METHOD AND APPARATUS FOR THE MEASUREMENT OF ELECTRICAL CURRENT BY MEANS OF A SELF-COMPENSATING CONFIGURATION OF MAGNETIC FIELD SENSORS

Abstract

Two magnetic field sensors, ratiometric with respect to their common supply and featuring matched thermal coefficients, are inserted in the two airgaps of a magnetic circuit arranged so that said airgaps appear in series with respect to the magnetic flux generated by the current to be measured, while appearing in parallel with respect to the reference flux generated by a stable permanent magnet. The output signal of one of the sensors is thus proportional to the sum of said fluxes, the other to their difference. Adding and subtracting said signals produces two outputs, one proportional solely to the current to be measured, and the other solely to the reference flux. A feedback loop acts on the common supply of the two sensors in order to hold constant the output proportional to the reference flux, thus producing the effect that drifts with temperature of the magnetic sensitivities are intrinsically compensated for.

Claims

1. A method of measuring electrical current by computing a function of the output signals of two magnetic field sensors (H1, H2) placed at the two airgaps of a magnetic circuit, both sensors belonging to the type ratiometric with respect to the bias current or both belonging to the type ratiometric with respect to the supply voltage and selected to feature matched temperature coefficients of the respective magnetic field sensitivity; wherein a magnetic flux is established in said magnetic circuit by a first winding carrying the current to be measured (I.sub.M), and thus resulting, at both magnetic field sensors locations, in a first magnetic field component (B.sub.M) proportional to said current; said method further comprising the step of establishing at said two airgap locations a second magnetic field component (B.sub.R) possessing a constant reference value independent from the value of the current to be measured, and in such a way that said second magnetic field component adds to said first component at one of said two locations, while subtracting from it at the other location; wherein said reference field component can be established by means of at least one permanent magnet or by means of at least one second winding carrying a current of a constant reference value; said method further comprising the step of processing through and adder circuit and a subtractor circuit the signals generated by said two magnetic field sensors, so as to produce an output sum signal and an output difference signal, and thus resulting in one of said output signals being proportional to the reference field component solely and the other output signal being proportional solely to the magnetic field component established by the current to be measured; whereby the output signal proportional to the reference field component is then used in a feedback loop arranged to control the common biasing current, in case the magnetic field sensors belong to the type ratiometric with respect to the biasing current, or the common supply voltage, in case the magnetic field sensors belong to the type ratiometric with respect to the supply voltage, of said magnetic field sensors in such a way that said output signal is held at a constant reference voltage value, consequently resulting in an automatic compensation of the impact of temperature drifts of the magnetic sensitivities on the accuracy of the output signal proportional to the current to be measured.

2. An apparatus for implementing the method as claimed in claim 1, comprising two identical magnetic circuits featuring one airgap each, the first winding carrying the current to be measured (I.sub.M) having the function to establish at both of said airgaps identical values of said first magnetic field component (B.sub.M) proportional to said current, and further comprising two identical auxiliary windings, each winding coupling to just one of said magnetic circuits, serially connected in order to carry the same constant reference current whose function is to establish in both of said airgaps the reference magnetic field component (B.sub.R), wherein said reference current flows through one of said auxiliary windings in the direction resulting in said reference field component adding to the first magnetic field component at the corresponding airgap, while flowing in the opposite direction through the other winding and thus resulting in the reference field component subtracting from the first magnetic field component at the other airgap; whereby said two magnetic field sensors (H1, H2) are placed at said two airgaps, and their common supply and their output signals are then processed according to the method.

3. An apparatus for implementing the method as claimed in claim 1, comprising a magnetic circuit so arranged that the two airgaps appears in series with respect to the magnetic flux generated by the first winding carrying the current to be measured (I.sub.M) and featuring a second winding wound around a parallel branch of said magnetic circuit, and wherein said parallel branch is positioned in such a way that said two airgaps appear in parallel with respect to the magnetic flux generated by said second winding, while the net magnetic flux fraction generated by the first winding and which closes through said parallel branch is null; wherein said second winding carries a constant reference current having the function to establish in both of said airgaps a reference magnetic field component (B.sub.R) which adds to the first magnetic field component at one of the airgaps, while subtracting from it at the other airgap; whereby said two magnetic field sensors (H1, H2) are placed at said two airgaps, and their common supply and their output signals are then processed according to the method.

4. An apparatus for implementing the method as claimed in claim 1, comprising a magnetic circuit so arranged that the two airgaps appears in series with respect to the magnetic flux generated by the winding carrying the current to be measured (I.sub.M), wherein said magnetic circuit features at least one parallel magnetic circuit branch excited by at least one permanent magnet and positioned in such a way that the two airgaps both appear in parallel with respect to the magnetic flux generated by said permanent magnet(s), so that one of the airgaps receives one half of said flux while the other airgap receives the other half; wherein said magnetic flux generated by said permanent magnet(s) has the function to establish in both of said airgaps a reference magnetic field component (B.sub.R) which adds to the first magnetic field component at one of the airgaps, while subtracting from it at the other airgap; whereby said two magnetic field sensors (H1, H2) are placed at said two airgaps and their common supply and their output signals are then processed according to the method.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0013] FIG. 1 depicts two gapped cores encircling a bar carrying the current to be measured, I.sub.M, both excited by the same reference current, I.sub.R, flowing through two series connected auxiliary windings. The magnetic field in the two airgaps is measured by the two Hall probes H1 and H2.

[0014] FIG. 2 depicts the two gapped cores of the configuration of FIG. 1 as seen from the opposite side.

[0015] FIG. 3 illustrates the equivalent electrical representation of the two gapped cores configuration of FIG. 1 and FIG. 2.

[0016] FIG. 4 presents the basic principle of operation of the feedback loop implementing the proposed self-compensating principle of operation.

[0017] FIG. 5 presents the symmetric implementation of the feedback loop of FIG. 4.

[0018] FIG. 6 illustrates a magnetic circuit configuration whereby only one auxiliary coil is sufficient.

[0019] FIG. 7 presents a magnetic circuit configuration whereby the central column and auxiliary coil visible in FIG. 6 have been replaced by a permanent magnet.

[0020] FIG. 8 depicts a configuration whereby the reference field in the two airgaps is obtained by means of two permanent magnets instead of one.

[0021] FIG. 9 presents the front and back perspective views of an advantageous variant embodiment of the configuration of figure FIG. 8.

[0022] FIG. 10 illustrates how, by adding an additional coil in a classical current transformer configuration, higher frequency components can also be measured.

DESCRIPTION

[0023] The object of the present invention is to provide a magnetically coupled electrical current measuring method arranged in order to obtain a self-compensating effect for drifts with temperature and aging of key electrical characteristics of the magnetic field sensors utilized.

[0024] FIG. 1 (a perspective view from the opposite side is shown in FIG. 2) and FIG. 3 depict a possible current sensor apparatus implementing the proposed self-compensating measurement method. Two magnetic field sensors, H1 and H2, are located in the airgaps of two identical magnetic circuits made of soft magnetic material (such as for example ferrite, silicon steel laminated stack, amorphous materials, etc.). These sort of magnetic circuits are known to the skilled in the art by the descriptive name of “gapped magnetic cores”. In the example of FIG. 1 the current to be measured, I.sub.M, is carried by the rectangular conductor (e.g. a bus bar) passing through both said gapped magnetic cores. Being the two cores identical, the value of the corresponding B fields established in the two airgaps, B(I.sub.M), will also be identical. Each core additionally carries an auxiliary winding. Said two identical windings are then electrically connected in series, in such a way that an accurate reference current, I.sub.R, flowing through both of them will result in an additional component of the B field, B(I.sub.R), summing to B(I.sub.M) in one of said airgaps, while subtracting from it in the other airgap, as exemplified in the electrically equivalent schematic diagram of FIG. 3. The basic principle of operation of the proposed self-compensating measurement method is then illustrated in FIG. 4.

[0025] H1 and H2 are two serially connected Hall effect probes, of respective magnetic sensitivities K.sub.M and K.sub.H2. For the sake of simplicity, we will now refer to I.sub.M and I.sub.R as to actually represent the respective N.sub.M I.sub.M and N.sub.R I.sub.R products (N.sub.M=number of turns of the main excitation winding, and N.sub.R=number of turns of one auxiliary winding). A1 and A2 represent signal conditioning electronics serving the purpose of pre-amplifying, with respective gains G1 and G2, the differential output voltage generated by H1 and H2, while providing compensation of the Hall probes respective zero field offset voltage values. Effective offset voltage compensation techniques are well known to the skilled in the art, including: dynamic offset cancellation, bias current spinning, and a technique whereby two Hall effect probes with matched offsets are mated back to back and electrically connected in such a way that when summing their output signals the respective offsets subtracts from each other.

[0026] Thus, the signals at the output of A1 and A2 become:

V1=G1K.sub.CI.sub.B(I.sub.M+I.sub.R)=K(I.sub.M+I.sub.R)  (4)

V2=G2K.sub.CK.sub.H2I.sub.B(I.sub.M−I.sub.R)=K(I.sub.M−I.sub.R)  (5)

whereby the identity G1 K.sub.C K.sub.H1 I.sub.B=G2 K.sub.C K.sub.H2 I.sub.B≡K can be obtained after adjusting the individual gains G1 and G2 for possibly different magnetic sensitivity K.sub.H values, in order that G1 K.sub.H1=G2 K.sub.H2 (K.sub.C and I.sub.B then have the same values in both expressions because of the identical cores and of the series connection). It is straightforward to verify that the voltage at the midpoint of the resistive divider (representing a very simple way to implement an accurate adder) is proportional to the current to be measured:

K(I.sub.M+I.sub.R+I.sub.M−I.sub.R)/2=KI.sub.M  (6)

while the voltage at the output of the subtractor is proportional to the reference current:

K(I.sub.M+I.sub.R−I.sub.M+I.sub.R)=2KI.sub.R  (7)

[0027] FIG. 5 depicts the equivalent self-compensating feedback loop configuration in case one of the Hall effect sensor is mounted the other way around with respect to the direction of B(I.sub.M), so that the signal proportional to I.sub.R is now available at the output of the adder instead than at the output of the subtractor. In other words, depending on the relative mounting orientation of the magnetic field sensors, the voltage proportional to I.sub.R will be available at the output of the adder or at the output of the subtractor. Either ways, the feedback loop is arranged in order to maintain at a constant reference voltage V.sub.ref the signal proportional to I.sub.R (i.e. the signal named K I.sub.R in FIG. 4 and FIG. 5). The operational amplifier integrator configuration visible in FIG. 4 and FIG. 5 schematically represents, in a simplified way, a conventional Proportional-Integral-Derivative (PID) circuital block. For most applications it will simply be sufficient to implement the Proportional-Integral block only. However, depending on the specific dynamical response requirements, the skilled in the art can then decide whether or not to implement also the Derivative block. For magnetic field sensors of the type ratiometric with respect to the supply current, such as Hall effect probes, the following example will aid in understanding the details of the principle of operation: [0028] the two Hall probes are selected in order to be matched in their respective drifts with temperature of the magnetic sensitivity, i.e. in order to feature very similar values of the temperature coefficient of the magnetic sensitivity; [0029] the reference current, I.sub.R, flowing through the auxiliary winding is generated by means of any of the circuital solutions known to the skilled in the art for being able to provide a very accurate and constant current value; [0030] would the magnetic sensitivity of both Hall probes drift with temperature by for example +10%, then the feedback loop will automatically react decreasing by 10% the value of I.sub.B, so as to hold at a constant value the product K I.sub.R, and thus K (because I.sub.R is already held at a constant value by independent means); [0031] in this way the output signal proportional to I.sub.M has now become largely independent from drifts with temperature of the magnetic sensitivity of the two Hall probes.

[0032] It will be apparent to those skilled in the art how to modify FIG. 4 and FIG. 5 in order to suit magnetic field sensors belonging to the type ratiometric with respect to the supply voltage, namely, by connecting said sensors supply inputs in parallel instead than in series.

[0033] Typically, sensors from the same manufacturing wafer lot would feature very similar behavior in terms of magnetic field sensitivity drift with temperature and aging, thus providing a sufficient degree of matching for an accurate current sensor.

[0034] A preferred embodiment of apparatus implementing the self-compensating method just described is illustrated in FIG. 1, FIG. 2, and FIG. 3. The fact that the two auxiliary winding carrying the reference current I.sub.R are serially connected in such a way that B(I.sub.R) adds to B(I.sub.M) in one of the cores, while subtracting from it in the other core, conveniently implies that only purely D.C. performances are required from the current source generating it, as any higher frequency component (of the current to be measured I.sub.M) would induce, in said two auxiliary windings, voltages of equal amplitude but opposite signs, and thus cancelling out.

[0035] FIG. 6 illustrates a second preferred embodiment of apparatus in accordance with the invention, whereby the magnetic circuit is so arranged that the two airgaps appear in series with respect to the magnetic flux generated by the winding carrying I.sub.M, while appearing in parallel with respect to the flux generated by the auxiliary winding carrying I.sub.R. Furthermore, the turns of the main excitation winding carrying I.sub.M are distributed along the magnetic circuit in order that the net generated flux closing through the parallel magnetic circuit branch carrying the auxiliary winding is zero (it is easy to verify that, through the central column, the turns at the right generate a magnetic flux which exactly subtracts from the flux generated by the turns at the left). Thus, also this embodiment is characterized in that it requires only purely D.C. performances from the current source generating I.sub.R, greatly simplifying its design. The reference B(I.sub.R)≡B.sub.R field then adds to B(I.sub.M)≡B.sub.M in one of the two airgaps, while subtracting from it in the other. Granted that the reference B.sub.R field needs to be a D.C. field only, it is just natural to try to replace said auxiliary winding with a permanent magnet. In fact, there exist special formulations of Samarium Cobalt (SmCo) permanent magnet materials featuring excellent temperature stability performances, with Grade 2:17TC-16 achieving −0.001%/° C.

[0036] Thus, in a third preferred embodiment of apparatus in accordance with the invention, schematically represented in FIG. 7, the central column and auxiliary winding of FIG. 6 has been replaced by the permanent magnet PM. In one of the two airgaps said permanent magnet establishes a reference B.sub.R field adding to B.sub.M, while subtracting from it in the other airgap, and thus allowing the application of the self-compensating feedback loop method described heretofore.

[0037] In another advantageous embodiment of apparatus in accordance with the invention, the reference B.sub.R field at the two airgaps is established by the magnetic fluxes, Φ.sub.1 and Φ.sub.2, generated by a pair of permanent magnets arranged with their respective North and South poles oriented as illustrated in FIG. 8. In this way about half of Φ.sub.1 will add to about half Φ.sub.2 in both airgaps. Thus, the resulting reference B.sub.R field will add to B.sub.M in one of the airgaps, while subtracting from it in the other, in turn allowing the application of the self-compensating feedback loop described heretofore. It shall be remarked that a certain fraction of the flux Φ.sub.M established by the winding carrying the current to be measured I.sub.M, will also interest both PM1 and PM2, slightly shifting their respective working points. However, because the result would be to slightly increase Φ.sub.1 (or Φ.sub.2), but while symmetrically decreasing the Φ.sub.2 (or 101 .sub.1), the negative impact on the constancy of B.sub.R (i.e. its independence from I.sub.M) is significantly reduced (as in each airgap Φ.sub.R=Φ.sub.1/2+Φ.sub.2/2).

[0038] FIG. 9 illustrates the front and back perspective views of an advantageous variant embodiment of the configuration of FIG. 8, whereby a plurality of pairs of permanent magnets are conveniently located as to directly surround the two airgaps.

[0039] Similarly to the technique disclosed in U.S. Pat. No. 4,961,049, for electrical current measurements requiring large bandwidth a further winding can be added, as illustrated in FIG. 10, mutually coupling with the winding carrying the current to be measured I.sub.M according to a conventional current transformer configuration. The voltage across the shunt resistor R, proportional to the higher frequency components of I.sub.M, can then be added to the voltage proportional to the lower frequency components produced by the self-compensating feedback loop of the method described heretofore.

[0040] As can be seen from the above, the invention is not limited to the embodiments and the applications which have been described in detail. On the contrary, the invention extends to any variant which may occur to the person skilled in the art without going beyond its scope.