JPS601780B2 - two-way switch - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a bidirectional switch for controlling AC power, especially an IG
This invention relates to a bidirectional switch using an FET (insulated gate field effect transistor).

More specifically, this switch is used, for example, to control motors, lighting equipment, etc. that operate on commercial power. I.G.
MOSFET, which is one type of FET, has been researched and developed as an element that can be integrated, and has been widely put into practical use in large-scale integrated circuits for calculators and memory. MOSFETs used in integrated circuits are low-power elements with a drain breakdown voltage of 10V and a drain current of 10-odd milliamps or less.

However, IGFETs have advantages such as the temperature coefficient of the drain current is negative, so there is no destruction due to thermal runaway, high-speed switching is possible because they do not use minority carriers, and the input impedance is high and gate drive power is low. It is expected to be used as a switching element for a large military force.

In response to the demand for higher power MOSFETs,
Offset gate type MOSFETs and VMOSFETs have been researched and developed as devices that can obtain high drain breakdown voltages and high drain currents, and devices with drain breakdown voltages of 200 V or more and on-resistance numbers Q or less are being put into practical use.

Below, Fig. 1 shows a conventional example in which the on-state of a switch is controlled using positive and negative voltages larger than the amplitude of the AC power source as control signal voltages, and Fig. 3 shows the dark value voltage of the transistor. A conventional example of a switch that controls the supply of power in a pulse manner by applying and releasing a voltage pulse of a certain degree of unipolarity will be described.

Figure 1 shows a VMOS with a drain breakdown voltage of 90V and an on state of 30V.
This is a conventional example of a bidirectional switch configured using two FETs.

In FIG. 1, source electrodes 1, 5 and substrate electrodes 4, 8 of two transistors Tr, Tr2 are commonly connected.

The drain electrode 3 of the transistor Tr is connected to one terminal 16 of an AC power supply 15 having an amplitude of Vp, and the drain electrode 7 of the transistor Tr 2 is connected to one terminal 13 of the load 12.

The gate electrodes 2 and 6 of the Tr, Tr2 are connected to one terminal of the control signal source 9. AC power supply 15, load 1
2, and the other terminals 17 and 14 of each of the control signal source 9
, 11' are commonly connected.

The operating principle of the conventional example shown in FIG. 1 will be explained using FIG. 2.
FIG. 2A shows an equivalent circuit in a period in which the terminal 16 of the AC power supply 15 has a positive potential difference with respect to the terminal 17.
In this case, the pn junction formed by the drain electrode 7 and substrate electrode 8 of Tr2 is forward biased, so Tr2 becomes conductive. Therefore, as shown in Figure 2 a, Tr
, performs a switch operation. Now the amplitude of AC power supply 15 is Vp
The output voltage of the control signal source 9 is defined as the potential difference between the terminals 10 and 11. ．． Since the diagram a in FIG. 2 is a source follower circuit, in order to extract the maximum amplitude Vp of AC to the load 12, the output voltage Von of the control signal source 9 must be Von,〉Vp+V, where the dark value of the transistor is Vt.
t...It is necessary to satisfy [1}.

On the other hand, in order to turn Tr, off, the output voltage Voff, of the control signal source 9 is VoH, <VT
...Just drop [2].

Next, consider a period in which the terminal 17 of the AC power supply has a positive potential difference with respect to the terminal 16.

In this case, the pn junction formed by the drain electrode 3 and substrate electrode 4 of the Tr is forward biased, and the Tr. becomes conductive.

Therefore, the circuit of FIG. 1 can be expressed as shown in FIG. 2, b.

In order to turn on Tr2, the gate electrode 6 of Tr2 is
Since it is sufficient that the potential difference between the voltage and the source electrode 5 is equal to or higher than the dark value Vt of the transistor, the output voltage Vo& of the control signal source 9 is Vo~>VT...
Just fill the lake.

On the other hand, in order to turn Tr2 off, the output voltage VoH2 of the control signal source 9 must be set to Voff2<VT-Vp.
・・・・・・You need to choose {4}.

m~'4} formula, T over the entire AC cycle. , Tr2 is turned on, the output voltage of the control signal source 9 is set to Vo.
n>Vt+Vp ・・・・・・
'51, and in order to turn off Tr, Tr2, the output voltage of the control signal source 9 is set to Voff<Vt-Vp.
...You need to choose a side.

That is, the slave column shown in FIG. 1 has a major drawback in that it requires a positive and negative voltage larger than the amplitude of the AC power source 15 as the output voltage of the control signal source.

Next, a conventional example in which a pulse transformer is used to isolate and separate the control signal source and the AC power supply circuit will be described with reference to FIG. In the dependent example shown in FIG. 3, compared to the conventional example shown in FIG. The difference is that the voltage is applied via 18.

As in the case of FIG. 1, when the terminal 16 of the AC power supply 15 is positive, Tr2 performs the switching function, and when it is negative, Tr2 performs the respective switch functions.

Assuming that the winding ratio of the pulse transformer 18 is 1, the output voltage of the control signal source 9 is equal to that of the gate electrodes 2 and 6 of Tr2.
and the source electrodes 1 and 5.

Therefore, in order to turn on the transistors Tr, .

On the other hand, in order to keep the Tr, . Since the conventional example shown in Fig. 3 uses a pulse transformer, Tr,, Tr2
cannot be kept on.

That is, Tr, Tr2 can only be turned on in a pulsed manner.

However, if the frequency response of the load 12 is slow, even if Tr,, Tr2 are repeatedly turned on in a pulsed manner over a certain period, it is assumed that Tr,, Tr2 are essentially always in the on state. I can imitate it. In this way, the conventional example shown in Fig. 3 is
Compared to the conventional example shown in Fig. 1, the output voltage of the control signal source is at most 1 ping of the threshold value Vt of the transistor, and it has the advantage that it is possible to isolate the AC circuit and the control signal source. On the other hand, pulse transformers are expensive, small,
The major drawback was that it required the use of parts that were difficult to reduce weight. SUMMARY OF THE INVENTION The object of the present invention is to eliminate the above-mentioned drawbacks and to provide a bidirectional switch for controlling alternating current power that is small, lightweight, and especially low-cost through hybrid integration.

According to the present invention, the gate electrodes, source electrodes, and substrate electrodes of two insulated gate field effect transistors of the same conductivity type are connected together, and the source electrode and the substrate electrode are connected, and the gate electrode and the source electrode are connected together. A bidirectional switch is provided in which the electrodes are connected with a resistor, a capacitance is connected to the gate electrode to serve as an input terminal to which one end of the control signal source is connected, and two drain terminals are used as controlled terminals. A capacitance or an equivalent stray capacitance for passing the control pulse signal is provided between the midpoint of the AC power supply connected between the control mode and the load and the other end of the control signal source. You can get a two-way switch featuring:

- The operating principle of the present invention will be explained below based on an embodiment of the present invention shown in FIG.

FIG. 4 shows an embodiment of the present invention in which n-channel MOS type field effect transistors are used as two insulated gate type field effect transistors.

Gate electrodes 2 and 6 of transistors Tr, Tr2,
Source electrodes 1 and 5 and substrate electrodes 4 and 8 are connected. The drain electrode 3 of the transistor Tr is connected to one end 16 of an AC power supply 15 (for example, 100V on a fifth circuit), and the drain electrode 7 of the transistor Tr2 is connected to one end 13 of a load 12.

Source electrodes 1 and 5 are connected to substrate electrodes 4 and 8.
Gate electrodes 2 and 6 and source electrodes 1 and 5 are connected via a resistor 19.

The other end 17 of the AC power source 15 and the other end 14 of the load 12 are connected.

The gate electrodes 2 and 6 are connected to one end of the control signal source 9 via a capacitance 20, and the other end 17 of the AC power supply 15 and the other end 11 of the control signal source 9 are connected to each other via an electrostatic capacitance 20. Capacity 21
connected via. Resistance 19, capacitance 20,2
The impedance determined by the series connection of 1 and the control signal source 9 can be chosen sufficiently high that leakage of AC power through this path can be ignored.

For example, if the resistor 19 is 100 KQ and the capacitances 20 and 21 are each 50 F, the impedance of the above-mentioned path for commercial frequency 5 plates z is approximately determined by the capacitances 20 and 21,
It is a sufficiently large value, ○ for 1 round. Normally, the impedance of the load is sufficiently small, so the impedance looking into the circuit from the control signal source 9 is mainly the capacitance 21, AC power supply 15,
The so-called drain capacitance between the drain electrode 3 and source electrode 1 of the Tr, a resistance 19, and a capacitance 20 are determined by a closed circuit connected in series.

Therefore, it is considered that the control signal mainly flows through this closed circuit. Next, estimate the drain capacity of Tr. When Tr,, Tr2 are in the off state, the substrate electrodes 4 and 8 are
Due to the rectifying action of the junction between the drain electrodes 3, 7 and the substrate electrodes 4, 8, the voltage amplitude Vp of the AC wire is charged. Since charging occurs in a direction that reverse biases the drain junctions of Tr, Tr2, even if voltages that change from positive to negative are applied to each drain junction from the AC power supply 15, the junctions will not be forward biased.

At this time, the drain capacitance changes depending on the voltage change of the AC power supply 15, but the range is generally from OV to V at the drain junction.
It is equivalent to the change in drain capacitance that occurs when a voltage of p is applied, and ranges from about one F to several tens of F. In the aforementioned closed circuit, in order for the output voltage of the control signal 9 to appear across the resistor 19 without changing its pulse waveform,
A time constant determined by the capacitances 20 and 21, the harmonic mean of the drain capacitances, and the resistance 19 is transmitted from the control signal source 9 to Tr, , T
It only needs to be sufficiently larger than the pulse width of the control pulse signal applied to turn r2 on.

If the resistor 19 is estimated to be 10 Cr and the drain capacitance is estimated to be 1 F in the worst case, the time constant will be 1 French sec, so it is sufficient to select the pulse width to be 0.5 French sec or less.

From another perspective, choosing the time constant as described above is
At the frequency of the control signal, the impedance of the resistor 19 is sufficiently large compared to the impedance of the capacitance, that is, it can be considered as open circuit. In such a state, the pulse output voltage of the control signal source 9 is equal to the capacitance 21, 20, T
It is divided into the gate input capacitance of r, and the drain capacitance of Tr. Normally, the capacitance is 21 compared to the gate input capacitance of a Tr.
, 20 are sufficiently large, and the drain capacity of the Tr is also sufficiently larger than the gate input capacitance when its changes are considered on average, so it can be considered that the pulse output voltage is almost applied to the gate. To turn on Tr,, Tr2, gate electrode 2
, 6 and the source electrodes 1, 5.
It is sufficient to apply a sufficiently large voltage of about 10 times t.

Therefore, if Vt is about IV, it is sufficient that the output voltage of the control signal source 9 is about 10V. By the way, when the above-mentioned time constant is not very large and the duty ratio D of the control signal source 9 is large, due to the nature of the CR circuit, T
The signal applied to the gates of r, , Tr2 has a waveform in which a negative DC voltage is superimposed on the output voltage of the control signal source.

Since the more positive voltage is applied to the gates of Tr, Tr2, the stronger the on state becomes, the on state becomes weaker by this negative DC voltage.

Therefore, in such a case, it is necessary to increase the output voltage of the control signal source 9. According to experiments, Vt is IV
When the duty ratio ◯ was about 10%, the output voltage of the control signal of 10V was enough to turn on Tr, Tr2.

During a period when no pulse output voltage is applied from the control signal source 9, the gate electrodes 2, 6 and the source electrodes 1, 5 are short-circuited by the resistor 19, so that the transistors Tr, Tr2 are in an off state. The resistor 19 is desirably larger in order to increase the above-mentioned time constant, but it also serves to reduce the input impedance of the gate electrodes 2 and 6 to prevent malfunctions due to noise, so it is necessary to take both into consideration when deciding on the resistor 19. .

The capacitances 20 and 21 have a large impedance with respect to the frequency of the AC power supply 15, and are substantially connected to the control signal source 9.
and AC power supply 15 are insulated and separated.

If there is no need for insulation separation, it will work even if the terminal 9 of the control signal source and the terminal 17 of the AC power supply 15 are directly connected, as long as the capacitance 20 is present. Further, the capacitance 21 does not necessarily need to be connected, and a stray capacitance between the terminal 11 of the control signal source 15 and the terminal 17 of the AC power supply 15 or the terminal 14 of the load 12 is sufficient for operation. As described above, according to the present invention, a control signal voltage that is sufficiently small compared to the voltage of the AC power source and larger than the threshold value of the transistor is applied in a pulse manner to turn the transistor on and off in a pulse manner, so that the voltage can be removed from the AC power source. A bidirectional switch that controls the supply of energy to a load and is inexpensive, small, lightweight, and easy to integrate into a hybrid can be obtained.

[Brief explanation of drawings]

1, FIGS. 2a and 2b, and FIG. 3 show conventional examples of bidirectional switches, and FIG. 2 is a diagram for explaining the operating principle of the conventional example of FIG. FIG. 4 shows an embodiment of the present invention. In the figure, 1,6
are source electrodes, 2 and 6 are gate electrodes, 3 and 7 are drain electrodes, 4 and 8 are substrate electrodes, 9 is a control signal source, 10 and 11
is a terminal of a control signal source, 12 is a load, 13 and 14 are load terminals, 15 is an AC power supply, 16 and 17 are terminals of an AC power supply,
18 is a pulse transformer, 19 is a resistor, and 20 and 21 are capacitors. O Figure O 2 Figure Juice 3 Figure 4 Ji

Claims (1)

[Claims] 1 Connect the gate electrodes, source electrodes, and substrate electrodes of two insulated gate field effect transistors of the same conductivity type, connect the source electrodes and the substrate electrodes, and connect the gate electrodes and the source electrodes with a resistor. In a bidirectional switch, the gate electrode is connected to a capacitor, and one end of the control signal source is connected as an input terminal, and the two drain terminals are controlled terminals. A bidirectional switch characterized by providing a capacitance or an equivalent stray capacitance for passing a control pulse signal between the midpoint between an AC power source and a load and the other end of the control signal source. .