JP2001518603A - Electric arc monitoring system - Google Patents

Abstract

(57) [Abstract] Electric arc monitoring uses the discovery that electric arcs are fractal phenomena that contain all the essential information that means the "arc" contained within each fractal subset. Achieved. These fractal subsets are logarithmically distributed over the arc spectrum. The monitoring of the arc is such that the amplitude is higher according to the l / f characteristic of the electric arc (22), the mutual induction between adjacent circuits is lower, and the movement between the arc and the arc signature pickup (23) is used for arc detection. It is most advantageously achieved with a low log order fractal subset longer than the customary high frequency. Fractal subset transformation reduces the risk of false alarms. The arc signature portion may be processed on an out-of-phase path or treated as a modulated carrier.

Description

DETAILED DESCRIPTION OF THE INVENTION

TECHNICAL FIELD [0001] The technical field of the invention includes methods and equipment for monitoring, detecting, indicating, evaluating, and signaling electric arcs or sparks.

BACKGROUND OF THE INVENTION Disordered electromagnetic emanations, manifested as electric arcs or sparks, are closely linked to matter, where electromagnetic interactions couple electrons to atoms and nuclei within molecules, causing the generation of electromagnetic radiation. The fundamental unit is a photon.

[0003] In practice, the spectrum of electric arcs and sparks actually extends from DC through the entire radio frequency spectrum and through the microwave infrared and optical spectrums.

The effective use of electric arcs and spark phenomena include electric arc lamps, electric welding, electric arc type metallurgical furnaces, arc type ion generators for propulsion in satellite reaction propulsion engines and in outer space, internal combustion Includes spark plug type ignition in engines and electric spark ignition in gas appliances.

[0005] Unfortunately, the same qualities of electric lighting, electric arc welding and metallurgy, as well as electric arcs or sparks, which have led to the ignition of internal combustion engines, have caused explosion or destruction by the generation or sparking of chaotic arcs. It has catastrophic effects in terms of electrical failures that cause fires.

[0006] By way of example, electric arc monitors are garages, automobile or electric vehicle repair facilities, gasoline ("petrol" in British English) storage or dispensing facilities, and accidental electric arcs are costly. It will work in other areas that can cause an explosion.

[0007] Fuses and circuit breakers can also prevent severe overload conditions, but they generally involve electrical fires and, typically, blown fuses and circuit breakers. It is not effective in preventing accidental arcs and other damage from sparks that generate enough heat for fires at electrical current levels below the tripping level. Therefore, reliable arc monitoring would be highly desirable in most diverse electrical circuits.

These are, of course, only representative examples of the areas where reliable arc or spark monitoring is effective.

A major stagnant problem in this regard is that the development with the prior art followed a natural course in fear of false alarms. Of course, false alarms are a bane of alarm systems, since frequent occurrences of false alarms can negate the utility of the alarm system.

[0010] Thus, in an effort to reduce the possibility of false alarms arising from radio broadcast and radio frequency security system signals, an arc detection system such as that disclosed in International Patent Publication No. WO 90/04278 by HAMPSHIRE, Michael John. Reject frequencies below about 160 kHz and above about 180 kHz of the arc signal signature, leaving electrical fault detection only to a narrow 20 kHz band at a center frequency of about 170 kHz.
However, this leaves an arc detection sample that is several tens of times too small in the 100 kHz range to reliably detect the occurrence of the arc signature while at the same time reliably preventing false alarms.

An arc detection system that avoids its shortcomings is hereby incorporated by reference for the United States, and for all other countries where incorporation by reference is permitted, Oct. 24, 1990. And submitted to WO92 / 08
No. 143, Hendry Mechanical Works (Hendry M)
technical Works), HAM, Jr. , Howard M
, And KEENAN, James J. et al. PCT / US90 / 06113, and US Pat. No. 5,37, issued Dec. 13, 1994.
3,241 and 5,477,15 issued on December 19, 1995.
No. 0 is evident. Also, references are incorporated by reference for all countries where incorporation by reference is permitted herein, together with their corresponding EPO 507,782 (9099177.88.8), And in accordance with the resulting European Patent and their corresponding Australian Patent No. 65
No. 6128, Canadian Patent Application No. 2,093,420, Chinese Patent Application No. 9210
No. 253.3, Japanese Patent Application No. 500428/91, Korean Patent Application (PCT)
No. 701219/93, and Mexican Patent No. 178914 (No. 92015)
No. 30). The system provides a false alarm by converting the instantaneous arc signature frequency to a combined frequency where the signal indicative of the arc is detected in contrast to an extraneous narrowband signal that may cause a false alarm. To avoid.

[0012] Contrary to this background, the frequency-selective arc detection system of the prior art filed subsequently filed emerges as a typical representative of the prior art approach to arc detection. Thus, it presents a wide variety of approaches to arc detection that primarily look at frequencies within the upper kilohertz range, such as from 100 kHz to 1 megahertz. However, this is not the case with the public A.I.A. M. It covers not only the majority of the radio broadcast band, but also the type of radio frequency band of the control system or security system referred to in the aforementioned WO 90/04278. Thus, in some places, one had to compete with dozens of extraneous signal interferences.

[0013] Similarly, a comb filter array consisting essentially of four bandpass filters, each of which has a passband of 50 kHz, three of which have center frequencies of 225 kHz, 525 kHz, and 825 kHz, respectively. This applies to another embodiment in terms of its prior art proposals which propose the use of A. M
. In the radio frequency band portion of the broadcast and its control and security systems described above, the 50 kHz sample only represents a minor piece of the disordered arc signature, and the danger of false alarms from contemporaneous extraneous signals. Enhance. Also, this is because the prior art proposal suggested that the 55 kHz bandpass filter in the comb filter array be used to continuously rotate the detection process between the four filter components of the comb filter array. Affects potency.

[0014] "Arc-generating fault detection using low-frequency current components" by B, D, Russell et al. Shows that previous efforts at arc detection that have taken the plunge into the low-frequency region have been unsuccessful. Techniques-Performance Evaluation Using Recorded Field Data "and" Operation of the Low Frequency Spectrum During Arcing Faults and Switching Events "(IEEE Minutes on Power Delivery, Vol. 3, No. 4, Oct. 1988)
(Monday, pages 1485-1500) achieved monitoring in a variety of low frequency bands that were too narrow for reliable arc detection.

In retrospect, these developments emerge primarily as a response to the perception of electric arcs as a very random phenomenon arising from the disordered nature of arc signatures. However, this prior art perception suggests that chaotic systems have a deterministic quality, so that if one were to understand the underlying principles and how to use them effectively, I ignore the fact that if I can find out what to do, I can cope safely.

In fact, even disordered electrical lightning exhibits some self-similarity during its arborsque night discharge and within the branched configuration of its lightning volts.

In this regard, by Benjamin Franklin and Ge
Pioneering work done in the 18th century by org Christoph Lichtenberg casts a long shadow on the subject invention.

In particular, Franklin has proven through lightning experiments in thunderstorms that lightning is an electrical phenomenon. Lichtenberg then created the famous "Lichtenberg figure" in 1777 by sprinkling a fine powder, such as sulfur, over the insulating surface where the electrical discharge occurred. Many of these Lichtenberg numbers of electrical discharges resemble lightning in appearance, and otherwise exhibit significant self-similarity in their branch line patterns and within such branch lines themselves. ing. Manfred Schr
Oeder describes this as "the law of fractals, chaos, and power (FRACTALS).
, CHAOS, POWER LAWS) "(WH Freeman (WHF)
Leeman and Company), 1991, 196, 197, 21.
5 and 216) in the book diffusion-limited set (diffusio
Compared to n-limited aggregation (DLA). Also,
Kenneth Falconer also applied the DLA model to electrical discharges in gases in a book entitled "Fractal Geometry" (John Wiley & Sons, 1990), pages 270-273.

By way of background, fractals were introduced in 1975 by Benoit Mandelbro.
It is a phenomenon in fractal geometry that was considered, named, and first described by t. Fractal geometry is actually a manifestation of the fact that nature does not conform to Euclidean geometry. Euclidean geometry is based on characteristic size and scaling. Nature is not restricted to any particular size or scaling. Euclidean geometry is suitable for artificial objects, but cannot realistically represent natural configurations. Euclidean geometry is described by formulas, whereas the mathematical language of natural phenomena is a recursive algorithm.

Such recursiveness is a natural expression of the entire destructive effect, even if not disordered, such as self-similarity or self-affinity
) Appear in a persistent invariance to changes in size and scaling, called). Fractals are self-similar in that each of the various smaller parts of the fractal represents an entire reduced replica. Such small portions are referred to herein as "fractal subsets."

[0021] Electrical arc or spark monitoring generally refers to the electromagnetic spectrum of the arc or spark, which itself lies in a frequency band well below the light, thermal radiation, and microwave spectrum. Turn to so-called arc signatures that are part of

Problems in this area include false alarms from mutual induction between adjacent monitored circuits. In this regard, the reference is a standard for mutual guidance, such as between a monitored circuit where an arc is occurring and an adjacent monitored circuit where the arc is not occurring at that time. May be held according to equations. I _n = 2πfMI _as / Z _n (1) where I _as = arc signature current flowing in the monitored circuit _{where an} electric arc occurs at the moment; I _n = arc occurs at the moment and it is induced by the arc signature in adjacent circuit being monitored no current; M = mutual inductance; Z _n = impedance of the adjacent circuits; and f = frequency

In adjacent monitored circuits, such as between adjacent circuits, the current I _n induced by the current I _as flowing in the monitored circuit where the arc is occurring is , Decreases with the decreasing frequency of the current _Ias . However, electromagnetic arc signatures are characterized by a special shape that approximates the inverse frequency (l / f) progression of their amplitude. If this were put into equation (1) above, the following would be obtained: I _n = (2πfMI _as / f) / Z _n (2) where “f” is obliterated, so that I _n = 2πMI _as / Z _n (3) that is, no mutual inductance and the secondary current I _n related to frequency. Such considerations have led to the prior art conclusion that reducing the frequency of the arc signature band for which the arc is being monitored will not effectively reduce cross-induction and false alarms therefrom.

[0024]

SUMMARY OF THE INVENTION

SUMMARY OF THE INVENTION It is a general object of the present invention to provide an improved electrical arc monitoring system that utilizes new circuitry and / or takes advantage of the characteristics of previously unused electrical arc signatures. is there.

It is a related object of embodiments of the present invention to provide a longer arc from the electrical arc and reliable arc monitoring with less crosstalk or induction.

In this regard, and generally, the expression “monitoring” includes monitoring, detecting, indicating, evaluating, and / or signaling an electrical arc or spark, where Although used broadly, the term "arc" is used generically herein to cover electrical arcs and sparks so that they can be interchanged as being essentially the same phenomenon.

From one aspect, the subject invention is not limited to the fact that the electric arc is not only a fractal phenomenon in the visible bright part of the electromagnetic radiation as previously thought, but also in fact a spatial It exploits the discovery that it is a fractal phenomenon that falls far down to the very low frequency of its electromagnetic emanation into or along the wires of the circuit where the particular arc occurs. Therefore, for arc monitoring purposes, it is sufficient to monitor the fractal subset of the electromagnetic emanation of the arc, since all the essential information meaning "arc" is contained in each fractal subset. .

The realization according to the subject invention, in which the fractal nature of the arc is not restricted to its visible range, but actually extends down to several cycles per second of its signature, has been achieved with at least one fundamental feature and at least one One criterion, namely the following, is added to the previously known electrical arc characteristics.

[0029] 1. All essential information for effective electric arc monitoring is contained in any fractal subset of the arc signature, thereby: The choice of monitoring frequency band for each purpose is relieved of the limitations of the prior art, and is truly true of sensitivity, detection speed, prevention or rejection of spurious signals, desired travel length, and arc signatures in different environments. An optimal tradeoff in terms of the mode of transmission from the arc to the monitoring circuit can be the result.

In accordance with these principles, the subject invention resides in a system for monitoring an electrical arc having an arc signature typified by a broadband range of frequencies of random nature in the circuit being monitored, and more particularly, The task is then to select a fractal subset of the arc signature that is characterized by relatively long travels along the circuit being monitored and monitor that fractal subset of the arc signature.

The expression “relatively” in this context refers to the fact that the possible length of the movement of the arc signal is inversely proportional to the frequency of the arc signature. In this regard, reference may be made to well-known algebraic equations of electrical current. I = E / [R ² + (2πfL−1 / 2πfC) ² ] ^1/3 (4) where I = electric current E = voltage or potential R = resistance f = frequency L = inductance, and C = electric circuit Capacitance

From this equation, the relevant advantages of embodiments of the present invention can be ascertained. That is, the selection of the lowest frequency or longest wavelength fractal is, in fact, substantially equal to the selection of the longest remnants of the various fractals of the arc signature traveling along the circuit being monitored. To a certain point, one can say that the circuit being monitored thus performs the function of a low-pass filter for the arc detection monitor itself. Thus, embodiments of the present invention include the ability to survey large circuits and the convenience of providing a central arc detection and monitoring station for a plurality of different circuits.
In practice it allows arc monitoring at a considerable distance from the occurrence of an arc in a circuit that is useful for several reasons.

In any event, at low arc signature frequencies, the possible movement of the arc signal along the circuit being monitored is longer compared to higher arc signature frequencies.

Also, false alarms from mutual induction between adjacent monitored circuits, where the frequency coefficient “f” in the denominator will obliterate the “f” in the numerator of equation (2) It can also be seen that, contrary to what the prior art showed according to equations (1) to (3) described above, it is lowest at low arc signature frequencies.

However, misleading conclusions become apparent when certain possible radiation effects are considered. In this regard, the λ / 2 and λ / 4 antennas have excellent Hertzian and Marconi types.
It is well known to constitute types of electromagnetic radiators. The wiring in many telephone exchanges, electrical power supplies and other equipment is, in fact, often such antennas at the type of radio frequency selected by the prior art for electrical arc detection. To form The constant reactance in the circuit makes relatively short conductors an effective radiator, even if the wiring lengths in the equipment are not sufficient to make up a quarter-wave antenna. An adjustment or "load" of the lumped impedance type can be provided.

As a result, unless the segment of the arc signature picked up for monitoring is a very long wave (VLF) in accordance with one aspect of the present invention, the picked up electromagnetic arc signature propagates between adjacent circuits. And give rise to false alarms.

Thus, lower frequency fractals in accordance with embodiments of the present invention induce less spurious signals through mutual induction in adjacent circuits than arc signatures having higher frequencies. Low frequency fractals more effectively avoid false alarms from mutual inductance between adjacent circuits than arc signatures at higher frequencies.

As a result, embodiments of the present invention not only allow arc monitoring a significant distance from the occurrence of an arc in the monitored circuit, but also provide false alarms in adjacent monitored circuits. To avoid.

According to a related embodiment of the invention, an electric arc is detected from a fractal subset of the arc signature at a frequency below 30 kHz. New IEEE Electrical and Electronic Terminology Standard Dictionary, 5th Edition (The New IEEE Standard D)
Ictionary of Electric and Electron
ics Terms, Fifth Edition) (IEEJ, 199)
3 years), this is the upper limit of the very long wave (VLF) band.

A currently preferred embodiment of the present invention is a fractal part in which an electric arc is detected up to an ELF (extremely low frequency) defined in the IEEE standard dictionary as extending from 3 Hz to 3 kHz. Restrict sets.

Another embodiment of the present invention is to monitor the fractal being monitored to an arc signature frequency below the audio frequency band (vf) defined in its IEEE standard dictionary as extending from 200 Hz to 3500 Hz. Limit the fractals that are present.

In its vein, additional embodiments of the present invention limit the fractal subset being monitored to an arc signature frequency below a first high frequency of the standard line frequency in an AC power system.

Embodiments of the present invention also select an arc signature fractal subset that is being monitored from a frequency that is approximately a standard line frequency in an AC power supply system.

According to a related aspect of the invention, an apparatus for monitoring an electric arc having an arc signature categorized by a broadband range of frequencies of the disordered nature of the circuit being monitored is combined with the arc. Having a passband according to a fractal subset of the arc signature characterized by a relatively long travel along the circuit being monitored, the fractal portion of the arc signature An electrical filter having an output for assembly is provided. Such equipment includes a chaotic theory broadband signal detector having a detector input for its fractal subset of the arc signature coupled to the output of the electrical filter.

From another aspect, the present invention is a method of monitoring an electric arc having an arc signature extending over a broadband range of frequencies of the disordered nature of the circuit being monitored. The invention in accordance with this aspect more particularly relates to the processing of the arc signature in two paths, out of phase with each other, in combination, and the monitoring of the electric arc from such out of phase portions of the arc signature There is a remedy that includes

From its related aspect, the present invention resides in an apparatus for monitoring an electric arc having an arc signature typified by a broadband range of disordered frequencies in a circuit being monitored. The invention according to this aspect, more particularly,
In combination, an electrical filter having an input coupled to the arc, having a passband corresponding to a portion of the arc signature, and having an output for such portion of the arc signature, connected to an output of the electrical filter. An inverting amplifier having an input and an amplifier output, a non-inverting (non-inverting) amplifier having an input connected to the output of the electrical filter, having an amplifier output, and in parallel with the inverting amplifier. Inverting amplifiers and chaotic theoretical broadband signal detectors with detector inputs coupled to the amplifier outputs of the inverting and non-inverting amplifiers.

From another aspect, the present invention resides in a method of monitoring an electric arc having an arc signature extending over a broadband range of disordered frequencies in a circuit being monitored, and more specifically, There are improvements that include treating the arc signature as a modulated carrier having a modulation indicative of an electric arc in combination, and monitoring the electric arc by monitoring the modulation of the modulated carrier.

From its related aspects, the present invention resides in an apparatus for monitoring an electric arc having an arc signature typified by a broadband range of disordered frequencies in a circuit being monitored, and more specifically, Specifically, there is an improvement comprising, in combination, a modulated carrier detector having an arc signature input and a carrier modulation output.

From its analogous aspects, the present invention resides in an apparatus for monitoring an electric arc having an arc signature typified by a broadband range of frequencies of disordered nature in a circuit being monitored, and more particularly to an apparatus for monitoring an electric arc. Thus, in combination there is an improvement comprising a combined modulated carrier detector having an arc signature input and a combined carrier modulation output.

FIG. 1 of the accompanying drawings is copyright as the first creation under the Bern Convention and all corresponding national laws, and is provided by Hendry Mechanical Works, Goleta, Calif., United States of America. Hendry Mechanical Works) is the copyright owner who understands that this figure is published by the World Intellectual Property Organization and subsequently by patent offices worldwide.

[0051]

Mode for carrying out the present invention

The drawings also illustrate some basic modes and preferred modes of implementing certain aspects of the present invention. Since fractal geometry is a technique as visual as mathematical science, FIG. 1 illustrates the practice of the invention in terms of logarithmic spirals (worki).
ngs). This is a novel aspect because the arc signature spectrum is conventionally plotted in Cartesian coordinates in terms of frequency. Though the frequency of the arc signature is largely independent of the medium in which it travels, to a large extent the thinking and plotting in terms of frequency were justified, but the wavelength of the arc signature traversed more directly. Depends on the media being used.

The traditional focus of the prior art on frequency has been to initially obscure the visualization of the arc signature as a logarithmic phenomenon of fractal nature, but in terms of wavelength according to the presently described aspects of the present invention. Thinking and plotting in, leads to visualization, graphical notation, and useful use of the phenomenon.

To this effect, the polar coordinate system notation of FIG. 1 results from the following basic equations: r = exp (qλ _w ) (5) where r = radius, growth factor greater than q = 1, and λ _w = polarity angle in terms of wavelength at the particular medium, such as the circuit wire being monitored. is there.

The polar plot of FIG. 1 shows that algae and the first terrestrial (teres) were about 380 million years ago.
tial) Represents logarithmic growth spirals arising in countless natural objects, including spiral ammonites that appeared on the surface during the Devonian period, when plants also grew.
They grew into ferns with a very pronounced fractal structure during the Carboniferous period of about 260 million years ago, where each leaf structure was a reduced replica of the branch structure, The branches are replicas of ferns or shrubs. Ammonites became extinct at the end of the Cretaceous period, about 65 million years ago, but ferns are very alive with millions of fractal-structured natural objects.

Fortunately, the insight needed for the subject invention is a pioneering mathematician, Jaco.
b The area where Bernoulli lived until 1705 was the marine environment millions of years ago. This provided that area with abundant petrified ammonite, along with numerous amounts of other petrogens.

Bernoulli was fascinated by the logarithmic spiral devoted to his famous treaty, entitled “Spira Mirabilis” (a mysterious spiral). One such surprising property is that the logarithmic spiral is a perfect fractal in that the variety persists through change. It reacts elegantly with rotational displacement to enlargement and reduction, thereby preserving its shape unaffected. This and other well-known properties of logarithmic spirals are that they are truly fractal phenomena that would be considered similar in electric arc signature when viewed in polar coordinates in terms of wavelength, such as in FIG. To reveal.

In this regard, the ammonite shellfish has an internal compartment which is most closely located at the center of the shellfish,
It had a chambered structure, separated by a septum, a series of spaced plates whose logarithm increased logarithmically along the growth spiral of the shellfish. Thus, the size of the chamber between adjacent septum increased logarithmically along the growth helix.

Similar to the ammonite shell compartment, the size of the frequencies or wavelength spacing 12 between the indicated wavelengths or frequencies 13 is also similar to the electric arc signature or spectrum 1.
It increases logarithmically in terms of wavelength at zero.

The core of the electromagnetic arc emanation was labeled <μ at the pole in FIG. 1 indicating a wavelength below 1 micron. That is, it represents the well-known visible light emitted by the arc. In subsequent log turns, the symbol>
Infrared radiation can cause μ to cause sparks in electric welding, metallurgical furnaces, and internal combustion engine ignition and, conversely, in explosions and the initiation of catastrophic fires by electric arcs And have been shown to show microwaves.

Thereafter, FIG. 1 shows a particular part of the electromagnetic arc signature in terms of frequencies including the following frequency bands in which the sequence increases.

[0061] GHz = gigahertz, MHz = arc signature detection implemented by conventional technology, megahertz range where extraneous signals from television broadcasts and radio signals are abundant, 100 kHz = arc signature detection also wide range by conventional technology A 100 kilohertz range where radio broadcast signals are abundant, 30 kHz = lower limit of arc detection according to the prior art, VLF = "ultra long wave" defined as extending from 3 kHz to 30 kHz by the aforementioned IEEE dictionary, ELF = "extremely low frequency" defined by the IEEE dictionary as extending from 3Hz to 3kHz, vf = "sound frequency" in the range of 200 to 3500Hz according to the IEEE dictionary, if = 50Hz in European systems Or It is 60Hz in the Rica system "line frequency".

Needless to say, patent drawings cannot actually depict the disordered nature of electric arcs. Rather, FIG. 1 as a minimum must be seen in terms of instantaneous moments in the disorderly occurrence of the arc signature. Nevertheless, FIG. 1 shows the statistical self-similarity of the log fractal subset of the depicted arc signature.

To summarize the similarity of ammonites, the logarithmic nature of the depicted arc signatures is not only in terms of the evolution of the growth spiral 10 but also individually separated by those corresponding to the aforementioned septum of the ammonite shell. And for the logarithmically proceeding length of the frequency interval 12 in terms of wavelength. In terms of electric arc signatures, such septum is constant such that the spacing 12 between such frequency points is logarithmically distributed along the arc signature in terms of increasing or decreasing wavelengths. Corresponds to a radial line 13 indicating the frequency of

FIG. 1 also depicts the 1 / f dependence of the electric arc signature in terms of inverse frequency or amplitude. In the case of FIG. 1, this is shown as an amplitude that increases with wavelength up to a value of a _max . This is another indication of the fractal nature of the electric arc.

Regarding the subject of 1 / f noise, see Res. rr. nat. Heinz
-Otto Peitgen and Dietmar Saupe, "The Science of Fractal Imaging (THE SCIENCE OF FRACTAL IMAGS)
) "(Springer-Verlag, New York 1988), on pages 39-44, a simple mathematical equation that produces such noise, except for the homologous repetitive assumption of a particular distribution of time constants. Point out that there is no such model.

This is very similar to white noise on the one hand and Brownian motion on the other.
In this regard, they speak of the discovery that almost all musical melodies mimic l / f noise. For those serious studies, this suggests that "music mimics the characteristic way our world changes over time." In fact, music with its many variations on one theme is full of fractals.

Dres. Peitgen and Sauppe also point out that both music and 1 / f noise are intermediate between randomness and predictability. Even the smallest phrases reflect the whole. Therefore, it is also the case with electric arc signatures,
Such a minimal phrase is referred to herein as a "fractal subset".

As with light motifs in music, such fractal subsets may have the property of an attractor as a restriction map of fractal iterations, as described more fully below.

In the context of the representation of FIG. 1, a preferred embodiment of the present invention detects electric arcs from a fractal subset below 30 kHz of the broad range of arc signature frequencies. This includes the VLF (Very Long Wave) range described above and frequencies below that range.

More specifically, embodiments of the present invention provide for detecting an electric arc therein,
A fractal in which the electric arc is monitored to an ELF (extremely low frequency) band in which the electric arc is defined as extending from 3 Hz to 3 kHz according to the new IEEE Electrical and Electronic Terminology Standard Dictionary 5th Edition (IEEJ, 1993). Restrict a subset.

Another embodiment of the present invention is to reduce the monitored fractal from 200 Hz to 35
Limit fractals being monitored to arc signature frequencies below the audio frequency band (vf) defined in the IEEE Standard Dictionary as extending to 00 Hz. To that effect, an additional embodiment of the present invention limits the fractal subset being monitored to arc signature frequencies below a first high frequency of the standard line frequency in the AC power system. Embodiments of the present invention also select a subset of the arc signature fractals being monitored from frequencies that are approximately the standard line frequency (If) in the AC power system.

Thus, the present invention, in its embodiment, determines the optimal signal-to-noise ratio by _max The best total in a situation where it is specified all the way up to the maximum arc signature amplitude
You can select a fractal that will exhibit physical performance. That is, like this
In the case where the signal is_w-1 / f of the arc indicated above as noise-
Noise. On the other hand, noise in the expression "signal-to-noise ratio" is
And brown noise and false alarms such as those created by electronic circuits
Or contains extraneous signals that may generate readings.

It goes without saying that, depending on the application, there may be goals other than reaching a _max , but as shown in FIG. 1, selected according to the subject invention in a larger part of the rotation outside the growth spiral 10. The arc signature amplitudes at the various frequency fractals being described are still significantly better than the prior art had to study.

In this regard, selecting a fractal from the arc signature that produces an amplitude of a _max or a comparable amplitude in accordance with embodiments of the present invention
Cross-induction of arc signatures has another advantage that cross-induction can be a problem. For example, taking the case where multiple electrical circuits are monitored for electrical arcs by multiple corresponding arc detectors, an arc occurs in one of these circuits, and the arc detector associated with that circuit is Suppose it is to respond.

By selecting a high-amplitude (ie, long-wave or low-frequency) fractal subset for the arc monitoring process, according to equations (1) through (4) and those previously described thereafter. A preferred embodiment of the present invention minimizes, if not practically eliminates, the prior art danger of false alarms from mutual induction between individually monitored adjacent circuits.

FIG. 2 shows the electric conductor 20 of the electric circuit component 21 in which the electric arc 22 occurs.

By way of example, circuit component 21 may be part of a telephone exchange or may be another of a wide variety of electrical circuits or loads, including the following examples.

In research, development, and maintenance of an internal combustion engine, it is important to establish and maintain an optimal spark within each cylinder. Therefore, a reliable spark monitoring system is highly desirable, if not potentially essential, in state-of-the-art internal combustion engine technology.

In a similar vein, electric welding is increasingly robotized and reliable monitoring of welding arcs or sparks is largely due to the quality control and quality assurance of research, development and assembly lines in automated electric arc welding. Benefit. Furthermore, many modern electric spot welding processes rely on the immediate application of electrical energy to the workpieces to be joined without the intervention of an electric arc. In practice, for example, the occurrence of an electric arc due to imperfect contact between the workpieces degrades the resulting weld in such a Joule effect welding process. Thus, the load at 21 may be, for example, a robot or other spot welding equipment. In that case, an electric arc monitor could supervise the spot welding process and signal when a substandard weld is being caused by an intervening arc. This would instead signal the need for a remedy, including, for example, better cleaning of the workpiece prior to welding, or better compression of the workpiece during welding for more intimate contact.

Also, modern offspring of the first electric arc lamps, such as mercury lamps and sodium lamps, provide a reliable arc monitoring system for research and development and to allow for electric arc types in metallurgical and other furnaces. It will be helpful in maintaining high quality performance from.

Similarly, arc and spark monitoring systems would be effective to detect and, if necessary, eliminate faults in electrical circuits and equipment that create radio interference due to excessive sparking or arcing.

As an additional example, some gas heating devices have gaseous fuel igniters that handle electric sparks. In such a case, it is often important to know if the desired spark has occurred for proper ignition, especially when the thermostat is far from the heating device. Also, the electrical spark monitor will indicate when the igniter needs replacement before failure and expensive outages occur.

In addition, electric arcs, for example, to re-stabilize sanitation in geosynchronous orbit,
Or for use in satellite recoil propulsion engines, such as for propelling satellite and space probes on their journey, and in ionizing devices such as ammonia arc and other ion generators that will be used in space propulsion systems. In such a case, the electric arc monitor will be used to study such ion generators,
Will be effective in development, maintenance and operation.

Alternatively, the machine, circuit component or equipment at 21 that produces a normal spark during its operation could be monitored for harmful arcing. One of many examples
First, the rotating rectifier relates to a rectifier of an electric motor that is often damaged when its carbon brush is worn because it rubs against a metal brush holder spring.
Since such wear is followed by severe arcing, early detection of such severe arcing signals the need for preventive action, as distinguished from standard rectifier sparking. Notice, equipment could be saved from breakdowns and severe damage. The same is true for relays and contactors that produce sparks and arcs in their normal operation, but are subject to excessive arcing in the event of malfunction or excessive wear.

Similarly, the recurrence of electric vehicles as more environmentally friendly vehicles than gasoline-powered vehicles or oil-powered electric vehicles makes reliable monitoring even more important. In particular, such electric vehicles typically have to be recharged at night, and include large batteries that create flammable expectations, such as oxygen and hydrogen, during such recharging. It is clear that electric arcing can cause catastrophe in such atmospheres. Therefore, monitoring the environment for the occurrence of an electric arc, stopping the charging process, and issuing an alarm immediately after the detection of the arc may prevent a disaster.

In all these cases, the present invention selects a fractal subset of the signature of the electric arc 22 for arc detection purposes. In such an option, the present invention also targets relatively long travels of the arc signature along the circuit being monitored 20 and mutual guidance between adjacent circuits, including the circuit 20 being monitored. .

According to a preferred embodiment of the present invention, the fractal subset of the arc signature has a higher arc signature amplitude, a longer arc signature movement along the wire (20), and separately monitored neighbors. Circuits (21, 30)
Signal mutual induction between the lower, 30k
Selected in frequency bands below Hz.

In that case, the present invention detects the electric arc 22 from the fractal subset 16 of the arc signature. A preferred embodiment of the invention is a 30 kHz frequency band, such as the ELF (extremely low frequency) band defined above as extending from 3 Hz to 3 kHz.
An electric arc is detected from a fractal subset of the arc signature below z, or below 200 Hz, also below the audio frequency band (vf) defined above.

As an example, FIG. 3 shows such a fractal subset of the arc signature at 16 in band 15 shown in log memory. In practice, the selection of such a low-frequency fractal subset 16 avoids false alarms due to mutual induction, such as between the circuit 20 in which the arc 22 occurs and the adjacent circuit 30 being monitored separately, where the arc Does not occur instantaneously, or conversely, between an adjacent circuit monitored by another monitoring circuit 19 and the circuit 20 when the arc does not occur at that point.

The selection of such a low frequency fractal subset 16 also allows for the detection of the arc 22 from a remote location along the wire 20 over longer distances than is possible at high frequencies. By selecting such a low frequency band, the high amplitude input signal for the detection process according to the 1 / f or λ _W characteristics described above to determine the highest signal-to-noise ratio with the lowest exposure to false alarms is also obtained. Occurs.

The switching transients in the telephone exchange, the multiplexed voice, and otherwise, and the sensitivity of the AC power supply frequency to high frequencies may actually be eliminated,
The allowed travel distance of the arc signal between the arc 22 and the pick-up 23 is multiplexed relative to a high frequency detection system by selecting a narrower frequency band of approximately the standard line frequency in public AC power systems. Good. Preferably, according to that embodiment, the selected narrower frequency band is below the first-order high frequency of its standard line frequency.

According to the embodiment illustrated with the help of FIG. 3, the selected arc signature fractal subset is below 50 Hz for a European type system and for an American type system. The filter passband 15 may be below the line frequency, such as below 60 Hz.

Thus, arc monitoring according to a preferred embodiment of the present invention may be said to be centered on the low end of the arc signature spectrum.

The fractal subset of the arc signature 16 or passband 15 at which the detection of the arc 22 occurs covers at least a quarter of the logarithmic decimal of the broadband range of frequencies of the electric arc 22.

This is because of limiting the band of detection to within 20 kHz at the 170 kHz center frequency, the previously mentioned WO 90/0427 which lost the point.
8 overcomes the drawbacks of the appearance of the prior art approach. However, 170 kh
Considering that the 20 kHz range in the region of z is only a small part of the logarithmic decimal which contains information meaning "arc" as distinguished from the other signals, this is It left only a small percentage of the arc signature information within a particular logarithmic decimal available for detection.

In the case of most of the bandpass filter components of the comb filter disclosed in the above-mentioned later filed application, the situation is not very good. Everything except the first bandpass filter component is actually 100 kHz to 1
It has a center frequency that belongs to the fifth logarithmic decimal covering hertz. Since all of these components have a 50 kHz bandwidth, they provide arc signature information in a specific logarithmic decimal for arc detection as distinguished from other error signals or from pickup extraneous signals. Only a small percentage can be passed. By rotating the detection between the components of the comb filter array,
The prior art proposal also misses the opportunity to make best use of its lowest frequency component in the 55 kHz range.

As an example, the present invention may be practiced with the circuit components shown in FIG. In that circuit component, an electric arc 22 having an arc signature typified by a broadband range of frequencies of disordered nature has an input 26 coupled to the arc and a relative arc along the circuit being monitored. Filter 25 having a passband corresponding to the fractal subset 16 of the arc signature characterized by a relatively long travel and low mutual induction between adjacent circuits, and having an output 27 of that fractal subset of the arc signature. Detected with the help of A broadband signal detector circuit component of chaos theory having a detector input of its fractal subset 16 of the arc signature may be coupled to its output of an electrical filter, as disclosed in the additional course of FIG.

The arc signature pickup 23 may, for example, be powered by two high speed diodes connected back to back on one side to a ground and series current limiting resistor,
Clamp-on terminating at impedance 24, which may represent a conventional peak-to-peak limiter for clipping abnormally large input transients
n) It may be of a conventional type, such as a current transformer.

The pick-up or transformer 23 is, for example, 5 to 25 Hz with minimal insertion loss.
It may be wound to be frequency sensitive in the range of 0 Hz. Such a transformer 23 may be clamped around the monitored line 20. Hall effect sensors provide another example of an arc signal pickup that may be utilized in the practice of the present invention;
It extends to the use of other sensors, wired or wireless. Other monitored circuits 30 and the like may be provided with similar or similar pickups 123 and monitoring circuits 19.

The arc signature signal or fractal subset 16 being picked up
Is passed through a low pass filter 25 having an input 26 connected to the arc signature pickup 13. In the prototype of this embodiment, the configuration of this filter is that of two continuously connected third-order Butterworth filters,
Other configurations and other types of filters may be selected within the scope of the present invention.

The gain versus frequency plot of FIG. 3 shows the narrower frequency band 15 selected.
Shows a typical response characteristic 41 of such a filter with little attenuation.

Within the scope of the present invention, passband 15 could cover the entire logarithmic decimal, such as 10 Hz to 100 Hz. In accordance with an embodiment of the present invention, the monitored arc signature fractal 16 or passband 15 may include a D.C. C. From 1
As long as the decimal up to 0 Hz is standard decimal rather than logarithmic decimal, it covers at most logarithmic decimal of the broadband range of frequencies of the electric arc 22.

In this regard, FIG. 3 shows only about half of the upper logarithmic decimal of the filter characteristic 41 in the passband 15. This provided an illustrated embodiment of the present invention for clearly reliable arc detection. Arc Signature 1 depending on the situation
The fractal of FIG. 6 may be less than that shown in FIG. 3, but for reliable arc detection by simultaneous elimination of false alarms, at least a quarter of the broadband logarithmic decimal of the frequency of the electric arc. One needs to be covered.

In principle, the aspects of the present invention disclosed herein can be applied to frequency bands other than the preferred ELF (extremely low frequency) band, and the subject applicant has:
A model of the circuit components shown in FIG. 2 was built not only for the ELF band, but also for operation at multiple kilohertz as well as in the 10-20 kHz region.

According to a preferred embodiment, a narrower frequency band 15 or fractal subset 16 is selected, where there is less extraneous signal in the rest of the broadband range of frequencies of the arc signature. In this regard, the embodiment illustrated in FIG. 3 includes extensive switching transients in the telephone exchange, multiplexed audio signals, high frequency AC power, signals from control or security systems, and wireless broadcasts. A narrow frequency band 15 covering up to about 10 Hz or below 50 Hz without influence of a transmission signal is shown. However, not all embodiments of the invention are intended to be limited to operation within and below the very longwave range.

According to FIG. 2, the output 27 of the filter 25 is applied to a non-linear processing in what is here called a non-linear processor 42. By way of example, such a non-linear processor is also called a "modulated carrier detector" that demodulates the signal passed by the filter 25, including the picked-up electric arc signature segments of varying amplitude and frequency. A demodulator may be provided. This and other aspects of the invention treat the fractal subset of the arc signature as a modulated carrier having a modulation indicative of the electric arc 22 and monitor one or more modulations on such a modulated carrier. To monitor the electric arc.

Equipment for monitoring an electric arc having an arc signature typified by a broadband range of disordered frequencies in the circuit being monitored includes, for example, an arc signature input 27 and a carrier modulation output 43. And a modulated or modulated carrier detector in a non-linear processor 42 having

As an example, the fractal subset of the arc signature may be treated as an amplitude modulated carrier, in which case the electric arc may be restored, for example, by restoring the modulation on such an amplitude modulated carrier, and By detecting the amplitude from such recovered modulation, it may be monitored by monitoring the modulation of such an amplitude modulated carrier.

Accordingly, the non-linear processor 42 generates an AC signal at 43 as a function of such randomly varying amplitude at 43 in response to the randomly varying amplitude at 27.
It may be a detector or a demodulator. On the other hand, more stable signals erroneously picked up by circuit components 23 to 27 will not produce such an AC signal. Thus, the non-linear processor 42 is the first step in distinguishing the picked-up arc signature from signals originating from radio or television broadcasts, radio frequency security systems or other sources other than electric arcs.

Thus, the non-linear processor 42 can thus actually monitor the modulation by monitoring the modulation or amplitude, such as the detection of the monitored carrier electric arc 22. Handles arc signatures that have been amplitude modulated or picked up as AM carriers. At least in the case of AM detection, such "modulation" still includes its "carrier". Such a carrier may be removed from its modulation by means such as a non-linear processor or a bandpass filter 44 connected to the output 43 of the AM detector 42. By way of example, bandpass filter 44 may comprise a third-order Butterworth filter, or a filter with a similar response. The preferred response of such a filter may in principle follow the characteristic 41 shown in FIG. 3, except that the gain may be adjusted as desired or required. . As an example, the filter 44
May pass AC signals at frequencies below 20 or 30 Hz and reject spikes and other high-speed signals that may occur or occur in circuit components up to that point.

The recovered modulation representing the “arc” appears at the output 45 of the bandpass filter 44.

In accordance with an embodiment of this aspect of the invention, a combined modulated carrier detector having an arc signature input and a combined carrier modulation output may be used in monitoring the electric arc.

In this regard, such a combined modulated carrier detector may be a similar type of modulated carrier detector such as an AM detector or both FM detectors. This example is shown in FIG. 2 according to which a modulated carrier detector of a similar type is connected in series.

In particular, the output 45 of the bandpass filter 44 may be, for example, an A for generating a signal level that varies as a function of the arc signature picked up at its output 47.
It is connected to a second demodulator 46, which may be an M demodulator or a detector.

Within the scope of the present invention, similar types of modulated carrier detectors may be connected in parallel, as in non-linear processor 42.

Within the broad aspects of the invention, other types or types of demodulation techniques, such as, for example, those developed for frequency modulation, phase modulation, or carrier suppressed, or single sideband modulation It must be understood that may be used.

Thus, the non-linear processor 42 may include, for example, an FM demodulator at the output 43 that produces a signal in response to random changes in phase or frequency as occurs in an electric arc signature. In this regard, the combined modulated carrier detector may include various types of modulated carrier detectors, such as an FM detector at 42 and an AM detector at 46 connected in series.

In accordance with the presently described embodiments of the present invention, the arc signature comprises a first modulation and a second modulation of the carrier modulated in both the first mode and the second mode.
Monitored by monitoring the modulation.

By way of example, various types of modulated carrier detectors may include an AM detector and an FM detector.
A detector including such an AM detector and an FM detector are connected in parallel.

As an example, nonlinear processor 42 has 27 as its common input,
It may include AM and FM demodulators 412 and 413 connected in parallel, with individual outputs connected to AND elements 414 as shown in FIG.

The AND-element 414 is both where the AM demodulator 412 responds to random amplitude changes of the signal picked up at 27 and the FM demodulator 413 responds to random phase or frequency changes of the signal at 27. Only in case 43 will the output signal be provided.

In that case, this provides additional protection against spurious signals from extraneous signals, such as AM broadcast or control signals and FM broadcast or control signals,
It is assured that the picked-up signals are only represented as originating from the electric arc if they display not only chaotic amplitude fluctuations but also chaotic phase or frequency fluctuations characterizing the electric arc signature.

Preferably, instead of half-wave rectification or detection, full-wave rectification or detection is used at 42 and 46 to increase the speed of detection. Such accelerated velocities may instead be such as 15 shown in FIG. 3 where the signal-to-noise ratio is maximum and the extraneous signal 52 is minimal even in the deliberately exaggerated depiction of FIG. Allows for the selection of lower frequency bands. In other words, by selecting full-wave rectification or detection, trade-off of lower detection speed with lower detection frequency (trade).
off) is reduced. In this way, the higher and less error-sensitive detection sensitivity described above is achieved without an unpleasant delay in detection.

The time of the signal indicating the arc occurring at the output 47 from the detection process at 46 is adjusted and the level is detected at 48. The conventional RC type or other timing circuit component, and the conventional comparator circuit component are the same as those shown in FIG.
May be used at 47 to prevent switching transients, contact bouncing, normal rectifier arcing, and other transients, harmless arcs, as described more fully herein. In this way, the alarm circuit component 50
Responds only to arcs that maintain themselves only for the specified critical time period.

Circuit component 47 includes a conventional resettable latching device for latching in an alarm condition when arc 22 occurs and all detection criteria are met as described herein. Good. Various control circuits may be activated at this point to provide audible alerts, remote fire alerts, etc., or to shut down power within the affected line 20. For this type of circuit is essentially known,
Only block 50 has been shown to represent the possible presence of such control and alarm circuits.

The embodiment of FIG. 2 gently avoids false alarms from extraneous signals without reducing the bandwidth of the picked-up arc signal to much less than a fractal subset containing sufficient arc information. For example, such avoided extraneous signals include television signals, radio broadcast signals, various control signals, harmonics, and other signals that are relatively narrow in bandwidth compared to a wide band of chaotic arc signals. including.

Some extraneous signals, such as signals that are themselves disorderly in nature, must undergo some common mode rejection or other processing in order to avoid their confusion with the disordered arc signature. May not be possible. Such rejection of extraneous signals occurs slowly when the picked-up signals are beats for themselves, such as in the context of the following embodiments of the invention.

In this regard, an improvement in accordance with an embodiment of the present invention is that the fractal arc signature subset 16 in frequency band 15 is derived from arc 22 and its fractal portion as shown in FIG. The fractal arc signature subset 1 is added to an arc signal 17 in a frequency band 18 different from the set or frequency band 15.
6 is converted to its arc signal 1
7 to be detected. The preferred embodiment of the present invention is a fractal subset 16
Is subjected to a frequency conversion as shown at 17 in FIG. 3, and after the frequency conversion, the electric arc 22 is detected from the fractal subset. By way of example, and not by way of limitation, the fractal subset 16 is attached to itself, and the electric arc 22 is applied to the fractal portion attached to itself, as disclosed in FIGS. It may be detected from the set.

In particular, component 142 of the circuit components shown in FIG. 5 prevents extraneous signals occurring within the monitored frequency band from affecting the arc detection process. In accordance with that technique, the narrow-band extraneous signal in the monitored fractal subset 16 of the arc signature is reduced in energy relative to the rest of the fractal prior to detection of an electric arc from the fractal. Such a component has transducer inputs 127 and 227 for an arc signature fractal subset 16 coupled to an output 27 of an electrical filter circuit component 25, and has a passband 15 of the filter circuit component 25. There may be a frequency converter 142 having a converter output 43 for the arc signature segment 17 in different frequency bands 18. Then, the chaos theory broadband characteristic typical of an electric arc, for example,
Such a converted arc signal 17 is obtained by a broadband signal detector based on chaos theory.
Is detected from.

By way of example, and not by way of limitation, the input and output terminals 26 and 45
May be the same in FIGS. 2 and 5, so that the rest of the circuit may be the same with respect to FIG. 5, as in the apparatus of FIG. 2 described more fully above.

The illustrated embodiment of the present invention also mitigates the inherent tradeoff of slower detection at lower frequencies. In particular, the present invention of the invention is now described by transforming a selected arc signal fractal subset 16 from a lower frequency band 15 to a higher frequency band 18 or higher as shown in FIG. The embodiment has less extraneous signals and cross-coupling between circuits.
The detection speed corresponding to the higher frequency band 18 of the arc signature fractal subset 16 resulting from the lower frequency band 15 is lower, the actual distance traveled of the picked arc signal is lower at frequencies above band 15, and the actual distance traveled of the picked arc signal is lower. Realize. At the same time, the illustrated preferred embodiment of the present invention provides lower interconnection between circuits and lower frequency bands for faster detection speeds corresponding to higher frequency bands 18. A possible longer travel distance along the affected refining 20 corresponding to 15 is realized. The arc 22 generated in the line 20 corresponds to the circuit component in FIG. 2, FIG. 4, FIG. 5, or FIG. 6, but the arc pickup 123 and the arc monitoring circuit 19 in which no arc is generated at that time. Such a circuit 2 may be located at a considerable distance from the arc without release of the arc alarm condition on a separately provided adjacent line.
It may be picked up from zero.

As already indicated, the passband of the filter circuit component 25 may be located at a location with less extraneous signals than in the rest of the broadband range of frequencies. Such passbands may be within the ELF (extremely low frequency) band, or below the vf (voice frequency) band, or the first of the standard line frequencies in AC power systems.
The electric arc is below the high frequency, or near the standard line frequency in the AC power system, and / or provides sufficient information for reliable detection of the arc signal from its logarithmic decimal. May cover at least a quarter of the logarithmic decimal of the broadband range of the frequency.

The frequency converter 142 in FIG. 5 may constitute the nonlinear processor 42 in FIG.
By way of example, component 142 may include a first input 12 and a second input 12 connected to single line 27 for receiving a filtered picked-up arc signal.
A multiplier having 7 and 227 may be included. This has the net effect of mixing the signal with itself and producing at its output 43 the product of the sum and the difference. By way of example, component 142 may be a four-quadrant multiplier of type AD633. However, within the scope of the present invention, diodes or non-linear circuits may be used for intermodulation of the selected fractal subset 16 with itself.

As an example, FIG. 5 shows that in fact the arc signature fractal subset 16 is
Double the frequency band 15 of the fractal subset as a separate frequency band 18 of the arc signal 17 that may have been somewhat truncated, such as by subsequent filtering at the lower end at the higher band 18. To do
Add the arc signature fractal subset 16 to itself in the mixer 142. The filter may be similar to the previously described filter 44 in the circuit of FIG. 2, but may have a narrower band characteristic as shown at 51 in FIG.

In instrument terms (terms), a frequency converter, such as component 142, is coupled to two outputs, such as 127 and 227, for arc signature fractal subset 16 that is coupled to output 27 of electrical filter circuit component 25. It has a converter input and has a converter output 43 of the arc signature signal at twice or otherwise higher frequency band 15 of the arc signature fractal subset.

Component 142, in effect, dilutes a standard man-made non-chaotic theoretical signal that, if not picked up and undiluted, could cause a false alarm.
A multiplier or some maintenance stage 142 does more than simply double the frequency of the input. It also mitigates the effect of all frequencies occurring at one input 127 against all frequencies occurring at other outputs 227 (beat
s). This causes the output to display the sum of all individual frequencies at one input being added and subtracted from all frequencies at the other input. Since both inputs include consecutive frequencies, the result is a very rich mixture of frequencies at higher and lower frequencies than those contained in the inputs. Because artificial signals are usually discrete single frequencies or, at worst, narrow spreads of frequencies,
Such a signal has no spectral width to contribute strongly to the output at 43. Such an artificial signal constitutes a narrow source that mitigates the effect on wider noise or arc signature continuum, the result being a very weak component at the output.

In this way, the overall strength of the continuum spectrum from the arc is compared to the strength of discrete or artificial components passed by, for example, a limiter or filter circuit component such as at 24. To be enhanced. This is the essence of this embodiment's ability to reject signals that would cause false alarms. However, floating carriers that are noise-free at the input will still have the latter, even if their relative sensitivity to disordered noise is greater than to the modulated carrier, provided that they are strong enough. The output can be reached, thus creating a level that can be mistaken for an arc,
If it is strong enough, it will still be mistaken for an arc.

Thus, the output of the converter or multiplier 142 is sent to the input of a bandpass filter 144, which may have a response characteristic 51 as shown in FIG. The response characteristics of the type shown in FIG. 3 at 51 are, for example, by viable amplifier types of bandwidth filters, quartz filters, LC resonance filters, and other circuit components that accomplish these types of functions. It may be realized.

As can be seen, the characteristic 51 indicates the minimum response to the fundamental frequency passed by the low-pass filter 25, for example in band 15, and in our example, the sum or other modulation product is The picked-up arc signature fractal subset 16 corresponding to the pass band range 18 of the frequency-converted arc signature signal 17
Display the response for only the frequency of.

The bandpass filter 144 will only pass frequency components that are within a few hertz on either side of the passband, eg, 80 Hz. This delivers the higher frequency samples in the output of multiplier 142 to the next stage 46. Since the passband is 50 Hertz, or above any other selected cutoff of the first filter 25, this step 144 reduces the boosted signal in the mixing process to frequencies above such cutoff. Just pass it. This effectively prevents non-chaotic signals from passing through this stage.

In principle, the multiplier or stage 142 will have the first input if there is no DC at either input 127 and 227 and if such a multiplier or stage 142 is perfectly efficient. Do not pass any of the frequencies (eg, below 50 Hz). However, neither assumption is always correct in practice.
Thus, the use of a higher frequency passband filter 144 provides effective rejection of the first unprocessed or unmixed frequency.

If a higher detection frequency band, such as a band limited between 10 kHz and 20 kHz, is selected in the present invention, as an example, the alternate choice of band center frequency is the output of multiplier 142 There will be a much lower frequency (ie 500 Hz) to detect the difference component at 43. This will allow for a much wider separation in the response of the various filters utilized in circuit components such as filters 25 and 144.

In either case, frequency conversion or intermodulation at 142 significantly reduces the energy of the extraneous signal 52 picked up in the arc signature segment 16 due to intensive common mode or similar rejection.

The output 43 of the multiplier 142 represents a version of the disordered frequency sum or difference signal of the picked up and filtered disordered arc signature fractal subset 16.

Components 142 within the scope of the present invention also include the AM or FM detectors described above, or, for example, the combined AM and FM detectors 412 and 413 of FIG. In such a case, one or more detectors may be used for arc monitoring purposes, such as amplitude and / or frequency, or other types of transformed arc signatures, as shown at 17. Detect modulation.

FIG. 6 illustrates another embodiment of the present invention where the energy of the extraneous signal is exposed to an effective common mode or similar rejection by pairing the inverting and non-inverting amplifiers 242 and 342 with each other. Show. By way of example, the inverting amplifier 242 has an input 327 connected to the output 27 of the bandwidth filter 25, and the non-inverting amplifier 342 has the input 327 of the bandpass filter 25 described above through which the picked-up arc signal is processed. Input 4 connected to its output 27
27.

The inverting amplifier 242 has an output 421 connected to the mixer 442, and the non-inverting amplifier 342 has an output 422 connected to the mixer 442.
Having.

The electrical filter 25 in the embodiment of FIG. 6 may be the same as the electrical filter 25 in the embodiments of FIGS. Such a filter may be split in two, providing between terminal 26 a first filter for inverting amplifier 242 and a second filter for non-inverting amplifier 342. The filter 25 is divided between or for the embodiments of FIGS. 4 and 5 and similarly from the input terminal 26 to the detectors 412 and 413 or to the multiplier or frequency converter inputs 127 and 227. Provide a separate filter path.

By way of example, mixer 442 may be comprised of conventional components such as two diodes interconnected in an OR element configuration between the input terminals at 421 and 422 and the output terminal 43 described above. . However, within its implementation, component 442 includes a modulator, such as modulator 142 described above. In this regard, and generally, the circuit components of FIG. 6 between the filtered arc signature portion or segment at 27 and the terminal 43 may correspond to the non-linear processor 42 shown in FIG. . Such a processor 42 or 242 at 43
, 342 and 442 are filtered or otherwise 2
44. For example, circuit 244 may include a bandpass filter corresponding to bandpass filter 4426 or 144 described above with respect to FIGS. Alternatively or additionally, component 244 of FIG. 6 may include a standard IF amplifier, such as, for example, a commercially available 440 Hz IF amplifier.

In the method section, the embodiment of FIG. 6 handles fractal segments as shown at 16 or higher frequency segments of the arc signature in the two paths 327 and 427 that are out of phase with each other. , Detect arcs 22 from such out-of-phase segments or portions of the arc signature.

For equipment applications, the embodiment shown in FIG.
21 connected to the electrical filter output 27, here the "inverting amplifier" 2
42, having an input 327 of the first phase processor, referred to as
It has an input 427 of a second phase processor, here referred to as a “non-inverting amplifier” 342, connected at 7 to 422 to the electrical filter output 27. That is, such a first phase processor and a second phase processor are 180 ° out of phase with each other.

The more the outputs 421 and 422 are out of phase with each other, the greater the common mode or similar rejection of the extraneous narrowband signal 52 in the embodiment of FIG. The embodiments of FIGS. 2-5 thus provide for the detection of an electric arc 22 from such a fractal subset prior to the detection of the fractal subset 16 or other fractal 52 of the arc signature. 3 accordingly share with each other the energy-decreasing features with respect to the rest of such fractal subsets.

In at least the embodiments shown in FIGS. 3 and 5, the fractal subset 16 is subjected to a frequency conversion as shown in FIG.
2 are detected from such a fractal subset after such a frequency transformation. By way of example, the fractal subset 16 may be intermodulated or added to itself as disclosed above with respect to FIG. 5, or as mentioned with respect to FIG. It is detected from a fractal subset that is intermodulated or added to itself. The embodiment of FIG. 4 adds a variant of parallel different types of detection or demodulation, and the embodiment of FIG. 6 adds a variant of out-of-phase processing.

In either case, the fractal subset of the arc signature portion may be treated as a modulated carrier having a modulation indicative of the electric arc 22, such an electric arc being so due to additional rejection of the extraneous signal. May be detected from any modulated carrier. The demodulator system disclosed above with respect to FIG. 2 can also be used in the embodiment of FIG. 6, the terminal 47 of which is connected to the time and time of FIG. Level detection alarm circuit component 4
The input terminals 47 of 8 to 50 may be the same.

Needless to say, a wideband signal is not an arc signature unless it indicates a disorderly frequency change. Thus, not only is the broadband in the region of interest, but also chaotic in nature, as are the arc signatures, for example,
A stage including a comparator 55 may be provided as shown in FIG. 7 to detect and indicate disturbance or signal pickup sustained for a time period determined by the RC component 58.

FIG. 7 illustrates a possible or actual arcing optical indicator that can be used at various stages in the systems of FIGS. 2, 4, 5 and 6 in accordance with additional embodiments of the present invention. It is a circuit diagram.

By way of example, an indication of the progress of a signal through the arc monitoring circuit component may be achieved by three similar functional blocks or difference indicators whose prototype is shown at 54 in FIG.

Such a circuit 54 includes a workable amplifier 55 having its non-inverted input 56 connected to a circuit input 57 through a low-pass filter and RC timing component 58 to prevent response to short-term transients. Including. The invert input 60 of the operational amplifier is connected to comparator level resistors 61 and 62. The operational amplifier 55 includes a unidirectional current conducting element such as that shown in feedback capacitors or other impedances 65 and 66 for purposes such as noise reduction, preventing premature or excessive switching. There is a feedback circuit 64 that may be included. As an example, the operational amplifier 55 is an LM3
It may be a 5BAN type.

The indicator circuit 54 comprises light emitting diodes or LEDs 68 and 69 which are switched by transistors 71 and 72 which are deflected through resistors including series resistors 73 and 74 and a set of resistors 75 and 76. . Transistors 71 and 72 may be, for example, of the type 2N2222.

[0160] A resistor 78 connects the transistor 72 to the output of the comparator enabled amplifier 55. Transistor 71 is normally deflected ON through resistors 75 and 76 connected in series. This is the first LE which may be, for example, a green LED
Turn on D68. Conversely, the second LED 69 may be a red LED. However, the second transistor 72 and thus the red LED 69 are deflected off at that point.

The signals having frequencies in the fractal subset or other bands of interest that are being monitored may be transmitted to the nonlinear processor 42 or modulator 142 or mixer 442 in the embodiments of FIGS. 2, 4 and 5. At the input 43 of the circuit 54 connected thereto, the output of the comparator 55 goes positive, turning on the transistor 72 and turning off the transistor 71. This is,
The red LED 69 is turned on and the green LED 68 is turned off, so that the frequency of interest for arc detection to the observer may indicate, for example, but not necessarily the electric arc 22. Indicates that it is occurring through disturbance. The gain of circuit 54 may be adjusted to avoid sharp transitions when switching states. This helps the user gain a "qualitative sensation" for the amplitude of the signal at that point by measuring the mixture of red and green LED colors. The unidirectional current conducting element 66 is similar to the circuit of FIG.
If so used in any of the circuits of FIGS. 5 and 6, it may be omitted.

Alternatively, terminal 57 of display circuit 54 may be connected to terminal 45 of FIG. 2, 5, or 6, or a copy of the circuit shown in FIG.
And thereby a bandpass filter or IF amplifier 44, 144 or 244
May be so connected.

[0163] In that case, such a display step 54 can be performed through the red LED, for example, in the fractal subset being monitored, ie the well-known criterium of the arc signature, at the bandwidth of interest. 4 illustrates the generation of a wideband signal. Adjusting the gain in such a circuit 54 again provides the user with a red and green L.
Measuring the color mixture of the ED may give a "qualitative sensation" with respect to the broadband signal picked up at that point.

However, a wideband signal is not an arc signature unless it indicates a chaotic frequency change. Thus, the circuit components shown in FIG. 7 may be used as a final display stage in the monitoring circuit shown in FIGS. 2, 4, 5, and 6. For example, input terminal 57 of circuit 54 may be connected to alarm output terminal 49 shown in FIG. In effect, the circuit components of FIG. 7 from terminal 57 to operational amplifier 55 are used as timing circuit 58 in the previously described timing and level sensing circuit component 48 shown in FIG. May be.

The circuit component 54 is thus not only broadband in the region of interest, but also chaotic in nature as the arc signature is, for example, as determined by the RC component 58. It may be used to detect and display disturbances or signal pickups that are sustained for a period of time.

Since the disturbances or signals at wideband frequencies in the range of interest occur and change randomly, the output of comparator 55 at circuit component 48 and terminal 49 goes positive, turning on transistor 72, The transistor 71 is turned off. This turns on the red LED 69 and turns off the green LED 68, thereby indicating to the observer the occurrence of the arc 22 on line 20.

The embodiment shown with the help of FIG. 7 thus provides a pre-warning of possible electric arcs, preferably in two or three phases, and the fractal being monitored Indication of the occurrence of disordered broadband signals at the bandwidth of the subset, or otherwise at the bandwidths of interest described above with respect to, for example, FIGS. 2, 4, 5, 6, and 7. At the climax.

The principles and circuit components disclosed herein may be utilized in a variety of arc monitoring functions as described above. In the case of use such as research, development and maintenance in internal combustion engines or fields such as electric ignition, electric welding, and electric lighting as described above, circuit components of the type shown in FIG. Terminals 47 may be utilized with and without circuit components of the type shown in FIGS. 4, 5 and 6. The types of signal display steps shown in FIG. 7 and more precise signal display and evaluation steps may be used in such cases.

This extensive disclosure will make or suggest various modifications and variations within the spirit and scope of the invention to those skilled in the art.

[Brief description of the drawings]

The subject invention and its various aspects and objects will be more readily apparent from the following detailed description of a preferred embodiment thereof, illustrated by way of example in the accompanying drawings that also constitute a written description of the invention. Where similar reference numbers indicate similar or equivalent parts.

FIG. 1 is a polar coordinate system representation of an electric arc signature spectrum in terms of wavelength, illustrating the selection of a fractal subset for arc monitoring according to an embodiment of the present invention.

FIG. 2 is a block diagram of an electric arc monitoring system according to an embodiment of the present invention.

FIG. 3 shows a frequency versus gain graph illustrating a currently preferred embodiment of the present invention.

FIG. 4 is a schematic diagram of circuitry for monitoring an arc that may be used in the system of FIG. 2 or otherwise in accordance with an embodiment of the present invention.

FIG. 5 is a schematic diagram of another circuit configuration that may be used in the system of FIG. 2 or otherwise monitor an arc in accordance with an embodiment of the present invention.

FIG. 6 is a schematic diagram of additional circuitry that may be used in the system of FIG. 2 or otherwise for monitoring an arc, also in accordance with an embodiment of the present invention.

FIG. 7 is a circuit diagram of an arcing or actual arcing optical indicator that can be used at various stages of the system of FIGS. 2, 4, 5 and 6, according to additional embodiments of the present invention. .

──────────────────────────────────────────────────続 き Continuation of front page (81) Designated country EP (AT, BE, CH, DE, DK, ES, FI, FR, GB, GR, IE, IT, LU, MC, NL, PT, SE), OA (BF, BJ, CF, CG, CI, CM, GA, GN, ML, MR, NE, SN, TD, TG), AP (GH, KE, LS, MW, SD, SZ, UG, ZW) , EA (AM, AZ, BY, KG, KZ, MD, RU, TJ, TM), AL, AM, AT, AU, AZ, BA, BB, BG, BR, BY, CA, CH, CN, CU, CZ, DE, DK, EE, ES, FI, GB, GE, GH, HU, IL, IS, JP, KE, KG, KP, KR, KZ, LC, LK, LR, LS , LT, LU, LV, MD, MG, MK, MN, MW, MX, NO, NZ, PL, PT, RO, RU, SD, SE, SG, SI, SK, SL, TJ, TM, TR, TT, UA, UG, US, UZ, VN, YU, ZW (72) Inventor Benoit Luke Pierre 9101 California La Canada Stardust Road 5322 Starlight Mesa F-term (reference) 2G014 AA00 AB00 AC18

Claims (65)

[Claims] 1. A method for monitoring an electric arc having an arc signature categorized by a broadband range of frequencies of disordered nature in the circuit being monitored, wherein an improvement is combined with the method of monitoring the electric arc. Selecting a fractal subset of the arc signature that is characterized by relatively long travels along a circuit and low mutual induction between adjacent circuits; and selecting the electrical fractal subset from the fractal subset of the arc signature. Including monitoring the arc. 2. The method of claim 1 wherein said fractal subset is selected from the logarithmic decimal of said wideband range of frequencies. 3. The method of claim 1, wherein said fraction covers at least a quarter of the logarithmic decimal of said broadband range of electric arc frequencies. 4. The method of claim 1, wherein the fractal subset is selected from a frequency band below 30 kHz. 5. The method of claim 1, wherein said selection of a fractal subset is frequency limited to an ELF (extremely low frequency) band. 6. The method of claim 1, wherein said selection of a fractal subset is frequency limited until below a vf (speech frequency) band. 7. The method of claim 1, wherein the fractal subset is selected below a first high frequency of a standard line frequency in an AC power system. 8. The method of claim 1 wherein said fractal subset is selected from a frequency band at approximately standard line frequency of an AC power system. 9. The narrowband extraneous signal in the fractal subset of the arc signature decreases in energy with respect to the remainder of the fractal subset prior to detection of the electric arc from the fractal subset. Claims 1, 2, 3
The method according to 4, 5, 6, 7 or 8. 10. The fractal subset is subjected to a frequency transform, and the electric arc is detected from the fractal subset after the frequency transform. Or the method described in 8. 11. The method of claim 10, wherein the fractal subset is applied to itself, and wherein the electric arc is detected from a fractal subset applied to itself. 12. The fractal subset is processed on two paths out of phase with each other, and the electric arc is monitored from the fractal subset being processed on the two paths out of phase with each other. A method according to claim 1, 2, 3, 4, 5, 6, 7 or 8. 13. The method of claim 1, wherein the fractal subset is treated as a modulated carrier having a modulation indicative of the electric arc, and the electric arc is monitored by monitoring modulation of the modulated carrier. The method according to 2, 3, 4, 5, 6, 7, or 8. 14. The method of claim 13, wherein the fractal subset is treated as an amplitude modulated carrier, and wherein the electric arc is monitored by monitoring modulation of the amplitude modulated carrier. . 15. The method of claim 14, wherein the electric arc is monitored by restoring modulation on the amplitude modulated carrier and by detecting amplitude from the restored modulation. 16. The method of claim 13, wherein the fractal subset is treated as a frequency modulated carrier, and wherein the electric arc is monitored by monitoring the modulation of the frequency modulated carrier. 17. The fractal subset is treated as a carrier modulated in both a first mode and a second mode different from the first mode, and wherein the electric arc is modulated in both the first mode and the second mode. Monitoring the first and second modulations of the carrier being monitored.
3. The method according to 3. 18. The method of claim 1, 2, 3, 4, 5, 6, 7, or 8 including providing a prior warning of possible electric arcs. 19. The method of claim 16, wherein said advance warning is provided in stages. 20. A method as claimed in claim 1, comprising indicating the occurrence of a signal having a frequency in the bandwidth of the fractal subset. 21. The method of claim 1, 2, 3, 4, 5, 6, 7, or 8 comprising indicating the occurrence of a wideband signal in the bandwidth of the fractal subset. 22. A method as claimed in claim 1, 2, 3, 4, 5, 6, 7, or 8 indicating the occurrence of a broadband signal of chaotic theory in the bandwidth of the fractal subset. 23. A method of monitoring an electric arc having an arc signature that extends over a wide range of frequencies of disordered nature in the circuit being monitored, an improvement wherein the remedy is combined in the two paths out of phase with each other. Processing a portion of the arc signature; and monitoring the electric arc from the out-of-phase portion of the arc signature. 24. A method for monitoring an electric arc having an arc signature extending over a wide range of frequencies of disordered nature in the circuit being monitored, wherein a remedy is to combine the arc signature as a modulated carrier. Handling; having a modulation indicative of the electric arc; and monitoring the electric arc by monitoring the modulation of the modulated carrier. 25. The method of claim 24, wherein the arc signature is treated as an amplitude modulated carrier, and wherein the electric arc is monitored by monitoring modulation of the amplitude modulated carrier. 26. The method of claim 25, wherein the electric arc is monitored by restoring modulation on the amplitude modulated carrier and by detecting amplitude from the restored modulation. 27. The method of claim 24, wherein the arc signature is treated as a frequency modulated carrier, and wherein the electric arc is monitored by monitoring modulation of the frequency modulated carrier. 28. The method of claim 28, wherein the arc is treated as a carrier modulated in both a first mode and a second mode different from the first mode, and wherein the electric arc is modulated in both the first mode and the second mode. 25. The method of claim 24, wherein the method is monitored by monitoring a first modulation and a second modulation of the first. 29. In an apparatus for monitoring an electric arc having an arc signature typified by a broadband range of frequencies of disordered nature in the circuit being monitored, remedies are combined with said arc. Passage corresponding to a fractal subset of the arc signature characterized by relatively long travel along the monitored circuit and low mutual induction between adjacent circuits having inputs coupled thereto. An electrical filter having a band and having an output of the fractal subset of the arc signature; and a chaotic theory broadband signal having a detector input for the fractal subset of the arc signature coupled to the output of the electrical filter. A detector. 30. The apparatus of claim 29, wherein said passband is within logarithmic decimal of said wideband range of frequencies. 31. The apparatus according to claim 29, wherein said pass band is below 30 kHz. 32. The apparatus of claim 29, wherein said passband is where less extraneous signals are in the remainder of said wideband range of frequencies. 33. The system according to claim 2, wherein said pass band is within an ELF (extremely low frequency) band.
9. The device according to 9. 34. The system according to claim 2, wherein said pass band is lower than a vf (voice frequency) band.
9. The device according to 9. 35. The system according to claim 31, wherein said pass band is a first line frequency of a standard line frequency of an AC power supply system.
30. The device of claim 29, which is below a high frequency. 36. The apparatus according to claim 29, wherein said pass band is approximately a standard line frequency of an AC power supply system. 37. The apparatus of claim 29, wherein said passband covers at least a quarter of the logarithmic decimal of said broadband range of electric arc frequencies. 38. The apparatus of claim 29, wherein said passband covers less than the logarithmic decimal of said wide range of frequencies of the electric arc. 39. An input having an input connected to the output of the electrical filter;
An inverting amplifier having an amplifier output connected to the detector input; a non-inverting amplifier having an input connected to the output of the electrical filter and having an amplifier output connected to the detector input. 30. The device of claim 29, comprising: 40. The apparatus of claim 29, wherein said chaotic theory broadband signal detector comprises a modulated carrier detector coupled to said output of an electrical filter. 41. The modulated carrier detector is an AM detector.
Equipment described in item 0. 42. The modulated carrier detector is an FM detector.
Equipment described in item 0. 43. The apparatus of claim 40, wherein said chaotic theory broadband detector comprises a coupled carrier detector. 44. A converter input for the fractal subset and for a narrowband extraneous signal at the arc signature segment coupled to the output of an electrical filter, for the arc signature segment. An energy converter having a transducer output for a narrowband extraneous signal of reduced energy with respect to a fractal subset and connected to the chaotic theory broadband signal detector. Claims 29, 30, 31, 32, 33, 34, 35, 36, 37, 38
, 39, 40, 41, 42, or 43. 45. A converter output for said fractal subset coupled to said output of an electrical filter circuit component, said converter output of said fractal subset in a frequency band different from said passband. 29. The device of claim 29, 30, 31, 31, 32, 33, 34, 3 connected to the chaotic theory broadband signal detector.
The device according to 5, 36, 37, 38, 39, 40, 41, 42 or 43. 46. The fractal subset having two transducer inputs coupled to the output of an electrical filter, wherein the different frequency bands of the arc signal are twice the frequency band of the fractal subset. 29. The apparatus of claim 29, 30, 31, 32, 33, 34, 35, 36, having a transducer output for the fractal subset in a frequency band of 37,
The device according to 38, 39, 40, 41, 42 or 43. 47. A modulator having a modulator input for said fractal subset coupled to said output of an electrical filter and having a modulation indicative of said electric arc connected to said chaotic theory broadband signal detector. 36. A modulator having a modulator output for a modulated carrier; and a broadband signal detector of the chaos theory including a modulator detector. , 36, 37, 38
, 39, 40, 41, 42, or 43. 48. An amplitude modulator having a modulator input for said fractal subset coupled to said output of an electrical filter, said amplitude modulation indicative of said electric arc being connected to said chaotic theory broadband signal detector. 29. A modulator having a modulator output for an amplitude modulated carrier having the modulator, and a broadband signal detector of the chaos theory including an amplitude modulation detector. 35, 36, 37, 38
, 39, 40, 41, 42 or 43. 49. The amplitude modulation detector includes a first step of restoring modulation on the amplitude modulated carrier and a second step of detecting an amplitude indicative of the arc signature from the restored modulation. Equipment marked in. 50. An electric arc pre-warning indicator coupled to said electric filter, comprising: an electric arc pre-warning indicator.
The device according to 38, 39, 40, 1, 42 or 43. 51. An electric arc pre-warning indicator coupled to the chaotic theory broadband signal detector.
, 35, 36, 37, 38, 39, 40, 41, 42, or 43. 52. A broadband signal indicator coupled to said chaotic theory broadband signal detector.
The device according to 36, 37, 38, 39, 40, 41, 42 or 43. 53. The apparatus of claim 52, wherein said indicator is a broadband chaos theory signal indicator coupled to said chaos theory broadband signal detector. 54. In an apparatus for monitoring an electric arc having an arc signature typified by a broadband range of frequencies of disordered nature in the circuit being monitored, the remedies are combined with the arc. An electrical filter having a coupled input, a passband corresponding to a portion of the arc signature, and an output having an output for the portion of the arc signature; and an electrical filter connected to the output of the electrical filter. An inverting amplifier having an input and an amplifier output; a non-inverting amplifier having an input coupled to the output of the electrical filter and having an amplifier output; and the amplifier output of the inverting amplifier and the non-inverting amplifier. And a chaotic theory signal detector having a detector input coupled to the input signal. 55. In an apparatus for monitoring an electric arc having an arc signature typified by a broadband frequency of random performance in the circuit being monitored, the remedies combine arc signature input. And a modulated carrier detector having a carrier modulation output. 56. The modulator of claim 5, wherein the modulated carrier detector is an AM detector.
The device according to item 5. 57. The modulated carrier detector is an FM detector.
The device according to item 5. 58. In an apparatus for monitoring an electric arc having an arc signature typified by a wideband range of random performance frequencies in the circuit being monitored, the remedies combine arc signature input. And a combined modulated carrier detector having a combined carrier modulation output. 59. The apparatus of claim 58, wherein said combined modulated carrier detector is a similar type of modulated carrier detector. 60. The apparatus according to claim 59, wherein said modulated carrier detectors of a similar type are connected in series. 61. The apparatus of claim 59, wherein said similar type of modulated carrier detectors are connected in parallel. 62. The apparatus of claim 58, wherein said coupled modulated carrier detector comprises various types of modulated carrier detectors. 63. The apparatus of claim 62, wherein the various types of modulated carrier detectors include an AM detector and an FM detector. 64. The apparatus according to claim 63, wherein said AM detector and FM detector are connected in parallel. 65. The apparatus of claim 64, further comprising an AND element having an input connected to said AM detector and said FM detector, and having an output as said coupled carrier modulation output.