WO2002037512A1 - Method and apparatus for controlling the magnetization of current transformers and other magnetic bodies - Google Patents
Method and apparatus for controlling the magnetization of current transformers and other magnetic bodies Download PDFInfo
- Publication number
- WO2002037512A1 WO2002037512A1 PCT/US2000/030358 US0030358W WO0237512A1 WO 2002037512 A1 WO2002037512 A1 WO 2002037512A1 US 0030358 W US0030358 W US 0030358W WO 0237512 A1 WO0237512 A1 WO 0237512A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- induction level
- voltage
- cuπent
- winding
- time
- Prior art date
Links
Classifications
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01F—MAGNETS; INDUCTANCES; TRANSFORMERS; SELECTION OF MATERIALS FOR THEIR MAGNETIC PROPERTIES
- H01F38/00—Adaptations of transformers or inductances for specific applications or functions
- H01F38/20—Instruments transformers
- H01F38/22—Instruments transformers for single phase ac
- H01F38/28—Current transformers
- H01F38/32—Circuit arrangements
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01F—MAGNETS; INDUCTANCES; TRANSFORMERS; SELECTION OF MATERIALS FOR THEIR MAGNETIC PROPERTIES
- H01F13/00—Apparatus or processes for magnetising or demagnetising
- H01F13/006—Methods and devices for demagnetising of magnetic bodies, e.g. workpieces, sheet material
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R15/00—Details of measuring arrangements of the types provided for in groups G01R17/00 - G01R29/00, G01R33/00 - G01R33/26 or G01R35/00
- G01R15/14—Adaptations providing voltage or current isolation, e.g. for high-voltage or high-current networks
- G01R15/18—Adaptations providing voltage or current isolation, e.g. for high-voltage or high-current networks using inductive devices, e.g. transformers
- G01R15/183—Adaptations providing voltage or current isolation, e.g. for high-voltage or high-current networks using inductive devices, e.g. transformers using transformers with a magnetic core
- G01R15/185—Adaptations providing voltage or current isolation, e.g. for high-voltage or high-current networks using inductive devices, e.g. transformers using transformers with a magnetic core with compensation or feedback windings or interacting coils, e.g. 0-flux sensors
Definitions
- This invention relates to the art of controlling the magnetic flux density within magnetic bodies.
- a varying voltage applied to a conductive winding causes the voltage induced in the winding to have such waveform and magnitude that the integral over time of the induced voltage correlates to desired changes of the magnetic induction level.
- Prior art methods that utilize a conductive winding to control magnetization, generally focus on controlling winding current to control the induction level of a magnetic body.
- the winding current directly correlates with magnetic field intensity. Changes in magnetic field intensity are coordinated with hysteresis-curve properties of the magnetic material to control the induction level.
- magnetization “magnetic induction,” “flux density” and "induction level”).
- demagnetizing methods that effectively reduce the induction level of a magnetic body to near zero
- the alternating magnetic field is usually produced by a conductive winding conducting a declining alternating current.
- Many different embodiments of this demagnetizing method are well established in the prior art.
- Faraday's Law basically states that the voltage induced around a closed path is proportional to the rate of change of magnetic flux within the closed path. It is well established in prior art how Faraday's Law can be used along with a conductive winding to measure changes of induction level in a magnetic body. However, the prior art does not appear to address how the induction level of a magnetic body can be controlled by controlling the voltage induced in a winding. The present invention has been developed primarily out of a need to demagnetize current transformers while the current transformers remain in service. The prior art does not address this problem very well.
- a primary winding of a current transformer is connected in series with a current-carrying conductor while a secondary winding is magnetically coupled to the primary winding by a suitable magnetic core.
- a current is induced in the secondary winding that is proportional to the primary current.
- the secondary current is isolated from the primary current and is smaller than the primary current by the turns ratio of the primary and secondary windings.
- the primary winding frequently consists of only one turn, which is often just the current-carrying conductor installed through an opening in the middle of the current transformer magnetic core.
- the secondary winding usually consists of multiple turns wrapped around the magnetic core.
- the impedance of the circuit connected to the secondary winding must be kept low so that current can flow freely.
- the impedance of the secondary circuit is often called the "burden.”
- a voltage In order for a current transformer to drive a secondary current through a non-zero burden, a voltage must be induced in the secondary winding.
- the induced voltage is proportional to secondary current and proportional to the burden, in accordance with Ohm's law.
- the induced voltage is induced in the secondary winding by a fluctuating induction level in the magnetic core.
- the fluctuating induction level is associated with a magnetizing current in accordance with well-known electromagnetic principles.
- the magnetizing current accounts for most of the error in the secondary current.
- the accuracy of a current transformer is inversely related to the burden of the secondary circuit. A higher burden causes the secondary current to be a less accurate representation of the primary current.
- the accuracy of the secondary current may also be adversely affected by either of the following:
- a primary current that is not symmetrical is intended to mean that the waveform has positive and negative half-cycles with the same waveform and magnitude.
- An alternating current that has a d-c (direct-current) component is a common example of a primary current that is not symmetrical.
- transient a-c fault currents are often not symmetrical.
- D-c currents are, by definition, not symmetrical.
- a varying voltage is applied to a conductive winding that magnetically interacts with a magnetic body.
- the varying voltage controls the voltage induced in the winding in such a way that the integral over time of the induced voltage correlates to desired changes of the induction level (magnetization) of the magnetic body.
- the invention may be used to control the induction level of a magnetic body in several ways: (a) The induction level of a magnetic body may be caused to transition from a known induction level to a preferred induction level. (A preferred induction level of zero may be chosen to demagnetize a magnetic body).
- a preferred induction level may be established by changing the induction level of the magnetic body from an unknown induction level to a known induction level and then to the preferred induction level.
- a preferred induction level may be maintained by causing the induced voltage across the winding to have an average value near zero (or by. causing the integral of induced voltage to not exceed a predetermined value).
- a preferred induction level may be more strictly maintained by causing the induced voltage across the winding to continuously be near zero volts, thereby reducing the amount that the induction level fluctuates.
- the induction level may be made to vary with time in a preferred manner, including matching a control signal that is proportional to a reference induction level.
- the varying voltage may be generated directly by an active voltage source (or, more generally, a controllable electric energy source) or the varying voltage may be generated indirectly, such as by current transformer secondary current flowing through an adjustable impedance.
- the key elements are a magnetic body, a conductive winding that magnetically interacts with the magnetic body, and a means of causing the induced voltage to have the appropriate waveform and magnitude.
- the invention is most readily applied to magnetic bodies that are configured to have a relatively uniform magnetic path, such as the magnetic cores of current transformers.
- the accuracy of a current transformer may be improved by the invention in three ways:
- a demagnetized current transformer can accurately measure d-c current and a-c current that has a d-c component.
- the invention is based on two principles of electromagnetism: (1) the relationship of magnetomotive force to flux density as defined by hysteresis curves of magnetic bodies, and (2) Faraday's Law applied to magnetic bodies.
- the second principle clarifies how a voltage induced in a winding is directly proportional to a changing induction level. While prior art shows how induced voltage can be used to measure changes in induction level, the present invention shows how induction level can be controlled by controlling induced voltage. The invention primarily utilizes this second principle relationship to control the induction level of a magnetic body. The application of this second principle is an integral part of all embodiments of the invention. BRIEF DESCRIPTION OF THE DRAWINGS
- FIGS. 1 through 12 illustrate how the invention may be applied to magnetic bodies in general.
- FIGS. 1 and 2 illustrate typical components and connections, while FIGS. 3 through 12 illustrate how a varying voltage may be utilized to control the induction level of a magnetic body.
- FIGS. 13 through 19 all illustrate how an adjustable impedance may be utilized in a current transformer secondary circuit to demagnetize a current transformer.
- FIGS. 20 through 25 all illustrate how an active voltage source may be utilized in a current transformer secondary circuit to control induction level.
- FIG. 1 illustrates the general concept by showing a magnetic body 60, a winding 61 wrapped around magnetic body 60, and a controllable voltage device 63 with a suitable control circuit 62.
- Magnetic body 60 and winding 61 are shown in a typical current transformer configuration with the magnetic core around an electric power system conductor 64.
- Controllable voltage device 63 may be an active electrical energy source or an adjustable impedance.
- FIG. 2 illustrates an alternate configuration in which magnetic material 72 is used to provide a closed magnetic path for a magnetic body 70.
- FIG. 3 shows a hysteresis curve sequence that maybe used to quickly change the induction level of a magnetic body from an unknown induction level to an induction level near zero.
- FIGS. 4A to 4D show one method of controlling voltages and currents to demagnetize a magnetic body utilizing the hysteresis curve sequence shown in FIG. 3.
- FIG. 5 shows a hysteresis curve sequence similar to FIG. 3, except now showing how the induction level of a magnetic body may be quickly changed from an unknown level to a preferred induction level (other than zero).
- FIGS. 6A to 6D show one method of controlling voltages and currents to establish a preferred induction level in a magnetic body utilizing the hysteresis curve sequence shown in FIG. 5.
- FIG. 7 shows a hysteresis curve sequence that may be used to automatically determine demagnetizing parameters and quickly demagnetize a magnetic body.
- FIGS. 8A to 8D show one method of controlling voltages and currents to automatically determine demagnetizing parameters and demagnetize a magnetic body utilizing the hysteresis curve sequence shown in FIG. 7.
- FIG. 9 shows a hysteresis curve sequence that illustrates how a magnetic body may be demagnetized when current is not adequate to drive the magnetic body into saturation.
- FIG. 10 shows another hysteresis curve sequence that illustrates how a magnetic body may be demagnetized without driving the magnetic body into saturation.
- FIG. 11 shows a very simple embodiment of a controllable electric energy source that can be used to carry out the hysteresis curve sequence shown in FIG. 10.
- FIGS. 12A to 12C show one way that an alternating voltage and current may be utilized to implement the hysteresis curve sequence of FIG. 10 using the controllable electric energy source shown in FIG. 11.
- FIG. 13 shows a functional schematic to illustrate how an adjustable impedance 2 may be used to demagnetize a current transformer while in service.
- An optional power supply circuit 4 is shown that can derive power from the current transformer secondary current to provide power for self-powered applications.
- FIGS. 14A to 14E show typical operating waveforms and control signals associated with preferred operation of the functional schematic shown in FIG. 13.
- FIGS. 15A to 15E show waveforms and control signals for an alternate method of operation for the functional schematic shown in FIG. 13. This method of operation does not require any feedback signal to the control circuit.
- FIG. 16 shows one way that the functional schematic of FIG. 13 may be embodied.
- the power supply circuit for self-powered operation is optional.
- FIGS. 17A to 17D show typical operating waveforms and control signals associated with the embodiment shown in FIG. 16.
- FIG. 18 shows an alternate embodiment of a current transformer demagnetizing circuit utilizing only one MOSFET transistor for the adjustable impedance.
- FIGS. 19A to 19E show typical operating waveforms and control signals associated with the embodiment shown in FIG. 18.
- FIGS. 20A to 20D illustrate how the configuration of FIG. 1 may be operated to demagnetize a current transformer and measure d-c current.
- FIG. 21 shows one way that control circuit 62 and voltage device 63 of FIG. 1 may be embodied to measure electric current, including d-c current.
- FIGS. 22A to 22E illustrate four different ways that the configurations shown in FIGS. 1 or 21 may be operated during a current-sensing mode after a demagnetizing sequence is completed.
- FIGS. 22B, 22D and 22E may be thought of as a continuation in time of FIGS. 20A, 20C and 20D respectfully, with enlarged vertical scales.
- FIG. 23 illustrates a variation of the configuration shown in FIG. 21 that may be used to improve the accuracy of a current transformer by reducing the amount that the induction level fluctuates, without provision for demagnetizing control.
- FIG. 24 shows a configuration that is similar to FIG. 21, but with a dedicated sensing winding providing a feedback signal that is proportional to induced voltage.
- FIG. 25 shows a configuration similar to FIG. 23, but modified to include demagnetizing control.
- FIG. 26 shows an embodiment configured to accurately calculate induced voltage V using analog calculation means.
- the calculation method shown takes into account the effect of various circuit impedances and stray winding impedances.
- the calculated induced voltage is configured as a feedback signal, thereby enabling the induction level to be controlled to closely follow a variable control signal.
- Sequence point identifiers A, B, C, D, E, F, and G are intended to indicate sequential progress, rather than a functional state.
- Point A indicates the first point of a sequence
- point B indicates the second point of a sequence, etc.
- the invention is based primarily on two principles of electromagnetism, herein referred to as the "first principle” and “second principle.” These two principles may be stated briefly as follows: (1) the relationship of magnetomotive force to flux density as defined by hysteresis curves of magnetic bodies, and (2) Faraday's Law applied to magnetic bodies.
- the "first principle,” the relationship of magnetomotive force (amp-turns) to flux density, is used by the invention to establish a known induction level in a magnetic body (when the initial magnitude of the induction level is not known).
- Hysteresis curves are commonly used to show the relationship of magnetic flux density (the y axis, also referred to as "induction level") to magnetic field intensity (the x axis) for a particular magnetic material.
- the magnetic field intensity within the magnetic core is fairly uniform and is proportional to the magnetomotive force (current magnitude times the number of winding turns).
- the x-axis of the hysteresis curve may be scaled to be proportional to the magnitude of current in the winding.
- a hysteresis curve may be utilized directly to find the induction level associated with a particular current magnitude.
- the same hysteresis curve could be used to find the magnitude of current (in a winding) that is necessary to drive the magnetic core to a specific induction level, and the minimum amount of current required to drive the magnetic body into saturation.
- the "second principle,” Faraday's Law applied to magnetic bodies, basically defines how a magnetic flux changing with time is associated with an electromotive force.
- emf -d ⁇ /dt (equation no. 1) where "emf' is the electromotive force (voltage caused by a changing magnetic flux) associated with a defined closed path, and -d ⁇ /dt is the time rate of change of magnetic flux ( ⁇ ) within the closed path.
- emf' is the electromotive force (voltage caused by a changing magnetic flux) associated with a defined closed path
- -d ⁇ /dt is the time rate of change of magnetic flux ( ⁇ ) within the closed path.
- the minus sign is usually included to show that the emf tends to generate a current that reduces the change in flux in accordance with Lenz's law. However, the minus sign will be dropped for the remainder of this description so that a positive change in flux will be associated with a positive induced voltage.
- a constant of proportionality may be required, depending on the system of units used.
- induced voltage is intended to refer to voltages generated by a changing flux in accordance with Faraday' s Law.
- Faraday's Law When applied to a winding with N turns positioned around a magnetic body, Faraday's Law may be modified to yield (with the minus sign dropped):
- V N (d ⁇ /dt) (equation no. 2) where V is the induced voltage (time-varying or constant) across the winding.
- Equation no. 4 shows that induced voltage V is proportional to the rate of change of induction level B. Taking the integral of both sides of equation no. 4 yields:
- Equation no. 5 which shows how a change in induction level B is related to the integral over time of induced voltage V.
- the left side of equation no. 5 will be referred to herein as a "volt-time integral.” Equation no. 5 clearly shows how the integral over time of induced voltage correlates to changes of induction level.
- the constant of proportionality (NA) will vary depending on the system of units utilized.
- Equation no. 4 and equation no. 5 are two different ways of viewing the same relationship between induced voltage and a changing induction level. This disclosure will treat these two viewpoints as equivalent, and will reference whichever one is most readily applied in a given situation. Both equations show how induced voltage may be controlled to bring about desired changes of induction level. This concept of controlling induced voltage in order to control the induction level of a magnetic body is common to all embodiments of the invention.
- the "volt-time integral" of the second principle leads to the concept that a particular "volt-time value" is associated with the change of induction level in a magnetic body from one induction level to another induction level.
- Equation no. 5 may then be written as:
- the units of this "volt-time value” will usually be volt-seconds, though other units are possible, of course.
- the volt-time value associated with a change in induction level from a "known" induction level to an induction level near zero is key to using the second principle to demagnetize a magnetic body. If the induction level is known, a volt-time value associated with a change to zero may be calculated. Alternatively, if a volt-time value associated with a change to zero is known, the actual induction level may be calculated. More specifically, let “Q 0 " designate a volt-time value associated with a change from an induction level B, to an induction level of zero. Then the relationship between volt-time value Q 0 and induction level B L may be expressed mathematically as:
- equation no. 8 may be used to determine the actual induction level in customary units.
- the induction level may be "known" only in a very general sense by knowing something about Q 0 , such as knowing the time period associated with a repeatable voltage that causes a transition between two opposite induction levels (the specific volt-time value of the voltage wave not being necessary for operation). This situation is illustrated by FIGS. 8B and 12A and the description of those figures.
- the magnitude of the volt-time value associated with a change of induction level from saturation to zero is normally constant for a particular magnetic body and winding combination.
- This constant value will be a positive number (the polarity of saturation is not included) designated “Q c " and will be referred to herein as a magnetic body's "volt-time constant.”
- This value may be calculated based on magnetic body and winding parameters, or may be determined by trial and adjustment. Alternatively, the "volt-time constant" may be determined automatically as described below under “Preferred Sequences.”
- the second principle also may be utilized to maintain a preferred induction level in a magnetic body. Considering equation no.
- the final induction level must be the same as the beginning induction level.
- an induction level may be fluctuating and causing a nonzero induced voltage, as long as the integral of induced voltage over time does not exceed a predetermined value, the preferred induction level will be maintained within acceptable limits.
- a less stringent requirement (relating to maintaining a preferred induction level) would be to require the average value (rather than the integral) of induced voltage over time to be near zero. Since average value may be determined by calculating an integral over a time period and dividing by the time period, as the time period becomes large the average value may be near zero while the integral may deviate significantly from zero. Thus, by merely requiring that the average value of induced voltage be near zero, a preferred induction level is prevented from changing quickly, but given sufficient time the induction level may drift away from the preferred induction level.
- a preferred induction level may be more strictly maintained if the induced voltage is caused to be continuously very near zero volts, in which case the induction level can change only very slowly, and only if the integral of induced voltage increases over time.
- This concept can be utilized to strictly maintain a preferred induction level in a magnetic body. Since some applications of magnetic bodies are adversely affected by a fluctuating induction level, this concept may be used to advantage to reduce the amount that the induction level fluctuates.
- current transformers are one such application. The accuracy of a current transformer may be improved by reducing induction level fluctuation and associated magnetizing current required for operation.
- the induction level of a magnetic body may also be controlled to vary in a preferred manner.
- the induced voltage may be controlled so that the integral over time of induced voltage is proportional to the control signal (per equation no. 5).
- the induced voltage may be controlled to be proportional to the derivative of the control signal (per equation no. 4).
- the induction level may be controlled so as to match the reference induction level.
- a feedback signal that is directly proportional to induced voltage may be produced by a dedicated sensing winding (as shown in FIG. 24), or a feedback signal may be derived from winding current (as in FIGS.21, 23, 5 and 26).
- the control method utilized may include a form of Proportional plus Integral plus Derivative (P.I.D.) control, or other control scheme such as fuzzy logic.
- P.I.D. Proportional plus Integral plus Derivative
- magnetic body is used in a general sense to refer to a mass of magnetic material.
- magnetic material refers to material that has a relative permeability significantly greater than a value of one.
- magnetic core or simply “core” is intended to refer to a magnetic body that is in a particular spatial relationship to one or more current-carrying conductors.
- induction level is intended to be synonymous with magnetic flux density.
- windings utilized throughout industry are most commonly made of copper or aluminum conductors, it should be understood that the present invention is not limited to standard types of windings. Windings of widely varying form made of almost any kind of conductive material may be utilized, including superconducting materials and semiconducting materials.
- Stray impedances associated with a winding may be defined to include not only the effects of wire resistance and stray inductance, but may also be defined to include the effects of eddy currents, core losses and other imperfections associated with a particular core/winding combination. Stated another way, the stray impedances of a winding may be defined in whatever way results in the most accurate determination of induced voltage
- induced voltage being understood to be approximately calculable as the voltage measured across a winding minus a voltage drop associated with winding current flowing through stray impedances.
- This issue is most applicable to embodiments of the invention that utilize a feedback signal that is derived from winding current.
- Different applications of the invention may find it advantageous to define stray impedances differently in order to optimize each application. It is not the intention of this disclosure to restrict the term "stray impedances" to a firm definition.
- current-sensing is used herein in a broad sense to refer not only to means of deriving information about a current, but also to refer to more direct uses, such as means of actuating a control mechanism based on current characteristics (such as a calibrated solenoid-type actuator that may be part of a protective relay).
- the phrase "known induction level” is intended to be synonymous with “determinate induction level.” Both phrases are used in a broad sense of having some kind of data that quantifies the state of magnetic flux density (induction level) in any way that can be used to implement the "second principle” (Faraday's Law applied to magnetic bodies) to move the induction level to a preferred induction level.
- the induction level is considered to be “known” (or “determinate") if the approximate volt-time value (equation no. 6) that is associated with a change of induction level to zero is known.
- the induction level is also considered to be “known” if the time period required to change the induction level to near zero is known.
- a parameter that is said to be a "function of time” is a parameter that may vary with time, even though the parameter may be constant for a period of time.
- varying voltage is intended to refer in general to a voltage that is a function of time. While relatively constant voltages are utilized at times in the preferred embodiments, it should be understood that non-constant voltages could also be used, and the term
- varying voltage is intended to include both possibilities.
- waveform is intended to mean the geometrical shape of a quantity when displayed as a --unction of time.
- alternating (as applied to current or voltage) is intended to indicate that the polarity changes between positive and negative values, without necessarily implying a consistent repeating waveform.
- a "half-cycle" of an "alternating" waveform refers to a part of the waveform continuously having the same polarity.
- a controllable active “voltage source” to produce “varying voltages”
- a controllable “electric energy source” may be utilized, meaning an active source that is able to generate suitable voltages and currents, while not necessarily being configured as a voltage source (the term “voltage source” usually indicates an electric energy source having output voltage that is relatively unaffected by current magnitude).
- a voltage source is usually the preferred form of electric energy source. Even so, the use of "voltage sources” in the preferred embodiments is not intended to restrict the invention from being embodied with other types of electric energy sources.
- the output should be current-limited to an appropriate value to ensure safe current levels during brief time periods that the magnetic bodies may be saturated.
- the preferred embodiment utilizes two sequential phases to establish a preferred induction level in a magnetic body when the initial magnitude of the induction level is not known.
- these two phases will be called the “first phase” and the “second phase.”
- the first phase primarily utilizes the "first principle” (the hysteresis-curve relationship of magnetomotive force to flux density) to establish a known induction level.
- the “second phase” utilizes the “second principle” (Faraday's Law applied to magnetic bodies) to change the induction level to a preferred induction level.
- Each of these two phases may involve several steps or just a single step, depending on the preferred operation for a particular application.
- a "third phase” may be applicable, during which the varying voltage may be turned off, or used to maintain the preferred induction level, or otherwise utilized depending on the particular application.
- first-phase method a first-phase metliod b
- first-phase method c first-phase method c
- First-phase method c (applying an alternating current and dete ⁇ nining a volt-time value) has the advantage of not requiring advance knowledge about the properties of the magnetic body.
- the second phase After a known induction level is reached by the first phase, the second phase starts, changing the known induction level to the preferred induction level (utilizing the second principle). This also may be done in several ways. However, all of these ways are constrained by the second principle to provide a voltage for a period of time such that the volt-time integral of induced voltage corresponds to the volt-time value necessary to change the induction level from the known induction level to the preferred induction level. Three ways to do this are:
- first voltage and second voltage of opposite polarities; the first voltage having such magnitude, duration and polarity so as to cause the induction level of the magnetic body to transition to and pass the preferred induction level; the second voltage having such magnitude, duration and polarity so as to cause the induction level of the magnetic body to transition to the preferred induction level.
- first voltage and second voltage may be referred to herein as a "first-step voltage” and “second-step voltage” respectively).
- second-phase method a second-phase method b
- second-phase method c second-phase method c
- “Second-phase method b” is an improvement over “second-phase method a" because the second voltage (with opposite polarity as the first voltage) can improve accuracy by eliminating a small d-c offset current that may be present after the first voltage.
- a first voltage of metliod a or b
- the second voltage used in method b is intended to eliminate this "coercive force error.”
- “Second-phase method c” is intended to cover all possibilities not covered by methods “a” and “b.” This includes alternating voltage pulses with declining volt-time values, which may also be used to eliminate any “coercive force error” (similar to method "b,” but with more than two pulses).
- a first mode also called a "demagnetizing mode”
- demagnetizing mode during which the current transformer is demagnetized (utilizing the first and second phases discussed previously).
- a second mode also called a "current-sensing mode"
- current transformer secondary current is proportional to primary current (this may be considered to be a "third phase”).
- the current transformer may become magnetized again due to the same problems that magnetized it in the first place. For this reason, preferred operation may sequence between the demagnetizing mode and the current-sensing mode.
- the demagnetizing aspect of the invention is best suited for applications that sense current periodically, with some time between sensing periods available for use with the demagnetizing mode. (In applications in which it is desirable to sense current continuously without using demagnetizing control, current transformer accuracy may still be improved by reducing the amount that the induction level fluctuates during current-sensing operation).
- three modes of operation may function sequentially (rather than just two modes as previously discussed):
- a power supply charging mode during which the current transformer may become magnetized.
- the magnetomotive force (mmf) applied to the magnetic core is the sum of the mmf of the primary current and the secondary current. These two mmf s are usually of opposite polarity and largely cancel each other during normal current transformer operation. These two mmf s do not normally cancel each other during the demagnetizing mode of the invention.
- FIG. 1 illustrates the general concept by showing a magnetic body 60, a winding 61 wrapped around the magnetic body, and a controllable voltage device 63 with a suitable control circuit 62.
- Magnetic body 60 and winding 61 are shown in a typical current transformer configuration with the magnetic body around an electric power system conductor 64 with an insulating covering 65.
- Power system conductor 64 functions as a primary winding with only one turn, with a primary current Jl flowing. Though shown with one end disconnected, power system conductor 64 is normally connected as part of an electric power system.
- Controllable voltage device 63 may be an active voltage source (or, more generally, a controllable electric energy source), in which case a primary winding is not required. In this case, power system conductor 64 (acting as a primary winding) may be omitted.
- Resistor Rl may be a current-sensing resistor, or it may represent the internal resistance of voltage device 63, or it may be a current-limiting resistor. Some applications may find it advantageous to replace resistor Rl with a complex impedance. Resistor Rl may be omitted for many applications.
- winding 61, voltage device 63, and resistor Rl are connected in series, so that current J2 flows through each of them.
- controllable voltage device 63 may be an adjustable impedance.
- resistor Rl is usually included as a current-sensing resistor with a low value of resistance, and controllable voltage device 63 may be an adjustable impedance rather than an active voltage source.
- current is sensed by a larger monitoring system as voltage V3 across resistor Rl, since this voltage is proportional to current J2.
- Winding 61 is shown with ten turns around magnetic body 60.
- the actual number of turns may vary widely depending on the application.
- Magnetic body 60 is shown as a toroid, though wide variation in magnetic body configurations is possible and the illustration is not intended to limit the breadth of application of the invention.
- Control circuit 62 is shown with three high-impedance voltage-sensing inputs (connected to conductors 66, 67, and 68) to enable sensing of voltage VI, voltage V2, and voltage V3.
- Voltage V3 across resistor Rl is proportional to current J2.
- Voltage V2 is the voltage across winding 61, which is the induced voltage generated by changing flux in magnetic body 60 plus any voltage drop associated with current J2 flowing through stray winding impedances. Often, the voltage drop associated with current J2 flowing through stray winding impedances is small compared to the induced voltage and may be ignored in some applications.
- Voltage VI is the output voltage of controllable voltage device 63.
- Voltage-sensing conductors 66, 67 and 68 are not required for the simplest embodiments of the invention, but are included to clarify the general concept. Alternatively, other means of sensing current and/or voltages may be utilized, as may be preferred for different embodiments.
- control conductors 69 act as an interface between control circuit 62 and controllable voltage device 63.
- the actual interface between control circuit 62 and controllable voltage device 63 may vary widely depending on the particular design.
- Control circuit 62 may be constructed utilizing prior art, with the control sequences being in accordance with the present invention.
- controllable voltage device 63 is an active voltage source, it also may be constructed utilizing prior art, with the control sequences being in accordance with the present invention.
- magnetic body 60 provides a closed magnetic path with no need for other magnetic components.
- FIG. 2 shows an alternate configuration, in which a cylindrical magnetic body 70 does not provide a closed magnetic path by itself. Instead, a closed magnetic path is provided by additional magnetic material 72.
- magnetic material 72 should be configured to have greater saturation flux capacity than magnetic body 70, so that the saturation characteristics of magnetic body 70 dominate the resulting magnetic circuit characteristics.
- One way this may be done is by configuring magnetic material 72 to have a greater cross- sectional area than magnetic body 70. Since the configuration shown in FIG. 2 does not have a primary winding (or other source of magnetic excitation), controllable voltage device 63 (shown in FIG. 1) is now shown as a controllable electric energy source 73. A winding 71 is wrapped around magnetic material 72. As in FIG.
- resistor Rl may be a current- sensing resistor, or it may represent the internal resistance of electric energy source 73, or it may be a current- limiting resistor.
- FIGS. 3 and 4A to 4D illustrate how a magnetic body may be demagnetized utilizing "first-phase method a" and "second-phase method b" previously discussed.
- FIG. 3 is a hysteresis curve that clarifies one way that the induction level of a magnetic body is controlled by the invention. These changes correlate to the waveforms shown in FIGS. 4A to 4D.
- the horizontal axis X of FIG. 3 is proportional to magnetomotive force (ampere-turns), and the vertical axis Y is proportional to the induction level (magnetic flux density) of the magnetic body. Magnetization of the magnetic body increases as the operating point moves away from axis X.
- a magnetic material with a relatively square hysteresis curve has purposely been chosen, as this simplifies the demagnetizing operation.
- the induction level of a magnetic body begins at a random magnetized state shown as point A of FIG 3.
- the "first phase” moves the induction level to saturation near point B.
- the “second phase” shown includes two steps. A “first step” moves the induction level somewhat passed an induction level of zero to point C. A “second step” moves the induction level to near zero at point D.
- FIGS. 4A to 4D show one method of controlling voltages and currents to demagnetize a magnetic body by using the sequence shown in FIG. 3.
- the waveforms shown in FIGS. 4 A to 4D are applicable to the configuration shown in FIG. 2, or to FIG. 1 when controllable voltage device 63 is an active voltage source and primary conductor 64 is either disconnected or omitted.
- the waveforms shown in FIGS. 4A to 4D correlate to the sequential changes of induction level shown in FIG. 3.
- Voltage V2 shown in FIG. 4B is the induced voltage across the winding, with any voltage drop associated with current flowing through stray winding impedances assumed to be negligible for the present discussion.
- FIG. 4D shows how the induction level varies with time.
- the vertical axis represents induction level, and is scaled similar to the vertical axis of the hysteresis curve shown in FIG. 3.
- the "first phase" of the demagnetizing cycle begins when voltage VI is driven from zero volts to a positive value by controllable voltage device 63 as controlled by control circuit 62.
- This positive voltage causes the induction level of magnetic body 60 to transition from point A of FIG. 3 to saturation at point B.
- Saturation begins at about time T42 when current J2 suddenly increases as shown in FIG. 4C.
- the transition point is also marked by voltage V2 suddenly decreasing, as shown by FIG. 4B.
- the "first step" of the second phase begins when the polarity of voltage VI is reversed and the induction level of magnetic body 60 begins a transition from point B of FIG. 3 to point C.
- control circuit 62 controls the time period of the first step in order to change the induction level by an amount somewhat greater than change Yl, stopping at point C at time T44.
- an optional "second step” begins when the polarity of voltage VI is reversed again and the induction level of magnetic body 60 begins a transition from point C of FIG. 3 to point D.
- the second step ends when voltage VI is changed to zero volts, and magnetic body 60 is left in a demagnetized state at point D (at an induction level near zero).
- Change Yl correlates to the "volt-time constant" previously discussed.
- the voltage and time periods associated with the first step and second step should be controlled such that the volt-time integral of voltage V2 is the same as the "volt- time constant" of magnetic body 60 and winding 61.
- the volt-time constant (of magnetic body 60 and winding 61) is related to areas Al and A2 (with the hatched areas calculated in units of volt-time). More specifically, area Al minus area A2 should be the same as the volt-time constant.
- control circuit 62 is configured beforehand to control voltage device 63 in the manner illustrated to effectively demagnetize magnetic body 60.
- FIG. 5 and FIGS. 6A to 6D illustrate the same sequence as FIG. 3 and FIGS. 4A to 4D, except that the preferred induction level is not zero.
- the preferred induction level is 50% of saturation on the positive side of the hysteresis curve.
- the sequence shown in FIG. 5 is the same as FIG. 3, except that in FIG. 5 the transition to sequence step
- FIGS. 7 and 8A to 8D illustrate "first-phase method c" and "second-phase method b" previously discussed.
- FIG. 7 shows a hysteresis curve sequence that can be used to automatically determine the volt-time constant of a magnetic body and winding and then demagnetize the magnetic body.
- FIGS. 8A to 8D show how an alternating voltage and current may be used to implement the sequence shown in FIG. 7.
- voltage VI (the output voltage of voltage device 63) between times T53 and T57 is the same as in FIG. 4A between times T41 and T45. In FIG. 8A, however, voltage VI begins the demagnetizing cycle with a negative polarity between times T51 and T53. This initial negative voltage moves the induction level from an unknown level shown as point A on the hysteresis curve of FIG. 7 to saturation at point B of FIG. 7. Saturation near point B is reached at about time T52. At time T53 of FIG. 8A voltage VI changes to positive polarity and drives the induction level toward point C of FIG. 7. Starting at time T53, control circuit 62 monitors either voltage V2 or current J2 to determine when saturation is reached.
- the volt-time constant of the magnetic body may be determined by calculating the volt-time integral of voltage V2 between times T53 and T54 (this calculation determines the volt-time area shown as crosshatched area A3) and dividing the result by two.
- voltage V2 is then controlled so that the volt-time integral from time T55 to time T57 has the same magnitude as the calculated volt-time constant.
- voltage V2 is controlled during the second phase so that area Al minus area A2 is half of area A3, Time T56 indicates the transition point between the first and second step of the second phase.
- the effect of winding resistance and current on voltage V2 may be automatically compensated for by the control circuit automatically determining winding resistance and calculating the volt-time integrals using corrected voltage.
- the winding resistance may be calculated as voltage V2 divided by current J2 between times T52 and T53 while the magnetic body is in saturation. Corrected voltage (induced voltage) is then calculated as voltage V2 minus (current J2 multiplied by winding resistance).
- total loop resistance may be calculated as voltage VI divided by current J2 (between times T52 and T53), and induced voltage may then be calculated as voltage VI minus current J2 multiplied by total loop resistance.
- FIG. 9 illustrates how a magnetic body may be demagnetized utilizing "first-phase method b" and "second- phase method b" previously discussed.
- FIG. 9 shows operation with a not-so-square hysteresis curve assumed for a magnetic body. Operation shown is similar to that described for FIG. 3 except that now current may not be large enough to drive the magnetic body to saturation. Similar to FIG. 3, horizontal axis X is proportional to magnetomotive force (ampere-turns), and vertical axis Y is proportional to the induction level (magnetic flux density) of the magnetic body. Magnetization of the magnetic body increases as the operating point moves away from axis X.
- the induction level of a magnetic body generally is not known. This corresponds to a randomly chosen point A on the hysteresis curve of FIG. 9.
- the "first phase" of the demagnetizing cycle causes a current to flow with peak magnitude corresponding to magnetomotive force X2. This causes a transition to point B, with an induction level of Y3 (which is not at saturation). Induction level Y3 may be calculated based on peak current and known characteristics of the magnetic body (in accordance with the "first principle").
- the degree of accuracy obtainable in determining induction level Y3 is somewhat dependent on the initial induction level, the magnitude of magnetomotive force X2, and the actual hysteresis characteristics of the magnetic body. Accuracy may be improved by cycling the current between opposite polarities once to remove the effects of an initial magnetization.
- a first voltage is applied to the winding for a time period such that the induction level of the magnetic body transitions from point B to point C (in accordance with the "second principle").
- a second voltage of the opposite polarity is then applied for a time period such that the induction level of the magnetic body transitions to point D.
- the magnetic body is demagnetized.
- FIGS. 10, 11 and 12A to 12C illustrate "first-phase method c" and "second-phase method b" previously discussed.
- FIG. 10 shows a method to demagnetize a magnetic body regardless of whether current is adequate to drive the magnetic body to saturation.
- sequence point A is assumed for a starting point.
- at least one complete cycle of an alternating current is first applied. This forces the magnetic core first to point B, then down to point C, cycling (almost) once around its hysteresis loop, thereby removing the effects of the initial induction level.
- FIG. 11 shows a relatively simple embodiment of the invention utilizing a controllable electric energy source 76.
- a winding 71A conducts current J2 and interacts with a magnetic body (not shown).
- the winding may be similar to windings 61 and 71 of FIGS. 1 and 2.
- the magnetic body (not shown) may be similar to either magnetic body 60 or 70 of FIGS. 1 and 2.
- a switch 78 is used to connect and disconnect an alternating voltage source 77 to winding 71A.
- Switch 78 will preferably be a solid-state electronic switch able to stop current flow mid-cycle.
- Alternating voltage source 77 may simply be a source of 50 Hertz or 60 Hertz electric power with suitable voltage magnitude, providing a consistent repeating waveform that causes induced voltage V2 to have a consistent repeating waveform.
- An optional resistor R77 may be included to limit peak current levels.
- An optional varistor 79, or other transient voltage suppression device may be included to limit transient voltages when switch 78 is opened.
- a control circuit 62A senses voltage V2 and controls switch 78. Zener diodes D21 and D22 have breakover voltages somewhat higher than the peak voltage of alternating voltage source 77, and are configured to limit transient voltages when switch 78 is opened.
- Zener diodes D21 and D22 may be omitted, in which case switch 78 should be configured to be to able to absorb the high momentary voltage associated with stopping current J2 very quickly.
- switch 78 should be configured to be to able to absorb the high momentary voltage associated with stopping current J2 very quickly.
- almost any kind of surge suppressor or impedance could be used in place of the zener diodes, but zener diodes are shown for ease of illustration and explanation.
- FIGS. 12A to 12C One simple method of operation is illustrated by FIGS. 12A to 12C.
- FIGS. 12A to 12C show one way that the configuration shown in FIG. 11 may utilize an alternating voltage and current to implement a sequence similar to the hysteresis curve sequence of FIG. 10.
- FIG. 12A shows voltage V2, which (prior to time T74) is merely the sinusoidal voltage generated by alternating voltage source 77 (less any voltage drop across resistor R77).
- FIG. 12B shows current J2, which is not quite sinusoidal due to hysteresis characteristics of the magnetic body.
- FIG. 12C shows induction level.
- Switch 78 (FIG. 11) is closed prior to time T71, and any transient d-c offset currents that may be associated with closing switch 78 have already decayed prior to time T71.
- FIGS. 10, 11, and 12A to 12C Between times T71 and T72 the operation is in a steady-state alternating cycle.
- the induction level is at a positive peak (FIG. 12C), corresponding to sequence point D of FIG. 10.
- the induction level then declines to an induction level corresponding to point F at time T74, at which time switch 78 is opened by control circuit 62A.
- a negative current is flowing at time T74, and this current is forced to flow through zener diodes D21 and D22 after switch 78 opens (the current keeps flowing due to the inductive nature of winding 71A).
- This causes a positive voltage pulse between times T74 and T75 with area A8 (FIG 12A), which is associated with a transition from point F to point G of FIG. 10, at which point current J2 has declined to zero and the induction level is near zero.
- area A6 of FIG. 12A is twice the volt- time value associated with a change in induction level from the peak induction level to an induction level near zero.
- the actual peak induction level may be calculated from the volt-time value using equation no. 8 (above).
- the timing of the opening of switch 78 should be such that area A7 minus area A8 equals half of area A6.
- time T74 (the point of time that switch 78 is opened) should be somewhat past the geometric middle of the half-cycle for best results.
- the demagnetizing operation will be satisfactory for many applications if the switch is simply opened near the geometric middle of the half-cycle.
- "Geometric middle of the half-cycle” is intended to mean the point of time at which the volt-time integral of the half-cycle would be divided into two equal parts (visually this would be the time at which area A7 would be exactly half of area A6).
- Preferred induction levels other than zero may also be established by opening switch 78 at other points of time during the same half-cycle. Equation no.
- FIG. 6 shows how a volt-time value may be calculated which may be used to determine the optimum point of time to open switch 78 to establish a preferred induction level. Opening switch 78 between times T73 and T74 will result in induction levels between point E and an induction level of zero, depending on the specific point of time that switch 78 is opened. Opening switch 78 after T74, but still during the same half-cycle, will result in induction levels with opposite polarity.
- the waveforms shown in FIGS. 12A to 12C are waveforms typically associated with peak induction levels being less than saturation induction levels.
- the voltage magnitude of alternating voltage source 11 is sufficient to drive the magnetic body to saturation, the alternating waveforms shown prior to T74 may be considerably different than those shown, but the principles of operation are similar.
- FIGS. 13 to 26 all illustrate how the invention may be applied to current transformers that are in service. The principles utilized, however, are applicable to other types of magnetic bodies as well.
- FIGS. 13 through 19 illustrate how an adjustable impedance may be utilized in a current transformer secondary circuit to demagnetize a current transformer.
- FIGS. 20 through 25 illustrate how an active voltage source may be utilized in a current transformer secondary circuit.
- FIG. 13 shows a functional schematic illustrating how voltage device 63 of FIG. 1 may be embodied by an adjustable impedance, which is part of a current transformer demagnetizing circuit 1.
- Demagnetizing circuit 1 includes a control circuit 3 and an adjustable impedance 2 comprising electronic switches SI, S2 and S3 and zener diodes Dl and D2.
- Electronic switches SI, S2 and S3 are controlled by control circuit 3 as indicated by the dashed lines between switches SI, S2 and S3 and control circuit 3.
- adjustable impedance 2 is for illustration purposes only, and is not intended to define all possible configurations of the present invention. It may be noted that switch SI is optional, since closing switches S2 and S3 at the same time approximates the effect of closing switch SI.
- FIG. 13 shows a current transformer CT1 which may have a magnetic core similar to magnetic body 60 of FIG. 1 and a secondary winding similar to winding 61 of FIG. 1.
- Conductor 64 is a primary conductor conducting a primary current Jl, also similar to FIG. 1.
- Current Jl is an alternating electric current flowing as part of a larger system.
- Current Jl causes secondary current J2 to flow by the transformer action of current transformer CT1.
- Current J2 is normally proportionally smaller than current Jl by the turns ratio of current transformer CT1. Under ideal conditions, the waveform of current J2 is virtually the same as the waveform of current Jl,
- Resistor Rl is a current-sensing resistor, connected in series with current J2, thereby producing a voltage signal across resistor Rl that is proportional to current J2.
- This voltage signal, conducted by conductor 15, is usually used as an input to some kind of current monitoring system as provided for by terminal 16.
- This voltage signal may also be used as an input to control circuit 3 as is presently shown.
- Other kinds of current-sensing means may be used in place of resistor Rl.
- a power supply 4 is configured to derive power from input current J2 whenever switch S4 is closed and switches SI, S2 and S3 are open. Power supply 4 and switch S4 are optional (switch S4 may be optional even if power supply 4 is included, depending on the configuration of power supply 4). If power is not derived from current J2, a separate source of power will usually be required to provide operating power to control circuit 3. Power is supplied to control circuit 3 via one or more conductors 13 and common conductor 11, the total number of power conductors being dependent on the specific design of control circuit 3 and power supply 4. Terminals 12 and 14 are included to provide for the possibility of power being transferred to or from other circuits.
- Conductor 17 and terminal 18 provide for a control signal to or from a larger system to coordinate current- sensing and demagnetizing modes.
- Voltage V2 is now the voltage across the secondary winding of current transformer CT1. This voltage is shown connected as an input to control circuit 3 via conductor 15 and conductor 19. For the present discussion, voltage V2 will be considered to be the induced voltage (the voltage drop associated with current flowing through stray winding impedances will be considered negligible).
- Optional ground connection 10 provides a stable signal reference potential for common conductor 11.
- FIGS. 14A to 14E illustrate the preferred operation of FIG. 13.
- the magnetic sequence utilized is similar to the sequence shown by the hysteresis curve of FIG. 3, utilizing "first-phase method a" and "second-phase method b" previously discussed.
- Three different operating modes are shown. First, from time Tl to time T2 the power supply charging mode is active. Second, from time T2 to time T10 the demagnetizing mode is active. Third, after time T10 the current-sensing mode is active. Actual test waveforms may vary somewhat from those shown depending on the specific power supply configuration, current transformer characteristics, and current magnitudes.
- FIG. 14A shows current transformer CT1 primary current Jl as a simple sine wave for simplicity of illustration, though in many applications it may be considerably distorted.
- FIG. 14B shows secondary current J2. Current magnitudes have been normalized for simplicity of illustration. Current J2 is normally many times smaller than current Jl, as influenced by the turns ratio of current transformer CT1.
- FIG. 14C shows operation of electronic switches SI, S2, S3 and S4 as controlled by control circuit 3. Dark lines indicate time periods during which each switch is closed. Blank spaces indicate time periods during which each switch is open.
- FIG. 14D shows secondary voltage V2. The voltage across resistor Rl is normally small compared to voltage V2, and no attempt has been made to show its minor influence on V2.
- FIG. 14E shows induction level with a scale similar to the hysteresis curve shown in FIG. 3.
- FIGS. 14A to 14E begins with the functional schematic of FIG. 13 being in the power supply charging mode. Between times Tl and T2, only switch S4 is closed, and power supply 4 is charging.
- the wavefo ⁇ ns shown in FIGS. 14B and 14D between times Tl and T2 are discussed in detail in U.S. Patent No. 6,018,700 for a "Self-Powered Current Monitor" to Thomas G. Edel, issued January 25, 2000.
- the first phase of the demagnetizing mode begins at time T2, a somewhat random point of time.
- the duration of the first phase lasts a predetermined time period that is adequate to drive the magnetic core to saturation as indicated by point B of FIG. 3.
- switch S2 closes and switch S4 opens to set adjustable impedance 2 to the nonlinear characteristics of zener diode Dl.
- This causes voltage V2 to be large and positive for current J2 with positive polarity, and small and negative for current J2 with negative polarity.
- the large positive voltage causes the induction level to transition to saturation.
- switch SI is optionally closed between times T3 and T4 and between times T5 and T6 to minimize the magnitude of negative voltages that occur during negative half-cycles of current J2.
- Switch SI is controlled during the first phase based on the polarity of current J2.
- the predetermined minimum time required for the first phase ends sometime between times T5 and T7. After this time period ends, control circuit 3 looks for the beginning of a negative half-cycle of current J2 to begin the second phase.
- the second phase of the demagnetizing mode begins at time T7 when control circuit 3 senses current J2 going negative.
- switch S3 closes and switch S2 opens.
- a first-step voltage pulse begins, the magnitude of which is the breakover voltage of zener diode D2.
- Control circuit 3 keeps switch S3 closed for a predetermined time period (until time T8) that correlates with the volt-time value associated with a transition from point B to point C of FIG. 3.
- An optional second step voltage is implemented between times T9 and T10 by closing only switch S2, thereby causing the transition from point C to point D of FIG. 3.
- the current-sensing mode begins a time T10 when switch SI is closed, thereby allowing current to flow freely.
- control circuit 3 While it is preferable for control circuit 3 to have current J2 and/or voltage V2 as a feedback signal, it is possible for control circuit 3 to operate without any feedback signal. This could be accomplished by modifying the first-step voltage pulse of the second phase (between times T7 and T8 of FIG. 14D) in either of the following ways:
- control circuit 3 may operate without feedback signals related to secondary voltage V2 or secondaiy current J2.
- FIGS. 15 A to 15E illustrate how the functional schematic of FIG. 13 may be operated without a feedback signal.
- FIGS. 14A to 14E operation is similar to FIGS. 14A to 14E.
- the second phase begins at a random time in the middle of a half-cycle when the first phase times out.
- Zener diode D2 is activated briefly several times over a time period one cycle in length (between times Til and T12) to generate a voltage V2 waveform (FIG. 15D) such that the integral of the voltage over the time period between times Til and T12 is approximately equal to that of FIG. 14D between times T7 and T8.
- zener diode D2 is made active with a pulse-width-modulated type of control (controlling switches SI and S3) to cause voltage V2 to have an average negative value suitable for the first step of the second phase (between times Til and T12).
- the demagnetizing effect of the multiple pulses shown in FIG. 15D is similar to the single voltage pulse shown in FIG. 14D (between times T7 and T8).
- the optional second-step voltage begins.
- zener diode Dl is made active with a pulse-width-modulated type of control (controlling switches SI and S2) to cause voltage V2 to have an average positive value suitable for the second step (between times T12 and T13).
- the effect of the multiple pulses shown in FIG. 15D is similar to the single voltage pulse shown in FIG. 14D (between times T9 and T10).
- the current-sensing mode begins.
- FIG. 16 shows one possible embodiment of a current transformer demagnetizing circuit based on the functional schematic of FIG. 13. Components that are common to FIG. 13 function in the manner previously described.
- the demagnetizing circuit shown is incorporated into a current monitoring circuit that derives operating power from current transformer CT1 secondary current J2.
- the electronic switches previously shown in the functional schematic of FIG. 13 are now implemented with field-effect transistors.
- the preferred embodiment utilizes N-channel enhancement mode devices with low drain-source on resistance and sensitive gates for operation at logic voltage levels.
- the power supply circuit includes a full wave bridge rectifier circuit consisting of diodes D5 and D6 and the drain-source diodes within field-effect transistors Q5 and Q6.
- the rectified current charges capacitor Cl, which provides an unregulated voltage for use by a d-c to d-c converter circuit 23. Voltage across capacitor Cl is limited by the breakover voltage of zener diodes Dll and D12.
- Resistor Rl is still a current-sensing resistor, but in this configuration current J2 flows through resistor Rl only when field-effect transistors Q5 and Q6 are actuated. Since the current through resistor Rl is not always the same as current J2, it is shown as current J3.
- Conductor 29 conducts a voltage signal that is proportional to the current through resistor Rl.
- Resistor R2 and zener diodes D7 and D8 form an optional surge suppression network to protect an analog-to-digital converter circuit 22.
- the resistance of resistor R2 must be small compared to the input resistance of analog-to-digital converter circuit 22 so that accuracy of the voltage signal will not be adversely affected.
- Analog-to-digital converter circuit 22 and a microcontroller 21 function in a manner similar to control circuit 3 of FIG. 13. Data is communicated between analog-to-digital converter circuit 22 and microcontroller 21 via serial or parallel communication utilizing several conductors 35.
- Zener diodes Dll and D12 function in a manner similar to zener diodes Dl and D2 of FIG. 13. Zener diodes Dll and D12 also act as voltage-limiting devices to limit the charge on power supply capacitor Cl.
- Field-effect transistors Q5 and Q6 perform the switching functions of switches SI, S2, and S3 of FIG. 13.
- Field-effect transistors Q5 and Q6 are individually controlled, with resistors R8 and R9 providing a discharge path for gate charge (these resistors may be optional depending on the configuration of microcontroller 21).
- Activating field-effect transistor Q6 is similar to closing switch S2 of FIG. 13.
- Activating field-effect transistor Q5 is similar to closing switch S3 of FIG. 13.
- Activating both field-effect transistors Q5 and Q6 is similar to closing switch SI of FIG. 13.
- switch S4 in FIG. 13 is not required in the embodiment of FIG. 16, since the power supply is configured to automatically charge whenever the current-sensing mode and demagnetizing mode are not active.
- the power supply may be easily eliminated if an alternate power source is available for control power. This may be done by removing diodes D5 and D6, Capacitor Cl, and d-c to d-c converter circuit 23.
- More than one current transformer may be connected to the configuration of FIG. 16 by duplicating the components within box 41 for each current transformer. These components may be connected to conductors 11A and 42, similar to the configuration shown for current transformer CT1. Additional connections to analog-to-digital converter circuit 22 and microcontroller 21 will also be needed for each current transformer, similar to conductors 29, 43, and 44 for current transformer CT1. With additional current transformers configured similar to current transformer CT1, each current transformer may contribute power to the power supply.
- More than one current transformer may be demagnetized at a time, assuming analog-to-digital converter circuit 22 and microcontroller 21 are capable of the faster processing and communication required for this (unique control signals are preferred for each current transformer in order to implement feedback control similar to FIGS. 14A to 14D). If a non-feedback type of control is utilized (similar to FIGS. 15A to 15E), then only two control conductors (43 and 44) may be used to simultaneously demagnetize all current transformers at once with little demand on the microcontroller.
- FIGS. 17A to 17D illustrate the operation of the embodiment shown in FIG. 16.
- the operation shown is similar to the operation described in FIGS. 14A to 14E.
- the embodiment shown in FIG. 16 may also be operated in a manner similar to that described for FIGS. 15A to 15E.
- FIG. 17A shows alternating electric current Jl, similar to FIG. 14A.
- FIG. 17B shows secondary current J2, similar to FIG. 14B.
- FIG. 17C shows the time periods that the field-effect transistors are actuated.
- a solid line indicates a high gate voltage, whereas a blank space indicates a low gate voltage.
- the switching sequence shown is functionally similar to FIG. 14C.
- FIG. 17D shows secondary voltage V2, similar to FIG. 14D.
- the voltage-limiting circuit is comprised of zener diodes Dll and D12, so the waveforms associated with the power supply charging cycle (between times Tl and T2) are different than in FIG. 14D. Also, the effect of capacitor Cl charging is evident at the beginning of each voltage pulse.
- FIG. 18 shows a veiy simple embodiment of a current transformer demagnetizing circuit, utilizing only one field-effect transistor for an adjustable impedance. All components required only for self-powered applications have been removed, so a separate source of operating power is required.
- a power supply 23A is now included, which provides regulated d-c operating power from a separate power source. To emphasize the simplicity of the present invention, surge suppression devices have been removed as well as gate resistors (all of which may be considered optional, depending of the particular application and microcontroller configuration). Operation is similar to FIG. 16, but now the breakover voltage of the internal body diode of transistor Q5 is utilized for voltage control during the first phase, and the forward voltage drop of the body diode is used for the voltage required by the second phase. This configuration also has the advantage of the current transformer being directly connected to grounded conductor 11, which is a requirement of some safety codes for some applications.
- FIG. 18 illustrates a very simple method of implementing an adjustable impedance, and illustrates the use of a single three-terminal device as an adjustable impedance.
- the main drawback to the configuration shown in FIG. 18 is that the low voltage available for the second phase will often require that the second phase be spread over several cycles.
- One way of increasing this voltage (and thereby reducing the time required) is to use two transistors in series with a common gate connection. This will double the second phase voltage to about 2 volts (the forward voltage drop of two diodes). Of course additional transistors could be added in series to increase the voltage still further. It should be noted from this that it is possible to make a single three-terminal semiconductor device (utilizing prior-art MOSFET technology) that has on/off parameters well suited for use as an adjustable impedance in this type of circuit.
- FIGS. 19A to 19E show typical operating waveforms and control signals associated with the embodiment shown in FIG. 18.
- the sequence shown utilizes "first-phase method a" and "second-phase method c" previously discussed.
- second phase voltage is merely the forward voltage drop of the body diode of transistor Q5.
- the first phase voltage is the breakover voltage of the internal body diode of transistor Q5, and this voltage is sufficient to drive the induction level to saturation in less than one half-cycle of current J2.
- the second phase voltage is not sufficient to complete the second phase transition (from saturation to an induction level near zero) in a single half-cycle of current J2.
- transistor Q5 must be turned on during the following positive half-cycle (from time T66 to time T67, to avoid going back to saturation), and turned off again at the beginning of the next negative half-cycle to continue moving the induction level toward zero. This type of cycling continues until time T68, at which time the induction level is near zero.
- the combined area of area A12, area A13, and area A14 should equal the volt-time constant of the current transformer.
- FIGS. 20 through 25 all illustrate how an active voltage source may be utilized in a current transformer secondary circuit to demagnetize a current transformer and also improve the accuracy of a current transformer by reducing the amount that the induction level fluctuates during current-sensing operation. These figures also show how a preferred induction level may be maintained by causing the average value of the induced voltage over time to have a value near zero. The principles illustrated are applicable to other magnetic bodies in addition to current transformers.
- FIGS. 20A to 20D show how the configuration of FIG. 1 may be used to demagnetize a current transformer using an active voltage source, thereby improving current transformer accuracy and enabling measurement of d-c current.
- the sequence shown utilizes "first-phase method a" and "second- phase method b," similar to the hysteresis curve of FIG. 3 and waveforms of FIGS. 4A to 4D (one difference being that the initial induction level of the present example is at -100% saturation instead of -50%).
- FIGS. 20A to 20D assume a varying d-c current for primary current Jl, though the sequence shown is equally applicable to a-c current.
- controllable voltage device 63 being a controllable voltage source.
- Resistor Rl is a current-sensing resistor with low resistance. After a demagnetizing cycle, the current through resistor Rl is proportional to the d-c primary current. As with previous examples, voltage V2 is considered to be the induced voltage.
- the magnetic core is saturated due to primary current Jl being a d-c current, with no recent demagnetizing cycle to prevent the magnetic core from being saturated.
- Current J2 is zero, since the magnetic core is unable to generate any voltage to drive a current of the same polarity as current Jl.
- the induction level at time T81 may be considered “known” from the perspective of this description, it is not known from the perspective of a current monitoring device.
- the present example will drive the induction level to a "known" induction level of +100% during the first phase, since this will be more informative than driving it to -100% (the induction level it is already at).
- the first-phase of a demagnetizing cycle begins at time T82 when control circuit 62 controls voltage device 63 to produce a large positive voltage. This voltage causes a transition from saturation at -100% to saturation at +100%. Saturation at +100% is reached at about time T83. During the transition, while the core is not in saturation between times T82 and T83, current J2 becomes somewhat proportional to current Jl, with an error related to hysteresis curve properties.
- a second phase begins at time T84, when control circuit 62 causes voltage device 63 to produce a large negative first-step voltage, followed by a shorter positive second-step voltage beginning at time T85.
- the second-step voltage is done, and control circuit 62 causes voltage device 63 to produce an output voltage near zero volts to allow secondary current to flow freely.
- control circuit 62 controls the first-step and second-step voltages so that the magnitude of the volt-time integral of voltage V2 during the second phase equals the volt-time constant of the current transformer (area A19 minus area A20 should equal the volt-time constant).
- the current transformer is demagnetized, and secondary current J2 is proportional to primary current Jl.
- the current transformer must generate a small induced voltage to keep current J2 flowing through the loop resistance (resistor Rl plus winding resistance plus other stray resistances)
- the induction level starts to drift back toward saturation after time T86 (see FIG. 20D). This transition toward saturation will cause an increasing error in current J2 until saturation is reached, at which point current J2 will cease to flow (as at time T81).
- the demagnetizing cycle must be repeated periodically (usually well before saturation is reached). For applications in which current flow is consistently one direction, it may be beneficial to move the initial operating point to an induction level other than zero to lengthen the time period between demagnetizing cycles.
- FIG. 21 shows one way that control circuit 62 and voltage device 63 of FIG. 1 may be embodied.
- the configuration shown is particularly suitable for measuring electric current, including d-c current.
- This co- ⁇ figuration may be operated in a manner similar to the manner described for FIGS. 20A to 20D.
- Many components are the same as FIG. 1, and these components function in the manner previously described.
- voltage device 63 of FIG. 21 is a controllable electric energy source, the same embodiment may be utilized for electric energy source 73 of FIG.2.
- control circuit 62 of FIG. 2 may also be embodied the same way - as shown in FIG. 21.
- resistor Rl is a current-sensing resistor with low resistance.
- Control circuit 62 has an analog-to-digital converter circuit 81 to sense current J2 (as voltage V3 across resistor Rl), a microcontroller 82 for data processing and control functions, and a digital-to-analog converter circuit 83.
- Digital-to-analog converter circuit 83 provides an analog voltage signal on conductor 85 that controls the voltage output of voltage device 63.
- Analog-to-digital converter circuit 81 and microcontroller 82 communicate via an interface shown as four conductors, and this interface may vary considerably depending on the particular design.
- microcontroller 82 communicates with digital-to-analog converter circuit 83 via an interface that may vary considerably depending on the particular design.
- Voltage device 63 has an operational amplifier 84, with resistors R81 and R82 configured to set the gain of operational amplifier 84.
- a separate power source (not shown) provides operating power for analog-to-digital converter circuit 81, microcontroller 82, digital-to-analog converter circuit 83, and operational amplifier 84.
- Ground connection 86 provides a common reference potential for the various circuits and power supply. If a particular application requires that winding 61 be directly grounded on one side, then resistor Rl may be relocated and connected in series with the opposite side of winding 61. This complicates the measurement of voltage V3 across resistor Rl somewhat, but prior-art differential voltage measurement methods are adequate.
- Operational amplifier 84 must be able to produce voltage in a circuit with relatively large current driven by a current source (a current transformer acts like a current source).
- a "power operational amplifier” will usually be required, such as model OPA548 made by Burr-Brown Corporation. This device has an adjustable current- limit feature and is rated for up to 3 amps of continuous current (other models are available with higher ratings). Tests found that these devices worked very well in a current transformer secondary circuit, as long as a "snubber circuit" was connected to the output to improve stability.
- the "snubber circuit" that was successfiilly utilized was a ten Ohm resistor in series with a 0.1 microfarad capacitor, connected between the operational amplifier output and the grounded conductor (as recommended by the device's data sheet).
- the OPA548 operational amplifier also has provision for adjustable current limit, which is beneficial for this application.
- FIG. 21 The type of control configuration shown in FIG. 21 (utilizing an analog-to-digital converter circuit to sense an input signal, a microcontroller to implement a control function based on the input signal, a digital-to-analog converter to produce an analog control signal, and an operational amplifier to produce a voltage proportional to the analog control signal) is well established in the prior art, so additional configuration details will not be described herein.
- Microcontroller 82 may be configured to implement the control sequence illustrated by FIGS. 20A to 20D.
- FIGS. 22A to 22E illustrate four different ways that the configurations of FIG. 1 or FIG. 21 may be operated during a current-sensing mode after a demagnetizing sequence is completed.
- FIGS. 22B, 22D and 22E may be thought of as a continuation in time of FIGS. 20A, 20C and 20D respectfully.
- Time T86 is common to both sets of figures.
- the vertical scales of FIGS. 22B, 22D, and 22E are magnified by factors of about 4, 4, and 8 respectively to facilitate display of wavefo ⁇ ns with relatively small magnitude.
- Voltage V2 is now considered to differ from the induced voltage V (see FIGS. 22B and 22C) by an amount that depends on current J2 and stray winding impedances.
- current J2 flows through winding 61, resistor Rl, voltage device 63, and various conductors that connect these different components.
- the "loop resistance" associated with current J2 is the sum of the resistances of all the components that current J2 flows through. With the assumption that operational amplifier 84 has near ideal properties, the resistance of voltage device 63 will be considered negligible. Then the loop resistance for current J2 of FIG. 21 is the sum of the resistances of winding 61, resistor Rl and various conductors that connect these different components. In the present example the total loop resistance will be assumed to be about twice the resistance of resistor Rl. For the present discussion, reactive impedances in the loop are considered negligible (see FIG.
- voltage device 63 may be controlled to produce a voltage with an average value equal to voltage V3 across resistor Rl. This is the case shown in FIGS. 22A to 22E between times T87 and T88. However, this is not enough voltage to counteract the voltage drop across the stray resistance in the loop (winding and conductor resistances), and the drift toward saturation continues at a slower rate (as shown by FIG. 22E).
- T88 and T89 voltage device 63 is controlled to produce a voltage equal to twice the average value of voltage V3 (since the total loop resistance is twice the resistance of resistor Rl), and this voltage causes induced voltage V to have an average value near zero, stopping the drift toward saturation (a larger voltage would cause induced voltage V to have an average positive value and cause the induction level to drift toward a positive direction).
- T89 and T90 voltage device 63 is controlled to produce a voltage equal to twice the instantaneous value of voltage V3, and this voltage also stops the drift toward saturation while causing induced voltage V to continuously be very near zero volts and greatly reducing induction level fluctuations.
- Either control method utilized between times T88 and T90 in FIGS. 22A to 22E may be used to maintain a prefe ⁇ ed induction level.
- the control method shown between times T89 and T90 has the added advantage of reducing the magnitude of the induced voltage to near zero (induction level fluctuations and magnetizing cu ⁇ ent are also reduced to near zero).
- the effective burden of the entire secondary circuit is reduced to near zero Ohms.
- a secondary circuit with a total effective burden near zero Ohms can significantly increase the accuracy of a cu ⁇ ent transformer as well as reduce the size of the magnetic core required to attain a particular level of accuracy. This is an improvement over the active load termination described in Reissued U.S. Patent Re. 28,851 (to)
- FIG. 23 A simple embodiment just for this purpose is illustrated in FIG. 23.
- the configuration of FIG. 23 is similar to FIG. 21 except that voltage device 63 is now controlled so that voltage VI is always proportional to voltage V3.
- the control means is now integral to voltage device 63 (adjustable resistor R83 and resistor R82 control the gain of operational amplifier 84).
- Conductor 87 conducts a voltage signal (voltage V3) which is proportional to cu ⁇ ent.
- Voltage V3 may also be communicated to a larger cu ⁇ ent-sensing system via terminals 88 and 89. Still referring to FIG. 23, if adjustable resistor R83 is set to zero ohms, the voltage gain of voltage device
- Adjustable resistor R83 may be adjusted to obtain any reasonable gain greater than one, in accordance with well-known design principles of operational amplifiers. By adjusting resistor R83 the gain is adjustable to match secondary circuit parameters for most secondary circuit configurations. Of course, resistor R83 may be a fixed resistor if the appropriate gain is predetermined. Care must be exercised, however, because if the gain is set to compensate for more than the total secondary circuit impedance, then the system becomes unstable and cu ⁇ ent will increase to the limit of the amplifier output circuit.
- resistors R82 and R83 may be replaced with complex impedances to better match secondary circuit parameters, thereby facilitating a total secondary circuit burden near zero ohms.
- Complex impedances may also be configured to compensate for cu ⁇ ent transformer imperfections, such as hysteresis effects and eddy cu ⁇ ents.
- a reactive impedance may be added to sensing resistor Rl (for example, like inductive reactance LI in FIG. 26) in order to provide a feedback voltage that is proportional to the voltage drop caused by cu ⁇ ent flowing through complex loop impedances.
- FIG. 24 shows a configuration that combines the benefits of a dedicated sensing winding 90 with the versatility of the control circuit shown in FIG. 21. Similar to FIG. 21, the configuration shown in FIG. 24 may be utilized for sensing both a-c and d-c cu ⁇ ents, since control circuit 62 is capable of demagnetizing the magnetic core.
- microcontroller 82 may be programmed to implement a burden-reducing method similar to FIG. 23 (controlling voltage device 63 based on current J2 and known loop impedances). Alternatively, microcontroller 82 may be programmed to control voltage device 63 to minimize the sensed voltage across winding 90. This may be done simply with a proportional type of control, controlling voltage device 63 to produce an output voltage proportionally many times larger than the sensed voltage, with polarity so as to reduce induced voltage.
- the control configuration shown in FIG.24 may be applied to many kinds of magnetic bodies in addition to cu ⁇ ent transformers. With the induced voltage across dedicated winding 90 utilized as a feedback control signal, the induction level of a magnetic body may be made to vary with time in almost any manner desired. Microcontroller 82 may be programmed to calculate changes in induction level as the integral of induced voltage over time, and utilize a P.I.D. (Proportional plus Integral plus Derivative) type of control to obtain the response desired. Other types of control, such as fuzzy logic, may also be utilized. These feedback control methods are well known in the prior art, so they will not be discussed in detail herein. Similar configurations are possible utilizing analog control means in place of the digital control means shown for control circuit 62.
- P.I.D. Proportional plus Integral plus Derivative
- FIG. 25 shows a variation of FIG. 23 with added provision for a demagnetizing sequence. This combines the accuracy of sensing cu ⁇ ent with near zero induction level fluctuation with a demagnetizing means. Similar to FIG. 21, the configuration of FIG. 25 may be used to measure d-c cu ⁇ ent as well as a-c current. FIG. 25 has the advantage of not requiring continual microcontroller supervision during the cu ⁇ ent-sensing phase.
- P- channel field-effect transistor Ql and N-channel field-effect transistor Q2 along with demagnetizing controls 91 have been added to enable voltage device 63 to produce the varying voltage required for a demagnetizing sequence. Positive voltage VP and negative voltage VN are supplied by a separate power supply (not shown).
- Demagnetizing controls 91 may be configured utilizing prior art.
- a microcontroller may be utilized for proper timing of the demagnetizing sequence, or an analog timing circuit may be utilized.
- the gate of field- effect transistor Ql should be driven low to cause voltage device 63 to produce a negative voltage, and the gate of field-effect transistor Q2 should be driven high to cause voltage source 63 produce a positive voltage. With both field-effect transistors Ql and Q2 off, the circuit functions similar to FIG. 23.
- Field-effect transistors Ql and Q2 should not be turned on at the same time, as this results in a relatively low impedance between power supply connections VP and VN.
- Resistors R82 and R83 should have resistance values set small enough that any leakage cu ⁇ ent through field-effect transistors Ql and Q2 during the cu ⁇ ent sensing mode will not adversely affect operation of voltage device 63.
- Surge protection for the configurations of FIGS. 21, 23, 24, and 25 may be provided by connecting a varistor, back-to-back zener diodes, or other voltage-limiting device between the output of operational amplifier 84 and the grounded conductor (thereby limiting voltage VI to a safe level and still provide for sensing of surge cu ⁇ ents).
- Dual silicon controlled rectifiers or a triac both with suitable trigger circuits may be particularly suitable.
- U.S. Patent No. 4,466,039 to Moran and Reis (1984) discloses a suitable triac circuit that may be used.
- FIG. 26 is a variation of FIG. 1 showing one way that induced voltage may be calculated taking into account the effects of stray impedances, and one way that calculated induced voltage may be utilized as a feedback signal to facilitate control of induced voltage.
- Inductive reactance LI has been added to resistor Rl, the combination being impedance Z5.
- the stray impedances of winding 61 are represented by winding stray impedance Z6, which is shown as a combination of resistance and inductive reactance.
- Other loop impedances are shown as impedance Z7, which is also shown as a combination of resistance and inductive reactance.
- Capacitive reactance could also be accounted for but it is assumed to be negligible for the present discussion.
- the total effective loop impedance is the sum of impedances Z5, Z6, and Z7.
- the voltage across resistor Rl is still directly proportional to current, and may be used by an external system to sense current via terminals 88 and 89.
- Voltage V2 is now strictly the measurable voltage across winding 61, with the actual induced voltage V differing from voltage V2 by a voltage drop V6 cause by current flowing through stray impedance Z6.
- Induced voltage V may be calculated as voltage V2 minus voltage drop V6.
- induced voltage V may be calculated as voltage VI minus voltages V5, V6, and V7 (in accordance with KirchofFs voltage law).
- inductor LI and resistor Rl are sized so that the ratio of the reactive component to resistive component of impedance Z5 is approximately the same as the ratio of the reactive component to resistive component of the series combination of winding stray impedances Z6 and impedance Z7.
- operational amplifier 95 and resistors R35 and R36 are configured to produce a voltage output equal to voltage V5 multiplied by a constant. Resistor R35 is adjustable to facilitate making the output approximately equal to the sum of voltage drops V5, V6 and V7.
- Operational amplifier 96 and resistors R37, R38, R39, and R40 are configured such that the output of operational amplifier 96 is approximately equal to voltage VI minus voltage V5 multiplied by a constant, which is approximately induced voltage V (resistors R37, R38, R39, and R40 should all have the same value). This calculated induced voltage is conducted by conductor 97 and is a feedback input to a controller 62B.
- Controller 62B may also (optionally) receive a variable control signal (from an external system) that is proportional to the induction level that is desired (the "reference induction level”).
- a control signal is shown in FIG. 26 as an input signal on conductor 98. Since the induced voltage is proportional to the rate of change (or the “derivative") of induction level (per equation no. 4), it may be beneficial to calculate the rate of change (the derivative) of the control signal. This calculation is performed by operational amplifier 99, capacitor C4, and resistor R38. Conductor 100 conducts the negative of the derivative as an input into controller 62B.
- Controller 62B is configured with suitable prior-art controls (such as a proportional plus integral plus derivative type of control system, or fuzzy logic controls, or any other suitable control means) to force the induced voltage to be proportional to the derivative of the control signal (with an appropriate constant of proportionality).
- suitable prior-art controls such as a proportional plus integral plus derivative type of control system, or fuzzy logic controls, or any other suitable control means
- the induced voltage could be integrated, and the controller could function to keep the integral of induced voltage proportional to the control signal (per equation no. 5).
- controller 62B may be configured to occasionally reset the induction level in magnetic core 60 to the preferred level as indicated by the control signal on conductor 98.
- a sequence similar to that shown in FIGS. 5 through 6D may be utilized to reestablish a prefe ⁇ ed induction level.
- analog calculations performed by the operational amplifiers in FIG. 26 could also be performed digitally utilizing an analog-to-digital converter and a microcontroller, similar to the configuration of FIG. 21. If the configuration of FIG. 21 is used, induced voltage may be calculated without measuring voltage VI, since microcontroller 82 controls this voltage and therefore knows its magnitude without measuring it.
- the invention provides for improved accuracy in cu ⁇ ent measurements for electric power monitoring and other cu ⁇ ent monitoring applications. Since almost all industries utilize electricity and are therefore concerned with accurate measurement of electric cu ⁇ ent, the invention has broad application. Also, many industries are concerned with controlling the magnetization of magnetic bodies other than cu ⁇ ent transformers, so the invention is expected to have diverse application.
- the present invention applies a varying voltage to a conductive winding to contiol the induction level of a magnetic body. Improvements over prior art include, but are not limited to, the following: (a) how induced voltage may be controlled to establish a preferred induction level,
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Measuring Magnetic Variables (AREA)
Abstract
A varying voltage is applied to a conductive winding (61) that magnetically interacts with a magnetic body (60). The varying voltage causes the voltage induced in the winding to have such waveform and magnitude that the integral over time of the induced voltage correlates to desired changes of the induction level (magnetic flux density). The induction level of a magnetic body may be controlled to transition from an initial induction level to a preferred induction level, to maintain a preferred induction level, to reduce fluctuations around a preferred induction level, or to vary with time in a preferred manner. The invention is especially applicable to ordinary current transformers, which may be demagnetized automatically while remaining in service. Once demagnetized, ordinary current transformers are able to accurately sense nonsymmetrical currents, including d-c currents and a-c currents that have d-c components. A demagnetizing mode, during which the current transformer is demagnetized, and a current-sensing mode, during which current transformer secondary current is sensed, are usually utilized sequentially. For a-c power system applications, a current transformer demagnetizing circuit may include an adjustable impedance with a suitable control circuit. A controllable active voltage source may alternately be used as part of a current transformer demagnetizing circuit, in which case an ordinary current transformer may be used to sense d-c current as well as a-c current. A controllable active voltage source may also be used to improve current transformer accuracy by reducing the amount that the induction level of the core fluctuates, regardless of whether the demagnetizing aspect of the invention is utilized.
Description
METHOD AND APPARATUS FOR CONTROLLING THE MAGNETIZATION OF CURRENT TRANSFORMERS AND OTHER MAGNETIC BODIES
TECHNICAL FIELD
This invention relates to the art of controlling the magnetic flux density within magnetic bodies. A varying voltage applied to a conductive winding causes the voltage induced in the winding to have such waveform and magnitude that the integral over time of the induced voltage correlates to desired changes of the magnetic induction level.
BACKGROUND ART
Prior art methods, that utilize a conductive winding to control magnetization, generally focus on controlling winding current to control the induction level of a magnetic body. The winding current directly correlates with magnetic field intensity. Changes in magnetic field intensity are coordinated with hysteresis-curve properties of the magnetic material to control the induction level. (As used throughout this disclosure, the following terms are synonymous: "magnetization," "magnetic induction," "flux density" and "induction level").
Many applications are particularly concerned with how a magnetic body may be demagnetized. Most prior- art demagnetizing methods (that effectively reduce the induction level of a magnetic body to near zero) utilize a declining alternating magnetic field. The alternating magnetic field is usually produced by a conductive winding conducting a declining alternating current. Many different embodiments of this demagnetizing method are well established in the prior art.
Other applications are concerned with how a magnetic body may be magnetized to a preferred induction level. To establish a preferred induction level, prior art methods generally apply a current with controlled magnitude to a winding (usually with the magnetic body initially in a demagnetized state). Many different embodiments of this magnetizing method are also well established in the prior art.
Of particular relevance to the present invention is the relationship between voltage induced in a winding and changing magnetic flux. The relationship between induced voltage and changing magnetic flux was originally stated in a general way by Michael Faraday, and is widely known as Faraday's Law. Faraday's Law basically states that the voltage induced around a closed path is proportional to the rate of change of magnetic flux within the closed path. It is well established in prior art how Faraday's Law can be used along with a conductive winding to measure changes of induction level in a magnetic body. However, the prior art does not appear to address how the induction level of a magnetic body can be controlled by controlling the voltage induced in a winding.
The present invention has been developed primarily out of a need to demagnetize current transformers while the current transformers remain in service. The prior art does not address this problem very well.
Most current monitoring systems for a-c (alternating-current) electric power systems utilize current transformers to provide input currents that are isolated from the electric power system conductors. A primary winding of a current transformer is connected in series with a current-carrying conductor while a secondary winding is magnetically coupled to the primary winding by a suitable magnetic core. A current is induced in the secondary winding that is proportional to the primary current. The secondary current is isolated from the primary current and is smaller than the primary current by the turns ratio of the primary and secondary windings. The primary winding frequently consists of only one turn, which is often just the current-carrying conductor installed through an opening in the middle of the current transformer magnetic core. The secondary winding usually consists of multiple turns wrapped around the magnetic core.
In order for the secondary current generated by a current transformer to be an accurate representation of the primary current, the impedance of the circuit connected to the secondary winding must be kept low so that current can flow freely. The impedance of the secondary circuit is often called the "burden." In order for a current transformer to drive a secondary current through a non-zero burden, a voltage must be induced in the secondary winding. The induced voltage is proportional to secondary current and proportional to the burden, in accordance with Ohm's law. The induced voltage is induced in the secondary winding by a fluctuating induction level in the magnetic core. The fluctuating induction level is associated with a magnetizing current in accordance with well-known electromagnetic principles. The magnetizing current accounts for most of the error in the secondary current. Generally speaking, the accuracy of a current transformer is inversely related to the burden of the secondary circuit. A higher burden causes the secondary current to be a less accurate representation of the primary current.
The accuracy of the secondary current may also be adversely affected by either of the following:
(a) A primary current that is not symmetrical. "Symmetrical" is intended to mean that the waveform has positive and negative half-cycles with the same waveform and magnitude. An alternating current that has a d-c (direct-current) component is a common example of a primary current that is not symmetrical. Also, transient a-c fault currents are often not symmetrical. D-c currents are, by definition, not symmetrical.
(b) A burden that is not a linear impedance. Nonlinear burdens are common in applications that derive power from the secondary current. In either of these cases, the current transformer core may become magnetized. This magnetization may cause significant error in the secondary current. This error may include distortion of the secondary current, including the loss of any d-c component that is present in the primary current.
BRIEF SUMMARY OF THE INVENTION
In accordance with the present invention, a varying voltage is applied to a conductive winding that magnetically interacts with a magnetic body. The varying voltage controls the voltage induced in the winding in such a way that the integral over time of the induced voltage correlates to desired changes of the induction level (magnetization) of the magnetic body. The invention may be used to control the induction level of a magnetic body in several ways:
(a) The induction level of a magnetic body may be caused to transition from a known induction level to a preferred induction level. (A preferred induction level of zero may be chosen to demagnetize a magnetic body).
(b) When the induction level is not known, a preferred induction level may be established by changing the induction level of the magnetic body from an unknown induction level to a known induction level and then to the preferred induction level.
(c) A preferred induction level may be maintained by causing the induced voltage across the winding to have an average value near zero (or by. causing the integral of induced voltage to not exceed a predetermined value).
(d) A preferred induction level may be more strictly maintained by causing the induced voltage across the winding to continuously be near zero volts, thereby reducing the amount that the induction level fluctuates.
(e) The induction level may be made to vary with time in a preferred manner, including matching a control signal that is proportional to a reference induction level.
The varying voltage may be generated directly by an active voltage source (or, more generally, a controllable electric energy source) or the varying voltage may be generated indirectly, such as by current transformer secondary current flowing through an adjustable impedance. The key elements are a magnetic body, a conductive winding that magnetically interacts with the magnetic body, and a means of causing the induced voltage to have the appropriate waveform and magnitude.
The invention is most readily applied to magnetic bodies that are configured to have a relatively uniform magnetic path, such as the magnetic cores of current transformers. The accuracy of a current transformer may be improved by the invention in three ways:
(a) By demagnetizing a current transformer, inaccuracies associated with core magnetization are removed. A demagnetized current transformer can accurately measure d-c current and a-c current that has a d-c component.
(b) By keeping the integral over time of induced voltage near zero, a current transformer is better able to measure unsymmetrical currents without quickly transitioning to saturation.
(c) By reducing the amount that the induction level in the core fluctuates, inaccuracies associated with magnetizing current may be greatly reduced and the accuracy of a current transformer may be greatly improved.
The invention is based on two principles of electromagnetism: (1) the relationship of magnetomotive force to flux density as defined by hysteresis curves of magnetic bodies, and (2) Faraday's Law applied to magnetic bodies.
The second principle (Faraday's law applied to magnetic bodies) clarifies how a voltage induced in a winding is directly proportional to a changing induction level. While prior art shows how induced voltage can be used to measure changes in induction level, the present invention shows how induction level can be controlled by controlling induced voltage. The invention primarily utilizes this second principle relationship to control the induction level of a magnetic body. The application of this second principle is an integral part of all embodiments of the invention.
BRIEF DESCRIPTION OF THE DRAWINGS
FIGS. 1 through 12 illustrate how the invention may be applied to magnetic bodies in general. FIGS. 1 and 2 illustrate typical components and connections, while FIGS. 3 through 12 illustrate how a varying voltage may be utilized to control the induction level of a magnetic body. FIGS. 13 through 19 all illustrate how an adjustable impedance may be utilized in a current transformer secondary circuit to demagnetize a current transformer.
FIGS. 20 through 25 all illustrate how an active voltage source may be utilized in a current transformer secondary circuit to control induction level.
FIG. 1 illustrates the general concept by showing a magnetic body 60, a winding 61 wrapped around magnetic body 60, and a controllable voltage device 63 with a suitable control circuit 62. Magnetic body 60 and winding 61 are shown in a typical current transformer configuration with the magnetic core around an electric power system conductor 64. Controllable voltage device 63 may be an active electrical energy source or an adjustable impedance.
FIG. 2 illustrates an alternate configuration in which magnetic material 72 is used to provide a closed magnetic path for a magnetic body 70.
FIG. 3 shows a hysteresis curve sequence that maybe used to quickly change the induction level of a magnetic body from an unknown induction level to an induction level near zero.
FIGS. 4A to 4D show one method of controlling voltages and currents to demagnetize a magnetic body utilizing the hysteresis curve sequence shown in FIG. 3. FIG. 5 shows a hysteresis curve sequence similar to FIG. 3, except now showing how the induction level of a magnetic body may be quickly changed from an unknown level to a preferred induction level (other than zero).
FIGS. 6A to 6D show one method of controlling voltages and currents to establish a preferred induction level in a magnetic body utilizing the hysteresis curve sequence shown in FIG. 5.
FIG. 7 shows a hysteresis curve sequence that may be used to automatically determine demagnetizing parameters and quickly demagnetize a magnetic body.
FIGS. 8A to 8D show one method of controlling voltages and currents to automatically determine demagnetizing parameters and demagnetize a magnetic body utilizing the hysteresis curve sequence shown in FIG. 7.
FIG. 9 shows a hysteresis curve sequence that illustrates how a magnetic body may be demagnetized when current is not adequate to drive the magnetic body into saturation.
FIG. 10 shows another hysteresis curve sequence that illustrates how a magnetic body may be demagnetized without driving the magnetic body into saturation.
FIG. 11 shows a very simple embodiment of a controllable electric energy source that can be used to carry out the hysteresis curve sequence shown in FIG. 10. FIGS. 12A to 12C show one way that an alternating voltage and current may be utilized to implement the hysteresis curve sequence of FIG. 10 using the controllable electric energy source shown in FIG. 11.
FIG. 13 shows a functional schematic to illustrate how an adjustable impedance 2 may be used to demagnetize a current transformer while in service. An optional power supply circuit 4 is shown that can derive
power from the current transformer secondary current to provide power for self-powered applications.
FIGS. 14A to 14E show typical operating waveforms and control signals associated with preferred operation of the functional schematic shown in FIG. 13.
FIGS. 15A to 15E show waveforms and control signals for an alternate method of operation for the functional schematic shown in FIG. 13. This method of operation does not require any feedback signal to the control circuit.
FIG. 16 shows one way that the functional schematic of FIG. 13 may be embodied. The power supply circuit for self-powered operation is optional.
FIGS. 17A to 17D show typical operating waveforms and control signals associated with the embodiment shown in FIG. 16.
FIG. 18 shows an alternate embodiment of a current transformer demagnetizing circuit utilizing only one MOSFET transistor for the adjustable impedance.
FIGS. 19A to 19E show typical operating waveforms and control signals associated with the embodiment shown in FIG. 18. FIGS. 20A to 20D illustrate how the configuration of FIG. 1 may be operated to demagnetize a current transformer and measure d-c current.
FIG. 21 shows one way that control circuit 62 and voltage device 63 of FIG. 1 may be embodied to measure electric current, including d-c current.
FIGS. 22A to 22E illustrate four different ways that the configurations shown in FIGS. 1 or 21 may be operated during a current-sensing mode after a demagnetizing sequence is completed. Conceptually, FIGS. 22B, 22D and 22E may be thought of as a continuation in time of FIGS. 20A, 20C and 20D respectfully, with enlarged vertical scales.
FIG. 23 illustrates a variation of the configuration shown in FIG. 21 that may be used to improve the accuracy of a current transformer by reducing the amount that the induction level fluctuates, without provision for demagnetizing control.
FIG. 24 shows a configuration that is similar to FIG. 21, but with a dedicated sensing winding providing a feedback signal that is proportional to induced voltage.
FIG. 25 shows a configuration similar to FIG. 23, but modified to include demagnetizing control.
FIG. 26 shows an embodiment configured to accurately calculate induced voltage V using analog calculation means. The calculation method shown takes into account the effect of various circuit impedances and stray winding impedances. The calculated induced voltage is configured as a feedback signal, thereby enabling the induction level to be controlled to closely follow a variable control signal.
Items that are common to more than one figure are identified by the same reference characters. Sequence point identifiers A, B, C, D, E, F, and G (used in FIGS. 3, 5, 7, 9 and, 10) are intended to indicate sequential progress, rather than a functional state. Point A indicates the first point of a sequence, point B indicates the second point of a sequence, etc.
DETAILED DESCRIPTION OF THE INVENTION
Theory
The invention is based primarily on two principles of electromagnetism, herein referred to as the "first principle" and "second principle." These two principles may be stated briefly as follows: (1) the relationship of magnetomotive force to flux density as defined by hysteresis curves of magnetic bodies, and (2) Faraday's Law applied to magnetic bodies.
The "first principle," the relationship of magnetomotive force (amp-turns) to flux density, is used by the invention to establish a known induction level in a magnetic body (when the initial magnitude of the induction level is not known). Hysteresis curves are commonly used to show the relationship of magnetic flux density (the y axis, also referred to as "induction level") to magnetic field intensity (the x axis) for a particular magnetic material. For a reasonably uniform magnetic core that magnetically interacts with a winding made with N turns of conductive material, the magnetic field intensity within the magnetic core is fairly uniform and is proportional to the magnetomotive force (current magnitude times the number of winding turns). Since the number of winding turns is usually constant, the x-axis of the hysteresis curve may be scaled to be proportional to the magnitude of current in the winding. With this understanding, a hysteresis curve may be utilized directly to find the induction level associated with a particular current magnitude. Alternatively, the same hysteresis curve could be used to find the magnitude of current (in a winding) that is necessary to drive the magnetic core to a specific induction level, and the minimum amount of current required to drive the magnetic body into saturation. The "second principle," Faraday's Law applied to magnetic bodies, basically defines how a magnetic flux changing with time is associated with an electromotive force. In simple form, Faraday's Law is often stated as: emf = -dΦ/dt (equation no. 1) where "emf' is the electromotive force (voltage caused by a changing magnetic flux) associated with a defined closed path, and -dΦ/dt is the time rate of change of magnetic flux (Φ) within the closed path. The minus sign is usually included to show that the emf tends to generate a current that reduces the change in flux in accordance with Lenz's law. However, the minus sign will be dropped for the remainder of this description so that a positive change in flux will be associated with a positive induced voltage. A constant of proportionality may be required, depending on the system of units used.
As used herein, the term "induced voltage" is intended to refer to voltages generated by a changing flux in accordance with Faraday' s Law.
When applied to a winding with N turns positioned around a magnetic body, Faraday's Law may be modified to yield (with the minus sign dropped):
V = N (dΦ/dt) (equation no. 2) where V is the induced voltage (time-varying or constant) across the winding. Now, with the assumption (and limitation) that the induction level (magnetic flux density) within a magnetic body is reasonably uniform, then induction level "B" and magnetic body cross-sectional area "A" are related to flux Φ by the simple equation:
Φ = AB (equation no. 3)
Combining equation no. 2 with equation no. 3 yields:
V = N A ( dB / dt) (equation no. 4)
Equation no. 4 shows that induced voltage V is proportional to the rate of change of induction level B. Taking the integral of both sides of equation no. 4 yields:
/ V dt = N A ΔB (equation no. 5) which shows how a change in induction level B is related to the integral over time of induced voltage V. The left side of equation no. 5 will be referred to herein as a "volt-time integral." Equation no. 5 clearly shows how the integral over time of induced voltage correlates to changes of induction level. The constant of proportionality (NA) will vary depending on the system of units utilized.
Equation no. 4 and equation no. 5 are two different ways of viewing the same relationship between induced voltage and a changing induction level. This disclosure will treat these two viewpoints as equivalent, and will reference whichever one is most readily applied in a given situation. Both equations show how induced voltage may be controlled to bring about desired changes of induction level. This concept of controlling induced voltage in order to control the induction level of a magnetic body is common to all embodiments of the invention. The "volt-time integral" of the second principle leads to the concept that a particular "volt-time value" is associated with the change of induction level in a magnetic body from one induction level to another induction level. For example, the volt-time integral of voltage induced in a winding by a change of induction level from saturation to zero will always have the same value regardless of how many steps are involved or how long it takes. Likewise, a change between any other two levels of induction will also be associated with a volt-time value that is not dependent on the sequence used to cause the change. For the sake of this description, let "Q" designate the "volt-time value" associated with the volt-time integral calculated for a change in induction level from a first induction level B, to a second induction level B2. Equation no. 5 may then be written as:
Q = J* V dt = N A (B2 - Bj) (equation no. 6)
The units of this "volt-time value" will usually be volt-seconds, though other units are possible, of course. Of particular interest is the volt-time value associated with a change in induction level from a "known" induction level to an induction level near zero, as this is key to using the second principle to demagnetize a magnetic body. If the induction level is known, a volt-time value associated with a change to zero may be calculated. Alternatively, if a volt-time value associated with a change to zero is known, the actual induction level may be calculated. More specifically, let "Q0" designate a volt-time value associated with a change from an induction level B, to an induction level of zero. Then the relationship between volt-time value Q0 and induction level B L may be expressed mathematically as:
Q0 = - N A B, (equation no. 7) or
B, = - Q0 / (N A) (equation no. 8)
If a volt-time value Q0 is determined for a particular induction level, and if the cross-sectional area and number of winding turns are known, equation no. 8 may be used to determine the actual induction level in customary units. However, in many demagnetizing applications, it is often sufficient to merely quantify Q0 in a general sense in order to facilitate a transition to an induction level of zero. For example, the induction level may be "known" only in a very general sense by knowing something about Q0, such as knowing the time period
associated with a repeatable voltage that causes a transition between two opposite induction levels (the specific volt-time value of the voltage wave not being necessary for operation). This situation is illustrated by FIGS. 8B and 12A and the description of those figures.
The magnitude of the volt-time value associated with a change of induction level from saturation to zero is normally constant for a particular magnetic body and winding combination. This constant value will be a positive number (the polarity of saturation is not included) designated "Qc" and will be referred to herein as a magnetic body's "volt-time constant." This value may be calculated based on magnetic body and winding parameters, or may be determined by trial and adjustment. Alternatively, the "volt-time constant" may be determined automatically as described below under "Preferred Sequences." In addition to utilizing the second principle to change between two induction levels as described above, the second principle also may be utilized to maintain a preferred induction level in a magnetic body. Considering equation no. 5, if the integral of induced voltage over a particular period of time is zero, the final induction level must be the same as the beginning induction level. Thus, even though an induction level may be fluctuating and causing a nonzero induced voltage, as long as the integral of induced voltage over time does not exceed a predetermined value, the preferred induction level will be maintained within acceptable limits.
A less stringent requirement (relating to maintaining a preferred induction level) would be to require the average value (rather than the integral) of induced voltage over time to be near zero. Since average value may be determined by calculating an integral over a time period and dividing by the time period, as the time period becomes large the average value may be near zero while the integral may deviate significantly from zero. Thus, by merely requiring that the average value of induced voltage be near zero, a preferred induction level is prevented from changing quickly, but given sufficient time the induction level may drift away from the preferred induction level.
Still referring to equation no. 5, it is clear that there can be no change in induction level without an induced voltage. Therefore, a preferred induction level may be more strictly maintained if the induced voltage is caused to be continuously very near zero volts, in which case the induction level can change only very slowly, and only if the integral of induced voltage increases over time. This concept can be utilized to strictly maintain a preferred induction level in a magnetic body. Since some applications of magnetic bodies are adversely affected by a fluctuating induction level, this concept may be used to advantage to reduce the amount that the induction level fluctuates. As already mentioned, current transformers are one such application. The accuracy of a current transformer may be improved by reducing induction level fluctuation and associated magnetizing current required for operation.
The induction level of a magnetic body may also be controlled to vary in a preferred manner. Given a control signal that is proportional to a reference induction level that is a function of time, the induced voltage may be controlled so that the integral over time of induced voltage is proportional to the control signal (per equation no. 5). Equivalently, the induced voltage may be controlled to be proportional to the derivative of the control signal (per equation no. 4). By selecting an appropriate constant of proportionality, the induction level may be controlled so as to match the reference induction level.
Many applications will benefit from some form of feedback control method. A feedback signal that is directly proportional to induced voltage may be produced by a dedicated sensing winding (as shown in FIG. 24),
or a feedback signal may be derived from winding current (as in FIGS.21, 23, 5 and 26). For these applications the control method utilized may include a form of Proportional plus Integral plus Derivative (P.I.D.) control, or other control scheme such as fuzzy logic. These feedback control schemes, and others, are well established in the prior art, so they will not be discussed in detail herein. The particular control parameters will depend on the specific application. FIGS. 20 through 26 and the accompanying description for those figures provide some examples of simple control methods that may be utilized for current transformers and other applications with similar configurations.
Clarifications
Throughout this disclosure, the term "magnetic body" is used in a general sense to refer to a mass of magnetic material. The term "magnetic material" refers to material that has a relative permeability significantly greater than a value of one. The term "magnetic core" or simply "core" is intended to refer to a magnetic body that is in a particular spatial relationship to one or more current-carrying conductors. The term "induction level" is intended to be synonymous with magnetic flux density.
While windings utilized throughout industry are most commonly made of copper or aluminum conductors, it should be understood that the present invention is not limited to standard types of windings. Windings of widely varying form made of almost any kind of conductive material may be utilized, including superconducting materials and semiconducting materials.
Unless winding current is kept very small, the voltage measurable across a winding deviates from the induced voltage by an amount that depends on characteristics of winding current and characteristics of stray winding impedances. Strictly speaking, the voltage used in equations 2, 4, 5, and 6 is the induced voltage (rather than the measurable voltage across a winding). For simplicity of illustration, much of the description herein assumes that voltage drops associated with stray winding impedances are small enough to be ignored.
Stray impedances associated with a winding may be defined to include not only the effects of wire resistance and stray inductance, but may also be defined to include the effects of eddy currents, core losses and other imperfections associated with a particular core/winding combination. Stated another way, the stray impedances of a winding may be defined in whatever way results in the most accurate determination of induced voltage
(induced voltage being understood to be approximately calculable as the voltage measured across a winding minus a voltage drop associated with winding current flowing through stray impedances). This issue is most applicable to embodiments of the invention that utilize a feedback signal that is derived from winding current. Different applications of the invention may find it advantageous to define stray impedances differently in order to optimize each application. It is not the intention of this disclosure to restrict the term "stray impedances" to a firm definition.
The term "current-sensing" is used herein in a broad sense to refer not only to means of deriving information about a current, but also to refer to more direct uses, such as means of actuating a control mechanism based on current characteristics (such as a calibrated solenoid-type actuator that may be part of a protective relay).
As used herein, the phrase "known induction level" is intended to be synonymous with "determinate induction level." Both phrases are used in a broad sense of having some kind of data that quantifies the state of
magnetic flux density (induction level) in any way that can be used to implement the "second principle" (Faraday's Law applied to magnetic bodies) to move the induction level to a preferred induction level. For example, the induction level is considered to be "known" (or "determinate") if the approximate volt-time value (equation no. 6) that is associated with a change of induction level to zero is known. In the case of repeatable voltages being utilized, the induction level is also considered to be "known" if the time period required to change the induction level to near zero is known.
A parameter that is said to be a "function of time" is a parameter that may vary with time, even though the parameter may be constant for a period of time. As used herein, the term "varying voltage" is intended to refer in general to a voltage that is a function of time. While relatively constant voltages are utilized at times in the preferred embodiments, it should be understood that non-constant voltages could also be used, and the term
"varying voltage" is intended to include both possibilities. Also, the term "waveform" is intended to mean the geometrical shape of a quantity when displayed as a --unction of time. The term "alternating" (as applied to current or voltage) is intended to indicate that the polarity changes between positive and negative values, without necessarily implying a consistent repeating waveform. A "half-cycle" of an "alternating" waveform refers to a part of the waveform continuously having the same polarity.
Although some of the preferred embodiments of the invention utilize a controllable active "voltage source" to produce "varying voltages," in more general terms a controllable "electric energy source" may be utilized, meaning an active source that is able to generate suitable voltages and currents, while not necessarily being configured as a voltage source (the term "voltage source" usually indicates an electric energy source having output voltage that is relatively unaffected by current magnitude). However, from a practical point of view, since the invention primarily involves the control of induced voltage rather than current (in accordance with the second principle), a voltage source is usually the preferred form of electric energy source. Even so, the use of "voltage sources" in the preferred embodiments is not intended to restrict the invention from being embodied with other types of electric energy sources. When utilizing an active voltage source, the output should be current-limited to an appropriate value to ensure safe current levels during brief time periods that the magnetic bodies may be saturated.
Preferred Sequences
The preferred embodiment utilizes two sequential phases to establish a preferred induction level in a magnetic body when the initial magnitude of the induction level is not known. For purposes of this description these two phases will be called the "first phase" and the "second phase." The first phase primarily utilizes the "first principle" (the hysteresis-curve relationship of magnetomotive force to flux density) to establish a known induction level. The "second phase" utilizes the "second principle" (Faraday's Law applied to magnetic bodies) to change the induction level to a preferred induction level. Each of these two phases may involve several steps or just a single step, depending on the preferred operation for a particular application. For applications in which the initial magnitude of the induction level is known, only the second phase is required to establish a preferred induction level (though the first phase may still be utilized). Once a preferred induction level is established, a "third phase" may be applicable, during which the varying voltage may be turned off, or used to maintain the preferred induction level, or otherwise utilized depending on the particular application.
Establishing a known induction level during the first phase (utilizing the "first principle") may be done in several ways:
(a) Provide a current of sufficient magnitude to drive the magnetic body to saturation, in which case the induction level is known to be the saturation induction level of the magnetic body. (b) Provide a current that is not sufficient to saturate the magnetic material, and utilize known hysteresis-curve characteristics to determine the induction level of the magnetic body. Accuracy may be improved by cycling the current between opposite polarities once to remove the effects of any initial magnetization, (c) Provide a current that alternates between positive and negative values and determine a "volt-time value" from the resulting voltage waveform. This "volt-time value" may be combined with known geometries of the magnetic body and winding to determine the actual peak induction level (per equation no. 8 above), or it may be utilized directly by the second phase to reduce the induction level to zero. In the case that the current is sufficiently large to drive the magnetic body to saturation, the "volt-time value" determined in this way has the same magnitude as the "volt-time constant" previously defined. The first phase ends (and the second phase begins) with the induction level at a peak induction magnitude. These three methods will be referred to herein as "first-phase method a," "first-phase metliod b" and "first-phase method c."
"First-phase method c" (applying an alternating current and deteπnining a volt-time value) has the advantage of not requiring advance knowledge about the properties of the magnetic body.
After a known induction level is reached by the first phase, the second phase starts, changing the known induction level to the preferred induction level (utilizing the second principle). This also may be done in several ways. However, all of these ways are constrained by the second principle to provide a voltage for a period of time such that the volt-time integral of induced voltage corresponds to the volt-time value necessary to change the induction level from the known induction level to the preferred induction level. Three ways to do this are:
(a) Provide voltage all of one polarity for a period of time such that the volt-time integral approximately equals the volt-time value.
(b) Provide a first voltage and a second voltage of opposite polarities; the first voltage having such magnitude, duration and polarity so as to cause the induction level of the magnetic body to transition to and pass the preferred induction level; the second voltage having such magnitude, duration and polarity so as to cause the induction level of the magnetic body to transition to the preferred induction level. (The first voltage and second voltage may be referred to herein as a "first-step voltage" and "second-step voltage" respectively).
(c) Provide a voltage made up of several sequential voltage pulses of varying form, magnitude, duration, and polarity; the integral of this induced voltage over the second-phase time period being approximately equal to the volt-time value.
These three methods will be referred to herein as "second-phase method a," "second-phase method b" and "second-phase method c."
"Second-phase method b" is an improvement over "second-phase method a" because the second voltage (with opposite polarity as the first voltage) can improve accuracy by eliminating a small d-c offset current that may be present after the first voltage. After a first voltage (of metliod a or b), there is a small current flowing that has a magnitude roughly corresponding to the coercive force of the magnetic body. This small current may
adversely affect the magnetic state of the magnetic core at the end of the demagnetizing sequence. The second voltage used in method b is intended to eliminate this "coercive force error."
"Second-phase method c" is intended to cover all possibilities not covered by methods "a" and "b." This includes alternating voltage pulses with declining volt-time values, which may also be used to eliminate any "coercive force error" (similar to method "b," but with more than two pulses).
Current Transformer Considerations
Current monitoring devices that incorporate the demagnetizing aspect of the invention and use standard current transformers will usually utilize two modes of operation sequentially:
(1) A first mode (also called a "demagnetizing mode"), during which the current transformer is demagnetized (utilizing the first and second phases discussed previously).
(2) A second mode (also called a "current-sensing mode"), during which current transformer secondary current is proportional to primary current (this may be considered to be a "third phase").
Once demagnetized, the current transformer may become magnetized again due to the same problems that magnetized it in the first place. For this reason, preferred operation may sequence between the demagnetizing mode and the current-sensing mode. The demagnetizing aspect of the invention is best suited for applications that sense current periodically, with some time between sensing periods available for use with the demagnetizing mode. (In applications in which it is desirable to sense current continuously without using demagnetizing control, current transformer accuracy may still be improved by reducing the amount that the induction level fluctuates during current-sensing operation). In a-c power system applications that derive operating power from the current transformer secondary current, three modes of operation may function sequentially (rather than just two modes as previously discussed):
(1) A power supply charging mode, during which the current transformer may become magnetized.
(2) A demagnetizing mode, during which the current transformer is demagnetized.
(3) A current-sensing mode, during which current transformer secondary current is accurately proportional to primary current.
In some applications it may be advantageous for secondary current to be sensed during the first two modes, even though current transformer accuracy is not as good during these modes.
In implementing the "first principle" with a current transformer that is in service, it should be kept in mind that the magnetomotive force (mmf) applied to the magnetic core is the sum of the mmf of the primary current and the secondary current. These two mmf s are usually of opposite polarity and largely cancel each other during normal current transformer operation. These two mmf s do not normally cancel each other during the demagnetizing mode of the invention.
Detailed Description of the Drawings
FIG. 1 illustrates the general concept by showing a magnetic body 60, a winding 61 wrapped around the magnetic body, and a controllable voltage device 63 with a suitable control circuit 62. Magnetic body 60 and winding 61 are shown in a typical current transformer configuration with the magnetic body around an electric power system conductor 64 with an insulating covering 65. Power system conductor 64 functions as a primary winding with only one turn, with a primary current Jl flowing. Though shown with one end disconnected, power system conductor 64 is normally connected as part of an electric power system.
Controllable voltage device 63 may be an active voltage source (or, more generally, a controllable electric energy source), in which case a primary winding is not required. In this case, power system conductor 64 (acting as a primary winding) may be omitted. Resistor Rl may be a current-sensing resistor, or it may represent the internal resistance of voltage device 63, or it may be a current-limiting resistor. Some applications may find it advantageous to replace resistor Rl with a complex impedance. Resistor Rl may be omitted for many applications.
As shown in FIG. 1, winding 61, voltage device 63, and resistor Rl are connected in series, so that current J2 flows through each of them.
If another source of magnetic excitation is available (such as current flowing in a primary winding), then controllable voltage device 63 may be an adjustable impedance. In the case of a current transformer in service, resistor Rl is usually included as a current-sensing resistor with a low value of resistance, and controllable voltage device 63 may be an adjustable impedance rather than an active voltage source. Often, current is sensed by a larger monitoring system as voltage V3 across resistor Rl, since this voltage is proportional to current J2.
Winding 61 is shown with ten turns around magnetic body 60. The actual number of turns may vary widely depending on the application. Magnetic body 60 is shown as a toroid, though wide variation in magnetic body configurations is possible and the illustration is not intended to limit the breadth of application of the invention.
Control circuit 62 is shown with three high-impedance voltage-sensing inputs (connected to conductors 66, 67, and 68) to enable sensing of voltage VI, voltage V2, and voltage V3. Voltage V3 across resistor Rl is proportional to current J2. Voltage V2 is the voltage across winding 61, which is the induced voltage generated by changing flux in magnetic body 60 plus any voltage drop associated with current J2 flowing through stray winding impedances. Often, the voltage drop associated with current J2 flowing through stray winding impedances is small compared to the induced voltage and may be ignored in some applications. Voltage VI is the output voltage of controllable voltage device 63.
Voltage-sensing conductors 66, 67 and 68 are not required for the simplest embodiments of the invention, but are included to clarify the general concept. Alternatively, other means of sensing current and/or voltages may be utilized, as may be preferred for different embodiments.
Several control conductors 69 act as an interface between control circuit 62 and controllable voltage device 63. The actual interface between control circuit 62 and controllable voltage device 63 may vary widely depending on the particular design. Control circuit 62 may be constructed utilizing prior art, with the control sequences being in accordance with the present invention. In the case that controllable voltage device 63 is an active voltage source, it also may be constructed utilizing prior art, with the control sequences being in
accordance with the present invention.
In the example of FIG. 1, magnetic body 60 provides a closed magnetic path with no need for other magnetic components.
FIG. 2 shows an alternate configuration, in which a cylindrical magnetic body 70 does not provide a closed magnetic path by itself. Instead, a closed magnetic path is provided by additional magnetic material 72. For best operation, magnetic material 72 should be configured to have greater saturation flux capacity than magnetic body 70, so that the saturation characteristics of magnetic body 70 dominate the resulting magnetic circuit characteristics. One way this may be done is by configuring magnetic material 72 to have a greater cross- sectional area than magnetic body 70. Since the configuration shown in FIG. 2 does not have a primary winding (or other source of magnetic excitation), controllable voltage device 63 (shown in FIG. 1) is now shown as a controllable electric energy source 73. A winding 71 is wrapped around magnetic material 72. As in FIG. 1, resistor Rl may be a current- sensing resistor, or it may represent the internal resistance of electric energy source 73, or it may be a current- limiting resistor. FIGS. 3 and 4A to 4D illustrate how a magnetic body may be demagnetized utilizing "first-phase method a" and "second-phase method b" previously discussed.
FIG. 3 is a hysteresis curve that clarifies one way that the induction level of a magnetic body is controlled by the invention. These changes correlate to the waveforms shown in FIGS. 4A to 4D. The horizontal axis X of FIG. 3 is proportional to magnetomotive force (ampere-turns), and the vertical axis Y is proportional to the induction level (magnetic flux density) of the magnetic body. Magnetization of the magnetic body increases as the operating point moves away from axis X. A magnetic material with a relatively square hysteresis curve has purposely been chosen, as this simplifies the demagnetizing operation.
The induction level of a magnetic body begins at a random magnetized state shown as point A of FIG 3. The "first phase" moves the induction level to saturation near point B. The "second phase" shown includes two steps. A "first step" moves the induction level somewhat passed an induction level of zero to point C. A "second step" moves the induction level to near zero at point D.
FIGS. 4A to 4D show one method of controlling voltages and currents to demagnetize a magnetic body by using the sequence shown in FIG. 3. The waveforms shown in FIGS. 4 A to 4D are applicable to the configuration shown in FIG. 2, or to FIG. 1 when controllable voltage device 63 is an active voltage source and primary conductor 64 is either disconnected or omitted. The waveforms shown in FIGS. 4A to 4D correlate to the sequential changes of induction level shown in FIG. 3. Voltage V2 shown in FIG. 4B is the induced voltage across the winding, with any voltage drop associated with current flowing through stray winding impedances assumed to be negligible for the present discussion.
FIG. 4D shows how the induction level varies with time. The vertical axis represents induction level, and is scaled similar to the vertical axis of the hysteresis curve shown in FIG. 3.
Referring to FIGS. 1, 3 and 4A to 4D, at time T41 the "first phase" of the demagnetizing cycle begins when voltage VI is driven from zero volts to a positive value by controllable voltage device 63 as controlled by control circuit 62. This positive voltage causes the induction level of magnetic body 60 to transition from point A of
FIG. 3 to saturation at point B. Saturation begins at about time T42 when current J2 suddenly increases as shown in FIG. 4C. The transition point is also marked by voltage V2 suddenly decreasing, as shown by FIG. 4B. At time T43 the "first step" of the second phase begins when the polarity of voltage VI is reversed and the induction level of magnetic body 60 begins a transition from point B of FIG. 3 to point C. In the example shown, control circuit 62 controls the time period of the first step in order to change the induction level by an amount somewhat greater than change Yl, stopping at point C at time T44. At time T44 an optional "second step" begins when the polarity of voltage VI is reversed again and the induction level of magnetic body 60 begins a transition from point C of FIG. 3 to point D. At time T45 the second step ends when voltage VI is changed to zero volts, and magnetic body 60 is left in a demagnetized state at point D (at an induction level near zero). Change Yl correlates to the "volt-time constant" previously discussed.
If the optional "second step" of the second phase was not included between times T44 and T45 of FIGS. 4A to 4D, then a small d-c current would continue flowing after time T44, with a natural decay rate determined mostly by inductance associated with magnetic body 60 and the resistance of winding 61 and resistor Rl. The final state of magnetic body 60 after this decay will also be an induction level near zero (at or near point D) as long as point C is close to an induction level of zero. Ideally, with or without the second step, point C of FIG. 3 should be at a point from which the natural decay of d-c current (without the second-step voltage) will leave the magnetic body at an induction level of zero. This naturally decaying d-c current flowing through resistance provides an uncontrolled second-step voltage that should be considered in volt-time calculations for the first step. As indicated previously regarding the second phase, the voltage and time periods associated with the first step and second step should be controlled such that the volt-time integral of voltage V2 is the same as the "volt- time constant" of magnetic body 60 and winding 61. Referring to FIG. 4B, the volt-time constant (of magnetic body 60 and winding 61) is related to areas Al and A2 (with the hatched areas calculated in units of volt-time). More specifically, area Al minus area A2 should be the same as the volt-time constant. For the sequence shown in FIG. 3 and FIGS. 4A to 4D, the appropriate value of the volt-time constant is predetermined, and control circuit 62 is configured beforehand to control voltage device 63 in the manner illustrated to effectively demagnetize magnetic body 60.
FIG. 5 and FIGS. 6A to 6D illustrate the same sequence as FIG. 3 and FIGS. 4A to 4D, except that the preferred induction level is not zero. In the example shown in FIG. 5 and FIGS. 6A to 6D the preferred induction level is 50% of saturation on the positive side of the hysteresis curve. The sequence shown in FIG. 5 is the same as FIG. 3, except that in FIG. 5 the transition to sequence step
C only changes the induction level by an amount slightly greater than change Y2. This change moves the induction level somewhat past +50%, and the transition to sequence step D then leaves the final induction level at about 50% of saturation. The smaller change in FIG. 5 corresponds to a smaller area A4 in FIG. 6B (compared to area Al of FIG. 4B). Change Y2 corresponds to the volt-time value of a change in induction level from +100% to +50%, which correlates to the volt-time value of area A4 minus area A5 of FIG. 6B. Times T46 and T47 are included for ease of reference, and indicate the end of the first and second steps of the second phase.
While the subsequent figures and description focus mostly on the demagnetizing function of the invention, it should be noted that the sequences required to establish a non-zero preferred induction level are almost the
same as the sequences utilized to establish an induction level near zero. The principles and embodiments discussed below are not intended to be limited only to the demagnetizing function of the invention.
FIGS. 7 and 8A to 8D illustrate "first-phase method c" and "second-phase method b" previously discussed. FIG. 7 shows a hysteresis curve sequence that can be used to automatically determine the volt-time constant of a magnetic body and winding and then demagnetize the magnetic body. FIGS. 8A to 8D show how an alternating voltage and current may be used to implement the sequence shown in FIG. 7.
In FIG. 8A, voltage VI (the output voltage of voltage device 63) between times T53 and T57 is the same as in FIG. 4A between times T41 and T45. In FIG. 8A, however, voltage VI begins the demagnetizing cycle with a negative polarity between times T51 and T53. This initial negative voltage moves the induction level from an unknown level shown as point A on the hysteresis curve of FIG. 7 to saturation at point B of FIG. 7. Saturation near point B is reached at about time T52. At time T53 of FIG. 8A voltage VI changes to positive polarity and drives the induction level toward point C of FIG. 7. Starting at time T53, control circuit 62 monitors either voltage V2 or current J2 to determine when saturation is reached. Saturation is reached at time T54, which corresponds to saturation near point C of FIG. 7. The volt-time constant of the magnetic body may be determined by calculating the volt-time integral of voltage V2 between times T53 and T54 (this calculation determines the volt-time area shown as crosshatched area A3) and dividing the result by two. During the second phase (the transition from point C to point D and point E of FIG. 7) voltage V2 is then controlled so that the volt-time integral from time T55 to time T57 has the same magnitude as the calculated volt-time constant. Another way of saying this is that voltage V2 is controlled during the second phase so that area Al minus area A2 is half of area A3, Time T56 indicates the transition point between the first and second step of the second phase.
For improved accuracy, the effect of winding resistance and current on voltage V2 may be automatically compensated for by the control circuit automatically determining winding resistance and calculating the volt-time integrals using corrected voltage. The winding resistance may be calculated as voltage V2 divided by current J2 between times T52 and T53 while the magnetic body is in saturation. Corrected voltage (induced voltage) is then calculated as voltage V2 minus (current J2 multiplied by winding resistance). Alternatively, total loop resistance may be calculated as voltage VI divided by current J2 (between times T52 and T53), and induced voltage may then be calculated as voltage VI minus current J2 multiplied by total loop resistance.
FIG. 9 illustrates how a magnetic body may be demagnetized utilizing "first-phase method b" and "second- phase method b" previously discussed.
FIG. 9 shows operation with a not-so-square hysteresis curve assumed for a magnetic body. Operation shown is similar to that described for FIG. 3 except that now current may not be large enough to drive the magnetic body to saturation. Similar to FIG. 3, horizontal axis X is proportional to magnetomotive force (ampere-turns), and vertical axis Y is proportional to the induction level (magnetic flux density) of the magnetic body. Magnetization of the magnetic body increases as the operating point moves away from axis X.
Referring to FIG. 9, when the demagnetizing mode begins, the induction level of a magnetic body generally is not known. This corresponds to a randomly chosen point A on the hysteresis curve of FIG. 9. The "first phase" of the demagnetizing cycle causes a current to flow with peak magnitude corresponding to magnetomotive
force X2. This causes a transition to point B, with an induction level of Y3 (which is not at saturation). Induction level Y3 may be calculated based on peak current and known characteristics of the magnetic body (in accordance with the "first principle"). The degree of accuracy obtainable in determining induction level Y3 is somewhat dependent on the initial induction level, the magnitude of magnetomotive force X2, and the actual hysteresis characteristics of the magnetic body. Accuracy may be improved by cycling the current between opposite polarities once to remove the effects of an initial magnetization. During the "second phase," a first voltage is applied to the winding for a time period such that the induction level of the magnetic body transitions from point B to point C (in accordance with the "second principle"). A second voltage of the opposite polarity is then applied for a time period such that the induction level of the magnetic body transitions to point D. At point D, the magnetic body is demagnetized.
FIGS. 10, 11 and 12A to 12C illustrate "first-phase method c" and "second-phase method b" previously discussed.
FIG. 10 shows a method to demagnetize a magnetic body regardless of whether current is adequate to drive the magnetic body to saturation. Again, sequence point A is assumed for a starting point. In order to be certain that the magnetic core is not originally magnetized to a level higher than the induction level attainable during the "first phase," at least one complete cycle of an alternating current is first applied. This forces the magnetic core first to point B, then down to point C, cycling (almost) once around its hysteresis loop, thereby removing the effects of the initial induction level. Then, the current is made to alternate again to drive the induction level back to point D, but this time the induced voltage of the winding is integrated over the time required for the transition, and a "volt-time value" is calculated from this to determine the magnitude and duration of a voltage pulse that will move the induction level from point D to point F. The following "second phase" uses the "volt-time value" just calculated to move the induction level from point D (through point E) to point F and then to point G, which is close to an induction level of zero. Point E is the zero-crossing point of current.
FIG. 11 shows a relatively simple embodiment of the invention utilizing a controllable electric energy source 76. A winding 71A conducts current J2 and interacts with a magnetic body (not shown). The winding may be similar to windings 61 and 71 of FIGS. 1 and 2. The magnetic body (not shown) may be similar to either magnetic body 60 or 70 of FIGS. 1 and 2. A switch 78 is used to connect and disconnect an alternating voltage source 77 to winding 71A. Switch 78 will preferably be a solid-state electronic switch able to stop current flow mid-cycle. Alternating voltage source 77 may simply be a source of 50 Hertz or 60 Hertz electric power with suitable voltage magnitude, providing a consistent repeating waveform that causes induced voltage V2 to have a consistent repeating waveform. An optional resistor R77 may be included to limit peak current levels. An optional varistor 79, or other transient voltage suppression device may be included to limit transient voltages when switch 78 is opened. A control circuit 62A senses voltage V2 and controls switch 78. Zener diodes D21 and D22 have breakover voltages somewhat higher than the peak voltage of alternating voltage source 77, and are configured to limit transient voltages when switch 78 is opened. Zener diodes D21 and D22 may be omitted, in which case switch 78 should be configured to be to able to absorb the high momentary voltage associated with stopping current J2 very quickly. Alternatively, almost any kind of surge suppressor or impedance could be used in place of the zener diodes, but zener diodes are shown for ease of illustration and explanation. One simple
method of operation is illustrated by FIGS. 12A to 12C.
FIGS. 12A to 12C show one way that the configuration shown in FIG. 11 may utilize an alternating voltage and current to implement a sequence similar to the hysteresis curve sequence of FIG. 10. FIG. 12A shows voltage V2, which (prior to time T74) is merely the sinusoidal voltage generated by alternating voltage source 77 (less any voltage drop across resistor R77). FIG. 12B shows current J2, which is not quite sinusoidal due to hysteresis characteristics of the magnetic body. FIG. 12C shows induction level. Switch 78 (FIG. 11) is closed prior to time T71, and any transient d-c offset currents that may be associated with closing switch 78 have already decayed prior to time T71.
Refer now to FIGS. 10, 11, and 12A to 12C. Between times T71 and T72 the operation is in a steady-state alternating cycle. At time T72 the induction level is at a positive peak (FIG. 12C), corresponding to sequence point D of FIG. 10. The induction level then declines to an induction level corresponding to point F at time T74, at which time switch 78 is opened by control circuit 62A. A negative current is flowing at time T74, and this current is forced to flow through zener diodes D21 and D22 after switch 78 opens (the current keeps flowing due to the inductive nature of winding 71A). This causes a positive voltage pulse between times T74 and T75 with area A8 (FIG 12A), which is associated with a transition from point F to point G of FIG. 10, at which point current J2 has declined to zero and the induction level is near zero.
Similar to area A3 of FIG. 8B being twice the volt-time constant, area A6 of FIG. 12A is twice the volt- time value associated with a change in induction level from the peak induction level to an induction level near zero. The actual peak induction level may be calculated from the volt-time value using equation no. 8 (above). For optimum demagnetizing operation, the timing of the opening of switch 78 should be such that area A7 minus area A8 equals half of area A6.
Referring to FIG. 12A, time T74 (the point of time that switch 78 is opened) should be somewhat past the geometric middle of the half-cycle for best results. However, the demagnetizing operation will be satisfactory for many applications if the switch is simply opened near the geometric middle of the half-cycle. "Geometric middle of the half-cycle" is intended to mean the point of time at which the volt-time integral of the half-cycle would be divided into two equal parts (visually this would be the time at which area A7 would be exactly half of area A6). Preferred induction levels other than zero may also be established by opening switch 78 at other points of time during the same half-cycle. Equation no. 6 shows how a volt-time value may be calculated which may be used to determine the optimum point of time to open switch 78 to establish a preferred induction level. Opening switch 78 between times T73 and T74 will result in induction levels between point E and an induction level of zero, depending on the specific point of time that switch 78 is opened. Opening switch 78 after T74, but still during the same half-cycle, will result in induction levels with opposite polarity.
The waveforms shown in FIGS. 12A to 12C are waveforms typically associated with peak induction levels being less than saturation induction levels. In the case that the voltage magnitude of alternating voltage source 11 is sufficient to drive the magnetic body to saturation, the alternating waveforms shown prior to T74 may be considerably different than those shown, but the principles of operation are similar.
FIGS. 13 to 26 all illustrate how the invention may be applied to current transformers that are in service. The principles utilized, however, are applicable to other types of magnetic bodies as well. FIGS. 13 through 19
illustrate how an adjustable impedance may be utilized in a current transformer secondary circuit to demagnetize a current transformer. FIGS. 20 through 25 illustrate how an active voltage source may be utilized in a current transformer secondary circuit.
FIG. 13 shows a functional schematic illustrating how voltage device 63 of FIG. 1 may be embodied by an adjustable impedance, which is part of a current transformer demagnetizing circuit 1. Demagnetizing circuit 1 includes a control circuit 3 and an adjustable impedance 2 comprising electronic switches SI, S2 and S3 and zener diodes Dl and D2. Electronic switches SI, S2 and S3 are controlled by control circuit 3 as indicated by the dashed lines between switches SI, S2 and S3 and control circuit 3.
The configuration shown for adjustable impedance 2 is for illustration purposes only, and is not intended to define all possible configurations of the present invention. It may be noted that switch SI is optional, since closing switches S2 and S3 at the same time approximates the effect of closing switch SI.
FIG. 13 shows a current transformer CT1 which may have a magnetic core similar to magnetic body 60 of FIG. 1 and a secondary winding similar to winding 61 of FIG. 1. Conductor 64 is a primary conductor conducting a primary current Jl, also similar to FIG. 1. Current Jl is an alternating electric current flowing as part of a larger system. Current Jl causes secondary current J2 to flow by the transformer action of current transformer CT1. Current J2 is normally proportionally smaller than current Jl by the turns ratio of current transformer CT1. Under ideal conditions, the waveform of current J2 is virtually the same as the waveform of current Jl,
Resistor Rl is a current-sensing resistor, connected in series with current J2, thereby producing a voltage signal across resistor Rl that is proportional to current J2. This voltage signal, conducted by conductor 15, is usually used as an input to some kind of current monitoring system as provided for by terminal 16. This voltage signal may also be used as an input to control circuit 3 as is presently shown. Other kinds of current-sensing means may be used in place of resistor Rl.
A power supply 4 is configured to derive power from input current J2 whenever switch S4 is closed and switches SI, S2 and S3 are open. Power supply 4 and switch S4 are optional (switch S4 may be optional even if power supply 4 is included, depending on the configuration of power supply 4). If power is not derived from current J2, a separate source of power will usually be required to provide operating power to control circuit 3. Power is supplied to control circuit 3 via one or more conductors 13 and common conductor 11, the total number of power conductors being dependent on the specific design of control circuit 3 and power supply 4. Terminals 12 and 14 are included to provide for the possibility of power being transferred to or from other circuits.
Conductor 17 and terminal 18 provide for a control signal to or from a larger system to coordinate current- sensing and demagnetizing modes.
Voltage V2 is now the voltage across the secondary winding of current transformer CT1. This voltage is shown connected as an input to control circuit 3 via conductor 15 and conductor 19. For the present discussion, voltage V2 will be considered to be the induced voltage (the voltage drop associated with current flowing through stray winding impedances will be considered negligible).
Optional ground connection 10 provides a stable signal reference potential for common conductor 11. FIGS. 14A to 14E illustrate the preferred operation of FIG. 13. The magnetic sequence utilized is similar
to the sequence shown by the hysteresis curve of FIG. 3, utilizing "first-phase method a" and "second-phase method b" previously discussed. Three different operating modes are shown. First, from time Tl to time T2 the power supply charging mode is active. Second, from time T2 to time T10 the demagnetizing mode is active. Third, after time T10 the current-sensing mode is active. Actual test waveforms may vary somewhat from those shown depending on the specific power supply configuration, current transformer characteristics, and current magnitudes.
FIG. 14A shows current transformer CT1 primary current Jl as a simple sine wave for simplicity of illustration, though in many applications it may be considerably distorted. FIG. 14B shows secondary current J2. Current magnitudes have been normalized for simplicity of illustration. Current J2 is normally many times smaller than current Jl, as influenced by the turns ratio of current transformer CT1. FIG. 14C shows operation of electronic switches SI, S2, S3 and S4 as controlled by control circuit 3. Dark lines indicate time periods during which each switch is closed. Blank spaces indicate time periods during which each switch is open. FIG. 14D shows secondary voltage V2. The voltage across resistor Rl is normally small compared to voltage V2, and no attempt has been made to show its minor influence on V2. FIG. 14E shows induction level with a scale similar to the hysteresis curve shown in FIG. 3.
The operating cycle illustrated in FIGS. 14A to 14E begins with the functional schematic of FIG. 13 being in the power supply charging mode. Between times Tl and T2, only switch S4 is closed, and power supply 4 is charging. The wavefoπns shown in FIGS. 14B and 14D between times Tl and T2 are discussed in detail in U.S. Patent No. 6,018,700 for a "Self-Powered Current Monitor" to Thomas G. Edel, issued January 25, 2000. The first phase of the demagnetizing mode begins at time T2, a somewhat random point of time. The duration of the first phase lasts a predetermined time period that is adequate to drive the magnetic core to saturation as indicated by point B of FIG. 3. At time T2, switch S2 closes and switch S4 opens to set adjustable impedance 2 to the nonlinear characteristics of zener diode Dl. This causes voltage V2 to be large and positive for current J2 with positive polarity, and small and negative for current J2 with negative polarity. The large positive voltage causes the induction level to transition to saturation. During this first phase, switch SI is optionally closed between times T3 and T4 and between times T5 and T6 to minimize the magnitude of negative voltages that occur during negative half-cycles of current J2. Switch SI is controlled during the first phase based on the polarity of current J2. The predetermined minimum time required for the first phase ends sometime between times T5 and T7. After this time period ends, control circuit 3 looks for the beginning of a negative half-cycle of current J2 to begin the second phase.
The second phase of the demagnetizing mode begins at time T7 when control circuit 3 senses current J2 going negative. At time T7, switch S3 closes and switch S2 opens. This sets adjustable impedance 2 to the nonlinear characteristics of zener diode D2. This causes voltage V2 to be large and negative for current J2 with negative polarity, and small and positive for current J2 with positive polarity. At time T7 a first-step voltage pulse begins, the magnitude of which is the breakover voltage of zener diode D2. Control circuit 3 keeps switch S3 closed for a predetermined time period (until time T8) that correlates with the volt-time value associated with a transition from point B to point C of FIG. 3. An optional second step voltage is implemented between times T9 and T10 by closing only switch S2, thereby causing the transition from point C to point D of FIG. 3.
The current-sensing mode begins a time T10 when switch SI is closed, thereby allowing current to flow freely.
Operation without power supply 4 is similar, except that the initial power supply charging cycle will not be part of the sequence. In FIG. 14D, voltage pulses between times T7 and T8 and between times T9 and T10 cause the induction level to transition from saturation to an induction level near zero. Area A9 minus area AlO should correspond to the volt-time constant of current transformer CT1. It is important that control circuit 3 be able to control adjustable impedance 2 so that the integral of voltage over time for each second-phase voltage pulse is a particular value for each particular type of current transformer. To facilitate this, it may be preferable to configure contiol circuit 3 to directly monitor voltage V2 via conductors 15 and 19 (as shown in FIG. 13).
However, it will usually suffice to monitor only current J2 (via the voltage signal on conductor 15, the voltage across resistor Rl), with the magnitude of voltage V2 known to be the applicable zener diode voltage drop, as determined by the status of the electronic switches (SI, S2, S3, S4) and the direction (polarity) of current J2. With this type of operation, the conductor 19 input connection to control circuit 3 may be eliminated, and voltage V2 is not an input to control circuit 3. Current J2 is then the only feedback signal to control circuit 3 used to accurately time voltage V2 reset pulses within particular half-cycles.
While it is preferable for control circuit 3 to have current J2 and/or voltage V2 as a feedback signal, it is possible for control circuit 3 to operate without any feedback signal. This could be accomplished by modifying the first-step voltage pulse of the second phase (between times T7 and T8 of FIG. 14D) in either of the following ways:
(a) The time period that zener diode D2 is active could be extended to one or more complete cycles, with the zener diode breakover voltage adjusted so that the net integral of voltage over time is the same as the previous value. Lengthening duration to one or more complete cycles in this way allows the switching to occur without regard to actual half-cycle time intervals. This method has the drawback that each kind of current transformer utilized may require different voltage magnitudes for the reset pulses.
(b) The negative voltage pulse could be split into several shorter intervals divided over one or more complete cycles. Spreading out the time that zener diode D2 is active over one or more complete cycles in this way allows the switching to occur without regard to actual half-cycle time intervals. This method of operation is illustrated by FIGS. 15A to 15E. In both cases (a) and (b), control circuit 3 may operate without feedback signals related to secondary voltage V2 or secondaiy current J2.
FIGS. 15 A to 15E illustrate how the functional schematic of FIG. 13 may be operated without a feedback signal.
From time Tl to T6, operation is similar to FIGS. 14A to 14E. At time Til, the second phase begins at a random time in the middle of a half-cycle when the first phase times out. Zener diode D2 is activated briefly several times over a time period one cycle in length (between times Til and T12) to generate a voltage V2 waveform (FIG. 15D) such that the integral of the voltage over the time period between times Til and T12 is approximately equal to that of FIG. 14D between times T7 and T8. Stated another way, zener diode D2 is made
active with a pulse-width-modulated type of control (controlling switches SI and S3) to cause voltage V2 to have an average negative value suitable for the first step of the second phase (between times Til and T12). The demagnetizing effect of the multiple pulses shown in FIG. 15D (between times Til and T12) is similar to the single voltage pulse shown in FIG. 14D (between times T7 and T8). At time T12 of FIGS. 15A to 15E, the optional second-step voltage begins. Similar to the first-step voltage between times Til and T12, zener diode Dl is made active with a pulse-width-modulated type of control (controlling switches SI and S2) to cause voltage V2 to have an average positive value suitable for the second step (between times T12 and T13). The effect of the multiple pulses shown in FIG. 15D (between times T12 and T13) is similar to the single voltage pulse shown in FIG. 14D (between times T9 and T10). At time T13 of FIGS. 15A to 15E, the current-sensing mode begins.
FIG. 16 shows one possible embodiment of a current transformer demagnetizing circuit based on the functional schematic of FIG. 13. Components that are common to FIG. 13 function in the manner previously described. The demagnetizing circuit shown is incorporated into a current monitoring circuit that derives operating power from current transformer CT1 secondary current J2. The electronic switches previously shown in the functional schematic of FIG. 13 are now implemented with field-effect transistors. The preferred embodiment utilizes N-channel enhancement mode devices with low drain-source on resistance and sensitive gates for operation at logic voltage levels.
The power supply circuit includes a full wave bridge rectifier circuit consisting of diodes D5 and D6 and the drain-source diodes within field-effect transistors Q5 and Q6. The rectified current charges capacitor Cl, which provides an unregulated voltage for use by a d-c to d-c converter circuit 23. Voltage across capacitor Cl is limited by the breakover voltage of zener diodes Dll and D12.
Resistor Rl is still a current-sensing resistor, but in this configuration current J2 flows through resistor Rl only when field-effect transistors Q5 and Q6 are actuated. Since the current through resistor Rl is not always the same as current J2, it is shown as current J3. Conductor 29 conducts a voltage signal that is proportional to the current through resistor Rl. Resistor R2 and zener diodes D7 and D8 form an optional surge suppression network to protect an analog-to-digital converter circuit 22. The resistance of resistor R2 must be small compared to the input resistance of analog-to-digital converter circuit 22 so that accuracy of the voltage signal will not be adversely affected. Analog-to-digital converter circuit 22 and a microcontroller 21 function in a manner similar to control circuit 3 of FIG. 13. Data is communicated between analog-to-digital converter circuit 22 and microcontroller 21 via serial or parallel communication utilizing several conductors 35.
Zener diodes Dll and D12 function in a manner similar to zener diodes Dl and D2 of FIG. 13. Zener diodes Dll and D12 also act as voltage-limiting devices to limit the charge on power supply capacitor Cl.
Field-effect transistors Q5 and Q6 perform the switching functions of switches SI, S2, and S3 of FIG. 13. Field-effect transistors Q5 and Q6 are individually controlled, with resistors R8 and R9 providing a discharge path for gate charge (these resistors may be optional depending on the configuration of microcontroller 21).
Activating field-effect transistor Q6 is similar to closing switch S2 of FIG. 13. Activating field-effect transistor Q5 is similar to closing switch S3 of FIG. 13. Activating both field-effect transistors Q5 and Q6 is similar to closing switch SI of FIG. 13. When both field-effect transistors Q5 and Q6 are deactivated the system is in the
power supply charging mode and zener diodes Dll and D12 limit the voltage on capacitor Cl.
The function of switch S4 in FIG. 13 is not required in the embodiment of FIG. 16, since the power supply is configured to automatically charge whenever the current-sensing mode and demagnetizing mode are not active. The power supply may be easily eliminated if an alternate power source is available for control power. This may be done by removing diodes D5 and D6, Capacitor Cl, and d-c to d-c converter circuit 23.
More than one current transformer (each normally having a unique primary current) may be connected to the configuration of FIG. 16 by duplicating the components within box 41 for each current transformer. These components may be connected to conductors 11A and 42, similar to the configuration shown for current transformer CT1. Additional connections to analog-to-digital converter circuit 22 and microcontroller 21 will also be needed for each current transformer, similar to conductors 29, 43, and 44 for current transformer CT1. With additional current transformers configured similar to current transformer CT1, each current transformer may contribute power to the power supply. More than one current transformer may be demagnetized at a time, assuming analog-to-digital converter circuit 22 and microcontroller 21 are capable of the faster processing and communication required for this (unique control signals are preferred for each current transformer in order to implement feedback control similar to FIGS. 14A to 14D). If a non-feedback type of control is utilized (similar to FIGS. 15A to 15E), then only two control conductors (43 and 44) may be used to simultaneously demagnetize all current transformers at once with little demand on the microcontroller.
FIGS. 17A to 17D illustrate the operation of the embodiment shown in FIG. 16. The operation shown is similar to the operation described in FIGS. 14A to 14E. The embodiment shown in FIG. 16 may also be operated in a manner similar to that described for FIGS. 15A to 15E.
FIG. 17A shows alternating electric current Jl, similar to FIG. 14A.
FIG. 17B shows secondary current J2, similar to FIG. 14B.
FIG. 17C shows the time periods that the field-effect transistors are actuated. A solid line indicates a high gate voltage, whereas a blank space indicates a low gate voltage. The switching sequence shown is functionally similar to FIG. 14C.
FIG. 17D shows secondary voltage V2, similar to FIG. 14D. However, now the voltage-limiting circuit is comprised of zener diodes Dll and D12, so the waveforms associated with the power supply charging cycle (between times Tl and T2) are different than in FIG. 14D. Also, the effect of capacitor Cl charging is evident at the beginning of each voltage pulse.
FIG. 18 shows a veiy simple embodiment of a current transformer demagnetizing circuit, utilizing only one field-effect transistor for an adjustable impedance. All components required only for self-powered applications have been removed, so a separate source of operating power is required. A power supply 23A is now included, which provides regulated d-c operating power from a separate power source. To emphasize the simplicity of the present invention, surge suppression devices have been removed as well as gate resistors (all of which may be considered optional, depending of the particular application and microcontroller configuration). Operation is similar to FIG. 16, but now the breakover voltage of the internal body diode of transistor Q5 is utilized for voltage control during the first phase, and the forward voltage drop of the body diode is used for the voltage
required by the second phase. This configuration also has the advantage of the current transformer being directly connected to grounded conductor 11, which is a requirement of some safety codes for some applications.
Though somewhat limited by the low voltage available for implementing the second phase (only about one volt), FIG. 18 illustrates a very simple method of implementing an adjustable impedance, and illustrates the use of a single three-terminal device as an adjustable impedance.
The main drawback to the configuration shown in FIG. 18 is that the low voltage available for the second phase will often require that the second phase be spread over several cycles. One way of increasing this voltage (and thereby reducing the time required) is to use two transistors in series with a common gate connection. This will double the second phase voltage to about 2 volts (the forward voltage drop of two diodes). Of course additional transistors could be added in series to increase the voltage still further. It should be noted from this that it is possible to make a single three-terminal semiconductor device (utilizing prior-art MOSFET technology) that has on/off parameters well suited for use as an adjustable impedance in this type of circuit.
FIGS. 19A to 19E show typical operating waveforms and control signals associated with the embodiment shown in FIG. 18. The sequence shown utilizes "first-phase method a" and "second-phase method c" previously discussed.
Referring to FIGS. 18 and 19A to 19E, operation is similar to that just described for FIGS. 16 and 17A to 17D, with the clarification that second phase voltage is merely the forward voltage drop of the body diode of transistor Q5. The first phase voltage is the breakover voltage of the internal body diode of transistor Q5, and this voltage is sufficient to drive the induction level to saturation in less than one half-cycle of current J2. However, the second phase voltage is not sufficient to complete the second phase transition (from saturation to an induction level near zero) in a single half-cycle of current J2. For this reason transistor Q5 must be turned on during the following positive half-cycle (from time T66 to time T67, to avoid going back to saturation), and turned off again at the beginning of the next negative half-cycle to continue moving the induction level toward zero. This type of cycling continues until time T68, at which time the induction level is near zero. The combined area of area A12, area A13, and area A14 should equal the volt-time constant of the current transformer.
FIGS. 20 through 25 all illustrate how an active voltage source may be utilized in a current transformer secondary circuit to demagnetize a current transformer and also improve the accuracy of a current transformer by reducing the amount that the induction level fluctuates during current-sensing operation. These figures also show how a preferred induction level may be maintained by causing the average value of the induced voltage over time to have a value near zero. The principles illustrated are applicable to other magnetic bodies in addition to current transformers.
FIGS. 20A to 20D show how the configuration of FIG. 1 may be used to demagnetize a current transformer using an active voltage source, thereby improving current transformer accuracy and enabling measurement of d-c current. To demagnetize magnetic core 60, the sequence shown utilizes "first-phase method a" and "second- phase method b," similar to the hysteresis curve of FIG. 3 and waveforms of FIGS. 4A to 4D (one difference being that the initial induction level of the present example is at -100% saturation instead of -50%). FIGS. 20A to 20D assume a varying d-c current for primary current Jl, though the sequence shown is equally applicable to a-c current. The waveforms shown are based on controllable voltage device 63 being a controllable voltage
source. Resistor Rl is a current-sensing resistor with low resistance. After a demagnetizing cycle, the current through resistor Rl is proportional to the d-c primary current. As with previous examples, voltage V2 is considered to be the induced voltage.
At time T81 of FIGS. 20A to 20D the magnetic core is saturated due to primary current Jl being a d-c current, with no recent demagnetizing cycle to prevent the magnetic core from being saturated. Current J2 is zero, since the magnetic core is unable to generate any voltage to drive a current of the same polarity as current Jl. Though the induction level at time T81 may be considered "known" from the perspective of this description, it is not known from the perspective of a current monitoring device. The present example will drive the induction level to a "known" induction level of +100% during the first phase, since this will be more informative than driving it to -100% (the induction level it is already at).
The first-phase of a demagnetizing cycle begins at time T82 when control circuit 62 controls voltage device 63 to produce a large positive voltage. This voltage causes a transition from saturation at -100% to saturation at +100%. Saturation at +100% is reached at about time T83. During the transition, while the core is not in saturation between times T82 and T83, current J2 becomes somewhat proportional to current Jl, with an error related to hysteresis curve properties.
A second phase begins at time T84, when control circuit 62 causes voltage device 63 to produce a large negative first-step voltage, followed by a shorter positive second-step voltage beginning at time T85. At time T86 the second-step voltage is done, and control circuit 62 causes voltage device 63 to produce an output voltage near zero volts to allow secondary current to flow freely. As with previous examples, control circuit 62 controls the first-step and second-step voltages so that the magnitude of the volt-time integral of voltage V2 during the second phase equals the volt-time constant of the current transformer (area A19 minus area A20 should equal the volt-time constant).
At time T86 the current transformer is demagnetized, and secondary current J2 is proportional to primary current Jl. However, since the current transformer must generate a small induced voltage to keep current J2 flowing through the loop resistance (resistor Rl plus winding resistance plus other stray resistances), the induction level starts to drift back toward saturation after time T86 (see FIG. 20D). This transition toward saturation will cause an increasing error in current J2 until saturation is reached, at which point current J2 will cease to flow (as at time T81). To maintain accuracy, the demagnetizing cycle must be repeated periodically (usually well before saturation is reached). For applications in which current flow is consistently one direction, it may be beneficial to move the initial operating point to an induction level other than zero to lengthen the time period between demagnetizing cycles. For example, in the example of FIGS. 20A to 20D, it may be preferable to shorten the first-step voltage (between times T84 and T85) so that the induction level is left initially at a positive level, say +25%. The magnetic core will still drift toward saturation at -100%, but it will take more time. FIG. 21 shows one way that control circuit 62 and voltage device 63 of FIG. 1 may be embodied. The configuration shown is particularly suitable for measuring electric current, including d-c current. This co-αfiguration may be operated in a manner similar to the manner described for FIGS. 20A to 20D. Many components are the same as FIG. 1, and these components function in the manner previously described. Since voltage device 63 of FIG. 21 is a controllable electric energy source, the same embodiment may be utilized for
electric energy source 73 of FIG.2. Of course, control circuit 62 of FIG. 2 may also be embodied the same way - as shown in FIG. 21.
Still referring to FIG. 21, resistor Rl is a current-sensing resistor with low resistance. Control circuit 62 has an analog-to-digital converter circuit 81 to sense current J2 (as voltage V3 across resistor Rl), a microcontroller 82 for data processing and control functions, and a digital-to-analog converter circuit 83.
Digital-to-analog converter circuit 83 provides an analog voltage signal on conductor 85 that controls the voltage output of voltage device 63. Analog-to-digital converter circuit 81 and microcontroller 82 communicate via an interface shown as four conductors, and this interface may vary considerably depending on the particular design. Likewise microcontroller 82 communicates with digital-to-analog converter circuit 83 via an interface that may vary considerably depending on the particular design. Voltage device 63 has an operational amplifier 84, with resistors R81 and R82 configured to set the gain of operational amplifier 84. A separate power source (not shown) provides operating power for analog-to-digital converter circuit 81, microcontroller 82, digital-to-analog converter circuit 83, and operational amplifier 84.
Ground connection 86 provides a common reference potential for the various circuits and power supply. If a particular application requires that winding 61 be directly grounded on one side, then resistor Rl may be relocated and connected in series with the opposite side of winding 61. This complicates the measurement of voltage V3 across resistor Rl somewhat, but prior-art differential voltage measurement methods are adequate.
Operational amplifier 84 must be able to produce voltage in a circuit with relatively large current driven by a current source (a current transformer acts like a current source). A "power operational amplifier" will usually be required, such as model OPA548 made by Burr-Brown Corporation. This device has an adjustable current- limit feature and is rated for up to 3 amps of continuous current (other models are available with higher ratings). Tests found that these devices worked very well in a current transformer secondary circuit, as long as a "snubber circuit" was connected to the output to improve stability. The "snubber circuit" that was successfiilly utilized was a ten Ohm resistor in series with a 0.1 microfarad capacitor, connected between the operational amplifier output and the grounded conductor (as recommended by the device's data sheet). The OPA548 operational amplifier also has provision for adjustable current limit, which is beneficial for this application.
The type of control configuration shown in FIG. 21 (utilizing an analog-to-digital converter circuit to sense an input signal, a microcontroller to implement a control function based on the input signal, a digital-to-analog converter to produce an analog control signal, and an operational amplifier to produce a voltage proportional to the analog control signal) is well established in the prior art, so additional configuration details will not be described herein. Microcontroller 82 may be configured to implement the control sequence illustrated by FIGS. 20A to 20D.
FIGS. 22A to 22E illustrate four different ways that the configurations of FIG. 1 or FIG. 21 may be operated during a current-sensing mode after a demagnetizing sequence is completed. Conceptually, FIGS. 22B, 22D and 22E may be thought of as a continuation in time of FIGS. 20A, 20C and 20D respectfully. Time T86 is common to both sets of figures. However, the vertical scales of FIGS. 22B, 22D, and 22E are magnified by factors of about 4, 4, and 8 respectively to facilitate display of wavefoπns with relatively small magnitude. Voltage V2 is now considered to differ from the induced voltage V (see FIGS. 22B and 22C) by an amount that
depends on current J2 and stray winding impedances.
Referring to FIG. 21, current J2 flows through winding 61, resistor Rl, voltage device 63, and various conductors that connect these different components. The "loop resistance" associated with current J2 is the sum of the resistances of all the components that current J2 flows through. With the assumption that operational amplifier 84 has near ideal properties, the resistance of voltage device 63 will be considered negligible. Then the loop resistance for current J2 of FIG. 21 is the sum of the resistances of winding 61, resistor Rl and various conductors that connect these different components. In the present example the total loop resistance will be assumed to be about twice the resistance of resistor Rl. For the present discussion, reactive impedances in the loop are considered negligible (see FIG. 26 for how reactive impedances may be accounted for). Referring now to FIGS. 21 and 22A to 22E, from time T86 to time T87 voltage device 63 is controlled by control circuit 62 to produce an output voltage of zero volts. This allows cuπent J2 to flow freely, but since cuπent J2 is not symmetrical, induced voltage V has an average negative value to keep cuπent J2 flowing. Induced voltage V is equal and opposite to the voltage drop of cuπent J2 around the loop, which is approximately equal to the magnitude of cuπent J2 multiplied by the loop resistance (neglecting stiay inductances and capacitances of loop components). This induced voltage, with an average negative value, coπelates with the induction level of magnetic core 60 drifting toward saturation as shown in FIG. 22E. To compensate for this drift toward saturation, voltage device 63 may be controlled to produce a voltage with an average value equal to voltage V3 across resistor Rl. This is the case shown in FIGS. 22A to 22E between times T87 and T88. However, this is not enough voltage to counteract the voltage drop across the stray resistance in the loop (winding and conductor resistances), and the drift toward saturation continues at a slower rate (as shown by FIG. 22E).
Between times T88 and T89 voltage device 63 is controlled to produce a voltage equal to twice the average value of voltage V3 (since the total loop resistance is twice the resistance of resistor Rl), and this voltage causes induced voltage V to have an average value near zero, stopping the drift toward saturation (a larger voltage would cause induced voltage V to have an average positive value and cause the induction level to drift toward a positive direction). Between times T89 and T90 voltage device 63 is controlled to produce a voltage equal to twice the instantaneous value of voltage V3, and this voltage also stops the drift toward saturation while causing induced voltage V to continuously be very near zero volts and greatly reducing induction level fluctuations.
Either control method utilized between times T88 and T90 in FIGS. 22A to 22E may be used to maintain a prefeπed induction level. The control method shown between times T89 and T90 has the added advantage of reducing the magnitude of the induced voltage to near zero (induction level fluctuations and magnetizing cuπent are also reduced to near zero). For cuπent transformer applications, one could say that the effective burden of the entire secondary circuit is reduced to near zero Ohms. A secondary circuit with a total effective burden near zero Ohms can significantly increase the accuracy of a cuπent transformer as well as reduce the size of the magnetic core required to attain a particular level of accuracy. This is an improvement over the active load termination described in Reissued U.S. Patent Re. 28,851 (to
Milkovic, reissued 1976, for a "Cuπent Transformer with Active Load Termination"), since the configuration disclosed in that patent only reduces the burden of the sensing device to near zero (rather than the burden of the whole secondary circuit).
In order to operate the configuration of FIG. 21 in either manner illustrated between times T88 and T90 of FIGS. 22A to 22E, it is necessary to determine the loop resistance for current J2. The loop resistance may be predetermined and microcontroller 82 may be preconfigured with the proper operating parameters. Alternatively, the loop resistance may be determined automatically by microcontroller 82 operating to drive the magnetic core to saturation with a known voltage VI and then measuring cuπent J2 after the magnetic core is saturated. The loop resistance is then simply the magnitude of voltage VI divided by cuπent J2.
The contiol method utilized between times T89 and T90 in FIGS. 22A to 22E may be used in a-c current- sensing systems to virtually eliminate the burden of cuπent transformer secondary circuits (without necessarily utilizing the demagnetizing sequence shown in FIGS. 20A to 20D). A simple embodiment just for this purpose is illustrated in FIG. 23. The configuration of FIG. 23 is similar to FIG. 21 except that voltage device 63 is now controlled so that voltage VI is always proportional to voltage V3. The control means is now integral to voltage device 63 (adjustable resistor R83 and resistor R82 control the gain of operational amplifier 84). Conductor 87 conducts a voltage signal (voltage V3) which is proportional to cuπent. Voltage V3 may also be communicated to a larger cuπent-sensing system via terminals 88 and 89. Still referring to FIG. 23, if adjustable resistor R83 is set to zero ohms, the voltage gain of voltage device
63 is one, and the circuit has a similar effect as the active load disclosed in Reissued U.S. Patent Re. 28,851. With a gain of one, voltage VI is equal to voltage V3, so that the combination of resistor Rl and voltage device 63 has a combined voltage near zero volts and therefore has a combined effective burden near zero ohms. Similar to the discussion for FIGS. 22A to 22E, if the total loop resistance for current J2 is twice the resistance of resistor Rl, and if voltage device 63 is controlled so that voltage VI is twice as large as voltage V3, the voltage drop of the entire loop would be compensated for. This can be done by making the resistance of adjustable resistor R83 in FIG. 23 equal to the resistance of resistor R82. This causes the induced voltage of winding 61 to be very small, and the effective burden of the entire secondary circuit is reduced to almost zero ohms. Adjustable resistor R83 may be adjusted to obtain any reasonable gain greater than one, in accordance with well-known design principles of operational amplifiers. By adjusting resistor R83 the gain is adjustable to match secondary circuit parameters for most secondary circuit configurations. Of course, resistor R83 may be a fixed resistor if the appropriate gain is predetermined. Care must be exercised, however, because if the gain is set to compensate for more than the total secondary circuit impedance, then the system becomes unstable and cuπent will increase to the limit of the amplifier output circuit.
For secondary circuits with non-negligible inductance or capacitance, resistors R82 and R83 may be replaced with complex impedances to better match secondary circuit parameters, thereby facilitating a total secondary circuit burden near zero ohms. Complex impedances may also be configured to compensate for cuπent transformer imperfections, such as hysteresis effects and eddy cuπents. Alternatively, a reactive impedance may be added to sensing resistor Rl (for example, like inductive reactance LI in FIG. 26) in order to provide a feedback voltage that is proportional to the voltage drop caused by cuπent flowing through complex loop impedances.
FIG. 24 shows a configuration that combines the benefits of a dedicated sensing winding 90 with the
versatility of the control circuit shown in FIG. 21. Similar to FIG. 21, the configuration shown in FIG. 24 may be utilized for sensing both a-c and d-c cuπents, since control circuit 62 is capable of demagnetizing the magnetic core. With the configuration shown in FIG. 24, microcontroller 82 may be programmed to implement a burden-reducing method similar to FIG. 23 (controlling voltage device 63 based on current J2 and known loop impedances). Alternatively, microcontroller 82 may be programmed to control voltage device 63 to minimize the sensed voltage across winding 90. This may be done simply with a proportional type of control, controlling voltage device 63 to produce an output voltage proportionally many times larger than the sensed voltage, with polarity so as to reduce induced voltage.
The control configuration shown in FIG.24 may be applied to many kinds of magnetic bodies in addition to cuπent transformers. With the induced voltage across dedicated winding 90 utilized as a feedback control signal, the induction level of a magnetic body may be made to vary with time in almost any manner desired. Microcontroller 82 may be programmed to calculate changes in induction level as the integral of induced voltage over time, and utilize a P.I.D. (Proportional plus Integral plus Derivative) type of control to obtain the response desired. Other types of control, such as fuzzy logic, may also be utilized. These feedback control methods are well known in the prior art, so they will not be discussed in detail herein. Similar configurations are possible utilizing analog control means in place of the digital control means shown for control circuit 62.
FIG. 25 shows a variation of FIG. 23 with added provision for a demagnetizing sequence. This combines the accuracy of sensing cuπent with near zero induction level fluctuation with a demagnetizing means. Similar to FIG. 21, the configuration of FIG. 25 may be used to measure d-c cuπent as well as a-c current. FIG. 25 has the advantage of not requiring continual microcontroller supervision during the cuπent-sensing phase. P- channel field-effect transistor Ql and N-channel field-effect transistor Q2 along with demagnetizing controls 91 have been added to enable voltage device 63 to produce the varying voltage required for a demagnetizing sequence. Positive voltage VP and negative voltage VN are supplied by a separate power supply (not shown). Demagnetizing controls 91 may be configured utilizing prior art. A microcontroller may be utilized for proper timing of the demagnetizing sequence, or an analog timing circuit may be utilized. The gate of field- effect transistor Ql should be driven low to cause voltage device 63 to produce a negative voltage, and the gate of field-effect transistor Q2 should be driven high to cause voltage source 63 produce a positive voltage. With both field-effect transistors Ql and Q2 off, the circuit functions similar to FIG. 23. Field-effect transistors Ql and Q2 should not be turned on at the same time, as this results in a relatively low impedance between power supply connections VP and VN. Resistors R82 and R83 should have resistance values set small enough that any leakage cuπent through field-effect transistors Ql and Q2 during the cuπent sensing mode will not adversely affect operation of voltage device 63.
Surge protection (not shown) for the configurations of FIGS. 21, 23, 24, and 25 may be provided by connecting a varistor, back-to-back zener diodes, or other voltage-limiting device between the output of operational amplifier 84 and the grounded conductor (thereby limiting voltage VI to a safe level and still provide for sensing of surge cuπents). Dual silicon controlled rectifiers or a triac (both with suitable trigger circuits) may be particularly suitable. U.S. Patent No. 4,466,039 to Moran and Reis (1984), discloses a suitable triac circuit that may be used.
FIG. 26 is a variation of FIG. 1 showing one way that induced voltage may be calculated taking into account the effects of stray impedances, and one way that calculated induced voltage may be utilized as a feedback signal to facilitate control of induced voltage.
Inductive reactance LI has been added to resistor Rl, the combination being impedance Z5. The stray impedances of winding 61 are represented by winding stray impedance Z6, which is shown as a combination of resistance and inductive reactance. Other loop impedances are shown as impedance Z7, which is also shown as a combination of resistance and inductive reactance. Capacitive reactance could also be accounted for but it is assumed to be negligible for the present discussion. The total effective loop impedance is the sum of impedances Z5, Z6, and Z7. The voltage across resistor Rl is still directly proportional to current, and may be used by an external system to sense current via terminals 88 and 89.
Voltage V2 is now strictly the measurable voltage across winding 61, with the actual induced voltage V differing from voltage V2 by a voltage drop V6 cause by current flowing through stray impedance Z6. Induced voltage V may be calculated as voltage V2 minus voltage drop V6. Alternatively, induced voltage V may be calculated as voltage VI minus voltages V5, V6, and V7 (in accordance with KirchofFs voltage law). In FIG. 26, inductor LI and resistor Rl are sized so that the ratio of the reactive component to resistive component of impedance Z5 is approximately the same as the ratio of the reactive component to resistive component of the series combination of winding stray impedances Z6 and impedance Z7. Then the instantaneous magnitude of the sum of voltages V5, V6 and V7 will be approximately proportional to the instantaneous magnitude of voltage V5. Induced voltage V may then be calculated as voltage VI minus voltage V5 multiplied by a constant. The operational amplifiers shown in FIG. 26 are configured to calculate induced voltage V in this manner.
More specifically, operational amplifier 95 and resistors R35 and R36 are configured to produce a voltage output equal to voltage V5 multiplied by a constant. Resistor R35 is adjustable to facilitate making the output approximately equal to the sum of voltage drops V5, V6 and V7. Operational amplifier 96 and resistors R37, R38, R39, and R40 are configured such that the output of operational amplifier 96 is approximately equal to voltage VI minus voltage V5 multiplied by a constant, which is approximately induced voltage V (resistors R37, R38, R39, and R40 should all have the same value). This calculated induced voltage is conducted by conductor 97 and is a feedback input to a controller 62B.
Controller 62B may also (optionally) receive a variable control signal (from an external system) that is proportional to the induction level that is desired (the "reference induction level"). Such a control signal is shown in FIG. 26 as an input signal on conductor 98. Since the induced voltage is proportional to the rate of change (or the "derivative") of induction level (per equation no. 4), it may be beneficial to calculate the rate of change (the derivative) of the control signal. This calculation is performed by operational amplifier 99, capacitor C4, and resistor R38. Conductor 100 conducts the negative of the derivative as an input into controller 62B. Controller 62B is configured with suitable prior-art controls (such as a proportional plus integral plus derivative type of control system, or fuzzy logic controls, or any other suitable control means) to force the induced voltage to be proportional to the derivative of the control signal (with an appropriate constant of proportionality). Alternatively, the induced voltage could be integrated, and the controller could function to keep the integral of
induced voltage proportional to the control signal (per equation no. 5).
Since there may be some inaccuracies in the contiol loop, the induction level may drift over time away from the level coπesponding to the control signal on conductor 98. For this reason, controller 62B may be configured to occasionally reset the induction level in magnetic core 60 to the preferred level as indicated by the control signal on conductor 98. A sequence similar to that shown in FIGS. 5 through 6D may be utilized to reestablish a prefeπed induction level.
Of course, the analog calculations performed by the operational amplifiers in FIG. 26 could also be performed digitally utilizing an analog-to-digital converter and a microcontroller, similar to the configuration of FIG. 21. If the configuration of FIG. 21 is used, induced voltage may be calculated without measuring voltage VI, since microcontroller 82 controls this voltage and therefore knows its magnitude without measuring it.
INDUSTRIAL APPLICABILITY
The invention provides for improved accuracy in cuπent measurements for electric power monitoring and other cuπent monitoring applications. Since almost all industries utilize electricity and are therefore concerned with accurate measurement of electric cuπent, the invention has broad application. Also, many industries are concerned with controlling the magnetization of magnetic bodies other than cuπent transformers, so the invention is expected to have diverse application.
CONCLUSIONS, RAMIFICATIONS, AND SCOPE
The present invention applies a varying voltage to a conductive winding to contiol the induction level of a magnetic body. Improvements over prior art include, but are not limited to, the following: (a) how induced voltage may be controlled to establish a preferred induction level,
(b) how induced voltage may be controlled to maintain a prefeπed induction level,
(c) how induced voltage may be controlled to reduce the amount that the induction level fluctuates (thereby improving accuracy in cuπent transformer applications, and benefitting other applications as well),
(d) how induced voltage may be controlled to cause the induction level of a magnetic body to vary with time in a prefeπed manner,
(e) how induced voltage and induction level may be controlled by utilizing an adjustable impedance in the secondary circuit of a cuπent transformer (or other multi-winding application), and
(f) how induced voltage may be controlled to enable standard cuπent transformers to measure d-c cuπents, as well as a-c cuπents with d-c components. Several different embodiments have been described and illustrated. There are many other embodiments possible that will be apparent to those skilled in the art. It is not the intent of this disclosure to limit the invention to the embodiments that have been illustrated. The components and configurations utilized in this disclosure are intended to be illustrative only, and are not intended to limit the scope of the appended claims. While only certain prefeπed features of the invention have been shown by way of illustration, many modifications and changes will occur to those skilled in the art. It is, therefore, to be understood that the appended claims are intended to cover all such modifications and changes as fall within the true spirit of the invention.
Claims
1. A method for controlling the induction level of a magnetic body; said induction level being a function of time; said magnetic body being positioned relative to a conductive winding in such a way that a change of said induction level is associated with an induced voltage across said winding; said induced voltage being proportional to the rate of change of said induction level; said method comprising the application of a voltage to said winding; said voltage controlling said induced voltage; said voltage and said induced voltage both being functions of time; said voltage causing said induced voltage to have such waveform and magnitude that the integral over time of said induced voltage correlates to desired changes of said induction level.
2. The method of claim 1 wherein said voltage is controlled for a period of time in such a way as to cause said induction level to transition from a determinate induction level at the beginning of said period of time to a prefeπed induction level at the end of said period of time; said voltage causing said waveform and magnitude of said induced voltage to have such characteristics that the integral of said induced voltage over said period of time is approximately equal to a volt-time value corresponding to a change of said induction level from said determinate induction level to said prefeπed induction level, thereby causing said induction level to transition from said determinate induction level to said prefeπed induction level.
3. The method of claim 2 wherein said preferred induction level is an induction level near zero, and said method is thereby a method to demagnetize said magnetic body.
4. The method of claim 2 wherein said induced voltage during said period of time is entirely of one polarity.
5. The metliod of claim 2 wherein said induced voltage during said period of time sequentially comprises a first voltage and a second voltage of opposite polarities; said first voltage having such magnitude, duration and polarity so as to cause said induction level to transition to and pass said prefeπed induction level; said second voltage having such magnitude, duration and polarity so as to cause said induction level to transition to said prefeπed induction level.
6. The metliod of claim 2 wherein said induced voltage during said period of time comprises a plurality of sequential voltage pulses of varying form, magnitude, duration, and polarity.
7. The method of claim 2 wherein the initial magnitude of said induction level is not known, and said method further comprises a first phase prior to said period of time during which said voltage is controlled in such a way as to cause cuπent flowing in said winding to have such waveform and magnitude so as to cause said induction level to transition to said determinate induction level; operation previously described during said period of time comprising a second phase; said first phase and said second phase occurring sequentially, thereby providing a method of establishing said prefeπed induction level when said initial magnitude of said induction level is not known.
8. The method of claim 7 wherein said prefeπed induction level is an induction level near zero, and said method is thereby a method to demagnetize said magnetic body.
9. The metliod of claim 7 wherein said magnetic body comprises saturable magnetic material; said cuπent during said first phase has sufficient magnitude to drive said induction level to an induction level near saturation; and said determinate induction level is approximately the saturation induction level of said magnetic material.
10. The method of claim 7 wherein said determinate induction level is calculated from known characteristics of both said magnetic body and said cuπent.
11. The method of claim 7 wherein during said first phase said cuπent is an alternating cuπent and said induced voltage is an alternating induced voltage; said alternating cuπent and said alternating induced voltage having waveforms and magnitudes that are related by the properties of said magnetic body and said winding; said alternating cuπent and said alternating induced voltage each having various characteristics herein refeπed to in general as wave parameters; said alternating electric cuπent causing said induction level to transition between opposite peak induction levels that are approximately equal in magnitude; one or more of said wave parameters being utilized to determine said determinate induction level.
12. The method of claim 11 wherein said determinate induction level is the induction level at the end of a positive or negative half-cycle of said alternating induced voltage; said determinate induction level being determined by determining the magnitude of a second volt-time value; said second volt-time value being approximately equal to the value of the volt-time integral of a positive or negative half-cycle of said alternating induced voltage divided by two; said second phase further beginning at the end of a positive or negative half-cycle of said alternating induced voltage.
13. The method of claim 7 wherein said magnetic body further interacts with a second winding, herein refeπed to as a primary winding; said primary winding comprising one or more turns of conductive material; said primary winding being positioned relative to said magnetic body in such a way that said primary winding magnetically interacts with said magnetic body; said primary winding conducting a primary cuπent; said method further comprising a third phase during which said cuπent is approximately proportional to said primary cuπent.
14. The method of claim 13 wherein said primary cuπent is not symmetrical, and said induced voltage during said third phase is controlled so that the average value of said induced voltage over time is near zero, thereby preventing said induction level from quickly transitioning away from said prefeπed induction level.
15. The method of claim 13 wherein said voltage is controlled during said third phase in such a way as to reduce the magnitude of said induced voltage, thereby making said cuπent a more accurate representation of said primary cuπent during said third phase.
16. The method of claim 13 wherein said magnetic body is the magnetic core of a cuπent transformer, said winding is the secondary winding of said current transformer, and said primary cuπent is a d-c cuπent; said method being utilized as a method of producing a secondary d-c cuπent that is proportional to said d-c cuπent during said third phase.
17. The metliod of claim 1 wherein said induction level is influenced by a magnetic field caused by a magnetomotive force independent of said winding and said voltage is controlled for a period of time in such a way as to maintain said induction level near a prefeπed induction level; said prefeπed induction level being present at the beginning of said period of time; said voltage further causing said induced voltage to have such waveform and magnitude during said period of time that the integral of said induced voltage from the beginning of said period of time to any point of time within said period of time does not exceed a predetermined value, thereby causing said induction level to be maintained near said prefeπed induction level.
18. The method of claim 1 wherein said induction level is influenced by a fluctuating magnetic field caused by a magnetomotive force independent of said winding, and said voltage is controlled for a period of time in such a way as to reduce the amount that said induction level fluctuates; said voltage being controlled during said period of time in such a way as to cause the magnitude of said induced voltage to be significantly smaller than would be the case without said method, thereby reducing the amount that said induction level fluctuates.
19. The method of claim 1 wherein said method causes said induction level to approximately match a reference induction level for a period of time; said reference induction level being a function of time; said method-further comprising an initial step of determi-αing a constant of proportionality; said constant of proportionality having such value that said induced voltage is approximately equal to said rate of change of said induction level multiplied by said constant of proportionality; said voltage further causing the instantaneous magnitude of said induced voltage to continuously be approximately equal to the rate of change of said reference induction level multiplied by said constant of proportionality, thereby causing said induction level to approximately match said reference induction level.
20. The method of claim 1 wherein not more than one additional winding magnetically interacts with said magnetic body.
21. The method of claim 1 wherein information related to said induced voltage is derived from characteristics of current flowing in said winding, and said information is utilized as a feedback signal to adjust the waveform and magnitude of said voltage, thereby improving the control of said induction level.
22. The method of claim 1 wherein said magnetic body is positioned relative to a second winding in such a way that a change of said induction level is associated with a second voltage across said second winding; said second voltage being approximately proportional to said induced voltage; said second voltage being utilized as a feedback signal to adjust the waveform and magnitude of said voltage, thereby improving the control of said induced voltage and said induction level.
23. The method of claim 1 wherein said voltage is produced by a cuπent flowing through an adjustable impedance, the waveform and magnitude of said voltage being controlled by controlling the characteristics of said adjustable impedance.
24. The method of claim 1 wherein said magnetic body is the magnetic core of a current transformer, and said winding is the secondary winding of said current transformer; said method being used to control said induction level in such a way that the accuracy of said cuπent transformer is improved.
25. Apparatus for controlling the induction level of a magnetic body; said apparatus comprising
(a) a winding,
(b) a controllable voltage device connected to said winding, and
(c) a suitable control means; said winding comprising one or more turns of conductive material positioned relative to said magnetic body in such a way that a change of induction level of said magnetic body is associated with an induced voltage across said winding; said induced voltage being proportional to the rate of change of said induction level; the output of said voltage device controlling said induced voltage; said control means controlling said voltage device in such a way that the integral over time of said induced voltage correlates to desired changes of said induction level.
26. The apparatus of claim 25 wherein said apparatus operates for a period of time to cause said induction level to transition from a determinate induction level at the beginning of said period of time to a preferred induction level at the end of said period of time; said control means controlling said voltage device so as to produce an output voltage with such waveform and magnitude so as to cause the integral of said induced voltage over said period of time to be approximately equal to a volt-time value coπesponding to a change of said induction level from said determinate induction level to said prefeπed induction level, thereby causing said induction level to transition from said determinate induction level to said preferred induction level.
27. The apparatus of claim 26 wherein said control means further operates during a first phase prior to said period of time to control said voltage device in such a way as to cause cuπent flowing in said winding to have such waveform and magnitude so as to cause said induction level to transition to said determinate induction level; operation previously described during said period of time being a second phase; said first phase and said second phase occurring sequentially, said apparatus thereby functioning to establish said prefeπed induction level whether or not the initial magnitude of said induction level is known.
28. The apparatus of claim 27 wherein said prefeπed induction level is an induction level near zero, and said apparatus thereby functions to demagnetize said magnetic body.
29. The apparatus of claim 27 wherein said magnetic body further interacts with a second winding, herein refeπed to as a primary winding; said primary winding comprising one or more turns of conductive material; said primary winding being positioned relative to said magnetic body in such a way that said primary winding magnetically interacts with said magnetic body; said primary winding conducting a primary cuπent; said control means further operating during a third phase to control said voltage device so as to allow said cuπent to flow freely, said cuπent being approximately proportional to said primary cuπent during said third phase.
30. The apparatus of claim 29 wherein said primary cuπent is not symmetrical, and said control means further operates during said third phase to control said voltage device so as to cause the average value of said induced voltage over time to be near zero, thereby preventing said induction level from quickly transitioning away from said prefeπed induction level, thereby making said current a more accurate representation of said primary cuπent during said third phase.
31. The apparatus of claim 29 wherein said voltage device comprises a controllable electric energy source, and said control means further operates during said third phase to control said electric energy source in such a way that said induced voltage is reduced, thereby making said cuπent a more accurate representation of said primary current during said third phase.
32. The apparatus of claim 29 wherein said voltage device comprises a controllable electric energy source; and said control means urther operates during said third phase to control said electric energy source so as to produce an output voltage that counteracts the voltage drop caused by said cuπent flowing through loop impedances, thereby reducing the magnitude of said induced voltage during said third phase, thereby making said cuπent a more accurate representation of said primary cuπent during said third phase.
33. The apparatus of claim 25 wherein said induction level is influenced by a magnetic field caused by a magnetomotive force independent of said winding, and said apparatus operates for a period of time to maintain said induction level near a prefeπed induction level; said prefeπed induction level being present at the beginning of said period of time; said contiol means controlling said voltage device during said period of time so as to cause said induced voltage to have such waveform and magnitude that the integral over time of said induced voltage from the beginning of said period of time to any point of time within said period of time does not exceed a predetermined value, thereby causing said induction level to be maintained near said prefeπed induction level.
34. The apparatus of claim 25 wherein said induction level is influenced by a fluctuating magnetic field caused by a magnetomotive force independent of said winding, and said apparatus operates for a period of time to reduce the amount that said induction level fluctuates; said control means controlling said voltage device during said period of time in such a way as to cause the magnitude of said induced voltage to be significantly smaller than would be the case without said apparatus, thereby reducing the amount that said induction level fluctuates.
35. The apparatus of claim 25 wherein said apparatus operates for a period of time to cause said induction level to approximately match a reference induction level; said induction level and said reference induction level both being functions of time; said induced voltage being approximately equal to said rate of change of said induction level multiplied by a constant of proportionality; said contiol means controlling said voltage device during said period of time so as to cause said induced voltage to continuously be approximately equal to the rate of change of said reference induction level multiplied by said constant of proportionality, thereby causing said induction level to approximately match said reference induction level.
36. The apparatus of claim 25 wherein not more than one additional winding magnetically interacts with said magnetic body.
37. The apparatus of claim 25 wherein said apparatus further comprises a sensing means providing information related to said induced voltage; said control means receiving said information and utilizing said information as a feedback signal to improve the control of said induction level; said sensing means comprising a voltage-measuring means measuring a first voltage across a first impedance connected in series with both said winding and said voltage device; said first voltage being directly related to a cuπent flowing in series through said winding and said first impedance; said induced voltage being calculable from the output voltage of said voltage device, characteristics of impedances through which said current-flows, and said first voltage.
38. The apparatus of claim 25 wherein said apparatus further comprises a sensing means providing information related to said induced voltage; said contiol means receiving said information and utilizing said information as a feedback signal to improve the control of said induction level; said sensing means comprising a second winding maghqtically coupled to said magnetic body, a second voltage measurable across said second winding being approximately proportional to said induced voltage.
39. The apparatus of claim.25 wherein said winding will also be refeπed to as a secondary winding; said secondaiy winding conducting a secondary cuπent; said magnetic body further magnetically interacting with a primary winding; said primary winding comprising one or more turns of conductive material; said primary winding conducting an alternating cuπent; said primary winding being positioned relative to said magnetic body in such a way that said alternating current influences said induction level of said magnetic body; said voltage device comprising an adjustable impedance connected in series with said secondary winding; said secondary cuπent flowing through said adjustable impedance; said control means controlling the output voltage of said voltage device by controlling the characteristics of said adjustable impedance; said output voltage being a function of the magnitude and polarity of said secondary cuπent and said characteristics of said adjustable impedance.
40. The apparatus of claim 39 wherein said magnetic body is the magnetic core of a cuπent transformer, said secondary winding is the secondary winding of said cuπent transformer, and said apparatus operates to demagnetize said magnetic core; said secondary winding conducting said secondary cuπent in one of two directions at a time, herein called direction one and direction two; said control means acting sequentially, first causing said adjustable impedance to be a relatively high impedance for said secondary cuπent flowing in direction one, thereby causing said magnetic core to transition to a determinate induction level; said control means then causing said adjustable impedance to be a relatively high impedance for said secondary current flowing in direction two, thereby causing said induced voltage to have such magnitude and waveform that said induction level transitions to an induction level near zero, thereby demagnetizing said magnetic core.
41. The apparatus of claim 39 wherein said magnetic body is the magnetic core of a current transformer, said secondary winding is the secondary winding of said current transformer, and said apparatus operates to demagnetize said magnetic core; said secondary winding conducting said secondary current in one of two directions at a time, herein called direction one and direction two; said control means acting sequentially, first causing said adjustable impedance to be a relatively high impedance for said secondary cuπent flowing in said direction one, thereby causing said magnetic core to transition to a determinate induction level; said control means then causing said adjustable impedance to be a relatively high impedance for said secondary cuπent flowing in said direction two, thereby causing said induced voltage to have such magnitude and waveform that said induction level transitions somewhat past an induction level of zero; said control means then causing said adjustable impedance to be a relatively high impedance for said secondary cuπent flowing in said direction one, thereby causing said induced voltage to have such magnitude and waveform that said induction level transitions to an induction level near zero, thereby demagnetizing said magnetic core.
42. The apparatus of claim 25 wherein said magnetic body is the magnetic core of a cuπent transformer, and said winding is the secondary winding of said cuπent transformer; said cuπent transformer unctioning to provide a secondary current that is approximately proportional to a primary cuπent; said secondary current flowing in said secondary winding; said primary cuπent flowing in a conductor configured as a primary winding of said current transformer; said control means controlling said voltage device in such a way as to make said secondary current a more accurate representation of said primary current than would be the case without said voltage device.
43. The apparatus of claim 42 wherein the accuracy of said cuπent transformer is improved by causing said magnetic core to transition to a prefeπed induction level; said control means operating during a first phase to cause said voltage device to produce an output that causes said secondary ciurent to have such waveform and magnitude that said induction level transitions to a determinate induction level; said control means then operating during said second phase to cause said voltage device to produce an output that causes said induced voltage to have such waveform, magnitude and duration that the value of the integral over time of said induced voltage is approximately equal to a volt-time value coπesponding to a change in said induction level from said determinate induction level to said prefeπed induction level, thereby causing said induction level to transition from said determinate induction level to said prefeπed induction level. said control means then operating during a third phase to cause said voltage device to produce an output that allows said secondary cuπent to flow freely; said secondary cuπent being approximately proportional to said primary current during said third phase.
44. The apparatus of claim 43 wherein said voltage device is a controllable electric energy source and said primary cuπent is a d-c cuπent.
45. The apparatus of claim 42 wherein the accuracy of said cuπent tiansformer is improved by reducing the amount that said induction level fluctuates; said apparatus further comprising a sensing means providing information related to said induced voltage; said sensing means comprising a first impedance connected in series with said secondary winding; said secondary cuπent flowing through said first impedance; a first voltage across said first impedance providing said information related to said induced voltage. said voltage device comprising a controllable electric energy source; said induced voltage and associated fluctuation of said induction level adversely affecting the accuracy of said cuπent transformer; said control means receiving said information from said sensing means and controlling said electric energy source in such a way that the magnitude of said induced voltage is reduced, thereby reducing the amount that said induction level fluctuates, thereby improving the accuracy of said cuπent transformer.
46. The apparatus of claim 45 wherein said first impedance is a resistor, and said control means causes said electric energy source to produce a second voltage that is proportional to said first voltage across said resistor with magnitude approximately equal to the magnitude of said secondary cuπent multiplied by the loop resistance of said secondary circuit, and with polarity so as to facilitate the flow of said secondary current, thereby reducing the amount that said induction level fluctuates.
47. The apparatus of claim 45 wherein a total effective loop impedance is defined as the sum of said first impedance plus the stray impedances of said winding plus any additional impedances which are connected in series and which conduct said secondary cuπent; a voltage drop being associated with said secondary cuπent flowing through said total effective loop impedance; said total effective loop impedance comprising a second reactive component and a second resistive component; said first impedance comprising a first reactive component and first resistive component such that the ratio of said first reactive component to said first resistive component is approximately equal to the ratio of said second reactive component to said second resistive component; said first voltage therefore being approximately proportional to said voltage drop; said control means causing said electric energy source to produce a second voltage that is approximately equal to said first voltage multiplied by a constant of proportionality; said constant of proportionality having such magnitude and polarity that said second voltage causes a significant reduction in the amount that said induction level fluctuates.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/US2000/030358 WO2002037512A1 (en) | 2000-11-03 | 2000-11-03 | Method and apparatus for controlling the magnetization of current transformers and other magnetic bodies |
AU2001214613A AU2001214613A1 (en) | 2000-11-03 | 2000-11-03 | Method and apparatus for controlling the magnetization of current transformers and other magnetic bodies |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/US2000/030358 WO2002037512A1 (en) | 2000-11-03 | 2000-11-03 | Method and apparatus for controlling the magnetization of current transformers and other magnetic bodies |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2002037512A1 true WO2002037512A1 (en) | 2002-05-10 |
Family
ID=21741961
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/US2000/030358 WO2002037512A1 (en) | 2000-11-03 | 2000-11-03 | Method and apparatus for controlling the magnetization of current transformers and other magnetic bodies |
Country Status (2)
Country | Link |
---|---|
AU (1) | AU2001214613A1 (en) |
WO (1) | WO2002037512A1 (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
RU2674580C1 (en) * | 2017-12-27 | 2018-12-11 | федеральное государственное бюджетное образовательное учреждение высшего образования "Ивановский государственный энергетический университет имени В.И. Ленина" (ИГЭУ) | Method for determining saturation of the magnetic wire of current transformer |
CN109378155A (en) * | 2018-12-27 | 2019-02-22 | 国网宁夏电力有限公司 | A kind of converter power transformer eraser system |
JP2021170041A (en) * | 2017-07-20 | 2021-10-28 | 株式会社トーキン | Current sensor |
WO2023189593A1 (en) * | 2022-03-29 | 2023-10-05 | 株式会社トーキン | Electric current sensor device |
KR102671333B1 (en) * | 2024-01-03 | 2024-06-04 | 주식회사 코본테크 | DC current sensor circuit that eliminates error caused by offset voltage |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4456875A (en) * | 1981-07-07 | 1984-06-26 | Hydro-Quebec | Demagnetization circuit for current transformers |
-
2000
- 2000-11-03 AU AU2001214613A patent/AU2001214613A1/en not_active Abandoned
- 2000-11-03 WO PCT/US2000/030358 patent/WO2002037512A1/en active Application Filing
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4456875A (en) * | 1981-07-07 | 1984-06-26 | Hydro-Quebec | Demagnetization circuit for current transformers |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2021170041A (en) * | 2017-07-20 | 2021-10-28 | 株式会社トーキン | Current sensor |
JP7132409B2 (en) | 2017-07-20 | 2022-09-06 | 株式会社トーキン | current sensor |
RU2674580C1 (en) * | 2017-12-27 | 2018-12-11 | федеральное государственное бюджетное образовательное учреждение высшего образования "Ивановский государственный энергетический университет имени В.И. Ленина" (ИГЭУ) | Method for determining saturation of the magnetic wire of current transformer |
CN109378155A (en) * | 2018-12-27 | 2019-02-22 | 国网宁夏电力有限公司 | A kind of converter power transformer eraser system |
CN109378155B (en) * | 2018-12-27 | 2024-05-17 | 国网宁夏电力有限公司 | Converter transformer degaussing system |
WO2023189593A1 (en) * | 2022-03-29 | 2023-10-05 | 株式会社トーキン | Electric current sensor device |
KR102671333B1 (en) * | 2024-01-03 | 2024-06-04 | 주식회사 코본테크 | DC current sensor circuit that eliminates error caused by offset voltage |
Also Published As
Publication number | Publication date |
---|---|
AU2001214613A1 (en) | 2002-05-15 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US6522517B1 (en) | Method and apparatus for controlling the magnetization of current transformers and other magnetic bodies | |
US6160697A (en) | Method and apparatus for magnetizing and demagnetizing current transformers and magnetic bodies | |
US7242157B1 (en) | Switched-voltage control of the magnetization of current transforms and other magnetic bodies | |
USRE49449E1 (en) | Charge pump with temporally-varying adiabaticity | |
US9188610B1 (en) | Apparatus for measuring multiple electric currents utilizing multiple current transformers | |
US6479976B1 (en) | Method and apparatus for accurate measurement of pulsed electric currents utilizing ordinary current transformers | |
CA2670907C (en) | Magnetizing inrush current suppression device and method for transformer | |
EP2710389B1 (en) | Apparatus and a method for testing winding resistances of transformers | |
US6018700A (en) | Self-powered current monitor | |
De Leon et al. | Elimination of residual flux in transformers by the application of an alternating polarity DC voltage source | |
US6181586B1 (en) | Current-to-voltage converter and associate closed-loop control circuit | |
WO2002037512A1 (en) | Method and apparatus for controlling the magnetization of current transformers and other magnetic bodies | |
Zhang et al. | Improved flux-controlled VFCV strategy for eliminating and measuring the residual flux of three-phase transformers | |
Vulin et al. | Determining the residual magnetic flux value of a single-phase transformer using a minor hysteresis loop | |
JPH04231026A (en) | Gradient speeding up circuit for nmr apparatus | |
CZ2014827A3 (en) | Device for testing control transformers | |
de Castro Cezar et al. | Elimination of inrush current using a new prefluxing method. Application to a single-phase transformer | |
JPH0630579A (en) | Current detecting circuit | |
US5642025A (en) | Process and apparatus for the reduction of surges in switch-on current in the operation of a rotary current load | |
US5469321A (en) | Magnetizing device having variable charge storage network and voltage control | |
Gukov et al. | Reduction of magnetizing current of power transformer due to optimal location of concentric windings | |
JP2006223083A (en) | Device to control magnetic flux of transformer in each iron core using direct current | |
Bhadkamkar et al. | Application of zig-zag transformers in a three-wire three-phase dynamic sag corrector system | |
RU2422839C1 (en) | Method of determining induction motor stator winding phase dissipation and device to this end | |
Negara et al. | Demagnetization method for reducing inrush current of single phase 1 kVA transformer |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
REG | Reference to national code |
Ref country code: DE Ref legal event code: 8642 |
|
122 | Ep: pct application non-entry in european phase |