What is claimed is:
1 . An AFCI tester adapted to be coupled to an AC power line having an AFCI, the tester comprising:
a voltage regulator having an input adapted to be coupled to the AC power line and an output; a gated switch having a gate input; a spark gap circuit coupled to the gated switch and coupled to the AC power line; a controller having an output coupled to the gate input; means for activating the controller to switch the controller output and close the gated switch; and means for checking for a trip by the AFCI within a specified time period after the gated switch is closed, whereby failure to trip with in a specific time period results in a visual indication of the failure.
2 . An arc fault current interrupter (AFCI) testing device, comprising:
hot and neutral terminals for respective connection to hot and neutral conductors of a branch circuit; a spark gap having first and second ends respectively coupled to the hot and neutral terminals; a gate interposed between a preselected one of said hot and neutral terminals and the spark gap, the gate having a current path and a control electrode; and a selector coupled to the control electrode of the gate, actuation of the control electrode by the selector rendering the current path of the gate conductive, thereby permitting an arc to be induced across the spark gap suitable for testing an AFCI connected to the branch circuit.
3 . The AFCI testing device of claim 2 , wherein the selector comprises an integrated circuit with an output coupled to the control electrode of the gate and an input coupled to a user-operable switch.
4 . The AFCI testing device of claim 3 , wherein the selector further comprises an optoisolator having an output coupled to the control electrode of the gate, a control electrode of the optoisolator coupled a control output of the integrated circuit.
5 . The AFCI testing device of claim 2 , wherein the current path of the gate is interposed between the hot terminal and the spark gap.
6 . The ACFI testing device of claim 2 , wherein the gate is a triac.
7 . The ACFI testing device of claim 2 , and further comprising a power supply having hot, neutral and ground terminals for connection to hot, neutral and ground terminals of an outlet and a power output, the power supply producing power on the power output even where the outlet terminals have been miswired, as long as one of the terminals of the outlet is connected to a hot voltage and another of the terminals of the outlet is connected to either neutral or ground.
8 . A method for testing an arc fault current interrupter (AFCI) installed on an alternating current (AC) power circuit, comprising the steps of:
connecting a hot terminal of an arc fault simulation device to an energized conductor of the AC power circuit; connecting a neutral terminal of the arc fault simulation device to a second conductor of the AC power circuit, the second conductor not being energized but providing a return path; providing a spark gap with a first end coupled to a preselected one of the hot terminal and the neutral terminal; coupling a first switch to a second end of the spark gap such that upon actuation the switch establishes a current path between the hot and neutral terminals; and actuating the first switch so as to draw an arc across the spark gap, thereby causing a detectable and real arc fault condition in the AC power circuit.
9 . The method of claim 7 , and further comprising the steps of:
coupling a control electrode of said switch to an output of a logic circuit; coupling a user-actuable, second switch to an input of the logic circuit; and relating the state of the logic circuit input to the logic circuit output such that actuation of the second switch causes actuation of the first switch.
10 . The method of claim 7 , and further comprising the steps of optically coupling a current path of the first switch to a control electrode thereof.
11 . The method of claim 7 , and further comprising the step of
prior to said step of actuating the first switch, ascertaining whether the AC power circuit has been correctly wired or whether it has been miswired.
12 . Apparatus for testing an arc fault current interrupter (AFCI) connected to an AC power circuit, comprising:
means for coupling a hot terminal of the apparatus to an energized conductor of the AC power circuit; means for coupling a neutral terminal of the apparatus to a second conductor of the AC power circuit, the second conductor not being energized but providing a return path; and first switch means for selectively coupling a spark gap across the hot terminal and the neutral terminal, actuation of the switch means causing an arc across the spark gap and a detectable and real arc fault condition on the AC power circuit.
13 . The apparatus of claim 12 , and further comprising
user-actuable second switch means coupled to a logic circuit, an output of the logic circuit controlling said first switch means.
FIELD OF INVENTION
 The present invention relates generally to AC power line test equipment and, more particularly, to a power line testing apparatus capable of detecting various wiring conditions and ground integrity, capable of generating an arc fault and capable of transmitting a tracer signal easily distinguishable from environmental noise.
BACKGROUND OF THE INVENTION
 When electrical power line outlets (receptacles) are installed in residential, commercial or industrial environments, it is extremely important that they are properly wired. If such outlets are incorrectly wired, they may be absolutely useless and/or can cause significant damage to equipment and property or, even worse, they can result in electrical shock, thereby possibly causing serious harm to, or death of, individuals. A further extremely hazardous situation occurs, even if the wiring scenario is correct, but where the integrity of the connected earth ground conductor is compromised. A proper ground connection is meant to shunt leakage current, which can appear at conductive enclosures of appliances due to defects or poor insulation, to earth ground. The integrity of earth ground conductors can be compromised by an increase in electrical impedance due to corrosion of conductors and/or conduits (when used as earth ground) or by degradation of conductor-to-conductor connections in commonly used wire nuts or screw terminals, which may loosen over time. If sufficient impedance is present in a ground system, the ground conductor is no longer a real earth ground.
 Ground fault interrupting devices are known in the art. Those devices are usually equipped with an internal test function which works quite reliably, yet they cannot test the integrity of fixed or temporary branch circuit extensions. Therefore, numerous external testing devices have become available which are to be connected to such extensions and to perform the required test from there. Almost all known devices in this art generate a leakage current flow of a certain amount from the hot to the ground conductor, but they ignore the time that elapses until the device under test reacts. This leads in many cases to the false assumption that the device under test works properly and reliably while, in fact, in many cases it does not since the device indeed reacts, but exceeds the permissible period of time for reaction.
 Fairly new in the art are so-called arc-fault interrupting devices. These devices are meant to interrupt an electrical branch circuit if serial or parallel arcing occurs along or between hot and neutral conductors in the power line system. Like ground fault circuit interrupting devices, they usually have an internal test function. In practice, there is a significant problem with these test functions because the activation of such a test is limited to either giving an incorporated microcontrolleri a command to activate the associated trip mechanism, or “inject” an electronic signal to the system in order to simulate arcing. As a result of the first such “test”, it only gets proven that the microcontroller is “alive” and that the mechanical parts of the apparatus work properly. A second type of “test” is apparently more enhanced than the first. However, in both of these tests, there is no reliable confirmation whatever that the devices will work properly and reliably under real arcing conditions.
 When work needs to be performed on an electrical branch circuit, it is first necessary to unenergize that circuit or interrupt the current flow to the circuit. This is commonly achieved by opening or unscrewing the circuit-interrupting device in the distribution panel that is associated with that particular branch circuit. In many cases, it is not known which circuit-interrupting device, out of a plurality of such devices (as commonly occurs on a panel of circuit breakers), is actually the one in question. Absolute methods of determining the correct device are often not only inconvenient and time consuming but, in many circumstances, not even feasible. A “classic” method used to find the associated device needed is to sequentially unscrew fuses or to open circuit breakers in a distribution panel until the one in question has been found. Subsequently, all outlets in the branch circuit under test need to be checked for an unenergized condition. In certain environments, such as hospitals or manufacturing plants, this or similar methods are totally impractical. In other environments (e.g. where computers without backup power are used), such methods can be, at the very least, disturbing and/or annoying. As alternatives to these methods a variety of electronic devices have been developed that accurately determine which is the particular circuit-interrupting device in question. By examining those devices, it becomes apparent that they all have significant drawbacks.
 The invention disclosed in Virgilio U.S. Pat. No. 5,625,285 describes a device for monitoring the present wiring scenario and acceptable grounding properties on a standard 3-Wire 120 volt AC electrical outlet. The circuit for this device is reproduced as FIG. 2. The described device can detect and indicate the following wiring possibilities:
 1. Correct wiring
 2. Defective ground
 3. Open neutral
 4. Hot and neutral reversed
 5. Hot on neutral with open neutral
 One of the most hazardous situations, hot and ground reversed, cannot be detected and indicated. Further, it sometimes occurs that a second hot wire is mistakenly connected to a receptacle. In such a case, the voltage across the two hot conductors is twice that of the nominal voltage. If someone intends to perform work on such a miswired circuit, he or she faces an undesirable and potentially dangerous situation, since the circuit is still energized even if one associated circuit-interrupting device has been deactivated.
 In order for the Virgilio structure to analyze the integrity of the earth ground conductor, a high current pulse of short duration is drawn over the power line system. A pickup coil senses the strength of the thereby generated magnetic field. The induced voltage in the pickup coil is then used to trigger a semiconductor device, which then activates a visible indicator. This circuit requires exact calibration during the manufacturing process. If parasitic impedance is present in the ground conductor, the magnetic field loses strength, therefore the induced voltage in the pickup coil is no longer sufficient to trigger the semiconductor control device. In practice, line resistance and capacitance are subject to continuous changes due to frequent on and off switching of heavy electrical loads. This has major impact on the proper performance of that device. These line and ground impedances impact the reliability of Virgilio's ground integrity test.
 Another structure, disclosed in Robitaille U.S. Pat. No. 4,929,887, describes an electrical outlet monitor that recognizes the fact that more than one hot conductor can be, possibly, connected to an outlet. However, the unit does not indicate in detail what the present wiring scenario is, but only that it is incorrect. The circuit for this unit is shown in FIG. 1.
 No patented prior art is known to the applicant which covers GFCI testing devices that include the measurement and consideration of elapsed time as part of testing criteria. However, there is a device available, designed and manufactured by a German company named BEHA GmbH and distributed in North America by Greenlee which does take elapsed time into consideration.
 Another structure, disclosed in Spencer et al. U.S. Pat. No. 5,875,087, describes an enhanced digital circuit breaker that includes an AFCI and an associated AFCI test function. As it is apparent from the block diagram of this breaker (FIG. 14), activation of the incorporated momentary push button 20 does not in any way introduce an actual arcing condition in the power line system under test, but rather, simply instructs the microcontroller to mechanically trip the device through D 2 and thyristor 22 . In effect, this is not at all a “test” of the reliability of the device to respond to a genuinely dangerous condition; it is simply a demonstration of the “desired” response to such a dangerous condition.
 A structure disclosed in MacKenzie et al. U.S. Pat. No. 5,459,630 depicts an enhanced circuit breaker with incorporated GFCI and AFCI functions. It also includes a “self-test” for those two functions. This self-test function still fails to introduce an actual arcing condition to the power line system for the purpose of testing. MacKenzie et al. depicts two different ways by which the incorporated arc-fault detector can be tested. First, it generates an electronic signal with pre-determined parameters in order to electronically simulate an arcing condition, thereby forcing the arc-fault detector to respond in the desired way. Secondly, and in a nearly similar way, it generates a test signal of a different shape and takes more parameters into consideration. Here, again, there is no real or actual arcing introduced into the power line system for the purpose of testing the reliability of the AFCI.
 U.S. Pat. No. 4,906,938 (Konopka), U.S. Pat. No. 5,497,094 (George), and U.S. Pat. No. 5,969,516 (Wottrich) all introduce devices consisting of two separate units in order to locate a particular circuit-interrupting device, among a number of such devices in an electrical distribution panel. All three devices draw a current spike over the power line system of sufficient strength and short duration that it can be used as an identification signal. The associated receivers for all three devices require manual adjustment in order to evaluate the detected identification signal in quantitative terms. None of those devices has the capability to evaluate an identification signal in terms of “quality”. Power line systems are, to a great degree, polluted by numerous and various electrical signals that are generated arbitrarily, accidentally, or deliberately.
 Many of those signals are sharp rising and of short duration, thereby having very similar shapes as the identification signals of the above-mentioned three devices. Since, as mentioned, those three devices do not evaluate any detected signal in terms of quality, they often get triggered by any number of polluting signals rather than by the intended identification signal. This makes those units, to a certain degree, unreliable. The technologies disclosed in U.S. Pat. Nos. 4,906,938, and 5,497,094 are of quite simple design, while the technology disclosed in Wottrich U.S. Pat. No. 5,969,516 is somewhat more enhanced. The transmitter of the Wottrich device has a wiring monitor incorporated that is, to some degree, similar to the wiring monitor disclosed in Virgilio U.S. Pat. No. 5,625,285 (as mentioned above). Furthermore, this structure is claimed to have the capability to generate an identification signal under all scenarios that are detected and indicated by the incorporated wiring monitor, except an unenergized circuit. In fact, it generates, under some wiring conditions, not only one but two identification signals. This gives rise to false identification. Practical evaluations of this device show that GFCI units, if associated with a branch circuit under test, get tripped immediately when the transmitter of this device gets connected to the electrical branch circuit.
 FIGS. 11 and 12 show the schematics of two conventional devices for locating circuit-interrupting devices associated with a particular electrical branch circuit. The design in FIG. 11 is of a quite simple structure and is disclosed in U.S. Pat. No. 4,906,938. U.S. Pat. No. 5,969,516 discloses a design that is illustrated in FIG. 12. As mentioned above (prior art), this device can generate an identification signal under different wiring scenarios, yet has the drawback as depicted above.
SUMMARY OF INVENTION
 The intended purpose of the herein described invention is to provide a tool, which can be used by either a professional or a layman, that allows work to be done on a power line system in the safest manner possible. This invention is an electronic testing device that includes two separate units. These units can be used independently or in conjunction with each other.
 This device provides the following specific functions:
 It is an AC outlet wiring monitor that analyzes and indicates the wiring scenario present in a standard power line system, testing for both correct wiring and the most common of the various incorrect wirings.
 It is a self-adjusting testing unit with regards to different power line systems (110V/60 Hz, 220V/50 Hz, 440V/50 Hz), or any other situation caused by improper wiring.
 It analyzes and indicates the integrity of a given earth ground system by measuring the amount of parasitic ohmic resistance.
 It provides the capability to test the reliability of an associated GFCI by not only generating a leakage current across the hot and ground conductor, but also measuring the time that elapses until reaction as an essential part of the testing criteria.
 It provides the capability to test the reliability of an associated AFCI by generating a real, physical, sputtering arc in the power line system while also measuring the time that elapses until reaction as an essential part of the testing criteria.
 It is a device that can be used to locate a particular circuit-interrupting device (E.g. circuit breaker or fuse) associated with the particular electrical branch circuit under test. Unlike all other known units, this device does not require any manual adjustments. Also, unlike all other known units, this device analyzes the quality of the generated identification signal, in addition to the quantity, in order to avoid false indications.
 It senses the presence of an electrical field in common 50/60 Hz power systems and indicates the strength of the field by increasing and/or decreasing the duty cycle of a visual and/or audible indicator.
 According to one aspect of the present invention, a signal generator adapted to be placed on a branch circuit includes a spark gap and a switch connected in series with a spark gap. When the switch is actuated, an arc will appear across the spark gap, thereby injecting an actual arcing condition into the circuit to be tested. Preferably, the switch is an optoisolator that is controlled by a logic circuit, which, in turn, is manually operated by a user.
 According to another aspect of the invention, the power circuit testing device transmitter includes a transformerless power supply which can derive DC power from any AC electrical outlet, even one which is miswired, as long as one of the conductors is energized and another of the conductors provides a return path. This power supply uses full wave rectification.
 According to a third aspect of the invention, a signal transmitter of a branch circuit locating system generates a signal in which the period of the signal is different from a period of an AC carrier on which the signal resides. This permits the easy identification of the signal as coming from an artificial source and makes it easy to discriminate this signal from other system conditions. In a preferred embodiment, the signal is amplitude-modulated and consists of voltage spikes on maxima and minima of the carrier wave through two cycles, with an intentional omission of such spikes on at least one following cycle. The signal generated according to the invention preferably has both negative and positive going components.
 According to yet a further aspect of the invention, the power line testing device according to the invention determines the integrity of the ground connection. It does this by measuring the parasitic impedance of the ground connection. In a first embodiment of the ground integrity test function, this is done by a voltage divider. In a second embodiment of the ground integrity test function according to the invention, a phase shift is measured taking into account both real and imaginary components of the parasitic ground impedance, and the degree of phase shift of a varying signal is ascertained and compared against a stored criterion.
 According to yet another aspect of the invention, the transmitter according to the invention is capable of producing a fault signal across any two of the three conductors typically found in an AC power circuit, as long as the first of those conductors is energized and a second of those conductors provides a return path.
 According to yet a further aspect of the invention, a receiver of the testing system measures incoming signals over a predetermined period of time and averages signal strength. This signal strength average is compared against a stored reference in order to determine whether or not the AFCI or GFCI interrupters are correctly operating.
BRIEF DESCRIPTION OF THE DRAWINGS
 FIG. 1 is a schematic diagram of a receptacle wiring monitor as disclosed in U.S. Pat. No. 4,929,887.
 FIG. 2 is another schematic diagram of a receptacle wiring monitor and a ground integrity tester as disclosed in U.S. Pat. No. 5,625,285.
 FIG. 3 is a schematic diagram of a ground integrity tester basic subcircuit in accordance with the present invention.
 FIG. 4 is a schematic diagram of a ground-fault tester basic subcircuit in accordance with the present invention.
 FIG. 5 is a schematic diagram of an arc-fault tester basic subcircuit in accordance with the present invention.
 FIG. 6 is a schematic diagram of a ground-fault tester used in numerous existing devices.
 FIG. 7 is a schematic diagram of a transformerless power supply basic subcircuit in accordance with the present invention.
 FIG. 8 is a schematic diagram of a preferred embodiment of a transformerless power supply in accordance with the present invention.
 FIG. 9 is a schematic diagram of a first embodiment of the signal generator in accordance with the present invention.
 FIG. 10 is a schematic diagram of a signal-transmitting basic subcircuit in accordance with the present invention.
 FIG. 11 is a schematic diagram of a transmitter and receiver of a conventional circuit-interrupting device locator as described in U.S. Pat. No. 4,906,938.
 FIG. 12 is a schematic diagram of a transmitter and receiver of a conventional circuit-interrupting device locator as described in U.S. Pat. No. 5,969,516.
 FIG. 13 is a schematic diagram of a preferred embodiment of a signal-detector in accordance with the present invention.
 FIG. 14 is a block diagram of a conventional arc fault interrupting device as described in U.S. Pat. No. 5,875,087.
 FIG. 15 is a block diagram of the signal generator in accordance with the present invention.
 FIG. 16 is a graph of voltage versus time of a preferred identification signal generated in accordance with the present invention.
 FIG. 17 is a block diagram of the signal detector in accordance with the present invention.
 FIGS. 18 a through 18 f is a software flowchart of a signal generator in accordance with a first embodiment of the present invention.
 FIGS. 19 a through 19 c is a software flowchart of a signal detector in accordance with a first embodiment of the present invention.
 FIG. 20 is a detailed schematic electrical diagram of a power supply according to a further embodiment of the invention.
 FIG. 21 is a detailed electrical schematic diagram of a signal generator according to a second preferred embodiment of the present invention.
 FIG. 22 is a schematic electrical diagram of a battery driven power supply component of a signal generator according to the invention.
 FIG. 23 is a schematic electrical diagram of a further power supply according to the invention.
 FIG. 24 is a schematic electrical diagram of a signal detector according to a second preferred embodiment of the invention.
 FIG. 25 is a software flow diagram of an algorithm used by the microcontroller in the signal generator shown in FIG. 21.
 FIG. 26 is a flow chart of a testing algorithm according to the invention.
 FIG. 27 is a flow chart of a TRACER subroutine used by the signal detector shown in FIG. 24.
 FIG. 28 is a LOWBATTERY subroutine according to the invention.
 FIG. 29 is a STANDBY subroutine according to the invention.
 FIG. 30 is a flow chart of a SENSOR subroutine used in the second preferred embodiment of the invention.
 FIG. 31 is a flow chart of a POWEROFF subroutine used in the invention.
 FIG. 32 is a schematic electrical diagram of a further embodiment of an arc fault current interrupt testing device according to the invention.
 In the above drawings, like characters denote like components where possible.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT
 Signal Generator
 This unit performs five different tasks:
 1. It analyzes and indicates the present wiring scenario or condition on a given power outlet (receptacle) and also analyzes which one of the commonly known power line systems it is attached to (110V/60 Hz, 220V/50 Hz, 440V/50 Hz).
 2. It checks for and indicates the electrical integrity of the branch circuit's “earth ground”.
 3. On command, it tests the reliability of a possibly associated ground fault circuit interrupter (GFCI).
 4. On command, it tests the reliability of a possibly associated arc fault circuit interrupter (AFCI) by generating an actual arc across a spark gap.
 5. It generates a distinct identification signal, which is sent over the power line system, allowing the below described signal detector to absolutely identify the particular circuit interrupting device associated with the branch circuit being tested.
 The signal generator is controlled by a microcontroller. Since it is considered to be a “testing” device, it must work under various possible wiring scenarios that may exist on the outlet under test. Thus, the incorporated microcontroller must be powered under all these conditions. This unit is energized directly by the power present in the outlet being tested. This means that the employed power supply needs to be used, in part, three times.
 The power supply schematic in FIG. 7 shows three bridge rectifiers (D 1 , D 5 , and D 6 ) which are connected across the different conductor terminals in order to obtain DC power under various different wiring scenarios. A two-pole filter, comprised of electrolytic capacitor C 1 , resistors R 7 and R 8 , capacitor C 2 and electrolytic capacitor C 3 , is employed to smooth the pulsing DC current and to limit the applied current. A voltage regulator U 1 , whose output is further smoothed by capacitor C 5 and electrolytic capacitor C 4 , follows this arrangement.
 This power supply has been designed for a first herein-described unit, but is not limited for use in this application. It can be used in any other application where low voltage DC needs to be generated from a standard power line system. FIG. 8 shows the power supply of a first preferred embodiment of the invention. Basically, it is the same arrangement as depicted in FIG. 7, but with some additional components that are necessary in order to meet the requirements for the preferred embodiment. Three resistor networks comprised of R 1 /R 2 , R 3 /R 4 , and R 5 /R 6 are configured as voltage dividers with a ratio of 1:100. The voltages obtained at the junctions 24 , 32 , 34 of those voltage dividers are fed to analog inputs of a PIC16C73 microcontroller U 2 (FIG. 9). The junction 24 of network R 1 /R 2 is connected to pin 2 of U 2 ; the junction 32 of R 3 /R 4 is connected to pin 3 of U 2 ; the junction 34 of R 5 /R 6 is connected to pin 4 of U 2 . Three additional diodes, D 2 , D 4 and D 6 , are employed in order to isolate the positive path of the three power supply subcircuits from each other in order to avoid feedback from one part to another. The cathodes of those three diodes are connected together at a node 36 , thereby forming the final positive supply. This positive supply is then fed through a two-pole filter (R 7 , R 8 and C 2 ) to a positive voltage regulator U 1 for 5VDC regulated output at node 38 as described above. The hot, neutral, and ground terminals are fed through three current-limiting resistors, R 9 , R 10 , and R 11 , to nodes 26 , 28 and 30 , which are respectively connected to digital input pins 5 , 6 , and 11 of microcontroller U 2 (FIG. 9). In order to avoid damage to microcontroller U 2 , nodes 26 , 28 and 30 are respectively clamped by diode pairs to Vdd and Vcc. The diode pairs are comprised of D 7 /D 8 for node 26 , D 9 /D 10 for node 28 , and D 11 /D 12 for node 30 .
 In practice, it happens that conductors in power line systems build up parasitic impedance over a period of time. This is due to corrosion of material, poor mechanical connections (screw terminals), and other minor influences. If it happens in the ground conductor, this conductor is no longer a real earth ground and builds up AC current which creates potentially hazardous conditions.
 The device shown in U.S. Pat. No. 5,625,285 (FIG. 2) putatively analyzes and indicates this condition. The reliability of this technology may be questioned. Since the magnetic field which is generated by sending a high-frequency current pulse towards ground is used in order to trigger any kind of indication device, reliability suffers because of continuously changing power line impedance and capacitance.
 The circuit shown in FIG. 3 needs to be powered by DC current, therefore any kind of commonly known DC power supply—which uses a bridge-rectifier in order to obtain a separate signal ground different from earth ground—can be used. In the illustrated embodiment a general-purpose operational amplifier (op amp) U 1 A is configured as a voltage comparator. A resistor network comprised of resistors R 1 and R 2 forms a slightly unbalanced voltage divider that provides the half of the supply voltage plus an additional 1% of the supply voltage on the junction 42 between those two resistors. This voltage is fed to the non-inverting input of the operational amplifier U 1 A. Another balanced voltage divider, comprised of resistors R 4 and R 5 , provides exactly the half of the supply voltage on the junction 44 between those two resistors. This voltage is fed to the inverting input of the operational amplifier U 1 A. If parasitic resistance is present in the ground conductor, it adds on to resistor R 5 , thereby forcing the voltage divider out of balance. This results in an increase of the voltage at junction 44 . If this voltage rises above the voltage on the non-inverting input of the operational amplifier (node 42 ), then the output of the op amp U 1 A goes high and lights LED D 1 through current-limiting resistor R 3 . This alert makes the user aware of improper grounding conditions. An advantage of this approach is that since this technique is based on pure resistive (ohmic) testing, neither changes in power line inductance or capacitance, nor any accidentally present high-frequency signals, have any impact on the result.
 Ground-fault circuit interrupting devices (GFCI) are widespread in the marketplace. Even though these devices are equipped with an internal test function, a variety of other testing devices for GFCIs are available. FIG. 6 shows an example of an electronic circuit as it is utilized in most of the existing GFCI testers. A leakage current flows from hot to ground through current-limiting resistor R 1 if push-button S 1 is activated. The value of resistor R 1 is chosen so that a leakage current in range between 6 and 9 mA flows to ground. This method is reliable, yet totally ignores consideration of the time elapsed until a GFCI reacts.
 FIG. 4 shows an arrangement that remedies this problem. In the same manner as mentioned above, any common DC power supply can be used in order to energize the circuit shown in FIG. 4. A versatile 555 timer (U 1 ) is configured as a monostable multi-vibrator, in which resistor R 1 in series with capacitor C 2 are the time determining components. Capacitor C 1 is not really necessary, but helps to give the circuitry better stability. If pushbutton S 1 gets activated, then timer U 1 switches its output (pin 3 ) high. This lights LED D 1 through current-limiting resistor R 2 and also triggers thyristor Q 1 into conduction. When thyristor Q 1 is conductive, a current flows through current-limiting resistor R 4 from hot to ground in the same manner as in commonly-known GFCI testers (see above). If a predetermined time (R 1 , C 2 ) elapses, then timer U 1 's output goes low, thereby deactivating thyristor Q 1 and so terminating the current flow from hot to ground. Furthermore, since U 1 's output goes low, LED D 1 no longer lights. This is the indication that the attempted test has failed.
 There have been arc-fault breakers for only a short period of time in the marketplace. Those devices are also equipped with an internal “test function”. The way this “test function” works is that the incorporated microcontroller recognizes that a test is requested and, therefore, activates one of its outputs in order to trip the breaker. This certainly gives an indication that the mechanical part of the breaker works properly, but it does not by any means indicate whether or not the unit would react reliably if an actual arcing condition were present or introduced to the power line system. The block diagram shown in FIG. 14 shows such a design as it is disclosed in U.S. Pat. No. 5,875,087.
 FIG. 5 shows a schematic diagram of a circuit according to the invention which indeed causes actual controlled arcing in the power line system. The circuit illustrated in FIG. 5 is similar to the circuit shown in FIG. 4. However, in the current-drawing path of the circuit comprised of thyristor Q 1 and current-limiting resistor R 3 , a spark gap GAP 1 has been employed to cause real arcing in the system.
 Numerous devices are also available in the marketplace for locating a particular circuit breaker in a distribution panel. All those devices consist of two separate units, usually referred to as transmitter and receiver. The device of the invention also includes two separate units. There are, however, significant differences between the two unit device of the invention and the known prior art. Those differences will be described in greater detail further below. FIG. 10 shows a design of a signal generator that generates a very strong identification signal over the power line system in order to be detected by the later-described signal detector. Resistor R 1 , in series with capacitor C 1 connected across the hot and neutral conductors of a power line system, forms a phase-shifting network that serves the purpose of activating thyristor Q 1 through trigger diode D 1 . An LED has been chosen as the trigger diode. The LED D 1 will light during triggering, thereby indicating to the user that the device is active. The electrical values of resistor R 1 and capacitor C 1 determine at what electrical phase angle triac Q 1 goes into and out of conduction. When Q 1 becomes conductive in the positive half-cycle of the sine wave, capacitor C 2 (with a large electrical value) gets charged instantly to the peak voltage level. This causes the appearance of a very strong electromagnetic field that surrounds the power line. When capacitor C 2 (FIG. 8) is fully charged, thyristor Q 1 (FIG. 9) goes out of conduction. During the negative half-cycle of the power line sine wave, the same sequence is repeated but with opposite polarity, thereby discharging capacitor C 2 instantly and causing the appearance of an electromagnetic field again. The electromagnetic field is of sufficient strength to be detected later by the signal detector.
 The preferred embodiment of the invention uses all above-mentioned subcircuits (FIGS. 3, 4, 5 , 7 and 10 ), with certain changes, in order to meet the particular needs of the final design. FIG. 8 shows a schematic of the power supply for the preferred embodiment's signal generator.
 Microcontroller U 2 (FIG. 9), preferably a Microchip PIC16C73, is the core of the first preferred design. Any other microcontroller that has similar performance capabilities can be used. Microcontroller U 2 has its own oscillator, power-on reset circuit (POR), ROM- and RAM memory on board. Therefore, only an RC network, consisting here of resistor R 14 and capacitor C 6 , connected to pin 9 of U 2 , is necessary to obtain the operating clock for the microcontroller U 2 . With the selected values for R 14 and C 2 (5 KΩ and 20 pf, respectively) and an operating voltage of 5VDC, microcontroller U 2 runs with an operating speed of approximately 4.12 MHz. Pin 1 (/MCLR) is connected directly to the 5V power supply since no additional circuitry is needed for POR. Digital input pins 13 and 14 are held by respective pull-up resistors R 12 and R 13 to +5VDC. Two momentary push buttons (S 1 and S 2 ) are employed in order to pull those input pins ( 13 and 14 ) to signal ground in order to indicate to microcontroller U 2 that a particular task is required. The activation of switch S 1 requires the task of performing a GFCI test, while the activation of switch S 2 requires the task of performing an AFCI test.
 The functionality of microcontroller U 2 is shown in flowchart form in FIGS. 18 a - 18 f . Referring to FIG. 18 a , the microcontroller U 2 at step 102 initializes and configures itself in accordance with values that are set in the software. Then, at subroutine 104 (FIG. 18 b ) it reads the values, as they are present, on its analog input pins 2 , 3 and 4 , respectively connected to voltage divider nodes 24 , 32 and 34 (FIG. 8). By reading the voltages present on those analog input pins, the type of branch circuit connected to the unit can be evaluated (E.g. 110V, 220V, or 440V). Furthermore, it is possible to analyze the wiring scenario or condition present at the receptacle where the device is connected. If a voltage is present on pins 2 and 4 of U 2 , but not on pin 3 , the wiring condition is correct. If voltage is present on pin 2 only, then the ground conductor is not connected. If voltage is present on pin 4 only, then the neutral conductor is not connected. If voltage is present on pins 2 and 3 , but not on pin 4 , then the hot and neutral conductors are reversed. If voltage is present on pins 3 and 4 , but not on pin 2 , then the hot and ground conductors are reversed. If voltage is present on pin 3 only, then the hot conductor is connected to the neutral terminal and the neutral conductor is not connected at all. If voltage is present on all three analog input pins ( 2 , 3 , and 4 ), then it is apparent that two hot conductors (two different phases) are connected to the outlet terminals. If the voltage on pin 2 is the double of the voltages on pins 3 and 4 , then two hot conductors are connected to the hot and neutral terminals in the outlet. If the voltage on pin 3 is the double of the voltage on pins 2 and 4 , then two hot conductors are connected to the neutral and ground terminals in the outlet. If the voltage on pin 4 is the double of the voltages on pins 2 and 3 , then two hot conductors are connected to the hot and ground terminals in the outlet. The result of this analysis will be stored in an internal register of microcontroller U 2 aptly named “wiring register”.
 Voltage present in any terminal of the outlet can also be read by the microcontroller U 2 on its digital input pins 5 , 6 , and 11 . This is done at steps 106 , 108 and 110 in FIG. 18 b . If voltage is present on any of those three input pins, but it falls periodically below 2 V (logical low), then zero crossing of the power line's sine wave cycle has been detected. If this is the case, then microcontroller U 2 starts immediately to count the time that elapses until another zero crossing is detected. If this time is approximately 8 ms, then a 60 Hz cycle is present. If this time is approximately 10 ms, then a 50 Hz cycle is present. The result of this evaluation is also stored in an internal register of U 2 named “frequency register”. The evaluation of the frequency is necessary in order to provide the capability for microcontroller U 2 to select the appropriate software part so that correct timing is established for the generation of the below-described identification signal.
 In the next task (FIG. 18 a ), the signal generator unit checks the electrical integrity of the ground conductor at subroutine 132 if (as determined at decision 134 ) a hot conductor is not connected to the ground terminal of the outlet. If a hot conductor is mistakenly connected to the ground terminal of the outlet, then this task will be skipped and microcontroller U 2 continues with other tasks as described below. The ground integrity test subroutine is shown in FIG. 18 e . In order to check the integrity of the ground conductor, at step 136 microcontroller U 2 's output pin 15 goes high, thereby triggering the gate of logical triac Q 6 . This causes current flow from +5V to ground through resistors R 15 and R 16 (FIG. 9). Those two resistors form a balanced voltage divider whose junction 112 is connected to microcontroller U 2 's analog input pin 7 . If the ground system is not compromised, then the voltage present on pin 7 of U 2 is exactly the half of the supplied 5VDC. If, due to material corrosion, poor mechanical connections (screw terminals), or other impacts, parasitic resistance is present in the grounding system, then this parasitic resistance adds on to resistor R 16 . If this is the case, then the voltage divider (R 15 /R 16 plus parasitic resistance) becomes unbalanced and the voltage on its junction 112 increases. If the degree of imbalance exceeds a stored threshold value which is equivalent to a predetermined resistance (such as 4 Ω), the program will declare that the ground lacks sufficient integrity. The 5 purpose of checking the integrity of the ground system is to make the user aware of a potentially hazardous condition, since the connected ground conductor is not a real earth ground which means that fairly high AC voltage can be present in the ground conductor.
 In the next task, all results of the above-described tasks will be displayed in distinct patterns at the incorporated LEDs D 7 , D 8 , and D 9 (steps 114 , 116 and 118 in FIG. 18 b ). Table I below illustrates the different lighting sequences of those three LEDs in accordance with the results of the above-described analyses.
TABLE I GREEN YELLOW RED LED D7 RED D8 LED D9 1. Correct wiring with fully ON ON OFF functioning earth Ground 2. Correct wiring but with earth ON FLASH- OFF Ground compromised ING 3. Open Ground ON OFF OFF 4. Open Neutral OFF ON OFF 5. Reversed Hot and Neutral ON OFF ON 6. Reversed Hot and Ground OFF ON ON 7. Hot on neutral with open neutral OFF OFF ON wire but an intact ground 8. Hot on neutral with open neutral OFF FLASH- ON wire but compromised ground ING 9. Two hot wires on the hot and FLASH- FLASH- OFF neutral terminal ING ING 10. Two hot wires on the hot and OFF FLASH- FLASH- ground terminals ING ING 11. Two hot wires on the neutral ON FLASH- FLASH- and ground terminals ING ING 12. Unenergized circuit OFF OFF OFF
 The unit is equipped with the capability to perform a reliability test of an eventually associated GFCI and/or an eventually associated AFCI. Those tests are useful only if the present wiring scenario is correct. If at step 122 (FIG. 18 a ) this is the case, then at step 124 the two incorporated dual color LEDs, D 10 and D 11 (FIG. 9, step 126 in FIG. 18 a ), will be lit in green, thereby indicating to the user that the two test functions are available. If the wiring scenario is not correct for any reason, then at steps 128 and 130 LEDs D 10 and D 11 will be lit in yellow, thereby indicating to the user that those two test functions are not available. If the user still tries to activate either of the two tests by pushing the momentary push buttons S 1 or S 2 , then microcontroller U 2 will ignore this request and will light either LED D 10 (S 1 ) or LED D 11 (S 2 ). If the two test functions are available and at step 160 (FIG. 18 c ) the user activates momentary push button S 1 (for the GFCI test), then microcontroller U 2 's output pin 15 goes high at step 164 , thereby (at 166 ) triggering the gate of thyristor Q 1 , thus allowing a current flow from hot to ground through current-limiting resistor R 17 . This current will be in the range required by appropriate authorities for the performance of GFCI tests. At the same time output pin 15 goes high, at step 168 microcontroller U 2 starts an internal timer with a resolution of 0.1 ms. If a predetermined time (such as 4 ms.; step 170 ) elapses (times out) from the time momentary push button S 1 is activated, then at step 172 the test function will be terminated by switching output pin 15 low, turning off thyristor Q 1 at 174 , and at step 176 LED D 16 b will be lit in red (step 178 ) in order to indicate to the user that the GFCI unit being tested has failed.
 If at step 190 (FIG. 18 a ) the user activates AFCI momentary push button S 2 , then at step 192 the ACFI test subroutine is executed (FIG. 18 d ). Microcontroller U 2 's output pin 12 goes high at step 196 , thereby (at 198 ) triggering the gate of thyristor Q 2 (FIG. 9), thus allowing a current flow from hot to neutral through current-limiting resistor R 25 and through spark gap GAP 1 . This causes the occurrence of real, sputtering arcing in the hot conductor of the electrical branch circuit under test. At the same time output pin 12 goes high, at step 200 microcontroller U 2 starts an internal timer with a resolution of 0.1 ms. If at step 202 a predetermined e.g. time (such as 4 ms.) elapses (times out) from the time momentary push button S 2 is activated, then at step 204 the test function will be terminated by switching output pin 12 low, turning off thyristor Q 2 at step 206 , and at step 210 LED D 17 b will be lit in red in order to indicate to the user that the AFCI unit being tested has failed to trip the circuit within the required time.
 The device also has the capability to locate the particular circuit-interrupting device that protects the electrical branch circuit under test. In order to perform this task, it is necessary to transmit an identification signal over the power line system, performed at subroutine 240 in FIG. 18 a and detailed in FIG. 18 f . This signal can be detected and evaluated by the later-described signal detector.
 Three subcircuits, comprised of (FIG. 9) thyristor Q 3 and capacitor C 7 , thyristor Q 4 and capacitor C 8 , and thyristor Q 5 and capacitor C 9 , are employed in order to generate this identification signal. The reason why three similar subcircuits are used is that any one of those subcircuits will be activated in accordance with the previously analyzed wiring scenario. Thyristor Q 3 and capacitor C 7 work from hot with respect to ground. Thyristor Q 4 and capacitor C 8 work from hot with respect to neutral. Thyristor Q 5 and capacitor C 9 work from neutral with respect to ground. When a zero crossing of a hot conductor has been detected in the above-described manner (steps 250 ), microcontroller U 2 starts an internal timer at 252 with a resolution of 0.1 ms in order to detect the phase angle of the power line sine wave. When the positive half-cycle of the sine wave reaches approximately 44 electrical degrees, then at step 258 any one of the microcontroller U 2 's output pins ( 16 , 17 or 18 ) goes high, thereby (at 260 ) triggering the gate of any of the three thyristors (Q 3 , Q 4 , or Q 5 ). When this happens, the associated capacitor (C 7 , C 8 , or C 9 ) will be charged relatively instantly, thereby causing the occurrence of a strong electromagnetic field for a duration of approximately 0.63 ms. When at step 266 the phase angle reaches approximately 90 electrical degrees, at step 268 the microcontroller U 2 switches the previously activated output pin low and, therefore, the energized thyristor goes out of conduction. The algorithm then loops. When a phase angle of approximately 224 electrical degrees has been reached, U 2 again switches its output pin high in order to trigger the gate of the thyristor, thereby forcing the thyristor into conduction. Since 224 degrees is in the negative half-cycle of the sine wave, the capacitor now relatively instantly discharges, thereby causing another occurrence of a strong electromagnetic field for a duration of approximately 0.63 ms. At 270 electrical degrees, the microcontroller's output pin will be switched low again and, therefore, the thyristor goes out of conduction. The same sequence happens during the following sine wave cycle. However, after the fourth half-cycle (step 272 ) the counter is cleared and no such pulsing sequence occurs. This three-wave cycle is repeated continuously in order to generate an obviously distinct and artificial pattern which is readily recognized by the later-described detector.
 Signal Detector
 This unit has the following features:
 1. It is a wireless AC voltage sensor.
 2. It is an automatic circuit-interrupting device locator.
 3. It features a “low battery” indicator.
 4. It features an automatic shut-off function.
 5. It features an automatic alert function.
 6. It features automatic switchover between different functions.
 The unit is a battery-powered hand-held device. As it is apparent from FIG. 13, it can be divided into six different function blocks that are named as follows:
 Power Supply 300 .
 Voltage Sensor 302 .
 RF Detector 304 .
 Peak Voltage Detector 306 .
 Microcontroller U 4 .
 Audible and Visible Indicator 308 .
 The core component of the unit conveniently is a Microchip PIC 16C710 microcontroller U 4 with the same features as the microcontroller utilized in the above-described signal generator, but with fewer I/O pins. Since the operating voltage of microcontroller U 4 needs to be in the range between 4 and 6VDC, a positive 5V voltage regulator (U 6 ) is employed. Electrolytic capacitor C 8 and capacitor C 9 , connected to the input of U 6 , are used in order to filter any possible ripple out of the 9V battery voltage. Electrolytic capacitor C 10 filters any possible ripple that might be present on the output of microcontroller U 4 . SPST switch S 3 serves as a simple power-on/power off switch. Components S 1 , C 8 , C 9 , C 10 and U 2 form a battery power supply 300 for the unit.
 A broadly tuned resonance tank, comprised of RF choke L 1 and capacitor C 10 in parallel, picks up the identification signal that is drawn over the power line system by the earlier-described signal generator. Transistor Q 7 and its associated components—capacitor C 11 , resistor R 26 , resistor R 27 , resistor R 28 and diode D 18 —act as a RF amplifier. The resonance tank and the amplifier together form an RF detector 304 . The output of this detector is, in one way, coupled through capacitor C 11 to the non-inverting input of operational amplifier U 3 A. In another way, it is also coupled through capacitor C 19 to pin 17 of microcontroller U 4 . Operational amplifier U 3 A—in conjunction with resistor R 29 , resistor R 30 , diode D 19 , diode D 20 , capacitor C 13 and electrolytic capacitor C 14 —form the above-mentioned peak voltage detector 306 . The output signal of the peak voltage detector is a DC voltage equivalent to the peak voltage of the RF signal which has been picked up and amplified by the RF detector. This output signal is fed through operational amplifier U 3 B to analog input pin 1 of microcontroller U 4 . Op Amp U 3 B is configured as a unity-gain amplifier (voltage follower) in order to provide high impedance for microcontroller U 4 's analog input. A resistor network, comprised of resistor R 31 and R 32 , acts as a balanced-voltage-divider and is connected across the battery voltage. Since it is a balanced voltage divider, the voltage present at the junction of this resistor network is exactly half the battery voltage. This voltage is fed to analog input pin 18 of microcontroller U 4 in order to enable U 4 to detect when this voltage falls below a predetermined value, which indicates a low battery voltage. An RC combination, comprising resistor R 33 and capacitor C 15 , is connected to pin 16 of U 4 . This RC combination determines the clock speed for the microcontroller. A dual-color LED (D 21 and D 22 ) is connected through two current-limiting resistors (R 35 and R 36 ) to U 4 's output pins 8 and 9 . Microcontroller U 4 's output pin 7 is connected to the base of transistor Q 8 . Transistor Q 8 acts as an amplifier for piezo buzzer BZ 1 . LEDs D 21 and D 22 , resistors R 35 and R 36 , transistor Q 8 and piezo BZ 1 form the above-mentioned audible and visual indicator 308 . Transistor Q 9 , whose base is connected to U 4 's output pin 6 , acts as a semiconductor switch in order to disconnect the analog part of the circuit from the battery power if required.
 A CMOS inverter (U 5 A), in conjunction with resistors R 37 and R 38 and tantalum capacitor C 20 , forms a low-frequency oscillator, which is triggered by varying an electromagnetic field received on antenna ANTI. Antenna ANTI is simply a small piece of metal or a copper zone on a printed circuit board (PCB). This antenna picks up a 50/60 Hz electromagnetic field. U 5 A's output is an oscillating signal whose frequency is a function of the strength of the 50/60 Hz electromagnetic field sensed by antenna ANTI. Resistor R 39 , connected across U 5 A's output pin 2 and signal ground, determines the sensitivity of the low-frequency oscillator. The oscillating signal is fed through four additional CMOS inverters (U 5 B, U 5 C, U 5 D and U 5 E) to U 4 's input pin 10 . The additional CMOS inverters are employed in order to amplify the oscillator's output. Antenna ANT 1 is used to determine whether an adjacent circuit is energized with AC power, while tuned tank L 1 , C 10 , etc. is meant to receive the identifying signal generated by the signal generator (FIG. 9).
 The flow diagram in FIGS. 19 ( a )-( c ) provides a better understanding of how the receiver unit works.
 When power gets applied to the unit through S 3 , microcontroller U 4 immediately starts operating in the same manner as the microcontroller U 2 in the above-described signal generator. After initialization at step 400 , U 4 at step 406 first measures the voltage present on its analog pin 18 in order to check if the battery voltage is sufficient. If the battery voltage is too low, then at 408 U 4 activates its output pins 7 and 8 (see FIG. 19 b ) in a predetermined pattern at 410 for approximately 5 seconds. This results in an intermittent flashing of LED D 22 (red) and an intermittent sounding of piezo BZ 1 ; this is the alert for low-battery status. After those 5 seconds, at step 412 microcontroller U 2 gets into a “sleep” mode, thereby drawing minimum current. If the user does not manually switch the unit off (S 3 ), then U 4 periodically—approximately every five minutes—activates LED D 22 (red) and piezo BZ 1 in a distinct predetermined pattern lasting approximately 5 seconds at step 416 . This is meant to be an alerting signal to the user in the event that the unit has been misplaced or forgotten.
 Returning to FIG. 19 a , if the battery voltage is sufficient, microcontroller U 4 switches its output pin 6 high, thereby turning on transistor Q 9 . This permits the battery voltage to be applied to the rest of the circuit. Microcontroller U 4 next switches its output pin 9 high, thereby lighting LED D 21 (green) to indicate that the unit is in standby mode. Then, at step 420 microcontroller U 4 checks for the presence of a signal at its input pin 10 in order (at 422 ) to detect if a 50/60 Hz electromagnetic field of a predetermined strength is present. If such a field is present, then the unit goes into AC voltage sensor mode, which is indicated by turning off green LED D 21 and turning on red LED D 22 . U 4 then generates a distinct sound and light pattern for LED D 22 and piezo trigger BZ 1 . The duty cycle of this pattern increases or decreases with respect to the strength of the detected electromagnetic field. The perceived effect of this indication sequence is similar to that of a Geiger counter. One technical advantage of this indication methodology, in addition to the flexibility of being able to sense electromagnetic fields of varying strengths, is that the user accomplishes it without the need for continuous, or any, manual adjustment. If, however, such an electromagnetic field is not present or detectable within a timeframe of approximately one second, then the unit remains or returns to standby mode. When the unit is in voltage sensor mode, it also checks for the presence of a signal on its input pin 17 . If there is a signal on this input pin, then U 4 measures the duration and rate of repetition of the signal. If those two parameters match the pattern of the identification signal generated by the signal generator, then the unit switches into breaker-locator mode. An LED will light in yellow to indicate this. Processor U 4 then measures the voltage present on its analog input pin 1 . If the voltage on pin 1 is equal to or greater than a predetermined threshold value, then this threshold value will be overwritten with the value currently present on pin 1 and a visible and audible alerting signal will be generated (D 21 , D 22 and BZ 1 ). Once a complete scan of the circuit-interrupting devices in question is done, as by moving the receiver to be proximate to each candidate circuit interrupting device, the highest detected voltage level becomes the current threshold value. During a second—or any subsequent—scan, this threshold value will be equal to the picked-up value of the identification signal only when the detector is adjacent to the particular circuit-interrupting device which had the strongest identification signal during the first scan. This identifies the circuit-interrupting device for the branch to which the transmitter is connected. If an identification signal cannot be detected anymore for a period of time of approximately four seconds, then the unit automatically switches back to voltage-sensor mode.
 If the unit is in standby mode for approximately five minutes, then U 4 switches its output pin 6 low, which switches transistor Q 9 off, thereby disconnecting the power for the entire circuitry, except the microcontroller itself. Microcontroller U 4 gets into “sleep” mode and the same process as mentioned above takes place.
 FIGS. 20 - 31 illustrate a further and even more particularly preferred embodiment of the invention. In this embodiment, the core component of the signal generator continues to be a microcontroller U 100 (FIG. 21) which provides a variety of digital and analogue input and output ports. FIG. 20 displays the schematic of a preferred transformerless power supply for the unit. Three full-wave bridge rectifiers D 100 , D 101 , and D 102 are connected across the HOT and NEU conductors (nodes 450 and 452 ), the HOT, and GROUND conductors (nodes 450 and 454 ) and the NEU and GND conductors (nodes 452 and 454 ). The employment of three separate bridge rectifiers is necessary because the unit needs to be powered under different possibly present wiring scenarios. Three voltage dividers, each of them comprised by a pair of resistors (R 100 and R 101 for D 100 , R 102 and R 103 for D 101 , and R 104 and R 105 for D 102 ) provide, at their respective center junctions 456 , 458 , 460 , DC voltages which are fed through three current limiting resistors (R 106 , R 107 , and R 108 ) to the input pins or nodes 462 , 464 , 466 of a digital to analogue converter which is incorporated into microcontroller U 100 (FIG. 21). The DC voltages present at the center junctions of those voltage dividers vary in accordance with the wiring scenario present at the power line outlet (receptacle) under test. Table II shows the different DC voltage values at multiple wiring scenarios.
TABLE II H-N H-G N-G Correct wiring with proper 2.09 VDC 2.05 VDC 1.04 VDC Earth-Ground Integrity Correct wiring with compromised 2.09 VDC 2.05 VDC 1.04 VDC Earth-Ground Open Earth-Ground 2.09 VDC 1.04 VDC 1.04 VDC Open Neutral 1.08 VDC 2.07 VDC 1.05 VDC Reversed Hot and Neutral 2.12 VDC 1.05 VDC 2.05 VDC Reversed Hot and Earth-Ground 1.07 VDC 2.08 VDC 2.08 VDC Hot on Neutral with open 1.18 VDC 1.05 VDC 2.07 VDC Neutral and proper Earth-Ground Hot on Neutral with open Neutral and 1.18 VDC 1.05 VDC 2.07 VDC compromised Earth-Ground Two hot wires on the Hot and Neutral 4.25 VDC 3.11 VDC 3.10 VDC terminals Two hot wires on the Hot and 3.17 VDC 4.16 VDC 3.12 VDC Earth-Ground terminals Unenergized circuit 0.00 VDC 0.00 VDC 0.00 VDC
 The positive outputs 468 , 470 , 472 of the bridge rectifiers D 100 , D 101 , D 102 are connected together, through three isolating diodes D 103 , D 104 , and D 105 . The three isolating diodes D 103 -D 105 are necessary in order to avoid a DC voltage feedback from any of the three sub circuits to another. Regardless of what wiring scenario is present, a DC voltage is always provided at the common junction 474 of those diodes which is further processed to power the entire circuitry. The only exception is if the outlet or line under test is not energized at all.
 The unit is alternatively powered by a 9V battery as mentioned above in order to insure that the unit can indicate an unenergized electrical branch circuit. FIG. 22 displays the schematic of such a battery power supply. This power supply is optional and could be omitted to save space and cost. The battery BT 1 is automatically switched on through optoisolator/voltage regulator U 101 if the unit is connected to an energized electrical branch circuit. Single pole, single throw (SPST) switch S 100 is employed in order to power the unit alternatively if the electrical branch unit under test is absolutely unenergized as when the unit is not connected to an electrical branch circuit at all. Electrolytic capacitor C 100 is employed in order to filter any low frequency ripple out of the battery supply voltage. Another voltage divider comprised by resistors R 109 and R 110 provides at its center junction 480 a DC voltage with a maximum value of +5V. This voltage is fed through current limiting resistor R 111 to an analogue input pin 482 of microcontroller U 100 (FIG. 21). If the supply battery voltage falls below a predetermined voltage (5.5V) the controller U 100 indicates the fact that the battery is low and switches after a time period of approximately 30 seconds the entire unit off because a reliable operation is not guaranteed in such a case. Returning to FIG. 20, voltage regulator U 101 provides at its output pin 484 a stable DC voltage of 5.1V which is the required operating voltage for microcontroller U 100 . Capacitor C 101 filters any high frequency ripple out of the supply voltage.
 All three power line conductors (HOT, NEU, and GND) are connected through current limiting resistors R 112 , R 113 , and R 114 to digital input pins 486 , 488 , 490 of microcontroller U 100 (FIG. 21). Three pairs of diodes (D 106 -D 107 , D 108 -D 109 , and D 110 -D 111 ) limit the power line AC voltage to a maximum voltage of 5.1V +/−0.7V (the breakover voltage of the diodes). This arrangement assures that microcontroller U 100 can detect when zero crossing of the power line sine wave occurs at either one or several of the three power line conductors.
 Referring to FIG. 21, an RC network comprised of resistor R 115 and capacitor C 102 in series determines the clock frequency for microcontroller U 100 (preferably 4.0 MHz). Another network comprised of resistors R 116 and R 117 , capacitor C 103 and diode D 112 , forms a subcircuit which assures that microcontroller U 100 is provided with power on reset (POR).
 The preferred embodiment of the transmitter unit includes three light emitting diodes (D 113 , D 114 , and D 115 ) of different colors (green, yellow, and red) which are lit in certain patterns by microcontroller U 100 in order to indicate the present wiring scenario or condition attached to the unit in accordance with Table I. Resistors R 118 , R 119 , and R 120 are respectively connected in series with LEDs D 113 , D 114 , and D 115 to limit the current for those three LEDs.
 The indication of a given wiring condition preferably also includes an indication of either a proper or a compromised earth ground connection of the electrical branch circuit under test. The analysis of earth ground integrity preferably works in the following way. The software for microcontroller U 100 provides a pulse-width modulated (PWM) signal of a frequency of approximately 450 kHz. This signal is fed through current limiting resistor R 121 to the anode of an LED which is incorporated into an optoisolator SO 100 . Optoisolator SO 100 provides a thyristor output having a terminal is connected to the HOT terminal of the power outlet being tested. The other terminal of the current path of this thyristor is connected through a coupling capacitor C 104 to one terminal of a primary winding of a transformer T 100 . The other terminal of the primary winding is directly connected to earth ground. Since the output thyristor of optoisolator SO 100 is controlled by its internal LED, which in turn is controlled by the above mentioned PWM signal, it follows the pattern of that PWM signal.
 Capacitor C 104 , the primary winding of transformer T 100 and the impedance of the earth ground, considered as a lossy conductor, form a phase shifting network. If parasitic impedance, which may have both real (resistance) and imaginary (reactance) components, and which can be caused by e.g. corroded or loose connections, is present at the earth ground connection, it adds on to the phase shifting network, thereby increasing the frequency which is seen at the primary winding of transformer T 100 . Transformer T 100 is a step-up transformer, which means that its secondary winding provides an output voltage which is higher than the voltage on the primary winding yet has the same frequency as the voltage on the primary winding. The signal present on the secondary winding of transformer T 100 is connected to a digital input pin 492 of microcontroller U 100 .
 Microcontroller U 100 's software counts the frequency and compares it to the frequency of the PWM signal. By analyzing the difference of those two frequencies, it can determine the value the impedance of the earth ground conductor. This in turn provides the capability to decide, by comparing the detected value to a predetermined stored value, whether the earth ground impedance exceeds permissible levels.
 Push buttons S 101 and S 102 are connected through current limiting resistors R 124 and R 125 to digital input pins 18 and 17 of microcontroller U 100 . Pull up resistors R 122 and R 123 hold those two digital input pins at a logical high level. If either of the two push buttons S 101 , S 102 gets activated, the level at that digital input pin which is associated with the activated push button gets pulled to a logical low level. This in turn activates a software routine, which triggers a test function for a possible associated ground fault circuit interrupter circuit (GFCI) or arc fault circuit interrupter circuit (AFCI). If push button S 101 gets activated, microcontroller U 100 's output pin 15 lights, through current limiting resistor R 127 , the LED which is enclosed in optoisolator SO 102 thereby forcing the thyristor output of optoisolator SO 102 into conduction. This causes the flow of a leakage current from the HOT conductor to the GND conductor through resistor R 133 . The value of resistor R 133 determines the amount of leakage current.
 LED D 116 is a dual color (green and red) LED with a common cathode. If the wiring scenario or condition present at the electrical outlet under test is correct, then microcontroller U 100 's output pin 25 goes high and lights the green part of LED D 116 a through current limiting resistor R 128 , thereby indicating that the GFCI test function is available. If the present wiring condition is incorrect, the LED D 116 a remains in an off state meaning the GFCI test function is not available. If the GFCI test function is available and gets requested by having push button S 101 activated then optoisolator SO 102 gets activated for a predetermined period of time. During this period of time microcontroller U 100 's output pins 24 and 25 both go high, thereby lighting both parts of LED D 116 a , D 116 b , which results in D 116 emitting a yellow color that indicates that the GFCI test function is activated. If the predetermined time elapses and the GFCI under tests does not trip, microcontroller U 100 's output pin 25 goes low while output pin 24 remains high for a predetermined period of time, and causing only red LED D 16 to light up, thereby indicating that the test has failed.
 Push button S 102 , in conjunction with microcontroller U 100 , current limiting resistors R 130 and R 131 and LED D 117 operates in the same manner as described above to show and indicate a test function for an AFCI device. If this test function is requested by having push button S 102 activated, optoisolator SO 101 goes into conduction, which in turn forces alternistor Q 100 into conduction. When alternistor Q 100 is in a conductive state then spark gap GAP 100 conducts as well, thereby causing current flow through load resistor R 132 from the HOT conductor to the NEU conductor. The value of load resistor R 132 determines the amount of current flowing. While this current is flowing a real sputtering arc is present at the electrical branch circuit under test. If the AFCI being tested does not trip within a predetermined period of time, the unit indicates that the AFCI has failed the test in the same manner as described for the GFCI test.
 Depending on the present wiring condition at the electrical branch circuit under test, one or several of microcontroller U 100 's digital input pins 11 , 12 , or 14 will see a logical high as long as the positive half cycle of the power line sine wave lasts. Microcontroller U 100 constantly monitors those three pins and as soon as it detects a zero crossing of the sine wave it starts counting the time in increments of microseconds. This allows calculating the current phase angle of the power line sine wave. When the phase angle reaches approximately 92 electrical degrees then microcontroller U 100 activates one of optoisolators SO 101 , SO 102 , or SO 103 depending on the wiring condition present. Activation of optoisolator SO 101 causes triac Q 100 to sink current from HOT, across GAP 100 , through resistor R 132 and to NEU. Activation of optoisolator SO 102 sinks current from HOT, directly from the emitter to the collector of the bipolar transistor incorporated therein, and thence through resistor R 133 to GND. Activation of optoisolator SO 103 in turn switches on triac Q 101 , which will then gate current from NEU, through resistor R 134 , to GND. The gating current to triacs Q 100 and Q 101 are respectively regulated by current-limiting resistors R 135 and R 136 . This in turn causes the current of significant magnitude and short duration (approximately 2 microseconds) to flow from HOT to NEU, from HOT to GND, or from NEU to GND depending on which one of the mentioned triacs or transistors is currently active. This procedure repeats for three positive half cycles of the power line sine wave and then pauses for another 2 positive half cycles. This pattern gets repeated continuously as long as the electrical branch unit in test is energized. As a result of this activity a unique identification signal is present at the electrical branch circuit being tested and at its associated circuit interrupting device (e.g. circuit breaker or fuse).
 FIG. 23 is an alternative embodiment of a transformerless power supply useful with the present invention. HOT, NEU and GND terminals are respectively connected through fuses F 150 , F 152 and F 154 to nodes 500 , 502 and 504 . Each of these nodes is in turn coupled to both digital and analog inputs of the microcontroller U 100 (see FIG. 21) through a combination of current limiting resistors and diodes, as affected by respective voltage divider circuits.
 Node 500 is connected through a current limiting resistor R 150 to a node 506 . An anode of a diode D 150 is connected to node 506 , which is also connected to a cathode of a diode D 152 . The anode of diode D 152 is connected to signal ground V ss which the cathode of diode D 150 is connected to positive DC voltage V dd . Node 506 is connected to a digital input pin of microcontroller U 100 (FIG. 21). Node 500 is also connected to the anode of a diode D 154 and the cathode of a diode D 156 . The anode of diode D 156 is connected to signal ground V ss . The cathode of diode D 154 is connected to a node 508 . The anode of a diode D 158 is connected to the node 508 , while its cathode is connected to a common node 510 which in turn serves as the input to a voltage regulator or power supply integrated circuit U 150 . Node 508 is connected through a resistor R 152 to a node 512 , which is connected to signal ground by a capacitor C 150 and a further resistor R 154 . A resistor R 156 connects node 512 to an analog input of the microcontroller U 100 (FIG. 21).
 The connections of the NEUTRAL and GROUND terminals to the power supply are analogous. Node 502 is connected through a resistor R 158 to a microcontroller digital input node 514 . The node 514 is clamped between diodes D 160 and D 162 , similar to the clamping diodes D 150 and D 152 of node 506 . Node 502 is also fed to the anode of a diode D 164 , the cathode of which is connected to a node 516 . Node 516 is connected to the anode of a diode D 166 which is connected in turn to common power input node 510 . Node 516 is further connected through a resistor R 160 to a node 518 . Node 518 is connected to signal ground by a capacitor C 152 and a resistor R 162 . A current limiting resistor R 164 connects node 518 to an analog input of the microcontroller U 100 .
 Node 504 is connected through a current limiting R 166 to a node 520 , which is clamped between diodes D 168 and D 170 that are respectively connected to the positive DC voltage V dd and the signal ground V ss . Node 520 is connected to a digital input pin of the microcontroller U 100 . Node 504 is further connected to the anode of a diode D 172 and the cathode of a diode D 174 . The anode of diode D 174 is connected to signal ground. The cathode of diode D 172 is connected to a node 522 , which in turn is connected through a resistor R 168 to a supply node 524 . Node 524 is connected through a current limiting resistor R 170 to an analog input of the microcontroller U 100 . The node 524 is also connected by a capacitor C 154 and a resistor R 172 to signal ground. Node 552 is connected to the anode of a diode D 176 , the cathode of which is connected to the common node 510 .
 Node 510 is further connected to the cathode of a Zener diode D 178 and to a capacitor C 156 . The anode of the Zener diode 178 and the capacitor C 156 are both connected to signal ground. The power supply integrated circuit U 150 has an output node 526 that provides positive DC voltage to the rest of the transmitter circuit. Filtering capacitors C 158 and C 160 connect the output node to signal ground.
 The power supply circuit in FIG. 23 will supply positive DC voltage to the rest of the circuit as long as one of the hot, neutral and ground conductors is energized and another one provides a return path. The analog and digital inputs provided by the power supply permit the microcontroller to sense any of twelve different wiring conditions present at the outlet, as above described.
 All of the above described functions are available for both commonly known power line systems which are 60 Hz/120 V for North America and Asia and 50 Hz/220V for Europe, Australia and Africa. The unit adjusts itself automatically to the power line system where it gets connected.
 An alternative and particularly preferred embodiment of a signal detector 550 according to the invention is shown in FIG. 24. Unit 550 is a microprocessor controlled, battery powered hand held apparatus. Pick up coil L 200 in conjunction with capacitor C 200 forms a tank whose resonance frequency is broadly tuned to the frequency of the identification signal which is generated by the above described generator (see FIGS. 20 - 23 ). The sensed frequency is AC coupled through capacitor C 202 to the inverting input of operational amplifier U 200 that is configured as a noninverting preamplifier with a gain of 100. Operational amplifier U 201 is another noninverting amplifier which further amplifies the signal which is available at the output 552 of op amp U 200 with a gain of 20. The signal available at the output of op amp U 201 is fed through a diode D 200 to the inverting input of operational amplifier U 202 that is configured as a unity gain voltage follower. The output of op amp U 202 is connected to a capture and compare input pin of microcontroller U 204 . The software associated with microcontroller U 204 compares the appearance pattern of the signal with the expected pattern as it was generated in the manner described above by the signal generator. If the two patterns match it is assumed that the detected signal is truly the identification signal generated by the signal generator. If this is the case microcontroller U 204 activates LED D 201 and buzzer BZ 200 . The visual and audible alert is meant to indicate that the circuit interrupting device in question has been located. If the two patterns do not match it is assumed that the detected signal is not the identification signal generated by the signal generator but rather some kind of electrical power line noise. If this is the case the signal will be ignored.
 Inverter U 205 is configured as a low frequency square wave generator. Antenna ANT 200 is a simple metal pad with a resonance band in the range of approximately 45 to 65 Hz. If the unit approaches an electrical power line circuit the antenna senses the magnetic field generated by the power line sine wave. This causes a frequency shift of the square wave generated by oscillator U 205 . The frequency shift increases equivalently to the strengths of the sense magnetic field. The inverters U 206 , U 207 , U 208 , and U 209 are simple frequency amplifiers. The output signal of inverter U 209 is connected through triac Q 200 to a digital input pin of microcontroller U 204 that is configured as a frequency counter. If a magnetic power line field has been detected, microcontroller U 204 again activates LED D 201 and buzzer BZ 200 yet with a different pitch in order to distinguish the two different signals from each other. The duty cycle of the controller frequency changes in accordance to the strengths of the detected magnetic field, which means that the control frequency for LED D 201 and buzzer BZ 200 increases as stronger as the sensed magnetic field appears.
 FIG. 25 is a flow chart showing the operation of microcontroller U 100 in FIG. 21. From a start condition 600 the program will query whether pin 11 (hot), pin 12 (NEU) and pin 14 (GND) are high at decision steps 602 , 604 and 606 . Depending on the voltage states which are sensed, LEDs D 114 , D 115 and D 116 a will be switched on or off at steps 608 - 618 .
 At decision step 620 the program analyzed whether the wiring scenario or condition is as it should be, i.e., with pin 11 high and pins 12 and 14 low. If this is the case, the program branches to step 624 , at which the GFSI “ready” flag is set, the AFCI “ready” flag is set, and LEDs D 116 b and D 117 b are switched on. If, on the other hand, the program determines that the wiring condition is incorrect, all “ready” and “switch” flags are cleared at step 626 , and both elements of LED pairs D 116 a , D 116 b , D 117 a and D 117 b are turned on; the energization of the red and green LED elements will produce a yellow color.
 Assuming that the wiring condition is correct, at step 628 the microcontroller U 100 will determine whether or not the operator has pushed push button S 101 , indicating that a GFCI test ought to be performed at step 630 . The program next inspects push button S 102 to determine whether the operator has pressed it. If so, the program branches to step 634 for the initiation of an AFCI test. Subroutines for these tests are similar to those shown in FIGS. 18 c and 18 d.
 FIG. 26 shows a TEST subroutine 640 that is implemented by microcontroller U 204 (FIG. 24) upon power up. At step 642 , D 202 and a first DTMF tone will be generated for about two seconds. At step 644 the other LED D 201 will be actuated, and a second DTMF tone will be emitted from the buzzer BZ 2000 for about two seconds. Then, at step 646 , both LEDs D 202 and D 201 will be actuated and a DTMF tone will be generated by the buzzer BZ 200 for about two seconds. The procedure then returns at step 648 .
 FIG. 28 is a flow chart of subroutine TRACER which begins at step 650 . This subroutine is entered once an energized circuit has been found by SENSOR (described below).
 A standby timer, average counter, and average buffer are cleared at steps 652 , 654 and 656 by microcontroller U 204 (FIG. 24). At step 658 , analog input pin 17 is read and stored to a value ADRes. This value is added to the average buffer at step 660 . At step 662 , the average counter is incremented and at step 664 is compared against a count value of four. Steps 658 - 664 are repeated for four cycles. A step 666 , the average of four readings is calculated. A step 668 , the subroutine asks whether this average is greater than or equal to the value presently stored in the tracer buffer. If not, the program returns at 670 . If, however, the sensed and averaged value exceeds the present value in the tracer buffer, this new value is stored in the tracer buffer at step 672 . At step 674 , the program announces to the user the presence of the strongest-yet magnetic field detected. At step 676 , a low battery subroutine is called. This low battery subroutine is illustrated in FIG. 28. At step 678 , pin 14 (V dd ) is read and at step 680 compared against a stored minimum DC voltage value such as 3 volts. If sufficient DC voltage is being supplied to the microcontroller U 204 , the subroutine returns at step 682 . If the voltage has decreased below 3 volts, the power is shut off at step 684 . Both LEDs D 201 and D 202 are flashed at step 686 and the piezo buzzer BZ 200 is actuated. The program then goes to a READTIMER( ) subroutine at step 688 .
 Assuming that the subroutine LOWBATTERY returns after making sure that the DC voltage has not declined below 3 volts, at step 690 in FIG. 27, the program goes to a STANDBY mode. This subroutine is shown in FIG. 29. In STANDBY mode, the program switches LEDs D 202 and D 201 on but switches the piezo buzzer BZ 200 off.
 FIG. 30 illustrates a SENSOR subroutine, which begins at step 700 . This subroutine is used to find, and to quantify, a nearby energized circuit picked up by antenna ANT 200 (FIG. 24). At step 702 , the standby timer is cleared and at step 704 an analog input of the microcontroller U 204 is read. At step 706 , variable ADRes is prepared as a proportional value for the sound pattern. That is, a lookup table is accessed which correlates different values of ADRes to different sound frequencies. At step 708 , LED 201 is flashed and at the same time a DTMF tone is generated. After calling the LOWBATTERY subroutine, the procedure exits to STANDBY mode.
 FIG. 31 is a flow diagram of a POWEROFF subroutine which begins at step 750 . LEDs D 201 and D 202 are shut off, and the piezo buzzer is shut off, at step 752 . The processor is put to sleep at step 754 .
 FIG. 32 is a further embodiment of an arc fault testing apparatus according to the invention. A spark gap GAP 250 has a first terminal connected to a HOT terminal of the device. A second terminal of the spark gap GAP 250 is connected through a resistor R 250 to a node 800 . The current path of a triac Q 250 connects the node 800 to a node 802 , which in turn is connected to the NEUTRAL terminal of the device. A capacitor C 250 connects node 802 to node 804 . A resistor R 252 connects node 804 to node 800 . A further resistor R 254 connects node 804 to one side of a thyristor component of an optoisolator U 250 ; the second end of this current path is connected as a gate to triac Q 250 . A fuse F 250 connects the HOT terminal to a node 806 , which is connected by a capacitor C 252 and a resistor R 256 to a node 808 . This is connected to a first AC input of a diode bridge D 250 . NEUTRAL node 802 serves as a second AC input to the diode bridge D 250 . The diode bridge D 250 provides a full wave rectified DC signal on a positive output node 810 as referenced against a negative DC output voltage node 812 . A Zener diode D 252 , a capacitor C 254 , and a capacitor C 256 connect positive and reference DC nodes 810 and 812 together. DC positive power node 810 is fed to a power input of an integrated circuit U 252 . Node 812 is connected to signal ground and a further input of the integrated circuit 252 . Integrated circuit U 252 has a pair of LED driving outputs which drive LEDs D 254 and D 256 through respective resistors R 258 and R 260 . Pin 5 of the integrated circuit U 252 is connected through a resistor R 262 back to node 806 .
 In operation, the integrated circuit U 252 is able to sense whether or not a voltage exists across the hot and neutral terminals, and selectively actuates diodes D 254 and D 256 to tell the user whether or not there is a difference in electric potential between these two points. When the user wishes to simulate an arc fault for testing purposes, the user presses push button S 250 , which will cause a signal to be received through resistor R 264 to an input of the controller U 252 . Upon actuation of the signal the controller will bring its pin 3 high. Current will then flow through a current limiting resistor R 266 to the light emitting diode component of optoisolator U 250 and thence to ground. This actuates the thyristor component of optoisolator U 250 to gate triac Q 250 to an on condition. This causes a short circuit between hot and neutral terminals and an arc across GAP 250 , thereby transmitting an arcing condition back to the branch circuit being tested.
 The relatively simple embodiment shown in FIG. 32 is suitable for inclusion with third party wiring testing circuitry.
 While this invention is susceptible of embodiments in many different forms, only certain embodiments of the invention have been described, with the understanding that the present disclosure is to be considered as exemplary illustrations of the principles of the invention and is not intended to limit the broad aspects of the invention to the embodiments illustrated. The present invention is limited only by the scope and spirit of the appended claims.