JP2003222534A - Phase difference error detection device and interpolation error estimation device using it - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a phase difference error detection device capable of detecting, simply and accurately, a phase difference error of a two-phase sine- wave signal of an encoder. <P>SOLUTION: This phase difference error detection device is equipped with major/minor axis detection means 22, 24 for determining the major axis length a and the minor axis length b of a Lissajous' waveform formed by synthesizing sine waves A<SB>1</SB>, B<SB>1</SB>, when the two-phase sine waves A<SB>1</SB>, B<SB>1</SB>outputted from the encoder include substantially only the phase difference error and can be expressed by formula 19, and with a major/minor axis ratio operation means 58 for determining the ratio k between the major axis a and the minor axis b determined by the major/minor axis detection means 22, 24. The device is characterized by determining the ratio k between the major axis length and the minor axis length of the Lissajous' waveform. [Formula 19] sine wave A<SB>1</SB>=sin(u), sine wave B<SB>1</SB>=cos(u+ε)...(19). <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a phase difference error detecting device and an interpolation error estimating device using the same, and more particularly to an improvement of a phase difference error detecting mechanism of an encoder signal which originally has a phase difference of 90 degrees.

[0002]

2. Description of the Related Art Conventionally, in order to obtain high resolution phase angle data by digitally interpolating a two-phase sine wave signal of an encoder for detecting position, angular velocity, angular velocity, etc., an output signal of the encoder A processor is used. Since there is a processing limit on the spacing of the grating formed on the scale of the encoder, in order to measure a finer spacing than the scale grating, the spatial period of the phase change of the sinusoidal signal output by the encoder is further subdivided and interpolated. There is a need.

Therefore, various interpolation circuits have been conventionally used. For example, the interpolation circuit by digital processing is originally 9
The phase angle data at each sampling point is calculated based on the digital data obtained by sampling the two-phase sine wave signal output from the encoder with a phase difference of 0 degree at a predetermined frequency, and the position information of the encoder is obtained. To get

By the way, in the interpolation of the two-phase sine wave signal, it is assumed that the signal is a sine wave signal containing no error. Since there is an error, this causes an interpolation error.
To avoid this, manual or automatic signal adjustment was performed to calibrate the interpolation error.

[0005]

By the way, from the past,
There was a technique to find the difference in offset and amplitude,
There is no suitable technique that can automatically and autonomously detect the phase difference error, and adjustment by an expert or a separate device is required to calibrate the interpolation error due to the phase difference error. For this reason, it is difficult for the user to easily calibrate the interpolation error due to the phase difference error, and there has been a strong demand for the development of a technique capable of easily and accurately detecting the phase difference error. There was no suitable technology that could solve this.

The present invention has been made in view of the above-mentioned problems of the prior art, and an object thereof is a phase difference error detecting device capable of easily and accurately detecting a phase difference error of a two-phase sine wave signal of an encoder, And to provide an interpolation error estimation device using the same.

[0007]

As a result of extensive studies made by the present inventor regarding the detection of a phase difference error, the formulation of the phase difference error was successful. By obtaining the phase difference error using this phase difference error formula, the phase difference error of the two-phase sine wave signal can be detected easily and accurately. As a result, it has been found that the interpolation error due to the phase difference error can be easily and accurately estimated and calibrated without adjustment by a skilled person or using a separate device, and the present invention has been completed.

That is, in order to achieve the above-mentioned object, the phase difference error detecting apparatus according to the present invention has a two-phase sine wave A ₁ , B which is originally output from an encoder with a phase difference of 90 degrees.
A phase difference error detecting device for detecting an error of the phase difference of ₁ , which comprises a long / short axis detecting means and a long / short axis ratio calculating means,
It is characterized in that a ratio k between the length of the major axis and the length of the minor axis of the Lissajous waveform is obtained.

When the two-phase sine waves A ₁ and B ₁ substantially include only a phase difference error and can be expressed by the following equation 5, the two-phase sine wave A ₁ , B
The position of the long axis and the position of the short axis of the Lissajous waveform formed by synthesizing ₁ are detected to obtain the length of the long axis and the length of the short axis. The long / short axis ratio calculating means calculates the length ratio k from the length of the long axis and the length of the short axis obtained by the long / short axis detecting means.
Ask for.

[0010]

Sine wave A ₁ = sin (u) sine wave B ₁ = cos (u + ε) (5) where u: 2πx / λ x: position of the encoder λ: cycle of the sine wave

The Lissajous waveform produced by synthesizing the two-phase sine waves A ₁ and B ₁ referred to here is as long as there is no phase difference error.
It becomes a perfect circle. When there is a phase difference error, it becomes an ellipse with a major axis and a minor axis in the direction inclined by 45 degrees with respect to the XY coordinate axes regardless of the magnitude, and the ratio of the major axis length to the minor axis length depends on the magnitude of the error. I say something that changes. In the present invention, it is preferable that the long / short axis detection unit includes a coordinate axis rotation unit, a switching point detection unit, and an axial length calculation unit.

Here, the coordinate axis rotating unit rotates the XY coordinate axis on which the Lissajous waveform is formed by 45 degrees and converts it into an X ₂ Y ₂ coordinate axis in which axes of the Lissajous waveform are parallel or orthogonal. Further, the switching point detection unit detects the X ₂ Y ₂ coordinate value of the switching point of each of the four quadrants of the Lissajous waveform divided by the X ₂ Y ₂ coordinate axis.

[0013] The axial length calculating unit, from X ₂ Y ₂ coordinate values of the switching point of the Lissajous waveform obtained by the switching point detector, the X ₂
The length of the long axis and the length of the short axis of the Lissajous waveform on the Y ₂ coordinate axis are obtained. Further, in the present invention, it is preferable that the long / short axis detection unit includes an absolute value equal point detection unit and an axial length calculation unit.

Here, the absolute value equal point detecting unit detects a point on the Lissajous waveform where the absolute values of the two-phase sine waves A ₁ and B ₁ are equal from four quadrants divided by XY coordinate axes. . Further, the axis length calculation unit obtains the length of the long axis and the length of the short axis of the Lissajous waveform on the XY coordinate axis from the XY coordinate value of each point obtained by the absolute value equal point detection unit.

Further, in the present invention, it is preferable to provide a phase difference error calculating means. Here, the phase difference error calculating means substitutes the length ratio k obtained by the long / short axis ratio calculating means into the following equation 6 to obtain the phase difference error ε of the two-phase sine waves A ₁ and B _1.
Ask for.

[0016]

Phase difference error ε = sin ⁻¹ ((1-k ² ) / (1 + k ² )) (6)

Further, in the present invention, it is preferable to include a program storage means and a calculation means. Here, the program storage means is a Lissajous waveform to be corrected based on the following equation 7 including the ratio k of the major axis length and the minor axis length of the Lissajous waveform obtained by the major / minor axis ratio calculating means. A calculation program for performing the phase correction of the above coordinate values and obtaining the coordinate values on the Lissajous waveform in which the phase difference error is corrected is stored in advance. Further, the calculation means obtains the coordinate value on the Lissajous waveform in which the phase difference error is corrected, based on the calculation program stored in the calculation program storage means.

[0018]

[Equation 7] Here, X and Y are coordinate values on the Lissajous waveform before correction, and X ′ and Y ′ are coordinate values on the Lissajous waveform with the phase difference error corrected.

Further, in the present invention, it is preferable to provide a program creating means for creating an arithmetic program stored in the program storage means.

Further, in the present invention, it is preferable that the program creating means includes an input section and a processing section. Here, the input unit inputs the calculation program. Further, the processing unit causes the program storage unit to store the arithmetic program input from the input unit. Examples of the input unit here include input devices such as a keyboard and a mouse. The processing unit may be, for example, a CPU or the like.

Further, in the present invention, it is preferable that the program creating means includes an output section for outputting a previously created calculation program to the program storing means. An example of the output unit mentioned here is an external program creating device including an external computer or the like. In addition, the term “creating a program” here includes a case where a calculation program is newly created and a case where an existing calculation program is modified.

Further, in order to achieve the above object, the interpolation error estimation device according to the present invention uses the phase difference error ε of the two-phase sine waves A ₁ and B ₁ obtained by the phase difference error detection device according to the present invention as follows. It is characterized by comprising an interpolation error calculating means for substituting into the equation (8) and obtaining an interpolation error E due to the phase difference error ε.

[0023]

Equation 8] the interpolation error E = (λ / 2π) ( - εsin 2 u) ... (8) where, lambda: the period of the sine wave epsilon: phase difference obtained from the phase difference error detecting apparatus according to the present invention Error u: 2πx / λ x: Position of the encoder

[0024]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A preferred embodiment of the present invention will be described below with reference to the drawings. FIG. 1 shows a schematic configuration of an output signal processing device of an encoder using an interpolation error estimating device according to an embodiment of the present invention.

The output signal processing device 10 shown in FIG.
A coder 12, an interpolation circuit 14, and an interpolation error estimation device 16
And an interpolation error correction device 18. And the interpolation circuit
14 is a two-phase sine output from the encoder 12, for example
Wave A ₁, B₁Is interpolated to obtain high-resolution phase angle data C
₁, D₁Is obtained and output to the interpolation error correction device 18.

The interpolation error estimator 16 first detects an error of, for example, an offset, an amplitude ratio and a phase difference of the two-phase sine waves A ₁ and B ₁ input from the encoder 12 to the interpolation circuit 14. The sine waves A ₁ and B ₁ include errors in the offset, the amplitude ratio, and the phase difference, and when it can be expressed by the following Expression 9, the interpolation error estimation device 16 detects the detected offset,
The error value of the amplitude ratio and the phase difference is substituted into the following formula 10, and the interpolation error E due to the error of the offset, the amplitude ratio and the phase difference is given.
Estimate.

[0027]

Sine wave A ₁ = As ₁ sin (2πx / λ) + Vs sine wave B ₁ = Ac ₁ cos (2πx / λ + ε) + Vc (9)

[0028]

[Equation 10] Interpolation error E = (λ / 2π) {(− Vs / As ₁ ) cosu + (Vc / Ac ₁ ) sinu + (Ac ₁ / As ₁ −1) sinucosu −εsin ² u} (10) Here, λ: period u of the sine waves A ₁ and B ₁ u: 2πx / λ x: positions As ₁ and Ac _{1 of the} encoder 12: amplitudes Vs and Vc of the sine waves A ₁ and B ₁ respectively: device 16 Two-phase sine wave A ₁ , B _{1 obtained in}
Each offset error Ac ₁ / As _{1 of} the two-phase sine wave A ₁ obtained by the device 16,
B ₁ amplitude ratio error epsilon: 2-phase sine wave _A 1 obtained by the apparatus 16, _{B 1} phase difference error

The interpolation error E obtained by the interpolation error estimation device 16 is sent to the interpolation error correction device 18. In the interpolation error correction device 18, the outputs C ₁ and D of the interpolation circuit 14 are
₁ is adjusted by the interpolation error E obtained by the interpolation error estimation device 16 and the position information X of the encoder 12 from which the influences of the offset, the amplitude ratio and the phase difference error are removed is output. By configuring the output signal processing device 10 in this way, the interpolation error of the encoder signal can be corrected.

By the way, in order to easily and accurately correct the interpolation error of the encoder signal, it is necessary to easily and accurately estimate the interpolation error. Then, in order to easily and accurately estimate this interpolation error, detection of the error parameter is very important. This error parameter includes an error of offset, amplitude ratio and phase difference, etc. A detection method for this offset and amplitude ratio has been established conventionally, but with respect to the phase difference error, it is automatic and autonomous. Since there is no suitable technique that can be used for detection, it requires adjustment by a skilled person and a separate device.
Therefore, it is difficult for the user to easily calibrate the interpolation error.

Therefore, a characteristic of the present invention is that
This is to formulate the phase difference error in order to easily and accurately detect the phase difference error of the two-phase sine wave of the encoder signal without adjustment by a skilled person or using a separate device. Therefore, in the present embodiment, the interpolation error estimation device 16
However, the phase difference error detection device 20 is provided.

[0032]First embodiment 2 is used in the interpolation error estimation device 16 shown in FIG.
The schematic configuration of the phase difference error detection device is shown.
In this embodiment, the two-phase sine wave A₁, B₁Coordinate axes
Is parallel or orthogonal to the axis of the Lissajous waveform of the elliptical shape.
Rotate 45 degrees to find the phase difference error
I have decided. Also, the interpolation error estimation device shown in FIG.
Then, as parameters, offset, amplitude ratio and phase difference
The error is assumed to be
Error is adjusted by an arbitrary method, and the two-phase
Sine wave A ₁, B₁Substantially includes only the phase difference error,
The two-phase sine wave A when expressed by the following equation 11₁, B
₁Explains an example in which Lissajous waveform is created by synthesizing
To do.

[0033]

Sine wave A ₁ = sin (u) sine wave B ₁ = cos (u + ε) (11) where u: 2πx / λ x: position of the encoder λ: cycle of the sine wave

The phase difference error detection device shown in the figure is provided with a coordinate axis rotating section (long-short axis detection means) 22 as shown in FIG. The coordinate axis rotating unit 22 is connected to an encoder, and adds and subtracts the two-phase sine waves A ₁ and B ₁ output from the encoder to generate an XY coordinate axis on which the elliptical Lissajous waveform is created. It is rotated by 45 degrees and converted into X ₂ Y ₂ coordinate axes in which the axes of the elliptical Lissajous waveform are parallel or orthogonal to each other.
₂ and B ₂ .

The coordinate axis rotating section 22 shown in FIG.
It is connected to a switching point detection unit (long and short axis detection means) 24 as shown in FIG. Therefore, the coordinate rotation unit 22
The two-phase sine waves A2 and B2 from are input to the switching point detection unit (long and short axis detection means) 24.

[0036] The switching point detecting unit 24, the elliptical Lissajous waveform, wherein X ₂ Y ₂ 4 one for each quadrant, divided by the coordinate axes X ₂ Y ₂ coordinates the switching point (first quadrant through fourth quadrant) To detect. This switching point detection unit 24 includes coordinate quadrant detection elements 26, 28, 30, 32, a 0 vicinity detection element 34,
36, and sampling elements 38, 40, 42, 44 and the like.

The coordinate quadrant detection elements 26, 28, 30,
32 is composed of, for example, a comparator, and has a sine wave A ₂ ,
The coordinate quadrant of B ₂ is detected. The 0 vicinity detection element 34,
Reference numeral 36 is composed of, for example, a window comparator or the like, and detects that the sine waves A ₂ and B ₂ are near 0. The sampling elements 38, 40, 42, 44 are, for example, latches, and sample the sine wave A ₂ or B ₂ when the coordinate quadrant and the vicinity of 0 are simultaneously satisfied.

Further, the phase difference error detecting apparatus according to the present embodiment includes the averaging units 46, 48, 50 and 52, the axial length calculating units 54 and 56, the long / short axis ratio calculating unit 58, and the phase difference error calculating unit. 60 is provided.

The averaging units 46 to 52 are composed of, for example, dividers or bit shifters, and average the values at the sampling points. The axis length calculation units 54 and 56 are composed of, for example, subtracters, calculate the difference between the values obtained by the averaging units 46 to 52, and calculate the long axis length of the elliptical Lissajous waveform on the X ₂ Y ₂ coordinate axes. The length a and the minor axis length b are obtained.

The long / short axis ratio calculating means 58 is composed of, for example, a divider or the like, and obtains the ratio k of the long axis length a and the short axis length b of the Lissajous waveform on the X ₂ Y ₂ coordinate axis. The phase difference error calculating means 60 is, for example, a product-sum calculator,
By substituting the length ratio k of the long axis to the short axis of the Lissajous waveform on the X ₂ Y ₂ coordinate axis into the following equation 12, the two-phase sine wave A
The phase difference error ε of ₁ and B ₁ is obtained.

[0041]

Phase difference error ε = sin ⁻¹ ((1-k ² ) / (1 + k ² )) (12)

The phase difference error detecting apparatus according to the present embodiment is roughly configured as described above, and its operation will be described below. First, on a cathode ray tube oscilloscope surface, a Lissajous waveform in which the sine wave A ₁ is plotted on the X axis and the sine wave B ₁ is plotted on the Y axis is a Lissajous waveform, and when the phase difference error ε = 0,
It becomes a perfect circle. On the other hand, when the phase difference error ε ≠ 0, the ellipse has an axis in a direction inclined by 45 degrees with respect to the XY coordinate axes regardless of the size, but the length of the major axis and the length of the minor axis depend on the magnitude of the error. The ratio changes.

For this purpose, first, the coordinate axis rotating unit 22 sets the XY axes as shown in FIG.
When converted into X ₂ Y ₂ coordinates rotated by π / 4 (45 degrees) as shown in, the major axis and minor axis of the ellipse are parallel or perpendicular to the X ₂ Y ₂ axis. That is, the XY coordinate axes as shown in FIG.
Rotation) to obtain the X ₂ Y ₂ coordinate axes as shown in FIG. A ₂ = cos (π / 4) × A ₁ −sin (π / 4) × B ₁ = (A ₁
-B ₁ ) / √2 B ₂ = cos (π / 4) × A ₁ + sin (π / 4) × B ₁ = (A ₁
+ B ₁ ) / √2

Next, the following Table 1 is obtained by the long and short axis detecting means.
Therefore, the position and the length of the long axis and the short axis are detected. First X₂
Y₂Extract the values near the axis. For example, sine wave B_Two> 0
A coordinate quadrant detection element 30 for detecting, and | A_Two| <R (A _Two≒
X) is detected by the 0 proximity detection element 34 that detects 0).₂Y₂On-axis
Switching point P between the first and second quadrants of₂₁(First quadrant and second
Quadrant boundaries). This R is an arbitrary value, for example
For example, 5% of the amplitude a or the amplitude b (= 0.05 × a)
It

In the sampling elements 38 and 40, the value of the detected sine wave B ₂ is added to B ₂₁ (initial value = 0). Further, in order to obtain the number of additions N, N ₂₁ = N ₂₁
Increment +1 and N ₂₁ . By the long and short axis detection means,
In the first quadrant the same method to obtain the second quadrant of the switching point P _21, the second and third quadrants of the switching point P _32, third and fourth quadrants of the switching point P _{4 3,} fourth quadrants And the switching point P of the first quadrant
₁₄ is detected.

[0046]

[Table 1]

Next, the averaging units 46 to 52 average the sample values of the sample points P _{21 to} P ₁₄ . _{P 21 B max = S 21 /} N 21 P 32 A min = S 32 / N 32 P 43 B min = S 43 / N 43 P 14 A max = S 14 / N 14

Next, the length of the long axis is calculated by the axis length calculation units 54 and 56.
The length a and the minor axis length b are obtained. Length of long axis a = A_max-A_min Minor axis length b = B_max-B_min Next, by the long / short axis ratio calculating means 58, the axis length calculating units 54, 5
Of the Lissajous waveform as shown in FIG.
Ratio k (k = b / a) between the length a of the major axis and the length b of the minor axis
Ask for. Next, by the phase difference error calculating means 60, the above number
Substituting the ratio k obtained by the long / short axis ratio calculation means 58 into 12, the phase
Find the difference error ε.

In the present embodiment, the phase difference error calculating means 60 can easily and accurately obtain the phase difference error ε by directly using the equation 12, but the phase difference error ε can be calculated more easily. It is particularly preferable to use the result of the Taylor expansion of the above equation 12 to the eleventh order, for example, as shown in the following equation 13 in order to perform at high speed.
As a result, the calculation of the phase difference error ε can be performed easily and at high speed, as compared with the case where Equation 12 is used as it is. At this time, by increasing / decreasing the order of Taylor expansion according to the required accuracy, it is possible to make a trade-off with the calculation time and the current consumption.

[0050]

Ε = sin ⁻¹ ((1-k ² ) / (1 + k ² )) = π / 2-2k + (2/3) k ³⁻ (2/5) k ⁵ + (2/7) k ^{7 - (2/9) k 9 + (} 2/11) k 11 ... ... (13)

As described above, the phase difference error detecting apparatus according to the present embodiment succeeds in formulating the phase difference error. Then, the X-axis of the sine waves A ₁ and B ₁ is converted by the coordinate axis rotating unit
The Y coordinate axis is rotated 45 degrees so as to be parallel or orthogonal to the axis of the substantially elliptical Lissajous waveform, and the long / short axis detecting means detects the major axis length and the minor axis of the substantially Lissajous waveform on the X ₂ Y ₂ coordinate axis. Seeking ratio of length. Then, the phase difference error calculating means substitutes this into the above equations 12 to 13 for deriving the phase difference error ε characteristic of the present embodiment to obtain the phase difference error ε.

As a result, in this embodiment, the phase difference error ε of the two-phase sine wave can be automatically detected. Moreover,
In this embodiment, since the above equations 12 to 13 for deriving the characteristic phase difference error ε in this embodiment are used, even if noise is superimposed on the signal, the phase difference error of the two-phase sinusoidal wave. It is possible to accurately detect ε. Moreover, the external position reference is not necessary.

In this embodiment, since it is possible to perform only the averaging process for one of the sine waves A ₂ or B ₂ by the averaging unit, the process for obtaining the sine waves A ₂ , B ₂ is performed. It will be easy. Thereby, the phase difference error of the two-phase sine wave of the encoder signal can be detected more easily and accurately.

Further, according to the interpolation error estimation apparatus of this embodiment, since the phase difference error detection apparatus of this embodiment is used, the interpolation error E can be accurately and easily obtained. Then, according to the output signal processing device of the encoder of the present embodiment, the interpolation error E obtained by the interpolation error estimation device of the present embodiment adjusts the output of the interpolation circuit by the interpolation error correction device. Thereby, the interpolation error of the encoder can be corrected more easily and accurately.

Further, in the present embodiment, the interpolation error estimation device detects the maximum and minimum values of the sine waves A ₁ and B ₁ input from the encoder to the interpolation circuit, and from the average thereof, The DC offsets V _s and V _c can be obtained respectively.

In this embodiment, the interpolation error
The estimation device is input to the interpolation circuit from the encoder
Sine wave A₁, B₁The difference between the maximum and minimum values of
Sine wave A₁, B₁Twice the amplitude (2As₁, 2Ac₁)When
is doing. Therefore, the sine wave A₁, B₁Maximum value of
And the minimum value, the sine wave A₁, B₁Each amplitude of (A
s₁, Ac₁), Each amplitude (As₁, Ac₁)before that
Note 2 phase sine wave A₁, B ₁Amplitude ratio (As₁, Ac₁)
Can be turned on.

[0057]Second embodiment FIG. 4 shows the phase difference error detection according to the second embodiment of the present invention.
The schematic configuration of the device is shown. In addition, as shown in FIG.
Corresponding to the phase difference error detection device according to the first embodiment
A reference numeral 100 is added to the portion and the description thereof will be omitted. Na
In this embodiment, the coordinate rotation as in the first embodiment is performed.
Not performed, the Lissajous waveform has its axis relative to the XY coordinate axes.
It becomes an ellipse inclined at 45 degrees. And the 1st to 4th quadrants
The point where the sine wave value is | A | ≒ | B |
Then, the values of the XY axis components are averaged to obtain the ratio of the long axis to the short axis.
Are seeking.

The long / short axis detecting means shown in the figure is provided with an absolute value equal point detecting section 170, and the points on the Lissajous waveform where the absolute values of the two-phase sine waves A ₁ and B ₁ are equal are divided by XY coordinate axes. Each of the four quadrants (first quadrant, second quadrant, third quadrant, and fourth quadrant) is detected. Absolute value equal point detection unit 1
Reference numeral 70 denotes, for example, the determination element 172 and the sampling element 1
74, 176, 178, 180. The absolute value equal point detection unit 170 is connected to an encoder, and receives the two-phase sine waves A ₁ and B ₁ output from the encoder.

The judging element 172 obtains the absolute values of the sine waves A ₁ and B ₁ from the encoder and detects that they are equal to each other. In addition, the sampling elements 174, 1
Reference numerals 76, 178, and 180 are, for example, latches, and sample the coordinate values of the sine waves A ₁ and B ₁ when the determination element 172 satisfies the condition that the absolute values are equal.

Further, in the present embodiment, the averaging element 1
82, axial length calculation units 184, 186, axial length ratio calculation means 158, and phase difference error calculation means 160. The averaging element 182 is composed of, for example, a divider or a bit shifter, and has sampling elements 174, 176, 178,
The values sampled at 180 are averaged and the XY of four quadrants
Get coordinate values.

The axis length calculators 184 and 186 are composed of, for example, a divider of 4 = 2 bit shift and the like, and based on the XY coordinate values of the four quadrants obtained by the averaging element 182, the major axis of the elliptical Lissajous waveform is calculated. The length a and the minor axis length b are determined. The axial length ratio calculation means 158 is composed of, for example, a divider and the like, and calculates the ratio k of the lengths from the length a of the major axis and the length b of the minor axis obtained by the axial length calculators 184 and 186. The phase difference error calculating means 160 is composed of, for example, a sum of products calculating device, and the length ratio k obtained by the axial length ratio calculating means 158 is expressed by the above mathematical expression 1
Substituting into 2 to 13, the phase difference error ε is obtained.

The phase difference error detection apparatus according to this embodiment is roughly configured as described above, and its operation will be described below. In this embodiment, since the coordinate rotation is not performed, the Lissajous waveform is an ellipse inclined by 45 degrees from the XY coordinates. The points where | A | ≈ | B | are sampled in the first to fourth quadrants. That is, points near the straight line y = x or y = -x are extracted.

That is, the absolute values of the sine waves A ₁ and B ₁
The absolute value of the difference is calculated from | A | and | B |
According to 2, it is determined whether it is equal to or less than the constant R. || A | − | B || <R This condition is satisfied when the Lissajous waveform is near the straight line y = x or y = −x as shown in FIG. The values at that time are determined by the positive and negative values of X and Y as shown in Table 2 below, and the sample values of the sampling points when the condition that the absolute values are equal in the determination element 172 are satisfied are sampling elements 174, 176, 178, Sampling and calculation are performed by 180.

[0064]

[Table 2]

Next, the averaging element 182 is used to sample the sample.
Sample point P sampled with ₁~ P_FourSample of
Averaging the values. At this time, sample point P₁~ P_FourIs
Since it is on the straight line y = x or y = −x, X_p1= Y_p1= -X_p3= -Y_p3 X_p4= Y_p2= -X_p2= -Y_p4 To be used. X_p1= S_x1/ N₁, Y_p1= S_y1/ N₁ X_p2= S_x2/ N₂, Y_p2= S_y2/ N₂ X_p3= S_x3/ N₃, Y_p3= S_y3/ N₃ X_p4= S_x4/ N_Four, Y_p4= S_y4/ N_Four

Next, using the sample values averaged as described above by the axial length calculation units 184 and 186, the length a of the major axis of the elliptical Lissajous waveform as shown in FIG. The length b of the short axis is calculated. That is, since the long axis a and the short axis b are inclined 45 degrees with _respect to X _p1 , ... Y _p4 ,
Since its size is √2, the major axis length a = (X _p1 + Y _p1 −X _p3 −Y _p3 ) (√2) /
4 Length of minor axis b = (X _p4 + Y _p2 −X _p2 −Y _p2 ) (√2) /
Get 4.

Next, the elliptic Lissajous waveform as shown in FIG. 5 is obtained from the length a of the major axis and the length b of the minor axis obtained by the axial length computing units 184 and 186 by the major / minor axis ratio computing means 158. The ratio k between the length a of the long axis and the length b of the short axis is calculated. The ratio k = b / a = (X _p4 + Y _p2 −X _p2 −Y _p2 ) / (X _p1 + Y _p1 −X _p3 −Y _p3 ) is obtained.

Next, the phase difference error ε can be obtained by substituting the ratio k obtained as described above into the equations 12 to 13 by the phase difference error calculating means 160.
As described above, the phase difference error detection apparatus according to the present embodiment first succeeds in formulating the phase difference error. Then, the sampling element samples the points where the sine wave value is | A | ≈ | B |
The values of the Y-axis components are averaged by an averaging element, and the long / short axis ratio calculating means obtains the ratio of the long axis to the short axis.

In this embodiment, as in the first embodiment, the phase difference error calculating means substitutes this ratio into the above equations 12 to 13 for deriving the phase difference error ε characteristic of this embodiment. Then, the phase difference error ε is obtained. As a result, in this embodiment, the phase difference error ε of the two-phase sine wave can be automatically detected. Moreover, in this embodiment, since the equations 12 to 13 for deriving the phase difference error ε characteristic of this embodiment are used, even if noise is superimposed on the signal, the level of the two-phase sine wave is reduced. The phase difference error ε can be accurately detected. Moreover, the external position reference is not necessary.

Further, the phase difference error ε obtained in each of the above-mentioned configurations can be substituted into the above equation 8 to obtain the interpolation error E by the phase difference error ε. Here, the equation 10 assumes offset, amplitude ratio, and phase difference errors as parameters, but the two-phase sine waves A ₁ and B ₁ substantially include only the phase difference error. When expressed as
Can be transformed into

[0071]

Equation 14] the interpolation error E = (λ / 2π) ( - εsin 2 u) ... (14) where, lambda: the period of the sine wave epsilon: phase difference error obtained from the phase difference error detection apparatus of the present invention u: 2πx / λ x: position of the encoder

The phase difference error ε thus obtained in each configuration is
By substituting the equation 10 into the simplified equation 14, the interpolation error E due to the phase difference error ε can be easily and accurately obtained.

In each of the above-mentioned structures, the formula 12 for deriving the phase difference error ε, that is, ε = sin ⁻¹
Although the example using ((1-k ² ) / (1 + k ² )) has been described, the following mathematical formula equivalent to the formula can be used. ε = cot ⁻¹ (k) −tan ⁻¹ (k) By using this equivalent formula, it is possible to obtain the same effect as that obtained by using the formula 12.

The interpolation error estimation apparatus according to the present embodiment can calculate the interpolation error caused by the parameter at high speed and easily without any other position reference, so that the interpolation correction can be performed even after the interpolation calculation. And can be applied to any interpolation error correction device.

For example, in the present embodiment, the interpolation error correction device creates correction information based on the interpolation error obtained by the interpolation error estimation means according to the present embodiment, and uses this as a look-up table (LUT). The correction information of the LUT is accessed at a predetermined time such as when the power of the apparatus is turned on, and the output of the interpolation circuit is corrected using the correction information.
It can be applied to a static correction that outputs the position information of the encoder with the interpolation error removed. Alternatively, the correction information created on the basis of the interpolation error obtained by the interpolation error estimation means is not stored in the LUT, and the interpolation error is autonomously corrected during normal operation such as operation of the interpolation circuit. Can be applied to dynamic correction. Alternatively, it can be applied to a combination of the static correction and the dynamic correction. By using the correction information based on the interpolation error obtained by the interpolation error estimation device according to the present embodiment, the interpolation error can be corrected easily and accurately.

[0076]Phase correction When correcting the phase, generally calculate the phase difference error,
Correction was made based on the phase difference error. However,
Since the amount of calculation is huge, it is strongly desired to simplify the calculation.
Although it was done, it can solve this in the past.
There was no suitable technology. Therefore, in the present invention
Is the calculation method of phase correction for simplification of calculation.
Adopted the method of introducing the ratio k of the long axis and the short axis of the Lissajous waveform.
And programmed instructions for performing phase correction.
And.

The programming of the phase correction will be described below. FIG. 6 shows the arrangement of the phase correction mechanism programmed so as to perform the phase correction based on the ratio k of the long axis to the short axis of the Lissajous waveform. The parts corresponding to those in FIG. 4 are designated by the reference numeral 100 and the description thereof is omitted.

In the figure, the phase correction mechanism 288 is connected to the output side of the long-axis / short-axis ratio calculating means 258 and to the output side of the averaging element 282. Then, the ratio k from the long axis / short axis ratio calculating means 258 and the XY coordinate values of the _four quadrants (P ₁ , P ₂ , P ₃ , P ₄ ) from the averaging element 282 are input to the phase correction mechanism 288. To be done. The phase correction mechanism 288 is composed of, for example, a computer as shown in FIG. 7, and includes a storage unit 290, a CPU (calculation unit) 292,
Equipped with. The storage unit 290 includes a data memory 294 and a program memory (program storage unit) 296.

The data memory 294 is connected via the CPU 292.
The ratio k (b /
a) and four elephants from the averaging element to be corrected
Limit (P ₁, P_Two, P_Three, P_Four) XY coordinate values, that is, P
₁(X₁, Y₁), P_Two(X_Two, Y_Two), P_Three(X_Three，
y_Three), P_Four(X_Four, Y_Four) Is remembered. Program
Mori 296 is the Lisa obtained by the long / short axis ratio calculation means.
The following equation including the ratio k of the major axis and minor axis of the Rouge waveform
Coordinate values on the Lissajous waveform to be corrected based on
The phase difference of (X, Y) was corrected and the phase difference error was corrected.
To obtain the coordinate values (X ', Y') on the Lissajous waveform
A calculation program is stored in advance.

The CPU 292 has a program memory 296.
The value of the ratio k stored in the data memory 294 is substituted into the equation 15 based on the arithmetic program stored in, and the coordinate value (X,
Y) is phase-corrected to obtain the coordinate value (X ', Y') on the Lissajous waveform in which the phase difference error is corrected.

[Equation 15] Where X and Y are coordinate values on the Lissajous waveform to be corrected, and X ′ and Y ′ are coordinate values on the Lissajous waveform whose phase difference error has been corrected.

In the present embodiment, when the arithmetic program stored in the program memory 296 in advance is created using the internal computer which is the phase correction mechanism 288 (internal creating method), the external program creating device is used. There is a case where an arithmetic program created in advance is transferred to an internal computer (external creation method). <Internal Creation Method> In the present embodiment, it is also preferable that the internal computer has a function as a program creation unit that creates an arithmetic program stored in the program memory 296. That is, the program creating means is, for example, the CPU (processing unit) 2 of the internal computer.
92, an input device (input unit) 300 including a keyboard, a mouse, etc., connected to the CPU 292 via the interface 298, and a display 302.
Equipped with. The user inputs, for example, the equation 15 and an arithmetic program including instructions for causing the CPU 292 to perform an arithmetic operation based on the equation 15 from the input device 300.

The CPU 292 stores the calculation program input from the input device 300 in the program memory 296. Further, the calculation program input from the input device 300 is displayed on the display 302. Then, when the user inputs a calculation program from the input device 300 while looking at the display 302, the calculation program is stored in the program memory 2 by the CPU 292.
Stored in 96.

<External Creation Method> In this embodiment,
For example, it is also preferable to include an external program creation device (output unit) 304 that is composed of an external computer and that has a function of creating calculation parameters and outputting them. The external program creation device 304 is connected to the CPU 292 of the internal computer via the interface 298. As a connection method between the external program creation device 304 and the internal computer, for example, cable connection, wireless connection using a wireless communication device, or the like can be given. An arithmetic program created in advance by using the external program creating device 304 is transferred from the external program creating device 304 to the program memory 296 via the interface 298 and the CPU 292. Then, the user can display on the display 302 the calculation program and the like transferred from the external program creating device 304 and stored in the program memory 296. The display 302 can also display the measurement result before the phase correction, the measurement result after the phase correction, and the like.

Next, the manner of phase correction according to this embodiment will be described with reference to FIG. It should be noted that FIG. 9A shows a Lissajous waveform including noise before phase correction displayed on the display, and FIG. 9B shows a Lissajous waveform after phase correction displayed on the display.

As shown in FIG. 8A, the Lissajous waveform (signal) having a phase error is inclined by 45 degrees, and | X | = |
The coordinates P ₁ (x ₁ , y ₁ ), P ₂ (x ₂ ,
y ₂ ), P ₃ (x ₃ , y ₃ ), P ₄ (x ₄ , y ₄ ),
In order to remove noise, averaging is performed in the section of 2R in the figure, the length a of the long axis and the length b of the short axis are obtained, and the length ratio k (b / a) is obtained. That is, the coordinates (X, Y) on the Lissajous waveform
Where X = cos (u + ε), sin (u), the axial length calculation unit obtains the major axis a and the minor axis b by the following formula 16.

[0086]

[Equation 16]

Then, the long / short axis ratio calculating means has the length a of the long axis and the length b of the short axis obtained by the axis length calculating section.
Then, the ratio k (b / a) is calculated. The CPU, based on a calculation program including the following equation 17 of the ratio k, coordinates values (X, Y) on the Lissajous waveform shown in FIG.
Phase correction is performed. As a result, as shown in FIG. 7B, a Lissajous waveform in which the phase difference error is beautifully removed is obtained, and coordinate values (X ′, Y ′) on the Lissajous waveform in which the phase difference error is corrected are obtained. be able to.

[0088]

[Equation 17]

As described above, in the present embodiment, a simple arithmetic expression as shown in Expression 17 which uses the ratio of the length of the Lissajous waveform to the length of the short axis is used for the calculation of the phase correction. Since it is programmed, the calculation can be simplified in order to obtain the coordinate value with the phase error corrected.

That is, the phase difference error ε is obtained from the following formula 18, but it is not necessary in the phase correction of this embodiment, and its calculation can be omitted. As a result, in the present embodiment, the phase difference error is calculated from the ratio k between the long axis and the short axis of the Lissajous waveform, and the phase difference error is compared with that for phase correction based on the calculated phase difference error. Since the number can be greatly reduced, the simplification of the calculation required for the phase correction can be largely achieved.

[Equation 18]

[0091]

As described above, according to the phase difference error detection device of the present invention, the formulation of the phase difference error was first successful. Then, a long / short axis detecting means for obtaining the length of the long axis / short axis of the Lissajous waveform produced by synthesizing the two-phase sine wave of the encoder, and a long / short axis ratio calculating means for obtaining the ratio of the length of the long axis / short axis. Since the phase difference error information is obtained based on the length ratio k obtained by the long / short axis ratio calculating means, the phase difference error information of the encoder signal, which has been extremely difficult in the past, can be easily and accurately detected. be able to. Further, in the present invention, based on the ratio k of the long axis and the short axis of the Lissajous waveform, the phase of the coordinate value on the Lissajous waveform to be corrected is corrected, and the Lissajous waveform on which the phase difference error is corrected is corrected. The phase correction can be easily performed by including a program storage unit that stores in advance a calculation program for obtaining the coordinate value and a calculation unit that performs the phase correction based on the calculation program. Further, in the present invention, since the ratio of the lengths is substituted into the equation for obtaining the characteristic phase difference error in the present invention and the phase difference error calculating means for obtaining the phase difference error is provided, the encoder signal The phase difference error can be detected more easily and accurately. Further, according to the interpolation error estimation device of the present invention, since it is provided with the interpolation error calculation means for obtaining the interpolation error based on the phase difference error obtained from the phase difference error detection device of the present invention,
The interpolation error due to the phase difference error can be estimated easily and accurately.

[Brief description of drawings]

FIG. 1 is an explanatory diagram of a schematic configuration of an interpolation error estimation device according to an embodiment of the present invention.

FIG. 2 is an explanatory diagram of a schematic configuration of a phase difference error detection device according to the first embodiment of the present invention.

FIG. 3 is an explanatory diagram of a long-short axis detection method characteristic of the phase difference error detection device shown in FIG.

FIG. 4 is an explanatory diagram of a schematic configuration of a phase difference error detection device according to a second embodiment of the present invention.

5 is an explanatory diagram of a long-short axis detection method which is characteristic of the phase difference error detection device shown in FIG.

6 is a modification of the device shown in FIG.

7 is an explanatory diagram of a characteristic phase correction mechanism in the device shown in FIG.

8 is an explanatory diagram of a phase correction method by the phase correction mechanism shown in FIG.

[Explanation of symbols]

12 Encoder 14 Interpolation Circuit 16 Interpolation Error Estimator 22 Coordinate Axis Rotation Unit (Long / Short Axis Detection Means) 24 Switching Point Detection Unit (Long / Short Axis Detection Means) 58, 158 Long / Short Axis Ratio Calculation Means 60, 160 Phase Difference Error Calculation Means 172 Judging element (absolute value equal point detecting section, long / short axis detecting means) 174, 176, 178, 180 Sampling element (absolute value equal point detecting section, long / short axis detecting means) 184, 186 Axis length calculating section (long / short axis detecting means)

Claims (9)

[Claims] 1. A phase difference error detection device for detecting an error of a phase difference between two-phase sine waves A ₁ and B ₁ that are originally output from an encoder and have a phase difference of 90 degrees. The waves A ₁ and B ₁ substantially include only the phase difference error, and when expressed by the following formula 1, the two-phase sine wave A 1
_1. A long / short axis detection means for detecting the position of the long axis and the position of the short axis of the Lissajous waveform formed by synthesizing ₁ and B ₁ to obtain the length of the long axis and the length of the short axis; A long-short axis ratio calculating means for obtaining the length ratio k from the length of the long axis and the length of the short axis obtained by the means, and the ratio k of the long axis to the short axis of the Lissajous waveform A phase difference error detection device characterized by obtaining. Sine wave A ₁ = sin (u) sine wave B ₁ = cos (u + ε) (1) where u: 2πx / λ x: position of the encoder λ: cycle of the sine wave 2. The phase difference error detection device according to claim 1, wherein the long / short axis detection means rotates an XY coordinate axis on which the Lissajous waveform is formed by 45 degrees, and the axes of the Lissajous waveform are parallel or orthogonal. a coordinate axis rotation unit for converting the ₂ Y ₂ axis, and the Lissajous waveform, wherein X ₂ Y ₂ switching point detector for detecting the X ₂ Y ₂ coordinate values of the four quadrants of the switching point, divided by the coordinate axis, the X ₂ of the switching point of the Lissajous waveform obtained by the switching point detector
A phase difference error detection device comprising: an axis length calculation unit that obtains the length of the long axis and the length of the short axis of the Lissajous waveform on the X ₂ Y ₂ coordinate axis from the Y ₂ coordinate value. 3. The phase difference error detection device according to claim 1, wherein the long / short axis detection means uses, on the XY coordinate axes, points on the Lissajous waveform where the absolute values of the two-phase sine waves A ₁ and B ₁ are equal. The absolute value equal point detecting unit for detecting from each of the four divided quadrants and the XY coordinate values of each point obtained by the absolute value equal point detecting unit are used to determine the length and the length of the long axis of the Lissajous waveform on the XY coordinate axes. A phase difference error detection device comprising: an axis length calculator that determines the length of the axis. 4. The phase difference error detection device according to claim 1, wherein the ratio k between the major axis length and the minor axis length of the Lissajous waveform obtained by the major axis / minor axis ratio calculation means is calculated. Substituting in the following equation 2,
A phase difference error detecting device comprising phase difference error calculating means for obtaining a phase difference error ε between the two-phase sine waves A ₁ and B ₁ . Phase difference error ε = sin ⁻¹ ((1-k ² ) / (1 + k ² )) (2) 5. The phase difference error detection device according to claim 1, wherein the ratio k between the length of the major axis and the length of the minor axis of the Lissajous waveform obtained by the major / minor axis ratio calculating means is calculated. A program that pre-stores a calculation program for performing phase correction of the coordinate value on the Lissajous waveform to be corrected based on the following Equation 3 and including the coordinate value on the Lissajous waveform with the phase difference error corrected. A phase difference error comprising: a storage unit; and a calculation unit that obtains a coordinate value on the Lissajous waveform in which the phase difference error is corrected based on a calculation program stored in the program storage unit. Detection device. [Equation 3] Where X and Y are coordinate values on the Lissajous waveform to be corrected, and X ′ and Y ′ are coordinate values on the Lissajous waveform whose phase difference error has been corrected. 6. The phase difference error detecting device according to claim 5, further comprising a program creating unit for creating an arithmetic program stored in the program storing unit. 7. The phase difference error detection device according to claim 6, wherein the program creating unit stores an input unit for inputting the arithmetic program, and the arithmetic program input from the input unit in the program storage unit. A phase difference error detection device comprising: a processing unit that stores the phase difference error. 8. The phase difference error detection device according to claim 6 or 7, wherein the program creating means includes an output section for outputting the previously created operation program to the program storing means. And a phase difference error detection device. 9. The phase difference error ε of the two-phase sine waves A ₁ and B ₁ obtained by the phase difference error detection device according to claim 1 is substituted into the following expression 4 to obtain the phase difference error. An interpolation error estimator comprising an interpolation error calculation means for calculating an interpolation error E by ε. Equation 4] the interpolation error E = (λ / 2π) ( - εsin 2 u) ... (4) where, lambda: the period of the sine wave epsilon: phase difference error obtained from the phase difference error detection apparatus of the present invention u: 2πx / λ x: position of the encoder