US20140300349A1

US20140300349A1 - Coreless current probe and a method of measuring direct current

Info

Publication number: US20140300349A1
Application number: US13/795,827
Authority: US
Inventors: Ian James Walker; Karron Louis Law
Original assignee: GMW Associates Inc
Current assignee: GMW Associates Inc
Priority date: 2013-03-12
Filing date: 2013-03-12
Publication date: 2014-10-09
Also published as: EP2972425B8; EP2972425B1; US9952257B2; EP2972425A4; ES2871854T3; EP2972425A1; US20180231588A1; US20160033554A1; US10690701B2; WO2014164551A1

Abstract

A coreless current probe has a U-shaped body with arms, an open end and a cross piece forming a closed end. The opening formed in the U-shaped body has a length not less than the width between the arms. The U-shaped body can engage around a conductor carrying a current to be measured. The U-shaped body has a number of coreless single point magnetic field sensors, usually Hall devices, distributed around the opening. An arrangement with six sensors has two sensors at the ends of the arms of the U-shaped body, two sensors at the closed end, and two sensors mid-way along the arms. Sensing circuitry applies factors to the outputs of the sensors and sums the results to produce a measured current value. The factors are selected so that the summed result is zero in any externally generated homogeneous magnetic field.

Description

BACKGROUND

1. Field of the Invention
The present invention relates to current probes and methods of measuring direct current. In particular, the invention is concerned with coreless current probes, which do not contain a core or cores of material with relatively high magnetic permeability.
2. Background of the Invention
Probes and methods for measuring current flowing in a conductor are known which do not require any electrical contact to be made with the conductor. For measuring alternating currents in a conductor, contactless current probes typically provide a core of magnetic material completely embracing the conductor, in combination with a sense winding on the core, to sense alternating magnetic field generated in the core. Such alternating current probes are known as current transformers and a magnetic core completely surrounding the conductor carrying the current to be measured is desirable to ensure good flux linkage between the primary “winding” which is the conductor carrying the current to be measured, and the secondary winding which is the sense coil. It is known also to provide current transformer type current probes in which the magnetic core is in two parts, enabling the probe to be clamped around the conductor carrying the current to be measured. Further, it is known to use a Hall device to sense magnetic field in a small gap in the core surrounding the conductor.
AC current sensing is also known using an air core coil surrounding the conductor carrying the current to be measured. In particular, U.S. Pat. No. 5,057,769—Edwards discloses a C-shaped sensing coil mounted on a skeleton board to enable the coil to be located embracing a conductor between the arms of the C. Compensating coils are provided at the ends of the main C-shaped coil to provide some compensation for the effect of the gap in the main sensing coil.
Generally, use of current transformer type current probes with cores of magnetic material is unsuitable in regions of very high magnetic fields which may cause saturation of the magnetic core. Furthermore, inductively linked current sensing devices are not suitable for measuring DC current. Sensors are known which can measure the magnetic field intensity at a single point. Examples of such sensors include MEMS sensors, various kinds of magnetometer, and in particular Hall effect sensors. According to Ampere's Law, the line integral of magnetic fields around a closed loop is proportional to the total current embraced by the loop. This simple expression of the law is true in magneto static situations, when there is no time varying charge density or electro magnetic propagation. U.S. Pat. No. 4,625,166—Steingroever et al. discloses a DC current sensor formed as a ring of Hall devices surrounding the current conductor. The sum of the outputs of the ring of Hall devices provides an approximation to the line integral of magnetic field around the conductor being measured, so that a value for the current in the conductor is determined.
U.S. Pat. No. 7,321,226—Yakymyshyn et al. discloses a current sensor employing a ring of Hall devices mounted in hinged housings to enable the probe to be clipped around the conductor carrying the current to be measured. Again, by providing multiple Hall devices in a ring completely surrounding the conductor, the sum of the outputs of the Hall devices can provide good approximation to Ampere's Law, thereby providing a good measurement of current in the conductor.
Measuring current in a conductor using multiple coreless single point magnetic field sensors, such as Hall devices, presents problems when it is not possible to obtain access completely around the conductor in which the current is to be measured. U.S. Pat. No. 7,445,696—You et al. discloses a device for measuring electric current in a conductor, where the conductor is a bus bar feeding current to and from the electrodes of the electro-chemical cells in an electro-metallurgical system. Such electro-metallurgical systems include electro-refining and electrowinning systems for copper, zinc, and other metals. Although it may be desirable to monitor the current flowing in a single bus bar feeding a single electrode of such an electro-metallurgical system, the physical arrangement of such systems means that it is not practicable to obtain access for a current sensing probe completely around the bus bar. Furthermore, the presence of multiple current carrying bus bars in close proximity leads to relatively high magnetic fields in the vicinity of each bus bar, including high levels of external magnetic field which is not produced by a current to be measured flowing in a target bus bar. The patent to You et al. describes using multiple Hall effect sensors mounted immediately above the bus bar being monitored. A proximity sensor is also provided on the probe to ensure the probe is in close contact with the top of the bus bar being monitored.

BRIEF SUMMARY OF THE INVENTION

In one aspect, the invention provides a coreless current probe comprising a U-shaped body having arms, an open end and a cross piece forming a closed end opposite to the open end. The U-shaped body defines an opening having a width between the arms and a length from the closed end to the open end. The length is not less than the width and the width and the length define a plane of the opening as well as a central line normal to the plane. A conductor carrying a current to be measured can be engaged by the body of the probe so as to extend through the opening parallel to the central line. A plurality of coreless single point magnetic field sensors are distributed in the body around the opening. A respective one of the sensors is located at an end of each of the arms of the U-shaped body, so as to be on each side of the open end of the body. At least one further of the sensors is located at the closed end of the body. The sensors each have a respective axis of magnetic field sensitivity and are arranged in the body so that each of these axes is not co-planar with the central line. Sensing circuitry is connected to the sensors and is operative to produce for each of the sensors a respective sensor signal which is a measure of the angle component of magnetic field at the sensor aligned with its respective axis of magnetic field sensitivity. The sensors are arranged in the body in such a way that there can be found values of c_rfor which, in any homogeneous magnetic field,
$\sum_{r = 1}^{n} c_{r} h_{r} = 0,$
where n is the number of the magnetic field sensors, h_ris the measure of the magnetic field component for the r^thsensor, and c_ris a constant factor for the r^thsensor. The sensing circuitry is operative to combine the measures h_rto produce a measured current value representing current flowing in a conductor engaged by the U-shaped body of the probe.
The invention further provides a method of measuring direct current flowing in a conductor, where the conductor has minimum and maximum orthogonal cross-sectional dimensions. In the method, an Ampere's Law integration path is defined around the conductor in an integration plane perpendicular to a central line of the conductor, where the path has minimum and maximum orthogonal dimensions which exceed the dimensions of the conductor. A respective angle component of magnetic field is measured at each of a plurality of locations on this integration path. An adjacent pair of these locations is at one end of the maximum dimension of the path and spaced apart by the minimum dimension of the path. The third of these locations is at the other end of the maximum dimension of the path. The locations and the orientations of the respective angle components of magnetic field being measured are selected such that values can be found of c_rfor which, in any homogeneous magnetic field,
$\sum_{r = 1}^{n} c_{r} h_{r} = 0,$
where n is the number of the above referred locations, h_ris the measured value of the magnetic field component at the r^thlocation and c_ris a constant factor for the r^thlocation. In the method, the measured values h_rare combined to produce a value of the direct current being measured.

BRIEF DESCRIPTION OF THE DRAWINGS

Embodiments of the invention will be described below with reference to the following figures.

FIG. 1 is a schematic representation of an example of a coreless current probe embodying the present invention, using three magnetic field sensors.

FIG. 2 is a view of the current probe of FIG. 1 taken from one side and showing sensing circuitry housed in the probe.

FIG. 3 illustrates further examples of current probe embodying the present invention, using four magnetic field sensors, and using six magnetic field sensors.

FIG. 4 illustrates an electrowinning tank with cathodes and anodes, illustrating a preferred use of the current probe.

FIG. 5 is a perspective view of a current probe engaged on a bus bar whose current is to be measured.

FIG. 6 illustrates further examples of current probe embodying the present invention.

FIG. 7 is a hand-held current probe embodying the present invention.

FIG. 8 is a wireless enabled current probe embodying the present invention.

FIG. 9 is a view of the current probe of FIG. 8 taken from one side and showing sensing circuitry and wireless circuitry in the probe.

FIG. 10 is a schematic circuit diagram illustrating the sensing circuitry and wireless circuitry which may be incorporated in a wireless enabled current probe as illustrated in FIGS. 8 and 9.

FIG. 11 is a timing diagram for the circuit of FIG. 10.

DETAILED DESCRIPTION OF THE EMBODIMENTS

FIGS. 1 and 2 illustrate a basic embodiment of the invention. A coreless current probe is shown having a U-shaped body 10 with arms 11 and 12, an open end 13 and a cross piece 14 forming a closed end 15 which is opposite to the open end 13. The U-shaped body 10 defines an opening 16 having a width d between the arms 11 and 12, and a length l from the closed end 15 to the open end 13. As illustrated in FIG. 1 the length l is not less than the width d and in the example illustrated is in fact substantially greater than the width d. The width and length dimensions of the opening 16 on the U-shaped body 10 define a plane of the opening which is substantially the plane of the paper in FIG. 1, and is a plane perpendicular to the paper along the line 17 in FIG. 2. The opening 16 further defines a central line 18, represented in FIG. 1 by a dot 18 which is normal to the plane of the opening and located substantially centrally in the opening.
In use, the U-shaped body 10 of the current probe illustrated in FIG. 1 can be engaged around a conductor 19 carrying a current to be measured. When engaged by the U-shaped body of the probe, the conductor 19 extends through the opening parallel to the central line 18. The conductor 19 is shown in cross-section in FIG. 1 within a dashed outline. A short section of the conductor 19 is illustrated in FIG. 2 extending through the opening 16.
In FIGS. 1 and 2, a plurality of coreless single point magnetic field sensors are distributed in the U-shaped body 10 around the opening 16. In the example of FIGS. 1 and 2, three such single point magnetic field sensors are illustrated at 20, 21 and 22. In one example, the single point magnetic field sensors 20, 21 and 22 comprise Hall effect sensors. However, any type of sensor may be used which is capable of producing an electrical signal representing the intensity of magnetic field at the location of the sensor. The magnetic field sensors employed in the example of the invention are vector magnetometers, in the sense that each of the sensors has a respective axis of magnetic field sensitivity. Apart from Hall effect devices, alternative magnetic field sensors include magneto resistive devices.
In the example illustrated, a respective magnetic field sensor 21, 22 is located at an end of each of the arms 11 and 12 of the U-shaped body 10, at the open end 13 of the body. A third sensor 20 is located at the closed end 15 of the body. In the particular example, sensor 20 is located mid-way along the cross piece 14 at the closed end 15 of the body. The magnetic field sensors 20, 21 and 22 are intended to detect the magnetic field generated by current flowing in the conductor 19 in the direction of the conductor, that is to say parallel to the central line 18. Accordingly, it is important that the axis of magnetic field sensitivity of each of the sensors 20, 21 and 22 is not aligned in a plane containing the central line 18. Otherwise the sensors would have minimum sensitivity to any magnetic fields generated by current flowing in the conductor parallel to the central line 18.
Generally, the sensors 20, 21 and 22 are intended to detect magnetic field generated by the current flowing along the conductor 19 and should preferably be arranged in the body 10 so as to maximize detection of this magnetic field, whilst having a minimal response to external magnetic fields which are not produced by currents in the conductor 19. To this end, the location and orientation of the sensors 20, 21 and 22 in the U-shaped body 10 are selected so that it is possible to derive a signal combining the outputs of these sensors, which is insensitive at least to any homogeneous magnetic field in the region of the probe. Such a homogeneous magnetic field is a field which would be generated externally of the probe, so that the field lines are substantially linear with a uniform flux density everywhere over the probe.
Each of the magnetic field sensors 20, 21 and 22 is connected with sensing circuitry, which is shown by the box 25 located in a housing 26 connected to cross piece 14 of the U-shaped body 10 by a neck piece 27. The sensing circuitry 25 operates to produce for each of the sensors 20, 21 and 22 a respective sensor signal which is a measure of the angle component of magnetic field at the respective sensor which is aligned with the axis of magnetic field sensitivity of the sensor.
In order for it to be possible for the sensors 20, 21 and 22 of the probe to reject a homogeneous external magnetic field, the sensors must be arranged in the U-shaped body 10 such that values c_rcan be found for which, in any homogeneous magnetic field,
$\sum_{r = 1}^{n} c_{r} h_{r} = 0,$
where n is the number of magnetic field sensors, h_ris the measure of the magnetic field component for the r^thsensor and c_ris a constant factor for the r^thfactor. So long as the sensors in a probe are arranged and orientated such that the values of c_rcan be obtained to satisfy the summation equation above, it is possible to derive a combined signal from the outputs of the sensors which will reject external homogeneous magnetic fields.
In the example illustrated in FIG. 1, the sensor 20 at the centre of the cross piece 14 is illustrated with its axis of sensitivity 30 directed in the plane of the opening 16 of the U-shaped body 10, and normal to a central plane 33 which contains the central line 18 and is mid-way between the arms 11 and 12 of the U-shaped body 10. The field sensor 21 is shown with its axis of sensitivity 31 also in the plane of the opening 16 and at an angle θ to a transverse line 34 which is normal to the central plane 33. The sensor 22 is shown with its axis of sensitivity 32 again in the plane of the opening and at an angle φ to the transverse line 34. If the sensors 20, 21 and 22 have the same nominal sensitivity to magnetic field intensity aligned with a respective axis of the sensitivity of the sensors, it can be shown that for a uniform homogeneous horizontal magnetic field in the plane of the opening 16 and parallel to the transverse line 34, the summation
$\sum_{r = 1}^{3} c_{r} h_{r} = 0,$
is true if c₁=c₂Cos θ+c₃Cos φ. Similarly, in order for the above summation to be zero in the presence of a vertical homogeneous magnetic field in the plane of the opening 16 and normal to the transverse line 34, c₂Sin θ=c₃Sin φ. Accordingly, in the general case illustrated in FIG. 1, values of c₁, c₂and c₃can be identified for which the summation is zero in any homogeneous magnetic field.
In practice, it is convenient to orientate the sensors 21 and 22 so that θ=φ=45°, whereupon the summation is zero if c₁=✓2*c₂=✓2*c₃.
In the probe illustrated in FIGS. 1 and 2, the sensing circuitry 25 is arranged to be operative to combine the signals from the sensors 20, 21 and 22, which comprise measures h₁, h₂and h₃of the angle component of magnetic field at the respective sensors, to produce a measured current value representing current flowing in the conductor 19 engaged by the U-shaped body 10 of the probe. In order fully to reject external homogeneous magnetic fields, the sensing circuitry 25 will combine the measures h_rby performing the summation
$\sum_{r = 1}^{3} c_{r} h_{r},$
using the values of c_ras calculated above for rejecting homogeneous external fields.
FIGS. 1 and 2 illustrate a basic example of the invention employing only three magnetic field sensors 20, 21 and 22. In order for the probe accurately to measure the current flowing in conductor 19 engaged by the probe, in the presence of relatively strong external magnetic fields, more than three magnetic field sensors will normally be required in the U-shaped body 10 of the probe. Referring to FIG. 3, improved performance may be obtained using four magnetic field sensors distributed symmetrically in the U-shaped body 10 about the central plane 33, which constitutes a plane of symmetry. The sensors 21 and 22 at the ends of the arms 11 and 12 on each side of the opening end 13 of the U-shaped body 10 are provided, as in the example of FIGS. 1 and 2, with their axes of magnetic sensitivity aligned at the same angle to transverse line 34. The single sensor 20 at the centre of the cross piece 14 is replaced, in the example of FIG. 3, with a pair of sensors 41 and 42 mirroring the sensors 21 and 22. As can be seen, in each case the axes of magnetic sensitivity 31 and 32 of the sensors 21 and 22, and 51 and 52 of the sensors 41 and 42 are each at a respective acute angle to the plane of symmetry 33, so that each of the axes 21, 22, 41 and 52 is generally tangential to the opening 16 between the arms 11 and 12 of the U-shaped body 10. In this way, each of the axes of sensitivity 31, 32, 51 and 52 is generally aligned with the direction of the magnetic field that will be generated by a current flowing in the conductor 19 engaged by the probe. If the angles of the axes of sensitivity 31, 32, 51 and 52, with respect to lines normal to the plane of symmetry 33, are all the same, then the factors c₁, c₂, c₃and c₄applied by the sensing circuitry 25 to the measures h₁, h₂, h₃and h₄from the four sensors should again be the same, in order to reject homogeneous external fields.
FIG. 3 also illustrates a further preferred arrangement which uses six sensors distributed around the U-shaped body 10 of the probe. In this embodiment, an additional pair of sensors 43 and 44 is located in the U-shaped body 10 substantially midway along the arms 11 and 12. Accordingly, the six sensors shown in FIG. 3 comprise a first pair of sensors 21 and 22 at the ends of the arms of the U-shaped body on either side of the open end 13, a second pair 41 and 42 at the closed end of the U-shaped body 10, and a third pair comprising the additional sensors 43 and 44 midway along the arms 11 and 12. The sensors of each of the first, second and third pairs are disposed spaced uniformly apart symmetrically on opposite sides of the plane of symmetry 33 which contains the central line 18 and is equally spaced between the arms. Each of the sensors 21, 22, 41 and 42 of the first and second pairs is orientated in the body so that its axis of sensitivity is in the plane of the opening and at a respective acute angle to the plane of symmetry 33, so as to be generally tangential to the opening. Each of the sensors 43 and 44 of the third pair is orientated in the body to have its axis of sensitivity in the plane of the opening and parallel to the plane of symmetry 33. The axes of sensitivity of the additional third pair of sensors 43 and 44 are illustrated by the lines and arrows 53 and 54 respectively.
Although the U-shaped body of the probe illustrated in FIG. 3 is generally similar to that illustrated in FIGS. 1 and 2, the aspect ratio of the U-shaped body in FIG. 3 is somewhat greater, in that the arms 11 and 12 of the U-shaped body 10 are more than twice as long as the spacing between the arms, so that l>2 d.
If the six magnetic field sensors are numbered in order in a clockwise direction around the opening 16, starting with sensor 42 as number 1, then the components of the summation discussed above are c₁h₁for sensor 42, c₂h₂for sensor 44, c₃h₃for sensor 22, c₄h₄for sensor 21, c₅h₅for sensor 43 and c₆h₆for sensor 41. Again assuming that the sensitivity to magnetic field of each of the six sensors is the same, so that the measure h for each sensor would be the same if the sensor is in an identical magnetic field aligned with the respective axis of sensitivity of the sensor, then the requirement that the summation
$\sum_{r = 1}^{6} c_{r} h_{r} = 0,$
in order to reject homogeneous external fields is met so long as c₁=c₃=c₄=c₆and c₂=c₅. Note however, that there is no constraint on the relationship between the value of c for sensors 44 and 43 at the mid-points of the arms 12 and 11 (of c₂and c₅) and the value of c for the sensors 42, 22, 21 and 41 at the corners of the opening 16 (c₁, c₃, c₄, c₆). In order to improve the performance of the probe illustrated in FIG. 3 with six magnetic field sensors, the sensing circuitry is adapted to derive the summation
$\sum_{r = 1}^{6} c_{r} h_{r},$
using values of c_rwhich are selected to maximize rejection by the probe of external magnetic fields which are not produced by currents in the conductor probe 19 engaged by the probe. Accordingly, in this example, the common value of c₂and c₅is selected accordingly to be different from the common value of c₁, c₃, c₄and c₆in order to maximize rejection of external magnetic fields.
It should be understood that the current probe described above with a U-shaped body carrying plural magnetic field sensors can be used to measure the current flowing in any conductor engaged between the arms 11 and 16 of the U-shaped body. Factors c_rcan be applied to the signals from the magnetic field sensors to produce a combined measurement
$\sum_{r = 1}^{n} c_{r} h_{r}$
representing current flowing in the conductor and rejecting the effect of external fields. In order to reject external fields successfully, the summation above should, as far as possible, approximate to the line integral along a closed loop around the conductor carrying the current to be measured. In order to approximate to the line integral value most accurately, and to reject external fields most successfully, it is desirable that the magnetic sensors in the U-shaped body are located along a closed loop path which has the shortest possible length surrounding the conductor to be measured. Accordingly, best results are obtained if the opening 16 of the U-shaped body 10 is sized so as closely to fit around a conductor 19 carrying the current to be measured.
An example of conductor 19 as illustrated in FIG. 3 has a substantially rectangular cross-section with a minimum orthogonal dimension which is only slightly less than the width d between the inner faces of the arms 11 and 12 of the U-shaped body. Also, the maximum transverse cross-sectional dimension of the conductor 19 is only slightly shorter than the length l between the open and closed ends 13 and 15 of the U-shaped body 10. Then, a dashed line 60 represents a minimum length closed loop encircling the conductor 19, and magnetic sensors 21, 22, 41, 42 and 43, 44 are shown, each with its point of sensitivity located on the line 60.
According to Ampere's Law, the line integral of magnetic fields along the minimum length loop 60 provides a measure of current flowing in the conductor 19. However, in the example of FIG. 3, actual magnetic field measurements are made only at six points around the minimum length loop 60. The sensors 21, 22, 41, 42 and 43, 44 are arranged with their axes of sensitivity generally aligned with the local direction of the magnetic field which would be produced by current flowing in the conductor 19, in the absence of any external fields. With this orientation of the magnetic field sensors, the sensitivity of the sensors to the magnetic field to be measured, in effect to the required signal, is maximized. In the absence of any external fields, the summation
$\sum_{r = 1}^{6} h_{r}$
would be proportional to the current flowing in the conductor 19. However, in order to provide discrimination between the magnetic field generated by current flowing in the conductor 19 and external fields, it is desirable to calculate the summation
$\sum_{r = 1}^{6} c_{r} h_{r},$
where the values of c_rare selected to provide a better approximation to the calculated line integral of magnetic field along the minimum length path 60.
A useful approach to determining appropriate values of c_ris to assign to each of the sensors in FIG. 3, a line segment along the minimum length loop 60 on either side of the sensor. For example, a line segment 61 may be assigned to sensor 41, extending from the mid-point 62 between sensor 41 and sensor 43 and the mid-point 63 between sensor 41 and sensor 42. Similarly, line segment 64 is assigned to sensor 42 extending from mid-point 63 to a mid-point 65 between sensor 42 and 44. A line segment 66 is assigned to sensor 44 extending from mid-point 65 to a mid-point 67 between sensor 44 and sensor 22. Line segment 68 is assigned to sensor 22 extending from the mid-point 67 to a mid-point 69 between sensor 22 and sensor 21. Line segment 70 is assigned to sensor 21 extending from the mid-point 69 to a mid-point 71 between sensor 21 and sensor 43. Line segment 72 is then assigned to sensor 43 extending from the mid-point 71 to the mid-point 62.
In order to determine values of c_rin the above summation which provide a better approximation to the line integral around the minimum length loop 60, a computer model is made of the magnetic field generated by current flowing along conductor 19, in the absence of any extraneous magnetic fields. It is then computationally straightforward to calculate the line integral of magnetic field along each of the line segments 64, 66, 68, 70, 72 and 61. These calculated line segments integrals are identified respectively as s_r, where r is 1-6. At the same time, it is also straightforward to identify in the computer model of the magnetic field the magnetic field intensity values h′_rwhich would be determined by the six magnetic field sensors.
In order to provide a line integral value of s′_rover a line segment length a_rin a magnetic field produced by current in the conductor 19 and in the absence of any extraneous field, the product h′_ra_rshould be multiplied by the factor s′_r/h′_ra_r. In a more general magnetic field comprising not only the magnetic field produced by current flowing in the conductor 19 but also external magnetic field, the line integral s_rover the line segment for the r^thsensor may be expressed
$s_{r} = (s_{r}^{'} / h_{r}^{'} a_{r}) * h_{r} * a_{r} = (s_{r}^{'} / h_{r}^{'}) * h_{r},$
where h_ris the measured field at the r^thsensor. It can be seen therefore that a more accurate approximation to the line integral of magnetic field around the minimum length loop 60 illustrated in FIG. 3, in a magnetic field comprising not only the field generated by current in the conductor 19, but also external magnetic field, is represented by the above referred summation
$\sum_{r = 1}^{6} c_{r} h_{r},$
where c_r=s′_r/h′_r. Since s′_rand h′_rcan be calculated in a computer model of the field generated by current flowing in the conductor 19, calculated values can be obtained for c_r.
In the example illustrated in FIG. 3, the minimum length loop 60 is shown to be slightly asymmetrical from top to bottom, since the loop has an apex point at 63 mid-way between the upper sensors 41 and 42. This loop shape corresponds to the cross-sectional shape of the conductor 19. Because of this, the computation of s′_ras outlined above for the line segments corresponding to sensors 41 and 42 will be slightly different to the computations of s′_rfor the line segments corresponding to the sensors 21 and 22 at the open end of the U-shaped probe.
Nevertheless, the values of c_rshould be symmetrical so that in a homogeneous external field only, in the absence of any current flowing through the conductor 19, the summation of c_rh_ris zero. Accordingly, in order to achieve this full rejection of any external homogeneous field, an average is taken of the calculated values c_rfor the four corner sensors 41, 42 and 21, 22, in order to provide identical values of c_rfor these sensors.
Full rejection of a uniform external field is important because any external magnetic field can be expanded into a uniform field plus a series of spatial harmonics. In most cases, the uniform field component of any external field has the largest contribution to the external field.
It can be seen from the above discussion of a procedure for calculating values c_rfor use in the summation
$\sum_{r = 1}^{n} c_{r} h_{r},$
that the lengths a_rof the line segments is somewhat arbitrary. In particular, it can be seen that the location of the mid-point 63 between upper sensors 41 and 42 on the U-shaped probe is determined by the requirements for symmetry, as is the position of the mid-point 69 between the sensors 21 and 22. Again for symmetry end points 65 and 67 should be equally spaced on opposite sides of sensor 44, and mid-points 62 and 71 should be equally spaced on opposite sides of sensor 43. However, there is no clear indication for the overall length of the segments 66 and 72 associated with the mid-point sensors 43 and 44. In practice, it can be seen that the overall length of the line segments associated with the mid-point sensors 43 and 44 may be extended to accommodate a region along the flanks of the conductor 19 over which the magnetic field produced by currents flowing in the conductor 19 extends generally parallel to the these flanks, at least at locations closely spaced to the flanks.
In practice, the current probe can be optimized for a particular installation by determining empirically the length of the line segments 66 and 72 associated with the mid-point sensors 43 and 44, which will maximize rejection of unwanted external magnetic fields.
The coreless current probe described above can have general application for measuring currents flowing in conductors, particularly where the physical construction and arrangement of the conductors to be measured does not permit the current probe to be wrapped entirely around the conductor. The current probe is also especially suited to arrangements where there may be high levels of external magnetic field, for example in circumstances where current is to be measured in a single conductor of an array of conductors carrying substantial currents. In particular the probe can be used to measure DC currents.
In large scale electro-chemical processing plants, particularly plants for electrowinning metals, a typical installation may comprise multiple tanks containing arrays of cathodes and anodes. For example, a single row of electrodes may comprise 50 cathodes and 51 anodes arranged alternating across the tank. FIG. 4 illustrates schematically part of an array of cathodes and anodes for an electrowinning installation comprising anodes 70 a, 70 b, 70 c, 70 d, 70 e, 70 f, alternating with cathodes 71 a, 71 b, 71 c, 71 d and 71 e etc. Although only six anodes and five cathodes are illustrated in FIG. 4, it should be understood that these will comprise just part of a much larger array for example comprising 51 anodes and 50 cathodes.
Each anode and cathode comprises a plate electrode extending normal to the page of the drawing of FIG. 4 into an electrolyte solution in the processing tank. The plates of the anodes and cathodes generally extend between the dotted lines 72 and 73 shown in FIG. 4. Anode bus bars 75 a, 75 b, 75 c, 75 d, 75 e, 75 f are provided supporting the anodes 70 a-f and are each connecting along the right hand side in FIG. 4 to an anode supply connector 76. Similarly, cathode bus bars 77 a-e respectively support cathodes 71 a-e and are each connected along the left hand side of FIG. 4 to a cathode supply connector 78.
In an electrowinning installation such as illustrated in FIG. 4, neighboring anode and cathode bus bars are relatively close together and may be separated by a spacing which is no greater than the width of each bus bar. It is not, therefore, possible to access each of the bus bars to clip a current probe completely around the bus bar. FIG. 5 illustrates in perspective view a typical anode bus bar 75, which is shaped with a generally rectilinear cross-section having a relatively high aspect ratio. This shape allows the bus bar to carry the high levels of current needed for an electrowinning process, typically in excess of 1000 amps for each anode or cathode bus bar, while minimizing the spacing between adjacent bus bars. FIG. 5 also illustrates a coreless current probe 80, of the kind described above with a U-shaped body, engaged on the bus bar 75. For best performance, the current probe 80 is dimensioned specifically for the particular bus bar 75, so that the width d between the arms 11 and 12 of the U-shaped body 10 is just sufficient to slide over the minimum cross-sectional dimension of the bus bar 75. The internal maximum dimension l of the U-shaped body is sized so that the full maximum cross-sectional dimension of the bus bar 75 is accommodated in the opening of the U-shaped body, for example as shown in the cross-section in FIG. 3.
A separate current probe 80 a-f may be located engaged with each of the anode bus bars 75 a-f, as illustrated in FIG. 4.
As mentioned previously, each of the anode bus bars 75 a-f in an electrowinning installation may carry a current in excess of 1000 amps. Similar currents will be carried by the cathode bus bars 70 a-e. It can be seen, therefore, that each of the current probes 80 a-f will be in a region of substantial magnetic field in addition to magnetic field generated by current flowing in the respective bus bar 75 a-f, that is the current to be measured in each case. Furthermore, the nature of the external magnetic field experienced by each of the current probes 80 a-80 f will be different depending on the location of the probe across the array of bus bars. Nevertheless, it has been found that a current probe with a U-shaped body and six sensors distributed as illustrated in FIG. 3 can measure the current flowing in a respective bus bar to an accuracy of better than about 1%, at any position across an array of bus bars comprising for example 51 anodes and 50 cathodes.
Some improvement in the rejection of magnetic fields is obtained by increasing the number of sensors located in the U-shaped body 10 of the probe, along the minimum length loop 60 as defined previously. FIG. 6 represents the U-shaped body 10 of a probe containing additional magnetic field sensors. In FIG. 6, features common to the probe of FIG. 3 are given the same numerals.
In FIG. 6, three further magnetic field sensors are provided, including one sensor 81 located substantially mid-way along the cross piece 14 of the U-shaped body 10, effectively at the location of the mid-point 63 identified in the arrangement with six sensors shown in FIG. 3. Also, there is a further sensor 82 and 83 located at the end of each of the arms 11 and 12 of the U-shaped body on opposite sides of the open end 13 of the body. Each of the three further sensors 81, 82 and 83 is orientated to have its axis of sensitivity in the plane of the opening and also substantially normal to the plane of symmetry 33. As shown in FIG. 6, the direction of sensitivity for the further sensor 81 is indicated by the arrow 84 and opposes the directions of sensitivity of the sensors 82 and 83, as shown by the arrows 85 and 86 respectively.
In order to obtain the requirement for the summation
$\sum_{r = 1}^{n} c_{r} h_{r} = 0$
for uniform external fields, either the sensitivity of the sensors 82 and 83 are set to be half the sensitivity of sensor 81, or the factor c_rfor the sensors 82 and 83 is set to be half the factor c_rfor sensor 81.
Within these constraints, the factors c_rfor the three further sensors can be different from the common factor c_rfor the corner sensors 41 42, 21 22 and also from the factor for the mid-point sensors 43 and 44. Values for the factors c_rfor the nine sensors can be determined as before by performing line integral calculations over a predetermined line segment for each sensor in a model field corresponding to the field generated by current flowing in the conductor 19 and no external fields.
FIG. 6 also illustrates a probe with still further sensors to provide even greater accuracy and rejection of external magnetic fields. Thus, there may be a total of thirteen sensors in the U-shaped body 10 of the probe including four still further sensors 90, 91, 92 and 93, each located at a respective mid-point along an arm of the U-shaped body between an existing mid-point sensor 43 or 44 and a respective corner sensor 42, 22, 21 and 41. Again, these four still further sensors 90, 91, 92 and 93 are located on the minimum length loop 60 and orientated to have axes of sensitivity in the plane of the opening. Preferably, the axes of sensitivity of these still further sensors 90, 91, 92 and 93 are also orientated parallel to the plane of symmetry 33, so as to extend as shown by the arrows in FIG. 6 substantially along the minimum length loop 60.
In order to maintain the summation
$\sum_{r = 1}^{n} c_{r} h_{r} = 0$
for uniform external fields, the values of c_rfor each of the still further sensors 90, 91, 92 and 93 should be the same, assuming each sensor has the same sensitivity. Again the common value of c_rfor these four still further sensors may be selected relative to the values of c_rfor the mid-point sensors 43 and 44, and for the corner sensors 41, 42, 21 and 22, by performing the line integral calculations described previously.
FIG. 7 illustrates a practical hand carried current probe incorporating the U-shaped body with magnetic sensors distributed as described in the previous embodiments. In the Figure, the current probe instrument comprises the U-shaped body 10 which may be as illustrated in FIG. 1, 2, 3 or 6. The instrument includes a housing 100 which contains a measured current display 101 which is connected to sensing circuitry within the housing 100. The housing 100 with display 101 corresponds to the housing 26 illustrated in FIG. 2 including the display 101 connected to the sensing circuitry 25. A handle 102 is fixed to the housing 100 and a tube 103 connects the U-shaped body 10 to the housing 100. Connection cables can run inside the tube 103 to connect the sensors on the U-shaped body 10 to the sensing circuitry 25 within the housing 100. Generally, the instrument shown in FIG. 7 can be battery operated so the housing 100 includes a battery compartment which is not shown in the drawing.
The instrument can be operated by an operator holding the handle 102 and standing above an array of bus bars carrying currents to be measured. The operator locates the arms 11 and 12 over a bus bar to be monitored, slides the U-shaped body 10 down onto the bus bar and can then measure the current by pressing a button 104 on the handle 102 of the instrument. The sensing circuitry 25 is arranged to respond to pressing the button 104 by recording the output signals of the sensors in the U-shaped body, performing the summation
$\sum_{r = 1}^{n} c_{r} h_{r}$
as described previously and displaying the calculated current on the display 101. An indicator light 105 may be provided which is arranged to flash when the current has been taken and is recorded in a data logger contained in the sensing circuitry 25. The operator can then lift the U-shaped body 10 off the bus bar and engage the next bus bar to measure its current.
FIGS. 8 and 9 illustrate a further embodiment of the current probe. In this embodiment, a housing 110 is physically connected to U-shaped body 10. The housing 110 contains not only the sensing circuitry of the probe but also wireless signaling circuitry connected to the sensing circuitry for wireless signaling measured current values to a remote location. The housing additionally has a battery compartment 111 for a battery to power the sensing circuitry and the wireless signaling circuitry.
As shown in FIGS. 8 and 9, the probe is formed as a unitary structure incorporating the U-shaped body 10 and the housing 110 with the battery compartment 111. The housing and the battery compartment may be integral with the cross piece 14 of the U-shaped body.
The wireless signaling circuitry is illustrated in FIG. 9 by the box 112 shown connected to box 25 containing the sensing circuitry. The wireless signaling circuitry may be constituted by Wi-Fi circuitry using standard Wi-Fi protocols, so that the probe can be networked in a computer network.
The embodiments shown in FIGS. 8 and 9 permits a Wi-Fi enabled current probe as illustrated to be located on each of the anode bus bars of an electrowinning tank in an electrowinning insulation, that is to say there would be fifty-one such wireless enabled probes engaged with respective anode bus bars on a tank comprising fifty-one anodes and fifty cathodes. In a real installation there may be multiple rows of anodes and cathodes, for example eight rows each comprising fifty-one anodes and fifty cathodes. In order to monitor all the cathodes in the installation, this implies over four hundred individual Wi-Fi channels to be monitored. Ethernet gateway systems may be provided, each capable of monitoring a hundred Wi-Fi channels and providing these channels over an Ethernet connection to an Ethernet router, in turn connected to a computer system running the monitoring software. In this way a system can be devised enabling the currents in every one of the anodes of a substantial electrowinning installation to be monitored substantially in real time by a computer at a remote location. The monitoring computer itself may be connected to a further remote location by internet.
It will be understood by those experienced in the art of electrowinning, that the electrodes of an electrowinning tank must be removed regularly for processing and cleaning. In order to accommodate this, the wireless enabled current monitoring probes are made to be readily removable from the respective bus bars. As shown in FIG. 8, the arms 11 and 12 of the U-shaped body 10 of the probe have parallel internal faces providing a predetermined uniform spacing which is sized to accommodate a rectangular section bus bar engaged by the probe. At least one compression tab 115 is located on the internal face 113 of at least one of the arms. The tab 115 protrudes inwards from the internal face 113 and is adapted to be resiliently outwardly compressible on engagement with the bus bar. In the illustrated example, a second resilient tab 116 is provided on the opposite internal face 114. These resilient tabs 115 and 116 enable the probe to be slid over the bus bar, causing the tabs 115 and 116 to move outwardly when engaging the sides of the bus bar, so that when the probe is fully engaged over the bus bar, the tabs 115 and 116 apply a resilient force to the sides of the bus bar which will secure the probe in position. However, the probe can readily be removed again from the bus bar. Depressions 117 may be formed as illustrated at an upper part on opposite sides of the probe, to enable the probe to be gripped readily by the fingers of an operator to assist in removal from the bus bar.
Generally, the U-shaped body of the probes described above should be made of an electrically insulating material, at least where the probe is to be in contact with the bus bar whose current is to be measured. In practice, it is convenient to form the entire unitary body of the probes such as illustrated in FIGS. 8 and 9 of an electrically insulating plastics material.
FIG. 10 is a schematic diagram of the circuitry 25, 112, which is incorporated in the wireless enabled probe shown in FIGS. 8 and 9. In the circuit of FIG. 10, each of the Hall sensors of the probe is represented by the device 120, which in the illustrated example is a linear Hall IC, Part No. EQ-731L, manufactured by AKM (Asahi Kasei Microdevices). The Hall devices used in each probe are presorted to provide at least 1% sensitivity matching. Each Hall IC has an offset voltage which is adjusted out using an operational amplifier circuit incorporating a digital potentiometer 121, such as device AD5116 made by Analogue Devices. The resulting circuitry provides a sensor output voltage on line 122 which is a measure of the magnetic field intensity aligned with the axis of sensitivity of the device 120. For the current probe with six magnetic field sensors, the circuitry shown in FIG. 10 containing the device 120 and the Op amp circuitry including the digital potentiometer 121 is repeated six times, one for each of the Hall devices.
The six sensor outputs on respective lines 122 are then supplied to the summing inputs 123 of the summing amplifier containing operational amplifier 125. The input resistances shown in the six summing inputs of the summing amplifier are selected to apply the relative values c₁to c₆for the six magnetic field sensors, so that the output of the summing amplifier on line 124 from op amp 125 represents
$\sum_{r = 1}^{6} c_{r} h_{r} .$
This summed magnetic field sensor value is inverted by op amp 126, sampled by op amp 127 and then buffered by buffer amp 128 for supply to a sense input of a wireless sensor device indicated by the box 129. The wireless sensor device used in the example is an analogue voltage sensing device made by Monnit Corporation which can be interfaced in a wireless network to transmit the sensed voltage value (representing the magnetic field sensor summation) to a remote location.
The circuitry of FIG. 10 is powered by a coin battery shown at 130 via a switched regulator module 131, such as LP-2980 made by Texas Instruments. The unregulated voltage from the battery 130 is supplied to maintain power to the sample and hold op amp 127 and buffer amp 128, and also the wireless sensor module 129. The regulated output from voltage regulator 131 is controlled by a pulse generator incorporating bi-stable circuit 132, producing pulses of duration 1 mS. The bi-stable 132 is itself triggered by a bi-stable 133, connected with a timing RC circuit to cycle between states every one second, so that the 1 mS pulses from the bi-stable 132 are produced once every two seconds. A further bi-stable circuit 134 is connected to produce pulses of length 0.75 mS, to clock the sample and hold circuit 127. With this arrangement, the Hall devices and summing amplifier are powered for brief periods of 1 mS every two seconds during which the summed sensor value is captured by the sample and hold circuit 127 for buffering and subsequent wireless transmission by the wireless module 129. FIG. 11 is the timing diagram for the circuit.
It will be understood that FIG. 10 is only an example of circuits which may be used for determining and summing the magnetic field sensor outputs and, in the wireless enabled embodiment, for transmitting these wirelessly to a remote location. A similar functionality may be achieved using a microprocessor, enabling program controlled sensitivity calibration and offset adjustment.
When the wireless enabled current probe embodiment described with reference to FIGS. 8, 9 and 10 is used in a wireless network, suitable wireless routers and also network monitoring software may also be used as provided by Monnit Corporation. In order to reduce battery consumption for the circuitry of the wireless enabled current probe, the networking and monitoring software can be set to provide a “heartbeat” which activates the wireless circuitry in each current sensor probe only periodically, for example once every hour, and for just long enough to complete a wireless transaction supplying the currently buffered magnetic field sensor summation value.
In summary, an example of the invention provides a method of measuring direct current flowing in a conductor which has minimum and maximum orthogonal cross-sectional dimensions. In the method, an Ampere's Law integration path is effectively defined around the conductor. The integration path is perpendicular to a central line of the conductors and the path has minimum and maximum orthogonal dimensions which exceed the dimensions of the conductor. A respective angle component of magnetic field is then measured at each of a plurality of locations along this integration path. It is important that there is an adjacent pair of these magnetic field measuring locations at one end of the maximum dimension of the path, the locations of this pair being spaced apart by the minimum dimension of the path. There should also be a third location for the measurement which is located at the other end of the maximum dimension of the path. Referring to FIGS. 1 and 2, the Ampere's Law integration path comprises the dotted line 120 linking the measurement locations corresponding to the locations of sensors 20, 21 and 22. The minimum and maximum orthogonal dimensions of the path correspond respectively to the horizontal width and vertical height of the path 120 as illustrated in FIG. 1. The pair of locations at one end of the maximum dimension of the path 120 corresponds to the locations of the sensors 21 and 22 and the third location at the other end of the maximum dimension of the path corresponds to the location of the sensor 20.
It is then important that the locations and the orientations of the angle components which are measured are selected such that there exist values of c_rfor which in any homogeneous magnetic field
$\sum_{r = 1}^{n} c_{r} h_{r} = 0$
where n is the number of said locations, h_ris the measured value of the magnetic field component at the r^thlocation and c_ris a constant factor for the r^thlocation. Then, the measured values of h_rcan be combined to produce a value for the direct current to be measured.
The foregoing detailed description has described only a few of the many forms that this invention may take. For this reason the detailed description is intended by way of illustration and not by way of limitation. It is only the following claims, including all equivalents, which are intended to define the scope of the invention.

Claims

1. A coreless current probe comprising a U-shaped body having arms, an open end, and a cross piece forming a closed end opposite to said open end, said U-shaped body defining an opening having a width between the arms and a length from said closed end to said open end, wherein said length is not less than said width, said width and said length defining a plane of said opening and a central line normal to said plane, whereby a conductor carrying a current to be measured can be engaged by said body of said probe so as to extend through said opening parallel to said central line;

a plurality of coreless single point magnetic field sensors distributed in said body around said opening including a respective said sensor at an end of each of said arms on each side of said open end of said body and at least one said sensors at said closed end of said body, said sensors each having a respective axis of magnetic field sensitivity and being arranged in said body so that each of said axes is not co-planar with said central line;

and sensing circuitry connected to said sensors which is operative to produce for each of said sensors a respective sensor signal which is a measure of the angle component of magnetic field at the sensor aligned with the respective axis of magnetic field sensitivity of the sensor,

said sensors being arranged such that there are values of c_rfor which, in any homogeneous magnetic field,

\sum_{r = 1}^{n} c_{r} h_{r} = 0,

where n is the number of said magnetic field sensors,

h_ris said measure of said magnetic field component for the r^thsensor,

and c_ris a constant factor for the r^thsensor,

said sensing circuitry being further operative to combine said measures h_rto produce a measured current value representing current following in a conductor engaged by said body of said probe.

2. A coreless current probe as claimed in claim 1, wherein each of said magnetic field sensors is orientated to have its axis of sensitivity in said plane of said opening.

3. A coreless current probe as claimed in claim 1, wherein said sensing circuitry is adapted to derive as said measured current value the summation

\sum_{r = 1}^{n} c_{r} h_{r}

where the values of c_rare selected to maximise rejection by the probe of external magnetic fields which are not produced by currents in a conductor engaged by the probe.

4. A coreless current probe as claimed in claim 3, wherein the values of c_rare selected such that in any homogeneous magnetic field

\sum_{r = 1}^{n} c_{r} h_{r} = 0.

5. A coreless current probe as claimed in claim 3, wherein said length of said opening of said U-shaped body is not less than twice said width of said opening.

6. A coreless current probe as claimed in claim 5, wherein said plurality of said magnetic field sensors comprises at least six said sensors, including

a) a first pair constituted by said sensors at said ends of said arms of said U-shaped body,

b) a second pair of said sensors at said closed end of said U-shaped body, and

c) a third pair of said sensors located substantially mid-way along said arms;

the sensors of each of said first, second and third pairs of said sensors being disposed spaced uniformly apart symmetrically on opposite sides of a plane of symmetry containing said central line and equally spaced between said arms;

each of the sensors of said first and second pairs of said sensors being orientated in said body to have its axis of sensitivity in said plane of said opening and at a respective acute angle to said plane of symmetry so as to be generally tangential to said opening;

each of the sensors of said third pair of sensors being orientated in said body to have its axis of sensitivity in said plane of said opening and parallel to said plane of symmetry.

7. A coreless current probe as claimed in claim 6, wherein said length of said opening is greater than twice the width, and the spacing between each of said third pair of sensors and a neighbouring sensor of said first or second pair of said sensors is greater than the spacing apart of the sensors of said first pair of said sensors,

and said sensing circuitry is adapted such that the selected values of c_rfor the sensors of said third pair are greater than the selected values of c_rfor the sensors of said first and second pairs.

8. A coreless current probe as claimed in claim 1, wherein said plurality of said magnetic field sensors comprises at least four said sensors.

9. A coreless current probe as claimed in claim 8, wherein four said magnetic field sensors are distributed symmetrically in said U-shaped body, about a plane of symmetry which contains said central line and is equally spaced between said arms.

10. A coreless current probe as claimed in claim 9, wherein each of said four sensors is orientated in said body to have its axis of sensitivity in said plane of said opening and at a respective acute angle to said plane of symmetry so as to be generally tangential to said opening.

11. A coreless current probe as claimed in claim 10, wherein said plurality of said magnetic field sensors comprises at least six said sensors, including two additional said sensors located substantially mid-way along said arms and orientated in said body to have their axes of sensitivity in said plane of said opening and parallel to said plane of symmetry.

12. A coreless current probe as claimed in claim 11, wherein said plurality of said magnetic field sensors comprises at least nine said sensors including three further said sensors comprising one said further sensor located substantially mid-way along said cross piece of said body, and one said further sensor located at the end of each of said arms, said three further sensors being orientated to have axes of sensitivity in said plane of said opening and substantially normal to said plane of symmetry.

13. A coreless current probe as claimed in claim 12, wherein said plurality of said magnetic field sensors comprises at least thirteen said sensors including four still further said sensors located on said arms of said body, a respective said still further sensor being located on each side of each of said additional said sensors which are located mid-way along said arms, and said four still further said sensors being orientated to have axes of sensitivity in said plane of said opening.

14. A coreless current probe as claimed in claim 1, further including a housing containing a measured current display connected to said sensing circuitry, a handle fixed to said housing, a tube connecting said U-shaped body to said housing, and connection cables extending through said tube, whereby an operator holding said handle can engage a conductor between the arms of said body of said probe and read a measured current from said display.

15. A coreless current probe as claimed in claim 1, including a housing connected to said U-shaped body, wireless signalling circuitry contained in said housing, said wireless signalling circuitry being connected to said sensing circuitry for wireless signalling said measured current values to a remote location, and a battery compartment for a battery to power said sensing circuitry and said wireless signalling circuitry.

16. A coreless current probe as claimed in claim 15, wherein said probe is formed as a unitary structure incorporating said U-shaped body and said housing with said battery compartment.

17. A coreless current probe as claimed in claim 16, wherein said housing with said battery compartment are integral with said cross-piece of said U-shaped body.

18. A coreless current probe as claimed in claim 17, wherein said arms have parallel internal faces providing a predetermined uniform spacing sized to accommodate a rectangular section bus bar engaged by said probe, and at least one compression tab located on the internal face of at least one of said arms, said tab protruding inwards from said internal face and adapted to be resiliently outwardly compressible on engagement with said bus bar to hold the probe in position on the bus bar.

19. A coreless current probe as claimed in claim 1, wherein said U-shaped body has at least exterior surfaces which are electrically insulating.

20. A method of measuring direct current flowing in a conductor having minimum and maximum orthogonal cross-sectional dimensions, the method comprising the steps of:

defining an Ampere's Law integration path around the conductor in an integration plane perpendicular to a central line of the conductor, said path having minimum and maximum orthogonal dimensions which exceed said dimensions of said conductor,

measuring a respective angle component of magnetic field at each of a plurality of locations on said integration path, an adjacent pair of said locations being at one end of said maximum dimension of said path and spaced apart by said minimum dimension of said path, and a third of said locations being at the other end of said maximum dimension of said path, said locations and the orientations of said respective angle components being selected such that there are values of c_rfor which, in any homogeneous magnetic field,

\sum_{r = 1}^{n} c_{r} h_{r} = 0,

where n is the number of said locations,

h_ris the measured value of said magnetic field component at the r^thlocation, and c_ris a constant factor for the r^thlocation,

and combining said measured values h_rto produce a value of said direct current.

21. A method as claimed in claim 20, wherein the respective angle components are orientated in said integration plane.

22. A method as claimed in claim 20, wherein said step of combining performs the summation

\sum_{r = 1}^{n} c_{r} h_{r},

where the values c_rare selected to maximise rejection in the summation of the influence of external magnetic fields which are not produced by current flowing in the conductor.

23. A method as claimed in claim 22, wherein the values of c_rin said summation are selected such that in any homogenous magnetic field

\sum_{r = 1}^{n} c_{r} h_{r} = 0.

24. A method as claimed in claim 20, wherein said maximum orthogonal cross-sectional dimension of said conductor is not less than twice said minimum orthogonal cross-sectional dimension.

25. A method as claimed in claim 24, wherein a respective angle component of magnetic field is measured at at least six locations on said integration path, said locations including

a) a first pair corresponding to said adjacent pair of said locations at said one end of said maximum dimension of said path,

b) a second pair at said other end of said maximum dimension of said path, and

c) a third pair substantially mid-way along said maximum dimension of said path;

said locations of each of said first, second and third pairs of locations being spaced uniformly apart on opposite sides of a plane of symmetry containing a central line of said conductor and parallel to said maximum orthogonal cross-sectional dimension of said conductor;

the respective angle component of magnetic field being measured at each of the locations of said first and second pairs of locations being in said integration plane and at an acute angle to said plane of symmetry so as to be generally tangential to said conductor;

the respective angle component of magnetic field being measured at each of the locations of said third pair of locations being in said integration plane and parallel to said plane of symmetry.

26. A method as claimed in claim 25, wherein said maximum orthogonal cross-sectional dimension of said conductor is greater than twice said minimum dimension of the conductor, and the spacing between each of said third pair of locations and a neighbouring location along said path of said first or second pairs of locations is greater than the spacing apart of the locations of said first pair of locations,

and, in said step of combining the selected values of c_rfor the locations of said third pair are greater than the selected values of c_rfor the locations of said first and second pairs.