JP2008076199A - Encoder - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an encoder that provides a sine wave where the distortion by an easy-to-manufacture binarizing slit is extremely small, allows high interpolation division to be performed from the sine wave with an extremely small distortion, is inexpensive, and has high resolution. <P>SOLUTION: The slit of a component for canceling the harmonic distortion included in a rectangular wave slit of a main scale to an index scale in a harmonic of each order is constituted, the sine wave of a fundamental wave without distortion is taken out, and interpolation division is performed. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to an encoder, and more particularly to an encoder that can be used for rotational positioning of a motor and linear positioning of a stage and the like to obtain a sine wave electric signal with extremely small distortion.

In a conventional encoder, when the index scale having a binary rectangular slit and a main scale having a binary rectangular slit is relatively moved, the light receiving element integrates a transmission port proportional to the amount of light transmitted through each slit. The obtained triangular wave is output. This output waveform was interpolated as a pseudo sine wave and used as an encoder output. However, since the pseudo sine wave includes a distorted wave, it cannot be interpolated with high accuracy.
As another means for improving the accuracy, there is a method of changing the slit to a sine wave instead of a rectangular shape. When the light receiving element receives light transmitted through the slit, the output of the light receiving element is also a sine wave, and the interpolation error is reduced. However, since it is difficult to manufacture the slit with a sine wave with high accuracy, it is difficult to increase the accuracy.
As an improvement measure other than the above, use multiple rows of rectangular slits, shift the phase of each slit, cancel out the harmonic distortion generated in each slit by harmonic distortion generated in other slits, and produce a sine wave output with low distortion A method of obtaining is proposed (see, for example, Patent Document 1). This method requires a plurality of rows of slits, and as a result, the size of the encoder is limited in terms of size. Further, since the obtained fundamental sine wave is a combination of harmonics obtained from each slit, the phase shifts from the slit pattern, which becomes an obstacle when it is necessary to match the phase with another slit pattern such as an absolute encoder.

JP-A-3-48122

As described above, the method of obtaining a sine wave output with a small distortion by using a plurality of rows of slits whose phases are shifted not only obstructs downsizing of the encoder but also other slit patterns such as an absolute encoder. There is a problem that becomes an obstacle when it is necessary to match the phases.
Therefore, the problem to be solved by the present invention has been made in view of the above-mentioned problems of the prior art, and both the main scale and the index scale use a binary rectangular slit that is easy to manufacture and does not contain distortion. Get the waves. In addition, when a sine wave output is obtained with a plurality of slits, it is difficult to reduce the size, so that an output can be obtained even with a single slit. Furthermore, it solves that the phase of the production slit pattern and the waveform output is shifted.

The encoder according to claim 1, which achieves the object, includes a main scale and an index scale that transmit or shield light emitted from a light source while relatively moving, and corresponds to the amount of light received through both scales. An encoder for outputting the electrical signal, wherein the pitch of the bright slit through which the light of the main scale is transmitted is P, the width of the bright slit is P / 2, and the angle P is linearly associated with the pitch P. In this case, the bright slit through which the light of the index scale is transmitted is composed of a plurality of N partial rectangular slits, and the width of the partial rectangular slit is set to [A _n , π−A _n ] (n = 1, 2, 3,..., N), A _n (n = 1, 2, 3,..., N) is expressed as Σ _{k = 1} ^N cos (3A _k ), Σ _{k = 1} ^N cos ( _{5A k), ···, Σ k} = 1 N cos each Σ value of _{[(2M + 1) * a} k] is B, or N and M are selected to be as small as possible (however, N and M are natural numbers), and each of the partial rectangular slits covers the bright part slit of the main scale at the time of maximum overlap of both scales. And a sinusoidal electric signal with extremely small distortion is output.
In the above encoder, the waveform of the output signal is proportional to the amount of light received by the light receiving element, and the amount of received light is the waveform of the overlapping transmission port (common transmission port) between the bright part slit of the main scale and the bright part slit of the index scale. Is proportional to the integrated waveform integrated over one period (pitch P), and the integrated waveform is a triangular wave or a trapezoidal wave. Configure the integrated waveform to be the sum of multiple triangular waves, etc. instead of one, and configure the sum of the Fourier series of harmonics when each waveform is Fourier decomposed to be zero or as small as possible. Thus, the waveform of the output signal becomes a sine wave output waveform with extremely small distortion from which harmonic components are removed. In other words, the integral transmission port shape (slit pattern) formed by the main scale and the index scale is composed of a plurality of rectangular slits, and the rectangular width of each rectangular slit is integrated over one period of the rectangular slit. By appropriately selecting the sum of the Fourier series of harmonics when the waveform is Fourier decomposed to be zero or as small as possible, the waveform of the output signal is a sine wave with extremely low distortion from which harmonic components have been removed. Output waveform.
As a result of intensive studies on the output waveform of the light receiving element of the encoder, the inventor of the present application has determined that the bright slit of the index scale is a plurality of N partial rectangular slits (slit width is [A _n , π−A _n ] (n = 1, 2 , 3,..., N)) with a stepped or uneven slit pattern, the harmonic component of the output waveform is −4 / (9π) × (Σ _{k = 1} ^N cos (3A _k )) cos (3x) -4 / (25π) × (Σk _{= 1} ^N cos (5A _k )) cos (5x)... -4 / (2M + 1) ² / π × (Σk _{= 1} ^N cos [(2M + 1 ) * A _k ]) cos ((2M + 1) x) and Fourier series expansion. Note that the waveform obtained by integrating the overlapping transmission ports has a Fourier coefficient of sin of zero from the even function.
Therefore, each value of Σ _{k = 1} ^N cos (3A _k ), Σ _{k = 1} ^N cos (5A _k ),..., Σ _{k = 1} ^N cos [(2M + 1) * A _k ] is zero or infinite. By selecting each _An (n = 1, 2, 3,..., N) so that it becomes smaller (however, N and M are natural numbers), the harmonic component of each term is suitably selected. As a result, a sinusoidal signal with extremely small distortion is output from the light receiving element.

In the encoder according to claim 2, the slit width of the bright portion slit through which the light of the index scale is transmitted is P / 6 (= π / 3 = [π / 3, π−π / 3]), 3P / 10. (= 3π / 5 = [π / 5, π−π / 5]), 11P / 30 (= 11π / 15 = [2π / 15, π-2π / 15]) and P / 2 (= π = [0 , Π-0]).
In the encoder, the bright part slits of the index scale are constituted by four partial rectangular slits (N = 4 in claim 1). That is, when A ₁ = 0, A ₂ = 2π / 15, A ₃ = π / 5, and A ₄ = π / 3, Σ _{k = 1} ⁴ cos (3A _k ) = Σ _{k = 1} ⁴ cos (5A _k ) = 0, and the harmonic components up to the fifth order become zero.
As described above, the shape of the bright part slit of the main scale is rectangular, and the shape of the bright part slit of the index scale is constituted by a plurality of partial rectangular slits within the range of the width of the rectangular slit. The harmonics generated by the rectangular slits are canceled or reduced, and as a result, only the fundamental wave and the direct current component can remain in the output waveform from the light receiving element. Therefore, the remaining fundamental wave and the direct current component pass through the differential amplifier, thereby making it possible to obtain a sinusoidal waveform with extremely small distortion leaving only the fundamental wave. In other words, when light passes through the binary rectangular slit of the main scale, the index scale has an element slit that can remove the arbitrary distortion order of the generated distortion harmonics. A sinusoidal waveform can be obtained.

  According to the encoder of the present invention, it is possible to obtain a distortion-free sine wave with a binary rectangular slit, so that interpolation division can be performed with high accuracy, and a high-resolution encoder can be manufactured at low cost. Further, since distortion can be removed with a single slit, the encoder can be miniaturized. Further, when a plurality of the slits are arranged, non-uniformities such as manufacturing variations are averaged and the accuracy is further improved.

  Hereinafter, the present invention will be described in more detail with reference to embodiments shown in the drawings.

FIG. 1 is an explanatory view showing a main part of an encoder 100 according to the present invention.
The encoder 100 includes a light source 0 that emits parallel rays, a bright portion slit that transmits parallel rays and a dark portion slit that shields the main scale 1 that moves relative to the index scale, and a bright portion slit that transmits light. A stepped index scale 2 and a light receiving element 3 that receives light that has passed through the bright slits of both the main scale 1 and the index scale 2 and outputs an electrical signal corresponding to the amount of light received. Configured.

  The operation of the encoder 100 will be briefly described. First, parallel rays emitted from the light source 0 are incident on the main scale 1, and light transmitted through the bright slit of the main scale 1 is incident on the index scale 2. The light transmitted through the two bright slits enters the light receiving element 3. On the other hand, the light receiving element 3 outputs a voltage proportional to the amount of light determined by the overlapping transmission port of the bright part slit of the main scale 1 and the bright part slit of the index scale 2.

  FIG. 2 is an explanatory diagram showing slit patterns of the main scale 1 and the index scale 2. 2A shows the slit pattern of the main scale 1, and FIG. 2B is an explanatory view showing the slit pattern of the index scale 2. FIG.

  As shown in FIG. 2 (a), the slit pattern 4 of the main scale has a bright part slit 4a that transmits light, and a dark part slit 4b that shields light have the same slit width P / 2, and the bright part slit. This is a binary rectangular pattern in which the pitch 4a is P.

  The index scale slit pattern 5 will be described in detail later with reference to FIG. 3, but the bright portion slits 5 a are patterned stepwise within a width of ½ the pitch P of the main scale 1.

FIG. 3 is an explanatory diagram showing details of the bright slit 5a of the index scale.
The bright portion slit 5a has a plurality of rectangular slits arranged stepwise in the range of the width P / 2, for example, a step pattern composed of four rectangular slits from the first bright portion slit 6 to the fourth bright portion slit 9 It has become.

  Here, in consideration of the Fourier series expansion described later, the pitch 0 → P of this slit is linearly corresponded to the angle 0 → 2π (therefore, P / 2 becomes π. Similarly, 11P / 30, 3P / 10, P / 6 corresponds to 11π / 15, 3π / 5, and π / 3, respectively, and in the following description, the first bright partial slit 6 has a height of 1 and a width of π (= [ 0, π]) rectangular slit, the second bright partial slit 7 has a height of 1 and a width of 11π / 15 (= [2π / 15, 13π / 15]), and the third bright partial slit 8 has a height. 1. A rectangular slit having a width of 3π / 5 (= [π / 5, 4π / 5]) and a fourth bright partial slit 9 having a height of 1 and a width of π / 3 (= [π / 3, 2π / 3]) ) Rectangular slit. Although details will be described later with reference to FIG. 5, the bright portion slit 5a of the index scale 2 is thus formed of a stepped rectangular slit, and the bright portion slit 4a of the main scale 1 is formed with a width P / 2. Further, by forming the rectangular slit having a height of 4 or more, a sine wave signal with extremely small distortion from which the harmonic component is removed in the light receiving element 3 is output.

FIG. 4 is an explanatory diagram showing a change in the overlapping transmission port between the bright part slit 4a of the main scale and the first bright part slit 6 of the index scale. It is assumed that the index scale 2 is stationary and the main scale 1 is moving.
4A shows the relative displacement ΔS = 0, FIG. 4B shows the relative displacement ΔS = P / 4, FIG. 4C shows the relative displacement ΔS = P / 2, and FIG. When the relative displacement ΔS = 3P / 4, (e) is when the relative displacement ΔS = P. Further, (f) shows the amount of incident light in the light receiving element 3 with respect to the relative displacement of ΔS = 0 to P. From these figures, it can be seen that the amount of light incident on the light receiving element 3 is proportional to the overlapping transmission aperture area So between the bright portion slit 4 a of the main scale 1 and the first bright portion slit 6 of the index scale 2. In particular, from ΔS = P / 2 to P, the overlapping transmission port area So is inversely proportional to the relative displacement. The decreasing speed is the same as the increasing speed from ΔS = 0 to P / 2, and the direction is the same. Therefore, the overlapping transmission port area So is the first bright partial slit 6 and the rectangular shape opposite to that of the first bright partial slit 6. It can be seen that the slit is proportional to the integrated value obtained by integrating the slit over the relative displacement (ΔS: 0 → P). Therefore, the waveform of the incident light quantity is a triangular wave as shown in FIG.

  The same applies to the second bright partial slit 7 to the fourth bright partial slit 9. As will be described later, the waveform of the incident light quantity from the second bright partial slit 7 to the fourth bright partial slit 9 is a trapezoidal wave.

  FIG. 5 is a graph showing each partial slit and the amount of incident light corresponding thereto. 3 is decomposed into individual rectangular shapes of the first bright partial slit 6 to the fourth bright partial slit 9, and incident light received by the light receiving element 3 through each partial slit. It represents the amount of light and the relative position (relative displacement). The output of the light receiving element 3 is the total amount of light passing through each partial slit (rectangular slit). The relative position of the horizontal axis is indicated by the corresponding angle (in radians).

  As described above, the stepped slit pattern of the index scale 2 has a wave height of 1, a first bright partial slit 6 having a width π (= [0, π]), and a width 11π / 15 (= [2π / 15, 13π / 15). ]) Second bright partial slit 7, width 3π / 5 (= [π / 5, 4π / 5]) third bright partial slit 8, width π / 3 (= [π / 3, 2π / 3]) The fourth bright partial slit 9 is broken down into individual rectangular waves. The amount of light incident on the light receiving element is proportional to the common opening area (overlapping transmission port area) of the bright slit of the main scale and the individual rectangular slits (the first bright partial slit 6 to the fourth bright partial slit 9). When the bright part slit 4a (bright part width π) of the scale passes through the index scale 2, the amount of light incident on the light receiving element is an integral value of each individual rectangular waveform (first bright part slit 6 to fourth bright part slit 9). It becomes. Therefore, it is proportional (increases) to the relative position.

  The waveform corresponding to the amount of light shielded by the dark slit 4b of the main scale has a wave height of −1, a first dark partial slit 10 having a width π (= [π, 2π]), and a width 11π / 15 (= [17π). / 15,28π / 15]), second dark partial slit 11, width 3π / 5 (= [6π / 5, 9π / 5]), third dark partial slit 12, width π (= [4π / 3, 5π). / 3]) is an integral value of the individual rectangular wave decomposed into the fourth dark partial slit 13, and is inversely proportional (decreased) to the relative position.

In general, the bright part slit constituting the bright part slit of the index scale 2 is written as [A _n , π−A _n ] (n = 1, 2, 3,..., N) when the wave height is 1. I can do it. In the present embodiment, when N = 4, that is, the bright part slit of the index scale 2 is formed in a stepped slit pattern with four partial slits. As for the A _n, the first light portion slit 6 A ₁ = 0, the second light portion slit 7 A ₂ = 2π / 15, the in 3 bright portion slit 8 A ₃ = π / 5, 4 Ming In the partial slit 9, A ₄ = π / 5.

As shown in FIG. 5, the waveform ya14 integrated over one period of the first bright partial slit 6 and the first dark partial slit 10 becomes a triangular wave, while the second bright partial slit 7, the second dark partial slit 11, A waveform yd15, a waveform yc16, and a waveform yb17 obtained by integrating the waveforms of the third bright partial slit 8 and the third dark partial slit 12, and the fourth bright partial slit 9 and the fourth dark partial slit 13 over one period are all trapezoidal waves. Become. Each formula is expressed in [Formula 1].
[Equation 1]
Waveform 14: ya = x (0 ≦ x <π)
= 2π-x (π ≦ x <2π)
Waveform 17: yb = 0 (0 ≦ x <π / 3)
= X-π / 3 (π / 3 ≦ x <2π / 3)
= Π / 3 (2π / 3 ≦ x <4π / 3)
= 5π / 3-x (4π / 3 ≦ x <5π / 3)
= 0 (5π / 3 ≦ x <2π)
Waveform 16: yc = 0 (0 ≦ x <π / 5)
= X-π / 5 (π / 5 ≦ x <4π / 5)
= 3π / 5 (4π / 5 ≦ x <6π / 5)
= 9π / 5-x (6π / 5 ≦ x <9π / 5)
= 0 (9π / 5 ≦ x <2π)
Waveform 15: yd = 0 (0 ≦ x <2π / 15)
= X-2π / 15 (2π / 15 ≦ x <13π / 15)
= 11π / 15 (13π / 15 ≦ x <17π / 15)
= 28π / 15-x (17π / 15 ≦ x <28π / 15)
= 0 (28π / 15 ≦ x <2π)
ya, yb, yc, and yd are rectangular waves (first bright partial slit 6 and first dark partial slit 10, second bright partial slit 7 and second dark partial slit 11, third bright partial slit 8 and third dark portion It represents a value proportional to each light quantity (light receiving element output) obtained by integrating the partial slit 12, the fourth bright partial slit 9, and the fourth dark partial slit 13), and x represents a relative movement distance corresponding to an angle (0 → 2π).
Therefore, the total output Y of the light receiving element is the sum of the square wave integral values, that is, the total waveform (Y = ya + yb + yc + yd) of the waveforms ya, yd, yc, yb14-17.

By the way, the total waveform can be expressed by a harmonic order component of a sine wave and a cosine wave by Fourier series expansion (Fourier decomposition) of [Equation 2], a fundamental wave component, and a direct current component. Further, since the total waveform is the sum of the square wave integral values, it can be obtained by summing the waveform components obtained by Fourier decomposition of the square wave integral values.
[Equation 2]
_{U = a 0/2 + Σ} {a n * cos (n * x) + b n * sin (n * x)}
n: Harmonic order (indicated up to 1, 2, 3, 4, 5 5th order)
a ₀ : DC component a _n , b _n : Fourier coefficient
a _n = 1 / π * ∫ (y * cos (n * x)) dx) 0 → 2π integration
b _n = 1 / π * ∫ (y * sin (n * x)) dx) 0 → 2π integration
y: each integral value ya, yb, yc, yd
When a _n and b _n of each integral value y are obtained by the above formula and Fourier decomposition is performed, each value is as follows.

First, it is obtained from the Fourier coefficient for the waveform ya14. Since the waveform ya14 is an even function, b _n = 0. That is, Fourier coefficients for the waveform ya14: b ₁ = b ₂ = b ₃ = b ₄ = b ₅ = 0,
a ₀ = π,
a ₁ = −4 / π,
a ₂ = 0,
a ₃ = −4 / (9π),
a ₄ = 0,
a ₅ = −4 / (25π)
Therefore, Fourier decomposition of waveform ya14:
Ua = π / 2-4 / π * cos (x) -4 / (9π) * cos (3x) -4 / (25π) * cos (5x)

Next, a Fourier coefficient for the waveform yb17 is obtained. Since the waveform yb17 is also an even function, b _n = 0. That is, Fourier coefficients for the waveform yb17: b ₁ = b ₂ = b ₃ = b ₄ = b ₅ = 0,
a ₀ = π / 3,
a ₁ = −2 / π,
a ₂ = 0,
a ₃ = 4 / (9π),
a ₄ = 0,
a ₅ = −2 / (25π)
Therefore, Fourier decomposition of waveform yd15:
Ub = π / 6-2 / π * cos (x) + 4 / (9π) * cos (3x) −2 / (25π) * cos (5x)

Next, a Fourier coefficient for the waveform yc16 is obtained. Since the waveform yc16 is also an even function, b _n = 0. That is, Fourier coefficients for the waveform yc16: b ₁ = b ₂ = b ₃ = b ₄ , b ₅ = 0,
a ₀ = 3π / 5,
a ₁ = −4 * cos (π / 5) /π=−1.03,
a ₂ = 0,
a ₃ = −4 / 9 * cos (3π / 5) /π=0.437,
a ₄ = 0,
a ₅ = −4 / 25 * cos (5π / 5) / π = 4 / (25π),
Fourier decomposition of yc:
Uc = 3π / 10−1.03 * cos (x) + 0.0437 * cos (3x) + 4 / (25π) * cos (5x)

Next, a Fourier coefficient for the waveform yd15 is obtained. Since the waveform yd15 is also an even function, b _n = 0. That is, the Fourier coefficient for waveform yd15: b ₁ = b ₂ = b ₃ = b ₄ = b ₅ = 0,
a ₀ = 11π / 15,
a ₁ = −4 * cos (2π / 15) /π=−1.163
a ₂ = 0,
a ₃ = −4 / 9 * cos (3 * 2π / 15) / π = −4 / 9 * cos (2π / 5) /π=−0.0437,
a ₄ = (− 11/30 * cos (π / 30) + 11/30 * sin (7π / 15)) / π = 0,
a ₅ = −4 / 25 * cos (5 * 2π / 15) / π = −4 / 25 * cos (2π / 3) / π = 2 / (25π),
Fourier decomposition of yd:
Ud = 11π / 30−1.1631 * cos (x) −0.0437 * cos (3x) + 2 / (25π) * cos (5x)

As shown in [Equation 3] below, the total U of the waveforms obtained by Fourier-decomposing each waveform ya, yb, yc, yd obtained by integrating each bright part slit and dark part slit is proportional to the output of the light receiving element. Value.
[Equation 3]
U = Ua + Ub + Uc + Ud
= Π / 2 + π / 6 + 3π / 10 + 11π / 30
− (4 / π + 2 / π + 1.03 + 1.1631) * cos (x)
+ (− 4 / 9π + 4 / 9π + 0.0437−0.0437) * cos (3x)
+ (-4 / 25π-2 / 25π + 4 / 25π + 2 / 25π) * cos (5x)
From [Equation 3], the third harmonic contained in the output waveform by the first partial slit 6 is canceled out by the third harmonic of the opposite phase contained in the output waveform by the second partial slit 7. Further, the fifth harmonic contained in the first partial slit 6 is canceled out by the reverse fifth harmonic contained in the output waveform from the third partial slit 8, and only the fundamental wave is included in the output waveform from the first partial slit 6. appear.

  Similarly, the fifth harmonic contained in the output waveform from the second partial slit 7 is canceled out by the fifth harmonic contained in the output waveform from the fourth partial slit 9. Further, the third harmonic contained in the output waveform by the third partial slit 8 is canceled by the third harmonic contained in the output waveform by the fourth partial slit 9, resulting in an output waveform by a slit waveform other than the first partial slit 6. Only the fundamental wave appears.

  From the above, the light receiving element 3 can extract a sine wave having a slit pitch period P and having no DC component and harmonic distortion component. If only a sine wave is extracted and interpolated by a differential amplifier, a high-resolution encoder can be configured.

  The output of the light receiving element 3 has the same shape as the index slit of FIG. The embodiment is also performed in the combined slit shape of FIG. 3 and the disassembled slit shape of FIG.

  Although the above embodiment has shown a configuration that cancels the fifth-order harmonic distortion component, it is possible to cancel the seventh-order harmonic distortion by adding four slit patterns, and to extract a sine wave with smaller distortion. Can do.

In general, a waveform obtained by integrating a rectangular wave having a wave height of 1 and a width of [A, π-A], a wave height of −1 and a width of [π + A, 2π-A] over one period of relative displacement becomes an even function, and Fourier when the series expansion, Fourier coefficients _{a n} relates cos, is omitted for the derivation _process, a n = 0 (n: even number) or ^{_{a n = -4 / (n 2}} π) * cos (nA) (n : Odd). A is a constant that determines the slit width [A, π-A] of the bright partial slit. A = 0 in the waveform ya14, A = 2π / 15 in the waveform yd15, A = π / 5 in the waveform yc16, and the waveform. In yb17, A = π / 3.
Therefore, when the bright part slit of the index scale 2 is composed of N rectangular bright part slits, the total U of each waveform is
U = (DC component)
−4 / π * (cosA ₁ + cosA ₂ +... + CosA _N ) cos (x) −4 / 9 / π * (cos (3A ₁ ) + cos (3A ₂ ) +... + Cos (3A _N )) cos (3x) −4 / 25 / π * (cos (5A ₁ ) + cos (5A ₂ ) +... + Cos (5A _N )) cos (5x). Therefore, Σ _{k = 1} ^N cos (3A _k ), Σ _{k = 1} ^N cos (5A _k ),..., And therefore each value of Σ _{k = 1} ^N cos [(2M + 1) * A _k ] is zero or By selecting each A _n (n = 1, 2,..., N) so as to approach zero as much as possible, it is possible to obtain an extremely sine wave output with distortion in the light receiving element. When A _n (n = 1, 2,..., N) is determined, the rectangular partial slits constituting the bright slits of the index scale are determined, and as a result, the stepped slit pattern is suitably determined. The In the above embodiment, N = 4 and higher harmonic components of 7th order and higher are ignored.

FIG. 6 is an explanatory diagram illustrating details of the bright slits of the index scale according to the second embodiment.
The bright part slit of the index scale has a configuration in which the order of the first bright part slit 6 and the third bright part slit 8 is exchanged in the bright part slit of FIG. That is, even if the order of each bright part slit is changed in this way, the shape of the overlapping transmission port (common transmission port) between each bright part slit and the bright part slit 4a of the main scale does not change. In addition, the light transmitted through the main scale 1 and the bright slit of the index scale is incident on the light receiving element 3, and the output waveform output from the light receiving element 3 is a sine wave with extremely small distortion from which harmonic components are removed. Become.

  Therefore, as the slit pattern of the index scale 2 according to the present invention, when the bright part slit is composed of N bright part slits, N! There will be.

  Since a sine wave with extremely small distortion can be produced at low cost, it can be used for a high-resolution position detector of a rotary motor or linear motor. In addition, since the output wave pattern coincides with the slit pattern of the fundamental wave component, it is easy to manufacture when making an absolute encoder.

It is principal part explanatory drawing which shows the encoder which concerns on this invention. It is explanatory drawing which shows the slit pattern of a main scale and an index scale. It is explanatory drawing which shows the detail of the bright part slit of an index scale. It is explanatory drawing which shows the change of the overlapping permeation | transmission opening | mouth of the bright part slit of a main scale, and the 1st bright part slit of an index scale. It is a graph which shows each partial slit and the incident light quantity corresponding to it. It is explanatory drawing which shows the detail of the bright part slit of the index scale which concerns on Example 2. FIG.

Explanation of symbols

0 Light source 1 Main scale 2 Index scale 3 Light-receiving element 4 Main scale slit pattern 5 Index scale slit pattern P Main scale slit period 6 First bright part slit 7 Second bright part slit 8 Third bright part slit 9 Fourth Bright partial slit 10 First dark partial slit 11 Second dark partial slit 12 Third dark partial slit 13 Fourth dark partial slit 14 Waveform ya
15 Waveform yb
16 Waveform yc
17 Waveform yd

Claims (2)

An encoder having a main scale and an index scale that transmit or shield light emitted from a light source while moving relative to each other, and that outputs an electrical signal corresponding to the amount of light received through both scales. When the pitch of the bright part slits through which the light passes is P, the width of the bright part slits is P / 2, and the angle P is linearly associated with the pitch P, there are a plurality of bright part slits through which the light of the index scale passes. is composed of n partial rectangular slit, and the width of each of the partial rectangular slit _{[a n, π-a n} ] (n = 1,2,3, ···, n) when the, the a _n ( n = 1,2,3, ···, n) ^{_{is, Σ k = 1 n cos (}} 3A k), Σ k = 1 n cos (5A k), ···, Σ k = 1 n cos [( 2M + 1) * a _k] becomes smaller each Σ value without zero or only (but, N, M is a natural number The partial rectangular slits are configured to be covered by the bright slits of the main scale at the time of maximum overlap of both scales, and output a sine wave electric signal with extremely small distortion. The encoder characterized by doing.   The bright part slit through which the light of the index scale transmits has a slit width of P / 6 (= π / 3 = [π / 3, π−π / 3]), 3P / 10 (= 3π / 5 = [π / 5, π−π / 5]), 11P / 30 (= 11π / 15 = [2π / 15, π−2π / 15]) and P / 2 (= π = [0, π-0]) partial rectangles The encoder according to claim 1, wherein the encoder is decomposed into slits.

Images