The present invention relates to a hall electromotive force signal detection circuit and a current sensor thereof, and more specifically, to a hall electromotive force signal detection circuit and a current sensor thereof each of which can achieve wide-band characteristics and fast response as well as high accuracy in a magnetic sensor using a hall element, such as a current sensor.
A magnetic sensor using a hall element is widely used not only as a proximity sensor, a linear position sensor, a rotation angle sensor, or the like, each of which is a sensor for detecting positional information of a magnet, but also as a current sensor for measuring, in a non-contact manner, the amount of current flowing in a current conductor by detecting a magnetic field induced by the current flowing in the current conductor.
Particularly, it is required for the current sensor used to detect an inverter current of a motor, to highly precisely detect the inverter current that is switched at a high-speed frequency, for the purpose of improving the efficiency of motor control.
Since the hall element has a magnetoelectric conversion function to generate a hall electromotive force signal depending on an intensity of an input magnetic field, the hall element is widely used as a magnetic sensor. However an offset voltage (unbalanced voltage) exists in the hall element. The offset voltage is a non-zero finite voltage and is output even in a state where the magnetic field is zero, i.e., in a state of no magnetic field.
In view of this, in regard to the magnetic sensor using a hall element, in order to cancel an offset voltage of the hall element, there is a driving method for the hall element, which is generally known as spinning current method or connection commutation method. In this method, switching between a position of a terminal pair for flowing a driving current to the hall element and a position of a terminal pair for detecting a hall electromotive force signal is periodically performed synchronized with clock called chopper clock (see, for example, Non Patent Document 1).
The spinning current method for the purpose of cancelling the offset voltage is configurable by use of a switching circuit in a CMOS semiconductor circuit, and therefore, a hall electromotive force detecting circuit to realize a highly-accurate magnetic sensor generally includes a switching circuit in order to implement the spinning current method.
Further, since a hall electromotive force signal generated in the hall element is generally feeble, the hall electromotive force signal detection circuit includes an signal amplifier circuit so as to amplify the hall electromotive force signal. Here, in a case where there is a finite offset voltage in this signal amplifier circuit, it is also necessary to perform the offset cancellation of the offset voltage included in the signal amplifier circuit.
Under these circumstances, as for the hall electromotive force signal detection circuit for detecting and amplifying the hall electromotive force generated by the hall element, there has been a known circuit configuration of signal amplifier circuit using a current feedback amplifier circuit that is suitable for the combination with the spinning current method of the hall element. Among these current feedback amplifiers, the circuit configuration in which the offset voltage of the amplifier circuit is modulated by using a chopper clock to the chopper clock frequency is generally known as chopper amplifier.
It is known that, with the use of a hall electromotive force detecting circuit in which the circuit for implementing the spinning current method in the hall element is combined with the circuit configuration of the chopper amplifier in the signal amplifier circuit, it is possible to modulate both the offset voltage of the hall element and the offset voltage of the signal amplifier circuit at the frequency of the chopper clock (see, for example, Patent Document 1 and Non Patent Document 2).
The offset cancellation of the hall element by the spinning current method is described below.
In
By periodically switching the direction of a bias current of the hall element between 0 degree and 90 degrees synchronized with the chopper clock, it is possible to switch the polarity of the hall electromotive force signal Vsig (B) corresponding to the target magnetic field to be detected between inverted polarity and non-inverted polarity, thereby making it possible to perform frequency modulation of the hall electromotive force signal Vsig(B) corresponding to the target magnetic field to be detected, at the frequency f_chop of the chopper clock. On the other hand, as for the DC offset voltage Vos (Hall) of the hall element, even if the driving direction of the hall element is switched between 0 degree and 90 degrees, the DC offset Vos(Hall) retains its polarity. Therefore, Vos(Hall) is not frequency-modulated by the chopper clock.
Based on the foregoing, in a case where the driving current of the hall element is switched between 0 degree and 90 degrees synchronized with the chopper clock, the signal Vhall generated in the hall element exhibits a waveform as illustrated in
It is noted that the above description deals with an example in which the driving direction of the hall element having four terminals is switched between two directions, i.e., 0 degree and 90 degrees. Similarly, when the driving direction of the hall element is switched among four directions, i.e., 0 degree, 90 degrees, 180 degrees, and 270 degrees, it is also possible to separate the hall electromotive force signal and the offset voltage in the frequency domain.
Furthermore, a technique described in Patent Document 2 relates to a hall electromotive force signal detecting apparatus which is able to effectively minimize a pattern noise generated in a ΔΣ modulator used to detect a hall electromotive force signal of a hall element. Furthermore, a technique described in Patent Document 3 relates to a hall electromotive force signal detecting apparatus which performs a signal processing by use of a ΔΣ modulator and which is able to effectively minimize settling errors.
Furthermore, a technique described in Patent Document 4 relates to a current sensor, and discloses a method for reducing ripple-shaped noise by use of a filter called a Notch filter or a COMB filter having a zero point at the frequency of a chopper clock, in association with a circuit technique called chopper amplifier.
Furthermore, a technique described in Patent Document 5 relates to a chopper stabilized amplifier, and discloses a circuit configuration that suppresses the occurrence of ripple-shaped noise in an output of a signal amplifier circuit by feedback means.
Furthermore, a technique described in Patent Document 6 relates to an amplifier circuit that compensates an offset, in which a feedback circuit is prepared for the purpose of offset cancellation, and the feedback circuit herein is coupled to a signal node before the signal node in which the hall electromotive force signal modulated by the chopper clock is demodulated by a demodulator.
Further, a technique described in Patent Document 7 relates to, for example, a magnetic field sensor, which is provided with a sensor element for outputting a sensor signal including an interfering signal, an evaluation apparatus coupled to this sensor element, and a subtractor for subtracting the interfering signal from the sensor signal. The technique also relates to an interfering signal compensation method of the sensor and discloses an example in which a ΔΣ AD converter is used for signal processing of a hall element.
Further, a technique described in Patent Document 8 relates to a current measuring device for measuring a current flowing in the target object by means of measuring the magnetic field induced by the current flowing in the target object, and discloses an example in which double integral AD conversion is used for signal processing of a hall element.
However, the techniques described in Patent Documents 1 to 3 do not disclose anything about a difference calculation circuit for detecting, in the output voltage signals, a component synchronized with the chopper clock by performing sampling at a timing obtained from the chopper clock, as disclosed in the present invention.
Further, in the technique described in Patent Document 4, ripple-shaped noise of an offset voltage modulated by a chopper clock is reduced by a filter in a subsequent stage, but in this case, various problems can occur as described below.
One use of the magnetic sensor using a hall element is a current sensor. In the current sensor using a hall element, an intensity of a magnetic field induced by a target current to be measured is measured in order to measure an amount of current flowing in a current conductor, which makes it possible to measure even a large current in a non-contact manner. Accordingly, the current sensor using a hall element is a sensor very suitable for measurement of an inverter current of a motor, and the like.
This current sensor is used to detect an inverter current which is switched at a high frequency, and is generally required to have a fast response of 10μ-seconds or less in terms of a signal processing delay time, as well as a wide-band property of approximately 100 kHz.
In a case where the hall element is used for the current sensor, it is difficult to achieve high accuracy in regard to detection accuracy of the current detection unless measures to cancel the offset voltages (the offset voltage of the hall element and the offset voltage of the signal amplifier circuit) are taken. In view of this, in a case where the hall element is used as the current sensor, the above-mentioned offset cancellation method using a chopper clock is used.
In a case where the above-mentioned offset cancellation method is used in the current sensor, if a narrow-band lowpass filter is used to reduce ripple-shaped noise occurring in the output signal of the signal amplifier circuit, it becomes difficult to satisfy the above-mentioned requirements such as the wide-band property and fast response. This is because the narrow-band lowpass filter has a long delay time (group delay time) of the filter. In addition, a problem with an increase in noise may be also caused.
On the other hand, a circuit configuration is also known in which a feedback is prepared from the output of the signal amplifier circuit to the signal amplifier circuit so that ripples do not occur, instead of reducing ripple-shaped noise occurring in the output signal of the signal amplifier circuit by a filter in the subsequent stages. That is, the circuit configuration detects a DC offset that causes the ripple-shaped noise from the ripple-shaped noise occurring in the output signal of the signal amplifier circuit, and the detected DC offset is subtracted in the signal amplifier circuit, so as to prevent the occurrence of the ripple-shaped noise. An example of this circuit configuration is exemplified in Non Patent Document 4.
In the circuit configuration disclosed in Non Patent Document 4, capacitive coupling by use of capacitors is used so as to detect a DC offset voltage component in the ripple-shaped noise occurring in the output signal of the signal amplifier circuit. However, the configuration using such capacitors cannot be used when a target frequency bandwidth to be detected is wide as in the case of the current sensor.
In order to explain the reason thereof, the following describes a clear distinction between the hall electromotive force signal detection circuit for a current sensor, used for current detection of an inverter, and the signal amplifier circuit used for an instrumentation (e.g., see Non Patent Document 4).
In the case of the above-mentioned Non Patent Document 4 belonging to a technical field of the signal amplifier circuit for instrumentation, the signal amplifier circuit is designed for the purpose of detecting a low-frequency signal of a few Hz, and there is a sufficient difference of approximately 10000 times between the frequency (40 kHz) of the chopper clock described in Non Patent Document 4 and a frequency bandwidth (a few Hz) of a target signal to be detected. In view of this, even if the DC offset voltage is detected in the ripple-shaped noise by means of capacitive coupling using capacitors, the detected DC offset voltage is not mixed with the target signal to be detected.
On the other hand, in the current sensor used for current detection of an inverter, since a bandwidth of a target current signal to be detected is wide (approximately 100 kHz), a frequency of the chopper clock should be correspondingly increased to a higher-frequency side. However, due to limitations in circuit design, such as settling of the output signal of the signal amplifier circuit, the frequency of the chopper clock in the current sensor is around 500 kHz at most. As understood from this, in the current sensor, the ratio between the frequency bandwidth of the target signal to be detected and the frequency of the chopper clock is only around five times at most. Accordingly, when the DC offset voltage is detected in the ripple-shaped noise, the possibility of mixing the detected DC offset voltage with the target signal to be detected will increase. In view of this, in the current sensor used for current detection of an inverter, in order to detect the DC offset voltage from ripple-shaped noise, it is necessary to design a more accurate method, rather than a simple method such as capacitive coupling by use of capacitors as disclosed in Non Patent Document 4.
Further, the technique described in Patent Document 5 is disclosed as a continuous-time analog switch which performs switching control of the polarity of a differential signal between inversion and non-inversion, by use of a “third chopping circuit” which modifies the signal coupling of the differential signal. In contrast, the difference calculation circuit of the present invention performs discrete time sampling of the input signal to the difference calculation circuit, so as to calculate the time variation of the input signal in a given period of time.
Further, in the technique described in Patent Document 6, “DEM1” indicates a chopper demodulator, however, this technique is different from the configuration of the present invention in that feedback is performed before “DEM1” as the chopper demodulator.
Differences between the present invention and Non Patent Document 5 and Patent Document 6 described above are investigated below. In Non Patent Document 5 and Patent Document 6, a feedback circuit is prepared for the purpose of offset cancellation. However, the feedback in Non Patent Document 5 and Patent Document 6 is essentially different from that of the present invention in that the feedback is performed from a signal node before the hall electromotive force signal modulated by a chopper clock is demodulated by a demodulator. Further, Patent Documents 7 and 8 do not disclose anything about the characteristic configuration of the present invention at all.
Further, when the hall electromotive force signal detection circuit is used in an environment with high-frequency noise, such as for an inverter, the main signal path of the hall electromotive force signal detection circuit is preferably a circuit that is able to perform continuous-time signal processing, rather than discrete-time signal processing, in order to prevent fold over noise by discretization in time (sampling).
The present invention is accomplished in view of such circumstances, and an object of the present invention is to provide a hall electromotive force signal detection circuit and a current sensor thereof each of which has excellent accuracy, wide band characteristics, and fast response and further has a main signal path of the circuit configurable by use of a continuous-time signal processing circuit.
The present invention is accomplished in order to achieve the above object. According to an aspect of the present invention, there is provided a hall electromotive force signal detection circuit configured to suppress an offset signal component in a signal composed of a hall electromotive force signal generated in a Hall element and the offset signal component, and amplify the hall electromotive force signal to generate an output signal, the hall electromotive force signal detection circuit including: a signal amplifier circuit configured to amplify an output of the hall element and output a signal having the offset signal component modulated at a frequency of a chopper clock; a difference calculation circuit configured to sample a component synchronous with the chopper clock out of an output signal of the signal amplifier circuit at a timing synchronous with the chopper clock, and output a difference between sampling results which are sampled at different time; an integrating circuit configured to integrate the output of the difference calculation circuit in the time domain; and a feedback circuit configured to feedback an output signal of the integrating circuit to the signal amplifier circuit.
Further, the hall electromotive force signal detection circuit includes a first switching circuit configured to perform switching operations on terminal pairs of the Hall element in which a terminal pair for driving a driving current to the hall element and a terminal pair for detecting the hall electromotive force signal are interchanged, and to modulate the hall electromotive force signal generated in the hall element at the frequency of the chopper clock. The signal amplifier circuit amplifies the output signal from the first switching circuit and the output signal of the signal amplifier is modulated at the frequency of the chopper clock, so that the hall electromotive force signal is demodulated to a low-frequency component including a direct current component, and outputs a signal having the offset signal component modulated at the frequency of the chopper clock.
Further, the signal amplifier circuit includes: a first transconductance element configured to convert an output voltage signal of the first switching circuit into a current so as to generate a first current; resistors configured to perform voltage-division on a component included in the output voltage signal of the first switching circuit at a predetermined ratio; a second switching circuit configured to invert the polarity of a voltage generated by the voltage-division according to the chopper clock; and a second transconductance element configured to convert an output voltage from the second switching circuit into a current so as to generate a second current. The feedback circuit includes a third transconductance element configured to convert an output voltage from the integrating circuit into a current so as to generate a third current, and with regard to a current sum of the first current I1, the second current I2, and the third current I3, the feedback circuit performs feedback operation in which a DC component included in the sum of the first current and the second current in the signal amplifier circuit is added with the third current such that the current sum becomes zero (I1+I2+I3=0).
Further, a third switching circuit configured to modulate an input signal at the frequency of the chopper clock, and an output stage coupled to the third switching circuit are provided in a subsequent stage of the first transconductance element.
Further, a fourth transconductance element is provided between a subsequent stage of the third switching circuit and the difference calculation circuit.
Further, the difference calculation circuit and the integrating circuit are switched capacitor circuits.
Further, the difference calculation circuit is composed of AD converters, and the integrating circuit is constituted of digital circuits.
Further, the AD converter is a ΔΣ-modulation AD converter.
Further, the AD converter is a double-integral AD converter.
Further, according to another aspect of the present invention, there is provided a current sensor using the above-mentioned hall electromotive force signal detection circuit.
According to the present invention, it is possible to provide a hall electromotive force signal detection circuit and a current sensor thereof each of which has excellent accuracy, wide band characteristics, and fast response and further has a main signal path of the circuit is configurable by a continuous-time signal processing circuit.
With reference to drawings, the following describes embodiments of the present invention.
The above-mentioned description deals with offset cancellation technique by use of the driving method (the spinning current method) of a hall element using a chopper clock. As to the hall element that performs the spinning current method, the following describes a configuration of a hall electromotive force signal detection circuit, which will be a premise of the present embodiment, with reference to
In the hall electromotive force signal detection circuit illustrated in
Note that, in
Note that in the hall electromotive force signal detection circuit illustrated in
In the hall electromotive force signal detection circuit illustrated in
The transistor differential pair Gm, 2 in the signal amplifier circuit 3 is driven by the feedback voltage Vfb represented by Expression 3 from the output voltage Vout. However, since there exists an offset Vos(Gm, 2) in the transistor differential pair Gm, 2, a current I2 output from Gm, 2 is represented by Expression 4.
Note that, in
In the hall electromotive force signal detection circuit illustrated in
Here, as represented by Expression 5, it should be noted that the signal Vsig (B) which is modulated by the chopper clock of the chopper clock generation circuit 4 is demodulated by the third switching circuit 34 in
The polarity of the part which is placed between parentheses “{“and”}” in Expression 5 is inverted every time the phase of the chopper clock is switched between ϕ1 and ϕ2. As can be seen from this, in the hall electromotive force signal detection circuit illustrated in
Further, in a case where presence of the above-mentioned ripple-shaped noise is removed from Expression 5, Expression 6 is obtained. It is shown that, in the hall electromotive force signal detection circuit illustrated in
With reference to drawings, the following describes embodiments of the present invention.
The present embodiment is a hall electromotive force signal detection circuit for generating an output voltage signal obtained by amplifying a hall electromotive force signal generated in a hall element, with a predetermined signal amplification gain. The first switching circuit 12 is configured to perform switching operations on the terminal pairs of the hall element 11 in which a terminal pair for driving a driving current to the hall element 11 and a terminal pair for detecting a hall electromotive force signal are interchanged, and to modulate a hall electromotive force signal generated in the hall element 11 at the frequency of the chopper clock.
Further, the chopper clock generation circuit 14 generates a chopper clock for periodically driving the first switching circuit 12. Further, the signal amplifier circuit 13 amplifies an output voltage signal from the first switching circuit 12.
Further, the difference calculation circuit 15 samples a signal component synchronized with the chopper clock generated by the chopper clock generation circuit 14, out of the output voltage signal of the signal amplifier circuit 13, at a timing obtained from the chopper clock, so as to detect the signal component. Further, the integrating circuit 16 integrates an output of the difference calculation circuit 15 in the time domain. Further, the third transconductance element 17 converts an output voltage of the integrating circuit 16 into a current to generate a third current.
Further, the signal amplifier circuit 13 includes: the first transconductance element 131 for converting a hall electromotive force signal into a current so as to generate a first current; resistors R1 and R2 for performing voltage-division on the output voltage signal at a given ratio; the second switching circuit 132 for reversing a polarity of a voltage generated by the voltage-division, according to the chopper clock; the second transconductance element 133 for converting an output voltage of the second switching circuit 132 into a current so as to generate a second current; the third switching circuit 134 provided in the subsequent stage of the first transconductance element 131 and performing a demodulation operation at a chopper clock frequency; and the output stage 135 coupled to this third switching circuit 134.
With such a configuration, an output voltage signal from the integrating circuit 16 is fed back to the signal amplifier circuit 13 via the third transconductance element 17. That is, with regard to a current sum of the first current I1, the second current I2, and the third current I3, feedback is performed on a DC component included in a sum of the first current and the second current via the third transconductance element 17 which converts an output voltage from the integrating circuit 16 into a current to generate the third current, so that the sum including the third current becomes zero (I1+I2+I3=0).
Further, in the present embodiment, a current sensor may be constituted of the above-mentioned hall electromotive force signal detection circuit.
(Feedback Circuit for Suppressing Occurrence of Ripple)
The following describes the hall electromotive force signal detection circuit according to the present embodiment with reference to
This feedback circuit is constituted of the difference calculation circuit 15, the integrating circuit 16, and the third transconductance element 17. The current I3 output from this third transconductance element 17 acts to cancel a component Ios of a DC offset current included in a current I1+I2=Gm, 1·Vhall+Gm, 2·Vfb, represented by Expression 7.
Note that, in
[Math. 7]
DC Offset Current Ios Causing Ripple-Shaped Noise Ios=Gm,1·Vos(Hall)+Gm,1·Vos(Gm,1)+Gm,2·Vos(Gm,2) Expression 7:
The DC offset current Ios, represented by Expression 7, is a DC offset current resulted from the voltage to current conversion in which an offset voltage of the hall element, offset voltages in the transistor differential pair (Gm, 1) 131, and the transistor differential pair (Gm, 2) 133 are converted to an current signal by use of the transistor differential pair. When this DC offset current Ios is modulated at a frequency of a chopper clock in the third switching circuit 134, ripple-shaped noise (
(Difference Calculation Circuit)
For the purpose of detecting a component of a chopper clock frequency from the ripple-shaped noise in the signal amplifier output signal Vout with high accuracy, the above-mentioned feedback performs time discretization (sampling) on the signal amplifier output signal Vout at a timing just before the chopper clock changes.
That is, Vout(ϕ1) is detected by sampling at a timing just before a chopper clock phase changes from ϕ1 to ϕ2, Vout(ϕ2) is detected by sampling at a timing just before the chopper clock phase changes from ϕ2 to ϕ1, and a difference signal Vout(ϕ1)−Vout(ϕ2) between Vout(ϕ1) and Vout(ϕ2) is calculated as a discrete-time signal. Thus, it is the function of the difference calculation circuit 15 in the present embodiment to perform a calculation of the time variation of Vout between two timing separated at a certain time period (calculation corresponding to time derivatives).
(Integrating Circuit in Subsequent Stage of Difference Calculation Circuit)
In the example in
As described later, in the hall electromotive force signal detecting apparatus of the present embodiment, since the difference calculation circuit 15 is a circuit for performing time discretization (sampling), a switched capacitor circuit and a digital circuit can be used for the integrating circuit 16, and, as a result, various desirable features can be realized.
(Hall Electromotive Force Signal Detection Circuit According to Present Embodiment)
As for the circuit configurations which suppress the occurrence of ripple-shaped noise in the output of the signal amplifier circuit 13 by feedback means, these circuit configurations are described in Non Patent Document 4 and Patent Document 5 which are mentioned before. Although, each of these documents deals with a circuit of capacitive coupling using capacitors and a circuit to invert its signal polarity with switch circuit as a circuit measure for detecting a DC component in the ripple-shaped noise, there is no description in these documents about the concept of a difference calculation circuit which is a characteristic configuration of the present embodiment.
It is explained in the following that the difference calculation circuit, which is the characteristic configuration of the present embodiment, makes it possible to achieve high accuracy of feedback that suppresses the occurrence of ripple-shaped noise in an output of a signal amplifier circuit. Particularly, the difference calculation circuit 15, which is the characteristic configuration of the present embodiment, can be a suitable circuit configuration in the applications such as current sensor designed to detect a switching current of an inverter, where a frequency bandwidth of a target magnetic field signal to be detected is relatively wider than the chopper clock frequency.
Under these circumstances, in the case of Non Patent Document 4 and Patent Document 5, the unbalance between the period of rising voltage and the period of falling voltage in the ripple-shaped noise causes an error in feedback to suppress the occurrence of the ripple-shaped noise. As the chopper clock frequency is increased to a higher frequency, such unbalance becomes more significant.
In contrast to this, in the difference calculation circuit, which is the characteristic configuration of the present embodiment, sampling is performed at a timing just before the chopper clock changes. In view of this, as far as the ripple-shaped noise settles into a constant value within the half of the chopper clock period, the difference calculation circuit is not affected by such unbalance between the rising period and the falling period.
In such a case, a signal Vsig (B) corresponding to the time variation of the target magnetic field B to be detected is mixed into sampled Vout(ϕ1) and Vout(ϕ2) (Expression 8). Notwithstanding, the circuit configuration of the hall electromotive force signal detection circuit according to the present embodiment is a circuit configuration which can minimize the influence of Vsig (B).
In order to describe this feature of the present embodiment, the following deals with a case where a value of Vout(ϕ1) sampled in chopper clock phase ϕ1 at a time n is represented to be Vout(ϕ1) (n), similarly, a value of Vout(ϕ2) sampled in chopper clock phase ϕ2 at a time n is represented to be Vout(ϕ2) (n), and a difference signal Vout(ϕ1)(n)−Vout(ϕ2)(n) therebetween is integrated for N times with respect to the time n. Here, in a case where the target magnetic field B to be detected is a magnetic field induced by an inverter current that drives a motor, the signal Vsig(B) exhibits a sinusoidal wave form of which cycle is the same as the cycle of the inverter driving current of the motor. Accordingly, if an integration times N herein is set to be a sufficiently large number, it becomes possible to detect a ripple component (a DC offset component as a source of ripple) with high accuracy while eliminating the influence of Vsig (B), as understood from Expression 9.
Further, it is a well-known fact that switched capacitor circuits and digital circuits are suitable circuit configurations for the purpose of realizing integrating circuits and narrow-band filters.
The difference calculation circuit, which is the characteristic configuration of the present embodiment, has the following two advantages because digital circuits can be utilized in the subsequent stages of the difference calculation circuit.
The first advantage when the digital circuit is used for the circuit placed in the subsequent stage of the difference calculation circuit 15 is the reduction of an initial pull-in time of the feedback loop that suppresses the occurrence of ripple-shaped noise in the output of the signal amplifier circuit 13. As described above, this feedback loop is a feedback loop with a very narrow bandwidth, and therefore, a long time is generally required for the initial pull-in time, where the initial pull-in time means the time which is required for the feedback loop to converge and settle the ripple-shaped noise converged to zero after the initial operation of the hall electromotive force signal detection circuit is started after power-on. Here, if a digital circuit is used for the circuit in the subsequent stage of the difference calculation circuit 15, it is easy to implement, in the digital circuit, an adaptive algorithm in which the time constant of the feedback loop may be switched between the initial period before the initial pull-in operation is completed and the steady operation period after the initial pull-in operation is completed, thereby making it possible to realize the reduction of the initial pull-in time.
The second advantage obtained when a digital circuit is used for the circuit placed in the subsequent stage of the difference calculation circuit 15 is the realization of robustness of the feedback loop. The current sensor used for detecting a switching current of an inverter is used under an environment with strong disturbance noise such as electromagnetic induction noise and electrostatic induction noise which are generated at the time of the switching of the inverter. Because of this, in the current sensor used for detecting a switching current of an inverter, the disturbance noise is input into the feedback loop that suppresses the occurrence of ripple-shaped noise in the output of the signal amplifier circuit 13. However, in a case where a digital circuit is used for the circuit in the subsequent stage of the difference calculation circuit, it is possible to implement algorithms that are insusceptible to such disturbance noise in the digital circuit, thereby making it possible to realize the robustness of the feedback loop.
(Example of Difference Calculation Circuit and Integrating Circuit Using Switched Capacitor Circuits)
Further, the integrating circuit 16 is constituted of an operational amplifier 16a, capacitors C1, int and C2, int, and a switch. Vint is calculated by converting Vdiff sampled in the capacitor C1, int at the input side into a charge (C1, int*Vdiff) and transferring the charge to the capacitor C2, int to perform integral calculation (see Expression 11).
Note that the difference calculation circuit 15 and the integrating circuit 16 are switched capacitor circuits.
The difference calculation circuit 15, which is the characteristic configuration of the present embodiment, has a circuit configuration known as Peak-To-Peak circuit, and calculates Vdiff=(C1, diff/C2, diff)×(1 Vout(ϕ1)−Vout(ϕ2)) according to the transfer function Hdiff(z) of the difference calculation circuit, as represented by Expression 10.
Herein, Peak-To-Peak circuit is taken as an example of the difference calculation circuit to make explanation, but there are various configurations of the difference calculation circuits usable herein, as well as the Peak-To-Peak circuit.
The output signal Vdiff of the difference calculation circuit 15 in
(Circuit Configuration Using AD Converter, Integrator by Digital Circuit, and DA Converter)
A difference calculation circuit 15D is the M-bit AD converter 136, an integrating circuit 16D is constituted of a digital integrator 137, and the DA converter 138 constituted of the N-bit register 138a and the N-bit DA converter 138b.
According to the circuit configuration herein, the difference calculation circuit 15D in the feedback circuit according to the present embodiment is implemented with the AD converter 136, the integrating circuit 16D therein is implemented with a digital circuit constituted of the digital integrator 137 and the DA converter 138, and the output signal of the integrating circuit 16D is converted into an analog signal Vint by use of the DA converter 138b. That is, the difference calculation circuit 15D is constituted of the AD converter 136, the integrating circuit 16D is constituted as the digital circuits 137 and 138, and the voltage fed back to the third transistor differential pair (Gm, 3) 17 is generated by the DA converter 138b.
(Example Using Difference Calculation Circuit Illustrated in
In
In
As illustrated in
In the feedback circuit of the present embodiment, it is necessary to keep track of the time variation of the offset voltage of the hall element and the offset voltages of the transistor differential pairs. However, the time variation of the offset voltages that should be tracked herein is very slow, and therefore, an AD converter with a high-speed operation is not required for the AD converter. Consequently, as in the case of the exemplary 1st order ΔΣ modulator illustrated in
(Example Using Difference Calculation Circuit Including Double-Integral AD Converter Including Integrating Circuit Illustrated in
As described above, the feedback loop that suppresses the occurrence of ripples is intended to cancel a DC offset, and therefore becomes a loop with extremely narrow bandwidth. In view of this, it is also possible to use an AD converter known as double-integral type, as the AD converter. The double-integral AD converter is a type that is generally used in low-speed and high-accuracy measurement such as digital multimeter, and in the present embodiment, it is one of extremely preferable types of AD conversion as a type of the AD converter constituting the above-mentioned feedback.
The double-integral AD converter 151 has the circuit configuration including the difference calculation circuit 15 shown in
In
An AD conversion operation of the double-integral AD converter 151 will be described below, with reference to
The AD conversion result N2 obtained by the double-integral AD converter 151 is the result obtained by performing AD conversion on Vout(ϕ1)−Vout(ϕ2). This digital signal N2 is integrated by the digital integrator 137, and then converted into an analog signal Vint by use of the DA converter 138b.
Here, in the case of the double-integral AD converter 151, the bandwidth of the above-mentioned feedback can be controlled by variably controlling the integration count N1. That is, in an initial pull-in operation, the value of N1 is set to be a small value so as to reduce the pull-in time of the feedback, while the value of N1 is increased in the steady operation after the initial pull-in operation is completed, thereby making it possible to realize the robustness in which the feedback loop does not malfunction by disturbance noise.
That is, it is also possible to provide the fourth transconductance element 18 between the output of the third switching circuit 134 and the difference calculation circuit 15 so as to perform feedback. An advantage of this feedback circuit configuration is that the feedback circuit for cancelling ripples is insusceptible to the fluctuation of the load seen from the output in
As stated above, the hall electromotive force signal detection circuit according to the present embodiment includes a feedback circuit for canceling a DC offset component to cause ripple-shaped noise in the output voltage signal of a signal amplifier circuit. In view of this, it is possible to realize highly accurate magnetic field detection by dynamic offset cancellation means (a spinning current method and a chopper amplifier) for modulating an offset voltage of a hall element and an offset voltage of a signal amplifier circuit, by use of a chopper clock. Further, in comparison with a circuit configuration (e.g., see Patent Document 4) in which filters for reducing ripple-shaped noise are additionally placed, it is possible to achieve fast response as a magnetic sensor, and further, there is an advantage that no noise increase occurs due to the filters thus additionally placed. Further, the above-mentioned hall electromotive force signal detection circuit of the present embodiment is usable as a current sensor.
Number | Date | Country | Kind |
---|---|---|---|
2012-012774 | Jan 2012 | JP | national |
This application is a continuation of U.S. patent application Ser. No. 17/200,515, filed on Mar. 12, 2021, which is a continuation of U.S. patent application Ser. No. 15/250,398, filed on Aug. 29, 2016, which is a Divisional of U.S. patent application Ser. No. 14/127,280, filed on Dec. 18, 2013, which is a National Stage Entry of International Patent Application No. PCT/JP2013/000065, filed on Jan. 10, 2013, and claims priority to and the benefit of Japanese Patent Application No. 2012-012774, filed on Jan. 25, 2012, the entirety of each of which is hereby incorporated by reference.
Number | Date | Country | |
---|---|---|---|
Parent | 14127280 | Dec 2013 | US |
Child | 15250398 | US |
Number | Date | Country | |
---|---|---|---|
Parent | 17200515 | Mar 2021 | US |
Child | 17939107 | US | |
Parent | 15250398 | Aug 2016 | US |
Child | 17200515 | US |