HK1226194B - Commutating switch with blocking semiconductor - Google Patents

Description

Reversing switch with blocking semiconductor

Technical Field

The present invention relates to a commutating switch, such as a circuit breaker.

Background Art

In order to open any DC circuit, the induced energy stored in the magnetic field due to the flowing current must be absorbed; this induced energy can be stored in capacitors or dissipated in resistors (the arc formed during the opening of the circuit is a special case of a resistor in this sense). The main difficulties in defining the drop-out points (resistance levels) for commutating circuit breakers using ohmic resistors are that: (1) for each drop-out point, the increase in transient voltage depends on the flowing current and the resistance introduced during commutation, and (2) the rate of increase of the current (during the fault) or the rate of decay (after the introduction of the resistor) depends mainly on the inductance in the most severe fault type, a "dead short", where the system resistance is almost zero; the inductance and the system resistance (external to the circuit breaker) can vary greatly in real faults. Therefore, it would be ideal to calculate and define a suitable drop-out point to insert the resistor at each circuit breaker operation in order to obtain the target maximum transient voltage difference across the inserted resistor, but this is not practical if ohmic resistors are used.

When varistors, reverse Zener diodes or transoscillators are inserted in the circuit, they develop a reverse electromotive force (EMF) that absorbs the energy stored in the fault; this can be viewed as a highly nonlinear resistor, but it is also reasonable to view it as a battery that loses all the stored energy during charging but still manages to control the voltage of the "charging current" well.

Due to the rapid inrush of current during a short circuit, the induced energy can easily be much greater than the induced energy stored in the system at normal full load alone; if the current reaches five times the normal full load amperage before being controlled, the induced energy can reach up to twenty-five times the induced energy of the circuit at normal full load (depending on the location of the short circuit). Until recently, test standards for DC circuit breakers have assumed a slow operation corresponding to arc chute circuit breakers (the standard DC circuit breaker design since Edison's time), where the time to open the electrodes after receiving the trip signal is typically greater than or equal to three milliseconds (ms); reaching the point where the current begins to decrease can take even longer (up to ten milliseconds). This means that high currents can build up through arc chute circuit breakers during a short circuit, potentially reaching the maximum capability of the DC power supply. To this end, the DC circuit breaker standard for circuit breakers applicable to electric trains in the United States (ANSI/IEEE 37.20) requires the circuit breakers to be able to handle a current of 200,000 amperes (200 kiloamperes, "kA"), approximately the maximum short circuit current of a DC fault in electric train subway systems.

The second class of mechanically switched DC circuit breakers includes the innovative, fast-acting High Speed Vacuum Circuit Breaker (HSVCB) DC circuit breakers from Hitachi (see, for example, U.S. Patent 4,216,513), which are based on the use of inductors and capacitors to create an L-C resonant circuit, coupled to an AC vacuum circuit breaker, interrupting the current when it passes through zero. These circuit breakers expose the insulation and circuit components of a normal DC circuit to rapid voltage reversals and voltage spikes. For L-C resonant circuit breakers for DC rail applications, Japanese regulations (standard JEC-7152) allow for a lower maximum current (50 kA) compared to the 200 kA that slower arc-chute rail breakers must withstand. This is made possible by the faster circuit-breaking action of the L-C resonant circuit breaker. Essentially, in such a circuit breaker, the discharge of the capacitor (electronic triggering) creates an L-C resonance that causes the current to oscillate through zero (much like an AC circuit). This oscillation decays quickly, but during this decay the vacuum circuit breaker opens the circuit when the current crosses zero.A recent US patent application (13/697,204) shows that this mechanism is also applicable to high voltage DC (HVDC) circuits.

The fastest known way to cut off DC power is to open the circuit using switchable power electronics; these devices are typically semiconductors (thyristors or transistors), but vacuum tubes can also be used. In these designs, the resistance of the switch itself is an important consideration because the load of the entire circuit passes through the switch in the on state. In the case of the most commonly used type of power electronic switch, namely the integrated gate bipolar transistor (IGBT), typical losses when conducting can be 0.25-0.50% of the transmitted power, which is unacceptably large for many applications and also implies very large cooling loads for high-power circuits, which typically require pumped liquid coolant. The need for active cooling increases cost and environmental impact, and reduces the reliability of the switch.

ABB is a leading developer of another approach to accelerating the operation of DC switches. This approach involves a circuit breaker that is a hybrid of a power electronic switch and a mechanical switch, but maintains lower on-state losses than a purely power electronic circuit breaker. In this hybrid approach, there are at least two power electronic switches combined with a fast mechanical switch. The first power electronic switch is a low-loss, low-voltage switch that commutates the current to a second path through a second power electronic switch capable of withstanding higher voltages (but with higher on-state losses). The second power electronic switch can be composed of a stack of IGBT transistors, a stack of gate turn-off (GTO) thyristors, or a variety of other current-interrupting devices. Before the second power electronic switch can be electronically turned off, the first low-voltage electronic switch must be protected from the combined voltage surge by a series-connected mechanical switch; the second, high-voltage-capable circuit breaker cannot open the circuit until the movable electrode of the mechanical switch reaches a minimum spacing to prevent arcing or restrike. This series-connected mechanical switch is the slowest component in the switch, so making the mechanical switch faster can make the hybrid switch even faster. Currently used fast mechanical switches have electrodes that are accelerated magnetically via electromagnetic repulsion or by means of capacitor discharge through a Thompson coil (induced magnetic repulsion) and that are separated in a vacuum or in a gas, which may be sulphur hexafluoride gas or a gas mixture.

In a hybrid circuit breaker for medium voltage DC (MVDC), the first low-voltage switch is preferably an IGCT (integrated gate-commutated thyristor); for a high voltage DC (HVDC) hybrid circuit breaker, the first low-voltage switch is preferably a single-stage IGBT that commutates current to an IGBT array having a plurality of series-connected IGBTs, each of which is connected in parallel with a metal oxide varistor (MOV). The second high-voltage capable disconnect switch may include an array of series-connected IGBT transistors, a stack of gate turn-off thyristors (GTOs), a cold cathode vacuum tube, or a similar power electronic switch capable of interrupting power flow.

It has been reported that Thompson coil-actuated mechanical switches have a response delay time of approximately 100 microseconds due to the mechanical response of the connected electrodes.

If the hybrid switch also functions as a circuit breaker, an energy-absorbing damper (e.g., a semiconductor blocking device or a capacitor bank) is also necessary to absorb the inductive energy stored in the magnetic field created by the current. The hybrid circuit breaker described above is an example where a large portion of the inductive energy stored in the HVDC circuit (which can exceed 100 MJ) is absorbed by the semiconductor blocking devices during operation of the circuit breaker.

Summary of the Invention

The present disclosure includes a mechanical switch that opens an electrical circuit by commutating current to an energy-absorbing path, or a sequence of paths, through at least one blocking semiconductor, wherein the commutation is caused by the sliding motion of at least one shuttle electrode over at least one fixed electrode. The blocking semiconductor may comprise a varistor (e.g., a polymer-based varistor or a metal oxide varistor, "MOV"), a Zener diode (effective at blocking in only one direction (opposite), or a transient voltage suppressor diode (TVS) (blocking bidirectionally up to a breakdown voltage). The blocking semiconductor absorbs at least a portion of the stored inductive energy to enable circuit opening and has a controlled maximum voltage (TVS diodes are referred to herein as "transient absorbers"). To prevent arcing during the sliding switch when the electrodes separate, at least one of the electrodes preferably has a region of increased resistivity that forms the final portion of the electrode to electrically connect to a matching electrode, thereby defining a circuit through the on-state of the switch. In the normal on-state, current flows through the low-resistance portion of the matching electrode, but with the switch open, the current is commutated to at least one well-defined second energy absorption path through a non-linear, non-ohmic resistor that blocks current below a breakdown voltage threshold (e.g., a varistor (which may be a polymer-based varistor, or a metal oxide varistor, "MOV"), or a transient voltage suppressor diode or a Zener diode; all such voltage-limiting semiconductor devices are referred to herein as "blocking semiconductors").

The variable resistivity trailing edge portion of the electrode can be attached to the shuttle electrode, the stator electrode, or preferably both. The graded resistivity in the trailing edge of the electrode prevents arcing when the electrodes separate for a range of experimentally defined fault conditions for voltage, current, capacitance, and inductance; current and inductance are particularly important because they determine the amount of magnetic energy stored in the flowing current that must be dissipated or stored to open the circuit.

The switch commutates current to at least one parallel path through a blocking semiconductor. The use of two or more blocking semiconductors can distribute voltage, provide a useful safety margin, or reduce voltage excursion during switching. In some cases, it is also desirable that the blocking semiconductor device be integrally connected to the stator electrode on which the shuttle electrode moves, thereby generating a voltage gradient in the stator electrode. The switch of the present disclosure is equally applicable to AC or DC power, but has particular advantages for DC power.

The present disclosure features a reversing switch. The reversing switch may include a fixed portion having a fixed electrode and a movable portion having a movable electrode. When the fixed electrode and the movable electrode are in conductive contact, a closed switch position is defined. Furthermore, the movable portion is movable relative to the fixed portion to interrupt the conductive contact between the fixed electrode and the movable electrode, thereby defining an open switch position. When the switch is open, current is commutated into an electrical path that may also include a nonlinear, non-ohmic blocking semiconductor.

The movable portion may include a shuttle or a rotor. The fixed electrode and the movable electrode may be contained in a dielectric fluid, the dielectric fluid being under a hydraulic pressure of at least 1 MPa, more specifically greater than 10 MPa. The fixed portion may include two fixed, spaced-apart electrodes, and the separate electrical paths may be connected through the two fixed, spaced-apart electrodes.

The fixed electrode may include a plurality of adjacent isolated conductors. When the switch is open, each of the movable electrodes can be electrically contacted with one of the isolated conductors. Alternatively, when the switch is open, the movable electrode can be electrically contacted with at least two of the isolated conductors simultaneously.

The commutation switch may have a plurality of nonlinear, non-ohmic blocking semiconductors in an electrical path into which current is commutated when the switch is open. The plurality of nonlinear, non-ohmic blocking semiconductors may be arranged in a stack. The nonlinear, non-ohmic blocking semiconductors may be metal oxide varistors (MOVs) arranged in a stack, the MOVs being arranged in such a manner that movement of the commutation electrode causes current to move through an increasing number of MOVs, resulting in a progressively increasing voltage across the stack. The MOVs may be arranged so that the edges of the sheets supporting the MOVs extend to a region where the edges of the sheets are in direct contact with the moving shuttle electrode, such that the voltage variation between adjacent sheets does not exceed four volts under normal operating conditions.

The fixed part may be a stator and the movable part may be a rotor. The rotor may be held partially fixed by friction, the friction coming from the tightly fitting stator being in contact with the rotor over a substantial portion of the surface area of the rotor. The stator may surround the rotor and the stator may include an interchangeable keystone-shaped member. The keystone-shaped member may be held against the rotor by elastic force or external hydraulic pressure applied to an impermeable membrane, the impermeable membrane surrounding the keystone-shaped member. The stator may include a plurality of commutation stages, a plurality of stator electrodes, and a resistor: each of the commutation stages includes two commutation regions, each commutation region includes a conductive wire; each of the stator electrodes is electrically coupled to the conductive wire; the resistor is located between each stator electrode and the conductive wire, wherein the two conductive wires of the two regions of each stage are electrically connected by a blocking semiconductor. At least some of the stator electrodes may include liquid metal.

The electrodes may slide apart. One or both of the fixed electrode and the movable electrode may have a graded, increased resistivity region forming a rearmost portion of the electrode that is electrically connected to the other electrode when the switch moves from the closed position to the open position.

The reversing switch may include at least two blocking semiconductors in a series electrical path. The fixed portion may include a series of stacked metal oxide varistors. The varistors may be annular and have varying outer diameters. The movable portion of the switch may be under stress in the closed position. The blocking semiconductors may be selected from the group consisting of a varistor, a Zener diode, and a transient voltage suppressor diode.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 shows a rotary motion commutation switch having two blocking semiconductors connected to the stator electrodes carrying current when the commutator rotates;

Figure 2 shows a rotary commutation switch that commutates current to a single blocking semiconductor to open the circuit;

FIG3 shows comparative current-voltage characteristics of three types of blocking semiconductors;

FIG4 shows a multi-stage commutation circuit breaker with a rotary motion having six commutation zones;

FIG5 shows a multi-stage commutation circuit breaker with linear motion of four commutation zones with two stages;

FIG. 6 shows a reversing circuit breaker with a shuttle in the form of a rod, tube or wire.

DETAILED DESCRIPTION

The present disclosure relates to a solid-state mechanical switch capable of breaking an electrical circuit without arcing between separated electrodes. This disclosure builds upon the disclosure of PCT/US2012/058240, which is incorporated herein by reference in its entirety. The present disclosure includes blocking semiconductors that replace some or all of the ohmic resistors described in the PCT application. Such blocking semiconductors include, but are not limited to, varistors, Zener diodes, and transient absorbers; in all of these, current is champed at a breakdown voltage and defined by a critical current density (e.g., 0.001 amperes per square centimeter). All blocking semiconductors can be compared based on their breakdown voltage (the voltage at which current begins to flow), effective voltage control range (the voltage range over which the blocking semiconductor usefully controls the voltage across its terminals), energy absorption capacity, and life expectancy. The difference between metal oxide varistors (MOVs) and transistors such as Zener diodes and transient absorbers lies in fatigue life. MOVs degrade with each use. Large energy pulses cause MOVs to degrade more than small energy pulses, and keeping track of the MOV's condition can become a maintenance issue. With careful monitoring, MOVs can be reliably reused, but for designing low-maintenance switches or circuit breakers, transoscillators are technically preferable, although they are more expensive.

The switches of the present disclosure are particularly advantageous for DC power due to the arc-less mechanism that quenches the electrical energy, although they can also be used for AC power. The present disclosure also allows for more compact switches because the total electrode displacement required to achieve a given withstand voltage level can be reduced compared to designs using air, vacuum, or gas. In the particular case of rotary motion switches, a high-pressure dielectric liquid environment (which suppresses arc formation) can be maintained around the switch mechanism using only a relatively small volume of liquid.

The present innovation uses highly nonlinear resistors (blocking semiconductors) in the switch, eliminating the need for the switch to commutate through multiple resistive steps to open the circuit. Prior art switches have used varistors, transoscillators, or Zener diodes to perform final circuit disconnection, but only after a series of ohmic resistors have absorbed most of the stored inductive energy. The present innovation recognizes that it is desirable to utilize highly nonlinear resistive semiconductor devices (such as varistors or transoscillators) to absorb most of the stored inductive energy. Prior art commutating switches rely on multiple commutations of current through multiple paths to quench the inductive energy. In the switch of the present disclosure, a single commutation to the blocking semiconductor can open the circuit. A fundamental advantage of using a blocking semiconductor to achieve final circuit disconnection is that the voltage is nearly constant during the absorption of the inductive energy, whereas using an ohmic resistor to absorb most of the inductive energy requires multiple commutations of the resistor inserted into the circuit, with a voltage increase followed by an exponential voltage decay after each commutation. Besides the complexity of the mechanism for implementing multiple commutations of the resistor into the circuit, the repeated exponential decay necessarily has an average voltage below the maximum voltage, a significant factor in switch insulation. The maximum voltage must be limited to control damage to dielectric components from transient high voltages. Because the inductive energy is quenched as the integral of (voltage) × (current) over time, maintaining a constant high voltage close to the maximum voltage during quenching (which can be achieved using a blocking semiconductor) can result in faster quenching of the switch of the present disclosure compared to other switches. Alternatively, the maximum voltage can be reduced without extending the time it takes to quench the inductive energy.

Example

FIG1 shows a simple design of a single-pole rotary switch (e.g., a circuit breaker) according to the present disclosure. In this case, the circuit breaker's rotation is driven by a splined shaft 101, which rotates about its axis 100 in the direction of arrow 120. During operation of the switch, the rotor (comprising 103, 111, 131, 133, and 134) rotates clockwise through angle 120 or angle 135. Component 103 is a solid insulating material with good strength, such as a glass fiber reinforced polymer or a self-reinforcing polymer (e.g., a liquid crystal polymer); component 103 can also be formed from a composite foam that is cast around the rotor poles 111, 131 and the electrical connections 133, 134 that electrically connect the two rotor poles together. Such composite foams are desirable for 103 due to their combination of low density and high stiffness. The entire rotor moves as a rigid body inside a conformal shell formed by 24 keystone-shaped segments, each covering a 15-degree angle of the shell; different segments have different electrical properties: keystones 105, 107, 109, 125, 127, 142, and 144 comprise portions of the stator poles, through which current occasionally flows, and 140 is an insulating segment repeated for most of the conformal shell (only some of the insulating segments 140 are labeled in the figure). Elastic sleeves or fluid pressure may be used to push all the keystone segments against the rotor.

Electrons move through the switch of FIG1 from the relatively negative terminal 102 to the relatively positive terminal 122; electrode segments 105, 107, 109, and 144 are connected to terminal 102 via connecting wires 104, 106, and 108, while electrode segments 125, 127, 129, and 144 are connected to terminal 122 via connecting wires 124, 126, and 128. Electrode segments 107 and 127 are partially semiconducting, but have an insulating layer adjacent to 109 or 129, and may also be graded within themselves according to resistivity, such that the resistivity increases from the edge adjacent to 105 or 125 to the edge adjacent to 109 or 129. Segments 107 and 127 are electrically connected to segments 105 and 125 except for insulating boundary layers 136 and 138 that extend partially outward from the inner diameter where the stator and rotor electrodes meet along the boundary between the conductive stator electrodes 105, 125 and the semiconducting stator electrodes 107 and 127. Insulating boundary layers 136 and 138, which extend partially outwardly from the inner diameter where the stator and rotor electrodes meet along the boundaries between the conductive stator electrodes 105, 125 and the semiconducting stator electrodes 107 and 127, function to reduce significant local heating that would otherwise occur in semiconducting electrode 107 at the boundary between electrodes 105 and 107 as the trailing edge of rotor electrode 111 moves from 105 to 107, or in semiconducting electrode 127 at the boundary between electrodes 125 and 127 as the trailing edge of rotor electrode 131 moves from 125 to 127. The function of the insulating boundary layers can also be achieved by grading the resistivity of the semiconducting electrodes 107 and 127, or even more preferably, by grading the resistivity of the trailing edges of rotor electrodes 111 and 131. The grading of the electrode trailing edge is thoroughly discussed in PCT/US2012/058240, which is incorporated herein by reference. Outside the insulating boundary layers 136 and 138, there are conductive boundary layers 137 and 139 that extend outward from the outermost radius of the insulating boundary layers 136 and 138 to the outer edges of the stator electrodes 105 and 107 (for 137) or to the outer edges of the stator electrodes 107 and 109 (for 139).

There are two pairs of side-by-side stator electrode segments: 109 and 144, and a second pair of stator electrode segments 129 and 144, connected by blocking semiconductors 110 or 130. These side-by-side stator electrode segments are electrically connected to each other and to the blocking semiconductors 110 or 130 via connecting wires 108 and 128. The entire rotary switch is enclosed in a pressure vessel 141 filled with high-voltage dielectric insulating oil 143. In practice, the volume of high-voltage dielectric insulating oil is typically much smaller than shown in the figure, because the inner edge of the high-pressure vessel 141 is expected to fit almost tightly against the outer edge of the keystone (105, 107, 109, 125, 127, 129, 140, 142, and 144) forming the solid stator contacting the rotor, in order to minimize the volume of high-voltage dielectric fluid. A device (not shown) to hold the keystone against the rotor is also required, such as an extended elastomeric sleeve or a liquid-filled bladder (containing a fluid at a higher pressure than the fluid surrounding the electrodes), which is inserted between the pressure vessel 141 and the outside of the 24 keystone segments that make up the stator. In the case where the keystone segments are held against the rotor by one or more liquid-filled bladders, the pressure within the bladders can be adjusted to adjust the normal force of the keystone segments against the rotor.

FIG1 illustrates several aspects of the present disclosure. In this device, power is commutated via two blocking semiconductor devices 110 and 130. Consider a case where the two blocking semiconductors have a breakdown voltage 20% higher than the normal line voltage and are capable of controlling voltage during surges up to 150% of the normal line voltage. Consider a case where the rotor rotates through an angle 120 of 45 degrees. At the end of this displacement, the blocking semiconductors 110 and 130 remain in the circuit. In this given scenario, operating the DC switch of FIG1 in a circuit where the stored inductive energy is sufficient to drive current through the two blocking semiconductors would generate a voltage three times the normal line voltage during the circuit opening process. In this case, the two blocking semiconductor devices remain in the circuit in a fully off state, and the series connection of the two blocking semiconductors creates a safety redundancy where one of the blocking semiconductors can fail and the switch will still be off, resisting the large amount of inductive (stored magnetic) energy. Alternatively, the breakdown voltages of the two blocking semiconductors 110, 130 can be chosen so that blocking the current requires the series connection of the two devices to block the current; this produces a lower overvoltage when the circuit is opened (only 1.5 times the normal operating voltage, according to the above assumptions), but a lower degree of safety.

At the end of the rotor rotation angle of 120 degrees, both blocking semiconductor devices are in the circuit. In the case of a high-inductance fault, most of the inductive energy quenched during the circuit opening is absorbed by the blocking semiconductors, while a small amount of inductive energy is absorbed by the semiconducting stator electrodes 107 and 127. In the case of a low-inductance, low-current fault, the inductive energy can be absorbed primarily or even entirely by the semiconducting stator electrodes 107 and 127. During the opening of the switch, the device of FIG. 1 can also rotate through an angle of 135 degrees (=90 degrees), in which case the rotor electrodes 111 and 131 will rotate beyond the point where the electrical connection through the blocking semiconductor is lost (at 60 degrees of rotation). This provides some safety margin if the blocking semiconductor is an MOV (the most economical choice). When an MOV fails due to fatigue, a short circuit forms across the device due to repeated overheating of a specific path. For the initial milliseconds of this MOV electrical fatigue failure, only a small current typically flows, but as this current further damages the MOV, the fault current through the MOV rapidly increases. The arrangement of Figure 1 has a good chance of surviving such an MOV failure through a 90-degree rotation because the MOV is removed from the circuit within a few milliseconds of triggering the switch. This is desirable if all the inductive energy can be reliably dissipated during the connection-in-place period across the blocking semiconductor; otherwise, damaging overvoltage transients can occur when the rotor rotates beyond 60 degrees (for the design of Figure 1).

FIG2 is a comparable but simpler version of the single-pole switch of FIG1 . The circuit is opened by rotating the rotor 155 clockwise through an angle 150, which has an axis of symmetry 152 and a radius 151, beyond the angle at which the rotor and stator electrodes lose contact. In this case, a blocking semiconductor 160 is placed in parallel circuit with the switch; when the rotor's rotation causes the voltage between the two stator electrodes 162 and 166 to exceed the breakdown voltage of the blocking semiconductor 160, the current is commutated to the blocking semiconductor 160. In the conductive state, the stator electrodes 162 and 166 contact the rotor electrodes 154 and 156, as shown in FIG2 . All four conductive electrodes (154, 156, 162, 166) are highly conductive, such as copper or silver, or a copper- or silver-based composite structure, and each is bonded to a trailing edge electrode (157, 168, 155, or 164) of graded resistivity. As the highly conductive electrodes separate, the graded resistivity electrodes continue to carry current with increasing resistance until most of the current is commutated to the parallel path through the blocking semiconductor 160. When the graded resistivity electrodes separate, the current flowing between them is very small (less than 0.1 amps), and the voltage is controlled within the effective voltage control range of the blocking semiconductor 160. This prevents substantial arcing from forming when the electrodes separate, although small sparks may still occur.

FIG3 shows the relationship between the current per square centimeter on the vertical axis and the voltage on the horizontal axis for two types of MOVs. The MOV characteristics shown are for silicon carbide MOVs (SiC; 171, 181) and zinc oxide MOVs (ZnO; 172, 182); approximate characteristics of transient absorbers (173, 183) are also included for comparison. Compared to silicon carbide-based MOVs, zinc oxide-based MOVs exhibit significantly higher voltage sensitivity in the region just above the breakdown voltage. The transient absorber has higher voltage sensitivity and has a greater slope in the current-voltage curve just above the breakdown voltage than the zinc oxide-based MOV. In FIG3, the current-voltage curves 173, 183 of the transient absorber lie inside the zinc oxide-based MOV curves 172, 182, and similarly, the zinc oxide-based MOV curves lie inside the silicon carbide-based MOV curves 171, 181. The Zener diode will follow the curve of the transosber for negative voltages (183) (reverse bias), but will simply conduct current in the forward direction (positive voltage). The cost per unit of energy absorption capacity of both the transosber and the Zener diode is significantly higher than that of the varistor. There are solutions where each of a zinc oxide-based MOV, a silicon carbide-based MOV, a transosber, and a Zener diode may be suitable for use in at least some switches of the present disclosure.

FIG4 shows a conceptual rotary multi-pole commutated circuit breaker designed as a medium for a high voltage DC or AC power circuit breaker. In this case, six commutation regions are shown: Region 1 includes elements 221-229 (including rotor pole 221; stator poles 222, 223, 224, and 225; conductive lead 226; and resistors 227, 228, and 229); Region 2 includes elements 231-239 (including rotor pole 231; stator poles 232, 233, 234, and 235; conductive lead 236; and resistors 237, 238, and 239); Region 3 includes elements 241-249 (including rotor pole 241; stator poles 242, 243, 244, and 245; conductive lead 246; and resistors 247, 248, and 249); Region 4 includes elements 251-259 (including rotor pole 251; stator poles 252, 253, 254, and 255; conductive lead 256; and resistors 257, 258, and 259). 1; stator electrodes 252, 253, 254, and 255; conductive lead 256; and resistors 257, 258, and 259); region 5 includes elements 261-269 (including rotor electrode 261; stator electrodes 262, 263, 264, and 265; conductive lead 266; and resistors 267, 268, and 269); and region 6 includes elements 271-279 (including rotor electrode 271; stator electrodes 272, 273, 274, and 275; conductive lead 276; and resistors 277, 278, and 279). These regions are arranged in pairs comprising commutation stages: the first commutation region (defined by 221-229 in FIG. 4 ) is closest to pole A and is connected to the second commutation region (defined by 231-239 in FIG. 4 ) via an insulated conductor 220; the first commutation region is then connected to junction B through a set of resistors that vary depending on how far the rotor 280 has rotated. A blocking semiconductor 292 is inserted between pole A and junction B; the blocking semiconductor has the function of limiting the maximum voltage during the period when the circuit is open.

The first commutating region, the second commutating region, and insulated conductor 220 form the first of three commutating stages in the commutating circuit breaker of Figure 4. The other two stages include components 240-259 plus blocking semiconductor 294 inserted between junctions C and D, and 260-279 plus blocking semiconductor 296 inserted between junctions E and F. A stage is defined as a closed circuit that conducts power to the commutating rotor and then out of the rotor; there are three stages in Figure 4.

The multi-stage rotary commutation circuit breaker of FIG4 is actuated by the rotation of a cylindrical commutation rotor 280. The circuit breaker of FIG4 has six commutation zones, which commutate power via a series of conventional resistors. Such resistors are less expensive per unit of energy dissipation capacity than blocking semiconductors. The device of FIG4 can be economically designed so that 90-95% of the inductive energy can be absorbed in conventional ohmic resistors (which means that the MOV can be smaller and less expensive). This reduces acquisition, maintenance, and operating costs.

In the device of FIG4 , the commutating rotor takes the form of a rotor that rotates counterclockwise approximately 18.2 degrees to open the circuit and then further rotates 7.9 degrees to the final open position, such that the total rotation during actuation of the commutating circuit breaker is 26.1 degrees ( 281 ). The rotor is made of a strong, electrically insulating material (e.g., a glass fiber reinforced polymer composite, an industrial grade thermoplastic composite, or a polymer-based composite foam), except for the rotor electrodes 221 , 231 , 241 , 251 , 261 and 271 and the insulated conductive paths (represented by dark black lines ( 220 , 240 and 260 )) connecting pairs of rotor electrodes (e.g., 221 and 231) within the rotor. The shaft is desirably metallic but electrically insulated from the conductors 220 , 240 and 260. The entire rotating portion is surrounded by a stator 290 in which the stator electrodes are mounted. The resistors and blocking semiconductors are preferably located outside the stator to facilitate removal of heat after the circuit breaker trips.

The perspective view in FIG4 is an end view of a commutated rotor having a cylindrical shape. The length of the cylinder (perpendicular to the cross-section shown in FIG4 ) can be adjusted to keep the nominal full-load ampere value per square centimeter of electrode contact area within design limits; thus, depending on the current, cylinder 280 can be disk-shaped or barrel-shaped. The circumferential insulation distance between stator electrodes (e.g., 222 , 223 , 224 , 225 ) can be adjusted to handle the voltage gradient at each commutation stage; in principle, both the width of each stator electrode and the distance between each pair of adjacent stator electrodes can be adjusted to achieve an optimal design. It should be noted that by limiting the maximum voltage applied to each stage, blocking semiconductors also prevent arcing between adjacent stator electrodes. The distance between stator electrodes, the width of the stator electrodes, or the composition of the different stator electrodes need not be the same for any two stator electrodes. Furthermore, multiple series-connected commutating circuit breakers (such as the commutating circuit breaker in FIG4 ) can be mounted on a single shaft to produce more commutation stages (6, 9, etc.). In this case, each of the switch contacts 221, 231, 241, 251, 261 and 271, and their mating contacts 222 etc. span only a small portion of the length of the drive shaft separated by the introduction of the insulating section and/or torque drive section.

In the specific design of FIG. 4 , the stator electrodes 222 , 232 , 242 , 252 , 262 , and 272 in the conductive state are desirably liquid metal electrodes; these are the most suitable stator electrodes for carrying high currents in the conductive state. Based on contact resistance, liquid metal electrodes are approximately 10 ⁴ times more conductive than sliding solid metal electrodes. Liquid metal electrodes can therefore also be narrower than sliding solid contact electrodes, a major advantage for the initial commutation stages of a commutating circuit breaker. Liquid metal stator electrodes 222 , 232 , 242 , 252 , 262 , and 272 can be, for example, one-tenth the width of solid stator electrodes 223 , 224 , and 225 and still have one-thousandth the contact resistance of the solid stator electrodes. As a specific example, consider the case where the commutated rotor of FIG. 4 is a 31.5 cm diameter barrel-shaped commutated rotor designed for 30 kV DC or AC power. Making one of the liquid metal stator electrodes 222, 232, 242, 252, 262, and 272 one millimeter (mm) wide in the circumferential direction means that if the first stator electrode is aligned with the rotor electrode so that only one millimeter of movement is required to cause the first commutation (for example), the first commutation can be achieved by rotating the rotor 280 by only 0.36 degrees. This first commutation is very important in any circuit breaker where controlling the maximum fault current is critical because the fault current is controlled once the first resistor is inserted. Using narrow liquid metal electrodes is one way to speed up the first commutation by reducing the distance that the commutating rotor must move to achieve the first commutation.

The six commutation regions and three blocking semiconductors 292, 294, 296 of FIG4 give this design a high level of cutoff redundancy and reliability. If, as part of the design, the failure of one of the three blocking semiconductors is survivable, then only two series-connected blocking semiconductors must be able to block the current during a fault. Let's consider (as we have already considered above) the case where the blocking semiconductors are MOVs with a breakdown voltage 20% higher than the normal line voltage and the ability to control the voltage during switching of the switch within a voltage control range of 20% to 50% above the normal line voltage. This means that three MOVs (each effective between 0.60 and 0.75 times the normal system voltage) can protect the switch from exceeding 2.25 times the normal voltage during switching, while still allowing for the failure of one MOV. As the configuration exceeds three stages, the voltage across the MOVs continues to decrease as the voltage is distributed among more devices. In an extreme example, all eighteen resistors shown in FIG4 could be MOVs having a breakdown voltage of 0.0666 times (1.2/18) the normal voltage and a maximum voltage of 0.0833 times (1.5/18) the normal voltage under normal operation. It is also true that each of the three blocking semiconductors 292, 294, and 296 shown in FIG4 could be a series-connected MOV stack within itself, which is inherently more fault-tolerant than the case of three individual MOVs.

The trailing edges of the conductive electrodes of FIG4 are desirably graded in composition and resistivity to reduce the chance of arcing when the electrodes separate. Optimally, the outermost surfaces of the rotor electrodes are made of a highly conductive metal or composite material that is wear-resistant and does not oxidize, re-crystallize, or interdiffuse with the facing, conductive stator electrode during use. Oxidation can be prevented by excluding oxygen or using oxidation-resistant metals such as gold, platinum, or molybdenum. In the case of excluding oxygen, composite materials of soft metal matrices/particulate hard particles with good electrical conductivity are suitable, such as porous structures impregnated with silver or copper based on sintered metals; for example, chromium powder as described in U.S. Patent 7,662,208, or tungsten powder as used in commercial electrodes available from Mitsubishi Materials C.M.I. Co. Ltd. Aluminum/silicon carbide electrodes are also suitable in an oxygen-free environment. In the case of not excluding oxygen, molybdenum is a favorable contact surface material for the metal electrodes; molybdenum plasma-sprayed onto aluminum/silicon carbide electrodes is particularly advantageous.

To achieve the target on-state losses of 1.0 kW at 2000 amperes under closed-circuit conditions, the total resistance of the path from pole A to pole F in Figure 4 is at most 2.5E-4 ohms. Such low resistance is only possible with liquid metal on-state electrode connections, or solid metal electrodes with large contact areas. Achieving lower resistance via large contact areas requires using larger rotors, which require more torque to accelerate; there is an optimal design based on the on-state resistance target, which varies slightly for each specific case; in some cases, heat releases above one kW will be better managed by incorporating fans or liquid cooling, which makes it easier to make a working switch without relying on liquid metal electrodes for the on-state electrode connections.

The spring or other actuator used to induce the counterclockwise radial acceleration of FIG4 can accelerate the rotor during the entire time of commutation, or alternatively, a very stiff spring can apply an initial acceleration that accelerates only a small portion of the 18.2 degrees of radial motion of the commutating rotor during commutation. In this arrangement, the commutating rotor is free to rotate during most of the time that the switching rotor is moving and commutation is induced.

During the circuit opening period of FIG. 4 , the 18-ohm resistor is introduced; the resulting voltage, current, and induced energy vary depending on the initial current at the first commutation and the inductance and resistance of the fault system. If the precise initial conditions are known, a significant amount of the induced energy can be effectively absorbed by the ohmic resistors of FIG. 4 alone, while keeping the voltage within design limits. This is illustrated in the discussion of FIG. 6 in PCT/US2012/058240. However, more practically, the initial current at the first commutation and the inductance of the fault system are unknown. In this case, the current may be too high or too low compared to the assumed value used to calculate the target resistance level. If the current is too high, blocking semiconductors 292, 294, and 296 will be activated, limiting the maximum transient voltage generated by the switching action. If the actual current at the first commutation does not exceed 100 amperes, but the current at the first commutation is assumed to be 10 kiloamperes (kA), then the order of introducing the ohmic resistors of FIG. 4 is likely meaningless, except for the introduction of the smaller resistor at the end. In this case, the inductive stored energy is practically insignificant, and building a switch capable of handling it is uneconomical. In this low-current switching scenario, a simple commutation switch, as shown in FIG2 , minus the parallel blocking semiconductor 160, is a better choice. The design of FIG4 is particularly important for highly inductive HVDC faults, where there may be stored inductive energy ranging from several kilojoules to several hundred megajoules (MJ) to dissipate; using conventional ohmic resistors to absorb most of this energy, the cost of the multiple resistors required to dissipate the short-circuit inductive energy can be significantly reduced compared to the cost of MOVs or transistors to absorb all of this energy.

A useful modification to the design of FIG4 is to design only six commutations on conventional resistors to quench any remaining inductive energy before commutating to the blocking semiconductors. This allows the use of economical conventional ohmic resistors to absorb approximately 95% of the inductive energy, but simplifies the design by having only two stator electrodes per rotor electrode (this means that features 224, 225, 228, 229, 234, 235, 238, 239, 244, 245, 248, 249, 254, 255, 258, 259, 254, 255, 258, 259, 264, 265, 268, 269, 274, 275, 278, 279 are excluded from the design). As the trailing edge of the rotor electrode moves away from the trailing edge of a particular stator electrode, the first six commutations can be precisely timed by adjusting the exact rotation angle to where each of the first six separations of the stator and rotor electrodes occurs. Although Figure 4 shows the rotor electrodes located on the outer radius of the commutated rotor, it is equally possible to place the rotor electrodes on the flat ends of a cylindrical rotor. Both designs have advantages and disadvantages. The design of Figure 4 is similar to a drum brake, where the brake pads play a role similar to the stator electrodes, and the drum resembles the commutated rotor. The alternative design of having the rotor electrodes at the ends of the commutated rotor is similar to a disc brake.

FIG5 illustrates a two-stage linear motion reversing switch of the present disclosure, having a reversing shuttle 358 that moves a distance 405 to open the circuit. The reversing shuttle includes two shuttle electrode pairs, 410, 411, and 412 (shuttle electrode pair #1) and 415, 416, and 417 (shuttle electrode pair #2), embedded in a structural insulator 359. There are four reversing regions 361 through 364: 361 and 362 together form the first stage 357; 363 and 364 together form the second stage 419 of the two-stage reversing circuit breaker. Each stage has a blocking semiconductor: stage 357 has a parallel path through blocking semiconductor 420, and stage 419 has a parallel path through blocking semiconductor 421. Within each of these regions are four stator electrodes; for example, reversing region 361 includes stator electrodes 366, 368, 370, and 372; stator electrode 366 is connected to pole A via a low-resistance conductor 374. Stator electrode 368 is connected to pole A via resistor 376; stator electrode 370 is connected to pole A via resistors 378 and 376 connected in series; stator electrode 372 is connected to pole A via resistors 380, 378, and 376 connected in series; and similarly for the other commutation regions. Commutation region 362 includes stator electrodes 381, 383, 385, and 387. Stator electrode 381 is connected to stator electrode 389 via low-resistance conductor 382. Stator electrode 383 is connected to low-resistance conductor 382 via resistor 384; stator electrode 385 is connected to low-resistance conductor 382 via resistors 386 and 384 connected in series; and stator electrode 387 is connected to low-resistance conductor 382 via resistors 388, 386, and 384 connected in series. Commutation region 363 includes stator electrodes 389, 390, 392, and 394. Stator electrode 389 is connected to stator electrode 381 via low-resistance conductor 382; stator electrode 390 is connected to low-resistance conductor 382 via resistor 391; stator electrode 392 is connected to low-resistance conductor 382 via resistors 391 and 393 connected in series; and stator electrode 394 is connected to low-resistance conductor 382 via resistors 395, 393, and 391 connected in series. Commutation region 364 includes stator electrodes 396, 398, 400, and 402. Stator electrode 396 is connected to pole B via low-resistance conductor 397. Stator electrode 398 is connected to pole B via resistor 399; stator electrode 400 is connected to pole B via resistors 401 and 399 connected in series; and stator electrode 402 is connected to pole B via resistors 403, 401, and 399 connected in series.

When the circuit is closed, there is a low resistance path from pole A through the reversing circuit breaker to pole B in such a way that pole A is connected to stator electrode 366 through conductor 374 and to shuttle electrode 411, which is then connected through insulated conductor 410 to shuttle electrode 412, which is then connected to stator electrode 381 and from there through conductor 382. There is also a parallel connection from pole A through blocking semiconductor 420 to 382; this connection through 420 limits the maximum voltage across stage 357. Conductor 382 carries current to stator electrode 389, then to shuttle electrode 216, then through insulated conductor 415 to shuttle electrode 417, then to stator electrode 396, and then through conductor 397 to pole B. There is also a parallel connection from 382 through blocking semiconductor 421 to pole B; this connection through 421 limits the maximum voltage across stage 419.

In this case, the multi-stage linear motion switch is essentially a rigid body that maintains the set geometric relationship between the four shuttle electrodes 411, 412, 416, and 417 as it moves to the right to open the circuit. After the insertion of the twelve resistors implied in FIG5, the current is low enough that the shuttle electrodes can move past their final connected position through the resistors with a significantly reduced current, which is then cut off by the blocking semiconductors 420 and 421. The presence of the two blocking semiconductors 420 and 421 broadens the range of initial conditions to which the switch can adequately respond, while they will also control the voltage during the forward period through the insertion of various ohmic resistors.

A multi-stage long chain reversing circuit breaker as shown in FIG5 can be used to interrupt arbitrarily high voltages. In order to efficiently move a long reversing shuttle such as that described herein, it is desirable to use multiple actuators along the length of the reversing shuttle, such as multiple springs or cylinders positioned to accelerate the shuttle between reversing zones, or multiple linear motors acting between reversing zones. Long multi-stage circuit breakers with embedded permanent magnets can also be driven by, for example, known electromagnetic devices (however, a greater force can be applied using springs or electromagnets than by coupling to permanent magnets). Combinations of drive mechanisms can also be used to achieve greater accelerations than can be generated by a single mechanism. In such a multi-stage linear circuit breaker, various triggers and releases can be configured, as discussed in more detail in PCT/US2012/058240.

Compared with the linear motion commutation circuit breaker shown in for example Figure 5, it is easier to immerse the cylindrical commutation rotor of Figure 1, Figure 2 or Figure 4 in the arc extinguishing fluid of pressurization, because the rotation of the cyclically symmetrical cylinder does not produce form resistance, yet, linear motion in fluid necessarily involves form resistance, and described form resistance can significantly suppress the rapid motion of commutation shuttle in liquid, or cause cavitation, i.e., form the position that is easy to induce arc. Cylindrical design can also make liquid immersion system have very low liquid volume compared with linear actuation design. Spark can be well suppressed by the fluid around the separated electrode, especially in the case of the fluid of high pressure maintenance. Dielectric fluid is limited to only a few cubic centimeters, which is feasible in cylindrical commutation circuit breaker (such as Figure 1, Figure 2 or Figure 4), but is impossible in the linear motion switch of Figure 5. This means that fluid of high dielectric strength (such as perfluorocarbon fluid) can be economically used in rotary switch of the present disclosure. The main advantage of using high-pressure liquid dielectrics in reversing circuit breakers is that the standoff distance between adjacent stator poles can be reduced if the gaps between the solid dielectrics are filled with a high-pressure fluid of high dielectric strength. This will allow for more compact reversing circuit breakers. Operating the switchgear at high liquid pressures has not been commercially implemented in the prior art. The unique shape of the rotary reversing switch of Figures 1, 2 or 4 allows for very small volumes of high-pressure liquid, which is not dangerous in terms of stored energy. Water pressures from ten times atmospheric pressure (1.0 megapascals, MPa) to 200 times atmospheric pressure (20 MPa) are desirable and feasible.

MOVs can be conveniently formed as stacked varistor films on a thin metal layer. The next example shows how a stack of such varistor films can be incorporated into the stator of a reversing switch of the present disclosure. FIG6 shows a linear motion switch of the present invention, but such an MOV/sheet stack can also be incorporated into the stator of a rotary switch. In FIG6 , a stack of hollow, disc-shaped MOV layer assemblies 460 forms the variable resistor portion of the switch. A separate MOV layer assembly similar to 461 has metal washers 451 , 452 on either side and is bonded to a hollow, disc-shaped MOV core 450. An actual disc-shaped MOV layer assembly (e.g., 450), if it includes only one MOV layer, would be thinner than that shown in FIG6 ; however, the MOV disc (e.g., 450) itself typically includes multiple (printed and then fired) MOV layers on a metal sheet or printed conductive layer. In this process, a single layer of MOV ceramic is coated on a thin sheet and then passed through a ceramic manufacturing process, with the final MOV layer typically being 25-50 microns thick. The breakdown voltage of each single layer depends on its composition and is typically 3-3.5 volts, which results in an average voltage gradient of the order of 300 volts/mm along the edge of the MOV/thin sheet stack at breakdown. As shown in Figure 6, combining multiple MOV layers into a MOV disk can result in a voltage change of several hundred volts per MOV disk. Alternatively, a special assembly consisting of individual MOV layers of 3.5 volts can be used to form a pyramid MOV similar to Figure 6, but with very thin layers. This may have the following disadvantages: if a failure occurs, the entire MOV assembly has to be replaced, rather than replacing the disk component that failed, such as 461.

Each pair of adjacent disc-shaped MOV layer assemblies (e.g., 461) is bonded together by some suitable method (e.g., conductive adhesive, soldering, or brazing). The metal washers 451, 452 are very simple examples of stator electrodes, preferably with a hole 456 through the metal washer that is slightly smaller than the hole 455 through the MOV layer itself (e.g., 450), so that the metal washer protrudes slightly further into the intermediate cavity than the inner diameter of the MOV element; in this case, it is preferred to place a polymer with good electrical resistance, a resistivity of approximately 10 to 10^5 ohm-meters, low friction, and good tracking resistance between the inner edges of the metal disks (e.g., 451, 452). This protects the inner surface of the actual MOV (e.g., 450) from damage via direct contact with the moving shuttle electrode 465, which in this case is a simple metal rod or tube that extends completely through the stacked MOV layer assembly 460. At the bottom end of the shuttle electrode 465 is an optional end portion 466 of the commutation shuttle 465, which acts as an electrical stress control device similar to the function of the trailing edge resistors 155, 164, 157 and 168 of Figure 2 to prevent the occurrence of arcing, but the optional end portion may also have an additional function as described below by providing a clamping surface to hinder the shuttle electrode 465 in the conductive state (as shown in Figure 6).

In the on-state, electrical connection to pole A is made through low-resistance stator electrode 490, which can be a highly conductive metal or liquid metal electrode that mates with the end of shuttle electrode 465. A parallel path exists from pole A through electrical contact 485 to the bottom of the MOV stack, and then from the top edge of the MOV stack through connector 486 and conductor 487 to pole B. When shuttle electrode 465 is withdrawn during switch operation, the connection through the increased length of MOV stack 460 becomes the only electrical connection between the poles. This parallel path through MOV stack 460 remains connected before and after operation of the switch shown in FIG6 . (Alternatively, if 486 and 487 were not present, the MOV connection to pole B would be severed once shuttle electrode 465 is withdrawn from the MOV stack.) The connection from pole B to commutating shuttle 465 is made through electrical slip ring 470, although other devices may also be used. At the upper end of commutation shuttle 475 is a feature for connecting to force 480, which pulls the commutation shuttle out of the disc-shaped MOV layer assembly stack 460 to open the circuit. FIG6 shows that the first disc-shaped MOV layer assembly has the largest outer diameter, as it is the first disc-shaped MOV layer assembly connected to the circuit and typically experiences the highest current for the longest duration. The cross-sectional area of a disc-shaped MOV layer assembly generally varies in proportion to the amount of energy dissipated by a single disc-shaped MOV layer assembly and, therefore, decreases with the height of the stack 460.

It is desirable that the lowest disc-shaped MOV layer assembly in Figure 6 (which is the first one connected to the circuit) should have the greatest mass and therefore the largest outer diameter. It is important that the metal disks (e.g., 451, 452) cover the entire face of the MOV layer assembly to which they are attached so that current can flow evenly throughout the entire volume of each disc-shaped MOV layer assembly.

The circuit breaker of FIG6 has several unique features. It uses the simplest possible shuttle, a metal rod or tube. The maximum force 480 that can be applied to the rod or tube depends on the strength of the material and the cross-sectional area of the rod or tube wall. If all the forces on the shuttle come from acceleration, then the maximum possible acceleration for any given material is strictly a function of the strength/density ratio of the material forming the shuttle and the length of the shuttle. If σ is the tensile yield strength of the material (Pascals), D is the density of the material (kg/ ^m3 ), and _L is the shuttle length (meters), then the maximum acceleration (meters/ ^second2 ) that can be applied to a shuttle such as 465 is given by:

(3)A _max =σ/LD

The results obtained from this equation for a 2-meter-long metal column pulled from one end, as shown in Figure 6, are shown in Table 1; the maximum possible acceleration ranges from less than 1000 m/ ^s² for sodium to 114,000 m/ ^s² for aluminum-based alumina fiber wire. Table 1 also shows the mass of various materials required at 20°C to make a 2-meter-long, 25-micro-ohm column of material; at this loss level, a nominal 2-meter-long shuttle would deliver 2000 amps of current while generating 100 watts of ^I²R waste heat. (Waste heat scales linearly with conductor mass; for example, one-tenth the mass of the conductor means ten times the heat is generated.) The mass of metal required to form a two-meter-long, 25-micro-ohm column of material ranges from 3.7 kg for sodium to 1118 kg for molybdenum.

The best overall solution for the shuttle 465 in FIG6 depends on the relative cost of the conductive materials and the mechanical structure (including the springs and triggers, as well as the structural supports that maintain 465 under stress or apply stress to it), and critically on the required acceleration. The structural cost is proportional to the mass of the conductor that must be accelerated times the acceleration. Acceleration determines the time to the critical first commutation, so there is a good reason to push forward with a large acceleration to minimize the time to the first commutation, if and where that is important (reaching the first commutation very quickly is more important when the system inductance in the fault is low than when the system inductance in the fault is high). Simply pulling the conductive tubing quickly so that the tubing reaches the engineering limit of the material's maximum tensile strength (see the "Maximum Acceleration" column in Table 1) is the theoretically fastest way to accelerate a linear motion shuttle.

Table 1: Data related to the conductors in the acceleration diagram 6

The fastest-actuating reversing circuit breaker of FIG. 6 using the materials in Table 1 would be based on the material with the highest strength-to-density ratio, namely, aluminum-based alumina-fiber wire. This cermet wire is the mechanical strength element in 3M ^™ Aluminum Conductor Composite Reinforced (3M ACCR) wire (replacing the steel in the more standard ASCR aluminum-steel core reinforced wire), available from 3M. Using only the materials listed in Table 1, the desired combination of reasonably low overall mass for acceleration and fast actuation can also be achieved by manufacturing the reversing shuttle 465 from a high-strength titanium alloy shell with sodium inside. Among potential single-component material solutions for the reversing shuttle 465, pure aluminum and pure magnesium have essentially equal mass to meet the 25 microohm resistance target, but pure aluminum is stronger, making it a better solution for the reversing shuttle 465. High-strength aluminum's electrical conductivity is relatively less compromised than with pure metals, making high-strength aluminum alloy 6061T-6 a good choice for the reversing shuttle 465. The second to last column of Table 1 is a dimensionless quality factor M that ranks the candidate materials for the reversing shuttle 465 .

M = {(strength)/[density×resistivity]}/{(strength) of annealed copper/[density×resistivity]}

The quality factor, M, is indicated relative to the reference value of 1.00 for annealed copper; higher values of M are more desirable. Of the single-component or alloy materials (not composites or fabricated structures) shown in Table 1, cold-worked copper has a moderately improved quality factor, M (1.257), compared to copper. All forms of magnesium and aluminum examined also have slightly higher M values than annealed copper, with high-strength aluminum alloy 6061-T6 having M values ranging from 1.147 to 4.411. The material with the highest quality factor, M (6.424) in Table 1, is cermet wire, which consists of alumina glass fibers in a pure aluminum matrix. Such ceramic wire can be used as both the actuator and conductor for the motion of the commutation shuttle 465 in FIG. 6 .

Because the modulus of the ceramic wire (the core wire of the 3M ACCR) is so high (4550 MPa), stretching it by only a few percent can store a large amount of elastic energy (comparable to a very stiff spring) that can provide force 480, while eliminating the need for slip rings 470. This design can be used for very fast actuation designs capable of very high voltages. In very extreme cases, the ceramic ACCR wire can be stressed to near its breaking strength (1400 MPa), strung through an MOV stack as shown in Figure 6, and then released under the stack of MOV layer assemblies to break the circuit. This design, in which a high-strength fiber-reinforced wire 465 extends through the MOV layer assembly 460 stack and is constrained under the stack, provides a feature 466 that is securely attached to the tensioned wire 465, enabling the fastest known actuation of a linear motion reversing circuit breaker. There are several known options for quickly releasing this high-stress fiber-reinforced wire form 465:

1. Feature 466 can be a rigid, strong rod held in place by a ring of piezoelectric actuators that hold the wire end 466 in place via a normal force that can be released within 20 microseconds (the required normal force can be reduced if part of the constraint on the motion of 466 can be due to related magnetic domains on the surface of 466 that mate with similar magnetic domains printed on the surface of sleeve 490);

2. Line 465 or line end 466 may be cut using high explosives;

3. The breakage of the wire itself or the wire end 466 can be initiated with a pulsed laser.

This type of circuit breaker can be reset without replacing the components used only for option 1. The last two methods are also useful as a form of fast-acting fuse for HVDC circuits that rarely blows; they can also be reset, but a part (the fuse) needs to be replaced each time. If a piezoelectric clamp is used to hold the bottom end of the reversing shuttle 465, the reversing circuit breaker of Figure 6 can be reset by the abutting rod-shaped clamping surface provided by feature 466 in Figure 6.

The design of FIG6 minimizes the mass of non-essential portions of the commutation shuttle by eliminating the bulk of the insulator attached to the commutation shuttle and minimizing the mass of the trailing edge electric field control techniques described elsewhere in this disclosure. For the circuit breaker of FIG6 , only the conductors are absolutely required; while the optional graded resistivity trailing edge component 466 is intended to reduce arcing within the core of the MOV layer assembly stack during operation, it is not a requirement but a desirable feature. This design can also be configured with a high vacuum surrounding the commutation shuttle 465 and MOV layer assembly stack 460, or an arc quenching gas mixture containing sulfur hexafluoride.

All mechanical electrical switches face similar limitations on their maximum speed of operation. There is always a maximum speed at which the electrodes can move, and thus the circuit can be opened, based on how long it takes for the electrodes to move far enough apart to quench an arc or other current between them and prevent restriking. The switch of the present invention speeds up the operation of mechanical switches because the separated electrodes are insulated by a high-dielectric-strength solid and liquid that maximizes electrical resistance during disconnection of the electrodes under high contact or hydraulic pressure conditions. In some forms of the present design, there is no gas in the area between the separated electrodes, which reduces the electrode spacing required to prevent restriking. This is true even in the case of mechanical switches that open without power flow and close again shortly thereafter, as is common in various hybrid circuit breaker designs. This allows for faster operation by the mechanical switch portion of various hybrid switch designs because the fast mechanical switch of the present disclosure does not need to move as far as prior art switches to prevent restriking.

The concepts disclosed herein can be implemented in either rotary or linear motion designs, preferably using graded resistivity on the trailing edge of the electrode for commutating power with little or no arcing. PCT/US2012/058240 mentions positioning and restraining a tensioned shuttle electrode in the on-state (where the shuttle is under stress in the on-state) via piezoelectric actuators and associated magnetic domains, a feature also applicable to the switch disclosed herein. A further improvement added to the present disclosure is the use of friction, modulated by the normal force, to restrain the motion of the commutating shuttle in the on-state.

In the reversing switch of the present disclosure, arc suppression relies on three features:

1. Gradient resistivity in the separation electrode;

2. Control of transient voltages by blocking semiconductors during commutation;

3. Tight gap between the electrode and the surrounding dielectric;

4. High-pressure liquid dielectric surrounding the separation electrodes.

One aspect described in PCT/US2012/058240 provides a switch primarily insulated by a solid dielectric that fits tightly around the electrodes, minimizing the size of any fluid-filled gap that may form between the electrodes during separation; this increases the ability to withstand a given voltage between the electrodes. While it is noted that this implies a normal force between the shuttle and stator, this normal force is also useful for limiting the force applied to the stator to cause its movement. Such a switch, by its very nature, implies the presence of a significant frictional interaction between the movable shuttle or rotor electrode and a dielectric solid bonded to the movable electrode with the surrounding stator, including those portions of the stator electrode and the insulating solid dielectric portion of the stator, which locally constrains the movement of the movable electrode before and during switching. This frictional interaction can be advantageously used in a rapid method of triggering an electronic switch, as the sliding flutter characteristic of the frictional interaction can be used to locally constrain the movement of the shuttle electrode before triggering. Specifically, static friction is typically greater than kinetic or sliding friction, allowing a friction-locked stator, where the critical force to initiate motion F(CR) is greater than the actual applied force F(AP), to maintain its position securely for extended periods. Once motion is initiated, however, motion is maintained at the same applied force. This allows for an additional "kick" to trigger the shuttle's movement by providing a trigger force F(TR) to initiate the shuttle's movement. Once the shuttle is in motion, it will continue to move until the applied force F(AP) drops below the critical dynamic force F(DYN). Given that the critical force to initiate motion F(CR) is proportional to the normal (perpendicular) force between the shuttle and stator, it is practical to adjust F(CR) by adjusting the pressure around the flexible stator, which in turn adjusts the normal force between the shuttle and stator.

A hydraulic cylinder or a fluid-filled fiber-reinforced elastomeric bag (similar to a high-pressure hydraulic hose) may desirably be used to apply a normal force at the interface between the shuttle and stator in the switch of the present disclosure. This can be visualized by referring to FIG1 , which shows a large gap between the outside of the modular stator assembly (composed of keystone-shaped segments 105, 107, 109, 125, 127, 129, 142, 144, and many copies of 140) and the inside of the pressure vessel 141. In FIG1 , this area is filled with a dielectric fluid under pressure, but it is readily envisioned that additional features could be added to this area, such as an expanded fabric-reinforced elastomeric bag that pushes the modular stator assembly against the commutating rotor 445 to provide a friction-restricting force that helps maintain the rotor in the on state.

The modular stator assembly can also be held tightly against the commutating rotor by a stretched elastomeric sleeve that surrounds the outer circumference of the modular stator assembly. Although features of the claimed invention are shown in some drawings and not in others, this is not intended to limit the scope of the invention. Other embodiments will occur to those skilled in the art and are within the scope of the claims.

Claims (26)

1. A reversing switch, comprising: A fixed part with fixed electrodes; A movable part with a movable electrode; The switch is closed when the fixed electrode and the movable electrode are in conductive contact. The movable part can move relative to the fixed part to disconnect the conductive contact between the fixed electrode and the movable electrode, thereby defining the switch off position; An electrical path into which current is diverted when the switch is open, and A nonlinear, non-ohmic blocking semiconductor in the electrical path, wherein the nonlinear, non-ohmic blocking semiconductor absorbs most of the stored induced energy. 2. The reversing switch as claimed in claim 1, wherein the movable part includes a shuttle. 3. The reversing switch as claimed in claim 1, wherein the movable part includes a rotor. 4. The reversing switch of claim 3, wherein the fixed electrode and the movable electrode are contained in a dielectric fluid under a hydraulic pressure of at least one MPa. 5. The reversing switch as claimed in claim 4, wherein the hydraulic pressure is greater than 10 MPa. 6. The reversing switch of claim 1, wherein the fixed portion comprises two fixed, spaced electrodes, and wherein the disconnected electrical paths are connected through the two fixed, spaced electrodes. 7. The reversing switch of claim 1, wherein the fixed electrode comprises a plurality of adjacent separate conductors. 8. The reversing switch of claim 7, wherein when the switch is open, the movable electrode makes electrical contact with one of the separate conductors each time. 9. The reversing switch of claim 7, wherein when the switch is open, the movable electrode simultaneously makes electrical contact with at least two of the separated conductors. 10. The commutation switch of claim 1, comprising a plurality of nonlinear, non-ohmic blocking semiconductors located in an electrical path into which current is commutated when the switch is open. 11. The commutation switch of claim 10, wherein the plurality of nonlinear, nonohmic blocking semiconductors are arranged in a stack. 12. The commutation switch of claim 11, wherein the plurality of nonlinear, nonohmic blocking semiconductors are metal oxide rheostats (MOVs) arranged in a stack such that movement of the commutation electrode causes current to flow through an increasing number of MOVs, thereby causing a gradual increase in voltage across the stack. 13. The commutator of claim 12, wherein the metal oxide rheostat is arranged such that the edge of the sheet supporting the metal oxide rheostat extends to the region where the edge of the sheet directly contacts the moving shuttle electrode, such that the voltage change between adjacent sheets does not exceed four volts under normal operating conditions. 14. The reversing switch of claim 1, wherein the fixed portion comprises a stator and the movable portion comprises a rotor. 15. The commutator of claim 14, wherein the rotor is partially held in place by means of friction caused by a tightly fitted stator in contact with the rotor over a substantial portion of the rotor's surface area. 16. The commutator of claim 14, wherein the stator surrounds the rotor, and the stator includes interchangeable dome-shaped members. 17. The reversing switch of claim 16, wherein the dome-shaped member is held against the rotor by an elastic force or external hydraulic pressure applied to an impermeable membrane surrounding the dome-shaped member. 18. The commutation switch of claim 14, wherein the stator comprises a plurality of commutation stages, each commutation stage comprises two commutation regions, each commutation region comprises a conductive lead, a plurality of stator electrodes and a resistor, each stator electrode is electrically connected to the conductive lead, the resistor is located between each stator electrode and the conductive lead, wherein the two conductive leads of the two commutation regions of each commutation stage are electrically connected by a blocking semiconductor. 19. The commutation switch of claim 18, wherein at least some of the stator electrodes comprise liquid metal. 20. The reversing switch of claim 1, wherein the fixed electrode and the movable electrode are slidably separated. 21. The reversing switch of claim 20, wherein one or both of the fixed electrode and the movable electrode have graded, resistivity-increasing regions forming the final portions of the one or both of the fixed electrode and the movable electrode, the final portion of the one of the fixed electrode and the movable electrode being electrically connected to the corresponding other of the fixed electrode and the movable electrode when the switch moves from the closed position to the open position. 22. The reversing switch of claim 1, comprising at least two blocking semiconductors in a series electrical path. 23. The commutator of claim 1, wherein the fixed portion comprises a series of stacked metal oxide rheostats. 24. The commutator of claim 23, wherein the rheostat is annular and the rheostat has a different outer diameter. 25. The reversing switch as claimed in claim 1, wherein the movable part of the switch is under stress in the closed position. 26. The commutation switch of claim 1, wherein the blocking semiconductor is selected from the semiconductor group consisting of a rheostat, a Zener diode and a transient voltage suppressor diode.