layer P ₂₁ to the n type layer N ₃₁ . The injected holes diffuse from the n-type layer _N31 to the p-type layer _P11 ,
Further forward bias the p-n junction _J1 . This promotes injection of electrons from the n-type layer _N1 . The above operations are repeated, and the layers P ₁₁ , N ₃₁ ,
Sum of current amplification factors for each equivalent transistor consisting of P ₂₁ and layers N ₁ , P ₂₁ , and P ₃₁ α _P11N31P21 + α _N1P
When _21N31 approaches 1, one of the four-layer regions 34 becomes conductive.

Next, one main terminal T is connected between the main terminals T1 and T2.
1 is applied with a voltage that is more positive than the other main terminal T2, pn junctions J ₂ and J ₃₂ are forward biased, and pn
The case where junction J ₄₂ is reverse biased will be explained. In this state, light is applied to one main surface 3 of the p-n junction J ₄₂ .
When the area to be irradiated including the exposed portion in No. 1 is irradiated, the other four-layer area 35 becomes conductive. The ignition function at this time will be explained. by irradiation with light
Electron-hole pairs generated near one main surface 31 of the p-n junction J ₄₂ are collected in the p-type layer P ₂₂ and electrons in the n-type layer N ₃₂ , respectively. The photocurrent due to the electron-hole pair is transmitted through the electrode 25 and the impedance device 37.
through to the main terminal _T2 . At this time, a potential difference generated across the impedance device 37 occurs in the direction of lowering the potential of the p-type layer P ₂₂ and increasing the potential of the n-type layer N ₂ . When this potential difference exceeds the barrier voltage of the p-n junction _J2 , the transition from the n-type layer _N2 to the p-type layer occurs.
Electron injection into P ₂₂ begins. Electrons injected into the p-type layer P ₂₂ pass through the p-n junction J ₄₂ and diffuse into the n-type layer N ₃₂ . The electrons collected by this electron diffusion and the electrons collected in the n-type layer N ₃₂ during light irradiation raise the potential in the n-type layer N ₃₂ and cause the voltage applied to the p-n junction J ₃₂ to exceed the barrier voltage. , holes are injected from the p-type layer P ₁₂ to the n-type layer N ₃₂ . The holes injected into the n-type layer N ₃₂ diffuse into the p-type layer P ₂₂ and further forward bias the p-n junction J ₂ . As a result, from the n-type layer _N2 to the p-type layer
Injection of electrons into _P22 is promoted. The above operations are repeated, and the layers N ₂ , P ₂₂ , N 32 and layers N 2 , P 22 , N ₃₂ and
The sum of the current amplification factors for each equivalent transistor consisting of P ₂₂ , N ₃₂ , and P ₁₂ α _N2P22N32 + α _P22N32P12 is 1
When approaching , the other four-layer region 35 becomes conductive.

Since the four-layer regions 34 and 35 are completely separated by the isolation region N ₀ , the problem of erroneous firing during commutation, which is a problem in the electric gate method, is solved. That is, for example, when the other four-layer region 35 is in a conductive state, a situation in which the accumulated charge in the conductive region flows into the vicinity of the pn junction J ₁ of one four-layer region 34 during commutation occurs in the isolated region. This can be completely prevented by N ₀ , and therefore, one of the four-layer regions 34 can be prevented from being erroneously turned on and turned into a conductive state by the leaked charge before being irradiated with the optical signal. Therefore, a planar bidirectional photothyristor suitable for high current and high speed applications can be obtained.

Further, the main electrode layer 22a on the side to which the optical signal is irradiated,
22b is divided into two, and the four-layer regions 34 and 35 are completely separated by the isolation region _N0 , which prevents dispersion of photocurrent generated by light irradiation and improves photosensitivity. can be achieved.

Next, the gate electrode 2 shown in FIGS. 3 and 4
The roles will be explained in 4 and 25. When this device is applied to the input/output coupling part of an optically coupled solid-state switch with a zero voltage switching function, it is necessary to devise a method to prevent the thyristor regions 34 and 35 from igniting even when irradiated with light. . for that purpose,
In the 4-layer region 34, the n-type layer N ₁ and the p-type layer P ₁₁ are
In the layer region 35, the n-type layer N ₂ and the second intermediate layer P ₂₂ are short-circuited, and the current amplification factor α of the transistor portion of N ₁ , P ₁₁ , N ₃₁ or N ₂ , P ₂₂ , _{N 32} is α _N2P22N32 must be disabled. For this purpose, in the 4-layer region 34, the p-type layer P ₁₁ and the 4-layer region 35 are
Then, an electrode from which an external terminal can be taken out is required on the p-type layer _P22 . In the case of the 4-layer region 34, there is no problem since the structure takes the p-gate terminal in the case of a normal electric gate system, but in the case of the 4-layer region 35, the region 3
If the terminal is taken out in a manner symmetrical to that in case 4, the external terminal must be taken out from the other main surface 32. However, when this element is mounted, a supporting electrode is provided on the other main surface 32 side, so this external terminal must be taken out while being electrically insulated from the main electrode 23b. The law is extremely difficult and has little practicality. If the electrodes are removed from the surface exposed on the main surface 31 side of the p-type layer _P21 as in this embodiment, mounting becomes extremely simple.

In addition, the electrical characteristics such as light ignition sensitivity or switching sensitivity, breakover voltage, and static dv/dt tolerance are determined by the n-type layer N ₁ and the p-type layer P ₁₁ (or the n-type layer N ₂ and the p-type layer P ₂₂ ). Since the characteristics vary greatly depending on the manner of short-circuiting, and since both characteristics are in a contradictory relationship, it is necessary to appropriately adjust the degree of short-circuiting depending on the use of the device.

Therefore, the n-type layer N ₁ has N ₂ and the p-type layer P ₁₁ or P ₂₂
In a case like this example where the and is not internally shorted,
Auxiliary electrode 24 from which auxiliary terminals T3 and T4 can be taken out,
25 is extremely useful when this device is used in an optically coupled solid-state switch circuit having a zero voltage switching function, and furthermore, it is possible to connect impedance devices 26 and 27 to adjust the impedance value appropriately. , it is possible to obtain light ignition sensitivity and electrical characteristics suitable for each application, which has the great advantage of widening the range of application and making it easier to use.

Next, with reference to FIGS. 5 and 6, the present invention will be described as follows.
An embodiment will be described in which the present invention is applied to a portion connecting input and output of an optically coupled solid-state switch circuit having a zero voltage switch function.

FIG. 5 shows an optically coupled solid-state switch circuit having a zero voltage switching function in which a planar type bidirectional photothyristor according to the first embodiment of the present invention is applied to the input/output coupling portion. A resistor for adjusting the firing sensitivity and electrical characteristics between one main terminal T1 and one auxiliary terminal T3 and between the other main terminal T2 and the other auxiliary terminal T4 of the bidirectional photothyristor PTY. Transistor with R _g1 , R _g2 and zero voltage switching function
TR1 and TR2 are added respectively. Resistors R5 and R6 are inserted between the base and emitter of these transistors TR1 and TR2, respectively, and when the voltage of the AC power supply 10 is high, a voltage is generated in R5 and 6R,
This voltage supplies base current to transistors TR1 and TR2, respectively. Incidentally, the capacitors C1 and C2 function as follows. To prevent unstable switching such as chattering, a difference is created between the input voltage when the switch is turned on and the voltage when it is turned off, and a so-called hysteresis function is provided to prevent the switch from reversing easily once it has been switched. It is something. In this circuit, hysteresis is obtained using the charge charged in C1 or C2. The operation of this circuit will be briefly explained. When DC input is applied, forward current flows through the light emitting diode LD, causing it to emit light.
In response to this light, one thyristor region of the bidirectional photothyristor RTY is fired, causing current to flow from the AC power supply 10 to the load 12. On the other hand, transistor
A base current determined by resistors R5 and T6 and the AC power supply voltage flows through TR1 (or TR2). Therefore, when the AC power supply voltage is high, transistor TR1 (or TR2) becomes conductive, and the main terminal and auxiliary terminal of one thyristor of the bidirectional photothyristor PTY are short-circuited, and the input signal is added to both. Thyristor PTY does not fire even when light is irradiated to the tropic photothyristor PTY. In other words, the bidirectional photothyristor is activated only when the AC power supply voltage is close to zero voltage and when light is irradiated.
PTY conducts and can supply current to the load 12, a so-called zero voltage switching function.

According to this first embodiment, since the bidirectional photothyristor according to the present invention, which can be bidirectionally triggered by light, is used in the coupling part between the input and output, it is possible to use the optically coupled solid-state relay shown in the conventional example (Fig. 1). The diode bridge DB used for this can be removed, and the forward voltage drop of two diodes is eliminated, so the forward loss can be reduced. In addition, in the conventional example, control was performed using the bidirectional thyristor AS of the main switch and the photothyristor TY of the optical coupling part, but in this example, both roles are performed by one bidirectional thyristor.
Since it is supplemented by PTY, the number of parts can be reduced, and the main switch is an electrically gated bidirectional thyristor.
Reliability can be improved because the problems associated with AS during commutation can also be resolved. Even when compared, the total number of parts can be reduced in this embodiment, which is advantageous when modularizing these parts. Furthermore, since one light emitting diode LD only needs to irradiate one light irradiation area of the photothyristor PTY according to the present invention, it is necessary to take special measures to prevent malfunctions such as those that were a problem in the conventional example shown in Fig. 2. Therefore, the driving power of the light emitting diode can be reduced and its life span can be extended, which is extremely advantageous in practice.

Next, a circuit according to a second embodiment of the present invention is shown in FIG. The principle of operation is exactly the same as that of the first embodiment shown in FIG. 5, so a description thereof will be omitted. The difference from the first embodiment is that diodes D1 and D2 and resistors R7 and R8 are added. Diode D1
and the role of D2 will be explained. transistor
Since the reverse withstand voltage between the base and emitter of TR1 and TR2 is about several volts, this is to prevent the transistors TR1 and TR2 from being destroyed by providing reverse voltage from the AC power supply voltage with the diodes D1 and D2. Next, the roles of resistors R7 and R8 will be explained. In the first embodiment (FIG. 5), since the circuit is symmetrical, the timing of the zero voltage switch is determined by the resistors R5 and R6, and there is little freedom. The timing for this zero voltage switching is set by the resistor R.
It is easy to use the circuit because it can be changed freely by inserting 7 and R8. Although the number of parts is increased compared to the first embodiment, the diodes D1 and D2 are not inserted in positions that cause forward loss, so
The advantages mentioned in the embodiment remain the same.

Next, the effects of the present invention will be explained numerically.

Compared to conventional optically coupled solid-state switch circuits, there is no need for rectifying diodes, so the forward voltage drop can be reduced from about 2.5V to about 1.5V. In addition, the electric gate type 10A, 400V, which is a conventional main switch.
Maximum voltage increase rate dv/dt and maximum current decrease rate without erroneous firing during commutation of class bidirectional thyristor itself
di/dt are 2V/μs and 5A/ms, respectively, but in this example they were 100V/μs or more.

[Brief explanation of the drawing]

Figures 1 and 2 are circuit diagrams showing conventional zero voltage switching circuits, and Figures 3 and 4 are circuit diagrams showing conventional zero voltage switching circuits.
The figures show a top view and a cross-sectional view taken along the line 2, respectively, of a bidirectional photothyristor according to an embodiment of the present invention, and FIGS. 5 and 6 show a zero-voltage switching circuit using the thyristor according to the present invention. It is a diagram. 20...Semiconductor substrate, 22a, 22b, 23
a, 23b... Main electrode layer, 24, 25... Auxiliary electrode layer, 26... Trigger light source, 36, 37...
Impedance device for adjusting switching sensitivity,
PTY...Bidirectional photothyristor, T1, T2
...Main terminal, T3, T4...Auxiliary terminal.

Claims (1)

[Scope of Claims] 1 (a) The base between the cathode electrode provided on the first main surface of the semiconductor substrate and the anode electrode provided on the second main surface opposite to the first main surface. a first thyristor element consisting of an n-type emitter layer, a p-type base layer, an n-type base layer and a p-type emitter layer which are sequentially stacked in the body; an anode electrode provided on the first main surface; and the second thyristor element. and a second thyristor element consisting of a p-type emitter layer, an n-type base layer, a p-type base layer and an n-type emitter layer sequentially laminated in the base between the cathode electrode provided on the main surface of the substrate. are arranged electrically separated from each other, and are terminated at each pn junction between the p-type layer and n-type layer of the first thyristor element and the first main surface, and the junction of the second thyristor element p-type emitter layer,
Between each of the n-type base layer and the p-type base layer
A p-n junction is terminated at the first main surface, a p-n junction between the p-type base layer and the n-type emitter layer is terminated at the second main surface, and the p-n junction is terminated at a portion near the boundary between the first and second thyristor elements. means for irradiating the exposed portion near the end of each pn junction with light;
the cathode electrode of the thyristor element and the anode electrode of the second thyristor element are connected to a common first main terminal; The cathode electrodes of the thyristor elements are connected to a common second main terminal, and the first and second thyristor elements are exposed at a position on the first main surface that does not interfere with light irradiation to the portion near the boundary. Auxiliary electrodes for switching sensitivity adjustment are provided on the exposed portions of both p-type base layers, and electrodes are provided between the auxiliary electrodes on the first thyristor element side and the first main terminal, and on the second thyristor element side. (b) a bidirectional photothyristor comprising first and second variable impedance devices connected between an auxiliary electrode and the second main terminal, respectively; circuit means for generating a zero voltage indication signal each time the instantaneous value of the alternating current voltage applied between the main terminals becomes substantially zero; the first and second variable impedance devices are applied as respective control inputs to the devices to cause the first and second variable impedance devices to transition from a low impedance state to a high impedance state each time the instantaneous value of the alternating current voltage becomes substantially zero; A zero voltage switching circuit characterized by maximizing the switching sensitivity of the elements.