WO2009036989A1

WO2009036989A1 - Physical quantity measuring unit and sensor for non-contact electric or magnetic field measurements

Info

Publication number: WO2009036989A1
Application number: PCT/EP2008/007900
Authority: WO
Inventors: Didier Robbes
Original assignee: Inphynix; L'universite De Caen Basse-Normandie
Priority date: 2007-09-21
Filing date: 2008-09-19
Publication date: 2009-03-26

Abstract

The present invention relates to physical unit measuring unit comprising: a sensing means (3) comprising a first (23) and a second (21) transducer, wherein the first transducer is configured to provide a first output (27) and comprises a first balancing input (19) and the second transducer is configured to provide a second output (25) and comprises a second balancing input (17), an amplifier means (5) comprising a first (31) and a second (29) amplifier, wherein the first output of the sensing means is connected to the input of the first amplifier and the second output of the sensing means is connected to the input of the second amplifier, wherein furthermore the first and second amplifier are coupled by a gain setting means(33), a first feedback loop(9) connecting the output of the first amplifier with the balancing input of the second transducer and a second feedback loop connecting the output of the second amplifier with the balancing input of the first transducer. The invention furthermore relates to a probe and a sensor for measuring electric fields, more specifically to a probe that in combination with an electrometer allows measurement of an electric field with high sensitivity, large bandwidth and low frequencies without making electrical contact.

Description

Physical Quantity Measuring Unit and Sensor for Non-contact Electric or

Magnetic Field Measurements

The present invention relates to a physical quantity measuring unit, in particular comprising a probe or a sensor for measuring electric fields, more specifically to a probe that; in combination with an electrometer; allows measurement of an electric field with high sensitivity, large bandwidth and low frequencies without making electrical contact. The invention furthermore relates to a sensor for measuring magnetic fields.

Analog signal processing and sensing of various physical quantities, the measure of which being a number with an attached physical unit in the case of a scalar physical quantity, or giving up to three numbers and one attached physical unit to account for a vector quantity in the ordinary 3D space is known in the art and is described using the general description of signals, linear circuits and systems, especially making use of negative feedback from the system output towards the input.

The feedback theory of linear circuits and systems introduces three main functional features between the input and output signals of a servo-loop: an error detection giving the difference between the system input and output, a signal amplification of that error signal, and a feedback unit which picks up the output signal in order to drive the error detection unit. According to the conventional feedback theory, many advantages are obtained, especially good linearity, dynamic range, and bandwidth of the closed loop system versus those of the open loop ones. Furthermore, a better immunity to gain variations and ageing of the signal amplification chain is obtained, in the sense that at very large forward gain values, the overall, closed loop, system transfer function between input and output becomes mainly dependent on the return feedback chain, chosen in such ways and technologies that they are less sensitive to parameters inducing forward gain variations.

Besides that feedback process in simple one input - one output circuit, analog signal processing is known that starts around differential topologies dealing with comparison between two signals, which means both of these signals are composed of a common component to which a differential component is added. In most cases, only the differential component is relevant to the user, and many detecting units and amplifiers that are following differential sensors provide a strong amplification of the signal difference and try to keep the common mode amplification as low as possible.

In addition, many physical principles of sensors lead to practical devices having at least two electrical contacts, and for which a functional relationship exist between a physical quantity X and a pair of electrical variables: voltage and current (resistive sensors), or electrical charge and voltage (capacitive sensors), or magnetic flux and current (inductive sensors), that can be defined between the two electrodes. Two such terminal devices, known as controlled dipoles, can be arranged in various bridge configurations, including the conventional Wheatstone bridge, to provide sensors with a differential output.

Scientific measuring instruments are making use of sensors to measure physical quantities to a certain degree of accuracy, and good results are obtained by mixing both of the above mentioned techniques, that are a differential amplification and a feedback loop. The former is obtained using instrumentation amplifiers, and to apply the latter, an additional device is introduced to convert the electrical output into amounts of the physical quantity X that are further locally added to the external measured one, already present at the sensor input. Such a device has the generic name of transducer. In the field of magnetometers use is made of various magnetic sensors, such as flux gates, magneto-resistive sensors, SQUIDs. In the case of magnetometers, the usual scheme to return the input physical quantity is by passing an electrical current through a coil, which is obtained using the output signal picked up after the differential - non differential signal conversion, commonly done in the instrumentation amplifier.

International patent application WO 06/027459 describes such a magnetic field sensor comprising a sensor for transforming a magnetic field to be measured into an electrical variable, a reaction field winding for forming a counterbalancing magnetic field and only one linear integrated amplifier for amplifying the electrical variable, wherein a counter reaction loop is provided between the output of the amplifier and the reaction field winding. This device has nevertheless a negative drawback that any fluctuations in the power supply of the transducer causes a reaction in the counter reaction loop, which deteriorates the properties of the magnetic field sensors and a poor common mode signal rejection. In the case of conventional electrometers very small currents are measured to render the measurement of an electric field possible. The electrometers have an ultra high input impedance in order to minimize the influence of the measurement on the field to be measured.

Electric field measurements that require electrical contact are often not possible or desirable, for example in the field of geophysics (mining prospecting, hydrography), biomedical technique (general electrophysiology, electrocardiography, and electroencephalography), measuring of electromagnetic fields, the detection of moving objects and for educational purposes.

International patent application WO 03/048789 describes an example of an electrodynamic sensor that does not need electrical contact with an object under test. The sensor has two probes connected to an electrometer. The electrometer comprises a differential amplifier to reduce noise and high input impedance so that little power is drawn from the received field. The high input impedance is obtained by combining different circuit techniques, such as guarding, bootstrapping and neutralisation.

Improvements of sensors known from the prior art are desirable where sensitivity, bandwidth and the measurement of fields with very low frequencies down to static fields are concerned. Further improvement of performance can be obtained when the disturbance of the physical quantity to be measured by the probe can be reduced.

The first object of the present invention is to provide a physical quantity measuring unit with improved properties.

The second object of the present invention is to provide a probe and a sensor for electric and/or magnetic field measurements that improves the above-mentioned aspects of the prior art.

This first object is achieved with the physical quantity measuring unit according to claim 1 comprising: a sensing means comprising a first and a second transducer, wherein the first transducer is configured to provide a first output and comprises a first balancing input and the second transducer is configured to provide a second output and comprises a second balancing input, an amplifier means comprising a first and a second amplifier, wherein the first output of the sensing means is connected to the input of the first amplifier and the second output of the sensing means is connected to the input of the second amplifier, wherein furthermore the first and second amplifier are coupled by a gain setting means, a first feedback loop connecting the output of the first amplifier with the balancing input of the second transducer and a second feedback loop connecting the output of the second amplifier with the balancing input of the first transducer.

In this context, a transducer is a physical device that converts a physical quantity to another one, while a sensor is a transducer mainly dedicated to the process of measuring, with electrical quantities as the result of the conversion. On the opposite side, a transducer that converts electrical signal to another physical quantity is known as an actuator. Besides a physical unit is attached to the result of its measure, a physical quantity has two main features: it is a scalar quantity, like voltage or temperature if its magnitude is only dependent on time and position in space, or it is a vector quantity, like an electric or magnetic field, if its magnitude is dependent on time and both on position and direction in space. The invention is thus applicable to any kind of scalar or vector physical quantity.

With the inventive circuit topology, the feedback takes advantage of two differential output signal sources instead of the single one used in the prior art. It is the single wiring reversal in the feedback loop design driving the two local counterbalancings which allow the needed sign reversal of the servo loop. It is important to mention that this is achieved without any supplementary poles in the feedback transfer functions thus producing an improved matching of that twofold sources servo loop with respect to stability conditions.

Advantageously the sensing means of the physical quantity measuring unit can be configured such that the first and second output of the sensing means form a differential output signal.

Preferably the sensing means can comprise scalar sensors in a gradiometric arrangement or vector sensors, in particular with reverse sensing reference directions. Using these kind of sensors, the advantages of the feedback loop arrangement are therefore applicable to scalar quantities but also vector quantities. In this context, a gradiometric arrangement of scalar sensors is a combination of two sensors which simultaneously sense a scalar quantity around two definite points separated within a certain distance d, generally named the baseline d of the gradiometer. The results are then expressed as the sensor's output differences divided by the base line d, which is an experimental making of the gradient.

In the case of a pair of scalar sensors, one output of the physical quantity measuring unit scales with the gradient of the scalar physical quantity and gives the result of the difference amplification while another output of the amplifier scales with the value of the scalar physical quantity just reached in the middle of the baseline d linking the two sensors of the sensing means. The latter, that is physically the half sum of the sensors output, is generally available using a proper connecting of a pair of resistors that defines the instrumentation amplifier gain. The corresponding output of the sensing means is a voltage, low impedance, copy of the common mode signal, which becomes useful when dealing with sensors having very high output impedance, to provide common signal guarding of some electrical shields of the circuits.

Another feature must be pointed out to highlight one important feature of the invention. Considering a differential signal which scales with the gradient of a scalar quantity and which is formed symmetrically around the common value that holds in the middle of the baseline d, the difference with respect to the conventional feedback as described above, is illustrated as follows. Using the conventional feedback, only one sensor is used which measures the value of the scalar quantity in a position corresponding to the middle of the baseline d. Additional conventional feedback processing then corresponds to the locking of the system to a reference value, independent of the source producing the measured signal. On the other hand, according to the invention, a gradiometric arrangement of scalar sensors together with the amplifier means is used in which the common mode signal acts as the reference signal to which the system is further locked in a differential way. That differential locked loop mode with crossed signal lines is characterized by the fact that both of the gradiometer output values are together getting very close to a reference value, the latter being strongly related to the value of the measured physical quantity just in the middle point of the baseline.

Another feature of the claimed invention applied to those scalar sensors in gradiometric arrangement is its convenient matching of the amplifier means output to physical quantities able to locally modify the measured physical quantity X. As a matter of fact, general physics principles say that, given a scalar physical quantity X, its gradient, or some mathematical operations upon that gradient, are commonly related to some transport phenomena that are capable of inducing modifications of the local value of that physical quantity. As an example, it is the case of heat conduction through the Fourier's law that links temperature gradient to heat flux, or it is the case of the electric field related to the gradient of the voltage and to electrical charges through the Gauss theorem. In both cases, moving heat or moving charges in the vicinity of the sensor can be used to change locally the value of the physical quantity measured by the sensor. These actions can be induced and controlled by any sources of signal, in such a way that the associated changes of the locally applied strength of the physical quantity counterbalance the variations of the sensed input X. Especially, the above sources of signal can be directly obtained from outputs of the amplifier means, then producing the outsets of the differential feedback process. As a matter of fact, in the proposed structure of the transducers, a sensor and an actuator part are integrated to provide a so called "sensactor" structure.

In the variant of using vector sensors, it has to be taken into account that a vector sensor is such that its output delivers an electrical signal, the magnitude of which is scaling with the scalar product of the measured vector quantity together with a unity vector aligned along a reference direction typically defined with respect to a geometrical marker upon the sensor. Given a vector sensor, and given a direction in the space, which may be, for example, indicated using a theodolite, there are two measuring positions of the sensor along the defined direction that lead to equal values, but sign opposite, of the measured quantity. These two positions differ by half a turn of the sensor around any axis crossing the reference direction at right angle. As a matter of fact, the two measured values may not be exactly the same, due to some uncertainties in the measuring system, but it is known that half of the difference between these two values is a much better estimate of the measured vector quantity. Due to the general miniaturization of sensors, it can be remarked that pairs of vector sensors can be formed, measuring along the same orientation but in opposite directions, in such a way that the magnitude of their outputs have equal values within the sensor accuracy.

In this variant, forming the difference in the outputs of the two sensors of the sensing means leads to a better measure of the vector physical quantity at the input, in the point just placed in the middle of the baseline defined by the geometrical arrangement of the vector sensor pair. According to the invention, the measurement can be performed through amplifier means, and the differential signals at the differential outputs are used to drive actuator parts of the sensing means that return local amounts of the physical quantity counterbalancing the input signal.

According to a preferred embodiment, the unit can further comprise a signal summing means with an inverting input, preferably an amplifying signal summing means. The inputs of the signal summing means are coupled to the outputs of the amplifier means so that, at the output of the signal summing means, with inverting input, a value proportional to the projection of a vector physical quantity onto the measuring axis can be directly obtained, whereas for a scalar physical quantity the information about the gradient is obtained.

Preferably, the first and the second transducer can have the same properties, in particular essentially have the same transfer parameter (α). and/or feedback transfer parameter (β). In this case a more effective signal treatment can be carried out.

According to a preferred embodiment the unit can be configured such that the parameter l αβA_d l » 1 , with A_d being the differential gain of the amplifier means. Under this configuration, the signal treatment and analysis is further simplified.

Preferably the common mode reduction ratio (CMRR) of the amplifier means can be chosen to be at least 12OdB, in particular 12OdB to 13OdB. Again, under this configuration, the signal treatment and analysis are further optimized.

Preferably the sensing means can be at least one of a magnetic field sensing means, a electric field sensing means, a voltage measuring means, a pressure sensing means, and a temperature sensing means. For example, sensors exhibiting these features can be found for magnetic sensing, namely sensors using magneto resistive effects that are currently built in the so-called barber pole configuration, the latter being further arranged in standard Wheatstone bridges, especially using magnetic sensor technologies based upon magnetosresistive effects, including the giant magnetoresistive effect. The advantages of the invention can thus be applied to a whole variety of different physical quantities. The second object is achieved by a probe comprising a means for concentrating the electric field to be measured and a signal capacitor, wherein the means for concentrating the electric field to be measured comprises a structure of two tapered bodies. The tapered shapes have the advantage that the electric field to be measured becomes concentrated between the plates of the signal capacitor. Due to this concentration, it becomes possible to measure low field values. Indeed, a probe of high sensitivity being capable of measuring electric fields from as low as a few μVm^"1 is achieved.

In a preferred embodiment of the invention, the tapered bodies of the means for concentrating the electric field can have a symmetry axis, a base, and a truncation section with a surface area smaller than that of the base, the symmetry axis connecting the centre of the truncation section and the centre of the base of the tapered body, the truncation section and the base being perpendicular to the symmetry axis, the tapered bodies being arranged with the truncation sections opposing each other at a certain distance and their axes coinciding, the signal capacitor being electrically connected to the truncation sections. In this configuration, the sensitivity is further improved.

Preferably, the tapered bodies of the means for concentrating the electric field can be in the form of truncated cones or pyramids. Adopting that shape for the tapered bodies optimizes the amplifying properties and facilitates manufacturing of the tapered bodies.

According to a preferred embodiment, the tapered bodies have essentially the same shape. This provides mirror symmetry with respect to the plane of the signal capacitor to the probe further improving the sensing properties of the probe.

Advantageously, the means for concentrating the electric field to be measured can be formed of an isolating dielectric material with a relative dielectric constant of at least unity, preferably of the order of 100. This choice of material provides a probe with high sensitivity, wide bandwidth and operational frequencies down to quasi-static fields.

According to a variant, the means for concentrating the electric field to be measured is formed of a conducting material with a conductivity of at least 10⁶ Ωm, in particular 10⁶ Ωm to 10⁷ Ωm. This choice of material provides a probe with high sensitivity, wide bandwidth and operational frequencies down to low frequencies. In a preferred embodiment of the invention, the signal capacitor can be provided in the form of a surface mounted component. Such capacitors are available with very small capacitances so that the sensitivity of the probe can be enhanced.

In another preferred embodiment, the signal capacitor can be formed by metalized parts of the truncation sections facing each other. This choice of capacitor has the advantage, that it is easily manufactured, while preserving the probe symmetry.

In another preferred embodiment, the signal capacitor can be provided on an insulating membrane, wherein the metallic electrodes of the signal capacitor are formed on the membrane and positioned in contact with the truncated sections. This provides a simplified arrangement, which is easy to realize and renders the device robust.

In another embodiment of the invention, the probe can further comprises one or more bodies having a shape complementary to that of the tapered bodies, the bodies being made of a material with a relative dielectric constant close to unity, and in contact with the tapered surface of the tapered bodies. This feature provides additional rigidity to the structure and at the same time a perturbation of the electric field can be kept low.

In a preferred embodiment of the invention, two capturing means, in particular metallic electrodes, can be provided, one on the base of each tapered body, so that the electromagnetic field to be measured is more effectively captured, thus improving the sensitivity of the probe.

Advantageously, the capturing means can be replaced by a dielectric material. The dielectric has the advantage that the operational frequencies of the probe are not restricted in the low frequency to static field case based on the skin effect in conducting media.

According to another preferred embodiment, a compensation electrode can be provided near the base of each tapered body. Preferably, the compensation electrodes can have an opening in the centre, and both the outer shape of the compensation electrodes and the shape of the opening match the shape of the capturing means. The compensating electrodes allow nulling the electric field inside the volume that is delimitated by the probe and allows for improved field determination.

Preferably, the compensation electrodes can be provided in a coplanar manner with the capturing means. This offers the advantage of realizing small values of the coupling capacitance between the capture electrode and the adjacent compensation electrode, which are typically in the pF or sub pF range. Small capacitance values are important when a large voltage signal applied to the compensation electrodes is desired to counteract small external fields to be measured.

According to an alternative, the compensation electrodes can be provided on the side of the capturing means opposite to the tapered body with an intervening insulating layer. This configuration reduces the size of the probe, while allowing larger values of the coupling capacitance between the capture electrode and its front compensation electrode, when required by certain specifications of applications, especially those requiring the counteracting of large external fields. Furthermore, this way of realization is easier, reducing mechanical manipulations and assembly.

According to a further embodiment of the invention, the probe according to any of the embodiments described above, can be combined with an electrometer to form an electromagnetic sensor. This sensor shares all the advantages of the corresponding probe.

Preferably, the sensor may further comprise a means to vibrate or rotate the probe along an axis perpendicular to the axes of the tapered bodies. In this way it is possible to also determine static fields on limited bandwidth, in particular using spectral transformation techniques. It should be considered that any rotating or vibrating body in the vicinity of the probe would produce similar spectral transformation, acting in a way similar to a mechanical chopper in optics.

In another embodiment of the invention, the signal capacitor comprises a dielectric material with a nonlinear response to an electric field. In that way, measurements down to static fields become also possible, without moving parts, which enlarge the bandwidth. In an advantageous embodiment, the electrometer of the sensor can comprise a fully differential amplification circuit connected to the electrodes of the signal capacitor for amplifying the signal. In this embodiment of the invention, the compensation electrodes can be connected directly to the fully differential stage outputs of the amplification circuit. In that way, phase inversion by simple reversal of connecting wires, is achieved without adding active structures in the direct line, and the stability of the system is obtained in a simple and robust way, while a high gain is allowed on the amplifying stage. The latter is essential to make the output noise contribution of head amplifiers negligible.

A probe, according to another aspect of the invention, comprises a compensation electrode to nullify the local field inside the active volume of the probe. The compensation electrode's function is to null the local electric field inside the active volume of the probe, typically using a servo loop. The potential difference between the signals applied to the compensation electrodes to compensate the electric field in the probe, divided by a value of length, which can be determined out of the geometry of the probe, gives access to the strength of the electric field.

Preferably, the probe of this embodiment can further comprise a means for concentrating the electric field to be measured and a signal capacitor, wherein the means for concentrating the electric field to be measured preferably comprises a structure of two tapered bodies, in particular truncated cones or pyramids, each tapered body having a symmetry axis, a base, and a truncation section with a surface area smaller than that of the base, the symmetry axis connecting the centre of the truncation section and the centre of the base of the tapered body, the truncation section and the base being perpendicular to the symmetry axis, the tapered bodies being arranged with the truncation sections opposing each other at a certain distance and their axes coinciding, and wherein the signal capacitor is electrically connected to the truncation sections. In a probe with those features, the compensation electrodes are particularly effective and, due to the amplifying effect of the truncated structures, the probe has a high sensitivity and at the same time functions over a large range of electric fields (up to several thousands of V/m.

In this embodiment, the tapered bodies can be formed of an isolating dielectric material with a relative dielectric constant of at least unity, preferably of the order of 100, and two capturing means are provided, one on the base of each tapered body, the capturing means having the same shape as the base. With this structure an effective concentration can be achieved.

The first and/or second object of the invention is also achieved with the magnetic field sensor according to claim 9 or 10. This magnetic field sensor comprises a magnetometer and a fully differential amplification stage. In this context fully differential means that the input and output of the stage is differential which has the advantage that the device is insensitive to fluctuations in the supply voltage of the magnetometer. Preferably the magnetometer has differential outputs.

Preferably the fully differential amplification stage can comprise two parallel amplifiers, wherein the outputs of the two amplifiers are connected to a magnetic field compensation means, in particular an inductor, such that a differential feedback loop is obtained from the two amplifier outputs. In this configuration a stable compensation of the magnetic field to be measured can be achieved.

Advantageously, the magnetometer can comprise a full bridge magneto resistive field sensor, wherein the middle connection points of the full bridge are respectively connected to one input of the two amplifiers thereby forming a differential input. With this structure the properties of the Wheatstone bridge arrangement can be exploited such that the device is insensitive to fluctuations in the power supply.

According to a preferred embodiment, the second inputs of the amplifiers are connected via two first resistors, wherein the output of each amplifier is connected to the respective second input of the amplifiers via a first resistor and wherein the ratio of the second to first resistor is of the order of 10⁴ or more. With such a high resistance ratio, the behaviour of the amplification stage in the frequency domain becomes comparable to a integrated linear amplifier, commonly said an operational amplifier.

Advantageous embodiments will be described in the following in relation to the following figures, wherein:

Figure 1 illustrates a first embodiment of an inventive physical quantity measuring unit, Figure 2 illustrates the sensing means of the physical quantity measuring unit,

Figure 3 illustrates the amplifier means of the physical quantity measuring unit,

Figure 4 illustrates the impact of the unit on a scalar physical quantity x along a measuring axis

Figure 5 illustrates a probe for measuring electric fields according to embodiment 2,

Figure 6 illustrates equipotential lines of the electric field around the probe according to embodiment 2 showing the amplification effect in the centre of the double-cone device,

Figure 7 illustrates a variant of embodiment 2, wherein the electrodes of the signal capacitor are formed on a membrane, which is then positioned in contact with the truncated sections of the cones and the volume between the two capturing electrodes is filled with a material with a relative dielectric constant close to unity,

Figure 8 illustrates a probe for measuring electric fields according to embodiment 3,

Figure 9 illustrates the disturbing influence of the probe of embodiment 3 on an electric field to be measured when compensation electrodes are not activated,

Figure 10a illustrates the electric field around the probe of embodiment 3, when the compensation electrodes have a negative potential relative to the symmetry plane of the device, perpendicular to the symmetry axis,

Figure 10b illustrates the signal picked up at the signal capacitor, in case the signals applied to the compensation electrodes have a slightly different frequency than the one of the outside electric field,

Figure 11 illustrates another example of a probe according to embodiment 3, Figure 12 illustrates a fourth embodiment, namely a biaxial probe, adapted to measure electric field intensities in two orthogonal directions,

Figure 13 illustrates a fifth embodiment, namely a configuration for gradiometric measurements with two sensors,

Figure 14 illustrates a sensor according to embodiment 7 of the invention, comprising a probe and an amplifying circuit with a first fully differential stage with two operational amplifiers and a second, non-differential stage with one operational amplifier, and

Figure 15 illustrates a sensor according to embodiment 8 of the invention, comprising a magnetic field transducer and an amplifying circuit of embodiment 7.

Fig. 1 illustrates a first embodiment of the inventive physical quantity measuring unit 1 , which comprises a sensing means 3 which is coupled to an amplifier means 5. The amplifier means 5 in turn is coupled to a signal summing means 7. According to the invention, the two feedback loops 9 and 11 are connecting differential outputs 13 and 15 of the amplifier means 5 with differential balancing inputs 17 and 19 of the sensing means 3.

Fig.2 illustrates in more detail the structure of the sensing means 3. It comprises a first and a second transducer 21 and 23. In this embodiment, each of the transducers 21, 23 comprise a sensor which is either a scalar or a vector sensor. Preferably both sensors are identical, thus having the same transfer parameter α and the same reverse transfer coefficient β, to take account for the actuator parts of the transducers 21 and 23 which are entered via the balancing inputs 17 and 19. The arrangement of the transducers 21 and 23 illustrated in Fig. 2 corresponds to the representation of a so called "sensactor", typically fabricated using micro technology techniques and assemblies and which integrates a sensor part (transfer parameter α) and an actuator part (reverse transfer coefficient β) able to generate amounts of the physical quantity sensed by the sensor.

In the case of scalar, identical sensors, they are placed within a geometrical baseline d. A unity vector u is defined using u = d/|| d || where d denotes the geometric vector joining the scalar sensors. The two scalar sensors are thus arranged in a gradiometric arrangement. A scalar sensor can, for example, be a voltage, or a temperature sensor.

In the case of a pair of vector sensors, they operate using the same reference direction u, but with opposite sensing orientation +/-u with u being given by the vector sensor specifications. A vector sensor can be an electric or magnetic field sensor.

Introducing electrical variables "el", no matter of their true physical form (e.g. voltage, current, charge or magnetic flux) the sensing means 3, illustrated in Figure 2, the functional relationships between the differential outputs 25, 27 and the input 29 for the physical quantity x (scalar case), x (vector case) can be written as follows:

el_Out⁺ = α(grad(x).u - βel_r ⁺)/2 + el_mc, _ιn, elout^" = -α(grad(x).u + βel_r ^")/2 + el_mc, _ιn,

in the case of scalar sensors, and

elout⁺ = α(x.u - βel_r ⁺)/2 + el_mc, _ιn, elo_ut ^' = -α(x.u + βel_r-)/2 + el_mc, ,_n,

in the case of vector sensors.

In these expressions, bold letters refer to vectors, "grad" is the mathematical gradient and the dot "." indicates the scalar product. These expressions also show how the representation is unified between the two types of sensors integrated in the differential pair of transducers 21 and 23. Indeed, the two types - scalar or vector sensitivity - of differential pairs of transducers can be handled with the same formalism.

To the end, the signal representation of a differential pair of sensactors of Fig. 2 includes a term of common mode signal el_mc,_ιn that is added to both output representations. This common mode term takes into account effects of physical quantities other than the input one x, x which lead to correlated changes of both outputs 25 and 27, the magnitude of which can be large with respect to that associated to the input x, x. The Figure 3 illustrates the structure of the amplifier means 5 according to this embodiment. It comprises a first and a second amplifier 29 and 31 , referred to as "Signal Integrated Linear Circuit" SILC₁ in order to handle the various classes of amplifiers that link electrical variables "el" with the common frequency dependency of a standard operational voltage amplifier. The outputs 13 and 15 form the differential outputs 13 and 15 of the amplifier means 5 which are connected with the signal summing means 7, illustrated in Figure 1. The two amplifier means 5 are coupled via a gain setting means 33, e.g. a passive dipoles network.

Following instrumentation amplifier theory, the output signals el_g ⁺ and el_g ^" are copies of inputs signals el_ιn ⁺ and el_ιn ^" present at the inputs 35 and 37 of the amplifier means 5. In this representation, the output value el_{g mc} is the half sum of these copies and essentially contains the common mode signals. According to a preferred variant and as illustrated, these three output signals el_g,_mc , el_g ⁺ and el_g ^' can be used for the purpose of guarding the input signal carrying lines 39, 41 and/or the overall system 1 , using the design rules for very high impedance circuits.

The two differential signal outputs 13 and 15 are related to the inputs 35 and 37 by the rules:

at the inputs 35 and37, in terms of the common mode signal el_mc and the differential mode signal el_d and:

el_s ⁺ = A_mcel_mc+ A_deld/2 els^' = A_mcel_mc - Ad^βld/2

at the differential outputs 13 and 15 of the amplifiers 29 and 31 and thus of the amplifying means 5. Here A_d represents the differential mode gain, the physical dimension of which is related to the possible change of the nature of the physical signal support between the input and the differential outputs. In the same way, A_mc corresponds to the gain of the common mode signal. It is assumed that A_d and A_mc have the same physical dimension, but that their magnitude generally differs by a very large factor, which is known as the common mode rejection ratio as CMRR = 20log₁₀(Ad/A_mc).

Here the amplifiers 29 and 31 can be conventional voltage operation amplifiers, but also those known as transconductance operational amplifiers, charge amplifiers etc. can find their application. In the former case, the gain setting unit can be represented by resistors (as indicated in Figure 3), and the input impedance is typically chosen to be very high, with values currently exceeding 1GΩ, or even exceeding 1TΩ. The invention is nevertheless not limited to voltage amplifiers, any variables, in particular current, electrical charge or magnetic flux, which are able to carry signals fall under the common description of the invention.

The common features of these amplifiers, also called "signal integrated linear circuits" are: they handle variable electrical signals called "el_x,_y" above, have two signal inputs, commonly materialized by a pair of conducting terminals and presenting impedances Z₊ et Z- matched to the real physical nature of the incoming signals el+ and el.. Each amplifier has a non differential signal output scaling with the difference in the signal inputs with a finite gain at null frequency. They exhibit a first order behavior over several decades after a characteristic frequency fc and have an output impedance z_s, matched to the physical nature of the output signal. It should be noted that only one terminal of each terminal pair is represented, meaning that one terminal is grounded to a reference level. Circuits to be classified under the designation SILC are operational amplifiers with voltage gain, current gain, transconductance or transimpedance gains, and even power gain of traveling guided electromagnetic waves, as defined in the general quadrupole theory.

The summing signal amplifier 7 in this embodiment has an inverting input and at its output 40 one obtains el_s = el_s ⁺ - el_s ^' + ε(el_s ⁺ + el_s ^") , defining the non differential output signal of the unit 1 , where the parameter ε is associated to a possible imbalance between the two channels of the overall signal chain, eventually leading to a small, typically negligible, amount of common mode signal in that output 40.

As a consequence of the feedback inverting system, the combination of the input signal to the returned one can become close to zero. This is established as follows, by regrouping the above introduced equations: el_s ⁺ = A_mceU + Adβl_d/2 and el_s ^" = A_mcel_mc - Aoβl.,/2 el_Out⁺ = α(x.u - βel_r ⁺)/2 + el_mc, ,„ and el_Out^" = -α(x.u + βel_r ^")/2 + el_mc, _in, el_r ⁺ = el_s ^" and el_r ^" = el_s ⁺,

to give at the inputs 35 and 37 of the amplifier means 5:

el_Out⁺ = eUcin + α(x.u - β(A_mcel_mc - Ad^βld/2) )/2, and elout^" = eUc.in - α(x.u + β(A_mcel_mc + Ad^βld/2 )/2

which is achieved by the crossed feedback loop paths 9 and 11 connecting output 15 with the balancing input 17 and output 13 with the balancing input 19. The sum and the difference of the last expression can then be formed to obtain:

el_Out⁺ + elout^" = 2el_mc,_in - αβ A_mcel_mc/2, and ebut⁺ - elout^' = αx.u + αβA_del_d/2

Further knowing that el_Out ⁺ = el_mc + e\J2 and el_Out^' = el_mc - el_d/2, the unit 1 with that differential feedback structure then leads to the solutions at the amplifier inputs 35 and 37 of:

elout⁺ - elout^" = eld = αx.u/(1-αβA_d/2), and el_Out⁺ + elout^' = 2el_mc,_in/(1+αβA_mc/4)

Using the above established amplification rules concerning the amplifier means 5 and these solutions, one then obtains for the differential output el_s ⁺ - el_s ^' of the amplifier means 5:

el_s ⁺ - el_s ^" = A_dαx.u/(1-αβA_d/2) , which can be rewritten as: el_s ⁺ - el_s ^{" «} -2x.u/β, whenever the | αβA_d | product is much greater than the unity.

The above expressions have been established in the case of a vector physical quantity x, but by replacing x by grad(x) the case is also established for a scalar physical unit. In conclusion, a solution looking alike in non differential theory is obtained with the arrangement like illustrated in Figures 1 to 3.This is firstly through the fact that a closed loop differential gain exists, related to the forward differential gain [αβA_d], and also becoming almost independent of the forward transfer gain. It has to be remarked that this forward parameter includes the transfer parameter α of the transducer which means that the closed loop system according to the invention has a better immunity to the interchanging of the sensors, or to their ageing. This signal el_s ⁺ - el_s ^" ~ -2x.u/β is then available at the output 40 of the signal summing means 7 with the inverting input, because it eliminates most of the common mode el_Out⁺ + el_Out^".

In addition, the second part of the solution concerns the common mode gain, which also is affected by the differential feedback process. It must be pointed out, always in that practical limit | αβAd I » 1 , the differential signal at the amplifier inputs 35, 37 becomes small with respect to (αx.u), that is what would have been obtained in the open loop configuration and zero applied counter value on the balancing inputs 17, 19. This means that, in the closed loop configuration, the two transducer outputs 25, 27 are going to values that are very close one from the other. The value to which these differential outputs converge deals with the feedback effect on the common mode and the common gain.

Once the condition | αβA_d | » 1 is obtained, another one follows on the αβA_mc product, which writes:

| αβA_mc | » iσ ^(CMRR/20),

In this embodiment the CMRR values are in a range of 120 dB or more, in particular 120 to 130 dB. This implies that the above condition can be very well fulfilled, while still matching the one : | αβA_mc/41 « 1. Within this last condition, the common mode signal at the level of the two transducer outputs 25, 27 in the locked mode, just reaches the limit elmcin- In practical cases, where el_{mc ιn} is significantly different of zero, the above condition shows that the term (1+αβA_mc/4), that appears in the denominator of the expression defining the common mode in closed loop operation 2el_mc,_ιn/(1+ αβA_mc/4), will not go to zero, then avoiding a large amplification due to the eventuality of a positive feedback for common mode signals. As a consequence, from the above conditions, the inventive differential feedback process does not significantly affect the common mode signal propagation and its removing by the signal summing means 7.

The scalar case: Another advantage, with respect to the locking of the common mode signal, is that obtained for a pair of scalars sensors in the sensing means 3. In this case, a new component adds to the common mode signal el_mc,_in, strongly related to the strength of the measured scalar quantity at the middle point of the base line d.

As a matter of fact, the sensor pair measures the scalar physical quantity X in two points separated by the base line d, which means that X is subjected to a twofold sampling, giving the values X(d₀ +d/2) and X(d₀ - d/2), d₀ denoting the position of the barycenter of the sensor pair 21 and 23. The quantity X can be expanded by virtue of the Taylor's expansion formula to get:

X(d₀ +d/2) = X(do) + (grad(X(d_o)).u)(d/2), and X(d₀ - d/2) = X(do) + (grad(X(d_o)).u)(-d/2),

which implies that the common mode value of the two signals at the sensing means 5 outputs 25 and 27 must contain a value proportional to the scalar quantity right at the reference place, having the magnitude αX(d0).

The algebraic formalization of the locked loop arrangement that has already been given above in the case of vector pairs of sensors can be redone for the scalars sensors by including this transforming.

It follows that the common mode term αX(d0) is just viewed as the el_mc.in one, giving the total common mode signal that holds at the sensing means 3 outputs:

el_Out⁺ + elout^" = (2el_mc,in+ 2αX(d_o))/(1+αβA_mc/4)

In the ideal case, when the other sources of common signal can be neglected with respect to 2αX(dO), we obtain that: a) the differential outputs of the sensing means 3 are together getting a value very close to a reference value, the latter being strongly related to the value of the measured physical quantity just in the middle point of the baseline and b) that the signal at the output 40 of the signal summing means 7 is an image of the gradient of the scalar quantity around the middle point of the baseline, which means that it is fully dominated by the difference:

el_s ^« (X(do +d/2) - X(d₀ - d/2)) = (grad(X(d_o)).u)(d)

Fig. 4 illustrates this situation, which plots a scalar physical quantity 41 along a measurement axis. The way the pair of differential transducers 25 and 27 works in this case can be understood as follows. In fact the actuator parts of the sensing means 3 react to the sensor parts information. First, an internal reference level 43 (the common mode signal) adjusts itself to a value that is very close to that really occurs 45 for the location d₀ on the measurement axis. The actuating parts of the pair of transducers are locally working together 47, in an opposite way and ± d/2 apart from d₀, in order to compensate for the variation of the physical quantity X around its value 45 at location d₀. The amount of work 49 done by the actuating parts of the pair of transducers 25 and 27 automatically compensate for the small change in the measured quantity sampled ±d/2 apart the reference position, and it is directly reported in the main output 40. Because de baseline d is known, the space variation of the measured physical quantity X is available, to thereby produce additional advantages to the overall measuring process and further post-processings.

The present invention provides a probe for an electromagnetic sensor and an electromagnetic sensor comprising the probe, and an electrometer adapted to condition and amplify the signal provided by the probe. The invention is based on the laws of classical electrodynamics, making use of the theory of displacement currents. The electric field to be measured is concentrated using bodies with a tapered shape. The invention is exemplified by the following embodiments.

Embodiment 2

Fig. 5 illustrates a side cut view of a probe according to a first embodiment of the invention. The structure of the probe 100 of this embodiment can be described as a double cone, comprising two truncated cones 102 placed so that the truncated sections 103 face each other. Both the base 104 of the truncated cones 102 and the truncated section 103 of each cone are perpendicular to the axis a of the cone, wherein the axis a is defined as the line connecting the imaginary apex of the cone with the middle of the base 104. The truncated cones 102 are arranged so that their axes overlap, and the common axis a forms the symmetry axis of the double cone structure. The opening angle of each cone is in the order of 90°. The invention is however not limited to this angle and the opening angle can also be any one in the range of 30° to 120°.

In this embodiment, the area of the base 104 is in the order of 1 cm² and the smaller, truncated section 103 has an area of some mm². These values can be chosen from a large range, provided that the area of the base 104 is sufficiently, in particular be a factor of 50 to 100 different from that of the truncated section 103 of the cones.

The cones 102 are preferably formed from an insulating material with a high value of the relative dielectric constant of at least 100. In this embodiment, ceramic barium titanate (BaTiO3) was used, which has a relative dielectric constant of the order of 500. The leakage current in this material is considered extremely weak, the equivalent electric resistance being at least higher than 1 TΩ.

Between the two cones a capacitor 105 is provided. In this embodiment, a capacitor 105 in the form of a surface mounted component is chosen. Other possibilities are discussed below. The capacitor 105 typically has a capacitance of the order of 10 pF to some 10 nF.

The contact between the ceramic material of the cones 102 and the electrical connections of the capacitor 105 is provided by a very thin electric wire (not shown) with a diameter of within a range of 50 to 100 μm. On the truncated side of the cones 102, a small hollow is formed in which the thin wire is inserted. The hollow is subsequently filled with a metallic material of high conductivity and high ductility, like for example Indium or Wood's metal, so that an excellent contact is obtained between the cones 102 and the capacitor 5. When the truncated cones 102 and the capacitor 105 are assembled, the two wires are connected with the centre cores of two coaxial cables respectively. In this embodiment, the electrodes of the signal capacitor 105 are parallel to the base 104 and the truncated section, which ensures best results.

The base sections 104 of the cones 102 are levelled and polished, and a polished metallic capturing electrode 106 is stuck to each of them. The diameters of the two capturing electrodes 106 are identical and larger than the diameter of the base sections 104. Their thickness can be chosen in order to comply with the desired specifications.

The probe 101 functions in the following way: It amplifies an external electric field captured by the capturing electrodes 106 by concentrating the field between the electrodes of the signal capacitor 5 on the other end of the truncated cones 102. The amplification factor depends on the geometry of the truncated cones 102 and the metallic electrodes of the signal capacitor 105.

Fig. 6 demonstrates the amplification effect of the double cone structure of embodiment 2 on an electric field. The lines represent equipotentials of the electric field with a distance of 50 mV. The external field is 100 Vm^'1. The average field at the centre is about 3500 Vm^"1, so that an amplification factor of 35 is attained. That is close to the ratio of the distance between the capturing electrodes 106 and the distance between the electrodes of the signal capacitor 105, which in the shown example was 10mm/0.3mm thus corresponding to 33,3. Figure 6 also shows that the probe 101 does only cause a significant deformation of the electric field lines in the volume delimited by the two capturing electrodes 106 and essentially keeps the field unaffected outside this volume.

Certain variations of the probe 101 of embodiment 2 will be described in the following. In embodiment 2, the bodies of dielectric material forming the means for concentrating the captured electric field are in the form of truncated cones 102. The cone shape is particular effective for this purpose, as a high amplification can be obtained. The cone shape is also favourable for its highly symmetric geometry and the ease with which it can be realized and worked on. However, the bodies do not have to be exactly cone-shaped. The cones could be replaced by bodies of other shape, as long as the electric field captured on a section with a larger diameter is guided to a section with a smaller diameter. In general all tapered bodies can therefore be used. The bodies should preferably have a symmetry axis connecting the centre of the two end surfaces, which in analogy to the truncated cone 102 can be called the base and the truncated section. Those end surfaces are preferably perpendicular to the symmetry axis. Another example of such tapered bodies are truncated pyramids, which can also form the basis for a sensitive probe. The base of the pyramids can be square, polygonal or, for example, rectangle. The dielectric bodies of embodiment 2 are made of BaTiO₃. However, many other materials, like compounds with perovskite structure, e.g. ACu₃Ti₄Oi₂, or CaCu₃Ti₄Oi₂ which exhibit values higher than 10⁴, can be used for that purpose, as long as the permittivity of the material is high, together with high resistivity or those materials commonly used in ceramic capacitor manufacturing.

The choice of a highly polarisable material leads to a very low electric field within the cones. The internal field would be zero in the case of an infinite dielectric constant. The large relative reduction of the internal electric field, in combination with the conditions on the borders forming the interface between different materials, leads to a localized amplification of the electric field between the electrodes of the signal capacitor 105. Important characteristics of the material to be used are a high resistivity and a small loss in the high frequency part of the frequency spectrum of interest for a certain use of the invention. The mentioned parameters are important for the performance of the device, especially regarding its intrinsic noise.

The cones or otherwise shaped tapered bodies forming the means for concentrating the captured electric field do, however, not have to be made of a material with very high dielectric permittivity in any case. Alternatively, they could be made of a material with a very high electric conductivity (metallic or otherwise, like superconductive), e.g. of the order of or above 10⁶Ωm. In that case the surface charge density on the surface of the cones is distributed in such a way that the internal field in the conductor is very weak (zero in the ideal case of infinite conductivity). Therefore, the surface conditions for the electric field are the same in the case of a material of high electric conductivity as for a material of high permittivity and the device will function in a similar way.

There is a difference, however, because in materials with high but finite conductivity the skin effect affects the functioning at very low frequencies.

The probe of embodiment 2 also functions without capturing electrodes 106. However, such a small simplification will cause deterioration in performance, especially concerning the capturing of the electric field at the base 104 of the cones 102. Nevertheless the capturing properties of such a structure could be improved by optimizing the shape of the cones, in particular by enlarging the base of the cone to obtain a form of the modified cone being similar to the form of the cone and the capturing electrodes combined into one structure.

In a variation of embodiment 2, the conical parts 102 of the device may have a size that would allow forming a cavity therein. This cavity could, for example, be used to provide a conditioning circuit, an energy source and/or a radio communication means inside. In that way, the symmetry of the probe is not disturbed by connections to the electronic part of the sensor.

A further possible variation of embodiment 2 concerns the way the signal capacitor 105 is implemented in the probe 110. Instead of providing a separate capacitor in the form of a surface mounted component (SMC), the truncated sections 103 can each be covered with a metallic layer, in part or completely, and placed at a certain distance. The capacitor 105 is then defined by the metallic layer forming the electrodes and their separation through a medium with a low relative dielectric constant, e.g. A value much lower than that of the tapered bodies. In that case, the amplification factor depends directly on the ratio of the distances of the electrodes of the thus formed signal capacitor 105 on one side and the capturing electrodes 106 on the other side. This variation of embodiment 1 has the advantage that the delicate operations to put a SMC capacitor in place (3D arrangement) are not necessary.

Alternatively, the electrodes of the signal capacitor 105 can be formed on both sides of an insulating membrane 121 , which is then placed in between and in contact with the truncated sections 103 of the cones 102, as depicted in Fig. 7. Also this way to implement the signal capacitor 105 facilitates the manufacturing of the probe, here carrying the reference numeral 101'. A further advantage is that this solution does not disturb the symmetry of the probe 101' in the vicinity of the signal capacitor 105. Furthermore, the signal can be taken from the electrodes of the signal capacitor 105 by forming wiring on the membrane applying e.g. micro-photolithographic techniques. The membrane can be of Mylar, for example, or any membrane material commonly used in high frequencies applications.

In another variation of the probe 101 of embodiment 2 and which is also illustrated in Figure 7, all or part of the space around the tapered surfaces of the truncated cones 102 and between the capturing electrodes 106 can be filled with a material of low relative dielectric constant, which only introduces a slight perturbation of the electric field lines. By providing such counter forms 123 of the conically shapes of the guiding bodies 102, the rigidity of the probe 101' is significantly improved.

The truncated cones 102 of embodiment 2 have the same shapes, so that mirror symmetry with respect to the signal capacitor is achieved, which further improves the qualities of the probe. This additional symmetry is, however, not mandatory as long as the desired amplification can be achieved, so that the form of both cones may vary.

It should be understood that each one of the variations on embodiment 2 described above could be implemented separately or in combination with one or more of the other variations.

Embodiment 3

A probe 101", according to embodiment 3 of the invention is depicted in Figs. 8 to 10. These figures are compatible with certain variations on embodiment 2 described above. However, the features characterizing embodiment 3 can be combined with each of the variations described and a combination thereof. Elements carrying the same reference numerals as in embodiment 2 have the same properties and their description is not repeated again in detail but incorporated herewith by reference.

Embodiment 3 differs from embodiment 2 in that additional compensation electrodes 108 are provided in order to essentially nullify the electric field inside the volume that is delimited by the probe 101", to improve the determination of the electric field to be measured which is based on the potentials applied to the compensation electrodes.

In the embodiment depicted in Fig. 8, a thin dielectric plate 107, for example made of glass or epoxy, is fixed to each one of the two capturing electrodes 106. Each dielectric plate 107 supports a compensation electrode 108 in the form of a conducting ring, centred on the symmetry axis a of the probe 101". The inner diameter of the compensation electrodes 108 is chosen to be larger than the external diameter of the capturing electrodes 106 and each compensation electrode 108 is provided in a coplanar manner with the corresponding capturing electrode 106. Figs. 9, 10a and 10b show the effect of the compensation electrodes 108. Contrary to Fig. 6, Figs. 9 and 10a show the local field direction and strength using arrows. In Fig. 9, the potential of the compensation electrodes 108 provided around the capturing electrodes 106 is left floating. In Fig. 10a, the compensation electrodes 108 are given a negative potential relative to the plane of symmetry positioned at the left border of the figure. The external electric field is not equal to zero. The value of the potential difference of the compensation electrodes 108 is chosen such that the electric field in the centre of the probe (lower left corner of the figure) is compensated. The value of the effective field within the cylindrical volume delimited by the compensating electrodes 108 is close to zero, showing the nulling properties of the device, while the outer field lines are not so changed with respect to Fig. 9. This means that the outer disturbance to the measured field is still low.

Figure 10b illustrates the signal obtained at the signal capacitor 105, when the probe 101" is positioned in an external alternating electric field and a compensation signal with a slightly different frequency is applied to the compensation electrodes 108. The clear beating of the signal illustrates that the amplitude of the potentials applied to the compensation electrodes 108 was correctly chosen such that the electric field was compensated at the signal capacitor 105. From the signals applied to the compensation electrodes the electric field properties can then be extracted.

Fig. 11 shows a variation of embodiment 3. The compensation electrodes 108' are provided in a sandwich structure of capturing electrode 106, insulating layer 125 and compensation electrode 108". The advantage of this configuration is that the outer dimensions of the probe are smaller, together with improving the fabrication process.

With both configurations it is possible to nullify the electric field inside the volume of the probe 101", 101"' with an electric field equivalent to the disturbance in the opposite direction using the compensation electrodes 108.

The potential difference between the signals returned to the compensation electrodes 108 divided by a transfer parameter L₁ with the dimension of a length, yields a numerical value of the projection of the electric field on the symmetry axis of the probe. The value of the transfer parameter L₁ depends essentially on the geometric characteristics of the electrodes. In order to facilitate the use of the probe 101", 101"', it is possible to determine the geometric conditions in such a way that the measured electric field corresponds directly to the opposite of the difference of the potential between the compensation electrodes 108 divided by the physical length separating the capturing electrodes 106.

The compensation in the form of an electric field as described here achieves advantages concerning dynamic performance and bandwidth. The thickness t of the capturing electrodes 106 and the compensation electrodes 108 is one of the parameters of the device that can be adjusted, as well as the distance w between those electrodes, the important factor being the ratio w/t. In that manner, the numerical value of the compensation capacitance defined between a capturing electrode 106 and its compensation electrode 108 can be adjusted over a large range including several orders of magnitude, at least from a fraction of a pF to several tens of nF. Numerical estimates are obtained directly from the ratio w/t and the plan capacitor formulae. Improved estimates are obtained using numerical tools. The range is wide, because the ratio w/t may be varied on a large range.

As shown, the probe 101", 101"' of embodiment 3, functioning as an electromagnetic sensor, only slightly deforms the electric field lines around it and has an active volume comparable to its physical dimensions, thereby further improving the characteristics of the probe of embodiment 2.

According to variants, additional electrodes could be supplied as secondary compensation electrodes in order to compensate a possible offset and/or to supply a modulation signal for the capturing electrodes 106. That could be necessary in view of certain signal treatments, for example in order to reduce the background noise.

When using the counter forms 123 of one of the variants of embodiment 2, additional electrodes could also be arranged by applying an electrode layer around the counter forms 123, in particular in a cylindrical manner

Embodiment 4

According to the fourth embodiment, two or three probes with the double cone structure according to embodiments 2 or 3 are combined in such a fashion that their sensitivity directions (symmetry axes) are different, and in particular orthogonal with each other, so that a probe is obtained that can measure both intensity and direction of the electric field in the vicinity of the probe. This arrangement is illustrated in Fig. 12, which represents a side cut view of the probe 131 according to the third embodiment and shows the two orthogonally arranged pairs of cones 133 and 135 with their respective signal capacitors 137 and 139.

Embodiment 5

According to the fifth embodiment, illustrated in figure 13, two or more probes 141 , 143 according to embodiments 2 or 3 could be placed on a straight line 145 with the symmetry axes of the probe parallel to that line. With this configuration it becomes possible to determine the derivative of the electric field component in the direction of the line, as one probe measures E₂ (z) and the other one E_z(z+dz), with z being the direction along line 145. By turning the arrangement by 90° with respect to line 145 the derivative along the other direction can be obtained in the same way. In that way, contributions of distant field sources can be greatly eliminated.

According to a variant the probes could be arranged in a matrix and either all aligned along the same direction or with some probes being aligned in a direction perpendicular

Embodiment 6

The probe according to embodiments 2 or 3 has a symmetry plane perpendicular to its symmetry axis. According to embodiment 6, the probe is furthermore configured such that it can be vibrated or rotated around an axis in that symmetry plane and by using spectral transformation techniques the field strength of static electric fields and fields varying with very low frequencies (<10Hz) can be measured.

In case that the electric field measurement is desired for Ia large bandwidth, a further improvement consists in using non linear dielectric materials with a low coercive field.

It should be notified, that the inventive probe topology comprising capturing means made of metallic electrodes 106 and no compensation electrodes, has the topology of an ac current transformer whose ratio is unity, and that acts just like a known voltage transformer made from magnetic materials and copper wiring. Higher current transformer ratios are obtained by serially adding proper ceramic bodies with their electrodes and a proper wiring between the said electrodes.

Embodiment 7

Figure 14 illustrates a sensor according to embodiment 7 of the invention. The sensor 151 comprises one of the probes according to embodiments 2 to 6 and, in addition, an inventive electrometer, which amplifies and transforms the signal on the signal capacitor 105 to a value of the measured electric field. Thus a structure according to the physical quantity unit 1 according to the invention illustrated in Fig. 1 to 3 is realized with this embodiment in the field of measuring electric fields.

Indeed, the first and second transducers 21 and 23 of the sensing means 3 with their sensing parts (α) and their counterbalancing actuating part (β) are formed by the sensor 151. One plate of the signal capacitor 105, the cone 102 in contact with this plate and the related capturing electrode 106 form the sensing part of one transducer 21 , 23 whereas one compensating electrode 108 and the related capturing electrode 106 (see figure 8) form the actuating part of one transducer 21 , 23. The other electrodes and the second cone form the second transducer.

The outputs 25 and 27 (connected to the signal capacitor 105) of the sensing means 3 are connected to the inputs of the amplifier means 5, here the two amplifiers 161 of the amplification circuit 157. The outputs 165a and 165b of these two amplifiers 161 are then connected to the compensating electrodes 108 with an inversion of the wiring to achieve the above described advantageous effects.

Finally the signals from outputs 165a and 165b are treated by a signal summing means 7 with inverting input, here amplifier 163.

The centre cores of coaxial cables 153, 155, connected to the electrodes of the signal capacitor 105 are connected to the two inputs of the inventive amplification circuit 157. The structure of the amplifier circuit 157 is an instrumentational amplifier with three operational amplifiers. This structure satisfies the requirements of an amplifier of the electrometer type: high input impedance (> 10 TΩ), low polarisation current (< 1 pA), low current noise (1 < fA/ VHz), low voltage noise (< some tens of nV/ VHz). The prepolarisation of the inputs of the amplifier is obtained in the classical fashion by resistance components of high values (> 50 GΩ), and a classical boot strapping technique, which allows keeping the major part of the input impedance of the amplifying structure (not shown). A standard guarding 159 is applied to the common mode signal and the differential signals, in order to limit the effects of imperfections of the wiring.

In this electrometer amplifier circuit 157, the invention introduces a feature that is necessary in order to provide a voltage for the compensation electrodes 108. In this embodiment, the compensation signals are obtained from the electrometer circuit before the transformation from differential amplification 161 to non-differential amplification 163. Taking the signal directly at the output 165a, 165b of the first fully differential amplification stage 161 with a single operational amplifier per input, has the advantage that the signals can be directly applied to the compensation electrodes 108. The necessary inversion of the phase is obtained only by inversing the wiring, without having to add an active element and its inherent supplementary pole, the position of which limiting the gain of the direct chain, given suitable gain and phase margins. Thus, the stability of the system is achieved in a convenient way.

The sensor according to the invention, particularly the one comprising the probe of embodiment 3 including compensation electrodes 018, has number of advantages of technical and economical nature over those known from the prior art. The probe only causes a very small amount of deformation of the electric field lines outside of its immediate vicinity. The sensor can be used in two modes: in direct mode without activating the compensation electrodes 108, or in compensation mode, counteracting the electric field. In both modes the sensor indicates the value of the projection of the electric field vector on its symmetry axis. In particular, the compensation mode brings the invention the advantages of a stable base line, linearity, less susceptible to aging. Numerous prior art devices need at least one galvanic contact in order to provide a reference voltage. The present device operates without any contact. It automatically obtains a reference voltage from the guarding provided to shield the signal lines.

The sensor has good dynamical characteristics. In the compensation mode the following specifications were obtained for the device: Bandwidth: at least 8 decades of frequency;

Sensitivity: S_n ^E = 100 μVm^'VVHz;

Dynamical range: 180 dB attainable.

Embodiment 8

It is important to mention that the inventive amplifier circuit 157 cannot only advantageously be used in applications used to measure electric fields, but by exchanging the probe 151 by a magnetometer, the same advantageous effects, like the stability of the system due to the inversing of wiring - achieved in the magnetic field application by e.g. choosing the direction of the wiring of a feedback coil (clockwise or counterclockwise) - can be achieved when measuring magnetic fields. Thus the invention also relates to a sensor comprising a magnetometer and the amplifier circuit 157 as described above. Thus again a structure according to the physical quantity unit 1 according to the invention illustrated in Fig. 1 to 3 is realized with this embodiment in the field of measuring magnetic fields.

Indeed, the first and second transducers 21 and 23 of the sensing means 3 with their sensing parts (α) and their counterbalancing actuating part (β) are formed by a full bridge magneto resistive magnetic field sensor 175. Resistive elements 181 and 185 form the sensing part of the first transducer 21 and elements 183 and 187 form sensing part of the second one 23. The actuating part is realized with a common compensation means 179, here an inductor, but could of course also be realized out of two distinct compensation means.

The outputs 25 and 27 (here 195 and 197) of the sensing means 3 are connected to the inputs of the amplifier means 5, here the two amplifiers 201 and 205 of the amplification circuit 157.

The outputs 177a and 177b of these two amplifiers 201 and 205 are then connected to the compensating means 179. In the embodiment illustrated in Figure 15 the differential compensation feedback is realized by appropriately choosing the wiring sense of the inductor 179. Thus unlike in the situation as illustrated in Figure 1 and 14 the crossing of the feedback loop 9 and 11 is here realized in an alternative way.

Figure 15 illustrates this situation and shows a magnetic field sensor 171. Its amplifier circuit 173 is based on amplifier circuit 157, but has been adapted to the use of a transducer (magnetometer) 175 for transforming a magnetic field B into an electric signal. Similar to what is shown in Figure 14, the output 177a and 177b of the fully differential amplification stage 77 (61 in Figure 10) is connected to a magnetic field compensation means, here inductor 179, forming a counter magnetic field B_r.

The transducer 175 in this embodiment is a full (Wheatstone) bridge magneto resistive magnetic field sensor, comprising 1^st to 4^th resistive elements 181 , 183, 185, 187. At least one of them is a magneto-resistive element, like a GMR (giant magneto resistance) element. Preferably, the four elements have magnetoresistive properties, arranged in a barber pole configuration to get a well balanced bridge and properly eliminate temperature drifts. The first and second resistive elements 181 , 183 are serially connected between a supply voltage V₃ 191 and ground 193. Also resistive elements 185 and 187 are serially connected between V₃ 191 and ground 193.

The middle connection points 195 and 197 of the transducer 175 are connected to the high impedance input of the fully differential amplification stage 177, namely middle connection point 197 with the positive input 199 of amplifier 201 and middle connection point 195 with the positive input 203 of amplifier 205.

The negative input 207 of amplifier 201 is connected with the negative input 209 of amplifier 205 via two first resistors 211 and 213. The negative inputs 207 and 209 are furthermore connected to the respective outputs 177b and 177a via second resistors 215 and 217.

Finally the outputs 177a and 177b are connected with the non differential amplification stage 219 just like in Figure 14. This one provides a signal at its output which is representative of the magnetic field B.

This magnetic field sensor functions the following way. In equilibrium the tension V_d of the Wheatstone bridge between middle connection points 195 and 197 is zero. In case the magnetic field B≠O the two amplifiers 201 and 205 are arranged such that a current I_R passes through the magnetic field compensation means 179. This current creates an opposing magnetic field B_R such that V_d stays zero. The remaining error (V_d - 0) stays low as long as the resistance ratio of resistance 215/211 and 217/213 is much larger than one, typically in a range of 10^Λ5 or more. In particular the resistance values of 215 and 217 are the same and the ones of resistance 211 and 213 are the same.

The symmetrical arrangement of the device has the advantage that fluctuations in the power supply 191 are equally distributed between the two branches of the Wheatstone bridge, such that the contribution of any power fluctuation contribute in the same way to the tension at middle connection points 197 and 195. As a consequence V_d sensing the potential difference is insensitive to such fluctuations.

The fully differential feedback, thus differential input and output concerning the fully differential amplifier circuit 177, provides stability to the system as during operation V_d is, or at least close to, zero and with the above mentioned resistance ratio leads to a response in frequency comparable to that of a single operational amplifier and a single feedback loop.

The dimensions and values of components given above are not limiting, but can be varied. Since the effect of such variation can in principle be calculated, specifications of the sensor can be made to match those needed for specific applications of the sensor.

Claims

1. Physical quantity measuring unit comprising: a sensing means (3) comprising a first (23) and a second transducer (21), wherein the first transducer (23) is configured to provide a first output and comprises a first balancing input (19) and the second transducer (21) is configured to provide a second output and comprises a second balancing input (17), an amplifier means (5) comprising a first (31) and a second (29) amplifier, wherein the first output (27) of the sensing means (3) is connected to the input of the first amplifier (31) and the second output (25) of the sensing means (3) is connected to the input of the second amplifier (29), wherein furthermore the first and second amplifier (29, 31) are coupled by a gain setting means (33), a first feedback loop (9) connecting the output of the first amplifier (31) with the balancing input (17) of the second transducer (21) and a second feedback loop (11) connecting the output of the second amplifier (29) with the balancing input (19) of the first transducer (23).

2. Physical quantity measuring unit according to claim 1 , wherein the sensing means is configured such that the first and second output of the sensing means form a differential output signal.

3. Physical quantity measuring unit according to claim 1 or 2, wherein the sensing means comprises scalar sensors in a gradiometric arrangement or vector sensors, in particular with reverse sensing reference directions.

4. Physical quantity measuring unit according to one of claims 1 to 3, further comprising a signal summing means (7) with an inverting input, preferably an amplifying signal summing means.

5. Physical quantity measuring unit according to one of claims 1 to 4, wherein the first and the second transducer have the same properties, in particular essentially have the same transfer parameter (α)and/or feedback transfer parameter (β).

6. Physical quantity measuring unit according to one of claims 1 to 5, configured such that the parameter | αβA_d I » 1 , with Ad being the differential gain of the amplifier means.

7. Physical quantity measuring unit, according to one of claims 1 to 6, wherein the common mode reduction ratio (CMRR) of the amplifier means is at least 12OdB.

8. Physical quantity measuring unit according to one of claims 1 to 7, wherein the sensing means is at least one of a magnetic field sensing means, an electric field sensing means, a voltage measuring means, pressure sensing means and a temperature sensing means.

9. Magnetic field sensor comprising a magnetometer (175) and a fully differential amplification stage (177), arranged in particular according to one of claims 1 to 8.

10. Magnetic field sensor, wherein the fully differential amplification stage (177) comprises two parallel amplifiers (201 , 205) wherein the outputs of the two amplifiers are connected to a magnetic field compensation means (177), in particular an inductor, such that a differential feedback loop is formed, in particular according to one of claims 1 to 8.

11. Magnetic field sensor according to claim 9 or 10, wherein the magnetometer comprises a full bridge magneto resistive field sensor (175), wherein the middle connection points (195, 197) of the full bridge are respectively connected to one input (199, 203) of the two amplifiers (201 , 205) thereby forming a differential input.

12. Magnetic field sensor according to one of claims 9 to 11 , wherein the second inputs (207, 209) of the amplifiers (201 , 205) are connected via two first resistors (211 , 213), wherein the output of each amplifier (201 , 205) is connected to the respective second input (207, 209) of the amplifiers (201 , 205) via a first resistor (215, 217) and wherein the ratio of the second to first resistor is of the order of 10^Λ4 or more.