Method and apparatus for processing a measurement signal

Abstract

A system for processing a signal s(t) from a sensor to recover sensed signal information within the bandwidth of the summation signal, wherein the signal s(t) includes a sensed signal m(t) and an offset signal having a first frequency f1. The system comprises: a sampling device for sampling the signal s(t) at a second frequency f2 that is a multiple of the first frequency f1, to create a sequence of sampled values; and an averaging device for averaging the sequence of sampled values to provide a sequence of averaged sampled values indicative of the sensed signal m(t).

Description

BACKGROUND OF THE INVENTION

The present invention relates to a technique for processing a measurement signal.

Measurement devices, such as sensors, bridge circuits, or the like, have a disadvantage that they often superpose a first offset signal on the measurement signal of interest to the user. This offset signal is due to the mechanical sensitivity of the measurement device, inaccuracies in the manufacturing technique of the measurement devices, and/or the electronics. The magnitude of this offset signal is often unknown to the user and varies in time.

Some measurement devices, such as Hall sensors or Wheatstone bridges can be driven in such a way that the offset signal is alternately added to and subtracted from the measurement signal. Therefore, the offset signal is superposed on the measurement signal as an AC signal, preferably a square-wave signal. Driving the sensors or bridge circuits in this way is often referred to as “chopping”. The frequency that the offset signal changes sign is called the “chopper frequency”.

The chopper frequency is typically selected to have a much higher value than the maximum frequency of the measurement signal, so the high frequency offset signal can be separated from the measurement signal by a low-pass filter to obtain an offset-free measurement signal.

A disadvantage of this technique is the complication associated with the circuitry involved in filtering the signal, and the fact that filtering usually creates a new offset. Therefore, there is a need for a system and method for obtaining the measurement signal without the need for filtering.

SUMMARY OF THE INVENTION

Briefly, according to the present invention, a summation signal delivered by a measurement device, for example a Hall sensor or a Wheatstone bridge, and composed of a measurement signal and an offset signal with a first frequency (chopper frequency), is sampled at the rate of a second frequency so as to provide sample values, the second frequency being a multiple of the first frequency. To create measurement values from the measurement signal, the average of at least two sample values is then formed. These sample values are formed by sampling the summation signal at the interval of a half period of the first frequency or at the interval of an odd multiple of half the period of the first frequency. In forming the average, one always uses at least one sample value for which the value of the offset is added to the measurement value and always at least one sample value for which the value of the offset is subtracted from the measurement value. The offset is thus eliminated by forming an average.

Assuming that the measurement signal changes slowly compared to the first frequency of the offset signal, forming the average of two sample values at an interval of half the period of the offset signal would have no effect on the measurement value. The time delay of the measurement values caused by forming the average, or the time delay of a measurement signal processed and formed from the sample values, relative to the original measurement signal, can often be tolerated in actual practice.

Sampling the summation signal at the rate of the second frequency, which is a multiple of the first frequency, and creating the measurement values by forming averages also at the rate of the second frequency, improves the time resolution of the measurement signal. This is especially necessary for applications in which the zero passage of the measurement signal is to be evaluated.

In one embodiment, the summation signal that includes the measurement signal and the offset signal is amplified before being sampled. This amplification is often necessary if the amplitude of the first summation signal (i.e., the output signal of the measurement device) is too small for sampling and forming an average. Amplification of the first summation signal results in a second summation signal, which is then sampled, and the average of at least two sample values is then formed.

Problems may arise when using amplifiers that add a second offset to their input signal, in the present case the summation signal includes the measurement signal and the first offset signal.

This can be dealt with by changing the sign of the input signal of the amplifier at periodic intervals, or changing the phase of the input signal by 180 degrees, at the rate of a third frequency. This causes the second offset of the amplifier to be added to the negative input signal during intervals in which the sign of the input signal is changed, and to be added to the positive input signal during intervals in which the sign of the input signal is retained.

The sign change of the input signal at periodic intervals at the rate of the third frequency is reversed by changing the sign of the output signal at periodic intervals at the rate of this same frequency, such that the sign of the output signal is changed in phase with the sign change of the input signal. An output signal is thus obtained that corresponds to the input signal amplified by the amplification factor of the amplifier, and to which the second offset of the amplifier has been superposed as a square-wave signal of the third frequency. The method technique is referred to as chopping the amplifier with the third frequency.

The third frequency of the amplifier offset signal preferably corresponds to the first frequency of the offset signal of the measurement circuit or is an even multiple of this first frequency. This ensures that during the subsequent sampling and formation of an average, the value of the second offset of the amplifier is also eliminated.

If the chopper frequency of the amplifier (the third frequency) is identical to the chopper frequency of the measurement device (the first frequency), the signal present at the output of the amplifier corresponds to the amplified measurement signal, on which a total offset signal of the first frequency is superposed. The amplitude of the total offset signal includes the amplified first offset signal of the measurement circuit and the second offset signal of the amplifier. This total offset signal is eliminated by forming an average following the sampling.

The amplification of the amplifier is preferably regulated according to the amplitude of the input signal, so that the amplitude of the amplifier output signal remains constant, and thus essentially independent of the amplitude of the input signal.

In one embodiment, a system according to the present invention includes a sampling device for sampling an input signal, which depends on the summation signal, at the rate of a second frequency to produce a sequence of sample values. The sequence of sample values are input to a device for forming an average of at least two sample values.

The device for forming an average preferably has a memory device for storing sample values, and an adder for adding at least two memory values (i.e., past values) or for adding a memory value (i.e., a past value) and a sample value present at the input of the device.

The number of memory cells of the memory device preferably corresponds to the number of sample values obtained within half a period of the first frequency of the first offset signal.

In another design of the invention, the memory cells store the sample values of an entire period of the offset signal. To form the average, the values of two memory cells are here always added, these values having been obtained within a half period of the offset signal.

These and other objects, features and advantages of the present invention will become apparent in light of the following detailed description of preferred embodiments thereof, as illustrated in the accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWING

FIG. 1 is a plot of a measurement signal, a summation signal and a sequence of sampled measurement values as a function of time;

FIG. 2 is a functional block diagram illustration of a system for processing a measurement signal according to the present invention;

FIG. 3 is a functional block diagram illustration of an alternative embodiment system for processing a measurement signal that includes an amplifier;

FIG. 4 is a schematic illustration of an amplifier suitable for the amplifier of FIG. 3 ;

FIGS. 5A-5C are plots of the amplifier input and output signals as a function of time;

FIG. 6 is a plot of various system signals as a function of time; and

FIG. 7 is a functional block diagram of a device for averaging sequence signal values.

DETAILED DESCRIPTION OF THE INVENTION

FIG. 1 is a plot of a measurement signal m(t), a summation signal s 1 (t) and a sequence of sampled measurement values w 1 (k) as a function of time. Measurement devices, such as sensors, Wheatstone bridges, or the like, which are used to measure such signals, frequently have the disadvantage that they superpose a first offset signal on the measurement signal m(t) of interest to the user. The offset signal is often caused by the mechanical sensitivity of the measurement circuit, inaccuracies in the manufacturing technique of the measurement equipment, and/or the electronics. Some sensors (e.g., Hall sensors or Wheatstone bridges) used for measurement purposes can be driven such that essentially constant first offset signals are alternately added to and subtracted from the measurement signal m(t) at a first frequency f 1 . In the case of Hall sensors, for example, the measurement equipment can be driven in this way by feeding current to the sensor alternately at different inputs.

Referring to FIG. 1 , the output signal of a measurement device driven in this way is the first summation signal s 1 (t), which is the sum of the measurement signal m(t) and the square-wave offset signal that has a signal amplitude ot 1 and a frequency equal to a first frequency value f 1 .

The first frequency f 1 may be referred to as a chopper frequency of the measurement device, had has a frequency value that is greater than the maximum frequency of the measurement signal m(t). Ideally, the highest frequency within the bandwidth of the measurement signal m(t) and the first frequency f 1 of the offset signal differ by a factor of ten or more. Referring still to FIG. 1 , T 1 designates the period of the offset signal, where T 1 =1/f 1 .

According to an aspect of the present invention, the first summation signal s 1 (t) is sampled at regular intervals T 2 =1/f 2 , where the sampling frequency f 2 is preferably an integer multiple of the first frequency value f 1 . As shown in FIG. 1 , the first frequency f 1 and the second frequency f 2 differ by a factor of 8 (i.e., f 2 =8*f 1 ) so the first summation signal s 1 (t) is sampled eight times during one period T 1 of the offset signal to provide eight sampled signal values 20 - 27 (FIG. 1 ). The sequence of these sample values is designated as a 1 (k), where a 1 (k)=s 1 (k·T 2 )and k is an integer.

To eliminate the first offset ot 1 from the sequence of sampled values a 1 (k), the average of at least two sample values a 1 (k) is formed, wherein the sample values were obtained during a half period T 1 /2 or an odd multiple of the half period T 1 /2 of the offset signal. For example, referring still to FIG. 1 , w 1 (k) represents a sequence of averages that were obtained by averaging two sample values from the sequence a 1 (k) at an interval of a half period T 1 /2. The sequence w 1 (k) can be expressed as w 1 (k)=0.5[a 1 (k)+a 1 (k−T 1 /(2·T 2 ))], which can be rewritten as w 1 (k)=0.5[a 1 (k)+a 1 (k−n)], where n is the number of sample values created during the half period T 1 /2 of the first offset signal.

The error that the averaging step imposes on the actual measurement value of the measurement signal m(t) can be neglected, assuming that the minimum period of the measurement signal m(t) is much longer than the half period T 1 /2, which specifies the time spacing of the sample values a 1 (k), a 1 (k−n) to form the average, and also assuming that the slope of the measurement signal m(t) can be taken as approximately constant during the interval T 1 /2.

The step of forming an average causes the sequence of the measurement values w 1 (k) to be time shifted by about T 1 /4 relative to the measurement signal m(t). This constant time shift can be taken into account during the subsequent analysis of the measurement values, and consequently is non-critical for most applications.

The greater the sampling frequency f 2 , the more measurement values w 1 (k) will be generated per unit time by the sampling process and by the formation of an average, and the greater the time resolution that can be achieved by the inventive method. For many applications, the moment of the zero passage of the measurement signal m(t) or of the sequence of measurement values w 1 (k)—from which one can conclude to the measurement signal m(t) by taking into account the time shift due to the formation of an average—is of special interest. The moment of the zero passage is here delimited by the two measurement values of the measurement value sequence w 1 (k), between which a sign change has occurred. The maximum time difference here is 1/T 2 and increases with increasing sampling frequency f 2 .

FIG. 2 is a functional block diagram illustration of a system for processing the measurement signal m(t). The system 30 includes a measurement device S that provides the first summation signal s 1 (t). As set forth above, the first summation signal s 1 (t) includes the superposition of the measurement signal m(t) and of an—at least approximately—square-wave offset signal of a first frequency value f 1 and an amplitude value ot 1 . To superimpose the AC offset signal on the measurement signal m(t), the sensor S receives a clock pulse signal TS operating at first frequency f 1 , by which the measurement circuit S is “chopped”.

The first summation signal s 1 (t) is input to a sampling device AV, which samples the signal at intervals of the period T 2 =1/f 2 and outputs sampled values a 1 (k) at a cycle time T 2 given by the sampling process. The sequence of sampled values a 1 (k) is input to an averaging device MW that forms the average of two sampled values a 1 (k), a 1 (k−n) obtained by sampling during a half period T 1 /2 of the offset signal or during an odd multiple of the half period T 1 /2 of the offset signal.

FIG. 3 is a functional block diagram illustration of an alternative embodiment system for processing a measurement signal that includes an amplifier V. Since the amplitude of the summation signal s 1 (t) often is too small for the signal to be properly sampled, this system includes an amplifier V that receives the summation signal s 1 (t) and provides a second summation signal s 2 (t). The second summation signal S 2 (t) is input to the sampling device AV with the cycle period T 2 to provide a sequence of sampled values a 2 (k). The sequence of sampled values a 2 (k) is input to the averaging device MW to generate a sequence of values w 2 (k) formed by averaging of two sample values a 2 (k), a 2 (k−n).

A disadvantage of the embodiment illustrated in FIG. 3 is that the amplifier V may add an undesirable second offset to the summation signal s 1 (t). To avoid introducing this second offset, the amplifier V may be configured as the amplifier circuit 60 illustrated in FIG. 4 .

The amplifier circuit 60 includes an amplifying element OP (e.g., op-amp), which amplifies an inputted signal s 11 (t) by a factor v and undesirably adds a second offset ot 2 to the amplified signal. To remove the second offset, a first multiplication circuit MUL 1 receives the input signal s 1 (t) and is multiplied by either the factor “+1” or “−1” as determined by a clock pulse signal TS 2 . This shifts the phase of the signal by 180 degrees, according to the specification of the clock pulse signal TS 2 . The phase shift of the input signal s 1 (t) is reversed by a second multiplication circuit MUL 2 , which is connected in series after the amplification element OP. The amplification element OP provides an output signal s 21 (t) that is multiplied by the factor “+1” or “−1” according to the specification of the clock pulse signal TS 2 . The multiplication circuits MUL 1 , MUL 2 have a switch, which can be switched over in the cycle of the clock pulse signal TS 2 . This switch alternately applies the value “+1” or “−1”. The switch is switched over and the phase of the input signal s 1 (t) and of the output signal s 21 (t) is shifted, preferably at the rate of the first frequency f 1 or at the rate of an even multiple of the first frequency.

FIGS. 5A-5C are plots of the amplifier input and output signals as a function of time. Specifically, FIG. 5A illustrates the input signal s 1 (t) as a function of time, while FIG. 5B illustrates the signal s 11 (t) output from the first multiplier MUL 1 , and the signal s 21 (t) output from the amplifier, both as a function of time. Due to the multiplication by “1”, the signal s 21 (t) coincides section by section with the input signal s 1 (t), and, due to the multiplication by “−1”, in the other sections it mirrors the input signal s 1 (t) in the time axis. The amplifier OP amplifies its input signal s 11 (t) by an amplification factor v and adds an offset of amplitude ot 2 , resulting in the signal s 21 (t). In the interest of ease of illustration, an amplification factor of unity was selected for the purposes of FIGS. 5A-5C . However, one of ordinary skill will recognize that the amplification factor of unity provides no gain, and that in practice amplification factors typically exceed one-hundred (i.e., v=100).

After the phase shift of the signal s 1 (t) is reversed by an appropriate phase shift of the signal s 21 (t) by the multiplier MUL 2 , one obtains the output signal s 2 (t) illustrated in FIG. 5C , which appears as an amplified input signal vs 1 (t), on which a square-wave offset signal with an amplitude ot 2 is superposed.

If an output signal of the measurement device S is selected as the input signal s 1 (t) for the amplifier V, which consists of a measurement signal m(t) superposed with an offset signal of the first frequency f 1 due to the chopping of the measurement device S, and if the phase of the signal s 1 (t) or s 21 (t) is shifted at the rate of the first frequency f 1 and in-phase with the first offset signal, then the output signal s 2 (t) of the amplifier V will be the amplified measurement signal vm(t) with a superposed offset signal of frequency f 1 and amplitude vot 1 +ot 2 , as illustrated in FIG. 6 .

The remaining processing of this signal s 2 (t) corresponds to that of the unamplified signal s 1 (t) explained in connection with FIGS. 1 and 2 . The signal s 2 (t) is sampled at the rate of the second frequency f 2 to produce a sequence of sample values s 2 (k). The average is always formed of two sample values obtained over a half period T 1 /2 of the offset signal, to provide a sequence of measurement values w 2 (k) in synchronism with the sample frequency f 2 that corresponds to measurement values of the measurement signal m(t), which are amplified by a factor v, where the values w 2 (k) being time-shifted relative to the actual measurement values approximately by T 1 /4.

If the amplifier is chopped with a third frequency, which is an even multiple of the first frequency f 1 , the subsequent sampling process and formation of an average likewise eliminates the second offset ot 2 , which was inserted by the amplification element OP. The chopping of the amplifier need not be in phase with the first offset signal.

Amplification elements, such as operational amplifiers, frequently have an inverting and a non-inverting input as well as an inverting and a non-inverting output. The chopping of the amplifier can be effected with such amplifiers by applying the input signal alternately to the inverting and non-inverting input, according to the specification of a clock pulse signal, and by tapping the output signal alternately from the inverting and non-inverting output, according to the specification of the same clock pulse signal.

FIG. 7 is a functional block diagram of a device MW for averaging sequence signal values. The device includes a memory device SP, preferably a shift register, in which a number of sample values of the sample sequence a 1 (k), a 2 (k) can be stored. The number of memory cells preferably corresponds to the number n of sample values, which are created in the time interval T 1 /2, which lies between the sample values a 1 (k), a 1 (k−n); a 2 (k), a 2 (k−n), which were used to form an average. The device also includes an adder that adds the current sample value a 1 (k); a 2 (k) and the previously created and stored sample value a 1 (k−n); a 2 (k−n). The output signal of the adder is multiplied by the factor 1/2 to form a correct average. Since amplification of the measurement values is frequently desired in any case, this multiplication by 1/2 can also be omitted.

Although the present invention has been shown and described with respect to several preferred embodiments thereof, various changes, omissions and additions to the form and detail thereof, may be made therein, without departing from the spirit and scope of the invention.

Claims

1. A method of processing a summation signal s(t) from a sensor to recover sensed signal information within the bandwidth of the summation signal, wherein the summation signal s(t) includes a sensed signal m(t) and a first offset signal having a first frequency f1, the method comprising the steps of:sampling the summation signal s(t) at a second frequency f2 that is a multiple of the first frequency f1, to create a sequence of sampled values; and averaging said sequence of sampled values to provide a sequence of averaged sampled values indicative of the sensed signal m(t).

2. The method of claim 1, comprising the step of amplifying said summation signal s(t) prior to said steps of sampling and averaging.

3. The method of claim 1, wherein said step of averaging includes summing two signals within said sequence of sampled values during a half period (T1/2) of the offset signal or an odd multiple of the value T1/2, wherein T1 is the period of the offset signal.

4. The method of claim 2, wherein said step of amplifying includes:shifting the phase of said summation signal s(t) by 180 degrees at periodic time intervals given by a third frequency value, and providing a signal s11(t) indicative thereof; amplifying said signal s11(t) by an amplification factor, and providing an amplified signal s21(t) indicative thereof; and shifting the phase of said amplified signal s21(t) by 180 degrees at periodic time intervals given by said third frequency.

5. The method of claim 4, wherein said third frequency value is identical to said first frequency value f1 or an even multiple thereof.

6. The method of claim 5, wherein the phase shift of said summation signal s1(t) occurs in phase with said first offset signal.

7. A system for processing a signal s(t) from a sensor to recover sensed signal information within the bandwidth of the summation signal, wherein the signal s(t) includes a sensed signal m(t) and a first offset signal having a first frequency f1, said system comprising:means for sampling the signal s(t) at a second frequency rate f2 that is a multiple of the first frequency f1, to create a sequence of sampled values; and means for averaging said sequence of sampled values to provide a sequence of averaged sampled values indicative of the sensed signal m(t).

8. The system of claim 7, wherein said means for sampling comprises an amplifier that amplifies said summation signal s(t) prior to sampling.

9. The system of claim 8, wherein said amplifier comprises:means for shifting the phase of said summation signal s(t) by 180 degrees at periodic time intervals given by a third frequency value, and providing a signal s11(t) indicative thereof; means for amplifying said signal s11(t) by an amplification factor, and providing an amplified signal s21(t) indicative thereof; and means for shifting the phase of said amplified signal s21(t) by 180 degrees at periodic time intervals given by said third frequency to provide a signal indicative of said sampled values.