Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
ELECTRONIC SHOCK PROTECTION IN CODSETS AND ELECTRICAL DISTRIBUTION SYSTEMS
Document Type and Number:
WIPO Patent Application WO/1998/024161
Kind Code:
A1
Abstract:
An apparatus and method for the avoidance of electrical shock in an electrical system. For a two wire appliance, the device consists of a current interrupting circuit that bidirectionally impedes current flow for short time intervals in each half cycle. If a fault to ground (24, 25) occurs during these short time intervals, an increased current flow through the plug (13) is detected and this is recognized as a fault, causing a circuit interrupter (18, 19) to open and removing current from the load (20) during the remainder of the half cycle. When the fault is removed, that event is detected within one half cycle and power is restored to the load (20). The apparatus can provide thermal control in the plug (13). The apparatus enables the use of a low current, low voltage switch (41) to control high voltages and high currents in the appliance. The apparatus can transmit fault status information or other information to a remotely located controller and can receive control signals from that remotely located controller. As used for electrical arc fault protection in an electrical distribution system, the invention is well suited for the retrofit of existing electrical distribution systems using the existing wiring. Additional features include the ability to detect an open neutral condition or a miswired outlet.

Inventors:
NEMIR DAVID C (US)
HIRSH STANLEY S
Application Number:
PCT/US1997/022529
Publication Date:
June 04, 1998
Filing Date:
November 21, 1997
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
NEMIR DAVID C (US)
International Classes:
H02H11/00; H02J13/00; H02H3/00; H02H3/04; H02H3/05; H02H3/16; H02H5/04; (IPC1-7): H02H3/00; H02H3/14; H02H3/20; H02H3/22; H02H3/24; H02H3/16; H02H3/26; H02H3/32; H02H3/38; H02H5/04; H02H7/00; H02H9/00; H02H9/02
Foreign References:
US3997818A1976-12-14
US4649454A1987-03-10
Attorney, Agent or Firm:
Myers, Jeffrey D. (Myers & Adams P.O. Box 2692, Albuquerque NM, US)
Download PDF:
Claims:
CLAIMSWhat is claimed is:
1. An electrical shock protection apparatus for an alternating current electrical appliance comprising an electrical load, a plug, and two electrical current carrying conductors connecting said electrical load to said plug; said electrical shock protection apparatus comprising: means for bidirectionally blocking electrical current flow through said two electrical current carrying conductors to said electrical load for predetermined time intervals; means for detecting electrical current flow in said two electrical current carrying conductors; and means for interrupting electrical current flow in said two electrical current conductors if an appreciable current in one of said two conductors is detected during said predetermined time intervals, said appreciable current being indicative of a potentially dangerous electrical fault; and optionally means for said shock protection apparatus to communicate to a remote controller information selected from the group consisting of presence of a fault condition, absence of a fault condition, appliance identification information, and appliance status information.
2. The electrical shock protection apparatus of Claim 1 wherein said bidirectionally blocking means, said electrical current flow detecting means and said electrical current flow interrupting means are all disposed within a plug.
3. The electrical shock protection apparatus of Claim 1 wherein said bidirectionally blocking means permits said detecting means to detect a fault selected from the group consisting of load to ground faults, hot to ground faults, and neutral to ground faults.
4. The electrical shock protection apparatus of Claim 1 wherein: while a fault is absent, a charge storage capacitor is charged to a voltage sufficient to trigger a first switch which allows flow of electrical current to said electrical load; and while a fault is present, said charge storage capacitor is forced to be in a discharged condition, thus preventing triggering of said first switch and preventing flow of electrical current to said electrical load.
5. The electrical shock protection apparatus of Claim 1 wherein no modification to said electrical load is required.
6. An electrical shock and arc fault protection device for an alternating current electrical distribution system comprising a load center, at least one load, and at least two conductors connecting said load center with said at least one load; said apparatus comprising: means for bidirectionally blocking substantially all of the electrical current flow to the one or more loads for predetermined time intervals; means for detecting electrical current in one of said at least two conductors; and means for interrupting said electrical current in said at least two conductors if an appreciable current in one of said at least two conductors is detected during one of said predetermined time intervals.
7. The electrical shock and arc fault protection apparatus of Claim 6 wherein said bidirectionally blocking means is configured as a two terminal circuit element that is inserted in a light socket.
8. The electrical shock and arc fault protection apparatus of Claim 6 wherein the bidirectionally blocking means is disposed within a wall outlet.
9. The electrical shock and arc fault protection apparatus of Claim 6 wherein said at least one load is an appliance that plugs into a wall outlet.
10. The electrical shock and arc fault protection apparatus of Claim 6 wherein if a given conductor is designed to carry electrical current during normal operation and is designed to be electrically grounded but is not electrically grounded, the electrical current interrupting means is triggered.
Description:
ELECTRONIC SHOCK PROTECTION IN CORDSETS AND ELECTRICAL DISTRIBUTION SYSTEMS BACKGROUND OF THE INVENTION 1. FIELD OF THE INVENTION This invention relates to shock protection circuits for electrical appliances and more particularly, to an improved protection device that can detect and eliminate ground faults, has an automatic reset upon the removal of the fault, and can also serve as a means for bidirectional communication and control between the appliance and a remote computer.

Furthermore, this invention relates to electrical shock and electrical arc fault protection circuits for electrical distribution systems and more particularly, to an improved protection system that can detect and eliminate many shock hazards and arcing faults within an electrical distribution system and which automatically resets itself upon the removal of the shock hazard and/or arcing fault.

2. BACKGROUND OF THE INVENTION Any electrical device ("the load") requires the flow of electrical current in order to operate. An analogy is the flow of water through an aquarium filter. A pump takes in water from the aquarium and increases the pressure (analogous to an increase in electrical voltage) to force the water through a tube (the tube is analogous to the electrical conductor or wire) to the filter (analogous to the electrical load). The flow of water current through the tube is analogous to the flow of electrical current in a wire. Most of the water pressure is "used up" in passing through the filter so that the water coming out of the filter has a relatively low pressure. A hose conveys the low pressure water from the filter outlet back to the aquarium. If the tube connecting the pump and the filter has a hole then some of the water will pass through this hole from the high pressure in the tube to a lower pressure outside the tube. This constitutes a water leak.

In somewhat the same way, an electrical device or load receives electrical energy from one terminal of an electrical outlet or source (the so-called high voltage or "hot" side), electrical current flows to the device through an electrical conductor or wire (the hot conductor), this current passes through the load and is then returned to another terminal of the electrical outlet through another wire called the neutral wire. The neutral wire will have a very low voltage (electrical pressure) because most of the voltage will be "used up" in the act of forcing electrical current through the load.

The two wires that connect source and load may have a coating of rubber or some other electrical insulating material or they may be bare, in which case air, which is a good insulator, functions to inhibit electrical current flow outside of the wire. Since the human body can conduct the flow of electrical current, if a person comes into contact with one electrified object such as the hot conductor, while also making contact with a second object having a substantially different voltage, then an electrical leakage current that is proportional to the voltage difference will flow through the person and may cause injury or death. If the second object that the person comes in contact with is electrically connected to the earth (ground) then this is called a ground fault. If the person makes contact between the neutral wire and ground this is called a neutral to ground fault. Since the voltage difference between neutral and ground is generally small (because the neutral line is connected to ground at a breaker box), a neutral to ground fault is generally considered to be less hazardous than a hot to ground or a hot to neutral fault. However, if the neutral line is not connected to ground at a breaker box, or if the neutral and hot lines are misconnected, a neutral to ground fault can be as dangerous as a hot to ground fault.

Electrical current is the flow of electrons. Since electrons are not created or destroyed, any functioning electrical appliance will require both an entry path and an exit path in order for electrical current to flow. In an electrical appliance, electrons may exclusively enter on one path and exit on a second (direct current or DC operation) . This is analogous to the aquarium leak example. For most household appliances that operate from a plug, electrons will sometimes enter path one and exit path two and sometimes enter path two and exit path one. This is known as alternating current or AC operation. A special case of AC operation is the so-called half wave AC (also known as pulsating DC) whereby over regularly repeating periods of time (or cycles), electrical current will flow in one direction for a portion of the cycle and will be blocked from flowing for another portion of the cycle. Although the direction of current flow in half wave AC operation is unidirectional, it is considered to be alternating current since the magnitude of the electrical current varies in a cyclical fashion.

Although the two conductors coming out of an electrical circuit are called "hot" and "neutral", in an AC system, the hot conductor will cyclically have a more positive voltage than the neutral for half the time and will have a more negative voltage than neutral for half the time, having a momentary value of zero as the voltage passes from positive to negative and negative to positive.

A common source of electrical injuries in the home occurs when people place radios or similar electrical devices that are operated using household AC electrical current near their pool or bath tub while swimming or bathing. If the radio is knocked into the water, it can create electrical leakage current through the water to ground. A ground fault can also occur when a person touches an electrically hot conductor while standing on or touching a grounded conductive surface. When sufficient current passes through a person, electrical burns or electrocution may result. Many electrical appliances such as heaters, hair dryers, electric razors and pumps are used near water and present this type of hazard.

A device called an immersion detection circuit interrupter (IDCI) detects the occurrence of electrical leakage due to water immersion and opens a circuit breaker, indirectly providing protection against a potential ground fault that would occur, for example, by dropping an IDCI equipped hair dryer into a bathtub. Such immersion protection devices are described in U.S. Patent Nos. 5,159,517 (Bodkin) 4,797,772 (Kaplanis) and 5,184,271 (Doyle and Rivera). They do not directly provide ground fault protection as they require immersion in water to operate.

Protection circuits for ground faults, commonly known as ground fault interrupters or GFI's, are presently required by code for the bathrooms of most new homes and commercial buildings. Such circuits are also required for Underwriter's Laboratories approval in hair dryers and are built into the plug of U.L. approved hair dryers. Of the GFI circuits presently available on the market, all use a current imbalance in a current sense transformer as the means of detecting a fault. Such circuits are described, for example, in U.S. Patent Nos. 4,216,515 (Van Zeeland), 4,353,103 (Whitlow), 4,979,070 (Bodkin) and 5,200,873 (Glennon).

U.S. Patent No. 3,997,818, to Bodkin, discloses a load selective power system whereby a load is made into a "proper load" by including a unidirectional current device in series with the load. Full wave AC power may be supplied to the load with two such current devices and three terminals. A power monitoring circuit indicates certain faults in the supply circuit. The disadvantage to this approach for AC loads is that three wires are required to connect source to load, it will not detect a fault from within the load to ground, and it requires a modification to the load to make it "proper".

A common source of electrical injuries in the home occurs when people place radios or similar electrical devices that are operated using household AC electrical current near their pool or bath tub while swimming or bathing. If the radio is knocked into the water, it can create electrical leakage current through the water to ground creating a ground

fault. A ground fault can also occur when a person touches an electrically hot conductor while standing on or touching a grounded conductive surface.

When sufficient current passes through a person, electrical burns or electrocution may result. Many electrical appliances such as heaters, hair dryers, electric razors and pumps are used near water and can present this type of hazard. Even a relatively low level of electrical current leakage can be hazardous to a human. Underwriters Laboratories, in their 943 standard for ground fault interrupt devices, requires that listed devices must open in response to any leakage current exceeding six milliamperes.

Another type of undesirable operating condition occurs when an electrical spark jumps between two conductors or from one conductor to ground. This spark represents an electrical discharge through the air and is objectionable because heat is produced as a byproduct of this unintentional "arcing" path. This occurrence is known as an arc fault and is a leading cause of, and contributor to, electrical fires. The electrical current levels drawn by an arcing fault are generally several orders of magnitude higher than the six milliampere standard for the Underwriters Laboratories 943 specification.

The U.S. Consumer Products Safety Commission estimates that there were approximately 41,000 fires involving electrical wiring systems in 1992. These fires resulted in 320 deaths, 1600 injuries and $511 million in property losses. The CPSC studies also show that the occurrence of wiring system fires is disproportionately high in homes more than 40 years old [Source: Technology for Detecting and Monitoring Conditions That Could Cause Electrical Wiring System Fires, Contract Number CPSC-C-94-1112, prepared by Underwriters Laboratories, September 1995]. The disproportionately high incidence of wiring system fires in older homes may be attributed to poor wiring techniques, outdated electrical components and aged or damaged insulation on conductors. The present invention could be applied as a retrofit to older homes, affording protection against the arcing faults that are a major source of wiring system fires.

Arcing faults can occur in the same places that ground faults can occur -- in fact, a ground fault would be called an arcing fault if it resulted in an electrical discharge across an air gap (a spark). As such, circuits that protect against ground faults can also prevent many classes of arcing faults. However, many of the circuits that purport to protect against arcing faults cannot detect low (albeit hazardous) level ground faults.

Protection circuits known as ground fault interrupters or GFI's, are presently required by code for the bathrooms of most new homes and commercial buildings. Similar circuits are also required for Underwriter's

Laboratories approval in hair dryers sold in the United States. Of the GFI circuits presently available on the market, all use a current imbalance in a current sense transformer as the means of detecting a fault. Such circuits are described, for example, in U.S. Patent No. 3,683,302 (Butler et al) and U.S. Patent No. 4,216,515 (Van Zeeland). One problem with approaches based upon a current sense transformer is that the magnetic core can acquire a residual magnetic state subsequent to one or more differential current conditions. This can result in a circuit which is too sensitive and is subject to nuisance tripping. In order to avoid nuisance trip problems, Underwriters Laboratories, in their 943 standard, requires that ground fault interrupt circuits not trip in response to a fault current of less than 4 milliamperes, thereby reducing the incidences of nuisance trippings. While existing ground fault interrupter circuits can detect and interrupt an arcing fault from the hot conductor to ground, they cannot detect and prevent an arcing fault that occurs between a hot conductor and a neutral conductor, a source of many electrical fires within homes and businesses. The present invention can detect such an occurrence within the household wiring. In addition, the present invention is an improvement over the current sense transformer based approaches in that it does not require the transformer and, since it is has a fast autoreset, nuisance tripping is not as great a concern and the device can be made responsive to a very low level of fault condition.

A spin-off of the conventional GFI circuit that has been applied to arc fault detection/avoidance in an appliance cordset is described in U.S.

Patent No. 4,931,894 (Legatti). This device works by using a conductive metal sheath to surround each current-carrying conductor individually. An arc protection winding is located on the core of the GFI current sense transformer and is connected in series with a resistance between the metal sheath and a neutral or return line. A fault involving any current carrying conductor will involve the shield before it involves any other conductive surface and the fault current will be sensed as a ground fault, thereby tripping the interrupting device in the plug. Problems with this approach, over and above the problems inherent in any current sense transformer based approach, is that this technology (U.S. Patent No. 4,931,894) requires special wires that have a grounded conductive shield and this prevents its use in a system where it is desirable to use existing wiring.

The late 1980's and early 1990's saw a great deal of effort directed at arc fault detection/protection circuits. For example, U.S. Patent No. 5,224,006 (Mackenzie and Engel) describes a system whereby the magnitude and rate of change of current is monitored. If the rate of

change of current has a profile characteristic of a sputtering are fault, a circuit breaker relay is tripped. U.S. Patent 4,878,144 (Garin) uses a light sensitive are detector to detect an arcing phenomenon and then trips a circuit breaker. U.S. Patent 4,658,322 (Rivera) and U.S. Patent No. 4,903,162 (Kopelman) use heat sensing elements to detect an overtemperature condition, such as that occurring due to an arcing fault, in electrical wiring and trigger a current interrupting circuit breaker in response thereto. U.S. Patent 4,848,054 (Franklin) discloses a protective circuit that trips a circuit breaker upon the detection of an overload current condition which exceeds the maximum expected during normal transient conditions of operation, such overload current condition said to be characteristic of an arcing fault. The problems with all of the above cited arc fault protection technologies is that they cannot detect a low level (non-arcing) fault current which, although of relatively small value, can still result in a painful or even lethal electrical shock.

Additional arc fault detection circuits have been proposed that look for a specific signature characteristic of the current, voltage or electromagnetic field associated with arcing faults. These technologies concentrate on detecting a specific signature characteristic because many electric devices produce arcing during normal operations. An example is an electrical light switch which may draw a spark (an arc) when opened or an electric igniter for a gas furnace. Another example of a device that arcs in normal operation is a commutated motor which will spark continuously at the brushes when energized. An arc fault detector/interrupter would be useless if it detected and tripped in response to all arcs, both good (corresponding to normal operation( and bad (corresponding to a fault condition). Examples of technologies wherein the current flow is monitored, filtered and processed to detect an arcing fault include U.S.

Patent No. 4,639,817 (Cooper and South) wherein voltage signals are measured between the three phases in an AC three phase network and bandpass limited to the frequency band between 1 and 100 kilohertz, the band which is alleged to contain harmonic frequencies indicative of an arcing fault.

U.S. Patents 5,047,724 (Boksinger and Parente) and U.S. Patent No. and U.S.

Patent No. 5,280,404 (Ragsdale) reveal methods for detecting arc faults in an electrical circuit by comparing the spectral frequency makeup of voltages and/or currents within the circuit conductors to the spectrum characteristic of an arcing event. Examples of technologies wherein an electromagnetic field is monitored, filtered and processed to detect an arcing fault include U.S. Patent 5,185,684 (Beihoff, Tennies, Richards and O'Neil), U.S. Patent 5,185,685 (Tennies, Beihoff, Hastings, Clarey and O'Neil), U.S. Patent 5,185,686 (Hansen, Beihoff, Tennies and Richards),

U.S. Patent 5,185,687 (Beihoff, Tennies, Richards and O'Neil) and U.S.

Patent No. 5,208,542 (Tennies, Beihoff, Hansen). The present invention is an improvement over the above referenced inventions in that the above inventions are directed at detecting an arcing condition but would be unresponsive to a low level fault current which, although not resulting in an arcing fault, could deliver a painful or lethal electrical shock if the fault path was a human body. Furthermore, most of the filtering algorithms proposed by these arc fault directed inventions require a signal analysis over multiple cycles and cannot detect and respond to a fault in less than one cycle of the fundamental AC source frequency.

3. OBJECTS AND ADVANTAGES The present invention has the following objects and advantages.

First, when implemented in an appliance cordset, the invention: a) requires only a two wire electrical cord connecting the appliance to the plug; b) does not require a current sense transformer; c) interrupts power to the appliance in less than one half cycle from the occurrence of the fault; d) electrical power is automatically restored to the device upon removal of the fault (auto-reset); e) can detect and respond to a hot to ground fault; f) can detect and respond to a hot to AC neutral fault; g) can detect and respond to an appliance neutral to AC neutral fault; h) can detect and respond to an appliance neutral to ground fault; i) can detect and prevent an overtemperature condition at the plug; j) its all solid state design (no current sense transformer, no moving parts) allows it to be built into a small, conventionally sized plug; k) can be used to control power to the appliance using a low current, low voltage switch located at the plug; 1) appliance can be turned on or off in response to a command signal broadcast over the household wiring; m) the appliance can identify itself or pass status information to a central controller via the household wiring; n) the appliance can use a simple two pronged non-polarized plug; and o) no modification of the load is required.

Second, as implemented in an electrical distribution system, the invention: a) requires only two electrical conductors connecting the parts of an electrical distribution system (no ground wire required) b) does not use a current sense transformer for fault detection;

c) can detect and interrupt arcing faults from the hot conductor to ground; d) can detect and interrupt shock hazards from the hot conductor to ground; e) can detect and interrupt an arcing fault from the hot conductor to the neutral conductor within the wiring connecting the load center and the outlets/lights in an electrical distribution system; f) interrupts current flow within one half cycle from the occurrence of a fault; g) electrical power is automatically restored to the device upon removal of the fault (auto-reset); h) can be easily retrofit into existing houses or other installations without requiring a restringing of existing wiring within that house or other installation; i) can be used to detect and prevent an open ground condition or miswired neutral at wall outlets within a home; j) can be made to provide short circuit current limiting and protection.

Further objects and advantages of the present invention will become apparent from a description of the drawings and ensuing description.

SUMMARY OF THE INVENTION As applied to an appliance cordset, the present invention comprises a circuit interrupter for use in an alternating current (AC) electrical system having one hot and one neutral connection (a two wire appliance).

Although additional wires could be present, for example, adding a ground wire to make a three wire cord, in most applications extra wires would be superfluous and would add unwanted additional cost. It is the principle objective of the invention to provide protection against a ground fault anywhere in the load or in the conductors connecting the plug and the load.

It is another objective to provide protection against an overtemperature condition in the plug. It is another objective to provide a low current, low voltage switch in the plug that can control power to the load. It is another objective to provide a means to communicate bidirectionally from the appliance to a controlling device via the household wiring and to thereby communicate status information and control signals.

The objectives of the cordset embodiment are obtained by using a fault sense/current interrupt circuit at the plug. The plug receives a sinusoidal alternating current (AC current) from a power delivery source or electrical outlet. Under normal operation, the interrupt circuit prevents all but a small sense current from flowing in the load during a short time interval around the zero crossing of the applied AC current cycle. If,

during this time interval when the interrupt circuit is blocking load current, a ground fault occurs, electrical current will flow at the plug due to the sensing current flowing through the fault. This will be detected by the fault sense circuitry in the plug, which will maintain the interrupt circuit in an open condition, thus preventing appliance operation and preventing dangerous levels of electrical current from being applied to the fault. In each subsequent zero crossing of the AC cycle, the appliance and cord will be tested briefly to establish if a fault condition is still present. If a fault condition is still present, the current interrupt remains open. If the fault condition is no longer present, the current interrupt is closed and full electrical energy is delivered to the appliance. In this way, the device is self resetting.

In addition to fault protection, this invention will also provide temperature control at the plug. This is done by using a temperature dependent resistance or other thermally sensitive element in the plug in such a way that when it reaches a sufficiently elevated temperature, it causes the circuit interrupter mechanism at the plug to open, preventing load current from flowing through the plug and thereby allowing the plug to cool for a half cycle or one or more full cycles. If on subsequent cycles, the thermally sensitive element remains too hot, the circuit interrupter remains open. This feature is important in order to keep the current interrupter from being damaged in the case of a persisting overcurrent condition. When the overtemperature condition is ended, and assuming that no fault exists, electrical current is allowed to flow to the load.

An additional feature offered by this invention is the ability to control possibly high voltage, high current power to the appliance by using a low voltage, low current switch in the plug in an appliance cordset.

Additional features include the ability to control the appliance under remote command from a controller that communicates over the household wiring. This is a feature that is important in implementing so-called "smart appliances". Smart appliances are appliances that can be controlled either internally or remotely without direct human intervention. A related idea is the so-called "Smart House" which is the prototypical home of the future. A number of protocols for the smart house have been established by U.S. and foreign manufacturers including the so-called Consumer Electronics Bus or CEBus. The smart house has a central computer that manages the functioning of all household appliances via a modified wall socket. In order to manage a given appliance, however, this central computer has to be able to identify what appliance is plugged into which socket. At a minimum, smart appliances will need the ability to identify themselves to the smart house via the household wiring. This is easily implemented as a

byproduct of the circuit operation of the present invention. If a ground fault interrupt circuit attached to an appliance has detected a fault condition, this represents an event that might be important to communicate back to a central controller. In addition, other types of status information could be transmitted to a remotely located controller over the, for example, household wiring. In addition, control signals could be passed to the appliance over the household wiring. An ON/OFF function, for example, could be easily implemented by adding or removing a simulated fault condition within the plug, thereby shutting off or turning on appliance power. The present invention incorporates the ability for this type of communication and control.

The present invention is unique in that it uses an all solid state (electronic) design to do a job that has been traditionally done by mechanical relays and differential sense transformers. As such, it has the potential to be implemented almost entirely on a single integrated circuit.

Silicon, the building block of integrated circuits, is cheap, and features such as control functions and signal conditioning can be added in silicon at little additional cost. This is particularly important from the standpoint of the smart appliance. Not enough smart houses are in existence in 1996 to justify a major investment by appliance manufacturers to make their appliances "smart house ready". However, with the availability of solid state fault protection where bidirectional communication and control features can be added at little additional cost, it becomes economically feasible to invest in solid state fault protection for the near term, while building in features that will make the appliances smart-house-ready in the future.

As applied to an electrical distribution system, the present invention comprises an electrical system for arc fault and shock protection. It is the principal objective of the invention to provide a device which can be used to retrofit existing distribution systems using the existing wiring in that installation. It is another objective to provide short circuit protection. It is another objective to provide fault protection in conjunction with light dimmer circuits or motor speed controls.

When applied to an electrical distribution system, the objectives of this invention are obtained by using a fault sense/current interrupt circuit at an electrical load center (breaker box) within a home, office or building, together with a load conditioning circuit located at the outlets and lights that are serviced by said electrical distribution center. Under normal operation, the load conditioning circuit(s) prevent substantially all current from flowing to the light(s) and/or the load(s) at the

outlet(s) during a brief time interval around the zero crossing of the applied AC power cycle. If, during this time interval when the load conditioning circuit is blocking load current, a ground fault or arc fault occurs, electrical current will flow at the distribution center due to current flowing through the fault. This will be detected by the fault sense circuitry at the distribution center, which will force the interrupt circuit to maintain an open condition, thus preventing substantial electrical current from flowing to the electrical outlets and lights that are fed by the distribution center and thereby preventing dangerous levels of electrical current from being applied to the fault during the balance of that half cycle. In each subsequent zero crossing of the AC cycle, the outlet and/or lights, along with the wiring connecting them to the distribution center will be tested briefly to establish if a fault condition is still present. If a fault condition is still present, the current interrupt remains open. If the fault condition is no longer present, the current interrupt is closed and full electrical energy is delivered to the appliance for the balance of that half cycle. In this way, the device is self resetting.

As an alternative, when a fault is detected, the current interrupt at the distribution center may be activated for a number of cycles, effectively preventing current flow for that number of cycles, irrespective of whether the fault condition is still present. This might be useful in that it would allow the circuit interrupt circuitry to cool after the occurrence of a hard short circuit.

An advantage of this type of fault protection for an electrical distribution system is that it can be added to the system without modifying or adding to existing wiring. An additional feature of the invention is that it allows for the incorporation of the detection and prevention of unsafe operating conditions such as when the neutral is not tied to ground or the electrical outlets are miswired.

BRIEF DESCRIPTION OF THE DRAWINGS The above mentioned and other features and objects of this invention and the manner of obtaining them will become more apparent and the invention itself will be better understood by reference to the following description of the embodiments of the invention taken together with the accompanying drawings, wherein: FIG. 1 depicts an aquarium pump/filter system which is analogous to the situation of sourcing electrical current to a load and furthermore depicting various analogs of electrical faults.

FIG. 2 is a block diagram the cordset embodiment of the invention, illustrating the approximate connections of fault sensing circuitry and circuit breakers within the plug/cord/appliance unit.

FIG. 3 is an electrical schematic of the first embodiment of this invention using a two-prong plug and able to detect a fault to ground from the hot side of the line cord, from the load, or from the neutral side of the line cord and to respond appropriately.

FIG. 4 is a sketch of approximately one and one fourth cycles of the AC voltage applied at the plug on the hot prong with respect to the neutral prong.

FIG. 5 is a sketch of the windows wherein all but a small sense current is inhibited from flowing within the load.

FIG. 6 is an electrical schematic of a second embodiment of this invention wherein circuit interruption occurs at both the hot side of the line as well as the neutral side of the line upon the detection of a fault condition.

FIG. 7 is an electrical schematic of a third embodiment of this invention wherein current interruption occurs at both hot and neutral sides of the line and which utilizes optocoupling to control the neutral side circuit interrupter.

FIG. 8 is an electrical schematic of a fourth embodiment of this invention wherein current interruption occurs at both hot and neutral sides of the line and which utilizes a pulse transformer to control the neutral side circuit interrupter.

FIG. 9 is an electrical schematic of a fifth embodiment of this invention which provides fault detection/protection and also incorporates a carrier current transmitter/receiver to either receive commands from a remotely located controller via carrier currents on the household wiring and/or can communicate fault, identification or other status information back to that controller.

FIG. 10 is an electrical schematic of a sixth embodiment of this invention which provides fault detection/protection and also incorporates a carrier current transmitter/receiver in the plug and at the load to transmit status and control information bidirectionally between plug and load and to transmit status and control information bidirectionally over the household wiring between the plug and a remotely located controller.

FIG. 11 is an electrical schematic of a seventh embodiment of this invention which provides fault detection/protection and which further provides an equivalent level of fault protection regardless of the polarity of the plug.

FIG. 12 depicts a block diagram of a representative connection of a building's electrical distribution system and various faults that can be present.

FIG. 13 depicts a block diagram of a building's electrical distribution system with modifications as per the present invention to provide fault protection.

FIG. 14 depicts a block diagram of the theory of operation of the invention as applied to an electrical distribution system.

FIG. 15 illustrates one approach underlying the creation of a load conditioning circuit whereby a dead zone is imposed at all cyclic zero crossings of current.

FIG. 16 depicts the electrical schematic of a distribution system specific embodiment of the load conditioning circuit whereby a bilateral trigger switch is used to trigger a triac during all but the intervals around a zero crossing.

FIG. 17 is a mechanical drawing of a "button" style load conditioning circuit attachment that could be inserted into a light bulb socket to impose a dead zone.

FIG. 18 depicts the electrical schematic of a second distribution system specific embodiment of the load conditioning circuit whereby the load conditioning circuit can serve a dual role of providing the dead zone for fault protection as well as providing a load controlling function.

FIG. 19 illustrates an approach to making a load conditioning circuit which is useful for inductive loads.

FIG. 20 is an electrical schematic of a specific embodiment of a load conditioning circuit that is suitable for inductive loads.

FIG. 21 depicts example control signals that might be generated by the load conditioning circuit controller.

FIG. 22 is an electrical schematic of an embodiment of the fault detect circuitry wherein circuit interruption occurs at the hot side of the line at the load center upon the detection of a fault condition.

FIG. 23 is an electrical schematic of an embodiment similar to that depicted in FIG. 14 except that the circuit also has a short circuit current limiting feature that inhibits large surge currents from occurring prior to the detection of a fault condition.

FIG. 24 is an electrical schematic of a general load conditioning circuit having additionally a circuit that can detect an open neutral or miswiring condition and thereupon generate a fault.

DESCRIPTION OF THE PREFERRED EMBODIMENTS In order to describe the functioning of the present invention, it is helpful to look at the analogous situation of an aquarium pump and filter combination. FIG. 1 depicts a situation where water is pumped from the aquarium 1 through a filter 2 and is then returned to the aquarium 1. The pump 3 takes low pressure water from the aquarium 1 using a hose 4 and increases the pressure so that the water flows in the high pressure hose 5 from the pump 3 to the filter 2 through the pump outlet valve 7 and the flowmeter 6. The pump outlet valve 7 controls the flow of water out of the pump 3. The aquarium inlet valve 8, serves to allow or to block water flow into the aquarium 1. Both valves 7 and 8 have only two possible positions, fully open or fully closed. Leaks 10,11, 12 can occur in a number of places within the system, resulting in water flowing in undesirable paths (faults) instead of through the hoses and filter.

Consider the flowmeter 6 installed at the outlet of the pump. This flowmeter 6 detects if any water is passing from the pump 3 into the high pressure hose 5. If pump outlet valve 7 is in an open position and aquarium inlet valve 8 is in a closed position, then in the absence of leaks 10, 11 and 12 there will be no flow of water out of the pump 3 and the flowmeter 6 will not detect any flow of water. If, however, there are leaks 10 and/or 11 and/or 12 then there will be some water flowing out of the pump 3 into the high pressure hose 5 and this event can be detected by the flowmeter 6. If this occurs, then the leaks 10 and/or 11 and/or 12 may be halted by closing the pump outlet valve 7. In this way, the flowmeter 6 together with the aquarium inlet valve 8 serve to detect the presence of leaks 10 and/or 11 and/or 12. Once a leak is detected, the pump outlet valve 7 can be closed to stop the leaks by interrupting the flow of water to the high pressure hose 5.

The water leakage scenario described above is helpful in understanding the functioning of the present invention for the protection against shocks and burns due to electrical current leakage in an electrical appliance or in an electrical distribution system. The key feature of this invention is that the majority of electrical current in the load is intentionally blocked from flowing during a portion of each half cycle.

During this portion of each cycle when current flow to the load is inhibited except for possibly a low level sensing current, if any electrically conductive path is present between the hot wire and ground, then electrical current will flow out of the source due to this conductive path and this "fault" current may or may not pass through the load. This

represents a fault condition and, after sensing, causes an open circuit to prevent substantial current from flowing out of the plug.

FIG. 2 depicts a block diagram of this approach to fault detection/protection in an electric appliance. The plug 13 connects to the hot 14 and neutral 15 prongs. The hot 14 and neutral 15 prongs are the blades of metal seen on any conventional electrical appliance plug. The plug 13 also connects to the hot side 16 of the line cord and neutral side 17 of the line cord. Switch 18 is located between prong 14 and conductor 16. Switch 19 is located between prong 15 and conductor 17.

These switches 18 and 19 are used to allow (when closed) or interrupt (when open) the current flow to the load 20.

The load 20 represents any conventional electrical appliance load and could be, for example, a hair dryer, an aquarium heater, a curling iron, or a string of lights. A fault 25 from the hot conductor 16 to ground 35 or a fault 24 from the neutral conductor 17 to ground 35 or a fault 55 from within the load 20 to ground 35 consists of an unintentional path for electrical current to travel and represents, for example, a human coming in contact with an exposed wire while standing on a grounded surface. In order to detect a fault, switches 18 and 19 will be opened at the same time with the fault detect/circuit breaker trigger 23 providing a limited current path around switch 18 and switch 19, through which fault current could be sensed. At any time that a fault 24 or 25 or 55' exists, this condition will be detected by the fault detect/circuit breaker trigger 23 which then causes switches 18 and/or 19 to maintain an open position which will be maintained as long as faults 24 and/or 25 and/or 55' exist.

Accordingly, if the neutral to ground fault 24 or hot to ground fault 25 or load to ground fault 55' represents a human body making accidental contact with dangerous voltages then this invention will protect against electrical injury by detecting this as a fault condition and will prevent substantial current from flowing from the plug, thereby implementing shock protection.

The first preferred embodiment is shown in FIG. 3. The power applied to the plug prongs 14 and 15 is sinusoidal alternating (AC) current.

FIG. 4 depicts approximately 1 and 1/4 cycle of this AC voltage at the hot prong 14 relative to the neutral prong 15. The negative half cycle 33 is defined as the time interval during each cycle of AC when the voltage at the hot prong 14 has a value that is negative with respect to the neutral prong 15. For example, FIG. 4 depicts a waveform corresponding to 60 cycles per second (a period of 16.66 milliseconds). In FIG. 4, the

first 8.33 milliseconds corresponds to the positive half cycle 34 while the interval going from 8.33 milliseconds to 16.66 milliseconds corresponds to a negative half cycle 33. This then repeats.

In the absence of a fault, during the negative half cycle 33 a charge storage capacitor 26 in the plug 13 is positively charged through resistor 27. When the voltage on capacitor 26 exceeds the diac 28 turn-on voltage, the diac 28 fires, triggering hot side triac 29. Although FIG. 3 shows a diac 28, it could be replaced by an equivalent element that serves to block current flow through the element until a voltage threshold or time interval is reached and then conducts current when that voltage threshold or time interval is exceeded. Hot side triac 29 then turns on and allows electrical current to flow through the plug to the load 20 through the hot conductor 16. In the absence of a fault, during the positive half cycle 34 the charge storage capacitor 26 in the plug 13 is negatively charged through resistor 27. When the voltage on this capacitor exceeds in magnitude the diac 28 turn-on voltage, the diac 28 fires, triggering hot side triac 29. Hot side triac 29 then turns on and allows electrical current to flow in the plug 13 to the load 20.

In this embodiment, the neutral side switching is handled by triac 31 and diac 30. During the portion of the AC cycle in which the magnitude of the voltage between conductor 17 and the neutral prong 15 is less than the diac 30 trigger voltage, the neutral side triac 31 will be inhibited from conducting and no electrical current will be delivered to the load 20 through the neutral side triac 31. FIG. 5 depicts the electrical current delivered to the load 20 in the absence of a fault. The intervals 32 are the times during which the current is blocked from flowing in the load 20 due to triacs 29 and 31 being unfired and representing open switches. This bidirectional blocking is a key element of the present invention. During these intervals, a small current may still flow out of the plug through the series connection of resistors 36 and 37 if a fault 24 and/or 25 is present. Resistors 36 and 37 would typically be chosen to be of high value to minimize this sense current. If fault 24 and/or 25 and/or 55' is present during time interval 32 at the beginning of the negative half cycle, transistor 38 in the plug 13 will conduct through its collector diode 40 due to base current generated by the fault current flowing through a resistor divider 36,37. Transistor 38 will discharge the capacitor 26 preventing diac 28 from firing the triac 29. Hot side triac 29 will thus remain off for the rest of the negative half cycle 33. During the time

interval 32, prior to the positive half cycle 34 of the input power, if fault 24 and/or 25 is present, transistor 39 will conduct through its collector diode 40 due to base current generated by the fault current flowing through resistor divider 36,37. Transistor 39 will discharge capacitor 26, preventing diac 28 from firing. Hot side triac 29 will thus remain off for the remainder of the positive half cycle. Because hot side triac 29 is prevented from firing, current flow through the plug is prevented. Whenever a fault is detected, whether at the beginning of a positive half cycle 34 or a negative half cycle 33, triac 29 functions as a current interrupter. After interval 32 is over, and during the remainder of the positive half cycle, the current flowing through the fault resistances 24 and/or 25 will keep either transistor 38 or 39 in a conducting mode thus preventing capacitor 26 from charging enough to fire diac 28 to subsequently trigger triac 29.

When neither fault 24 nor fault 25 is present, the capacitor 26 will be allowed to charge through resistor 27. By doing so, the capacitor 26 and diac 28 will enable the power triac 29 to conduct current to the load 20 on subsequent half cycles, in this way serving to automatically reset the load within one half cycle from the removal of the fault. The neutral triac 31 and diac 30 serve to remain open during the time intervals 32 around zero crossing. In this way they serve to distinguish the legitimate load 20 from a fault load 24 and/or 25.

To summarize, if no fault is present the load will receive power during both the positive and the negative half cycles. At any time that a fault 24 and/or 25 occurs, it will be recognized at the beginning of the next occurring half cycle during time interval 32 and all power (except for perhaps a low level test current) will be shut off to the appliance for the remainder of that half cycle. Accordingly, fault detection occurs within one half cycle of the occurrence of a fault and power is removed from the load and the faults within one half cycle of the occurrence of the fault.

This fast fault detection/protection and equally fast reset upon fault removal is an important feature of the present invention.

It should be recognized that a leakage path 21 to ground from within the load would be detected in an identical manner to faults 24 and 25.

This might occur, for example, if the load was a lightbulb with a broken glass bulb and a person touched the exposed, electrically live filament while standing on a grounded surface.

In FIG. 3 an optional temperature dependent element 44 may be used to shut off current flow through the plug if operating temperatures in the

plug become excessive. The temperature dependent element could be, for example, a negative temperature coefficient resistor (a so-called NTC thermistor) that decreases in resistance as the temperature is increased.

Under nominal operating temperatures the temperature dependent element 44 has a sufficiently high resistance that negligible current flows through it from capacitor 26 to the bases of transistors 38 and 39 and so element 44 has no impact on the circuit operation. If, however, element 44 has a reduced resistance due to a high temperature then the current flow from capacitor 26 to the bases of the transistors 38 and 39 is appreciable, causing one of the transistors to conduct and thereby discharging capacitor 26 and preventing it from firing diac 28 to turn on triac 29.

The effect would be to turn off power through the plug 13 whenever the plug 13 gets too hot, thereby allowing the plug to cool.

For some applications it might be important to turn off power at the plug when the plug became too cool, in which case temperature dependent element 44 could be a so-called positive temperature coefficient (PTC) thermister that increases in resistance as temperature is increased and would function in an inverse manner to the NTC thermistor in that appliance power delivery would be inhibited for cool temperatures.

In FIG. 3 an optional switch 41 may be used to implement a low power ON/OFF switch at the plug. When the switch is closed, it furnishes current to the bases of the transistors 38 and 39, causing them to discharge capacitor 26 and thereby causing the hot side triac 29 to interrupt current flow to the load 20. The advantage to controlling appliance power in this way is that switch 41 can have a current and voltage rating much lower than the rating of the load.

A second preferred embodiment is shown in FIG. 6. It functions in a similar manner to the first preferred embodiment with the primary difference being that the neutral triac 31 is now turned on at the same time as the hot triac 29. In this way, both triacs 29 and 31 function to block the current during time intervals when checking for a fault and both triacs function to inhibit current flow to the load and faults when a fault is detected.

At the beginning of every half cycle, hot triac 29 and neutral triac 31 are open and in the absence of a fault, the only path for electrical current to flow out of the plug is from the hot prong 14 through resistors 36 and 37, through the load 20 and through temperature sensing element 44. Under normal operating conditions (no fault), the temperature sensing element 44 will have a relatively high electrical impedance and very little current will flow out of the plug. During the negative half

cycle (33 on FIG. 5) diac 28 blocks the flow of current until the voltage at node 46 at one end of diac 28 exceeds a diac threshold voltage. When this occurs, the diac 28 fires, allowing current to pass through it, and turning on the NPN transistor 43 which then turns on the hot triac 29 and the neutral side triac 31. This neutral side triac is turned on due to current provided through the gate of triac 29 via resistor 45 and steering diodes 40. Current can then flow to the load for the remainder of the negative half cycle. In a similar way, when no fault is present, during the positive half cycle, diac 28 blocks current flow until the magnitude of the voltage at node 46 exceeds the diac turn-on voltage. When that happens, the diac conducts, turning on the PNP transistor 42 which turns on the hot side triac 29 and the neutral side triac 31.

At the beginning of the negative half cycle 33, if faults 24 and/or 25 of sufficiently low resistance are present, a significant leakage current will flow from ground 35 through fault 24 in series with load 20 and/or through fault 25, then through resistor 36 and 37 to the hot prong 14. If the fault is sufficiently severe (that is, sufficiently low in value), enough electrical current will flow so that the voltage developed across resistor 37 will exceed the turn-on voltage of the NPN Darlington transistor 38, causing it to turn on and effectively causing the diac node 46 to have a voltage magnitude lower than the diac 28 turn on voltage with respect to the hot prong 14. As a result, diac 28 will not fire and will thus not allow triacs 29 and 31 to turn on.

At the beginning of the positive half cycle 34, if faults 24 and/or 25 of sufficiently low resistance are present, a significant leakage current will flow from the hot prong 14 through resistors 37 and 36 and through the the load 20 and fault 24 to ground 35, and/or through the fault 25 to ground 35. When the voltage across resistor 37 exceeds the turn-on voltage of the PNP Darlington transistor 39, the PNP Darlington will turn on, causing the diac node 46 to have a very low magnitude voltage with respect to the hot prong 14. As a result, diac 28 will not fire and will thus not allow triacs 29 and 31 to turn on.

If no fault is present the load will receive power during both the positive and the negative half cycles. At any time that a fault 24 and/or 25 occurs, it will be recognized at the beginning of the next occurring half cycle and load power will remain off for that half cycle.

Accordingly, fault detection occurs within one half cycle of the occurrence of a fault and power is removed from the load and the faults within one half cycle of the occurrence of the fault. This fast fault

detection/protection is an important feature of the present invention. The relatively small fault sense current is not controlled by triacs 29 and 31 but will be limited by resistors 36 and 37 to a small, safe value.

In FIG. 6 an optional temperature dependent element 44 may be used to shut off current flow through the plug if operating temperatures become excessive. The temperature dependent element could be, for example, a negative temperature coefficient resistor (a so-called NTC thermistor) that decreases in resistance as the temperature is increased. Under nominal operating temperatures the temperature dependent element 44 has a sufficiently high resistance that negligible current flows through it and so element 44 has no impact on the circuit operation. If, however, element 44 has a reduced resistance due to a high temperature then during the time interval 32 when triacs 29 and 31 are turned off, there will be a significant current flow through the sensing element 44 and to the load 20 and resistors 36 and 37 to the hot prong 14. The effect will be the same as that of a fault current. Either transistor 38 or 39 will be turned on (depending on which half cycle), inhibiting diac 28 from firing and preventing triacs 29 and 31 from being turned on during the remainder of the half cycle. The plug then is allowed to cool for the remainder of the half cycle.

In FIG. 6 an optional switch 41 may be used to implement a low power ON/OFF switch at the plug. When the switch is closed, it prevents diac 28 from ever receiving a sufficiently high magnitude voltage at node 46 to turn on, forcing triacs 29 and 31 to remain off. The advantage to controlling appliance power in this way is that switch 41 can have a current and voltage rating much lower than the current and voltage rating of the load itself.

A third specific embodiment is shown in FIG. 7 wherein a photo triac driver 47 is used to control the neutral triac 31 simultaneously with the hot triac 29. The functioning of the circuit is almost identical to that of the second specific embodiment (FIG. 6) in the detection of a fault condition and the conditions under which diac 28 is fired or is prevented from firing. In the third specific embodiment, when diac 28 is fired, it provides gate current to the hot triac 29 through the AC controlled photo triac driver 47 and thereby turns on triac 29. At the same time, by means of an optical signal transmitted from the light emitting diode side of the photo triac driver 47 to the light activated triac side. The light

activated triac side of the photo triac driver 47 triggers the neutral triac 31.

A fourth specific embodiment is shown in FIG. 8. This circuit uses a pulse transformer 48 to trigger the neutral triac 31 at the same time that the hot triac 29 is triggered. The pulse transformer 48 is triggered by a pulse of current from capacitor 26 through diac 28 and the gate of triac 29 whenever current is to be allowed to flow out of the plug. Fault sensing is accomplished as with earlier described embodiments. This embodiment depicts a neon indicator light 64 which together with a limiting resistor 63 is in parallel across the hot side triac 29. During the time intervals 32 when the triac is turned off, or at any time when the triac 29 is turned on, there is an insufficient voltage across the triac to turn on the neon 64. If, however, triac 29 is turned off over one or more full half cycles, a sufficient voltage is built up across it to turn on the neon light, thus giving a visual indication of a fault condition. Current limiting resistor 63 is a very high value resistance that would limit to safe levels the amount of current that could flow through the neon in a path around triac 29 in the case of a fault condition.

A fifth specific embodiment is shown in FIG. 9. This embodiment contains the features of the third embodiment except that a carrier current transmitter has been added with direct electrical connection to the hot prong 14, a direct electrical connection to node 46 and a capacitive coupling to the neutral prong 15. The carrier current transmitter can communicate bidirectionally with a remote computer controller over the AC line. This is done by using radio frequency carrier currents which travel along the household wiring. This communication could be used, for example, to alert the remote computer system that a fault had been sensed in the appliance. This information would come from monitoring node 46.

Additional information that could be sent to a remote location might include a characteristic code that would identify what type of appliance the plug is connected to as well as information about the operating status of the device such as temperature, etc. In addition, the remote computer system could send a signal to tell the appliance to turn off, in which case the carrier current transmitter/receiver module 49 would prevent capacitor 26 from charging by providing a discharge path from node 46 to the hot prong 14, thus preventing the diac 28 from firing and thus preventing triacs 29 and 31 from turning on.

A sixth specific embodiment is shown in FIG. 10. This embodiment utilizes carrier currents to transmit information bidirectionally between plug and load over the two conductors 16 and 17 connecting the plug to the

load. The appliance 54 is depicted as consisting of a load 20 and a carrier current transmitter 52. The carrier current transmitter 52 is connected to neutral through a coupling/power capacitor 53. The carrier current transmitter 52 can accept an arbitrary number of inputs from one or more sensors within the appliance and can encode that information for transmission on the two wires 16 and 17 connecting the plug 13 to the appliance 54. Sensory information is received by the carrier current transmitter/receiver 49 which can use that information to control the power to the appliance over control line 50. The carrier current transmitter/receiver 49 can also accept an arbitrary number of inputs from one or more sensors within the plug 13. The carrier current transmitter/receiver can also encode the sensory information gathered within the plug, as well as sensory information received from the appliance 54, and can transmit that information to a central computer over the household wiring. As in the fifth embodiment, the transmitter/receiver can receive control signals from the remote household controller and shut off appliance power by controlling line 50.

A seventh specific embodiment is depicted in FIG. 11. This embodiment is very much like the third specific embodiment (FIG. 7) except that it is designed to work with a nonpolarized plug, that is, it will detect a ground fault even if the hot and neutral prongs are swapped. The upper half of the circuit operates in an identical fashion to the previous (FIG. 7) embodiment when the plug prong 14 is actually plugged into the hot side of the source and the plug prong 15 is plugged into the neutral side of the source. The only difference in operation from the third embodiment is that resistor 57 is used to charge capacitor 26. The path for charging is from the neutral 15 through capacitor 56 through resistor 57.

Capacitor 56 is charged at the same time as capacitor 26. As discussed when describing the third specific embodiment, in the absence of a fault, during each half cycle when capacitor 26 charges to a voltage magnitude sufficient to fire diac 28, it will turn on triac 29 and, through the photo triac driver 47, will turn on triac 31, and both triacs 29 and 31 will conduct electrical current for the balance of the half cycle. When a fault 24 and/or 25 occurs, it serves to cause a sufficiently high voltage at the bases of Darlington transistors 38 and 39 to cause them to turn on, which in turn causes capacitor 26 to discharge so that the diac 28 voltage does not reach its turn on voltage and triacs 29 and 31 are not turned on.

If the hot and neutral prongs are swapped, then in the absence of a fault, the circuit functions as described above with capacitor 26 charging

to a voltage magnitude sufficient to trigger diac 28 on each half cycle, in turn triggering triacs 29 and 31 and allowing current to flow to the load.

If a fault 24 and/or 25 is present at the beginning of a given half cycle, then the bottom half of the circuit becomes active. Now a fault current flows through the path from prong 15 (which is now the hot prong) through resistors 61 and 62 through fault resistors 24 and/or 25 to ground. If the fault impedance 24 and/or 25 is sufficiently low in value then an appreciable current will flow at the beginning of the half cycle and a voltage will be generated at the base of Darlington transistors 59 and 60 which will be sufficient in magnitude to trigger one of these transistors, causing capacitor 56 to discharge through an LED on the photo triac driver 55 thus turning on the triac in the photo triac driver 55. When this triac turns on it discharges capacitor 26, preventing it from charging to a sufficiently high voltage to turn on diac 28. Since diac 28 is not fired, triac 29 remains in an off condition and triac 31 remains off and no current is supplied to the load for the balance of the half cycle.

The water leakage scenario, described in conjunction with FIG. 1, is helpful in understanding the functioning of the present invention for the protection against shocks and burns due to electrical current leakage in an electrical distribution system. The key feature of this invention is that the majority of electrical current in the load is intentionally blocked from flowing during a portion of each half cycle. During this portion of each cycle when current flow to the load is inhibited, if any electrically conductive path (a leak) is present between the hot wire and ground, then electrical current will flow through this conductive path and will be sensed as a fault condition.

FIG. 12 depicts a block diagram of a representative connection of one branch in a building's electrical distribution system. The load center 14' serves as a connection location between the outside electrical service hot 15' and outside electrical service neutral 16' to the load center hot 17' and load center neutral 18'. The ground 19' will generally be connected via connection 20' to outside electrical service neutral 16' at a point outside of the load center 14' and will generally also be connected via connection 20'' to the load center neutral 18'. In most installations and as required by code, a circuit breaker 60' is located within the load center 14' and serves to break the connection between the outside electrical service hot 15' and the load center hot 17'.

The hot conductor 22' exiting the load center 14' connects to one or more wall outlets 23' and to one or more switched lights 24'. The optional

ground 21' is a conductive path between ground 19' and the one or more outlet(s) 23'. The National Electric Code mandates that this connection be present on all new construction. This was not required on buildings that were constructed prior to about 1950 and so this connection is shown as a dotted line since it may not be present and the electrical outlets in older construction may not have a ground connection.

A hot to ground fault 25' and hot to neutral fault 26' are anomalous conditions that represent an undesirable electrical leakage path and may result in a shock or fire hazard. These faults may occur when the electrical insulation on the conductors 22' and 28' are frayed or otherwise damaged. Such damage could, for example, occur due to aging (the insulation wears out over time), due to being chewed upon by rodents and insects, or resulting from disruptions due to earthquake or other natural disaster. The hot to ground fault 25' could occur when a conductor 22' having damaged insulation makes electrical contact with a grounded conduit.

The hot to ground fault 25' might also occur if a human comes into contact with an exposed conductor 22' while standing on a grounded surface in which case an electrical path would be established through the human between the hot conductor 22' and ground 19' resulting in a shock hazard. The hot to neutral fault 26' might occur if an exposed hot conductor 22' were to touch the neutral conductor 28' in which case an electrical arc might occur and this arcing could result in an electrical fire. Alternatively, a human coming in contact between the hot 22' and neutral 28' conductor would represent a hot to neutral fault 26'. In a properly wired building, a neutral to ground fault 27' is not likely to result in a dangerous condition since the electrical potential difference between neutral 28 and ground 19' should be small. However, in the case of miswiring, the conductor 28' that should be at a so-called neutral potential can actually have an electrically "hot" potential, in which case a neutral to ground fault 27' can be hazardous. It is one object of the present invention to recognize a miswiring condition as a fault, thereby avoiding a hazardous condition.

FIG. 13 depicts a block diagram of a building's electrical distribution system with modifications as per the present invention to provide fault protection within the electrical distribution system in a building. The fault protection is implemented by adding a fault detection/circuit interrupt module 29' within the load center (a breaker box or the like) 14' and by adding a load conditioning module 30' at each outlet 23' and at each light 24' that is serviced by conductors 22' and 28'. FIG. 13 depicts a single light 24' and a single outlet 23',

although it should be recognized that the invention will work for any arbitrary number n of lights and any arbitrary number m of outlets (n not necessarily equal to m) connected in parallel to conductors 22' and 28'.

One advantage to this implementation is that it can be used to retrofit existing buildings, using existing wiring by simply adding the fault detection/circuit interrupt module 29' and one or more load conditioning modules 30'. In addition to shock and arc fault protection within the branch electrical distribution wiring, an added benefit is that all appliances that are serviced by this branch will also accrue shock and arc fault protection. In FIG. 13, the appliance 54' is connected via a plug 53' and appliance wiring conductors 56' and 57' to an electrical outlet 23'. A fault (current leakage path) 51' between the hot side 56' of the appliance cord and ground 19' or a fault (current leakage path) 52 between the neutral side 57' of the appliance cord and ground 19' is detected at the fault detection/circuit interrupt module 29', and power is removed from the branch conductors 22' and 28 until the fault condition is removed. In a similar way, a fault to ground 55' within the appliance load 54' will be detected and power will be removed from the branch until all faults are removed. Even if an electrical outlet 23' that is equipped with a load conditioning module 30' does not have an appliance load 54', a fault 25', 26' or 27' within the wiring connecting the load center 14' to the outlet 23' will be detected by the fault detection/circuit interrupt module 29' in the load center 14'.

FIG. 14 illustrates the way in which the fault protection/detection system works. The fault detection/circuit interruption module 29' connects to the load center hot 17' and load center neutral 18'. A switch 35' connects the load center hot conductor 17' to the hot conductor 22' connecting load center to outlets/lights. Switch 35' is used to allow (when closed) or interrupt (when open) the current flow in the conductor 22'. In this way, switch 35' functions as a circuit breaker. In some embodiments it might be desirable to add a second switch to break the neutral side of the line. This optional second switch would be located between nodes 38' and 39' and would operate synchronously with switch 35'.

In order to detect a fault, switch 35' will be opened during short time intervals with the fault detect/circuit breaker trigger 36 providing a high impedance current path around switch 35 through which a low level fault current could be sensed if it were present. At any time that a fault condition 25' and/or 26' is sensed, this condition will be detected by the fault detect/circuit breaker trigger 36' which then causes switch 35' to

maintain an open position which will be maintained at least as long as faults 25' and/or 26' exist.

At the input to wall outlet 23', there is a load conditioning module 30' which serves to block current flow to the outlet 23' for short time intervals. In the absence of a fault condition, when the load conditioning module 30' blocks current flow, there will be no substantial current flowing in the hot conductor 22'. However, if faults 25' and/or 26' exist, then even when the load conditioning module 30' blocks current flow to the appliance, there will be a sense current flow in hot conductor 22' through a high impedance in the fault detect/circuit breaker trigger 36' and through the faults 25' and/or 26'. This current flow is recognized as a fault condition and causes the switch to remain open.

Accordingly, if the hot to ground fault 25' or hot to neutral fault 26' represents a human body making accidental contact with dangerous voltages then this invention will protect against electrical injury by detecting this as a fault condition and will prevent substantial current from flowing in hot conductor 22', thereby implementing shock protection.

If the hot to ground fault 25' or hot to neutral fault 26' represents two conductors making a short circuit or arcing fault, for example through a carbonized path, then this invention will recognize a fault condition and will prevent substantial current from flowing in hot conductor 22', thereby implementing arc fault protection and preventing the dangerous heat generation that can cause electrical fires.

Although FIG. 14 depicts a single load, the theory extends to multiple loads such as outlets with appliances connected thereto and such as switched lights, provided that each light and each outlet so connected has a built-in load conditioning module 30'.

FIG. 15 illustrates the basic approach underlying the creation of a load conditioning module 30' whereby a "dead zone" is imposed at all bidirectional cyclic zero crossings of voltage. A load conditioning circuit controller 39' controls switch 40' depending upon the magnitude and polarity of the voltage sensed between conductor 42' and conductors 41' or between conductor 42' and conductor 48'. Switch 40' is normally in a closed position. When the voltage between conductors 42 and 41' (or 48') is negative and increases to zero (a so-called zero crossing with positive slope), the switch 40' is opened for a preset time interval, thus breaking the series connection and forcing the current to be zero through the load 43' for that time interval. When the time interval is over, the load conditioning circuit controller 39' closes the switch 40' to enable current

flow to the load 43'. In the same manner, when the voltage between conductors is positive and decreases to zero (a so-called zero crossing with negative slope), switch 40' is opened for a preset time interval. The electrical load 43' in FIG. 15 could be a purely resistive load such as a switched incandescent light, in which case the electrical current flow through the load would be exactly like that depicted in FIG. 5 for a 60 hertz applied voltage where the operation of the switch (40' in FIG. 15) serves to implement the dead zone (32 in FIG. 5). The electrical load 43' in FIG. 15 could also be any resistive or capacitive appliance plugged into a wall outlet or the wall outlet itself (the wall outlet without an appliance plugged in would be a load 43' modeled as an infinite resistance) FIG. 16 depicts the electrical schematic of a specific embodiment of the load conditioning module 30' whereby a bilateral trigger switch 44' is used to trigger a triac 45' to conduct electrical current through a load 62' during all but the intervals around a zero crossing. The bilateral trigger switch 44' is a bidirectional thyrister that is triggered from a blocking-to-conduction state for either polarity of applied voltage whenever the magnitude of the applied voltage exceeds a characteristic breakover voltage. The triac 45' turns off at zero crossings of current and remains off until triggered by the bilateral trigger switch 44' which turns on the triac 45' whenever the magnitude of the difference in polarity between conductors 42' and 48' is greater than the breakover voltage. Once turned on, the triac 45' conducts and allows current to flow through the load 62' for the remainder of the half cycle until the next zero crossing, at which time the triac 45' turns off. Accordingly, the load conditioning module 30' serves to implement bidirectional blocking.

FIG. 17 is a mechanical drawing of a "button" style load conditioning circuit 73' attachment that could be inserted into a light bulb socket 74' to impose a dead zone in the electrical current drawn by an incandescent light bulb 72'. This is one example approach wherein the implementation of the fault protection system at an electric light may be carried out without the need for special sockets or refixturing. The load conditioning circuit 73' consists of a triac 45' triggered by a bilateral trigger switch 44'. Triac 45' is installed in the "button shaped" load conditioning circuit 73' to create a two terminal circuit element. The MT2 terminal of the triac is exposed on one side of load conditioning circuit 73'. The MT1 terminal of the triac 45' is exposed on the other side of load conditioning circuit 73' (it does not matter which terminal

goes where). Internal to the button load conditioning circuit 73', a bilateral trigger switch connects the gate of triac 45' to the MT2 terminal of triac 45'. When installed in the socket between the light bulb 72' and the socket 74', the load conditioning module 73' is now series connected in the conductors that connect the load center to the light.

FIG. 18 depicts the electrical schematic of another specific embodiment of the load conditioning circuit whereby the load conditioning circuit can serve a dual role of providing the dead zone for fault protection as well as providing a load controlling function. The load 43 could be, for example, an incandescent light in which case the circuit can provide a dimming function, or the load 43' could be an electric fan, in which case the circuit can provide a speed control function. In FIG. 18 back to back zener diodes 49', serve to block the current flow through the variable resistance 46' to the capacitor 47' when the voltage magnitude between conductors 41' and 42' is less than the zener voltage. When the voltage magnitude between conductors 41' and 42' exceeds the zener voltage, capacitor 47' begins to be charged through variable resistor 46', the back to back zener diodes 49', and the load 43'. Until the triac 45' is triggered by the bilateral trigger switch 44', the load current is limited to that current which charges the capacitor 47' . When the voltage on capacitor 47' exceeds the breakover voltage on the bilateral trigger switch 44', the triac 45' is triggered and substantial load current is allowed to flow through the load. By adjusting the value of the variable resistor 46', phase control of the load may be implemented, essentially extending the dead zone in the waveform of the current flow through the load 43'. If the variable resistor 46' has a minimum value, the load 43' will receive maximum rms current. If the variable resistor 46' has maximum value, the load 43' will receive minimum rms current.

FIG. 19 illustrates an approach to making a load conditioning module 30' which is useful for inductive loads 61' and will also work with capacitive and resistive loads. In this embodiment, a load conditioning circuit controller 39' controls switch 40'. Possible faults that could be detected using this embodiment of the load conditioning module are depicted by dotted boxes and connections (dotted because they may or may not be present at a given time) include a hot to neutral fault 26', a hot to ground fault 51', a neutral to ground fault at the load 52', and a ground fault within the load 55'. Recalling FIG. 5, wherein the sinusoidal current waveform within a conditioned load was distinguished as having positive values over time intervals denoted as 34, negative values over the

time intervals denoted as 33, and an essentially zero value during the dead zones (interval 32), switch 40' would be in position B connecting the neutral side conductor 42' to the neutral side 48' of the load 61' for time intervals 33 and 34. During the time interval 32, switch 40' is placed in position A which connects the neutral side 48' of the load 61' to the hot side conductor 41', thereby shunting or "recycling" the current in the inductive load 61' back into itself. During this time interval 32, any or all of the faults 26', 51', 52' and 55', that were present would cause a current to flow in the hot side conductor 41', thus indicating a fault condition. For an inductive load 61' switch 40' will be controlled to be at all times either in position A or position B, otherwise a large commutating voltage could be developed across the inductance of the load due to the rapid change in current when switch 40' was opened. In the cases of purely resistive, purely capacitive, or mixed resistive-capacitive loads, position A can be open rather than connected to the hot side conductor 41', in which case FIG. 19 becomes identical to FIG. 15 and faults 26', 51', 52', and 55' can still be detected.

FIG. 20 is an electrical schematic of a specific embodiment of a load conditioning circuit that is suitable for all loads. This is a specific implementation of the approach suggested in FIG. 19. To reduce any comutating voltage developed across an appliance load 54', a commutating capacitor 71 is used. Metal oxide semiconductor field effect transistors (MOSFET's) are used as the series switch of the load conditioning circuit.

MOSFET transistors 63' and 64' can only control current flow in one direction. They generally have a body diode 65' internally connected which conducts current flowing in the reverse direction. To prevent reverse current flow, current steering diodes 66' are used. Using both an NMOS transistor 63' and a PMOS transistor 64' allows a simple method to switch current flow in both directions where the NMOS transistor 63' controls current during the positive half cycle 34 and the PMOS device controls current during the negative half cycle 33. The load conditioning circuit controller 39' receives its power from the hot side conductor 41' and neutral side conductor 42'. The load conditioning circuit controller 39' might be a microcontroller. The load conditioning circuit controller 39' must be chosen so that it does not itself draw appreciable current during the deadzone interval 32 or it would be recognized as a fault.

FIG. 21 depicts the control signals that must be furnished by the load conditioning circuit controller 39' in order to implement the embodiment in FIG. 20 for making a load conditioning circuit. The top

trace in FIG. 21 depicts the turn-on voltage applied to the gate 68' of the PMOS transistor 64', with respect to the neutral side conductor 42'. The PMOS transistor 64', is an enhancement device that will conduct when a sufficiently negative gate voltage is applied, but is nonconducting with zero or positive gate voltages. The voltage at gate 67' for the NMOS device 63' is shown as the middle trace on FIG. 21. The NMOS transistor 63', is an enhancement device which becomes conducting with a sufficiently positive gate voltage but is nonconducting with zero or negative gate voltages. The bottom trace depicts the direction (but not relative magnitude) of current flow in the hot side conductor 41' under one example control profile. The dead zones 32 indicate intervals wherein current is blocked from flowing into the load in the absence of a fault condition.

FIG. 22 is an electrical schematic of one embodiment of the fault detect circuitry wherein circuit interruption occurs at the hot side of the line at the load center upon the detection of a fault condition. The fault detect/circuit interrupt module 29' has load center hot 17' and load center neutral 18' supplying the input power. The hot conductor connecting load center to outlets/lights 22' and neutral conductor exiting load center 28' are the means by which power is delivered from the load center to the load conditioning module 30' and the load 78', the load 78' representing, for example, a light or an outlet or an arbitrary number of appliances connected to outlets. At the beginning of any half cycle, the current interrupting triac 79' is nonconducting until such time as it is fired by the bilateral trigger switch 44'. This serves to give a dead zone just after the zero crossing, until such time (in the absence of a fault) as the applied voltage exceeds the bilateral trigger switch triggering voltage.

At the beginning of any half cycle, the fault latch triac 88' is unfired.

At the beginning of the positive half cycle, with triacs 79' and 88' unfired, a small detection current through the series combination of resistors 82', 83' and any fault resistances 26' and/or 51', and/or 52' if present, will serve to fire the PNP Darlington transistor 84' which serves to trigger the fault latch triac 88' which in turn discharges the triac trigger capacitor 89', thereby preventing the bilateral trigger switch 44' from attaining the breakover voltage and thereby preventing the current interrupting triac 79' from firing and preventing current from being supplied to the load for the balance of the positive half cycle. On the other hand, if no fault resistance is present, the triac triggering capacitor 89' charges through charging resistor 90' until its voltage

exceeds the breakover voltage of the bilateral trigger switch 44', causing current to be delivered to the gate of the current interrupting triac 79', causing the triac 79' to turn on and to deliver power to the load. Once the current interrupting triac 79' has been fired, the circuit is insensitive to a fault condition until the beginning of the next half cycle.

For the negative half cycle, the circuit functions in an identical manner as for the positive half cycle except that the polarities are reversed through all devices and the NPN Darlington 85' is active in the case of a fault condition instead of the PNP Darlington 84'.

Once fired, in order for current interrupting triac 79' to stay fired, load current must be conducted by the triac 79'. In the event that the load conditioning circuit 30' does not draw a current, even though the current interrupting triac 79 has been fired, the triac trigger capacitor 89' will begin to charge up again through charging resistor 90'.

If no fault is detected, when the voltage on the triac trigger capacitor 89' reaches the firing voltage of the bilateral trigger switch 44', it will again fire and deliver gate current to the current interrupting triac 79'. This process will repeat until the load draws sufficient current to latch the current interrupting triac 79' in an on condition for the remainder of the half cycle. In the absence of back to back zeners 87' and control resistor 91' it might be possible for the load conditioning circuit 30' to draw a low level current while the triac trigger capacitor 89' is being recharged after a failed pervious attempt.

This condition would then be detected as a fault with fault latch triac 88 being fired and inhibiting any further attempted firing of the current interrupting triac 79'. To prevent this condition, the back to back zener diodes 87' will conduct at a predetermined voltage. Once this voltage is reached, current is conducted through the control resistor 91, thereby developing a voltage across the emitter resistor 86'. As the magnitude of the voltage across the emitter resistor 86' increases, the sensing voltage threshold at the base of the Darlington transistors also increases until the voltage at the emitters of the Darlington transistors exceeds the sensing range of the circuit, thus preventing any additional fault sensing until near the very end of the half cycle. Consequently, the role played by opposing zeners 87', together with control resistor 91', is to allow a predefined current sense window during which a fault will be detectable.

This is important if there are multiple load conditioning circuits 30', each of which having a different firing voltage and consequently a

different dead zone length. This is also important if the load conditioning circuit 30' is used in conjunction with phase power control as depicted in FIG. 18, because the dead zone during phase control may be as wide as the entire half cycle.

In FIG. 22, if a very low impedance fault such as a short circuit occurs, the current interrupting triac 79' is required to carry this current for as much as one half cycle before a fault condition is recognized. This half cycle short circuit current can be limited by modifying the circuit shown in FIG. 22 to that depicted in FIG. 23. In FIG. 23, two NMOSFETs 92' and 93' have their gates connected together and their sources connected together. The drain of NMOSFET 92' is connected to the incoming hot conductor 17'. The drain of NMOSFET 93' is connected to the hot input to the fault detecting circuit 94'. Zener diode 95' connects between the gates and sources on the NMOSFETs 92' and 93' and regulates the positive gate voltage applied to NMOSFETs 92' and 93'. This voltage is provided through resistor 96' from capacitor 97' which is in turn charged through resistor 98' and diode 99' from the load center neutral wire 18'.

During the negative half cycle (the period when the hot conductor 17' is more negative in potential than the neutral conductor 18') diode 99' will conduct and charge capacitor 97' through resistor 98', diode 99 and body diode 100'. Diode 99' will prevent charge capacitor 97' from discharging during subsequent positive half cycles. The voltage maintained during both half cycles at the gates of the two NMOSFETs will be a fixed DC value due to zener regulator 95'. During the negative half cycle, NMOSFET 93' will conduct current through body diode 100'. However, the amount of current that can be drawn will be limited by the gate voltage (the voltage on zener regulator 95') and by the internal drain resistance. In a similar way, during the positive half cycle, NMOSFET 92' will conduct current through body diode 101' and will limit current according to the value of the gate voltage and the drain resistance characteristic of the NMOSFET.

Consequently, the circuit elements between conductors 17' and 94' in FIG. 23 serve one role only and that is to limit the maximum current that can flow in conductor 94 in the case of an overload current condition such as a short circuit.

FIG. 24 depicts a load conditioning circuit (which in its most general form consists of a load conditioning circuit controller 39' controlling a switch 40' (see FIG. 15) at a load 43'), together with an open ground and wiring transposition detector circuit. Resistor 102' must have a value higher than that of the maximum detectable fault resistance

and resistor 105' must have a value well under the maximum detectable fault resistance. The gate of faulting triac 104' is connected to an assumed ground 107'. This is a point within an outlet or light which is assumed to be grounded. The dotted line 103' indicates a connection to ground. If this connection does in fact exist, the faulting triac 104' is unable to fire because its gate voltage will have the same potential as the neutral conductor 42'. Consequently, the triac 104' acts as an open circuit and does not impact the operation of the system. If there is no connection 103' (corresponding to an open ground condition) then ground sensing resistor 102' provides gate current to fire the fault triac 104', which in turn simulates a hot to neutral fault by placing fault limiting resistor 105' across the lines 41' and 42' thereby creating a fault. This in turn causes current interruption at the load center for the entire branch. Equivalently, if the ground conductor 19' is wired to the hot conductor 41', it will cause a current to flow through the fault triac 104', simulating a fault condition and causing current interruption at the load center for the entire branch. The role of fault limiting resistor 105' is to protect the gate of the fault triac 104'. If the hot and neutral wires are transposed at the load or throughout the branch, regardless of whether the ground is connected, the circuit in FIG. 24 will detect the condition and trigger the fault triac 104', thereby simulating a fault condition and causing current interruption of the hot conductor at the load center for the entire branch.

The invention may be embodied in other specific forms without departing from its spirit or essential characteristics. The described embodiments are to be considered in all respects only as illustrative and not restrictive and the scope of the invention is, therefore, indicated by the appended claims rather than by the foregoing description. All changes which come within the meaning and range of the equivalency of the claims are to be embraced within their scope.




 
Previous Patent: CABLE SUPPORT SYSTEM

Next Patent: OVERCURRENT PROTECTION SYSTEMS