SENSOR DEVICE AND SYSTEM WITH NON-LINEARITY COMPENSATION

Abstract

A sensor circuit for measuring a physical quantity including: a signal acquisition circuit having a sensor to provide an input signal related to the physical quantity; a processing circuit to receive the input signal and for providing an output signal representative of the physical quantity; the processing circuit comprising a closed loop comprising: a first sub-circuit arranged for receiving the input signal and a feedback signal, and configured for providing a first signal; a frequency dependent filter for receiving and filtering the first signal, and for providing the output signal; a second sub-circuit for receiving and converting the filtered signal into the feedback signal using a non-linear function.

Claims

1. A sensor circuit for measuring a physical quantity, the sensor circuit comprising: a signal acquisition circuit comprising at least one sensor configured to provide at least one input signal related to the physical quantity; a processing circuit configured to receive said at least one input signal and for providing an output signal representative of the physical quantity; the processing circuit comprising a closed loop, comprising: i) a first sub-circuit arranged for receiving said at least one input signal and at least one feedback signal, and comprising a combiner, and configured for providing a first signal; ii) a frequency dependent filter configured for receiving and filtering said first signal, and for providing a filtered signal or a signal derived therefrom as the output signal; and iii) a second sub-circuit configured for receiving said filtered signal, and for converting said filtered signal into said at least one feedback signal using a non-linear function.

2. The sensor circuit according to claim 1, wherein the non-linear function is defined by a plurality of parameters determined during a calibration procedure.

3. A sensor circuit according to claim 1, wherein the signal acquisition circuit comprises at least one magnetic sensor configured to measure a magnetic field signal associated with the physical quantity to be measured; or wherein the signal acquisition circuit comprises at least two magnetic sensors, each configured to measure a magnetic field associated with the physical quantity to be measured; or wherein the signal acquisition circuit comprises at least three magnetic sensors, each configured to measure a magnetic field associated with the physical quantity to be measured.

4. The sensor circuit according to claim 1, wherein the physical quantity has an input phase, and wherein the output signal is a phase signal indicative of the input phase; and wherein the signal acquisition circuit comprises a plurality of sensors configured to provide a plurality of input signals, each being a function of said input phase; and wherein the first sub-circuit further comprises a phase generator configured to provide an estimate of said input phase; and wherein the processing circuit is configured to provide the output phase so as to have an improved accuracy with respect to the input phase.

5. The sensor circuit according to claim 4, wherein the first sub-circuit comprises said phase estimator followed by said combiner, and wherein the phase estimator is arranged for receiving and converting said at least one input signal into said phase signal, and wherein said combiner is configured for receiving and combining said phase signal and said feedback signal.

6. The sensor circuit according to claim 5, wherein the second sub-circuit comprises a nonlinear function block arranged for receiving and modifying said filtered signal using a non-linear function; and where the at least one feedback signal is derived from the modified signal.

7. The sensor circuit according to claim 1, wherein the filter has a frequency dependent transfer function T(s), and wherein T(s) is chosen such that [T(s)−1] has at least one zero at DC, and wherein the nonlinear function block is configured to output a feedback signal in the form of θfb1=f(θo)−θo, where θfb1 is the feedback signal, f( ) is a predefined non-linear function, and θo is the output value.

8. The sensor circuit according to claim 7, wherein the transfer function T(s) is chosen such that [T(s)−1] has at least two zeros at DC.

9. The sensor circuit according to claim 1, wherein the filter has a frequency dependent transfer function H(s), and wherein H(s) has at least one pole at DC, and wherein the nonlinear function block is configured to output a feedback signal in the form of θfb=f(θo), where θfb is the feedback signal, f( ) is a predefined non-linear function, and θo is the output value.

10. The sensor circuit according to claim 9, wherein the transfer function H(s) has at least two poles at DC.

11. The sensor circuit according to claim 1, wherein the first sub-circuit comprises said combiner followed by said phase generator, and wherein the combiner is arranged for receiving and combining said at least one input signal and said at least one feedback signal, and wherein the phase generator is configured for converting the combined signal into said phase signal.

12. The sensor circuit according to claim 11, wherein the second sub-circuit comprises a nonlinear function block arranged for receiving and modifying said filtered signal using a non-linear function; and wherein the second sub-circuit further comprises a phase-to-I/Q convertor configured for receiving and converting this modified signal into at least two component signals; and wherein the feedback signal are derived from these component signals.

13. The sensor circuit according to claim 1, wherein the processing circuit introduces a non-linearity; and wherein the nonlinear function block is configured to reduce or substantially eliminate at least the non-linearity introduced by the processing circuit.

14. A position sensor system comprising: a magnetic source configured for generating a magnetic field having a phase indicative of a mechanical position; and a magnetic sensor circuit according to claim 3; wherein the input phase, and a nonlinear error function of said mechanical position; and wherein the nonlinear function block is configured to reduce or substantially eliminate said nonlinear error function.

15. The position sensor system according to claim 14, wherein the second sub-circuit is configured for reducing or substantially eliminating errors, e.g. higher harmonics, related to or caused by mechanical non-idealities of the magnetic source, situated outside of the magnetic sensor circuit.

16. The position sensor system according to claim 14, wherein the magnetic source is a permanent magnet which is movable relative to the magnetic sensor circuit, or vice versa; and wherein the second sub-circuit is configured for reducing or substantially eliminating errors, e.g. higher harmonics, related to mechanical mounting aspects, e.g. selected from the group consisting of: position offset, tilt, eccentricity.

17. The position sensor system according to claim 14, comprising a printed circuit board comprising a plurality of coils; and comprising a target, which is movable relative to said plurality of coils; and comprising said magnetic sensor circuit; and wherein the second sub-circuit is configured for reducing or substantially eliminating errors, e.g. higher harmonics, related to mechanical mounting aspects of the target, e.g. selected from the group consisting of: position offset, tilt, eccentricity, and/or related to layout aspects of the plurality of coils.

18. A current sensor system, comprising: a current conductor for conducting a current to be measured; and a sensor circuit according to claim 3, configured for measuring a magnetic field generated by the current to be measured.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0087] FIG. 1(a) is a schematic representation of an angular position sensor system according to an embodiment of the present invention, comprising a permanent magnet, and a sensor circuit configured for measuring an angular position of the magnet. Nonlinearities (or non-idealities) may be introduced at several places. In the example, a first nonlinearity may be introduced when converting a mechanical angle into magnetic field values, and a second nonlinearity may be introduced inside the sensor device when converting the magnetic signals into electrical signals by the magnetic sensor elements or circuit, and a third non-linearity may be introduced when calculating or estimating an angular position based on the signals provided by the sensor elements. The sensor circuit comprises a correction circuit operating in the angular domain and configured for reducing these nonlinearities.

[0088] FIG. 1(b) shows an example of the combination of the first, second and third non-linearity, illustrated as a non-linear relationship between the (magnetic) angle θcalculated and the mechanical angle θmechanical.

[0089] FIG. 2(a) shows a block diagram of a classical correction circuit for reducing the non-linearity described in FIG. 1(a) and FIG. 1(b).

[0090] FIG. 2(b) shows the inverse of the function shown in FIG. 1(b), which, when combined with the non-linearity function of FIG. 1(b) is highly linear.

[0091] FIG. 3(a) and FIG. 3(b) show a block diagram of a first and a second processing circuit proposed by the present invention, configured for calculating an angle, and reducing or correcting a nonlinearity error. This processing circuit may be used in the system of FIG. 1(a). In the embodiment of FIG. 3(a) and FIG. 3(b), the processing circuit comprises a phase calculator, followed by a correction circuit comprising a combiner and a filter and a non-linear function block arranged in a closed loop. The correction circuit is arranged downstream (after) the phase calculator, hence the combiner and the filter and the non-linear function block operate in the angular domain.

[0092] FIG. 4(a) shows an illustrative example of a circuit having a topology as shown in FIG. 3(a), wherein the processing circuit is configured to determine the angle using a “tracking loop”.

[0093] FIG. 4(b) shows an illustrative example of a typical error between the calculated angular value θcalc provided by the angle calculator 427 of FIG. 4(a).

[0094] FIG. 4(c) shows an illustrative example of a typical error between the corrected angle θcorr provided by the correction circuit 428 of FIG. 4(a) as a function of the input angle.

[0095] FIG. 5(a) is a schematic representation of an inductive angular position sensor system according to an embodiment of the present invention, comprising: a printed circuit board (PCB) comprising a plurality of coils, and a conductive target, and a sensor circuit configured for measuring an angular position of the target. Nonlinearities (or non-idealities) may be introduced at several places. In the example, a first nonlinearity may be introduced when converting a mechanical angle into magnetic field values, a second nonlinearity may be introduced in the receiver and demodulation circuit, and a third nonlinearity may be introduced by the phase calculator or phase estimator inside the sensor device.

[0096] FIG. 5(b) shows an illustrative block diagram of an inductive angular position sensor system comprising a printed circuit board comprising one transmitter coil and three receiver coils. The system further comprises a sensor circuit configured for calculating the angular position in a manner that reduces the above-mentioned nonlinearities, and for outputting the angular position to an external processor (e.g. to an Electronic Control Unit, ECU).

[0097] FIG. 6 shows another embodiment of the present invention, having a signal acquisition circuit similar to that of FIG. 1(a), but having another processing circuit for determining the angular position and reducing the non-linearity error. The processing circuit comprises a combiner, a phase estimator, a filter, a non-linear function block and a phase-to-I/Q convertor arranged in a closed loop, wherein the combiner operates in the I/Q domain; and wherein the filter and the non-linear function block operate in the angular domain. The I/Q domain is located upstream (before) the phase estimator and downstream (after) the phase-to-I/Q convertor. The angular domain is located downstream (after) the phase estimator and upstream (before) the phase-to-I/Q convertor.

[0098] FIG. 7 shows another embodiment of the present invention, having a signal acquisition circuit similar to that of FIG. 5(a), and having the processing circuit of FIG. 6.

[0099] The drawings are only schematic and are non-limiting. In the drawings, the size of some of the elements may be exaggerated and not drawn on scale for illustrative purposes. Any reference signs in the claims shall not be construed as limiting the scope. In the different drawings, the same or like reference signs (e.g. having same modulo 100) typically refer to the same or analogous elements.

DETAILED DESCRIPTION OF ILLUSTRATIVE EMBODIMENTS

[0100] The present invention will be described with respect to particular embodiments and with reference to certain drawings, but the invention is not limited thereto but only by the claims.

[0101] The terms first, second and the like in the description and in the claims, are used for distinguishing between similar elements and not necessarily for describing a sequence, either temporally, spatially, in ranking or in any other manner. It is to be understood that the terms so used are interchangeable under appropriate circumstances and that the embodiments of the invention described herein are capable of operation in other sequences than described or illustrated herein.

[0102] The terms top, under and the like in the description and the claims are used for descriptive purposes and not necessarily for describing relative positions. It is to be understood that the terms so used are interchangeable under appropriate circumstances and that the embodiments of the invention described herein are capable of operation in other orientations than described or illustrated herein.

[0103] It is to be noticed that the term “comprising”, used in the claims, should not be interpreted as being restricted to the means listed thereafter; it does not exclude other elements or steps. It is thus to be interpreted as specifying the presence of the stated features, integers, steps or components as referred to, but does not preclude the presence or addition of one or more other features, integers, steps or components, or groups thereof. Thus, the scope of the expression “a device comprising means A and B” should not be limited to devices consisting only of components A and B. It means that with respect to the present invention, the only relevant components of the device are A and B.

[0104] Reference throughout this specification to “one embodiment” or “an embodiment” means that a particular feature, structure or characteristic described in connection with the embodiment is included in at least one embodiment of the present invention. Thus, appearances of the phrases “in one embodiment” or “in an embodiment” in various places throughout this specification are not necessarily all referring to the same embodiment but may. Furthermore, the particular features, structures or characteristics may be combined in any suitable manner, as would be apparent to one of ordinary skill in the art from this disclosure, in one or more embodiments.

[0105] Similarly, it should be appreciated that in the description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure and aiding in the understanding of one or more of the various inventive aspects. This method of disclosure, however, is not to be interpreted as reflecting an intention that the claimed invention requires more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of this invention.

[0106] Furthermore, while some embodiments described herein include some, but not other features included in other embodiments, combinations of features of different embodiments are meant to be within the scope of the invention, and form different embodiments, as would be understood by those in the art. For example, in the following claims, any of the claimed embodiments can be used in any combination.

[0107] In the description provided herein, numerous specific details are set forth. However, it is understood that embodiments of the invention may be practiced without these specific details. In other instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description.

[0108] In this document, the terms “phase calculator” and “angle calculator” can be used interchangeably, unless explicitly mentioned otherwise.

[0109] In this document, the expression “phase generator” is used as a general term and encompasses the expression “phase calculator” and “phase estimator”. While the expression “phase calculator” suggests that the result is more accurate than for a “phase estimator”, the three expressions can be used interchangeably in this document.

[0110] The expression “reducing the error over a measurement range”, may mean: reducing the maximum error over said measurement range, or an average error, or a total squared error (e.g. sum of integral of the squared error).

[0111] With “improving linearity” is meant: reducing an error between a measured curve and the ideal curve.

[0112] In embodiments of the present invention, “improved linearity” can mean “having an improved linear regression coefficient”, meaning, “having a linear regression coefficient closer to 1.000”.

[0113] With “5 krpm electrical” is meant “5000 revolutions or rotations per minute of the electrical signal”. For example, if a target of an inductive position sensor system contains five lobes and is mounted on a shaft that rotates at an angular speed of 1000 rotations per minute (mechanical), the electrical signal has an angular speed of 5 krpm electrical.

[0114] The present invention relates to the field of sensor systems and devices, and may comprise various types of sensors, e.g. one or more light sensor, one or more magnetic sensor, one or more accelerometer, etc. The principles of the present invention will be explained mainly for a magnetic position sensor system comprising a permanent magnet, and for an inductive angular position sensor system, but the present invention is not limited thereto, and is more generally applicable to other types of sensors and systems, especially magnetic sensors and systems.

[0115] FIG. 1(a) is a schematic representation of an angular position sensor system 100, comprising a permanent magnet 110, and a sensor circuit 120. The permanent magnet 110 may be a two-pole magnet, or a four-pole magnet, or a magnet comprising more than four poles. The magnet 110 shown in FIG. 1(a) is rotatable about a rotation axis, and the sensor circuit 120 is configured for determining the mechanical angle ° mechanical.

[0116] Circuits and methods for determining the angular position ° mechanical of the magnet 110 by measuring characteristics of the magnetic field generated by the magnet 110 using one or more magnetic sensor, are known in the art, for example from WO9854547(A1) or from WO2014029885(A1), just to name a few. Without going into the details, it suffices to know that the angular position of the magnet 110 can be determined for example by measuring a first magnetic signal behaving like a sine-function of the mechanical position, and a second magnetic signal behaving like a cosine function of the mechanical position, and by calculating an arctangent of the ratio of the sine and cosine signal.

[0117] This can for example be accomplished by the sensor circuit 120 shown in FIG. 1(a), comprising a first sensor element 121 providing a first sensor signal s1, and comprising a second sensor element 122 providing a second sensor signal s2. The first and second sensor element 121, 122 may for example be a horizontal Hall element, a vertical Hall element, a magneto-resistive element (MR element), etc.

[0118] Biasing and readout-circuits of magnetic sensor elements are well known in the art, and are not the focus of the present invention, and hence do not need to be explained in more detail here. In the example of FIG. 1(a), the readout circuit 126 may comprise at least one analog-to-digital convertor ADC, and the readout circuit 126 may provide two digital signals v1, v2 to the processing block 150.

[0119] In practice, however, the signals v1 and v2 are typically not a perfect sine and cosine signal, for various reasons. One reason may be that the first magnetic field component measured at the first sensor location, at a predefined distance from the magnet (e.g. at about 1.0 mm), may not be a perfect sine signal, and/or the second magnetic field component measured at the second sensor location may not be a perfect cosine signal. This can happen for example if the sensor device is not perfectly aligned to the magnet, or is tilted, or if the magnet is not perfect, etc., causing a non-linearity. This is schematically indicated in FIG. 1(a) by the symbol “NL1” representing a first nonlinearity or a first non-ideality.

[0120] But also, the biasing and readout circuit is typically not perfect, e.g. due to amplifier offset or gain offset, or non-linearity caused by the analog-to-digital convertor. This is schematically indicated in FIG. 1(a) by the symbol “NL2”.

[0121] In the sensor circuit of FIG. 1(a), the signals v1 and v2 are applied to the processing circuit 150. The processing circuit 150 comprises a “phase generator”, also referred to herein as a “phase calculator”, or “angle calculator”. This block 127 may be configured for determining (e.g. calculating) an angular value θcalc based on an arctangent of a ratio of v1 and v2. In practice, however, the value θcalc provided by the phase calculator 127 is only an approximate value which may deviate more or less from arctan(v1/v2) depending on the implementation, e.g. due to a limited number of bits, and/or due to the way in which the division is implemented, and/or due to the way in which the arctangent function is implemented (e.g. using an iterative formula, or as a Taylor series, or as a look-up table with/without interpolation, etc.). In other words, the phase calculator 127 may also introduce a non-linearity (or non-ideality), schematically represented by the symbol NL3.

[0122] In practice, the effects of the various nonlinearities NL1, NL2 and NL3 typically cumulate, and may be represented by a combined or overall non-linearity NLo, an example of which is shown in FIG. 1(b), as an illustrative example only.

[0123] The inventors of the present invention wanted to find a solution to improve the accuracy of the sensor device (excluding the magnet) and/or the accuracy of the sensor system (including the magnet), by reducing at least one of the non-linearity errors, preferably reducing all of the non-linearity errors, and more preferably substantially eliminating or compensating the overall non-linearity error.

[0124] FIG. 2(a) illustrates a classical solution to this problem.

[0125] FIG. 2(a) shows a block diagram of a correction circuit 228 arranged for correcting the signal provided by the phase calculator 227.

[0126] FIG. 2(b) shows the inverse of the function shown in FIG. 1(b).

[0127] The correction circuit 228 is configured for receiving the calculated angle θcalc from the phase calculator block 227, and for converting this value into a corrected value θcorr using a predefined remapping function, e.g. using a piecewise linear correction function interconnecting a limited number of points (e.g. from 4 to 32 points, e.g. about 8 points, or about 12 points, or about 16 points, or about 24 points). The non-linearity function, or the coordinates of these points may be determined for example during a calibration test, in which the magnet is positioned at a plurality of predefined angular positions, (e.g. at multiples of 30°), and by storing the calculated angular values θcalc in a non-volatile memory of the sensor device. Later, during normal operation, the sensor device can then use a piecewise-linear interpolation between these point to correct the calculated value θcalc in order to reduce the non-linearity error.

[0128] By choosing the piecewise-linear correction function as an approximation of the inverse function of the combination of nonlinearities NL1, NL2 and NL3 described above, the non-linearity can be largely reduced. This technique is used in commercial products, and is well known in the art, and thus does not need to be described in more detail here.

[0129] The correction circuit 228 may optionally comprise a noise filter 230, e.g. having a low-pass characteristic. This may be helpful to smooth the output signal in case of a relatively slowly rotating magnet (e.g. at an angular speed lower than 100 rpm). This noise filter may for example be implemented as a low pass FIR filter (Finite Impulse Response), or as an Exponential Moving Average filter (EMA). In a practical implementation, this noise filter may be disabled or bypassed.

[0130] During the design of a new sensor device, the inventors of the present invention discovered however, that under certain conditions, the solution proposed in FIG. 2(a) and FIG. 2(b) did not work very well, and actually made the results less accurate than without the correction circuit 228.

[0131] After further investigating this problem, it was found that the correction circuit 228 of FIG. 2(a) appears to work quite well for relatively low speed, but does not work very well at relatively high speed, e.g. in the order of about 100 krpm electrical, or higher.

[0132] Desiring to reduce one or more, and preferably all of the nonlinearities also at such a high speed, the inventors started experimenting using various circuit topologies, and they surprisingly found four circuit topologies, illustrated in FIG. 3(a), FIG. 3(b), FIG. 6 and FIG. 7, that give satisfactory results.

[0133] FIG. 3(a) shows a block diagram of a first processing circuit 350 proposed by the present invention, comprising a phase calculator 327 followed by a correction circuit 328 for reducing or correcting one or more nonlinearity errors. This arrangement can be used in sensor circuits like the one shown in FIG. 1(a), for reducing one or more or all of the nonlinearities NL1, NL2, NL3.

[0134] As can be seen, the correction circuit 328 is arranged for receiving an uncorrected signal θcalc from the phase calculator 327, and for providing a corrected output signal θcorr. The correction circuit 328 comprises: [0135] i) a combiner 331 arranged for receiving and combining said uncorrected signal θcalc and a feedback signal θfb1, and [0136] ii) a frequency dependent filter 332 configured for receiving and filtering an output of said combiner 331, and for generating said corrected output signal θcorr, and [0137] iii) a nonlinear function block 333 configured for receiving said corrected output signal θcorr, and for generating said feedback signal θfb1 using a non-linear function f(θ), e.g. the inverse function shown in FIG. 2(b), or an approximation thereof, e.g. a piecewise-linear approximation.

[0138] The frequency dependent filter 332 has a transfer characteristic T(s), wherein [T(s)−1] has at least one zero at DC (i.e. at 0 Hz), but preferably has at least two zeros at DC. The latter is called a Zero Latency Filter (ZLF).

[0139] The feedback signal θfb1 provided by the nonlinear function block 333 may be expressed by the following formula: θfb1=f(θ.sub.0)−θ.sub.0, where Nis the output value provided by the filter T(s), and f(θ.sub.0) is a non-linear transformation of said output value.

[0140] The combiner 331 may be configured to provide an output as a linear combination of the uncorrected value θcalc and the feedback signal θfb1, for example as an addition or a subtraction.

[0141] Simulations have demonstrated that this circuit works very well, even for relatively high speed up to about 200 krpm electrical, or even up to 300 krpm, or even up to 400 krpm, or even up to 500 krpm, or even up to 600 krpm, or even up to 700 krpm.

[0142] It is noted that the correction circuit 328 can be implemented completely in the digital domain.

[0143] In an embodiment, the phase calculator circuit 327 may be further configured for determining an angular speed w and/or an angular acceleration a, e.g. as a first and second time derivative of the angular position, and one or both of these values may optionally be provided to the correction circuit 328. The latter may be configured for adjusting the filter characteristic T(s) and/or the non-linear function f( ) taking into account one or both of said angular speed ω and/or angular acceleration a. For example, in a particular implementation, the correction circuit 328 comprises two filters, one of which is chosen dependent on whether the angular speed is higher or lower than a predefined threshold value.

[0144] In FIG. 3(a) the processing block 327 that provides an estimate of the physical value to be measured is a phase calculator, which may perform an arctangent function of a ratio of the signals v1, v2, but the present invention is not limited hereto, and another processing block 327 may also be used.

[0145] FIG. 3(b) shows a block diagram of a second processing circuit 350′ proposed by the present invention, comprising a phase calculator 327′ followed by a correction circuit 328′ for reducing or correcting a nonlinearity error. This arrangement can also be used in sensor circuits like the one shown in FIG. 1(a), for reducing one or more or all of the nonlinearities NL1, NL2, NL3. The processing circuit 350′ is a variant of the processing circuit 350. The phase calculator 327′ of FIG. 3(b) may be identical to the phase calculator 327 of FIG. 3(a), but the correction circuit 328′ is different.

[0146] As can be seen, the correction circuit 328′ is arranged for receiving an uncorrected value θcalc from the phase calculator 327′, and for providing a corrected output signal θcorr. The correction circuit 328′ comprises: [0147] i) a combiner 331′ arranged for receiving and combining said uncorrected signal θcalc and a feedback signal θfb2, and [0148] ii) a frequency dependent filter 332′ configured for receiving and filtering an output of said combiner 331′, and for generating said corrected output signal θcorr, and [0149] iii) a nonlinear function block 333′ configured for receiving said corrected output signal θcorr, and for generating said feedback signal θfb2 using a non-linear function f(θ).

[0150] The frequency dependent filter 332′ has a transfer characteristic H(s) with at least one pole at DC (i.e. at 0 Hz), but preferably having at least two poles at DC.

[0151] The feedback signal θfb2 provided by the nonlinear function block 333′ may be expressed by the following formula: θfb2=f(θ.sub.0), where θ.sub.0 is the output value provided by the filter H(s), and f(θ.sub.0) is a non-linear transformation of said output signal.

[0152] The combiner 331′ may be configured to provide an output as a linear combination of the uncorrected signal θcalc and the feedback signal θfb2, for example as an addition or subtraction.

[0153] In a variant (not shown), the block 333 of FIG. 3(a) and/or the block 333′ of FIG. 3(b) may comprise one or more storage elements, and/or one or more delay elements, and/or one or more scalers, and may provide a combination of a non-linear function of the value θ.sub.0 at its input and one or more previous values.

[0154] In an embodiment, the block 333 of FIG. 3(a) and/or the block 333′ of FIG. 3(b) may comprise a piecewise linear interpolator and a FIR filter and be configured for providing a combination of a piecewise linear transformed version and a FIR filtered version of the value θ.sub.0 applied at its input.

[0155] In an embodiment, the block 333 of FIG. 3(a) and/or the block 333′ of FIG. 3(b) do not contain a FIR filter or a delay element, but only contain said piecewise linear interpolator.

[0156] Also in this case, the phase calculator 327′ may optionally provide one or both of an angular speed w and/or an angular acceleration a to the correction circuit 328′, which may adjust its behaviour depending on one or both of these values (speed and/or acceleration).

[0157] In FIG. 3(b) the phase calculator 327′ that provides an estimate of the physical value to be measured, may perform an arctangent function of a ratio of the signals v1, v2, but the present invention is not limited hereto, and another processing block may also be used.

[0158] FIG. 4(a) shows an illustrative example of an implementation of a circuit as shown in FIG. 3(a), wherein the “phase generator” or “angle calculator” 427 is implemented using a so called “tracking loop”. In a particular embodiment, this tracking loop may be implemented as described in more detail in EP3528388(A1), incorporated herein by reference in its entirety, in particular the circuit shown in FIG. 2 thereof, and the corresponding text.

[0159] As can be seen, the processing circuit 450 comprises an angle calculator 427 comprising a tracking loop, and the processing circuit 450 further comprises a correction circuit 428, but the correction circuit 428 is arranged downstream of the angle calculator 427, outside of the tracking loop.

[0160] It is a major advantage of this “tracking loop” that it can be used to calculate or estimate or approximate or emulate an arctangent function without actually requiring a programmable DSP (digital signal processor) or arithmetic unit.

[0161] FIG. 4(b) shows an illustrative example of a typical error between the calculated angular value θcalc provided by the tracking loop 427 (e.g. implemented as described in EP3528388), and the angular value corresponding to the one or more sensor signals applied to its input. As can be seen, there is a predefined relationship between the angular value corresponding to the input signal(s) and the calculated value. This error can be regarded as a “non-linearity error” NL3 as described above.

[0162] FIG. 4(c) shows an illustrative example of a typical error between the corrected angle θcorr provided by the correction circuit 428 as a function of the input angle “input”. As can be seen, the correction circuit 428 is capable of reducing the error of about ±4.0° illustrated in FIG. 4(b) to an error of only about ±0.2° illustrated in FIG. 4(c), i.e. is capable of reducing the error by a factor of more than 10.

[0163] As already mentioned above, the principles of the present invention are not limited to magnetic position sensor systems comprising a magnet, for example as illustrated in FIG. 1(a), but can also be used for other kind of sensor systems and circuits, e.g. comprising one or more magnetic sensors, one or more light sensors, one or more pressure sensors, one or more accelerometers, etc.

[0164] FIG. 5(a) and FIG. 5(b) show a schematic representation of an inductive angular position sensor system 500, comprising a printed circuit board (PCB) 513, and a sensor circuit 520 and a conductive target 511. The sensor circuit 520 comprises a signal acquisition circuit 540 and a processing circuit 550. The signal acquisition circuit 540 comprises a plurality of coils, e.g. one transmitter coil and three receiver coils, the latter functioning as sensor elements 521, 522, 523. The sensor circuit 520 is configured for determining an angular position θmech of the target 511.

[0165] Nonlinearities (or non-idealities) may be introduced at several places. For example, a first nonlinearity NL1 may be introduced when converting the mechanical angle θmech into electrical or magnetic signals (e.g. into three modulated AC signals); a second nonlinearity NL2 may be introduced by the “Receiver and Demodulation circuit” 526. This circuit may also include at least one analog-to-digital convertor ADC, and optionally also a Clarke transformation circuit; A third nonlinearity NL3 may be introduced inside the processing circuit 550 itself, e.g. by the phase calculator 527, when calculating an angular position θcalc based on the input signals v1, v2. The nonlinearities NL1, NL2, NL3 are only shown for illustrative purposes, and in practice there may be more than three or less than three sources of nonlinearities, or there may be other nonlinearities, but the principles explained above remain the same.

[0166] Inductive position sensors are well known in the art, and hence need not be explained in full detail here. In order to understand the present invention, it suffices to know that typically an alternating signal (AC signal) is applied to an excitation coil (also referred to as transmitter coil), which will generate an alternating magnetic field. The transmitter coil may have a circular shape surrounding the receiver coils, as is known in the art. This magnetic field is modulated by the angular position of the conductive target 511, which influences the signals s1, s2, s3 obtained by the receiver coils 521, 522, 523, acting as sensor elements. These signals can be converted into an angular position, e.g. in the “Receiver and Demodulation circuit” 526, using known techniques. Some inductive sensor systems contain only two receiver coils, which are arranged such that the received and demodulated signals are substantially 90° phase shifted.

[0167] In the example shown in FIG. 5(b), there are three receiver coils, which are arranged such that the received and demodulated signals are substantially 120° phase shifted. These three signals may be converted into quadrature signals v1, v2 in known manners, e.g. using a Clark transformation. A phase calculator 527 can then determine an angular position of the target based on these signals v1, v2. And a correction circuit 528 can reduce any nonlinearity errors and provide the angular position θcorr with an improved accuracy to an external processor 540, for example to an electronic control unit (ECU). The correction circuit 528 can have a topology as illustrated in FIG. 3(a) or as illustrated in FIG. 3(b) described above.

[0168] This is of course only one illustrative example of an inductive position sensor, and many variations are possible. For example, many shapes of the target 511 are possible, and various layouts of the coils are possible, etc. (for example: C-shape, O-shape). The phase calculator 527 may be implemented using a digital processor (e.g. a DSP), or using a look-up table, or using a tracking loop, e.g. as described above in FIG. 4(a).

[0169] In certain embodiments, the corrected angle θcorr is converted into a sine and a cosine signal, which are then provided to the ECU. Such implementation is particularly useful for backwards compatibility with existing systems.

[0170] In a variant of the system shown in FIG. 5(b), the correction circuit 528 is not implemented inside the sensor device, but in the external processor 540.

[0171] Referring back to FIG. 1(a) to FIG. 5(b), in FIG. 1 to FIG. 5 sensor circuits (e.g. 120) are disclosed which are capable of reducing or substantially compensating nonlinearities introduced inside or outside of the sensor circuit up to a very high speed. These sensor circuits (e.g. 120) comprise a signal acquisition circuit (e.g. 140), and a processing circuit (e.g. 150). In these embodiments, the processing circuit (e.g. 150) is configured to receive at least one input signal v1, v2, for example two sinusoidal input signals substantially 90° phase shifted, or three sinusoidal signals substantially 120° phase shifted. In these embodiments, the processing circuit (e.g. 150) is configured for providing an output signal θcorr representative of the physical quantity to the measured, e.g. an angular position of a shaft connected to a magnet or connected to a conductive target.

[0172] Using the reference numbers of FIG. 3(a), the processing circuits of FIG. 1(a) and FIG. 5(a) comprise a closed loop formed by: a first sub-circuit 360, an output of which is connected to a filter 332, an output of which is connected to a second sub-circuit 333, an output of which is fed back to the first sub-circuit 360. More specifically, in these embodiments:

the first sub-circuit 360 is arranged for receiving said at least one input signal v1, v2 and at least one feedback signal θfb1. The first sub-circuit comprises a phase generator 327 and a combiner 331 and provides phase signal θcalc. The phase estimator 327 is followed by the combiner 331.
the filter 332 is a frequency dependent filter, is configured for receiving and filtering the phase θcalc, and is configured for providing the filtered signal as the output signal θcorr;
the second sub-circuit 333 consists of a nonlinear function block 333, arranged for receiving said filtered signal θcorr as an input signal, and for converting this signal into a feedback signal θfb1 using a non-linear function, e.g. using a piecewise-linear approximation passing through a limited number of 8 to 32 predefined points.

[0173] As can be appreciated, this closed loop functions completely in the angular domain.

[0174] As mentioned above, the inventors also found another topology, which is also capable of reducing or substantially removing nonlinearities up to very high speed, e.g. up to 200 krpm electrical, or even up to 400 or 500 or 600 or 700 krpm electrical, as will be described next, illustrated by FIG. 6 and FIG. 7.

[0175] FIG. 6 shows another embodiment of a processing circuit 620 proposed by the present invention, having a signal acquisition circuit 640 and a processing circuit 650. The signal acquisition circuit 640 may be similar or identical to that of FIG. 1(a), or variants thereof. But the processing circuit 650 is different from those described above.

[0176] The processing circuit 650 comprises a closed loop formed by: a first sub-circuit 660, an output of which is connected to a filter 632, an output of which is connected to a second sub-circuit 670, an output of which is fed back to the first sub-circuit 660. More specifically, in the specific embodiment shown in FIG. 6, the processing circuit 650 comprises a combiner 631, a phase estimator 627, a filter 632, a non-linear function block 633 and a phase-to-I/Q convertor 634 arranged in a closed loop. The combiner 631 operates in the I/Q domain. The filter 632 and the non-linear function block 633 operate in the angular domain.

[0177] In the specific embodiment of FIG. 6, the first sub-circuit 660 is arranged for receiving said at least one input signal v1, v2 and two feedback signals fbq, fbi. The first sub-circuit 660 comprises a combiner 631 followed by a phase estimator 627. The combiner 631 is configured for combining the input signals v1, v2 and the feedback signals fbq, fbi, for example by performing a complex multiplication of the input signals with the complex conjugate of the feedback signals. the phase estimator 627 is configured for converting the combined signals into a phase signal θest. The phase estimator 627 may be an I/Q-to-phase convertor;

[0178] the filter 632 is a frequency dependent filter, is configured for receiving and filtering the estimated phase signal θest, and is configured for providing the filtered signal as the output signal θcorr;

[0179] the second sub-circuit 333 comprises a nonlinear function block 633 followed by a phase-to-I/Q convertor 634. The nonlinear function block 670 is arranged for receiving and modifying said filtered signal θcorr using a (e.g. memoryless) non-linear function, e.g. using a piecewise-linear approximation function passing through a limited number of 8 to 32 predefined points which may be stored in a non-volatile memory. The phase-to-I/Q convertor 634 is configured for receiving and converting this modified signal into at least two component signals, from which two feedback signals fbi, fbq are derived, which are provided to the combiner 631.

[0180] As can be appreciated, this closed loop functions partly in the angular domain, and partly in the I/Q domain.

[0181] FIG. 7 shows another embodiment of a processing circuit 720 proposed by the present invention, having a signal acquisition circuit 740 similar to that or identical to that of FIG. 5(a), and having the same or similar processing circuit 750 as that of FIG. 6.

[0182] If the three sensor signals s1, s2, s3 are obtained from an inductive sensor having three receiver coils, these modulated signals are preferably demodulated, digitized and converted into two-phase signals v1, v2, e.g. by using a Clarke Transformation, and then provided to the processing circuit 750.

[0183] The principles of the present invention are explained for four magnetic position sensor circuits, illustrated mainly in FIG. 1(a), in FIG. 5(a), in FIG. 6 and in FIG. 7, but the present invention is not limited thereto, and many variants are also contemplated. For example:

in FIG. 1(a) and FIG. 5(a) the readout circuits 126, 526 provide two signals v1, v2 to a phase calculator 127, 527, and the latter calculates (or estimates) a phase or angle based on these signals. But the invention also works if a physical value is measured using a single sensor, e.g. an amplitude of a current to be measured. In this case, the phase calculator may be replaced by another processing block 527, e.g. an analog or a digital scaling block;
in FIG. 1(a) and FIG. 5(a) the signals(s) v1, v2 are digital signals, but that is not absolutely required, and the step of digitization may also be performed after the processing block 527;

[0184] As mentioned before, the present invention also works in other systems, where another source (e.g. gravity force, light source) and/or another kind of sensor (e.g. an accelerometer, a light sensor) is used. The present invention therefore also discloses:

a linear position sensor system, and an angular position sensor system, comprising a position sensor circuit or device, and a magnetic source, e.g. a permanent magnet.
a current sensor system comprising a current sensor circuit or device, and an electrical conductor, such as an integrated conductor or an external conductor, e.g. a bus bar.
a proximity sensor system comprising a sensor circuit or device, and a conductive target, which is movable relative to the sensor device.