JPH11271026A - Dimension measuring apparatus using optical scale - Google Patents

Abstract

PROBLEM TO BE SOLVED: To realize a dimension measuring apparatus whose constitution is simple and whose a precision is high by a method wherein, while only two fixed scales are used, the phase of a two-phase sine-wave signal is detected directly by a numerical- value computing operation and the dividing precision between pitches of a lattice is increased. SOLUTION: A light receiving part 12 is constituted of a photodetector 120 and a photodetector 125 which are used to detect beams of light which are transmitted through a lattice group 110 and a lattice group 115 in a fixed scale 11. The photodetector 120 outputs an A-phase signal 122, and the photodetector 125 outputs a B-phase signal 127. The signals 122, 127 are sine-wave signals whose phase is different by π/2, and they are signals which form one cycle when a lattice is moved by one pitch. A counter part 13 judges the relationship between the advance and the delay of phases in the signals 122, 127 it detects the movement direction of a moving scale 10, it performs a counting operation, and it detects the number od movements of the lattice in the moving scale 10. For example, while the A-phase signal 122 is used as a reference, one pulse is generated in every cycle of sine waves, and the number of pulses is counted by a counter circuit. Thereby, a movement distance which corresponds to the integral multiple of the number of movements of the lattice is detected.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a dimension measuring apparatus using an optical scale on which a large number of grating patterns are formed, and to an arrangement for detecting the phase of a detected sine wave signal with high resolution to enhance the dimension measuring accuracy. .

[0002]

2. Description of the Related Art In a production line, there is a strong need to measure the size, shape, and the like of a workpiece on the spot with an accuracy in the micrometer range. As a simple measuring device for this purpose, a stylus-type dimension measuring device in which a linear encoder having a grid pattern formed thereon is attached to a stylus and moved is often used. This is an apparatus that irradiates light to a grid pattern, detects light intensity that changes with movement of a stylus, and measures dimensions. FIG. 5 shows an example of the configuration of a conventional dimension measuring device, and its operation will be described. Light emitted from a light source 50 such as a white lamp or a light emitting diode is collimated by a collimator lens 51.
The moving scale 52 is illuminated via. Moving scale 5
Reference numeral 2 denotes a large number of grids having a binary (0 and 1) light transmission intensity distribution formed on a glass substrate. The grid width of a black pattern that does not transmit light and the gap width of a white pattern that transmits light are different. Both are a / 2, and one pitch length of the grating is a. A normal pitch length of one grating is about 10 μm, and the length a is a reference scale for dimension measurement. The moving scale 52 moves in the up and down direction indicated by the arrow with the movement of the stylus (not shown).

[0003] Behind the movable scale 52, there are provided four fixed scales 53, 54, 55 and 56 which are fixed and do not move. These four fixed scales 53 to 56 and the movable scale 52 are collectively called an optical scale. The above fixed scales are individually formed at different positions on one glass substrate. The grid pitch length of each fixed scale is a, and both the grid width and the gap width are a / 2. When the grid of the fixed scale 53 is set as a reference, the grid position of the fixed scale 54 is set to be shifted by a / 4 from the grid of the fixed scale 53, and the phase shift with respect to the fixed scale 53 is set to π /. Let it be 2. Similarly, the fixed scales 55 and 56 have a / 2,
The lattice position is shifted by 3a / 4, and the phase shift is π,
3π / 2. Behind the above four fixed scales, four light receivers 535, 545, 555 and 5
65, the moving scale 52 and the fixed scale 53
The light intensity transmitted through .about.56 is individually detected by each light receiver.

The light intensity of each of the four signals (four-phase signals) detected by the above configuration changes in a sinusoidal manner in accordance with the movement of the moving scale 52. 4 fixed scales 53-5
6, the phases of the four-phase signals also differ by π / 2. The signal processing unit 57 performs data processing on the four-phase signals having different phases, detects the moving direction of the moving scale 52 and the distance moved, and measures the dimensions. Since it is not possible to detect both the moving direction and the moving distance of the moving grating 52 with only one-phase sine wave signal, it is necessary to create the four-phase signals having different phases.

FIG. 6 shows an example of a light intensity signal detected by the above method, and a conventional signal processing operation will be described. Sine wave signals 60, 61, 62, 6 whose phases are shifted by π / 2
3 is a moving scale 52 and fixed scales 53, 54, 5
5 and 56, signals obtained by passing through phase A and phase B, respectively.
Phase, C-phase, and D-phase signals. Of these four signals, for example, the moving direction indicated by the arrow of the moving scale 52 is determined from the relationship between the lead and lag of the phases of the A-phase signal 60 and the B-phase signal 61 (the phase difference is π / 2). The example in the figure is an example in which the phase of the B-phase signal 61 is advanced. For example, it is determined that the moving scale 52 is moving in the upward direction indicated by the arrow. By moving in the opposite direction, the phase relationship between the A-phase signal 60 and the B-phase signal 61 is reversed. In accordance with the moving direction, the number of grids moved on the moving scale 52 is detected by an up / down type counter circuit. In this case, for example, when one pulse is generated in one cycle of the A-phase signal 60 and the number of grid movements is measured, the counter circuit can detect a moving distance that is an integral multiple of one pitch length a of the grid.

In order to increase the resolution of dimension measurement, one of the gratings is required.
It is necessary to precisely measure the moving distance of the pitch length a or less. For that purpose, it is important how to finely divide one pitch of the grating and measure the moving distance of one grating length or less from the stop position of the grating to the next grating. For example,
In the case of a = 10 μm, in order to realize a resolution of 0.1 μm, it is necessary to detect a grid position by dividing one pitch of the grid by 100. The operation of the conventional division method will be described below.
The signal 64 is a rectangular signal obtained by binarizing the A-phase signal 60.
The signal 65 is a signal obtained by comparing the intensities of the A-phase signal 60 and the D-phase signal 63. The signal 65 has an H level when the intensity of the A-phase signal 60 is higher than the intensity of the D-phase signal 63, and has an L level when the intensity is lower. This is a rectangular signal. The signal 66 is a rectangular signal obtained by binarizing the B-phase signal 61. The signal 67 is a signal obtained by comparing the intensities of the C-phase signal 62 and the D-phase signal 63. The signal 67 has an H level when the intensity of the C-phase signal 62 is higher than the intensity of the D-phase signal 63, and has an L level when the intensity is lower. This is a rectangular signal.

The above four rectangular signals 64-67 have a phase of π.
This signal is shifted by / 4 (（of one pitch of the grating). For example, when the strength levels of the rectangular signals 64, 65, 66, 67 at the position where the moving scale 52 starts to move from the stop are H, H, H, L, the phase of the A-phase sine wave signal 60 is 90 degrees. Can be determined to be in the region between 135 degrees
One stop pitch of the grid can be detected by dividing one pitch of the grid of the moving scale 52 into eight. This phase domain detection
Any area within the range of 0 to 2π of the sine wave can be detected. In this way, by comparing the intensity of the rectangular signal obtained from the mutual intensity relationship of the four-phase sine wave signal, it is determined which phase region of the sine wave the detected light intensity is in, and it is more than one pitch length a of the grating. The stop position of the grid is detected with fine positional accuracy. The above is an example in which one cycle is divided into eight. However, one cycle of a sine wave signal is further divided into eight by similar circuit means using a resistance division method or the like.
It is also possible to divide more than division.

[0008]

An A phase, a B phase, a C phase, and an A phase which are detected by being transmitted through a moving scale 52 that moves together with a stylus and four fixed scales 53 to 56 whose positions are fixed.
The D-phase four-phase sine wave signals need to have mutually different phases by π / 2. Therefore, the grid position of another fixed scale with respect to the grid position of the fixed scale 53 is na /
4 (n = 1, 2, 3) so that
It is necessary to adjust the mutual grid positions of the fixed scales. For example, when one pitch length of the grating is 8 μm, it is necessary to shift each grating position of the four fixed scales at a pitch of 2 μm to create a grating pattern. for that reason,
Each of the four gratings has to be machined with an accuracy of about 1 μm, which makes it difficult to manufacture the gratings and increases the cost.

The most important factor for improving the dimension measurement accuracy is how to detect one period of the grating by finely dividing it. In the conventional detection method shown in FIG. 6, four phases A, B, C, and D are used.
Detect two sinusoidal signals and compare their intensity levels,
One period of the grating is divided into eight by circuit means for creating four rectangular signals. In this case, the pitch length a of the grating 1 is 8
In the case of μm, the resolution of dimension measurement is 1 μm, and there is a problem in accuracy that resolution in submicron cannot be obtained. When dividing into 16 to increase the resolution, about 8 rectangular wave signals are required, and there is also a problem that a circuit configuration therefor becomes complicated.

In the above lattice division, the A phase, the B phase, the C phase
It is necessary that the intensity level of each of the phase and D phase signals be equal. If the intensity (amplitude) of the four-phase signal is different, the phase of the rectangular signal obtained by comparing the two sine wave intensities is shifted as compared with the normal case. As a result, the phase division accuracy of one period of the grating is deteriorated and the dimension measurement accuracy is reduced. 4
In order to equalize the phase signal intensities, it is necessary to adjust the amplifier gain of the photoelectric conversion of the light intensity, and there is a problem in that the adjustment becomes complicated. Further, the necessity of four-phase signals requires four light receivers corresponding to four fixed scales, thus increasing the cost of the apparatus.

The main causes of the above-mentioned problems are that a four-phase sine wave signal is required to detect the phase of the grating, and that a rectangular signal is created in a circuit by binarization. is there. Therefore, the phase of the rectangular signal is affected by the sine wave signal intensity, and the division of one pitch of the grating is coarse. Therefore, in order to solve the above-mentioned problems, the present invention directly detects the phase of a two-phase sine wave signal by numerical operation using only two fixed scales, and improves the division accuracy between one pitch of the grating. It is an object of the present invention to realize a highly accurate dimension measuring device having a simple configuration.

[0012]

In order to solve the above-mentioned problems, a dimension measuring apparatus according to the present invention has the following arrangement. It is composed of a moving scale on which a grid having a constant pitch that moves with the stylus is formed, and a fixed scale fixed and installed near the moving scale with the same pitch as the grid of the moving scale. In a dimension measuring apparatus using an optical scale, the fixed scale has a phase of π / 2.
When the moving scale moves and outputs two sine-wave signals of A phase and B phase which are different from each other by π / 2 when the moving scale moves, the A scale outputs when the moving scale moves. A counter unit that counts the number of movements of the grating according to the relationship between the advance and delay of the mutual phase of the phase signal and the B-phase signal, and a specific period within a period from when the moving scale starts moving to when it stops. The A-phase signal and the B-phase signal are represented by A
A light intensity storage unit that stores the light intensity in a memory circuit after performing the D / D conversion, and the A-phase signal and the B signal that are stored in the light intensity storage unit.
A phase quadrant detection unit that detects a phase quadrant of a sine wave when the moving scale is stopped from the light intensity of the phase signal, a trigonometric function table unit that provides values of sin and cos in a specific range of the sine wave, The light intensity of the A-phase signal or the B-phase signal in the phase quadrant detected by the phase quadrant detector is compared with the value stored in the trigonometric function table to determine the phase when the moving scale is stopped. From the phase calculation unit to be detected and the count value counted by the counter unit,
A dimension calculating unit is provided for detecting a moving distance of an integral multiple of the pitch length and for detecting a moving distance of one grating length or less from the phase detected by the phase calculating unit, and measures a dimension from the moving distance of the moving scale. .

At this time, the phase quadrant detection section detects the amplitudes of the A-phase signal and the B-phase signal stored in the light intensity storage section, and normalizes the amplitude into a signal having an amplitude of ± 1. And a unit for judging the leading and lagging of the mutual phases of the standardized A-phase signal and the B-phase signal, and determining the standardized A-phase signal and the B-phase signal when the moving scale is stopped. It comprises an intensity code determination unit for comparing intensity codes, and detects the phase of the standardized A-phase signal or B-phase signal in a quadrant in units of π / 2. Further, the trigonometric function table section calculates the s of the sine wave signal when the sine wave phase is within 0 to π / 2.
Give the values of in and cos. Further, the phase calculation unit includes:
When the absolute value of the light intensity of any one of the normalized A-phase signal and the B-phase signal is close to 1, the phase is detected from the sin or cos value of the other signal. It is.

[0014]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention uses a moving scale and an optical scale consisting of two fixed scales, calculates two-phase sine wave signals, directly detects the phase, and divides the grid into one pitch. It is the structure which raises. Both lattice pitches are a
Then, when two fixed scales whose lattice positions are shifted by a / 4 (phase difference of π / 2) are set (fixed) behind the moving scale (lattice pitch = a), and the moving scale moves. , Two-phase sine wave signals (A-phase signal and B-phase signal) having phases different by π / 2 are detected. The A-phase and B-phase signals are generally signals having different intensity levels due to a difference in amplification gain of the photoelectric conversion circuit. Two different operations are performed on the detected A-phase and B-phase signals. One operation is similar to the conventional device,
The number of moving grids on the moving scale is counted using a counter circuit, and a moving distance that is an integral multiple of the grid pitch a is detected. The other operation relates to the present invention, and the position (phase) of the grating when the moving scale is stopped can be determined from the light intensities of the A-phase and B-phase two-phase signals without creating the rectangular signal described above. Detect directly. In this phase detection, sine wave intensity data for several cycles of the A-phase signal and the B-phase signal are A / D converted and stored in the memory circuit for two specific periods, ie, when the moving scale starts and stops moving. Then, the phase of the grating stop position is detected by numerical processing.

Since the amplitudes of the A-phase and B-phase signals stored in the memory circuit are generally different, the respective amplitudes are detected and normalized to a sine wave signal having both amplitudes of ± 1, and the A-phase and B-phase signals are Equal strength. This normalization is performed by numerical operation using a microprocessor or the like. Next, the phase quadrants (1 to 4) at the stop position of the grating, that is, the phase, are determined from the positive and negative signs (four combinations) of the normalized A-phase and B-phase signals at the position where the grating is stopped. Is detected in which area is in units of π / 2. The phase quadrant at this time is a value based on one of the two-phase signals, for example, the A-phase signal, and represents a coarse phase region in which one pitch of the grating is divided into four. When viewed in one cycle of a sine wave, since two phases correspond to the same intensity, the phase cannot be determined from the intensity alone. However, once the phase quadrant is determined, the phase can be directly detected from the intensity value of the sine wave at that time. It should be noted that even if the standardized two-phase signals have the same sign
Since the phase quadrant is different depending on the advance and delay of the phase of the B-phase signal with respect to the phase signal, it is necessary to determine the phase quadrant by detecting the relationship between the advance and delay of the phase of the two-phase signal.

Next, a phase is detected from one of the intensity values of the normalized two-phase signal corresponding to the phase quadrant. At this time,
A numerical table of trigonometric functions of sin and cos is prepared in advance, and the phase is determined by comparing the normalized intensity of one of the A-phase and B-phase signals with the numerical table of trigonometric functions. This phase detection is performed when the moving scale starts moving,
This is performed twice when stopping. The reason why the A-phase and B-phase signals stored in the memory circuit are normalized is that the normalized strength and the numerical value of sin or cos can be directly corresponded. The numerical table of sin and cos is 0 to π /
The range may be up to 2 in the first quadrant. In the case of a phase exceeding π / 2, the phase value in the first quadrant may be corrected from the relationship of the complement of the trigonometric function. Absolute value of normalized intensity of A-phase signal is 1
, The detection accuracy of the phase obtained from sin is low. In this case, for example, the value of the normalized intensity of the B-phase signal (0
Is compared with the value of cos to detect the phase. By detecting the phase with a resolution of one degree in this way, it is possible to detect one pitch of the grating by dividing it into 360. Next, the dimension is measured with high accuracy from the sum of the moving distance of an integral multiple of the grating one pitch length detected by the counter circuit and the moving distance of one grating length or less detected by the phase.

FIG. 1 is a block diagram showing the configuration of a dimension measuring apparatus using an optical scale according to the present invention. The illumination light source 100 includes a light emitting diode, a white lamp, and the like, and illuminates, via a collimating lens 102, a movable scale 10 that moves with a stylus (not shown) and a fixed scale 11 fixed and installed near the movable scale 10. I do. The moving scale 10 and the fixed scale 11 are called an optical scale. The moving scale 10 has one kind of lattice group 105, and the fixed scale 11 has two kinds of lattice groups 110 and 115. At this time, the lattice groups 105, 110, 11
Each of the gratings that constitutes No. 5 is composed of a white element that transmits light and a black element that does not transmit light. The widths of the white element and the black element are equal and a / 2. One set of white elements and black elements forms one grid, and one pitch length of each grid is a. Usually, one pitch length of the grating is about 10 μm.

Two grid groups 1 of the fixed scale 11
Although the grating pitch lengths of 10 and 115 are equal, the grating of one group has a phase of π with respect to the grating of another group.
/ 2 shifted. That is, the lattice positions are formed shifted from each other by 1/4 pitch. Light receiving section 12
Are the two grid groups 110 and 1 of the fixed scale 11
It comprises two photodetectors 120 and 125 for individually detecting the light transmitted through the photodetector 15. The light receiver 120 outputs an A-phase signal 122. The light receiver 125 has a B-phase signal 127
Is output. The A-phase signal 122 and the B-phase signal 127 are sine wave signals having phases different by π / 2, and are shifted by one pitch of the grating.
This is a signal that forms a cycle. Generally, the A-phase signal 122 and B
The phase signal 127 has a different sine wave intensity due to a difference in the amplification gain of the photoelectric conversion circuit.

The counter section 13 determines the relationship between the leading and lagging phases of the A-phase signal 122 and the B-phase signal 127 to detect the moving direction of the moving scale 10, performs a counter operation, and performs a grid operation of the moving scale 10. Detect the number of movements. Since the A-phase signal 122 and the B-phase signal 127 are out of phase by π / 2, for example, when the A-phase signal is used as a reference, the grid is detected by detecting whether the B-phase signal is moving forward or backward. Can be determined. In the detection at this time, for example, one pulse is generated for each cycle of the sine wave based on the A-phase signal 122, and the number of pulses is counted by a counter circuit. When the moving scale 10 moves upward as indicated by the arrow, an up-counter operation (count increase) is performed, and when it moves downward, a down-counter operation (count decrease) is performed. In this counter operation, a moving distance corresponding to an integral multiple of the number of moving grids is detected.

In order to measure the dimensions precisely, it is important to detect the moving distance of the grating less than one pitch with high resolution.
Therefore, it is necessary to detect the phase of the sine wave signal. The present invention has a configuration in which a phase at a position where the grating is stopped is detected by a numerical calculation process from two-phase sine wave signals of A phase and B phase having phases different by π / 2. The light intensity storage unit 14 stores A
The phase signal 122 and the B-phase signal 127 are A / D converted and stored in the memory circuit as digital data. At this time,
During the period from when the moving scale 10 starts moving to when it stops, the above-mentioned storage is performed during a period of several cycles before the movement starts and immediately before the movement stops. For periods other than the above,
The operation of successively storing new data while resetting the stored contents taken into the memory circuit is repeatedly performed.
The stop of the stylus is detected by detecting a mechanical shock at the time of stop using an electric signal using a piezoelectric element,
It can be determined by detecting that the phases of the phase and phase B signals have been inverted from the normal changes.

Since a sine wave signal has two phases for a specific intensity within one cycle, the phase cannot be detected only from the intensity value. Therefore, for the two times of the position where the grating starts to move and the position where the grating stops, the rough phase quadrant at the stopping position of the grating is determined based on the intensity data of several periods of the sine wave signal, and then the phase is determined. Is detected. The phase quadrant detection unit 15 controls the light intensity storage unit 14 for the above two periods.
Detects the phase quadrant when the grating is stopped based on the light intensities of the A-phase signal 140 and the B-phase signal 145 stored in the storage unit, and includes a normalization processing unit 150 and an intensity code determination unit 155. The normalization processing unit 150 stores the A-phase signal 14
The amplitudes of the 0 and B phase signals 145 are detected, and both are converted to a normalized signal whose amplitude is normalized to ± 1. In addition, the intensity code determining unit 155 determines the sign of the intensity of the standardized A-phase signal and the B-phase signal (four combinations), and determines the phase quadrant from the intensity when the grating is stopped. This phase quadrant is a quadrant from 1 to 4 in units of π / 2. It is necessary to determine the phase quadrant according to the phase advance and delay of the standardized A-phase signal and B-phase signal. This is because the phase quadrants differ depending on the phase relationship between the two-phase signals. Note that the above-described counter unit 13 includes the light receivers 120, 1
Although the example in which the operation is performed using the two-phase signal output from the control unit 25 has been described, the operation may be performed using the standardized two-phase signal shown here.

The trigonometric function table section 16 stores a signal in a specific range within one cycle of the sine wave, for example, s in the range of 0 to π / 2.
The values of in and cos, or one of them, are stored. Each value may be stored as numerical data, or a function form of sin and cos may be stored, and the values of sin and cos may be calculated by a function expansion method. The phase calculation unit 17 calculates the intensity value of the A-phase or B-phase normalized sine wave signal at the grating stop position based on the phase quadrant detected by the phase quadrant detection unit 15 as a reference.
The phase is determined by comparing with the value in. When the normalized absolute value of the intensity of the A-phase signal at the lattice stop position is close to 1, A
If the phase is detected from the value of the sine of the phase signal, the phase detection accuracy is reduced. In this case, since the value of the normalized B-phase signal is close to 0, the phase is detected from the B-phase signal to improve the phase detection accuracy. At this time, when the phase of the A-phase signal is ahead of the phase of the B-phase signal, the phase is calculated from the value of cos of the B-phase signal, and when the phase is opposite, the phase is detected from the value of sin of the B-phase signal. In this way, sin
And cos detection. Further, depending on the phase quadrant, the stored sin and c in the range of 0 to π / 2 are stored.
By using the supplementary relationship of the trigonometric function to the value of os, any phase in the range of 0 to 2π can be detected.

The dimension calculating section 18 detects a moving distance of an integral multiple of the pitch length of the grating 1 from the information on the number of moving grids detected by the counter section 13, and detects the pitch of the grating 1 pitch from the phase information detected by the phase calculating section 17. The moving distance equal to or less than the length is detected, and the dimension is calculated from the sum of the moving distances. In this system, the phase detection resolution becomes the resolution of the dimension measurement, and it is necessary to detect the phase with a resolution of about 1 degree to achieve the submicron measurement resolution. According to the method of the present invention described above, the phase can be detected within an error of one degree by detecting the intensity of the sine wave signal within an error of about 2%.

FIG. 2 shows an example of the relationship between the intensity of the two-phase sine wave signal and the phase quadrant. Waveform 21 is a normalized A-phase signal, waveform 22
Is a normalized B-phase signal, and the phase of the A-phase signal 21 is the B-phase signal 2
This is the case where the phase is more advanced than phase 2. Phase quadrant diagram 23
Is a phase quadrant based on a combination of positive and negative signs of the intensities of the A-phase signal 21 and the B-phase signal 22. For example, A-phase signal 2
If the intensity of 1 is negative and the intensity of B-phase signal 22 is positive, it is determined that the phase of A-phase signal 21 is between 3π / 2 and 2π in the fourth quadrant. As described above, the phase quadrant can be determined from the combination of the four intensity codes, and the phase can be detected with further finer precision from the intensity value at that time. The phase shown in the figure is based on the A-phase signal 21.

FIG. 3 shows that the phase of the A-phase signal 31 is
7 is an example of a phase quadrant when the delay is longer than the above. In this case, the phase quadrant 33 differs from the phase quadrant 23 described above. For example, when the intensity of the A-phase signal 31 is negative and the intensity of the B-phase signal 32 is positive, the phase of the A-phase signal 31 is between π and 3π / 2 in the third quadrant. As described above, since the phase quadrants differ depending on the leading and lagging phases of the two-phase sine wave signal,
It is necessary to determine the advance or delay of the phase in advance.

FIG. 4 shows an example of a two-phase sine wave signal detected with the movement of the grating, and the phase detection method will be described. A
The phase signal 41 and the B-phase signal 42 are generally signals having different amplitudes. The left end of the waveform is the position where the stylus starts moving, and the right end is the position where the stylus stops. Waveforms 43 (A phase), 44 (B
Phase) is a waveform whose amplitude has been normalized to ± 1 over almost one cycle from the position where the stylus has started moving. Waveform 45 (A
Phases) and 46 (phase B) are waveforms whose amplitudes are normalized to ± 1 over almost one cycle from before the position where the stylus stops.
In both cases, the phase of the A-phase signal is ahead of the phase of the B-phase signal.

In the case of the waveforms 43 and 44, it is determined that the phase quadrant at the start of the movement of the left end is in the second quadrant (A-phase signal reference). In this case, since the intensity of the A-phase signal 43 is close to 1,
From the A-phase signal 43, the phase detection accuracy decreases. Therefore, the phase is detected from the intensity of the B-phase signal 44. Waveform 4
In the case of 5, 46, it is determined that the phase quadrant at the rightmost stop position is the fourth quadrant. In this case, the phase is detected from the intensity of the A-phase signal 45. In this way, the phase can be detected on the order of one degree, and one pitch of the grating can be divided into 360. Therefore, the moving distance of one pitch length or less of the grating can be measured with a submicron resolution. Note that a part of the actually measured A-phase signal and B-phase signal may be modulated from an ideal sine wave. In this case, the phase may be detected after correcting the normalized signal with an ideal sine wave.

[0028]

As described above, the dimension measuring apparatus according to the present invention has a configuration in which the phases of two sinusoidal signals having phases different from each other by π / 2 are directly determined by numerical calculation processing. Compared with the conventional method using a four-phase sine wave signal, the present invention uses only a two-phase signal, so that the configuration of the fixed scale and the detection of the optical signal are simplified. In the conventional method of creating a rectangular signal by comparing the intensity levels of four-phase signals, the circuit configuration for creating a rectangular signal is complicated if the phase detection accuracy is to be improved. Since the present invention directly detects the phase of the sine wave signal,
There is no need to create a rectangular signal, and the hardware configuration can be simplified. As described above, according to the present invention, the apparatus is simplified and the cost of the apparatus is reduced as compared with the conventional method.

In the phase detection according to the conventional method, the resolution of the phase detection is about 45 °, and the division between one pitch of the grating is about 8. Therefore, the resolution of dimension measurement is 1 μm,
Submicron resolution was not obtained. In the direct phase detection method according to the present invention, the intensity level of the detected two-phase sine wave signal is normalized to detect the phase quadrant of the grating stop position in advance, thereby selecting a sin or cos table and selecting a phase. Can be detected. Therefore, the reliability of the phase detection is improved, and the phase can be measured with a resolution of 1 °, and the distance between one pitch of the grating can be detected by dividing it by 360. As a result, there is an effect that dimensions can be measured with a resolution in the submicron region.

[Brief description of the drawings]

FIG. 1 is a system block diagram illustrating a configuration example of a dimension measuring device according to an embodiment of the present invention.

FIG. 2 is a first example showing a relationship between a two-phase sine wave signal and a phase quadrant in an embodiment of the present invention.

FIG. 3 is a second example showing a relationship between a two-phase sine wave signal and a phase quadrant in the embodiment of the present invention.

FIG. 4 is a diagram illustrating an example of phase detection in the embodiment of the present invention.

FIG. 5 is a diagram showing a configuration example of a conventional dimension measuring device.

FIG. 6 is a waveform chart showing an example of detecting a phase by a conventional dimension measuring device.

[Explanation of symbols]

 Reference Signs List 10 Moving scale 11 Fixed scale 12 Light receiving unit 13 Counter unit 14 Light intensity storage unit 15 Phase quadrant detection unit 16 Trigonometric function table unit 17 Phase calculation unit 18 Dimension calculation unit 150 Normalization processing unit 155 Intensity code determination unit

Claims (4)

[Claims] 1. A moving scale on which a grating having a constant pitch that moves with a stylus is formed, and a fixed scale fixedly installed near the moving scale with the same pitch as the grating of the moving scale. And a fixed scale composed of two types of lattice groups having phases different by π / 2, and two phases A and B having different phases π / 2 when moved. A moving scale that outputs two sine wave signals, and a counter that counts the number of grid movements according to the relationship between the leading and lagging phases of the A-phase signal and the B-phase signal when the moving scale is moving. A light intensity storage unit for A / D converting the A-phase signal and the B-phase signal for a specific period within a period from when the moving scale starts moving to when the moving scale stops, and stores the light intensity in a memory circuit A phase quadrant detector for detecting a phase quadrant of a sine wave when the moving scale is stopped, based on the light intensities of the A-phase signal and the B-phase signal stored in the light intensity storage unit; A trigonometric function table section for providing values of sin and cos in the range of: and a light intensity of an A-phase signal or a B-phase signal in the phase quadrant detected by the phase quadrant detection section stored in the trigonometric function table section. A phase calculating unit for detecting a phase when the moving scale is stopped as compared with a phase calculating unit for detecting a moving distance of an integral multiple of one pitch length of a grating from a count value counted by the counter unit; A dimension calculating unit for detecting a moving distance of one grating length or less from the phase detected in the step (a), and measuring a dimension from the moving distance of the moving scale. Place. 2. A normalization processing unit for detecting an amplitude of the A-phase signal and the B-phase signal stored in the light intensity storage unit and normalizing the amplitude to a signal having an amplitude of ± 1. When,
By judging the leading and lagging of the mutual phases of the standardized A-phase signal and the B-phase signal, the intensity codes of the standardized A-phase signal and the B-phase signal when the moving scale is stopped are determined. And a strength code judging section for comparison.
2. The dimension measuring device using an optical scale according to claim 1, wherein the phase of the phase signal or the phase B signal is detected in a quadrant in units of π / 2. 3. The trigonometric function table section calculates the sin and cos of a sine wave signal when the phase of the sine wave is within 0 to π / 2.
The dimension measuring apparatus using the optical scale according to claim 1, wherein the value is given by: 4. A phase calculation section, wherein the light intensity of one of the normalized sine wave signals of the A-phase signal and the B-phase signal is 1
2. A dimension measuring apparatus using an optical scale according to claim 1, wherein when the absolute value of the other signal is close to the absolute value, the phase is detected from the value of sin or cos of the other signal.