Shorting switch and system to eliminate arcing faults in power distribution equipment

申请号 US10172651 申请日 2002-06-14 公开(公告)号 US06657150B1 公开(公告)日 2003-12-02
申请人 John J. Shea; Jeffrey A. Miller; 发明人 John J. Shea; Jeffrey A. Miller;
摘要 A shorting switch includes a vacuum switch having fixed and movable contact assemblies and a driven member. A spring cover tube and bushing mount the driven member for linear movement along a path substantially parallel to a longitudinal axis of the movable contact assembly. The driven member moves the movable contact assembly between open and closed circuit positions. A compression spring has a compressed state and a released state, which moves the driven member and movable contact assembly to the closed circuit position. A release bolt has an opening therein and is coupled to the driven member to normally maintain the compression spring in the compressed state. A charge is disposed in the opening of the release bolt and is actuated to fracture the release bolt and release the compression spring to the released state. First and second terminals are respectively electrically interconnected with the fixed and movable contact assemblies.
权利要求

What is claimed is:1. A shorting switch for eliminating arcing faults in power distribution equipment, said shorting switch comprising:a vacuum switch comprising a vacuum envelope containing a fixed contact assembly and a movable contact assembly movable along a longitudinal axis between a closed circuit position in electrical contact with the fixed contact assembly and an open circuit position spaced apart from the fixed contact assembly;a driven member;a mount mounting the driven member for linear movement along a path substantially parallel to the longitudinal axis of the movable contact assembly, said driven member coupled to the movable contact assembly to move the movable contact assembly between the open circuit position and the closed circuit position with the linear movement of the driven member;a spring member having a compressed state and a released state, which moves said driven member and the movable contact assembly to the closed circuit position;a release member having an opening therein, said release member coupled to said driven member and normally maintaining said spring member in the compressed state;a charge disposed in the opening of said release member, said charge being actuated to fracture said release member and release said spring member to the released state; andfirst and second terminals respectively electrically interconnected with said fixed contact assembly and said movable contact assembly.2. The shorting switch of claim 1 wherein said vacuum envelope is a sealed vacuum chamber; wherein said fixed contact assembly comprises a fixed contact sealed within the sealed vacuum chamber and an electrical conductor connected to the fixed contact at one end thereof, said electrical conductor sealably penetrating the sealed vacuum chamber and terminating at said first terminal at the other end of said fixed contact; wherein said movable contact assembly comprises a moveable contact sealed within the sealed vacuum chamber and moveable between a first position, out of electrical communication with the fixed contact, and a second position, in electrical communication with the fixed contact, said movable contact assembly further comprises a movable electrical stem connected to the moveable contact at one end thereof, said movable electrical stem sealably penetrating the sealed vacuum chamber and terminating at said second terminal at the other end of said moveable contact; and wherein said driven member is coupled to the movable electrical stem.3. The shorting switch as recited in claim 1 wherein said spring member is a compression spring having a first end and a second end; wherein said release member is a release bolt having a first end and a second end; and wherein said mount comprises a bushing having a longitudinal opening and further comprises a longitudinal tube having a closed end and an open end, said longitudinal tube housing said compression spring, said release bolt and said charge, the first end of said compression spring engaging the closed end of said longitudinal tube, the first end of said release bolt coupled to the closed end of said longitudinal tube, the first end of said driven member coupled to the second end of said release bolt, said bushing resting in the open end of said longitudinal tube, said driven member resting in the longitudinal opening of said bushing, the second end of said compression spring biasing said driven member to move the movable contact assembly to the closed circuit position after actuation of said charge.4. The shorting switch as recited in claim 1 wherein said charge is an electrically activated, chemical charge.5. The shorting switch as recited in claim 1 wherein said first and second terminals are adapted for electrical connection to first and second power lines, respectively.6. The shorting switch as recited in claim 1 wherein said first terminal is adapted for electrical connection to a power line; and wherein said second terminal is adapted for electrical connection to a flexible conductor.7. The shorting switch as recited in claim 1 wherein said first and second terminals are adapted for electrical connection to a power line and a neutral or ground.8. The shorting switch as recited in claim 1 wherein said release member has a longitudinal axis; and wherein the opening of said release member is a longitudinal cavity along said longitudinal axis of said release member.9. The shorting switch as recited in claim 8 wherein said charge is activated to provide a shock wave to fracture said release member; and wherein said release member is a release bolt having a body and a breakline disposed thereon to locate and control fracture of said release bolt responsive to said shock wave.10. The shorting switch as recited in claim 9 wherein the breakline has a predetermined depth in the body of said release bolt; and wherein said spring member has a predetermined compression force, with said release bolt being structured to maintain at least said compression force until after said charge is activated.11. A shorting system for eliminating arcing faults in power distribution equipment, said shorting system comprising:a vacuum switch comprising a vacuum envelope containing a fixed contact assembly and a movable contact assembly movable along a longitudinal axis between a closed circuit position in electrical contact with the fixed contact assembly and an open circuit position spaced apart from the fixed contact assembly;a driven member;a mount mounting the driven member for linear movement along a path substantially parallel to the longitudinal axis of the movable contact assembly, said driven member coupled to the movable contact assembly to move the movable contact assembly between the open circuit position and the closed circuit position with the linear movement of the driven member;a spring member having a compressed state and a released state, which moves said driven member and the movable contact assembly to the closed circuit position;a release member having an opening therein, said release member coupled to said driven member and normally maintaining said spring member in the compressed state;a charge disposed in the opening of said release member, said charge being actuated to fracture said release member and release said spring member to the released state;first and second terminals respectively electrically interconnected with said fixed contact assembly and said movable contact assembly; andmeans for detecting an arcing fault and responsively activating the charge disposed in the opening of said release member,wherein said activated charge fractures said release member, which releases said spring member, which drives said driven member to move the movable contact assembly to the closed circuit position to eliminate said arcing fault.12. The shorting system as recited in claim 11 wherein said charge includes an electrical input; and wherein said means for detecting an arcing fault and responsively activating the charge comprises means for detecting the arcing fault and responsively outputting a trigger signal, and means for detecting the trigger signal and responsively outputting an activation signal to the electrical input of said charge.13. The shorting system as recited in claim 12 said charge is an electrically activated, chemical charge.14. The shorting system as recited in claim 11 wherein said vacuum switch is adapted for operation with medium voltage power distribution equipment.15. The shorting system as recited in claim 11 wherein said means for detecting an arcing fault and responsively activating the charge has an activation delay of about 550 &mgr;s after initiation of an arcing fault event.16. The shorting system as recited in claim 12 wherein said means for detecting the arcing fault and responsively outputting a trigger signal comprises at least one photovoltaic sensor adapted to detect the arcing fault and output an optical signal as said trigger signal.17. The shorting system as recited in claim 16 wherein said arcing fault includes arc light; and wherein said at least one photovoltaic sensor is a photovoltaic sensor, which is self-powered from the arc light, said photovoltaic sensor having an output with said trigger signal, which is responsive to said arc light.18. The shorting system as recited in claim 12 wherein said means for detecting the arcing fault and responsively outputting a trigger signal comprises a plurality of photovoltaic sensors adapted to detect the arcing fault and output a plurality of first optical signals on a plurality of outputs, an optical multiplexer having a plurality of inputs and an output, a plurality of fiber optic cables connected between the outputs of said photovoltaic sensors and the inputs of said optical multiplexer, the output of said optical multiplexer outputting a second optical signal as said trigger signal.19. The shorting system as recited in claim 12 wherein said means for detecting the trigger signal and responsively outputting an activation signal to the electrical input of said charge comprises means for generating a charge voltage, a spark gap adapted to electrically connect the charge voltage to the electrical input of said charge, said spark gap having a control input, and means for detecting the trigger signal and responsively outputting a control signal to the control input of said spark gap, said spark gap outputting the activation signal to the electrical input of said charge responsive to said control signal.20. The shorting system as recited in claim 19 wherein said means for detecting the trigger signal and responsively outputting a control signal comprises means for detecting the trigger signal, and means for outputting the control signal responsive to the detected trigger signal.21. The shorting system as recited in claim 20 wherein said means for detecting the trigger signal and responsively outputting a control signal further comprises means for tripping an upstream circuit breaker for said power distribution equipment responsive to the detected trigger signal.22. The shorting system as recited in claim 12 wherein said charge includes an electrical input; and wherein said means for detecting an arcing fault and responsively activating the charge comprises means for detecting the arcing fault and responsively outputting a plurality of trigger signals, and means for detecting the trigger signals and responsively outputting an activation signal to the electrical input of said charge.23. A shorting switch for eliminating arcing faults in power distribution equipment, said shorting switch comprising:a vacuum switch comprising a vacuum envelope containing a fixed contact assembly and a movable contact assembly movable along a longitudinal axis between a closed circuit position in electrical contact with the fixed contact assembly and an open circuit position spaced apart from the fixed contact assembly;a driven member having a longitudinal opening with a circumferential groove therein;a mount mounting the driven member for linear movement along a path substantially parallel to the longitudinal axis of the movable contact assembly, said driven member coupled to the movable contact assembly to move the movable contact assembly between the open circuit position and the closed circuit position with the linear movement of the driven member;a spring member having a compressed state and a released state, which moves said driven member and the movable contact assembly to the closed circuit position;a ball-lock member having a plurality of ball bearings and a push rod with a circumferential groove therein, said ball bearings engaging the circumferential groove of the longitudinal opening of the driven member to hold said spring member in the compressed state and to hold said movable contact assembly in the open circuit position;means for driving the push rod of said ball-lock member into the longitudinal opening of said driven member to cause said ball bearings to engage the circumferential groove of the push rod of said ball-lock member and to release said driven member, in order to move the movable contact assembly to the closed circuit position; andfirst and second terminals respectively electrically connected to said fixed contact assembly and said movable contact assembly.24. The shorting switch of claim 23 wherein said means for driving the push rod is a solenoid having a plunger, which drives the push rod of said ball-lock member into the longitudinal opening of said driven member when said solenoid is actuated.25. The shorting switch of claim 23 wherein said spring member is a compression spring having a first end and a second end; wherein said ball-lock member has a first end and a second end; wherein said driven member has a first end with said longitudinal opening and a second end; and wherein said mount comprises a longitudinal tube having a first end and an open second end, said longitudinal tube housing said compression spring and said ball-lock member, the first end of said compression spring engaging the first end of said longitudinal tube, the first end of said ball-lock member coupled to the first end of said longitudinal tube, the second end of said driven member coupled to said movable contact assembly, the first end of said driven member passing through the open second end of said longitudinal tube, the second end of said compression spring biasing said driven member to move the movable contact assembly to the closed circuit position.26. The shorting switch of claim 25 wherein the push rod of said ball-lock member passes through the first end of said longitudinal tube; and wherein said means for driving the push rod is a solenoid having a plunger, which drives the push rod of said ball-lock member into the longitudinal opening of said driven member when said solenoid is actuated.27. The shorting switch as recited in claim 23 wherein said first terminal is adapted for electrical connection to a power line; and wherein said second terminal is adapted for electrical connection to a flexible conductor.28. A shorting system for eliminating arcing faults in power distribution equipment, said shorting system comprising:a housing;a vacuum switch comprising a vacuum envelope containing a fixed contact assembly and a movable contact assembly movable along a longitudinal axis between a closed circuit position in electrical contact with the fixed contact assembly and an open circuit position spaced apart from the fixed contact assembly;a driven member;a mount mounting the driven member in said housing for linear movement along a path substantially parallel to the longitudinal axis of the movable contact assembly, said driven member coupled to the movable contact assembly to move the movable contact assembly between the open circuit position and the closed circuit position with the linear movement of the driven member;a spring member having a compressed state and a released state, which moves said driven member and the movable contact assembly to the closed circuit position;a latch member, which normally latches said driven member to hold said spring member in the compressed state and to hold said movable contact assembly in the open circuit position, said latch member releasing said driven member to move the movable contact assembly to the closed circuit position, said latch member having an end, which engages said driven member, and a pivot in said housing;means for detecting an arcing fault and responsively unlatching the end of said latch member to release said driven member to move the movable contact assembly to the closed circuit position; andfirst and second terminals respectively electrically interconnected with said fixed contact assembly and said movable contact assembly.29. The shorting system of claim 28 wherein the end of said latch member is a first end; wherein said latch member also has a second end; and wherein said means for detecting an arcing fault and responsively unlatching said latch member to release said driven member comprises a solenoid having a plunger, which moves the second end of said latch member to pivot said latch member about said pivot and to cause the first end of said latch member to release said driven member.30. The shorting system as recited in claim 28 wherein said first terminal is adapted for electrical connection to a power line; and wherein said second terminal is adapted for electrical connection to a flexible conductor.

说明书全文

CROSS-REFERENCE TO RELATED APPLICATIONS

This application is related to commonly assigned, concurrently filed:

U.S. patent application Ser. No. 10/172,208, filed Jun. 14, 2002, entitled “Shorting Switch And System To Eliminate Arcing Faults In Power Distribution Equipment;

U.S. patent application Ser. No. 10/172,826, filed Jun. 14, 2002, entitled “Shorting Switch And System To Eliminate Arcing Faults In Power Distribution Equipment”;

U.S. patent application Ser. No. 10/172,238, filed Jun. 14, 2002, entitled “Shorting Switch And System To Eliminate Arcing Faults In Power Distribution Equipment”;

U.S. patent application Ser. No. 10/172,622, filed Jun. 14, 2002, entitled “Bullet Assembly For A Vacuum Arc Interrupter”;

U.S. patent application Ser. No. 10/172,080, filed Jun. 14, 2002, entitled “Vacuum Arc Interrupter Having A Tapered Conducting Bullet Assembly”;

U.S. patent application Ser. No. 10/172,209, filed Jun. 14, 2002, entitled “Vacuum Arc Interrupter Actuated By A Gas Generated Driving Force”;

U.S. patent application Ser. No. 10/172,628, filed Jun. 14, 2002, entitled “Blade Tip For Puncturing Cupro-Nickel Seal Cup”; and

U.S. patent application Ser. No. 10/172,281, filed Jun. 14, 2002, entitled “Vacuum Arc Eliminator Having A Bullet Assembly Actuated By A Gas Generating Device”.

BACKGROUND OF THE INVENTION

1. Field of the Invention

The invention is directed to shorting switches and, in particular, to shorting switches for eliminating arcing faults in low voltage power distribution equipment. The invention is also directed to shorting systems for eliminating arcing faults in power distribution equipment.

2. Background Information

There is the potential for an arcing fault to occur across the power bus of a motor control center (MCC), another medium voltage (MV) enclosure (e.g., a MV circuit breaker panel) and other industrial enclosures containing MV power distribution components. This is especially true when maintenance is performed on or about live power circuits. Frequently, a worker inadvertently shorts out the power bus, thereby creating an arcing fault inside the enclosure. The resulting arc blast creates an extreme hazard and could cause injury or even death. This problem is exacerbated by the fact that the enclosure doors are typically open for maintenance.

A high-speed shorting switch is needed for medium voltage as an alternative to arc proofing switchgear enclosures. Presently, manufacturers are developing more robust enclosures, which contain and direct the hot gases and flames out the top of the enclosure upon the occurrence of an internal arcing fault (e.g., a short across the bus bar, breaker, cable phase-to-phase or phase-to-ground). These faults can occur from a wide variety of sources, such as, for example, animals that crawl into the enclosure, tools left behind from maintenance crews, insulation failure, earthquakes, and other mechanical damage.

Rather than trying to contain and direct the blast, a new idea has been developed by others for eliminating the arcing fault altogether. This is done by shorting out the high-voltage bus either phase-to-phase or phase-to-ground. Known shorting switches use closing and holding techniques which are very expensive to buy and to maintain.

It is known to employ a high-speed shorting switch, placed between the power bus and ground, or from phase-to-phase, in order to limit or prevent equipment damage and personnel injury due to arc blasts. Such switches, which are large and costly, are located on the main power bus to shut down the entire power bus system when a fault occurs even if the fault is only on the load side of a branch circuit.

It is also known to employ various types of crowbar switches for this purpose. The switches short the line voltage on the power bus, eliminating the arc and preventing damage. The resulting short on the power bus causes an upstream circuit breaker to clear the fault.

Examples of medium voltage devices include a stored energy mechanism with vacuum interrupter contacts, and a mechanism to crush a conductor magnetically.

An example of a low voltage device is a stored energy air bag actuator, which drives a conductive member having a pin and a flange, in order to short two contacts. The first contact is in the form of a receptor for capturing the pin of the driven conductive member. The second contact has an opening, which allows the pin to pass therethrough, but which captures the flange of the driven member.

There is room for improvement in shorting switches and systems that respond to arcing faults and switch fast enough in order to protect workers and equipment from arc blasts associated with power distribution equipment.

SUMMARY OF THE INVENTION

These needs and others are met by the present invention, which provides a shorting switch and system for eliminating arcing faults in power distribution equipment. The shorting switch includes a vacuum switch having fixed and movable contact assemblies, a driven member, and a mount mounting the driven member for linear movement along a path substantially parallel to a longitudinal axis of the movable contact assembly. The driven member is coupled to the movable contact assembly to move the movable contact assembly between open and closed circuit positions with the linear movement of the driven member. A spring member has a compressed state and a released state, which moves the driven member and the movable contact assembly to the closed circuit position. A release mechanism holds and releases the driven member and the spring member. First and second terminals are respectively electrically interconnected with the fixed contact assembly and the movable contact assembly.

As one aspect of the invention, a shorting switch for eliminating arcing faults in power distribution equipment comprises: a vacuum switch comprising a vacuum envelope containing a fixed contact assembly and a movable contact assembly movable along a longitudinal axis between a closed circuit position in electrical contact with the fixed contact assembly and an open circuit position spaced apart from the fixed contact assembly; a driven member; a mount mounting the driven member for linear movement along a path substantially parallel to the longitudinal axis of the movable contact assembly, the driven member coupled to the movable contact assembly to move the movable contact assembly between the open circuit position and the closed circuit position with the linear movement of the driven member; a spring member having a compressed state and a released state, which moves the driven member and the movable contact assembly to the closed circuit position; a release member having an opening therein, the release member coupled to the driven member and normally maintaining the spring member in the compressed state; a charge disposed in the opening of the release member, the charge being actuated to fracture the release member and release the spring member to the released state; and first and second terminals respectively electrically interconnected with the fixed contact assembly and the movable contact assembly.

The spring member may be a compression spring having a first end and a second end. The release member may be a release bolt having a first end and a second end. The mount may comprise a bushing having a longitudinal opening and a longitudinal tube having a closed end and an open end, the longitudinal tube housing the compression spring, the release bolt and the charge. The first end of the compression spring may engage the closed end of the longitudinal tube. The first end of the release bolt may be coupled to the closed end of the longitudinal tube. The first end of the driven member may be coupled to the second end of the release bolt. The bushing may rest in the open end of the longitudinal tube, the driven member may rest in the longitudinal opening of the bushing, and the second end of the compression spring may bias the driven member to move the movable contact assembly to the closed circuit position after actuation of the charge.

The charge may be an electrically activated, chemical charge. The charge may be activated to provide a shock wave to fracture the release member. The release member may be a release bolt having a body and a breakline disposed thereon to locate and control fracture of the release bolt responsive to the shock wave. The breakline may have a predetermined depth in the body of the release bolt. The spring member may have a predetermined compression force, with the release bolt being structured to maintain at least the compression force until after the charge is activated.

As another aspect of the invention, a shorting system for eliminating arcing faults in power distribution equipment comprises: a vacuum switch comprising a vacuum envelope containing a fixed contact assembly and a movable contact assembly movable along a longitudinal axis between a closed circuit position in electrical contact with the fixed contact assembly and an open circuit position spaced apart from the fixed contact assembly; a driven member; a mount mounting the driven member for linear movement along a path substantially parallel to the longitudinal axis of the movable contact assembly, the driven member coupled to the movable contact assembly to move the movable contact assembly between the open circuit position and the closed circuit position with the linear movement of the driven member; a spring member having a compressed state and a released state, which moves the driven member and the movable contact assembly to the closed circuit position; a release member having an opening therein, the release member coupled to the driven member and normally maintaining the spring member in the compressed state; a charge disposed in the opening of the release member, the charge being actuated to fracture the release member and release the spring member to the released state; first and second terminals respectively electrically interconnected with the fixed contact assembly and the movable contact assembly; and means for detecting an arcing fault and responsively activating the charge disposed in the opening of the release member, wherein the activated charge fractures the release member, which releases the spring member, which drives the driven member to move the movable contact assembly to the closed circuit position to eliminate the arcing fault.

The charge may include an electrical input, the means for detecting an arcing fault and responsively activating the charge may comprise means for detecting the arcing fault and responsively outputting a trigger signal, and means for detecting the trigger signal and responsively outputting an activation signal to the electrical input of the charge.

As another aspect of the invention, a shorting switch for eliminating arcing faults in power distribution equipment comprises: a vacuum switch comprising a vacuum envelope containing a fixed contact assembly and a movable contact assembly movable along a longitudinal axis between a closed circuit position in electrical contact with the fixed contact assembly and an open circuit position spaced apart from the fixed contact assembly; a driven member having a longitudinal opening with a circumferential groove therein; a mount mounting the driven member for linear movement along a path substantially parallel to the longitudinal axis of the movable contact assembly, the driven member coupled to the movable contact assembly to move the movable contact assembly between the open circuit position and the closed circuit position with the linear movement of the driven member; a spring member having a compressed state and a released state, which moves the driven member and the movable contact assembly to the closed circuit position; a ball-lock member having a plurality of ball bearings and a push rod with a circumferential groove therein, the ball bearings engaging the circumferential groove of the longitudinal opening of the driven member to hold the spring member in the compressed state and to hold the movable contact assembly in the open circuit position; means for driving the push rod of the ball-lock member into the longitudinal opening of the release member to cause the ball bearings to engage the circumferential groove of the push rod of the ball-lock member and to release the driven member, in order to move the movable contact assembly to the closed circuit position; and first and second terminals respectively electrically connected to the fixed contact assembly and the movable contact assembly.

The means for driving the push rod may be a solenoid having a plunger, which drives the push rod of the ball-lock member into the longitudinal opening of the release member when the solenoid is actuated.

As another aspect of the invention, a shorting system for eliminating arcing faults in power distribution equipment comprises: a housing; a vacuum switch comprising a vacuum envelope containing a fixed contact assembly and a movable contact assembly movable along a longitudinal axis between a closed circuit position in electrical contact with the fixed contact assembly and an open circuit position spaced apart from the fixed contact assembly; a driven member; a mount mounting the driven member in the housing for linear movement along a path substantially parallel to the longitudinal axis of the movable contact assembly, the driven member coupled to the movable contact assembly to move the movable contact assembly between the open circuit position and the closed circuit position with the linear movement of the driven member; a spring member having a compressed state and a released state, which moves the driven member and the movable contact assembly to the closed circuit position; a latch member, which normally latches the driven member to hold the spring member in the compressed state and to hold the movable contact assembly in the open circuit position, the latch member releasing the driven member to move the movable contact assembly to the closed circuit position, with the latch member having an end, which engages the driven member, and a pivot in the housing; means for detecting an arcing fault and responsively unlatching the end of the latch member to release the driven member to move the movable contact assembly to the closed circuit position; and first and second terminals respectively electrically interconnected with the fixed contact assembly and the movable contact assembly.

The end of the latch member may be a first end, and the latch member may also have a second end. The means for unlatching the latch member to release the driven member may be a solenoid having a plunger, which moves the second end of the latch member to pivot the latch member about the pivot and to cause the first end of the latch member to release the driven member.

BRIEF DESCRIPTION OF THE DRAWINGS

A full understanding of the invention can be gained from the following description of the preferred embodiments when read in conjunction with the accompanying drawings in which:

FIG. 1

is an exploded front elevation view of a single phase, spring-loaded, high-speed vacuum shorting switch employing a single vacuum interrupter (VI) in accordance with the present invention.

FIG. 2

is a plan view of the release bolt of

FIG. 1

, which is employed to hold the spring compressed, shown as being fractured after the charge is activated.

FIG. 3

is a plot of breaking torque versus breakline depth for the release bolt of FIG.

1

.

FIG. 4

is a front elevation view of a three-phase, spring-loaded, high-speed vacuum shorting switch employing three of the shorting switches of FIG.

1

.

FIG. 5A

is a schematic diagram of an arcing fault sensor suitable for use with the shorting switch of FIG.

1

.

FIG. 5B

is a schematic diagram of another arcing fault sensor suitable for use with the shorting switch of FIG.

1

.

FIG. 5C

is a schematic diagram of a modified form of the arcing fault sensor of FIG.

5

B.

FIG. 6

is a block diagram of a shorting system including the shorting switch of FIG.

1

.

FIG. 7A

is a plot of bus voltage and current over five cycles at 500V and 38 kA for a shorting switch similar to the shorting switch of FIG.

1

.

FIG. 7B

is a plot of bus voltage and current over 27 cycles at 500V and 38 kA for a shorting switch similar to the shorting switch of FIG.

1

.

FIG. 7C

is a plot of bus voltage and current over 27 cycles at 500V and 38 kA for a shorting switch similar to the shorting switch of

FIGS. 7A and 7B

except that symmetrical length movable and stationary VI electrical stems are employed.

FIG. 8

is a block diagram in schematic form of the detection circuit of FIG.

6

.

FIGS. 9A-9C

are a block diagram in schematic form of the activation circuit of FIG.

6

.

FIG. 10

is a cross-sectional view of a single phase, spring-loaded, high-speed vacuum shorting switch employing a single vacuum interrupter (VI) and a ball-lock mechanism in accordance with another embodiment of the present invention

FIG. 11A

shows the ball-lock mechanism of

FIG. 10

with the VI contacts open, the solenoid plunger stationary, and the ball-lock not released.

FIG. 11B

shows the ball-lock mechanism of

FIG. 10

with the VI contacts open, the solenoid plunger striking the ball-lock push rod, and the ball bearings sliding on the push rod shaft.

FIG. 11C

shows the ball-lock mechanism of

FIG. 10

with the VI contacts open, the solenoid plunger pushing the ball-lock push rod, and the ball bearings rolling down the circumferential groove of the push rod shaft.

FIG. 11D

shows the ball-lock mechanism of

FIG. 10

with the VI contacts closing, the solenoid plunger stopping, and the ball-lock push rod shaft releasing the spring.

FIG. 12

is block diagram of a single phase, high-speed vacuum shorting switch employing a mechanical latch release mechanism, a solenoid, and a single vacuum interrupter (VI) in accordance with another embodiment of the present invention.

FIG. 13

is a schematic diagram of a circuit for driving the solenoids of

FIGS. 10 and 12

.

FIGS. 14A-14C

are plots of solenoid gap, plunger force and solenoid coil current for analyzing the motion of the solenoid of FIG.

12

.

DESCRIPTION OF THE PREFERRED EMBODIMENTS

Referring to

FIG. 1

, a single phase, spring-loaded, high-speed vacuum shorting switch

2

eliminates arcing faults in power distribution equipment (not shown). The shorting switch

2

includes a single vacuum switch, such as a conventional vacuum interrupter (VI)

4

(e.g., a 3″ VI bottle made by Eaton/Cutler-Hammer). As is well known, the vacuum interrupter

4

includes a vacuum envelope or sealed vacuum chamber (e.g., vacuum bottle

6

) containing a fixed contact assembly

8

and a movable contact assembly

10

movable along a longitudinal axis between a closed circuit position (not shown) in electrical contact with the fixed contact assembly

8

and an open circuit position (as shown in

FIG. 1

) spaced apart from the fixed contact assembly

8

.

The fixed contact assembly

8

includes a fixed contact

12

sealed within the sealed vacuum bottle

6

and an electrical conductor

14

connected to the fixed contact at one end thereof. The electrical conductor

14

sealably penetrates the sealed vacuum bottle

6

and terminates at a first terminal

16

at the other end of the fixed contact assembly

8

. The movable contact assembly

10

includes a moveable contact

18

sealed within the sealed vacuum bottle

6

and moveable between a first position (as shown in FIG.

1

), out of electrical communication with the fixed contact

12

, and a second position (not shown), in electrical communication with the fixed contact

12

. The movable contact assembly

10

further includes a movable electrical stem

20

connected to the moveable contact

18

at one end thereof. The movable electrical stem

20

sealably penetrates the sealed vacuum bottle

6

and terminates at a second terminal

22

at the other end of the moveable contact assembly

10

. Preferably, the second terminal

22

is a copper stem including a plurality of threads

24

.

Although a conventional VI

4

is shown, the shield (not shown) and the contacts

12

,

18

may be removed (e.g., as a cost reduction), the length of the movable electrical stem

20

may be adjusted (e.g., shortened), and the length of the fixed or stationary stem or terminal

16

may be adjusted (e.g., lengthened) in order to provide a wide range of lengths (e.g., asymmetrical lengths).

Other modifications to the conventional VI

4

may be made to further reduce the moving mass and/or the cost of the shorting switch

2

. For example, reducing the mass will reduce the time to close. This may involve making a relatively short movable electrical stem

20

, lengthening the stationary or fixed terminal

16

, removing the contacts

12

,

18

, removing the shield (not shown), and/or employing a reduced diameter of the terminals

16

,

22

(e.g., about ⅝″ diameter).

A driven member such as, for example, an aluminum adapter shaft

26

having a threaded longitudinal opening

28

is threadably coupled at one end to the threads

24

of the movable electrical stem

20

. The aluminum adapter shaft

26

is also threadably coupled at its other end to the threads

30

of a release member, such as release bolt

32

.

A suitable mount

34

, which includes a bushing

36

(e.g., nylon) and a spring cover tube

38

, mounts the adapter shaft

26

for linear movement along a path substantially parallel to the longitudinal axis of the movable contact assembly

10

. The adapter shaft

26

is coupled to the movable contact assembly

10

to move the same between the open and closed circuit positions of the vacuum interrupter

4

with the linear movement of such shaft.

A spring member, such as compression spring

40

, has a compressed state (as shown in

FIG. 1

) and a released state (not shown), which moves the adapter shaft

26

and the movable contact assembly

10

to the closed circuit position. The release bolt

32

, which is coupled to the shaft

26

, normally maintains the compression spring

40

in the compressed state. The two ends of the compression spring

40

are disposed between a pair of washers

42

,

44

(e.g., steel). The head

46

of the adapter shaft

26

normally engages the washer

42

and the head

48

of the release bolt

32

engages the washer

44

.

The longitudinal tube

38

has a closed end

50

and an open end

52

with a flange portion

53

(e.g., steel). An opening

54

in the end

50

threadably receives and is closed by a threaded retainer bolt

56

(e.g., steel), which is threadably coupled to a threaded portion

57

of a longitudinal opening

58

of the release bolt

32

, thereby coupling the head

48

of the release bolt

32

to the closed end

50

of the longitudinal tube

38

. The lower (with respect to

FIG. 1

) end of the compression spring

40

engages the washer

44

(and, thus, the head

48

of the release bolt

32

at the closed end

50

of the longitudinal tube

38

).

Disposed within the release bolt opening

58

is a suitable charge, such as an electrically activated, chemical charge

60

. The charge

60

is actuated to fracture the release bolt

32

and release the compression spring

40

to the released state. The longitudinal tube

38

houses the compression spring

40

, the release bolt

32

and the charge

60

, which is, of course, advantageous during the activation of such charge.

The bushing

36

includes an upper portion

62

and a lower portion

64

(e.g., with respect to FIG.

1

). The upper portion

62

rests on the flange portion

53

of the longitudinal tube

38

and the lower portion

64

rests in the open end

52

of such tube. A longitudinal opening

66

passes through the upper and lower portions

62

,

64

of the bushing

36

. As shown in

FIG. 4

, the adapter shaft

26

rests in the longitudinal opening

66

of the bushing

36

.

The opening

58

of the release bolt

32

provides a longitudinal cavity

68

(shown in

FIG. 4

) along the longitudinal axis of such release bolt. The charge

60

activated to provide a shock wave to fracture the release bolt

32

. Preferably, as shown in

FIGS. 2 and 4

, the body

70

of the release bolt

32

has a breakline

72

disposed thereon to locate and control fracture of the release bolt

32

responsive to the shock wave. The breakline

72

has a predetermined depth in the release bolt body

70

, and the compression spring

40

has a predetermined compression force, with the release bolt

32

being structured to maintain at least the compression force until after the charge

60

is activated.

The release bolt

32

normally compresses the compression spring

40

. After activation of the charge

60

inside the release bolt

32

, such bolt fractures at or about the breakline

72

(as shown in FIG.

2

), thereby releasing the compression spring

40

. In turn, the upper end (with respect to

FIG. 1

) of the spring

40

biases the washer

42

and the adapter shaft

26

to move the movable contact assembly

10

to the closed circuit position after actuation of the charge

60

.

The exemplary charge

60

is a small electrically activated, chemical charge, such as model number RP-501 made by Reynolds Industries Systems, Inc. RISI). The RP-501 is a standard, end lighting, exploding bridge wire (EBW) detonator for use in general purpose applications (e.g., it is capable of detonating compressed TNT and COMP C-4). Although an exemplary detonator charge is employed, any suitable charge may be employed to fracture any suitable release member.

The release bolt

32

is employed to hold open the separable contacts

12

,

18

and to compress the spring

40

as shown in FIG.

1

. Upon activation of the charge

60

inside the bolt

32

, such bolt preferably fractures at a predetermined location, such as the breakline

72

, thereby releasing the energy of the compression spring

40

. The exemplary spring

40

closes and holds the contacts

12

,

18

closed with about 512 lbs. of force. This holding force prevents such contacts from reopening and vaporizing, while maintaining a suitably low contact resistance. For example, it is believed that at least about 300 lbs. of force is desired to hold the contacts

12

,

18

closed at a rated current of about 38 kA

RMS

symmetrical.

As shown in

FIG. 4

, the charge

60

includes an electrical input, such as a pair of conductors

74

, which pass through the opening

58

of the release bolt

32

and through an opening

76

of the bolt

56

. The charge

60

is suitably activated by an electrical signal on the conductors

74

to provide a shock wave to fracture the release bolt

32

.

In the exemplary embodiment, the bolt body

70

has a 0.5-inch diameter and the bolt cavity

68

has a 0.295-inch diameter. The exemplary bolt

32

is 4.5 inches in length, with the cavity

68

being 2.0 inches deep from the bolt head

48

, and the breakline

72

being 1.9 inches deep from the bolt head

48

. The exemplary breakline

72

is employed to locate and control the fracture zone when the shock wave, created from the charge

60

, fractures the metal release bolt

32

.

The first and second terminals

16

,

22

, which are respectively electrically interconnected with the fixed contact assembly

8

and the movable contact assembly

10

, are adapted for electrical connection to first and second power lines

78

,

80

, respectively. For example, the first power line

78

may be a copper power bus (e.g., a single phase; one phase of a three-phase power bus) and the second power line

80

may be a copper ground bus. Although a ground bus is shown, a neutral bus or a different phase may be employed.

As shown in

FIG. 1

, the vacuum interrupter

4

has a pair of mounting studs

82

,

84

(e.g., steel), which pass through respective openings

86

,

88

of the second power line

80

and through respective openings

90

,

91

of a bushing

96

and which are secured thereto by suitable fasteners

92

. The second terminal

22

of the vacuum interrupter

4

passes through a suitably sized opening

94

of the second power line

80

and through an opening

95

of the bushing

96

(e.g., nylon). The bushing

96

is secured with respect to the second terminal

22

by a nut

98

(e.g., brass). The nut

98

is suitable attached (e.g., welded; brazed) to a flexible shunt

99

(e.g., a copper laminate), which is suitable electrically connected to the second power line

80

by a pair of fasteners

100

(e.g., brass) at opposite ends of the shunt

99

. As discussed above, the second terminal

22

is coupled to the adapter shaft

26

for movement therewith and is electrically connected to the second power line

80

by the flexible shunt

99

. The flexible shunt

99

is movable between and is preferably insulated by the upper and lower nylon bushings

96

,

36

.

FIG. 3

shows the result obtained from testing the torque required to fracture a ½″ diameter bolt with a 0.295″ diameter hole for the charge. A breakline (e.g.,

72

of

FIGS. 2 and 4

) is employed to locate and control the fracture zone when the shock wave, created from the charge

60

, fractures the metal. The vertical line (MT) represents the minimum torque on the release bolt

32

suitable to fully compress the compression spring

40

. The plot shows the maximum depth of the breakline

72

while still maintaining a spring force of about 1200 pounds plus a suitable safety factor. The exemplary release bolt

32

is “grade 5” and can safely withstand a tensile stress of about 120,000 PSI without fracturing. An optimum breakline depth of about 0.025 inch or 0.03 inch is preferably employed to reliably fracture the exemplary bolt

32

with the exemplary charge

60

and still allow the spring

40

to be compressed solid and held with a suitable safety margin.

FIG. 4

shows a three-phase, spring-loaded, high-speed vacuum shorting switch

101

employing three of the shorting switches

2

of FIG.

1

. For example, the three first terminals

16

of the three shorting switches

2

may be respectively electrically connected with three corresponding power busses (e.g., phases A, B and C). The three second terminals

22

of the three shorting switches

2

may be electrically connected to a common ground bus (not shown) by the common flexible shunt

99

′. The three-phase shorting switch

101

may employ, for example, a conventional molded housing

102

without an operating mechanism.

The closing times of the shorting switches

2

of

FIG. 1

depend upon the amount of mass being moved and on the force applied by the compression springs

40

. For example, the exemplary shorting switches

2

,

101

are capable of being activated in the presence of an arcing fault in medium voltage switchgear and are able to maintain contact closure under medium voltage operating circuit parameters (e.g., 15 kV

RMS

at 38 kA

RMS

).

One form of an arcing fault sensor unit suitable for use with the shorting switches

2

,

101

is shown in FIG.

5

A. The sensor unit

103

includes the first photovoltaic device

104

including at least one, or a plurality of series connected photovoltaic cells

105

, and a first filter

107

which filters light incident upon the photovoltaic cells

105

. This first filter

107

has a passband centered on the characteristic wavelength, e.g., 521.820 nm, of the arcing material.

The sensor

103

includes a second photovoltaic device

109

, which also includes one or more series connected photovoltaic cells

111

, and a second filter

113

which filters light incident upon the photovoltaic cells

111

and has a passband that does not include the characteristic wavelength of the arcing material, e.g., centered on about 600 nm in the exemplary system.

The first photovoltaic device

104

generates a sensed light electrical signal in response to the filtered incident light, and similarly, the second photovoltaic device

109

generates a background light electrical signal with an amplitude dependent upon the irradiance of light in the passband of the second filter

113

. An electric circuit

115

, having a first branch

115

1

connecting the first photovoltaic cells

104

in series and a second branch

115

2

similarly connecting the second photovoltaic cells

111

in series, connects these two electrical signals in opposition to a light-emitting device such as a light-emitting diode (LED)

117

. When arcing is present, the sensed light electrical signal generated by the first photovoltaic device

104

exceeds the background light electrical signal generated by the second photovoltaic device

109

by a threshold amount sufficient to turn on the LED

117

. While in the absence of arcing, the first photovoltaic device

104

will generate a sensed light electrical signal due to some irradiance in the passband of the first filter

107

, it will be insufficient to overcome the reverse bias effect of the background light signal generated by the second photovoltaic device

109

on the LED

117

. In fact, where the background light is fluorescent, from an incandescent bulb or a flashlight all of which have very low irradiance in the passband of the first filter

107

, but significant irradiance in the passband of the second filter

113

, the background light electrical signal will significantly exceed the sensed light electrical signal and strongly reverse bias the LED

117

. The filters

107

and

113

can be interference filters, although lower cost bandpass filters could also be utilized.

An alternate embodiment of the sensor unit

103

′ shown in

FIG. 5B

adds a bias generator

119

in the form of one or more additional photovoltaic cells

121

connected in series with the first photovoltaic device

104

in the first branch

115

1

of the electrical circuit

115

. This puts a forward bias on the LED

117

so that fewer or smaller filtered photovoltaic cells

105

and

111

can be used. This also reduces the size and therefore the cost of the filters

107

and

113

. As the additional photovoltaic cells

121

are not provided with filters, the total cost of the sensor is reduced. The embodiment of

FIG. 5B

can be modified as shown in

FIG. 5C

to place the bias generating cells

121

of the sensor

103

″ in series with both filtered photovoltaic cells

105

and

111

, but still provide the same effect of forward biasing the LED

117

.

Through their utilization of photovoltaic cells

105

,

111

and

121

, the sensors

103

and

103

′ of

FIGS. 5A-5C

are self-energized.

FIG. 6

shows a shorting system

140

including one or more shorting switches

2

of

FIG. 1

(only one switch (SW)

2

is shown in FIG.

6

). The shorting system

140

eliminates an arcing fault

142

in medium voltage power distribution equipment

144

(e.g., switchgear). The shorting system

140

also includes a detection and activation circuit

146

for detecting the arcing fault

142

and responsively activating the shorting switch charge (C)

60

, in order that the activated charge

60

results in the elimination of the arcing fault as discussed above in connection with

FIGS. 1-3

. The circuit

146

includes a detection (OD) circuit

148

for detecting the arcing fault

142

and responsively outputting one or more trigger signals

150

, and an activation circuit (ACT)

152

for detecting the one or more trigger signals

150

and responsively outputting the activation signal

154

to the electrical inputs

155

of the charges

60

. The detection circuit

148

utilizes photovoltaic cells in a sensor unit, such as one of the sensor units

103

,

103

′,

103

″ of

FIGS. 5A-5C

.

FIGS. 7A-7C

are test waveforms showing that a shorting switch similar to the shorting switch

2

of

FIG. 1

is capable of operating within a suitable time (e.g., without limitation, less than about 4 ms) and can hold closed at about 38 kA

RMS

fault current for a duration of about 0.5 second. The conventional molded housing

102

of

FIG. 4

makes for convenient mounting to the bus bar (e.g.,

78

of

FIG. 1

) of the switchgear (e.g.,

144

of

FIG. 6

) without any additional mold costs.

FIGS. 7A and 7B

show plots of bus voltage and current over five cycles and 27 cycles, respectively, at 500V and 38 kA. In the test of

FIG. 7A

, the window in a molded case circuit breaker (not shown) used to generate arc light for testing purposes was clear (clean).

For the test of

FIG. 7B

, the window of the sensor unit is blocked to prevent arc light from reaching the detector, which is aimed at the arc-viewing window. A circuit breaker (not shown) arcs phase-to-phase on the line side and the sensor unit detects this. The sensor did not respond to the arc through the window in the circuit breaker since the light was blocked. But, since the breaker arced phase-to-phase, externally, the arc light was subsequently detected, albeit at a later time (4.4 ms total response time versus 3.20 ms). In response, the shorting switch

2

responsively quenches the arcing fault, thereby saving the circuit breaker.

FIG. 7C

shows a plot of bus voltage and current over 27 cycles at 500V and 38 kA on a shorting switch, similar to the shorting switch for

FIGS. 7A and 7B

, except that symmetric length movable and stationary electrical stems are employed in the vacuum interrupter. In this test, the window of the sensor unit is darkened. The sensor response time is longer (0.82 ms versus 0.60 ms) and the total response time is less than the response time for

FIGS. 7A and 7B

since a smaller moving mass is employed.

Table 1 summarizes the shorting system operating times for the examples of

FIGS. 7A-7C

.

TABLE 1

Total

Sensor

Switch

Operating

Time (ms)

Time (ms)

Time (ms)

Arc Source

0.60

2.60

3.20

Clean window

1.58

2.84

4.42

Window black, phase-phase

0.82

2.06

2.88

Dark window

The exemplary spring-loaded vacuum shorting switch

2

successfully operates within about 3.2 ms (FIG.

7

A), including sensing time, and holds closed for 27 cycles at 38 kA

RMS

(FIGS.

7

B and

7

C). As shown by Table 1, the shorting switches, switch, after triggering, between about 2.06 ms and about 2.84 ms and each one remains closed for the duration of the fault current.

Referring to

FIG. 8

, the detection circuit

148

is shown. In the exemplary embodiment, the medium voltage power distribution equipment

144

of

FIG. 7

includes two circuit breaker cells

156

,

157

, two upper cable cells

158

,

159

, and two lower cable cells

160

,

161

, although the invention is applicable to a wide range of medium voltage power distribution equipment having any count (e.g., one or more) of cells in which an arcing fault may occur. As another example, U.S. Pat. No. 6,229,680, which is incorporated by reference herein, discloses a switchgear cabinet having a forward compartment, a middle compartment and a rear compartment. The forward compartment is divided vertically into three cells in which are housed electrical switching apparatus such as circuit breakers.

The detection circuit

148

includes six photovoltaic sensors

162

,

164

,

166

,

168

,

170

,

172

adapted to detect arcing faults in the cells

156

,

158

,

160

,

157

,

159

,

161

, and output optical trigger signals

174

,

176

,

178

,

180

,

182

,

184

, respectively. These photovoltaic sensors

162

,

164

,

166

,

168

,

170

,

172

are self-powered from arc light and have an output

186

(as shown with sensor

162

) with the respective optical trigger signals

174

,

176

,

178

,

180

,

182

,

184

, which are responsive to the arc light. In the exemplary embodiment, suitable photovoltaic sensors are shown in

FIGS. 5A-5C

, although any suitable sensor for detecting any characteristic of an arcing fault may be employed. In the exemplary embodiment, the detection circuit

148

is employed for each switchgear enclosure (not shown), with three photovoltaic sensors for each circuit breaker cell.

The detection circuit

148

further includes a suitable optical multiplexer

188

having a plurality of fiber optic inputs

190

,

192

,

194

,

196

,

198

,

200

and a fiber optic output

202

. A plurality of suitable fiber optic cables

204

,

206

,

208

,

210

,

212

,

214

are connected between the outputs

186

of the photovoltaic sensors

162

,

164

,

166

,

168

,

170

,

172

and the inputs

190

,

192

,

194

,

196

,

198

,

200

, respectively, of the optical multiplexer

188

. The fiber optic cables (as shown with cable

210

) include a first connector

216

attached to the corresponding photovoltaic sensor output (as shown with the output

186

of sensor

168

) and a second connector

218

attached to the corresponding optical multiplexer input (as shown with input

196

).

The output

202

of the optical multiplexer

188

outputs an optical trigger signal

220

to another fiber optic cable

222

, which includes a first connector

224

attached to the multiplexer output

202

. The other end (as shown in

FIG. 9A

) of fiber optic cable

222

includes a second connector

226

, which is attached to the activation circuit

152

. The optical multiplexer

188

functions to repeat any of the first optical trigger signals

174

,

176

,

178

,

180

,

182

,

184

to the second optical trigger signal

220

. In operation, the photovoltaic sensors

162

,

164

,

166

,

168

,

170

,

172

and the optical multiplexer

188

have a detection time of about 550 &mgr;s after initiation of an arcing fault event to the activation of the second optical trigger signal

220

, although the invention is applicable to a wide range of response times. The detection time varies (e.g., about 300 &mgr;s to about 2 ms) as a function of the arc current magnitude and line of sight orientation of the sensor to the arc.

The detection circuit

148

further includes a suitable power supply, which in the exemplary embodiment is an AC/DC power supply

228

, which inputs an AC line voltage

230

and outputs a suitable DC voltage

240

to the optical multiplexer

188

at node

241

. For example, for input

194

(A), the optical multiplexer

188

has an electrical output

242

, which is electrically connected to the gate of transistor

244

through resistor

261

. The other five multiplexer inputs

192

,

190

,

196

,

198

,

200

have similar outputs

242

B,

242

C,

242

D,

242

E,

242

F, respectively. The emitter of the transistor

244

is electrically connected to the common

246

of the power supply

228

, and the collector of the transistor

244

is electrically connected through a resistor

248

to the cathode of an output photodiode

250

of the optical multiplexer

188

. The anode of the photodiode

250

and the collectors of the input phototransistors

251

A-

251

F (e.g., similar to the phototransistor

408

of

FIG. 9A

) of the optical multiplexer

188

are electrically connected to the DC voltage node

241

. A resistor

252

is electrically connected between the gate and emitter of the transistor

244

. A zener diode

254

is electrically connected in parallel with the resistor

252

.

The detection circuit

148

preferably includes a plurality of suitable indication circuits, such as

258

for multiplexer input A, having an input

259

and one or more indication outputs

260

for indicating the occurrence of an arc fault trigger event as determined by a suitable voltage at the optical multiplexer output

242

as measured across resistors

261

and

252

. Other circuits

258

are provided for the other multiplexer inputs, which have resistors

261

B,

261

C,

261

D,

261

E,

261

F electrically connected between the respective outputs

242

B,

242

C,

242

D,

242

E,

242

F and the gate of transistor

244

. Normally, the flip-flop (FF)

262

has a reset state, which is established by pushbutton

264

. Otherwise, in response to an arc trigger event, the positive voltage across the resistors

261

and

252

, as buffered by operational amplifier

266

, causes the flip-flop outputs

268

,

270

to change state. In response, a green LED

272

is extinguished, and a red LED

274

is illuminated. Following manual operation of the pushbutton

264

, the flip-flop outputs

268

,

270

change state, again, with the red LED

274

being extinguished and the green LED

272

being illuminated.

FIGS. 9A-9B

show the activation circuit

152

of

FIG. 7

, which detects one or more of the trigger signals, such as the optical trigger signal

220

of

FIG. 8

, and responsively outputs the activation signal

154

to the electrical inputs

155

of the charges

60

. In the exemplary embodiment, the activation circuit

152

is employed with a three-phase circuit breaker (not shown) having a pair of charges

60

for each of the shorting switches

2

for each of the three phases A,B,C, although the invention is applicable to any count of shorting switches

2

(e.g., one or more), any count of phases (e.g., one, two, three or more), any count of shorting switches per phase (e.g., one, two, or more), and any count of charges (e.g., one or more) per shorting switch.

The activation circuit

152

includes a suitable high voltage power supply

280

(FIGS.

9

B-

9

C), which generates a suitable charge voltage

282

(e.g., without limitation, 2000 VDC) between nodes

284

,

286

from an input voltage

288

(e.g., without limitation, 120 VAC

RMS

). A triggerable spark gap

290

, such as a model GP-486 marketed by Perkin Elmer of Salem, Mass., has an input

292

electrically connected to the node

284

, an output

294

electrically connected to a first input

296

of the charges

60

, and a trigger control input

298

. The other node

286

is electrically connected to the second input

300

of the charges

60

. Normally, the spark gap

290

presents an open circuit between the input

292

and output

294

. In response to a suitable trigger voltage between the trigger control input

298

and the output

296

(with each of the charges essentially providing a near short circuit between the inputs

296

,

300

), the spark gap

290

becomes a short circuit, thereby electrically connecting the node

284

to the first charge input

296

and, thus, electrically connecting the charge voltage

282

to the charge electrical inputs

155

.

A circuit

302

detects the one or more trigger signals

220

and responsively outputs a control signal

304

(with respect to the node

286

) to the control input

298

of the spark gap

290

. As discussed above, in responsive to the control signal

304

, the spark gap

290

outputs the activation signal

154

to the electrical inputs

155

of the charges

60

.

The circuit

302

, which is preferably housed in a suitable EMI/magnetic shield (not shown), includes one or more circuits

306

,

306

A,

306

B for detecting the one or more trigger signals

220

(only one trigger signal is shown), respectively. For example, zero, one or more (e.g.,

306

A,

306

B) additional circuits may be employed in the event that the medium voltage power distribution equipment (e.g.,

144

of

FIG. 7

) includes more than six exemplary arc fault detection locations. The circuits

306

,

306

A,

306

B include a common output

308

having a detected trigger signal

310

.

A circuit

312

has an input

314

, which is electrically connected to the common output

308

of the circuits

306

,

306

A,

306

B. The circuit

312

outputs the control signal

304

responsive to the detected trigger signal

310

.

The power supply

280

of

FIG. 9B

includes a line-conditioning filter

316

, an MOV

318

, a fuse

320

, a transformer

322

, a full wave bridge

324

, a capacitor

326

and a DC/DC regulator (REG)

332

. When contacts

334

,

336

are closed, the power supply

280

functions to output a suitable DC voltage

338

(e.g., +17 VDC) between power supply node

340

and common node

342

. In turn, the DC/DC regulator

338

outputs a suitable DC voltage

346

(e.g., +12 VDC) to a suitable high voltage DC/DC regulator

348

of FIG.

9

C. In turn, the regulator

348

generates the charge voltage

282

between the nodes

284

,

286

. The series combination of a resistor

350

and an LED

352

is electrically connected in parallel with the capacitor

326

to indicate the DC voltage

338

and the test mode as selected by the closed state of the contact

334

.

Preferably, for safety considerations, a suitable switch

354

is provided with the normally open (NO) contacts

334

,

336

and a suitable normally closed (NC) contact

358

(e.g., one NC contact or plural NC contacts electrically connected in series). The switch

354

preferably includes a key (not shown), which key has three positions, including a first off position

364

, a second test position

366

, and a third arm position

368

, and is removable in the off and arm positions

364

,

368

. The states of the contacts

334

,

336

,

358

are static in each of the three key positions

364

,

366

,

368

.

In the off position

364

, the NO contacts

334

,

336

are open, thereby disabling the power supply

280

, the regulators

332

,

348

and the circuit

302

, and the NC contact

358

is closed, thereby ensuring that the activation signal

154

is suitably electrically connected to the common node

286

.

In the test position

366

, the NO contact

334

is closed, thereby enabling the DC voltage

338

(e.g., +17 VDC) at the power supply node

340

, and the circuit

302

. However, since the NO contact

336

remains open, the regulators

332

,

348

remain disabled.

In the arm position

368

, first, the NC contact

358

is opened and, then, the NO contact

336

is closed. This safely removes the short circuit between the nodes

296

,

286

before the regulators

332

,

348

are enabled.

The circuit

306

preferably includes a suitable indication circuit

369

, which is similar to the indication circuit

258

of FIG.

8

. The circuit

369

has one or more indication outputs

370

for indicating that the trigger signal

220

has been received as determined by a suitable voltage at the node

410

. Normally, the flip-flop

372

has a reset state, which is established by pushbutton

374

. Otherwise, in response to the trigger signal

220

, the voltage

409

, as buffered by operational amplifier

376

, causes the flip-flop outputs

378

,

380

to change state. In response, a green LED

382

is extinguished, and a red LED

384

is illuminated. Following manual operation of the pushbutton

374

, the flip-flop outputs

378

,

380

change state, again, with the red LED

384

being extinguished and the green LED

382

being illuminated.

The circuit

302

includes a pair of power supplies

386

,

388

, both of which are powered from nodes

390

and

392

of the power supply

280

. The circuit

302

also includes a filter bushing

394

, which further filters the switched line voltage between the nodes

390

,

392

, and a fuse

396

. The first power supply

386

has an output

398

with a suitable unregulated DC voltage

399

(e.g., +170 VDC) with respect to a ground node

400

. The voltage

399

powers a portion of the circuit

312

. The second power supply

388

has an output

402

with a suitable unregulated DC voltage

403

(e.g., +24 VDC) with respect to the ground node

400

. The voltage

403

powers a DC/DC regulator

405

, which has an output

406

with a suitable regulated DC voltage

407

(e.g., +15 VDC) with respect to the ground node

400

. The voltage

407

powers the circuits

306

,

306

A,

306

B.

As discussed above, the circuits

306

,

306

A,

306

B detect the one or more optical trigger signals

220

(as shown with circuit

306

) and have a common output

308

with the detected trigger signal

310

. The circuit

306

includes a suitable phototransistor

408

, which in the exemplary embodiment is a model OP802WSL marketed by Honeywell of Morristown, N.J., although any suitable phototransistor for optical signals may be employed. Whenever the optical trigger signal

220

is active, the phototransistor

408

generates a corresponding voltage

409

at output

410

by sourcing current through resistor

412

to the ground node

400

. The voltage

409

is buffered by the non-inverting (+) input of operational amplifier

414

, which provides the corresponding detected trigger signal

310

at the common output

308

. The phototransistor

408

and the operational amplifier

414

are both powered from the power supply output

406

having the regulated DC voltage

407

. The circuits

306

A and

306

B are similar to the circuit

306

.

The circuit

312

, which outputs the control signal

304

responsive to the detected trigger signal

310

, includes a one-shot multivibrator

416

, a transistor

418

, resistor

419

, capacitor

420

, and a transformer

422

. In response to the rising edge of the detected trigger signal

310

, as detected at input

423

, the multivibrator

416

provides a suitable pulse at output

424

, which is connected by line

425

to the gate of the transistor

418

. In turn, the transistor

418

turns on, thereby effectively electrically connecting the collector side of resistor

419

to the ground node

400

. The step change in voltage (e.g., about −170 VDC) on the transformer side of capacitor

420

is communicated to the primary winding

426

of the transformer

422

. In response, the transformer (inverted) secondary winding

428

provides a suitable positive voltage pulse, which is the control signal

304

, to twisted pair conductors

430

, which are electrically connected to the control input

298

of the spark gap

290

and the second input

300

of the charges

60

. Preferably, the circuits

306

,

312

of the activation circuit

152

provide a minimal delay (e.g., without limitation, about 40 &mgr;s) between the activation of the optical trigger signal

220

and the activation of the control signal

304

.

Preferably, the circuit

302

includes a circuit

432

for tripping an upstream circuit breaker (not shown) in response to the detected trigger signal

310

and the output pulse of the multivibrator output

424

. The circuit

432

includes a transistor

434

and a relay

436

having a coil

438

and contacts

440

. The power supply

386

includes the series combination of a resistor

442

and a capacitor

444

, which combination is electrically connected between the nodes

398

,

400

. The common node

446

of the resistor-capacitor combination sources a suitable voltage

448

(e.g., about +170 VDC) to one side of the relay coil

438

. The other side of the relay coil

438

is switched by the transistor

434

to the ground node

400

in response to the output pulse of the multivibrator output

424

. In response to the energization of the coil

438

, the normally open (NO) contacts

440

are closed. The contacts

440

, in turn, are electrically connected to terminals

450

,

452

, which are adapted for electrical connection to the upstream circuit breaker by a suitable cable (not shown).

Preferably, the circuits

306

,

312

,

432

of the activation circuit

152

provide a suitable delay (e.g., without limitation, about 6 to 10 ms) between the activation of the optical trigger signal

220

and the closure of the NO contacts

440

. This permits the charges

60

to initiate the relatively high-speed shorting of the arcing fault through the shorting switch(s)

2

of

FIG. 1

, with the prompt, albeit relatively slower, clearing of the short circuit(s) caused by such switch(s)

2

, through the tripping of the upstream circuit breaker (not shown).

Preferably, the activation circuit

152

includes an output relay and contact (not shown) which is employed to enable an upstream circuit breaker (not shown) for the medium voltage power distribution equipment

144

of FIG.

7

. The activation circuit

152

preferably further includes an annunciator circuit (not shown) which annunciates when such upstream circuit breaker is not enabled. Preferably, such upstream circuit breaker is enabled whenever the various power supply voltages (e.g., +12 VDC, +17 VDC, +170 VDC, +2000 VDC) of the activation circuit

152

are suitably energized.

FIG. 10

shows a single phase, spring-loaded, high-speed vacuum shorting switch

502

employing the conventional vacuum interrupter (VI)

4

of

FIG. 1 and a

ball-lock mechanism

504

. The switch

502

includes a driven member

506

having a longitudinal opening

508

with a circumferential groove

510

therein. A suitable mount

512

, which includes a frame

514

and a spring cover tube

516

, mounts the driven member

506

for linear movement along a path substantially parallel to the longitudinal axis of the movable contact assembly

10

. The driven member

506

is suitably coupled (e.g., by threaded rod

518

) to the second terminal

22

of the movable contact assembly

10

to move such assembly between the open circuit position (as shown in

FIG. 10

) and the closed circuit position (shown in phantom line drawing in

FIG. 11D

) with the linear movement of the driven member

506

(as shown in FIGS.

11

B-

11

D).

A suitable spring member, such as compression spring

520

, has a compressed state (as shown in

FIG. 10

) and a released state (as shown in FIG.

11

D), which moves the driven member

506

and the movable contact assembly

10

to the closed circuit position.

The ball-lock mechanism

504

has a plurality of ball bearings, such as

522

,

524

, and a push rod

526

with a circumferential groove

528

therein. As shown in

FIG. 10

, the ball bearings

522

,

524

normally engage the circumferential groove

510

of the longitudinal opening

508

of the driven member

506

to hold the spring member

520

in the compressed state between a flange portion

530

of the ball-lock mechanism

504

and a flange portion

532

of the driven member

506

. This, also, holds the movable contact assembly

10

in the open circuit position of FIG.

10

.

A suitable electromechanical mechanism, such as a solenoid

534

, includes a plunger

536

for driving the ball-lock push rod

526

into the longitudinal opening

508

of the driven member

506

to cause the ball bearings

522

,

524

to engage the circumferential groove

528

of the ball-lock push rod

526

and to release the driven member

506

, in order to move the movable contact assembly

10

to the closed circuit position.

Whenever the solenoid

534

is activated by a suitable activation signal on leads

538

, which energize the solenoid coil

540

, the solenoid plunger

536

drives (as shown in

FIGS. 11B and 11C

) the ball-lock push rod

526

into the driven member longitudinal opening

508

.

The longitudinal spring tube

516

has a first end

542

and a second end

544

and houses the compression spring

520

and the ball-lock mechanism

504

. The first spring end

546

engages the flange portion

530

of the ball-lock mechanism

504

, which is suitably coupled to the first tube end

542

. The frame

514

has an opening

548

at the second tube end

544

with the upper (with respect to

FIG. 10

) end portion

550

of the driven member

506

passing through such opening. Whenever the ball-lock mechanism

504

releases the driven member

506

, the second spring end

552

suitably biases the driven member

506

to move the movable contact assembly

10

to the closed circuit position.

Similar to the shorting switch

2

of

FIG. 1

, the first VI terminal

16

is adapted for electrical connection (e.g., through a suitable conductive fastener

554

) to a first power line, such as power bus

78

, and the second VI terminal

22

is adapted for electrical connection to a suitable flexible shunt

556

(e.g., copper laminate). In this manner, the driven member

506

and the movable contact assembly

10

are free to move independent of a rigid second power line

558

(e.g., ground or neutral bus).

FIG. 11A

shows the locked position of the ball-lock mechanism

504

with the solenoid plunger

536

stationary, the ball bearings

522

,

524

in the circumferential groove

510

, and the separable contacts

12

,

18

of

FIG. 10

being open. The ball-lock push rod

526

passes through an opening

560

in the flange portion

530

of the ball-lock mechanism

504

for engagement by the solenoid plunger

536

as shown in

FIGS. 11B-11D

.

In

FIG. 11B

, the solenoid coil

540

is energized, the solenoid plunger

536

is actuated to engage the ball-lock push rod

526

, thereby causing the ball bearings

522

,

524

to slide on the shaft

562

of such push rod. At this stage, the driven member

506

has not yet been released and the separable contacts

12

,

18

of

FIG. 10

remain open.

Next, as shown in

FIG. 11C

, the solenoid coil

540

remains energized, the actuated solenoid plunger

536

continues to drive the ball-lock push rod

526

, thereby causing the ball bearings

522

,

524

to roll down into the push rod circumferential groove

522

and releasing the driven member

506

, although the separable contacts

12

,

18

of

FIG. 10

remain open. Hence, as shown by

FIGS. 11B and 11C

, the actuated solenoid plunger

536

strikes the push rod

526

causing the ball bearings

522

,

524

to enter the circumferential groove

522

, thereby releasing the driven member

506

and the compression spring

520

.

Finally, as shown in

FIG. 11D

, the compression spring

520

drives the driven member

506

to close the separable contacts

12

,

18

of FIG.

10

.

Although the ball-lock mechanism

504

is shown in FIGS.

10

and

11

A-

11

D, any suitable ball-lock configuration may be employed. As shown in

FIG. 10

, the ball-lock mechanism

504

includes a headed shank

564

having a central through-bore

566

. An end of the shank

564

opposite the head has a pair of radially opposite passageways

568

, which extend radially outward from the central through-bore

566

. Preferably, the passageways

568

are formed with a constant diameter bore and receive a similarly sized one of the ball bearings

522

,

524

. The ball bearings

522

,

524

are inserted into the corresponding passageway

568

and the outer edge of each such passageway is locally crimped. This traps the ball bearings

522

,

524

at their widest diameter such that the passageway

568

will still allow the radial outermost surface of the ball bearing to extend beyond the outer surface of the shank

564

. Subsequently, the push rod

526

is inserted into the central through-bore

566

, which traps the ball bearings

522

,

524

in a radially outward position such that their radial outermost feature extends beyond the radial outer surface of the shank

564

.

The push rod

526

is axially retained in the position of

FIG. 11A

by the solenoid plunger

536

. However, the radially inward extending circumferential groove

522

allows the ball bearings

522

,

524

to be radially inwardly displaced by axially displacing the push rod

526

with the solenoid plunger

536

as shown in

FIGS. 11B and 11C

in order to center the groove

522

beneath the pair of ball bearings

522

,

524

(FIG.

11

C).

FIG. 12

shows a single phase, spring-loaded, high-speed vacuum shorting switch

602

employing the conventional vacuum interrupter (VI)

4

of

FIG. 1

, a spring member such as compression spring

604

, a mechanical latch release mechanism

605

, and an electromechanical mechanism such as solenoid

606

.

A driven member

608

includes a circular flange portion

610

, an upper shaft portion

612

and a coupler

614

. The upper shaft portion

612

is suitably secured (e.g., by welding) to the circular flange portion

610

. The coupler

614

is suitably secured (e.g., threadably connected) to both the driven member

608

and the terminal

22

of the movable contact assembly

10

to move such assembly between the open circuit position and the closed circuit position with the linear movement of the driven member

608

.

A suitable mount

616

, which includes the frame

514

of

FIG. 10 and a

spring tube

617

, mounts the driven member

608

and, thus, the second terminal

22

with respect to a housing

618

for linear movement along a path substantially parallel to the longitudinal axis of the movable contact assembly

10

. The compression spring

604

is disposed within the spring tube

617

with the circular flange portion

610

at one end and a surface

620

of the housing

618

at the other end. The compression spring

604

has a compressed state (as shown in

FIG. 12

) and a released state (not shown), which moves the driven member

608

and the movable contact assembly

10

to the closed circuit position (not shown).

The mechanical latch release mechanism

605

includes a latch member, such as lower shaft portion

622

, having an upper hook portion

624

, which normally latches a corresponding lower hook portion

626

of the upper shaft portion

612

of the driven member

608

. This holds the compression spring

604

in the compressed state and holds the movable contact assembly

10

in the open circuit position. Otherwise, the lower shaft portion

622

releases the upper shaft portion

612

in order to release the driven member

608

to move the movable contact assembly

10

to the closed circuit position.

The lower shaft portion

622

includes a central opening

627

and a lower pin

628

. A spring

630

(e.g., about three pounds) is connected between the upper hook portion

624

and a surface

632

of the housing

618

in order to bias the lower shaft portion

622

clockwise with respect to

FIG. 12 and

, thereby, suitably engage the corresponding hook portions

624

,

626

. The solenoid

606

has a normally extended plunger

634

and coil

636

. Whenever the solenoid

606

is activated by a suitable activation signal on leads

637

, which energizes the solenoid coil

636

, such solenoid retracts the plunger

634

(e.g., toward the right of FIG.

12

).

The plunger

634

has an opening

638

, which captures therein the pin

628

of the lower shaft portion

622

. When the plunger

634

retracts, a plunger end portion

639

engages the pin

628

, which rotates the lower shaft portion

622

counter-clockwise about pivot point

640

of the housing

618

against the bias of the spring

630

. Upon suitable retraction of the plunger

634

and suitable corresponding counter-clockwise rotation of the lower shaft portion

622

, the upper hook portion

624

of the lower shaft portion

622

unlatches the lower hook portion

626

of the upper shaft portion

612

, thereby releasing the driven member

608

. In turn, the compression spring

604

drives the driven member

608

upward (with respect to

FIG. 12

) in order to close the separable contacts

12

,

18

of the shorting switch

602

. Similar to the shorting switch

502

of

FIG. 10

, the first terminal

16

is adapted for electrical connection to a power line, such as

78

, and the second terminal

22

is adapted for electrical connection to a flexible conductor, such as

556

.

The shorting switch

502

of FIG.

10

and the shorting switch

602

of

FIG. 12

are operable with a detection and activation circuit, which is similar to the circuit

146

of

FIGS. 6

,

8

and

9

A-

9

B. That circuit

146

includes the detection (OD) circuit

148

of

FIG. 8

, which detects the arcing fault

142

and responsively outputs one or more trigger signals

150

, and the activation circuit (ACT)

152

of

FIGS. 9A-9C

, which detects the one or more trigger signals

150

and responsively outputs the activation signal

154

. The difference, however, is that the shorting switches

502

,

602

employs the respective solenoids

534

,

606

, rather than the charges

60

.

FIG. 13

shows a suitable circuit

641

for driving either of the solenoids

534

,

606

from the output

424

of the one-shot multivibrator

416

of FIG.

9

A.

After the circuit

148

detects the arcing fault

142

and responsively outputs the trigger signal

150

, the multivibrator output

424

provides a pulse of suitable duration (e.g., about 1.2 &mgr;s). In turn, the circuit

642

triggers a thyristor

644

, which causes a capacitor (C)

646

to dump current into the solenoid coil

648

(e.g., coil

540

of

FIG. 10

or coil

636

of FIG.

12

).

Table 2 shows the results of a solenoid circuit analysis to determine the current needed to move the solenoid plunger

650

in less than about 1 ms. These assume that the plunger mass, m, is 0.016 kg; that the length of the core, l

CORE

, is 1.3 cm; and that the diameter of the core, d

CORE

, is 2.64 cm. Although exemplary values are shown, a wide range of values and/or other suitable electro-mechanical mechanisms may be employed.

TABLE 2

SYMBOL

VALUE

UNITS

DESCRIPTION

V

o

300

V

Initial Voltage of Capacitor C

C

3300

&mgr;F

Capacitance of Capacitor C

R

31

M&OHgr;

Resistance of Coil 648

Lc

3.9

&mgr;H

Inductance of Coil 648

Lstray

2

&mgr;H

Stray circuit inductance

&Dgr;t

0.000006

s

Time increment for circuit

analysis

&tgr;

0.0002

s

Time constant (L/R)

&ohgr;

6645.06

rad/s

Frequency

C1

7590.37

A

Coefficient = V

o

/&ohgr;L

N

20

#

Turns of Coil 648

D

COIL

2

cm

Diameter of Coil 648

L

COIL

4

cm

Length of Coil 648

&mgr;r

1

#

Permeability of Coil 648

The value of L in Table 2 is defined by Lc+Lstray.

FIGS. 14A-14C

show plots of the solenoid gap (g), force of the plunger

650

, and current of the coil

648

, respectively, for analyzing the motion of the solenoids

534

,

606

.

The exemplary shorting switches

2

,

101

,

502

,

602

employ relatively simple and low cost approaches in providing an energy storage and release mechanism. The compression springs

40

,

520

,

604

provide a suitable closing and holding force on the vacuum interrupter (VI)

4

or a modified VI (e.g., no shield; no contacts or modified contacts) (not shown) for use with medium voltage power distribution equipment, such as switchgear.

While specific embodiments of the invention have been described in detail, it will be appreciated by those skilled in the art that various modifications and alternatives to those details could be developed in light of the overall teachings of the disclosure. Accordingly, the particular arrangements disclosed are meant to be illustrative only and not limiting as to the scope of the invention which is to be given the full breadth of the claims appended and any and all equivalents thereof.

QQ群二维码
意见反馈