JPH04365542A - Automatic offset correction system for sensor signal - Google Patents

Abstract

PURPOSE:To perform accurate offset correction even when offset of a signal from a sensor is fluctuated owing to a fluctuation in temperature and an object to be detected is moved at a low speed. CONSTITUTION:A DC component is removed by high-pass filters 2a and 2b by means of phase A and Phase b signals SA and SB from a sensor generated by differential amplifiers 1a and 1b. From the signals generated through the high-pass filters 2a and 2b, a zero cross point is detected by means of comparators 5a and 5b. Phase A and phase B signals SA and SB at the zero cross point are inputted to offset correction calculating circuits 7a and 7b to determine an offset correction value. By means of the correction value, the phase A and phase B are corrected by adding circuits 10a and 10b. Even when offset of a signal from the sensor is fluctuated, since the fluctuated offset is detected for correction, constantly accurate offset correction is practicable.

Description

[Detailed description of the invention]

0001

[Industrial Application Field] The present invention relates to a detector for detecting the position and speed of a rotating shaft such as a motor or a main shaft of a machine tool, or a moving object.
The present invention relates to an automatic offset correction method for a sensor signal of a detector that generates a sine wave signal and uses the sine wave signal to detect the position and speed of a rotating axis or a moving object.

0002

[Prior Art] In a detector that detects the moving speed and position of a rotating shaft or a moving object, when the rotating shaft or moving object moves, the sensor outputs A-phase and B-phase sine waves with a 90 degree phase difference. It is known to detect the speed and position of a rotating shaft or a moving object from the A-phase and B-phase sine waves. In this case, if there is an offset in the A-phase and B-phase sine waves (if the neutral point voltage at the center of the amplitude deviates from the set value), accurate speed and position cannot be detected. Therefore, offset correction of the sensor signal must be performed.

Conventionally, this offset correction has been carried out by adjusting an offset adjustment variable resistor provided in a differential amplifier that amplifies the difference between a sine wave signal and its inverted signal, and adjusting the amplitude of the sine wave of the output voltage of the differential amplifier. A method is adopted in which the center becomes the neutral point voltage and the offset becomes 0V.

[0004] In the above method, the offset value is determined by the temperature at the time the offset correction is performed, the speed of the rotating axis, the moving object, etc., and is fixed, and the offset value shifts due to temperature changes and ceases to be 0V. This has the disadvantage that stable detection cannot be achieved because the detection speed and position also fluctuate.

[0005] Therefore, the applicant of the present application has proposed a method for automatically correcting the offset value of the sine wave signal from the sensor even if the offset value of the sine wave signal from the sensor fluctuates without being affected by temperature etc. We proposed a method to automatically correct the offset by sampling the signal and calculating the average of the sampled data to obtain a correction value, and then correcting the sine wave signal from the sensor using this correction value (Patent application No. 3-45335).

[0006]

The above-mentioned offset correction method using sampling data has the disadvantage that an accurate correction value cannot be determined when the rotating shaft or moving object moves at a low speed. For example, if the speed is slow,
If only half a period, 1.5 periods, etc. of a sine wave can be sampled, the average of the sampled data is used as the correction value, so this correction value does not indicate the value of the neutral point of the sine wave, so it is not possible to accurately offset the sine wave. There is a problem that correction cannot be performed. Furthermore, since a large amount of sampling data is required, there is also the problem that the configuration becomes complicated and expensive.

SUMMARY OF THE INVENTION Therefore, an object of the present invention is to provide an automatic offset correction method for sensor signals that can perform accurate offset correction even when a detected object is moving at a low speed.

[0008]

[Means for solving the problem] In an offset correction method for a detector that generates a sine wave signal and A-phase and B-phase signals of the sine wave signal having a phase difference of 90 degrees from the sine wave signal due to the movement of a moving object. , the present invention determines the neutral point (amplitude center position) of the signal from which the DC component of the A-phase and B-phase sine wave signals has been removed, and corrects the values of the A-phase and B-phase at the neutral point, respectively. offset correction is automatically performed by correcting the above correction values for the A-phase and B-phase signals, respectively.

[0009]

[Operation] The neutral point of the signal from which the DC component of the sine wave signal has been removed represents the amplitude center position of the sine wave, so the amplitude center of the sine wave signal of the A-phase and B-phase signals from the sensor at the neutral point It is the position and indicates the value of the DC component. The value of this DC component is an offset value, and if the A-phase and B-phase sine wave signals are corrected using this value as a correction amount, a signal from which the offset has been removed can be obtained.

0010

DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 is a block diagram of an automatic offset correction circuit according to an embodiment of the present invention, where MA is an A-phase sine wave signal from a sensor and RMA is an inverted signal of the sine wave signal MA. Furthermore, MB is a B-phase sine wave signal from the sensor, and RMB is its inverted signal. 1a and 1b are differential amplifiers, and the differential amplifier 1a amplifies the difference between the A-phase sensor signal MA and the A-phase sensor inverted signal RMA. Differential amplifier 1b amplifies the difference between B-phase sensor signal MB and B-phase sensor inverted signal RMB. VM is the reference voltage and each sensor signal MA, RMA
, MB, RMB (in the example, VM = Vcc/2), and the output signal S of the differential amplifier 1a
A and the output signal SB of the differential amplifier 1b, respectively. For example, MA=(k1+VM)+sinθ RMA=(k2+VM)+sin(θ+180)(k
1, k2 is the offset amount of each signal), then the A-phase signal SA is (amplification factor 1x in the example), SA = (MA-RMA) + VM
=α+2sinθ+VM
…(1) (Also α
=k1-k2 is the offset amount). Similarly,
The B-phase signal SB is: SB = β + 2 cos θ + VM
…(2)(
Note that β is the offset amount).

Reference numerals 2a and 2b are high-pass filters that remove DC components from the A-phase signal SA and B-phase signal SB, respectively, using a capacitor C, and add "Vcc/2" to them. Note that the voltage Vcc is set to 2VM. Also, when the frequency is f (Hz), the gain is -3db, and when the output is below this frequency f, the capacitor C and resistor R are connected so that they are between the upper limit value Vh and lower limit value VL of window comparators 4a and 4b, which will be described later. Value is set. 4a
, 4b are window comparators, and the outputs VA and VB of the high-pass filters 2a and 2b are set to the upper limit value V, respectively.
High level signal Ha when between H and lower limit value VL
, Hb. That is, VH ≧VA
, VB ≧VL, high level signals Ha, H
Output b.

Further, 5a and 5b are comparators, and outputs VA and V of the high-pass filters 2a and 2b, respectively.
When B is higher than the reference voltage VM (=Vcc/2), high level signals Ja and Jb are output to the AND circuits 8a and 8b. The output signal Ja of the comparator 5a, the inverted signal of the output Jb of the comparator 5b, and the inverted signal of the output Hb of the B-phase window comparator 4b are input to the AND circuit 8a. Further, the AND circuit 8b receives the output signal Jb of the comparator 5b, the inverted signal of the output Ja of the comparator 5a, and the inverted signal of the output Ha of the A-phase window comparator 4a. 6a and 6b are the outputs SA and SB of the differential amplifiers 1a and 1b.
This is an A/D converter that converts the signal into a digital signal.

7a and 7b are offset correction value calculation circuits, which add N registers and the values stored in the N registers, divide by N to obtain an average value, and calculate the complement of this average value as an offset correction value. It has a circuit for outputting data to D/A converters 9a and 9b. When the output signals of AND circuits 8a and 8b rise, the value of the register is shifted by one, and the A/D
The values of converters 6a and 6b are respectively taken into registers,
The complement of the average value of the values stored in the register is output as an offset correction value. That is, AND circuit 8
When the output signals of a and 8b rise, the value of register Rn-1 is stored in register Rn, and similarly, Rn-1 is set to Rn-1.
Value of n-2... Store the value of R1 in R2, store the output of the A/converters 6a, 6b in R1, add the values of each register R1 to Rn, divide by N to find the average, and calculate the average value. The value of each bit of the average value is reversed and "1" is added to the least significant bit to obtain the complement, which is used as an offset correction value. This correction value is converted into an analog signal by D/A converters 9a and 9b, and the addition circuit 10a
, 10b.

Addition circuits 10a and 10b combine the above correction values with A-phase signals SA and SB of the outputs of differential amplifiers 1a and 1b.
and subtracts the reference voltage VM, and outputs offset-corrected A-phase and B-phase signals NA and NB.

[0015] When the speed of the rotating shaft or moving body to which the sensor is attached is low, the sensor signals MA, RMA, MB, RMB
2 (
As shown by the broken lines in a) and (b), the high-pass filter 2
The output signals VA, VB of a, 2b do not exceed the upper limit values VH, VL of the window comparators 4a, 4b. Therefore, the output signal Ha of the comparators 4a and 4b
, Hb never go to low level, and the output of the A/D converter is never taken into the offset correction value calculation circuits 7a and 7b. However, if the speed of the detected object increases and the frequency of the sensor signal exceeds the frequency f, then
), (b), the high-pass filter 2a
, 2b exceed the upper limit values VH and VL of the window comparators 4a and 4b near 90 degrees and 270 degrees of the respective sine waves, and at this time, the
), the output signals Ha and Hb of the window comparators 4a and 4b become low level. On the other hand, since the A-phase signal SA and B-phase signal SB have a 90 degree phase difference, the B-phase signal VB passes through the neutral point position when the sine wave of the A-phase signal VA is near 90 degrees or 270 degrees. . That is, crossing the reference voltage VM is hereinafter referred to as a zero crossing. Also,
The zero-crossing point is called the zero-crossing point. Similarly, when the sine wave of the B-phase signal VB approaches 90 degrees or 270 degrees, the A-phase signal VA crosses zero. Further, the output signals of the comparators 5a and 5b, which compare the reference voltage VM with the A-phase signal VA and the B-phase signal VB, respectively, change at the zero cross point, and the A-phase signals VA, 5B exceed the reference voltage VM.
When the B-phase signal VB is reached, a high level signal is output as shown in FIGS. 2(e) and 2(f).

As a result, the output Ja of the comparator 5a,
The output of the AND circuit 8a which inputs the inverted signal of the output signal Hb of the B-phase window comparator 4b and the inverted signal of the output Jb of the B-phase comparator 5b is shown in FIG.
), the A-phase signal VA rises at the zero-crossing position, and this rising signal causes the A/D converter 6a to
The output of is taken into the offset correction value calculation circuit 7a. That is, the value of the zero cross point of the A-phase signal SA is taken in and stored in the offset correction value calculation circuit 7a. Similarly, the value of the zero cross point of the B-phase signal SB is taken into the B-phase offset correction value calculation circuit 7b.

Note that in this embodiment, the comparator 5a
, 5b of phase A and B at the zero cross point when the output rises.
Although the values of the phase signals SA and SB are sampled, when the outputs of the comparators 5a and 5b fall, it is also a zero cross point, and the A and B phase signals S at this point
The values of A and SB may also be sampled. In this case, for the A phase, an AND circuit is added that inputs the inverted signal of the output Ja of the comparator 5a, the output signal Jb of the B phase comparator 5b, and the inverted signal of the output Hb of the B phase window comparator 4b. death,
An OR circuit may be provided to OR the output of this AND circuit and the output of the AND circuit 8a, and the value of the A/D converter 6a may be sampled at the rising edge of the output signal of this OR circuit. Furthermore, for the B phase, an AND circuit is added which similarly inputs the inverted signal of the output Jb of the comparator 5b, the output signal Ja of the A-phase comparator 5a, and the inverted signal of the output Ha of the A-phase window comparator 4a. An OR circuit may be provided to OR the output of this AND circuit and the output of the AND circuit 8b, and the value of the A/D converter 6b may be sampled at the rising edge of the output signal of this OR circuit.

As described above, the A-phase signal SA and the B-phase signal S are supplied to the offset correction value calculation circuits 7a and 7b, respectively.
N zero-cross point voltages of B are sequentially stored, their average is taken, and the complement of the average value is output as an offset correction value and converted into an analog signal by D/A converters 9a and 9b. .

Next, it will be explained that the complement of the above average value becomes the offset correction value.

Since the values stored in the offset correction value calculation circuits 7a and 7b are the values at the zero cross points of the A and B phase signals SA and SB, in the first and second equations above, sin θ=
0, cos θ=0, and the offset value VM + α or VM + β, which is a DC component, is stored in the offset correction value calculation circuits 7a and 7b, respectively. Therefore, the outputs of the A/D converters 6a and 6b are
bit, and the above offset value VM + α, VM +
Assuming that the value of β is stored as an 8-bit digital value, and the digital value of the reference voltage VM is 28, that is, "10000000" in binary notation, then the complement of (VM + α) = 29 - (VM + α) = 29
−(28 + α) = 28 −α = VM − α Similarly, the complement of (VM + β) = VM − β. Using the complement of (VM + α) and the complement of (VM + β) as offset correction values, the D/A converters 9a and 9
As shown in the following third and fourth equations, adders 10a and 10b add the A-phase signal SA and the A-phase correction value (VM - α), respectively, and obtain the B-phase signal SB and By adding the B-phase correction value (VM - β) and subtracting the reference signal VM, it is possible to obtain the A-phase and B-phase signals NA and NB that have undergone offset correction with the reference signal VM as the neutral point. .

[0021]NA=SA+(VM−α)−VM
= VM + α + 2 sin θ + (VM - α)
−VM=VM+2sinθ
…(
3) NB =SB +VM -β-VM
= VM + β + 2 cos θ + (VM - β
)−VM=VM+2cosθ
…
(4) In the above embodiment, the high pass filter 2a,
2b is provided to attenuate the input signal with a frequency below a predetermined frequency f, so that the window comparators 4a and 4b always output a high level signal when this low frequency input is received. When the moving speed is slow and the sine wave frequency of the sensor signal is low, it is difficult to detect the zero cross point and there is a risk of false detection. This is to avoid sampling.

Further, the output HA of the A-phase window comparator 4a is sent to the B-phase AND circuit 8b, and the output HB of the B-phase window comparator 4b is sent to the A-phase AND circuit 8a.
The reason for inputting is that when the high-pass filters 2a and 2b are applied, the capacitor is charged when the detected object is stopped, and when the object starts moving, the high-pass filter 2a and 2b are input.
A, 2b outputs a voltage that changes rapidly and may cross zero, so in order to prevent sampling due to this, when the output of the window comparator of one phase is at a low level (at this time, the output of the window comparator of one phase is (the sine wave of the phase crosses zero), each signal SA, SB is sampled to prevent erroneous sampling.

Note that when this detector is first operated, a value corresponding to the reference voltage VM is initially set in each register of the offset correction value calculation circuits 7a and 7b. When the sine wave signal exceeds the frequency f, each signal SA at the zero cross point is stored in each register.
, SB are sequentially stored and updated, and the D/A converters 9a and 9b always output updated offset correction values, so that the sine wave signal from the sensor has a frequency f or less and the speed of the detected object becomes low, the A, B phase signals SA, SB will be corrected. Therefore, even if the offset of the signal from the sensor changes due to temperature or the like, the changed offset is always corrected.

[0024]

As described above, in the present invention, the value of the neutral point of the sine wave signal from the sensor is sequentially detected and updated, the offset correction value is determined from the value of the neutral point, and the offset correction value is calculated. Therefore, even if the offset of the sensor signal fluctuates due to temperature changes, the offset is corrected, and accurate offset correction can be performed at all times. Further, even in a low speed range, accurate offset correction can be performed without error, and the speed and position of the rotating shaft or moving object, which is the object to be detected, can be controlled accurately and evenly. Furthermore, since the data detected for offset correction is only the value of the neutral point, the configuration is simple.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of an automatic offset correction circuit in one embodiment of the present invention.

FIG. 2 is an explanatory diagram of operation timing in the same embodiment.

[Explanation of symbols]

1a, 1b Differential amplifier 2a, 2b High pass filter 4a, 4b Window comparator 5a, 5b Comparator 6a, 6b Analog/digital converter (A/D converter) 7a, 7b Offset correction value calculation circuit 8a, 8b
AND circuit 9a, 9b Digital/analog converter (D/A converter) 10a, 10b Adder circuit

Claims (4)

[Claims] Claim 1: A phase of a sine wave signal and a sine wave signal having a phase difference of 90 degrees from the sine wave signal due to the movement of a moving body;
In an offset correction method for a detector that generates a B-phase signal, the neutral point of the signal from which the DC component of the A-phase and B-phase sine wave signals has been removed is determined, and the A-phase and B-phase signals at the neutral point are determined.
An automatic offset correction method for sensor signals, characterized in that phase values are used as correction values, and the correction values are corrected for A-phase and B-phase signals, respectively. [Claim 2] A phase of a sine wave signal and a sine wave signal having a phase difference of 90 degrees from the sine wave signal due to the movement of the moving body;
In the offset correction method for a detector that generates a B-phase signal, the DC component is removed from the A-phase and B-phase sine wave signals by a high-pass filter, and one of the A-phase and B-phase signals exceeds a predetermined level. Find the neutral point of the other signal when Features an automatic offset correction method for sensor signals. 3. The A phase of the sine wave signal and the sine wave signal having a phase difference of 90 degrees from the sine wave signal due to the movement of the moving body,
In the offset correction method for a detector that generates a B-phase signal, a high-pass filter removes the DC component from the A-phase and B-phase sine wave signals, and the high-pass filter attenuates signals below a predetermined frequency. ,A
When one of the phase and B signals is at a predetermined level or higher, find the neutral point of the other signal, use the above values of the A phase and B phase at the neutral point as correction values, and calculate the values of the A phase and B phase. An automatic offset correction method for sensor signals, characterized in that each signal is corrected by the above correction value. 4. Automatic offset of a sensor signal according to claim 1, claim 2, or claim 3, wherein a plurality of values of the A phase and B phase at the neutral point are each stored, and the average thereof is used as the respective correction value. Correction method.