JP5628184B2 - switch - Google Patents

Description

[Description of related applications]
The present invention includes Japanese patent applications: Japanese Patent Application No. 2009-214415 (filed on September 16, 2009), Japanese Patent Application No. 2009-272435 (filed on November 30, 2009), Japanese Patent Application No. 2010-050648 (2010). Filed on March 8) based on the priority claim, and the entire description of the application is incorporated herein by reference.
The present invention relates to a switch, and more particularly to a switch suitable for application to a direct current switch (DC switch).

  In data centers where power consumption is increasing, it is planned that DC of about 400V will be used. For this purpose, it is necessary to simplify DC switches and circuit breakers. Unlike the AC switch, the DC switch does not have a current zero point, so that energy is constantly supplied to the arc plasma. For this reason, it is difficult to cut off the current. Conventionally, a direct current (DC) system has been frequently used, and various circuit breakers and switches have been developed.

  One of them is a magnetic field application method. This uses the principle that an electromagnetic force works when a magnetic field is applied perpendicularly to the direction of current flow and pushes arc plasma in the direction perpendicular to each direction. It has been used in large circuit breakers. Currently, small switches using this principle have been developed. Each will be described below. 1 is a diagram showing a cross-sectional structure of an arc extinguishing part of an air circuit breaker described in Non-Patent Document 1 (The Institute of Electrical Engineers, “Electrical Engineering Handbook (6th edition)”, p. 755, 2001. Quoted from year). The blowout coil strikes the arc against the bulkhead or grid to lower the plasma temperature and extinguish the arc. When the interruption operation is started, the switch portion is opened and an arc is generated between the electrodes. Arc plasma hits the “partition or folds” and the “diaion grid” in FIG. It is blown away by the magnetic field created by the arc current itself, the magnetic field created by the blowout coil, and the electromagnetic force generated by the arc plasma current. As for the “partition wall”, iron is used in a structure as shown in the figure for households with a low voltage and devices up to AC 600V. When the arc plasma collides, the arc is cooled rapidly, the temperature drops and the resistance of the arc plasma increases, and the ion plasma in the plasma is recombined to reduce current carriers due to recombination and increase resistance. Become. For this reason, the ON voltage rises and current limiting action appears. As the air circuit breaker, a device having a rated current of 200 A to 6 kA and a rated breaking current of 5 to 125 kA is used. The charging operation includes manual operation and electric operation, and devices that are easy to operate and maintain are widely used. This type of air circuit breaker has not only a breaker function but also a current limiting function through the arc / plasma extinguishing action.

  Patent Document 1 discloses a DC circuit breaker in which a capacitor and a switch for commutating / cutting off a direct current in a non-vibrational manner are connected in parallel.

JP 2004-14241 A

The Institute of Electrical Engineers of Japan, “Electrical Engineering Handbook (6th edition)”, p. 755, 2001

  In a mechanical switch using a magnetic field, arc plasma is generated between the contacts, and the electrodes are damaged. For this reason, the lifetime (the number of times the switch is opened and closed) is shortened. In particular, when the current value becomes large, the damage becomes large, and the contacts may be exchanged after several times of opening and closing the switch.

  A switch using a semiconductor device does not cause a problem of life as described above, but it is expensive, requires a control circuit, has a large ON resistance, generates heat, requires a radiator, etc. , Power consumption becomes large.

  An object of the present invention is to provide a DC switch that achieves miniaturization and power saving and extends the life of the switch.

  According to a first aspect of the present invention, there is provided a switch comprising a mechanical switch having a metal contact and a semiconductor switch connected in parallel to the mechanical switch and controlled to be turned on and off by an electric signal. .

  In the present invention, after the semiconductor switch is turned on from the non-conductive state when the switch is turned on, the mechanical switch is turned on from the non-conductive state, and when the switch is turned off, the mechanical switch is turned off from the conductive state. The semiconductor switch is changed from the conductive state to the non-conductive state.

  In the present invention, a series circuit of the second switch and the first resistor may be connected in parallel with the load.

  In this invention, it is good also as a structure which connected the series circuit of the 2nd switch and the 1st resistance in parallel with the said mechanical switch.

  In the present invention, a series circuit of a first capacitor and a first resistor may be connected in parallel with the mechanical switch.

  In the present invention, a voltage dividing resistor for dividing the power supply voltage may be provided, and the voltage at the voltage dividing point may be input to the control terminal of the semiconductor switch. A parallel circuit of a first capacitor and an auxiliary switch may be connected between the power source and the voltage dividing point in parallel with one of the voltage dividing resistors.

  In the present invention, after the auxiliary switch is turned off from the conductive state, the mechanical switch is turned on from the non-conductive state, and after the mechanical switch is turned off from the conductive state, the auxiliary switch is turned off from the non-conductive state. Let

  In the present invention, the mechanical switch is a cutoff mechanical switch.

  In the present invention, the semiconductor switch may be changed from the conductive state to the non-conductive state after the interruption mechanical switch is changed from the conductive state to the non-conductive state.

In the present invention, each of the first and second contacts of the mechanical switch has a copper contact and a refractory metal contact, and in the first contact, the surface of the refractory metal contact protrudes from the surface of the copper contact. And
The second contact comprises an elastic member behind the refractory metal contact;
At the time of closing, the copper contact and the refractory metal contact abut on the same surface, one of the first and second contacts is a fixed contact, and the other is a movable contact.

In the present invention, each of the first and second contacts has a copper contact and a refractory metal contact,
The first contact is a fixed contact, and the second contact is a movable contact;
In the first contact, the refractory metal contact surface protrudes from the copper contact surface,
The arm that movably supports the second contact is made flexible, and when closed, the copper contact and the refractory metal contact abut on the same surface.
Furthermore, in the present invention, when detecting the arc plasma generated between the contacts of the mechanical switch due to failure to open the mechanical switch, and detecting arc plasma generated between the contacts of the mechanical switch, And a means for erasing the arc plasma.
In the present invention, a current detector is provided in the current path of the mechanical switch, and when the current is detected by the current detector even though the mechanical switch is off, the semiconductor switch is turned on. The arc plasma generated between the contact points of the mechanical switch may be extinguished by commutating the current flowing through the mechanical switch to the semiconductor switch. Alternatively, the present invention may be configured such that after a current is further passed through the semiconductor switch for a predetermined period, the semiconductor switch is turned off and arc plasma generated between the contacts of the other mechanical switch is extinguished.

  According to the present invention, downsizing and power saving can be achieved, and the life of the switch can be extended.

It is the figure quoted from the nonpatent literature 1. It is a figure which shows the structure of the 1st Embodiment of this invention. It is a timing diagram which shows the operation example of the 1st Embodiment of this invention. It is a figure which shows the structure of the 2nd Embodiment of this invention. It is a figure which shows the structure of the 3rd Embodiment of this invention. It is a figure which shows the structure of the 4th Embodiment of this invention. It is a figure which shows the structure of the 5th Embodiment of this invention. It is a timing diagram which shows the operation example of the 5th Embodiment of this invention. It is a figure which shows the structure of the 6th Embodiment of this invention. It is a timing diagram which shows the operation example of the 6th Embodiment of this invention. It is a figure which shows the structure of the 7th Embodiment of this invention. It is a figure which shows the structure of the 7th Embodiment of this invention. It is a figure which shows DC circuit example. It is a figure which shows the C switch system in the 8th Embodiment of this invention. It is a figure which shows the structure of a switch. It is a figure which shows 2 contact switch (the 1). It is a figure which shows 2 contact switch (the 2). It is a figure which shows a 2 contact switch circuit (the 1). It is a figure which shows the operation | movement sequence of a circuit. It is a figure which shows a 2 contact switch circuit (the 2). It is a figure which shows a 2 contact switch circuit (the 3). It is a figure which shows a 3 contact switch circuit (the 1). It is a figure which shows a 3 contact switch circuit (the 2). It is a figure which shows a 3 contact switch circuit (the 3). It is a figure which shows a 3 contact switch circuit (the 4). It is a figure explaining the condition where the arc plasma is generated between the contacts at the time of interruption failure. It is a figure which shows the structure of another embodiment of this invention. It is a figure which shows the structure of the modification 1 of FIG. It is a figure which shows the structure of the modification 2 of FIG.

  An embodiment of the present invention will be described.

<Embodiment 1>
FIG. 2 is a diagram showing a configuration of the first exemplary embodiment of the present invention. In the switch of the present invention, a mechanical switch and a semiconductor device are connected in parallel between the contacts. A mechanical switch having a metal contact and an IGBT (insulated gate bipolar transistor) are connected in parallel. C, E, and G are the collector, emitter, and gate of the IGBT.

  FIG. 3 is a timing waveform diagram for explaining the operation of the first embodiment of FIG. Next, the operation will be described. When switching the switch from OFF (non-conducting) to ON (conducting), first, a signal is given to the gate (G) of the IGBT to turn the IGBT on. This timing is time t1. At time t2 delayed from t1, the mechanical switch SW1 is turned on.

  Next, the operation when the switch is turned from ON to OFF will be described. When the IGBT is in the ON state, the mechanical switch SW1 is turned off (time t3). Thereafter, the IGBT is turned OFF at time t4.

  In the configuration of FIG. 2, when operated as in FIG. 3, almost no arc plasma is generated between the electrodes of the mechanical switch SW1. This is because, since the IGBTs connected in parallel are in an ON state when the metal contacts of the mechanical switch SW1 are opened and closed, the voltage between the metal contacts of the mechanical switch SW1 is, for example, several volts or less, and the current flows to the IGBT. Because. The current flowing through the mechanical switch SW1 is commutated to the IGBT in, for example, several μs (microseconds). As a result, the life of the switch can be extended.

  The power consumption due to the ON resistance of the IGBT is not large. This is because the voltage between the collector and emitter when the IGBT is ON is 2-3V, and the voltage between the contacts when the mechanical switch SW1 is ON is nearly two orders of magnitude lower than this. When the IGBT and the mechanical switch SW1 are in the ON state at the same time, most of the current between the terminals flows to the mechanical switch SW1 side. That is, the loss is equivalent to that of a conventional mechanical contact switch. For this reason, the substantial IGBT energization time is short. Accordingly, the amount of heat generated by the IGBT is reduced, and a heat sink attached to the IGBT is not necessary.

  Further, as shown in FIG. 3, after the mechanical switch SW1 is turned on, the IGBT may be turned off from time t2x to t3x in order to prevent a current from flowing through the IGBT. In most of the energization period, no current flows through the IGBT. Therefore, the problem of heat generation due to the ON resistance of the IGBT is solved.

  According to this embodiment, intelligent control from the outside becomes possible. That is, the switch can be remotely controlled from an external terminal or the like via the Internet or the like.

<Embodiment 2>
FIG. 4 is a diagram showing a configuration of the second exemplary embodiment of the present invention. This embodiment is effective when applied to a circuit (LR load) including a large inductance in the load. That is, since the circuit has magnetic energy due to the inductance, when the IGBT is cut off in a short time, a large surge voltage higher than the power supply voltage is generated between the collector and the emitter of the IGBT. Then, the IGBT is damaged. In order to avoid this, it is necessary to use a high voltage IGBT or to select an IGBT having a large absorbed energy. However, it becomes expensive.

  In the present embodiment, as shown in FIG. 4, the resistor R and the second switch SW2 are inserted in the circuit in parallel with the load (LR load).

  When the switch is shut off, the switch SW2 is turned on. Note that the switch SW2 may not be inserted. When the switch SW2 is deleted, a current always flows through the resistor R, and power consumption increases. The switch SW2 is not inserted, for example, when the switch SW2 remains OFF and the occurrence of a situation where the switch fails is avoided. In order to avoid the failure of interruption, the mechanical switch SW1 and the switch SW2 are mechanically interlocked. The surge voltage in which the inductance component is generated by the interruption of the switch SW2 is absorbed by the resistor R (magnetic energy disappears from the circuit as heat energy by the resistor R2). When the mechanical switch SW1 is turned off, the switch SW2 is always turned on. In FIG. 4, the resistor R and the switch SW2 are arranged on the load side, but generally there may be a large inductance component on the power supply side. In this case, a series circuit of the resistor R and the switch SW2 is preferably arranged on the power supply side with respect to the IGBT. In this case, after the switch SW2 is turned on, the switch SW2 is turned off after a certain period of time to prevent the current from constantly flowing through the resistor R.

<Embodiment 3>
FIG. 5 is a diagram showing the configuration of the third exemplary embodiment of the present invention. This embodiment is effective when there is a large inductance on the power supply side. A series circuit of a resistor R and a switch SW2 is connected in parallel to the mechanical switch SW1 and the semiconductor switch IGBT. When the mechanical switch SW1 and the IGBT are OFF (when the switch SW2 is ON), a current flows through the resistor R. The magnetic energy of the inductance element (component) in the circuit is converted into thermal energy by the resistor R. The value of the flowing current is limited by the resistance R, and the current can be cut by a normal switch that is 10 mA or less. The switch SW2 is turned off. Note that the surge voltage can be reduced by connecting a capacitor in parallel with the switch SW2. This can be said to be an IGBT snubber circuit (protection circuit).

<Embodiment 4>
FIG. 6 is a diagram showing the configuration of the fourth exemplary embodiment of the present invention. When the load includes a large amount of inductance component, a voltage higher than the power supply voltage is generated in the switch unit at the time of interruption. As a circuit for avoiding this, in this embodiment, as shown in FIG. 6, a capacitor C1 is connected in parallel to the mechanical switch SW1 and the semiconductor switch IGBT. In FIG. 6, a series circuit of a resistor R and a capacitor C1 is connected between the collector and emitter of the semiconductor switch IGBT. The capacitor C1 cuts a direct current component. Therefore, this is equivalent to the presence of the switch SW2 in FIG. Therefore, the switch SW2 in FIG. 5 is not necessary. In this circuit, the magnetic energy due to the inductance on the power supply side can also be absorbed by this portion. However, the circuit of FIG. 4 is used to process large magnetic energy.

<Embodiment 5>
FIG. 7 is a diagram showing the configuration of the fifth exemplary embodiment of the present invention. The control of the IGBT gate signal will be described. In an AC100V switch, a control circuit is rarely included in the switch. The amplitude of the IGBT gate signal is about 20V. Referring to FIG. 7, when the DC voltage on the power source side is divided by resistors R1 and R2, and the voltage between terminals is 400V, the ratio of voltage dividing resistors R1 and R2 (voltage dividing ratio) may be 380: 20. In order to reduce the current flowing through the voltage dividing resistor, for example, 380 kOhm and 20 kOhm are used. In this case, the leakage current is 1 mA and the loss is 0.4 W.

  A parallel circuit of the capacitor C1 and the auxiliary switch SW3 is connected in parallel between the terminals of the resistor R1, and the divided voltage is connected to the gate G of the IGBT. The voltage dividing ratio of the voltage dividing resistors R1 and R2 corresponds to the threshold value of the gate voltage of the IGBT. When the auxiliary switch SW3 is ON, the gate and emitter of the IGBT are at the same voltage, the IGBT is cut off, and no collector-emitter current flows. The auxiliary switch SW3 controls ON / OFF of the IGBT. Alternatively, as another method, a DC / DC converter that outputs a DC voltage of 400V to 20V may be used.

  When the auxiliary switch SW3 is turned off, a voltage is generated between the gate G and the emitter E, and the IGBT is turned on. Since the capacitor C2 is inserted, after the auxiliary switch SW3 is turned on, the IGBT is turned on with a delay of the charging time (time constant) of the capacitor C2. Subsequently, the machine switch is turned on. At this time, the auxiliary switch SW3 and the mechanical switch SW1 are configured to work together. In addition, by inserting the capacitor C2, since the current does not rise immediately when the switch SW3 is turned on, the possibility of damaging the IGBT can be reduced. For example, it is desirable that the auxiliary switch SW3 and the mechanical switch SW1 be interlocked with each other with a predetermined time delay using a relay.

  FIG. 8 is a diagram illustrating an operation sequence of the switch of FIG. The horizontal axis represents time, and the time transition of ON / OFF of the auxiliary switch SW2 and the mechanical switch SW1 is shown.

  First, the auxiliary switch SW3 is turned off. As a result, the IGBT is turned on.

  Next, a little later, the mechanical switch SW1 is turned on in conjunction.

  Therefore, first, current flows through the IGBT, and then flows through the mechanical switch SW1. This is because the ON voltage of the mechanical switch SW1 is low.

  When turning off the switch, first, the mechanical switch SW1 is turned off. Thereafter, the auxiliary switch SW3 is turned off. Also in this case, the two switches SW3 and the mechanical switch SW1 operate in conjunction with each other. Even when the mechanical switch SW1 is turned OFF, since the IGBT is in the ON state, almost no arc is generated between the contacts. For this reason, the contact of the mechanical switch SW1 is not damaged. This has been confirmed by experiments by the inventor.

  In the above example, the IGBT is used. However, any semiconductor device having a self-extinguishing capability can be applied to this circuit. For example, a power MOSFET, GTO (Gate Turn Off Thyster), or the like may be used.

  The current flowing through the semiconductor device is limited to the time between t1-t2 and t3-t4. If this time is short, the current capacity of the IGBT is rated for a short time, so a small IGBT can be used. Further, it is not necessary to attach a heat radiation fin or the like to the semiconductor device. In general, mechanical switches have a withstand voltage of DC400V and current capacities from 10A to 20A, so a very large switch is not required, and there are contacts with current capacities comparable to AC220V (DC conversion 311V). Available.

<Embodiment 6>
FIG. 9 is a diagram showing the configuration of the sixth exemplary embodiment of the present invention. The DC switch can also be applied to a DC circuit breaker. In particular, it is difficult to cut off high voltage and large current direct current. Therefore, the embodiment of the present invention will be described again focusing on blocking. FIG. 9 has the same circuit configuration as FIG. 2, but the mechanical switch SW1 is large. FIG. 10 is a diagram showing the operation sequence of FIG.

  The embodiment of FIG. 9 is basically constructed to be arc-plasma resistant and typically uses breaker contacts including alternating current. However, basically, since arc plasma is hardly generated, a normal metal contact using copper can be used. In this case, it is important to use a large high-speed contact. The direct current flows through the cutoff machine switch SW11. In FIG. 9, an IGBT is used as the semiconductor switch, but a GTO having a self-extinguishing capability for a large current may be used.

  The current flowing through the semiconductor device is limited to the time t1-t3 before and after the interruption. If this time is short, an IGBT having a small current capacity can be used as a short-time rating. The short-time rating is determined by the material and heat capacity, and can be significantly increased from the rated value by using SiC or the like.

  Although a current flows through the breaking machine switch SW11, when the IGBT is turned on after the IGBT is turned on, arc plasma is generated between the contacts, and the voltage is increased. Then, by this voltage, the current of the cutoff mechanical switch SW11 is commutated to the IGBT. As a result, the arc plasma between the contacts is extinguished. The time during which this arc plasma is generated is determined by the time constant of the circuit and the characteristics of the semiconductor switch and the mechanical switch, but it is usually several μs (microseconds), and the damage to the contact is small. This phenomenon is the same as in FIG. Thereafter, the IGBT shuts off.

  In addition, when an element such as a thyristor having no self-extinguishing function is used instead of the IGBT, a current may be injected using a capacitor or the like from the opposite direction as is well known. A thyristor is cheaper than an IGBT or the like, so that even if a current injection mechanism is attached, the thyristor may be cheap as a whole.

<Embodiment 7>
FIG. 11 is a diagram showing a configuration of the present embodiment. Since arc plasma is generated even in a short time, a high melting point metal or the like is used for the contact portion in the circuit breaker. Since the contact point of the refractory metal has a high resistance, heat generation at the contact point is increased, and the power consumption is increased. In the circuit breaker, the arc plasma generated at the time of breaking is preferably generated at a refractory metal contact, and a current flows through the copper contact during normal energization. In this embodiment, the contact consists of two members. One is a copper contact with low electrical resistance, and the other is a refractory metal contact with high electrical resistance but resistance to arc plasma. Tungsten, molybdenum and their alloys are used. In FIG. 11, the direction of the current is arbitrary. On the fixed contact side of the switch, the surface of the refractory metal contact protrudes from the surface of the copper contact 13. On the other hand, on the movable contact side, the refractory metal contact 12 is provided with a spring 11 on the back side. For this reason, when the switch is closed, the copper contact 13 and the refractory metal contact 12 can also contact on the same surface. With such a structure, the contact pressure of the refractory metal contact 12 is higher than that of the copper contact 13, but the copper contact 13 side also takes a sufficient contact pressure that allows current to flow. For this reason, most of the current flows through the copper contact 13 side having a low contact resistance.

  FIG. 12 shows a state when the switch starts to open. First, the copper contact portion 13 is opened. However, the refractory metal contact portion 12 is still in contact. In this way, the current flowing between the copper contacts 13 is commutated to the refractory metal contact 12 side. Therefore, no arc plasma is generated between the copper contacts 13. As the opening progresses and the distance between the contacts becomes longer, the refractory metal contacts 12 are finally opened. For this reason, arc plasma is generated in this portion. However, since the refractory metal contact 12 is resistant to arc plasma, damage is suppressed. This is because, since semiconductor devices such as IGBTs are connected in parallel, the arc / plasma generated in this portion is limited, and only a few μs (microseconds) is generated. Thus, contact damage is significantly reduced. With the structure as shown in FIGS. 11 and 12, a long-life switch can be configured even if semiconductor devices are not connected in parallel.

  The contact structure of this embodiment is resistant to arc plasma. It can be easily understood that such a structure has the same effect even when the arm 10 is bent without using a spring. Using the flexible arm 10, the copper contact 13 and the refractory metal contact 12 are pressed so as to be on the same surface. This is because the surface of the refractory metal contact 12 protrudes from the surface of the copper contact 13, so that the refractory metal contact 12 is still in contact with the copper contact 13 when the contact is opened. In the present embodiment, the spring 11 is used on the movable contact side, but a spring may be used on the fixed contact side or both. According to this embodiment, direct current can be easily interrupted. Of course, the switch structure described in the above embodiment may be applied to an AC switch or the like.

  According to the present invention, a semiconductor switch is connected in parallel to a mechanical switch, and an arc is not generated between the mechanical contacts by performing a cooperative operation with each other.

  However, DC switches can be more dangerous than AC switches. As a specific example, in a short circuit accident or the like, in AC, an NFB (Non-Fuse Breaker) connected in series to a switch cuts off a current. FIG. 13 shows a configuration in which NFB is incorporated in the DC switch. A DC NFB connected in series with a DC switch is connected to the load. If the load causes a short circuit accident, the DC NFB will shut off. The DC NFB has the same structure as the current DC switch. That is, semiconductor switches are connected in parallel, and a cooperative operation as described above with respect to the DC switch is performed. In the circuit of FIG. 13, when the DC switch is turned off so that no current flows through the load, it is assumed that the semiconductor of the DC switch is damaged and the current interruption fails. Then, the same current continues to flow through the load. Then, the DC switch will eventually burn up due to arcing if it continues as it is. This leads to a fire. However, since the current flowing through the circuit is almost the same as when the DC switch is ON, the DC NFB does not operate. In the AC switch, since there is a current zero point, it is easy to cut off, and the above-described problems are unlikely to occur. However, since the occurrence of an arc leads to a fire, further improvement in safety is desired.

  Therefore, the configuration shown in FIG. That is, when the current cannot be cut off even when the DC switch is turned off, the DC switch informs the upstream DC NFB by communication and performs a shut-off. It would be natural to use a power line to which a switch is directly connected as a communication line for performing this communication. Such a system allows each switch to set what relationship the switches are to each other when configuring the circuit. For example, the DC NFB is a parent and the DC switch is a child. Then, when the child switch fails to shut down, the parent switch is immediately contacted so that the parent switch cuts off the current. Furthermore, the parent switch also has a function of communicating with the outside, and if the child switch has failed to shut down, the parent switch has a function of notifying the administrator (personal computer or mobile phone). If this is done on a regular basis, it can be seen that there is a problem with the communication function if there is no contact, which can also be useful for system maintenance. Further, if the parent switch has a function of monitoring ON / OFF of a plurality of child switches, the system can be operated more safely. Then, it is made clear which child switch has failed to be cut off. A function of notifying the administrator of such information is provided. These make the switch system safer. In the above description, the upstream parent switch is DC NFB. However, since this has only a large current capacity, a large switch may be used. Further, the semiconductor switch of the parent switch may fail at the same time. At that time, another large NFB switch is provided, and there is a system in which the semiconductor switches are not connected in parallel. This corresponds to a case where the semiconductor element of this circuit is completely destroyed by a surge such as lightning.

  When configuring a switch for power applications, a circuit that completely cuts off only with a metal contact may be required. That is, in the circuit as shown in FIG. 15, the semiconductor switch (IGBT) cuts off the current, but voltage is always applied between the collector (C) and the emitter (E), which may be considered dangerous. Because. That is, it means that it is necessary to put a mechanical switch in series with the IGBT. This automatically requires at least two mechanical switches. First, an example of a mechanical switch structure is shown in FIG. The cable, the rotating unit, and the arm in FIG. 16 correspond to the cable, the rotating unit, and the arm in FIGS. In the rotating part, the two movable electrodes are electrically insulated from each other (insulating layer 1), but are mechanically integrated. Moreover, the fixed contact side also has two contacts, and each is electrically insulated (insulating layer 2), but is mechanically integrated. And although it is electrically wired to each contact, because of such a structure, when operated, the two contacts have slightly different ON times. In the following, the two contacts are SW1 and SW2, and when ON, SW2 is turned ON slightly earlier than SW1, and then SW1 is turned ON. Further, at the time of OFF, SW1 is turned off first, and thereafter, SW1 is turned off mechanically. Still, this time difference is structured and adjusted so that there is not much variation.

  FIG. 17 shows another similar example. This is a push-type switch, and it is turned ON when pushed out. As in FIG. 16, there are two contacts, one of which protrudes more through the spring structure in the direction of the other contact, so that it contacts quickly when it is ON. When both are ON, the springs are contracted and contacted together. Further, when turned off, the side with the spring is delayed and turned off. Then, by adjusting the spring length or the like, the time difference between when the two switches are turned on and off can be adjusted.

  Next, FIG. 18 shows a DC switch circuit using the two-contact switch of FIG. Since the switch enters between the power source and the load and does not place the load (= device) on the high voltage side, the switch is inserted on the high voltage side when there is only one switch. The two contacts are SW1 and SW2, and SW2 is turned on earlier in time when turned on, and turned off later in time when turned off. The two metal contacts are mechanically integrated and move in conjunction with each other. As shown in FIGS. 16 and 17. In FIG. 18, the power supply terminal P1 of the timer (Timer) is connected to the high voltage side via the switch SW2, and the terminal P2 on the ground side is directly connected to the ground. For this reason, when the switch SW2 is closed, power is supplied to the timer (Timer). When the switch SW2 is closed and the switch SW1 is OFF and the IGBT is also OFF, a voltage is generated at both ends of the resistors R1 and R2. Therefore, a trigger signal (trig) of the timer (Timer) is supplied to the trigger signal (trig). Voltage) is applied. Then, an output signal from the output terminal out of the timer (Timer) is applied to the gate of the IGBT, and the IGBT is turned on (conductive). In the example of FIG. 18, the IGBT is connected to the high voltage side, but may be the ground side. The same applies to the following drawings. When an example of the ground side is written, it may be on the high voltage side. For example, FIGS. 20 to 24 correspond to this. In addition, two switches are used, and this circuit (FIG. 18) also includes a configuration in which independent switches are used with their time differences controlled separately from the switches as shown in FIG.

  19A and 19B are diagrams showing a time sequence of the operation of the circuit of FIG. The circuit operation will be described below using the circuit configuration of FIG. 18 and the time sequence of FIG. In addition to the two mechanical contact switches, the circuit includes a self-extinguishing semiconductor switch (which may be an IGBT or a MOSFET) and a timer (FIG. 18). The timer is operated by connecting terminals P1 and P2 to a power source and obtaining power. When a trigger signal is input to the trig terminal, a pulse (pulse width = T1, T2) is output to the gate of the IGBT immediately after the output terminal pulse signal is output. In the following, T1 and T2 and the pulse are written in two parts, but they may be the same time (may overlap). Thus, the IGBT is set to be in the ON state while the pulse is active. The resistors R1 and R2 (voltage dividing resistors) are connected between the switch power supply side and the switch load (EBT and E of the IGBT), and the middle point is connected to the trigger terminal of the timer (Timer). The voltage of the trigger signal that defines the trigger signal that enters the trig terminal is defined. The timing operation is as shown in FIG.

  First, the operation when ON is described. When the switch SW2 is first turned ON, the switch SW1 is still OFF. Then, the drive power is supplied to the timer (Timer), and the timer enters an operating state. At the same time, a trigger signal is input to the trigger terminal. Then, a pulse signal is immediately output from the output terminal of the timer (Timer). A pulse signal is input to the gate of the IGBT and the IGBT is turned on. The time is within a few microseconds. This is illustrated in FIG. 19B, in which the pulse signal input to the gate of the IGBT is turned ON (Gate ON) with a delay of Td after the trigger signal is input. The pulse width of the pulse signal input to the gate of the IGBT is T1. On the other hand, the switch SW1 is not turned on yet, but the switch SW1 is turned on while the pulse signal inputted to the gate of the IGBT is in the activated state (High) (T1). For this reason, no arc is generated between the contacts of the mechanical switch SW1, which is the main switch. As a result, the entire switch is turned on. T1 and T2 are assumed to be about several milliseconds to several tens of milliseconds.

  Thereafter, since the output pulse signal of the timer (Timer) (pulse to the gate to the IGBT) is not output, the IGBT is turned off. For this reason, the current flowing through the circuit flows through the switch SW1 and the switch SW2 is ON. However, since the IGBT is OFF, no current flows through the SW2 and IGBT paths.

  Next, the operation when the switch is OFF will be described. When the switch is OFF, first, the switch SW1 is turned OFF. At this time, the switch SW2 is still in the ON state, and the IGBT is in the OFF state (Gate OFF). The timer (Timer) is in a standby state because the terminals P1 and P2 are connected to the power source. When the switch SW1 is turned off, the divided voltage of the resistors R1 and R2 is applied as a trigger signal to the trigger terminal of the timer (Timer). As a result, a pulse (pulse width T2) is output from the out terminal of the timer (Timer) to the gate signal of the IGBT, and the IGBT is turned on. Then, an arc is hardly generated between the metal contacts of the switch SW1, and the switch SW1 is turned off. When the pulse width T2 elapses, the IGBT is turned off.

  At this time, since the IGBT and the switch SW1 are OFF, the divided voltage of the resistors R1 and R2 is again applied as a trigger signal to the trigger terminal, but the timer (Timer) once receives a signal at the trigger terminal. , T3 time is set to be insensitive. That is, T3> T2 is set, and the timer (Timer) outputs a terminal pulse signal by this trigger signal, and therefore does not output a pulse signal to the gate of the IGBT. Therefore, the current zero is maintained during the dead time. The switch SW2 is turned OFF during the current zero period. Therefore, no arc is generated between the contacts. Finally, no arc is generated at either contact point of the switches SW1 and SW2, and the current interruption is completed.

  When a timer is used, time such as T1, T2, and T3 can be set, so that the entire switch system can be configured in accordance with the characteristics of the mechanical switch.

  As another example of the similar two-contact circuit, the circuit configuration is shown in FIG. In FIG. 20, the high voltage side and the ground side are opposite to those in FIG. Moreover, although the electric current direction is shown by the arrow, this has shown that the high voltage | pressure side is a plus side. In DC, a high voltage may be taken on the negative side, but in this case, it is necessary to reverse the collector (C) and emitter (E) of the IGBT, so that a predetermined measure is required. The timing operation in FIG. 20 conforms to FIG.

  In the example shown in FIG. 20, the switch SW2 enters first when ON. When the switch SW2 is turned on, a slight arc is generated between the contacts. However, since the arc is extinguished when the switch SW2 is turned on, there is no problem. When the switch SW2 is turned ON, power is supplied to the timer (Timer) and operation starts. At this time, since the switch SW1 is OFF, a trigger signal is input to the trigger terminal of the timer (Timer), a pulse signal is output from the out terminal to the gate of the IGBT, and the IGBT is turned on.

  While the IGBT is ON, the switch SW1 is turned ON. For this reason, no arc is generated between the contacts in the switch SW1. When the pulse width T1 elapses, the IGBT is turned off. However, since the switch SW1 is ON, no trigger signal is input to the trig terminal, so that a pulse signal from the out terminal to the gate of the IGBT is not output, and the IGBT is maintained in the OFF state. However, since power is continuously supplied to the Timer, the timer (Timer) is in a standby state.

  When the switch is OFF, first, the switch SW1 is turned OFF. Electric power is supplied to the timer (Timer) and the standby state continues. Then, a trigger signal is input to the trigger terminal of the timer (Timer), a pulse signal is output from the out terminal, and the IGBT is turned on. For this reason, no arc occurs between the contacts of the switch SW1. Thereafter, when the period T2 elapses, the IGBT is turned off. As soon as the trigger signal is input to the trigger terminal of the timer (Timer), the timer (Timer) does not output a gate signal from the out terminal within the dead period T3, so that the IGBT is kept OFF. The circuit current is then held at zero during this period. During this period, the switch SW2 is turned off. Since no current flows at this time, no arc is generated between the contacts of the switch SW2.

  FIG. 21 is a diagram showing a modification of FIG. The timer (Timer) takes different power, and the connection parts of P1 and P2 are different. That is, the terminal P1 is the same as P1 in FIG. 20, but the ground side of P2 is connected to the ground side of the switch SW1 and the IGBT. The operation is the same as in FIG. Since the timer is directly connected to ground, it will be resistant to noise.

  Next, a circuit example using three contacts as an extension of the above circuit is shown below. This is because, in the switches described so far, the shut-off portion is partially connected including the power source and the ground. However, in order to be more secure as a switch, it may be desirable for the switch to be completely disconnected from the power source. For this reason, three contacts were used. In this example, three switches are used, but the discussion proceeds based on the assumption that each switch is controlled independently and operates in conjunction with the sequence. In these examples, the IGBTs are all connected to the ground side, but may be on the high voltage side as described above.

  In fact, the specification standard for NTT switches is that electrical circuit elements such as semiconductor switches, resistors, capacitors (capacitors), inductors, and the like must not be connected when the switches are OFF. However, as shown in the circuit of FIG. 22 and subsequent figures, by providing three switches, it is possible to completely disconnect from the power source, so that the circuit can comply with NTT standards.

  In the example of FIG. 22, a switch SW3 is provided as a mechanical contact on the ground side. As a result, the load (= device) is completely disconnected from the power supply side, so that safety is improved. The switch SW3 is also mechanically interlocked with the switches SW1 and SW2. For this reason, three contacts are used. Similarly, the mechanical switches SW1, SW2, and SW3 operate in conjunction with each other. An example of this is a switch used in an AC three-phase circuit. The operation is as follows. The load is assumed to be grounded. The operation is as follows.

  1) At the time of ON, the switches SW2 and SW3 are turned ON almost simultaneously first. Then, power is supplied to the timer (Timer), a trigger signal is input to the trigger terminal, a pulse signal input from the out terminal to the gate of the IGBT is output with a pulse width of T1, and the IGBT is turned on. During this time, the switch SW1 is turned on. For this reason, no arc occurs between the contacts of the switch SW1. An arc is generated between the contacts of the switches SW2 and SW3, but this is a timing at which the switch is turned on, so that the final contact is made, so that the arc is generated only for a short time, so there is no problem.

  2) When the ON state is held, current flows through the switches SW1 and SW3, but no current flows through the switch SW2 because the IGBT is OFF.

  3) At the time of OFF, switch SW1 is turned OFF first. Then, since the IGBT is OFF, a trigger signal is input to the trigger terminal of the timer (Timer), and a pulse signal input from the out terminal to the gate of the IGBT is immediately output for a period T2, the IGBT is turned ON, and the switch SW2 An arc is not generated between the metal contacts, and is turned off.

  When the pulse width T2 elapses, the IGBT is turned off, so that the circuit current becomes zero. Then, the trigger signal immediately enters the trigger terminal of the timer (Timer), but during the period T3, the timer (Timer) does not receive the pulse signal input from the out terminal to the IGBT gate even if the trigger signal is input. Since it is not output, the OFF state of the IGBT is maintained. While the circuit current during that time is zero, either one of the switches SW2 and SW3 is OFF. Since the current is zero, no arc is generated between the contacts. Since all of the switches SW1, SW2, and SW3 are turned off, the switch unit is disconnected from the high voltage side and the ground side of the power source, and the cutoff is completed.

  FIG. 23 is obtained by changing a part of FIG. This is only a change in the position of the switch SW2 in FIG. The operation is based on FIG. In this case, the two contacts of the switches SW2 and SW3 are in series with the current flowing through the main circuit. For this reason, if the interruption fails, the arc voltage is at least doubled, which may be safer than the circuit of FIG.

  FIG. 24 is a diagram showing a modification of FIG. This is different from FIG. 23 in taking the power of the timer, and corresponds to the relationship between FIG. 20 and FIG. Therefore, the operation is the same.

  FIG. 25 shows a configuration provided with a free wheeling diode corresponding to a case where the load has a large inductance component. The operation is the same as in FIG. The surge voltage generated in the switch part at the time of interruption can be absorbed by the reflux diode. In addition, the energy of a circuit is absorbed by putting resistance in series with a diode.

<Another embodiment>
In a DC switch, “failure to open” does not mean that a large current flows like a short circuit, but a situation where almost the same current flows. It is difficult for other devices to detect “shutdown failure”. When the DC switch fails to shut off, arc plasma is generated, so the electrode may melt and the switch may be destroyed, which may cause a fire (fire) or the like. If there is a disconnection failure or the like once during the life of the switch, it cannot be actually used. In addition, it is considered that the failure of shut-off can hardly cause a direct fire because the distance between the electrodes is increased and the current zero is generated many times in alternating current.

  The DC switch also needs a countermeasure against the failure of interruption. This will be described with reference to FIG. FIG. 26 shows a situation in which arc plasma is generated between the contacts of the three switches SW1, SW2, and SW3 due to the failure of interruption. FIG. 26 is the same as the configuration of FIG.

As arc plasma detection,
・ Detection of voltage between contacts,
・ Measurement of electrode temperature,
・ Measurement of light and current from arc plasma and magnetic field measurement.

  In the present embodiment, a current sensor CT (Current Transformer) is used for current measurement. The signal from the CT enters the Cin of the timer / control system. When the signal from the CT is not zero even though the switches SW1 and SW2 are in the OFF state, the timer / control system determines that the interruption has failed. At this time, if the arc plasma is cut off even at one of the switches SW1 to SW3, the current interruption is successful. Therefore, the following countermeasures are taken for the failure of interruption.

(1) The IGBT is fired again by a signal (gate voltage) supplied from the timer / control system out to the gate. As a result, the current flowing through the switch SW1 is commutated to the IGBT. Since the interruption has failed, the timer / control system is supplied with driving power and can operate normally.

(2) As a result, the arc plasma of the switch SW1 is extinguished. At this time, since current flows through the IGBT, the arc plasma generated between the contacts of the switches SW2 and SW3 continues to be generated.

(3) When a current flows through the IGBT for several milliseconds, for example, the withstand voltage between the contacts of the switch SW1 is recovered. Thereafter, the IGBT is turned OFF by a signal (gate voltage) supplied from the timer / control system out to the gate of the IGBT. As a result, the circuit is opened and shut off. As a result, the arc plasma generated between the contacts of the switches SW2 and SW3 is also extinguished.

(4) If the operation is not performed even if the operations (1) to (3) are performed, the operations (1) to (3) are further repeated. This repeats until the block is successful. This situation is equivalent to experiencing zero current many times with the AC switch contact open. It becomes possible to deal with the DC switch of the above embodiment.

(5) When interruption is not possible even if the above steps (1) to (3) are repeated a certain number of times, a signal is sent to the upstream DC switch to cut off the current on the upstream side.

  In addition, it is important to add a function of notifying the signal that there has been a disconnection failure through a communication such as the Internet, by a telephone or e-mail from an upstream DC switch or a switch that has failed.

  Of course, this embodiment can be applied to any configuration other than FIG. 24, such as FIGS. 18, 20, 21, 22, 23, and 25.

  Commercially available DC switches are generally of the same type as ordinary AC switches, and the electrodes are made of a refractory metal material, and measures are taken such as enclosing them in a special gas. I can't take it. For this reason, it is considered that the DC switch cannot be widely used in homes, offices, factories, etc., unless measures against interruption failures are realized.

  According to this embodiment, it is expected to make a technical contribution to widely use DC.

<Still another embodiment>
FIG. 27 is a diagram showing the configuration of still another embodiment of the present invention. Switches SW1, SW2, SW3, and SW are provided as mechanical switches, and two switches are used as self-extinguishing elements such as IGBTs. The switches SW1 and SW2 are inserted on the + high-voltage side, and the switches SW3 and SW4 are inserted on the negative-side high-voltage side. Are provided with timers (Timers) 1 and 2 for inputting a voltage divided by resistors R1 and R2 to a trigger terminal (trig). When ON, the switches SW1, SW2, SW3, and SW4 are turned on (conducted) almost simultaneously, so that current flows through the load between the + high voltage side and the ground and the load between the high voltage side and the ground. start.

  When the switches SW2 and SW4 on the + high voltage side and the −high voltage side are turned on, power is supplied to the timers (Timers) 1 and 2, and the operation is started. At the same time, since the switches SW1 and SW3 are OFF, a trigger signal is input to the trigger terminals of the timers (Timers) 1 and 2, a pulse signal is output from the out terminal to the gates of the IGBTs 1 and 2, and the IGBTs 1 and 2 are turned on. Become.

  While the IGBTs 1 and 2 are in the ON state, the switches SW1 and SW3 are turned on. For this reason, no arc is generated between the contacts in the switches SW1 and SW3. When the pulse width T1 elapses, the IGBTs 1 and 2 are turned off. However, since SW1 and SW3 are ON, the trigger signal does not enter the trigger terminals of timers 1 and 2, so the pulse signal from the out terminal to the gates of IGBTs 1 and 2 is not output, and the IGBT remains in the OFF state. Is done. However, since power is continuously supplied to the timers 1 and 2, the timers 1 and 2 are in a standby state.

  At the time of OFF, first, the switches SW1 and SW3 on the + high voltage side and the −high voltage side are turned OFF. Electric power is supplied to the timers (Timers) 1 and 2, and the standby state continues. Then, a trigger signal enters the trigger terminals of the timers (Timers) 1 and 2, a pulse signal is output from the out terminal, and the IGBTs 1 and 2 are turned on. For this reason, no arc is generated between the contacts of SW1 and SW3. Thereafter, when the period T2 elapses, the IGBTs 1 and 2 are turned off.

  When the switches SW1, SW2, SW3, and SW4 are ON, no current flows through the IGBT. For this reason, as in the above embodiments, the loss does not increase due to the current flowing through the semiconductor element.

  When the switch is OFF, a gate signal is applied to the IGBTs 1 and 2 and the IGBTs 1 and 2 are turned ON. The switches SW1 and SW3 are turned off immediately. Since current flows to the IGBT side, no arc is generated between the contacts. Since this time difference is shortened, there may be no problem even if an arc is generated between the electrodes of the metal contact switch for a short time. Thereafter, the IGBTs 1 and 2 are turned off, and the direct current is cut off. Finally, the switches SW2 and SW4 are turned off, and the load is disconnected from the power supply side. In the present embodiment, the metal wire switch is not included in the ground wire. However, when necessary, a switch that is simultaneously turned off may be provided in conjunction with the switches SW2 and SW4.

  When the inductance of the load is large, a large surge voltage is generated when the IGBT cuts off the current. In this case, as a first modification of FIG. 27, as shown in FIG. 28, a series circuit of a diode D and a resistor R that absorbs energy is connected between the + high voltage side and the ground, and between the −high voltage side and the ground. This reduces the electrical energy that must be processed by the IGBT and improves safety.

  FIG. 29 is a diagram illustrating a second modification of the present embodiment. FIG. 29 shows a configuration in which a fuse is used to cut off current when it becomes impossible to cut off current due to an IGBT failure. Referring to FIG. 29, the configuration of FIG. 28 further includes a fuse 1 for cutting off direct current between the collector terminal of the IGBT 1 and the connection point of the switches SW2 and SW1 on the + high voltage side. A DC current blocking fuse 2 is provided between the collector terminal of the IGBT 2 and the connection point of the switches SW4 and SW3. Since the ON time of the IGBT is short, the fuse does not operate during this time, and the current is not interrupted by fusing. In general, a semiconductor switch element or the like may be in an ON state at the time of a failure, and if the current cannot be cut off due to the failure, the current continues to flow through the fuse for a long time. This time can be adjusted by selecting a fuse. For this reason, if a current continues to flow through the IGBT for a predetermined period, it is determined that there is a failure and the fuse operates (melting). A recovery circuit using a fuse can be applied to any of the above embodiments.

  It should be noted that the disclosures of the above-mentioned patent documents and non-patent documents are incorporated herein by reference. Within the scope of the entire disclosure (including claims) of the present invention, the embodiments and examples can be changed and adjusted based on the basic technical concept. Various combinations and selections of various disclosed elements are possible within the scope of the claims of the present invention. That is, the present invention of course includes various variations and modifications that could be made by those skilled in the art according to the entire disclosure including the claims and the technical idea.

10 Arm 11 Spring 12 High melting point metal contact 13 Copper contact 14 Rotating part 15 Cable

Claims (3)

First and second mechanical switches having metal contacts connected in series between a first load having one end grounded and a positive power supply;
A first semiconductor switch connected in parallel to the first mechanical switch and controlled to be conductive or non-conductive by an electrical signal;
A first timer that receives a voltage divided by a voltage dividing resistor between a connection point of the second mechanical switch and the first mechanical switch and the first load side as a trigger signal and outputs a pulse having a predetermined pulse width. When,
Third and fourth mechanical switches having metal contacts connected in series between a second load having one end grounded and a negative power supply;
A second semiconductor switch connected in parallel to the third mechanical switch and controlled to be conductive or non-conductive by an electrical signal;
A second timer for receiving a voltage divided by a voltage dividing resistor between a connection point of the fourth mechanical switch and the third mechanical switch and the first load side as a trigger signal and outputting a pulse having a predetermined pulse width; When,
The output pulses of the first and second timers are supplied to the control terminals of the first and second semiconductor switches to control on / off of the first and second semiconductor switches. A first circuit in which a diode and a resistor are connected in series between a connection point of the first mechanical switch and the first load and the ground;
A second circuit in which a diode and a resistor are connected in series between a connection point between the third mechanical switch and the second load and the ground;
And a switch of claim 1, wherein. A first fuse connected between connection points of the first semiconductor switch and the first and second switches;
A second fuse connected between connection points of the second semiconductor switch and the third and fourth switches;
And a claim 1 or 2 switch according.

Images