JP2010216961A - Encoder output signal correction apparatus and method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an encoder output signal correction apparatus and a method capable of removing offsets, amplitude errors, phase errors, and tertiary harmonic components, according to the multiplex Lissajous signals, contained in two-phase sinusoidal signals of multiplex Lissajous signals constituted of two-phase sinusoidal signals. <P>SOLUTION: The encoder output signal correction apparatus 1 includes: an up/down counter 60 for specifying the period of two-phase sinusoidal signals; detection parts 31 and 41 for detecting errors from ideal Lissajous signals, for every value of information of each specified period, contained in Lissajous signals made of two-phase sinusoidal signals; and correction parts 30 and 40 for correcting two-phase sinusoidal signals for every value of information of each specified period by a correction coefficient based on errors detected by the detection parts 31 and 41. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an encoder output signal correction apparatus and method for correcting a two-phase sinusoidal signal of an encoder that detects position, angle, velocity, angular velocity, and the like.

  Since there is a processing limit on the interval of the grating formed on the encoder scale, in order to measure an interval finer than the scale grating, the spatial period of the phase change of the sinusoidal signal output from the encoder is further subdivided and interpolated. There is a need. For this reason, various interpolation circuits are conventionally used. An interpolation circuit by digital processing is, for example, an A / D converter that samples A and B phase sinusoidal signals output from an encoder with different 90 ° phases and converts them into digital data, and this A / D converter. And a memory storing a look-up table for obtaining the phase angle data PH at each sampling point based on the digital data DA and DB obtained by the above. The lookup table is created based on PH = ATAN (DB / DA) using an arctangent function (ATAN).

  The A and B phase sine wave signals output from the encoder are not usually perfect sine waves, and generally represent an elliptical Lissajous signal when expressed in orthogonal coordinates. If the amplitudes of the A and B phase sinusoidal signal voltages are different, the Lissajous signal becomes an ellipse, and the Lissajous signal becomes a circle or ellipse waveform deviated from the origin due to the offset value of each signal voltage. Also, if there is a phase error, the major and minor axes of the ellipse are not parallel to the coordinate axis and become 45 °. Since the interpolation circuit is made on the assumption that the A and B phase sine wave signals are sine waves, the deviation from the ideal sine wave adversely affects the interpolation accuracy. For this reason, for example, Patent Documents 1 and 2 propose apparatuses for correcting amplitude errors, phase errors, and offsets in the A and B phase sinusoidal signals.

  However, even in a two-phase sine wave signal in which such an amplitude error is corrected, the deviation from the ideal sine wave signal waveform, that is, the waveform distortion is large, and the distortion rate is particularly large between the main scale and the index scale. Fluctuates greatly due to fluctuations. Many of these waveform distortions are due to odd-order (third order, fifth order,...) Harmonic components. When measurement is performed using a two-phase sinusoidal signal having such distortion fluctuation, Measurement error occurs.

  Several techniques for outputting a sinusoidal signal excluding such harmonic components have been proposed. For example, Patent Document 3 proposes a technique in which two rectangular wave grating patterns with slightly shifted phases are provided on a scale, and their outputs are added to cancel out the harmonic components. Further, Patent Document 4 proposes that a sinusoidal signal from which harmonic components are removed by a combination of a uniform grating scale and a non-uniform grating scale is output.

  Furthermore, the apparatus described in Patent Document 5 detects an error from an ideal Lissajous signal included in a Lissajous signal formed by a two-phase sinusoidal signal and uses a correction coefficient based on the detected error to correct the two-phase sinusoidal signal. Correct. The apparatus detects an error from the ideal Lissajous signal included in the corrected two-phase sinusoidal signal, and cumulatively calculates the detected error to obtain a new correction coefficient. Update the correction factor. As a result, the apparatus described in Patent Document 5 has improved robustness by relatively simple digital arithmetic processing.

  However, the techniques described in Patent Documents 1 to 5 do not correspond to the double (or multiplexed) Lissajous signal, and the accuracy thereof is not sufficient. For example, even when only the secondary frequency of a predetermined signal is selected and measurement is desired, the primary frequency may remain in the selected signal depending on the measurement environment. In this case, the odd-numbered peak in the two-phase sinusoidal signal has a value different from the even-numbered peak, and the Lissajous signal composed of the two-phase sinusoidal signal is observed in a doubled manner.

Japanese Patent Laid-Open No. 10-311741 JP 2003-222534 A JP-A-3-48122 Japanese Patent No. 2695623 JP 2006-112862 A

  The present invention is an encoder capable of removing an offset, an amplitude error, a phase error, and a third harmonic component contained in a two-phase sinusoidal signal corresponding to a multiple Lissajous signal composed of the two-phase sinusoidal signal. An output signal correction apparatus and method are provided.

  An encoder output signal correction apparatus according to the present invention is an encoder output signal correction apparatus for correcting a phase-shifted two-phase sinusoidal signal output from an encoder, and a period specifying means for specifying a period of the two-phase sinusoidal signal; Detecting means for detecting an error from an ideal Lissajous signal included in the Lissajous signal formed by the two-phase sinusoidal signal for each specified period information value, and for each specified period information value And a correction means for correcting the two-phase sinusoidal signal with a correction coefficient based on the error detected by the detection means.

  The encoder output signal correction method according to the present invention includes a step of specifying a period of a phase-shifted two-phase sinusoidal signal output from an encoder, and the two-phase sinusoidal signal for each specified value of each period information. Detecting and correcting the offset included in the step, detecting and correcting the amplitude error included in the two-phase sinusoidal signal with the offset corrected for each specified period information value, A step of detecting and correcting a phase error included in the amplitude-corrected two-phase sinusoidal signal for each period information value, and a phase-corrected two-phase sinusoidal signal for each specified period information value And a step of detecting and correcting the third-order harmonic distortion included.

  According to the present invention, it is possible to remove the offset, amplitude error, phase error, and third harmonic component contained in the two-phase sinusoidal signal corresponding to the multiple Lissajous signals composed of the two-phase sinusoidal signal. An encoder output signal correction apparatus and method can be provided.

It is a block diagram which shows the basic composition of the encoder output signal correction apparatus 1 which concerns on 1st Embodiment of this invention. It is a figure which shows digital signal A1 (m) and B1 (m). It is a figure which shows the double Lissajous signal L1 (L1 (0), L1 (1)). It is a flowchart which shows the flow of a process of the correction apparatus. 5 is a flowchart showing details of offset correction, amplitude correction, and phase correction of FIG. 4. It is a figure which shows an example of the Lissajous signal observed. 3 is a flowchart showing details of third-order harmonic distortion correction (first method) in FIG. 2. In the third harmonic is a diagram for explaining a method of calculating the amplitude a _1, a _3. 5 is a flowchart showing details of third-order harmonic distortion correction (second method) in FIG. 4. It is a figure for demonstrating the coordinate rotation in the correction process of FIG. It is a graph which shows the relationship between the A phase (or B phase) and the voltage of the 3rd harmonic in the correction process of FIG. It is a graph which shows a mode that a correction value converges by dynamic correction. 3 is a circuit diagram showing specific circuit configurations of an offset / amplitude / phase correction unit 30 and a third harmonic distortion correction unit 40. FIG. It is a block diagram which shows the basic composition of the encoder output signal correction apparatus 2 which concerns on 2nd Embodiment of this invention.

[First Embodiment]
[Configuration of Encoder Output Signal Correction Device 1 According to First Embodiment]
A first embodiment of the present invention will be described in detail with reference to the drawings. FIG. 1 is a block diagram showing a basic configuration of an encoder output signal correction apparatus 1 according to the first embodiment of the present invention.

  The encoder output signal correction apparatus 1 performs correction to remove the third harmonic from the output signals A0 and B0 of the encoder 10. As shown in FIG. 1, the encoder output signal correction apparatus 1 includes A / D converters 20 and 21, an offset / amplitude / phase correction unit 30, an offset / amplitude / phase detection unit 31, and a third-order harmonic distortion correction. Unit 40, third harmonic calculation / detection unit 41, r-θ conversion unit 50, and UP / DN counter 60.

  The encoder 10 is not limited in its detection principle, but is, for example, a photoelectric type or a magnetic type. The A-phase sinusoidal signal and the B-phase sinusoidal signals A0 and B0 output from the encoder 10 usually include an amplitude error, a phase error, an offset, a third harmonic distortion, and the like.

  As shown in FIG. 1, the A / D converters 20 and 21 sample the signals A0 and B0 at a predetermined frequency and convert them into digital signals A1 (m) and B1 (m). The A / D converters 20 and 21 output the digital signals A 1 (m) and B 1 (m) to the offset / amplitude / phase correction unit 30.

  Here, the digital signals A1 (m) and B1 (m) have an amplitude with a phase difference of 90 ° as shown in FIG. In the digital signals A1 (m) and B1 (m), for example, the amplitude fluctuates due to the influence of noise or the like, and the peak in the even number (2k (k is a natural number)) period is the peak in the odd number (2k-1) period. The Lissajous signals of these digital signals A1 (m) and B1 (m) become double Lissajous signals L1 (L1 (0), L1 (1)) as shown in FIG. End up.

  As shown in FIG. 1, the offset / amplitude / phase correction unit 30 digitally calculates the correction coefficient calculated by the offset / amplitude / phase detection unit 31 and the period information m counted by the UP / DN counter 60. The output signals A4 (m) and B4 (m) are generated by correcting the offset, amplitude and phase of the signals A1 and B1 for each period information value. The offset / amplitude / phase correction unit 30 outputs the output signals A4 (m) and B4 (m) to the offset / amplitude / phase detection unit 31 and the third-order harmonic distortion correction unit 40.

  As shown in FIG. 1, the offset / amplitude / phase detection unit 31 uses the output signals A4 (m) and B4 (m) and the period information m based on the periods of the signals A1 (m) and B1 (m). For each information value, a correction coefficient is calculated, and the correction coefficient is output to the offset / amplitude / phase correction unit 30. The correction coefficient calculation method will be described later.

  As shown in FIG. 1, the output signals A4 (m) and B4 (m) are sinusoidal output signals in which the amplitude, phase, and offset are corrected, but the harmonic components including the third harmonic are still present. include. For this reason, the 3rd harmonic distortion correction | amendment part 40 outputs the 3rd harmonic component of output signal A4 (m) and B4 (m) for every value of each period information of signal A1 (m) and B1 (m). The output signals A7 (m) and B7 (m) are output after correction. The correction is executed based on the correction coefficient given from the third harmonic calculation / detection unit 41 and the period information m.

  As shown in FIG. 1, the third-order harmonic calculation / detection unit 41 has radius information r () given from the r-θ conversion unit 50 for each period information value of the signals A1 (m) and B1 (m). m) Based on the phase information θ (m), each correction coefficient is calculated, and the correction coefficient is output to the third harmonic distortion correction unit 40. The correction coefficient calculation method will be described later.

  As shown in FIG. 1, the r-θ conversion unit 50 determines the value of each period information of the signals A1 (m) and B1 (m) from the output signals A7 (m) and B7 (m) and the period information m. A Lissajous signal is generated, and a radius r (m) (radius information r (m)) for each phase θ (m) (phase information θ (m)) of each Lissajous signal is calculated. The r-θ conversion unit 50 outputs the phase information θ (m) and the radius information r (m) to the third harmonic calculation / detection unit 41.

  As shown in FIG. 1, the UP / DN counter (wave number counter) 60 specifies the periods of the digital signals A1 (m) and B1 (m) based on the phase information θ (m) and the phase threshold θth, and the period Information m is generated. The UP / DN counter (wave number counter) 60 includes an offset / amplitude / phase correction unit 30, an offset / amplitude / phase detection unit 31, a third harmonic distortion correction unit 40, and a third harmonic distortion detection unit 41. Is output.

  The phase threshold θth is, for example, “θth = 337.5 °” shown in FIG. When the Lissajous signal makes one rotation counterclockwise with reference to the phase threshold θth, the UP / DN counter 60 counts in the up direction. When the Lissajous signal makes one clockwise rotation with the phase threshold θth as a reference, the UP / DN counter 60 counts in the down direction. The phase threshold θth is an angle (0 ° (y = 0 above), 45 °, −45 ° (y = x above, y = −x above), 90 ° (x = 0) where the point used for correction calculation is located. If the phase threshold θth is set to the angle at which the point used for the correction calculation is located, the period of the point used for the correction calculation cannot be specified.

  The period information m is represented by 1 bit. When the Lissajous signal rotates an even number with reference to the phase threshold θth (in the case of the 2k cycle), the cycle information m is “0”. On the other hand, when the Lissajous signal rotates an odd number with reference to the phase threshold value θth (in the case of (2k−1) period), the period information m is “1”.

  Next, with reference to FIG. 4, the detail of the correction | amendment process using the encoder output signal correction apparatus 1 comprised in this way is demonstrated. FIG. 4 is a flowchart showing the correction process. The A-phase sine wave signal and the B-phase sine wave signals A0 and B0 output from the encoder 1 are first AD-converted (step S11), and then the digital A-phase sine wave signal and B-phase sine wave signal A1 (m), B1. (M). The signals A1 (m) and B1 (m) can be expressed as the following (Formula 1) and (Formula 2).

Here, a ₀ (m) and b ₀ (m) are offsets of the A phase and the B phase of the values (m = 0, 1) of each period information. a ₁ (m) and b ₁ (m) are amplitude errors of the A phase and the B phase of the values (m = 0, 1) of each period information. φ ₁ (m) is a phase error of the B phase with respect to the A phase of each period information value (m = 0, 1). a ₃ (m) and b ₃ (m) are the amplitudes of the third-order harmonics of the A phase and the B phase of each period information value (m = 0, 1). φ ₃ (m) is a phase error of the value of each period information (m = 0, 1) with respect to the fundamental wave of the third harmonic. u is “u = 2πx / λ”. x represents displacement, and λ represents signal pitch. Among these errors, the offset, amplitude error, and phase error are offset correction processing steps (step S12) and amplitude correction processing steps (steps) executed by the offset / amplitude / phase correction unit 30 and the offset / amplitude / phase detection unit 31, respectively. The third harmonic distortion is sequentially removed in S13) and the phase correction processing step (step S14), and the third harmonic distortion is executed by the third harmonic distortion correction unit 40 and the third harmonic calculation / detection unit 41. It is removed in the correction step (step S15). Subsequently, the phase information θ (m) is obtained by the r-θ conversion unit 50 using the two-phase sinusoidal signals A7 (m) and B7 (m) of the values (m = 0, 1) of each period information from which the error is removed. ) Is obtained (step S16). After step S16, a cycle specifying step (step S17) for specifying the cycle (m = 0 or 1) of the signals A1 (m) and B1 (m) is executed.

  In this embodiment, dynamic correction using a recurrence formula is performed in each of the correction processing steps (steps S12 to S15) described above.

[Offset correction]
FIG. 5 shows details of the correction processing steps (steps S12 to S14). First, as shown in FIG. 6, similarly to FIG. 3, the double Lissajous signals L1 (L1 (0), L1 (1)) are obtained from the A-phase and B-phase sinusoidal signals A1 (m) and B1 (m). ) Is obtained. In such a case, the Lissajous signal L1 (0) (L1 (1 )) In four zero cross points across the X-axis and Y-axis _{P12 m (x12 m, y12 m} ), P23 m (x23 m, y23 m) , P34 _m (x34 _m , y34 _m ), P41 _m (x41 _m , y41 _m ) [m = 0, 1], changes in offset correction values in the X-axis and Y-axis directions Δda1 (m), Δdb1 (m ) Is obtained as in (Equation 3) and (Equation 4) below (step S111).

Δda1 (m) and Δdb1 (m) obtained here are close to the offsets a ₀ (m) and b ₀ (m), but are not completely coincident due to the amplitude error and the phase error. Therefore, this error is gradually converged by repeating the feedback process several times. That is, the correction values da1 (m) and db1 (m) are obtained as cumulative addition values as in the following (Formula 5) and (Formula 6) (step S112).

  Then, the correction process for removing the offset from the signals A1 (m) and B1 (m) is executed according to the following (Expression 7) and (Expression 8) (Step S113).

(Amplitude correction)
Similarly to the above, from the four zero cross points P12 _m , P23 _m , P34 _m , P41 _m [m = 0, 1] crossing the X axis and the Y axis of the Lissajous signal, the amplitude correction values in the X axis and Y axis directions are calculated. Changes Δka1 and Δkb1 are obtained as follows (Equation 9) and (Equation 10) as follows (step S121).

  Also in this case, the error is gradually converged by repeating the feedback process several times. That is, the correction values ka1 (m) and kb1 (m) are obtained as cumulative division values as in the following (Formula 11) and (Formula 12) (step S122).

  Then, correction processing for removing the amplitude error from the signals A2 (m) and B2 (m) is executed according to the following (Equation 13) and (Equation 14) (step S123).

(Phase correction)
Similarly to the above, four points P1 _m crossing a straight line (y = x, y = −x) of 45 ° with respect to the X axis and the Y axis of the double Lissajous signal L (0) (L (1)) From (x1 _m , y1 _m ), P2 _m (x2 _m , y2 _m ), P3 _m (x3 _m , y3 _m ), P4 _m (x4 _m , y4 _m ) [m = 0, 1], A phase and B A change Δkp1 of the phase correction value of the phase is obtained as shown in the following (Equation 15) (step S131).

  Also in this case, the error is gradually converged by repeating the feedback process several times. That is, the correction value kp1 (m) is determined as a cumulative multiplication value as in the following (Formula 16), (Formula 17), and (Formula 18) (step S132).

  Then, the correction process for removing the phase error from the signals A3 (m) and B3 (m) is executed according to the following (Equation 19) and (Equation 20) (step S133).

[Third harmonic distortion correction]
The output signals A4 (m) and B4 (m) are sinusoidal output signals whose amplitude, phase, and offset are corrected, but still include harmonic components including the third harmonic. That is, assuming that the amplitude and phase of the third harmonic are equal, the amplitude of the fundamental wave is a ₁ (m) (= b ₁ (m)), and the amplitude of the third harmonic is a ₃ (m) (= b ₃ ( m)) When the phase of the third harmonic is φ ₃ (m), the output signals A4 (m) and B4 (m) are expressed by the following (Formula 21) and (Formula 22).

  Accordingly, the Lissajous radius r (m) of the signals A4 (m) and B4 (m) is obtained as shown in the following (Equation 23).

As apparent from the above equation, the Lissajous radius r (m) has a maximum value r _max (m) = a ₁ (m) + a ₃ (m) and a minimum value r _min (m) = a ₁ (m) −a ₃ cycles between the (m) λ / 4, changes in the phase 3 [phi] _3. Therefore, if a ₁ (m), a ₃ (m), and φ ₃ (m) are obtained, the third-order harmonic distortion can be corrected.

[Method 1: When φ ₃ (m) = 0]
First, FIG. 8 shows a simpler first method. As described above, the radius r (m) of each Lissajous signal L (0), L (1) changes with a period of λ / 4 due to third-order harmonic distortion, and its maximum value r _max (m) is “R _max (m) = a ₁ (m) + a ₃ (m)”, and the minimum value r _min (m) is “r _min (m) = a ₁ (m) −a ₃ (m)” ( (See FIG. 8). Accordingly, a ₁ (m) and a ₃ (m) can be calculated as in the following (Equation 24) and (Equation 25) using r _max (m) and r _min (m).

For the sake of brevity, it is assumed here that the phase φ ₃ (m) = 0 can be considered. For example, assuming that there is a minimum radius r _min (m) on the X and Y axes and a maximum value r _max (m) on a line forming 45 ° with respect to the X and Y axes, r _max (m), _rmin (m) can be obtained as follows (Equation 26) and (Equation 27) as follows (step S151).

Again, the error is gradually converged by repeating the feedback process several times. That is, the correction values a ₁ (m) and a ₃ (m) are obtained as cumulative addition values as in the following (Equation 28) and (Equation 29) (step S152).

  Then, correction processing for removing the third-order harmonic distortion from the signals A4 (m) and B4 (m) is executed according to the following (Equation 30) and (Equation 31) (step S153).

[Method 2: When φ ₃ (m) is arbitrary]
FIG. 9 shows method 2 in the case where φ ₃ (m) is arbitrary. Method 1 is simple and requires a small computer load. However, if φ ₃ (m) is arbitrary, the detection accuracy of phase φ ₃ can be reduced if the amplitude a ₃ (m) of the _third harmonic component is reduced. There is sex. Method 2 to be described next is a method by which φ ₃ (m) can be calculated more strictly. Hereinafter, Method 2 will be described. In this method 2, the amplitudes a ₁ (m), a ₃ (m) and the phase φ ₃ (m) are calculated using Fourier analysis. That is, the real part of the Fourier transform of the signal component of wavelength λ / 4 (spatial frequency 4 · 2π / λ) included in the Lissajous signal is Re (m) and the imaginary part is Im (m) for dynamic correction. Then, dRe (m) and dIm (m) represented by the following (Equation 32) and (Equation 33) are obtained from the real part and the imaginary part detected in the corrected waveform (step S154).

  Subsequently, Re (m) and Im (m) are updated by the recurrence formulas of the following (Formula 34) and (Formula 35) (step S155). By repeating this update several times, Re (m) and Im (m) converge to constant values, and the values are determined as Re (m) and Im (m).

In step S155, a ₁ (m) is obtained as in the following (Equation 36).

In step S155, based on Re (m) and Im (m) determined by the above (Equation 34) and (Equation 35), a ₃ (m) and φ ₃ (m) are Re (m), respectively. , Im (m), the distance and angle from the origin of the coordinates on the complex space are obtained by the following (Equation 37) and (Equation 38). Here, the reason why the coefficient √2 is multiplied in (Equation 37) is that the magnitude of the signal obtained by the Fourier transform is an effective value, and the amplitude is √2 times thereof.

In the third-order harmonic distortion correction process, the Lissajous signal L4 (m) of the signals A4 (m) and B4 (m) as shown on the left side of FIG. A Lissajous signal L5 (corresponding to signals A5 (m) and B5 (m) as shown on the right in FIG. 10 is rotated counterclockwise by an angle φ ₃ (m) corresponding to the phase of the third harmonic. m). The reason why the rotation of the angle φ ₃ (m) is performed is to create a state where the phase of the third harmonic is 0 ° or 90 ° on the Lissajous signal and execute the amplitude correction processing in this state.

In this state, the fundamental wave amplitude a ₁ (m) and the third harmonic amplitude a ₃ (m) are based on the relationship curve of the A phase (or B phase) voltage to the third harmonic voltage in FIG. ), The third harmonic component from the output signals A5 (m) and B5 (m) where the phase of the third harmonic is 0 ° or 90 ° is expressed by the following (Equation 40) and (Equation 41). Correct the direction of removal.

Finally, by performing the calculation shown in the following (Equation 42), the Lissajous signals of the signals A6 (m) and B6 (m) are reversely rotated (rotated by an angle −φ ₃ (m)), and the signal A7 (m ), B7 (m) is generated (S156). The Lissajous signals rotated by the angle −φ ₃ (m), that is, the output signals A7 (m) and B7 (m) include the same fundamental wave as that of the original output signals A4 (m) and B4 (m). And the third harmonic component is subtracted.

  In this embodiment, in order to speed up convergence, division is used for ka1 (m) and kb1 (m), and multiplication is used for kph (m). However, a method using addition / subtraction is also possible.

  In addition, each correction process described above needs to be executed after the Lissajous has made at least one rotation. In consideration of signal noise removal, it may be possible to obtain an average of N rotations. During the required rotation, correction calculation is performed with the correction values da1 (m), db1 (m),... Im (m) detected earlier. Accordingly, the initial value (all 0s, no correction) is started first. Then, a predetermined rotation is detected, and correction calculation is performed with the aforementioned da1 (m), db1 (m),..., Im (m), and correction calculation is performed with this correction value up to a specified rotation speed. Since this corrected Lissajous signal has a smaller error value, the next correction detection is performed using that value as a starting point. That is, Δda1 (m), Δdb1 (m),..., ΔIm (m) are obtained and integrated with da1 (m), db1 (m),. By repeating the above procedure indefinitely, the correction values da1 (m), db1 (m),... Im (m) approach the true value and eventually converge to the detection resolution.

  Although two methods have been shown for correcting third-order harmonic distortion, any of the recurrence formulas can be added, subtracted, multiplied, or divided. An optimal method may be selected depending on the calculation speed and convergence conditions.

  FIG. 12 is a diagram showing how the detected correction value converges to a constant value. In this way, if the dynamic correction is sufficiently converged and then interrupted and the value is stored in a nonvolatile memory or the like, it can be used for a self-calibration method for static correction.

[Specific Circuit Configurations of Offset / Amplitude / Phase Correction Unit 30 and Third Harmonic Distortion Correction Unit 40]
Next, specific circuit configurations of the offset / amplitude / phase correction unit 30 and the third-order harmonic distortion correction unit 40 will be described with reference to FIG.

  As shown in FIG. 13, the offset / amplitude / phase correction unit 30 includes an offset correction unit 301, an amplitude correction unit 302, and a phase correction unit 303. The offset correction unit 301 includes adders 310 and 311. The adders 310 and 311 respectively add the addition coefficients da1 (m) and db1 (m) given from the offset / amplitude / phase detection unit 31 to the signal A1 (m) for each value of each period information based on the period information m. Offset correction is executed by adding to B1 (m). The amplitude correction unit 302 includes multipliers 320 and 321. The multipliers 320 and 321 use the multiplication coefficients ka1 (m) and kb1 (m) given from the offset / amplitude / phase detection unit 31 for each value of each period information based on the period information m, respectively, as signals A2 (m), Amplitude correction is performed by multiplying B2 (m). The phase correction unit 303 includes multipliers 330 to 333 and adders 340 and 341. The multipliers 330 to 333 and the adders 340 and 341 use multiplication coefficients kph1 (m) and kph2 (m) given from the offset / amplitude / phase detection unit 31 for each value of each period information based on the period information m. And phase correction is performed by converting the signals A3 (m) and B3 (m) into output signals A4 (m) and B4 (m).

The third harmonic distortion correction unit 40 includes a coordinate rotation unit 401, an amplitude correction unit 402, and a coordinate reverse rotation unit 403. The coordinate rotation unit 401 includes multipliers 410 to 413 and adders 414 and 415. That is, the coordinate rotation unit 401 converts the Lissajous signal L4 (m) of the signals A4 (m) and B4 (m) to the third harmonic phase for each value of the period information based on the period information m. By rotating counterclockwise by φ ₃ (m), a Lissajous signal L5 (m) corresponding to the signals A5 (m) and B5 (m) is generated. The amplitude correction unit 402, for each period information value based on the period information m, calculates the fundamental wave amplitude a ₁ (m) calculated by the third harmonic calculation / detection unit 41 and the third harmonic amplitude a _3. (M) is used to correct the direction in which the third harmonic component is removed from the output signals A5 (m) and B5 (m) in which the phase of the third harmonic is 0 ° or 90 °. In this calculation, the amplitude correction unit 402 stores the relationship between A5 (m) and A6 (m) or the relationship between B5 (m) and B6 (m) expressed by (Equation 40) and (Equation 41). This can be realized by providing the look-up table 402T. That is, if each sample value of the output signal A5 (m) (or B5 (m)) is used as an index to the lookup table 402T and the value of the third harmonic component is read as an output, the output signal A6 ( m), B6 (m) are obtained.

The coordinate reverse rotation unit 403 includes multipliers 430 to 433 and adders 434 and 435. The multipliers 430 to 433 and the adders 434 and 435 perform the following calculation to coordinate the Lissajous signals of the signals A6 (m) and B6 (m) for each period information value based on the period information m. The signal is rotated clockwise by the rotation angle φ ₃ (m) in the rotation unit 401, that is, reversely rotated (rotated by angle −φ ₃ (m)) to generate signals A7 (m) and B7 (m). The Lissajous signals rotated by the angle −φ ₃ (m), that is, the output signals A7 (m) and B7 (m) include the same fundamental wave as that of the original output signals A4 (m) and B4 (m). And the third harmonic component is subtracted.

In this way, the third harmonic calculation / detection unit 41 calculates a ₁ (m), a ₃ (m), and φ ₃ (m) for each period information value based on the period information m, This is used for correction in the third harmonic distortion correction unit 40. Output is obtained by repeating the correction in the third harmonic distortion correction unit 40, the r-θ conversion in the r-θ conversion unit 50, and the calculation of the correction coefficient in the third harmonic calculation / detection unit 41 several times. The third harmonic components of the signals A7 (m) and B7 (m) are further removed, and the waveforms of the output signals A7 (m) and B7 (m) can be made closer to an ideal sine wave.

[Effect of the encoder output signal correction apparatus 1 according to the first embodiment]
Next, the encoder output signal correction apparatus 1 according to the first embodiment can execute correction independently for each of the double Lissajous signals by the above configuration. Thereby, the encoder output signal correction apparatus 1 according to the first embodiment can improve the measurement accuracy.

[Second Embodiment]
[Configuration of Encoder Output Signal Correction Device 2 According to Second Embodiment]
Next, the configuration of the encoder output signal correction apparatus 2 according to the second embodiment will be described with reference to FIG. FIG. 14 is a block diagram showing a basic configuration of the encoder output signal correction apparatus 2 according to the second embodiment of the present invention.

  The encoder output signal correction apparatus 2 according to the second embodiment stores the correction coefficient dynamically corrected by the offset / amplitude / phase detection unit 31 and the third harmonic calculation / detection unit 41 as shown in FIG. A memory 70 is provided. Timing for storing the correction coefficient in the memory 70 includes (1) when an external switch is pressed, (2) when the power is turned off (when finished), and (3) always (according to the operation clock, or each correction coefficient is It is conceivable that it has been updated to the correction unit 30, 40). Further, at the time of restart, the correction coefficient may be read from the memory 70 and stored in the detection units 31 and 41 as an initial value. Subsequent processing is the same as the dynamic correction described above. In addition, dynamic correction invalidity instruction means for invalidating the dynamic correction coefficient update operation is provided, and when the dynamic correction invalidity instruction means instructs to invalidate the dynamic correction, the correction units 30 and 40 The correction coefficient read from the memory 70 may be used to correct the two-phase sinusoidal signal.

[Effect of the encoder output signal correction apparatus 2 according to the second embodiment]
The encoder output signal correction apparatus 2 according to the second embodiment has substantially the same configuration as that of the first embodiment, and has the same effects as those of the first embodiment.

[Other Embodiments]
As mentioned above, although embodiment of invention was described, this invention is not limited to these, A various change, addition, etc. are possible within the range which does not deviate from the meaning of invention.

  For example, the apparatus according to the above embodiment is intended for the double Lissajous signals L1 (0) and L1 (1). However, the encoder output signal correction apparatus according to the present invention can also be applied to Lissajous signals exceeding four (4, 6, etc. Lissajous signals). For example, when correcting a quadruple Lissajous signal, “m” can be set to four states “m = 0, 1, 2, 3”, and correction may be performed accordingly. For example, when correcting a six-layer Lissajous signal, “m” can be set to six states “m = 0, 1, 2, 3, 4, 5”, and correction may be performed accordingly.

For example, in the above embodiment, the phase threshold θth is 337.5 °. However, in the present invention, the phase threshold θth only needs to satisfy “θth = 22.5 ° + 45 ° × N (N = 0, 1, 2,..., 7)”. For example, in the example shown in FIG. 6, the Lissajous signal L1 (0) and the Lissajous signal L1 (1) are continuous at points P4 ₀ and P4 ₁ . Therefore, in the example shown in FIG. 6, the phase threshold θth is preferably 292.5 ° in addition to 337.5 °. That is, the examples so shown in FIG. 6, a point that satisfies the above relationship before and after the point P4 ₀ consecutive Lissajous signal L1 (0) and Lissajous signal L1 (1), it is desirable that the phase threshold [theta] th.

  For example, in the above embodiment, the offset, amplitude and phase are first corrected for the A-phase and B-phase sinusoidal signals output from the encoder, and then the third harmonic is corrected. This order can be changed. That is, the correction of the third harmonic may be performed first, and the offset, amplitude, and phase may be corrected later. In the above embodiment, correction of amplitude, phase, and the like is executed by a digital circuit, but similar processing may be performed by a DSP, software, or the like.

  DESCRIPTION OF SYMBOLS 1, 2 ... Encoder output signal correction apparatus 10 ... Encoder 20, 21A ... A / D converter 30 ... Offset / amplitude / phase correction part 31 ... Offset / amplitude / phase detection part 40 ... Third harmonic distortion correction 41: Third harmonic calculation / detection unit 50: r-θ conversion unit 60: UP / DN counter 70: Memory 401: Coordinate rotation unit 402: Amplitude correction unit 403: Coordinate reverse rotation unit .

Claims (15)

In an encoder output signal correction device that corrects a phase-shifted two-phase sinusoidal signal output from an encoder,
Period specifying means for specifying a value of period information of the two-phase sinusoidal signal;
Detecting means for detecting an error from an ideal Lissajous signal included in the Lissajous signal formed by the two-phase sinusoidal signal for each specified period information value;
An encoder output signal correction apparatus comprising: correction means for correcting the two-phase sinusoidal signal with a correction coefficient based on an error detected by the detection means for each specified value of each period information. The detection means detects an error from the ideal Lissajous signal included in the two-phase sinusoidal signal after being corrected by the correction means, and cumulatively calculates the detected error to obtain a new correction coefficient. The encoder output signal correction apparatus according to claim 1, wherein the correction coefficient is dynamically updated. The detection means is capable of detecting at least one of an offset, an amplitude error, a phase error, and a third harmonic distortion included in a Lissajous signal formed by the two-phase sinusoidal signal. Or the encoder output signal correction apparatus of Claim 2. The detecting device, from the four zero-cross points across the X-axis and Y-axis of the Lissajous signal formed by two-phase sinusoidal signals _{P m 12, P m 23,} P m 34, P m 41, X -axis and Y 4. The encoder output signal correction apparatus according to claim 3, wherein the offset is detected by accumulatively calculating changes Δda1 (m) and Δdb1 (m) of an offset correction value in the axial direction. The detecting device, from the four zero-cross points across the X-axis and Y-axis of the Lissajous signal formed by two-phase sinusoidal signals _{P m 12, P m 23,} P m 34, P m 41, X -axis and Y The encoder output signal correction apparatus according to claim 3, wherein the amplitude error is detected by cumulatively calculating changes Δka1 (m) and Δkb1 (m) of an amplitude correction value in the axial direction. The detection means includes four points P _m 1, P that cross a 45 ° straight line (y = x, y = −x) with respect to the X axis and Y axis of the Lissajous signal formed by the two-phase sinusoidal signal. _The phase error is detected by accumulating a change Δkp1 (m) of the phase correction value of the two-phase sinusoidal signal from _m 2, P _m 3 and P _m 4. Item 4. The encoder output signal correction apparatus according to Item 3. The detection means sets the amplitude of the fundamental wave of the two-phase sinusoidal signal as a ₁ (m), the amplitude of the third harmonic as a ₃ (m), and the phase of the third harmonic as φ ₃ (m) = Assuming 0, four points P _m 1 and P _m crossing a 45 ° straight line (y = x, y = −x) with respect to the X axis and Y axis of the Lissajous signal formed by the two-phase sinusoidal signal. 2, P _m 3, P _m 4, and four zero cross points P _m 12, P _m 23, P _m 34, P _m 41 crossing the X axis and the Y axis, the maximum radius rmax (m) of the Lissajous signal And the minimum value rmin (m) is obtained, the change Δa1 (m) of the fundamental wave of the two-phase sinusoidal signal and the change Δa ₃ (m) of the _third harmonic correction value are cumulatively calculated, and the two-phase the value Va 1 for each time point of the sine wave signal _(m), the value of each point of the third harmonic component Va 3 and _(m) and When, _Va 3 (m) _{to, Va 3 (m) = K1} · Va 1 (m) 3 -K2 · Va 1 (m) ( where, K1, K2 _{is, a 1 (m), a} 3 (m) The encoder output signal correction device according to claim 3, wherein the third-order harmonic component is detected by a calculation using a coefficient determined by: The detecting means sets the amplitude of the fundamental wave of the two-phase sinusoidal signal as a ₁ (m), the amplitude of the third harmonic as a ₃ (m), and the phase of the third harmonic as φ ₃ (m). Then, a signal component of wavelength λ / 4 (spatial frequency 4 · 2π / λ) included in the Lissajous signal formed by the two-phase sinusoidal signal is Fourier-transformed to obtain amplitudes a ₁ (m) and a ₃ (m). , Φ ₃ (m)
The correction means rotates the Lissajous signal by φ ₃ (m), then sets the value at each time point of the two-phase sine wave signal as Va ₁ (m) and the value at each time point of the third harmonic component. When Va ₃ (m) is assumed, Va ₃ (m) is
Va ₃ (m) = K1 · Va ₁ (m) ³ −K2 · Va ₁ (m) (where K1 and K2 are coefficients determined by a ₁ (m) and a ₃ (m)), Correcting the third harmonic component,
The encoder output signal correction device according to claim 3, wherein the corrected Lissajous signal is reversely rotated by φ ₃ (m). The detection unit calculates a maximum value and a minimum value of a radius of the Lissajous signal constituted by the two-phase sinusoidal signal, and is included in the two-phase sinusoidal signal based on a difference between the maximum value and the minimum value. The encoder output signal correction apparatus according to claim 3, wherein an amplitude of the third-order harmonic component is calculated as the third-order harmonic distortion. The detection unit obtains the amplitude and phase of the third harmonic component included in the two-phase sinusoidal signal by obtaining a change in radius of the Lissajous signal constituted by the two-phase sinusoidal signal by Fourier analysis. The encoder output signal correction device according to claim 3, wherein the encoder output signal correction device is calculated as third-order harmonic distortion. The correction unit rotates the Lissajous signal of the two-phase sinusoidal signal including the third harmonic component by the phase calculated by the detection unit, so that the phase of the third harmonic component is 0 ° or 90 °. The value of the third harmonic component corresponding to each value of the two-phase sine wave signal is obtained, and the value of the third harmonic component is determined as the two-phase sine corresponding to the rotated Lissajous signal. The encoder output according to claim 3, wherein the amplitude is corrected by subtracting from the wave signal, and the Lissajous signal of the corrected two-phase sine wave signal is reversely rotated by the same angle as the rotated angle. Signal correction device. A memory for storing the correction coefficient;
The encoder output signal correction apparatus according to claim 1, wherein the detection unit reads a correction coefficient stored in the memory at the time of activation and uses the correction coefficient as an initial value of the dynamic update operation. A memory for storing the correction coefficient;
Dynamic correction invalidity instruction means for invalidating the dynamic correction coefficient update operation,
The correction means corrects the two-phase sinusoidal signal using a correction coefficient read from the memory when the dynamic correction invalidity instruction means instructs to invalidate dynamic correction. The encoder output signal correction apparatus according to claim 1. Identifying a period information value of a two-phase sinusoidal signal out of phase output from the encoder;
Detecting and correcting an offset included in the two-phase sinusoidal signal for each specified value of each period information;
Detecting and correcting an amplitude error included in the two-phase sinusoidal signal whose offset has been corrected for each specified period information value;
Detecting and correcting a phase error included in the amplitude-corrected two-phase sinusoidal signal for each specified period information value;
An encoder output signal correction method comprising: detecting and correcting third-order harmonic distortion included in a phase-corrected two-phase sinusoidal signal for each specified period information value. Each of the correction steps detects an error from the ideal Lissajous signal included in the corrected two-phase sinusoidal signal, and adds the detected error to a previously accumulated value to obtain a new correction coefficient. The encoder output signal correction method according to claim 14, wherein the correction coefficient is dynamically updated.

Images