JPH057641B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an optical scale reading device that uses optical interference, and more particularly to an optical scale reading device that uses amplitude-based interpolation to improve performance.

Various types of optical scale reading devices are conventionally known. FIG. 1 is a diagram showing the configuration of a conventional device of this type using amplitude-based interpolation. In the figure, a shows its mechanical configuration, and b shows its electrical configuration. In a, 1
2 is a light source, 2 is a beam splitter, 3 is a condenser lens, 4 is a light amount control detector that receives a portion of the reflected light from the beam splitter 2, 5 is a reference grating, 6 is a scanning grating, 7 is a condensing lens, 8 is a photoelectric conversion element. In b, 9 is a pickup that receives the output of the photoelectric conversion element 8 and outputs a sine wave and a cosine wave with different phases, and 10 is an analog synthesizer that receives both outputs of the pickup, synthesizes a triangular wave and a reference voltage, and outputs the result. A circuit 11 is a resistor voltage divider circuit that divides the reference voltage output from the analog synthesis circuit to create a plurality of reference voltages for comparison; A comparator circuit 13 outputs a logical value of "0" or "1" for each reference voltage after comparison, and a comparator circuit 13 receives the output of the comparator circuit and generates pulses corresponding to the moving distance of the scale 5 in each direction of rotation. This is the direction discrimination circuit that generates. The output of the direction discrimination circuit is counted and displayed by a counter. Conventional devices of this type are
It has the following defects.

(1) Since the sine wave and cosine wave are full-wave rectified to synthesize the DC reference voltage, the reference voltage is distorted and interpolation with good accuracy cannot be performed.

(2) Since the minimum resolution pulse is generated and then counted by a counter, errors during interpolation accumulate.

(3) Since the minimum resolution pulse must be output accurately even during high-speed movement, the analog synthesis circuit 10 is required to be high-speed.

(4) Increasing the resolution increases the frequency of the output pulse, and when moving at high speed, a high-speed counter is required. Furthermore, the movement speed is limited by the response speed of the counter.

The present invention was made in view of these points, and it creates a reference voltage using the square root of (sine wave) ² + (cosine wave) ² to make it a distortion-free direct current and improves the accuracy of analog division. , sine wave, and cosine wave output, whichever has better sensitivity is selected and input.
We achieved an ultra-high resolution optical scale reading device that eliminates the above drawbacks by dividing analog pulses using a D converter, counting pulses divided into 1/4 during high-speed movement, and performing analog interpolation during low-speed movement. It is something.

Hereinafter, the present invention will be explained in detail with reference to the drawings.

FIG. 2 is an electrical configuration diagram showing an embodiment of the present invention. In the figure, PD ₁ to PD ₃ are photoelectric conversion elements that each receive interference light with different phases and convert it into an electrical signal, 20 is a first amplifier that receives the output of the photoelectric conversion element PD ₃ , and 21 is a photoelectric conversion element PD _1. 22 is a photoelectric conversion element that receives the output of
This is the third amplifier that receives the output of PD ₂ . The output of the first amplifier 20 is transmitted to the second and third amplifiers 2
1 and 22 are input in common. 23 is a first comparator that receives the output of the second amplifier 21 and converts it into a pulse; 24 is a second comparator that receives the output of the third amplifier 22 and converts it into a pulse. 25 is a direction discrimination circuit which receives the outputs of the comparators 23 and 24 and divides the cycle into 1/4 and discriminates the moving direction of the scale; 26 is a counter which counts the pulse output of the direction discrimination circuit; and 27 is a second 28 is a second arithmetic unit that squares the output of the third amplifier 22.

29 is an adder that outputs the sum of squares of the arithmetic units 27 and 28; SW is a switch that selects one of the amplifiers 21 and 22; 33 is a square root circuit that calculates the output of the adder 29 and outputs the square root; 30 is an A/D whose reference voltage is the output of the square root circuit, and whose input unknown voltage is the sine wave sent via the switch SW.
A converter 31 receives the contents of the counter 26 and the output of the A/D converter 30, performs predetermined calculations, calculates the amount and direction of movement of the scale, and controls switching of the switch SW. 32 is a calculation control circuit. This is a display section that displays the output of the arithmetic control circuit. As the calculation control circuit 31, for example, a microcomputer is used. The operation of the device configured as described above will be explained as follows.

From the reading head (not shown), two sine waves with a phase difference of 90° are generated, corresponding to the amount of movement x of the scale.
In other words, sine waves and cosine waves are obtained. Photoelectric conversion element
Since the signals obtained by PD ₁ and PD ₂ each contain a DC bias, the DC output of amplifier 20 is added to each amplifier 21 and 22 to cancel it. If there is a difference in the bias values of PD ₁ and PD ₂ , the values of the input resistances R ₁ and R ₂ of the amplifiers 21 and 22 are adjusted to match the values. The outputs of the amplifiers 21 and 22 are converted into pulses by comparators 23 and 24, respectively, and then enter a direction discrimination circuit 25, where a signal indicating the moving direction and a pulse whose period is divided into 1/4 are generated. Output. These pulses are counter 2
The count is output at 6. By the method described above, PD ₁ ,
Since the output period of PD ₂ is 1/2 of the scale pitch, the resolution is 1/8 of the scale pitch. In the present invention,
The aim is to further interpolate the output pulses of the counter 26 using the A/D converter 30 to achieve ultra-high resolution.

An adder 29 that follows the squared outputs of the arithmetic units 27 and 28
The square root circuit 33 generates a reference voltage, and this reference voltage is used as the reference voltage of the A/D converter 30. The outputs e ₁ and e ₂ of the amplifiers 21 and 22 satisfy the following equation.

e ₁ = asinπx/d e ₂ = acosπx/d where a: amplitude, d: scale pitch, x: scale movement amount Therefore, the output of the square root circuit 33, that is, the reference voltage
e _r becomes as follows.

That is, the reference voltage of the A/D converter 30 always has an amplitude value regardless of the amount of movement of the scale, and even if the amplitude changes, accurate division is possible. on the other hand,
As the input unknown voltage of the A/D converter, a sine wave or a cosine wave, whichever has better sensitivity, is selected.
The selection is made by the arithmetic control circuit 31. That is, a switching signal is applied from the arithmetic control circuit 31 to the switch SW, so that the switch SW selects the one with higher sensitivity and inputs it to the A/D converter 30. When the scale is moving at high speed, the output pulses of the direction discrimination circuit 25 are counted, and when the scale is moving at low speed or stopped, the A/D converter 30 interpolates, so the reference voltage circuits 27 to 29 and the A/D conversion The device 30 may be of low speed operation. After the scale stops, the arithmetic control circuit 31 compares the value of the counter 26 and the A/
After receiving the value of the D converter 30, predetermined arithmetic processing is performed, and the moving distance and moving direction of the scale are displayed on the display unit 32, while the servo loop (not shown)
send data to.

FIG. 3 is a diagram showing the configuration of a reading head section used in the present invention.

In the figure, 41 is a coherent light source using a semiconductor laser or the like, 42 is a condenser lens that receives the emitted light from the light source, 43 is a reflective scale, 44 and 45 are mirrors that each receive the diffracted light reflected by the scale 43, 46 is a first half mirror that receives reflected light from these mirrors, 47 is a second half mirror that receives transmitted light from the first half mirror and mixes and interferes with it, and 48 and 49 are phases of the second half mirror. 50 and 51 are amplifiers that amplify the outputs of these light receiving elements, and 52 receives the outputs of these amplifiers and performs arithmetic processing to calculate the moving distance of the scale 43. A signal processing circuit, 53 is a display section that displays the output of the signal processing circuit, and 54 is a light receiving element that receives reflected light from the first half mirror 46. The operation of the device configured as described above will be explained as follows.

The output light of the semiconductor laser 41 is converted into an angle (or parallel light) that is focused by the lens 42 on the light receiving elements 48 and 49. At this time, the plane of polarization should be oriented as shown in the figure. scale this light 43
to project. As the scale, for example, a diffraction grating in which grooves are precisely carved at regular intervals or a diffraction grating made by holographic technology is used. The projected light is therefore diffracted. The diffraction angle θ at this time is the scale pitch d, and the wavelength of the semiconductor laser 41 is λ, the following equation holds true.

sinθ=mλ/d (m; integer) However, -90°≦θ90°, -1≦m/d≦+1, where, for example, if λ=0.78μm and d=0.83μm, m=0, ±1, θ=0° (0th order diffracted light when m=0), θ=±70.0° (±1st order diffracted light when m=±1). The ±1st-order diffracted light is mirrored by mirrors 44 and 4, respectively.
5, and after passing through a half mirror 46, they are mixed and interfered by a second half mirror 47.
This interfered light is transmitted to the light receiving element 48,
49, it is converted into an electrical signal. At this time, the interfering lights must have a phase difference of 90°.
The method is shown below. Figure 4 shows half mirror 4
FIG. 7 is a diagram illustrating a situation when interference occurs. In the figure, 60 is glass and 61 is a metal semi-transparent surface. Generally, when light is reflected from a metal surface, the phase is delayed, but when reflected or transmitted from a glass surface, the phase is not delayed.
In the figure, a half mirror 27 for −1st order diffracted light
Let δ _r1 be the phase lag due to reflection of the +1st-order diffracted light, δ _r2 be the phase lag due to reflection of the +1st-order diffracted light, and let δ _t1 to δ _t3 be the phase lag in the glass medium of the half mirror, respectively, as shown in the figure. The +1st order light is reflected and transmitted by the half mirror, and the light goes in the direction of the light receiving elements 48 and 49.
P ₊₁ , Q ₊₁ , −1st-order light that similarly travels in the direction of the light-receiving element are assumed to be P _-1 and Q _-1 . The phase delays of these four beams are as follows.

P ₊₁ ; δ _t1 + δ _r2 + δ _t2 , P _-1 ; δ _t3 Q ₊₁ ; δ _t1 , Q _-1 ; δ _r1 Therefore, the phase difference between P ₊₁ and P _-1 Δ ₁ , Q ₊₁ The phase difference Δ ₂ of Q _-1 is expressed by the following equations.

Δ ₁ = δ _t1 + δ _r2 + δ _t2 − δ _t3 Δ ₂ = δ _t1 − Δ _r1Here , if the optical paths of P ₊₁ and P _-2 are matched, δ _t2 =
δ _t3 . Therefore, the following equation holds.

Δ=δ _t1 +δ _r2 Now, P ₊₁ and P _-1 and Q ₊₁ and Q _-1 interfere with each other and enter the light receiving elements 48 and 49. At this time, if the phase difference between the outputs of the light receiving elements 48 and 49 is α, then α=Δ ₁ −Δ ₂ =δ _t1 +δ _r2 −δ _t1 +δ _r1 =δ _r1 +δ _r2 . Therefore, the phase difference between the light receiving elements 48 and 49 is determined only by δ _r1 and δ _r2 and is unrelated to the thickness of the glass of the half mirror 47. δ _r1 on the metal surface,
The value of δ _r2 is determined by the incident angle Φ and the angle of the plane of polarization of the incident light. δ _r1 and δ _r2 become maximum when the plane of polarization is oriented as shown in FIG. 4, and in this case, the following equation holds true from Fresnel's formula and the law of refraction.

R _p =tan(φ−x)/tan(φ+x)A _p sinx=Sinφ/n(1+ik) However, R _p ; Complex amplitude of reflected light A _p ; Complex amplitude of incident light x; Complex angle of refraction n; Refraction of metal Rate k: Attenuation constant If x is eliminated from the above equation, the phase delay δ of the reflected light is expressed by the following equation.

δ＝tan ^-1 R _p /A _p = tan ^-1 [2nktanφsinφ(tan ² φ+1
)／tan ² φ{n ² +(nk) ² }−sin ² φ(tan ² φ+1) ² ]
However, in the case of a half mirror, it is thought that there is reflection from the glass surface in addition to the metal surface.
The phase of reflection on the glass surface is reversed by 180° at the Brieuster angle.

Therefore, in the case of a half mirror, when the relationship between the incident angle Φ and the phase difference α between the light receiving elements 48 and 49 is actually measured, it is as shown in Fig. 5, which has both the characteristics of metal surface reflection and the characteristics of glass surface reflection. In the case of an Inconel half mirror, Φ = approximately 75° and α = 90°. In the figure, the horizontal axis represents the incident angle Φ, and the vertical axis represents the light receiving element 4.
8 and 49, respectively. Therefore, the direction of movement of the scale can be determined from this output, and the amount of movement can be determined by counting the number of waves of the sine wave. Since the output is a sine wave with an accurate 90° phase difference, further analog interpolation is performed to achieve a resolution of 1/100 to 1/100.
Ultra-high resolution of 0 μm can be obtained, which can be displayed or used for position control.
Various types of general configurations can be considered as the configuration for processing this 90° phase difference signal depending on the purpose. In this configuration, the diameter of the light beam projected onto the scale is approximately 4 to 5 mm, and the pitch d of the scale is approximately 4 to 5 mm.
If it is 0.8 μm, there are approximately 5,000 gratings within this beam diameter, and all of these gratings create one interference fringe. Therefore, the influence of lattice defects of the scale, small pitch irregularities, and dust and dirt attached to the scale can be greatly reduced. By the way, the first half mirror 46 and the light receiving element 54 in FIG. 3 are used to monitor the optical power of the ±1st order diffracted light, and create a voltage to remove the bias component of the output sine wave of the light receiving elements 48 and 49. It is something. In this way, even if the diffraction efficiency fluctuates at each location on the scale, or the intensity of the ±1st-order diffracted light changes due to dust or dirt, and the outputs of the light receiving elements 48 and 49 change, an accurate sine wave can be obtained. It can be accurately converted into pulses. However, if the scale is uniform and the change in position or angle during movement is small, it may not be necessary.

The features of the present invention described above are listed below.

(1) Since the counter only counts up to a resolution of 1/8 of the scale pitch, the capacity and speed of the counter may remain constant no matter how many times it is divided by the A/D converter. Therefore, high-speed response and ultra-high resolution can be achieved.

(2) Since the sine wave and cosine wave are switched and the parts with high sensitivity are A/D converted, even if the resolution of the A/D converter is low, even if the resolution is ultra-high (for example, when the resolution is 1/1000 μm, the (about a bit) good.

(3) Since the square root of sin ² θ + cos ² θ is used as the reference voltage of the A/D converter, an accurate DC voltage can be obtained, and even if the amplitude of the sine wave and cosine wave fluctuates, high accuracy can be achieved.
Ultra-high resolution can be obtained with a simple configuration.

(4) Since the pulse count of the counter and the division of the A/D converter are independent, errors when dividing by the A/D converter do not accumulate.

(5) Accurate amplitude values can be obtained over the entire time range from the moment the power is turned on, and physical quantities can be measured with high precision and high resolution without being affected by amplitude fluctuations.

In the above explanation, the switching signal for the switch SW is given from the arithmetic control circuit 31, but as shown in FIG. 2, a hardware logic circuit may be constructed and the switching control for the switch SW may be performed by the logic circuit. . Also, instead of switching with a switch SW, it is also possible to create a triangular wave by creating a difference between a sine wave and a cosine wave as in the conventional example, or to digitally convert each sine wave and cosine wave using a dedicated A/D converter as shown in Figure 6. Alternatively, the data may be converted into data, and the arithmetic control circuit may determine which data to select. In the figure, 70
71 is a first A/D converter that receives a sine wave; 71 is a second A/D converter that receives a cosine wave; 31 is an arithmetic control circuit that processes the output data of both of these A/D converters; They are the same as shown in the figure. The reading head is not limited to the one shown in Figure 3, but any linear scale such as an optical scale, moiré scale, magnetic scale, or electromagnetic scale, as long as it outputs a sine wave or cosine wave, Alternatively, it may be anything such as a rotary encoder. In addition to displacement sensors, we also offer optical waveguide type temperature sensors as shown in Figure 7, optical fiber type temperature sensors as shown in Figure 8, magnetometers using Faraday elements as shown in Figure 9, and current Any sensor may be used as long as it outputs a sine wave with a 90° phase difference, such as an optical waveguide type voltmeter as shown in FIG.

As explained in detail above, according to the present invention, the reference voltage is created using the square root of (sine wave) ² + (cosine wave) ² to improve the precision of analog division as a distortion-free direct current. The more sensitive one of the wave outputs is selected and input to the A/D converter, which divides it into analog signals.During high-speed operation, the divided pulses are counted by 1/4, and when moving at low speeds, analog interpolation is performed. It is possible to realize an optical scale reading device with ultra-high resolution and high performance.

[Brief explanation of drawings]

FIG. 1 is a diagram showing an example of a conventional device, FIG. 2 is an electrical configuration diagram showing an embodiment of the present invention, FIG. 3 is a diagram showing the configuration of a reading head section used in the present invention, and FIG. 4 is a half FIG. 5 is a diagram showing the phase relationship when interference occurs with a mirror. FIG. 5 is a diagram showing the relationship between the incident angle and the light receiving element. FIG. 6 is a diagram showing another embodiment of the present invention.
Figures 1 to 10 are diagrams showing various sensors. 1...Coherent light source, 2...Beam splitter, 3, 7...Lens, 5...Reference grating, 6...
Scanning grating, 8... Photoelectric conversion element, 9... Pickup, 10... Analog synthesis circuit, 11... Resistance voltage divider circuit, 12... Comparison circuit, 13... Direction discrimination circuit, 20-22... Amplifier, 23, 24... Comparator, 25... Direction discrimination circuit, 26... Counter, 27, 28... Arithmetic unit, 29... Adder, 3
0,70,71...A/D converter, 31...Arithmetic control circuit, 32...Display section, 33...Square root circuit, 34...Arithmetic circuit, 42...Lens, 44,
45... Mirror, 48, 49, 54... Light receiving element, 41... Coherent light source, 43... Reflective scale, 46, 47... Half mirror, 50, 51...
...Amplifier, 52...Signal processing circuit, 53...Display section, 60...Glass, 61...Metal anti-transparent surface.

Claims (1)

[Claims] 1. A sensor that outputs two sine waves whose phases differ by 90 degrees depending on changes in a physical quantity to be measured, or a portion of less than one cycle thereof, and a sensor that calculates the squares of the instantaneous values of the sine waves and adds them together. an analog circuit that takes the square root of the sine wave; an A/D converter that uses the output of the analog circuit as a reference voltage and inputs either one of the sine waves or the output of an arithmetic circuit that inputs the sine wave; a comparator circuit that converts the sine wave into a pulse in the case of a large number of repetitive signals; a direction discrimination circuit that discriminates the direction of change in the physical quantity from a 90° phase shift between the two sine waves; and the direction discrimination circuit. A physical quantity measuring system comprising: a counter that counts output pulses; and an arithmetic control circuit that interpolates the contents of the counter with the output of the A/D converter to calculate the amount of change in the physical quantity. 2 Any one of the two sine waves as the arithmetic circuit
2. The physical quantity measuring system according to claim 1, further comprising a switch for selecting one of the physical quantities. 3. The sensor is characterized by using an optical sensor that irradiates the scale with the output of a coherent light source, converts interference light between reflected diffracted lights into an electrical signal, and outputs a signal according to the amount of movement of the scale. A physical quantity measuring system according to claim 1.